Aberrant forms of α2-macroglobulin purified from patients with multiple sclerosis

Clinica Chimica Acta 295 (2000) 27–40www.elsevier.com/ locate /clinchim

Aberrant forms of a -macroglobulin purified from2

patients with multiple sclerosis

a a b ,*Martin Gunnarsson , Torgny Stigbrand , Poul Erik H. Jensena

˚ ˚Department of Immunology, Umea University, S-901 85 Umea, SwedenbThe Neuroimmunology Laboratory, The Centre for Neuroscience, Section 9202,

Copenhagen University Hospital, Rigshospitalet, Juliane Mariesvej 24, 2100 Copenhagen, Denmark

Received 13 September 1999; received in revised form 29 November 1999; accepted 14 December 1999

Abstract

The biochemical properties of a -macroglobulin were investigated in four patients with multiple2

sclerosis and compared to a -macroglobulin from healthy controls. An impaired stability of2

a -macroglobulin from the multiple sclerosis patients was demonstrated as a spontaneous2

conversion to an electrophoretic‘‘fast’’ form of a -macroglobulin upon purification and storage,2

with a concomitant decrease in functional capacity to inhibit proteinases. The ability to formcomplexes with proteinases was significantly reduced in a -macroglobulin purified from the2

multiple sclerosis patients. The aberrant molecular arrangements of the protein were not due toproteinase cleavages in the bait regions of a -macroglobulin, as demonstrated by gel electro-2

phoresis and protein sequencing. The number of functional thiol esters, however, was reduced ina -macroglobulin purified from the multiple sclerosis patients, an observation compatible with the2

impaired proteinase binding property. Furthermore, differences in isoelectric points were observedbetween a -macroglobulin from the multiple sclerosis patients and a -macroglobulin from healthy2 2

controls. The results suggest that aberrant forms of a -macroglobulin may be present in patients2

with multiple sclerosis. 2000 Elsevier Science B.V. All rights reserved.

Keywords: Multiple sclerosis; a -macroglobulin; Proteinase inhibitor; Thiol ester; Conformation2

*Corresponding author. Tel.: 1 45-35-456-701; fax: 1 45-35-456-713.E-mail address: [email protected] (P.E.H. Jensen)

0009-8981/00/$ – see front matter 2000 Elsevier Science B.V. All rights reserved.PI I : S0009-8981( 00 )00190-X

28 M. Gunnarsson et al. / Clinica Chimica Acta 295 (2000) 27 –40

1. Introduction

Multiple sclerosis (MS) is an inflammatory demyelinating disease of thecentral nervous system (CNS). Increasing evidence indicates that MS is ofautoimmune origin and both cell mediated and humoral immune mechanisms areactivated [1,2]. The mechanisms causing the primary demyelination are stillunknown, but different components of the myelin are postulated to be degradedand to act as target antigens. Both T- and B cell responses against such myelinantigens have been identified [3–5].

The exposure of antigenic target structures in the myelin can be due todifferent mechanisms. Several proteinases have been identified in elevated levelsin MS plaques, MS myelin and cerebrospinal fluid (CSF) from MS patients[6–8]. Appearance of such proteolytic enzymes may cause myelin degradationand release of immunogenic peptides. One mechanism by which increasedproteolytic activity within the CNS may be generated is by malfunctioningproteinase inhibition mechanisms. One of the most abundant proteinase in-hibitors in the CNS is a -macroglobulin (a M), which has a broad specificity2 2

and is capable of inhibiting proteinases from all of the four major classes [9,10].Moreover, it has been demonstrated that a M, after forming complexes with2

proteinases can activate and take part in antigen presentation by macrophages[11]. a M is furthermore a cytokine-binding protein (binds e.g. IL-1b, IL-6,2

TGF-b and TNF-a) [12,13]. The demonstration of changes in cytokine con-centrations in MS patients may alternatively be due to alterations in binding ofcytokines to a M. Some reports indicate that patients with MS have a M with2 2

altered characteristics, displayed as modified isoelectric properties, aberrations incharge, immunoprecipitation behaviour and carbohydrate content [14–17]. Noconclusive structural or functional investigations, however, of the a M from MS2

patients, have been made.Human a M is a glycoprotein composed of four identical 180 kDa subunits2

[18,19]. It is widely distributed in the extracellular fluid compartments of thebody [10]. The‘‘bait region’’ of the a M subunit denotes a region especially2

sensitive to proteolytic cleavage [9,20]. Proteinase cleavages within this regionexpose thiol ester bonds in a M, which then can react with the proteinase or2

other nucleophiles present [21,22]. The activation of a M induces conformation-2

al changes in the inhibitor and causes ‘‘trapping’’ of proteinases [9,23]. Primaryamines such as methylamine can hydrolyse the thiol ester bonds and inducesimilar conformational changes in absence of proteinases [24,25]. By changingconformation, a M exposes certain new sites, recognised by the LRP/a M-2 2

receptor [26], which results in rapid clearance of a M-proteinase complexes2

from the circulation [27,28]. The proteinase- and the methylamine treated a M2

are referred to as ‘‘fast’’ forms, because of their different electrophoreticmobility compared to the native ‘‘slow’’ form of a M [29].2

M. Gunnarsson et al. / Clinica Chimica Acta 295 (2000) 27 –40 29

The aim of this study was to investigate biochemically the putative structuralmodifications of a M from MS patients. For this purpose we used plasma2

collected from three MS patients and one family with individuals who developedMS.

2. Materials and methods

2.1. Materials

All buffer substances were of highest purity. Sodium dodecyl sulphate (SDS)was from Bio-Rad Laboratories (CA, USA). Methylamine, phenylmethylsul-phonyl fluoride, 2-mercaptoethanol, 5.59-dithiobis-(2-nitrobenzoic acid)(DTNB), bovine a-chymotrypsin (type VII), bovine trypsin (type XIII) andbovine serum albumin were from Sigma Chemicals Co. (MO, USA). Theconcentrations of trypsin and chymotrypsin were calculated using values of 15.4

1%and 20.5, respectively, for E [30,31] and 23.3 and 25.1 kDa, respectively, for1 cm

the molecular masses [32,33]. Chloramine T was from Merck (Germany).125Na[ I] and Hyperfilm MP were from Amersham (UK).

2.2. Patients

We had access to blood samples from the mother (51 years old), the son (20years old) and the daughter (26 years old) of a family, in which the son, at theage of 19, developed a primary progressive MS with rapid progress. The fatherof the son and daughter died from MS and no blood samples from him wereavailable. Furthermore, blood samples from two patients with relapsing-remit-ting MS (females, 27 and 39 years old) and one patient with secondaryprogressive MS (female, 41 years old) were obtained following informedconsent. All patients were examined at the Department of Neurology, Norrlands

˚Universitetssjukhus, Umea, Sweden, and fulfilled the Poser criteria of eitherclinical definite or laboratory supported definite MS [34]. Healthy blood donorsserved as non-MS controls.

2.3. Preparation of plasma

Peripheral bloods from MS patients and healthy controls were collected inEDTA tubes and were centrifuged at 1000 3 g for 15 min followed by collectionof the plasma layer.


2.4. Purification of a M and preparation of a M-methylamine2 2

a M was purified from plasma according to the method of Imber and Pizzo2

[27]. a M concentrations were determined spectrophotometrically, calculated21%with an E value of 8.93 and a molecular mass of 718 kDa [35]. The purified1 cm

21a M was stored at concentrations ranging from 0.3 to 3.5 mg ml in 0.12

21mol l sodium phosphate buffer, pH 8.0, at 2 238C. Purified a M from healthy2

individuals is stable for more than one year at these conditions, as judged bynon-denaturing polyacrylamide gel electrophoresis (PAGE) (demonstrating‘‘slow’’ mobility), thiol ester titration (demonstrating intact thiol esters) andincubation with proteinases (demonstrating preserved capacity of inhibitingproteinases) (not shown).

Methylamine treatment of a M was performed by incubating a M at a2 221 21concentration of 1.0 mg ml in 0.1 mol l sodium phosphate buffer, pH 8.0,

21with 0.4 mol l methylamine (final concentration) for at least 1 h at roomtemperature.

2.5. Radiolabelling

125Na[ I]-labelling of proteins was performed using the chloramine T methodaccording to Hunter and Greenwood [36]. For the removal of free iodine andreagents, a Sephadex G-50 column (0.8 3 24 cm) (Pharmacia, Sweden) was

21used. The specific activities (approximately 600 cpm ng ) were determinedusing a Unigamma g-counter (LKB, Sweden).

2.6. Incubation of chymotrypsin and trypsin with a M2

21In general, chymotrypsin and trypsin were incubated with a M (1.0 mg ml )221at a molar ratio of 2:1 in 0.1 mol l sodium phosphate buffer, pH 8.0, for 5 min

at room temperature, followed by inhibition of the active site of the proteinases21with 1 mmol l phenylmethylsulfonyl fluoride (final concentration) for at least

5 min.

2.7. Gel electrophoresis, autoradiography and isoelectric focusing

Non-denaturing PAGE and SDS–PAGE were carried out in mini-PROTEANII electrophoresis cells (Bio-Rad, USA) according to Laemmli [37], on 5% and

21 217.5% gels, respectively, in 25 mmol l Tris, 190 mmol l glycine, pH 8.6.Denaturation and reduction of a M were performed by incubation with 1% SDS2

and 2.5% 2-mercaptoethanol for 1 h at room temperature. Following electro-phoresis, gels were stained with Coomassie Brilliant Blue R.


Autoradiography was performed by exposing gels to Hyperfilm MP at2 708C. For laser scanning of gels and autoradiograms, a 2202 Ultra-Scan laserdensitometer (LKB, Sweden) was used.

Isoelectric focusing was performed using Phastsystem (Pharmacia, Sweden)following the instructions of the manufacturer. The samples were applied onPhastGel IEF with pH-gradient 3–9 and the broad pI-kit (Pharmacia, Sweden)was used as pH markers.

2.8. N-terminal sequence analysis

Sequence analysis was made on an Applied Biosystems 477A Pulsed LiquidPhase sequencer with an online PTH 120A Analyzer (Foster City, USA)according to Matsudaira [38]. Sequencing was performed with cycle programsadapted to the reaction cartridges and chemicals from the manufacturer. Theprotein was adsorbed on Immobilon (Millipore, USA) and stained withCoomassie Brilliant Blue R.

2.9. Thiol ester titration

The concentration of exposed sulfhydryl groups in a M was determined2

spectrophotometrically by titration with 5.59-dithiobis-(2-nitrobenzoic acid)(DTNB) according to Sottrup-Jensen et al. [21].

3. Results

3.1. Conformational states of a M in plasma and after purification2

A non-denaturing PAGE of plasma samples from the three individuals of thefamily investigated demonstrated no difference in either total amount of a M in2

the plasma samples or mobility of a M, which was of the ‘‘slow’’ mobility form2

characteristic for native a M (not shown). Following simultaneous and identical2

individual purification, however, the preparations of a M from the family2

behave differently as shown in Fig. 1. Purified a M can normally be stored for2

several years at 2 238C without any significant conversion from ‘‘slow’’ to‘‘fast’’ form (lane 1). a M purified from the mother (lane 4) represents2

the‘‘slow’’ type band of a M corresponding to normal native a M. The band of2 2

a M purified from the daughter also demonstrates the ‘‘slow’’ type, but contains2

some a M converted to‘‘fast’’ form (lane 3). In contrast a M purified from the2 2

son displays a different pattern, documented as a weak ‘‘slow’’ band andappearance of a ‘‘fast’’ form (lane 2). Upon storage for only two months, thea M from four MS patients was converted to a ‘‘ fast’’ form (lanes 5–8). These2


Fig. 1. Conformational states of a M purified from the family studied, three additional MS221patients, and a healthy control. The concentration of a M ranged from 0.30 to 0.64 mg ml .2

Following non-denaturing PAGE, the 5% gel was stained with Coomassie Brilliant Blue R. Lane 1is a M from a healthy control stored up to two years at 2 238C, lanes 2–4 are a M, directly after2 2

purification, from the family investigated; the MS son (lane 2), the daughter (lane 3) and themother (lane 4), and lanes 5–8 are a M, stored for two months at 2 238C, from four MS2

patients. Lane 8 is from the MS son.

results indicate that a M from these MS patients have properties rendering the2

proteins less stable upon purification and storage, and may reflect a biochemicalheterogeneity in the population of molecules.

3.2. Isoelectric focusing of a M purified from MS patients2

The a M purified from seven individuals (four MS patients and three healthy2

controls) were subjected to isoelectric focusing as shown in Fig. 2. Differences

Fig. 2. Isoelectric focusing of a M purified from MS patients and healthy controls. Samples were2

applied on PhastGel IEF with pH-gradient 3–9. Std. is standard; lanes 1–4 are a M from four2

different MS patients (lane 4 is a sample from the MS son of the family investigated); lanes 5–7are a M from three different controls (the a M in lane 6 is from an impure preparation).2 2


in isoelectric point were confirmed and all the samples from the MS patients(lanes 1–4) displayed two well separated diffuse bands with pI:s around 4.7 and5.8, while a M from the controls (lanes 5–7) presented one major band with an2

pI around 4.6, i.e. a M from the MS patients presented a higher pI. In both these2

regions, the samples from the MS patients demonstrated a distinct microhetero-geneity, comprising 6–10 closely positioned bands, which are typical forvariability in glycosylation of plasma proteins. These results indicate a charge-heterogeneity in the population of a M molecules from these MS patients.2

3.3. N-terminal sequence analysis of purified MS a M2

For the elucidation of the mechanisms, which cause the conformationalchanges of a M in the MS patients, it is crucial to know if cleavages of the bait2

regions have occurred, which would transform the molecule to ‘‘fast’’ form andfurthermore display new protein sequences downstream from the cleavage sites.Sequencing of a M purified from the MS son unequivocally demonstrated the2

beginning of the N-terminal sequence, Ser–Val–Ser–Gly–Lys, with no traces ofbait region derived or other protein sequences. This indicates that the baitregions of this a M are intact and that the instability is not due to cleavage by2

proteinases. This was furthermore supported by the demonstration of 180 kDasubunits of MS a M in SDS–PAGE (not shown).2

3.4. Thiol ester titration of purified MS a M2

In order to examine if the thiol esters of purified MS a M still were intact,2

titration of thiol groups with methylamine was performed. Following storage forone year at 2 708C (which normally permits stable storage), a M purified from2

21the MS son presented only 0.5 mol titratable thiol groups mol a M, while the2

corresponding value for healthy controls was reproducible approximately four.Titrations of a M from the son directly after several separate purifications2

21displayed values between 1 and 3 mol thiol groups mol a M. These results2

demonstrate that the thiol esters are partially hydrolysed or modified in thispopulation of a M molecules.2

3.5. Proteinase trapping capacity of purified MS a M2

As shown in the autoradiogram in Fig. 3, a M purified from the four MS2125patients and stored for two months at 2 238C is unable to trap Na[ I]-labelled

chymotrypsin (lanes 7–10), whereas a M from healthy controls and newly2

purified a M from the son are able to do so. Following scanning of the2

autoradiogram, the total relative trapping capacity of the newly purified a M2


125Fig. 3. Trapping of Na[ I]-labelled chymotrypsin by a M purified from MS patients and healthy2

controls. Autoradiogram of a 5% gel after non-denaturing PAGE, following incubation of a M2125with Na[ I]-labelled chymotrypsin. Lanes 1–3 are a M, stored at 2 238C for 1 to 2 years, from2

three different controls; lane 4–6 are a M from the MS son (lane 4), the daughter (lane 5) and2

the mother (lane 6) of the family investigated, directly after purification. Lanes 7–10 are a M,2

stored at 2 238C for two months, from four different MS patients (lane 10 is a sample from theMS son).

from the MS son was, however, only 38% compared to 100% of the controlsubjects. The daughter and the mother investigated displayed trapping values of54% and 100%, respectively.

3.6. Proteinase binding capacity of purified MS a M2

125Investigation of the covalent binding of Na[ I]-labelled chymotrypsin toa M purified from four MS patients and a healthy control (Fig. 4), following2

storage of purified a M at 2 238C for two months, demonstrated significantly2125reduced formation of covalent complexes of Na[ I]-labelled chymotrypsin and

125a M from the MS patients (lanes 1–4) than of Na[ I]-labelled chymotrypsin2

and normal a M (lane 5), using SDS–PAGE and autoradiography. By scanning2125of autoradiograms, only 30% relative binding of both Na[ I]-labelled chymo-

125trypsin and Na[ I]-labelled trypsin to newly purified a M from the MS son2

was determined, while a M from the daughter and the mother investigated2

bound around 60% and 100%, respectively, of the proteinases, compared to the100% of other controls.

3.7. Bait region cleavages in purified MS a M following treatment with2

proteinases

The appearances of cleavage product bands of a M (85–95 kDa) in SDS gels,2

following denaturation and reduction of the proteins, may be taken as proofs ofbait region cleavages of a M by proteinase. With similar treatment of samples,2

but without proteinase activity, all the purified a M investigated presented 1802

kDa subunits in SDS–PAGE (not shown). As seen in Table 1, 80–100% of the


125Fig. 4. Covalent binding of Na[ I]-labelled chymotrypsin to a M purified from MS patients and2

a healthy control. Autoradiogram of a 7.5% gel after SDS–PAGE, following incubation of a M2125with Na[ I]-labelled chymotrypsin, denaturation with SDS and reduction with 2-mercap-

toethanol. Lanes 1–4 are a M, stored at 2 238C for two months, from four MS patients (lane 4 is2

a sample from the MS son of the family investigated) and lane 5 is a M from a control.2aHMW5 high molecular weight.

Table 1aFractions of cleaved bait region (%)

Proteinase Chymotrypsin Trypsin

Control a M (n 5 3): 79 (65) 97 (64)2

MS son a M: 54 552

Non MS daughter a M: 60 682

Non MS mother a M: 70 882

a Calculated as fraction (in percent) N-terminal bait region fragment band of total bands(N-terminal bait region fragment and subunit band) after scanning of SDS gels stained withCoomassie Brilliant Blue R. The samples were denatured with SDS and reduced with mercap-toethanol prior to SDS–PAGE. Control is presented as mean value of experiments with a M2

preparations from three healthy controls.


bait regions in a M from healthy controls were cleaved following treatment with2

chymotrypsin or trypsin, whereas a M purified from the MS son only yielded2

values of 54–55%. The daughter and the mother investigated displayed values of60–68% and 70–88%, in three consecutive experiments respectively. Theseresults indicate that the structural arrangements of the a M from the MS son2

hamper an easy hydrolysis within the bait regions.

4. Discussion

Earlier investigations of a M at MS have indicated modified conformational2

or structural properties of this protein [14–17]. These results have, however, alsobeen questioned [39]. Until now, no investigations have been initiated toelucidate the nature of altered a M properties from MS patients. The major2

finding here is that a M in the MS patients presents several structural2

aberrations indicating heterogeneity within the population of a M molecules.2

This heterogeneity was documented as a faster conversion to conformationalforms resembling the classical ‘‘fast’’ forms of a M. Furthermore, the analysis2

of the charge distribution indicate that the population of a M from the MS2

patients demonstrate a heterogeneity in charge.The assembly of a M comprises two dimers which are held together by2

non-covalent bonds referred to as the contact zone [40]. The proteinaseinhibition mechanisms exerted by human a-macroglobulins are complex, asshown by Jensen et al. [23,41]. The cleavage of the thiol esters is enough togenerate conformational changes in a M. It is thus sufficient only to interact2

with the thiol ester in a M for causing conformational changes. In the present2

report, it is demonstrated that the level of reactive thiol esters in fact issignificantly reduced in the conformationally changed MS a M. The thiol ester2

may thus partially be consumed by some factor or mechanism, rendering theprotein unstable. Such a property would result in an impaired capacity to bindproteinases and generate a decreased formation of a M-proteinase complexes.2

In addition to proteinases, several other components have been shown to bindto a M, such as cytokines and growth factors [12], zinc [42,43], lectins [44],2

b-amyloid peptide [45], histone [46] and myelin basic protein (MBP) [47,48].Interestingly, it has been suggested that addition of MBP to a M diminishes its2

trypsin protecting capacity [49]. It is unclear, however, if binding of MBP toa M may influence the stability or cause conformational changes of a M.2 2

It is known that glycosylation patterns affect protein stability and may alsoalter the antigenic properties of some proteins [50,51]. Furthermore, abnormalprotein denaturation and glycosylation have been detected in patients withautoimmune diseases [52,53]. The differences in pI and the distinct a M2

microheterogeneity observed in MS patients in the present investigation and by


others [14,15,17] may indicate differences in glycosylation of a M in these2

patients. Abnormal glycosylation of a M in terms of increased concanavalin A2

reactivity has recently been reported in MS [54], and alterations of the sialic acidcontent of a M from MS patients have been observed [16]. The occurrences of2

minor as well as major differences in pI of the a M from MS patients compared2

to normal a M are of great interest. Further studies to explain these differences2

are in progress.Some publications have described polymorphism of the a M subunits [55,56].2

Recently, genetic associations of a M polymorphism and Alzheimer’s disease2

were reported [57,58]. A genetically derived structural modification in one of thea M alleles should cause a very complex phenotypic expression which may2

affect the exquisite conformational repertoire the molecule undergoes, whileexerting the proteinase inhibition. The consequences of such a heterogeneity insome of the tetrameric complexes should be expected to cause complexfunctional aberrations, due to appearance of different stoichiometry of subunits,differences in covalent binding between these subunits and interfering effects onthe cooperativity between the non-covalently linked subunits. Further studies ofthe structural properties of a M from patients with MS are in progress.2

Acknowledgements

¨Skillful technical assistance was provided by Lisbet Arlestig. We thank M.D.,¨Ph.D. Peter M. Andersen and M.D. Peter Sundstrom at the Department of

˚Neurology, Norrlands Universitetssjukhus, Umea, Sweden, for providing us withblood samples. This work was supported with grants from the Medical Faculty,

˚ ˚Umea University, Umea, Sweden.

References

[1] Steinman L. Multiple sclerosis: A coordinated immunological attack against myelin in thecentral nervous system. Cell 1996;85:299–302.

[2] Link H. In: Russel WC, editor, B cells and autoimmunity, Molecular biology of multiplesclerosis, Chichester: John Wiley and Sons, 1997, pp. 161–90.

[3] Olsson T. Critical influences of the cytokine orchestration on the outcome of myelinantigen-specific T-cell autoimmunity in experimental autoimmune encephalomyelitis andmultiple sclerosis. Immunol Rev 1995;11:245–68.

¨[4] Baig S, Olsson T, Jiang YP, Hojeberg B, Cruz M, Link H. Multiple sclerosis: Cells secretingantibodies against myelin-associated glycoprotein are present in CSF. Scand J Immunol1991;33:73–9.

¨[5] Olsson T, Baig S, Hojeberg B, Link H. Anti myelin basic protein and anti myelin antibodiesin multiple sclerosis. Ann Neurol 1990;27:132–6.


[6] Banik NL. Pathogenesis of myelin breakdown in demyelinating diseases: role of proteolyticenzymes. Crit Rev Neurobiol 1992;16:257–71.

[7] Cossins JA, Clements JM, Ford J et al. Enhanced expression of MMP-7 and MMP-9 indemyelinating multiple sclerosis lesions. Acta Neuropathol 1997;94:590–8.

[8] Cuzner ML, Davison AN, Rudge P. Proteolytic enzyme activity of blood leucocytes andcerebrospinal fluid in multiple sclerosis. Ann Neurol 1978;4:337–44.

[9] Barrett AJ, Starkey PM. The interaction of a -macroglobulin with proteinases: characteristics2

and specificity of the reaction, and a hypothesis concerning its molecular mechanism.Biochem J 1973;133:709–24.

[10] Sottrup-Jensen L. alpha -macroglobulin and related thiol ester plasma proteins. In: Putnam2

FW, editor, 2nd Edition, The plasma proteins,Vol. 5, Orlando, FL: Academic Press, 1987, pp.191–291.

[11] Chu CT, Pizzo SV. Receptor-mediated antigen delivery into macrophages. J Immunol1993;150:48–58.

[12] James K. Interactions between cytokines and a -macroglobulin. Immunol Today2

1990;11:163–6.[13] Wollenberg GK, LaMarre J, Rosendal S, Gonias SL, Hayes MA. Binding of tumor necrosis

factor-a to activated forms of human plasma a -macroglobulin. Am J Pathol 1991;138:265–2

72.[14] Rastogi SC, Clausen J. Abnormalities in native and dissociated a -macroglobulin isolated2

from multiple sclerosis serum. Clin Chim Acta 1980;107:141–4.[15] Rastogi SC, Clausen J, Fog T. Abnormal serum a -macroglobulin in multiple sclerosis. Eur2

Neurol 1981;20:33–9.[16] Rastogi SC, Clausen J. Effect of multiple sclerosis a -macroglobulin on mitogen induction2

of lymphocyte proliferation. Immunobiol 1985;169:339–45.[17] Back SA, Alhadeff JA. Differential isoform profiles of a -macroglobulin from plasma of2

patients with chronic-progressive or relapsing-remitting multiple sclerosis. Clin Chim Acta1992;211:27–36.

[18] Harpel PC. Studies on human plasma a -macroglobulin-enzyme interactions. J Exp Med2

1973;138:508–21.[19] Jensen PEH, Sottrup-Jensen L. Primary structure of human a -macroglobulin. J Biol Chem2

1986;261:15863–9.[20] Sottrup-Jensen L, Sand O, Kristensen L, Fey GH. The a-macroglobulin bait region.

Sequence diversity and localization of cleavage sites for proteinases in five mammaliana-macroglobulins. J Biol Chem 1989;264:15781–9.

[21] Sottrup-Jensen L, Petersen TE, Magnusson S. A thiol-ester in a -macroglobulin cleaved2

during proteinase complex formation. FEBS Lett 1980;121:275–9.[22] Sottrup-Jensen L, Petersen TE, Magnusson S. Trypsin-induced activation of the thiol esters

in a -macroglobulin generates a short-lived intermediate (‘nascent’ a -M) that can react2 2

rapidly to incorporate not only methylamine or putrescine but also proteins lackingproteinase activity. FEBS Lett 1981;128:123–6.

[23] Jensen PEH, Stigbrand T. Differences in the proteinase inhibition mechanism of humana -macroglobulin and pregnancy zone protein. Eur J Biochem 1992;210:1071–7.2

[24] Salvesen GS, Sayers CA, Barrett AJ. Further characterization of the covalent linking reactionof a -macroglobulin. Biochem J 1981;195:453–61.2

[25] Jensen PEH, Stigbrand T, Shanbhag VP. Use of hydrophobic affinity partitioning as a methodfor studying various conformational states of the human a-macroglobulins. J Chromatogr A1994;668:101–6.

[26] Van Leuven F, Cassiman JJ, Van den Berghe H. Demonstration of an a -macroglobulin2

receptor in human fibroblasts, absent in tumor-derived cell lines. J Biol Chem1979;256:5155–60.


[27] Imber MJ, Pizzo SV. Clearance and binding of two electrophoretic ‘fast’ forms of humana -macroglobulin. J Biol Chem 1981;256:8134–9.2

[28] Jensen PEH, Shanbhag VP, Stigbrand T. Methanethiolation of the liberated cysteine residuesof human a -macroglobulin treated with methylamine generates a derivative with similar2

functional characteristics as native a -macroglobulin. Eur J Biochem 1995;227:612–6.2

[29] Barrett AJ, Brown MA, Sayers CA. The electrophoretically ‘slow’ and ‘fast’ forms of thea -macroglobulin molecule. Biochem J 1979;181:401–18.2

[30] Robinson NC, Tye RW, Neurath H, Walsh KA. Isolation of trypsins by affinity chromatog-raphy. Biochemistry 1971;10:2743–7.

[31] Wilcox PE. Chymotrypsinogens-chymotrypsins. Methods Enzymol 1970;19:64–108.[32] Walsh KA, Neurath H. Trypsinogen and chymotrypsinogen as homologous proteins. Proc

Natl Acad Sci 1964;52:884–9.[33] Hartley BS, Kauffman DK. Corrections to the amino acid sequence of bovine chymo-

trypsinogen A. Biochem J 1966;101:229–31.[34] Poser CM, Paty DW, Scheinberg L et al. New diagnostic criteria for multiple sclerosis:

guidelines for research protocols. Ann Neurol 1983;13:227–31.[35] Hall PK, Roberts RC. Physical and chemical properties of human plasma a -macroglobulin.2

Biochem J 1978;171:27–38.[36] Hunter WM, Greenwood FC. Preparation of iodine-131 labelled human growth hormone of

high specific activity. Nature 1962;194:495–6.[37] Laemmli UK. Cleavage of structural proteins during the assembly of the head of bac-

teriophage T4. Nature 1970;227:680–5.[38] Matsudaira P. Sequence from picomole quantities of proteins electroblotted onto poly-

vinylidene difluoride membranes. J Biol Chem 1987;262:10035–8.[39] Bridges MA, Applegarth DA, Johannson J. Assessment of protease-binding by a -macro-2

globulin in multiple sclerosis. Clin Chim Acta 1982;121:167–72.[40] Shanbhag VP, Stigbrand T, Jensen PEH. The contact zones in human a -macroglobulin:2

Functional domains important for the regulation of the trapping mechanism. Eur J Biochem1997;244:694–9.

¨ ¨[41] Jensen PEH, Hagglof EM, Arbelaez LF, Stigbrand T, Shanbhag VP. Comparison ofconformational changes of pregnancy zone protein and human a -macroglobulin, a study2

using hydrophobic affinity partitioning. Biochim Biophys Acta 1993;1164:152–8.[42] Parisi AF, Vallee BL. Isolation of a zinc a -macroglobulin from human serum. Biochemistry2

1970;9:2421–6.¨[43] Osterberg R, Malmensten B. Methylamine-induced conformational change of a -macro-2

globulin and its zinc(II) binding capacity. Eur J Biochem 1984;143:541–4.[44] Ziska P, Mohr J. Affinity chromatography of serum proteins using immobilized lectin from

vicia faba. Acta Biol Med Ger 1970;36:1197–8.[45] Du Y, Ni B, Glinn M et al. a -macroglobulin as a b-amyloid peptide-binding plasma protein.2

J Neurochem 1997;69:299–305.[46] Stollar BD, Rezuke W. Separation of anti-histone antibodies from nonimmune histone-

precipitating serum proteins, predominantly a -macroglobulin. Arch Biochem Biophys2

1978;190:398–404.[47] McPherson TA, Marchalonis JJ, Lennon V. Binding of encephalitogenic basic protein by

serum a-globulins. Immunology 1970;19:929–33.[48] Gunnarsson M, Jensen PEH. Binding of soluble myelin basic protein to various conforma-

tional forms of a -macroglobulin. Arch Biochem Biophys 1998;15:192–8.2

[49] Van der Veen RC, Brinkman CJJ, Hommes OR, Lamers KJB. The effect of myelin basicprotein on the protease inhibitors a -antitrypsin and a -macroglobulin. Acta Neurol Scand1 2

1985;71:199–205.


[50] Woodward MP, Young Jr. WW, Bloodgood RA. Detection of monoclonal antibodies specificfor carbohydrate epitopes using periodate oxidation. J Immunol Methods 1985;78:143–53.

[51] Hud E, Cohen MP. Evaluation and performance characteristics of a novel ELISA usingmonoclonal antibody to glycated albumin. Clin Chim Acta 1989;185:157–64.

[52] Silvestrini B, Guglielmotti A, Saso L, Cheng CY. Changes in concanavalin A-reactiveproteins in inflammatory disorders. Clin Chem 1989;35:2207–11.

[53] Saso L, Silvestrini B, Guglielmotti A, Lahita R, Cheng CY. Abnormal glycosylation ofa -macroglobulin, a non-acute-phase protein, in patients with autoimmune diseases. Inflam-2

mation 1993;17:465–79.[54] Saso L, Valentini G, Leone MG, Grippa E, Guglielmi R, Paris L, Cantore G, Silvestrini B.

Changes in Concanavalin A-reactive Proteins in Neurological Disorders. J Clin Lab Anal1999;13:158–65.

[55] Matthijs G, Marynen P. A deletion polymorphism in the human a -macroglobulin (A2M)2

gene. Nucleic Acids Res 1991;19:5102.[56] Poller W, Faber JP, Klobeck G, Olek K. Cloning of the human a -macroglobulin gene and2

detection of mutations in two functional domains: the bait region and the thiolester site. HumGenet 1992;88:313–9.

[57] Blacker D, Wilcox MA, Laird NM et al. a -macroglobulin is genetically associated with2

Alzheimer disease. Nat Genet 1998;19:357–60.[58] Liao A, Nitsch RM, Greenberg SM et al. Genetic association of an a -macroglobulin2

(Val1000lle) polymorphism and Alzheimer’s disease. Hum Mol Genet 1998;7:1953–6.

Documents

Aberrant forms of α2-macroglobulin purified from patients with multiple sclerosis