Occurrence of oligoclonal IgM bands in the cerebrospinal fluid of neurological patients: An immunoaffinity-mediated capillary blot study

Occurrence of oligoclonal IgM bands in the cerebrospinal fluid of neurological patients: An immunoaffinity-mediated capillary blot study

JOURNAL OF THE NEUROLOGICAL SCIENCES E[£EVIER Jotirnal o! the Neurological Scictlccs 124 (Ic~tJ4) 215 21c; Occurrence of oligoclonal IgM bands in t...

541KB Sizes 0 Downloads 17 Views



Jotirnal o! the Neurological Scictlccs 124 (Ic~tJ4) 215 21c;

Occurrence of oligoclonal IgM bands in the cerebrospinal fluid of neurological patients: an immunoaffinity-mediated capillary blot study C.J.M. S i n d i c *, P. M o n t e y n c , E.C. k a t e r r e l . a h o r a t m v +~t N('t~rochcmistp3 mid DclJartmc#lt Of '~'rcItt'O]O'O) " ( 'lmiquc~ [till ctsihm-c~ 5ain/-l.u(, (at/toll+. ~m/~'r~it~ , q l <>m am, 5.7-5? A r c t u . , ,llmoticr. B- 1200 Brt+sscZv. Beh,,ium

(Rcccixcd 22 Octobcr 1993; revised I Pcbruai), 19c~4:accepted 15 February 19cJ4i

Abstract Wc dcvclopcd a highly scnsitivc and spccific immtinoaffinity-mcdiatcd capillar'> blot technique for the detection of oligoch+II;ll lgM bands in ('SI: and scra from patients with various neurological disorders. Prc-tt+catmcllt of the samples by dithiothrcitol x~as ncccssaty to obtain a migration of the lgM molcculc into the gel of isoclcctric focusing, lgM "~',is then transfcrrcd by immunoaffinit:> onto a polyvinylidcnc difluoridc sheet previously coatcLt with anti-lgM antiserum. The limit of detection was found to hc 6 ng in 15-/xl samples. The prcscncc of CSF-rcstrictcd lgM bands was considcrcd the result ot an intrathccal synthesis and was obscrvcd in 13 out of 46 (285{) patients with MS, more frcqucntly in acute relapses (tl out of 21, 43c~), including 6 cases out of 13 presenting the first bout of the discasc. Similar lgM b:_lnds wct+c also detected m 15 out of 40 (3sci) patients with ('NS irlfcctions, especially in cascs of ncurosyphilis and ncuroborrcliosis. The prcscncc of ('SF tdigochmal lgM bands was linked to an iricrcasc of the lgM iridcx in the MS group, but not in the group of infectious Ltiscascs a', a ~vholc. The t+ccut+rcrlcc of ('SF-rcstrictcd oligoclonal lgM bands seems to bc the most specific indicator of {111intrathcccil synlhcsis el this isotypc. Key word.v: lgM: Intrathccal immunity: Oligoclonal bend: Multiple sclerosis: CNS inlcctions

1. Introduction T h e p r e s e n c e of oligoclonal IgM b a n d s restricted to the C S F should be a sensitive m a r k e r of i n t r a t h e c a l synthesis, as already shown for oligoclonal IgG (Latcrre e t a l . 1970). It was indued r e p o r t e d that the qualitative detection of oligoelonal lgM b a n d s r e p r e s e n t s the best way for d e t e c t i n g the i n t r a t h c e a l IgM response b e c a u s e of a higher sensitivity than a q u a n t i t a t i v e assessment (Sharief c t a l . 1990). Despite a higher resolving capacity, isoelectric focusing ( I E F ) followed by blotting o n t o a native nitrocellulose sheet (Giles and W r o e 19911) was reported to bc less specific than agarose gel clectrophorcsis (Sharief and T h o m p s o n 1992) as some cross-reactions with lgG p a r a p r o t e i n b a n d s were still obscrved despite the use of a # - c h a i n specific first antibody.

+: ( ~ o r l c s p o n d i n g a u t h o r . Fax: ( + 32-21 7f~4 3679.

I)022-510X, cl4 SOT.Ill) c lt~C)4l:lscvicr Sdcncc B.V. All righls rcsct~cd

T h e aim of this work was three-fold: to analyse the frequency of oligochmal IgM b a n d s in the CSF of p a t i e n t s with '~arious rleurological disorders by using a highly sensitive and specific: capillary blot t e c h n i q u e based on all i n l m u n o a f f i n i i \ transfer after 11"]t:: to c o m p a r e these results w i t h q u a n t i t a t i v e l g M d e t e r m i nations; alld to c o r r e l a t e o u r findings w i t h the clinical course t'tf multiplc sclerosis (MS).

2. Patients Five groups of patients (#z 131) ~ c t e studied. G r o u p I (#l = 1()) consisted of n e l l - n e u r o l o g i c a l patients suffering f r o m m i n o r neurosis or tension headache but dcwfid of clinical signs of neurological disorders. They were n o r m a l by the criteria of e l c c t r o e n cepimlography, c o m p u t e d tomogr,lphy and ('St= amllysis. G r o u p I1 (it 46) consisted of patients with MS. All but one displayed oligoclonal IgG b a n d s restricted


C.J.M. Sindic et al. /Journal of the Neuroh)gical Sciences 124 (1994) 215-219

to the CSF and fulfilled the criteria for laboratory-supported MS, either probable or clinically definite; the last case had a clinically definite diagnosis based on a 20-year duration of the disease (Poser et al. 1983). Group III consisted of 46 patients with various infectious disorders of the CNS: 18 with aseptic meningitis (including two cases of varicella zoster and 3 cases of ECHO 30 viral infections), 7 with AIDS and neurological impairment, two with herpetic encephalitis, one with post-varicella encephalitis, 4 with neurosyphilis, two with neuroborreliosis, 4 with pyogenic meningitis, 3 with tuberculous meningitis, 4 with fungal meningitis and one with in utero cytomegalovirus (CMV) infection. Group IV (n = 8) consisted of patients with acute inflammatory polyradiculoneuritis (Guillain-Barr6 syndrome) and Group V (n = 21) were patients with other neurological disorders (OND; 9 with degenerative disorders of the CNS, two with sciatica, two with cervicarthrosis myelopathy, two with neurolupus, one with stroke, two with peripheral neuropathies, 3 with carcinomatous meningitis).

3. Materials and methods

Samples All paired samples of CSF and serum were collected for diagnostic purposes and aliquots were stored at - 2 0 ° C . IgM concentration in CSF was determined by particle counting immunoassay (Sindic et al. 1982). CSF and serum albumin and serum IgM were assayed by immuno-turbidimetry (Turbitimer, Behring, Marburg, Germany). The integrity of the blood-CSF barrier was assessed by the CSF albumin/serum albumin quotient (QAlb); its upper reference value was 8" 10 -3. The upper reference level of the IgM index was 0.055, Detection of CSF restricted oligoclonal IgG bands was performed by an immunoaffinity-capillary blot technique (Sindic and Laterre 1991).

thawed at 37°C for 15 min, vortexed and diluted as described under Results. To each 45 txl sample was then added 5/xl of dithiothreitol (DTT, Calbiochem) at a concentration of 0.4 mM in distilled water. After an incubation of 30 min at 37°C, 15-/,1 paired samples were applied onto the middle of the gel, side by side, and isoelectrically focused for 70 min at 10°C in a LKB Multiphor Unit. The gel was blotted onto a polyvinylidene difluoride (PVDF) sheet (Immobilon, 0.45 #m, Millipore, Bedford, MA, USA) that had been coated overnight with rabbit anti-IgM (m-chains) antiserum at a concentration of 30 /xg/mt of the IgG fraction (1 ml/10 cm 2 of PVDF area). The immunoaffinity-mediated capillary blotting was performed under a uniform weight of 1 kg for 40 min at 10°C. The immunoblot was then washed in Trisbuffered saline (TBS. Tris 20 raM, NaCI 500 mM, pH adjusted to 7.5 with HCl) containing 0.1% Tween-20 (Technicon, Tarrytown. USA) and afterwards dipped into a solution of 0.25% glutaraldehyde (Merck) in PBS for 20 min at 4°C. The immunoblot was further washed in TBS-Tween before incubation for 60 min at 20°C with peroxidase-conjugated rabbit anti-human IgM antiserum diluted 800 times in TBS containing 0.3% (w/v) defatted milk powder. After three 15-min washings with TBS, the immunoblots were stained by 3,Y-diaminobenzidine (DAB. Sigma) (50 mg in 100 ml PBS with 0.03% H202).

4. Results

DTT treatment A slight reduction of the samples by DTT was necessary to obtain a migration of the IgM molecule into the gel before its transfer by immunoaffinity on the PVDF sheet. The results were similar over a wide range of DTT concentration (between 0.01 and 0.2 mM) and we arbitrarily chose a DTT concentration of 0.04 raM.

Antisera Rabbit anti-human IgM and peroxidase-conjugated rabbit anti-human IgM specific for/,-chains were obtained from DAKO (Copenhagen, Denmark) (code A 091 and P 215, respectively).

Immunoaffinity-mediated capillary blotting Agarose gel plates (1 × 110 × 240 mm) were prepared in 30 ml of distilled water with 0.36 g agarose-IEF (Pharmacia) and 4.3 g sorbitol (Merck) containing 1 ml ampholine, pH 3.5-9.5, and 1 ml amphotine, pH 4-6 (Pharmalytes, Pharmacia). After each of 12 consecutive IEF runs, the pH gradient was measured directly on the gel by a surface electrode and was found to be linear and highly reproducible (data not shown): For the study of the total IgM pattern, CSF and sera were

Sensitivity and specificity of the method The limit of detection was found to be 6 ng in 15/,1 of samples put on the gel, corresponding to an IgM concentration of about 0.40/,g/ml. CSF samples with elevated IgM concentration were first diluted to 3-4 /zg/ml of IgM (i.e., 45-60 ng IgM in 15 /,l) before DTT treatment. Corresponding serum samples were always diluted to the same IgM concentration as the CSF ones. Cross-reactions with monoclonal IgG components were not observed (data not shown).

IgM detection in pathological samples Three types of results were obtained by eye reading of the immunoblots (Table 1): (a) undetectable IgM (Table 1, column 2): the immunoblot remained un-

('.,/.M. Simlic et al. /,Iourmd ~d the Netu'oh)gical Sciem'cs 124 (1994) 215 217 Table 1 P a t t e r n s of lgM i n m m n o b l o t s in ,~arious n e u r o l o g i c a l d i s o r d e r s Undetccb able (lroupl:('onlrols


( ; r o u p I1: MS


(h-oup 111: infectious d i s e a s e s aseplic meningitis viral e n c e p h a l i t i s in utero ( ' M V infection AIDS ncurosyphilis netlroborrcliosis pyogenicmeningitis tubcrcuhmsmeningitis fungalmcningitis






13 (18)



15 1

18 3 1 7 4 2 4 3


3 2 3

3 (15) 2 (2) 1 (1) I (6) 3 (4) 2 (2) 1 (4) I (3) 1 (2)


15 (39)






Oligochmal pattern (highlgM index)

{~ 1

all cast's








MS& S' ?SF

pH 6.8


(~rotip IV: (}BS






( h o u p V: a N D



-' (4)





30 (47)


stained because of the low amounts of lgM in the samples studied (below (I.40/xg/ml). This was the case for the 10 patients from thc control Group I, but also for 21 out of 46 patients with MS (Group II), 2 out of 8 with GBS (Group IV) and 12 out of 21 with a N D (Group V). (b) Polyclonal IgM (Table 1, column 3): staining of thc immunoblot revealed a diffuse polyclonal background identical in CSF and serum. No intratheeal production was dcmonstrated in such cases and a passive transudation through an impaired b l o o d CSF barrier was assumed to be present, as all these samples displayed an increase of their lgM concentration. For example, this polyclonal pattern was observed in 12 out of 4~ patients with MS, in 15 out of 18 with viral mcningitis, in 6 out of 8 with GBS and in 7 out of 21 with ONI). (c) Oligoclonal IgM (Table 1, column 4): the immunoblot was characterized in such cases by the presence of oligoclonal lgM bands in CSF, not present in serum samples, most often superimposed on a polyclonal background. Such a pattern was observed in 13 out of 46 (28C4) patients with MS (Fig. 1), in 15 out of 46 (380;) patients with CNS infections (Fig. 2) and in 2 out of 21 patients with e N D . It should be noted that three out of 4 patients with neurosyphilis, and the two patients with neuroborreliosis displayed CSF oligochmal IgM bands. The latter were rarely observed in AIDS with neurological involvemcnt ( 1 out of 8) and in aseptic meningitis (3 out of 18). Interestingly, two cases out of 3 with culture-proven ECHO-30 virus meningitis displayed oligoclonal lgM bands. Such bands were also detected in one case each of herpetic encephalitis, post-wtricella encephalitis, in utero CMV infcction,


A.8 &3

a n t i - I g M coated PVDF Fig. 1. l n m m n o b l o t s of lgM in the ( ' S F alld st.'rtilll (S) from 4 MS p a t i e n l s . In each case. the s e r u m was first d i h i t e d to the same lgM c o n c e n t r a t i o n as the c o r r c s p o n d i l l g ('SF. and both fluids were pro t r e a t e d ~,,ith D T T b e f o r e focusing. Note thc p r e s e n c e of oligoclonal IgM b a n d s r e s t r i c t e d to the ( ' S F m MS patienb, 1. 3 and 4. "l'hcsc h a n d s tire g e n e r a l l y more a n o d i c than the pol),clonal lgM pl¢Sulll i11 the serum. The total ttmOtllltS J,~l lgM s t l h m i t t c d to l'octisiilg wcrc 5. 14, 2(~ and 23 ng. respectively.

pyogenic, tuberculous and fungal meningitis, but not in the GBS group (Table 1). Finally the two positive patients from Group V suffered from neurolupus and idiopathic relapsing radicuhmeuritis, rcspcctivcly.

Relationship between the qtmlitatit'e (oligoc/onal lgM) and the qua#ttitatit'e (lgM index) a~se.vvm'#tt o1" lgM intrathecal sy#tthesis The IgM index was increased in 18 cases (39c7) of MS patients whereas oligoclonal igM bands wcrc detectable in 13 (28c<:). All cases with oligochmal lgM


GBS 1 ~






Neurosyphitis 2 3 \





CMV /.

-, !



68 6.3 5.8 53 /..8 A.3 a n t i - IgM coated PVDF Fig. 2. lmnnm,,)blc.ts of lgM in the CSF and ~,erum (S) h-,mn t pLtticllt with (;BS, 2 patients with neurc,syphilis and I nc,.,,born v,ith an in utero ('MV infection. The total ~llllOtll]ts of IgM subnlittcd to locusing w e r e It). {~£~, OI and 4 ng. i¢specti~

C.J.M. Sindic et al. /Journal qf the Neurological Sciences 124 (1994) 215-2t t)

bands also displayed an increased IgM index. This high concordance in MS was not observed in other diseases often characterized by a more severe impairment of the blood-CSF barrier. The most discrepant results were observed in aseptic meningitis (increase of the IgM index in 15 cases, oligoclonal IgM bands in only 3 cases), in AIDS (6 versus 1, respectively), in pyogenic meningitis (4 versus 1) and in GBS (4 versus 0).

Relationship between clinical course and oligoclonal IgM bands in MS Out of the 46 MS patients under study, 21 were in acute relapses at the time of the lumbar puncture and had not yet received corticosteroids or immunosuppressive drugs; 13 cases out of these 21 experienced the first bout of the disease and the interval between the first clinical signs and the CSF collection was less than 3 months. Six presented a secondary progressive course of the disease 5-20 years after onset, and 4 of these had previously received cyclophosphamide. Twelve presented a primary progressive course of the disease and 7 were clinically stable at the time of the lumbar puncture; these 19 patients have never received immunosuppressive drugs. Oligoclonal IgM bands were detectable in 9 out of 21 with acute relapses (43%) but in only 4 out of the 25 patients with other forms of the disease (16%). However, these differences did not reach a statistical significance. Similarly, an high IgM index was also more frequently present in acute relapses: 11 out 21 (52%) versus 7 out 25 (28%) with other forms of the disease. Among the t3 patients experiencing the first bout of their disease, 6 (46%) displayed CSF oligoclonal IgM bands and 7 (54%) an increased IgM index.

5. Discussion

From a technical point of view, our immunoaffinity mediated capillary blot method was a synthesis, at least in part, from a number of published techniques dealing with the detection of oligoelonal immunoglobulins in the CSF. In order to avoid protein loss or aggregation, we did not concentrate CSF samples as did Giles and Wroe (1990). The overall sensitivity of the immunoblot assay was increased by the glutaratdehyde fixation of the primary antibody-antigen complex (Ikegaki and Kenneth,1989). The pH gradient was extended between 4 and 6 by a mixture of ampholines and samples were pre-treated by DTF in order to obtain a better IgM penetration and migration into the gel (Kaiser 1991). Finally, the specificity was increased by the double use of anti-IgM (~-chain specific) antisera, first on the PVDF sheet for the immunoaffinity transfer, and then for the immunodetection with the peroxidase conjugate (McLaehlan 1989).

CSF restricted oligoclonal IgM bands were detected in most cases of neuroborreliosis and neurosyphilis, these infections being characterized by a strong IgM response (Felgenhauer and Sch~idlich 1987). In MS, CSF oligoclonal IgM bands were detected in 13 out of 46 patients (28%). This percentage is lower than the figure of 55% reported by Sharief and Thompson (1991). However, we, as well as these authors, find that detection of oligoclonal IgM bands varies with the clinical activity of the disease. Specifically, in first bouts of the disease as well as in acute relapses, oligoclonal IgM bands were present in about 45% of the time. In initial MS, Lolli et al (1990) also found an increased IgM index in 14 out 33 (42%) MS patients. Similarly, Frequin and coworkers (1992) found a trend towards higher CSF IgM levels and IgM indexes in relapsing MS, with a decrease of both parameters after treatment by high dose intravenously administered methylprednisolone. In any case, these data imply a specific and new antigenic challenge concomitant to an acute bout, especially at the beginning of the MS process. In the MS group, all the 13 patients with oligoclonal IgM bands also displayed an high IgM index, the frequency of the latter being somewhat higher (18 positive, that is 39%). This high concordance in MS is strikingly different of what is observed in the group of infectious diseases: 39 patients displayed a high IgM index but only 15 had CSF oligoclonat IgM bands. For example, within the 15 cases of aseptic meningitis with a high IgM index, only 3 had an oligoclonal IgM pattern. This discrepancy may have two explanations: (1) as CSF samples are only collected within a few days after onset of meningitis, the intrathecal production of IgM may be still polyclonal before its restriction to oligoclonality; (2) in such cases, the IgM index may be misleadingly high because of a more severe blood-CSF impairment resulting in a tack of linearity between transudated albumin and IgM. The latter assumption is supported by the fact that we failed to detect oligoctonal IgM bands in GBS patients in spite of the increase of the IgM index in 4 cases whose the blood CSF barrier was severely impaired (QAIb ranging from 13 to 65.10-3). The IgM index has been shown to increase with progressive dysfunction of this barrier (Felgenhauer 1982) and in such cases, quantitative linear formulae may yield falsely abnormal results. In MS however, the blood-CSF barrier is either normal or minimally disturbed and the concordance between an increased IgM index and the presence of oligoclonal IgM bands is high. In conclusion, the detection of CSF-restricted oligoclonal IgM bands is the best indicator of an intrathecal synthesis of this isotype. Their presence indicates a recent immunological stimulation, as observed in early MS and in acute relapses, or an active infection as

C.J.),I. Sindic et al. /Journal ~ f the Neurological 5;cience~ 124 ( lg~/4t 215 210

mainly observed in neuroborreliosis and neurosyphilis (Sharief and Thompson 1989).

Acknowledgements We arc indebted to Prof. C.J. Pfau for his critical reading of the manuscript. The skilful technical assistance of Mrs. M.P. Van Antwerpen and Ms. R. Marichal is gratefully acknowledged. This work was supported by grants from the "Fondation B6n6" and "Les Amis rdunis de Philippeville".

References Felgenhauer, K. (1t~82) Differentiation of the humoral immune response in inflammatory diseases of the central nervous system. J. Neurol., 228:223 237. Felgenhauer, K. and H.J. Sch~idlich (19871 The compartmental lgM and lgA response within the central nervous system. J. Neurol. Sci., 77:125-135. Frequin, S.T.F.M., F. Barkhof, K.J.B. Lamers, O.R. H o m m e s and G.F. Borm (1992) Cerebrospinal fluid myelin basic protein, lgG and lgM levels in 1(11 multiple sclerosis patients before and after treatment wilh high-dose intravenous methylprednisolone. Acta Neurol. Stand., 86:291 297. Giles, P.D. and S.J. Wroe (19901 Cerebrospinal fluid oligockmal IgM in multiple sclerosis: analytical problems and clinical limitations. Ann. Clin. Biochem., 27: 199-207. Ikegaki N. and R.II. Kennett (1989) Glutaraldehyde fixation of thc primary antibody-antigen complex on nitrocellulose paper increases the overall sensitivity of immunoblot assay. J. lmmunol. Methods. 124: 205-210.


Kaiser, R. (1091) Affinity immunoblolling: rapid and sensitive dolce tion of oligochmal lgG, IgA and IgM ill unconcentratcd CSF by agarosc isoelectric focusing. J. Neurol. Sci., 103: 21~ 225. Lalerre, E.('., A. Callewaert, J.F. l t e r e m a n s and Z. Sfacllo (1~1701 Etectrophoretic morphology o | gamma-globulins in ccrebrospinal fluid of multiple sclerosis and other diseases of lhc ncr',ou,~ system. Neurology, 2(1:982 000. Lolli, F., G, Sirucusa, M.P. Amato. l.. Fratiglioni. (;. l)al Pozzt~, [-L. Galli and L. Amaducci (1991) Intrafllecal symhesis ol free immunoglobulin light chains and IgM in inilial mulliplc sclerosis. Aeta Neurol. Stand., 83:239 243. McLachlan, R. (1081) Monochmal lnmmnogh~bulins: affinity blouing for low concentrations in serum. Clin. ('hem., 35:47S 4Sl. Poser, C.M., D.W. Paty, L. Scheinberg, W.I. McDonald, F.A. Davis, G.('. Ebers, K.P. Johnson, W.A. Sibley, 1).It. Silbcrbcrg and W.W. Tourtellotte (1983) Nev, diagnostic criteria to~ multiple sclerosis: guidelincs for research protocols. Ann. Ncuro].. 13: 227 23 I. Sharief, M.K. and E.J. T h o m p s o n (1089) Immunoglobulin M in the cerebrospinal fluid: an indicator ol recent immunological slimuh> tion. J. Neurol. Neurosurg. Psychiat., 52:040 ~53. Sharief, M.K. and E.J. T h o m p s o n (1091) lntrathecal immunogh)bulin M synthesis in multiple sclerosis. Relationship wilh clinical and cerebrospinal fluid parameters. P,rain, 114:181 1~15. Sharief. M.K. and E.J. T h o m p s o n (1002) l)istribulion of ccrcbrospinal fluid oligoclonal lgM bands ill neurological diseases: a comparison between agarose electrophorcsis and isoelectric locusing. J. Ncurol. Sci., 109:83 87. Sharief, M.K., G. Keir and E.J. T h o m p s o n (19911) lntrathecal synthc sis of lgM in neuroh)gical diseases: a comparison between detection of oligoclonal bands and quanlilalivc cstimalion..I. Ncurol. Sci.. 96:131 142. Sindic. ('.J.m., ('.L. Cambiaso, A. Dcpr~, E.('. [,alcrrc and P.l,. Mascon (19821 The concentration of lgM in lhc cercbrospinal fluid of neurological patients. J. Ncurol. Sci., 55: 37,t~ 35(I. Sindic, C.J.M. and E.C. Latcrre (1991) Oligoclonal free kappa and lambda bands in the cerebrospinal fluid of palicnts with multiple sclerosis anti other neurological diseases. An inmmnoaffinity mediated capillary blot study. J, Neunfimnmnol.. 33: 6~ 72.