http://jhc.sagepub.com/Journal of Histochemistry & Cytochemistry

http://jhc.sagepub.com/content/28/7/636The online version of this article can be found at:

 DOI: 10.1177/28.7.6967079

1980 28: 636J Histochem CytochemE C Azmitia and W F Marovitz

pharmacological approach to specificity.In vitro hippocampal uptake of tritiated serotonin (3H-5HT): A morphological, biochemical, and

  

Published by:

http://www.sagepublications.com

On behalf of: 

  Official Journal of The Histochemical Society

can be found at:Journal of Histochemistry & CytochemistryAdditional services and information for     

  http://jhc.sagepub.com/cgi/alertsEmail Alerts:

 

http://jhc.sagepub.com/subscriptionsSubscriptions:  

http://www.sagepub.com/journalsReprints.navReprints:  

http://www.sagepub.com/journalsPermissions.navPermissions:  

by guest on July 14, 2011jhc.sagepub.comDownloaded from

‘This work was supported by Grant BNS-79-06474 from the

National Science Foundation and a Hirschl Career Scientist Award to

E.C.A.

636

0022- 1 554/80/0�0#{244}3609S02. 25

The Journal of Histochemistry and CytochemistryCopyright © 1980 by The Histochemical Society, Inc.

4

Vol. 28, No. �‘, pp. 636-644, 1980

Przntedin U.S.A.

In Vitro Hippocampal Uptake ofTritiated Serotonin(3H-5HT):

A Morphological, Biochemical, and Pharmacological Approach to

EFRAIN C. AZMITIA and WILLIAM F. MAROVITZ

Department ofAnatomy. Mount Sinai School of Medicine, Nets York, Nets York 10029

Received for publication September 18, 1979 and in revised formJanuary 10, 1980 (MS 79-226)

The in vitro uptake of tritiated serotonin (3H-5HT) intohippocampal slices was measured in Ringer’s solution

(37#{176}C)containing pargyline, ascorbic acid, and dextrose.The specific uptake of 3H-5HT rose asymptotically as the3H-5HT molarity was increased from 5 x 1O�#{176}to 1.5 x

lO� M. Linear regression analysis gave a Km value for the

specific uptake of 1.4 x 1O� M. The nonspecific binding

(NSB) was the amount of 3H-5HT retained by the slices

following incubation in a medium with a very large excess

of unlabeled 5-HT added to dilute the specific uptake of

3H-5HT. This NSB increased with increasing molarity of

3H-5HT, and was linearly related to 3H-5HT concentra-

tions between 5 x 1O� and 1.5 x 1O� M. The ratio of

specific uptake to NSB was highest at 5 x 1O� M (2.75)

and lowest at 1.5 x 1O� M of3H-5HT (0.54). Competition

studies with noradrenaline, desipramine (a noradrenergic

uptake blocker), fluoxetine (a 5-HT uptake blocker), and

tryptophan confirmed the specificity of the 3H-5HT up-

take mechanism. Radioautographic studies of in vitro

incubated hippocampal slices showed silver grain aggre-

gates at 3H-5HT specific uptake sites. Addition of an

excess of unlabeled 5-HT to the slices, or the use of

IntroductionThe visualization of chemically identified neurons provides

the structural basis for the interpretation of biological func-

tions unique to the brain. The class of neurons which use

serotonin (5-HT) as a transmitter substance was first identified

by histochemical methods (fluorescence (20), potassium di-

chromate (40,41)). Although other techniques have been suc-

cessfully applied for the visualization of these monoaminergic

hippocampi from 5,7-dihydroxytryptamine intracerebral

microinjected rats (5 �Lg/4OO nI into the fornix-fimbria

and the cingulum bundle, 6 day survival) caused a dra-

matic decrease in these aggregates. The distribution of

hippocampal 5-HT axons and terminals, inferred from the

pattern of silver grain aggregates, is more widespread

than previously described. 5-HT varicosities were clearly

seen in all layers of Ammon’s horn, dentate gyrus, and the

subicular cortex. Innervation routes were seen to the

stratum radiatum and stratum lacunosum from stratum

oriens in Ammon’s horn, and to the polymorphic layer of

the dentate gyrus from the subicular cortex and from the

fimbria. Semiquantitation ofthe occurrence ofsilver grain

aggregates was done in the various hippocampal regions.

The highest density in Ammon’s horn was 1 19.5 boutons/10,000 .tm2, in the dentate gyrus it was 67.4 boutons/

10,000 �m2, and in the subicular cortex it was 79.2boutons/1O,000 �am2. These results are consistent with

previous quantitative results. KEY WORDS: Serotonin;Hippocampus; In vitro; Specific uptake; Radioautography;

5,7-Dihydroxytryptamine; Fluoxetine; Semiquantitation.

neurons, radioautographic procedures based on in vivo uptake

of tritiated SHT (3H-SHT) are reputed to be most sensitive

and specific for descriptive and quantitative work.

The radioautographic localization of serotonergic neurons

depends on the uptake and storage of exogenous 3H-SHT.

The in vivo administration of 3H-SHT into the brain can be

achieved by intracisternal injection (9), intraventricular injec-

tions (2, 13, 15, 25), topical application (14, 18), or direct

intracerebral injection (25, 26). High molarity S-HT solutions

are required to insure maximum labeling of the neurons in a

given area of tissue. Nevertheless, the diffusion of 3H-SHT

through brain tissue is slow and uneven. The penetration of

3H-SHT after intraventricular injections may be confined to

the periventricular zones (1, 15). Thus, topographical and

by guest on July 14, 2011jhc.sagepub.comDownloaded from

HIPPOCAMPAL UPTAKE OF SEROTONIN 637

quantitative analysis of regions at different distances from the

source of the exogenous 3H-SHT can not be directly corn-

pared.

A further constraint is establishing the specificity of the

radioautographic labeling. The high affinity uptake of 3H-SHT

has a K,, equal to 1.4-1.7 x 10� M in the brain (38). At a

5-HT molarity near this K,,,, the in vitro uptake process has

been shown to be preferentially blocked by selective uptake

blockers. For example, fluoxetine is effective in blocking

3H-SHT binding by brain slices (19, 39), while desiprarnine is

ineffective against the uptake of serotonin (29). These

biochemical and pharmacological studies are not directly

applicable to tissue labeled by in vivo administration of

3H-SHT, since the molarity neither of the 3H-SHT nor of

uptake blocker can be controlled.

The specificity of 3H-SHT uptake can be established for

these in vivo experiments by the use of selective neurotoxic

drugs or localized lesions of particular monoaminergic nuclei.

However, these controls depend on a double negative result:

the uptake sites must be absent in the morphological study and

this absence must be verified by biochemical measures of the

transmitter chemical ( 18).

An approach to circumvent many of these problems is in

vitro incubation of tissue slices with a known molarity of

chemical transmitter. Such an approach has lead to the locali-

zation of a variety of transmitters such as GABA (22), glycine

(21, 22, 31), glutamate (23), and norepinephrine (NE) (27). In

vitro localization of3H-5HT has been done in the snail central

nervous system (CNS) (33), in nuclei-free homogenates of the

rat brain (24), in the median emminence ofseveral species (13,

14), and in the spinal cord of the cat (37). This latter study

provided good topographic and quantitative data with a molar-

ity of 3H-5HT of iO� M. Furthermore, control experiments

with concomitant 106 M of NE or 2 x 10_6 M of fluoxetine

produced an unmodified or a negative radioautographic pat-

tern, in that order. We have undertaken to study the biochem-

ical and pharmacological properties of 3H-5HT retention and

uptake by hippocampal slices, and to directly correlate these

parameters with morphological observations following

radioautography.

Materials and Methods

Sprague-Dawley, female rats weighing 220-250 g were decapitated

under light ether anesthesia and the brain was rapidly removed and

placed in cold oxygenated Ringer’s bicarbonate solution (containing

pargyline (10� M), ascorbic acid (10� M), and dextrose (102 M)).

The hippocampus was removed by blunt dissection and was cut into

0.2-0.3 mm slices using a double edge razor blade. These slices were

incubated in 2 ml of fresh oxygenated Ringer’s solution at 37#{176}Cin a

shaking water bath for 1 5 mm with various concentrations of 3H-5HT.

The slices were then processed either for scintillation counting or for

radioautography.

Scintillation counting. At the end of the incubation, the vials

containing the tissue slices were placed on ice and 2 ml of cold

Ringer’s solution (0#{176}C)was added to each vial. The slices were briefly

rinsed in a large volume of cold Ringer’s solution and weighed. The

samples were dissolved in 0.2-0.5 ml of Protosol (New England

Nuclear) and counted in 9 ml of Econofluor (New England Nuclear).

Radioautography. The vials were removed from the shakingwater bath and the slices were carefully transferred to vials containing

a solution of 2.57r glutaraldehyde in 0.1 M cacodylate buffer (or

phosphate buffer), pH 7.2, supplemented with 0.05% MgC12 and 3%

sucrose. The slices were fixed for at least 3 hr and then placed in the

same supplemented buffer used for the fixative. The next day the

slices were dehydrated and embedded in paraffin. Sections were cut at

5 /L, mounted on clean slides using albumin as an adhesive, and were

allowed to dry overnight at 37#{176}C.The deparaffinized, and rehydrated

slides were then coated with NTB-2 nuclear track emulsion (Eastman

Kodak, diluted 1:1 with a 5#{231}/�solution ofgelatine). The slides were

immediately cooled on a tray placed over ice and dried overnight in a

horizontal position. The slides were exposed for 1-3 months at 5#{176}C.

They were then developed for 2 mm in freshly prepared D-170

developer (Eastman Kodak formula), fixed in Kodak Rapid fix, and

washed. The slides were counterstained with cresyl violet, dehy-

drated, and cover slipped.

Neurotoxin injections. Rats were injected with desipramine

hydrochloride ( 10 mg/kg) 45 mm before intracerebral-microinjection

of 5,7-dihydroxytryptamine (5,7-DHT) (5 �g/400 nI of Ringer’s

solution containing ascorbic acid, 0.2 mg/mI). Three microinjections

were made in each rat: one into the midline of the fornix-fimbria (6.9

mm anterior to lambda suture, lateral (15#{176})1.0 mm, and ventral (15#{176})

4.5 mm); and two, one on each side, into the cingulum bundle (6.9

mm anterior to lambda, lateral (15#{176})1.0 mm, and ventral (15#{176})2.8

mm). The animals were kept for 6 days before the hippocampi were

removed for uptake studies.

Microscopy. Slides were viewed under bright field illuminationusing a Leitz Orthoplan microscope with a Leitz 25X PlanApo and a

Zeiss 63X Planapo objective. Photomicrographs were taken with a

Wild MPS S 1 automatic camera using 35 mm (Pan X 32 ASA) or 5

in. X 4 in. (Kodak Ektapan 100 ASA) negatives.

The “semiquantitative” results on bouton density were obtainedby counting the number of silver grain aggregates (SGA) present in

the 5 �a paraffin sections from 3 rats. The number of varicosities per

unit area was calculated from the number of SGA found in different

regions of the hippocampus using an ocular reticle delimitating an area

1,590 �am2. At least 10 counts were made in each hippocampal region

in each rat.

Drugs and chemicals. Desipramine hydrochloride was provided

by Merrell Research Center of Richardson-Merrell Inc., and

fluoxetine hydrochloride was provided by Lilly Research Laboratories

of Eli Lilly and Company. 3H-SHT (hydroxytryptamine creatinine

sulfate 5-( 1,2-3H(N)), 29.2 Ci/mmol) was purchased from New

England Nuclear. All other drugs and chemicals were obtained from

Sigma.

Results

Biochemical/Pharmacological Studies

Kinetic analysis. Hippocampal slices were incubated at

37#{176}Cfor 20 mm with different molarities of 3H-SHT (Table

1). The total binding of3H-SHT measured in the slices rose as

the molarity of 3H-5HT was increased from 5 x 10i0 M (16

prnol/g) to 1.5 X 1O_6 M (5,335 prnol/g) (Figure 1).

This represents the binding of 3H-SHT to a variety of

compartments. To calculate the specific reuptake compart-

ment of serotonergic neurons, the slices were incubated with a

large excess ofunlabeled 5HT (1O� M) added to the medium

containing the 3H-SHT. Thus, the saturable specific uptake

by guest on July 14, 2011jhc.sagepub.comDownloaded from

TOTAL

I

SPECIFIC

638 AZMITIA, MAROVITZ

5HT Concentration xi08

Table 1. In zitro hippocampal uptake of3H-5HT

Concentration of3H-5HT#{176}

(M)Total

(pmolIg)

Nonspecific”(NS)

Specifict(5)

Ratio

(S/NS)

5 x 10b0 16 ± 7 7.6 ± 0.7 8.4 1.1

5 x 10’ 82 ± 14 27 ± 2 55 2.03

6 x 10_8 502 ± 109 134 ± 26 368 2.75

1.5 x 10� 1315 ± 163 400 ± 50 915 2.28

1.5 x 106 5335 ± 483 3460 ± 117 1875 0.54

‘Concentration of 3H-5HT (29.2 Ci/mmol) in 2 ml of oxygenated Ringer’s

solution containing pargyline, ascorbic acid, and dextrose. Incubation for 1 5 mmat 37#{176}C.Three determination at each concentration, averages and standard

deviation given.

b10-4 M unlabeled 5-HT.

�Specific Uptake = Total Uptake - Nonspecific Uptake.

would be greatly diluted, while the nonsaturable nonspecific

binding (NSB) of 3H-SHT would not be altered. The NSB

increased linearly with increasing 3H-SHT concentration (Fig-

ure 1). No evidence for saturability was found within the

concentration range examined. The 3H-SHT specific uptake

was defined as total tissue binding minus the NSB. Plotting

this value against 3H-5HT molarity produced an asymptotic

curve with the uptake being essentially saturated by 1.5 x

106 M concentration of 3H-SHT (Figure 1). Converting the

values for the specific uptake into their reciprocal, a

Lineweaver-Burk analysis was performed (Figure 2). The

straight line was determined by linear regression. This gave a

Km value of 1.4 x 1O� M and a Vmax value of 1,562 pmol/g/

15 mm of 3H-5HT. The ratio of specific uptake to NSB of

3H-SHT in the hippocampus was the highest at 5.0 X 10� M

(2.75), while the lowest ratio occurred at 1.5 X 106 M of

3H-5HT (0.54).

Nonspecific binding. Incubation of the slices at 0#{176}Chas

been suggested as an appropriate means to differentiate

between specific uptake and NSB (28). However, incubating

hippocampai slices at 0#{176}Cresulted in a 95% reduction in

3H-SHT tissue retention when compared to incubating at

37#{176}C. These results suggest a 95:5 ratio between specific

uptake and nonspecific retention, as contrasted to a 73:27

ratio when excess 5-HT is used. To determine if the 5%

3H-5HT retention at 0#{176}Cwas indeed nonsaturable and non-

specific, an excess of unlabeled 5-HT was added to the 0#{176}C

medium containing 3H-5HT. A further reduction in tissue

retention to 1% of the control value was obtained; indicating

that the 0#{176}Cincubation was not an accurate measure of the

nonspecific uptake of3H-5HT by hippocampal slices.

The measure of specific uptake can also be approached by

in vivo destruction of the S-HT fibers with a specific

neurotoxic drug. S,7-DHT was injected into the fornix-

fimbria and cingulum bundle of rats 6 days before death. The

binding of 3H-SHT (S x 1O_8 M) was reduced to 3 1% of

control levels, giving a specific uptake value of 69% (Table 2).

This measure of specific uptake is in good agreement with that

determined with excess S-HT (73%) but not with the specific

uptake determined by 0#{176}C(95%).

A test of the selectivity of uptake was performed by

incubation of hippocampal slices in the presence of 1O� M of

unlabeled tryptophan. This had no effect on 3H-5HT uptake

(Table 2). This result indicated that the uptake mechanism in

the hippocampus was not a general indole transport system.

Competition with norepinephrine. Incubation of hip-

pocampal slices was performed with various concentrations of

unlabeled NE (Table 2). At 10� and 10_6 M of NE, no

significant competition was detected in the binding of

S x 10_8 M of3H-SHT (87 and 80% ofcontrol). However, at

higher molarities of NE (1 x 1O� and 2 x 1O� M) significant

competition did occur (66 and 37% of control levels, respec-

tively).

Figure 1. Graphic representation of the veloc-

ity of 5-HT accumulation at varying 5-HTconcentrations as shown in Table 1. The totalbinding is the amount of 5-HT retained after

15 mm incubation at 37#{176}C.The NSB is the

fraction of the total binding that remains within

the tissue after incubation with a large excess of

unlabeled 5-HT. This represents a nonsatura-

ble compartment. Finally the specific uptake is

the total binding minus the NSB. Slices were

incubated in triplicate.

by guest on July 14, 2011jhc.sagepub.comDownloaded from

3H 5H1 Uptake ii Hippocimpal slices

.

50

01�’5IIT x 1O�

HIPPOCAMPAL UPTAKE OF SEROTONIN 639

Figure 2. Graphic analysis of the reciprocals of

3H-5HT molarity and the specific uptake intohippocampal slices as shown in Figure 1 and

Table 1. Linear regression analysis gave a

Km 1.4 x 10� M and a Vmax 1569 pmol/g/1 5 mm. No evidence of a second low afinityuptake can be seen.

100

I ��o4L�dake

Table 2. 3H-5HT (5 x 108M) uptake in hippocampa/slicesa

Treatment Concentration (M) pmol/g Percentage

Normal(12) ---- 245 ± 59 100

S-HT(9) I X 10-s 67 ± 18 27

S,7-DHT(3) INVIVO 77 ± 9 31

Fluoxetine(3)

(3)

(4)

1 X 10�

1 x 10-#{176}

1 x 10-v

91 ± 7

111 ± 17

140 ± 29

37

45

57

Tryptophan(3) 1 x 10� 233 ± 27 95

Norepinephrine(4)

(4)

(4)(3)

2 x 10�

1 x 10-i

1 x 10-s

1 x 10-s

90 ± 13

161 ± 20195 ± 40213 ± 33

37

66

80

87

Desipramine (4)

(DM1) (4)

(3)

(3)

2 x 10-i

1 x 10�

1 x 10-s

1 x 10-i

84 ± 7

177 ± 19

222 ± 6186 ± 51

34

72

9176

5-HTand DM1 (5) 1 x 10� 46 ± 27 19

Cold (7) ---- 13± 5 5

Cold + 5-HT(3) 1 X 10� 3.4 ± 0.3 1

#{176}Tliedrugs and chemicals areadded before #{176}H-5HT, incubation for 1 5 mmnin oxygenated Ringer’s solution with pargyline, ascorbic acid, and dextrose at

37#{176}C.The percentages are based on 100% control uptake. Number ofdeterminations given in brackets. pmollg values are the averages and the

standard deviations.

Uptake blockers. Further evidence for the specificity of

the uptake was obtained by using fluoxetine (Lilly 1 10, 140), a

reputed 5-HT uptake blocker ( 19, 39). This compound pro-

duced a strong inhibition of 3H-SHT binding in hippocampal

slices at a molarity as low as 10� M (Table 2). The 3H-SHT

tissue binding inhibited by this drug (63%) was close to that

found after excess S-HT (73%) or in vivo 5,7-DHT microin-

jection (6 1%).

The role of noradrenergic fibers in binding of 3H-SHT was

tested by using desipramine (DM1), a NE uptake blocker (29).

This compound, at a molarity as high as 10� M (Table 2), had

no significant effect on the uptake of 3H-SHT at S x 1O� M.

However, 2 x 10-i M of desipramine did cause a substantial

inhibition of 3H-5HT binding (66%), most likely resulting

from a loss of selectivity of the drug itself rather than from the

uptake mechanism.

A nonsaturable, low affinity uptake process for 3H-5HT

has been found in the hypothalamus and striatum. This

uptake, which may be a component of the 3H-SHT NSB, is

quite sensitive to inhibition by norepinephrine (38). There-

fore, the proportion of3H-SHT NSB blocked by desipramine

was calculated in a medium containing 3H-5HT and an excess

of unlabeled 5-HT. The results (Table 2) showed that the

NSB was nonsignificantly reduced from 27 to 19% when

DM1 was added. Therefore, retention of 3H-5HT at 5 X

108 M by noradrenergic sites appears to be negligible in the

hippocampus.

Anatomical Findings

Paraffin embedment preparations. The accumulation ofsilver grains in distinct patterns was clearly seen in all regions

of the hippocampus (cornu Ammonis (CA), fascia dentata, and

subiculum). Dense SGA have been shown to represent the

by guest on July 14, 2011jhc.sagepub.comDownloaded from

CA1

so�sP

SR

51

SM

640 AZMITIA, MAROVITZ

concentration of 3H-5HT in axons and terminals (8, 13-18,

25, 26). These labeled boutons stood out against the lighter,

random distribution of silver grains in the background. The

course of the fibers could be easily determined even at

relatively low magnifications (Figure 3). The fibers in the CA,

region of Ammon’s horn extended from the stratum oriens

through the pyramidal cell layer, and turned at right angles

after passing through the stratum radiatum and entering the

stratum lacunosum (terminology: Lorenti de No (30) and

Ramon y Cajal (34)). Numerous fibers were labeled in the

stratum oriens of CA3-CA,, in the stratum radiatum of CA3,

and in all layers of the dentate gyrus. In addition, numerous

varicosities were evident in the subicular cortex, especially

around the pyramidal neurons.

The fibers passing through the cellular layers of Ammon’s

horn and dentate gyrus illustrate the route by which the

deeper layers of the hippocampus are innervated. Conven-

tional neuroanatomical 5-HT tracing techniques (anterograde

transport (4, 5, 10, 1 1, 17, 32) and fluorescence (3, 20)) did

not fully reveal this pattern of innervation.

Neurotoxic microinjections. The conditions of the invitro incubation with a 3H-5HT level of 5 x 10_8 M argue

against nonspecific uptake into nonserotonergic fibers. How-

ever, two controls were used to verify the specificity of the

labeling seen in the radioautographic preparations: in vitro

incubation of normal tissue with an excess amount of un-

labeled 5-HT, and the in vitro incubation of hippocampal

slices from animals previously injected with 5,7-DHT (a

neurotoxic drug (6, 7)). There was a dramatic decrease in the

number of SGA when excess 5-HT was added. Furthermore,

examination of tissue slices from rats injected 6 days previ-

ously with 5,7-DHT into the cingulum bundle and the fornix-

fimbria, also showed a marked suppression of SGA, although

background diffuse reaction (individual silver grains) re-

mained relatively high (Figure 4).

Semiquantitation. The number of SGA differed in the

various hippocampal regions. It was the highest in Ammon’s

horn, 1 1,950 boutons/mm2, in the dentate gyrus it was 6,740

boutons/mm2, and in the subicular cortex 7,920 boutons/mm2.

However, within each region the distribution of SGA was

heterogeneous. For example, in the pyramidal cell layer of

CA1 certain nerve cell bodies had a large number of boutons

surrounding them (8 boutons/962 j.am2); whereas others

nearby had none (Figure 3).

Discussion

The identification and characterization of serotonin nerve

terminals in the hippocampus were achieved with an in vitro

incubation technique coupled with radioautography. This

Figure 3. 5-HT innervation pattern in

the CA1 region. Lower power photo-

micrograph of radioautographic pro-

cessed hippocampal slices using bright

field illumination. The 3H-5HT con-

centrating fibers are easily visible as a

series of black silver grain aggregates

(SGA). The SGA are seen in all layersof the CA1 field. On the right side of

the figure is a schematic drawing,

based on Golgi preparations, showing

a pyramidal cell in this area (27, 30).

On the left side of the figure, the5-HT axons are seen entering from

the stratum oriens (SO) through the

stratum pyramidal (SP). The axons oc-

casionally follow the outline of the

pyramidal neurons. The 5-HT fibers

continue deeper into stratum radiatum

(SR) where numerous short branchesare given off by the main axon. These

axons appear to follow the apical den-

drites of the pyramidal neuron (see

arrows). Finally, in the stratum

lacunosum (SL) the pyramidal den-

drites branch laterally and, likewise,

the SGA fibers appear to change

orientation in the same direction (see

arrows). Picture on left from normal

hippocampus, incubated in vitro with

5 x 10-s M, 3H-5HT for 15 mm, ex-

posed to NTB-2 (1: 1) for 2 months,

Leitz Orthoplan, Leitz 25 X Plan Apo

(NA=0.65), 4 in. x 5 in. (Kodak, Ek-

tapan 100 ASA).

-�-

by guest on July 14, 2011jhc.sagepub.comDownloaded from

HIPPOCAMPAL UPTAKE OF SEROTONIN 641

method avoided several of the disadvantages found after in

vivo uptake studies with 3H-5HT. First, precise control of the

concentration of the 3H-5HT and test substances is main-

tamed with the in vitro method. The importance of this can be

seen by noting that incubation with 10� M NE produced only

a small inhibition of the uptake of 5 x 10_8 M 3H-5HT, but

10-i M produced a 34% inhibition of the 3H-5HT binding.

Second, direct comparison between morphological and

biochemical results are possible. For example, it has been

proposed that the dorsal hippocampus (septal pole) in the rat

is exclusively innervated by 5-HT fibers traveling in the

cingulum bundle and in the fornix-fimbria (4-6). Six days after

5,7-DHT microinjections into these pathways the specific

uptake in the dorsal hippocampus was eleminated, leaving

only the NSB component. Morphological studies confirmed

this. Radioautographs of dorsal hippocampal slices from 5,7-

DHT-treated rats showed a virtual disappearance of SGA

without any reduction in the diffuse distribution of silver

grains (Figure 4). The final advantage of in vitro labeling is

that all regions of the brain can be incubated under identical

conditions to ensure uniform exposure to 3H-5HT. This

allowed radioautographic detection of 5-HT axons and termi-

nals in all areas of the hippocampus. The sensitivity of this

technique can be appreciated by noting the description of

fibers traveling through the granular layer of the dentate gyrus

in route to terminate in the polymorphic layer. This 5-HT

connection has not been previously described by workers

usi ng rad ioautographic anterograde traci ng, histochemical

fluorescence, or silver degeneration methods. Furthermore,

preliminary studies in collaboration showed good visualization

of 5-HT processes in other rodent and primate brain regions

(olfactory bulb, caudate, hypothalmus, and brain stem) and in

peripheral tissue (pancreas, pituitary gland, and salivary gland)

(unpublished observations).

Biochemical Pharmacological Studies

The 5-HT uptake characteristics in the hippocampus are

consistent with those described in the hypothalamus and

striatum. In these areas the K,, for the specific uptake was

1.7 x 10� M in the striatum and 1.4 x 10� M in the

hypothalamus (38). Hippocampal uptake of 3H-5HT has an

uptake Km of 1.4 x 10-i M. This uptake showed the best

specific to nonspecific ratio at 5 x 10_8 M. The specificity of

this uptake in the hippocampus was determined by using 1)

excess of unlabeled 5-HT, 2) the uptake inhibitor fluoxetine,

3) the 5-HT precursor tryptophan, and 4) intracerebral injec-

tion of 5,7-DHT, a 5-HT specific neurotoxic drug. All four

procedures produced results consistent with the uptake pro-

cess being specific for 5-HT. These results agree with the

findings of several other laboratories ( 19, 37-39). Further-

more these in vitro studies of hippocampal slices with

5 x 10� M 3H-5HT indicate that 66-73% of the 3H-5HT

binding is due to the specific reuptake process of serotonergic

fibers.

The uptake of 3H-5HT by noradrenergic fibers has been

proposed by a number of investigators to explain low affinity

uptake of 5-HT seen in the brain (see ref. 38). The molarity of

3H-5HT (5 x 10-8 M) used in the majority of our experi-

ments, should avoid any significant uptake by the low affinity

sites which have a Km of 8 x 10_6 M (38). Nevertheless

the amount of interspecific competition between 5-HT and

NA was studied for both the specific uptake and the non-

specific binding. The addition of noradrenaline (NA) at

20-200 times the 3H-5HT concentration blocked 13-20% of

the total binding. Likewise, desipramine, a specific NA uptake

blocker at low concentrations, blocked a similar amount of the

total 3H-5HT binding (9-24%). NA and DM1 reduce total

tissue retention of 3H-5HT, probably by affecting the NSB

rather than the specific uptake mechanism. This was deduced

from the 9% reduction in the NSB by concomitant incubation

ofdesipramine and a large excess ofunlabeled 5-HT (Table 2).

This nonspecific nonsaturable binding of 3H-5HT probably

involves a non-monoaminergic compartment, rather than up-

take into NA fibers, since as pointed out by Shaskan and

Snyder (38) the interspecific uptake into noradrenergic fibers

should be saturated in the same range as the specific uptake

into serotonergic fibers. In support of this hypothesis, recent

in vitro radioautographic experiments using molarities of

3H-5HT which favor the low affinity uptake sites ( 10�- 10’M) do show some accumulation oi tracer in intrinsic neuronal

elements of the hippocampus (Azmitia and Descarries, in

preparation).

The measure of the nonspecific uptake of 3H-5HT was

assessed by incubating the slices in a large excess of unlabeled

5-HT. This uptake was nonsaturable at the concentration

range tested and was linearly related to the molarity of

3H-5HT in the medium. It had been suggested that incubating

the slices at 0#{176}Cinstead of 37#{176}Cwould provide an index of

nonspecific uptake (28). However, in hippocampal slices the

NSB of 3H-5HT at 0#{176}Cwas partially blocked by excess 5-HT

(Table 2). Other workers have also questioned the validity of

using 0#{176}Ccontrol, and have employed neurotoxic drugs to

estimate the extraneuronal uptake (36). Radioautographic

examination of slices incubated at 0#{176}C(unpublished data)

show that the diffusion into the tissue of the 3H-5HT is

markedly reduced. Therefore, this control reduces all types of

3H-5HT binding, and is not selective for the specific uptake

mechanism of serotonergic neurons. In our studies, approxi-

mately 30% of total binding of 3H-5HT was retained after

both in vivo neurotoxic depletion and in vitro incubation with

an excess of unlabeled 5-HT. The NSB value of 30% of total

binding is applicable at low 3H-5HT concentration (5 X

108 M) and will increase proportionally as the 3H-5HT con-

centration is raised (Table 1, Figure 1).

Morphology

The light microscopic visualization of the 5-HT concentrating

fibers in the hippocampus is an improvement in many respects

when compared to other histochemical techniques. The to-

pographic distribution after 3H-5HT uptake radioautography,

in addition to being more directly specific than 3H proline

injection into the median raphe nucleus (4, 5, 1 1), does

represent a significant increase in resolution. Individual fibers

by guest on July 14, 2011jhc.sagepub.comDownloaded from

642

CONTROL

AZMITIA, MAROVITZ

DG

CA4

CA1

Sub

p� #{149} � I,

‘. �#{176}::i:�%’� � :� � :::2.:.:. .S � �? ‘ ‘�‘. .-

3�-:�.’’.. ,.- . . .. . .‘ ,.

� . . .. � :�

�

‘�‘� 4. . . � ‘ �: � :.�

� ..�

.:‘:: �

.. � � , :� �

�

:�T � .‘ �

. . ..‘ ‘. �:‘ :� � “ ,

“.1 �, I:�it.,., ,1�mt �S j .�l ..4,z’ i ‘- �

*t� “iij’ .i.,. p t

;::: � f�’�*�:�’

. . L;-::, � ‘s;’, � ‘� ,6) ‘ �, . , . . . �. #{149}.� ., , � . ;,, . . . . ‘/‘ . .. -;:

� � -% �4I �!

) r’ ���1#{216}” � _,

by guest on July 14, 2011jhc.sagepub.comDownloaded from

HIPPOCAMPAL UPTAKE OF SEROTONIN 643

Figure 4. Hippocampal 5-HT terminals in normal and 5,7-DHT-

injected rats. Radioautography performed on hippocampal slices

incubated for 15 mm at 37#{176}Cwith 5 x 10-8 M 3H-5HT. Thephotomicrographs on the left side (control) show the SGA (black

densities) in the dentate gyrus (DG), cornu Ammonis (CA), and in the

subicular cortex (Sub). In the dentate gyros, two rows of varicosities

are seen in the granular layer. In the CA4 region, numerous large

boutons are visible. In the CA4 region, certain pyramidal neurons

appear to be surrounded by SGA. Finally, in the subicular cortex,varicosities are seen near the cortical cell bodies. On the right side of

the figure, the same regions are shown in the dorsal hippocampusfrom a rat injected 6 days previously with 5,7-DHT into the fornix-

fimbria and the cingulum bundle. The background level of individual

silver grains is similar to the control tissue, but few SGA are evident.Photographs taken on a Leitz Orthoplan, Zeiss 63X oil objective(NA= 1.4), Leitz oil condensor (NA 1.25) using 35 mm film (Pan X,

32 ASA). No filter, bright field illumination.

and their boutons were easily traced in all regions of the

hippocampal complex. Labeled boutons were seen in close

apposition to pyramidal cells as they passed from the stratum

oriens to the stratum radiatum and the stratum lacunosum in

Ammon’s horn. A heavy concentration of SGA surrounding

some of the perikarya of the pyramidal and granule cells was

evident. Individual fibers with an orderly array of boutons

could be easily identified coursing through the granular layer.

SGA were also apparent in the subicular cortex. Furthermore,

individual fibers could be followed from this cortical area to

the polymorphic layer of the dentate gyrus, a connection

originally described by Ramon y Cajal (Figure 493, p. 770 of

ref. 34).

The counting of boutons was done on 5 j.t paraffin-

embedded tissue after a 3 month exposure period. It has been

reported that only 1-2 j.a of the section will contribute to the

density of radioactivity seen (35, p. 237). Therefore, the

density of SGA reported may be an overestimate by as much

as 50% when compared to studies using 1 �.a sections. Previ-

ously, Leger and Descarries (25) reported that the 5HT

terminals in the locus coeruleus had a density of 186

boutons/10,000 �.am2 using 1 �.a thick Epon-embedded tissue

sections. In the fronto-parietal neocortex the distribution of

5-HT varicosities varied within the upper five cortical layers

examined and the highest density, in the first layer, was 29

boutons/10,000 j.amt (8). Furthermore, the density of 5-HT

boutons in the lumbar periependymal regions after in vitro

incubation of spinal cord with 3H-5HT was 66 boutons/

10,000 j.tm2 (37). In the present study, the density of 5-HT

boutons was the highest in Ammon’s horn ( 1 19 boutons/

10,000 �am2), it was 67 boutons/10,000 j.am2 and 79 boutons/

10,000 .am2 in the dentate gyrus and subicular cortex, respec-

tively. Neighboring areas within the same regions of the

hippocampus had much lower densities of SGA. These results

suggest that the innervation level in the hippocampus is

substantially lower than that of the locus ceruleus, but higher

than the fronto-parietal neocortex. This regional distribution

pattern is in agreement with the 5-HT biosynthetic capacity of

different brain regions. Measurement of tryptophan by-

droxylase in the rat brain gave 0.70 mg of 5-HTP/mg protein/hr

in the locus ceruleus, 0.37 in the hippocampus, 0.22 on the

dentate gyrus, and 0.20 in the parietal cortex (12).

Ultrastructural studies are presently under way to inves-

tigating the 5-HT axons and terminals within the hippocam-

pus.

Acknowledgments

The authors wish to thank Tibor Barka for his helpful advice in planning

certain experiments and Laurent Descarries for his careful and detailed

comments during the preparation of this manuscript. They also uish to

thank Ms. Vernell G. Daniels for her technical assistance, Mr. Patrick].

Gannonfor his photographic andartistic work, and Ms. Grace Friscia for

her typing ofihe many drafts ofthis manuscript.

Literature Cited

1. Aghajanian GK, Bloom FE, Lovell RA, Sheard MJ, Freedman

DX: The uptake of 5-hydroxytryptamine 3-H from cerebralventricles: autoradiographic localization. Biochem Pharmacol

15:1401, 1966

2. Aghajanian GK, Bloom FE: Localization of tritiated serotonin inrat brain by electron-microscopic autoradiography. J PharmacolExpTher 156:23, 1967

3. Azmitia EC, Henriksen SJ: A modification of the Falck-Hillarp

technique for 5-HT fluorescence employing hypertonic formal-

dehyde perfusion. J Histochem Cytochem 24: 1286, 1976

4. Azmitia EC: The serotonin-producing neurons of the midbrain

median and dorsal raphe nuclei. In Handbook of Psychophar-

macology, vol. 9. Edited by LL Iversen, SD Iversen, SH Snyder.Plenum Press, New York, ch 5, pp. 233-314

5. Azmitia EC, Segal M: An autoradiographic analysis of the differ-ential ascending projections of the dorsal and median raphe

nuclei intherat.JComp Neurol 179:641, 1978

6. Azmitia EC, Buchan AM, Williams JH: Structural and functionalrestoration by collateral sprouting of hippocampal 5-HT axons.

Nature274:374, 1978

7. Baumgarten HG, Lachenmayer L: 5,7-Dihydroxytryptamine: im-

provement in chemical lesioning of indoleamine neurons in themammalian brain. Zellforsch Z 135:399, 1972

8. Beaudet A, Descarries L: Quantitative data on serotonin nerve

terminals in adult rat neocortex. Brain Res 1 1 1 :301, 1976

9. Bloom FE, HotTer BJ, Siggins GR, BarkerJL, Nicoll RA: Effectsof serotonin on central neurons: microiontophoretic administra-

tion. Fed Proc 31:97, 1972

10. Bobillier PF, Petejean D, Slavert M, Ligier, Seguin 5: Differential

projections of the nucleus raphe dorsalis and nucleus raphe

centralis as revealed by autoradiography. Brain Res 85:205, 1975

I 1. Bobillier PF, Seguin 5, Degueurce A, Lewis BD, Pujol JF: The

efferent connections ofthe nucleus raphe centralis superior in the

rat as revealed by radioautography. Brain Res 166:1, 1979

12. Brownstein MJ, Palkovits M, SaavedraJM, KizerJS: Tryptophan

hydroxylase in the rat brain. Brain Res 97: 163, 1975

13. Calas A, Alonso G, Arnauld E, Vincent JD: Demonstration ofindolaminergic fibers in the median eminence of the duck, rat,

and monkey. Nature (Lond) 250:241, 1974

14. Calas A, Besson MJ, Gaughy C, Alonso G, Glowinski J, CheramyA: Radioautographic study of in vivo incorporation of 3H-monoamines in the cat caudate nucleus: identification of

serotoninergic fibers Brain Res 1 18: 1, 1976

1 5. Chan-Palay V: Serotonin axons in the supra- and subependymal

plexuses and in the leptomeninges; their roles in local alterations

ofcerebrospinal fluid and vasomotor activity. Brain Res 102: 103,

1976

by guest on July 14, 2011jhc.sagepub.comDownloaded from

644 AZMITIA, MAROVITZ

16. Chan-Palay V: Fine structure of labelled axons in the cerebeller

cortex and nuclei of rodents and primates after intraventricular

infusions with tritiated serotonin. Anat Embryol 148:235, 1975

17. Conrad LCA, Leonard CM, PfaffDW: Connections ofthe median

and dorsal raphe nuclei in the rat; an autoradiographic and

degeneration study. J Comp Neurol 156: 179, 1974

18. Descarries L, Beaudet A, Watkins KC: Serotonin nerve terminals

in adult rat neocortex. Brain Res 100:563, 1975

19. Fuller RW, Perry KW, Molloy BB: Effect of an uptake inhibitoron serotonin metabolism in rat brain: studies with 3-(p-trifluoro-methylphenoxy)N-methyl-3-phenylpropylamine (Lilly 1 10140)LifeSci 15:1161, 1974

20. Fuxe K: Distribution ofmonoamine nerve terminals in the central

nervous system. Acta Physiol Scand 64 (Suppl 247): 37, 1965

2 1. Hokfelt T, Ljungdahl A: Light and electron microscopic au-

toradiography on spinal cord slices after incubation with labelled

glycine. Brain Res 32: 189, 1971

22. Iversen LL, Bloom FE: Studies on the uptake of 3H-GABA and

(3H) glycine in slices and homogenates of rat brain and spinalcord by electron microscopic autoradiography. Brain Res 4 1 : 1 3 1,

1972

23. Iversen LL, Storm-Mathisen J: Uptake of (3H) glutamate inexcitatory nerve endings in the hippocampal formation of the rat.

Acta Physiol Scand 96:22A, 1976

24. Kuhar MJ, Aghajanian GK: Selective accumulation of 3H-serotonin by nerve terminals of raphe neurons: an autoradio-

graphic study. Nature 24 1: 187, 1973

25. Leger L, Descarries L: Serotonin nerve terminals in the locus

coeruleus of adult rat: a radioautographic study Brain Res 145:1,

1978

26. L#{233}ger L, Mouren-Mathieu AM, Descarries L: Identification

radioautographique de neurones monoaminergiques centraux par

micro-instillation locale de s#{233}rotonine ou de noradrenaline triti#{233}e

chez le chat. CR Acad Sci (D) (Paris) 286(2 1 ): 1 523, 1978

27. Lenn NJ: Localization of uptake of tritiated norepinephrine by ratbrain in vivo and in vitro using electron microscopic autoradio-

graphy. AmJ Anat 120:377, 1967

28. Lidbrink P. Jonsson G: Noradrenaline nerve terminals in the

cerebral cortex: effects on noradrenaline uptake and storage

following axonal lesion with 6-hydroxydopamine. J Neurochem22:617, 1974

29. Lidbrink P, Jonsson G, Fuxe K: The effect of imipramine-like

drugs and antihistamine drugs on uptake mechanisms in the

central noradrenaline and 5-hydroxytryptamine neurons.

Neuropharmacol 10:52 1, 1971

30. Lorente de No R: Studies on the structure of the cerebral cortex

II. Continuation of the study of the ammonic system. J PsycholNeurol 46: 1 13, 1934

31. Matus Al, Dennison ME: Autoradiographic localization of

tritiated glycine at ‘flat-vesicle’ synapses in spinal cord. Brain Res

32:195, 1971

32. Moore RY, Halaris AE: Hippocampal innervation by serotonin

neurons of the midbrain raphe in the rat. J Comp Neurol164:171, 1975

33. Penreath VW, Cottrell GA: Selective uptake of 5-hydroxytrypta-

mine by a.xonal processes in Helix pomatia. Nature New Biol

239:213, 1972

34. Ramon y Cajal 5: Histologie du Systeme Nervuex de L’Homme

et des Vertebres II. Translated into French by L Azoulay. A

Malone,Paris, 1911

35. Rogers AW: Techniques of Autoradiography. Elsevier North

Holland, New York, 1967

36. Sachs CH: Noradrenaline uptake mechanisms in the mouse

atrium. A biochemical and histochemical study. Acta Physiol

Scand (Suppl 341):1, 1969

37. Segu L, Calas A: The topographic distribution of serotoninergic

terminals in the spinal cord of the cat: quantitative radioauto-graphic studies. Brain Res 153:449, 1978

38. Shaskan EG, Snyder SH: Kinetics of serotonin accumulation into

slices from rat brain: relationship to catecholamine uptake. JPharmacolExpTher 175:404, 1970

39, Wong DT, Horng JS, Bymaster FP, Hauser KL, Molloy BB: A

selective inhibitor of serotonin uptake: Lilly 1 10140, 3-(p-trifluoro-methylphenoxy)-N-methyl-3-phenylpropylamine. Life Sci 15:47 1,

1974

40. Wood JG: Electron microscopic localization of amines in central

nervous tissue. Nature 209: 1 1 3 1 , 1966

41. Wood JG: Electron microscopic localization of 5-hydroxytrypta-mine (5-HT). Tex Rep Biol Med 23:828, 1965

by guest on July 14, 2011jhc.sagepub.comDownloaded from