J Neurol (1981) 225 : 135-143 Journal of

Neurology Springer-Verlag 1981

Measles Antibodies, x-2 Light Chain Distribution and Immunoglobulins in Serum, Cerebrospinal Fluid and Brain of a Patient Affected with Multiple Sclerosis

F. Bollengier ~*, A. Mahler 1, and G. Clinet 2

Laboratorium Fysiopathologie van het Zenuwstelsel, Vrije Universiteit Brussel, B-1090 Brussels, and

2 Institut Pasteur du Brabant, B-1040 Brussels, Belgium

Summary. The serum, cerebrospinal (CSF) and brain of a patient (NAG) affect- ed with multiple sclerosis (MS) were examined for measles antibodies with CF and HI techniques, and the x-2 light chain ratios of all samples available were evaluated. ;<-2 populations of the matched serum, CSF and brain specimens were all 2-predominant and in agreement with each other; the light chain distribution of the brain specimens confirmed previous findings [3].

Only the serum immunoglobulins showed significant measles antibody titers, but slightly increased measles antibody titers were also observed in ventricutar plaques.

The amount of immunoglobulin G (IgG) synthesized per day by t h e central nervous system (CNS) was estimated.

The IgG synthesis in CNS NAG (11.6mg/day) was above the upper limit of the normal range (3.3mg/day), but apparently there was no positive correlation between the intracerebral IgG synthesis and specific anti-measles IgG.

Key words: Multiple sclerosis -- CNS immunoglobulins - Light chain ratios - Measles antibodies

Zusammenfassung. Das Serum und der Liquor sowie das Gehirn eines Patien- ten (NAG),der an Multipler Sklerose (MS)litt, wurden auf das Vorhandensein von Masernantik6rper mit der CF- und HI-Technik untersucht und die x-2- Relation der leichten Ketten aller vorhandenen Proben wurde ausgewertet. In allen Proben lag eine 2-Dominanz vor, sowohl im Serum, im Liquor als auch im Gehirn. Die Verteilung der leichten Ketten in den Gehirnproben best~itigt frtihere Ergebnisse [3].

Nur in den Serumimmunoglobulinen konnten signifikante Masernanti- k6rpertiter nachgewiesen werden, jedoch waren leicht erh6hte Masernanti- k6rpertiter auch in den ventrikelnahen Plaques vorhanden.

* Corresponding author

0340-5354/81/0225/0135/$ 01.80

136 F. Bollengier et al.:

Es wurde die Menge des tfiglich durch das zentrale Nervensystem syntheti- sierte Immunoglobulin G (IgG) geschfitzt.

Die IgG-Synthese im Zentralnervensystem der Patientin NAG (11.6mg/ Tag) lag fiber der tiblichen Normgrenze (3.3 mg/Tag) aber es schien keine positive Korrelation zwischen der intrazerebralen IgG-Synthese einerseits und den spezifischen Antimasern-IgG andererseits vorzuliegen.

Introduction

In previous papers we examined for measles antibodies in brain extracts and demyelination plaques of several MS patients [3], and for measles antibodies and the K-2 light chain distribution in fractionated serum immunoglobulins of patients affected with MS [4]. Both reports emphasized that there is very little evidence to suggest that measles antigen is a general agent responsible for MS, although the serum of some of the patients studied, yielded immunoglobulin fractions with slightly elevated measles antibodies.

We were unfortunately not able to correlate the findings in the MS brain with those from matched serum and CSF. In the present paper, however, we were able to examine for measles antibodies in matched total brain, topographically local- ized brain fractions, CSF, and immunoglobulin G-fractions obtained by liquid isoelectric focusing from the serum of a patient affected with MS. The fc-). light chain distribution of all the samples was evaluated and the amount of IgG synthesized by the CNS was estimated according to Tourtellotte's new formula [21].

Material and Methods

The diagnosis of MS had been established in patient NAG (28 years, (2) at the age of 14 years in 1965, on the basis of clinical findings and thorough routine examination of the serum and CSF. The onset of the disease was characterized by a staggering gait and frequent falls; the evolution was progressive without remission. The patient was ultimately hospitalized in 1979 with severe dyspnea and dysphagia.

On admission she was unable to walk, had ataxic nystagmus, bilateral Babinski reflexes, and a massive bronchopneumonia to which she soon succumbed.

The brain was removed 2h after death and one half frozen at -80°C until used. The other half served for macroscopic and microscopic examination which revealed several old demyelin- ated plaques with axonal preservation and gliosis. There was gliosis in the hemispheres and brainstem and plaques were observed in the medulla oblongata and spinal cord at every level. Here demyelination was accompanied by Wallerian degeneration.

Serum and CSF were obtained when the patient was in the terminal stage of the disease; this material was preserved at - 35°C for later study.

The serum immunoglobulins were prepared by Na2SO4 precipitation at concentrations of 18~/~ and 12% [11].

The lyophilized material was submitted to liquid isoelectric focusing in an LKB 8100 ampholine column (LKB-Produkten, AB Bromma, Sweden); the pH gradient was established with LKB ampholines (pH: 9.5-8.0 (60%) - pH: 8.0-5.0 (40%)) and isotocusing carried out for about 50h at 4°C.

The proteins were studied by agar gel electrophoresis [22] and thin layer isoelectric focusing [lOl.

Measles Antibodies

+

A

137

B

C

Fig. 1. A: agar gel electrophoresis of serum NAG. B: agar gel electrophoresis of CSF NAG. C: agar gel electrophoresis of brain extract NAG. Oligoclonal IgG is on the side of the cathode ($)

Table 1, pH-dependent immunoglobul in fractions obtained by column isofocusing in serum from multiple sclerosis patient N A G

pH range K/2 CF a HI b adsorbed HI RBC c +kaolin non adsorbed

Iso 1 8.90-8.70 n.d. n.d. n.d. n.d.

Iso 2 8.61-8.41 1.2 1/32 1/128 1/128

Iso 3 8.35-8.29 1.1 - - 1/16 1/32

Iso 4 8.19-8.04 1.3 - - 1/16 1/16

. Iso 5 7.97-7.71 1.4 - - 1/16 1/16

Iso 6 7.62-7.58 1.4 1/4 1/8 1/8

Iso 7 7.50-7.41 1.4 - - 1/8 1/4

Total IgG 1.1 1/4 1/16 1/16

Serum 0.7 1/32 1/128 1/256

CSF a 0.7 1/4 1/16 1/16

Brain 0.6 - - - - - -

a CF: complement fixation b HI: hemagglutinat ion inhibition c RBC: red blood cells « 10x concentrated CSF: results were negative when assayed unconcentrated n.d.: not determined

138 F. Bollengier et al.:

The brain specimens (50% w/v homogenate) were homogenized in 0.01 M phosphate-NaCl pH 7.4 (PBS) and centrifuged at 20,000 rpm. When necessary supernates were concentrated by pressure ultrafiltration.

Hemagglutination inhibition and complement fixation tests were carried out as previously described [18,23] and light chain ratios were determined as reported elsewhere [2].

The amount of IgG synthesized per day by the CNS was estimated according to the method of Tourtellotte [20, 21].

R e s u l t s

W h e n e x a m i n e d in aga r gel e l e c t r o p h o r e s i s the s e r u m s h o w e d one fa in t b a n d , a n d

the C S F f o u r M - c o m p o n e n t s . In b r a i n ex t rac t s , t o o , o l i goc lona l cha rac t e r i s t i c s

were a lso ev iden t , a n d p a r t o f the b r a i n M - c o m p o n e n t s r e p r e s e n t i m m u n o -

g lobu l in s (see be low) (Fig. 1).

Table 2. x/2 light chain ratios. CF and HI measles antibody titers of whole white and grey brain matter and plaques from patient NAG

K/2 CF a HI b adsorbed HI RBCC+kaolin non adsorbed

Total brain 0.6 - - - - - -

White matter 1.1 1/4 1/16 1/16

Grey matter 1.0 1/4 1/8 1/8

Ventricular plaque 1 1.0 1/16 1/8 1/8 Surrounding white matter 1.1 1/32 1/32 1/32

Ventricular plaque 2 1.1 1/4 1/8 1/8 (ventficular junction)

Temporal plaque (temporal horn) 1.1 1/4 1/8 1/8 Surrounding white matter 1.1 1/8 1/8 1/8

Temporo-orbital plaque 1.0 1/4 1/8 1/8 (ventricular junction)

Surrounding white matter 0.9 1/4 1/8 1/8

Caudal plaque 1.0 1/4 - - - - Surrounding white matter 0.6 1/4 - - - -

Brainstem plaque 0.4 - - - - - - Surrounding white and grey matter 0.6 - - - - - -

Bulbar plaque 0.5 1/4 1/8 1/8 Surrounding white matter 0.5 1/4 1/8 1/8

Album cerebellum plaque n.d. n.d. n.d. n.d. Surrounding white matter 0.9 1/2 - - - -

a CF: complement fixation b H I : hemagglutination inhibition « RBC: red blood cells n.d.: not determined

Measles Antibodies 139

Table 3. Estimate of the amount of IgG synthesized per day in the CNS of patient NAG

Controls Patient NAG mg/100 ml mg/100 ml

Mean value IgG serum (n = 20) 1,196

Mean value IgG CSF (n = 24) 1.85

Mean value Alb. serum a 4,000 Mean value Alb. CSF (n = 20) 15.3

IgG serum 1,500 IgG CSF 6.0 Alb. serum 2,700 Alb. CSF 16.0 Total protein serum 6,990 Total protein CSF 64.9

De novo IgG synthesis in CNS NAG: ll.6mg/day

" Mean value published in [13]

Immunoglobulin G was isolated from the serum of NAG by sodium sulphate precipitation and submitted to column isofocusing. The pH-dependent immuno- globulin fractions were grouped with regard to their measles antibody activity and their x-2. The findings in serum immunoglobulins of NAG confirm those previously reported [4].

Table 1 presents the ~c/2 and the measles antibody activity of the pH- dependent serum immunoglobulin fractions, of the CSF and brain. The CF and HI measles antibody activity of the serum seems to be concentrated in one cathodic fraction i.e. iso 2, and there is a general increase of the Kl2 ratlos when compared to the total serum K/2, which is clearly 2-predominant.

A very good correlation between the x/2 of matched serum, CSF and brain is also observed.

The slightly increased measles antibody activity seems to be mainly confined to the serum, since the CF and HI response in unconcentrated CSF was negative; concentration was necessäry to obtain a positive result which most probably means that the major part of the IgG present in the CSF does not represent measles antibody.

Table 2 reports the ~c-2 light chain ratios and the measles antibody activity in whole white and grey matter and topographically localized plaque material from the brain.

x/2 ratios ranging from 1.I to 0.4 are considerably 2-predominant and in most specimens there is a striking similarity between plaque material and sur- rounding white matter. The lowest x/2 values are found in the brainstem. 2- predominant light chain ratios seem to be characteristic of old plaques. In a study of an MS brain where most plaques were old, we found a similar light chain distribution [3]. Esiri [6] reported ~c/2 of 1.53 in non plaque material and 1.6 in old plaques; on the contrary, in moderately recent plaques she reported K/2 of 1.9 and in most recent plaques of 2.3.

Measles antibody activity in the brain was far from outstanding, being perhaps significant only in ventricular plaque 1 and its surrounding white matter.

The de novo CNS IgG synthesis was calculated according to Tourtellotte's new formula [21].

140

+

F. Boltengier et al.:

A B C D E F G H I J K L M N O P

Fig. 2. Thin layer isoelectric focusing of brain specimens NAG A: white matter surrounding B B: ventricular plaque 1 C: white matter surrounding D D: temporo-orbital plaque E: white matter surrounding F F: plaque temporat horn G: ventricular plaque 2 H: white matter surrounding I

I: caudal plaque J: white and grey matter surrounding K K: brainstem ptaque L: white matter surrounding M M: bulbar plaque N: white matter surrounding cerebellar plaque O: normal looking grey matter P: normal looking white matter

Since our mean values for both cont ro l serum and C S F - I g G , and also cont ro l serum and C S F a lbumin were different f rom those found by Tour te l lo t te , we ca lcula ted CNS IgG synthesis using the convers ion factors we der ived f rom our own mean values (Table 3).

F ina l ly all the different samples were submi t ted to thin layer isoelectric focusing (Fig. 2). Whi te , grey and p laque mater ia l all showed very s imilar mult i - band ing and clearly had most componen t s in common .

N o ma jo r qual i ta t ive specific MS changes were observed in the p H range used. However , c o m p o n e n t 4, which was present in grey mat te r and plaques, appea red d iminished or absent in white ma t t e r and su r round ing white mat ter , but these changes were not quant i ta ted . C o m p o n e n t 9 cor responds to a neuropep t ide which we recent ly isola ted f rom con t ro l bra in and whose molecu la r weight is -23 ,500 . This neu ropo lypep t ide par t ia l ly contr ibutes to the ol igoclonal results from bra in (in press). F r o m the topograph ica l d i s t r ibu t ion of the different bra in specimens, it is concluded that this neu ropo lypep t ide is ubiqui tous in bo th MS and cont ro l brain.

Discussion

In a recent pape r [4] we emphas ized that there is little g round for regarding measles ant igen as a general agent responsible for MS, and we wondered whether in certain MS pat ients , there might be a pa r t i cu la r immunolog ica l response to

Measles Antibodies 141

measles virus other than seen in measles itself. The data here assembled do not add up to a convincing argument for a measles virus etiology of MS hut leave the mechanism controversial.

Only the serum immunoglobulins showed non-normal measles antibody titers both in CF and HI. Several authors [8, 12, 15] have reported MS patients with significantly raised measles antibody titers both in HI and CF in unconcentrated CSF.

The present case does not belong to this category of MS patients, the CSF giving only slightly increased titer on concentration, with the only significant increases in the brain, i.e. in ventricular plaques. This could indicate that in measles positive MS, next to the synthesis of non-measles CNS IgG, measles antibody IgG diffuses from the serum to the CNS compartment. A serum component in the CSF and brain immunoglobulins in MS is also suggested by the very good K/2 correlation between matched serum, CSF and brain, the K/2 of the measles antibody positive ventricular plaque, and the ventricular plaques in general being almost identical to the ~¢/2 of the measles positive serum immuno- globulin fraction iso 2. The mechanism might be the following:

a) measles specific serum I g G ~ C S F ~ C N S b) de novo synthesized CSF I g G ~ d e novo synthesized CNS IgG This would be in good agreement with earlier findings of: 1) selectively increased permeability to specific IgG in the filtration of plasma

proteins during the formation of the CSF, or migration ofsmall inactive lympho- cytes from the blood into the subarachnoid space.

2) localized synthesis of IgG by lymphocytes that are situated within the brain [19].

On the whole, it has never been very clear whether the CSF is entirely generated from the choroid plexus in the lateral ventricles, or whether there is an important participation from the blood by means of the blood capillaries.

Tourtellotte et al. [20, 21], and Ewan et al. [7], demonstrated increased intra- cerebral IgG synthesis in MS and markedly elevated levels were mainly found during relapse and in chronic progressive cases. In a recent paper on antibodies to measles and distemper virus in SSPE and MS, Gorman et al. [9] confirmed the significantly elevated intracerebral IgG values in MS during relapse, but they also demonstrated that the level of CSF and serum measles was not of diagnostic significance in MS.

In the present case, although de novo CNS IgG synthesis has been demon- strated, there was no clear correlation between intracerebral IgG synthesis and specific anti-measles IgG, so that it cannot be maintained that this particular IgG is in fact measles antibody.

Since SSPE is characterized by significant intracerebral IgG synthesis [7,9] and by hyperimmunization against measles or measles-like virus [1, 17], this strongly suggests that in SSPE intracerebral IgG synthesis is a response to persistent antigenic stimulus. It might be so in MS brain, too, but the nature of the antigenic stimulus is unknown.

In agreement with Nordal et al. [12], if there is local antibody production, it "could well be a result of non-specific mitogenic stimulation associated in some unknown way with the disease process in the CNS."

142 F. Bollengier et al.:

Several authors [5, 14, 16] failed to demons t ra te measles antigen in brain

lesions, both in active and chronic MS, and were thus not able to confirm the measles virus et iology of MS. The findings in our patient confirm our earlier

reports [3, 14]; most p robably the et iology of MS is not specifically related to measles antigen, a l though one must bear in mind the possibility of an infection

with a defective or a related pa ramyxo virus. Clearly the basic p rob lem of identifying a tr iggering mechanism for the immunologica l response in MS

remains to be elucidated.

Acknowledgement. We are greatly indebted to Dr.P. Dajez (Brugmann Hospitaal--Vrije Universiteit Brussel) and the Department of Anatomo-Pathology (Brugmann Hospitaal--Vrije Universiteit Brussel) for the precious gift of the biological material. We are also very grateful to Prof. Dr. J. Perier (Laboratoire de Neuropathologie, Université Libre de Bruxelles) who carried out the localization and dissection of plaque material.

References

1. Adels BR, Gajdusek DC, Gibbs CJW, Albrecht P, Rogers N (1968) Attempt to transmit SSPE and isolate measles related agent, with a study of the immune response in patients and experimental animals. Neurology 18:30-51

2. Bollengier F, Lowenthal A, Henrotin W (1975) Bound an free light chains in subacute sclerosing panencephalitis and multiple sclerosis serum and cerebrospinal fluid. Z Klin Chem Klin Biochem 13 : 305-310

3. Bollengier F, Mahler A, Clinet G, Lowenthal A (1978) Multiple sclerosis: oligoclonal IgG, t¢-2 light chain distribution and measles antibodies in brain extracts. Brain Res 152: 133-144

4. Bollengier F, Mahler A (1979) Measles antibodies and ~c-2 light chain distribution in immuno- globulins of patients affected with multiple sclerosis. J Neurol 220:105-112

5. Dubois-Dalcq M, Schumacher G, Sever JL (1973) Acute multiple sclerosis. Electron micro- scopic evidence for and against a viral agent in the plaques. Lancet II: 1408-1411

6. Esiri MM (1977) Immunoglobulin containing cells in multiple sclerosis plaques. Lancet II: 478-480

7. Ewan PW, Lachmann PJ (1979) IgG synthesis within the brain in multiple sclerosis and subacute sclerosing panencephalitis. Clin Exp Immunol 35:227-235

8. Forghani B, Cremer NE, Johnson KP, Ginsberg AH, Likosky WH (1978) Viral antibodies in cerebrospinal fluid of multiple sclerosis and control patients: comparison between radio- immunoassay and conventional techniques. J Clin Microbiol 7:63-69

9. Gorman NT, Habricht J, Lachmann PJ (1980) Intracerebral synthesis of antibodies to measles and distemper viruses in patients with SSPE and MS. Clin Exp Immuno139 : 44-52

I0. Karlsson C, Davis H, Ohman J, Andersson UB (1973) LKB 2117 Multiphor I. Analytical thin layer gel electrofocusing in polyacrylamide gel. LKB Application Note--March 29

11. Keckwick RA (1940) The serum proteins in multiple myelomatosis. Biochem J 34:1248 -1257

12. Nordal HJ, Vandvik B, Norrby E (1978) Multiple sclerosis: local sYnthesis of electro- phoretically restricted measles, rubella, mumps and herpes simplex virus antibodies in the central nervous system. Scand J Immunol 7:473-479

13. Putnam FW (1975) The plasma proteins, vol II. Academic Press, NewYork 14. Raine CS, Prineas JW, Sheppard RD, Bornstein MB, Dubois-Dalcq M (1977) Immuno-

chemical studies for the localization of measles antige n MS plaques and measles virus- infected CNS-tissue. J Neurol Sci 33:13-20

15. Reunanen M, Arstila P, Hakkarainen H, Nikoskelainen J, Salmi A, Panelius M (1976) A longitudinal study on antibodies to measles and rubella viruses in patients with multiple sclerosis. A preliminary report. Acta Neurol Scand 54:1-12

Measles Antibodies 143

16. Tanaka R, Iwasaki Y, Koprowski H (1975) Ultrastructural studies of perivascular cuffing cells in multiple sclerosis brain. Am J Pathol 8:467-474

17. ter Meulen V, Müller D, Joppich G (1967) Fluorescence microscopy studies of brain tissue from a case of SSPE. Germ Med Mth 12:438-441

18. Thiry L, Dachy A, Lowenthal A (1969) Measles antibodies in patients with various types of measles infection. Arch Ges Virusforsch 28:278-284

19. Thompson EJ (1977) Laboratory diatgnosis of multiple sclerosis: immunological and bio- chemical aspects. Br Med Bull 33:28-33

20. Tourtellotte W (1970) On cerebrospinal fluid immunoglobulin G (IgG) quotients in multiple sclerosis and other diseases. A review and a new formula to estimate the amount of IgG synthesized per day by the central nervous system. J Neurol Sci 10:279-304

21. Tourtellotte WW, Booe IA (1978) Multiple sclerosis: The blood brain barrier and the measurement of de novo central nervous system IgG synthesis. Neurology 28 : 76-83

22. Wieme RJ (1964) Agar gel electrophoresis. Elsevier, Amsterdam 23. Zissis G, Clinet G (1974) Viral antibody detection by a more sensitive complement fixation

reaction. Lancet II: 754-755

Received October 20, 1980