Embed Size (px)
Citation preview
A KOPONYATRAUMA ÁLTAL KIVÁLTOTT DIFFÚZ AXONÁLIS
ÉS NEURONÁLIS KÁROSODÁS PATHOMECHANIZMUSÁNAK
VIZSGÁLATA ÉS TERÁPIÁS BEFOLYÁSOLÁSÁNAK
LEHETŐSÉGEI
Doktori (PhD) értekezés
Dr. Farkas Orsolya
Témavezető: Dr. Gallyas Ferenc
Dr. Büki András
Elméleti Orvostudományok, Doktori Iskola vezetője: Dr. Szolcsányi János
Kísérletes Neurológia, Programvezető: Dr. Gallyas Ferenc (2006.03.31-ig)
Klinikai Orvostudományok, Doktori Iskola vezetője: Dr. Nagy Judit
Klinikai Idegtudományok, Programvezető: Dr. Komoly Sámuel
Pécsi Tudományegyetem, Általános Orvostudományi Kar
Pécs, 2007
TARTALOMJEGYZÉK
RÖVIDÍTÉSEK ................................................................................................................................. 4
AZ ÉRTEKEZÉST KÉPEZŐ SAJÁT KÖZLEMÉNYEK LISTÁJA ............................................... 5
1. BEVEZETÉS ................................................................................................................................. 7
1.1. A baleseti agysérülések epidemiológiai jelentősége ....................................................... 7
1.2. A baleseti agysérülések osztályozása.............................................................................. 8
2. IRODALMI HÁTTÉR................................................................................................................. 10
2.1. A diffúz axonális károsodás (diffuse axonal injury, DAI)............................................ 10
2.1.1. Az axonduzzadás/axonballon-képződés (AD/B) ......................................... 11
2.1.1.1. Az axonduzzadás/axonballon-képződés felfedezése....................... 11
2.1.1.2. Az axonduzzadás/axonballon-képződés mechanizmusa................. 12
2.1.2. Az ultrastrukturális (neurofilament) kompakció (UC, NFC)....................... 13
2.1.3. A DAI terápiájának, illetve klinikai-kémiai diagnosztizálhatóságának
biokémiai alapjai .......................................................................................... 14
2.1.3.1. Az axolemmális permeabilitási zavar okozta Ca2+ beáramlás ........ 15
2.1.3.2. A calpain aktiváció és a calpain-mediált spektrin bontás ............... 15
2.1.3.3. A mitokondriális károsodás és a caspase-3 aktiváció ..................... 16
2.1.4. A DAI terápiás befolyásolását célzó vizsgálatok........................................... 17
2.2. A diffúz neuronális károsodás....................................................................................... 18
2.2.1. A „sötét”-idegsejt képződés ........................................................................... 19
2.2.2. A neuronális mechanoporáció........................................................................ 20
3. CÉLKITŰZÉSEK .......................................................................................................... 21
4. ANYAGOK ÉS MÓDSZEREK .................................................................................................. 24
4.1. A diffúz agysérülés modellezése................................................................................... 24
4.1.1. Az impact-accelerációs koponyatrauma modell ............................................ 24
4.1.2. Szelektív ultrastrukturális kompakciót kiváltó koponyatrauma modell ........ 25
4.2. Szövettani feldolgozás .................................................................................................. 25
4.2.1. Az argyrophil-III típusú ezüstözési eljárás .................................................... 26
4.2.2. Immunhisztokémiai technikák ....................................................................... 26
4.2.3. Elektronmikroszkópos előkészítés ................................................................. 27
4.3. Western blot .................................................................................................................. 27
4.4. Permeabilitás-vizsgálat ................................................................................................. 28
2
5. EREDMÉNYEK ÉS KÖVETKEZTETÉSEK............................................................................. 29
5.1. A traumás axonkárosodás vizsgálata ............................................................................ 29
5.1.1. Az ultrastrukturális axon-kompakció vizsgálata............................................ 29
5.1.1.1. Az ultrastrukturális axon-kompakció és az argyrophil
axonkárosodás kapcsolata ................................................................ 29
5.1.1.2. Ultrastrukturális kompakció szelektív előidézésére alkalmas
koponyatrauma modell kidolgozása................................................. 31
5.1.1.3. A kompaktálódott axonok sorsának vizsgálata szelektív
ultrastrukturális kompakciót kiváltó koponyatrauma modellben..... 32
5.1.1.4. Az ultrastrukturális axon- illetve idegsejt-kompakció
mechanizmusának tisztázása............................................................ 34
5.1.2. Terápiás beavatkozások tesztelése ................................................................. 37
5.1.2.1. Egy sejt-permeábilis calpain-inhibitor, az MDL-28170, hatásának
vizsgálata a diffúz axonkárosodásra................................................. 37
5.1.2.2. A trauma előtt illetve után iv- illetve icv-adott PACAP
hatásának vizsgálata; valamint a hatásos dózis megállapítása......... 38
5.1.2.3. A terápiás ablak meghatározása icv PACAP-kezelés esetén .......... 40
5.2. A neuronális mechanoporáció, a resealing és a CMSP vizsgálata................................ 41
5.3. A calpain- és caspase-aktiválódás szerepe a humán traumás agykárosodásban ........... 43
6. A TÉMÁBAN ELÉRT, NEMZETKÖZI-JELENTŐSÉGŰ EREDMÉNYEK ........................... 45
IRODALOMJEGYZÉK .......................................................................................................... 49
KÖSZÖNETNYILVÁNÍTÁS ......................................................................................................... 59
AZ ÉRTEKEZÉST KÉPEZŐ SAJÁT KÖZLEMÉNYEK.............................................................. 60
3
RÖVIDÍTÉSEK
AD/B Axonduzzadás/axonballon-képződés
APP Amyloid Precursor Protein
CMSP Calpain-mediált spektrin-bontás (Calpain-mediated Spectrin Proteolysis)
CsA Cyclosporin A
CSF Cerebrospinális folyadék (liquor)
CSpT Kortikospinális pálya (Corticospinal Tract)
DAB Diamino-benzidin
DAI Diffúz axonális károsodás (Diffuse Axonal Injury)
EM Elektronmikroszkóp
GCS Glasgow kóma skála (Glasgow Coma Scale)
GOS Glasgow kimenetel skála (Glasgow Outcome Scale)
IAT Axonális transzport-zavar (Impaired Axonal Transport)
ICP Intrakraniális nyomás (Intracranial Pressure)
icv intracerebroventrikuláris
iv intravénás
KIR Központi Idegrendszer
MLF Mediális hosszanti köteg (Medial Longitudinal Fascicle)
MPT Mitokondriális permeabilitási átmenet (Mithochondrial Permeability
Transition)
NF Neurofilament
NFC Neurofilament kompakció (Neurofilament Compaction)
PACAP Hipofízis adenilát-cikláz aktiváló polipeptid (Pytuitary Adenilate Cyclase
Activating Polypeptide)
SBDP Spektrin degradációs termék (Spectrin Breakdown Product)
TAI Traumás axonkárosodás (Traumatic Axonal Injury; a DAI állatkísérletes
megfelelője)
UC Ultrastrukturális kompakció (Ultrastructural Compaction)
4
AZ ÉRTEKEZÉST KÉPEZŐ SAJÁT KÖZLEMÉNYEK LISTÁJA
I. Gallyas F, Farkas O, Mazlo M. Traumatic compaction of the axonal cytoskeleton
induces argyrophilia: histological and theoretical importance. Acta Neuropathol (Berl).
2002 Jan;103(1):36-42.
IF: 2.283
II. Pal J, Toth Z, Farkas O, Kellenyi L, Doczi T, Gallyas F. Selective induction of
ultrastructural (neurofilament) compaction in axons by means of a new head-injury
apparatus. J Neurosci Methods. 2006 Jun 15;153(2):283-9. Epub 2005 Dec 27.
IF: 1.784
III. Gallyas F, Pal J, Farkas O, Doczi T. The fate of axons subjected to traumatic
ultrastructural (neurofilament) compaction: an electron-microscopic study. Acta
Neuropathol (Berl). 2006 Mar;111(3):229-37.
IF: 2.527
IV. Gallyas F, Farkas O, Mazlo M. Gel-to-gel phase transition may occur in
mammalian cells: Mechanism of formation of "dark" (compacted) neurons. Biol Cell.
2004 May;96(4):313-24.
IF:2.233
V. Farkas O, Lifshitz J, Povlishock JT. Mechanoporation induced by diffuse
traumatic brain injury: an irreversible or reversible response to injury? J Neurosci. 2006
Mar 22;26(12):3130-40.
IF: 7.506
VI. Buki A, Farkas O, Doczi T, Povlishock JT. Preinjury administration of the calpain
inhibitor MDL-28170 attenuates traumatically induced axonal injury. J Neurotrauma. 2003
Mar;20(3):261-8.
IF: 2.587
VII. Farkas O, Tamas A, Zsombok A, Reglodi D, Pal J, Buki A, Lengvari I,
5
Povlishock JT, Doczi T. Effects of pituitary adenylate cyclase activating polypeptide in a
rat model of traumatic brain injury. Regul Pept. 2004 Dec 15;123(1-3):69-75.
IF: 2.531
VIII. Tamas A, Zsombok A, Farkas O, Reglodi D, Pal J, Buki A, Lengvari I,
Povlishock JT, Doczi T. Postinjury administration of pituitary adenylate cyclase
activating polypeptide (PACAP) attenuates traumatically induced axonal injury in rats. J
Neurotrauma. 2006 May;23(5):686-95.
IF: 2.574
IX. Farkas O, Polgar B, Szekeres-Bartho J, Doczi T, Povlishock JT, Buki A. Spectrin
breakdown products in the cerebrospinal fluid in severe head injury--preliminary
observations. Acta Neurochir (Wien). 2005 Aug;147(8):855-61.
IF:1.064
6
1. BEVEZETÉS
1.1. A baleseti agysérülések epidemiológiai jelentősége
A balesetek során bekövetkező agysérülés gyakorisága a motorizációval
párhuzamosan emelkedik; a fejlett ipari országokban a 40 év alatti lakosság körében a
vezető halálokot képezi. A WHO felmérései szerint 10 millió baleseti agysérülésből
következő haláleset, vagy kórházi ápolás történik világszerte minden évben, és a világon
kb. 57 millió ember szenvedett élete során legalább egyszer traumás agysérülést [64]. A
National Institute of Health felmérései alapján az USA-ban 1.4 millió baleseti agysérülés
történik évente, ebből 50 000 halálos kimenetelű, 235 000 kórházi kezelést igényel, és 5.3
millió ember él baleseti agysérülések következtében kialakult tartós korlátozottsággal [64].
Követéses vizsgálatok igazolták, hogy a baleset után 1-3 évvel a normál populációhoz
képest a baleseti agysérülést szenvedettek körében 1.8-szoros a rizikó alkohol-abúzus
kialakulására [50], míg 11-szeres epilepszia (P. L. Ferguson, written communication,
February 2006), 7.5-szörös halál, 1.5-szörös depresszió [49], illetve 2.3-4.5-szörös
Alzheimer kór előfordulására [103].
A „silent epidemic”-ként emlegetett népbetegség, a baleseti agysérülés,
Magyarországon is hasonló nagyságrendű epidemiológiai probléma. Egy 2002-ben
készült, a 2001-es év adataira vonatkozóan retrospektív, valamint egy három-hónapos
időszakra vonatkozóan prospektív esetelemző tanulmány [24] szerint hazánkban a
kórházba kerülő koponya-agysérültek évi száma közel 14 000, ebből 71.3% enyhe, 19.4%
közepes, és 9.4% a súlyos kategóriába tartozik. A sérültek 67%-a férfi, 33%-a nő, 78%-a
60 év alatti, 36%-a 40 éves kor alatti. A prospektív adatok szerint a kórházakba súlyosként
(GCS≤8) került koponya-agysérültek 55%-a, míg a felvételkor közepesen súlyosnak
(GCS=9-12) illetve enyhének (GCS=13-15) ítélt, de az első 24 órában tovább súlyosbodó
állapotú sérültek 35%-a hal meg az akut ellátás során; ami jóval magasabb, mint a világ
fejlett részén közölt halálozási adatok. Ennek hátterében a prehospitális és a hospitális
ellátás közötti kommunikációs zavar, a definitív ellátásig, illetve az első CT-vizsgálatig
eltelő extrém hosszú idő, továbbá az állhat, hogy a magyar és a nemzetközi ellátási
irányelvek nem kerülnek széles körben alkalmazásra. A túlélők közül az elbocsátáskor
40% tartós vegetatív állapotú, vagy súlyos maradványtüneteket mutat, ami szintén jóval
magasabb, mint a nemzetközi irodalomból ismert hasonló adatok. Az akut ellátást
7
túlélőknek csak 45%-a gyógyul enyhe maradványtünetekkel vagy maradványtünetek
nélkül. A túlélők hosszú távú életminőségéről nem állnak rendelkezésre pontos adatok, de
becslések szerint hazánkban a súlyos koponyasérülteknek csak mintegy negyede
illeszkedhet vissza a társadalomba [24].
A fenti ijesztő epidemiológiai adatok hangsúlyozzák a baleseti agysérülések
klinikai és társadalmi jelentőségét, és ezáltal a koponyatrauma különféle vonatkozásai
irányába történő kutatások fontosságát.
1.2. A baleseti agysérülések osztályozása
A koponyatrauma által kiváltott agysérülések igen heterogén betegségcsoportot
képviselnek. Osztályozásuk számos szempont szerint történhet. A klinikai kép súlyossága
alapján megkülönböztetünk enyhe, középsúlyos és súlyos sérüléseket. A koponyatrauma
mechanizmusa alapján beszélhetünk nyílt illetve zárt sérülésekről. A koponyatrauma által
okozott intrakraniális elváltozások lehetnek elsődlegesek, vagyis a mechanikai behatás
közvetlen következményei, illetve másodlagosak; ezek a koponyatrauma
következményeként kialakult különböző agyi szövődményekből eredeztethetők [45, 85,
119]. Az osztályozás szempontjait – a jelen dolgozat tárgyát képező kutatási területeket
vastag betűvel kiemelve – az alábbiakban foglalhatjuk össze:
I. A baleseti agysérülések osztályozása a klinikai kép súlyossága szerint (Glasgow
Coma Scale alapján):
Enyhe: GCS = 13-15
Közepesen súlyos: GCS = 9-12
Súlyos: GCS ≤ 8
II. A baleseti agysérülések osztályozása a koponyatrauma mechanizmusa alapján:
1. Nyílt vagy penetráló: a dura mater sérül
2. Zárt vagy tompa: a dura mater intakt marad
A.) Statikus: lassú erőbehatás
B.) Dinamikus: hirtelen erőbehatás
a.) Impakt típusú: hirtelen tompa ütés a fejre, amely
ennek hatására alig mozdul el; általában kontakt erők
okozzák az agysérülést
b.) Impulzív típusú: a fej hirtelen mozgásba kerül vagy
8
mozgásból hirtelen megáll; általában tehetetlenségi
erők okozzák az agysérülést (acceleráció-
deceleráció)
III. A baleseti agysérülések osztályozása a koponyatrauma hatására kialakuló
morfológiai elváltozások alapján:
1. A koponyatrauma hatására kialakuló elsődleges elváltozások
A.) Fokális (gócos)
Laceráció
Kontúzió
Intrakraniális vérzések (állományi, epidurális, subdurális
vagy subarachnoidális vérzés
B.) Diffúz
Axonális károsodás
Axonduzzadás/axonballon-képződés
Ultrastrukturális (Neurofilament) kompakció
Neuronális károsodás
2. A koponyatrauma hatására kialakuló másodlagos patológiás állapotok és
következményeik (fokális vagy diffúz)
Ödéma
Ischemia
Emelkedett intrakraniális nyomás
A disszertáció első része e rendkívül összetett kórkép-csoportnak csak egyetlen
alcsoportjával, az elsődleges diffúz morfológiai károsodásokkal foglalkozik, és
vizsgálataink elsősorban állatkísérletes aspektusokra szorítkoznak. Kutatásaink másik
része pedig – állatkísérletes adatokra támaszkodva – a humán baleseti agysérülések
közvetlen vizsgálatát tűzte ki céljául.
A jelen dolgozat a témához kapcsolódó szakirodalom összefoglalása (2. fejezet) és
a célkitűzések ismertetése (3. fejezet) után röviden tárgyalja a kísérletek során alkalmazott
módszereket (4. fejezet), a kísérletek eredményeit és legfontosabb következtetéseit (5.
fejezet), külön kiemelve a témában elért legfontosabb új eredményeket (6. fejezet). Az
egyes kísérletekben alkalmazott módszerek, eredmények és következtetések részletesen a
kísérletekből született és a jelen értekezés alapját képező különlenyomatokban kerülnek
ismertetésre.
9
2. IRODALMI HÁTTÉR
2.1. A diffúz axonális károsodás (diffuse axonal injury, DAI)
A diffúz axonális károsodás (diffuse axonal injury, DAI) a koponyatrauma hatására
kialakult primer axonális elváltozások összessége, melyek az agy egész állományára
kiterjedve, ép axonok között elszórtan figyelhetők meg, egyébként többnyire ép
parenchimális környezetben. A DAI klinikai jelentőségét alátámasztó adatok szerint e
kórkép felelős 50 %-ban a tartós tudatzavarért, illetve 35 %-ban a mortalitásért, a nem-
térfoglaló jellegű koponyasérülésekben [39]. A diffúz agysérülések kialakulásában
acceleráció-deceleráció hatására ébredő nyíróerők játszanak szerepet, amelyek tipikusan
motorbicikli, személygépkocsi, vagy gázolásos balesetek során fordulnak elő [1]. A
kórkép klinikai megjelenésére a tudatzavart magyarázó térfoglaló elváltozás vagy
metabolikus zavar nélkül fennálló komatózus tudatállapot a jellemző [39, 41, 44]. A CT-n
csak kb. 20 %-ban detektálható elváltozás kicsi, petechiális vérzések formájában,
elsősorban a szürke- és a fehérállomány határán, illetve a corpus callosumban és a hosszú
pályák agytörzsi szakaszán [3]. A korai stádiumban végzett MRI a CT-nél hatékonyabb az
elváltozások kimutatására. A nem-haemorrhagiás léziók a T2-súlyozott és a
protondenzitású képeken kis, ovális vagy kerek hiperintenzív jelek formájában láthatók,
míg a vérzéses léziók centrális területén hipodenzitás figyelhető meg. Ezek az
elváltozások a krónikus fázisban nagyon alacsony intenzitású léziókként jelennek meg,
mert hemosziderint tartalmaznak [47]. Annak ellenére, hogy a fejlettebb MRI eljárások
(diffúzió-súlyozott képalkotás, diffúziós tensor képalkotás 3 dimenziós tractográfiával)
megkönnyítik a kimutatást [3], a diffúz axonális károsodás teljes bizonyossággal még ma
is elsősorban szövettani vizsgálattal (általában post mortem) diagnosztizálható.
A diffúz axonális károsodásnak két, morfológiailag jól elkülöníthető megjelenési
formája ismert: az axonduzzadás/axonballon-képződés és az ultrastrukturális
(neurofilament) kompakció. A DAI-kutatások kezdetben az axonduzzadás vizsgálatára
irányultak; az ultrastrukturális kompakció jelensége később került csak leírásra [98]. A
klasszikus elképzelés szerint e két morfológiai jelenség kiváltásában ugyanazok a
tényezők játszanak szerepet, ezáltal ugyanazokat az axonokat érintik; a legújabb nézetek
szerint azonban – a legtöbb esetben – a kétféle elváltozás két jól-elkülöníthető axon-
populációban figyelhető meg [73, 135].
10
Meg kell jegyezni, hogy a traumás axonkárosodás kutatások túlnyomó többsége
különböző állatkísérletes modelleket alkalmaz a kórkép feltérképezésére és
befolyásolásának vizsgálatára. Ezek a modellek – természetesen – nem képesek a humán
viszonyokat és ezáltal a humán DAI teljes spektrumát, kiterjedését és időbeni lefolyását
maradéktalanul reprodukálni, ezért helyesebb az állatkísérletekben modellezett
axonkárosodást traumás axonkárosodásként (traumatic axonal injury, TAI) említeni, a
DAI kifejezést pedig a humán esetekre alkalmazni [80]. Korlátai ellenére azonban az
állatkísérletes modellek egyedülálló mértékben járultak hozzá a humán DAI kialakulása
alapvető elemeinek feltérképezéséhez, segítségükkel ugyanis a humán DAI számos
aspektusa – fokális elváltozásokkal kísérve vagy anélkül – megbízhatóan vizsgálható.
2.1.1. Az axonduzzadás/axonballon-képződés (AD/B)
2.1.1.1. Az axonduzzadás/axonballon-képződés felfedezése
A diffúz axonkárosodás fogalma régóta ismert az irodalomban. Több, mint 50 éve
Strich és munkatársai fedezték fel és írták le, mint a súlyos traumás agysérülések
következetes velejáróját. Súlyos koponya-agysérülésben meghalt betegek post mortem
vizsgálata során – számottevő fokális léziók hiányában az etiológiát kutatva – Strich
ballonszerű axontágulatokat figyelt meg a fehérállományban „normális” axonok között
elszórtan, amely tágulatoktól disztális axonszakasz a túlélési idő növekedésével Waller-
féle degenerációt és myelin-degradációt szenvedett [137, 138]. Ezt az axonballon-
képződést és az azt követő fehérállományi degenerációt Strich a trauma pillanatában
kialakult azonnali axonszakadás következményének tartotta. Hipotézise szerint a trauma
által okozott mechanikus erők az axon szakadásához, az érintett proximális axonvég
következményes retrakciójához és az axoplasma kiboltosulásához vezetnek, létrehozva
ezzel a „retrakciós axonballont” a kórkép korai szakaszában; ehhez társul a későbbiekben
a disztális axonszakasz Waller-féle degenerációja [137]. E korai megfigyelések óta számos
kutató megerősítette, hogy a traumás fehérállományi károsodás a koponyasérülések
konzekvens velejárója azokban az esetekben, amelyekben térfoglaló agyi elváltozások
nem mutathatóak ki, a beteg mégis klinikailag súlyos állapotban van. Strich a megfigyelt
elváltozásokat még mint a fehérállomány diffúz degenerációját vagy traumás idegrost
szakadást említi. A kórkép ma is használatos elnevezése, a diffúz axonális károsodás,
Adamstől származik [1], és arra utal, hogy a károsodott axonok elszórtan, ép axonok
11
között az agy egész állományára kiterjedve fordulnak elő. Adams és kollégái
hangsúlyozták Strich óta először, hogy a kórkép kialakulásában az axonkárosodás a
koponyasérülés elsődleges következménye, szembeszállva ezzel sok támadással, amelyek
az axonális károsodást másodlagos – ödéma, hypotensio és/vagy hypoxia következtében
kialakult – elváltozásoknak tartották [44]. Adams és munkatársai ugyancsak úttörő munkát
végeztek a humán diffúz axonkárosodás anatómiai lokalizációjának, illetve súlyossági
fokozatainak leírásában [1, 2]. E szerint a DAI típusosan a féltekei fehérállományban, a
corpus callosumban, az agytörzsi hosszúpályákban, illetve kisebb gyakorisággal a
kisagyban fordul elő fokális károsodás nélkül (1-es fokozat) vagy fokális léziókkal a
corpus callosumban (2-es fokozat) és a rostralis agytörzsben (3-as fokozat).
A kórkép Strich és Adams szerinti pathomechanizmusa – azaz, hogy az
axonszakadás a trauma pillanatában azonnal bekövetkezik és emiatt terápiásan nem
befolyásolható – évtizedekig elfogadott volt, ami a kórkép kutatását és kezelését hosszú
ideig hátráltatta [6].
2.1.1.2. Az axonduzzadás/axonballon-képződés mechanizmusa
Az axonduzzadás/axonballon-képződés kutatásának következő mérföldkövét
Povlishock és munkatársai 1983-ban közölt munkája jelentette, amely – szemben a
korábban elfogadott elképzeléssel – kimutatta, hogy a DAI nem a trauma pillanatában
azonnal és véglegesen kialakuló axonszakadásként jelentkezik, hanem a károsodott
axonok döntő többségében egy, időben fokozatosan progrediáló folyamatként [106]. Azóta
ez az elmélet számos állatfajban és traumamodellben, valamint humán szövetmintákon is
igazolódott [6, 7, 17, 19, 30, 31, 39, 42, 43, 84, 98, 108-111, 128, 129].
A TAI kialakulásának Povlishock-féle elmélete szerint ezen axonális károsodást a
középsúlyos illetve a súlyos koponyatrauma által kiváltott nyíróerők indítják el, amelyek
számos fény- és elektronmikroszkóp segítségével vizsgálható intraaxonális elváltozást
eredményeznek, mégpedig: A károsodott axonszakaszok a sérülést követően azonnal (<5
perc) fokális axolemmális permeabilitási zavart mutatnak („mechanoporáció”), amely
folyamatot az szemlélteti és igazolja, hogy a károsodott axonok nagy molekulasúlyú
anyagokat (pl. tormagyökér-peroxidázt, vagy dextránokat) vesznek fel, amelyek az ép
axolemmán keresztül az ép axonokba illetve ép neuronokba nem tudnak bejutni [98, 99,
112-114, 136]. Az axolemma fokális sérülését egyéb morfológiai jellemzők kísérik:
mitokondrium duzzadás [77, 98, 99], neurotubulus veszteség [78, 79, 99], neurofilament-
módosulás [51, 52, 93, 99, 112, 113, 134], axonális transzportzavar [99, 112, 114, 136],
12
amely utóbbi az előre irányuló intraaxonális transzport – egy ponton való – fokális
leállásában nyilvánul meg, és az ilyen módon szállítódó sejtalkotók illetve egyéb anyagok
felhalmozódásához (organellum-akkumuláció) és következményes axonduzzadáshoz vezet
[75, 79, 106, 107]. A traumától eltelt idő előrehaladtával az axonduzzadás egyre nagyobb
mértékű lesz, és végül az érintett axonszakasz ballon formájában lefűződik, létrehozva
ezzel a kórkép korai leírásaiból ismert proximális axonballonokat, míg a disztális
axonszakasz Waller-féle degenerációt szenved [80, 106, 109, 110, 114, 115]. Az axonális
elváltozások kezdetétől az axonballon kialakulásáig illetve az axonszakadásig eltelt idő a
trauma súlyossága és a species függvényében változik; emberben tipikusan órákban,
napokban mérhető [128, 129, 148]. Ez a tény elvi lehetőséget biztosít az axonszakadás
terápiás megelőzésére (terápiás ablak). Hangsúlyozni kívánjuk, hogy a TAI-ra jellemző
fenti morfológiai elváltozások a humán DAI későbbi tanulmányozása során szintén
leírásra kerültek [5, 6, 19, 42, 43, 92, 128, 129].
A szóban-forgó AD/B kimutatása az előre irányuló axontranszport által szállítódó
és annak károsodásakor fokálisan felhalmozódó anyagok elleni antitestek alkalmazásával
végzett immunhisztokémiai módszerekkel lehetséges. A legelterjedtebb ezek közül a béta
amyloid precursor protein (APP) felhalmozódásának kimutatása, amely viszonylag
egyszerű, ezért a DAI diagnózisának legszélesebb körben alkalmazott módjává vált [42,
67, 128, 129], annak ellenére, hogy a legújabb vizsgálatok szerint valószínűleg alulbecsüli
a DAI mértékét [73, 115, 135].
2.1.2. Az ultrastrukturális (neurofilament) kompakció (UC, NFC)
A „nagy-múltú” axonballon képződéssel ellentétben az intraaxonális citoszkeleton
kompakciója (a neurofilamentumok közötti távolságok lényeges csökkenése,
neurofilament kompakció, NFC), a mikrotubulusok számának csökkenése, a
neurofilamentumok oldalkarjainak módosulása csak a 90-es években kerültek leírásra [19,
43, 51, 52, 98, 99, 113, 150]. Kezdetben úgy gondolták, hogy a citoszkeletális
elváltozások ugyanazokat az axonokat érintik, mint az AD/B. A hipotézis szerint az NFC
következményeként áll le az axonduzzadáshoz vezető előreirányuló intraaxonális
transzport [79, 80, 98, 99, 112]. Ezt támasztották alá azok a kísérletek is, melyek szerint az
intraaxonális transzportzavart mutató APP-pozitív axonok és a neurofilament kompakciót
mutató RMO-14-pozitív axonok ugyanazon anatómiai régiókban fordulnak elő és
befolyásolásuk ugyanazon intervenciókkal volt lehetséges [8, 9, 58, 94, 95]. Mára azonban
13
már világossá vált, hogy TAI két jól elkülöníthető morfológiai jellemzője, tehát az AD/B,
illetve az NFC két különböző axonpopoulációt érint az esetek túlnyomó többségében, és
csak kis részben figyelhető meg egyazon károsodott axonon belül, leginkább a lemniscus
medialis nagy kaliberű axonjaiban, és itt is csak immunhisztokémiai módszerekkel [73,
135]. Az NFC kimutatására a közepes méretű NF alegység (NF-M) elleni RMO-14
antitestet alkalmazzák, amely az axonális citoszkeleton strukturális átrendeződése
folyamán (a calpain által mediált strukturális fehérjebontás következményeként,
részletesen lásd később) szabaddá vált NF-M „rod domén”-hez képes kötődni [65, 113].
Az immunhisztokémiai technikát alkalmazó vizsgálatok minimum 15 perccel a
koponyatrauma kiváltása után tudtak axon-kompakciót kimutatni. Ugyanakkor egy
speciális ezüstözési eljárással a trauma után azonnal leölt, illetve post mortem
koponyatraumát szenvedett állatokban is kimutatható egy – hosszú axonszakaszok
megvékonyodásával járó – argyrophil károsodás, amely fénymikroszkópos megjelenése
lehet az NFC-nek [33-35]. E jelenség más megvilágításba helyezi az NFC kialakulásának
mechanizmusában feltételezett enzimatikus reakciók kizárólagos szerepét.
Az ultrastrukturális kompakció jelensége tehát ismert, azonban kialakulása pontos
mechanizmusáról, a humán neuropathológiában betöltött morfológiai és neurológiai
szerepéről, illetve a kompakciót mutató axonok sorsáról viszonylag keveset tudunk.
Itt kívánjuk megjegyezni, hogy a nemzetközi szakirodalom az ultrastrukturális
kompakció megjelölésére axonok esetén a neurofilament kompakció kifejezést használja.
Ez a terminus technikus idegsejtek és dendritek esetén azonban nem használható. Ezért a
továbbiakban axonális kompakció esetére fenntartjuk az NFC megjelölést, hangsúlyozva,
hogy ez megegyezik az általunk egyébként használt UC-vel.
2.1.3. A DAI terápiájának, illetve klinikai-kémiai diagnosztizálhatóságának
biokémiai alapjai
A fentebb részletezett morfológiai vizsgálatok kiderítették, hogy a koponyatrauma
által kiváltott axonszakadás nem azonnal megy végbe, hanem a trauma pillanatától
számított órák, napok múlva, ami lehetőséget biztosít a terápiás megelőzésre. Ehhez a
morfológiai elváltozások hátterében lejátszódó biokémiai folyamatok megismerése
elengedhetetlenül szükséges.
14
2.1.3.1. Az axolemmális permeabilitási zavar okozta Ca2+ beáramlás
A mára elfogadottá vált nézet szerint, az axolemma sérülése (mechanoporáció) és a
következményes permeabilitási zavar a trauma által kiváltott erők hatására azonnal
bekövetkezik [115, 133]. Ez az érintett axonokban az ion-homeosztázis felborulásához
vezet, és lehetővé teszi, hogy az extracelluláris térből Ca2+-ionok áramoljanak az
intraaxonális térbe. A megnövekedett intraaxonális Ca2+-koncentráció különböző
fehérjebontó enzimrendszerek aktiválódásához vezet, illetve a mitokondriumokban
akkumulálódva azok duzzadását, funkcionális károsodását okozhatja. Az intracelluláris
térben megnövekedett Ca2+-koncentráció jelenlétére eleinte csak közvetlen bizonyíték
létezett a Ca2+-függő enzimrendszerek aktiválódásának bizonyításával, mára azonban már
közvetlenül is igazolt a folyamat lejátszódása [149].
2.1.3.2. A calpain aktiváció és a calpain-mediált spektrin-bontás (calpain mediated
spectrin proteolysis, CMSP)
A Ca2+-dependens folyamatok közül egyik legfontosabb a calpain aktiválódása,
amelynek számos idegrendszeri kórkép kialakulásában – köztük az utóbbi évek kutatásai
alapján a traumás agysérülések fokális [46, 54, 87, 88, 100-102, 105, 122] illetve diffúz [8,
10, 61, 62, 125] típusában is – fontos szerepe van. A calpain a cisztein proteázok
családjába tartozó enzim. Több fajtája ismert, ezek közül a calpain-1 vagy µ-calpain és a
calpain-2 vagy m-calpain a központi idegrendszerben mindenhol előfordul [147]. Számos
szubsztrátja ismert (pl. citoszkeletális fehérjék) [15, 18, 147], és élettani szerepe van az
idegsejt fejlődés és a szinaptikus struktúra kialakításában [18]; túlműködése azonban a sejt
számára végzetes lehet [18, 147]. Szerepe elsősorban a necrotikus sejthalál kialakulásában
nagy jelentőségű, de a legújabb kutatások szerint egyes apoptotikus folyamatok során is
aktiválódik [86].
Az agyi elváltozások vizsgálata során legnagyobb figyelmet kapott calpain-
specifikus folyamat a citoszkeletális fehérjék, elsősorban a spektrin specifikus bontása
(calpain-mediated spectrin proteolysis, CMSP). E folyamat eredményeként a
citoszkeletont alkotó nem-erythroid alfa-II spektrin calpain-specifikus 145 kD és 150 kD
nagyságú lebontási termékei (spectrin breakdown products, SBDP) a bontás helyén
felszaporodnak, és specifikus antitestek segítségével kimutathatóak [120, 130].
Traumás axonkárosodás során a sérült axolemmán beáramló Ca2+ a CMSP
aktiválódásához vezet. A spektrin bontása kezdetekben csak a subaxolemmális térre
korlátozódik (15-30 perc); a subaxolemmális citoszkeletális hálózat emésztődése később
15
további permeabilitási zavart okoz, és órák alatt egy generalizált intraaxonális
proteolitikus kaszkád aktiválásához vezet, ami végső soron az axon visszafordíthatatlan
károsodását eredményezi [10, 11]. A calpain-specifikus spektrin-degradációs termékek
kimutathatóak a sérülés helyén a fokális koponyatrauma állatkísérletes modelljeiben [46,
53, 87, 100, 122], diffúz traumás agysérülések esetén a károsodott axonokban [10, 62, 63,
125], fokális traumát szenvedett állatok cerebrospinális folyadékában [102], illetve
traumás agysérülés következtében elhunyt egyének agyában a corpus callosum területén
[83].
Az a tény, hogy a spektrin-degradációs termékek nemcsak a trauma helyén, hanem
a cerebrospinális folyadékban is kimutathatóak, megteremti a lehetőséget olyan potenciális
biomarkerek azonosítására koponyatraumát szenvedett betegekben, melyek specifikusan
kapcsolódnak a baleseti agysérülések kialakulása során lejátszódó patobiokémiai
folyamatokkal, és így potenciálisan összefüggésben lehetnek az agysérülés klinikai
képével, például a súlyosságával vagy kimenetelével.
2.1.3.3. A mitokondriális károsodás és a caspase-3 aktiváció
A megnövekedett intraaxonális Ca2+-koncentráció a fehérjebontó folyamatok
aktiválása mellett további káros folyamatokat indukálhat. Ismert, hogy az intraaxonális
mitokondriumok fiziológiás körülmények között nem akkumulálják a Ca2+-ionokat. Ha
azonban az intraaxonális Ca2+-koncentráció meghaladja az 5 µM-t a Ca2+-többlet a
mitokondriumokban szekvesztrálódik [80]. Ez a folyamat a mitokondriális
transzmembrán-potenciál összeomlásához és az ún. mitokondriális tranziciós
permeabilitási (mitochondrial permeability transition, MPT) pórus kinyílásához vezet,
lehetővé téve víz beáramlását, a mitokondrium duzzadását és szétrepedését, valamint a
lokális energiaháztartás zavarát [94, 80, 151]. A mitokondriumokban akkumulálódó Ca2+
nemcsak a lokális energiaháztartást veszélyezteti, hanem e folyamat során lehetővé válik,
hogy a károsodott mitokondriális membránon keresztül citokróm-c áramoljon a
citoplazmába, amely a caspase enzimcsalád aktiválódásához vezet [11, 12, 60]. A caspase-
ok – a calpainhoz hasonlóan – a cisztein proteázok családjába tartoznak és az apoptotikus
sejthalál szabályozásában és végrehajtásában játszanak szerepet [147]. A caspase-3, a
közös apoptózis végrehajtó enzim, képes a calpain hatásának felerősítésére a calpain sejten
belüli inhibitorának, a calpastatinnak gátlásával [145], illetve irreverzibilisen – és a
calpainhoz hasonlóan specifikusan – hasítja a spektrint [146]. A mitokondriumok
károsodása következtében kialakult citokróm-c felszabadulás, illetve a calpain és a
16
caspase-3 egyidejű aktiválódása TAI során is megfigyelhető [11]. A caspase-specifikus
spektrin-degradációs termékek (egy 150 kD és a 120 kD molekulasúlyú SBDP) specifikus
antitestek segítségével szintén kimutathatóak a koponyatrauma in vitro modelljében [101],
továbbá állatkísérletesen az agyszövetben mind fokális, mind diffúz traumás agysérülések
esetén [11, 63, 100], illetve állatkísérletekben a cerebrospinális folyadékban fokális
traumás agysérülés esetén [102].
2.1.4. A DAI terápiás befolyásolását célzó vizsgálatok
Miután egyértelművé vált, hogy a DAI nem azonnal létrejövő axonszakadás,
hanem egy órák alatt progrediáló folyamat, a komplex pathomechanizmusnak
megfelelően, több támadásponton is kutatások kezdődtek a károsodás mérséklésére, illetve
kivédésére. Az első sikeres beavatkozás a hypothermia volt [70]. A trauma után rövid időn
belül (< 25 perc) alkalmazott moderált hypothermia (32 Celsius) csökkentette az AD/B-t
mutató axonok számát kontúziós modellben. A korai kontrollált hypothermia szignifikáns
axonális védelmet nyújtó hatása később a diffúz agykárosodást kiváltó modellekben is
igazolást nyert mind az AD/B (APP-pozitivitás) [58, 81], mind a CMSP és az NFC
tekintetében [8, 81]. Ugyanakkor az is bizonyítottá vált, hogy a felmelegedési periódusnak
lassú, gradált formában kell történnie, a túl gyors visszamelegítés ugyanis az
axonkárosodás újbóli felerősödését okozza [82, 139].
A terápiás jellegű kutatások egy másik válfaja a mitokondriumok integritásának
megőrzését, ezáltal a lokális energiaháztartás fenntartását, illetve a citokróm-c
felszabadulásának és a következményes caspase aktivációnak a megakadályozását célozta.
E tekintetben a figyelem egy, a klinikumban már immunszupresszánsként bevált szernek,
a cyclosporine-A-nak (CsA) irányába fordult; e vegyület ugyanis képes kötődni a MPT-
pórushoz, megakadályozza annak kinyílását, ezáltal csökkentve a mitokondriális
károsodást [94]. A további kísérletek azt is igazolták, hogy a mitokondriumok
funkciójának fennmaradása esetén az excesszív mértékben beáramló Ca2+ eltávolítását
végző folyamatok energiaigénye fedezhető, és így a Ca2+-indukálta strukturális
fehérjebontás fékezhető, következményesen az axonális citoszkeletális elváltozások
megelőzhetőek [9, 95]. Az NFC befolyásolásában megfigyelt jótékony hatás mellett, a
CsA az intraaxonális transzportzavart mutató axonok számát is szignifikánsan
csökkentette [9, 94, 96]. Az MPT-pórusokra kifejtett hatása mellett a CsA és számos
17
egyéb, hozzá hasonlóan az immunophillinek családjába tartozó vegyület képes a
calcineurin gátlására is. Ismert, hogy a citoszkeleton integritásának fenntartásában
elengedhetetlenül fontos a neurofilamentek foszforilált állapotának fenntartása. A
calcineurin egy foszfatáz, amely olymódon segíti az axonális citoszkeleton károsodását,
hogy a neurofilamenteket – defoszforilásuk útján – hozzáférhetőbbé teszi a calpain
számára [97], így gátlása jótékonyan befolyásolhatja az axonkárosodás folyamatát [26, 27,
126]. E vegyületek közül az FK506 -ról sikerült kimutatni, hogy az axonduzzadást mutató
axonok számának csökkentése révén kedvező hatása van diffúz traumás axonkárosodásra
[74, 131].
A diffúz axonkárosodás mechanizmusának ismeretében további terápiás
lehetőségnek tarthatjuk a calpain és a calpain-mediált spektrin proteolízis gátlását.
Korábban kontúziós traumamodellben a calpain gátlásával az NFC és spektrin-degradációs
termékek csökkent jelenléte volt megfigyelhető ˙[105]. A calpain gátlása ugyancsak
javította a trauma utáni neurológiai (magatartási) kimenetelt fokális agykárosodás esetén
[123, 124], ugyanakkor e jótékony hatás mögött nem sikerült Western blottal és
hisztológiai módszerekkel csökkent CMSP-t kimutatni [124]. Hasonlóan
ellentmondásosak az eredmények a diffúz koponyatrauma egér-modelljében, amennyiben
ez esetben sem kísérte a javult neurológiai kimenetelt csökkent neuronális CMSP az
agykéregben illetve a hippocampusban [61]. A calpain-gátlás közvetlen axonális hatását
azonban sem az idézett tanulmányok sem mások idáig még nem vizsgálták.
2.2. A diffúz neuronális károsodás
Az utóbbi évtizedekben számos kutatócsoport foglalkozott a koponyasérülések
során kialakuló diffúz axonkárosodás vizsgálatával. Ugyanakkor kevés figyelmet
szenteltek annak, hogy ugyanazok az erők, melyek axonkárosodást váltanak ki, okoznak-e
diffúz eloszlásban, egyébként ép parenchimális környezetben neuronális sérüléseket.
Fokális traumás agykárosodás esetén kialakuló regionális nekrotikus és apoptotikus
neuron pusztulás jól ismert az irodalomban [20, 22, 28, 89, 116]. Fokális agysérülésekhez
társuló, de a kontúziós régiótól távol eső idegsejt-károsodásokat is leírtak már [21, 23, 48],
ugyanakkor a diffúz agysérülések során kialakuló neuronális károsodás szerepét kevesen
vizsgálták. A meglévő kutatási eredmények is inkább leíró jellegűek és nem terjednek ki a
sejtkárosodás kialakulásának mechanizmusára [16, 69, 121]. A fent említett kísérletek azt
18
feltételezték, hogy a megfigyelt sejthalál a trauma által indukált neuroexcitáció, oxigén-
szabadgyök képződés vagy egyéb másodlagos elváltozások következménye, és nem vették
figyelembe azt a lehetőséget, hogy a koponya trauma által kiváltott erők közvetlenül
okoznának idegsejt-károsodást.
2.2.1. A „sötét”-idegsejt képződés
Kontúziós és nem-kontúziós koponyatrauma által kiváltott agyi morfológiai
elváltozások vizsgálata során Gallyas és munkatársai hívták fel a figyelmet először arra,
hogy a trauma képes – axonális károsodás mellett – diffúz idegsejt-és dentritikus-
károsodás közvetlen létrehozására is; ugyanis bizonyos argyrophil morfológiai
elváltozások a trauma kiváltása után 1 percen belül leölt kíséreti állatokban már jelen
voltak a behatás helyétől távoli agyterületeken is, míg a kontrol állatokból hiányoztak [33,
34]. E megfigyelés ellene szól annak az elképzelésnek, amely szerint a neuronális
károsodás másodlagosan alakul ki a trauma által generált egyéb elváltozások (pl.
ischemia) következtében. A trauma hatására kialakult korai diffúz eloszlású idegsejt-
elváltozások fénymikroszkóppal vizsgálva a sejttest-dendrit egység zsugorodásában és
bennük argyrophilia, illetve hyperbasophilia megjelenésében nyilvánulnak meg. Az ilyen
morfológiai jeleket mutató idegsejteket tradicionálisan „sötét”-idegsejteknek nevezik [13].
Ultrastrukturálisan a „sötét”-idegsejtekre emelkedett elektrondenzitás, UC, a Golgi
ciszternák tágulata és a nukleáris kromatin nem-apoptotikus aggregációja jellemző.
Hasonló sejt-elváltozások nemcsak trauma, hanem egyéb tényezők – hypoglicaemia,
ischemia, status epilepticus – hatására is kialakulnak [32], sőt képződésük post mortem is
kiváltható [35], ami felveti annak lehetőségét, hogy a „sötét”-idegsejt képződés nem
enzim-mediált pathobiokémia folyamatok eredménye. A kísérletek során azt is
megfigyelték, hogy az idegsejt-dendrit egységben, valamint az axonokban létrejött
elváltozások egymástól függetlenek, ami azt jelezi, hogy a sejt különböző doménjei
szelektíven érzékenyek a kiváltó traumából eredő hatásokra. A „sötét”-idegsejtek további
sorsát vizsgálva az is leírásra került, hogy a kiváltó tényezők hatására azonnal kialakult
sötét idegsejteknek csak egy hányada pusztul el, többségük azonban regenerálódik, és a
kialakulás után 4 órán belül visszanyeri normális morfológiai és festődési tulajdonságait
[25].
19
2.2.2. A neuronális mechanoporáció
In vitro módszerek alkalmazásával megfigyelték, hogy sejtkultúrák idegsejtjei
nyújtás hatására nagy molekulasúlyú anyagokat vesznek fel a sejthártya sérülésének
következtében. Ez az idegsejthártya-sérülés és permeabilitás-változás a kiváltó trauma
után azonnal észlelhető volt, ami újra megkérdőjelezte az idegsejtkárosodás
kialakulásában feltételezett másodlagos tényezők kizárólagos kóroki szerepét [38]. E
kísérlet során a plazmamembrán-permeabilitás növekedése (mechanoporáció) az
erőbehatás nagyságával és az alkalmazott anyagok molekulasúlyával arányosnak és időben
átmenetinek bizonyult: a legkisebb anyagok felvétele a sérülést követően 5 perc múlva is
megfigyelhető volt, míg a legnagyobb mólsúlyú anyagok csak a sérülés kiváltását
követően azonnal jutottak át a sejthártyán [38]. A mechanoporáció jelensége in vivo
modellben is megfigyelhető volt. A diffúz koponyatrauma egy patkány-modelljében nagy
molekulasúlyú anyagok intracerebroventrikuláris alkalmazásával igazolódott, hogy – az
axonok mellett – az agykéregben, a hippocampusban és a thalamusban is találhatók
plazmamembrán permeabilitási zavart szenvedett neuronok; e sejtek közvetlenül a kiváltó
trauma után (túlélési idő < 5 perc) felvesznek az ép sejthártyán keresztül át nem jutó nagy
molekulasúlyú anyagokat [133]. Az idegsejt-elváltozások további vizsgálata során az is
bebizonyosodott, hogy a kiváltó trauma nem egyformán érinti az idegsejteket, hanem a
neuronkárosodás különböző, egymástól függetlenül kialakuló elváltozások egymás
melletti (azonos agyi régiókban megtalálható) megjelenéséből tevődik össze. Meglepő
megfigyelés volt, hogy a perisomatikus axonkárosodást szenvedett APP-pozitív axonok
neuronjai nem mutattak sejthártya-permeabilitási zavart vagy súlyos sejtkárosodásokra
utaló ultrastrukturális elváltozásokat [132, 133]. Ez annyit jelent, hogy az axonkárosodás
és axonszakadás nem vezet az érintett idegsejt gyors pusztulásához, szemben a korábbi
feltételezésekkel [4, 76]. A megfigyelt neuronális elváltozások spektruma a
mechanoporációt mutató neuronok körében a gyors nekrózistól az ultrastrukturálisan ki
nem mutatható elváltozásokig terjedt [133]. Ez utóbbi jelenség hátterében – azaz abban,
hogy a sejtmembrán-sérülés nem feltétlenül vezet az érintett idegsejt pusztulásához – a
szerzők az idegsejt-membrán gyors helyreállítódását (resealing) feltételezték. Ez irányú
kísérletek azonban vizsgálatainkig nem történtek.
20
3. CÉLKITŰZÉSEK
A dolgozat célkitűzéseit – a kísérleti eredményeket megjelentető tudományos
közlemények felsorolásával együtt – az alábbiakban foglaljuk össze:
1. A traumás axonkárosodás vizsgálata
1.1. Az ultrastrukturális axon-kompakció vizsgálata
1.1.1. Az ultrastrukturális kompakció és az argyrophil axonkárosodás kapcsolatának
tisztázása
(Gallyas F, Farkas O, Mazlo M. Traumatic compaction of the axonal
cytoskeleton induces argyrophilia: histological and theoretical importance. Acta
Neuropathol (Berl). 2002 Jan;103(1):36-42.)
1.1.2. Olyan koponyatrauma modell kidolgozása, melynek segítségével az
ultrastrukturális kompakció szelektíven idézhető elő
(Pal J, Toth Z, Farkas O, Kellenyi L, Doczi T, Gallyas F. Selective induction of
ultrastructural (neurofilament) compaction in axons by means of a new head-
injury apparatus. J Neurosci Methods. 2006 Jun 15;153(2):283-9.)
1.1.3. A kompaktálódott axonok sorsának vizsgálata az előző pontban ismertetett modell
segítségével
(Gallyas F, Pal J, Farkas O, Doczi T. The fate of axons subjected to traumatic
ultrastructural (neurofilament) compaction: an electron-microscopic study. Acta
Neuropathol (Berl). 2006 Mar;111(3):229-37.)
1.1.4. Az ultrastrukturális axon- és idegsejt-kompakció mechanizmusának tisztázása
(Gallyas F, Farkas O, Mazlo M. Gel-to-gel phase transition may occur in
mammalian cells: Mechanism of formation of "dark" (compacted) neurons. Biol
Cell. 2004 May;96(4):313-24.)
21
1.2. Terápiás beavatkozások tesztelése
1.2.1. Egy sejt-permeábilis calpain inhibitor (MDL-28170) hatásának vizsgálata a
diffúz axonkárosodásra
(Buki A, Farkas O, Doczi T, Povlishock JT. Preinjury administration of the
calpain inhibitor MDL-28170 attenuates traumatically induced axonal injury. J
Neurotrauma. 2003 Mar;20(3):261-8.)
1.2.2. A – más központi idegrendszeri betegségekben már hatásosnak bizonyult –
PACAP (Hypophysis Adenilát Cikláz Aktiváló Polypeptid) hatásának vizsgálata a
diffúz axonkárosodásra, a trauma előtt illetve után iv illetve icv adva, továbbá
hatásos dózisának megállapítása
(Farkas O, Tamas A, Zsombok A, Reglodi D, Pal J, Buki A, Lengvari I,
Povlishock JT, Doczi T. Effects of pituitary adenylate cyclase activating
polypeptide in a rat model of traumatic brain injury. Regul Pept. 2004 Dec
15;123(1-3):69-75.)
1.2.3. A terápiás ablak meghatározása PACAP kezelés esetén, az 1.2.2. kísérletben
hatásosnak bizonyult dózis alkalmazásával
(Tamas A, Zsombok A, Farkas O, Reglodi D, Pal J, Buki A, Lengvari I,
Povlishock JT, Doczi T. Postinjury administration of pituitary adenylate cyclase
activating polypeptide (PACAP) attenuates traumatically induced axonal injury in
rats. J Neurotrauma. 2006 May;23(5):686-95.)
2. A diffúz idegsejt-károsodás vizsgálata
A neuronális mechanoporáció, a potenciális membrán-helyreállítás (resealing) és
következményeik, illetve az ezekhez kapcsolódó calpain-mediált strukturális
fehérjebontás szerepének vizsgálata
(Farkas O, Lifshitz J, Povlishock JT. Mechanoporation induced by diffuse traumatic
brain injury: an irreversible or reversible response to injury? J Neurosci. 2006 Mar
22;26(12):3130-40.)
22
3. A calpain- és caspase-aktiválódás szerepe a humán traumás
agykárosodásban
A calpain- és caspase-specifikus spektrin-degradációs termékek humán liquormintákból
történő kimutatása Western blot segítségével súlyos koponyatraumát szenvedett betegek
esetén
(Farkas O, Polgar B, Szekeres-Bartho J, Doczi T, Povlishock JT, Buki A. Spectrin
breakdown products in the cerebrospinal fluid in severe head injury--preliminary
observations. Acta Neurochir (Wien). 2005 Aug;147(8):855-61.)
23
4. ANYAGOK ÉS MÓDSZEREK
A jelen fejezetben röviden ismertetjük azokat a módszereket, amelyeket több
kísérletsorozatban is használtunk, míg részletes ismertetésük céljából utalunk a
disszertáció tárgyát képező közleményekre.
Az állatkísérletek mindegyikéhez felnőtt hím patkányokat használtunk; a Baranya
Megyei Állategészségügyi Igazgatóság (illetve a 2. téma kísérletei során a Virginia
Commonwealth University Institutional Animal Care and Use Committee, IACUC)
engedélye alapján, a PTE ÁOK OEC Regionális Kutatási Etikai Bizottság által elfogadott
normákat alkalmazva.
A patkányok a kísérletek teljes ideje alatt mély intratracheális (30 % O2 és 70 %
N2O keverékéhez adott 4 % Isoflurane) vagy intraperitoneális (Thiopental és Seduxen 1:1
arányú keveréke) narkózisban voltak.
Az egyes kísérletek során különböző vitális paraméterek – hőmérséklet, artériás
vérnyomás, artériás vérgázok, intrakraniális nyomás – kerültek monitorozásra.
4.1. A diffúz agysérülés modellezése
4.1.1. Az impact-accelerációs koponyatrauma modell
A kísérletek egy része során a diffúz axonális és neuronális károsodás kiváltására
a Marmarou által leírt, nemzetközileg elfogadott és széles körben használt impact-
accelerációs koponyatrauma modell alkalmazásával került sor. A módszer lényege, hogy
altatott állatok koponyacsontjára a bregma és a lambda varratok közé egy fém-korongot
erősítünk, majd az állatokat egy speciális szivacs-ágyra helyezzük; a fejre erősített
korongot egy plexicső alá centrálva. A trauma kiváltására a plexicsőn keresztül az állat
fejére 450 g súlyt ejtünk. A szivacs-ágy trauma-utáni gyors eltávolításával akadályozzuk
meg, hogy az állat fejét újabb ütés érje. E módszer koponyatöréstől és fokális
agykárosodástól mentes koponyatraumát vált ki, ezáltal alkalmas a diffúz traumás agyi
24
elváltozások vizsgálatára [31, 72]. Az alkalmazott trauma középsúlyos/súlyos agysérülés
kiváltását eredményezi, amely az állatok döntő többségénél átmeneti légzés-leállást okoz.
Ezért a koponyatraumát követően minden esetben az állatot rövid ideig 100 % O2-nel
lélegeztettük. Koponyacsont-törés esetén az állatot a további kísérletekből kizártuk.
4.1.2. Szelektív ultrastrukturális kompakciót kiváltó koponyatrauma modell
Az ultrastrukturális kompakció szelektív kiváltására laboratóriumunkban
kifejlesztettünk egy olyan modellt, amely a rugalmas koponyatetőt egy állítható
mélységig pillanatszerűen benyomja, de nem töri be. A modell lényege, hogy az altatott
állat fejét a készülékhez csatlakoztatható fej-tartóba rögzítjük, majd a fej-tartó
mozgatásával a készülék „impaktor” részét kiválasztott koordinátáknak megfelelően a
szabaddá tett koponyacsont fölé centráljuk. Az impaktort addig mozgatjuk lefelé, amíg az
a koponyát éppen eléri, majd egy csavar segítségével a benyomódás kívánt mértékét
beállítjuk. Ezután egy fémcsőn keresztül 200 g-os acél-súlyt ejtünk az impaktorra, amely
a beállított mértékben benyomja a koponyacsontot. Ez a készülék alkalmas
ultrastrukturális kompakció kiváltására úgy, hogy közben fokális agysérülés, illetve a
diffúz axonkárosodás egyéb formái – elsősorban AD/B – nem következnek be, így
segítségével a kompaktálódott axonok sorsa szelektíven vizsgálható.
4.2. Szövettani feldolgozás
A kísérletek végén a kívánt – az agysérülés kiváltásától eltelt – túlélési idő
elérésekor az állatok mély altatásban szíven keresztüli perfúziós fixálásra kerültek. A
perfúzió aldehid-tartalmú fixáló oldatokkal történt, amelyek alkalmassá teszik az
agyszövetet fény-mikroszkópos (formaldehid tartalmú fixáló) vagy elektron-
mikroszkópos (glutáraldehid tartalmú fixáló) feldolgozásra.
A szíven keresztüli fixálás után az agyat az agytörzzsel és a gerincvelő felső nyaki
szegmentumával együtt a koponyából eltávolítottuk, majd további – immerziós – fixáció
céljából a perfúzióhoz használt oldatba tettük. Ezt követően az agyból a kívánt részt
kivágtuk és vibratóm segítségével a nagyagyból koronális, az agytörzsből szagittális
25
metszeteket készítettünk.
4.2.1. Az argyrophil-III típusú ezüstözési eljárás
Ez az eljárás alkalmas a különféle behatások következtében a központi
idegrendszerben kialakult bizonyos idegsejt- és axon-elváltozások érzékeny kimutatására
[36], és – ellentétben a hagyományos ezüstözési eljárások közismert
megbízhatatlanságával – jól-reprodukálható eredményeket biztosít [90]. A módszer
lényege, hogy a metszeteket felszálló alkohol-sorban vízmentesítjük, majd 16 h-n át 56
°C fokon 0.8 % kénsavat és 2 % vizet tartalmazó l-propanol oldatban inkubáljuk
(észterifikáció). Ezután a metszeteket rehidráljuk, 10 percig 1 %-os ecetsavba helyezzük,
majd egy speciális fizikai előhívó oldatba merítjük, amíg világos barnává válnak. Az
előhívást 1 %-os ecetsavval állítjuk le.
4.2.2. Immunhisztokémiai technikák
Az immunhisztokémia során alkalmazott antitestek a következők voltak:
Target Elsődleges Antitest Másodlagos Antitest
Kísérlet
Intraaxonális transzportzavar
Anti-APP Biotinilált anti-nyúl 1.1.2., 1.2.
Neurofilament kompakció
RMO-14 Biotinilált anti-egér 1.1.2., 1.2.1. és 1.2.3.
Calpain-specifikus SBDP 150
Ab38 Biotinilált anti-nyúl 2.
Calpain-specifikus SBDP 150
Ab38 Alexa Fluor 647 jelölt anti-nyúl
2.
Alexa Fluor 488 jelölt dextrán
Anti-Alexa Fluor 488
Biotinilált anti-nyúl 2.
A fény-mikroszkópos technikák alkalmazása során az immunjelölések láthatóvá
tételére avidin/biotin-peroxidáz komplex alkalmazását követően 0.05 % diaminobenzidint
(DAB) és 0.01 % hidrogén-peroxidot tartalmazó 0.1 mólos foszfát puffert használtunk,
26
szemkontroll mellett. A 2. téma kísérletei során különböző fluorescens festékeket
tartalmazó dextránokat juttattunk az agyszövetbe; ezek agyszöveti megoszlásának
vizsgálata a metszést követően közvetlenül – antitestek használata nélkül – fluorescens
mikroszkóppal történt.
4.2.3. Elektronmikroszkópos előkészítés
Festetlen vagy elektrondenz terméket (DAB, az immunhisztokémia végterméke,)
tartalmazó metszetekből a szomszédos ezüstözött, vagy egyéb módon festett metszeteken
érdekesnek ítélt területeket kivágtuk, majd ozmium-tertoxiddal vagy ozmium-tetroxid és
kálium-hexacyanoferrát(II) keverékével tovább fixáltuk. (Az utóbbi vegyület a myelin
szerkezetének megtartásában fontos; alkalmazása az axon-vizsgálatok esetén különösen
indokolt volt.) A fixálás után a metszeteket Durcupan gyantába ágyaztuk. Félvékony
metszetek toluidin-kékkel való festésével azonosítottuk a kívánt területeket, majd
ultravékony szériákat metszettünk, amelyeket „gridekre” húztunk. A grideket – az
elektrondenzitás növelése céljából – uranil-acetáttal és ólom-citráttal kontrasztosítottuk.
Néhány esetben a már gyantába ágyazott metszeteken ezüstözést végeztünk, majd újra-
beágyazás után „ultra-metszettünk”.
4.3. Western blot
A 3. altéma kísérletei során kamrai katéteren keresztül vagy lumbál-punkcióval
nyert liquor-mintákat a calpain- és caspase-specikifus SBDP-k kimutatására dolgoztuk
fel. A mintákat 100 kD-s szűrőt használva Amicon ultrafiltrációs készüléken átszűrtük,
majd a fehérje-koncentrációikat fotometriás módszerrel megmértük. 20 μg fehérjét
tartalmazó mintákkal 6.5 %-os gélen nátrium-dodecil-szulfátos poliakrilamid gél-
elektroforézist végeztünk, majd a gélről a fehérjéket nitrocellúlóz membránra
transzferáltuk. A membránokat az alfa-II spektrint és lebontási termékeit kimutató egér
antitesttel, majd biotinilált anti-egér másodlagos antitesttel, végül streptavidin-biotinilált
peroxidáz komplex-szel inkubáltuk. Az immunjelölést kemiluminescens reagens
segítségével Kodak filmeken tettük láthatóvá. Az eredményeket denzitometriás
27
módszerrel értékeltük.
4.4. Permeabilitás-vizsgálat
Az intrakraniális nyomás monitorozására az oldalkamrába vezetett kanülön
keresztül a membrán-permeabilitás változás kimutatására alkalmas (az ép sejtekbe az ép
sejthártya által be-nem-engedett) nagy molekulasúlyú dextránokat infundáltunk az
oldalkamrába. Az első dextrán infúzió Alexa Fluor 488 fluorescens festékkel jelölt 10
kDa molsúlyú dextránnal 2 h-val a koponyatrauma előtt történt. A trauma kiváltása után 2
illetve 6 h-val a megnövekedett permabilitás elhúzódásának, illetve az esetleges
sejthártya-helyreállításnak (resealing) kimutatására újabb dextrán-infúzió történt Texas
Red fluorescens festékkel jelölt 10 kDa mólsúlyú dextránnal. A második dextrán-infúziót
követően 2 h-val az állatokat perfundáltuk.
28
5. EREDMÉNYEK ÉS KÖVETKEZTETÉSEK
Az alábbiakban – a Célkitűzések fejezetben vázoltaknak megfelelően –
kísérleteink hátterét, témáját, módszereit valamint legfontosabb eredményeit és
következtetéseit foglaljuk össze. További részletek a disszertáció tárgyát képező
közlemények megfelelő fejezeteiben találhatók.
5.1. A traumás axonkárosodás vizsgálata
5.1.1. Az ultrastrukturális axon-kompakció vizsgálata
5.1.1.1. Az ultrastrukturális axon-kompakció és az argyrophil axonkárosodás
kapcsolata (Gallyas F, Farkas O, Mazlo M. Traumatic compaction of the axonal
cytoskeleton induces argyrophilia: histological and theoretical importance. Acta
Neuropathol (Berl). 2002 Jan;103(1):36-42.)
Háttér: Különféle módon előidézett koponyatrauma után (i) az alkalmazott
argyrophil-III típusú speciális, és reprodukálható eredményű ezüstöző eljárás [32] hosszú
axon-szakaszokat fest feketére, nem-festődő axonok közt sporadikus (diffúz) eloszlásban
[33], (ii) valamint ultrastrukturális (neurofilament) kompakciót hoz létre számos elektron-
mikroszkópos axon-profilban, nem-kompaktálódott axon-profilok közt sporadikus
(diffúz) eloszlásban [98].
Cél: Annak a – tapasztalati úton nyert feltételezésnek – igazolása, hogy az
ezüstöződő axon-szakaszok azonosak az UC-t szenvedett axon-profilokkal.
Módszer: A Marmarou-féle készülékkel végzett koponya-trauma után, a szíven
keresztüli perfúzióval azonnal fixált patkányok agyából készített 150 µm-es vibratómos
sorozat-metszetek közül minden ötödiket megezüstöztük. (A) vizsgálat: Az ezüstözött
axon-szakaszt tartalmazó területeket elektronmikroszkópos megfigyelésre ozmifikáltuk,
Durcupanba ágyaztuk, „ultra-metszettük” és kontrasztoztuk. (B) vizsgálat: Nem-
ezüstözött vibratómos metszetek ugyanezen területeiből ozmifikálás és Durcupanba
29
ágyazás után 1 µm-es metszeteket készítettünk, ezeket megezüstöztük, majd újra
beágyaztuk, „ultrametszettük” és kontrasztoztuk.
Eredmények: (A) vizsgálat: Az ezüstözés erősen roncsolja az ultrastruktúrát;
fellazítja a myelin-hüvelyek lemezeit, feloldja a normális axonokat, de - homogenizált
formában - megőrzi a kompaktálódott axon-profilokat. Ezüst-szemcséket csak
kompaktálódott axon-profilokban találtunk; másrészt nem találtunk olyan kompaktálódott
axon-profilt, amely nem tartalmaztak ezüst-szemcséket. (B) vizsgálat: Minthogy az
ezüstözést ozmifikálás és Durcupanba ágyazás előzte meg, az ultrastruktúra kiválóan
megőrződött. Sok, és viszonylag nagy ezüst-szemcsét találtunk a kompakciót szenvedett
axon-profilokban; másrészt nem találtunk olyan kompaktálódott axon-profilt, amely nem
tartalmazott sok nagy ezüst-szemcsét.
Következtetések: (1) Az alkalmazott ezüstöző módszer lehetővé teszi az UC-t
szenvedett axon-szakaszok fénymikroszkópos feltüntetését vastag vibratómos
metszeteken; ezáltal vizsgálhatóvá teszi ezen károsodást elszenvedett axonok térbeli
elhelyezkedését, valamint az elektronmikroszkópos vizsgálatra érdemes területek
kiválasztását. Továbbá, szemben a vonatkozó korábbi elképzelésekkel, (2) az UC a
trauma pillanatában következik be, nem pedig 5-15 perces késéssel, (3) a citoszkeletális
kompakció viszonylag hosszú, akár 1 mm-es axon-szakaszokra (de általában nem a teljes
axon-hosszúságra) is kiterjedhet, azaz nem EM-szinten „fokális” elváltozás, (4) a
neurotubulusok eltűnése, a neurofiament-oldalkarok részleges feloldódása, valamint a
mitokondriumok duzzadása nem tartozik az UC elsődleges folyamatába.
Megjegyzések: (I) A kompaktálódott axonok feltüntetésére más szerzők által
alkalmazott RMO-14 immunfestéssel kapcsolatos megfigyelésünkről az 5.1.1.2.
alfejezetben lesz szó. (II) Ezüstöződést előidéző UC idegsejtek szóma-dendrit
doménjében is végbe tud menni koponyatrauma hatására [25], ennek, és az axonális
kompakciónak közös – a korábbi nézeteknek részben ellentmondó – mechanizmusáról az
5.1.1.4. alfejezetben lesz szó.
30
5.1.1.2. Ultrastrukturális kompakció szelektív előidézésére alkalmas
koponyatrauma modell kidolgozása (Pal J, Toth Z, Farkas O, Kellenyi L, Doczi T,
Gallyas F. Selective induction of ultrastructural (neurofilament) compaction in axons by
means of a new head-injury apparatus. J Neurosci Methods. 2006 Jun 15;153(2):283-9.)
Háttér: A diffúz axonkárosodás koponyatraumát szenvedett kísérleti állatokban
két jól elkülöníthető morfológiai elváltozás formájában jelentkezik: (i) az előre-irányuló
intraaxonális transzport-zavar által előidézett axonduzzadás/ballon-képződés (AD/B), (ii)
az ezüstöződéssel járó ultrastrukturális kompakció (UC) ( lásd 5.1.1.1. alfejezet) [135].
Míg az előbbi jól ismert humán vonatkozásban is, az utóbbi szerepe a DAI-ban kevésbé
tisztázott. A meglévő koponyatrauma-modellek az említett axonális elváltozások – és
egyéb károsodások (1.2 alfejezet a Bevezetésben) – együttes létrehozására, és ezáltal a
TAI komplex vizsgálatára alkalmasak [40], az egyes morfológiai jellemzők szelektív
vizsgálata azonban használatukkal nem lehetséges.
Cél: Olyan koponyatrauma-modell kidolgozása, melynek segítségével az UC
kialakulása és a kompaktálódott axonok sorsa – más morfológiai elváltozások
befolyásoló hatása nélkül – vizsgálható.
Módszer: Patkányok puha koponya-tetején kontrollálható-mélységű benyomódást
hoztunk létre pillanat-szerűen, majd a patkányokat azonnal, illetve 1 vagy 4 órával
később, formalinos vagy glutáraldehides fixálóval perfundáltuk. A koponyatrauma előtt,
illetve a túlélés időtartamában monitoroztuk a légzés-számot, a pulzusszámot, a
vérnyomást és a koponyaűri nyomást. Az 1 nap múlva kivett agyakból készített
vibratómos metszetek közül minden harmadikat megezüstöztük; a formalinban
fixáltaknál a „közbeeső” metszetek APP immun-hisztokémiára vagy RMO-14 immun-
hisztokémiára kerültek, míg a glutáraldehides „közbeeső” metszeteknek az ezüstözés
alapján vizsgálni érdemesnek tartott területeit elektron-mikroszkópiára ágyaztuk be.
Eredmények: 0.75 mm-es benyomódás esetén a vizsgált fiziológiai paraméterek a
koponyatrauma után 1 perccel visszatértek a koponyatrauma előtti értékekre; ezüstözött
illetve UC-t szenvedett axonok csak a benyomódás alatti agykéregben voltak
megfigyelhetők, normális axonok között elszórtan, ép parenchimális környezetben.
Egyéb morfológiai elváltozást nem találtunk. A károsodott axonok száma a
koponyacsont-benyomódás mértékével párhuzamosan változott. Azonnal, illetve 1 órával
31
a trauma után a kompaktálódott axonok homogénen ezüstöződtek, míg 4 órával a traumát
követően pontszerű ezüstöződés volt megfigyelhető. Intraaxonális transzportzavart
mutató (anti-APP-pozitív) axonok csak 1 mm-es, vagy nagyobb benyomódás esetén
voltak láthatók; RMO-14 pozitivitást egyetlen esetben sem észleltünk.
Következtetések: (1) A kidolgozott koponyatrauma-modell 0.75 mm-es
koponyacsont-benyomódás esetén alkalmas az ultrastrukturális axon-kompakció szelektív
előidézésére, ezáltal alkalmas a kompaktálódott axonok sorsának – más morfológiai
elváltozások vagy pathofiziológiai tényezők befolyásoló hatása nélküli – vizsgálatára
(lásd az 5.1.1.3. alfejezetet), terápiás lehetőségek tesztelésére vagy akár ismételt enyhe
koponyatrauma hatásának tanulmányozására. (2) Az APP-pozitivitás hiánya megerősíti
azt az elképzelést, hogy az intraaxonális transzportzavar és az axon-kompakció két
különböző axon-populácót érint. (3) A kompaktálódott axonok az agykéregben több óra
elteltével sem festődnek az agytörzsi axon-kompakció kimutatására széles körben
elterjedt markerrel, az RMO-14 antitesttel. A jelenség valószínűleg azzal magyarázható,
hogy – ellentétben az agytörzs nagy-kaliberű, myelinizált axonjaival – az agykéreg
vékony axonjaiban igen kevés az axon-kalibert meghatározó NF-M alegység [29],
amelynek a calpain-aktiválódás hatására szabaddá vált „rod doménjét” mutatja ki az
RMO-14 antitest [10].
Megjegyzések: (I) 1 mm-es koponyacsont-benyomódás, továbbá alacsonyabb
sebességű koponyacsont benyomás, illetve egyéb koponya-trauma előidézési-módok [pl.
25] esetén nemcsak axonok, hanem idegsejtek és dendritfájuk is ezüstözhetővé válik,
illetve kompaktálódik. Erről a jelenségről, illetve az ultrastrukturális axon- illetve
idegsejt-kompakció háttérében feltételezett mechanizmusról az 5.1.1.4. alfejezetben lesz
szó. (II) Az ismertetett koponya-trauma módszer egy fontos alkalmazási területével az
5.1.1.3. alfejezetben foglalkozunk.
5.1.1.3. A kompaktálódott axonok sorsának vizsgálata szelektív
ultrastrukturális kompakciót kiváltó koponyatrauma modellben (Gallyas F, Pal J,
Farkas O, Doczi T. The fate of axons subjected to traumatic ultrastructural
(neurofilament) compaction: an electron-microscopic study. Acta Neuropathol (Berl).
2006 Mar;111(3):229-37.)
32
Háttér: Az ultrastrukturális axon-kompakció a TAI konzekvens morfológiai
velejárója, ennek ellenére eddig nem vizsgálták az érintett axonok sorsát. Ugyanakkor
ismert, hogy UC-t szenvedett idegsejtek egy hányada néhány óra alatt spontán
„meggyógyul” (recovery), másik hányada viszont elpusztul [25]. Minthogy az axon-
kompakció illetve az idegsejt-kompakció feltételezhetően azonos mechanizmussal megy
végbe (lásd 5.1.1.4. alfejezet), feltételezhető, hogy a kompaktálódott axonoknak is csak
egy hányada pusztul el, a többi viszonylag gyorsan visszanyeri eredeti morfológiáját.
Cél: Az 5.1.1.2. alfejezetben ismertetett koponyatrauma módszer segítségével
szelektíven kompakttá tett axonok sorsának nyomon követése.
Módszer: 0.75 mm-es koponyacsont-benyomódást létrehozó koponyatraumát
szenvedett patkányokat a trauma kiváltása után azonnal illetve 6 hónapig terjedő túlélési
idő elteltével glutáraldehid-tartalmú fixálóval perfundáltunk, az agyakból készült
metszetek egy hányadát ezüstöztük, a szomszédos metszetek EM feldolgozásra kerültek.
Egy kiválasztott agykérgi területen a kompaktálódott, illetve a nem-kompaktálódott
myelinizált axonok számát kvantitatív analízis céljából megállapítottuk.
Eredmények: Az agykéreg vizsgált területén a kompaktálódott myelinizált axonok
száma az azonnal-perfundált patkányokban több, mint kétszer annyi volt, mint az 1 napot
vagy az 1 hetet túléltekben. Az azonnal, illetve a 4-órás túlélés után leölt patkányokban
számos olyan myelinizált axon-profil volt megfigyelhető – az 5.1.1.1. és 5.1.1.2.
alfejezeteknek megfelelően –, amelyekben a hosszanti ultrastrukturális elemek közötti
távolságok drámai módon lecsökkentek (kompakció), míg az egyéb ultrastrukturális
elemek épnek tűntek. A túlélési idő növekedésével egy sor egyéb ultrastrukturális
elváltozás került megfigyelésre; így (i) membrán-örvények megjelenése az első két
túlélési napban egyébként ép ultrastruktúrájú axon-profilokban, (ii) a kompaktálódott
axon-profilok ultrastruktúrájának homogenizációja az első két túlélési napban, (iii)
később a homogenizált ultrastrukturális területek fragmentációja, (iv) majd a
fragmentumok felcserélődése oligodendroglia-citoplazmával, és (v) végül axolemmával
körülvett, neurofilamenteket tartalmazó profilok megjelenése szokatlanul sok
oligodendroglia-citoplazmát tartalmazó myelinizált axonokban. (vi) Kiemelendő, hogy
makrofág infiltrációra, mikroglia proliferációra és a Waller-féle degeneráció késői,
előrehaladott stádiumára utaló jeleket nem láttunk még 6 hónap elteltével sem.
33
Következtetések: (1) A kompaktálódott axonok >50%-a 1 napon belül, míg <10%-
a 1 nap és 1 hét között visszanyeri eredeti ultrastruktúráját (meggyógyul); ennek jele
– hasonlóan az idegsejtek szómája és dendritjei esetében találtakhoz [25] – a „membrán-
örvények” megjelenése. (2) A többi kompaktálódott axon néhány hónap alatt
regenerálódik (lásd az eredmények (ii)-(vi) pontjait). (3) Feltételezhető, hogy a fentiek
szerepet játszanak a koponyatraumát szenvedett betegek fizikai és szellemi
teljesítményeinek gyors (az első trauma-utáni nap folyamán végbemenő) illetve lassú (a
néhány hónappal később végbemenő) javulásában. (4) A Waller-féle (irreverzibilis)
degenerációval szembeni eltérés magyarázata abban keresendő, hogy a vizsgált esetben
nem szakad meg a kapcsolat a kompaktálódott axon myelin-burka, és az azt tápláló
oligodendroglia közt; ezért a myelin-burok nem pusztul el, és így pályát képez a
regenerálódó axon számára. (5) Az alkalmazott modellről ismert, hogy nem vált ki tartós
rosszabbodást a fiziológiai jellemezőkben (lásd 5.1.1.2. alfejezet). Ennek tudható be a
talált spontán-gyógyulási képesség, valamint a degeneráció reverzibilitása.
Megjegyzés: Feltételezhető azonban, hogy tartósan-kedvezőtlen pathofiziológiai
körülmények között – pl. megnövekedett ICP, ischemia – az axonoknak az általunk
megfigyeltnél nagyobb hányada nem gyógyul meg, illetve a spontán nem-gyógyuló
axonok egy hányada nem regenerálódik (elpusztul), amint ez a kompaktálódott idegsejtek
esetében kimutatható volt [37].
5.1.1.4. Az ultrastrukturális axon- illetve idegsejt-kompakció
mechanizmusának tisztázása (Gallyas F, Farkas O, Mazlo M. Gel-to-gel phase
transition may occur in mammalian cells: Mechanism of formation of "dark" (compacted)
neurons. Biol Cell. 2004 May;96(4):313-24.)
Háttér: Különböző patobiokémiai tényezők (köztük hypoglicaemia, ischemia és
epilepsia), továbbá fizikai behatások (elektromos sokk, koponyatrauma) [25, 32] képesek
azonos morfológiai elváltozások (ultrastrukturális kompakció) létrehozására.
Fénymikroszkópos szinten ismert – de elektron-mikroszkóppal korábban nem vizsgált – ,
hogy ugyanilyen morfológiai elváltozás jön létre post mortem is idegsejtekben a nem-
fixált, vagy rosszul fixált agynak a koponyából való kivétele során elszenvedett
mechanikai megterhelés hatására [14]. Az azonos morfológia a kiváltó károsító hatástól
34
független közös kialakulási mechanizmust feltételez; a post mortem kialakulás lehetősége
pedig nem-enzim-közvetített folyamatra utal.
A polimer gél-kémiából ismert egy, a kiváltó tényezőtől független mechanizmusú,
és makromolekulák nem-kovalens kötései formájában tárolt energia hatására végbemenő,
egy pontból kiindulóan tovaterjedő folyamat, a gélből gél fázisátalakulás [140], melynek
szerepét az élő sejt egyes mechanizmusaiban már feltételezték [104, 141, 144].
Cél: (1) In vivo, illetve az enzim-közvetített folyamatok számára rendkívül
kedvezőtlen körülmények között végzett, azonos típusú mechanikai behatás által kiváltott
ultrastrukturális idegsejt- illetve axon-kompakció morfológiai jellemzőinek
összehasonlítása annak bizonyítására, hogy a kompakció kialakulásának mechanizmusa
nem enzim-közvetített folyamat. (2) Annak valószínűsítése, hogy ez a folyamat mind
energetikai, mind strukturális szempontból gélből-gél fázisátalakulás.
Módszerek: A Marmarou-féle készülékkel koponyatraumát hajtottunk végre
egyrészt post mortem (glutáraldehides fixálóval végzett 30-perces, szíven-keresztüli
perfúziót követően 0-fok közelébe hűtött patkányokon), másrészt in vivo (közvetlenül a
perfúziós fixálás előtt a Marmarou-féle készülékkel traumatizált patkányokon). Az egy
nappal később kivett agyakból készített vibratómos metszetek egy hányada ezüstözésre,
másik hányada EM-feldolgozásra került. Kvantitatív analízis céljából meghatároztuk a
hosszanti ultrastrukturális elemek közti átlag-távolságot mind kompaktálódott, mind
pedig normál axonokban.
Eredmények: (A) Az in vivo és a post mortem koponyatrauma egyforma
morfológiai elváltozásokat eredményezett, mind a (i) a hyperbasophilia, (ii) a III-típusú
argyrophilia, (iii) a hiper-elektrondenzitás, mind pedig (iv) az UC jellegzetességeinek
tekintetében. (B) Az ultrastrukturális axon-kompakció foka a post mortem trauma esetén
nem különbözött az in vivo trauma esetén meghatározottól. (C) A trauma-indukálta
ezüstöződés (az UC fénymikroszkópos jele, lásd 5.1.1.1. alfejezet) mindkét esetben
„minden vagy semmi” jellegű volt, azaz az érintett szóma-dendrit domének teljes
egészére, illetve hosszú axon-szakaszokra terjedt ki, de általában nem egyazon idegsejt
szóma-dendrit doménjét és axonját érintette.
Következtetések: Az ultrastrukturális idegsejt- és axon-kompakció
mechanizmusában nem enzim-közvetített folyamatok játszanak szerepet, hiszen (1)
35
valószínűtlen, hogy ugyanolyan morfológiai eredménnyel folynak le biokémiai folyamat-
kaszkádok in vivo és post mortem (előzetes aldehides fixálás és 0 C fokra való lehűtés
után), vagyis az enzimatikus folyamatoknak kedvező illetve rendkívül kedvezőtlen
körülmények között. (2) A “minden-vagy-semmi” jelenség csak úgy magyarázható, hogy
a kompakció egyetlen kezdeti pontból terjed ki az érintett idegsejt szóma-dendrit
doménjének egészére, vagy egy hosszú axonszakaszra. Elképzelhetetlen ugyanis, hogy a
kompakció az érintett idegsejt szóma-dendrit doménjének valamennyi pontján egyszerre
iniciálódjon a – néhány idegsejt környezetében kvázi-homogén mechanikai paraméter-
változásokat okozó – pillanatszerű koponyatrauma hatására, a szomszédos idegsejtek
szóma-dendrit doménjének pedig egyetlen pontján sem. Nagyon valószínűtlen az is, hogy
egy enzim-közvetített reakció-kaszkád csillapodás nélkül végigterjedjen az érintett
idegsejt szóma-dendrit doménjének egészén vagy hosszú axon-szakaszokon, különösen
az enzimatikus reakcióknak rendkívül kedvezőtlen körülmények között.
Az UC jelenségének magyarázatára feltételeztük, hogy az ismert ultrastrukturális
elemek (organellumok, citoszkeleton) közti terekben egy összefüggő, fehérje-mátrixú
gél-struktúra található, amely energiát tárol nem-kovalens kötések formájában. Ebben a
gél-struktúrában a gél mátrixát felépítő fehérje-molekulák kooperatív (egyetlen pontból
kiindulva tovaterjedő) konformáció-változása következtében fázisátalakulás megy végbe,
miközben a megkötött vízmolekulák illetve K+-ionok egy jelentős hányada felszabadul a
gél összehúzódását (térfogatcsökkenés) eredményezve. Az összehúzódó gél a hozzákötött
ultrastrukturális elemeket magával vonja (kompakció). A polimer-kémiából ismert, hogy
a gélből-gél fázisátalakulás iniciálható mind fizikai, mind kémiai behatásokkal, és
reverzibilis lehet, összhangban a „sötét” idegsejtek és axonok ismert regenerációs
képességével.
Megjegyzések: (I) A spontán regenerációra képes idegsejtek valószínűleg
megegyeznek azokkal, amelyek a trauma hatására plazmamembrán-permeabilitási zavart
szenvednek, de nem mutatnak súlyos sejtkárosodásra utaló ultrastrukturális jeleket (lásd
az 5.2. alfejezetet). (II) Ellentétben az axonokkal, az idegsejtek szómájában illetve
dendritjeiben – rendezetlen irányaik miatt – nem vitelezhető ki az ultrastrukturális
elemeik közti átlagtávolságok meghatározása. (III) Elektromos sokkal [56], valamint 0.1
mM SDS-t tartalmazó fiziológiás só szíven-keresztüli perfúziójával [59] is sikerült
36
ultrastrukturális kompakciót létrehozni post mortem körülmények közt. (IV) Az
ultrastrukturális axon-kompakció mechanizmusában korábban feltételezett calpain-
mediált strukturfehérje-bontás feltételezhetően nem oka, hanem következménye az axon-
kompakciónak. Elképzelésünk szerint a gélből gélbe történő átalakulással létrejött
ultrastrukturális kompakció során az axolemma vongálódása révén axolemmális
permeabilitási zavar alakul ki, amely azután a jól ismert mechanizmussal a calpain
aktiválódásához és ezáltal citoszkeletális fehérje-bontáshoz vezet (lásd 5.1.2.1. alfejezet).
5.1.2. Terápiás beavatkozások tesztelése
5.1.2.1. Egy sejt-permeábilis calpain-inhibitor, az MDL-28170, hatásának
vizsgálata a diffúz axonkárosodásra (Buki A, Farkas O, Doczi T, Povlishock JT.
Preinjury administration of the calpain inhibitor MDL-28170 attenuates traumatically
induced axonal injury. J Neurotrauma. 2003 Mar;20(3):261-8.)
Háttér: A koponyatrauma által kiváltott axolemma-permeabilitási zavar
(„mechanoporáció” – a megnövekedett intracelluláris Ca2+-koncentráció révén – calpain
aktiválódáshoz, az axonális citoszkeleton részleges lebontásához és következményes
NFC-hez vezet [10]. Kontúziós koponyatrauma-modellekben a calpain gátlása jótékony
hatásúnak bizonyult az agykérgi citoszkeletális károsodások, és funkcionális károsodások
csökkentésében [105, 123, 124], illetve a használni kívánt vegyület neuroprotektív hatású
volt agyi ischemia során [68, 71].
Cél: Egy sejt-permeábilis calpain-inhibitor, az MDL-2817, hatásának vizsgálata a
diffúz axonkárosodás két morfológiai jellemzőjére, a csökkent intraaxonális transzportra
(IAT) és a neurofilament kompakcióra (NFC).
Módszerek: A Marmarou-féle készülékkel kiváltott koponyatrauma előtt 30
perccel MDL-28170-t, vagy vivőanyagot (kontrol) iv-kapott patkányokat a trauma után 2
órával a szíven keresztül fixálószerrel perfundáltuk. A szagittális agytörzsi metszetek egy
hányadán az IAT markerével (anti-APP), a szomszédos metszeteken pedig az NFC
markerével (RMO-14) immunhisztokémiát végeztünk. A károsodott axonok számát a
kortikospinális pályában (CSpT) és a mediális hosszanti kötegben (MLF)
37
összehasonlítottuk a kezelt és a kontrol patkányok vonatkozásában.
Eredmények: Az anti-APP-pozitív és az RMO-14-pozitív axon-profilok
morfológiai jellemzőiben nem volt különbség megfigyelhető a kezelt illetve a kontrol
patkányokban. A kvantitatív analízis szerint a trauma előtt végzett MDL-28170 kezelés
szignifikánsan csökkentette az RMO-14-pozitív axonok számát mind a CSpT (p<0.02),
mind pedig az MLF területén (p<0.03). Az anti-APP-pozitív axonok száma is
statisztikailag szignifikáns csökkenést mutatott az MDL-28170-nel való kezelést
követően mindkét vizsgált agytörzsi területen (p<0.03 a CSpT estén és p<0.01 az MLF
esetén).
Következtetések: A calpain-gátlásnak az NFC kialakulására kifejtett jótékony
hatása magyarázható a calpain-mediált spektrin-bontás és az NFC ismert kapcsolatával.
Bár a legújabb kísérleti eredmények azt mutatják, hogy az NFC és az IAT különböző
axon-populációkat érint [73, 135], a calpain-gátlásnak az IAT-ra kifejtett pozitív hatása
azt bizonyítja, hogy – ha a kompakció nem is társul transzport-zavarral – a calpain
aktiválódása mindkét folyamatban jelentős szerepet játszik.
Megjegyzés: Bár kísérleteink tovább erősítik a calpain és a CMSP szerepének az
NFC-ben betöltött jelentőségét, ez nincs ellentmondásban az axon-kompakció általunk
feltételezett mechanizmusával (lásd 5.1.1.4. alfejezet).
5.1.2.2 A trauma előtt illetve után iv- illetve icv-adott PACAP neuroprotektív
hatásának vizsgálata; valamint a hatásos dózis megállapítása (Farkas O, Tamas A,
Zsombok A, Reglodi D, Pal J, Buki A, Lengvari I, Povlishock JT, Doczi T. Effects of
pituitary adenylate cyclase activating polypeptide in a rat model of traumatic brain injury.
Regul Pept. 2004 Dec 15;123(1-3):69-75.)
Háttér: A PACAP számos neurotrófikus és neuroprotektív hatással rendelkezik in
vitro. In vivo jótékony hatásúnak bizonyult mind a globális, mind a fokális agyi ischemia
kezelésében [117, 118, 142], emellett traumás gerincvelő-sérülésekben [55], továbbá
látóideg- illetve arcideg-átvágás esetén is [57, 127]. Az ischemiás és a traumás
agykárosodás kialakulásában szerepet játszó – részben hasonló – mechanizmusokból
kiindulva feltételezhető, hogy ez a vegyület traumás agykárosodás esetén is hatásos lehet.
Cél: A PACAP esetleges neuroprotektív hatásának vizsgálata a traumás
38
agysérülések során kialakuló diffúz axonkárosodásra; illetve a vegyület hatásos
dózisának megállapítása.
Módszerek: (A) A Marmarou-féle készülékkel kiváltott koponyatrauma előtt
PACAP-ot iv-kapott patkányokat 2 illetve 6 órával a koponyatrauma után perfúziósan
fixáltuk, a szagittális agytörzsi metszeteket anti-APP immunhisztokémiával vizsgáltuk, a
károsodott axonok számát a kezelt illetve a kontrol patkányokban a CSpT és az MLF
területén összehasonlítottuk. (B) A hatásos dózis megállapítása céljából 1 μg, 10 μg és
100 μg PACAP-ot adtunk be az (A) kísérlet eredményeiből kiindulva icv; a túlélési idő 2
óra volt.
Eredmények: (A) Intravénás adagolás esetén – más KIR betegségekben
hatásosnak bizonyult dózisban – a PACAP nem bizonyult neuroprotektívnek (sem a
CSpT sem pedig az MLF területén nem csökkentette szignifikánsan az anti-APP
immunpozitív axonok számát). (B) A PACAP intracerebroventrikuláris adagolásban 1
μg-os és 10 μg-os dózis esetén szintén nem bizonyult hatásosnak, 100 μg-os dózis
azonban szignifikánsan csökkentette a CSpT területén talált anti-APP immunpozitív
axonszakaszok számát (p<0.05).
Következtetések: (1) A PACAP az apoptotikus és a gyulladásos mediátorok
gátlása, illetve bizonyos mitokondriális enzimek befolyásolása révén fejthet ki
neuroprotektív hatást. E folyamatok – mind az elsődleges, mind a másodlagos (lásd az
1.2. alfejezetet a Bevezetésben) patológiás folyamatok következtében – traumás
agysérülések esetén is megfigyelhetők [11, 66, 143]. (2) Az intravénás adagolás
sikertelensége abban keresendő, hogy – bár a PACAP bizonyítottan átjut a vér-agy gáton
– transzportjának mértéke az agytörzsben sokkal alacsonyabb, mint más agyterületeken
[91]. (3) Hasonlóan, az agytörzs területén a feltételezett alacsony mértékű liquor-agy
transzport lehet a magyarázata a más kísérletekben tapasztaltabbaknál magasabb hatásos
dózis igénynek icv alkalmazás esetén.
Megjegyzés: A jelen kísérletek célja pusztán a PACAP traumás axonkárosodásra
gyakorolt hatásának felmérése volt, ezért adása közvetlenül a trauma kiváltása után
történt. A klinikai alkalmazhatóság feltételeként késleltetve adott PACAP
hatékonyságának vizsgálatával az 5.1.2.3. alfejezetben foglalkozunk.
39
5.1.2.3. A terápiás ablak meghatározása icv PACAP-kezelés esetén (Tamas A,
Zsombok A, Farkas O, Reglodi D, Pal J, Buki A, Lengvari I, Povlishock JT, Doczi T.
Postinjury administration of pituitary adenylate cyclase activating polypeptide (PACAP)
attenuates traumatically induced axonal injury in rats. J Neurotrauma. 2006
May;23(5):686-95.)
Háttér: Egy korábbi kísérlet szerint (lásd 5.1.2.2.) a koponyatrauma után azonnal
icv-beadott PACAP csökkenti az axonális transzport fokális leállása által okozott
axonkárosodást.
Cél: (A) A trauma kiváltását követően késleltetve icv-adott PACAP
hatékonyságának tesztelése diffúz axonkárosodásban az intraaxonális transzportzavar és a
neurofilament kompakció vonatkozásában; (B) továbbá a terápiás ablak meghatározása.
Módszerek: A Marmarou-féle készülékkel kiváltott koponyatrauma után 30
perccel illetve 1 órával 100 μg icv-beadott PACAP- vagy vivőanyag-kezelésben részesült
patkányokat a traumát követően 2 órával perfundáltuk. A szagittális agytörzsi metszetek
egy hányadán az IAT markerével (anti-APP), a szomszédos metszeteken pedig az NFC
markerével (RMO-14) immunhisztokémiát végeztünk. A károsodott axonok számát a
kortikospinális pályában (CSpT) és a mediális hosszanti kötegben (MLF)
összehasonlítottuk a kezelt illetve a kontrol patkányok vonatkozásában.
Eredmények: A PACAP 100 μg-os dózisban (az 5.1.2.2. alfejezet során
megfigyelt hatásos dózis) mind 30 perccel (p<0.01), mind pedig 1 órával a
koponyatrauma kiváltása után icv.-beadva szignifikánsan (p<0.001) csökkentette a CSpT
területén az anti-APP-pozitív axonok számát. Nem volt azonban hatással – a fenti
időpontokban egyikében sem – az anti-APP-pozitív axonok számára az MLF területén,
illetve az RMO-14-pozitív axonok számára egyik vizsgált agytörzsi területen sem.
Következtetések: (1) A koponyatrauma által okozott DAI progressziójának
PACAP-kezeléssel való részleges gátlására jelentős nagyságú terápiás ablak áll
rendelkezésre, ami a klinikai alkalmazhatóság alapfeltétele. (2) Az NFC-ra gyakorolt
jótékony hatás elmaradása megerősíti azokat a megfigyeléseket, amelyek szerint az IAT
és az NFC részben különböző axon-populációkat érint [73, 135], és kialakulásukban
részben különböző mechanizmusok játszanak szerepet (lásd pl. 5.1.1.2. alfejezet). (3) Az
IAT-ra gyakorolt PACAP-hatás elmaradása az MLF területén valószínűleg annak
40
köszönhető, hogy (i) az MLF területén sokkal kevesebb a károsodott axonok száma, mint
a CSpT-ben, ami megnehezíti a statisztikai kimutathatóságot, illetve (ii) az MLF területén
a traumát követően 2 órával az NFC-t mutató károsodott axon-profilok dominálnak.
5.2. A neuronális mechanoporáció, resealing és CMSP vizsgálata (Farkas O,
Lifshitz J, Povlishock JT. Mechanoporation induced by diffuse traumatic brain injury: an
irreversible or reversible response to injury? J Neurosci. 2006 Mar 22;26(12):3130-40.)
Háttér: A koponyatrauma hatására ébredő intracerebrális erők a plazmamembrán
mechanoporációját válthatják ki [38, 133]. Ez lehetőséget nyit arra, hogy az érintett
idegsejtek nagy mólsúlyú anyagokat tudjanak felvenni az extracelluláris térből, szemben
más idegsejtekkel, amelyek ép plazma-membránja nem engedi át ezeket az anyagokat. A
tartós permeabilitás-zavar az ion-homeosztázis felborulása – elsősorban az intracelluláris
Ca2+-koncentráció megnövekedése – miatt, különböző enzimkaszkádok – például a
calpain – aktiválódása révén sejthalálhoz vezethet.
Cél: Annak a hipotézisnek a tesztelése, hogy (I) a mechanoporációt szenvedett
neuronok egy része képes a sejtmembrán regenerációjára és a túlélésre, illetve, hogy (II)
a permeabilitás-zavar calpain-aktiválódással és következményes spektrin-bontással társul
a koponyatrauma által kiváltott diffúz idegsejt-károsodás esetén.
Módszerek: A koponyatrauma előtt illetve különböző idő elteltével a trauma után
adott fluorescens festékkel konjugált, nagy-mólsúlyú dextrán icv infúziójával teszteltük a
tartós permeabilitási zavar fennállását és az esetleges membrán-helyreállítódást. (A)
Formalinos perfúziós fixálás és vibratómos metszés után konfokális mikroszkóppal
vizsgáltuk a metszeteket. (B) Dextránnal konjugált festék ellenes antitest segítségével
végzett immunhisztokémia és EM-feldolgozás útján vizsgáltuk a permeabilitás-
változással asszociált ultrastrukturális elváltozásokat. (C) CMSP kimutatására képes
antitesttel végzett fluorescens immunhisztokémia után konfokális mikroszkópiával
vizsgáltuk a permeabilitás-változás és a calpain-aktiválódás ko-lokalizációját. A calpain-
asszociált ultrastrukturális elváltozások vizsgálatára a metszetek egy része EM
feldolgozásra került. Kvantitatív analízist végeztünk a permeabilitás-változást és/vagy
CMSP-t mutató idegsejtek arányának a vizsgált időpontokban való összehasonlítására.
41
Eredmények: A permeabilitás-változást szenvedett idegsejteknek kb. fele mutatott
tartósan fennálló permeabilitási zavart a vizsgált időpontokban. E sejtek egy hányadánál
súlyos sejtkárosodás jeleit észleltük, más hányaduk azonban csak minimális
ultrastrukturális elváltozásokat mutatott. A tartós permeabilitási zavar csak kevesebb,
mint 15 %-ban társult calpain aktiválódással. A sejtek egy hányadánál megtörtént a
membrán helyreállítódása, ezek nem mutattak súlyos sejtkárosodás. Olyan idegsejtek is
megfigyelhetőek voltak, melyek késői permeabilitási zavart mutattak. A CMSP-t mutató
sejtek kb. harmada nem szenvedett permeabilitás-változást. E sejtek moderált
ultrastrukturális károsodást mutattak, a nekrotikus sejtek nagy részében CMSP nem volt
kimutatható.
Következtetések: (1) A membrán-sérülés nem vezet feltétlenül gyors sejthalálhoz,
szemben azzal, amit korábban feltételeztek. A membrán-regenerációt mutató sejtek
mellett valószínűleg a tartós permeabilitási zavart mutató sejtek egy hányada is képes a
regenerációra. (2) A késői permeabilitás-zavar hátterében a trauma után órákig tartó
mérsékelten emelkedett intrakraniális nyomásnak lehet szerepe. (3) Kvantitatív
elemzéseink azt mutatják, hogy a túlélési idő növekedésével a különféle membrán-
sérülésben (elhúzódó permeabilitás-zavar, helyreállítódás, késői permeabilitási zavar)
érintett neuronok között átrendeződés megy végbe, mely során a helyreállítódott-
membránú neuronok egy hányadának újra kinyílik a membránja, míg a folyamatban
korábban nem érintett neuronok egy hányada később permeabilitás-zavart szenvedhet. Ez
a megfigyelés a késői permeabilitás-zavar jelentőségét hangsúlyozza, és felhívja a
figyelmet a trauma által kiváltott másodlagos patológiás tényezők (lásd az 1.2. alfejezetet
a Bevezetésben) szerepére. (4) A membrán-sérülés nem feltétlenül szükséges vagy
elegendő triggere a calpain aktiválódásnak, a trauma indukálta sejthalál –az adott
időkereten belül – nincs feltétlenül összefüggésben a calpain aktiválódással.
Megjegyzés: Noha megfigyeléseink felhívják a figyelmet arra, hogy a CMSP nem
döntő jelentőségű a diffúz traumás idegsejt-károsodásban, a calpain-aktiválódást a diffúz
agysérülések fontos részfolyamatának tartjuk. Ennek bizonyítékai többek között a
calpain-inhibitorokkal elért sikerek a TAI kezelésében (lásd 5.1.2.1. alfejezetet), illetve a
calpain-specifikus spektrin-lebontási termékek jelenléte a cerebrospinális folyadékban
(lásd 5.3. fejezetet).
42
5.3. A calpain- és caspase-aktiválódás szerepe a humán traumás
agykárosodásban (Farkas O, Polgar B, Szekeres-Bartho J, Doczi T, Povlishock JT,
Buki A. Spectrin breakdown products in the cerebrospinal fluid in severe head injury--
preliminary observations. Acta Neurochir (Wien). 2005 Aug;147(8):855-61.)
Háttér: Habár a koponyatrauma által kiváltott fokális illetve diffúz agysérülések
eltérő módon jönnek létre és klinikai manifesztációjuk is különböző, hasonló biokémiai
folyamatok mindkét forma progressziójában szerepet játszanak. Ilyen folyamat például a
calpain és a caspase aktiválódása [10, 11, 53, 54, 87, 100, 122]. Ezek következtében
specifikus spektrin-degradációs termékek (SBDP) felszaporodása figyelhető meg az
agyszövetben mind állatkísérletekben [10, 62, 63, 125], mind humán koponyatraumát
követően [83], illetve kísérleti állatokban a cerebrospinális folyadékban (CSF) [102].
Cél: (I) A calpain- és caspase-mediált folyamatok részvételének igazolása a
humán koponyatraumában specifikus spektrin-degradációs termékeknek a humán CSF-
ből (liquorból) való kimutatásával; (II) továbbá patomechanizmus-specifikus biokémiai
markerek azonosítása céljából annak vizsgálata, hogy ezen termékek liquor-
koncentrációja mutat-e összefüggést az agysérülés klinikai képével, például
súlyosságával és kimeneteléve
Módszerek: 9 súlyos koponyatraumát (GCS<9), 3 subarachnoidális vérzést (SAV)
illetve 3 intraventrikuláris vérzést (IVV) szenvedett, továbbá 3 agytumoros beteg – ICP
kontroll céljából behelyezett icv katéteren keresztül lebocsátott – liquor-mintáin, illetve 5
diagnosztikus-célú lumbál-punkción átesett beteg liquor-mintáin a calpain- és caspase-
specifikus SBDP-k kimutatására Western blottot végeztünk. A kapott eredményeket
denzitometriás módszerrel hasonlítottuk össze. Néhány beteg esetén ezen kóros fehérjék
jelenlétét a liquorban a traumától eltelt idő függvényében is meghatároztuk.
Eredmények: (A) A 280 kD molsúlyú intakt, és a 120 kD súlyú caspase-specifikus
spektrin-degradációs termék a koponyatraumát szenvedettek szignifikánsan nagyobb
hányadában volt jelen, mint más betegek esetén. A diagnosztikus lumbál-punkción átesett
betegek CSF-mintái nem tartalmaztak SBDP-t. (B) Az intakt spektrin, valamint a 150 kD
és a 120 kD nagyságú SBDP-k szintjét szignifikánsan magasabbnak találtuk
43
koponyatrauma esetén, mint a vizsgált nem-traumás agysérülésekben. (C) A fehérjék
liquor-szintje a trauma utáni 2-3. napon tetőzött, majd visszatért a kiindulási szintre, ezt
követően egy újabb emelkedés volt tapasztalható. (D) A spektrin-szintek és a klinikai
paraméterek összehasonítása során sem a súlyossági fokkal (GCS alapján), sem a
kimenetellel (GOS alapján), sem pedig az ICP-emelkedés mértékével nem találtunk
szignifikáns összefüggést.
Következtetések: (1) A lumbál-punkción átesett betegek esetén tapasztalt negatív
eredmény igazolja, hogy az SBDP-k megjelenése a CSF-ben agysérülés és/vagy
emelkedett intrakraniális nyomás következménye. (2) A koponyatrauma során – más
ICP-emelkedéssel járó állapotokkal szemben – tapasztalt magasabb SPDP-szint arra utal,
hogy a megfigyelt jelenség a sérülés direkt következménye is, és nem csupán az
emelkedett ICP eredménye. (3) A vizsgált fehérjék liquor-szintjének jellegzetes idő-
összefüggése alapján feltételezhető, hogy a calpain- és caspase-specifikus SBDP-k
vizsgálata a jövőben hasznos pathomechanizmus-specifikus biomarker lehet a
koponyatraumát szenvedettek állapotának nyomon követésében.
Megjegyzés: (I) A kísérletbe bevont betegek viszonylag alacsony száma
egyértelműen jelzi munkánk korlátait, a kísérletek pusztán „tájékozódó” jellegét.
Valószínűleg a kis esetszám az oka annak is, hogy a megfigyelt tendenciák ellenére az
SBDP-k liquor-szintje és a klinikai paraméterek között szignifikáns összefüggés nem volt
megfigyelhető. (II) Klinikánkon megkezdődött a súlyos koponyatraumát szenvedett és
ICP-monitorozás céljából behelyezett kamradrénes betegek CSF-mintáinak
szisztematikus gyűjtése, ami a jövőben nagymértékben hozzájárulhat a különböző
biomarkerek azonosítása és diagnosztikus/prognosztikus jelentőségük felmérése céljából
indított vizsgálataink kiterjesztéséhez.
44
6. A TÉMÁBAN ELÉRT, NEMZETKÖZI-JELENTŐSÉGŰ EREDMÉNYEK
6.1. A traumás axonkárosodás
6.1.1. Az ultrastrukturális axon-kompakció
I. Elsőként igazoltuk, hogy a koponyatrauma következtében kialakuló, argyrophil-
III típusú ezüstözési technikával festődő axonok azonosak a koponyatrauma
következtében ultrastrukturális kompakciót szenvedett axonokkal.
II. A traumás axon-kompakció mechanizmusának vizsgálata során elsőként
igazoltuk, hogy
(A) az ultrastrukturális axon-kompakció a trauma pillanatában következik be, nem
pedig 5-15 perces késéssel, ahogy azt korábban feltételezték,
(B) a citoszkeletális kompakció viszonylag hosszú, akár 1 mm-es axon-szakaszokra is
kiterjedhet, azaz a korábbi feltételezésekkel ellentétben nem EM-szinten „fokális”
elváltozás,
(C) az in vivo és a post mortem koponyatrauma egyforma morfológiai elváltozásokat
eredményez, mind a (i) a hyperbasophilia, (ii) a III-típusú argyrophilia, (iii) a hiper-
elektrondenzitás, mind pedig (iv) az ultrastrukturális kompakció tekintetében,
(D) az ultrastrukturális axon-kompakció foka a post mortem trauma esetén nem
különbözik az in vivo trauma esetén meghatározottól,
(E) a trauma-indukálta ezüstöződés minkét esetben ”minden vagy semmi” jellegű,
azaz az érintett szóma-dendrit domének teljes egészére, illetve hosszú axon-szakaszokra
terjed ki, de általában nem egyazon idegsejt szóma-dendrit doménjét és axonját érinti.
III. Kísérleteink során kidolgoztunk egy olyan új koponyatrauma modellt, mely
alkalmas az ultrastrukturális axon-kompakció szelektív előidézésére.
IV. A koponyatrauma hatására kompaktálódott axonok sorsának vizsgálatával
45
elsőként írtuk le
(A) a kompakciót szenvedett axonokban hónapok alatt lejátszódó morfológiai
elváltozásokat, továbbá igazoltuk, hogy
(B) az érintett axonok >50%-a 1 napon belül, míg <10%-a 1 nap és 1 hét között
regenerálódik, azaz visszanyeri eredeti ultrastruktúráját. A többi kompaktálódott axon
néhány hónap alatt regenerálódik,
(C) a kompaktálódott axonok az agykéregben több óra elteltével sem festődnek az
agytörzsi axon-kompakció kimutatására széles körben elterjedt markerrel, az RMO-14
antitesttel.
(D) A kompaktálódott axonok nem mutatnak pozitivitást az IAT markerével, ami
megerősíti azt az elképzelést, hogy az intraaxonális transzportzavar és az axon-
kompakció döntően két különböző axon-populácót érint.
6.1.2. Terápiás beavatkozások
I. Kísérleteinkkel elsőként igazoltuk, hogy a sejt-permeábilis calpain-inhibitor,
MDL-28170 szignifikánsan csökkenti az ultrastrukturális kompakciót szenvedett, illetve
az intraaxonális transzport-zavart mutató axonok számát a TAI kialakulására jellegzetes
agytörzsi régiókban, azaz a CSpT és az MLF területén.
II. Elsőként igazoltuk, hogy a PACAP traumás axonkárosodás esetén szignifikánsan
csökkenti az intraaxonális transzport-zavart mutató axonok számát az agytörzsben.
III. Traumás axonkárosodás esetén kísérleteink során elsőként határoztuk meg
PACAP-kezelés tekintetében
(A) a hatásos dózist
(B) a terápiás ablakot.
6.2. A diffúz idegsejt-károsodás
46
Kísérleteink során elsőként vizsgáltuk a koponyatrauma által kiváltott idegsejt-
membrán permeabilitási zavar alakulását órákkal a traumát követően in vivo modellben.
Elsőként igazoltuk, hogy
(A) a traumás sejtmembrán-károsodás a korábbi nézetekkel ellentétben, az idegsejtek
egy részében nem vezet súlyos ultrastrukturális elváltozásokhoz és gyors sejthalálhoz,
(B) a tartós sejtmembrán permeabilitási zavar csak kevesebb, mint 15 %-ban társul
calpain aktiválódással és következményes strukturális fehérjebontással,
(C) a károsodott sejtmembrán a sejtek egy részénél helyreállítódik, ezek a sejtek nem
szenvednek súlyos károsodást,
(D) sejtmembrán permeabilitási zavar az idegsejtek egy részénél órákkal a traumát
követően alakul ki, melynek hátterében feltételezhető a kísérleteink során észlelt, tartósan
20 Hgmm körüli ICP emelkedés,
(E) a calpain-aktiválódást és következményes CMSP-t mutató sejtek kb. harmada
nem szenved permeabilitás-változást. E sejtek közepes fokú ultrastrukturális károsodást
mutatnak, a nekrotikus sejtek nagy részében nem igazolható CMSP.
(F) a túlélési idő növekedésével a membrán-sérülésben érintett neuronok között
(elhúzódó permeabilitás-zavar, helyreállítódás, késői permeabilitási zavar) átrendeződés
megy végbe, mely során a helyreállítódott-membránú neuronok egy hányadának újra
kinyílik a membránja, míg a folyamatban korábban nem érintett neuronok egy hányada
később szintén permeabilitás-zavart szenvedhet.
6.3. A calpain- és caspase-aktiválódás szerepe a humán traumás
agykárosodásban
I. Kísérleteink során elsőként vizsgáltuk a calpain és caspase aktiváció és
következményes strukturális fehérjebontás során létrejött spektrin degradációs termékek
jelenlétét humán cerebrospinális folyadékban. Megerősítettük, hogy
II. a calpain- és caspase-aktiváció fontos patobiokémiai tényező a traumás
agykárosodás kialakulásában, továbbá elsőként igazoltuk, hogy
47
III. a specifikus SBDP-k megjelenése a CSF-ben agysérülés és/vagy emelkedett
intrakraniális nyomás következménye,
IV. az intakt spektrin, valamint a 150 kD és a120 kD nagyságú SBDP-k CSF-szintje
szignifikánsan magasabb koponyatrauma esetén, mint a vizsgált nem-traumás
agysérülésekben,
V. a spektrin-degradációs termékek liquor-szintje jellegzetes idő-kinetikát mutat,
azaz a trauma utáni 2-3. napon tetőzik, majd visszatér a kiindulási szintre, ezt követően
egy újabb emelkedés tapasztalható.
VI. Kísérleteink nagyban hozzájárultak egy nemzetközi, multicentrikus, a PTE ÁOK
Idegsebészeti Klinika részvételével folyó vizsgálat indításához, amely különböző
biomarkerek azonosítása és diagnosztikus/prognosztikus jelentőségük felmérése céljából
– több más fehérjével együtt – a spektrin degradációs termékek vizsgálatát is célul tűzte.
48
IRODALOMJEGYZÉK
1. Adams JH (1982). Diffuse axonal injury in non-missile head injury. Injury 13: 444-445.
2. Adams JH, Doyle D, Ford I, Gennarelli TA, Graham DI and McLellan DR (1989). Diffuse axonal injury in head injury: definition, diagnosis and grading. Histopathology 15: 49-59.
3. Barr RM, Gean AD and Le TH (2006). Craniofacial Trauma. In: Fundamentals of Diagnostic Radiology (Brant, W. E and Helms, C. A), Philadelphia: Lippincott Williams & Wilkins pp 55-85.
4. Barron KD (1983). Comparative Observations on the Cytologic Reactions of Central and Peripheral Nerve Cells to Axotomy. In: Spinal Cord Reconstruction (Kao, C. C, Bunge, R. P, and Reier, P. J.), New York: Raven Press. pp 7-40.
5. Blumbergs PC, Jones NR and North JB (1989). Diffuse axonal injury in head trauma. J.Neurol.Neurosurg.Psychiatry 52: 838-841.
6. Blumbergs PC, Scott G, Manavis J, Wainwright H, Simpson DA and McLean AJ (1994). Staining of amyloid precursor protein to study axonal damage in mild head injury. Lancet 344: 1055-1056.
7. Bramlett HM, Kraydieh S, Green EJ and Dietrich WD (1997). Temporal and regional patterns of axonal damage following traumatic brain injury: a beta-amyloid precursor protein immunocytochemical study in rats. J.Neuropathol.Exp.Neurol. 56: 1132-1141.
8. Buki A, Koizumi H and Povlishock JT (1999). Moderate posttraumatic hypothermia decreases early calpain-mediated proteolysis and concomitant cytoskeletal compromise in traumatic axonal injury. Exp.Neurol. 159: 319-328.
9. Buki A, Okonkwo DO and Povlishock JT (1999). Postinjury cyclosporin A administration limits axonal damage and disconnection in traumatic brain injury. J.Neurotrauma 16: 511-521.
10. Buki A, Siman R, Trojanowski JQ and Povlishock JT (1999). The role of calpain-mediated spectrin proteolysis in traumatically induced axonal injury. J.Neuropathol.Exp.Neurol. 58: 365-375.
11. Buki A, Okonkwo DO, Wang KK and Povlishock JT (2000). Cytochrome c release and caspase activation in traumatic axonal injury. J.Neurosci. 20: 2825-2834.
12. Cai J, Yang J and Jones DP (1998). Mitochondrial control of apoptosis: the role of cytochrome c. Biochim.Biophys.Acta 1366: 139-149.
13. Cammermeyer J (1960). The post mortem origin and mechanism of neuronal Hyperchromatosis and nuclear pyknosis. Exp. Neurol. 2:379-405.
14. Cammermeyer J (1961). The importance of avoiding "dark" neurons in experimental neuropathology. Acta Neuropathol (Berl) 1:245-270.
15. Carafoli E and Molinari M (1998). Calpain: a protease in search of a function? Biochem. Biophys. Res.Commun. 247: 193-203.
16. Cernak I, Chapman SM, Hamlin GP and Vink R (2002). Temporal characterisation of pro- and anti-apoptotic mechanisms following diffuse traumatic brain injury in rats. J.Clin.Neurosci. 9: 565-572.
17. Cernak I, Vink R, Zapple DN, Cruz MI, Ahmed F, Chang T, Fricke ST and Faden AI (2004). The pathobiology of moderate diffuse traumatic brain injury as identified using a new experimental model of
49
injury in rats. Neurobiol.Dis. 17: 29-43.
18. Chan SL and Mattson MP (1999). Caspase and calpain substrates: roles in synaptic plasticity and cell death. J.Neurosci.Res. 58: 167-190.
19. Christman CW, Grady MS, Walker SA, Holloway KL and Povlishock JT (1994). Ultrastructural studies of diffuse axonal injury in humans. J.Neurotrauma 11: 173-186.
20. Clark RS, Chen J, Watkins SC, Kochanek PM, Chen M, Stetler RA, Loeffert JE and Graham SH (1997). Apoptosis-suppressor gene bcl-2 expression after traumatic brain injury in rats. J.Neurosci. 17: 9172-9182.
21. Colicos MA, Dixon CE and Dash PK (1996). Delayed, selective neuronal death following experimental cortical impact injury in rats: possible role in memory deficits. Brain Res. 739: 111-119.
22. Conti AC, Raghupathi R, Trojanowski JQ and McIntosh TK (1998). Experimental brain injury induces regionally distinct apoptosis during the acute and delayed post-traumatic period. J.Neurosci. 18: 5663-5672.
23. Cortez SC, McIntosh TK and Noble LJ (1989). Experimental fluid percussion brain injury: vascular disruption and neuronal and glial alterations. Brain Res. 20;482: 271-282.
24. Csepregi G, Buki A, Futo J, Sandor J, Gobl G and Doczi T (2007). Management of patients with severe head injury in Hungary, in 2002. Orv Hetil. 148(17): 771-777.
25. Csordás A, Mázló M, Gallyas F (2003). Recovery versus death of "dark" (compacted) neurons in non-impaired parenchymal environment. Light and electron microscopic observations. Acta Neuropathol.(Berl) 106.37-49.
26. Dawson TM, Steiner JP, Dawson VL, Dinerman JL, Uhl GR and Snyder SH (1993). Immunosuppressant FK506 enhances phosphorylation of nitric oxide synthase and protects against glutamate neurotoxicity. Proc.Natl.Acad.Sci.U.S.A 90: 9808-9812.
27. Dawson TM, Steiner JP, Lyons WE, Fotuhi M, Blue M and Snyder SH (1994). The immunophilins, FK506 binding protein and cyclophilin, are discretely localized in the brain: relationship to calcineurin. Neuroscience 62: 569-580.
28. Dixon CE, Clifton GL, Lighthall JW, Yaghmai AA and Hayes RL (1991). A controlled cortical impact model of traumatic brain injury in the rat. J.Neurosci.Methods 39: 253-262.
29. Elder GA, Friedrich VL Jr, Bosco P, Kang C, Gourov A, Tu PH, Lee VM, Lazzarini RA (1998). Absence of the mid-sized neurofilament subunit decreases axonal calibers, levels of light neurofilament (NF-L), and neurofilament content. J Cell Biol. 141(3):727-739.
30. Erb DE and Povlishock JT (1988). Axonal damage in severe traumatic brain injury: an experimental study in cat. Acta Neuropathol.(Berl) 76: 347-358.
31. Foda MA and Marmarou A (1994). A new model of diffuse brain injury in rats. Part II: Morphological characterization. J.Neurosurg. 80: 301-313.
32. Gallyas F, Güldner FH, Zoltay G, Wolff JR (1990). Golgi-like demonstration of "dark" neurons with an argyrophil III method for experimental neuropathology. Acta Neuropathol.(Berl) 79:620-628.
33. Gallyas F and Zoltay G (1992). An immediate light microscopic response of neuronal somata, dendrites and axons to non-contusing concussive head injury in the rat. Acta Neuropathol.(Berl.) 83: 386-
50
393.
34. Gallyas F, Zoltay, G and Balas I (1992). An immediate light microscopic response of neuronal somata, dendrites and axons to contusing concussive head injury in the rat. Acta Neuropathol.(Berl.) 83: 394-401.
35. Gallyas F, Zoltay G and Horvath Z (1992). Light microscopic response of neuronal somata, dendrites and axons to post-mortem concussive head injury. Acta Neuropathol.(Berl.) 83: 499-503
36. Gallyas F, Hsu M, Buzsaki G (1993). Four modified silver methods for thick sections of formaldehyde-fixed mammalian central nervous tissue: 'dark' neurons, perikarya of all neurons, microglial cells and capillaries. J Neurosci Methods. 50(2):159-64.
37. Gallyas F, Gasz B, Szigeti A, Mazlo M (2006). Pathological circumstances impair the ability of "dark" neurons to undergo spontaneous recovery. Brain Res. 1110(1):211-20.
38. Geddes DM, Cargill RS and LaPlaca MC (2003). Mechanical stretch to neurons results in a strain rate and magnitude-dependent increase in plasma membrane permeability. J.Neurotrauma 20: 1039-1049.
39. Gennarelli TA, Thibault LE, Adams JH, Graham DI, Thompson CJ and Marcincin RP (1982). Diffuse axonal injury and traumatic coma in the primate. Ann.Neurol. 12: 564-574.
40. Gennarelli TA (1994). Animate models of human head injury. J Neurotrauma 11:357-368.
41. Gennarelli TA and Graham DI (1998). Neuropathology of the Head Injuries. Semin.Clin.Neuropsychiatry 3: 160-175.
42. Gentleman SM, Nash MJ, Sweeting CJ, Graham DI and Roberts GW (1993). Beta-amyloid precursor protein (beta APP) as a marker for axonal injury after head injury. Neurosci.Lett. 160: 139-144.
43. Grady MS, McLaughlin MR, Christman CW, Valadka AB, Fligner CL and Povlishock JT (1993). The use of antibodies targeted against the neurofilament subunits for the detection of diffuse axonal injury in humans. J.Neuropathol.Exp.Neurol. 52: 143-152.
44. Graham DI (1996). Neuropathology of Head Injury. In: Neurotrauma (Narayan, R. K., Wilberger, J. E., and Povlishock, J. T.), New York: McGraw-Hill. pp 43-59.
45. Graham DI, McIntosh TK, Maxwell WL and Nicoll JA (2000). Recent advances in neurotrauma. J.Neuropathol.Exp.Neurol. 59: 641-651.
46. Hall ED, Sullivan PG, Gibson TR, Pavel KM, Thompson BM and Scheff SW (2005). Spatial and temporal characteristics of neurodegeneration after controlled cortical impact in mice: more than a focal brain injury. J.Neurotrauma 22: 252-265.
47. Hankins L, Taber KH and Yeakley J (1996). Magnetic resonance imaging in head injury. In: Neurotrauma (Narayan, R. K., Wilberger, J. E., and Povlishock, J.), New York: McGraw-Hill. pp 151-161.
48. Hicks R, Soares H, Smith D and McIntosh T (1996). Temporal and spatial characterization of neuronal injury following lateral fluid-percussion brain injury in the rat. Acta Neuropathol.(Berl) 91: 236-246.
49. Holsinger T, Steffens DC, Phillips C, Helms MJ, Havlik RJ, Breitner JC, Guralnik JM and Plassman BL (2002). Head injury in early adulthood and the lifetime risk of depression. Arch.Gen.Psychiatry 59: 17-22.
51
50. Horner MD, Ferguson PL, Selassie AW, Labbate LA, Kniele K and Corrigan JD (2005). Patterns of alcohol use 1 year after traumatic brain injury: a population-based, epidemiological study. J.Int.Neuropsychol.Soc. 11: 322-330.
51. Jafari SS, Maxwell WL, Neilson M and Graham DI (1997). Axonal cytoskeletal changes after non-disruptive axonal injury. J.Neurocytol. 26: 207-221.
52. Jafari SS, Nielson M, Graham DI and Maxwell WL (1998). Axonal cytoskeletal changes after nondisruptive axonal injury. II. Intermediate sized axons. J.Neurotrauma 15: 955-966.
53. Kampfl A, Posmantur R, Nixon R, Grynspan F, Zhao X, Liu SJ, Newcomb JK, Clifton GL. and Hayes RL (1996). mu-calpain activation and calpain-mediated cytoskeletal proteolysis following traumatic brain injury. J.Neurochem. 67: 1575-1583.
54. Kampfl A, Posmantur RM, Zhao X, Schmutzhard E, Clifton GL and Hayes RL (1997). Mechanisms of calpain proteolysis following traumatic brain injury: implications for pathology and therapy: implications for pathology and therapy: a review and update. J.Neurotrauma 14: 121-134.
55. Katahira M, Yone K, Arishima Y, Nagamine T, Komiya S, Iwata S et al (2003). The neuroprotective effects of PACAP on spinal cord injury. Regul Pept 115:49 (abstract)
56. Kellermayer R, Zsombok A, Auer T, Gallyas F (2206). Electric-shock experiments support the gel-phase transition theory of the formation of “dark” neurons. Cell Biol Int 30:175-82.
57. Kimura H, Kawatani M, Ito E, Ishikawa K (2003). Effects of pituitary adenylate cyclase activating polypeptide on facial nerve recovery in the guinea pig. Laryngoscope113:1000– 6.
58. Koizumi H and Povlishock JT (1998). Posttraumatic hypothermia in the treatment of axonal damage in an animal model of traumatic axonal injury. J.Neurosurg. 89: 303-309.
59. Kovács B, Bukovics P, Gallyas F (2007). Morphological effects of transcardially perfused sodium dodecylsulfate on the rat brain: Cell-biologic aspects. Biol Cell (in press)
60. Krajewski S, Krajewska M, Ellerby LM, Welsh K., Xie Z, Deveraux QL, et al (1999). Release of caspase-9 from mitochondria during neuronal apoptosis and cerebral ischemia. Proc.Natl.Acad.Sci.U.S.A. 96: 5752-5757.
61. Kupina NC, Nath R, Bernath EE, Inoue J, Mitsuyoshi A, Yuen PW, Wang KK and Hall ED (2001). The novel calpain inhibitor SJA6017 improves functional outcome after delayed administration in a mouse model of diffuse brain injury. J.Neurotrauma 18: 1229-1240.
62. Kupina NC, Detloff MR, Dutta S and Hall ED (2002). Neuroimmunophilin ligand V-10,367 is neuroprotective after 24-hour delayed administration in a mouse model of diffuse traumatic brain injury. J.Cereb.Blood Flow Metab 22: 1212-1221.
63. Kupina NC, Detloff MR, Bobrowski WF, Snyder BJ and Hall ED (2003). Cytoskeletal protein degradation and neurodegeneration evolves differently in males and females following experimental head injury. Exp.Neurol. 180: 55-73.
64. Langlois JA, Rutland-Brown W and Wald MM (2006). The epidemiology and impact of traumatic brain injury: a brief overview. J.Head Trauma Rehabil. 21: 375-378.
65. Lee VM, Carden MJ, Schlaepfer WW and Trojanowski JQ (1987). Monoclonal antibodies distinguish several differentially phosphorylated states of the two largest rat neurofilament subunits (NF-H and NF-M) and demonstrate their existence in the normal nervous system of adult rats. J.Neurosci. 7: 3474-
52
3488.
66. Leker RR, Shohami E (2002). Cerebral ischemia and trauma—different etiologies yet similar mechanisms: neuroprotective opportunities. Brain Res Rev 39:55– 73.
67. Lewen A, Li GL, Nilsson P, Olsson Y and Hillered L (1995). Traumatic brain injury in rat produces changes of beta-amyloid precursor protein immunoreactivity. Neuroreport 6: 357-360.
68. Li PA, Howlett W, He QP, Miyashita H, Siddiqui M and Shuaib A. (1998). Postischemic treatment with calpain inhibitor MDL 28170 ameliorates brain damage in a gerbil model of global ischemia. Neurosci.Lett. 247: 17-20.
69. Lin X, Zhi D and Zhang S. (2001). Inhibiting effect of moderate hypothermia on cell apoptosis after diffuse brain injury in rats. Chin J.Traumatol. 4: 14-19.
70. Marion DW and White MJ (1996). Treatment of experimental brain injury with moderate hypothermia and 21- aminosteroids. J.Neurotrauma. 13: 139-147.
71. Markgraf CG, Velayo NL, Johnson MP, McCarty DR, Medhi S, Koehl JR et al (1998). Six-hour window of opportunity for calpain inhibition in focal cerebral ischemia in rats. Stroke 29: 152-158.
72. Marmarou A, Foda MA, van den Brink W, Campbell J, Kita H, Demetriadou K. (1994). A new model of diffuse brain injury in rats. Part I: Pathophysiology and biomechanics. J Neurosurg. 80(2):291-300.
73. Marmarou CR, Walker SA., Davis CL and Povlishock JT (2005). Quantitative analysis of the relationship between intra- axonal neurofilament compaction and impaired axonal transport following diffuse traumatic brain injury. J.Neurotrauma 22: 1066-1080.
74. Marmarou CR and Povlishock JT (2006). Administration of the immunophilin ligand FK506 differentially attenuates neurofilament compaction and impaired axonal transport in injured axons following diffuse traumatic brain injury. Exp.Neurol. 197: 353-362.
75. Maxwell WL, Irvine A, Graham DI, Adams JH, Gennarelli TA, Tipperman R and Sturatis M (1991). Focal axonal injury: the early axonal response to stretch. J.Neurocytol. 20: 157-164.
76. Maxwell WL, Islam MN, Graham DI and Gennarelli TA (1994). A qualitative and quantitative analysis of the response of the retinal ganglion cell soma after stretch injury to the adult guinea-pig optic nerve. J.Neurocytol. 23: 379-392.
77. Maxwell WL, McCreath BJ, Graham DI and Gennarelli TA (1995). Cytochemical evidence for redistribution of membrane pump calcium- ATPase and ecto-Ca-ATPase activity, and calcium influx in myelinated nerve fibres of the optic nerve after stretch injury. J.Neurocytol. 24: 925-942.
78. Maxwell WL (1996). Histopathological changes at central nodes of Ranvier after stretch- injury. Microsc.Res.Tech. 34: 522-535.
79. Maxwell WL and Graham DI (1997). Loss of axonal microtubules and neurofilaments after stretch-injury to guinea pig optic nerve fibers. J.Neurotrauma 14: 603-614.
80. Maxwell WL, Povlishock JT and Graham DL (1997). A mechanistic analysis of nondisruptive axonal injury: a review. J.Neurotrauma 14: 419-440.
81. Maxwell WL, Donnelly S, Sun X, Fenton T, Puri N and Graham DI (1999). Axonal cytoskeletal responses to nondisruptive axonal injury and the short-term effects of posttraumatic hypothermia.
53
J.Neurotrauma 16: 1225-1234.
82. Maxwell WL, Watson A, Queen R, Conway B, Russell D, Neilson M and Graham DI (2005). Slow, medium, or fast re-warming following post-traumatic hypothermia therapy? An ultrastructural perspective. J.Neurotrauma 22: 873-884.
83. McCracken E, Hunter AJ, Patel S, Graham DI and Dewar D (1999). Calpain activation and cytoskeletal protein breakdown in the corpus callosum of head-injured patients. J.Neurotrauma 16: 749-761.
84. Meaney DF, Ross DT, Winkelstein BA, Brasko J, Goldstein D, Bilston LB, Thibault LE and Gennarelli TA (1994). Modification of the cortical impact model to produce axonal injury in the rat cerebral cortex. J.Neurotrauma 11: 599-612.
85. Morales DM, Marklund N, Lebold D, Thompson HJ, Pitkanen A, Maxwell WL, Longhi L, Laurer H, Maegele M, Neugebauer E, Graham DI, Stocchetti N and McIntosh TK (2005). Experimental models of traumatic brain injury: do we really need to build a better mousetrap? Neuroscience 136: 971-989.
86. Neumar RW, Xu YA, Gada H, Guttmann RP and Siman R (2003). Cross-talk between calpain and caspase proteolytic systems during neuronal apoptosis. J.Biol.Chem. 278: 14162-14167.
87. Newcomb JK, Kampfl A, Posmantur RM, Zhao X, Pike BR, Liu SJ, Clifton GL and Hayes RL (1997). Immunohistochemical study of calpain-mediated breakdown products to alpha-spectrin following controlled cortical impact injury in the rat. J.Neurotrauma 14: 369-383.
88. Newcomb JK, Pike BR, Zhao X, Banik NL and Hayes RL (1999). Altered calpastatin protein levels following traumatic brain injury in rat. J.Neurotrauma 16: 1-11.
89. Newcomb JK, Zhao X, Pike BR and Hayes RL (1999). Temporal profile of apoptotic-like changes in neurons and astrocytes following controlled cortical impact injury in the rat. Exp.Neurol. 158: 76-88.
90. Newman GR, Jasani AB (1998) Silver development in microscopy and bioanalysis: past and present J Pathol 186:119-125.
91. Nonaka N, Banks WA, Mizushima H, Shioda S, Morley JE (2000). Regional differences in PACAP transport across the blood – brain barrier in mice: a possible influence of strain, amyloid h protein and age. Peptides 23:2197– 202.
92. Oehmichen M, Meissner C, Schmidt V, Pedal I, Konig HG and Saternus KS (1998). Axonal injury--a diagnostic tool in forensic neuropathology? A review. Forensic Sci.Int. 95: 67-83.
93. Okonkwo DO, Pettus EH, Moroi J and Povlishock JT (1998). Alteration of the neurofilament sidearm and its relation to neurofilament compaction occurring with traumatic axonal injury. Brain Res. 784: 1-6.
94. Okonkwo DO and Povlishock JT (1999). An intrathecal bolus of cyclosporin A before injury preserves mitochondrial integrity and attenuates axonal disruption in traumatic brain injury. J.Cereb.Blood Flow Metab 19: 443-451.
95. Okonkwo DO, Buki A, Siman R and Povlishock JT (1999). Cyclosporin A limits calcium-induced axonal damage following traumatic brain injury. Neuroreport 10: 353-358.
96. Okonkwo DO, Melon DE, Pellicane AJ, Mutlu LK, Rubin DG, Stone JR and Helm GA (2003). Dose-response of cyclosporin A in attenuating traumatic axonal injury in rat. Neuroreport 14: 463-466.
54
97. Pant HC (1988). Dephosphorylation of neurofilament proteins enhances their susceptibility to degradation by calpain. Biochem.J. 256: 665-668.
98. Pettus EH, Christman CW, Giebel ML and Povlishock JT (1994). Traumatically induced altered membrane permeability: its relationship to traumatically induced reactive axonal change. J.Neurotrauma 11: 507-522.
99. Pettus EH and Povlishock JT (1996). Characterization of a distinct set of intra-axonal ultrastructural changes associated with traumatically induced alteration in axolemmal permeability. Brain Res. 722: 1-11.
100. Pike BR, Zhao X, Newcomb JK, Posmantur RM, Wang KK and Hayes RL (1998). Regional calpain and caspase-3 proteolysis of alpha-spectrin after traumatic brain injury. Neuroreport 9: 2437-2442.
101. Pike BR, Zhao X, Newcomb JK, Glenn CC, Anderson DK and Hayes RL (2000). Stretch injury causes calpain and caspase-3 activation and necrotic and apoptotic cell death in septo-hippocampal cell cultures. J.Neurotrauma 17: 283-298.
102. Pike BR, Flint J, Dutta S, Johnson E, Wang KK and Hayes RL (2001). Accumulation of non-erythroid alpha II-spectrin and calpain-cleaved alpha II-spectrin breakdown products in cerebrospinal fluid after traumatic brain injury in rats. J.Neurochem. 78: 1297-1306.
103. Plassman BL, Havlik RJ, Steffens DC, Helms MJ, Newman TN, Drosdick D et al (2000). Documented head injury in early adulthood and risk of Alzheimer's disease and other dementias. Neurology 55: 1158-1166.
104. Pollack GH (1996). Phase transitions and the molecular mechanism of contraction. Biophys Chem 59:315-28.
105. Posmantur R, Kampfl A, Siman R, Liu J, Zhao X, Clifton GL and Hayes RL (1997). A calpain inhibitor attenuates cortical cytoskeletal protein loss after experimental traumatic brain injury in the rat. Neuroscience 77: 875-888.
106. Povlishock JT, Becker DP, Cheng CL and Vaughan GW (1983). Axonal change in minor head injury. J.Neuropathol.Exp.Neurol. 42: 225-242.
107. Povlishock JT and Becker DP (1985). Fate of reactive axonal swellings induced by head injury. Lab Invest 52: 540-552.
108. Povlishock JT (1986). Traumatically induced axonal damage without concomitant change in focally related neuronal somata and dendrites. Acta Neuropathol.(Berl) 70: 53-59.
109. Povlishock JT (1992). Traumatically induced axonal injury: pathogenesis and pathobiological implications. Brain Pathol. 2: 1-12.
110. Povlishock JT, Erb DE and Astruc J (1992). Axonal response to traumatic brain injury: reactive axonal change, deafferentation, and neuroplasticity. J.Neurotrauma 9 Suppl 1:S189-200.: S189-S200.
111. Povlishock JT (1993). Pathobiology of traumatically induced axonal injury in animals and man. Ann.Emerg.Med. 22: 980-986.
112. Povlishock JT and Pettus EH (1996). Traumatically induced axonal damage: evidence for enduring changes in axolemmal permeability with associated cytoskeletal change. Acta Neurochir.Suppl (Wien.) 66: 81-86.
113. Povlishock JT, Marmarou A, McIntosh T, Trojanowski JQ and Moroi J (1997). Impact acceleration
55
injury in the rat: evidence for focal axolemmal change and related neurofilament sidearm alteration. J.Neuropathol.Exp.Neurol. 56: 347-359.
114. Povlishock JT, Buki A, Koiziumi H, Stone J and Okonkwo DO (1999). Initiating mechanisms involved in the pathobiology of traumatically induced axonal injury and interventions targeted at blunting their progression. Acta Neurochir.Suppl (Wien.) 73:15-20.: 15-20.
115. Povlishock JT and Stone J (2001). Traumatic axonal injury. In: Head Trauma: Basic, preclinical, and clinical directions (Miller LP, Hayers RL, and Newcomb JK), New York: John Wiley abd Sons, Inc. pp 281-301.
116. Raghupathi R, Graham DI and McIntosh TK (2000). Apoptosis after traumatic brain injury. J.Neurotrauma 17: 927-938.
117. Reglodi D, Somogyvari-Vigh A, Vigh S, Kozicz T and Arimura A (2000). Delayed systemic administration of PACAP38 is neuroprotective in transient middle cerebral artery occlusion in the rat. Stroke 31: 1411-1417.
118. Reglodi D, Tamas A, Somogyvari-Vigh A, Szanto Z, Kertes E, Lenard L, Arimura A and Lengvari I (2002). Effects of pretreatment with PACAP on the infarct size and functional outcome in rat permanent focal cerebral ischemia. Peptides 23: 2227-2234.
119. Rimel RW, Giordani B, Barth JT and Jane JA (1982). Moderate head injury: completing the clinical spectrum of brain trauma. Neurosurgery 11: 344-351.
120. Roberts-Lewis JM, Savage MJ, Marcy VR, Pinsker LR and Siman R (1994). Immunolocalization of calpain I-mediated spectrin degradation to vulnerable neurons in the ischemic gerbil brain. J.Neurosci. 14: 3934-3944.
121. Runnerstam M, Bao F, Huang Y, Shi J, Gutierrez E, Hamberger A, Hansson HA, Viano D and Haglid K (2001). A new model for diffuse brain injury by rotational acceleration: II. Effects on extracellular glutamate, intracranial pressure, and neuronal apoptosis. J.Neurotrauma 18: 259-273.
122. Saatman KE, Bozyczko-Coyne D, Marcy V, Siman R and McIntosh TK (1996). Prolonged calpain-mediated spectrin breakdown occurs regionally following experimental brain injury in the rat. J.Neuropathol.Exp.Neurol. 55: 850-860.
123. Saatman KE, Murai H, Bartus RT, Smith DH, Hayward NJ, Perri BR and McIntosh TK (1996). Calpain inhibitor AK295 attenuates motor and cognitive deficits following experimental brain injury in the rat. Proc.Natl.Acad.Sci.U.S.A 93: 3428-3433.
124. Saatman KE, Zhang C, Bartus RT and McIntosh TK (2000). Behavioral efficacy of posttraumatic calpain inhibition is not accompanied by reduced spectrin proteolysis, cortical lesion, or apoptosis. J.Cereb.Blood Flow Metab 20: 66-73.
125. Saatman KE, Abai, B, Grosvenor A, Vorwerk CK, Smith DH and Meaney DF (2003). Traumatic axonal injury results in biphasic calpain activation and retrograde transport impairment in mice. J.Cereb.Blood Flow Metab 23: 34-42.
126. Sabatini DM, Lai MM and Snyder SH (1997). Neural roles of immunophilins and their ligands. Mol.Neurobiol. 15: 223-239.
127. Seki T, Izumi S, Shioda S, Arimura A (2003). Pituitary adenylate cyclase activating polypeptide (PACAP) protects ganglion cell death against cutting of optic nerve in the rat retina. Regul Pept 115:55 (abstract)
56
128. Sherriff FE, Bridges LR, Gentleman SM, Sivaloganathan S and Wilson S (1994). Markers of axonal injury in post mortem human brain. Acta Neuropathol.(Berl) 88: 433-439.
129. Sherriff FE, Bridges LR and Sivaloganathan S (1994). Early detection of axonal injury after human head trauma using immunocytochemistry for beta-amyloid precursor protein. Acta Neuropathol.(Berl) 87: 55-62.
130. Siman R, Baudry M and Lynch G (1984). Brain fodrin: substrate for calpain I, an endogenous calcium-activated protease. Proc.Natl.Acad.Sci.U.S.A. 81: 3572-3576.
131. Singleton RH, Stone JR, Okonkwo DO, Pellicane AJ and Povlishock JT (2001). The immunophilin ligand FK506 attenuates axonal injury in an impact-acceleration model of traumatic brain injury. J.Neurotrauma 18: 607-614.
132. Singleton RH, Zhu J, Stone JR and Povlishock JT (2002). Traumatically induced axotomy adjacent to the soma does not result in acute neuronal death. J.Neurosci. 22: 791-802.
133. Singleton RH and Povlishock JT (2004). Identification and characterization of heterogeneous neuronal injury and death in regions of diffuse brain injury: evidence for multiple independent injury phenotypes. J.Neurosci. 24: 3543-3553.
134. Smith DH, Chen XH, Xu BN, McIntosh TK, Gennarelli TA and Meaney DF (1997). Characterization of diffuse axonal pathology and selective hippocampal damage following inertial brain trauma in the pig. J.Neuropathol.Exp.Neurol. 56: 822-834.
135. Stone JR, Singleton RH and Povlishock JT (2001). Intra-axonal neurofilament compaction does not evoke local axonal swelling in all traumatically injured axons. Exp.Neurol. 172: 320-331.
136. Stone JR, Okonkwo DO, Dialo AO, Rubin DG, Mutlu LK, Povlishock JT and Helm GA (2004). Impaired axonal transport and altered axolemmal permeability occur in distinct populations of damaged axons following traumatic brain injury. Exp.Neurol. 190: 59-69.
137. Strich SJ (1956). Diffuse degeneration of the cerebral white matter in severe dementia following head injury. J.Neurol.Neurosurg.Psychiatry 19: 163-185.
138. Strich SJ (1961). Shearing of nerve fibers as a cause of brain damage due to head injury: A pathological study of twenty cases. Lancet 2: 443-448.
139. Suehiro E and Povlishock JT (2001). Exacerbation of traumatically induced axonal injury by rapid posthypothermic rewarming and attenuation of axonal change by cyclosporin A. J.Neurosurg. 94: 493-498.
140. Tanaka T, Annaka M, Ilmain F, Ishi K, Kokufuta E, Suzuki A, Tokita M. (1992). Phase transitions of gels. In: Karalis TK. (ed.). Mechanics of Swelling. Berlin:Springer; NATO ASI Series, Vol. H64;. p. 683-703.
141. Tasaki I (1999). Rapid structural changes in nerve fibers and cells associated with thier excitation processes. Jpn J Physiol 49:125-138.
142. Uchida D, Arimura A, Somogyvari-Vigh A, Shioda S and Banks WA (1996). Prevention of ischemia-induced death of hippocampal neurons by pituitary adenylate cyclase activating polypeptide. Brain Res. 736: 280-286.
143. Vaudry D, Gonzalez BJ, Basille M, Pamantung TF, Fontaine M, Fournier A et al (2000). The neuroprotective effect of pituitary adenylate cyclase activating polypeptide on cerebellar granule cells is mediated through inhibition of the CED3-related cystein protease caspase-3/CPP32. Proc Natl Acad Sci U
57
S A 97:13390 –5.
144. Verdugo P, Deyrup-Olsen I, Martin AW, Luchtel DL (1992).Polymer gel phase transition: the molecular mechanism of product release in mucin secretion. In: Mechanics of Swelling. Karalis TK (ed.). NATO ASI Series, Vol. H64 Berlin: Springer; pp.671-81.
145. Wang KK, Posmantur R, Nadimpalli R, Nath R, Mohan P, Nixon RA, Talanian RV, Keegan M, Herzog L and Allen H (1998). Caspase-mediated fragmentation of calpain inhibitor protein calpastatin during apoptosis. Arch.Biochem.Biophys. 356: 187-196.
146. Wang KK, Posmantur R, Nath R, McGinnis K, Whitton M, Talanian RV, Glantz SB and Morrow JS. (1998). Simultaneous degradation of alphaII- and betaII-spectrin by caspase 3 (CPP32) in apoptotic cells. J.Biol.Chem. 273: 22490-22497.
147. Wang KK (2000). Calpain and caspase: can you tell the difference? Trends Neurosci. 23: 20-26.
148. Wilkinson AE, Bridges LR and Sivaloganathan S (1999). Correlation of survival time with size of axonal swellings in diffuse axonal injury. Acta Neuropathol.(Berl) 98: 197-202.
149. Wolf JA, Stys PK, Lusardi T, Meaney D and Smith DH (2001). Traumatic axonal injury induces calcium influx modulated by tetrodotoxin-sensitive sodium channels. J.Neurosci. 21: 1923-1930.
150. Yaghmai A and Povlishock J (1992). Traumatically induced reactive change as visualized through the use of monoclonal antibodies targeted to neurofilament subunits. J.Neuropathol.Exp.Neurol. 51: 158-176.
151. Zoratti M and Szabo I (1995). The mitochondrial permeability transition. Biochim.Biophys.Acta 1241: 139-176.
58
KÖSZÖNETNYILVÁNÍTÁS
Mindenek előtt, hálás köszönettel tartozom témavezetőimnek, Gallyas Ferenc Professzor
Úrnak és Dr. Büki Andrásnak a PhD éveim alatt nyújtott felbecsülhetetlen segítségükért,
támogatásukért, bizalmukért, bíztatásukért, konstruktív kritikáikért.
Köszönöm volt mentoromnak, Dr. Sándor Jánosnak, hogy elindított a tudomány rögös
útján, és azóta is szemeit rajtam tartva, töretlen bizalommal és bíztatással kíséri végig
tudományos életem alakulását.
Külön köszönet illeti John T. Povlishock Professzor Urat, a Virginia Commonwealth
University Anatómiai és Neurobiológiai Intézete vezetőjét, aki lehetővé tette, hogy 3
éven keresztül laboratóriumában dolgozhassam, és ott tartózkodásom ideje alatt illetve
hazajövetelem óta is támogatásával és bizalmával tisztel meg.
Köszönöm Dóczi Tamás Professzor Úrnak, hogy lehetővé tette, hogy a PTE ÁOK
Idegsebészeti Klinika kutató laboratóriumában folytathassam munkáimat.
Köszönet illeti a PTE ÁOK Idegsebészeti Klinikájának kutató laboratóriumában
dolgozókat, név szerint: Nyirádi Józsefet, Andok Csabánét és Nádor Andrásnét, illetve a
Virginia Commonwealth University Anatómiai és Neurobiológiai Intézetének
laboránsait, Lynn Davist és Sue Walkert az évek alatt nyújtott felbecsülhetetlen technikai
segítségükért.
Köszönöm a PTE ÁOK Anatómiai Intézetéből Dr. Reglődi Dórának és Dr. Tamás
Andreának közös munkáink során nyújtott együttműködését.
Köszönöm a Pécsi Tudományegyetem Idegsebészeti Klinika intenzív osztályán
dolgozóknak a súlyos koponyatraumás betegek liquor- és vér-mintagyűjtésében nyújtott
segítségét. Továbbá köszönöm a PTE ÁOK Mikrobiológiai Intézetében dolgozóknak,
elsősorban Dr. Szekeres Júlia Professzor Asszonynak, Dr. Polgár Beátának és Kiss
Ágnesnek, e minták feldolgozásában való áldozatos közreműködését.
Köszönetet mondok végül, de nem utolsó sorban minden kedves volt és jelenlegi
kollégámnak, családomnak és barátaimnak a sok bíztatásért és lelki támaszért.
Abstract It was earlier established that one of the pri-mary morphopathological consequences of experimentaltraumatic brain injury is a dramatic reduction in the dis-tances between the neurofilaments (cytoskeletal com-paction) inside a number of axon segments that appear tobe randomly distributed among normal axons in an other-wise undamaged parenchymal environment. The presentresults demonstrate that the cytoskeletal compaction in-stantly induces argyrophilia, thereby rendering possibleselective visualisation of the affected axon segments forlight microscopy through use of a special silver stainingmethod. On combination of this method with electron mi-croscopy, it was revealed that the cytoskeletal compactionis completed in much shorter times and extends to muchlonger axon segments than previously assumed.
Keywords Head injury · Axonal pathology · Cytoskeletal compaction · Silver staining · Electron microscopy
Introduction
In selected brain areas of experimental animals that sur-vive for 5 min or longer following fluid percussion, im-pact acceleration or non-disruptive axonal injury to cen-tral myelinated nerve fibres, the following ultrastructuralchanges have recently been discovered in a number ofaxon profiles scattered among the normal-looking axonprofiles: the mitochondria and myelin sheaths becomeswollen, a considerable proportion of the microtubulesand the side-arms of the neurofilaments disappear, periax-
onal water-filled spaces develop, and the distances be-tween the neurofilaments are dramatically reduced [11,12, 18, 19, 20]. In consequence of the latter feature, thismorphological phenomenon has been designated as neu-rofilament/cytoskeletal/axonal compaction/condensation/collapse. It has been hypothesised to precede axonal dis-connection in experimental diffuse traumatic axonal in-jury (TAI) [14].
Using a special silver staining method [6, 8] that yieldsreproducible results [15], we have previously demon-strated [5, 7] that millimetre-long axon segments werestained with silver in a diffuse distribution within certainbrain areas of rats that had been perfusion-fixed immedi-ately after suffering either contusing or non-contusingconcussive head injuries, whereas no axons were stainedwith silver in the control animals. This induced argyro-philia was suggested to be a manifestation of some, at thattime unknown, primary morphological disturbance.
Here we demonstrate that it is the axonal segmentswith a traumatically compacted cytoskeleton that arestained with silver by this method following a head injury.Furthermore, the histological and theoretical importanceof this finding is discussed.
Materials and methods
Animal experiments
Seven 400-g Sprague-Dawley rats were anaesthetised with a mix-ture of 4% isoflurane, 29% oxygen and 67% nitrous oxide using aSurgiVet respirator. The skull was exposed with a middle-line in-cision and a 4-mm-thick 10-mm-diameter brass disc was cementedwith dental acrylic to the exposed skull vault, between the coronalsuture and lambdoid. Thereafter, the animals were placed in aprone position on a 12-cm-thick foam bed under the impact-accel-eration head-injury device of Marmarou et al. [13]. An intracere-bral mechanical force was generated by dropping a metal rod witha mass of 450 g onto the brass disc from a height of 200 cmthrough a vertically positioned Plexiglas tube.
Immediately after administration of the head injury, the ratswere killed by transcardial perfusion with 50 ml physiologicalsaline, followed by 500 ml of an aldehyde fixative (100 ml 20%paraformaldehyde, 100 ml 25% glutaraldehyde, 50 ml 0.1 mol/lcalcium chloride, 500 ml 0.2 mol/l sodium cacodylate and 250 ml
F. Gallyas · O. Farkas · M. Mázló
Traumatic compaction of the axonal cytoskeleton induces argyrophilia:histological and theoretical importance
Acta Neuropathol (2002) 103 :36–42DOI 10.1007/s004010100424
Received: 27 November 2000 / Revised: 26 February 2001 / Accepted: 25 May 2001 / Published online: 4 September 2001
REGULAR PAPER
F. Gallyas (✉ ) · O. FarkasDepartment of Neurosurgery, Faculty of Medicine, Pécs University, Rét utca 2, 7623 Pécs, Hungarye-mail: [email protected], Fax: +36-72-535931
M. MázlóCentral Electron Microscopic Laboratory, Faculty of Medicine, Pécs University, Szigeti ùt 12, 7623 Pécs, Hungary
© Springer-Verlag 2001
distilled water were mixed, and the resulting solution was adjustedto pH 7.4 with a few drops of concentrated hydrochloric acid).Three additional animals, sham-operated and then perfusion-fixedwithout administration of the head injury, served as controls. Ani-mal care and handling were carried out in accordance with order243/1998 of the Hungarian Government, which is an adaptation ofdirective 86/609/EGK of the European Committee Council.
Tissue processing
All animals were left untouched at room temperature for 24 h afterthe start of fixation [3]. Thereafter the brains, together with a 5-mm-long stub of the spinal cord, were removed from the skulland then immersed in the fixative. Vibratome sections, 150 µmthick, were cut coronally from the caudal two-thirds of the cere-brum, and sagittally from the block including the cerebellum,pons, oblongata and spinal cord. Every fifth vibratome section wassilver-stained. For electron microscopy, those areas of the subcor-tical white matter, caudate putamen and pons which contained nu-merous argyrophilic axon segments in the silver-stained sectionswere dissected from the nearby vibratome sections of head-injuredanimals. The corresponding areas were also cut from the vibra-tome sections of the control animals. These pieces of vibratomesections were post-fixed with a 1:1 mixture of 2% osmium tetrox-ide and 3% potassium ferrocyanide for 1 h at room temperature,and then flat-embedded in Durcupan ACM. Semithin sectionswere cut at 0.5 µm and air-dried onto microscopic slides previ-ously coated with the Vectabond adhesive. Thin sections (50 nm)were contrasted with uranyl acetate and lead citrate.
Silver staining
Vibratome sections were dehydrated with graded 1-propanol(25%, 50%, 75% and 98%; 5 min in each), incubated for 16 h at56°C in 1-propanol containing 0.8% sulphuric acid and 2% water(esterification), rehydrated with graded 1-propanol (75%, 50%,25%) and distilled water (5 min in each), treated with 1% aceticacid for 10 min, and then immersed in a special physical developeruntil the background had become light-brown. The physical devel-oper was prepared just before use by pipetting one part of stock so-lution B into one part of vigorously stirred stock solution A. Stocksolution A consisted of 100 g anhydrous sodium carbonate dis-solved in 1,000 ml of distilled water; stock solution B contained2.5 g ammonium nitrate, 2.0 g silver nitrate, 100 ml 20% tungstosili-cic acid, 100 ml 20% paraformaldehyde and 890 ml distilled water.Physical development was terminated by washing with 1% aceticacid for 30 min. Sections were dehydrated, cleared and coveredwith Canada balsam (for trouble shooting see [8]). Several silver-stained sections were processed for electron microscopy.
Durcupan-embedded semithin sections were silver-stained withthe following modifications: (i) before esterification the sectionswere dehydrated by air-drying, and (ii) the treatment with aceticacid between the esterification and the physical development wasreplaced with a periodic acid treatment (1%, 10 min), followed bya short washing with distilled water (see discussion). Several sil-ver-stained semithin sections were re-blocked, thin-sectioned andcontrasted with uranyl acetate and lead citrate in the usual manner.Selected light microscopic areas of silver-stained semithin sectionswere compared with the electron microscopic pictures of the cor-responding areas in nearby thin sections.
Results
Control animals
Neither axons nor neurones were visualised by the silvermethod in the central nervous system of the control ani-mals. However, hypertrophic astrocytic fibres were stained
with silver, mainly in the white matter of the spinal cord(Fig.1a). The light microscopic distinction of solitary as-trocytic fibres from argyrophilic axons in the silver-stained sections of the head-injured animals can be diffi-cult even for experienced observers.
No axon profiles with cytoskeletal compaction werefound under the electron microscope in the brain areas ofthe control animals that corresponded to the areas con-taining numerous silver-stained axon segments and com-pacted axon profiles in the head-injured animals. In thewhite matter of the spinal cord, hyperplastic bundles ofglial filaments were found in several astrocytic processes(Fig.1b).
37
Fig.1A,B Fibrous astrocytes in the pyramidal tract of the spinalcord of a control rat. A Silver-stained hypertrophic processes in a150-µm-thick vibratome section; B hyperplastic glial filaments.The arrows in B point to dense bundles of glial filaments. Notethat solitary long silver-stained astrocytic processes (arrow in A)can not be easily distinguished from silver-stained axon segments.A ×225; bar B 500 nm
Head-injured animals
Gross morphological observations
Four of the seven head-injured animals suffered an im-pression fracture of the calvaria, with severe lacerationand bleeding underneath. In all animals, either with orwithout a skull fracture, massive haemorrhage occupiedthe basal cisterns and extended to the subarachnoid spacesaround the cerebral hemispheres, cerebellum and spinalcord.
Light microscopic observations
In all brains, long axon segments stained with silver in ascattered distribution or in dense bundles. The whole lengthof an axon was rarely stained. It was possible to follow in-dividual axons for a maximum length of about 1 mm. Thespatial distribution of silver-stained axon segments showedwide variations from one animal to another. The smallnumber of animals involved in the study did not allow a
statistical description of the distribution patterns of argyr-ophilic axon segments. For this reason, only those areas of the brain are demonstrated where silver-stained axonsegments were consistently found at an appreciable dis-tance from the focal parenchymal lesions, in a numberthat appeared sufficient for electron microscopic observa-tion. Such areas in the animals in which the skull was notfractured were the deep neocortex with the subcorticalwhite matter under the impact site (Fig.2a), the caudateputamen (Fig.2c) and the pons (Fig.2b). The distributionof the silver-stained axon segments in the two cerebralhemispheres was not symmetrical. Their incidence dis-played considerable differences in each of the above areasfrom animal to animal. The microphotos in Fig.2 weretaken from the most seriously affected animal. Pyramidalneurones were not demonstrated anywhere in the neocor-tex, but an insignificant number of interneurones werestained with silver (not shown).
In addition to silver-stained axon segments, the brainsof animals in which the skull was fractured contained nu-merous silver-stained neurones and dendrites, even in ar-eas far from the impact site (not shown). The abundanceof capillaries filled with erythrocytes indicated that thesebrains were imperfectly perfused with the fixative. Forthis reason, they were excluded from the following ultra-structural description, although they contained abundantaxon profiles with compacted cytoskeleton (not shown).
38
Fig.2 Silver-stained axon segments in 150-µm-thick vibratomesections cut from the subcortical white matter, corpus callosumand deep neocortex under the impact site (A), the reticular nucleusof the pons (B) and the caudate putamen (C) of a rat perfusion-fixed immediately after suffering a head injury. The arrow pointsto a silver-stained interneurone, and the arrowhead to a neurone inthe induseum griseum. Note that even millimetre-long axon seg-ments are silver-stained. A, B ×100; C ×210
Electron microscopic observations
In those brain areas of head-injured animals where argyr-ophilic axon segments were present, a number of axonprofiles displaying a dramatic decrease in the distancesbetween the individual neurofilaments (Fig.3) were scat-
tered among normal-looking axon profiles (Fig.4b). Themitochondria appeared to be non-swollen (Figs. 3, 4b).Lamellar separation of the myelin sheath was infrequent.Fluid-filled spaces of various shapes and volumes had de-veloped between the axolemma and the myelin sheath(Fig.4b). In compacted axon profiles with small periax-
39
Fig.3 Normal-looking (A, C)and compacted (B, D) axonprofiles in the reticular nucleusof the pons (A, B) and the sub-cortical white matter (C, D) ofa rat perfusion-fixed immedi-ately after suffering a head in-jury. A small periaxonal fluid-filled space is marked with anasterisk. Microtubuli in A andB are indicated by arrowheads.Note that the distances be-tween individual neurofila-ments are dramatically reducedin B and D. Bars A–D 200 nm
Fig.4 Comparison of the lightmicroscopic image of a se-lected area in a silver-stainedsemithin section (A) with theelectron microscopic image ofthe corresponding area in anearby ultrathin section (B) cutfrom the caudate putamen of arat perfusion-fixed immedi-ately after suffering a head in-jury. The short arrow points toa large fluid-filled periaxonalspace, and the long arrow to alamellar separation of themyelin sheath. Note that thepattern of the silver-stainedaxon profiles corresponds tothat of the compacted ones.Bars A, B 2 µm
onal fluid-filled spaces, the myelin sheath appeared to besomewhat deformed (Fig.4b).
In silver-stained, re-blocked and then contrasted semi-thin sections, numerous silver grains were found in axonprofiles with a compacted cytoskeleton, but there werehardly any in the normal-looking axons (Fig.5). The myelinsheath, axolemma and mitochondria also contained a fewsilver grains, of somewhat smaller sizes. They renderedthe myelin sheath brown in semithin sections, while thecompacted axon segments turned black and the non-com-pacted axons remained unstained.
A comparison of the light microscopic pictures of selected areas in the silver-stained semithin sections (Fig.4a) with the electron micrographs of the correspond-ing areas in nearby ultrathin sections (Fig.4b) showed thatthe pattern of the silver-stained axonal cross-sections cor-responded to that of the compacted axon profiles. Minutedifferences between the two pictures in the distances be-
tween the individual axons and in their shapes were prob-ably caused by the facts that their directions differedslightly from each other and, for technical reasons, not thenext neighbouring sections were compared.
In specimens silver-stained before being processed forelectron microscopy, silver grains were found only in ho-mogeneously electron-dense axon profiles. In contrast,the ultrastructure of axon profiles without silver grainswas disintegrated and electron lucent (Fig.6).
Discussion
Methodological considerations
It is generally believed that silver staining methods areunreliable. However, several recent methods employingphysical developers instead of the chemical developers
40
Fig.5 Axon profiles in thecaudate putamen of a rat perfu-sion-fixed immediately aftersuffering a head injury. Thespecimen was osmicated, Dur-cupan-embedded, and semi-thin-sectioned, silver-stained,re-blocked and thin-sectionedbefore being contrasted withuranyl acetate and lead citrate.Note that the normal-lookingaxon profiles do not containsilver grains. Bar 500 nm
Fig.6 Axon profiles in thecaudate putamen (A) and thesubcortical white matter (B) ofa rat perfusion-fixed immedi-ately after suffering a head in-jury. The specimen was silver-stained before being processedfor electron microscopy. Notethat the silver-containing axonprofiles are homogeneouslyelectron dense, while thosewithout silver grains have dis-integrated. Bars A, B 200 nm
applied in the traditional methods, do give reproducibleresults [15]. The silver technique used in the presentstudy, which employs a time-honoured physical developer[4], has been widely used to obtain information on neu-ronal death.
An important step in this silver staining method, ester-ification with strongly acidic n-propanol at 56°C for 20 h,considerably impairs the ultrastructure in non-osmicatedvibratome sections (Fig.6). Nevertheless, traumaticallycompacted axon profiles can be identified under the elec-tron microscope through their marking with silver grains.It appears that they are more resistant than the normal ax-ons to the structure-disintegrating action of esterification(Fig.6).
In a separate study (Gallyas et al., submitted for publi-cation), osmophilic structural elements were found to bemore argyrophilic than compacted neurones, dendrites oraxons. As a consequence, the microscopic images ofsemithin sections silver-stained according to the originalprocedure were dominated by silver-stained membranesand membranous structures such as myelin sheaths andmitochondria. Of the chemical agents proposed for the re-moval of bound osmium from Durcupan-embedded sec-tions [10], periodic acid proved to be the most effectivefor suppression of the co-staining of myelin sheaths andmitochondria.
Histological considerations
Compacted axon segments have so far been selectivelystained for light microscopic observation only in animalssurviving for at least 5 min [horseradish peroxidase(HRP) uptake] or for 15 min (Ab38 or RMO14 antibodypositivity) following a head injury. If injected in due timeinto the cerebrospinal fluid, HRP inundates the extracellu-lar space; it remains excluded from the normal axons, butit penetrates into those which display cytoskeletal com-paction [18, 19, 20]. Ab38 and RMO14 antibodies are ef-fective because calcium-mediated processes linked to theproteolysis of spectrin and to alterations in the neurofila-ment-M subunit, respectively, take place in the compactedaxon segments [1, 17, 20].
Our present findings clearly demonstrate that the silvermethod used here can also be utilised for demonstration ofaxon segments with a compacted cytoskeleton by lightmicroscopy. This applies both to animals perfusion-fixedimmediately following a head injury (Fig.2) and to thosesurviving for several minutes or hours thereafter (notdemonstrated). By virtue of the high selectivity and lowbackground staining of this method, compacted axons caneasily be located even in 150-µm-thick vibratome sections(Fig.2). Similar findings in the brains of mice, rats andpigs traumatised in three different ways, and also in surgi-cally removed human hippocampus [21], lead us to as-sume that this silver staining technique could be instru-mental (i) in mapping the distribution of compacted axonsegments throughout the brains of mammals, and (ii) inlocating areas in mammalian brains that could be investi-
gated in nearby vibratome sections by electron micro-scopy, immunocytochemistry or other techniques.
The occurrence of hypertrophic fibrous astrocytes,which may hamper the identification of silver-stainedaxon segments, is not necessarily associated with anyneurological disease. In the brains of control (unlesioned)rats weighing about 200 g, a few hypertrophic astrocyticfibres occur in subependymal and subpial positions (notdemonstrated). Their number increases with age, espe-cially in the spinal cord (Fig.1a). Non-hypertrophic fi-brous astrocytes are not stained with the silver methodused here (Fig.2).
Neuronal somata and dendrites with a traumaticallycompacted cytoskeleton (“dark” neurones) are also stainedby this silver method [5, 7]. In specific cases, it is not easyto distinguish solitary compacted dendrites from solitarycompacted axons.
Theoretical considerations
The prevailing concept of the traumatic compaction of theaxonal cytoskeleton [1, 2, 9, 11, 12, 14, 16, 17, 18, 19, 20]is based on the following experimental findings. In certainbrain areas of animals perfusion-fixed at or after 5 minfollowing a head injury, (i) HRP injected into the cere-brospinal fluid, (ii) calpain-mediated spectrin proteolysis,(iii) some kind of alteration in the neurofilament-M sub-unit, (iv) reduction of the extension of neurofilament side-arms, (v) the disappearance of neurotubuli and neurofila-ment side-arms, and (vi) the swelling of mitochondria aredemonstrable in the axon segments with a compacted cytoskeleton, but not inside the normal axons. Accord-ingly, some intracerebral force generated by the head in-jury is assumed to stretch a number of axons to such anextent that a focal axolemmal perturbation allowing un-controlled influx of ions including calcium occurs. Thelatter activates the proteolytic enzyme calpain. This medi-ates spectrin proteolysis and neurofilament-M subunit al-teration, which gradually disturb the side-arms of the neu-rofilaments, resulting in the compaction of the axonal cyto-skeleton.
The present study throws light on two previously un-considered features of the traumatic compaction of the ax-onal cytoskeleton:
1. To date, no data have been reported on either the pres-ence or the absence of compacted axon profiles in ani-mals perfusion-fixed immediately after a head injury.The reason for this is probably the lack of any light mi-croscopic technique that could indicate where to lookfor them under the electron microscope. Our presentdata clearly demonstrate that the cytoskeletal com-paction is already completed in numerous axons of an-imals surviving for less than 1 min after the head in-jury. Consequently, the pathometabolic cascade result-ing in the cytoskeletal compaction must proceed at amuch higher speed than previously assumed.
2. As demonstrated by the silver staining method usedhere, the cytoskeletal compaction extends to much
41
42
greater distances in the affected axons (Fig.2a, b) thanpreviously estimated on the basis of pictures obtainedusing HRP histochemistry, Ab38 and RMO14 im-munocytochemistry or electron microscopy. Thus, ei-ther the stretch-induced perturbation of the axolemmamust extend to long axon segments (i.e. it is not focal)or, in case of focal axolemmal perturbation, the path-ometabolic cascade resulting in the cytoskeletal com-paction must spread over long distances during a rela-tively short period of time.
Acknowledgements The authors express thanks to Andok Csabáné,Nyírádi József, Nádor Andrásné and Nagy Attila for their valuablehelp in the preparative work. This study was supported by Hun-garian grants from OTKA (T 029352) and from FKFP (493/1999).
References
1.Buki A, Siman R, Trojanowsky JQ, Povlishock JT (1999) Therole of calpain mediated spectrin proteolysis in traumaticallyinduced axonal injury. J Neuropathol Exp Neurol 58:365–375
2.Buki A, Onkowo DO, Wang KK, Povlishock JT (2000) Cyto-chrome c release and caspase activation in traumatic axonal in-jury. J Neurosci 20:2825–2834
3.Cammermeyer J (1961) The importance of avoiding “dark”neurons in experimental neuropthology. Acta Neuropathol (Berl)1:245–270
4.Gallyas F (1971) A principle for silver staining of tissue ele-ments. Acta Morphol Acad Sci Hung 19:57–71
5.Gallyas F, Zoltay G (1992) An immediate light microscopic re-sponse of neuronal somata, dendrites and axons to non-contus-ing concussive head injury. Acta Neuropathol 83:386–393
6.Gallyas F, Güldner FH, Zoltay G, Wolff JR (1990) Golgi-likedemonstration of “dark” neurons with an argyrophil III methodfor experimental neuropathology. Acta Neuropathol 79:620–628
7.Gallyas F, Zoltay G, Balás I (1992) An immediate light micro-scopic response of neuronal somata, dendrites and axons tocontusing concussive head injury. Acta Neuropathol 83:394–401
8.Gallyas F, Hsu M, Buzsáki G (1993) Four modified silvermethods for thick sections of formaldehyde-fixed mammaliancentral nervous tissue: “dark” neurons, perikarya of all neu-rons, microglial cells and capillaries. J Neurosci Methods 50:159–164
9.Graham DI, McIntosh TK, Maxwell WL, Nicoll JAR (2000)Recent advances in neurotrauma. J Neuropathol Exp Neurol59:641–651
10.Hayat MA (1981) Fixation for electron microscopy. AcademicPress, New York, p. 176
11. Jafari SS, Maxwell WL, Neilson M, Graham DI (1997) Axonalcytoskeletal changes after nondisruptive axonal injury. J Neu-rocytol 26:207–221
12. Jafari SS, Neilson, M, Graham DI, Maxwell WL (1998) Ax-onal cytoskeletal changes after nondisruptive axonal injury. II.Intermediate sized axons. J Neurotrauma 15:955–966
13.Marmarou A, Foda MA, Brink W van den, Campbell J, Kita H,Demetriadou K (1994) A new model of diffuse brain injury inrats. Part I. Pathophysiology and biomechanics. J Neurosurg80:291–300
14.Maxwell WL, Povlishock JT, Graham DL (1997) A mechanis-tic analysis of nondisruptive axonal injury: a review. J Neuro-trauma 14:419–440
15.Newman GR, Jasani A B (1998) Silver development in mi-croscopy and bioanalysis: past and present. J Pathol 186:119–125
16.Onkowo DO, Pettus EH, Moroi J, Povlishock JT (1998) Alter-ation of the neurofilament sidearm and its relation to neurofila-ment compaction. Brain Res 784:1–6
17.Onkowo DO, Büki A, Siman R, Povlishock JT (1999) Cy-closporin A limits calcium-induced axonal damage followingtraumatic brain injury. NeuroReport 10:353–358
18.Pettus EH, Povlishock JT (1996) Characterization of a distinctset of intra-axonal structural changes associated with traumati-cally induced alteration in axolemma permeability. Brain Res772:1–11
19.Pettus EH, Christman CW, Giebel ML, Povlishock JT (1994)Traumatically induced altered membrane permeability: its rela-tionship to traumatically induced reactive axonal change. J Neurotrauma 11:507–552
20.Povlishock JT, Marmarou A, McIntosh T, Trojanowsky JQ,Moroi J (1997) Impact acceleration injury in the rat: evidencefor focal axolemmal change and related neurofilament sidearmalteration. J Neuropathol Exp Neurol 56:347–359
21.Van den Pol AN, Gallyas F (1990) Trauma-induced Golgi-likestaining of neurons: a new approach to neuronal organizationand response to injury. J Comp Neurol 296:654–673
Journal of Neuroscience Methods 153 (2006) 283–289
Selective induction of ultrastructural (neurofilament) compactionin axons by means of a new head-injury apparatus
Jozsef Pal a,1, Zsolt Toth b,2, Orsolya Farkas b,3, Lorand Kellenyi b,Tamas Doczi a,b, Ferenc Gallyas b,∗
a Clinical Neuroscience Research Group of the Hungarian Academy of Sciences, Pecs University,Ret utca 2, H-7623 Pecs, Hungary
b Department of Neurosurgery, Pecs University, Ret utca 2, H-7623 Pecs, Hungary
Received 1 August 2005; received in revised form 10 November 2005; accepted 10 November 2005
Abstract
A new weight-drop head-injury apparatus is described that can produce a momentary depression of predetermined depth at a predeterminedsite of the elastic calvaria of scalped young adult rats. In Wistar rats weighing about 200 g, a 0.75-mm deep calvaria depression immediatelycaused ultrastructural (neurofilament) compaction in many long axon segments, which were diffusely scattered among non-compacted axons in awtnifa©
K
1
bCosni
((f
P
C
0d
ell-defined area of cortical layers IV and V under the impact site. Apart from these morphological changes and swollen astrocytic processes inheir vicinity, the brain tissue appeared non-impaired. The blood pressure, intracranial pressure, heart rate and respiration rate had returned to theormal range in 1 min. Diffuse axonal swelling caused by impaired axonal transport, ultrastructural compaction in neuronal soma-dendrite domains,mpression fracture and subarachnoid or subdural hemorrhages were observed only in rats with a calvaria depression of 1 mm or more. All theseeatures create favorable circumstances for study of various problems that are closely related to the ultrastructural (neurofilament) compaction inxons, such as the fate of the affected axons.
2005 Elsevier B.V. All rights reserved.
eywords: Head injury; Calvaria depression; Axonal injury; Staining with silver; Electron microscopy; Ultrastructural compaction
. Introduction
The morphopathological responses of the human brain tolunt head injuries are rather complex (Graham et al., 2000).ertain responses set in rapidly (e.g. focal contusions or hem-rrhages) whereas others are initiated primarily, but their finaltages develop slowly (e.g. “diffuse axonal injury” or apoptoticeuronal death). Additionally, traumatically induced pathophys-ological processes (e.g. ischemia or elevated intracranial pres-
∗ Corresponding author. Tel.: +36 72 535900; fax: +36 72 535931.E-mail addresses: [email protected] (J. Pal), [email protected]
Z. Toth), [email protected] (O. Farkas), [email protected]. Kellenyi), [email protected] (T. Doczi),[email protected] (F. Gallyas).
1 Tel.: +36 72 535930; fax: +36 72 535931.2 Present address: Department of Heart Surgery, University of Pecs, H-7634ecs, Ifjusag ut 13, Hungary. Tel.: +36 72 536390; fax.: +36 72 536399.3 Present address: Department of Anatomy, Medical College of Virginia,ampus of Virginia Commonwealth University, Richmond, VA, USA.
sure) may secondarily cause their own morphological responses,or may influence the outcome of those elicited primarily.
The term “diffuse axonal injury” (DAI) refers to the phe-nomenon of fatal damage to axons that are scattered amongapparently normal axons in a widespread distribution in thehuman brain. DAI is a primary cause of death and neurologi-cal disability following head injuries (Adams et al., 1983). Inhuman neuropathology, the pathomechanism of DAI cannot beelucidated mainly because of the complexity of traumaticallyinduced morphopathological damages.
Although the traumatic axonal injury observed in animalmodels (TAI) does not reproduce exactly the nature, extentand time course of axonal injury in DAI, the investigation ofTAI provides good insight into the axonal changes occurringin DAI (Maxwell et al., 1997). By means of animal mod-els, two distinct mechanisms of diffuse axonal damage wereobserved: focal impairment of the axonal transport and neu-rofilament (ultrastructural) compaction (Stone et al., 2001). Inhuman neuropathology, focal axonal swellings are indicativeof the impairment of axonal transport even for years post-
165-0270/$ – see front matter © 2005 Elsevier B.V. All rights reserved.oi:10.1016/j.jneumeth.2005.11.004
284 J. Pal et al. / Journal of Neuroscience Methods 153 (2006) 283–289
injury. Regarding the neurofilament (ultrastructural) compactionin axons, no persistent morphological change is known, whichcould facilitate the investigation of its participation in humancases. Since this kind of axonal damage was discovered only 10years ago in animal experiments (Pettus et al., 1994), its mor-phological, neurological and behavioural consequences have notbeen cleared even in experimental neurotraumatology, proba-bly because of the lack of a method capable of its selectiveproduction.
The widely used experimental head-injury devises producevarious sets of morphopathological responses such as thosementioned above (Gennarelli, 1994), thereby allowing the inves-tigation of multiple aspects of traumatic brain injury in thecomplexity that can occur in various conditions in human neuro-traumatology (Ellingson et al., 2005). However, another experi-mental approach, the investigation of a single morphopathologi-cal response without the influence of other responses, also holdsout the promise of the acquisition of new information on thisfield.
This paper describes a new weight-drop head-injury appara-tus that can induce neurofilament (ultrastructural) compaction inlong axon segments in the rat brain, but without inflicting otherkinds of immediate or delayed morphological damage to axonsand without pathophysiological processes ensuing in the post-injury period. Thus, various problems that are closely related tothe neurofilament (ultrastructural) compaction in axons can beie
twapa
Fig. 2. (a) The head-injury apparatus in action. (b) The machinery allowing theprecise positioning and fixation of the rat’s head under the impactor.
2. Materials and methods
2.1. The new rat-head-injury apparatus and its use
An earlier described weight-drop head-injury devicedesigned to produce a depression of controlled depth on theexposed brain surface (Feeney et al., 1981) was modified sub-stantially for the closed rat skull as follows (Figs. 1 and 2). Aflanged steel cylinder 65 mm in length, 45 mm in outer diameterand 25 mm in inner diameter (impact-depth adjuster housing) isfirmly fixed with a nut (fixing nut) into the bridge of a massiveiron frame, which is welded to a heavy three-legged baseboard.Round the flange, a rotatable hollow disc (scale disc) is placed,which is marked every 36◦. A second steel cylinder 95 mm inlength, 25 mm in outer diameter and 11 mm in inner diame-ter (impact-depth adjuster), the mantel of which is threaded atboth ends (10 threads/cm), is fitted into its housing. The ver-
atus.
nvestigated by means of this apparatus without outside interfer-nce.
Since microtubules, endoplasmic reticulum cisternae andhe axolemma are also involved in neurofilament compactione prefer the term “ultrastructural compaction”. As a further
rgument, head injuries can cause the same ultrastructural com-action in the neuronal soma-dendrite domain, where neurofil-ments are sparse (Csordas et al., 2003; Gallyas et al., 2004).
Fig. 1. Schematic drawing of the calvaria-depression head-injury appar
(a–c) Characteristic positions of the impactor relative to the rat’s head.
J. Pal et al. / Journal of Neuroscience Methods 153 (2006) 283–289 285
tical motion of the impact-depth adjuster is controlled with anut (adjusting nut) screwed on its upper end. From the man-tel of the impact-depth adjuster, a pin protrudes radially intoa vertical slot milled in its housing, which allows it to moveup or down, but not to rotate. The bottom end of a verticallypositioned iron tube 1800 mm in length and 15 mm in innerdiameter (guide tube) is soldered axially into the upper endof the impact-depth adjuster. Through the mantel of the guidetube, several vent holes are bored. An iron cylinder weighingabout 5 kg (ballast weight) containing an axial hole 19 mm indiameter is fixed concentrically to the guide tube. This cylinderpresses the impact-depth adjuster down, to prevent its possibleaxial play. A steel nose-piece containing an axial hole 5 mm indiameter is firmly screwed onto the bottom end of the impact-depth adjuster. The resulting sleeve (stopping sleeve) stops thedownward movement of an 80-mm long, 11-mm thick steel pis-ton (impactor). From its bottom (impactor shoulder), a 20-mmlong, 4.0-mm thick steel rod with a lentiform bottom (impactorhead) protrudes axially through the nose-piece. A bearing ballis brazed onto the upper end of the piston. Mechanical force isgenerated by dropping a 150-mm long, 13-mm thick steel barweighing 200 g through the guide tube onto the impactor, whichweighs 40 g. A head holder is fixed onto a dismounted micro-scope stage that is movable in two perpendicular directions bymeans of micrometer screws. The microscope stage is fastenedto the baseboard under the impactor head at an adjustable angle,iv
stfatTdthadsbpddsudwa(h
2
w
2 ml/kg of a 1:1 mixture of 25 mg/ml Thiopental (BiochemieGmbH, Austria) and 5 mg/ml Seduxen (Richter Gedeon Rt,Budapest). Nine groups of rats were studied, with four rats ineach. The group 1 rats were not subjected to head injury, but weresham-operated and then perfusion-fixed transcardially (controlrats). All other rats were subjected to the calvaria-depressionhead injury. The depression depth was set to 0.5 mm for the ratsin group 2, 0.75 mm for those in groups 3–6, and 1.0 mm forthose in groups 7–9. Thirty minutes before the administrationof the head injury, the femoral artery in each rat in group 6 wasexposed, channeled with polyethylene tubing 0.96 mm in outerdiameter and 150 mm in length (Becton, Dickinson, Cat. no.427411) and connected to the pressure-measuring transducer (B.Braun, Melsungen, Germany) of a custom-made device for mon-itoring blood pressure (BP-Monitor, designed by L. Kellenyi).Thereafter, their cisterna magna was channeled with polyethy-lene tubing of the above type, sealed hermetically with dentalacrylate and connected to the pressure-measuring transducer(B. Braun, Melsungen, Germany) of a custom-made device formonitoring intracranial pressure (ICP-Monitor, designed by L.Kellenyi). This device calculates the pulse rate and the respira-tion rate at intervals of 1 min by means of fast Fourier transfor-mation of the oscillating waveform (Fig. 3a) of the monitoredintracranial pressure. Before head injury, the head holder wastilted through 20◦ relative to the baseboard, and the axis of theimpactor head was centered over the left hemisphere of each rat3F38iwb
Fp(p
n order to permit the setting of a selected area on the rat’s cal-aria into the horizontal plane.
The head injury is administered in the following sequence ofteps: (1) the scalp of an anesthetized rat’s head is incised longi-udinally and stretched apart, and the soft tissue is then removedrom the calvaria. (2) The rat’s head is fixed in the head holder,nd a selected area on the calvaria is then set in the horizon-al plane and positioned under the impactor head (Fig. 1a). (3)he impact-depth adjuster, together with the impactor and theropping weight on its top, is moved downwards by rotatinghe impact-depth adjusting nut until the face of the impactoread just reaches the exposed calvaria (Fig. 1b). (4) There-fter, the adjusting nut is rotated again as far as the intendedepth of the calvaria depression is adjusted, as measured on thecale disc (the angular displacement between any two neigh-oring marks means a depression depth of 0.1 mm). During thisrocess, the rat’s calvaria does not allow the impactor to moveownwards. As a consequence, a distance equal to the intendedepression depth results between the impactor shoulder and thetopping sleeve (Fig. 1c). (5) The dropping weight is pulledp to the selected height by means of a light string and thanropped onto the impactor. This latter suddenly moves down-ards as far as allowed by the stopping sleeve, thereby producingmomentary depression of the intended depth in the calvaria.
6) Immediately thereafter, the rat is removed from the headolder.
.2. Animal experiments
A total of 36 Wistar rats weighing between 190 and 210 gere anesthetized by the intraperitoneal administration of
.5 mm caudal to the bregma and 2.5 mm lateral to the midline.ollowing the infliction of the head injury, the rats in groups 2,and 7 were perfusion-fixed immediately, those in groups 4 and, 1 h later, and those in groups 5, 6 and 9, 4 h later. In each ratn group 6, anesthesia was maintained until the start of fixation,hile the rats in groups 4, 5, 8 and 9 were allowed to recover,ut re-anesthetized before fixation.
ig. 3. Respiration-caused and heart beat-caused oscillations of the intracranialressure before and after the production of a 0.75-mm deep calvaria depression.a and b) Intracranial pressure, (c) heart rate, (d) respiration rate and (e) bloodressure. Arrows point to the time of the head injury.
286 J. Pal et al. / Journal of Neuroscience Methods 153 (2006) 283–289
From each group, two rats were perfused transcardially with500 ml of 4% paraformaldehyde in PBS buffer of pH 7.4 forimmunohistochemistry, while the remaining two rats were per-fused with an electron-microscopic fixative. This was preparedby mixing 250 ml of 0.2 mol/l sodium cacodylate, 50 ml of20% paraformaldehyde, 50 ml of 25% glutaraldehyde, 25 ml of0.1 mol/l calcium chloride and 125 ml of 10% polyvinylpyrroli-done K25, followed by adjustment of the mixture to pH 7.5with hydrochloric acid. Before fixation, the vascular system wasrinsed with physiological saline for 30 s. After fixation, the ratswere left untouched at room temperature for 24 h before removalof the brain from the skull (Cammermeyer, 1961). Animal careand handling were carried out in strict accordance with directive86/609/EGK of the European Committee Council.
2.3. Tissue processing
Brains were cut into three blocks close to the coronal planes1 and 7 mm caudal to the bregma. The middle blocks wereserial-sectioned on a vibratome in the coronal plane, while thecaudal blocks, which included a part of the brain stem, thecerebellum, the pons, the oblongata and the spinal cord wereserial-sectioned in the sagittal plane. From the glutaraldehyde-fixed blocks, sections were cut on a vibratome at 150 �m. Theformaldehyde-fixed blocks were embedded in 5% agarose andcut at 30 �m. For each block, every sixth of the serial sectionwa1a(wii2cc3atu
2
2
1gawfwabwwC
2.4.2. APP immunohistochemistryAfter the suppression of endogenous peroxidase activity by a
30-min treatment with 0.5% hydrogen peroxide in PBS followedby washing with PBS for three 10-min periods, the sections wereexposed to controlled-temperature microwave retrieval (Stone etal., 1999) and then incubated for 20 min in 0.2% Triton X-100(Sigma, Cat. no. T-8532) dissolved in PBS. Next, the sectionswere rinsed with PBS, immersed for 40 min in 1% normal horseserum (NHS, Vectastain Universal Elite ABC Kit, Vector PK-6200) diluted with PBS (henceforth 1% NHS/PBS), incubatedovernight in rabbit anti-�-APP (Zymed Laboratories Inc., Cat.no. 51–2700) diluted with 1% NHS/PBS at 1:250, and thenwashed with 1% NHS/PBS three times for 10 min each. There-after, the sections were subjected to the staining protocol ofthe Vectastain Universal Elite ABC Kit (Vector, Cat. no. PK-6200). Finally, the product of the immunological reaction wasvisualized with a diaminobenzidine (DAB) peroxidase substratekit (Vector, Cat. no. SK-4100) and then washed with PBS for10 min.
2.4.3. RMO14 immunohistochemistryThe procedure was similar to that detailed above for �-
APP, except that the primary antiserum was mouse monoclonalRMO14 antibody (Zymed Laboratories Inc., Cat. no. 34–1000)diluted with 1% NHS/PBS at 1:500.
3
3
ts
3o
vIs(dT(Niwv(p1t
wwm
as placed in the same collecting well. Sections from wells 1nd 4 were stained with a special silver method (Gallyas et al.,993) that selectively demonstrates compacted axons (Gallyas etl., 2002) and also compacted neuronal soma-dendrite domainsCsordas et al., 2003). All formaldehyde-fixed sections fromells 2 and 5 were processed for RMO14 immunohistochem-
stry while those from wells 3 and 6 for APP immunohistochem-stry. From the glutaraldehyde-fixed sections collected in wells, 3, 5 and 6, 2 mm × 2 mm areas corresponding to those whichontained many silver-stained axons in adjacent sections, wereut, postfixed with a 1:1 mixture of 2% osmium tetroxide and% potassium hexocyanoferrate(II) for 1 h at room temperature,nd then flat-embedded in Durcupan ACM. Thin (50 nm) sec-ions were contrasted with uranyl acetate and lead citrate in thesual manner.
.4. Staining procedures
.4.1. Silver stainingThe protocol described in a previous paper (Gallyas et al.,
993) was strictly observed. Briefly, following dehydration withraded 1-propanol, vibratome sections were incubated for 16 ht 56 ◦C in 1-propanol containing 0.8% sulfuric acid and 2%ater for glutaraldehyde-fixed sections, or 1.0% sulfuric acid
or formaldehyde-fixed sections (esterification), then rehydratedith graded 1-propanol, treated with 1% acetic acid for 10 min
nd finally immersed in a special physical developer until theackground had become light-brown. Thereafter, the sectionsere washed with three changes of 1% acetic acid, dehydratedith graded 1-propanol, cleared with clove oil and covered withanada balsam.
. Results
.1. Control rats
In the brains of the sham-operated rats, and also in the non-raumatized hemispheres of the head-injured rats, anomaloustaining properties or morphological features were not found.
.2. Head-injured rats, gross clinical and pathologicalbservations
None of the head-injured rats developed apnea or con-ulsions either immediately after the head injury or later.n the rats that suffered a 0.75-mm deep calvaria depres-ion, the intracranial pressure displayed a sudden increase14.3 ± 2.1 mmHg), which was immediately followed by a sud-en decrease (2.3 ± 0.4 mmHg) and then a gradually fading.he intracranial pressure returned to the normal range in 1 min
Fig. 3a). No other change was monitored thereafter (Fig. 3b).either the pulse rate nor the respiration rate could be measured
n the first minute post-insult due to the absence of an oscillatingaveform of the intracranial pressure. Otherwise, the post-insultalues were similar to those which were measured pre-insultFig. 3c and d). Immediately after the head injury, the bloodressure decreased slightly then returned to the normal range inmin (Fig. 3e). No other pathological event was monitored in
he 4-h survival period.Impression fracture of the skull was observed only in one rat
ith a 1-mm depression depth. Except for this rat, hemorrhagesere not present in the basal cisternae, the ventricles, the cisternaagna or over the surface of the affected hemisphere. In the rats
J. Pal et al. / Journal of Neuroscience Methods 153 (2006) 283–289 287
with a 1-mm deep calvaria depression, but without fracturedcalvaria, small subarachnoid hemorrhages were observed justunder the impact site.
3.3. Head-injured rats, immediate sacrifice, silver staining
In the rats subjected to a 0.5-mm deep calvaria depression, afew long (even about 1 mm) axon segments were evenly silver-stained in the cortical layers IV and V under the impact site(Fig. 4b), but nowhere else.
In each rat that received a 0.75-mm deep calvaria depres-sion, many more silver-stained axon segments were present inthe above location (Fig. 4c). These were diffusely distributedin a crescent-shaped area in the coronal plane under the cen-ter of the impact site (Fig. 4a). At increasing distances fromthis plane, the density of the silver-stained axons in the neo-cortex gradually decreased, reaching zero at about 1.5 mm ineither direction. Hardly any neuronal soma-dendrite domains
Fdscawad
were silver-stained either under the impact site or anywhere else.No intra- or extracellular edema, or any other morphologicaldamage was observed at all.
When the calvaria depressions were 1 mm deep there werehigher numbers of affected axons, predominantly in cortical lay-ers V and VI and the subcortical white matter (Fig. 4d and e)under the impact site. A few neuronal soma-dendrite domainswere also demonstrated, mainly just above the subcortical whitematter, but occasionally in cortical layers II–V, and also inthe hilus and the granule cell layers of the hippocampal den-tate gyrus (not demonstrated). Neither silver-stained axons norsilver-stained soma-dendrite domains were present in other brainareas.
3.4. Head-injured rats, 1 or 4-h survival, silver staining
In the rats with 1.0-mm deep calvaria depressions, evenlysilver-stained axons were scarce. However, many axons were
ig. 4. Light-microscopic (a–h) and electron-microscopic (i) damages in rats sacrificeeep (b), a 0.75-mm deep (a, c and i), a 1.0-mm deep (d–h) momentary calvaria detained with silver (a–e and g), anti-APP (f) or RMO14 antibody (h). Because of oveannot be seen in (a and e), and can be hardly discerned in (d and g). (a and e) Coronas that framed in (a). (f–h) Cortical areas in the same position as that framed in (e).ith dashed lines. (i) Cross-sectioned compacted (C) and normal (N) myelinated ax
xon segment in (b), to a subarachnoidal bleeding in (e) and an APP-positive axonalomains in d and to an axon traced out with silver-stained dots in (g). Scale bars: (a a
d immediately (a–e and i) or at 4 h (f–h) following the production of a 0.5-mmpression. Thirty micrometer (f–h) or 150-�m (a–e) vibratome sections were
rcrowding, individual silver-stained axons or neuronal soma-dendrite domainsl sections, 3 mm caudal to the bregma. (b–d) Cortical areas in the same positionThe border between the neocortex and the subcortical white matter is markedons in normal neuropil. Black arrowheads point to a long, curving compactedswelling in (f). White arrowheads point to compacted neuronal soma-dendritend e) 1 mm; (b–d and f–g) 50 �m; (i) 500 nm.
288 J. Pal et al. / Journal of Neuroscience Methods 153 (2006) 283–289
traced out with silver-stained dots (Fig. 4g; see also Fig. 4c inGallyas and Zoltay, 1992, and Fig. 2d in Csordas et al., 2003).With depressions that were less deep, there were fewer suchaxons.
3.5. Head-injured rats, immunohistochemical staining
Anti-APP stained several axonal swellings in the subcorticalwhite matter under the impact site, but only a few in the neocortexof the rats that survived for 1 or 4 h following the production ofcalvaria depressions of 1 mm in depth (Fig. 4f). However, hardlyany anti-APP-positive axon swellings were found in the ratswith depressions that were less deep. RMO14 antibody stainedno axons in the affected cortical area or the subcortical whitematter of any of the head-injured rats, but cell bodies of normalneurons were faintly stained (Fig. 4h).
3.6. Head-injured rats, electron microscopy
In the affected cortical area of each rat sacrificed eitherimmediately or 1–4 h after the production of 0.50 or 0.75-mm deep calvaria depressions, numerous myelinated (but nonon-myelinated) axons (Fig. 4i) displayed a marked reductionof the distances between any two neighboring parts of appar-ently intact ultrastructural elements, mainly neurofilaments.Thdidpmcsctefidcodu2
4
4
1paL(p
RMO14 immunostaining is a widely used tool for the light-microscopic identification of axons that have suffered traumaticneurofilament (ultrastructural) compaction (Lee et al., 1987;Stone et al., 2001). However, it does not work in animals thatsurvive for less than 15 min post-injury (Buki et al., 1999), norin thin (less than a few �m in diameter) compacted axons inthe rat neocortex, even after a survival period of 4 h (Fig. 4h).The target of the RMO14 antibody is the rod domain of themid-sized neurofilament protein, which is masked in normalaxons, but becomes unmasked as a result of calcium-initiated andcalpain-mediated spectrin proteolysis in the axons that displayneurofilament (ultrastructural) compaction (Buki et al., 1999).Mid-sized neurofilament protein is known to be the main regu-lator of axon thickness; its absence decreases the axon caliber,while its presence is required to achieve the maximal axon diam-eter (Elder et al., 1998). This is probably the reason for the lackof RMO-14 staining in thin axons in the cortex.
The difference in location between the anti-APP-stainedaxonal swellings and the silver-stained axonal segments in therats that survived for 4 h following the production of a 1.0-mmdeep calvaria depression confirms the observation of Stone et al.(2001) that the neurofilament (ultrastructural) compaction andthe impairment of axonal transport are independent phenomena.
4.2. Nature of the mechanical load delivered by the presenthead-injury apparatus
tqthwppidlm(ow
tuoprst
fiptos
he spaces among the compacted ultrastructural elements wereyper-electrondense, which gave the affected axon profiles aark appearance. The axolemma did not exhibit any abnormal-ty at an intermediate magnification of electron microscopy,espite the fact that a relatively large volume of fluid must haveassed through it. Some of this fluid was found between theyelin sheath and the axolemma, and some in nearby astro-
ytic processes. The compacted axonal profiles were randomlycattered in an otherwise intact environment. There was no indi-ation of axonal profiles characteristic of impaired intra-axonalransport (the focal accumulation of mitochondria, vesicles andndoplasmic reticulum intermingled with disorganized neuro-lament bundles; Stone et al., 2001). In the rats with a 1-mmeep calvaria depression, in addition to the above morphologi-al changes, a few axonal profiles displayed the characteristicsf impaired intra-axonal transport, and a few somal and/or den-ritic profiles displayed the characteristics of somato-dendriticltrastructural compaction described elsewhere (Csordas et al.,003; Gallyas et al., 2004).
. Discussion
.1. Methodological comments
Immediately after the administration of head injury and alsoh later, the silver method used here evenly blackens the axonsossessing compacted ultrastructure in a selective (Gallyas etl., 2002) and reproducible (Newman and Jasani, 1998) manner.ater, the affected axons are traced out with silver-stained dots
Gallyas and Zoltay, 1992). The dotted axonal staining disap-ears gradually in 2 days (Zsombok et al., 2005).
With the present head-injury apparatus, the mechanical loado the brain depends only on two reproducible, adjustable anduantifiable parameters: the depth of the calvaria depression andhe velocity of its production. The latter can be adjusted via theeight from which the weight is dropped. A height of 1.5 m,hich gives a final velocity of 5.44 m/s to the dropped weight,roved to be suitable for the selective induction of axonal com-action in the neocortex. Lower velocities tended to favor thenitiation of ultrastructural compaction in the neuronal soma-endrite domains (our unpublished observation), whereas a veryow velocity (0.5 mm/s), produces ultrastructural compaction
ainly in astrocytes, much less in neurons, but not at all in axonsToth et al., 1997). All these suggest that it may be the velocityf the calvaria depression, i.e. a dynamic factor that determineshich of the tissue elements mentioned will be damaged.A further factor that should be considered as the cause of
he axonal damage in our case is a strain in the cortical tissuender the impact site, which is evoked by the lentiform shapef depression in its superficial layers. This possibility is sup-orted by the fact that most affected axons in our experimentsun parallel to the outline of the depressed cortical surface. Con-equently, a sudden stretch in the cortical tissue may contributeo the development of the axonal damage in our case.
A third factor that should also be considered consists in theact that, in our case, the sudden increase in intracranial pressures immediately followed by a sudden decrease in intracranialressure, which is also able to cause morphological damage tohe rat brain (Shreiber et al., 1999). Since the experiments withur apparatus are not qualified to decide among these factors, wehare the non-committed opinion of Shreiber et al. (1999) that
J. Pal et al. / Journal of Neuroscience Methods 153 (2006) 283–289 289
some dynamic deformation in the cortical tissue is responsiblefor causing the axonal damage in our case.
Above a critical value (150 g, our unpublished observation)the mass of the dropping weight does not appreciably influencethe numbers of affected axons in the rats that suffered calvariadepressions of 0.75 mm in depth, probably because the energy inexcess to that needed for the production of the calvaria depres-sion is absorbed by the stopping sleeve when the dropping weightstrikes against it.
4.3. Possible fields of application of the head-injuryapparatus presented
In case of a depression depths of 0.75 mm, the head-injuryapparatus presented is unparalleled in four senses: (1) it pro-duces ultrastructural compaction in diffusely distributed axonsegments without co-production of other kinds of primary or sec-ondary axon damage, (2) the affected axon segments are presentin an apparently intact environment, (3) in a well-defined areaof the neocortex, but nowhere else and (4) the blood pressure,intracranial pressure, heart rate and respiration rate return tothe normal ranges in 1 min. These qualities make this apparatussuitable for the electron-microscopic investigation of the fate ofthe compacted axon segments during a long post-injury period,without the interference of other kinds of morphological dam-ages or of unfavorable post-injury circumstances. Furthermore,tcboqb
A
Nmt1A
R
A
B
C
Csordas A, Mazlo M, Gallyas F. Recovery versus death of “dark” (compacted)neurons in non-impaired parenchymal environment. Acta Neuropathol2003;106:37–49.
Elder GA, Friedrich Jr VL, Bosco P, Kang C, Gourov A, Tu PH, et al.Absence of mid-sized neurofilament subunit decreases axonal calibers,levels of light neurofilament (NF-L), and neurofilament content. J CellBiol 1998;141:727–39.
Ellingson BM, Fijalkowski LJ, Pintar FA, Yoganandan N, Gennarelli TA.New mechanism for inducing closed head injury in the rat. Biomed SciInstrum 2005;41:86–91.
Feeney DM, Boyeson MG, Linn RT, Murray HM, Dail WG. Responses tocortical injury. I. Methodology and local effects of contusion in the rat.Brain Res 1981;11:67–77.
Gallyas F, Zoltay G. An immediate light microscopic response of neuronalsomata, dendrites and axons to non-contusing consussive head injury.Acta Neuropathol 1992;83:386–93.
Gallyas F, Hsu M, Buzsaki G. Four modified silver methods for thick sec-tions of formaldehyde-fixed mammalian central nervous tissue: “dark”neurons, perikarya of all neurons, microglial cells and capillaries. J Neu-rosci Method 1993;50:159–64.
Gallyas F, Farkas O, Mazlo M. Traumatic compaction of the axonal cytoskele-ton induces argyrophilia: histological and theoretical importance. ActaNeuropathol 2002;103:36–43.
Gallyas F, Farkas O, Mazlo M. Whole-cell ultrastructural compaction (“dark”cell formation) in mammalian tissues may consist in gel-to-gel phasetransition. Biol Cell 2004;96:313–24.
Gennarelli TA. Animate models of human head injury. J Neurotrauma1994;11:357–68.
Graham DI, McIntosh TK, Maxwell WL, Nicoll JAR. Recent advances inneurotrauma. J Neuropathol Exp Neurol 2000;59:641–51.
Lee VM, Carden MJ, Schlaepfer WW, Trojanowsky JQ. Monoclonal anti-
M
N
P
S
S
S
T
Z
he effect of experimentally induced post-injury pathological cir-umstances (such as elevated intracranial pressure or impairedlood circulation) or the effect of future therapeutic experimentsn the fate of the compacted axons, or else the cumulative conse-uences of recurrent minor head injuries can also be investigatedy means of this apparatus.
cknowledgments
The authors thank Andok Csabane, Nyiradi Jozsef and Dr.ador Andrasne for their valuable help in the light- and electron-icroscopic, and the photographic preparative work, respec-
ively. This study was supported by the Hungarian grants NKFP/A 00026/2002 and ALK-00126/2002, and the Hungariancademy of Sciences.
eferences
dams JH, Graham DI, Gennarelli TA. Head injury in man and experimentalanimals. Acta Neurochir 1983;32(Suppl.):15–30.
uki A, Siman R, Trijanovsky JQ, Povlishock JR. The role of calpainmediated spectrin proteolysis in traumatically induced axonal injury. JNeuropathol Exp Neurol 1999;58:365–75.
ammermeyer J. The importance of avoiding “dark” neurons in experimentalneuropathology. Acta Neuropathol 1961;1:245–70.
bodies distinguish several differentially phosphorylated states of the twolargest rat neurofilament subunits (NF-H and NF-M) and demonstratetheir existence in the normal nervous system of adult rats. J Neurosci1987;7:3474–88.
axwell WL, Povlishock JT, Graham DL. A mechanistic analysis of nondis-ruptive axonal injury: a review. J Neurotrauma 1997;14:419–40.
ewman GR, Jasani AB. Silver development in microscopy and bioanalysis:past and present. J Pathol 1998;186:119–25.
ettus EH, Christman CW, Giebel ML, Povlishock JT. Traumatically inducedaltered membrane permeability; its relationship to traumatically inducedreactive axonal change. J Neurotrauma 1994;11:507–22.
hreiber DI, Bain AC, Ross DT, Smith DH, Gennarelli TA, McIntosh TK,et al. Experimental investigation of cerebral contusion: histopathologicaland immunohistochemical evaluation of dynamic cortical deformation. JNeuropathol Exp Neurol 1999;58:153–64.
tone JR, Walker SA, Povlishock JT. The visualization of a new class oftraumatically injured axons through the use of a modified method ofmicrowave antigen retrieval. Acta Neuropathol 1999;97:335–45.
tone JR, Singleton RH, Povlishock JT. Intraaxonal neurofilament compactiondoes not evoke local axonal swelling in all traumatically injured axons.Exp Neurol 2001;172:320–31.
oth Zs, Seress L, Toth P, Ribak CE, Gallyas F. A common morphologicalresponse of astrocytes to various injuries: “dark” astrocytes. A light andelectron microscopic analysis. J Brain Res 1997;38:173–86.
sombok A, Toth Z, Gallyas F. Basophilia, acidophilia and argyrophilia of“dark” (compacted) neurons in the course of their formation, recoveryor death in an otherwise undamaged environment. J Neurosci Method2005;142:145–52.
ORIGINAL PAPER
Ferenc Gallyas Æ Jozsef Pal Æ Orsolya Farkas
Tamas Doczi
The fate of axons subjected to traumatic ultrastructural(neurofilament) compaction: an electron-microscopic study
Received: 14 November 2005 / Revised: 7 December 2005 / Accepted: 9 December 2005 / Published online: 17 February 2006� Springer-Verlag 2006
Abstract By means of a new head-injury apparatus, a0.75-mm-deep depression was produced momentarily ata predetermined site of the rat calvaria. This immedi-ately evoked ultrastructural (neurofilament) compactionin many myelinated axon segments in layers IV and V ofthe neocortex under the impact site. The affected axonsegments run quasi-parallel to the brain surface in adiffuse distribution among normal axons. Other kinds ofdamage to the brain tissue were insignificant; the con-ditions were therefore favorable for investigation of thefate of the compacted axons. Quantitative analysis of thefindings on groups of ten rats that were sacrificed eitherimmediately after the head injury or following a 1 day ora 1 week survival period showed that around 50% of thecompacted axons recovered in 1 day, and a further lessthan 10% did so in 1 week. Electron microscopy re-vealed that the non-recovering compacted axonsunderwent a sequence of degenerative morphologicalchanges including homogenization, fragmentation andresorption of the fragments. However, the myelinsheaths around these degenerating axons remainedapparently unchanged even in the long-surviving rats,and hardly any phagocytotic cells were encountered. Onthe other hand, many such myelin sheaths containedaxolemma-bound, normal-looking axoplasm besides theabove morphological signs of axon-degeneration. It isconcluded that the non-recovering compacted axonsundergo an uncommon (non-Wallerian) kind of degen-eration, which is mostly reversible.
Keywords Recovery Æ Degeneration Æ Regeneration ÆElectron microscopy
Introduction
In human neurology, diffuse axonal injury (DAI) is aprimary cause of disability or death following headinjuries [1]. However, mainly because of the complexityof primary and secondary morphological damages to thebrain tissue [12], human neuropathology is not qualifiedfor the elucidation of its patho-mechanism. Although itsequivalent in experimental animals, traumatic axonalinjury (TAI), does not parallel exactly its nature, dis-tribution and time course, the investigation of TAI isconsidered [14] to improve our knowledge on DAI.
Two distinct mechanisms of TAI have been estab-lished: focal impairment of the intra-axonal transportleading to delayed axon disconnection, and neurofila-ment (ultrastructural) compaction in axons [21]. Thelatter phenomenon consists in a dramatic decrease of thedistances between adjacent ultrastructural elements,mainly neurofilaments. The affected axons are scatteredamong apparently normal axons, frequently in anapparently intact environment. Although this phenom-enon was discovered 10 years ago [19], its contributionto the morphological, neurological and behaviouralconsequences of head injury in human neuropathologyhas not been investigated at all. Even the fate (recoveryor death) of the affected axons in animal experimentshas not been cleared.
A newly elaborated rat head injury apparatus [17] isable to immediately produce ultrastructural (neurofila-ment) compaction in myelinated axons in a well-definedarea of the neocortex under the impact site, withoutgiving rise to any appreciable extent of other kinds ofmorphological damage to the brain. Thus, the ultra-structural appearance of the compacted axons, whichpresumably changes during a long post-injury period,cannot be confused with the ultrastructural formationsoriginating from the axonal degeneration initiated bydirect tearing or delayed disconnection, or death of theparent neuron. Further, under the circumstances used inthe present study, the blood pressure, intracranial pres-
F. Gallyas (&) Æ O. Farkas Æ T. DocziDepartment of Neurosurgery, Faculty of Medicine,Pecs University, Ret utca 2, 7623, Pecs, HungaryE-mail: [email protected].: +36-72-535930Fax: +36-72-535931
J. Pal Æ T. DocziClinical Neuroscience Research Group of the Hungarian Academyof Sciences, Pecs University, 7623, Pecs, Hungary
Acta Neuropathol (2006) 111: 229–237DOI 10.1007/s00401-006-0034-3
sure, heart-pulse rate and respiration rate return to thenormal range in 1 min. [17], which creates exogenouscircumstances favorable for a spontaneous recovery.
It has long been known that a proportion of neuronsthat have undergone somato-dendritic ultrastructuralcompaction, caused either by a pathometabolic condi-tion (such as hypoglycemia [2]) or by a physical insult(including head injury [6]) recover spontaneously. Sincethe mechanism of ultrastructural compaction in thesoma-dendrite domain is probably the same as that inthe axons [11], it is reasonable to assume that a pro-portion of traumatically compacted axons may also re-cover spontaneously.
By means of the head-injury apparatus mentioned,the present study follows for 6 months the fate of axonsegments subjected to traumatic ultrastructural com-paction. As no light-microscopic marker is available thatcould demonstrate the structural changes in the com-pacted axons for longer than a few days (see later),electron microscopy is used throughout.
Materials and methods
Animal experiments
A total of 48 Wistar rats weighing between 190 and210 g were anesthetized by the intraperitoneal adminis-tration of 2 ml/kg of a 1:1 mixture of 25 mg/ml Thio-pental (Biochemie GmbH, Austria) and 5 mg/mlSeduxen (Richter Gedeon Rt, Budapest). Nine groups ofrats were studied, with three rats in each of groups 1, 3,and 6–9, and 10 rats in each of groups 2, 4 and 5. Thegroup 1 rats were not subjected to head injury, but weresham-operated and then perfusion-fixed transcardially(control rats). In all the other rats, a 0.75-mm-deepmomentary calvaria depression was produced with anew rat head-injury apparatus [17]. Its head holder wastilted at 20� relative to the baseboard, and the axis of theimpactor head was centered over the left side of thecalvaria, 2.0 mm caudal to the bregma and 2.0 mmlateral to the midline. After the infliction of the headinjury, the rats in group 2 were perfusion-fixed imme-diately, whereas the rats in groups 3–9 were perfusion-fixed after survival for 4 h, 1 day, 1 week 1 month, or 2,4 and 6 months, respectively. The rats in groups 3–9were allowed to recover, but re-anesthetized beforefixation.
Each rat was perfused transcardially with 500 ml ofan electron-microscopic fixative, prepared by mixing250 ml of 0.2 mol/l sodium cacodylate, 50 ml of 20%paraformaldehyde, 50 ml of 25% glutaraldehyde, 25 mlof 0.1 mol/l calcium chloride and 125 ml of 10% poly-vinylpyrrolidone K25, followed by adjustment of themixture to pH 7.5 with a few drops of 0.1 mol/lhydrochloric acid. Before fixation, the vascular systemwas rinsed with physiological saline for 30 s. After fix-ation, the rats were left untouched at room temperaturefor 24 h before removal of the brain from the skull [4].
Animal care and handling were carried out in accor-dance with order 243/1998 of the Hungarian Govern-ment, which is an adaptation of directive 86/609/EGKof the European Committee Council.
Tissue processing and staining
The middle thirds of the brains were serial-sectioned at150 lm on a vibratome. The sections corresponding tothe coronal plane 2 mm caudal to the bregma (accordingto the atlas of Paxinos and Watson [16]) were selected.From each of them, the parts of the neocortex between1.5 and 2.5 mm, both right and left from the midline,together with the underlying white matter, were cut out.Additionally, from three rats in each traumatized group,three further parts of the left (traumatized) neocortexwere cut out, together with the subcortical white matter(0–1.5 mm, 2.5–4 mm and 4–5.5 mm with respect to themidline). All these specimens were postfixed with a 1:1mixture of 2% osmium tetroxide and 3% potassiumhexocyanoferrate(II) for 1 h at room temperature, andthen flat-embedded in Durcupan ACM. The specimensfrom the right (non-traumatized) hemisphere of thehead-injured rats served as ‘‘internal control’’. Fromeach specimen, semithin (1 lm) sections were cut andstained for 1 min. at 90�C in a solution containing0.05% toluidine blue, 0.05% sodium tetraborate and0.1% saccharose (pH 9.5). Thereafter, the cortical layersI and II were trimmed out from the specimens from thehead-injured rats. Thin (50 nm) sections were mountedon G300 square mesh grids (TAAB GG007/C) in such amanner that the direction perpendicular to the corticalsurface should be parallel to either of the grid bars.Sections were contrasted with uranyl acetate and leadcitrate in the usual manner.
From the remaining vibratome sections, every fifthone was stained with a silver method that selectivelydemonstrates freshly-compacted axons and neuronalsoma-dendrite domains [9]. Briefly, following dehydra-tion with graded 1-propanol, vibratome sections wereincubated for 16 h at 56�C in 1-propanol containing0.8% sulfuric acid and 2% water (esterification) thenrehydrated with graded 1-propanol, treated with 1%acetic acid for 10 min. and finally immersed in a specialphysical developer until the background had becomelight-brown. Thereafter, the sections were washed withthree changes of 1% acetic acid, dehydrated with graded1-propanol, cleared with clove oil and covered withCanada balsam.
Quantitative analysis
In the control rats, the myelinated axon profiles werecounted in 240 grid-squares belonging to ten adjacentcolumns arranged perpendicularly to the cortical sur-face. At a magnification of 3,000, the electron-micro-scopic image in the grid-square just examined was
230
moved at a right angle to a straight line drawn verticallythrough the center of the picture screen of the electronmicroscope, and the myelinated axons just reaching thisstraight line were counted. The counts belonging to therows of grid-squares at the same distance from thecortical surface were averaged for the three control ratsand plotted, together with the standard deviation,against the serial number of the grid-squares as countedfrom the cortical surface.
In the head-injured rats, all the myelinated axonprofiles and the compacted axon profiles were counted inthose grid-squares aligned in eight adjacent columnsperpendicular to the cortical surface that did not containmore than 150 or less than 40 myelinated axon profiles(for an explanation, see the Results). The resultingcounts were then totaled for each specimen examined.Thereafter, the totals for the rats that survived for 1 day(group 4) or 1 week (group 5) following the infliction ofthe head injury were compared with those for the ratsthat were sacrificed immediately (group 2), using theStudent’s t test.
Results
Control specimens
In all brain areas of the sham-operated rats and in thebrain areas outside the neocortex under the impact siteof the head-injured rats, the silver staining method didnot demonstrate either axons or neuronal soma-dendritedomains. Electron microscopically, none of the ultra-structural abnormalities that were found in the trau-matically damaged cortical areas were observed either inthe specimens obtained from the control rats or in the‘‘internal-control’’ specimens obtained from the head-injured rats.
Head-injured rats, light-microscopic observations
In the rats sacrificed immediately after the infliction ofthe head injury, as described and illustrated in a previous
paper [17], the silver method stained evenly numerouslong axon segments (Fig. 1a) in the neocortex under theimpact site, but nowhere else. Most affected axons runquasi-parallel to the brain surface, but no one into thesubcortical white matter. The frequency of silver-stainedaxons decreased with the distance from the center of theimpact site.
In the rats that survived for 4 h, evenly silver-stainedaxons were scarce. However, many axons (Fig. 1b) weretraced out by silver-stained dots. At 1 day post-injury,the incidence of axons displaying dotted silver stainingwas much lower (Fig. 1c). In the silver-stained sectionsof the rats with longer survival times, nothing wasindicative of a previous axonal disturbance.
Head-injured rats, electron-microscopic observations
In the rats that were sacrificed immediately, as de-scribed and illustrated in a previous paper [17], manymyelinated axons, but no non-myelinated axons orboutons displayed a marked reduction of the distancesbetween any two neighboring parts of apparently in-tact ultrastructural elements, mainly neurofilaments, inthe neocortex under the impact site, but nowhere else(compare Fig. 2a, b). However, the interior of mem-brane-bound organelles, such as the mitochondria,remained non-compacted. In parallel with the com-paction, the spaces surrounding the ultrastructuralelements visible in the transmission electron micro-scope became more electrondense, which gave the af-fected axon profiles a dark appearance. The axolemmadid not exhibit any abnormality at an intermediatemagnification of transmission electron microscopy,despite the fact that a relatively large volume of fluidmust have passed through it. A proportion of thisfluid was found between the myelin sheath and theaxolemma, and another proportion appeared to swellnearby astrocytic processes. As a result, the myelinsheaths of the affected axons became deformed andreduced in caliber. The compacted axon profiles wererandomly scattered among normal axon profiles in anotherwise intact environment.
Fig. 1 Light-microscopicappearance of compacted axonsin 150-lm-thick silver-stainedvibratome sections of layers IVand V of the neocortex in ratssacrificed immediately (a), or4 h (b) or 1 day (c) after theproduction of a 0.75-mm-deepmomentary calvaria depression.Arrowheads point tocontinuously silver-stainedaxons, and arrows to axons thatare outlined with silver-stainedpuncta. Scale bars: a–c 50 lm
231
In the rats that survived for 4 h, the ultrastructuralabnormalities in the affected cortical area were verysimilar to those found in the rats that were sacrificedimmediately. As regards the compacted axons, all theirindividual ultrastructural elements could be easily dis-cerned (Fig. 2b). Very infrequently, misaligned neuro-filaments running through loosely piled-upmitochondria, dense bodies, endoplasmic reticulum sacsand vesicles were found to fill slightly distended myelinsheath profiles (Fig. 2g).
In a few compacted axons of the rats that survived for1 day, individual ultrastructural elements could still bediscerned, whereas the overwhelming majority of thecompacted axons had become homogenized (Fig. 2c). Inthe latter, fluid-filled spaces were not found. Mito-chondrion-sized membranous whorls present in axonprofiles of otherwise normal appearance were frequentlyobserved (Fig. 2d, e). Very infrequently, considerablydistended myelin sheath profiles filled with some fluid(henceforth: watery axons; Fig. 2f) or with densely piled-up mitochondria, mitochondrion-sized dense bodies,
endoplasmic reticulum sacs and misaligned neurofila-ments (henceforth: ballooned axons; similar to thosedemonstrated in Fig. 2g) were observed. The astrocyticprofiles in the vicinity of the compacted axons appearednormal, but many astrocytic foot processes around theblood vessels in the affected cortical area were extremelyswollen. In other respects, the tissue structure was nor-mal.
In the rats that survived for 1 week or 1 month, inaddition to the ultrastructural features found in the1 day rats, four other types of morphological changeswere found. (1) In a large proportion of the com-pacted axons, the membrane-bound, homogenousinterior had become convoluted or even fragmented(Fig. 3a, b). The spaces around such formations ap-peared frequently (but not always) to be extensions ofthe space present between the axolemma and themyelin sheath in the non-compacted axons. Thus, theultrastructural elements in such spaces might be ofoligodendrocytic origin. (2) Ultrastructural elements ofsuch origin filled a large part (Fig. 3c, d) or the whole
Fig. 2 Electron-microscopicappearance of normal (a),compacted (b), homogenized(c), presumably recovering (d,e), watery (f) and ballooning (g)axons in layer IV or V of theneocortex in rats sacrificed 4 h(a, b, g) or 1 day (c–f) after theproduction of a 0.75-mm-deepmomentary calvaria depression.White triangles denotemitochondria, white asterisksmembranous whorls and whitecircles dense bodies. Scale bars:a–c, e and f 200 nm; d 500 nm;g 1 lm
232
(Fig. 3e) of the interior of several myelin sheath pro-files. (3) In a few cases, apparently normal, axolemma-bound, normal-looking axoplasm shared the interiorof the myelin sheath with one of the above-describedultrastructural formations (Fig. 3f–i). (4) Very infre-quently, myelin sheath profiles engulfed by astrocytes(Fig. 4a), lipoid-containing phagolysosomes in pericytes(Fig. 4b), and anomalous myelin profiles (Figs. 3i, 4c–e)were present.
In the rats that survived for 2, 4 or 6 months, all themorphological changes in the compacted axons dem-onstrated in Fig. 3a–i were still present, but far lessfrequent than in the 1 week or 1 month surviving rats.The myelin sheaths around each of such axon profileswere apparently normal even in the 6 month survivingrats (Fig. 4f). In the astrocytes, debris of engulfed myelinsheaths, phagolysosomes, tight bundles of glial fila-
ments, or other abnormalities were not present. Nothingindicated macrophage infiltration or microglial prolif-eration. Fat-containing phagolysosomes were observedin a negligible number of pericytes, but in none of theresting microglial cells.
It should be emphasized that compacted or degener-ated boutons, and degenerated myelin sheath formationssimilar to those in the advanced or chronic phases ofWallerian degeneration, were never observed in any ofthe cortical and subcortical areas examined, even in thelong-surviving rats.
Quantitative analysis
The thin sections of the neocortex of the control ratscovered about 28 rows of grid-squares counted from the
Fig. 3 Electron-microscopicappearance of compacted axonsin various stages ofdegeneration (a–e) andregeneration (f–i) in layer IV orV of the neocortex in ratssacrificed 1 week (a, b) or1 month (c–i) after theproduction of a 0.75-mm-deepmomentary calvaria depression.White asterisks denote changesin the compacted axonalultrastructure, black asterisksaxolemma-bound axoplasm,white squares some structure ofeither oligodendrocytic orneuronal origin, and whitetriangles mitochondria. In i, anarrowhead points to anapparently normal myelinsheath in an unusual position.Scale bars: a, c and h 500 nm; b,d–g and i 200 nm
233
brain surface. Cortical layers IV and V, where a majorityof the compacted axon segments resided in the rats thatwere sacrificed immediately, corresponded approxi-mately to the grid-squares in rows 9–22 (Fig. 5). Inthese, the numbers of myelinated axon profiles per grid-square were on average between 40 and 150. Thus, ineach head-injured rat, about 100 such grid squares werepresent in the eight columns examined.
As regards the head-injured rats involved in thequantitative analysis, the average and the standarddeviation of the numbers of all myelinated axon profileswere 8,451±908 in group 2, 8,494±960 in group 4 and8,549±924 in group 5. The differences between any twoof them were insignificant: t18=0.1028, P>90% be-tween groups 2 and 4; t18=0.2282, P>80% betweengroups 2 and 5; and t18=0.1331; P>80% between
groups 4 and 5. On the other hand, the average andstandard deviation of the numbers of compacted axonprofiles were 542±270 in group 2, 240±181 in group 4and 191±137 in group 5. The differences between therats that were sacrificed immediately (group 2) and thosewhich survived for 1 day (group 4) or 1 week (group 5)were significant: t18=2.917, P<1% and t18=3.905,P<1%, respectively. The average number of compactedaxon profiles was less in the 1 week rats than in the1 day rats, but the difference between them was notsignificant (t18=0.684, P�50%).
In the rats that were sacrificed immediately, no bal-looning axon was counted, whereas in the 1 day and the1 week rats, there were 3 and 9 ballooned axons,respectively. The incidences of these damaged structuresas compared to the total numbers of compacted axons in
Fig. 4 Electron-microscopicpictures in layer IV or V of theneocortex in rats sacrificed1 month (a–e) or 6 months (f)after the production of a 0.75-mm-deep momentary calvariadepression. The pictures in aand b indicate the phagocytosisof myelinated axons, in c–e amalfunction ofoligodendrocytes and in f thepossibility that apparentlynormal myelin sheaths mayexist around compacted andhomogenized axons even after a6 months survival period. In a,an arrow points to an engulfedmyelin sheath; A denotesastrocytic cytoplasm and Nastrocytic nucleus. In b, whitearrowheads point to basallaminae; black squares denotefat-containing phagolysosomes,white squares other lysosomesand A an extremely swollenastrocytic foot process. In c–e,arrows point to abnormalmyelin sheath figures. In d, Odenotes oligodendrogliacytoplasm and N anoligodendroglia nucleus. In e,an asterisk denotes compactedand homogenized axoplasm. Inf, arrowheads point to normalmyelin sheaths; the insert is amagnification of the compactedaxon denoted by an asterisk.Scale bars: a and b 500 nm;c 200 nm; d–f 1 lm
234
the same rat groups (henceforth: relative incidence) wereless than 0.5%.
Discussion
Methodological comments
One of our observations (discussed below) was thepresence of membranous whorls inside several myelin-ated axon profiles displaying an otherwise normalultrastructure. In some respects, these whorls resembledthe myelin bodies produced by inadequate brain fixa-tion. In order to avoid the artificial formation of myelinbodies, calcium chloride and polyvinylpyrrolidone wereadded to the fixative, and potassium hexacyanoferrat(II)was added to the osmicating solution [13]. It should benoted that the quality of the ultrastructural preservationin the surroundings of each electron-microscopic area inFigs. 2 and 3 was as good as that demonstrated inFig. 4f. This photo also demonstrates that both thecompacted and the non-compacted myelinated axonprofiles are clearly visible in the electron microscope at amagnification of 3,000, where they were counted for aquantitative analysis.
The number of axon segments stained either evenly ordotted with silver decreased to zero within a few days.This could be assumed to be a result of the spontaneousrecovery of all compacted axons or a decrease in time oftheir stainability with silver [22]. A comparison with theelectron microscopic observations and the quantitativedata suggests that both assumptions are partly valid.Furthermore, the compacted axons in the rat neocortexcannot be stained with the RMO14 antibody [17], whichantibody is widely used for their light-microscopicdemonstration in the brain stem, pons and oblongata[21]. This inconsistency may be explained by the differ-ence in thickness of the axons studied. Specifically, the
target of the RMO14 antibody is the rod domain of themid-sized neurofilament protein, which is masked innormal axons, but becomes unmasked as a result ofcalcium-initiated and calpain-mediated spectrin prote-olysis in the axons that display neurofilament (ultra-structural) compaction [3]. Mid-sized neurofilamentprotein is known to be the main regulator of axonthickness; its absence decreases the axon caliber, whileits presence is required to achieve the maximal axondiameter [7]. This is probably the reason for the lack ofRMO-14 staining in thin axons in the cortex. For allthese, only electron microscopy is qualified to furnish areliable answer regarding the fate of traumaticallycompacted axons. In this respect it should be noted thatthe decrease in distance between adjacent neurofilamentsin thin compacted axons in the neocortex immediatelyafter the head injury (58.4%, [11]) is similar to thatfound in thick axons in the brain stem (62.6%, [18]).
Besides the well-known ultrastructural features ofcrushed or dissected axons in the early phase of Walle-rian degeneration (the accumulation of mitochondria,endoplasmic reticulum cisternae, vesicles and mito-chondrion-sized dense bodies), two other types of axondegeneration occur in the central nervous system:‘‘dark’’ degeneration and watery degeneration [15]. Theelectron-microscopic features of ‘‘dark’’ degenerationare very similar to those demonstrated in our Figs. 2cand 3a, b, whereas the appearance of watery degenera-tion resembles that demonstrated in our Fig. 2f. This isthe reason why head-injury apparatuses that also pro-duce an immediate or a delayed disconnection in axonsare not suitable for clarification of the fate of the com-pacted axons.
Conclusions drawn from the quantitative analysis
The fact that the differences between any two of thethree rat groups concerning the numbers (mean ± -E.S.M.) of all (normal + compacted) axon profiles wereinsignificant indicates that similar populations of axonswere examined, and that the method of data samplingused is suitable for quantitative comparisons. As regardsthe number of compacted axons, there was some diver-sity among the individual rats in any of the three groupsexamined. In addition to chance error attributable to thehead-injury apparatus, this could be caused by variationin the thickness and elasticity of the rat’s calvaria, whichcertainly depends on the rat’s age and breed. However,the degree of this diversity does not contradict the use ofthree rat groups for qualitative purposes. In Wistar ratsweighing about 200 g, approximately 6% of all the ax-ons in the cortical area examined became compacted.
The significantly lower numbers of compacted axonprofiles in the rats that survived for 1 day as comparedwith those in the rats sacrificed immediately clearlyshowed that a large proportion (more than 50%) of thecompacted axons recover in less than 1 day while a smallproportion (less than 10%) between 1 day and 1 week.
Fig. 5 The average number and standard deviation of myelinatedaxons per grid-square in relation to the serial numbers of the grid-squares counted from the cortical surface
235
The ability of the compacted axons to recover spon-taneously indirectly supports our assumption [11] thattheir mechanism of formation is the same as that of thecompacted (‘‘dark’’) neurons, which are also able torecover spontaneously [6]. The low relative incidence ofballooned axons allows the supposition that neither thesomato-dendritic ultrastructural compaction nor thefocal impairment of intra-axonal transport (which leadsto the formation of axon balloons, [20]) influences theconclusions drawn from the present quantitative andelectron-microscopic findings. Moreover, the low rela-tive incidence of ballooned axons confirms the obser-vation [21] that axonal compaction and the focalimpairment of axonal transport are not directly related.
Conclusions drawn from the electron-microscopicobservations
Despite the momentary nature of the traumatic ultra-structural compaction in axons [11], the onset time of thesubsequent pathological changes (convolution, frag-mentation, resorption of the fragments) in the affectedaxons segments varied considerably. This made itmeaningless to attempt an exact correlation of the se-quence of morphological changes with the duration ofsurvival. For the same reason, the interpretation of theultrastructural findings could be composed only of amixed set of ‘‘mosaics’’. Overall, however, three path-ways appear to be available for the compacted thin ax-ons in the neocortex as concerns their fate: (1)spontaneous recovery, (2) irreversible degeneration and(3) reversible degeneration. It should be noticed that theaxon caliber, the anatomical region and the animalspecies may influence considerably the extent ofinvolvement of these pathways.
The only electron-microscopic sign of spontaneousrecovery, in which more than 50% of the compactedaxons is involved, is the presence of membranous whorlsin several axons displaying an otherwise normal-lookingultrastructure. Such membranous whorls were found inthe affected cortical areas of the rats that survived for1 day or 1 week, but not in their corresponding ‘‘inter-nal-control’’ areas or in the electron-microscopic speci-mens of the other head-injured rats. Similarmembranous whorls were found in otherwise normal-looking neuronal somata and dendrites in otherwisenormal-looking granule neurons in the hippocampaldentate gyrus where electrically compacted neurons hadrecovered [6]. Lamellated myelin sheaths invaginatedinto axons, which occasionally resemble to our mem-branous whorls, had been found by other authors andinterpreted as the response of myelin to the applicationof tensile strain to nerve fibers (reviewed in [14]). How-ever, since our membranous whorls are present in axonssurrounded by normal-looking myelin sheaths we as-sume that they are not myelin-derived abnormalities.
The chance for recovery probably exists as long as theultrastructural formations produced immediately after
the head injury persists, i.e., for several hours after thehead injury. This affords a good opportunity for futuretherapeutic attempts to increase the probability ofrecovery. On the other hand, pathophysiological con-ditions such as an increased intracranial pressure orischemia, if they exist in this period, may decrease theprobability of recovery. The persistence for hours of theprimary ultrastructure in traumatically compacted ax-ons in an animal brain was first observed by other au-thors [18].
The phenomenon of the irreversible degeneration ofthe compacted axons is indicated directly by the pres-ence of myelin debris in astrocytes, and indirectly by thepresence of fat-containing phagolysosomes in pericytes.The incidences of these abnormalities were estimated tobe very low as compared with those of the ultrastruc-tural changes of the compacted axons in long-survivingrats. The absence of microglial proliferation and mac-rophage infiltration indicates that, at least under non-impaired circumstances, appreciable proportions ofcompacted axons are not involved in this pathway.However, long-lasting pathophysiological conditionssuch as an increased intracranial pressure or ischemiamay increase the probability of irreversible degenera-tion. The abnormal myelin sheath configurations de-picted in Figs. 3i and 4c–e are probably related to somemalfunction of oligodendrocytes rather than to myelindegeneration concomitant to the Wallerian axondegeneration.
As regards reversible degeneration, a comparison ofthe message of the previous paragraph with the relevantquantitative data suggests that a considerable propor-tion of the compacted axons regain their normal ultra-structure within a few months following a degenerationprocess. The latter begins with the homogenization ofthe axonal ultrastructure, which is completed during thefirst day post-injury in most compacted axons. In themeantime, the fluid expressed from the axoplasm flowssomehow out of the myelin sheath, or becomes accu-mulated inside the myelin sheath, filling most part oreven the whole of the interior of a cross-sectioned myelinprofile. Thereafter, the homogenized axon structureundergoes convolution and fragmentation, which canstart at any time between 1 day and 6 months post-in-jury. The next process is the gradual resorption of thesefragments, accompanied by their replacement probablywith oligodendrocyte cytoplasm. In parallel with allthese processes, axolemma-bound normal axoplasm isassumed to grow into the compacted segment, settingout from its proximal end, and can be expected toestablish a functioning connection between the parentcell body and its boutons. The myelin sheaths aroundthe degenerating compacted axons remain non-impairedfor several months and ensure guide tubes for theregenerating axons. The reason why the axon degener-ation following ultrastructural compaction does not leadto myelin degeneration, which is characteristic of theWallerian degeneration, is probably that the affectedaxon segments remain in physical contact with the
236
parent neurons. Concerning the effort of mammalianbrain neurons to attain axon regeneration, it is knownthat an attempt at regeneration takes place in thedistal end of the proximal part even of the disconnectedaxons [5].
As revealed by the silver method used here, extensiveaxonal compaction in various brain areas of the rat,including the brain stem, cerebellum and pons, is a fre-quent concomitant of various head injuries [8, 10]. If thiswere the case in human head injuries too, the potentialof the compacted axons for a spontaneous recovery aswell as a reversible degeneration could be one of thecauses leading to an hours-long as well as a months-longimprovement of physical or mental capabilities that werelost immediately.
Acknowledgments The authors thank Andok Csabane, NyiradiJozsef, and Dr. Nador Andrasne for their valuable help in the light-microscopic, electron-microscopic and photographic preparativework, respectively. This study was supported by Hungarian grantsNKFP 1/A 00026/2002 and ALK-00126/2002, and by theHungarian Academy of Sciences.
References
1. Adams JH, Graham DI, Gennarelli TA (1983) Head injury inman and experimental animals. Acta Neurochir 32(Suppl):15–30
2. Auer RN, Kalimo H, Olsson Y, Siesjo BK (1985) The temporalevolution of hypoglycemic brain damage. I. Light- and elec-tron-microscopic findings in the rat cerebral cortex. ActaNeuropathol 67:13–24
3. Buki A, Siman R, Trojanovsky JQ, Povlishock JR (1999) Therole of calpain mediated spectrin proteolysis in traumaticallyinduced axonal injury. J Neuropathol Exp Neurol 58:365–375
4. Cammermeyer J (1961) The importance of avoiding ‘‘dark’’neurons in experimental neuropathology. Acta Neuropathol1:245–270
5. Christman CW, Salivan JB Jr, Walker SA, Povlishock JT(1997) Characterization of a prolonged regenerative attempt bydiffusely injured axons following traumatic brain injury in adultcat: a light and electron microscopic immunocytochemicalstudy. Acta Neuropathol 94:329–337
6. Csordas A, Mazlo M, Gallyas F (2003) Recovery versus deathof ‘‘dark’’ (compacted ) neurons in non-impaired parenchymalenvironment. Acta Neuropathol 106:37–49
7. Elder GA, Friedrich VL Jr, Bosco P, Kang C, Gourov A, TuPH, Lee VM, Lazzarini RA (1998) Absence of mid-sized
neurofilament subunit decreases axonal calibers, levels of lightneurofilament (NF-L), and neurofilament content. J Cell Biol141:727–739
8. Gallyas F, Zoltay G (1992) An immediate light microscopicresponse of neuronal somata, dendrites and axons to non-contusing consussive head injury. Acta Neuropathol 83:386–393
9. Gallyas F, Hsu M, Buzsaki G (1993) Four modified silvermethods for thick sections of formaldehyde-fixed mammaliancentral nervous tissue: ‘‘dark’’ neurons, perikarya of all neurons,microglial cells and capillaries. J Neurosci Meth 50:159–164
10. Gallyas F, Farkas O, Mazlo M (2002) Traumatic compactionof the axonal cytoskeleton induces argyrophilia: histologicaland theoretical importance. Acta Neuropathol 103:36–43
11. Gallyas F, Farkas O, Mazlo M (2004) Whole-cell ultrastruc-tural compaction (‘‘dark’’ cell formation) in mammalian tissuesmay consist in gel-to-gel phase transition. Biol Cell 96:313–324
12. Graham DI, McIntosh TK, Maxwell WL, Nicoll JAR (2000)Recent advances in neurotrauma. J Neuropathol Exp Neurol59:641–651
13. Hayat MA (1981) Fixation for electron microscopy Academic,New York, pp 88, 123, 212
14. Maxwell WL, Povlishock JT, Graham DL (1997) A mecha-nistic analysis of nondisruptive axonal injury: A review. JNeurotrauma 14:419–440
15. Narciso MS, Hokoc JN, Martinez AMB (2001) Watery anddark axons in Wallerian degeneration of the opossum’s opticnerve: different patterns of cytoskeletal breakdown? Ann AcadBras Cienc 73:231–243
16. Paxinos G, Watson C (1982) The rat brain in stereotaxiccoordinates Academic, Sydney, New York, plates 18 and 19
17. Pal J, Toth Zs, Farkas O, Kellenyi L, Doczi T, Gallyas FSelective induction of ultrastructural (neurofilament) compac-tion in axons by means of a new head injury apparatus. JNeurosci Meth (in press)
18. Pettus EH, Povlishock JT (1996) Characterization of a distinctset of intraaxonal ultrastructural changes associated withtraumatically induced alteration in axolemmal permeability.Brain Res 722:1–11
19. Pettus EH, Christman CW, Giebel ML, Povlishock JT (1994)Traumatically induced altered membrane permeability: itsrelationship to traumatically induced reactive axonal change. JNeurotrauma 11:507–552
20. Povlishock JT, Jenkins LW (1995) Are the pathobiologicalchanges evoked by traumatic brain injury immediate and irre-versible? Brain Pathol 5:415–426
21. Stone JR, Singleton RH, Povlishock JT (2001) Intraaxonalneurofilament compaction does not evoke local axonal swellingin all traumatically injured axons. Exp Neurol 172:320–331
22. Zsombok A, Toth Zs, Gallyas F (2005) Basophilia, acidophiliaand argyrophilia of ‘‘dark’’ (compacted) neurons in the courseof their formation, recovery or death in an otherwise undam-aged environment. J Neurosi Meth 142:145–152
237
Gel-to-gel phase transition may occur in mammalian cells:Mechanism of formation of “dark” (compacted) neurons
Ferenc Gallyas a,*, Orsolya Farkas a, Mária Mázló b
a Department of Neurosurgery, Pécs University, H-7623, Rét utca 2, Pécs, Hungaryb Central Electron Microscopic Laboratory, Faculty of Medicine, Pécs University, H-7623, Rét utca 2, Pécs, Hungary.
Received 2 October 2003; accepted 23 January 2004
Available online 12 April 2004
Abstract
In the course of many diseases, individual non-apoptotic cells that are randomly distributed among undamaged cells in various mammaliantissues become shrunken and hyperbasophilic (“dark”). The light microscopic shrinkage is caused by a potentially reversible, dramaticcompaction of all ultrastructural elements inside the affected cells, and escape of the excess water through apparently intact plasma membrane.In the case of neurons, the ultrastructural compaction rapidly involves the soma-dendrite domains in an all-or-nothing manner, and alsomm-long axon segments. The present paper demonstrates that such ultrastructural compaction in neurons, which affects the wholesoma-dendrite domain or long axon segments, can take place both immediately after an in vivo head injury and in rat brains perfusion-fixed for30 min., and then chilled to just above the freezing point before the same kind of head injury was inflicted. This argues strongly against anyenzyme-mediated compaction mechanism. On the analogy of gel-to-gel phase transitions in polymer chemistry, we hypothesize a purephysico-chemical compaction mechanism. Specifically, after initiation at a single site in each affected cell, the ultrastructural compaction ispropelled throughout the whole cell on the domino principle by the free energy stored in the form of non-covalent interactions among theconstituents of some cytoplasmic gel structure.© 2004 Elsevier SAS. All rights reserved.
Keywords: Head injury; Post mortem; Cytoplasmic gel; Mechanical energy; Energy storing
1. Introduction
Head injuries, electric shock and many diseases or path-ometabolic conditions can cause a common type of morpho-logical damage to a number of cells randomly scatteredamong morphologically intact cells of the same kind inmammalian tissues. The morphological features of the af-fected cells, which are traditionally called “dark”, are (a)hyperbasophilia of the cytoplasm, (b) nuclear pyknosis, (c)massive shrinkage of the whole cell, (d) increased electrondensity, (e) compaction of the ultrastructural elements, and(f) aggregation of the nuclear chromatin with a non-apoptoticpattern. “Dark” cell formation has been reported to occur inthe somatotrophic cells of the adenohypophysis, the reticularcells of the thymus and lymph nodes, the chief cells of thecarotid body, the cortical cells of the adrenal gland, the
epithelial cells of the cornea and the mammary gland, thetubule cells of the kidney, the keratinocytes of the epidermis,mastocytoma cells, hepatic cells and neurons (reviewed byHarmon, 1987), and also in astrocytes (Tóth et al., 1997). Asa possible mechanism of their formation, compression result-ing from overcrowding, i.e. a physical injury, has been sug-gested.
In the neurons, “dark” cell formation is initiated in manyhuman diseases, such as hypoglycemia, ischemia, status epi-lepticus and head injury, and it can be initiated in experimen-tal animals by excessive stimulation, deafferentation, hyper-osmotic shock, poisoning with folic acid, tunicamycin,methoxypyridoxine, 6-hy droxydopamine, 3-acetylpyridine,colchicine, 3-aminopyridine, kainic acid or 3-nitropropionicacid (reviewed by Gallyas et al., 1990), and also by electricshock (Csordás et al., 2003). Neurons are particularly suit-able objects for elucidation of the mechanism of “dark” cellformation, because they possess long, ramified dendrites thatare always involved in the ultrastructural compaction.
This paper compares the morphological changes caused inneuronal somata, dendrites and axons of the rat brain by a
* Corresponding author: Ferenc Gallyas, Ph.D., Department ofNeurosurgery, University of Pécs, H-7623 Pécs, Rét utca 2, Hungary;Tel.: +36/72/535900, Fax: +36/72/535931.
E-mail address: [email protected] (F. Gallyas).
Biology of the Cell 96 (2004) 313–324
www.elsevier.com/locate/biocell
© 2004 Elsevier SAS. All rights reserved.doi:10.1016/j.biolcel.2004.01.009
head injury under in vivo conditions with those caused underconditions extremely unfavorable for enzyme-mediated pro-cesses, and concludes from their similarity that the formationof “dark” neurons may not be a result of a cascade of (patho)-biochemical processes. It is hypothesized that the compac-tion of apparently intact ultrastructural elements, which is theessence of “dark” cell formation, consists in a gel-to-gelphase transition that extends to the whole of the cell.
As regards gel-to-gel phase transitions, polymer chemis-try has invented synthetic gels that can assume two or moremetastable phases, each having a distinct free-energy mini-mum, a distinct set of macromolecular conformations, adistinct degree of water content (Annaka and Tanaka, 1992),and therefore a distinct volume. Initiated at a single point,transition from one gel phase to another can spread through-out the gel, propelled on the domino principle by the differ-ence in free energy (Tanaka et al., 1992). The initiation can beperformed by a subtle change around a critical value ofchemical composition, pH or temperature of the surroundingmedium, and also by illumination, an electric field or amechanical stress (reviewed by Pollack, 2001). Such gel-to-gel phase transitions restricted to special intracellular com-partments have already been suggested to play essential rolesin certain activities of the living cell, such as muscle contrac-tion (Pollack, 1996), action potential (Tasaki, 1999) andmucin secretion (Verdugo et al., 1992).
2. Results
2.1. Control rats
Neither the traumatically induced anomalous light-microscopic staining properties nor the ultrastructuralanomalies described below were observed in the control ratsor in the rats that underwent head-injury 1 day after perfusionfixation. Special attention was paid to the neocortex underthe impact site.
2.2. In vivo experiments
Intracerebral mechanical force was generated by droppinga brass rod with a mass of 450 g from a height of 100 cmthrough a vertically positioned Plexiglas tube onto a metaldisc cemented to the exposed calvaria of anesthetized rats(impact acceleration head injury, Marmarou et al., 1994).
Light microscopy: In the neocortex under the impact sitein the rats transcardially perfused with a glutaraldehyde fixa-tive immediately after the infliction of the impact-acceleration head injury, several neuronal somata togetherwith their dendritic trees (henceforth: soma-dendrite do-mains) and numerous long axon segments were stained blackwith silver in a randomly scattered distribution among un-stained neuronal cell bodies, dendrites and axons (Fig. 1a).The nucleoli of normal neurons also became black, whereasthe nuclei stained less intensely than the light brown back-
ground. The numbers of affected axon segments and soma-dendrite domains (henceforth: affected neuronal parts) var-ied widely from rat to rat. The microphoto in Fig. 1a wastaken from the most seriously affected rat. On revolution ofthe micrometer screw, individual axons could be followed in150-µm-thick vibratome sections for a maximum length of1.4 mm. Interestingly, the axons relating to the silver-stainedsoma-dendrite domains usually remained unstained. Notonly the neocortex, but also the caudate putamen, brainstem,pons and oblongata contained silver-stained neuronal parts ina scattered distribution (for a detailed description of themorphological damage caused by the head injury paradigmutilized, see Foda and Marmarou, 1994). In Durcupan-embedded semithin sections, toluidine blue stained homog-enously and intensely the somata and dendrites of severalconsiderably shrunken neurons in the same locations as didthe silver method. Even their thin dendrites could be dis-cerned from the background under a high-power microscopiclens (Fig. 1b). On the other hand, only the nucleolus and the“rim” of the nucleus were visualized in the normal neurons,while their dendrites were much lighter than the background.Several axons were also stained with toluidine blue (notshown), but, in consequence of the co-staining of the myelinsheaths, this could be discerned only under a high-powermicroscopic lens (similar to that demonstrated in Fig. 5b).
Electron microscopy: In those brain areas where silver-stained neuronal parts were present in adjacent vibratome
Fig. 1. Light-microscopic appearance of compacted neuronal somata (thickarrows), dendrites (thick arrowheads) and axons (medium arrows) in thebrain cortex under the impact site following an in vivo head injury. (A)Silver-stained, 150-µm vibratome section. The nucleoli (thin arrows) ofnormal neurons are darker, while the nuclei lighter than the background. (B)Toluidine blue-stained, 0.5-µm Durcupan-embedded section. The nuclei (N)and main dendrites (asterisks) of normal neurons are much lighter, whereasthe nucleoli (thin arrow) are darker than the background. Scale bars:A = 50 µm, B = 10 µm.
314 F. Gallyas et al. / Biology of the Cell 96 (2004) 313–324
sections, a number of neuronal cell bodies (Fig. 2), dendrites(Fig. 3) and axons (Fig. 4) displayed a dramatic compactionof apparently intact ultrastructural elements and a consider-able volume decrease. The degree of compaction appeared tobe similar throughout each affected axonal, somatic anddendritic profile including the dendritic spines. In the com-pacted axons, the cross-section area relating to a single lon-gitudinal ultrastructural element was 880±59.4 nm2, whereasin the normal-looking axons 2116±245 nm2. Their quotient,i.e. the degree of compaction, was 0.416±0.037. The plasmamembrane did not exhibit any abnormality, despite the factthat a relatively large volume of water must have passedthrough it. Interestingly, the excess water was taken up bynearby astrocytic processes (but not by the extracellularspace). However, in several compacted axons, some excesswater remained between the axolemma and the myelinsheath (Fig. 4b). The affected neuronal parts were randomlyscattered among normal neuronal parts in an otherwise mor-phologically intact environment. The compaction involved amarked reduction of the distances between any two neighbor-ing parts of the ultrastructural elements that were “visible” inthe conventional transmission electron microscope (nuclearand plasma membranes, free ribosomes, ribosome rosettes,the outer surfaces of the endoplasmic reticulum cisternae,Golgi cisternae, mitochondria, lysosomes and multivesicularbodies, and also the components of the filamentous cytoskel-eton). The distances between the inner surfaces of the endo-plasmic reticulum cisternae were also reduced. All the ultra-structural elements involved in the compaction appeared to
be interconnected with fine side-arms. In parallel with thecompaction, these became shorter and more electron-dense,which gave a dark appearance to the affected neuronal parts.The nuclei of the neurons with a compacted cytoplasm werealso compacted. The nuclear chromatin aggregated to numer-ous small clumps with irregular outlines and a myriad ofminute granules, which is a non-apoptotic pattern. Nochanges were apparent in the volumes of the mitochondria,multivesicular bodies and lysosomes, whereas the interior ofthe Golgi cisternae was always dilated.
2.3. Post mortem experiments
Rats were transcardially perfused with a glutaraldehydefixative and then chilled to just above the freezing point bybeing immersed into icy water. Thereafter, a head injury wasinflicted by means of the same impact-acceleration paradigmas that used in the in vivo experiment. Brains were removed1 day later.
Light microscopy: In the neocortex under the impact site,similar results were found as in the in vivo experiment.Specifically, the traumatically induced stainability with sil-ver was an all-or-nothing (whole-cell) phenomenon, with thefollowing restrictions: (i) if the soma of a neuron was af-fected, a continuous network of its dendrites was also af-fected, whereas its axon was rarely; (ii) the affected axonsegments were always long (up to 1.2 mm); and (iii) theneuronal soma-dendrite domains relating to affected axonsegments remained predominantly unaffected (Fig. 5a).
Fig. 2. (A) Electron-microscopic appearance of normal (N) and compacted (D) neuronal cell bodies in the brain cortex under the impact site following an in vivohead injury. White, thick arrowheads point to dilated Golgi cisternae; n denotes nucleolus, black circles markedly swollen astrocytic processes, and small whitesquares staining artifacts. In (B), which is a magnification of the area framed in (A), black or white arrowheads point to endoplasmic reticulum cisternae, andblack or white arrows to ribosome rosettes; m denotes a mitochondrion. Scale bars: A = 2 µm, B = 500 nm.
315F. Gallyas et al. / Biology of the Cell 96 (2004) 313–324
However, all the brain areas other than the neocortex did notdemonstrate such effects, and the numbers of affected neu-ronal parts in the neocortex were much smaller. The micro-photo in Fig. 5a was taken from the most seriously affectedrat. Similarly as found in the in vivo case, the head injuryinduced stainability with toluidine blue in injured neuronalsomata, dendrites and axons (Fig. 5b).
Electron microscopy: Several neuronal somata, dendritesand axons revealed the same ultrastructural features (Figs6-8) as those observed in the in vivo head-injured rats. Noexceptions were found. In the compacted axons, the cross-section area relating to a single longitudinal ultrastructuralelement was 867±63.6 nm2, whereas in the normal-lookingaxons 2066±246 nm2. Their quotient, i.e. the degree of com-paction, was 0.420±0.039.
3. Discussion
3.1. Comments on the methods used and the resultsobtained
A special silver staining method was utilized to find brainareas worthy of electron-microscopic investigation, i.e. in
which compacted neuronal parts were present in sufficientnumbers (Figs 1a and 5a). Because of the poor reputation ofthe traditional silver impregnation methods, it should bestressed that various recent methods that employ physicaldevelopers, including the present method, are reliable (New-man and Jasani, 1998). In the rat brain, this method demon-strates ultrastructural compaction selectively and reproduc-ibly both in axons (Gallyas et al., 2002) and in neuronalsoma-dendrite domains (Csordás et al., 2003). No othermethod is available for the demonstration of freshly com-pacted neuronal dendrites and axons in thick vibratome sec-tions. In thin frozen or paraffin sections, in addition to the“usual” basophilic substances (nucleic and ribonucleic acidsand polyanions) neurons with compacted somata can also bestained at about pH 4 with cationic dyes such as toluidineblue. In osmicated and Durcupan embedded 1-µm-thick sec-tions, nothing can be stained with cationic dyes at about pH4, but both the “usual” basophilic substances and the com-pacted neuronal somata, dendrites and axon segments areselectively stained at pH 9.5 (see Figs. 1b and 5b). In surviv-ing animals, from the 5th min. post injury, compacted axonscan be demonstrated by the uptake of intraventricularly ad-ministered horseradish peroxidase (Pettus et al., 1994), andfrom the 15th min. on by Ab-38 and RMO-14 immunohis-
Fig. 3. Electron-microscopic appearance of longitudinally sectioned normal (d) and compacted (black arrows) neuronal dendrites in the brain cortex under theimpact site following an in vivo head injury. Black or white arrowheads point to microtubules; m denotes mitochondria, a black asterisk glial filaments, and smallwhite squares staining artifacts. Scale bar = 500 nm.
316 F. Gallyas et al. / Biology of the Cell 96 (2004) 313–324
tochemistry (Büki et al., 1999). Freshly-produced “dark”neurons are both strongly basophilic and slightly acidophilic,thereby stainable with many dyes (Brown, 1977). The acido-philia of the “dark” neurons that die gradually increases,while their basophilia decreases (Auer et al., 1984; Kiernanet al., 1998).
For the investigation of the ultrastructural compaction inneuronal somata, dendrites and axons without the influenceof any other kinds of traumatic damage to the brain tissue, itis of fundamental importance that the affected area shouldremain otherwise intact. This is, why the rats with a calvariafracture (which results in serious laceration of the braincortex) were discarded from the present investigation (seeMaterials and Methods).
The differences, between the normal and the affectedaxons, in the average cross-section area relating to a singlelongitudinal ultrastructural element were highly significantboth after an in vivo (t18 = 15.5; P < 0.1%), and a post mortem(t18 = 14.9; P < 0.1%.) head injury. On the supposition thatthe axonal length remains unchanged during the compaction,the degree of compaction is directly proportional to thedegree of reduction in the axon volume. Accordingly, theaverage volume of the compacted axons following an in vivohead injury is 41.6±3.7% of the average volume of thenormal-looking axons, while this parameter is 42.0±3.9% forthe post-mortem case. These values are very similar (thedifference between them is not significant: t18 = 0.235;10% < P <20%), indicating that the compaction may proceed
with the same mechanism both in vivo and post mortem. (Thevolume decrease in neuronal somata and dendrites could notbe analyzed quantitatively due to problems in sampling com-parable data.)
The essence of the morphological damage caused to neu-ronal somata, dendrites and axons by a head injury in thepresent post mortem experiment is the same as that caused inthe present in vivo experiment. Specifically, in both cases, (i)a high-grade compaction of apparently intact ultrastructuralelements occurred in the affected neuronal somata (Figs2 and 6) dendrites (Figs 3 and 7) and axons (Figs 4 and 8), (ii)the degree of volume reduction in the axons was similar (seeabove;), (iii) the interior of the membrane-bound ultrastruc-tural elements (mitochondria, lysosomes, vesicles and multi-vesicular bodies) remained non-compacted, (iv) the Golgicisternae became dilated, (v) the compacted neuronal partswere scattered among normal neuronal parts in an otherwiseintact environment, (vi) whole soma-dendrite domains andlong axon segments were always affected, and the compactedneuronal parts immediately acquired (vii) basophilia, (viii)argyrophilia and (ix) hyper-electrondensity.
Dissimilarities between the in vivo and the post mortemcases were found only in the number of affected neuronalparts and in the extent of the affected areas. Both weresmaller under the post mortem circumstances, probably be-cause of a fixation-derived decrease in the elasticity of thebrain tissue. However, these dissimilarities are not sufficientto negate the message of the present investigation, specifiedin the first sentence of the previous paragraph.
Fig. 4. Electron-microscopic appearance of cross-sectioned (A) and longitudinally sectioned (B) normal (white arrowheads) and compacted (white arrows)myelinated axons in the brain cortex under the impact site following an in vivo head injury. Black arrows point to neurotubulí in a normal dendrite (d); white mdenotes myelin sheaths, and black circles fluid-filled spaces between the axolemma and myelin sheath. Neurofilaments and microtubules appear as fine dotsinside the cross-sectioned axons, while as fine lines inside the longitudinally sectioned axons. Scale bars: A and B = 1 µm.
317F. Gallyas et al. / Biology of the Cell 96 (2004) 313–324
The sporadic formation of “dark” (strikingly shrunkenand hyperbasophilic) neurons as a result of post mortemmechanical injuries to unfixed or improperly fixed brains,and its avoidance by transcardial perfusion fixation and a1-day delay of the brain autopsy, have long been known(Cammermeyer, 1960 and 1961). Since these “dark” neuronshave been regarded as artifacts, their ultrastructure has notbeen investigated so far. The in vivo formation of compactedaxons scattered among normal axons was discovered a de-cade ago in experimental animals that survived for 5 min. orlonger following a head injury (Pettus et al., 1994). Shortersurvival periods were not investigated. It was demonstratedonly recently that the process of ultrastructural compactionboth in axon segments (Gallyas et al., 2002) and in neuronalsoma-dendrite domains (Csordás et al., 2003) is completedeven in rats that are perfused transcardially with an electron-microscopic fixative immediately after the infliction of ahead injury.
As regards the soma-dendrite compaction caused by an invivo head injury, no biochemical mechanism has been postu-lated so far. For axons, traumatic ultrastructural (neurofila-ment) compaction has been assumed by other authors (e.g.Buki et al., 1999) to be a morphological manifestation ofcalpain-mediated spectrin proteolysis, initiated by an uncon-
trolled influx of Ca++ through the axolemma, perturbed fo-cally by some traumatically generated intracerebral shearingforce.
3.2. Two arguments against any enzyme-mediatedcompaction mechanism
The first argument is based on the high degree of similar-ity between the pathomorphologic features found in our invivo and post mortem experiments. This suggests that themechanism of compaction is the same in both cases. How-ever, it appears improbable that any complex enzyme-mediated process should result in the same morphologicalfeatures under both optimum and extremely unfavorable con-ditions. Namely, the individual biochemical reactions thatrun in “cascade” and/or in parallel in any complex enzyme-mediated process would be slowed down to different degreesby the partial fixation and the marked temperature dropapplied in our post mortem experiment, which would sub-stantially disturb the in vivo harmony of their action.
The second argument is based on the all-or-nothing natureof the compaction. It is impossible that a head injury shouldsimultaneously produce ultrastructural compaction at eachpoint of a number of randomly distributed neuronal soma-dendrite domains and axons and, at the same time, shouldspare all points of the neighboring soma-dendrite domainsand axons. The all-or-nothing nature can be explained onlyby the assumption that the compaction spreads throughouteach affected soma-dendrite domain or axon from a singleinitiation point. It appears highly unlikely that any complexenzyme-mediated process should spread over long intracel-lular distances without fading under the extremely unfavor-able conditions applied in our post mortem experiment.
If the above train of thoughts is true, any post-traumaticenzyme-mediated pathological process should be assumed tobe a consequence, and not the cause of the in vivo ultrastruc-tural compaction. In accordance with this assumption,calpain-mediated spectrin proteolysis in compacted axonscould be detected only in rats that survived for at least15 min. after an in vivo head injury (Buki et al., 1999).
3.3. The hypothesized non-enzymatic mechanismof the whole-cell ultrastructural compaction
We propose here a non-enzymatic mechanism, first interms of the energy available to propel the whole-cell ultra-structural compaction both in vivo and post mortem, and thenin terms of the structure capable of executing the compactionat the expense of the available energy.
The free energy of this structure comprises that of thecovalent bonds and that of the non-covalent interactions(ion-ion, dipole-dipole, van der Waals, hydrophobic and hy-drogen bonding). In contrast with the release of the freeenergy of the covalent bonds, the release of the free energy ofthe non-covalent interactions does not necessarily need en-zyme mediation. If the free energy of the non-covalent inter-
Fig. 5. Light-microscopic appearance of compacted neuronal somata (thickarrows), dendrites (thick arrowheads) and axons (medium arrows) in thebrain cortex under the impact site following a post mortem head injury. (A)Silver-stained, 150-µm vibratome section. The nuclei of normal neurons aremuch lighter, whereas the nucleoli (thin arrows) are darker than the back-ground. (B) Toluidine blue-stained, 0.5-µm Durcupan-embedded section. Acompacted neuron (thick arrow), the compacted myelinated axons (mediumarrows), and also all myelin sheaths are intensely stained, whereas thenormal neurons (N) and the non-compacted myelinated axons (mediumarrowheads) and are unstained. Open circles denote nuclei of astrocytes, anda black square a capillary lumen. Scale bars: A = 25 µm, B = 10 µm.
318 F. Gallyas et al. / Biology of the Cell 96 (2004) 313–324
actions in the non-compacted state of this structure is higherthan that in the compacted state, the difference between theseenergy levels (henceforth: stored non-covalent free energy)can propel the compaction both in vivo and post mortem,provided that the following two energetic requirements aremet:
First, the non-compacted state of the structure responsiblefor the compaction must be metastable. Thus, enzyme-mediated energy is needed only for its building up, but not forits maintenance. Furthermore, some exogenous energy,called “activation energy”, is needed for the release of thestored energy. The fulfillment of this energetic requirement isplausible: without exogenous intervention, the neurons neverbecome compacted post mortem, even in the unfixed brain.
Second, in the non-compacted state, the structure respon-sible for the compaction must be built up in such a way thatthe stored non-covalent free energy, if released somehow atany point, inevitably serves as activation energy at the neigh-boring points (domino principle). If this requirement is ful-filled, the all-or-nothing nature of the compaction can beexplained as follows: Some intracerebral mechanical forceinduced by the head injury transmits activation energy toonly a single point in each somatodendritic domain or axonthat will become compacted (initiation phase). Thereafter,the structural transformation concomitant with the energy
release spreads over long intracellular distances, propagatedby the stored non-covalent free energy (spreading phase).
The scattered occurrence of the affected neuronal soma-dendrite domains or axon segments indicates that the prob-ability of initiation is very low, while the fact that wholesoma-dendrite domains or long axon segments are generallyaffected indicates that the process of spreading is relativelyrapid.
We suggested earlier (Gallyas et al., 1992) that the neu-rofilament network, which had been assumed to store tor-sional energy (Metuzals and Izzard, 1969; Gilbert, 1975), isresponsible for the compaction. However, the morphologicalsubstratum of the compaction cannot be any of the ultrastruc-tural elements visible in the conventional transmission elec-tron microscope (such as the neurofilaments), since intracel-lular spaces devoid of them (the dendritic spines, the interiorof the endoplasmic reticulum cisternae, or the cell nucleus)are also involved in the compaction. Conversely, this struc-ture must be present and continuous in all intracellular spacesinvolved in the compaction. We propose here that this struc-ture consists in a gel built up as outlined below.
Numerous non-covalent binding sites in numerous proteinmolecules that are present in the “aqueous cytoplasm” (allcytoplasmic spaces not occupied by the ultrastructural com-ponents “visible” in the conventional transmission electron
Fig. 6. (A) Electron-microscopic appearance of normal (N) and compacted (D) neuronal cell bodies in the brain cortex under the impact site following a postmortem head injury. White, thick arrows point to dilated Golgi cisternae; black circles denote markedly swollen astrocytic processes. In (B), which is amagnification of the area framed in (A), black or white thin arrows point to endoplasmic reticulum cisternae, black or white thin arrowheads to ribosome rosettes,and a thick arrow to a multivesicular body; white asterisks denote mitochondria, and black circles swollen astrocytic processes. Scale bars: A = 2 µm,B = 500 nm.
319F. Gallyas et al. / Biology of the Cell 96 (2004) 313–324
microscope; Clegg, 1984) are engaged not in maintaining thesecondary and tertiary structures of their native (in-vitro-active) state, but in mutually chaperoning their in vivo con-formation, anchoring the resulting protein matrix to the ultra-structural elements visible in the conventional transmissionelectron microscope, binding K+ (Ling and Ochsenfeld1973) and adsorbing water molecules in multiple orderedlayers (Ling and Walton, 1975). In this case, the whole-cellultrastructural compaction (“dark” cell formation) couldconsist in a gel-to-gel phase transition, i.e. in a co-operativeconformational change in the protein molecules building upthe matrix of this gel, which conformational change is ac-companied by the release of bound water molecules (com-paction) and K+ (hyperbasophilia). Namely, the gel-to-gelphase transition (for its relevant qualities see the Introduc-tion) can (i) fulfil the two energetic requirements discussedearlier, (ii) cause a dramatic volume reduction, (iii) be initi-ated by both chemical and physical injuries and (iv) bereversible (see in section 3.6.), like the whole-cell compac-tion.
In aldehyde-fixed specimens, the microtrabecular latticeobservable in the high-voltage stereo electron microscope(Porter, 1989) might be the post-fixation morphological
equivalent of the matrix of this gel. The absence of somato-dendritic or axonal compaction from the brains of rats head-injured 24 h after transcardial perfusion with an aldehydefixative can be explained as follows. By gradually formingchemical bridges between protein molecules, the aldehydefixation stabilizes the matrix of the presumed intracellular gelin 1 day at room temperature, but not in much shorter periodsof time.
3.4. Supporting arguments drawn from the post mortemdevelopment of hyperbasophilia and argyrophilia
Two all-or-nothing phenomena accompany the whole-cellcompaction both in vivo and post mortem: hyperbasophilia(Figs 1b and 5 b) and argyrophilia (Figs 1a and 5a). Baso-philia is the histochemical manifestation of structure-boundanionic groups that are capable of binding positively chargeddye molecules. In the normal neurons, only the Nissl gran-ules, the nucleoli and the rim of the nucleus (nucleic acids)are basophilic (see Fig. 1a and 5a). As a consequence of theultrastructural compaction, the affected neuronal somata,dendrites and axons homogeneously acquire excess baso-philia (become hyperbasophilic). Enzymatic digestion of
Fig. 7. (A) Electron-microscopic appearance of a longitudinally sectioned compacted neuronal dendrite in the brain cortex under the impact site following a postmortem head injury. Black arrows point to dendritic spines; black circles denote swollen astrocytic processes. (B) A normal (left side) and a compacted (rightside) dendrite in juxtaposition. Black or white arrowheads point to microtubules; black circles denote swollen astrocytic processes. Scale bars: A = 5 µm,B = 200 nm.
320 F. Gallyas et al. / Biology of the Cell 96 (2004) 313–324
nucleic acids cannot remove the excess basophilia, whereas itremoves completely the basophilia of the normal neurons(our unpublished observation). Consequently, hyperbaso-philia of the “dark” (compacted) neurons can not be thesimple consequence of the condensation of preexisting baso-philic substances. Furthermore, it appears improbable thatany enzyme-mediated process could form negatively chargedsubstances in excess under extremely unfavorable condi-tions. Conversely, as a result of compaction-derived confor-mational changes in the affected protein molecules, pre-existing structure-bound anionic groups can become
detached from the counter-ions (K+ ?), escape from a hydro-phobic environment or become more electronegative.
The silver staining method utilized demonstrates type-IIIargyrophilia. In its case, the formation of silver grains ismediated by tissue-related catalytic sites, which consist of afew side-groups of protein molecules in a favorable spatialarrangement (Gallyas, 1982). It appears improbable that anyenzyme-mediated process could form new proteins or pro-tein side-groups in excess under extremely unfavorable con-ditions. Conversely, catalytic sites may result fromcompaction-derived conformational changes in the affected
Fig. 8. (A) Electron-microscopic appearance of cross-sectioned normal (arrowhead) and compacted (arrow) myelinated (white m) axons in the brain cortexunder the impact site following a post mortem head injury. A black circle denotes a swollen astrocytic process. (B) and (C) are enlargements of the normal andthe compacted axons, respectively, demonstrated in (A). Black or white arrows point to microtubules, and black or white arrowheads to endoplasmic reticulumcisternae; white m denotes myelin sheaths. Neurofilaments appear as small dots. Scale bars: A = 500 nm, B and C = 200 nm.
321F. Gallyas et al. / Biology of the Cell 96 (2004) 313–324
protein molecules that lead to a favorable change in thespatial arrangement of their pre-existing side-groups.
3.5. Two previous conceptions of phase transition inthe aqueous cytoplasm
The association-induction theory of Ling (1962) postu-lates the existence of a physicochemical process, called “car-dinal adsorption”, which is in fact micro-compartmentalizedgel-to-gel phase transition. Ling assumes that the binding of“cardinal adsorbents”, such as ATP or Ca++, donates addi-tional free energy to an association of fully extended pro-teins, inorganic ions and water molecules. This induces aspreading change in protein conformation accompanied bythe selective adsorption of K+ and the large-scale polariza-tion of water molecules into multilayers. The reverse of thisprocess supplies free energy to various physiological pro-cesses.
The Pollack (2001) conception considers that gel-to-gelphase transition may be a “central mediator” of physiologicalcell functions, especially those in which chemical energy istransformed to physical work. The different physiologicalneeds are met by various types of phase transition, which areconfined to intracellular compartments of diverse size andgel composition.
Mammalian cells could undergo both the above physi-ological types and the currently reported pathological type ofgel-to-gel phase transition, provided that the hypothesizedcytoplasmic gel has at least three metastable phases repre-senting different levels of molecular organization. It shouldbe stressed that complex synthetic gels that possess such aproperty do exist (Annaka and Tanaka, 1992).
3.6. Presumed function of whole-cell gel-to-gel phasetransition in pathology
“Dark” cells are widely believed to be fated inevitably todie. However, as concerns “dark” neurons, strong evidencehas been presented in the past twenty years that, underfavorable circumstances, a varying proportion of them canspontaneously recover (reviewed by Csordás et al., 2003).Since the dying “dark” neurons are removed from an excito-toxic or a necrotic environment through a necrotic-like se-quence of ultrastructural changes (gross swelling and disin-tegration of the organelles and the gradual disappearance ofthe plasma and nuclear membranes), it has generally beenconcluded that their death is caused by necrosis (e.g. Ingvaret al., 1988). However, a recent publication (Csordás et al.,2003) demonstrates that, from an intact environment, “dark”neurons are removed through a sequence of ultrastructuralchanges similar to that for apoptotic neurons. Specifically:marked convolution of the nuclear and cellular outlines,separation of the pedunculated protuberances formingmembrane-bound bodies of compacted ultrastructure, whichare later engulfed by phagocytotic cells.
After completion of the programmed biochemical cascadeand segregation of the nuclear chromatin into sharply delin-
eated masses, apoptotic cells are known to undergo a suddencytoplasmic shrinkage (e.g. Wyllie, 1987). Ultrastructuralsimilarities between the shrunken cytoplasm of apoptoticneurons and that of the “dark” neurons produced in a non-compromised environment (our unpublished observation)suggest that their mechanisms of formation are the same. Ifthis is true, the formation of “dark” cells might be a malfunc-tion of the currently hypothesized metastable intracellulargel structure, the genuine function of which is to executewhole-cell ultrastructural compaction in an advanced mor-phological phase of apoptotic cell death. The capacity of“dark” neurons to recover indicates that at least the neuronsare “prepared” to attempt to remedy the damage caused bythis malfunction. It should be stressed that no idea of thefunction of “dark”-cell formation (the ultrastructural com-paction in non-apoptotic cells) has been suggested so far.
3.7. Concluding remarks
It remains the task of further investigations to establishhow this conception of the whole-cell ultrastructural com-paction can be harmonized with the prevailing conceptionsof the physical state of the living cell and concomitant fea-tures such as membrane pumps and ion channels. Evenwithin the realm of the current conception, tremendouscomplementary problems demanding a solution will emerge,such as the very low probability of initiation of whole-cellcompaction or its stop at the axon hillock in the case ofneurons. Regardless of whether or not the present conceptionholds true, some overall cell-biological phenomenon is lyinghidden behind whole-cell ultrastructural compaction, andthis must not be passed over in silence.
4. Materials and methods
4.1. Animal experiments
A total of 29 Sprague-Dawley rats weighing about 200 gwere anaesthetized with a mixture of 4% isoflurane, 29%oxygen and 67% nitrous oxide, using a SurgiVet respirator(Waukesha, WI, USA). Each of them was scalped and a4-mm-thick metal disc 10 mm in diameter was cemented tothe exposed skull vault between the coronal suture and lamb-doid. Seven rats were placed in a prone position on a 12-cm-thick foam bed under an impact-acceleration head injurydevice (Marmarou et al., 1994). An intracerebral mechanicalforce was generated by dropping a brass rod with a mass of450 g onto the metal disc from a height of 100 cm through avertically positioned Plexiglas tube. Two of these rats suf-fered impression fracture of the calvaria and were excludedfrom further investigation (for the reason, see the Discus-sion). Immediately after the head injury, the 5 rats with anon-fractured calvaria were perfused transcardially at roomtemperature, first with 50 ml of physiological saline and then,for 30 min. with 500 ml of an aldehyde fixative (100 ml of
322 F. Gallyas et al. / Biology of the Cell 96 (2004) 313–324
20% paraformaldehyde, 100 ml of 25% glutaraldehyde,500 ml of 0.2 mol/l sodium cacodylate, and 300 ml ofdistilled water were mixed, and the resulting solution wasadjusted to pH 7.4 with a few drops of concentrated hydro-chloric acid).
Without any previous manipulation, each of the remaining22 rats was perfused with the above fixative for 30 min.Thereafter, the bead probe of a digital thermometer waspositioned deep in the rectum and the skin over the abdomenand thorax was tightly sutured. Finally, such rats were im-mersed into icy water until the rectal temperature fell to justabove the freezing point (about 90 min.). Twelve of themwere head-injured as described above. Seven of the head-injured rats suffered impression fracture of the calvaria, andwere excluded from the further investigation (for the reason,see the Discussion). The remaining 5 head-injured rats and10 rats without head injury were allowed to warm up to roomtemperature. In 5 of the latter rat group, the above head injurywas provoked 24 h later. None of them suffered calvariafracture. The non-traumatized 5 rats served as controls. Apartfrom the above manipulations, all rats were left untouched atroom temperature for 24 h after the start of fixation. There-after, the brains were removed from the skull and immersedin the fixative.
Animal care and handling were carried out in accordancewith order 243/1998 of the Hungarian Government, which isan adaptation of directive 86/609/EGK of the EuropeanCommittee Council.
4.2. Tissue processing and staining
Vibratome sections 150 µm in thickness were cut coro-nally from the caudal two-thirds of the cerebrum. Every fifthvibratome section was stained by a special silver technique,as described previously (Gallyas et al., 1990). Briefly, fol-lowing dehydration with graded 1-propanol, vibratome sec-tions were incubated for 16 h at 56 °C in 1-propanol contain-ing 0.8% sulfuric acid and 2% water (esterification),rehydrated with graded 1-propanol, treated with 1% aceticacid for 10 min. and then immersed in a special physicaldeveloper until the background had become light-brown.
For electron microscopy, 2x2-mm2 areas of the neocortexand the subcortical white matter were cut from the sectionsadjacent to those displaying the highest number of silver-stained neuronal parts. These specimens were postfixed witha 1:1 mixture of 2% osmium tetroxide and 3% potassiumhexocyanoferrate(II) for 1 h at room temperature, and flat-embedded in Durcupan ACM. Semithin (1-µm) sectionswere stained in a solution containing 0.05% toluidine blue,0.05 % sodium tetraborate and 0.1% saccharose (pH 9.5) for1 min. at 90 °C. Thin (50-nm) sections were contrasted with5% uranyl acetate in 50% methanol for 2 min. and then 0.5%lead citrate for 1 min. Ultrastructural investigations werecarried out with a Jeol JEM 1200EX transmission electronmicroscope.
4.3. Quantitative analysis
From each of the 5 in vivo and the 5 post mortem head-injured rats, cross-sections of 10 normal-looking and10 compacted axons were photographed at a magnification of25000, digitized at 1200 dpi with an Epson GT 9600 scannerand magnified electronically to 100000 times their actualsize. Thereafter, the longitudinal ultrastructural elements(neurofilaments and microtubules) were counted in them.From each 10-member group, 2 axons that contained longi-tudinal elements in similar numbers (between 195 and 216)were selected. In these axons, the area inside the axolemmabut outside the mitochondria and endoplasmic reticulumcisternae was measured, and then divided by 1000002, andfinally by the relevant number of longitudinal elements. Thequotients relating to each of the 4 axon groups of interest(normal axons, in vivo head injury; compacted axons, in vivohead injury, normal axons, post mortem head injury; com-pacted axons, post mortem head injury) were next averaged.These averages, which represent the cross section areas thatrelate to a single longitudinal ultrastructural element, werecompared using the Student t-test.
Acknowledgements
We thank Andok Csabáné, Nyírádi József and Nádor An-drásné for technical assistance. Supported by Hungariangrants from OTKA (T-029352) and from NKFP (1/A00026/2002).
References
Annaka, M., Tanaka, T., 1992. Multiple phases of polymer gels. Nature 355,430–432.
Auer, R.N., Wieloch, T., Olsson, Y., Siesjö, B.K., 1984. The distribution ofhypoglycemic brain damage. Acta Neuropathol. 64, 177–191.
Brown, A.W., 1977. Structural abnormalities in neurons. J. Clin. Pathol. 30(Suppl), 155–169 (Royal College of Pathologists) 11.
Buki, A., Siman, R., Trojanowsky, J.Q., Povlishock, J.T., 1999. The role ofcalpain mediated spectrin proteolysis in traumatically induced axonalinjury. J. Neuropathol. Exp. Neurol. 58, 365–375.
Cammermeyer, J., 1960. The post mortem origin and mechanism of neu-ronal hyperchromatosis and nuclear pycnosis. Exp. Neurol. 2, 379–405.
Cammermeyer, J., 1961. The importance of avoiding “dark” neurones inexperimental neuropathology. Acta Neuropathol. 1, 245–270.
Clegg, J.S., 1984. Properties and metabolism of the aqueous cytoplasm andits boundaries. Am. J. Physiol. 246, R133–R151.
Csordás, A., Mázló, M., Gallyas, F., 2003. Recovery versus death of “dark”(compacted) neurones in non-impaired parenchimal environment. Lightand electron microscopic observations. Acta Neuropathol. 106, 37–49.
Foda, M.M.A., Marmarou, A., 1994. A new model of diffuse brain injury inrats. Part II: Morphological characterization. J. Neurosurg 80, 301–313.
Gallyas, F., 1982. Physico-chemical mechanism of the argyrophil III reac-tion. Histochemistry 74, 409–421.
Gallyas, F., Güldner, F.H., Zoltay, G., Wolff, J.R., 1990. Golgi-like demon-stration of “dark” neurones with an argyrophil III method for experimen-tal neuropathology. Acta Neuropathol. 79, 620–628.
323F. Gallyas et al. / Biology of the Cell 96 (2004) 313–324
Gallyas, F., Zoltay, G., Dames, W., 1992. Formation of “dark” (argyrophilic)neurones of various origin proceeds with a common mechanism ofbiophysical nature. Acta Neuropathol. 83, 504–509.
Gallyas, F., Farkas, O., Mázló, M., 2002. Traumatic compaction of theaxonal cytoskeleton induces argyrophilia: histological and theoreticalimportance. Acta Neuropathol. 103, 36–42.
Gilbert, D., 1975. Axoplasm architecture as seen in the Myxicola giant axon.J. Physiol 253, 257–301.
Harmon, B.V., 1987. An ultrastructural study of spontaneous cell death inmouse mastocytoma with particular reference to dark cells. J. Pathol 153,345–355.
Ingvar, M., Morgan, P.F., Auer, R.N., 1988. The nature and timing ofexcitotoxic neuronal necrosis in the cerebral cortex, hippocampus andthalamus due to flurothyl-induced status epilepticus. Acta Neuropathol.75, 362–369.
Kiernan, J.A., Macpherson, C.M., Price, A., Sun, T., 1998. A histochemicalexamination of the staining of kainate-induced neuronal degeneration byanionic dyes. Biotechnic & Histochemistry 73, 244–254.
Ling, G.N., 1962. A Physical Theory of the Living State: The Association-Induction Hypothesis. Blaisdell, Waltham, pp. 1–68.
Ling, G.N., Ochsenfeld, M.M., 1973. Mobility of potassium ion in frogmuscle cells, both living and dead. Science 181, 78–81.
Ling, G.N., Walton, C.L., 1975. What retains water in living cells. Science191, 293–295.
Marmarou, A., Foda, M.A., Brink, W., Chambell, J., Khita, H., Demetri-adu, K., 1994. A new model of diffuse brain injury in rats. Part I:Pathophysiology and biomechanics. J. Neurosurg 80, 291–300.
Metuzals, J., Izzard, C.S., 1969. Spatial patterns of threadlike elements inthe axoplasm of the giant nerve fibre of the squid (Loligo pealei L) asdisclosed by differential interference microscopy and by electronmicroscopy. J. Cell Biol. 43, 456–479.
Newman, G.R., Jasani, A.B., 1998. Silver development in microscopy andbioanalysis: past and present. J. Pathol 186, 119–125.
Pettus, E.H., Christman, C.W., Giebel, M.L., Povlishock, J.T., 1994. Trau-matically induced altered membrane permeability: its relationship totraumatically induced reactive axonal change. J. Neurotrauma 11, 507–552.
Pollack, G.H., 1996. Phase transitions and the molecular mechanism ofcontraction. Biophys. Chem 59, 315 328.
Pollack, G.H., 2001. Cells, Gels and the Engines of Life. Ebner & Sons,Seattle, pp. 1–298.
Porter, K.R., 1989. The cytoplasm and its matrix. Prog. Clin. Biol. Res. 295,15–20.
Tanaka, T., Annaka, M., Ilmain, F., Ishii, K., Kokufuta, E., Suzuki, A.,Tokita, M., 1992. Phase transitions of gels. In: Karalis, T.K. (Ed.),Mechanics of Swelling, H64. Springer, Berlin, pp. 683–703 NATO ASISeries.
Tasaki, I., 1999. Rapid structural changes in nerve fibers and cells associatedwith their excitation processes. (Review). Jpn. J. Physiol. 49, 125–138.
Tóth, Z.S., Seress, L., Tóth, P., Ribak, C.E., Gallyas, F., 1997. A commonmorphopathological response of astrocytes to various injuries: “dark”astrocytes: A light and electron microscopic analysis. J. Brain Res. 38,173–186.
Verdugo, P., Deyrup-Olsen, I., Martin, A.W., Luchtel, D.L., 1992. Polymergel phase transition: the molecular mechanism of product release inmucin secretion. In: Karalis, T.K. (Ed.), Mechanics of Swelling, H64.Springer, Berlin, pp. 671–681 NATO ASI Series.
Wyllie, A.H., 1987. Apoptosis: Cell death in tissue regulation. J. Pathol. 153,313–316.
324 F. Gallyas et al. / Biology of the Cell 96 (2004) 313–324
JOURNAL OF NEUROTRAUMAVolume 20, Number 3, 2003© Mary Ann Liebert, Inc.
Preinjury Administration of the Calpain Inhibitor MDL-28170Attenuates Traumatically Induced Axonal Injury
A. BUKI,1 O. FARKAS,1 T. DOCZI,1 and J.T. POVLISHOCK2
ABSTRACT
Traumatic brain injury (TBI) evokes diffuse (traumatic) axonal injury (TAI), which contributes tomorbidity and mortality. Damaged axons display progressive alterations gradually evolving to ax-onal disconnection. In severe TAI, the tensile forces of injury lead to a focal influx of Ca21, initiat-ing a series of proteolytic processes wherein the cysteine proteases, calpain and caspase modify theaxonal cytoskeleton, causing irreversible damage over time postinjury. Although several studies havedemonstrated that the systemic administration of calpain inhibitors reduces the extent of ischemicand traumatic contusional injury a direct beneficial effect on TAI has not been established to date.The current study was initiated to address this issue in an impact acceleration rat-TBI model in or-der to provide further evidence on the contribution of calpain-mediated proteolytic processes in thepathogenesis of TAI, while further supporting the utility of calpain-inhibitors. A single tail vein bo-lus injection of 30 mg/kg MDL-28170 was administered to Wistar rats 30 min preinjury. After in-jury the rats were allowed to survive 120 min when they were perfused with aldehydes. Brains wereprocessed for immunohistochemical localization of damaged axonal profiles displaying either amy-loid precursor protein (APP)– or RMO-14–immunoreactivity (IR), both considered markers of spe-cific features of TAI. Digital data acquisition and statistical analysis demonstrated that preinjuryadministration of MDL-28170 significantly reduced the mean number of damaged RMO-14– as wellas APP-IR axonal profiles in the brainstem fiber tracts analyzed. These results further underscorethe role of calpain-mediated proteolytic processes in the pathogenesis of DAI and support the po-tential use of cell permeable calpain-inhibitors as a rational therapeutic approach in TBI.
Key words: axonal injury; calpain; MDL-28170; spectrin; traumatic brain injury
261
1Department of Neurosurgery, Medical Faculty of Pécs University, Pécs, Hungary.2Department of Anatomy and Neurobiology, Medical College of Virginia Campus of Virginia Commonwealth University,
Richmond, Virginia.
INTRODUCTION
TRAUMATIC BRAIN INJURY (TBI) has long been associ-ated with the generation of traumatic axonal injury
(TAI) both in animals and humans (Gennarelli et al.,1982; Blumbergs et al., 1989; Adams et al., 1977, 1980;Strich, 1956). In the last two decades, detailed investi-
gation of the pathobiology of traumatic axonal injury hasrevealed that the majority of traumatically injured axonsare not mechanically severed at the time of impact, in-stead showing progressive changes gradually evolving toaxonal disconnection (Povlishock et al., 1983, 1992;Povlishock, 1992; Maxwell et al., 1997; Meaney et al.,2000). Typically, this process of progressive axonal
change evolves over a several hour period, leading to anupstream swelling due to an impairment of axonal trans-port (Povlishock et al., 1983), although it is now recog-nized that not all injured axons go on to swell (Stone etal., 2000). The precise subcellular mechanisms responsi-ble for initiating some, if not all of these progressivechanges, have also been unmasked over the last severalyears. In vitro studies by Wolf and colleagues (2001)have shown that the traumatic event is associated withintraaxonal calcium influx, which was linked to trau-matically induced depolarization, activation of voltagesensitive calcium channels and the concomitant activa-tion of sodium/calcium exchangers. In the in vivo setting,the involvement of calcium in this progressive pathobi-ology has also been confirmed through several indirectapproaches. Maxwell and colleagues (1995) showed sig-nificant intraaxonal calcium increases through the use ofpyroantimonate in the injured optic nerve. Our own groupdemonstrated, in foci of traumatic axonal injury, alter-ations in axolemmal permeability that were directly cor-related with cytoskeletal and mitochondrial damage, afinding consistent with local calcium overloading, oc-curring as a result of alterated axolemmal permeability(Pettus et al., 1994; Pettus and Povlishock, 1996). Whenfirst observed, it was assumed that this altered axolem-mal permeability and presumed calcium influx wouldcause catastrophic and ultra-rapid degradation of the in-traaxonal cytoskeleton. However, continued work did notconfirm this. Detailed studies conducted by Büki and col-leagues (1999a) examining calcium-mediated pathologyin terms of calcium-mediated protease activation demon-strated a compartmentalized and progressive series ofchanges that led to axonal disconnection. Specifically,through the use of antibodies targeting calpain-mediatedspectrin-proteolysis Büki and colleagues (1999a) recog-nized, in the early post-traumatic period, that the calpain-mediated processes were confined to the sub-axolemmaldomain, progressing over time to involve the entire ex-tent of the axon diameter, thereby causing the progres-sive degradation of the axon cylinder. These studies con-firmed and extended earlier studies by Saatman andcolleagues (1996b), Newcomb and collaborators (1997)and McCracken et al. (1999) which also had alluded, inpart, to the occurrence of calpain-mediated axonal dam-age occurring together with other forms of TBI-inducedpathology. Because of these observed progressive patho-logical changes involving calpain-mediated damage, itappeared that therapeutic strategies targeting calcium andits downstream targets such as calpain would constituterational approaches to blunt the progression of axonaldamage, thereby translating into improved outcome. Thecorrectness of this assumption was supported, in part, byfindings employing hypothermia, wherein the use of post-
traumatic hypothermia followed by slow rewarming wasshown to translate into significant axonal protectionwhich was directly correlated with the attenuation ofmarkers targeting calpain- mediated proteolysis. (Koizumiand Povlishock, 1998; Buki et al., 1999b). The potentialusefulness of inhibitors of calpain-mediated proteolysisis not a new concept and, in fact, this approach has beenused in the study of ischemic brain injury as well as trau-matic brain injury paradigms focusing primarily uponcontusional change (Saatman et al., 1996a, 2000; Pos-mantur, 1997). In studies of cerebral ischemia, Bartus andcolleagues (1994) have reported that the systemic ad-ministration of calpain inhibitors reduced the overall de-gree of the ischemic alteration, with local cytoskeletalprotection also reported in a contusional model of TBI(Posmanter et al., 1997). Saatman and colleagues (1996a)also demonstrated, in the context of traumatic brain in-jury, that the use of calpain inhibitors translated into im-proved behavioral outcome. However, this occurred with-out concomitant neuroprotection in terms of reducedcontusional size or a blunting of apoptotic neuronal celldeath (Saatman et al., 2000). Unfortunately, as Saatmanand colleagues did not examine any potential reductionin the degree of axonal damage, the possibility remainedthat the observed behavioral protection resided in con-comitant axonal sparing. In that we have already con-firmed the presence of calpain-mediated axonal damagein well-characterized model of traumatic brain injury, itseemed reasonable and rational to revisit our animal mod-els, now using inhibitors of calpain-mediated damage todetermine if they significantly attenuate axonal damage.To this end, we used a rodent model of traumatic braininjury in concert with the administration of a cell-perme-able peptidyl-aldehyde calpain inhibitor (MDL-28170) todetermine whether this drug attenuates traumatically in-duced axonal damage. This approach was also initiated toprovide further evidence of the physical role of calpain-mediated damage in the pathogenesis of TAI while set-ting the stage for additional therapeutic studies on the util-ity of calpain inhibitors for the treatment of TAI.
MATERIALS AND METHODS
Drug Administration and Induction of TraumaticBrain Injury
In this study we utilized the well-characterized rodentmodel of impact acceleration head injury described in de-tail elsewhere (Marmarou et al., 1994; Foda and Mar-marou, 1994). In all, 12 Wistar rats (Charles River Hun-gary) weighing 385–405 g were used for the experiments;three animals served as sham-injured controls.
For the induction of anesthesia, the animals were ex-
BUKI ET AL.
262
posed to 4% isoflurane (Forane, Abbott, Hungary) in abell jar for 5 min, and then intubated and ventilated witha mixture of 1–2% isoflurane in 30% O2 and 70% N2O.
A single tail vein bolus injection of 30 mg/kg MDL-28170 (courtesy of Aventis Pharmaceuticals) dissolvedin 1 mL of the vehicle (PEG300/EtOH, 9:1) was admin-istered 30 min prior to injury (n 5 4). The dosage androute of administration were selected based on pharma-cokinetic studies described by Markgraf and colleagues(1998). Other rats were treated with a single tail vein bo-lus injection of the vehicle alone (n 5 5).
Next, the skull between the coronal and lambdoid su-tures was exposed with a midline incision. A metallicdisc-shaped helmet of 10mm diameter was firmly gluedto this point on the skull. The animal was placed in aprone position on a foam bed with the metallic helmetcentered under the edge of a Plexiglas tube. The rat wasprevented from falling by two belts secured to the foampad. Brass weights weighing 450 g were allowed to fallfrom a height of 200 cm through the Plexiglas tube di-rectly onto the metallic disc fixed to the animal’s skull—a setting that precludes cerebral contusion or subduralhemorrhage. Immediately after injury the animal wasventilated with 100% O2. The helmet was removed andthe skull was investigated for any sign of fracture which,if found, disqualified the animal from further evaluation.The scalp wound was sutured, with the animal remain-ing on artificial ventilation until the pre-determined sur-vival of 120 min after injury. This relatively brief sur-vival time was employed to optimize the usefulness ofthe antibodies described below, particularly those target-ing the altered rod domains of the NFM subunits whichyield the potential for false positives with increased sur-vival (Povlishock et al. 1997).
Physiological Assessments
The respiratory status was monitored through pulseoximetry via the footpad and/or the ear. Additionally,brain temperature was monitored by a temporalis muscleprobe, while core temperature was determined by a rec-tal probe.
Immunohistochemistry
At the designated survival time, the rats were re-anes-thetized with an overdose of sodium pentobarbital andtranscardially perfused with 4% paraformaldehyde inMillonig’s buffer. Brains were removed and immersed inthe same fixative overnight (16–18 h). On the basis ofpreviously published observations concerning the topog-raphy of diffusely injured axons in rat brain (Koizumiand Povlishock, 1998; Okonkwo et al., 1999; Okonkwoand Povlishock, 1999), a midline, 5-min-wide block of
the brain was removed using a sagittal brain blocking de-vice to include the region extending from the interpe-duncular fossa to the first cervical segment (World Precision Instruments, Sarasota, FL). The tissue was sec-tioned with Vibratome Series 1000 (Polysciences Inc.,Warrington, PA) at a thickness of 40 mm and collectedin 0.1 M phosphate buffer. The sections were collectedin two groups in a semi-serial fashion, rinsed three timesten minutes in phosphate buffered saline (PBS), andprocessed for immunohistochemical localization of dam-aged axonal profiles. Half of the sections were single la-beled for the detection of RMO-14 immunoreactivity.This antibody is known to exclusively target an epitopeon the rod domains of altered NFM subunits, that are ex-posed upon modification of the NF sidearms, an assumedconsequence of calcium induced enzymatic processes(Lee et al., 1987; Povlishock et al., 1997; Buki et al.,1999a; Okonkwo et al., 1998). Every second section wasprocessed for immunohistochemical detection of theamyloid precursor protein (APP). This classical markerof TAI is carried by fast axonal transport and pools atfoci affected by TAI (Gentleman et al., 1993; Sherriff etal., 1994; Stone et al., 1999). In our experiments we uti-lized a polyclonal antibody targeted to the C-terminus ofthe b-amyloid precursor protein. Characterization and de-tailed description of the advantages of this antibody wererecently published by Stone et al. (2000). After washingin PBS, sections from both groups were incubated for 35min with 0.2% Triton X (Sigma Chemical Co., St. Louis,MO [products of Sigma Chemical Co. were purchasedvia their Hungarian representative, Sigma–Aldrich, Hun-gary, Budapest]) in 10% normal goat or horse serum, re-spectively (NGS or NHS) (Sigma) in PBS. After twoquick rinses in PBS containing 1% NGS (or NHS in thecase of RMO14) the above-defined groups of sectionswere incubated overnight in rabbit anti-APP antibody(Zymed Laboratories, CT) at a dilution of 1:3000 or inmouse monoclonal RMO14 antibody at a dilution of1:500 (kindly provided by Dr. John Q. Trojanowski, Uni-versity of Pennsylvania Department of Pathology). After3 3 10 min washes in PBS containing 1% NGS (or NHS)sections were incubated in biotinylated anti-rabbit im-munoglobulin derived from goat (diluted 1:200 in 1%NGS/PBS; Vector, Burlingame, CA [products of Vectorwere purchased via their Hungarian representative, Bio-marker Ltd., Godollo]) or in 1:400 dilution of biotiny-lated, rat adsorbed anti-mouse immunoglobulin derivedfrom horse (Vector) for 60 min followed by 3 3 10 minrinses in PBS. After incubation in an avidin biotin–per-oxidase complex (ABC standard “Elite” kit, Vector,Burlingame, CA, at a dilution of 1:100) and rinsing inPBS and 0.1 M phosphate buffer 3 3 10 min and 2 3 10min, respectively, sections were processed for visualiza-
CALPAIN INHIBITOR PREVENTS TRAUMATIC AXONAL INJURY
263
tion of the immunohistochemical complex using 0.05%diaminobenzidine (DAB; Sigma) and 0.01% hydrogenperoxide in 0.1 M phosphate buffer. Next the sectionswere mounted and cleared for routine light microscopicexamination.
Immunohistochemical Controls
The employed antibodies have been well characterizedand extensively used. However, control studies were per-formed to further assure specificity. To this end all im-munocytochemical incubations included sections incu-bated without the addition of the primary or secondaryantibodies.
Image Analysis
Semi-serial brainstem sections (six from each animal)were examined with a NIKON light microscope inter-faced with a computer-assisted image analysis system(NIH 1.55) in a blinded fashion. At the pontomedullaryjunction, two adjacent grids of 40,000 mm2 were super-imposed over the corticospinal tracts (CSpT), imageswere captured and digitized at a magnification of 503
and the total number of damaged APP or RMO14 im-munopositive axonal profiles within this area weremarked and counted. In the same brainstem sections anadjacent image of the medial longitudinal fasciculus(MLF) was captured and digitized at 253, a 160,000 mm2
grid was superimposed over the region of the MLF andthe number of damaged axonal profiles within the grid
were counted and expressed as density of APP or RMO-14–immunopositive axons. For the purpose of imageanalyses each immunoreactive section was captured in ablinded fashion, with each section scanned to identify theplane of maximal immunoreactivity for counting. Onlystrongly positive immunoreactive profiles were includedin the data analyses.
Statistical Analysis
For each antibody and region twelve fields of theCSpT, and six of the MLF where counted in each animaland the mean number of immunopositive axons was com-puted as number of immunopositive axons per mm2. Themean number of immunopositive axons in both regionsin vehicle treated animals (n 5 5) was compared withmean density in MDL-28170 treated rats (n 5 4) with anindependent samples t-test. Differences were assumedsignificant at a level of p # 0.05.
RESULTS
In accordance with previous observations from our lab-oratories and others the use of the above-described ex-perimental protocols did not result in any significant al-teration in the physiological parameters monitored.
Light microscopic examination of vehicle and drugtreated animals subjected to TBI and reacted for the vi-sualization of the APP and RMO-14 antibodies showeddiscrete immunocytochemical reaction products within
BUKI ET AL.
264
FIG. 1. Traumatically injured axons displaying APP immunoreactivity indicating axonal injury (disturbed axonal transport) inlight micrographs of thecorticospinal tract (A,B) and medial longitudinal fasciculus (C,D) from injured animals. Note that thenumber of immunoreactive damaged axons appears dramatically reduced in the MDL-28170–treated (B,D) compared to the ve-hicle-treated (A,C) sections. (Bar indicates 50 mm.)
scattered axons localized in the CSPT and MLF. Mor-phological characteristics of the swollen, occasionallydisconnected APP-IR-axonal segments and the lobu-lated/vacuolated, partially or totally disconnected RMO-14 IR-axonal segments were entirely consistent with pre-vious descriptions of damaged axonal profiles 2 hpost-injury (Stone et al., 1999; Buki et al., 1999a,c). Thesham-injured animals displayed no immunoreactive ax-onal profiles.
In the MDL-28170–treated group (Figs. 1B,D and2B,D), both the localization and the appearance of APPand the RMO-14 immunopositive axonal profiles weresimilar to that observed in the vehicle treated group (Figs.1A,C and 2A,C); however, now the number of the dam-aged, immunopositive axonal profiles appeared dramati-cally decreased.
Digital acquisition followed by statistical analysis con-firmed that preinjury administration of MDL-28170 sig-nificantly reduced the mean number of damaged RMO-14-IR axonal profiles in both the corticospinal tract(CSpT) (from 1864.2 6 200.7RMO-IR (mean 6 SEM,/mm2) to 1155.0 6 54.7RMO-IR (t 5 3.05, df 5 7, p ,
0.02) and the medial longitudinal fasciculus (MLF; from109.7 6 16.8RMO-IR to 52.5 6 9.7RMO-IR (t 5 2.74, df 5
7, p , 0.03); Figs. 3 and 4). Similarly, pre-injury ad-ministration of MDL-28170 significantly reduced themean numbers of damaged APP-IR axonal profiles bothin the brainstem in the CSpT (from 1771.6 6 278.5APP-
IR to 757.5 6 194.3APP-IR (t 5 2.82, df 5 7, p , 0.03))
and at the MLF (from 85.7 6 7.1APP-IR to 50.8 6 1.6APP-
IR (t 5 4.25, df 5 7, p , 0.01); Figs. 3 and 4).
DISCUSSION
In the present study, we have demonstrated that pre-injury administration of cell permeable calpain-inhibitorMDL-28170 is capable of significantly reducing trau-matically evoked axonal injury as assessed by the im-munocytochemical markers APP and RMO-14, both ofwhich target specific features of TAI. Not only do thesestudies strongly support the potential utility of calpain-inhibitors in blunting the progression of traumatically in-duced axonal injury, but also they provide further evi-dence of the role of calcium-induced, calpain-mediatedstructural proteolysis in traumatic brain injury as previ-ously suggested through multiple lines of evidence in an-imals (Pike et al., 2001; Kampfl et al., 1997; Saatman etal., 1996b), and in man (McCracken et al., 1999). Whilethe present studies did not utilize parallel ultrastructuralanalyses to define the precise sub-cellular targets of thechosen calpain inhibitor, the study’s use of two differentantibodies targeting specific features of traumatic braininjury provides insight to its potential mode of action. Asit is recognized that RMO-14 immunoreactivity is asso-ciated with neurofilament compaction secondary tocleavage of the neurofilament sidearm (Povlishock et al.,1997), the observed reduction in the immunreactive prod-
CALPAIN INHIBITOR PREVENTS TRAUMATIC AXONAL INJURY
265
FIG. 2. Traumatically injured axons displaying RMO-14 immunoreactivity indicating axonal injury (neurofilament-compaction)in light micrographs of corticospinal tract (A,B) and medial longitudinal fasciculus (C,D) from injured animals. Note that thenumber of immunoreactive damaged axons appears dramatically reduced in the MDL-28170–treated (B,D) compared to the ve-hicle-treated (A,C) sections. (Bar indicates 50 mm.)
uct clearly speaks to the utility of the chosen calpain in-hibitor in blunting the calpain-mediated proteolytic cleav-age of the neurofilament sidearms. Similarly, the overallreduction in the APP immunreactive axonal swellingsmost likely speaks to a further attenuation of the calpain-mediated processes. As alluded to previously, the trau-matically induced intraaxonal rise in intraaxonal calcium,coupled with the increase in calpain proteolytic activity,contributes to microtubular loss and dispersion, disrupt-ing axonal transport kinetics, with the net effect of ax-onal transport impairment (Pettus and Povlishock, 1996).This leads to upstream intraaxonal swelling with the ac-cumulation of anterogradely transported products such asAPP. The attenuation of this APP accumulation via useof a calpain inhibitor speaks to protective effect that thisagent has in preventing calpain-linked microtubular dis-assembly and alterations in axonal transport. Until re-cently, it was assumed that the above described markersAPP and RMO-14, were co-localized in the same axonalpopulation. However, it is now understood that, in somecases, these markers can co-localize whereas in othercases, they may identify different populations of axons,with the RMO-14 fibers identifying more severely in-jured axons, with the APP immunreactivity confined tothose axons sustaining less severe insult (Stone et al.,2001). While in the current communication, we havemade no direct RMO-14/APP colocalization, theirchange parallels that described in our recent publication
(Stone et al., 2001), thereby confirming the utility of thecalpain-inhibitors against a broad spectrum of traumaticaxonal injury.
The results of the current communication are novel inthe context of traumatically induced axonal injury, yet,they do join a relatively large volume of literature sug-gesting the utility of calpain inhibitors in various centralnervous system disorders. Calpain-inhibitors have provenefficacious in reducing brain ischemic injury (Markgrafet al., 1998; Bartus et al., 1994; Hong et al., 1994; Leeet al., 1997) and spinal cord injury (Banik et al., 1998).The studies reported herein also extend other work con-ducted in traumatic brain injury wherein the use of cal-pain-inhibitors was observed to improve behavioral out-come although it did not exert neuroprotection in termsof contusional volume reduction and/or apoptotic change.As noted in the introduction to this manuscript, the cur-rently observed axonal protection may constitute a ma-jor substrate of behavioral protection described in theseoriginal studies of Saatman and colleagues (1996a; 2000),a point which requires further study.
The major limitation of the current investigation re-sides in the fact that the results rely on the use of a sin-gle pre-injury injection strategy, which was mandated bythe limited availability of the chosen drug. Ideally, todemonstrate potential pre-clinical relevance, it wouldhave been of interest to observe the axonal protection af-forded by calpain-inhibitor delivered in a delayed, post-injury injection schedule. Based upon our published work
BUKI ET AL.
266
FIG. 3. Bar graph of the mean numbers of immunopositiveaxons displaying APP and RMO-14 immunoreactivity in thecorticospinal tracts (CSpT) of vehicle and MDL-28170–treatedanimals subjected to impact acceleration injury. Black lines rep-resent standard error of mean, asterisks indicate a statisticallysignificant difference.
FIG. 4. Bar graph of the mean numbers of immunopositiveaxons displaying APP and RMO-14 immunoreactivity in themedial longitudinal fasciculus (MLF) of vehicle and MDL-28170–treated animals subjected to impact acceleration injury.Black lines represent standard error of mean, asterisks indicatestatistically significant difference.
looking at calpain-mediated spectrin proteolysis (Buki etal., 1999a), we would assume that the window of oppor-tunity for maximal protection would reside within thefirst hour or two postinjury, however, these issues requirefurther study.
Another issue that must be addressed concerns the factthat while compelling, the present study does not providedirect evidence that calpain inhibition attenuates TAI. Itis known that MDL 28170 is not solely calpain specific.Further, as the antibodies used did not specifically rec-ognize axonal breakdown products generated by calpainactivation, some caution must be exercised to precludeover interpretation of the data.
ACKNOWLEDGMENTS
We thank Aventis Pharmaceuticals for the kind dona-tion of the MDL-28170, John Trojanowski for the RMO-14 antibody, and Carrie G. Markgraf for her excellenttechnical advices. We also thank Csabáné Andok, An-drásné Nádor, and József Nyirádi for their technical as-sistance as well as Carolyn Davis and Susan Walker fortheir continuous support. Grants of the Hungarian Sci-ence Foundation (OTKA T 035195, ETT), Bólyai Schol-arship of the Hungarian Academy of Sciences and theFogarty International Research Collaboration Award (1-RO3-TW0131302A1) and NIH Grant NS20193 sup-ported this work.
REFERENCES
ADAMS, H., MITCHELL, D.E., GRAHAM, D.I., et al. (1977).Diffuse brain damage of immediate impact type. Its rela-tionship to “primary brain-stem damage” in head injury.Brain 100, 489–502.
ADAMS, J.H., GRAHAM, D.I., SCOTT, G., et al. (1980).Brain damage in fatal non-missile head injury. J. Clin. Pathol.33, 1132–1145.
BANIK, N.L., SHIELDS, D.C., RAY, S., et al. (1998). Role ofcalpain in spinal cord injury: effects of calpain and free rad-ical inhibitors. Ann. N.Y. Acad. Sci. 844, 131–7.
BARTUS, R.T., HAYWARD, N.J., ELLIOTT, P.J., et al.(1994). Calpain inhibitor AK295 protects neurons from fo-cal brain ischemia. Effects of postocclusion intra-arterial ad-ministration. Stroke 25, 2265–2270.
BLUMBERGS, P.C., JONES, N.R., and NORTH, J.B. (1989).Diffuse axonal injury in head trauma. J. Neurol. Neurosurg.Psychiatry 52, 838–841.
BUKI, A., KOIZUMI, H., and POVLISHOCK, J.T. (1999b).Moderate posttraumatic hypothermia decreases early calpain-
mediated proteolysis and concomitant cytoskeletal compro-mise in traumatic axonal injury. Exp. Neurol. 159, 319–328.
BUKI, A., OKONKWO, D.O., and POVLISHOCK, J.T.(1999c). Postinjury cyclosporin A administration limits ax-onal damage and disconnection in traumatic brain injury. J.Neurotrauma 16, 511–521.
BUKI, A., OKONKWO, D.O., WANG, K.K., et al. (2000). Cy-tochrome c release and caspase activation in traumatic ax-onal injury. J. Neurosci. 20, 2825–2834.
BUKI, A., SIMAN, R., TROJANOWSKI, J.Q., et al. (1999a).The role of calpain-mediated spectrin proteolysis in trau-matically induced axonal injury. J. Neuropathol. Exp. Neu-rol. 58, 365–375.
FODA, M.A., and MARMAROU, A. (1994). A new model ofdiffuse brain injury in rats. Part II: Morphological charac-terization. J. Neurosurg. 80, 301–313.
GENNARELLI, T.A., THIBAULT, L.E., ADAMS, J.H., et al.(1982). Diffuse axonal injury and traumatic coma in the pri-mate. Ann. Neurol. 12, 564–574.
GENTLEMAN, S.M., NASH, M.J., SWEETING, C.J., et al.(1993). Beta–amyloid precursor protein (beta-APP) as amarker for axonal injury after head injury. Neurosci. Lett.160, 139–144.
HONG, S.C., GOTO, Y., LANZINO, G., et al. (1994). Neuro-protection with a calpain inhibitor in a model of focal cere-bral ischemia. Stroke 25, 663–669.
KAMPFL, A., POSMANTUR, R.M., ZHAO, X., et al. (1997).Mechanisms of calpain proteolysis following traumatic braininjury: implications for pathology and therapy: implicationsfor pathology and therapy: a review and update. J. Neuro-trauma 14, 121–134.
KOIZUMI, H., and POVLISHOCK, J.T. (1998). Posttraumatichypothermia in the treatment of axonal damage in an animalmodel of traumatic axonal injury. J. Neurosurg. 89, 303–309.
LEE, K.S., YANAMOTO, H., FERGUS, A., et al. (1997). Cal-cium-activated proteolysis as a therapeutic target in cere-brovascular disease. Ann. N.Y. Acad. Sci. 825, 95–103.
LEE, V.M., CARDEN, M.J., SCHLAEPFER, W.W., et al.(1987). Monoclonal antibodies distinguish several differen-tially phosphorylated states of the two largest rat neurofila-ment subunits (NF-H and NF-M) and demonstrate their ex-istence in the normal nervous system of adult rats. J.Neurosci. 7, 3474–3488.
MARKGRAF, C.G., VELAYO, N.L., JOHNSON, M.P., et al. (1998). Six-hour window of opportunity for calpain inhibition in focal cerebral ischemia in rats. Stroke 29,152–158.
MARMAROU, A., FODA, M.A., VAN DEN BRINK, W., etal. (1994). A new model of diffuse brain injury in rats. PartI: Pathophysiology and biomechanics. J. Neurosurg. 80,291–300.
CALPAIN INHIBITOR PREVENTS TRAUMATIC AXONAL INJURY
267
MAXWELL, W.L., POVLISHOCK, J.T., and GRAHAM, D.L.(1997). A mechanistic analysis of nondisruptive axonal in-jury: a review. J. Neurotrauma 14, 419–440.
MCCRACKEN, E., HUNTER, A.J., PATEL, S., et al. (1999).Calpain activation and cytoskeletal protein breakdown in thecorpus callosum of head-injured patients. J. Neurotrauma 16,749–761.
MEANEY, D. F., MARGULIES, S.S., and SMITH, D. H.(2000). Diffuse axonal injury. J. Neurosurg. 95, 1108–1110.
OKONKWO, D.O., BUKI, A., SIMAN, R., et al. (1999). Cy-closporin A limits calcium-induced axonal damage follow-ing traumatic brain injury. Neuroreport 10, 353–358.
OKONKWO, D.O., PETTUS, E.H., MOROI, J., et al. (1998).Alteration of the neurofilament sidearm and its relation toneurofilament compaction occurring with traumatic axonalinjury. Brain Res. 784, 1–6.
OKONKWO, D.O., and POVLISHOCK, J.T. (1999). An in-trathecal bolus of cyclosporin A before injury preserves mi-tochondrial integrity and attenuates axonal disruption in trau-matic brain injury. J. Cereb. Blood Flow Metab. 19, 443–451.
PETTUS, E.H., CHRISTMAN, C.W., GIEBEL, M.L., et al.(1994). Traumatically induced altered membrane permeabil-ity: its relationship to traumatically induced reactive axonalchange. J. Neurotrauma 11, 507–522.
PETTUS, E.H., and POVLISHOCK, J.T. (1996). Characteri-zation of a distinct set of intra-axonal ultrastructural changesassociated with traumatically induced alteration in axolem-mal permeability. Brain Res. 722, 1–11.
PIKE, B.R., FLINT, J., DUTTA, S., et al. (2001). Accumula-tion of non-erythroid alpha II-spectrin and calpain-cleavedalpha II-spectrin breakdown products in cerebrospinal fluidafter traumatic brain injury in rats. J. Neurochem. 78,1297–1306.
PIKE, B.R., ZHAO, X., NEWCOMB, J.K., et al. (1998). Re-gional calpain and caspase-3 proteolysis of alpha-spectrin af-ter traumatic brain injury. Neuroreport 9, 2437–2442.
POSMANTUR, R., KAMPFL, A., SIMAN, R., et al. (1997).A calpain inhibitor attenuates cortical cytoskeletal proteinloss after experimental traumatic brain injury in the rat. Neu-roscience 77, 875–88.
POVLISHOCK, J.T. (1992). Traumatically induced axonal in-jury: pathogenesis and pathobiological implications. BrainPathol. 2, 1–12.
POVLISHOCK, J.T., BECKER, D.P., CHENG, C.L., et al.(1983). Axonal change in minor head injury. J. Neuropathol.Exp. Neurol. 42, 225–242.
POVLISHOCK, J.T., ERB, D.E., and ASTRUC, J. (1992). Ax-onal response to traumatic brain injury: reactive axonalchange, deafferentation, and neuroplasticity. J. Neurotrauma9, S189–S200.
POVLISHOCK, J.T., MARMAROU, A., MCINTOSH, T., etal. (1997). Impact acceleration injury in the rat: evidence forfocal axolemmal change and related neurofilament sidearmalteration. J. Neuropathol. Exp. Neurol. 56, 347–359.
SAATMAN, K.E., BOZYCZKO-COYNE, D., MARCY, V., etal. (1996b). Prolonged calpain-mediated spectrin breakdownoccurs regionally following experimental brain injury in therat. J. Neuropathol. Exp. Neurol. 55, 850–860.
SAATMAN, K.E., MURAI, H., BARTUS, R.T., et al. (1996a).Calpain inhibitor AK295 attenuates motor and cognitivedeficits following experimental brain injury in the rat. Proc.Natl. Acad. Sci. U.S.A. 93, 3428–3433.
SAATMAN, K.E., ZHANG, C., BARTUS, R.T., et al. (2000).Behavioral efficacy of posttraumatic calpain inhibition is notaccompanied by reduced spectrin proteolysis, cortical lesion,or apoptosis. J. Cereb. Blood Flow Metab. 20, 66–73.
SHERRIFF, F.E., BRIDGES, L.R., GENTLEMAN, S.M., et al.(1994). Markers of axonal injury in post mortem humanbrain. Acta Neuropathol. (Berl.) 88, 433–439.
STONE, J.R., SINGLETON, R.H., and POVLISHOCK, J.T.(2000). Antibodies to the C-terminus of the beta-amyloid pre-cursor protein (APP): a site specific marker for the detectionof traumatic axonal injury. Brain Res. 871, 288–302.
STONE, J.R., SINGLETON, R.H., and POVLISHOCK, J.T.(2001). Intra-axonal neurofilament compaction does notevoke local axonal swelling in all traumatically injured ax-ons. Exp. Neurol. 172, 320–331.
STONE, J.R., WALKER, S.A., and POVLISHOCK, J.T.(1999). The visualization of a new class of traumatically in-jured axons through the use of a modified method of mi-crowave antigen retrieval. Acta Neuropathol. (Berl.) 97,335–345.
STRICH, S.J. (1956). Diffuse degeneration of the cerebral whitematter in severe dementia following head injury. J. Neurol.Neurosurg. Psychiatry 19, 163–185.
WOLF, J.A., STYS, P. K., LUSARDI T., et al. (2001). Trau-matic Axonal Injury induces calcium influx modulated bytetrodotoxin-sensitive sodium channels. J. Neurosci. 21,1923–1930.
Address reprint requests to:John T. Povlishock, Ph.D.VCU Neuroscience Center
Medical College of Virginia CampusVirginia Commonwealth University
P.O. Box 9807091101 E. Marshall St.
Richmond, VA 23298-0709
E-mail: [email protected]
BUKI ET AL.
268
www.elsevier.com/locate/regpep
Regulatory Peptides 123 (2004) 69–75
Effects of pituitary adenylate cyclase activating polypeptide
in a rat model of traumatic brain injury
Orsolya Farkasa, Andrea Tamasb,*, Andrea Zsomboka,c, Dora Reglo}dib, Jozsef Pala,Andras Bukia, Istvan Lengvarib, John T. Povlishockd, Tamas Doczia
aDepartment of Neurosurgery, University of Pecs, Medical Faculty, HungarybDepartment of Anatomy (Neurohumoral Regulations Research Group of the Hungarian Academy of Sciences), University of Pecs,
Medical Faculty, Szigeti u 12, Pecs 7624, HungarycDepartment of Central Laboratory of Animal Research, University of Pecs, Medical Faculty, HungarydDepartment of Anatomy and Neurobiology, Virginia Commonwealth University, Richmond, VA, USA
Available online 28 July 2004
Abstract
Pituitary adenylate cyclase activating polypeptide (PACAP) is a widely distributed neuropeptide that has numerous different actions.
Recent studies have shown that PACAP exerts neuroprotective effects not only in vitro but also in vivo, in animal models of global and focal
cerebral ischemia, Parkinson’s disease and axonal injuries. Traumatic brain injury has an increasing mortality and morbidity and it evokes
diffuse axonal injury which further contributes to its damaging effects. The aim of the present study was to examine the possible
neuroprotective effect of PACAP in a rat model of diffuse axonal injury induced by impact acceleration. Axonal damage was assessed by
immunohistochemistry using an antiserum against beta-amyloid precursor protein, a marker of altered axoplasmic transport considered as key
feature in axonal injury. In these experiments, we have established the dose response curves for PACAP administration in traumatic axonal
injury, demonstrating that a single post-injury intracerebroventricular injection of 100 Ag PACAP significantly reduced the density of
damaged, beta-amyloid precursor protein-immunoreactive axons in the corticospinal tract.
D 2004 Elsevier B.V. All rights reserved.
Keywords: Neuroprotection; Traumatic brain injury; Beta-amyloid precursor protein; Corticospinal tract; Traumatic axonal injury
1. Introduction few studies have shown that this effect is also present in
Pituitary adenylate cyclase activating polypeptide
(PACAP) is a member of the vasoactive intestinal peptide
(VIP)/secretin/glucagon peptide family [1]. PACAP was
discovered as a hypothalamic peptide on its potential of
increasing adenylate cyclase activity in the pituitary gland.
Since its discovery, various distinct effects in the central
and peripheral nervous systems have been described [2–4].
Among them, PACAP has neurotrophic and neuroprotec-
tive actions [2–4]. Numerous studies have proven its
neuroprotective effects in vitro, where PACAP protects
neurons against different neurotoxic agents [5–10]. A
0167-0115/$ - see front matter D 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.regpep.2004.05.014
* Corresponding author. Tel.: +36-72-536001/5398; fax: +36-72-
536393.
E-mail address: [email protected] (A. Tamas).
vivo, in optic nerve axotomy, spinal cord injury, facial
nerve transection, and 6-hydroxydopamine-induced lesion
of the substantia nigra [11–14]. Also, PACAP has neuro-
protective effects in animal models of global and focal
ischemia [15–17].
Although cerebral ischemia and traumatic brain injury
are triggered by different events, the majority of cellular
events leading eventually to death of neurons are shared
by both ischemic and traumatic brain injuries [18–21].
These include excitotoxicity, nitric oxide, inflammation,
reactive oxygen species, elevated Ca levels and apoptosis
[18–21]. The presence of these common pathways sug-
gests that those interventions that target the consensus
pathways of these two major diseases of the central
nervous system should be effective in both types of brain
injury. Based on the several lines of evidence on the
neuroprotective effects of PACAP in cerebral ischemia,
O. Farkas et al. / Regulatory Peptides 123 (2004) 69–7570
the aim of the present study was to investigate the possible
protective effect of this neuropeptide in a rat model of
traumatic brain injury.
2. Methods
2.1. Induction of traumatic brain injury
Adult Wistar rats (375–400 g) were subjected to impact
acceleration traumatic brain injury in accordance with
previous descriptions [22,23]. Animals were anesthetized
in a bell jar for 5 min under 4% isoflurane and a 2:1
mixture of N2O/O2, endotracheally intubated and main-
tained under 1.5–2% isoflurane. Rectal and temporal
temperatures were monitored and maintained at 37 jCwith a ventral heating pad. A midline scalp incision was
made and the skull was exposed with blunt dissection.
Parietal bones were exposed, cleaned and dried with bone
wax. A 10-mm stainless steel disc helmet was secured with
dental acrylic to the skull between bregma and lambda.
Rats were then fastened prone to a foam bed beneath the
injury tower and injured from 2 m with a 450 g weight.
After injury, animals were immediately ventilated with
100% O2, the helmet was removed, and the skull examined
for sign of fracture, which, if found, disqualified the rat
from further evaluation. Selected animals underwent mon-
itoring of other physiological parameters including inva-
sive blood pressure, pulse oxymetry and arterial blood gas
analysis.
2.2. Drug administration
In the first study, a single bolus of 125 Ag/kg PACAP
dissolved in physiological saline (n = 11) or the same
volume of vehicle in control animals (n = 13) was given
intravenously before the induction of traumatic brain injury.
The dose of PACAP was selected based on previous studies
[16]. These animals were sacrificed 2 h (n = 5 and 7 in
PACAP-treated and control groups, respectively), or 6
h (n = 6 in both groups) after brain injury.
Since intravenous PACAP pre-treatment did not prove to
be effective (see Results), intravenous administration as a
post-treatment did not seem reasonable. Due to the failure of
the intravenous treatment, PACAP was administered intra-
cerebroventricularly in the second study. This model of
traumatic brain injury requires intact skull; therefore, ad-
ministration was started post-injury. Animals received 1 Ag(n = 8), 10 Ag (n = 11) 100 Ag (n = 13) PACAP, or vehicle
(n = 16) intracerebroventricularly. Sham-injured rats (n = 4)
were anesthetized, intubated and the parietal bones were
exposed without performing the impact acceleration injury.
In the first experiment, no significant increase of immuno-
positive axonal profiles could be observed 6 h after injury
compared to the results at 2 h; therefore, histological
assessment was done only 2 h after the induction of
traumatic brain injury or sham operation in the second
study.
2.3. Immunohistochemistry
At the designed survival time, animals were reanesthe-
tized and transcardially perfused with Zamboni’s fixative
[24]. The brains were immersed in the same fixative
overnight (16–18 h). Based upon previously published
observations concerning the topography of diffusely injured
axons in the rat brainstem [25], a median 5 mm wide block
of the brain was removed using a sagittal brain blocking
device (Braintree Scientific) to include the region extending
from the interpeduncular fossa to the first cervical segment.
The brainstem was blocked just lateral to the pyramids to
encompass known vulnerable fiber tracts in this model of
traumatic brain injury: the corticospinal tract and the medial
longitudinal fascicle.
Immunohistochemistry was used to detect axonal dam-
age. In the present study, we utilized immunostaining with
a polyclonal antiserum targeting the C-terminus of beta-
amyloid precursor protein, a marker of altered axoplasmic
transport [26,27]. Vibratome sections were cut serially at
40 Am and microwaved for optimizing antigenicity with
the retention of excellent ultrastructural detail according to
former publications [28]. Sections were then washed in
phosphate-buffered saline (PBS) containing 1% normal
goat serum (Fluca), and were incubated overnight in rabbit
anti-beta-amyloid precursor protein antibody at a dilution
of 1:3000 (Zymed). Having been rinsed in PBS, sections
were incubated in biotinylated anti-rabbit immunoglobulin
(Sigma) for 60 min. After incubation in avidin biotin–
peroxidase complex (ABC kit, Vector) and rinsing, sec-
tions were processed for visualization of the immunohis-
tochemical complex using 0.05% diaminobenzidine and
0.01% hydrogen peroxide in 0.1 M PBS. The sections
were then mounted and cleared for routine light micro-
scopic examination.
2.4. Image analysis
Damaged immunopositive axon profiles in the area of the
corticospinal tract and the medial longitudinal fascicle were
captured and digitized at a magnification of 50� . The total
number of damaged, immunoreactive axons were counted
using an NIH Image J software in a blinded fashion.
Student’s t-test was used to compare the density of immuno-
positive profiles (expressed as number/mm2) between the
control and PACAP-treated animals.
3. Results
In accordance with previous observations from our
laboratories and others’, the use of the above-described
experimental protocols did not result in any significant
O. Farkas et al. / Regulatory Peptides 123 (2004) 69–75 71
alteration in the physiological parameters monitored during
the experiment.
Light microscopic examination of the animals subjected
to traumatic brain injury and reacted for the visualization of
beta-amyloid precursor protein antibody revealed discrete
focal immunoreactivity within scattered axons in the corti-
cospinal tract and medial longitudinal fascicle. Sham-in-
jured animals did not display immunoreactive axonal
profiles (Fig. 1A). Morphological characteristics of swollen,
occasionally disconnected immunoreactive axonal segments
appeared entirely consistent with previous descriptions of
damaged axonal profiles 2–6 h post-injury (Fig. 1B)
[21,26,29]. There was no significant difference between
the number of immunopositive axonal profiles 2 or 6 h after
the injury.
Results of pre-injury intravenous PACAP administration
revealed that the mean densities of beta-amyloid precursor
D
E F
A
C
B
Fig. 1. Representative beta-amyloid precursor protein-immunostained sections of t
animal receiving pre-injury intravenous PACAP (C), and animals treated with 1 Asome of the immunopositive axon profiles) is proportional to the mean densities
protein-immunopositive axon profiles in PACAP-treated
animals in the corticospinal tract and medial longitudinal
fascicle were not significantly different from those in control
animals at either 2 or 6 h post-injury (Fig. 1C). When
PACAP was administered intracerebroventricularly after the
induction of traumatic brain injury, treatment with 100 AgPACAP resulted in a significant reduction of beta-amyloid
precursor protein-immunopositive axon profiles in the cor-
ticospinal tract when compared to control animals 2 h post-
injury (Figs. 1F and 2A). Treatment with lower doses of
PACAP (1 and 10 Ag) did not result in significant reduction
of damaged axons (Figs. 1D,E and 2A). In animals treated
with 1 Ag PACAP, even an elevation in immunopositive
axon profiles could be observed, which was not significant-
ly different from control animals, but significantly higher
when compared to animals treated with 100 Ag PACAP. In
contrast to the corticospinal tract, no statistical significance
he corticospinal tract from a sham-operated animal (A), a control rat (B), an
g (D), 10 Ag (E) and 100 Ag (F) PACAP. The number of arrows (indicating
of immunostained axon profiles.
Fig. 2. Mean densities of beta-amyloid precursor protein (APP)-immunoreactive axon profiles in the corticospinal tract (A) and medial longitudinal fascicle (B)
of control animals and those treated with different doses of PACAP. *P < 0.05 vs. control group.
O. Farkas et al. / Regulatory Peptides 123 (2004) 69–7572
could be established between the various treatment groups
in the medial longitudinal fascicle (Fig. 2B).
4. Discussion
In the present study we demonstrated that post-traumatic
treatment with high dose of PACAP significantly reduced
the density of damaged axons in the most vulnerable
corticospinal tract.
Our results showing that PACAP is neuroprotective in
vivo are in accordance with numerous other studies that
prove protective effects of this pleiotropic neuropeptide.
VIP, which shows close homology to PACAP, has also been
shown to exert neuroprotective effects in various in vitro
and in vivo systems [30–33]. Soon after its discovery of
PACAP, several studies demonstrated that PACAP also has
neurotrophic actions in vitro, enhancing survival and neurite
outgrowth of cultured neuroblasts [4,34,35]. Later, it was
demonstrated that PACAP is able to protect neurons against
various neurotoxic agents. Its protective effects have been
shown against glutamate-induced neurotoxicity in cortical
and retinal neurons [5,6,8], in PC12 cells against beta-
amyloid-induced toxicity [7], against lipopolysaccharide-
induced toxicity in cortical neurons [9] and against HIV
envelope protein in PC12 cells [4]. A few studies have
demonstrated that PACAP is able to exert neuroprotection in
vivo, in animal models of various brain pathologies. PACAP
shows neuroprotection in rat global cerebral ischemia [15]
and focal ischemia in rats and mice [16,17], in facial nerve
injury [12] and in optic nerve transection [13].
Considering traumatic injuries, there are only a few
studies that indicate the potentially protective effect of
PACAP. In a rat model of spinal cord injury, induced by
extradural static weight compression, post-injury PACAP
treatment significantly reduced the number of apoptotic
cells assessed by TUNEL staining in the damaged spinal
cord [11]. Pro-inflammatory cytokines are expressed in
O. Farkas et al. / Regulatory Peptides 123 (2004) 69–75 73
spinal cord within 1–2 h after traumatic injury, and this
temporal profile of cytokine production is mimicked in
spinal cord slices, where PACAP has been shown to
eliminate the increase of tumor necrosis factor after spinal
cord transection [36]. In a model of traumatic brain injury,
similar to the one used in our study, PACAP mRNA was
induced after injury, and the temporal profile of its upregu-
lation paralleled the decrease in the number of apoptotic
cells [37]. These studies indicate that PACAP is a promising
therapeutic agent in traumatic brain and spinal cord injuries
which is further supported by our observations.
The exact mechanism underlying the in vivo neuro-
protective effect of PACAP has not been clarified so far.
In vitro studies however indicate that various effects can
play a role, including antiapoptotic and anti-inflammatory
actions. The antiapoptotic effects of PACAP have been
proven in different neuronal cultures in control conditions
and against several apoptosis-inducing agents [10,38–40].
In cerebellar granule cells, PACAP efficiently reduces
caspase-3 activation, the final effector of apoptotic cell
death, which is activated also in traumatic brain injury
[21,39]. Inflammation accompanies and further aggravates
neuronal injury in most brain pathologies, including trau-
matic brain injury [18]. PACAP is a potent inactivator of
induced microglial release of pro-inflammatory cytokines
and nitric oxide [9,36,41]. Mitochondrial damage has long
been associated with the pathogenesis of traumatic brain
injury [20]. The mitochondrial failure caused by calcium-
induced opening of the permeability transition pore is a key
phenomenon in multiple neurological insults, including
axonal injury [21,42]. Aconitase is a key mitochondrial
enzyme influencing the viability of neurons, and it is
inactivated upon mitochondrial calcium influx. Recently, it
has been shown that the inactivation of aconitase is atten-
uated by PACAP [43].
We attempted to show the possible protective effect of
PACAP as an intravenous injection because this route of
administration is clinically more relevant. As far as the pre-
injury treatment is concerned, we followed the generally
used routine to test candidate neuroprotective drugs first as
pre-treatment. Several examples show that pre- and post-
treatments can be equally effective [29,44] in models of
brain injuries. Due to the failure of intravenous PACAP
treatment given pre-injury, giving PACAP as intravenous
post-treatment did not seem reasonable. The failure may be
due to the low concentration reaching the brainstem.
PACAP crosses the blood–brain barrier relatively well in
comparison with other neuropeptides both in the normal and
injured brain [45,46]. However, results of the regional
differences in brain transport revealed that the transport rate
of PACAP is lower or not measurable in the area examined
in our study (corticospinal tract and medial longitudinal
fascicle), in contrast to other brain areas [47]. Furthermore,
recent studies have demonstrated that intravenous efficacy
of PACAP requires continuous infusion—an observation
explained by the presence of a serum-binding protein
[48,49]. To this end, in a model of focal cerebral ischemia,
a single injection was not effective; nevertheless, continuous
intravenous infusion significantly attenuated the ischemic
brain damage [16]. Due to the failure of intravenous
administration of PACAP, we chose to administer PACAP
intracerebroventricularly, where significant neuroprotection
could be achieved by high concentrations of PACAP.
Several studies have proven the dose dependence of the
neuroprotective effect of PACAP [6,7]. It has also been
postulated that different doses of PACAP act through
different signal transduction pathways [50,51]. Our results
indicate that relatively high concentration is required to
achieve significant neuroprotection in this rodent model of
diffuse traumatic brain injury. Although most of the in vivo
studies have shown protective effects of PACAP at lower
doses [14–16,38], our results are in accordance with some
other studies where neuroprotection by PACAP was reached
by relatively high concentrations, such as in spinal cord
injury [11], in optic nerve transection [13] and in facial
nerve injury [12]. According to these few in vivo studies, it
seems that protective effects can be reached by only higher
doses of PACAP in traumatic injuries. No data are available
on the passage of PACAP from the cerebrospinal fluid to the
brain; therefore, the exact tissue concentration reached in
our study is not known. It is possible, however, that the
passage of PACAP to the brainstem is relatively low
similarly to the low transport rate across the blood–brain
barrier in this area [46]. This may also explain why higher
concentrations were required to reach significant neuro-
protection in the corticospinal tract. Further examination
of the exact dose dependence and therapeutic window is
reasonable in models of traumatic brain injury, where our
observations suggest a promising therapeutic value for
PACAP.
Acknowledgements
This work was supported by the National Science
Research Fund (OTKA T034491/046589, T035195, ETT),
Fogarty IRCA (1-RO3-TW0131302A1), NIH Grant
NS20193, National Science Projects NFKP1A/00026/2002
and ALK 000126/2002, the Hungarian Academy of
Sciences, Bolyai and Bekesy Scholarships.
References
[1] Miyata A, Arimura A, Dahl RR, Minamino N, Uehara A, Jiang L,
et al. Isolation of a novel 38-residue hypothalamic polypeptide
which stimulates adenylate cyclase in pituitary cells. Biochem Bio-
phys Res Commun 1989;164:567–74.
[2] Arimura A. Perspectives on pituitary adenylate cyclase activating
polypeptide (PACAP) in the neuroendocrine, endocrine, and nervous
systems. Jpn J Physiol 1998;48:301–31.
[3] Vaudry D, Gonzalez BJ, Basille M, Yon L, Fournier A, Vaudry H.
Pituitary adenylate cyclase activating polypeptide and its receptors:
from structure to functions. Pharmacol Rev 2000;52:269–324.
O. Farkas et al. / Regulatory Peptides 123 (2004) 69–7574
[4] Arimura A, Somogyvari-Vigh A, Weill C, Fiore RC, Tatsuno I, Bay
V, et al. PACAP functions as a neurotrophic factor. Ann N YAcad Sci
U S A 1994;739:228–43.
[5] Frechilla D, Garcia-Osta A, Palacios S, Cenarruzabeitia E, Del Rio J.
BDNF mediates the neuroprotective effect of PACAP-38 on rat cor-
tical neurons. NeuroReport 2001;12:919–23.
[6] Morio H, Tatsuno I, Hirai A, Tamura Y, Saito Y. Pituitary adenylate
cyclase activating polypeptide protects rat-cultured cortical neurons
from glutamate-induced cytotoxicity. Brain Res 1996;741:82–8.
[7] Onoue S, Endo K, Ohshima K, Yajima T, Kashimoto K. The neuro-
peptide PACAP attenuates h-amyloid (1–42)-induced toxicity in
PC12 cells. Peptides 2002;23:1471–8.
[8] Shoge K, Mishima HK, Saitoh T, Ishihara K, Tamura Y, Shiomi H,
et al. Attenuation by PACAP of glutamate-induced neurotoxicity in
cultured retinal neurons. Brain Res 1999;839:66–73.
[9] Kong LY, Maderdrut JL, Jeohn GH, Hong JS. Reduction of lipopoly-
saccharide-induced neurotoxicity in mixed cortical neuron/glia cul-
tures by femtomolar concentrations of pituitary adenylate cyclase
activating polypeptide. Neuroscience 1999;91:493–500.
[10] Vaudry D, Pamantung TF, Basille M, Rouselle C, Fournier A, Vaudry
H, et al. PACAP protects cerebellar granule neurons against oxidative
stress-induced apoptosis. Eur J Neurosci 2002;15:1451–60.
[11] Katahira M, Yone K, Arishima Y, Nagamine T, Komiya S, Iwata S,
et al. The neuroprotective effects of PACAP on spinal cord injury
(SCI) in rats. Regul Pept 2003;115:49 [abstract].
[12] Kimura H, Kawatani M, Ito E, Ishikawa K. Effects of pituitary adeny-
late cyclase activating polypeptide on facial nerve recovery in the
guinea pig. Laryngoscope 2003;113:1000–6.
[13] Seki T, Izumi S, Shioda S, Arimura A. Pituitary adenylate cyclase
activating polypeptide (PACAP) protects ganglion cell death against
cutting of optic nerve in the rat retina. Regul Pept 2003;115:55
[abstract].
[14] Reglo}di D, Lubics A, Tamas A, Szalontay L, Lengvari I. Pituitary
adenylate cyclase activating polypeptide protects dopaminergic neu-
rons and improves behavioral deficits in a rat model of Parkinson’s
disease. Behav Brain Res 2004;151:303–12.
[15] Uchida D, Arimura A, Somogyvari-Vigh A, Shioda S, Banks W.
Prevention of ischemia-induced death of hippocampal neurons by
pituitary adenylate-cyclase activating polypeptide. Brain Res 1996;
736:280–6.
[16] Reglo}di D, Somogyvari-Vigh A, Vigh S, Kozicz T, Arimura A.
Delayed systemic administration of PACAP38 is neuroprotective in
transient middle cerebral artery occlusion in the rat. Stroke 2000;31:
1411–7.
[17] Ohtaki H, Dohi K, Yin T, Takaki A, Nakajo S, Shioda S. Neuro-
protective effect of PACAP38 through IL-6 after focal ischemia in
mouse. Regul Pept 2003;115:53 [abstract].
[18] Leker RR, Shohami E. Cerebral ischemia and trauma—different eti-
ologies yet similar mechanisms: neuroprotective opportunities. Brain
Res Rev 2002;39:55–73.
[19] Povlishok JT. Traumatically induced axonal injury: pathogenesis and
pathobiological implications. Brain Pathol 1992;2:1–12.
[20] Maxwell WL, Povlishok JT, Graham DL. A mechanistic analysis of
nondisruptive axonal injury: a review. J Neurotrauma 1997;14:
419–40.
[21] Buki A, Okonkwo DO, Wang KK, Povlishok JT. Cytochrome c re-
lease and caspase activation in traumatic axonal injury. J Neurosci
2000;20:2825–34.
[22] Marmarou A, Foda M.A.A.-E, Van Den Brink W, Campbell J, Kita H,
Demetriadou K. A new model of diffuse brain injury in rats: Part I.
Pathophysiology and biomechanics. J Neurosurg 1994;80:291–300.
[23] Foda MA, Marmarou A. A new model of diffuse brain injury in rats:
Part II. Morphological characterization. J Neurosurg 1994;80:301–13.
[24] Zamboni L, Martino L. Buffered picric acid– formaldehyde: a new
rapid fixative for electron microscopy. J Cell Biol 1967;35:148A.
[25] Povlishok JT, Marmarou A, McIntosh T, Trojanowski JQ, Moroi J.
Impact acceleration injury in the rat: evidence for focal axolemmal
change and related neurofilament sidearm alteration. J Neuropathol
Exp Neurol 1997;56:347–59.
[26] Stone JR, Singleton RH, Povlishok JT. Antibodies to the C-terminus of
the beta-amyloid precursor protein (APP): a site specific marker for the
detection of traumatic axonal injury. Brain Res 2000;871:288–302.
[27] Gentleman SM, Nash MJ, Sweeting CJ, Graham DI, Roberts GW.
Beta-amyloid precursor protein (beta-APP) as a marker for axonal
injury after head injury. Neurosci Lett 1993;160:139–44.
[28] Stone JR, Walker SA, Povlishok JT. The visualisation of a new class
of traumatically injured axons through the use of a modified method
of microwave antigen retrieval. Acta Neuropathol 1999;97:335–45.
[29] Buki A, Okonkwo DO, Povlishok JT. Postinjury cyclosporin A ad-
ministration limits axonal damage and disconnection in traumatic
brain injury. J Neurotrauma 1999;16:511–21.
[30] Gozes I. Neuroprotective drug delivery and development: potential
new therapeutics. Trends Neurosci 2001;24:700–5.
[31] Gozes I, Perl O, Giladi E, Davidson A, Ashur-Fabian O, Rubinraut S,
et al. Mapping the active site in vasoactive intestinal peptide to a core
of four amino acids: neuroprotective drug design. Proc Natl Acad Sci
U S A 1999;96:4143–8.
[32] Dibbern DA, Glazner GW, Gozes I, Brenneman DE, Hill JM. Inhibi-
tion of murine embryonic growth by human immunodeficiency virus
envelope protein and its prevention by vasoactive intestinal peptide
and activity-dependent neurotrophic factor. J Clin Invest 1997;99:
2837–41.
[33] Gozes I, Divinsky I, Pilzer I, Fridkin M, Brenneman DE, Spier AD.
From vasoactive intestinal peptide (VIP) through activity-dependent
neuroprotective protein (ADNP) to NAP: a view of neuroprotection
and cell division. J Mol Neurosci 2003;20:315–22.
[34] Deutsch PJ, Sun Y. The 38-amino acid form of pituitary adenylate
cyclase activating polypeptide stimulates dual signaling cascades in
PC12 cells and promotes neurite outgrowth. J Biol Chem 1992;267:
5108–13.
[35] Waschek JA. Multiple actions of pituitary adenylate cyclase activating
peptide in nervous system development and regeneration. Dev Neuro-
sci 2002;24:14–23.
[36] Kim WK, Kan Y, Ganea D, Hart RP, Gozes I, Jonakait M. Vasoactive
intestinal peptide and pituitary adenylate cyclase activating polypep-
tide inhibit tumor necrosis factor-a production in injured spinal cord
and in activated microglia via cAMP-dependent pathway. J Neurosci
2000;20:3622–30.
[37] Skoglosa Y, Lewen A, Takei N, Hillered L, Lindholm D. Regulation
of pituitary adenylate cyclase activating polypeptide and its receptor
type 1 after traumatic brain injury: comparison with brain-derived
neurotrophic factor and the induction of neuronal cell death. Neuro-
science 1999;90:235–47.
[38] Shioda S, Ozawa H, Dohi K, Mizushima H, Matsumoto K, Nakajo S,
et al. PACAP protects hippocampal neurons against apoptosis: in-
volvement of JNK/SAPK signaling pathway. Ann N Y Acad Sci
1998;865:111–7.
[39] Vaudry D, Gonzalez BJ, Basille M, Pamantung TF, Fontaine M, Four-
nier A, et al. The neuroprotective effect of pituitary adenylate cyclase
activating polypeptide on cerebellar granule cells is mediated through
inhibition of the CED3-related cystein protease caspase-3/CPP32.
Proc Natl Acad Sci U S A 2000;97:13390–5.
[40] Vaudry D, Rouselle C, Basille M, Falluel-Morel A, Pamantung TF,
Fontaine M, et al. Pituitary adenylate cyclase activating polypeptide
protects rat cerebellar granule neurons against ethanol-induced apo-
ptotic cell death. Proc Natl Acad Sci U S A 2002;99:6398–403.
[41] Delgado M, Leceta J, Ganea D. Vasoactive intestinal peptide and
pituitary adenylate cyclase activating polypeptide inhibit the produc-
tion of inflammatory mediators by activated microglia. J Leukoc Biol
2003;73:155–64.
[42] Pettus EH, Christman CW, Giebel ML, Povlishok JT. Traumatically
induced altered membrane permeability: its relationship to traumat-
ically induced reactive axonal change. J Neurotrauma 1994;11:
507–22.
O. Farkas et al. / Regulatory Peptides 123 (2004) 69–75 75
[43] Tabuchi A, Funaji K, Nakatsubo J, Fukuchi M, Tsuchiya T, Tsuda M.
Inactivation of aconitase during apoptosis of mouse cerebellar granule
neurons induced by a deprivation of membrane depolarization. J Neu-
rosci Res 2003;71:504–15.
[44] Okonkwo DO, Buki A, Siman R, Povlishock JT. Cyclosporin A limits
calcium-induced axonal damage following traumatic brain injury.
NeuroReport 1999;10:353–8.
[45] Banks WA, Kastin AJ, Komaki G, Arimura A. Passage of pituitary
adenylate cyclase activating polypeptide 1–27 and pituitary adenylate
cyclase activating polypeptide 1–38 across the blood–brain barrier.
J Pharmacol Exp Ther 1993;267:690–6.
[46] Somogyvari-Vigh A, Pan W, Reglo}di D, Vigh S, Kastin AJ, Arimura
A. The passage of pituitary adenylate cyclase activating polypeptide
across the blood–brain barrier during focal cerebral ischemia. Regul
Pept 2000;91:89–95.
[47] Nonaka N, Banks WA, Mizushima H, Shioda S, Morley JE. Regional
differences in PACAP transport across the blood–brain barrier in
mice: a possible influence of strain, amyloid h protein and age. Pep-
tides 2002;23:2197–202.
[48] Somogyvari-Vigh A, Svoboda-Teet J, Vigh S, Arimura A. Is an in-
travenous bolus injection required prior to initiating slow intravenous
infusion of PACAP38 for prevention of neuronal death induced by
global ischemia? The possible presence of a binding protein for
PACAP38 in blood. Ann N Y Acad Sci 1998;865:595–600.
[49] Tams JW, Johnsen AH, Fahrenkrug J. Identification of pituitary
adenylate cyclase activating polypeptide 1–38-binding factor in
human plasma, as ceruloplasmin. Biochem J 1999;341:271–6.
[50] Arimura A. Perspectives on the development of a neuroprotective
drug based on PACAP. Regul Pept 2003;115:40 [abstract].
[51] Zhou CJ, Shioda S, Yada T, Inagaki N, Pleasure SJ, Kikuyama S.
PACAP and its receptors exert pleiotropic effects in the nervous sys-
tem by activating multiple signaling pathways. Curr Prot Pept Sci
2002;3:423–39.
686
JOURNAL OF NEUROTRAUMAVolume 23, Number 5, 2006© Mary Ann Liebert, Inc.Pp. 686–695
Postinjury Administration of Pituitary Adenylate CyclaseActivating Polypeptide (PACAP) Attenuates Traumatically
Induced Axonal Injury in Rats
ANDREA TAMÁS,1 ANDREA ZSOMBOK,2,3 ORSOLYA FARKAS,2 DÓRA REGLÖDI,1JÓZSEF PÁL,2,5 ANDRÁS BÜKI,2 ISTVÁN LENGVÁRI,1 JOHN T. POVLISHOCK,4
and TAMÁS DÓCZI2,5
ABSTRACT
Pituitary adenylate cyclase activating polypeptide (PACAP) has several different actions in the ner-vous system. Numerous studies have shown its neuroprotective effects both in vitro and in vivo. Pre-viously, it has been demonstrated that PACAP reduces brain damage in rat models of global andfocal cerebral ischemia. Based on the protective effects of PACAP in cerebral ischemia and the pres-ence of common pathogenic mechanisms in cerebral ischemia and traumatic brain injury (TBI), theaim of the present study was to investigate the possible protective effect of PACAP administered 30min or 1 h postinjury in a rat model of diffuse axonal injury. Adult Wistar male rats were subjectedto impact acceleration, and PACAP was administered intracerebroventricularly 30 min (n � 4), and1 h after the injury (n � 5). Control animals received the same volume of vehicle at both time-points(n � 5). Two hours after the injury, brains were processed for immunhistochemical localization ofdamaged axonal profiles displaying either �–amyloid precursor protein (�-APP) or RMO-14 im-munoreactivity, both considered markers of specific features of traumatic axonal injury. Our re-sults show that treatment with PACAP (100 �g) 30 min or 1 h after the induction of TBI resultedin a significant reduction of the density of �-APP–immunopositive axon profiles in the corticospinaltract (CSpT). There was no significant difference between the density of �-APP–immunopositiveaxons in the medial longitudinal fascicle (MLF). PACAP treatment did not result in significantlydifferent number of RMO-14–immunopositive axonal profiles in either brain areas 2 hours post-in-jury compared to normal animals. While the results of this study highlighted the complexity of thepathogenesis and manifestation of diffuse axonal injury, they also indicate that PACAP should beconsidered a potential therapeutic agent in TBI.
Key words: �-amyloid precursor protein; corticospinal tract; neuroprotection; PACAP; traumatic braininjury
Departments of 1Anatomy (Neurohumoral Regulations Research Group of the Hungarian Academy of Sciences), 2Neurosurgeryand 3Central Laboratory of Animal Research, University of Pécs, Medical Faculty, Pécs, Hungary.
4Department of Anatomy and Neurobiology, Virginia Commonwealth University, Richmond, Virginia.5Clinical Neuroscience Research Group of the Hungarian Academy of Sciences, Department of Neurosurgery, University of
Pécs, Medical Faculty, Pécs, Hungary.
INTRODUCTION
NEUROPATHOLOGICAL STUDIES, clinical observations,and other data derived from animal models have sig-
nificantly increased the understanding of both primaryand secondary brain damage over the past decades. In up to 45% of severely head-injured patients, the neuro-pathological changes after head injury may be due to theeffects of secondary mechanisms. The early insults, suchas transient global ischemia, haematoma, or diffuse ax-onal injury, are followed and complicated by secondarymechanisms, which may be mediated by complex cas-cades of biochemical processes. Many of these secondaryposttraumatic events have been targeted as potential sitesfor pharmacological interventions to prevent or to sig-nificantly reduce secondary brain damage (Bullock,1993).
Numerous neuroprotective drugs and therapeutic in-terventions have been tested in different animal modelsof traumatic brain injury (TBI). Despite the obvious dif-ferences in their pathogenesis, TBI and ischemic braindamage do share some common pathways providing asolid basis for the application of—at least partially—sim-ilar treatment strategies. Such mechanisms include exci-totoxicity, overproduction of free radicals, inflammation,and apoptosis (Bramlett and Dietrich, 2004; Leker andShohami, 2002).
One of the most widely used therapeutic approachesis hypothermia. Postinjury hypothermia is extensivelystudied and utilized for neuroprotection in several mod-els of ischemia and traumatic injuries of the nervous sys-tem (Bethea and Dietrich, 2002; Bramlett et al., 1995;1997; Koizumi and Povlishock, 1998; Lyeth et al., 1993).Hypothermia prevents free radical production (Globus etal., 1995), nitric oxide synthesis (Sakamoto et al., 1997),intracellular calcium accumulation (Mitani et al., 1991),and the rise in excitatory amino acid secretion (Koizumiet al., 1997).
There are several other therapeutic interventions effec-tive both in cerebral ischemia and TBI. The efficacy of posttraumatic Mg2� as a neuroprotective agent has been shown in both diffuse axonal injury and brain ischemia (Heath and Vink, 1999; Tsuda et al., 1991). Sim-ilarly, treatment with NMDA channel blockers limits ex-citotoxin-induced secondary neuronal damage in modelsof TBI (Goda et al., 2002) and cerebral ischemia (Dirnaglet al., 1990). Inhibition of inflammatory mediators and cal-pain has also been proven effective in both models (Bukiet al., 1999b, 2003; Cernak et al., 2002; Hong et al., 1994;Kadoya et al., 1995; Nakayama et al, 1998; Sanderson etal., 1999; Yoshimoto and Siesjo, 1999).
Several neurotrophic factors, such as brain-derivedneurotrophic factor, nerve growth factor and fibroblast
growth factor, have been shown to play an important rolein the endogenous response following both cerebral ischemia and traumatic brain injury, and to be effectivein both brain pathologies when given exogenously (Di-etrich et al., 1996; Kawamata et al., 1997; Leker andShohami, 2002; Truettner et al., 1999). Pituitary adeny-late cyclase-activating polypeptide (PACAP) is a neu-ropeptide structurally belonging to the secretin/glucagon/VIP family. PACAP was isolated from ovine hypothala-mus based on its ability to activate adenylate cyclase inthe pituitary gland. PACAP is a bioactive peptide withdiverse activities in the nervous system. In addition to itsmore classic role as a neurotransmitter, PACAP functionsas a neurotrophic factor (Somogyvari-Vigh and Reglodi,2004). Numerous studies have shown its neuroprotectiveeffect in vitro and in vivo (Arimura, 1998; Somogyvari-Vigh and Reglodi, 2004; Vaudry et al., 2000a).
Previously, it has been demonstrated that PACAP hasneuroprotective effects in a rat model of global and focalcerebral ischemia treated before and after the injury(Uchida et al., 1996; Reglodi et al., 2000, 2002). Based onthe findings with PACAP in cerebral ischemia and thepresence of common pathogenetic mechanisms in cerebralischemia and TBI, we applied PACAP therapy in a ratmodel of TBI in a previous study. It was found that PACAPhad neuroprotective effects when administered immedi-ately after the injury (Farkas et al., 2004). In acute cere-bral injuries, it is of utmost importance to determine thetherapeutic time window of candidate neuroprotectiveagents, since the immediate therapy is not possible in mostcases. Of similar importance is the identification of thera-peutic interventions that could be applied at the scene ofthe accident in an attempt to prolong the time window forfurther neuroprotective interventions and measures to in-hibit progression to irreversible structural changes. Theaim of the present study was therefore, to investigate thepossible protective effect of PACAP in a rat model of TBIgiven 30 min and 1 h postinjury in order to examine thepossibility of delaying the therapeutic window.
MATERIALS AND METHODS
Induction of Traumatic Brain Injury
Adult Wistar male rats (375–400 g) were subjected toimpact acceleration TBI in accordance with previous de-scriptions (Marmarou et al., 1994; Foda and Marmarou,1994). Animals were anesthetized in a bell jar for 5 minunder 4% isoflurane (Forane, Abott, Hungary) and a 2:1mixture of N2O:O2, endotracheally intubated, and main-tained under 1.5–2% isoflurane. Rectal and temporal tem-perature were monitored and maintained at 37°C with aventral heating pad (FHC BOWDOINHAMME 04008
POSTINJURY ADMINISTRATION OF PACAP IN TBI
687
USA Temperature Control). A midline scalp incision wasmade and the skull was exposed with blunt dissection. Pari-etal bones were exposed, cleaned, and dried with bone wax.A 10-mm stainless steel disc helmet was secured with den-tal acrylic to the skull between bregma and lambda. Ratswere then fastened prone to a foam bed beneath the injurytower and injured from 2 m with a 450-g weight. After in-jury, animals were immediately ventilated with 100% O2,the helmet was removed, and the skull examined for signof fracture, which, if found, disqualified the rat from fur-ther evaluation. Selected animals underwent monitoring ofother physiological parameters including invasive bloodpressure, pulse oxymetry and arterial blood gas analysis.All procedures were performed in accordance with the eth-ical guidelines of NIH and guidelines approved by the Uni-versity of Pécs (no. BA02/2000-31/2001) to minimize painand suffering of the animals. The minimal number of ani-mals required to achieve statistically meaningful results wasused in all cases.
Drug Administration
A single bolus of 100 �g PACAP (Sigma, Hungary)dissolved in 5 �L of physiological saline was adminis-tered intracerebroventriculary 30 min after the injury inone group of rats (n � 4), and 1 h after the injury in an-other group (n � 5). Control animals received the samevolume of vehicle at both time-points (n � 5). The doseof PACAP was selected based on previous studies de-scribing dose-dependent characteristics of PACAP ad-ministration (Farkas et al., 2004). Sham-injured rats (n �4) were anesthetized, intubated and the parietal boneswere exposed without performing the impact accelera-tion injury. The animals were sacrificed for the histolog-ical assessment 2 h after brain injury.
Immunohistochemistry
At the designed survival time, animals were reanes-thetized with an overdose of sodium pentobarbital andtranscardially perfused with Zamboni’s fixative (Zam-boni and Martino, 1967). The brains were removed andimmersed in the same fixative overnight (16–18 h). Basedupon previously published observations concerning thetopography of diffusely injured axons in the rat brain-stem (Povlishock et al., 1997), a median 5-mm-wideblock of the brain was removed using a sagittal brainblocking device (Braintree Scientific Inc.) to include theregion extending from the interpeduncular fossa to thefirst cervical segment. The brainstem was blocked justlateral to the pyramids to encompass known vulnerablefiber tracts in this model of TBI: the corticospinal tract(CSpT) and the medial longitudinal fascicle (MLF). Thetissue was sectioned with Vibrotome Series 1000 (Poly-
sciences Inc., Warrington, PA) at a thickness of 40 �mand collected in 0.1 M phosphate buffer. The sectionswere collected in two groups in a semi-serial fashion,rinsed three times for 10 min in phosphate-buffered saline(PBS), and processed for immunhistochemical localiza-tion of damaged axonal profiles.
Half of the sections were single labeled for the detec-tion of RMO-14 immunoreactivity. This antibody inknown to exclusively target an epitope on the rod domainsof altered NF-M subunits, which are exposed upon mod-ification of the NF sidearms, and assumed consequence ofcalcium induced enzymatic processes (Lee et al., 1987;Okonkwo et al., 1998; Povlishock et al., 1997). Every sec-ond section was processed for immunohistochemical de-tection of the �–amyloid precursor protein (�-APP). Thisclassical marker of traumatic axonal injury (TAI) is car-ried by fast axoplasmic transport and will pool at foci af-fected by TAI. In the present study, we utilized a poly-clonal antiserum targeting the C-terminus of �-APP(Gentleman et al., 1993; Stone et al., 2000). Vibratomesections microwaved for optimizing antigenicity with theretention of excellent ultrastructural detail according toformer publications (Stone et al., 1999). Sections fromboth groups were then washed in PBS containing 1% nor-mal goat serum (Fluca), and were incubated for 35 minwith 0.2% Triton X (Sigma-Aldrich, Hungary) in 10% nor-mal goat or horse serum, respectively (NGS or NHS;Sigma Chemical Co., St. Louis, MO) in PBS. After twoquick rinses in PBS containing 1% NGS (or NHS in thecase of RMO-14) the above defined groups of sectionswere incubated overnight in rabbit anti-APP antibody at adilution of 1:3000 (Zymed Laboratories, Hartford, CT) orin mouse monoclonal RMO-14 antibody (kindly providedby Dr. John Q. Trojanowski, University of PennsylvaniaDepartment of Pathology) at the dilution of 1:500. Hav-ing been rinsed 3 � 10 min in PBS containing 1% NGS(or NHS), sections were incubated in biotinylated anti-rab-bit immunoglobuline (Sigma) derived from goat or in1:400 dilution of biotinylated, rat adsorbed anti-mouseimmunoglobulin derived from horse (Vector) for 60 min.After incubation in avidin biotin-peroxidase complex(ABC kit, Vector) and rinsing in PBS and 0.1 M phos-phate buffer 3 � 10 min and 2 � 10 min respectively, sec-tions were processed for visualization of the immunohis-tochemical complex using 0.05% diaminobenzidine(Sigma) and 0.01% hydrogen peroxide in 0.1 M PBS. Thesections were then mounted and cleared for routine lightmicroscopic examination.
Image Analysis
Semi-serial brainstem sections were examined with aNikon light microscope interfaced with a computer-as-
TAMÁS ET AL.
688
sisted image analysis system (NIH Image J software) ina blinded fashion. At the pontomedullary junction, twoadjacent grids of 40,000 �m2 were superimposed overthe CSpT and the MLF, images were captured and digi-tized at a magnification of �50 and the total number ofdamaged APP and RMO-14 immunopositive axonal pro-files within this area were marked and counted. The meandensity of immunopositive axons was computed as anumber of immunopositive axons per mm2.
Statistical Analysis
Student t-test was used to compare the density of im-munopositive profiles (expressed as mean number/mm2)between the control and PACAP-treated animals.
RESULTS
Neither the use of the above-described experimentalTBI protocol nor the administration of PACAP resultedin any significant alteration in the physiological parame-ters monitored during the experiment (temporal and rec-tal temperature, blood pressure, and blood gases), con-sistent with previous observations from our laboratoriesand others’ (Buki et al., 1999a, 2003; Otani et al., 2002;Reglodi et al., 2000, 2002; Singleton et al., 2001). Pre-vious observations with PACAP treatment found a tran-sient, slight drop in blood pressure only when the pep-tide was administered intravenously: fall in the bloodpressure was observed 5 min after intravenous injection,but it soon returned to normal (Reglodi et al., 2000). Con-sistent with the current study, no such alteration wasfound in previous studies when PACAP was given icv(Reglodi et al., 2002).
Light microscopic examination of vehicle- and drug-treated animals subjected to TBI and reacted for the vi-sualization of �-APP and RMO-14 antibodies, revealeddiscrete focal immunoreactivity within scattered axons inthe CSpT and the MLF. Sham-injured animals did notdisplay immunoreactive axonal profiles. Morphologicalcharacteristics of swollen, occasionally disconnected �-APP–immunoreactive axonal segments and the lobu-lated/vacuolated, partially or totally disconnected RMO-14–immunoreactive axonal segments appeared entirelyconsistent with previous descriptions of damaged axonalprofiles 2 h post-injury (Buki et al., 1999a,b, 2000; Stoneet al., 2000).
Treatment with PACAP (100 �g) at 30 min or 1 hafter the induction of TBI resulted in a significant re-duction of �-APP–immunopositive axon profiles in theCSpT when compared to control animals 2 h postinjury.There was no significant difference between the num-ber of �-APP–immunopositive axons in the MLF andRMO-14–immunopositive axonal profiles in both brainareas at 2 h after the injury compared to normal animals(Figs. 1,2,3).
Quantitative findings followed by statistical analysisconfirmed that postinjury administration of 100 �g ofPACAP significantly reduced the mean density of dam-aged �-APP–immunoreactive axonal profiles in theCSpT from 812.7 � 29.0 to 501.9 � 81.4 (p � 0.001) orto 459.7 � 27.7 (p � 0.0001) 30 min or 1 h after injury,respectively (Figs. 1,3). There was no significant changein the MLF, where the mean number of �-APP–im-munoreactive axon profiles varied from 89.8 � 14.9 inthe control group to 99.3 � 60.4 and to 53.0 � 5.0 ingroups treated with PACAP at 30 min and 1 h postin-jury, respectively (Fig. 1).
POSTINJURY ADMINISTRATION OF PACAP IN TBI
689
FIG. 1. Mean densities of amyloid precursor protein (APP)–immunopositive axons in the corticospinal tract (CSpT) and themedial longitudinal fascicle (MLF) in control and pituitary adenylate cyclase activating polypeptide (PACAP)–treated animals.**p � 0.01, ***p � 0.001 versus control group.
In contrast, the mean density of RMO-14–immunorec-tive axonal profiles showed no significant differences ineither brain areas. In the CSpT, the mean density of RMO-14–immunorective profiles varied from 60.7 � 12.6 in thecontrol group to 117.8 � 38.5 and to 86.5 � 20.9 after ad-ministration of PACAP at 30 min and 1 h postinjury. Themean number of RMO-14–immunoreactive profiles in theMLF varied from 70.2 � 10.2 in the control group to
43.0 � 9.0 and to 54.5 � 8.6 in groups treated withPACAP at 30 min and 1 h after trauma (Fig. 2).
DISCUSSION
In the present study, we demonstrated that adminis-tration of PACAP 30 as late as 60 min post-injury sig-
TAMÁS ET AL.
690
FIG. 2. Mean densities of RMO-14–immunopositive axons in the corticospinal tract (CSpT) and the medial longitudinal fas-cicle (MLF) in control and pituitary adenylate cyclase activating polypeptide (PACAP)–treated animals.
FIG. 3. Traumatically injured axons displaying amy-loid precursor protein (APP) immunreactivity in the cor-ticospinal tract (CSpT) from injured animals. (A) Con-trol animals. (B) Animals treated with pituitary adenylatecyclase activating polypeptide (PACAP) at 30 min afterthe injury. (C) Animals treated with PACAP at 1 h postin-jury. Arrows indicate some immunostained axons, andthe number of arrows is proportional to the mean densi-ties of immunostained axon profiles. Higher magnifica-tion of the same areas is seen in the inserts.
nificantly reduced traumatically evoked axonal injury inthe most vulnerable CSpT in the brainstem. Thus, ourstudy provides further evidence for the previously de-scribed neuroprotective effects of PACAP.
Recently, many therapeutic agents have been used inmodels of TBI, such as antiinflammatory and antiapop-totic drugs, as well as substances protective against ex-citotoxicity, to attenuate the main pathogenetic mecha-nisms present in brain injuries (Leker and Shohami,2002). Most recent results indicate that PACAP may alsobe a promising therapeutic agent in injuries of the ner-vous system. PACAP is thought to interact with growthfactors during development and following injury (Somo-gyvari-Vigh and Reglodi, 2004). In a model of TBI, in-crease of PACAP mRNA has been observed in the vul-nerable cortex and dentate gyrus concomitant with thedecrease of TUNEL-positive apoptotic cells, which mayindicate a role for PACAP in reducing the number ofapoptotic neurons (Skoglosa et al., 1999). Since its dis-covery, numerous in vitro studies have demonstrated thatPACAP has neurotrophic and neuroprotective effects.The first reports of the in vivo efficacy of PACAP showedthat the peptide exerts neuroprotection in a rat model ofglobal and focal cerebral ischemia (Ohtaki et al. 2003;Reglodi et al., 2000, 2002; Uchida et al., 1996). Sub-sequently, it has been shown that local injections ofPACAP enhance facial nerve recovery after transection,and is neuroprotective in models of optic nerve transec-tion, spinal cord injury, and 6-OHDA–induced lesion(Katahira et al., 2003; Kimura et al., 2003; Reglodi et al.,2004; Seki et al., 2003) Based on the findings withPACAP in nervous injuries and the presence of commonfeatures of pathogenetic mechanisms in cerebral ischemiaand TBI, we administered PACAP in a rat model of TBI.
The application of PACAP was considered on the ba-sis of widespread mechanisms of action. Glutamate ex-citotoxicity plays a role in the damage caused by TBI,and NMDA-antagonists have been shown to reduce theneuronal damage after impact-acceleration brain injury(Goda et al., 2002). The survival promoting effect ofPACAP has been demonstrated against glutamate-in-duced toxicity in cultured cortical and retina neurons(Frechilla et al., 2001; Morio et al., 1996; Shoge et al.,1999). A recent study has shown that PACAP upregu-lates several genes that increase resistance to neurotoxicagents, including cytochrome P450 and FGF regulatedprotein (Vaudry et al., 2002b). PACAP also increases thevelocity of glutamate uptake by cortical astroglial cellsby promoting the expression of glutamate transportersand by inducing the expression of glutamine synthetase(Figiel and Engele, 2000).
Inflammation is not restricted to infectious or autoim-mune disorders of the nervous system, but occurs in cere-
bral ischemia, trauma and neurodegenerative disorders(Leker and Shohami, 2002). Inhibition of COX-2, anenyzme responsible for the production of prostaglandinsat sites of inflammation, improves cognitive and motoroutcome following diffuse TBI in rats (Cernak et al.,2002). Interleukin-1 receptor antagonists attenuate re-gional neuronal cell death and cognitive dysfunction af-ter experimental brain injury (Sanderson et al., 1999). Inaddition, postinjury hypothermia, which is widely uti-lized for neuroprotection in TBI, also has anti-inflam-matory effect (Deng et al., 2003). PACAP has beenproven to be a potent inactivator of induced microglialrelease of proinflammatory cytokines and chemokines,such as TNF�, IL-1�, IL-6, IL-12, and NO (Delgado etal., 2002, 2003; Kim et al., 2000). Moreover, PACAPstimulates the anti-inflammatory cytokine IL-10 (Ganeaand Delgado, 2002). Recently, it has also been shownthat PACAP has inhibitory action on chemotaxis and neu-trophil migration which may play a role in its neuropro-tective effect (Kinhult et al., 2001). A few studies havedemonstrated that the anti-inflammatory actions ofPACAP exist also in vivo (Delgado et al., 1999).
Apoptotic cell death plays a role in tissue injury fol-lowing brain trauma and can be prevented by manipula-tion of the steps that lead to activation of the caspase cas-cade (Leker and Shohami, 2002). Recently, it has beendemonstrated that inhibition of the caspase cascade re-duces post-TBI (Fink et al., 1999). PACAP decreases thenaturally occurring apoptotic cell death as well as apop-tosis induced by various agents in neuronal cell lines(Canonico et al., 1996; Shioda et al., 1998; Somogyvari-Vigh and Reglodi, 2004; Tanaka et al., 1997; Vaudry etal., 2000b, 2002a). In cerebellar granule cells, PACAPreduces caspase-3 activation, which is also activated inTBI (Buki et al., 2000; Vaudry et al., 2000b).
Mitochondrial damage has long been associated withthe pathogenesis of TBI (Maxwell et al., 1997; Okonkwoet al., 1999). The mitochondrial failure purportedlycaused by calcium-induced opening of the permeabilitytransition pore is a key phenomenon in multiple neuro-logical insults, including axonal injury (Buki et al., 2000;Pettus et al., 1994). Cyclosporin A, a potent inhibitor ofCa2�-induced mitochondrial permeability transition,blunts the progression of calcium-mediated cytoskeletalchange and reduces the number of disconnected/dys-functional axons in rat TBI (Buki et al., 1999b; Okonkwoet al., 1999). Aconitase, a key mitochondrial enzyme in-fluencing the viability of neurons in response to oxida-tive stress, is inactivated by a deprivation of Ca2� influxinto neurons and PACAP attenuates this inactivation(Tabuchi et al., 2003).
In the present study we demonstrated that administra-tion of PACAP at 30 min and 1 h after the injury reduced
POSTINJURY ADMINISTRATION OF PACAP IN TBI
691
traumatically evoked axonal injury as assessed by im-munocytochemical marker �-APP in the CSpT, whilethere was no significant changes in the MLF. Similar sur-prising finding of this study was the lack of therapeuticeffect in the case of damaged RMO-14–immunoreactivefibers. These findings might be explained by several rea-sons: first, the design of this study may not have allowedenough statistical power to establish significant differ-ence in the MLF where the density of damaged axonalprofiles was found in previous studies about 10% of thatof the CSpT. Further, recent observations by Stone et al.(2000) indicated that APP and RMO-14 purportedly la-bel—at least partially—different axon populations; 2 hpost-injury, the former is more frequently localized to theCSpT while the latter is prevalent in the MLF at this timepoint. Similar results have been obtained by Suehiro etal. (2001) where the immunophilin ligand FK506 was ad-ministered following posttraumatic hypothermia associ-ated with rapid rewarming. Recently, it has been demon-strated that the ability to protect mitochondria fromundergoing mitochondrial permeability transition is veryimportant in the neuroprotection following posthy-pothermic rewarming (Okonkwo and Povlishock, 1999),and RMO-14 immunoreactivity correlates with the per-centage of calpain-mediated spectrin proteolysis and mi-tochondrial damage (Buki et al., 1999a; Povlishock et al.,1997; Povlishock and Stone, 2001). Based on the obser-vation that FK506 did not decrease the number of RMO-14–positive fibers, the authors confirmed that FK506 hasno known action on mitochondrial permeability transi-tion (Liu et al., 1991; Suehiro et al., 2001). Presently,there are no data available on the effects of PACAP oncalpain-mediated spectrin proteolysis, which should cor-relate RMO-14 immunoreactivity. The differences be-tween the structure of fibers situated in the CSpT and theMLF, and the different pathogenesis of axonal damagein these structures could also be the reason for differentprotective effects (Povlishock and Stone, 2001; Suehiroet al., 2001).
A major finding in our study is that PACAP was ef-fective even when given delayed after the trauma. Nu-merous investigations have shown the protective effectof different treatments applied before injury (Buki et al.,2003; Singleton et al., 2001) and immediately after theinjury (Koizumi and Povlishock, 1998). Clinical thera-pies on the other hand focus on delayed administrationstrategies post-injury. It has been shown that cyclosporinA has neuroprotective effect when given 30 min after TBI(Buki et al., 1999b), while MgSO4 improves motor out-come when administered up to 24 h after injury in ratTBI (Heath et al., 1999). Previously, we have demon-strated that PACAP is effective when applied immedi-ately after the injury (Farkas et al., 2004), and in this
study we have proven that PACAP has neuroprotectiveeffect also 30 min and 1 h after TBI. These findings in-dicate a considerable therapeutic window existing for theuse of PACAP in TBI, implicating PACAP to the ther-apy of traumatic axonal injury.
ACKNOWLEDGMENTS
This work was supported by the National Science Re-search Fund (OTKA T034491, T035195, T046589,T048724), Fogarty IRCA (1-RO3-TW0131302A1), NIH(grant NS20193), National Science Projects (NFKP1A/00026/2002 and ALK 000126/2002), and the HungarianAcademy of Sciences, Bolyai and Bekesy Scholarships.
REFERENCES
ARIMURA, A. (1998). Perspectives on pituitary adenylate cy-clase activating polypeptide (PACAP) in the neuroendocrine,endocrine, and nervous systems. Jpn. J. Physiol. 48, 301–331.
BETHEA, J.R., and DIETRICH, W.D. (2002). Targeting thehost inflammatory response in traumatic spinal cord injury.Curr. Opin. Neurol. 15, 355–360.
BRAMLETT, H.M., GREEN, E.J., DIETRICH, W.D., BUSTO,R., GLOBUS, M.Y., and GINSBERG, M.D. (1995). Post-traumatic brain hypothermia provides protection from sen-sorimotor and cognitive behavioral deficits. J. Neurotrauma12, 289–298.
BRAMLETT, H.M., DIETRICH, W.D., GREEN, E.J., andBUSTO, R. (1997). Chronic histopathological consequencesof fluid-percussion brain injury in rats: effects of post-trau-matic hypothermia. Acta Neuropathol. (Berl.) 93, 190–199.
BRAMLETT, H.M., and DIETRICH, W.D. (2004). Patho-physiology of cerebral ischemia and brain trauma: similari-ties and differences. J. Cereb. Blood Flow Metab. 24, 133–150.
BUKI, A., KOIZUMI, H., and POVLISHOCK, J.T. (1999a).Moderate posttraumatic hypothermia decreases early calpain-mediated proteolysis and concomitant cytoskeletal compro-mise in traumatic axonal injury. Exp. Neurol. 159, 319–328.
BUKI, A., OKONKWO, D.O., and POVLISHOCK, J.T.(1999b). Postinjury cyclosporin A administration limits ax-onal damage and disconnection in traumatic brain injury. J.Neurotrauma 16, 511–521.
BUKI, A., OKONKWO, D.O., WANG, K.K., and POVLI-SHOCK, J.T. (2000). Cytochrome c release and caspase ac-tivation in traumatic axonal injury. J. Neurosci. 20, 2825–2834.
BUKI, A., FARKAS, O., DOCZI, T., and POVLISHOCK, J.T.(2003). Preinjury administration of the calpain inhibitor
TAMÁS ET AL.
692
MDL-28170 attenuates traumatically induced axonal injury.J. Neurotrauma 20, 261–268.
BULLOCK, R. (1993). Opportunities for neuroprotective drugsin clinical management of head injury. J. Emerg. Med. 11,Suppl 1, 23–30.
CANONICO, P.L., COPANI, A., D’AGATA, V., et al. (1996).Activation of pituitary adenylate cyclase activating polypep-tide receptors prevents apoptotic cell death in cultured cere-bellar granule cells. Ann. N.Y. Acad. Sci. 805, 470–472.
CERNAK, I., O’CONNOR, C., and VINK, R. (2002). Inhibi-tion of cyclooxygenase 2 by nimesulide improves cognitiveoutcome more than motor outcome following diffuse trau-matic brain injury in rats. Exp. Brain Res. 147, 193–199.
DELGADO, M., POZO, D., MARTINEZ, C., et al. (1999). Va-soactive intestinal peptide and pituitary adenylate cyclase-ac-tivating polypeptide inhibit endotoxin-induced TNF-� pro-duction by macrophages: in vitro and in vivo studies. J.Immunol. 162, 2358–2367.
DELGADO, M., JONAKAIT, G.M., and GANEA, D. (2002).Vasoactive intestinal peptide and pituitary adenylate cyclaseactivating polypeptide inhibit chemokine production in acti-vated microglia. Glia 39, 148–161.
DELGADO, M., LECETA, J., and GANEA, D. (2003). Va-soactive intestinal peptide and pituitary adenylate cyclase ac-tivating polypeptide inhibit the production of inflammatorymediators by activated microglia. J. Leukoc. Biol. 73,155–164.
DENG, H., HAN, H.S., CHENG, D., SUN, G.H., and YENARI,M.A. (2003). Mild hypothermia inhibits inflammation afterexperimental stroke and brain inflammation. Stroke 34,2495–2501.
DIETRICH, W.D., ALONSO, O., BUSTO, R., and FINKLE-STEIN, S.P. (1996) Posttreatment with intravenous basic fi-broblast growth factor reduces histopathological damage fol-lowing fluid-percussion brain injury in rats. J. Neurotrauma13, 309–316.
DIRNAGL, U., TANABE, J., and PULSINELLI, W. (1990).Pre- and post-treatment with MK-801 but not pretreatmentalone reduces neocortical damage after focal cerebral ischemia in the rat. Brain Res. 527, 62–68.
FARKAS, O., TAMAS, A., ZSOMBOK, A., et al. (2004). Ef-fects of pituitary adenylate cyclase activating polypeptide(PACAP) in a rat model of traumatic brain injury. Regul.Peptides 123, 69–75.
FIGIEL, M., and ENGELE, J. (2000). Pituitary adenylate cy-clase activating polypeptide (PACAP), a neuron-derived pep-tide regulating glial glutamate transport and metabolism. J.Neurosci. 20, 3596–3605.
FINK, K.B., ANDREWS, L.J., BUTLER, W.E., et al. (1999).Reduction of post-traumatic brain injury and free radical pro-duction by inhibition of the caspase-1 cascade. Neuroscience94, 1213–1218.
FODA, M.A., and MARMAROU, A. (1994). A new model ofdiffuse brain injury in rats. Part II. Morphological character-ization. J. Neurosurg. 80, 301–313.
FRECHILLA, D., GARCIA-OSTA, A., PALACIOS, S., CEN-ARRUZABEITIA, E., and DEL RIO J. (2001). BDNF me-diates the neuroprotective effect of PACAP-38 on rat corti-cal neurons. Neuroreport 12, 919–923.
GANEA, D., and DELGADO, M. (2002). Vasoactive intesti-nal peptide (VIP) and pituitary adenylate cyclase activatingpolypeptide (PACAP) as modulators of both innate and adap-tive immunity. Crit. Rev. Oral Biol. Med. 13, 229–237.
GENTLEMAN, S.M., NASH, M.J., SWEETING, C.J., GRA-HAM, D.I., and ROBERTS, G.W. (1993). Beta-amyloid pre-cursor protein (beta-APP) as a marker for axonal injury af-ter head injury. Neurosci. Lett. 160, 139–144.
GLOBUS, M.Y., ALONSO, O., DIETRICH, W.D., BUSTO,R., and GINSBERG, M.D. (1995). Glutamate release and freeradical production following brain injury: Effects of post-traumatic hypothermia. J. Neurochem. 65, 1704–1711.
GODA, M., ISONO, M., FUJIKI, M., and KOBAYASHI, H.(2002). Both MK801 and NBQX reduce the neuronal dam-age after impact-acceleration brain injury. J. Neurotrauma 19,1445–1456.
HEATH, D.L., and VINK, R. (1999). Improved motor outcomein response to magnesium therapy received up to 24 hoursafter traumatic diffuse axonal brain injury in rats. J. Neuro-surg. 90, 504–509.
HONG, S.C., GOTO, Y., LANZINO, G., SOLEAU, S., KAS-SELL, N.F., and LEE, K.S. (1994). Neuroprotection with acalpain inhibitor in a model of focal cerebral ischemia. Stroke25, 663–669.
KADOYA, C., DOMINO, E.F., YANG G., STERN J.D., andBETZ, A.L. (1995). Preischemic but not postischemic zincprotoporphirin treatment reduces infarct size and edema ac-cumulation after temporary focal cerebral ischemia in rats.Stroke 26, 1035–1038.
KATAHIRA, M., YONE, K., ARISHIMA, Y., et al. (2003).The neuroprotective effects of PACAP on spinal cord injury(SCI) in rats. Regul. Peptides 115, 49(abst).
KAWAMATA, T., DIETRICH, W.D., SCHALLERT, T., et al.(1997). Intracisternal basic fibroblast growth factor enhancesfunctional recovery and up-regulates the expression of a mo-lecular marker of neuronal sprouting following focal cere-bral infarction. Proc. Natl. Acad. Sci. USA 22, 8179–8184.
KIM, W.K., KAN, Y., GANEA, D., HART, R.P., GOZES, I.,and JONAKAIT M. (2000). Vasoactive intestinal peptide andpituitary adenylate cyclase activating polypeptide inhibit tu-mor necrosis factor-� production in injured spinal cord andin activated microglia via cAMP-dependent pathway. J. Neu-rosci. 20, 3622–3630.
KIMURA, H., KAWATANI, M., ITO, E., and ISHIKAWA, K.(2003). Effects of pituitary adenylate cyclase activating
POSTINJURY ADMINISTRATION OF PACAP IN TBI
693
polypeptide on facial nerve recovery in the guinea pig. Laryn-goscope 113, 1000–1006.
KINHULT, J., UDDMAN, R., LAAN, M., LINDEN, A., andCARDELL, L.O. (2001). Pituitary adenylate cyclase acti-vating polypeptide inhibits neurophil chemotaxis. Peptides22, 2151–2154.
KOIZUMI, H., and POVLISHOCK, J.T. (1998). Posttraumatichypothermia in the treatment of axonal damage in an animalmodel of traumatic axonal injury. J. Neurosurg. 89, 303–309.
KOIZUMI, H., FUJISAWA, H., ITO, H., MAEKAWA, T., DI,X., and BULLOCK, R. (1997). Effects of mild hypothermiaon cerebral blood flow-independent changes in cortical ex-tracellular levels of amino acids following contusion traumain the rat. Brain Res. 747, 304–312.
LEE, V.M., CARDEN, M.J. SCHLAEPFER, W.W., and TRO-JANOWSKI, J.Q. (1987). Monoclonal antibodies distinguishseveral differentially phosphorylated states of the two largestrat neurofilament subunits (NF-H and NF-M) and demon-strate their existence in the normal nervous system of adultrats. J. Neurosci. 7, 3474–3488.
LEKER, R.R., and SHOHAMI E. (2002). Cerebral ischemiaand trauma—different etiologies yet similar mechanisms:neuroprotective opportunities. Brain Res. Rev. 39, 55–73.
LIU, J., JR., FARMER, D.J., LANE, W.S., FRIEDMAN, J.,WEISSMAN, I., and SCHREIBER, S.L. (1991). Calcineurinis a common target of cyclophilin-cyclosporin A and FKBP-FK506 complexes. Cell 66, 807–815.
LYETH, B.G., JIANG, J.Y., and LIU, S. (1993). Behavioralprotection by moderate hypothermia initiated after experi-mental traumatic brain injury. J. Neurotrauma 10, 57–64.
MARMAROU, A., FODA, M.A., VAN DEN BRINK, W.,CAMPBELL, J., KITA, H., and DEMETRIADOU, K.(1994). A new model of diffuse brain injury in rats. Part I:Pathophysiology and biomechanics. J. Neurosurg. 80, 291–300.
MAXWELL, W.L., POVLISHOCK, J.T., and GRAHAM, D.L.(1997). A mechanistic analysis of nondisruptive axonal in-jury: a review. J. Neurotrauma 14, 419–440.
MITANI, A., KADOYA, F., and KATAOKA, K. (1991). Tem-perature dependence of hypoxia-induced calcium accumula-tion in gerbil hippocampal slices. Brain Res. 562, 159–163.
MORIO, H., TATSUNO, I., HIRAI, A., TAMURA, Y., andSAITO, Y. (1996). Pituitary adenylate cyclase activatingpolypeptide protects rat-cultured cortical neurons from glu-tamate-induced cytotoxicity. Brain Res. 741, 82–88.
NAKAYAMA, M., UCHIMURA, K., ZHU, R.L., et al. (1998).Cyclooxygenase-2 inhibition prevents delayed death of CA1hippocampal neurons following global ischemia. Proc. Natl.Acad. Sci. USA 95, 10954–10959.
OHTAKI, H., DOHI, K., YIN, T., TAKAKI, A., NAKAJO, S.,and SHIODA S. (2003). Neuroprotective effect of PACAP38
through IL-6 after focal ischemia in mouse. Regul. Peptides15, 53(abst).
OKONKWO, D.O., PETTUS, E.H., MOROI, J., and POVLI-SHOCK, J.T. (1998). Alteration of the neurofilament sidearmand its relation to neurofilament compaction occurring withtraumatic axonal injury. Brain Res. 784, 1–6.
OKONKWO, D.O., and POVLISHOCK, J.T. (1999). An in-trathecal bolus of cyclosporin A before injury preserves mi-tochondrial integrity and attenuates axonal disruption in trau-matic brain injury. J. Cereb. Blood Flow Metab. 19, 443–451.
OKONKWO, D.O., BUKI, A., SIMON, R., and POVLI-SHOCK, J.T. (1999) Cyclosporin A limits calcium-inducedaxonal damage following traumatic brain injury. Neuroreport10, 353–358.
OTANI, N., NAWASHIRO, H., FUKUI, S., et al. (2002). Tem-poral and spatial profile of phosphorylated mitogen-activatedprotein kinase pathways after lateral fluid percussion injuryin the cortex of the rat brain. J. Neurotrauma 12, 1587–1596.
PETTUS, E.H., CHRISTMAN, C.W., GIEBEL, M.L., andPOVLISHOCK, J.T. (1994). Traumatically induced alteredmembrane permeability: its relationship to traumatically in-duced reactive axonal change. J. Neurotrauma 11, 507–522.
POVLISHOCK, J.T., and STONE, J.R. (2001). Traumatic ax-onal injury, in: Head Trauma: Basic, Preclinical and Clini-cal Directions. L.P. Miller and R.L. Hayes (eds), Wiley: NewYork.
POVLISHOCK, J.T., MARMAROU, A., MCINTOSH, T.,TROJANOWSKI, J.Q., and MOROI, J. (1997). Impact ac-celeration injury in the rat: evidence for focal axolemmalchange and related neurofilament sidearm alteration. J. Neu-ropathol. Exp. Neurol. 56, 347–359.
REGLODI, D., SOMOGYVARI-VIGH, A., VIGH, S., KOZ-ICZ, T., and ARIMURA, A. (2000). Delayed systemic ad-ministration of PACAP38 is neuroprotective in transient mid-dle cerebral artery occlusion in the rat. Stroke 31, 1411–1417.
REGLODI, D., TAMAS, A., SOMOGYVARI-VIGH, A., et al.(2002). Effects of pretreatment with PACAP on the infarctsize and functional outcome in rat permanent focal cerebralischemia. Peptides 23, 2227–2234.
REGLODI, D., LUBICS, A., TAMAS, A., SZALONTAY, L.,and LENGVARI, I. (2004). Pituitary adenylate cyclase acti-vating polypeptide protects dopaminergic neurons and im-proves behavioral deficits in a rat model of Parkinson’s dis-ease. Behav. Brain Res. 151, 303–312.
SAKAMOTO, K.I., FUJISAWA, H., KOIZUMI, H., et al.(1997). Effects of mild hypothermia on nitric oxide synthe-sis following contusion trauma in the rat. J. Neurotrauma 14,349–353.
SANDERSON, K.L., RAGHUPATHI, R., SAATMAN, K.E.,MARTIN, D., MILLER G., and MCINTOSH, T.K. (1999).Interleukin-1 receptor antagonist attenuates regional neuronal
TAMÁS ET AL.
694
cell death and cognitive dysfunction after experimental braininjury. J. Cereb. Blood Flow Metab. 19, 1118–1125.
SEKI, T., IZUMI, S., SHIODA, S., and ARIMURA A. (2003).Pituitary adenylate cyclase activating polypeptide (PACAP)protects ganglion cell death against cutting of optic nerve inthe rat retina. Regul. Peptides 115, 55(abst).
SHIODA, S., OZAWA, H., DOHI, K., et al. (1998). PACAPprotects hippocampal neurons against apoptosis: involvementof JNK/SAPK signaling pathway. Ann. N. Y. Acad. Sci. 865,111–117.
SHOGE, K., MISHIMA, H.K., SAITOH, T., et al. (1999). At-tenuation by PACAP of glutamate-induced neurotoxicity incultured retinal neurons. Brain Res. 839, 66–73.
SINGLETON, R.H., STONE J.R., OKONKWO D.O., PELLI-CANE, A.J., and POVLISHOCK, J.T. (2001). The im-munophilin ligand FK506 attenuates axonal injury in an im-pact-acceleration model of traumatic brain injury. J.Neurotrauma 18, 607–614.
SKOGLOSA, Y., LEWEN, A., TAKEI, N., HILLERED, L.,and LINDHOLM, D. (1999). Regulation of pituitary adeny-late cyclase activating polypeptide and its receptor type 1 af-ter traumatic brain injury: comparison with brain-derivedneurotrophic factor and the induction of neuronal cell death.Neuroscience 90, 235–247.
SOMOGYVARI-VIGH, A., and REGLODI, D. (2004). Pitu-itary adenylate cyclase activating polypeptide: a potentialneuroprotective peptide. Curr. Pharm. Des. 10, 2861–2889.
STONE, J.R., WALKER, S.A., and POVLISHOCK, J.T.(1999). The visualisation of a new class of traumatically in-jured axons through the use of a modified method of mi-crowave antigen retrieval. Acta Neuropathol. 97, 335–345.
STONE, J.R., SINGLETON, R.H., and POVLISHOCK, J.T.(2000). Antibodies to the C-terminus of the beta-amyloid pre-cursor protein (APP): a site-specific marker for the detectionof traumatic axonal injury. Brain Res. 871, 288–302.
SUEHIRO, E., SINGLETON, R.H., STONE, R.J., and POVLI-SHOCK, J.T. (2001). The immunophilin ligand FK 506 at-tenuates the axonal damage associated with rapid rewarmingfollowing posttraumatic hypothermia. Exp. Neurol. 172,199–210.
TABUCHI, A., FUNAJI, K., NAKATSUBO, J., FUKUCHI,M., TSUCHIYA, T., and TSUDA, M. (2003). Inactivationof aconitase during apoptosis of mouse cerebellar granuleneurons induced by a deprivation of membrane depolariza-tion. J. Neurosci. Res. 71, 504–515.
TANAKA, J., KOSHIMURA, K., MURAKAMI, Y., SOHMIYA,M., YANAIHARA, N., and KATO, Y. (1997). Neuronal pro-
tection from apoptosis by pituitary adenylate cyclase activatingpolypeptide. Regul. Peptides 72, 1–8.
TRUETTNER, J., SCHMIDT-KASTNER, R., BUSTO, R., et al.(1999) Expression of brain-derived neurotrophic factor, nervegrowth factor, and heat shock protein HSP70 following fluidpercussion brain injury in rats. J. Neurotrauma 16, 471–486.
TSUDA, T., KOGURE, K., NISHIOKA, K., and WATAN-ABE, T. (1991). Mg2� administered up to twenty-four hoursfollowing reperfusion prevents ischemic damage of the CA1neurons in the rat hippocampus. Neuroscience 44, 335–341.
UCHIDA, D., ARIMURA, A., SOMOGYVARI-VIGH, A., SHI-ODA, S., and BANKS, W. (1996). Prevention of ischemia-induced death of hippocampal neurons by pituitary adenylate-cyclase activating polypeptide. Brain Res. 736, 280–286.
VAUDRY, D., GONZALEZ, B.J., BASILLE, M., YON, L.,FOURNIER, A., and VAUDRY, H. (2000a). Pituitary adeny-late cyclase activating polypeptide and its receptors: fromstructure to functions. Pharmacol. Rev. 52, 269–324.
VAUDRY, D., GONZALEZ, B.J., BASILLE, M., et al.(2000b). The neuroprotective effect of pituitary adenylate cy-clase activating polypeptide on cerebellar granule cells is me-diated through inhibition of the CED3-related cystein pro-tease caspase-3/CPP32. Proc. Natl. Acad. Sci. USA 97,13390–13395.
VAUDRY, D., ROUSELLE, C., BASILLE, M., et al. (2002a).Pituitary adenylate cyclase activating polypeptide protects ratcerebellar granule neurons against ethanol-induced apoptoticcell death. Proc. Natl. Acad. Sci. USA 99, 6398–6403.
VAUDRY, D., CHEN, Y., RAVNI, A., HAMALINK, C.,ELKAHLOUN, A.G., and EIDEN, L.E. (2002b). Analysisof the PC12 cell transcriptome after differentiation with pi-tuitary adenylate cyclase activating polypeptide (PACAP). J.Neurochem. 83, 1272–1284.
YOSHIMOTO, T., and SIESJO, B.K. (1999). Posttreatmentwith the immunsuppressant cyclosporin A in transient focalischemia. Brain Res. 839, 283–291.
ZAMBONI, L., and MARTINO, L. (1967). Buffered picricacid–formaldehyde: a new rapid fixative for electron mi-croscopy. J. Cell Biol. 35, 148A.
Address reprint requests to:Andrea Tamas, M.D.
Department of AnatomyUniversity of Pécs
Pécs, Hungary
E-mail: [email protected]
POSTINJURY ADMINISTRATION OF PACAP IN TBI
695
Neurobiology of Disease
Mechanoporation Induced by Diffuse Traumatic BrainInjury: An Irreversible or Reversible Response to Injury?
Orsolya Farkas,1,2 Jonathan Lifshitz,1 and John T. Povlishock1
1Department of Anatomy and Neurobiology, Medical College of Virginia Campus of Virginia Commonwealth University, Richmond, Virginia 23298, and2Department of Neurosurgery, Medical Faculty of Pecs University, H-7623 Pecs, Hungary
Diffuse traumatic brain injury (DTBI) is associated with neuronal plasmalemmal disruption, leading to either necrosis or reactive changewithout cell death. This study examined whether enduring membrane perturbation consistently occurs, leading to cell death, or if thereis the potential for transient perturbation followed by resealing/recovery. We also examined the relationship of these events to calpain-mediated spectrin proteolysis (CMSP). To assess plasmalemmal disruption, rats (n � 21) received intracerebroventricular infusion 2 hbefore DTBI of a normally excluded 10 kDa fluorophore-labeled dextran. To reveal plasmalemmal resealing or enduring disruption, ratswere infused with another labeled dextran 2 h (n � 10) or 6 h (n � 11) after injury. Immunohistochemistry for the 150 kDa spectrinbreakdown product evaluated the concomitant role of CMSP. Neocortical neurons were followed with confocal and electron microscopy.After DTBI at 4 and 8 h, 55% of all tracer-flooded neurons contained both dextrans, demonstrating enduring plasmalemmal leakage, withmany demonstrating necrosis. At 4 h, 12.0% and at 8 h, 15.7% of the dual tracer-flooded neurons showed CMSP, yet, these demonstratedless advanced cellular change. At 4 h, 39.0% and at 8 h, 24.4% of all tracer-flooded neurons revealed only preinjury dextran uptake,consistent with membrane resealing, whereas 7.6 and 11.1%, respectively, showed CMSP. At 4 h, 35% and at 8 h, 33% of neuronsdemonstrated CMSP without dextran flooding. At 4 h, 5.5% and at 8 h, 20.9% of tracer-flooded neurons revealed only postinjury dextranuptake, consistent with delayed membrane perturbation, with 55.0 and 35.4%, respectively, showing CMSP. These studies illustrate thatDTBI evokes evolving plasmalemmal changes that highlight mechanical and potential secondary events in membrane poration.
Key words: diffuse traumatic brain injury; neuron; dextrans; membrane disruption; membrane resealing; calpain
IntroductionOur understanding of diffuse traumatic brain injury (DTBI) haslong moved on the assumption that the forces of injury evokefocal as well as diffuse neuronal death, contributing to morbidityand mortality. However, little work exists to support thispremise. In animals and humans, contusions are associated withregional neuronal death, together with the finding of scatteredneuronal death in the neocortex, hippocampus, and diencepha-lon (Cortez et al., 1989; Dietrich et al., 1994a,b; Colicos et al.,1996; Hicks et al., 1996; Saatman et al., 1996). In these cases, bothapoptotic and necrotic cell death have been identified, with theassumption that the observed neuronal death proceeds fromtraumatically induced neuroexcitation, oxygen radical-mediateddamage and/or secondary insults. Few have considered the po-tential that the forces of injury directly cause mechanically in-duced neuronal perturbation or death. Recently, after DTBI, weidentified axotomy-mediated change in the neuronal somata as-
sociated with impaired, yet transient protein translation (Single-ton et al., 2002). We also observed progressive necrosis in neu-rons adjacent to the axotomized neurons. Studies usingextracellular tracers normally excluded by intact neuronal mem-branes revealed that these necrotic neurons suffered alteredtracer permeability at the moment of injury most likely allowingfor influx of damaging ions (Singleton and Povlishock, 2004).Other neurons in the same field, however, also took up thesenormally excluded tracers, yet reorganized them over time, man-ifesting no evidence of cell death. These observations led to thepremise that the DTBI evoked direct mechanical poration of theneuronal cell membrane, which, if enduring, contributed to celldeath. Conversely, it was posited that those neurons that con-tained tracers, yet remained viable, resealed their mechanicallydamaged membranes. In this communication, we tested this hy-pothesis by reexamining the potential for traumatically inducedmechanical poration of neuronal membranes while evaluatingwhether continuing membrane perturbation leads to cell death.We also explored the potential that some neuronal membranesmay reseal and recover. Fluorescently labeled tracers were admin-istered both before and after injury to evaluate poration and itscontinuance. Parallel immunocytochemical investigations exam-ined calpain-mediated spectrin proteolysis (CMSP) long associ-ated with the damaging proteolytic death cascades activated inDTBI (Kampfl et al., 1996, 1997; Saatman et al., 1996; Newcombet al., 1997; Pike et al., 1998; Buki et al., 1999; McCracken et al.,
Received Aug. 18, 2005; revised Jan. 16, 2006; accepted Feb. 8, 2006.This work was supported by National Institutes of Health Grants R01 NS-045834, F32 HD-49343, and NS-47463
and by the Commonwealth Center for Traumatic Brain Injury. We thank Sue Walker, Lynn Davis, and Tom Coburn fortheir technical assistance. We also thank for Dr. Robert Hamm for his advice on statistical analysis and Dr. ScottHenderson for his help with confocal microscopy.
Correspondence should be addressed to Dr. John T. Povlishock, Department of Anatomy and Neurobiology,Medical College of Virginia Campus of Virginia Commonwealth University, P.O. Box 980709, Richmond, VA 23298.E-mail: [email protected].
DOI:10.1523/JNEUROSCI.5119-05.2006Copyright © 2006 Society for Neuroscience 0270-6474/06/263130-11$15.00/0
3130 • The Journal of Neuroscience, March 22, 2006 • 26(12):3130 –3140
1999). In fact, the significance of CMSP is highlighted by exper-imental and clinical studies that have suggested that CSF CMSPbreakdown products are surrogate markers of morbidity andmortality (Pike et al., 2001; Farkas et al., 2005). Confocal micros-copy and concomitant ultrastructural analyses using antibodiestargeting the fluorophore-tagged tracers confirmed that TBIcaused direct mechanical poration in numerous neuronal somatascattered throughout the neocortex. Over 50% of the tracer-containing neurons were labeled with both tracers with many ofthese revealing necrotic cell death. Other flooded neurons con-tained only one tracer, consistent either with membrane resealingor delayed opening. Surprisingly, CMSP did not frequently colo-calize with neurons, demonstrating tracer influx and/or necrosis.Collectively, these studies illustrate the complex and evolvingplasmalemmal changes ongoing with DTBI.
Materials and MethodsIn all, 21 adult male Sprague Dawley rats weighing 380 – 400 g were usedfor the experiments (n � 10 in 4 h survival group; n � 11 in 8 h survivalgroup). Four animals served as tracer-infused, sham-injured controls todemonstrate that infusion into the lateral ventricle itself does not lead toneuronal cellular membrane disruption. Ten noninjected, injured ani-mals were used as intracranial pressure (ICP) controls to obviate theconcern that any tracer infusion-induced rise in ICP could constitute aconfound.
Surgical preparation and injury induction. For the induction of anes-thesia, animals were exposed to 4% isoflurane in a mixture of 30% O2 and70% N2O for 5 min, then intubated and ventilated with 1–2% isofluranein the mixture of 30% O2 and 70% N2O. Anesthesia was maintainedthroughout the duration of the experiment until transcardial perfusion.Animals were placed on a feedback-controlled heating pad (HarvardApparatus, Holliston, MA) to maintain body temperature at 37°C duringsurgery. A PE50 polyethylene tube (Becton Dickinson, Sparks, MD) wasplaced into the left femoral artery to monitor the mean arterial bloodpressure (MABP) and blood gases. After cannulation animals wereplaced in a stereotaxic frame (David Kopf Instruments, Tujunga, CA).The skull between the coronal and lambdoidal sutures was exposed witha midline incision. A 2 mm burr hole was drilled in the right parietalbone. A 26 gauge needle connected to a pressure transducer and a micro-infusion pump via a PE50 polyethylene tube was placed into the lateralventricle 0.5 mm posterior and 1.3 mm lateral from bregma through theburr hole. During the needle placement, sterile saline was infused with a3 �l/min rate within a closed fluid-pressure system, while intracranialpressure monitoring was obtained via a PowerLab system (ADInstru-ments, Colorado Springs, CO). With the advancement of the needle, thedetection of a significant pressure drop indicated the moment that theneedle had broached the ventricular wall (Amorini et al., 2003). Thiscontrolled and monitored needle placement assured consistent needleplacement within the ventricular compartment without the potentialconfound of brain placement and its potentially damaging consequenceson tracer infusion. After placing the needle into the ventricle, 75 �l ofAlexa Fluor 488-labeled 10 kDa molecular weight dextran (Invitrogen,San Diego, CA) was infused into the ventricle at 2 �l/min rate using aCMA/100 microinjection pump (Carnegie Medicin, Stockholm, Swe-den), with continuous intracranial pressure monitoring. The dextranwas diluted in sterile saline in 26.67 mg/ml concentration; the finalamount infused was 5 mg/kg. The needle was allowed to remain in theventricle for 10 min after completion of the infusion, after which theneedle was slowly removed. Gelfoam and bone wax were used to restoreskull integrity before the induction of closed head injury. Next, the ani-mal was prepared for injury as described in previous protocols (Foda andMarmarou, 1994; Marmarou et al., 1994). Briefly, a metallic disc-shapedhelmet of 10 mm diameter was firmly glued between the bregma andlambdoid sutures of the skull. The animal was disconnected from theventilator and placed in a prone position on a foam bed with the metallichelmet centered under the edge of a Plexiglas tube. The rat was preventedfrom falling by two belts secured to a foam bed of known spring constant
(Type E bed; Foam to Size, Ashland, VA). Brass weights weighing 450 gwere allowed to fall from a height of 2 m through the Plexiglas tubedirectly to the metallic disc fixed to the animal’s skull, a setting thatprecluded cerebral contusion or subdural hemorrhage. Rebound impactwas prevented by sliding the foam bed containing the animal away fromthe tube immediately after the initial impact. Each animal sustained in-jury 2 h after preinjury tracer administration. Noninjured controls wereprepared for injury in the same manner but were not injured. Immedi-ately after the injury, the animal was reconnected to the ventilator. Thehelmet was removed and the skull was studied for any sign of fracture,which, if found, disqualified the animal from further evaluation. Thescalp wound was then sutured while the animal remained on artificialventilation.
Delayed tracer infusion. Before the second tracer infusion, animalswere returned to the stereotaxic device, the skull was exposed, and an-other hole was drilled through the left parietal bone. Texas Red-labeled10 kDa dextran (Invitrogen) was then infused at the same concentrationas the preinjury dextran. Once again, this was accomplished via a needleplaced in the left ventricle 0.5 mm posterior and 1.3 mm lateral to bregmawith 2 �l/min infusion rate with continuous ICP monitoring. This tracerwas infused either at 2 h after injury (n � 10) or 6 h after injury (n � 11).The tracer was then allowed to diffuse for additional 2 h before perfusionwith fixative, resulting in the respective 4 and 8 h postinjury survivalgroups. Sham-injured animals were infused either 4 or 8 h after the firstinfusion. Non-injected-injured animals (n � 10) were prepared in thesame manner for ICP monitoring and injury, but no tracers were infusedin their ventricles. Rather, ICP was monitored through a needle placed inthe ventricle at the same time points used in injected animals. Five min-utes before transcardial perfusion, the animals were injected with 150mg/kg euthasol. At either 4 or 8 h after injury or at either 6 or 10 h afterthe first infusion (sham injury), the animals underwent transcardial per-fusion with saline and then 4% paraformaldehyde in Millonig’s buffer.
Physiological assessment. MABP was monitored via a femoral line ineach animal for the duration of the surgical procedures. Blood gases wereanalyzed before the tracer infusions as well as before and after injury. Asnoted above, ICP was also monitored before, during, and after tracerinfusion or throughout the duration of all procedures in noninjectedanimals. Body temperature was monitored via a rectal probe and main-tained at 37°C using a feedback control heating pad. PowerLab was usedto monitor all physiological parameters (MABP, ICP, temperature).
Tissue processing. After perfusion, the brains were removed and stored in4% paraformaldehyde in Millonig’s buffer at 4°C for 24 h, after which theywere coronally blocked at the optic chiasm and midbrain and sectioned in0.1 M phosphate buffer with a vibratome (Leica, Banockburn, IL) at a thick-ness of 40 �m. Sections were collected serially in 24-well culture plates andstored in Millonig’s buffer at 4°C until further processing.
Multiple fluorescent analysis. Because dextrans infused to the brainwere labeled with fluorophores (Alexa Fluor 488 for the preinjury dex-tran and Texas Red for postinjury dextran), no additional steps wereneeded to visualize the tracer or any tracer-containing neurons. In addi-tion to the identification of tracer-containing neurons, antibody againstthe 150 kDa calpain-specific spectrin breakdown product was used toexamine the link between any observed membrane perturbation and theensuing induction of calpain-mediated spectrin proteolysis, previouslylinked to TBI-induced proteolytic cascades. The antibody Ab 38 was usedfor this approach, because it is well characterized, targeting theN-terminal fragment of the 150 kDa calpain-specific breakdown of �II-spectrin, a cytoskeletal protein known to be cleaved after traumatic braininjury (Siman et al., 1989; Roberts-Lewis et al., 1994). In this procedure,some sections were washed in PBS three times for 10 min and thensubjected to temperature-controlled microwave antigen retrieval as de-scribed previously (Stone et al., 1999). After retrieval, sections were againrinsed in PBS three times for 10 min and treated for 60 min in 10%normal goat serum (NGS) with 0.2% Triton X-100 in PBS and 5% BSA.The tissue was then incubated in a 1:5000 dilution of Ab 38 (generous giftfrom Dr. R. Siman, University of Pennsylvania, Philadelphia, PA) in 1%NGS/PBS with 1% BSA overnight. The following day, sections wererinsed three times for 10 min 1% NGS/PBS and then incubated for 2 h ina 1:1500 dilution of Alexa Fluor 647-labeled anti-rabbit IgG (Invitrogen)
Farkas et al. • The Consequences of Mechanoporation in DTBI J. Neurosci., March 22, 2006 • 26(12):3130 –3140 • 3131
in 1% NGS/PBS. After rinsing three times for 10 min in PBS and twice for10 min in 0.1 M phosphate buffer, sections were mounted on nontreatedslides and coverslipped (ProLong; Invitrogen).
Visualization of fluorophores at ultrastructural level. Because the fluo-rescent dyes that were used are not detectable at electron microscopiclevel, for parallel assessments of the ultrastructural change, conversion offluorophores to a stable, electron-dense product was performed. Thiswas achieved through the use of an antibody against the fluorophoresthemselves. The same sections previously scanned with confocal micro-scope for tracer uptake or individually captured with a Spot RT cameraattached to an epifluorescent Nikon (Tokyo, Japan) Eclipse E800 micro-scope were used to provide a visual fluorescent map to which the subse-quently prepared chromagen-based sections could be directly comparedat the LM and EM levels. After capture, the tissues were soaked off theslides, and brain regions demonstrating preinjury and/or postinjurytracer uptake were removed and processed to investigate any ultrastruc-tural change related to membrane disruption by using antibodies tar-geted to Alexa Fluor 488. This method allowed the identification of thesame cells that previously revealed evidence of the uptake of either one orboth dextran tracers, allowing the ultrastructural assessment of any al-teration associated with enduring membrane change and/or membraneresealing. In this approach, sections were postfixed with 0.1% glutaral-dehyde in 4% paraformaldehyde, then rinsed five times for 10 min inMillonig’s buffer and three times for 10 min in PBS. Endogenous perox-idase activity within the tissue was blocked with 0.5% H2O2 in PBS for 30min. Sections were processed using the temperature-controlled micro-wave antigen retrieval approach described above. After microwave anti-gen retrieval, sections were preincubated for 60 min in 10% NGS with0.2% Triton X-100 in PBS. The tissue was incubated overnight in a1:5000 dilution of rabbit anti-Alexa Fluor 488 antibody (Invitrogen) in1% NGS in PBS. Sections were then incubated for 1 h in biotinylatedanti-rabbit antibody (IgG; Vector Laboratories, Burlingame, CA) diluted1:2000 in 1% NGS in PBS and then for 1 h in a 1:200 dilution of anavidin– horseradish peroxidase complex (ABC Standard Elite kit; VectorLaboratories). The reaction product was visualized with 0.05% diamino-benzidine, 0.01% hydrogen peroxide, and 0.3% imidazole in 0.1 M phos-phate buffer for 10 –20 min.
Visualization of calpain-mediated spectrin proteolysis at the ultrastruc-tural level. Brain regions demonstrating evidence for tracer uptake withconfocal microscopy were removed and processed to investigate CMSP
using the previously described Ab 38 antibody. In this approach, afterpostfixation tissues were processed for EM analysis as described above,using Ab 38 as primary antibody in 1:8000 dilution and biotinylatedanti-rabbit antibody in 1:2000 dilution.
Electron microscopy. Using confocal and routine fluorescent images asdescribed above, the once-fluorescent cells, which now contained a per-oxidase reaction product, were prepared for EM. In this approach, thepreviously prepared fluorescent maps were used to identify specific sitesof interest. These were dissected out of the tissue sections, and thenosmicated, dehydrated, and embedded in epoxy resins on plastic slideswith plastic coverslips. After resin curing, the plastic slides were studiedwith routine LM to identify the precise neurons of interest. Once identi-fied, these sites were removed, mounted on plastic studs, and thick sec-tioned to the depth of the immunoreactive sites of interest. Serial 70 nmsections were cut and picked up on to Formvar-coated slotted grids. Thegrids then were stained in 5% uranyl acetate in 50% methanol for 2 minand 0.5% lead citrate for 1 min. Ultrastructural analysis was performedusing a JEM 1230 electron microscope (JEOL-USA, Peabody, MA). Spe-cifically, at 1000�, the grid field was scanned to identify the cell(s) ofinterest. Once identified, these cells were digitally acquired at 5000� as amontage. These digital montages were copied to a digital video disk andtransferred to a Dell Dimension XPS Gen 4 computer (Dell ComputerCompany, Round Rock, TX), on which the images were further enlargedand assessed for any evidence of cellular/subcellular change.
Confocal fluorescent microscopy. Zeiss (Oberkochen, Germany) 510Meta and Leica TCS SP2 AOBS laser scanning confocal microscope sys-tems were used to analyze double- and/or triple-labeled sections. Appro-priate laser lines (argon 488 nm for Alexa Fluor 488, He-Ne 594 nm forTexas Red, and He-Ne 633 nm for Alexa Fluor 647) and emission filters(Zeiss) or emission bands (Leica) were used for exciting and detectingfluorescent signals.
Semiquantification of fluorescent images. The TBI model used in thecurrent study evokes diffuse injury throughout the neocortex, resultingin scattered neuronal injury with heterogeneous pathologies. Thesemodest numbers of damaged neurons, particularly those revealing onlyone tracer, would violate a basic tenant of stereology and preclude further
Figure 1. Bar charts representing the mean levels of intracranial pressure measured 1 and2 h before injury as well as 2, 3, 6, and 7 h after injury (black column, noninfused injured, n�10;gray column, infused injured, 4 h survival, n � 10; white column, infused injured, 8 h survival,n � 11). No elevation in ICP was found in injected-non-injured sham animals, indicating thatcontrolled infusion into the lateral ventricle does not increase ICP (data not shown). No signif-icant difference in ICP was observed between injected-injured and non-injected-injured ani-mals at 4 or 8 h after injury, indicating that infusion into the lateral ventricle does not complicateany ICP change associated with closed head injury. Note, however, that in this injury model, thetraumatic episode was associated with a significant elevation of ICP. The ICP was elevated ateach time point measured postinjury, particularly at 2, 3, 6, and 7 h after injury. Data are shownas mean � SEM. The asterisk indicates statistical significance ( p � 0.001).
Figure 2. This double-labeled confocal image from a sham-injured animal receivingfluorophore-labeled dextran infusions demonstrates the diffusion of the tracers throughout theinterstices of the neocortex without any evidence of neuronal uptake of any of the adminis-trated dextrans. Note that in addition to their passage through the interstices of the brainparenchyma, the labeled dextrans can be easily visualized in the perivascular regions, consis-tent with a control pattern of dextran distribution (arrows). Scale bar, 100 �m.
3132 • J. Neurosci., March 22, 2006 • 26(12):3130 –3140 Farkas et al. • The Consequences of Mechanoporation in DTBI
stereological analysis of the number of neurons with membrane pertur-bation at 4 or 8 h after injury. Densitometric evaluation of injured neu-rons showing different dextran uptake profiles could not be performed,considering the high and variable background level of the fluorophoresassociated with the intraventricular administration of the labeled dex-trans, making the necessary standardization difficult. Based on thesedeficiencies, we conducted traditional cell counting using images cap-tured by confocal microscopy with rigorous inclusion criteria to semi-quantify neurons showing primary and/or secondary dextran uptakeboth at 4 and 8 h after injury. Sections captured at 40� magnification andcontaining at least three fluorescent labeled neurons were included(number of animals at 4 h, n � 8; at 8 h, n � 7; mean number of sections
per animal, 7.8). The definition for positive labeled neurons was all neu-rons showing any dextran flooding. Neurons showing preinjury dextranuptake, postinjury dextran uptake, or both preinjury and postinjury dextranuptake were quantified as a percentage of the total number of neurons dem-onstrating any tracer uptake in each section and averaged for each animal.Similar methods were used to quantify calpain mediated spectrin proteolysisand its relationship to membrane disruption.
Statistical analysis. Quantitative data are presented as mean � SEM.One-way ANOVA was used to compare the physiological parameters(MABP, pO2, pCO2, body temperature, and ICP) between the differentgroups or between different time points. For further analysis, Duncanpost hoc tests were performed, and a p value �0.05 was considered statis-
tically significant. To detect a redistribution oftracer-filled neurons over time after injury, se-rial � 2 tests between proportions of neuronswith specific membrane perturbations wereperformed. The three � 2 tests (resealed vs en-during, enduring vs delayed, resealed vs de-layed) were evaluated at p � 0.0167 to correctfor multiple analyses.
ResultsPhysiological assessmentsMean arterial blood pressure, pCO2, pO2,and pH were monitored and maintained atphysiological levels in each animal for theduration of the experiments. There wereno differences in these physiological pa-rameters between injected and nonin-jected animals, demonstrating no systemiceffects of the intracerebroventricular injec-tions (data not shown).
ICP was also monitored to reveal theeffects of preinjury and postinjury infu-sions on ICP as well as any ICP changerelated to diffuse traumatic brain injury.No elevation in ICP was found in the in-jected, noninjured sham animals. This in-dicated that the intraventricular infusioninto the lateral ventricle did not increaseICP itself (data not shown). No significantdifferences in ICP over time were observedbetween injected-injured and non-injected-injured animals at any time point,indicating that intracerebroventricular in-fusion had no effect on ICP change relatedto DTBI. However, it is important to notethat in this injury model, like the humancondition, the traumatic episode itself wasassociated with ICP elevation. Specifically,ICP was elevated 2, 3, 6, and 7 h after injurycompared with preinjury levels both innon-injected-injured and injected-injuredanimals (Fig. 1). In some cases, the ICPapproached 20 mmHg, a value consistentwith that seen in brain-injured humans(Miller et al., 1977; Marshall et al., 1979).
Evidence for altered neuronalmembrane permeability after DTBIAdministration of high-molecular-weightdextrans normally excluded from the neu-ronal and glial cytoplasm by intact cellmembranes was used to investigate alteredmembrane permeability related to diffuse
Figure 3. Confocal images of tracer-infused-injured animals. A, This double-labeled image demonstrates numerous corticalneurons (arrows) revealing intracellular tracer flooding with both the preinjury and postinjury infused dextrans. Scale bar, 100�m. B–D, These confocal images, revealing the initially administered dextran flooding (B), the postinjury infused dextranflooding (C), and their overlay (D), demonstrate that the majority of the neurons sustaining membrane disruption and traceruptake are double labeled both with preinjury- and postinjury-infused dextrans (arrows). Such double labeling was observed atboth 4 and 8 h after injury. Note that some neurons flooding with the preinjury dextran alone can also be observed (arrowheads).Scale bar, 100 �m.
Figure 4. A–D, Double-labeled confocal images. In A–C, neurons flooding with both dextrans reveal evidence of concomitantcellular injury, reflected in their irregular, distorted profiles and vacuolization (arrows). In those cells showing the most severedamage, dextrans are also typically found within the nucleus (arrowhead). Note that other double-labeled neurons (D) demon-strate little or no pathological damage and that despite homogenous tracer uptake, no nuclear accumulation or vacuolizationoccurs (arrows). Scale bars, 100 �m.
Farkas et al. • The Consequences of Mechanoporation in DTBI J. Neurosci., March 22, 2006 • 26(12):3130 –3140 • 3133
brain injury. Based on our previous obser-vation that the infusion of 10 kDa dextraninto the lateral ventricle is a reliable ap-proach for the detection of plasmalemmalalteration related to closed head injury(Singleton and Povlishock, 2004), AlexaFluor 488-conjugated 10 kDa dextran wasinfused 2 h before injury. Additionally,Texas Red-conjugated 10 kDa dextran wasused either 2 or 6 h after TBI to examinepotential membrane resealing and/or en-during membrane disruption. The ratio-nale here was that neurons containingboth fluorophores sustained enduringmembrane damage over the experimentaltime course, whereas the presence of onefluorophore alone would argue for eithermembrane resealing in the case of thepreinjury-administered fluorophore ordelayed opening in the case of thepostinjury-administered fluorophore-labeled dextran. Sham-injured animalsreceiving fluorophore-labeled dextran in-fusions demonstrated interstitial orperivascular diffusion of the tracersthroughout the neocortex without any ev-idence of cellular uptake. This indicatedthat controlled dextran infusion into thelateral ventricles did not lead to cellularuptake of the tracers (Fig. 2). This patternof tracer distribution was consistent withthat described with the use of comparablemolecular weight tracers described by ourlaboratory and others (Singleton and Pov-lishock, 2004). In contrast, the injured an-imals revealed tracer-containing neuronsindicative of membrane disruption scat-tered bilaterally throughout the rostrocau-dal extent of the neocortex, primarilywithin layers IV and V (Fig. 3A). The rela-tively large number of dextran-floodedneurons in injured tissue indicated thatthe membrane disruption was not an iso-lated phenomenon but rather occurredthroughout the extent of the neocortexscattered among other apparently unal-tered neurons. In all cases, based on atboth light and electron microscopy evalu-ation, there was no evidence of dextran-flooded glia. At 4 and 8 h after injury, re-spectively, 55.4% � 7.4 and 54.7% � 6.4of neurons sustaining any membrane dis-ruption and tracer flooding were doublelabeled both with the preinjury- andpostinjury-infused dextrans (Alexa Fluor488 and Texas Red conjugated). Thesefindings suggest that neuronal cell mem-branes were opened initially after injuryand maintained continued membrane dis-ruption detected by the administration ofthe second tracer (Fig. 3B– D). With con-focal microscopy, many of the neuronsflooded with both dextrans demonstrated
Figure 5. Ultrastructural analysis of the dual tracer-containing neurons reveals various forms of pathologic change. A, Thefluorescent image obtained via routine fluorescent microscopy. B, The anti-Alexa Fluor 488-immunostained image of the sameregion. C, D, An electron micrograph of the same neuron. Note that this neuron demonstrates moderately increased electrondensity, cytoplasmic and nuclear tracer flooding (arrows), and perinuclear organelle vacuolization (arrowheads) without overtmitochondrial damage (double arrowheads), best shown in the enlarged panel D. Scale bar, 2 �m. E, Three tracer-floodedneurons, again confirmed by routine fluorescent microscopy and followed via EM. The most severely damaged neuron (asterisk)demonstrates increased electron density, organelle vacuolization, and perisomatic glial ensheathment best illustrated in enlargedpanel F. The two other cells (double and triple asterisks) demonstrate little or no pathological change. Note that the surroundingneuropil demonstrates little overt pathologic change consistent with the confocal observations. Scale bar, 5 �m
3134 • J. Neurosci., March 22, 2006 • 26(12):3130 –3140 Farkas et al. • The Consequences of Mechanoporation in DTBI
evidence of concomitant cell injury reflected in their irregularprofiles and vacuolization, all of which were suggestive of ne-crotic alteration leading to cell death (Fig. 4A– C). In those neu-rons showing the most dramatic structural change, the dextranstypically were found within the nucleus. In contrast to these find-ings, a limited number of double-labeled neurons at both timepoints demonstrated little or no pathological damage. These neu-
rons revealed homogenous tracer floodingwithout nuclear accumulation or vacuol-ization (Fig. 4D). Concomitant ultrastruc-tural analyses confirmed and supple-mented the morphological changesobserved with confocal microscopy. Withelectron microscopy, many of the double-flooded neurons appeared damaged withevidence of necrosis. Such necrotic changewas reflected in occasional nuclear pykno-sis, dilation of the Golgi apparatus and thesmooth endoplasmic reticulum (SER), to-gether with occasional vesicular swelling,isolated mitochondrial change, increasednuclear and cytoplasmic electron density,and perisomatic glial swelling in the mostseverely necrotic neurons (Fig. 5E,F).However, as noted above, a limited num-ber of double-flooded cells demonstratedonly modest ultrastructural change, con-sistent with a less severe form of cellularperturbation (Fig. 5C– E).
Within these injury foci, CMSP re-vealed a complex and unanticipatedrepertoire of responses in relation to theabove-described pathologies. No CMSP-immunopositive neurons were observedin sham-injured animals. However, atboth 4 and 8 h after injury, CMSP-immunoreactive neurons were observedthroughout the neocortex (Fig. 6a). At 4 hafter injury, 12.0% � 8.1 of neurons dem-onstrating enduring membrane disrup-tion, evidenced by their content of bothtracers, were also found to be Ab 38 posi-tive. By 8 h after injury, 15.7% � 4.3 of thedouble-flooded neurons revealed evidenceof CMSP. Furthermore, at 4 and 8 h re-spectively, 34.3% � 3.1 and 35.4% � 6.3of all neurons analyzed revealed CMSPimmunoreactivity without concomitanttracer flooding (Fig. 6B– D). Ultrastruc-tural analysis of these Ab 38-positive neu-rons revealed only moderate subcellulardamage (Fig. 7A– C). Typically, the CMSPreaction product was found scatteredthroughout the cytoplasm, with local con-centrations around dilated mitochondriadispersed throughout the neuronal cellbody. In contrast, neurons demonstratingsevere ultrastructural alterations failed toshow CMSP immunoreactivity (Fig.7A,B).
Evidence for membrane resealingafter DTBI
In contrast to those neurons demonstrating enduring membranealteration as evidenced by their content of both dextrans, at 4 h,39.1% � 6.4 and at 8 h, 24.3% � 9.4 of cortical neurons, showingany evidence of tracer uptake, flooded with the preinjury infuseddextran alone, thereby suggesting membrane resealing (Fig. 8A–C).With confocal microscopy, these neurons displayed normal cel-lular detail, an observation consistent with post-TBI membrane
Figure 6. A, Triple-labeled confocal image demonstrating dextran flooded neurons as well as CMSP-immunopositive neurons.Note that at both 4 and 8 h after injury, CMSP-immunoreactive neurons were observed throughout the neocortex (arrows). Scalebar, 100 �m. B–E, Confocal images of preinjury dextran flooding (B), postinjury dextran flooding (C), CMSP immunopositivity(D), and their overlay (E) demonstrate neurons showing enduring membrane disruption reflected in their content of both tracers.Note that some neurons colocalize with CMSP immunopositivity (arrow), whereas other neurons demonstrate tracer floodingwithout CMSP (arrowheads). Also note that at these same time points, several CMSP-immunoreactive neurons also could beidentified without concomitant tracer flooding (double arrowhead). Finally, note that in the same region, a neuron demonstratingonly the initial tracer flooding (big arrow), as well as a neuron revealing only secondary tracer flooding (triple arrowhead), can alsobe seen. Scale bar, 50 �m.
Farkas et al. • The Consequences of Mechanoporation in DTBI J. Neurosci., March 22, 2006 • 26(12):3130 –3140 • 3135
resealing and cell survival. Parallel ultrastructural analyses failedto reveal significant pathological change within these neurons(Fig. 8F,G). Furthermore, at 4 and 8 h respectively, 7.6% � 5.5and 11.1% � 2.9 of these resealed neurons showed evidence ofcalpain-mediated spectrin proteolysis.
Evidence for delayed membrane perturbation after DTBIDespite the consistent finding of neurons containing the initiallyinfused tracer without any delayed tracer flooding, at 4 h, 5.5% �2.7 and at 8 h, 20.9% � 9.1 of all tracer-containing neurons werefound to flood with the postinjury infused dextran alone (Fig.9A– C). At 4 h, 55.1% � 12.9 and at 8 h, 32.5% � 11.8 of neuronsdemonstrating delayed membrane perturbation also showedCMSP.
Importantly, all of the above-described populations of neu-rons could be found in the same neocortical loci, excluding thepossibility that different regions exposed to different rates of dex-tran diffusion could have sustained different degrees of dextranflooding.
Redistribution of neuronal membrane perturbationafter DTBIAs noted above, neurons scattered throughout the neocortex ex-hibited varying degrees of membrane perturbation and cellularpathology after DTBI. In the present study design, three catego-ries of neurons with membrane perturbation were detected: (1)resealed neurons, (2) neurons with enduring membrane pertur-bation, (3) neurons with delayed membrane damage. Similarly,the experimental design separated each survival group (4 and 8 hafter injury) into periods before and after the availability of thepostinjury tracer. In this way, the relative proportions of neuronswith resealed, enduring, and delayed membrane perturbationscould be analyzed in relationship to time periods of tracer avail-
ability, to provide a model of membrane perturbation pathologyafter DTBI (Fig. 10).
Of all the neurons that suffer membrane perturbation by 4 hafter injury, 95% are damaged within the first 2 h. However, 40%of the neurons reseal by 2 h, leaving 60% with enduring or de-layed membrane perturbations at 4 h after injury (Fig. 10). Fromthe 8 h experimental group, 80% of all neurons with membraneperturbation suffer damage within the first 6 h, with 25% of theneurons resealing before the postinjury tracer administration at6 h, and 75% of the population manifests enduring or delayedmembrane perturbation (Fig. 10). Because the percentages of thepopulation are normalized and derived from the same data set,statistical tests cannot be performed within a time point. Betweentime points, the proportion of neurons with delayed membraneperturbation differs significantly from the proportion of resealedneurons and the proportion of enduring membrane perturbation(� 2, p � 0.016). The change in the proportion of neurons withresealed and enduring membrane perturbation over time was notsignificant (� 2, p � 0.19). Together, these tests confirm a signif-icant redistribution of the neuronal population with membraneperturbations between 4 and 8 h. These analyses were based onfractions of all neurons that demonstrate perturbation, but thetotal populations were not necessarily equal.
DiscussionUsing tracer exclusion and immunohistochemical techniques,together with confocal and electron microscopy, the currentstudy provides new evidence for the diversity of the neuronalcellular responses to DTBI, emphasizing the significance of neu-ronal membrane mechanoporation in DTBI pathobiology.
Membrane disruption and enduring membrane permeabilityThe present study is the first to demonstrate, in an in vivo modelof DTBI, the potential for either enduring membrane perturba-tion progressing to cell death or neuronal membrane resealingcompatible with recovery. These neuronal changes were bothwidespread and diffusely distributed, interspersed with appar-ently uninjured neurons, a finding consistent with the pathobi-ology of DTBI in humans (Povlishock and Katz, 2005).
These data emphasize that the mechanical forces initiate theobserved pathology. The early, diffuse damage identified is notconsistent with focal ischemia that would have precipitated amore delayed and highly focal neuronal response (Dietrich et al.,1987; Dereski et al., 1993; Dietrich, 1998; Mennel et al., 2000;Bramlett and Dietrich, 2004). Although in this study, DTBIevoked elevated ICP, this elevation never exceeded 20 mmHg,suggesting no significant alteration in cerebral perfusion pressurewith the normal arterial blood pressure described in this study.This does not preclude, however, the potential that over time theelevated ICP evoked secondary neuronal damage (vide infra).
Using lactate dehydrogenase (LDH) release or high-molecular-weight tracer uptake to evaluate membrane damage invarious TBI models, our laboratory and others have establishedthat membrane disruption can occur at the moment of injury,suggesting that the traumatic event evokes direct mechanical po-ration of the neuronal cell membrane (Pettus et al., 1994; LaPlacaet al., 1997; Geddes and Cargill, 2001; Geddes et al., 2003; Pradoand LaPlaca, 2004; Prado et al., 2004; Singleton and Povlishock,2004). In these studies, it was suggested that the altered neuronalmembrane permeability generated ionic disturbances to precip-itate cell death. This premise, however, was neither directly ad-dressed nor followed comprehensively via multiple endpoints(LaPlaca et al., 1997; Geddes et al., 2003; Prado et al., 2004). The
Figure 7. A–C, A toluidine blue-stained 1-�m-thick section reacted for the visualization ofCMSP is shown in A, whereas B and C reveal an EM of the same neurons. Note that in B, oneCMSP-immunopositive neuron demonstrates moderate subcellular damage. Note that C, takenfrom that area blocked off in B, reveals the CMSP reaction product around swollen mitochondria(arrows) and dispersed throughout the cytoplasm (double arrowheads). A neuron demonstrat-ing overt necrosis with no evidence of CMSP is also shown (arrowhead). Scale bar, 5 �m.
3136 • J. Neurosci., March 22, 2006 • 26(12):3130 –3140 Farkas et al. • The Consequences of Mechanoporation in DTBI
current study significantly extends previous observations, dem-onstrating that membrane perturbation persists for, at least, sev-eral hours after injury in many injured neurons. Significance isalso attached to the LM observation that these neurons revealedmorphological features consistent with irreversible damage, in-cluding cellular distortion, vacuolization, and pyknotic nuclei.The parallel EM findings of increased electron density, togetherwith the dilation of the Golgi and SER and increased numbers ofdilated vesicles, were also consistent with a necrotic cascade trig-gered by the translocation of lysosomal cathepsins and other hy-drolytic enzymes (White et al., 1993; Chan, 1996; Yamashima etal., 1998; Yamashima, 2004). However, we cannot rule out thatthis organelle and vesicular dilation are linked to an albeit abor-
tive attempt at membrane repair mediatedby the lysosomes and the synaptotagminfound on their surfaces (Andrews, 2005).These issues will require highly targetedinvestigations for their resolution. Al-though our finding of isolated neuronsdemonstrating enduring membrane per-meability without severe ultrastructuraldamage at both 4 and 8 h after injury sug-gests that membrane perturbation maynot always lead to cell death, we cannotpreclude the possibility that with increasedsurvival these same neurons would die.
Membrane resealingIn addition to enduring membrane alter-ation, other neurons displayed evidence ofmembrane resealing via their uptake of onlythe preinjury tracer. Previous studies in non-neuronal cells demonstrated that, after me-chanical injury, membrane resealing can oc-cur within seconds to minutes (McNeil andIto, 1989; McNeil and Khakee, 1992; Yu andMcNeil, 1992; Steinhardt et al., 1994; Ter-asaki et al., 1997; McNeil et al., 2000). Thesestudies investigated cardiac muscle (Clarkeet al., 1995), skeletal muscle (McNeil andKhakee, 1992), skin (McNeil and Ito, 1990),vascular endothelium (Yu and McNeil,1992), or gastrointestinal cells (McNeil andIto, 1989), which frequently experience me-chanical stress/strain under normal condi-tions. To date, only limited studies have in-vestigated membrane resealing in the CNS.In vitro or nonmammalian models have fo-cused primarily on resealing after axonaltransection (Yawo and Kuno, 1983, 1985;Fishman et al., 1990; Xie and Barrett, 1991;Godell et al., 1997; Shi and Pryor, 2000; Shi etal., 2001; Geddes et al., 2003). In contrast tonon-neuronal cells, these studies suggestedthat neuronal cell membrane resealing takesminutes to hours rather than seconds de-scribed in non-neuronal cells. Our finding ofneurons labeled with the preinjury traceralone is consistent with membrane repair;however, the sampling period occurredover several hours after injury preclud-ing the determination of whether this re-sealing was immediate or rather oc-
curred at a later time point.
Delayed membrane perturbationOur observation of neurons flooded with the postinjury infuseddextran alone suggests the potential for delayed membrane dis-ruption occurring several hours after injury. The precise mecha-nism(s) for this delayed opening is (are) unknown; however, thepossibility exists that the sustained, elevated ICP occurring afterinjury could alter the neuronal cell membrane. This issue is ofmore than mere academic interest, because sustained elevatedICP is one of the major adverse prognostic factors for recoveryafter TBI and obviously this issue will merit continued evaluation(The Brain Trauma Foundation et al., 2000).
Figure 8. A–C, Confocal images of preinjury-infused dextrans (A), postinjury-infused dextrans (B), and their overlay (C)demonstrate some cortical neurons flooding with the preinjury-infused dextran alone without concomitant flooding with thepostinjury-administrated tracer (arrows). Note that one double-flooded neuron demonstrating severe damage (arrowhead) andone double-labeled neuron demonstrating less severe pathology (double arrowhead) are also shown. Scale bar, 50 �m. D, Aroutine fluorescent image that reveals a single labeled cell. E, The same neuron as in D is visualized through the use of antibodiesto the fluorophore and then carried to the EM level (F, G). Note that neurons flooding with the preinjury dextran alone do not showovert pathological damage. Immunoreactive products (anti-Alexa Fluor IR) are labeled with arrows in G, which is an enlargementof the area, blocked out in F. Scale bar, 2 �m.
Figure 9. A–C, Confocal images of preinjury-infused dextran (A), postinjury-infused dextran (B), and their overlay (C) dem-onstrate scattered neurons flooding with the postinjury-infused dextran alone (arrows), among with double-flooded neurons(arrowheads). Scale bar, 100 �m.
Farkas et al. • The Consequences of Mechanoporation in DTBI J. Neurosci., March 22, 2006 • 26(12):3130 –3140 • 3137
The coincidence of CMSP with alteredmembrane permeabilityThe present study also examined the potential role of calpainactivation and its relation to membrane permeability and/or neu-ronal damage. Calpain activation and the subsequent CMSP havelong been associated with the pathogenesis of different brain in-juries and TBI cell death (Kampfl et al., 1996, 1997; Saatman et al.,1996; Newcomb et al., 1997; Pike et al., 1998, 2000, 2001; Buki etal., 1999; McCracken et al., 1999; Kupina et al., 2002, 2003).Calpain activation occurs at elevated intracellular calcium levels.Membrane perturbation can lead to excessive calcium influxfrom the extracellular space, resulting in calpain activation (Pov-lishock et al., 1997; Buki et al., 1999), that then causes directsubcellular damage or lysosome membrane rupture with the re-lease of damaging enzymes such as cathepsins (Yamashima,2004). The cysteine proteases, such as calpain and caspase, canalso maintain increased membrane permeability, theoreticallypermitting additional pathology (Liu and Schnellmann, 2003;Liu et al., 2004). Furthermore, it has been demonstrated, in vitro,that CMSP correlates with LDH release and propidium iodideuptake, markers of increased membrane permeability afterstretch injury (Pike et al., 2000). Other studies, however, havealso suggested that calpain might have an important role in mem-brane resealing (Godell et al., 1997; Howard et al., 1999; Shi et al.,2000).
In the present study, the majority of cells demonstratingCMSP did not reveal tracer flooding either at 4 or 8 h after injury,suggesting that calpain activation occurs independent of cellmembrane disruption. Furthermore, the majority of CMSP-positive neurons revealed only limited ultrastructural change,suggesting that calpain activation is not necessarily associatedwith cell death after DTBI, at least within the time frames as-sessed. Alternatively, the initiation of neuronal death may awaitthe activation of a secondary calpain cascade, a premise sup-ported by several studies (Saido et al., 1993; Yokota et al., 1995;
Neumar et al., 2001; Saatman et al., 2003; Czogalla and Sikorski,2005). Equally surprising was the finding that the majority of cellsflooding with both dextrans and manifesting necrotic change didnot show CMSP. Conceivably, the calpain-specific spectrinbreakdown products in the necrotic neurons were cleaved beforethe assessment, yet, this argument is countered by the fact thatspectrin fragments are highly stable in vivo and in vitro (Wang etal., 1989; Czogalla and Sikorski, 2005). Neurons triple labeledwith both dextrans and Ab 38 were observed, yet they representedthe smallest population of injured neurons. Collectively, theseresults suggest that the observed membrane disruption is notnecessary or sufficient for calpain-mediated spectrin proteolysis.Together with other studies (Saatman et al., 1996; Brana et al.,1999), these findings emphasize the need for caution in interpret-ing the occurrence of CMSP and its overall implications for neu-ronal injury and death.
Redistribution of neuronal membrane perturbation after TBIIn addition to the above provocative information regardingmembrane perturbation, resealing, and delayed opening, theshifts over time in the percentages of neuronal populations show-ing membrane damage provide compelling evidence for a redis-tribution of the membrane damaged population over time. Spe-cifically, the fraction of neurons that reseal early postinjury maysuffer secondary membrane reopening, converting these neuronsto those linked with enduring membrane perturbation. Further-more, with increased survival, proportionally more neurons suf-fer delayed membrane perturbation. Theoretically to achievesuch a distribution from 4 – 8 h after injury, 25% of the resealedneurons would be required to reopen, with an additional 20% ofthe previously uninjured neurons now suffering delayed mem-brane perturbation. Together with our previous findings, theseobservations confirm that the plasmalemmal changes evoked byDTBI are complex, evolving, and potentially subject to modifica-tion by secondary mechanisms.
In sum, this study demonstrates the heterogeneity of injury-related membrane perturbation and the diversity of the neuronalresponses to this perturbation, while also demonstrating the po-tential of delayed membrane repair after TBI. Furthermore, ourfindings indicate that calpain-mediated spectrin proteolysis canoccur independent of membrane disruption and that membranedisruption itself does not necessarily lead to calpain activation.These findings highlight the complexity of DTBI, its dissimilarityto most CNS disorders that evoke homogenous, focal change,and its role in dynamic mechanical poration of the neuronalplasmalemma.
ReferencesAmorini AM, Dunbar JG, Marmarou A (2003) Modulation of aquaporin-4
water transport in a model of TBI. Acta Neurochir Suppl 86:261–263.Andrews NW (2005) Membrane resealing: synaptotagmin VII keeps run-
ning the show. Sci STKE 2005:e19.Bramlett HM, Dietrich WD (2004) Pathophysiology of cerebral ischemia
and brain trauma: similarities and differences. J Cereb Blood Flow Metab24:133–150.
Brana C, Benham CD, Sundstrom LE (1999) Calpain activation and inhibi-tion in organotypic rat hippocampal slice cultures deprived of oxygen andglucose. Eur J Neurosci 11:2375–2384.
Buki A, Siman R, Trojanowski JQ, Povlishock JT (1999) The role of calpain-mediated spectrin proteolysis in traumatically induced axonal injury.J Neuropathol Exp Neurol 58:365–375.
Chan PH (1996) Role of oxidants in ischemic brain damage. Stroke27:1124 –1129.
Clarke MS, Caldwell RW, Chiao H, Miyake K, McNeil PL (1995)Contraction-induced cell wounding and release of fibroblast growth fac-tor in heart. Circ Res 76:927–934.
Figure 10. Distribution of neurons with resealed, enduring, or delayed membrane pertur-bation within the population of neurons with DTBI-induced membrane perturbation. As mem-branes remain closed, open, or reseal in response to injury, tracer availability will determine thetype of membrane perturbation categorically assigned to each neuron (table within figurelegend). The preinjury administration of Alexa Fluor 488 and postinjury administration of TexasRed-conjugated dextrans permit the evaluation of neuronal membrane perturbation distribu-tion at 4 and 8 h after injury. Across time points, the proportion of neurons with delayedmembrane perturbation is significantly different from the proportion of resealed neurons andthe proportion of enduring membrane perturbation (�2, p � 0.016). The change in the pro-portion of neurons with resealed and enduring membrane perturbation over time was notsignificant (�2, p � 0.19). The redistribution of the types of membrane perturbation between4 and 8 h after injury may result from resealed neurons reopening to become neurons withenduring damage and/or additional neurons suffering delayed membrane perturbation. SeeResults for more details.
3138 • J. Neurosci., March 22, 2006 • 26(12):3130 –3140 Farkas et al. • The Consequences of Mechanoporation in DTBI
Colicos MA, Dixon CE, Dash PK (1996) Delayed, selective neuronal deathfollowing experimental cortical impact injury in rats: possible role inmemory deficits. Brain Res 739:111–119.
Cortez SC, McIntosh TK, Noble LJ (1989) Experimental fluid percussionbrain injury: vascular disruption and neuronal and glial alterations. BrainRes 482:271–282.
Czogalla A, Sikorski AF (2005) Spectrin and calpain: a “target” and a“sniper” in the pathology of neuronal cells. Cell Mol Life Sci62:1913–1924.
Dereski MO, Chopp M, Knight RA, Rodolosi LC, Garcia JH (1993) Theheterogeneous temporal evolution of focal ischemic neuronal damage inthe rat. Acta Neuropathol (Berl) 85:327–333.
Dietrich WD (1998) Neurobiology of stroke. Int Rev Neurobiol 42:55–101.Dietrich WD, Busto R, Yoshida S, Ginsberg MD (1987) Histopathological
and hemodynamic consequences of complete versus incomplete ischemiain the rat. J Cereb Blood Flow Metab 7:300 –308.
Dietrich WD, Alonso O, Busto R, Globus MY, Ginsberg MD (1994a) Post-traumatic brain hypothermia reduces histopathological damage follow-ing concussive brain injury in the rat. Acta Neuropathol (Berl)87:250 –258.
Dietrich WD, Alonso O, Halley M (1994b) Early microvascular and neuro-nal consequences of traumatic brain injury: a light and electron micro-scopic study in rats. J Neurotrauma 11:289 –301.
Farkas O, Polgar B, Szekeres-Bartho J, Doczi T, Povlishock JT, Buki A (2005)Spectrin breakdown products in the cerebrospinal fluid in severe headinjury–preliminary observations. Acta Neurochir (Wien) 147:855– 861.
Fishman HM, Tewari KP, Stein PG (1990) Injury-induced vesiculation andmembrane redistribution in squid giant axon. Biochim Biophys Acta1023:421– 435.
Foda MA, Marmarou A (1994) A new model of diffuse brain injury in rats.Part II: Morphological characterization. J Neurosurg 80:301–313.
Geddes DM, Cargill RS (2001) An in vitro model of neural trauma: devicecharacterization and calcium response to mechanical stretch. J BiomechEng 123:247–255.
Geddes DM, Cargill RS, LaPlaca MC (2003) Mechanical stretch to neuronsresults in a strain rate and magnitude-dependent increase in plasmamembrane permeability. J Neurotrauma 20:1039 –1049.
Godell CM, Smyers ME, Eddleman CS, Ballinger ML, Fishman HM, BittnerGD (1997) Calpain activity promotes the sealing of severed giant axons.Proc Natl Acad Sci USA 94:4751– 4756.
Hicks R, Soares H, Smith D, McIntosh T (1996) Temporal and spatial char-acterization of neuronal injury following lateral fluid-percussion braininjury in the rat. Acta Neuropathol (Berl) 91:236 –246.
Howard MJ, David G, Barrett JN (1999) Resealing of transected myelinatedmammalian axons in vivo: evidence for involvement of calpain. Neuro-science 93:807– 815.
Kampfl A, Posmantur R, Nixon R, Grynspan F, Zhao X, Liu SJ, Newcomb JK,Clifton GL, Hayes RL (1996) mu-Calpain activation and calpain-mediated cytoskeletal proteolysis following traumatic brain injury. J Neu-rochem 67:1575–1583.
Kampfl A, Posmantur RM, Zhao X, Schmutzhard E, Clifton GL, Hayes RL(1997) Mechanisms of calpain proteolysis following traumatic brain in-jury: implications for pathology and therapy: implications for pathologyand therapy: a review and update. J Neurotrauma 14:121–134.
Kupina NC, Detloff MR, Dutta S, Hall ED (2002) Neuroimmunophilin li-gand V-10,367 is neuroprotective after 24-hour delayed administration ina mouse model of diffuse traumatic brain injury. J Cereb Blood FlowMetab 22:1212–1221.
Kupina NC, Detloff MR, Bobrowski WF, Snyder BJ, Hall ED (2003) Cy-toskeletal protein degradation and neurodegeneration evolves differentlyin males and females following experimental head injury. Exp Neurol180:55–73.
LaPlaca MC, Lee VM, Thibault LE (1997) An in vitro model of traumaticneuronal injury: loading rate-dependent changes in acute cytosolic cal-cium and lactate dehydrogenase release. J Neurotrauma 14:355–368.
Liu X, Schnellmann RG (2003) Calpain mediates progressive plasma mem-brane permeability and proteolysis of cytoskeleton-associated paxillin,talin, and vinculin during renal cell death. J Pharmacol Exp Ther304:63–70.
Liu X, Van Vleet T, Schnellmann RG (2004) The role of calpain in oncoticcell death. Annu Rev Pharmacol Toxicol 44:349 –370.
Marmarou A, Foda MA, van den BW, Campbell J, Kita H, Demetriadou K
(1994) A new model of diffuse brain injury in rats. Part I: Pathophysiol-ogy and biomechanics. J Neurosurg 80:291–300.
Marshall LF, Smith RW, Shapiro HM (1979) The outcome with aggressivetreatment in severe head injuries. Part I: The significance of intracranialpressure monitoring. J Neurosurg 50:20 –25.
McCracken E, Hunter AJ, Patel S, Graham DI, Dewar D (1999) Calpainactivation and cytoskeletal protein breakdown in the corpus callosum ofhead-injured patients. J Neurotrauma 16:749 –761.
McNeil PL, Ito S (1989) Gastrointestinal cell plasma membrane woundingand resealing in vivo. Gastroenterology 96:1238 –1248.
McNeil PL, Ito S (1990) Molecular traffic through plasma membrane dis-ruptions of cells in vivo. J Cell Sci 96:549 –556.
McNeil PL, Khakee R (1992) Disruptions of muscle fiber plasma mem-branes. Role in exercise-induced damage. Am J Pathol 140:1097–1109.
McNeil PL, Vogel SS, Miyake K, Terasaki M (2000) Patching plasma mem-brane disruptions with cytoplasmic membrane. J Cell Sci 113:1891–1902.
Mennel HD, El Abhar H, Schilling M, Bausch J, Krieglstein J (2000) Mor-phology of tissue damage caused by permanent occlusion of middle cere-bral artery in mice. Exp Toxicol Pathol 52:395– 404.
Miller JD, Becker DP, Ward JD, Sullivan HG, Adams WE, Rosner MJ (1977)Significance of intracranial hypertension in severe head injury. J Neuro-surg 47:503–516.
Neumar RW, Meng FH, Mills AM, Xu YA, Zhang C, Welsh FA, Siman R(2001) Calpain activity in the rat brain after transient forebrain ischemia.Exp Neurol 170:27–35.
Newcomb JK, Kampfl A, Posmantur RM, Zhao X, Pike BR, Liu SJ, Clifton GL,Hayes RL (1997) Immunohistochemical study of calpain-mediatedbreakdown products to alpha-spectrin following controlled cortical im-pact injury in the rat. J Neurotrauma 14:369 –383.
Pettus EH, Christman CW, Giebel ML, Povlishock JT (1994) Traumaticallyinduced altered membrane permeability: its relationship to traumaticallyinduced reactive axonal change. J Neurotrauma 11:507–522.
Pike BR, Zhao X, Newcomb JK, Posmantur RM, Wang KK, Hayes RL (1998)Regional calpain and caspase-3 proteolysis of alpha-spectrin after trau-matic brain injury. NeuroReport 9:2437–2442.
Pike BR, Zhao X, Newcomb JK, Glenn CC, Anderson DK, Hayes RL (2000)Stretch injury causes calpain and caspase-3 activation and necrotic andapoptotic cell death in septo-hippocampal cell cultures. J Neurotrauma17:283–298.
Pike BR, Flint J, Dutta S, Johnson E, Wang KK, Hayes RL (2001) Accumu-lation of non-erythroid alpha II-spectrin and calpain-cleaved alpha II-spectrin breakdown products in cerebrospinal fluid after traumatic braininjury in rats. J Neurochem 78:1297–1306.
Povlishock JT, Katz DI (2005) Update of neuropathology and neurologicalrecovery after traumatic brain injury. J Head Trauma Rehabil 20:76 –94.
Povlishock JT, Marmarou A, McIntosh T, Trojanowski JQ, Moroi J (1997)Impact acceleration injury in the rat: evidence for focal axolemmal changeand related neurofilament sidearm alteration. J Neuropathol Exp Neurol56:347–359.
Prado G, LaPlaca MC (2004) Neuronal plasma membrane is transiently dis-rupted by mechanical trauma: An insight into the mechanisms involved.J Neurotrauma 21:1331.
Prado G, Caltagirone CL, LaPlaca MC (2004) Neuronal plasma membranedisruption in the traumatically injured brain. J Neurotrauma 21:1331.
Roberts-Lewis JM, Savage MJ, Marcy VR, Pinsker LR, Siman R (1994) Im-munolocalization of calpain I-mediated spectrin degradation to vulnera-ble neurons in the ischemic gerbil brain. J Neurosci 14:3934 –3944.
Saatman KE, Bozyczko-Coyne D, Marcy V, Siman R, McIntosh TK (1996)Prolonged calpain-mediated spectrin breakdown occurs regionally fol-lowing experimental brain injury in the rat. J Neuropathol Exp Neurol55:850 – 860.
Saatman KE, Abai B, Grosvenor A, Vorwerk CK, Smith DH, Meaney DF(2003) Traumatic axonal injury results in biphasic calpain activation andretrograde transport impairment in mice. J Cereb Blood Flow Metab23:34 – 42.
Saido TC, Yokota M, Nagao S, Yamaura I, Tani E, Tsuchiya T, Suzuki K,Kawashima S (1993) Spatial resolution of fodrin proteolysis in postisch-emic brain. J Biol Chem 268:25239 –25243.
Shi R, Pryor JD (2000) Temperature dependence of membrane sealing fol-lowing transection in mammalian spinal cord axons. Neuroscience98:157–166.
Farkas et al. • The Consequences of Mechanoporation in DTBI J. Neurosci., March 22, 2006 • 26(12):3130 –3140 • 3139
Shi R, Asano T, Vining NC, Blight AR (2000) Control of membrane sealingin injured mammalian spinal cord axons. J Neurophysiol 84:1763–1769.
Shi R, Qiao X, Emerson N, Malcom A (2001) Dimethylsulfoxide enhancesCNS neuronal plasma membrane resealing after injury in low tempera-ture or low calcium. J Neurocytol 30:829 – 839.
Siman R, Noszek JC, Kegerise C (1989) Calpain I activation is specificallyrelated to excitatory amino acid induction of hippocampal damage.J Neurosci 9:1579 –1590.
Singleton RH, Povlishock JT (2004) Identification and characterization ofheterogeneous neuronal injury and death in regions of diffuse brain in-jury: evidence for multiple independent injury phenotypes. J Neurosci24:3543–3553.
Singleton RH, Zhu J, Stone JR, Povlishock JT (2002) Traumatically inducedaxotomy adjacent to the soma does not result in acute neuronal death.J Neurosci 22:791– 802.
Steinhardt RA, Bi G, Alderton JM (1994) Cell membrane resealing by avesicular mechanism similar to neurotransmitter release. Science263:390 –393.
Stone JR, Walker SA, Povlishock JT (1999) The visualization of a new classof traumatically injured axons through the use of a modified method ofmicrowave antigen retrieval. Acta Neuropathol (Berl) 97:335–345.
Terasaki M, Miyake K, McNeil PL (1997) Large plasma membrane disrup-tions are rapidly resealed by Ca 2�-dependent vesicle-vesicle fusionevents. J Cell Biol 139:63–74.
The Brain Trauma Foundation, The American Association of NeurologicalSurgeons, The Joint Section on Neurotrauma and Critical Care (2000)Indications for intracranial pressure monitoring. J Neurotrauma17:479 – 491.
Wang KK, Villalobo A, Roufogalis BD (1989) Calmodulin-binding proteinsas calpain substrates. Biochem J 262:693–706.
White BC, Daya A, DeGracia DJ, O’Neil BJ, Skjaerlund JM, Trumble S,Krause GS, Rafols JA (1993) Fluorescent histochemical localization oflipid peroxidation during brain reperfusion following cardiac arrest. ActaNeuropathol (Berl) 86:1–9.
Xie XY, Barrett JN (1991) Membrane resealing in cultured rat septal neu-rons after neurite transection: evidence for enhancement by Ca 2�-triggered protease activity and cytoskeletal disassembly. J Neurosci11:3257–3267.
Yamashima T (2004) Ca 2�-dependent proteases in ischemic neuronaldeath: a conserved “calpain-cathepsin cascade” from nematodes to pri-mates. Cell Calcium 36:285–293.
Yamashima T, Kohda Y, Tsuchiya K, Ueno T, Yamashita J, Yoshioka T, Komi-nami E (1998) Inhibition of ischaemic hippocampal neuronal death inprimates with cathepsin B inhibitor CA-074: a novel strategy for neuro-protection based on “calpain-cathepsin hypothesis.” Eur J Neurosci10:1723–1733.
Yawo H, Kuno M (1983) How a nerve fiber repairs its cut end: involvementof phospholipase A2. Science 222:1351–1353.
Yawo H, Kuno M (1985) Calcium dependence of membrane sealing at thecut end of the cockroach giant axon. J Neurosci 5:1626 –1632.
Yokota M, Saido TC, Tani E, Kawashima S, Suzuki K (1995) Three distinctphases of fodrin proteolysis induced in postischemic hippocampus. In-volvement of calpain and unidentified protease. Stroke 26:1901–1907.
Yu QC, McNeil PL (1992) Transient disruptions of aortic endothelial cellplasma membranes. Am J Pathol 141:1349 –1360.
3140 • J. Neurosci., March 22, 2006 • 26(12):3130 –3140 Farkas et al. • The Consequences of Mechanoporation in DTBI
Acta Neurochir (Wien) (2005) 147: 855–861
DOI 10.1007/s00701-005-0559-6
Clinical ArticleSpectrin breakdown products in the cerebrospinal fluid in severehead injury – preliminary observations
O. Farkas1;3, B. Polgar2, J. Szekeres-Barth�oo2, T. D�ooczi1, J. T. Povlishock3, and A. Buki1
1 Department of Neurosurgery, Center for Medical and Health Sciences, P�eecs University, P�eecs, Hungary2 Clinical Microbiology and Immunology, Center for Medical and Health Sciences, P�eecs University, P�eecs, Hungary3 Department of Anatomy and Neurobiology, Medical College of Virginia, Virginia Commonwealth University, Richmond, VA
Received May 19, 2004; accepted April 28, 2005; published online June 9, 2005
# Springer-Verlag 2005
Summary
Background. Calcium-induced proteolytic processes are considered
key players in the progressive pathobiology of traumatic brain injury
(TBI). Activation of calpain and caspases after TBI leads to the cleavage
of cytoskeletal proteins such as non-erythroid alpha II-spectrin. Recent
reports demonstrate that the levels of spectrin and spectrin breakdown
products (SBDPs) are elevated in vitro after mechanical injury, in the
cerebrospinal fluid (CSF) and brain tissue following experimental TBI,
and in human brain tissue after TBI.
Methods. This study was initiated to detect spectrin and SBDP accu-
mulation in the ventricular CSF of 12 severe TBI-patients with raised
intracranial pressure (ICP). Nine patients with non-traumatically elevat-
ed ICP and 5 undergoing diagnostic lumbar puncture (LP) served as
controls. Intact spectrin and calpain and caspase specific SBDPs in CSF
collected once a day over a several day period were assessed via Western
blot analysis. Parameters of severity and outcome such as ICP, Glasgow
Coma Scale and Glasgow Outcome Scale were also monitored in order
to reveal a potential correlation between these CSF markers and clinical
parameters.
Results. In control patients undergone LP no immunoreactivity was
detected. Non-erythroid alpha-II-spectrin and SBDP occurred more fre-
quently and their level was significantly higher in the CSF of TBI
patients than in other pathological conditions associated with raised
ICP. Those TBI patients followed for several days post-injury revealed
a consistent temporal pattern for protein accumulation with the highest
level achieved on the 2nd–3rd days after TBI.
Conclusion. Elevation of calpain and caspase specific SBDPs is a
significant finding in TBI patients indicating that intact brain spec-
trin- and SBDP-levels are closely associated with the specific neuro-
chemical processes evoked by TBI. The results strongly support the
potential utility of these surrogate markers in the clinical monitoring
of patients with severe TBI and provide further evidence of the role
of calcium-induced, calpain- and caspase-mediated structural prote-
olysis in TBI.
Keywords: Traumatic brain injury; calpain; caspase; spectrin; human;
cerebrospinal fluid.
Introduction
Traumatic brain injury (TBI) is the primary cause of
death in the population under the age of 40 representing
an extreme challenge for individuals and families affected
as well as the society at large [10]. Thus, understanding
the pathobiochemical processes evoked by or operant in
TBI is integral to developing new therapeutic strategies
while exploring potential biochemical markers that can
help assess the severity of the injury and predict outcome.
Traumatic brain injuries are typically classified on
whether the primary damage evokes more localized brain
damage and=or mass lesion formation (focal injury), or
more scattered, non-focal injuries typically involving
scattered axonal and=or microvascular damage (diffuse
injury). Although focal and diffuse injuries can be
evoked by different mechanism and their clinical mani-
festations can be different, several studies to date have
shown that similar biochemical cascades take part in the
pathobiology of these different injury types. These stud-
ies have demonstrated increased pathological activa-
tion of calpain and=or caspase-3 both after focal TBI
[8, 9, 12, 16] and diffuse injuries [5, 6, 18]. Calpain and
caspase-3, both members of cysteine protease family,
have also been shown to play an important role in the
proteolytic cascades associated with several other cen-
tral nervous system disorders such as stroke, hypoxia-
ischemia [3, 13], experimental hydrocephalus [7] and
spinal cord injury [1]. Non-erythroid alpha II-spectrin,
major component of the cytoskeleton, constitutes a
substrate for both calpain and caspase-3 [21]. Calpain-
mediated cleavage of intact spectrin (280 kDa) results in
150 and 145 kDa-fragments specific for calpain, whereas
the caspase-3-specific products are linked to 150 and
120-fragments [20]. Using antibodies targeting alpha
II-spectrin breakdown products numerous recent reports
have demonstrated that non-erythroid alpha II-spectrin
and spectrin breakdown products (SBDPs) are elevated
in vitro after mechanical injury [15, 4], also being found
in the cerebrospinal fluid and brain following experi-
mental TBI [16, 12, 8, 18], as well as human brain tissue
post–TBI [11].
The goal of the current study was to further define the
role of calcium-induced, calpain- and caspase-mediated
structural proteolysis in the pathobiology of TBI, iden-
tifying potential biomarkers specifically associated with
the pathological processes evoked by TBI and thereby
purportedly capable of predicting the severity of the
initial injury and outcome to help in the better clinical
management of severely brain-injured patients.
In this study, the levels of non-erythroid alpha II-
spectrin and SBDPs in the CSF of TBI patients sustain-
ing severe injury were compared to two control groups:
one with other disorders associated with elevated intra-
cranial pressure (ICP) and the other with normal ICP.
Clinical parameters of severity and outcome such as in-
tracranial pressure, Glasgow Coma Scale and Glasgow
Outcome Scale were monitored.
Materials and methods
The study included twelve patients with severe TBI (GCS<9). The
first control group consisted of patients with comparably raised ICP not
associated with TBI and included 3 patients with subarachnoid haemor-
rhage (SAH), 3 with intraventricular haemorrhage (IVH), and 3 with
brain tumours. The second control group consisted of 5 patients under-
going diagnostic LP that subsequently proved negative for SAH and=or
meningitis (clinical data of all patients are presented in Table 1). All
CSF harvesting was part of routine patient management, in accordance
with institutional guidelines; the study protocol was approved by the
Regional Ethics Committee (NIH-approved Institutional Review Board)
of the Centre of Medical and Health Sciences of Pecs University. In the
traumatically brain–injured group all patients were equipped with intra-
ventricular catheters for the control of ICP. Via these catheters ventri-
cular CSF samples were collected once a day, in some patients for
several days following ventriculostomy. They were centrifuged at
6000 g for 6 min and stored at �80 �C. After defrosting samples were
washed in fivefold PBS on Amicon ultrafiltration cell using poly-
ethersulfone ultrafiltration membranes with 100 kDa nominal molecular
weight limit. During this procedure the samples were purified from
serum albumin and other proteins with less than 100 kDa molecular
weight and concentrated to about five-fold. Protein concentrations were
determined using Bio-Rad Protein Assay. Samples containing less than
1 mg=ml protein were dried by Heto dry Winner. The dehydrated sub-
stances were dissolved in distilled water to get a protein concentration of
1 mg=ml. Sodium dodecyl sulfate-polyacrilamide gel electrophoresis
(SDS-PAGE) was performed according to the method of Laemmli using
a mini-gel apparatus (Bio-Rad). Samples containing 20 micrograms of
Table 1.
Group Age Sex Diagnosis CT GCS o.a. ICP or OP Date GOS Date
T-1 41 M SDHþCont – 4 35Hgmm D1- 2 28
T-2 79 FM Cont IVH 4 20Hgmm D1- 2 12
T-3 28 M Cont IVH 5 20 Hgmm D3 3 18
T-4 48 M SDH – 4 25 Hgmm D3 3 25
T-5 51 M EDHþCont – 6 10 Hgmm D1 1 31
T-6 21 M Cont – 7 20 Hgmm D1 4 9
T-7 51 M SDHþCont – 5 40 Hgmm D4 1 90
T-8 68 M SDH – 6 20 Hgmm D4 2 14
T-9 51 FM Cont IVH 4 40 Hgmm D1 1 3
T-10 18 M EDH – 4 35 Hgmm D1 4 37
T-11 19 M SDHþCont – 6 35 Hgmm D3 3 17
T-12 72 FM EDHþCont – 5 10 Hgmm D2 2 16
C-1-1 69 FM SAH HC 9 20 Hgmm D10 4 28
C-1-2 65 M SAH HC 8 15 Hgmm D14 4 21
C-1-3 32 FM SAH OE 4 30 Hgmm D2 2 22
C-1-4 46 FM SAH IVH 10 20 Hgmm D3 5 23
C-1-5 43 M SAH IVHþHC 6 20 Hgmm D10 3 43
C-1-6 30 M IVH HC 7 40 Hgmm D2 1 7
C-1-7 66 FM Tu. V-3 HC 9 25 Hgmm D2 1 10
C-1-8 57 FM Tu. V-4 HC 8 10 Hgmm D1 5 30
C-1-9 28 M Tu. V-4 HC 10 25 Hgmm D2 3 16
C-2-1 18 M MGITIS? Neg. 15 6 Hgmm NA NA NA
C-2-2 43 FM SAH? Neg. 15 8 Hgmm NA NA NA
C-2-3 19 M MGITIS? Neg. 15 8 Hgmm NA NA NA
C-2-4 59 M MGITIS? Neg. 15 5 Hgmm NA NA NA
C-2-5 72 FM SAH? Neg. 15 10 Hgmm NA NA NA
856 O. Farkas et al.
protein (equal loading from all samples) were electrophoresed in 6.5%
Tris=glycine gels. Next, separated proteins were transferred to nitrocel-
lulose membranes. Blots were blocked for 1 hour in 1% non-fat milk in
TBS with 0.1% Tween-20 (TBST). The presence of non-erythroid alpha
II-spectrin and its breakdown products were examined with monoclonal
mouse anti-spectrin antibody (Chemicon) capable of detecting intact
non-erythroid alpha II-spectrin (280 kD) and 150, 145 and 120 kD clea-
vage fragments. (These proteins in the CSF detected by the antibody are
exclusively derived from the central nervous system as erythroid spectrin
is not recognized by the antibody applied. This premise excludes the
possibility of false positive reactions due to intraventricular haemor-
rhages and focal lesions leading to the accumulation of erythroid spec-
trin in the CSF.)
Following 1 hour incubation at room temperature with this primary
antibody (1:500 in 1% non-fat milk TBST) blots were incubated with
biotinylated anti-mouse Ig (Amersham) in 1:400 dilution in 1% non-fat
milk TBST for 1 hour at room temperature, next were incubated with
Streptavidin-Biotinylated Horseradish Peroxidase Complex (Amersham,
1:1000) for half an hour at room temperature. Western LightingTM
Chemiluminescence reagents were used to visualize immunolabelling
on Kodak films. Semi-quantitative evaluation of detected proteins was
performed by computer-assisted densitometric scanning (Bio-Profile
Bio-1Dþþ, Vilber Lourmat). Data were normalized to background
density as well as negative control reacted without the presence of the
antigens. Densitometric data gained from CSF samples of TBI-patients
and controls was compared with an independent samples t-test.
Differences were assumed significant at a level of p � 0.05. Correla-
tion between protein levels and clinical parameters was tested by
linear regression.
Results
Thirty-four CSF samples from the 12 brain-injured pa-
tients in addition to 15 samples from 9 control patients
were evaluated in the current investigation (Table 1).
Analysis of all protein bands examined demonstrated
that intact (280 kDa) spectrin as well as the 120 kDa
breakdown product was present in significantly higher
percentage of TBI-patients than in patients with raised
ICP of different etiologies (Table 2).
The CSF samples from patients undergoing routine,
diagnostic LP (second control group) contained neither
intact spectrin nor SBDP.
Selected patients (on Table 1 these patients could be
identified as T-1 and T-2) with severe TBI were followed
for several days to explore the time course of spectrin re-
lease in CSF. One patient (GCS¼ 4, GOS¼ 2, ICP¼ 35
post admission) revealed a peak of both intact spectrin
and SBDPs on the third day (Table 3 and Fig. 1a and c).
The density of all proteins decreased to the baseline
levels by the day 7–8 followed by another increase in
the density of 150 kDa spectrin at the 8th day post-injury.
Similarly, another patient (GCS¼ 4, GOS¼ 2, ICP¼ 20
at admission) had a peak of SBDPs at the second day
followed by a density decrease, however by the fourth
day postinjury another increase was observed in the
150 kDa spectrin level as well as in the density of intact
and 145 kDa spectrin (Table 3 and Fig. 1b and c).
Independent samples t-test was used to compare the
density of protein bands taken from head injured patients
versus those suffering from other central nervous system
disorders (clinical data of all patients are summarized in
Table 1). One sample from each patient, taken on the first
Table 2.
Band (kDa) Occurrence in trauma 95% CI value Occurrence in control 95% CI value Significance
280 66% �26.7% 11% �20.4% �150 100% – 88% �21.2% –
145 75% �24.5% 44% �32% –
120 75% �24.5% 22% �27% �
Table 3.
Date Band
280 kDa 280 kDa 150 kDa 150 kDa 145 kDa 145 kDa 120 kDa 120 kDa
patient 1 patient 2 patient 1 patient 2 patient 1 patient 2 patient 1 patient 2
12 hrs 0 0 309 0 0 0 0 0
24 hrs 178 0 357 382 87 119 79 0
2 days 523 580 749 1383 749 1383 181 357
3 days 908 155 1406 240 1406 79 351 122
4 days 575 243 723 221 723 605 0 121
5 days 421 0 555 0 555 0 0 0
6 days 245 0 333 382 333 119 0 0
7 days 118 580 293 1383 0 1383 0 357
8 days 0 155 398 240 0 79 0 122
9 days 0 243 318 221 0 605 0 121
Spectrin breakdown products in the CSF in severe head injury – preliminary observations 857
to third day post-injury, was analyzed based upon the
observations described above regarding the time course
of SBDP levels in ventricular CSF. In those TBI patients,
where multiple samples were available, maximal levels
have been considered. In the first control group, where
ventriculostomy was performed to decrease ICP primarily
evoked by occlusive or malresorptive hydrocephalus and
oedema, maximal levels detected at the insertion of the
ventricular drainage were considered.
The result of the densitometric analysis and statistical
comparison is summarized in detail in Table 4. In sum,
Table 4.
Band Diagnosis Mean
density
SEM P Significance
280 kDa TBI 251.083 �42.41 0.031 �Control-I 33.111 �16.55
150 kDa TBI 622.5 �64.40 0.023 �Control-I 268.55 �28.25
145 kDa TBI 414.33 �80.65 0.055 –
Control-I 68 �13.75
120 kDa TBI 179.91 �19.57 0.012 �Control-I 44.66 �14.84
Fig. 2. Characteristic Western blots (a) and bar charts (b) indicating ventricular CSF level of non-erythroid alpha II-spectrin and its breakdown
products in TBI and non-injured control patients with other pathological conditions associated with raised ICP. (Blots from patients underwent
diagnostic lumbar punction are not shown.) Asterisks indicate significance of difference, error bars represent standard error of means, data derive
from TBI and non-injured control patients, 34 versus 15 samples, respectively). P < 0.05
Fig. 1. Western blots ((a) patient 1, (b) patient 2) and charts (c)
indicating time course in densitometric units of the accumulation of non-
erythroid alpha II-spectrin and its breakdown products in ventricular CSF
of two severe head injured patients. (solid line: patient 1, dotted line:
patient 2, &: 280 kDa intact spectrin, ^: 150 kDa SBDP, ~: 145 kDa
SBDP, �: 120 kDA SBDP)
858 O. Farkas et al.
non-erythroid alpha II-spectrin and SBDP protein band
densities were significantly higher in CSF of TBI pa-
tients than non-TBI controls (Fig. 2a, b).
Association between Glasgow Coma Scale (GSC),
Glasgow Outcome Scale (GOS), ICP and protein levels
was also examined to evaluate the potential correlation
between clinical and neurobiochemical parameters.
Most probably due to the relatively low number of
patients linear regression analysis did not reveal a sig-
nificant correlation either between the accumulation of
various SBDPs and injury severity (defined by GCS),
outcome (defined by GOS) or ICP.
However, it is of note that the negative association
between injury severity (defined by GCS) was close to
the level of significance (280 kDa: r2¼ 0.318 P¼ 0.07;
150 kDa: r2¼ 0.245 P¼ 0.121; 145 kDa: r2¼ 0.209
P¼ 0.156).
Discussion
This is the first study to demonstrate the accumulation
of non-erythroid alpha II-spectrin and its breakdown
products in the ventricular CSF of patients with severe
TBI.
It is of particular importance that CSF-samples from
control patients, undergone diagnostic lumbar puncture
without raised ICP=opening pressure=were proved nega-
tive for intact, non-erythroid alpha II-spectrin as well as
its breakdown products, a finding clearly indicating that
the accumulation of these substances is associated with
brain injury and=or raised ICP.
Interestingly, non-erythroid alpha II-spectrin and
SBDPs reached significantly higher levels in TBI than
in other pathological conditions associated with com-
parably elevated ICP, indicating that these changes were
in direct response to the injury and not a secondary re-
sponse to intracranial hypertension. The fact that only
two patients were followed for several days in this study
clearly reflects the preliminary nature of our observation
however it is of note that the peak values as well as the
time course of those markers (280 kDa and 120 kDa)
which displayed significant alterations in some aspects
of our analysis were rather consistent in both cases.
Recent studies in animal models of head injury provide
further credit to this observation in terms of describing
similar, reliable time course of SBPD accumulation in
head injured rodents [17].
To date, experimental studies have demonstrated that
CSF levels of non-erythroid alpha II-spectrin and SBDPs
are elevated following ischemic stroke [2, 12]. In these
ischemic studies it was suggested by Bartus and col-
leagues that calpain-mediated structural proteolysis was
responsible for brain damage associated with cerebral
ischemia [2]. These observations reported in cerebral
ischemia suggested that traumatically induced brain tis-
sue destruction could lead to the CSF accumulation of
these substances following TBI, a premise consistent
with our current study.
Recent studies provided further evidence that calpain-
and caspase-mediated proteolytic cleavage fragments
should be considered future surrogate markers in central
nervous system injury of various origin [19].
Moreover, based upon our previous observations re-
garding the role of calpain- and caspase-mediated struc-
tural proteolysis in the pathogenesis of experimental TBI
[4, 5] we anticipated that the magnitude of spectrin
breakdown in TBI would exceed those levels observed
in other CNS-disorders associated with raised ICP.
Again, these assumptions were supported by our find-
ings that revealed significant elevation of calpain and
caspase-3 specific SBDPs in TBI patients versus patients
suffering from other CNS disorders. These observations
clearly indicate that intact brain spectrin – and SBDP-
levels are closely associated with those specific neuro-
chemical processes evoked by and=or operant in TBI.
What remains to be determined in larger population of
TBI patients is whether or not differing types of TBI are
associated with specific patterns and degrees of cysteine-
protease mediated damage. For example would a patient
sustaining primary focal lesion and=or mass lesion forma-
tion show a different response than a patient demonstrating
primary diffuse injury. This issue will be considered in
future studies in a larger population of brain – injured
patients who have sustained different types of primary
injury.
Most probably due to the relatively small number of
patients enrolled to this study we were not able to estab-
lish statistically significant association between injury-
severity, ICP values and outcome and the accumulation
of SBDPs, thus potential correlations between SBDPs
and clinical parameters require further study in a larger
patient population. Nevertheless, the observations sum-
marized in the above passages as well as some of the
results of the analysis of the correlation between
SBDP-levels in CSF and injury severity indicate that
non-erythroid alpha II-spectrin and SBDPs should be
considered as candidate biomarkers in TBI.
The current observations, however, do provide evi-
dence for the contribution of calcium-induced calpain-
and caspase-mediated proteolysis in the pathogenesis of
Spectrin breakdown products in the CSF in severe head injury – preliminary observations 859
human TBI, as previously suggested in multiple animal
studies [13, 9, 18].
In sum, the current, preliminary studies indicate the
utility of cerebrospinal fluid monitoring for cysteine-
protease-mediated proteolytic damage as a potential
marker for evaluating TBI patients. Obviously, these
studies require expansion to address some of the ques-
tions alluded to above. Further, they might require the
use of alternative approaches including but not limited to
ELISA analyses, to further enhance the usefulness of
these measurements in the routine clinical setting.
Aknowledgement
The authors are grateful to Dr. Istvan M�eeszaros, Dr. Zsolt Horvath and
Mrs. Paula Turschl-Juhasz (Neurosurgical ICU, Medical Center, P�eecs
University) and �AAgnes Kiss, Csaban�ee Andok, J�oozsef Nyiradi and Ern€ooBognar for their technical assistance as well as Dr. Laszl�oo Bors
(Department of Neurology) for kindly providing CSF from ‘‘negative’’
LPs.
This work was supported by the National Science Research Fund
(T035195, OTKA T047824), Fogarty IRCA (1-RO3-TW0131302A1),
NIH Grant NS20193, National Science Projects NFKP1A=00026=
2002 and ALK 000126=2002, the Hungarian Academy of Sciences,
Bolyai and Bekesy Scholarships.
References
1. Banik NL, Shields DC (2000) The role of calpain in neurofilament
protein degradation associated with spinal cord injury. Methods
Mol Biol 144: 195–201
2. Bartus RT (1997) The calpain hypothesis of neurodegeneration:
evidence for a common cytotoxic pathway. The Neuroscientist 3:
314–327
3. Bartus RT, Dean RL, Mennerick S, Eveleth D, Lynch G (1998)
Temporal ordering of pathogenic events following transient global
ischemia. Brain Res 790: 1–13
4. Brana C, Benham CD, Sundstrom LE (1999) Calpain activation and
inhibition in organotypic rat hippocampal slice cultures deprived of
oxygen and glucose. Eur J Neurosci 11: 2375–2384
5. Buki A, Okonkwo DO, Wang KK, Povlishock JT (2000)
Cytochrome c release and caspase activation in traumatic axonal
injury. J Neurosci 20: 2825–2834
6. Buki A, Siman R, Trojanowski JQ, Povlishock JT (1999) The role
of calpain-mediated spectrin proteolysis in traumatically induced
axonal injury. J Neuropathol Exp Neurol 58: 365–375
7. Del Bigio MR (2000) Calcium-mediated proteolytic damage in
white matter of hydrocephalic rats. J Neuropathol Exp Neurol 59:
946–954
8. Kampfl A, Posmantur R, Nixon R, Grynspan F, Zhao X, Liu SJ,
Newcomb JK, Clifton GL, Hayes RL (1996) Mu-calpain activation
and calpain-mediated cytoskeletal proteolysis following TBI.
J Neurochem 67: 1575–1583
9. Kampfl A, Posmantur RM, Zhao X, Schmutzhard E, Clifton GL,
Hayes RL (1997) Mechanisms of calpain proteolysis following
TBI: implications for pathology and therapy: implications for
pathology and therapy: a review and update. J Neurotrauma 14:
121–134
10. Lewin ICF (1992) The Cost of Disorders of the Brain, Washington
DC: The National Foundation for Brain Research. (Update of
Max W, MacKenzie EJ, Rice DP (1991) Head Injuries: Cost and
Consequences. J Head Trauma Rehab
11. McCracken E, Hunter AJ, Patel S, Graham DI, Dewar D (1999)
Calpain activation and cytoskeletal protein breakdown in the corpus
callosum of head-injured patients. J Neurotrauma 16: 749–761
12. Newcomb JK, Kampfl A, Posmantur RM, Zhao X, Pike BR, Liu SJ,
Clifton GL, Hayes RL (1997) Immunohistochemical study of
calpain-mediated breakdown products to alpha-spectrin following
controlled cortical impact injury in the rat. J Neurotrauma 14:
369–383
13. Pike BR, Flint J, Dave JR, Lu XC, Wang KK, Tortella FC, Hayes RL
(2004) Brain-derived alpha II-spectrin in cerebral spinal fluid after
middle cerebral artery occlusion in rats. J Cereb Blood Flow Metab
24(1): 98–106
14. Pike BR, Flint J, Dutta S, Johnson E, Wang KK, Hayes RL (2001)
Accumulation of non-erythroid alpha II-spectrin and calpain-
cleaved alpha II-spectrin breakdown products in cerebrospinal fluid
after TBI in rats. J Neurochem 78: 1297–1306
15. Pike BR, Zhao X, Newcomb JK, Glenn CC, Anderson DK, Hayes
RL (2000) Stretch injury causes calpain and caspase-3 activation
and necrotic and apoptotic cell death in septo-hippocampal cell
cultures. J Neurotrauma 17: 283–298
16. Pike BR, Zhao X, Newcomb JK, Posmantur RM, Wang KK, Hayes
RL (1998) Regional calpain and caspase-3 proteolysis of alpha-
spectrin after TBI. Neuroreport 9: 2437–2442
17. Ringger MC, Silver X, O’Steen B, Brabham JG, De Ford SM, Pike
BR, Pineda J, Hayes RL (2002) CSF accumulation of calpain-
specific aII-spectrin breakdown products are associated with injury
magnitude and lesion volume after traumatic brain injury in rats.
J Neurotrauma 19: 1302
18. Saatman KE, Bozyczko-Coyne D, Marcy V, Siman R, McIntosh TK
(1996) Prolonged calpain-mediated spectrin breakdown occurs
regionally following experimental brain injury in the rat. J Neuro-
pathol Exp Neurol 55: 850–860
19. Siman R, McIntosh TK, Soltesz KM, Chen Z, Neumar RW, Roberts
VL (2004) Proteins released from degenerating neurons are surro-
gate markers for acute brain damage. Neurobiol Dis 16: 311–320
20. Wang KK (2000) Calpain and caspase: can you tell the difference?Trends Neurosci 23: 20–26
21. Wang KK, Posmantur R, Nath R, McGinnis K, Whitton M, Talanian
RV, Glantz SB, Morrow JS (1998) Simultaneous degradation of
alpha II- and beta II-spectrin by caspase 3 (CPP32) in apoptotic
cells. J Biol Chem 273(35): 22490–22497
Comments
Traumatic brain injury is a very complex pathology due to the vari-
ability of tissue damage encountered in the individual patient, and also for
the multitude of biochemical cascades activated generating a variety of
secondary insults. It is for these reasons, that the identification of bio-
markers specific to selective types of brain damage (focal vs diffuse) is
of great importance to 1) discriminate the biochemistry involved, 2) be
applied for diagnostic purposes, 3) for monitoring the onset of secondary
damage, and 4) as predictors of outcome. Most studies published
recently have shown a direct correlation between the levels of S-100
and the presence and size of focal brain damage, however to date there
are no reliable markers to estimate the degree of axonal damage.
Previous studies by Farkas and collaborators based on an animal
model of axonal injury in the rat, showed an association with foci of
axonal damage with mitochondrial dysfunction, calpain and caspase-3
activation and calpain-mediated spectrin proteolysis, suggesting their
participation not only in neuronal cell death but also in axonal dis-
connection (Buki et al, J Neuropathol Exp Neurol 58: 365, 1999 &
J Neurosci: 20: 2825, 2000). Furthermore, the same group showed a
860 O. Farkas et al.
decreased number of injured axons when rats were treated with a calpain
inhibitor (Buki et al J Neurotrauma 20: 261, 2003). The present study,
reports for the first time that spectrin breakdown products are elevated in
human CSF in response to different neuropathologies including trauma
as well as focal brain injury. It remains to be established whether these
levels are related to axonal injury possibly coexistent with focal lesions.
The patient population analysed is undersized to establish any associa-
tion; therefore, further investigation on larger groups of patients with
either type of brain damage is warranted before drawing any conclusions.
M. C. Morganti-Kossman
Melbourne
This manuscript reports on analysis of cerebrospinal fluid (CSF)
samples collected from a small group of patients who sustained a TBI
compared to two control groups. The levels of non-erythroid spectrin
and spectrin breakdown products (SBDPs) were measured by western
blotting. This is essentially a confirmatory study, using human samples,
of a body of evidence from animal studies implicating calpain- and
caspase-mediated spectrin breakdown in the pathophysiology of TBI.
However, it is essential that such human studies are performed to assess
the relevance of measures made in animal models and therefore the
study is of clinical significance. The data indicate that after human
TBI, levels of spectrin and SBDPs in CSF are elevated compared to a
group which had raised ICP due to causes other than TBI and samples
taken for diagnostic lumbar puncture.
Debbie Dewar
Glasgow
Correspondence: Orsolya Farkas, Department of Anatomy and Neu-
robiology, Medical College of Virginia, Virginia Commonwealth Uni-
versity, Richmond, VA 23298-0709. e-mail: [email protected]
Spectrin breakdown products in the CSF in severe head injury – preliminary observations 861
![szövettani metszetek I [Kompatibilis mód]Oedema pulmonis Makroszkópia Lokalizáció Diffúz (basalisan kifejezettebb) Mintázat Homogén Szín Vörös Konzisztencia Kissé tömött](https://img.pdfslide.net/doc/110x75/5f22db1dfcb3514d0f55442d/szvettani-metszetek-i-kompatibilis-md-oedema-pulmonis-makroszkpia-lokalizci.jpg)
