- Email: [email protected]
Antibodies against sulfated glycosphingolipids of peripheral nerve myelins detected in patients with human cytomegalovirus infection
Journal of Neuroimmunology ELSEVIER
Journal of Neuroimmunology55 (1994) 55-60
Antibodies against sulfated glycosphingolipids of peripheral nerve myelins detected in patients with human cytomegalovirus infection K. Ogawa-Goto *, K. Kubota, A. Kurotani, T. Abe Department of Pediatrics, Teikyo University School of Medicine, 2-11-1 Kaga, Itabashi-ku, Tokyo 173, Japan
Received 21 March 1994; revision received and accepted 12 July 1994
Abstract
In this paper we studied whether cytomegalovirus (CMV) infection could induce a production of antibodies against PNS glycosphingolipids (GL). Sera from patients with congenital CMV infection were tested for IgM and IgG antibodies against acidic and neutral GL purified from human PNS. TLC-immunostaining assay revealed that the CMV-infected patients' sera contained antibodies against sulfoglucuronyl glycosphingolipids (SGGL), which also bound to sulfatide with a low affinity. No reactivity was observed to PNS gangliosides or neutral GL. The antibody also bound to other sulfated glycolipids including seminolipid and LacCer-sulfate, but not to cholesterol-sulfate, suggesting that a sulfated sugar chain may be important for their low-affinity binding. Furthermore, both anti-sulfatide and anti-SGGL antibodies were absorbed with sulfatide-conjugated octyl-Sepharose and heparin-Sepharose columns, whereas CMV-specific IgG titer was not decreased by the absorption of anti-sulfated GL antibody. These results suggest that CMV infection might specifically induce production of antibodies against sulfated GL, whereas these antibodies differed from CMV-specific antibody. Keywords: Cytomegalovirus; Sulfated glycosphingolipids; Antibody, HNK-1
1. Introduction
Glycosphingolipids (GL) are components of plasma m e m b r a n e s and are considered to play important roles in m e m b r a n e function (Yamakawa and Nagai, 1978), especially in the nervous system by their enrichment and localization in the neural m e m b r a n e (Norton et al., 1975, Ogawa-Goto et al., 1993). Over the last few years, much interest has been directed to the potential immunological role of G L antigens in the pathogenesis of peripheral neuropathy including Guillain-Barr6 syn-
* Corresponding author. Abbreviations: PNS, peripheral nervous system; HPLC, high-performance liquid chromatography; GL, glycosphingolipids; ELISA, enzyme-linked immunosorbent assay; CMV, cytomegalovirus; GBS, Guillain-Barr6 syndrome; SGGL, sulfated glucuronyl glycosphingolipids; SGPG, SO3-V3-GIcUfll-3nLc4Cer; SGLPG, SO3-VII 3GlcU/~l-3nl.,c6Cer; nLc4Cer, Gall31-4GlcNAc[31-3Gal~I-4Glcl31l'Cer; nLc6Cer, Galfll-4GlcNAcfll-3Galfll-4GlcNAcfll-3Gal/314Glcfll-l'Cer; LacCer, Gal/31-4Glcl-l'Cer; Gb3Cer, Galal-4Galfll4Glcl-YCer; Gb4Cer, GalNAc/31-3Galal-4Gal/31-4Glcl-l'Cer; sulfatide (CSE), GalCer-Ia-sulfate; lacotosylsulfatide, LacCer-II 3sulfate; 0165-5728/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0165-5728(94)00102-2
drome (GBS). GBS is characterized as an acute inflammatory demyelinating polyneuropathy, in which myelin is thought to be the target of immune attack. The disease often occurs between 1 and 6 weeks after an infection such as respiratory infection and gastroenteritis. From these evidences, it was suggested that autoimmune mechanisms may be involved in this disease. Antibodies which are produced against infective agents may cross-react with peripheral nerve myelin antigens. Cytomegalovirus (CMV) was reported to be one of the frequent antecedent infections associated with autoimmune inflammatory demyelinating neuropathies (Winer et al., 1988; R o p p e r et al., 1991), while no reports have so far been published about shared antigens between C M V and peripheral nervous system (PNS) myelin. The purpose of this study was to find out whether CMV infection could induce antibody productions against PNS GL. Sera from patients with congenital C M V infection were tested for IgM and I g G antibodies against acidic and neutral G L purified from human PNS, demonstrating that the antibodies bound to sulfated glycolipids which were considered to be myelinassociated.
56
K. Ogawa-Goto et al. /Journal of Neuroimmunologs, 55 (1994) 55- 60
2. Materials and methods
2.1. Patients Sera were obtained from infants with symptomatic congenital HCMV infection over a period of months. Sera were also obtained from mothers of the HCMVinfected infants, and normal adults (laboratory workers), and were frozen at -75°C prior to analysis.
2.2. Materials Human root nerves without neurological disease were dissected at autopsy and stocked at -80°C until use. Standard GL (LacCer, Gb3Cer, Gb4Cer) were prepared from human erythrocyte membrane. Monoclonal antibody HNK-1 hybridoma clone was obtained from the American Type Culture Collection (Rockville, MD). Cholesterol sulfate was purchased from Sigma (St. Louis, MO). LacCer-sulfate and seminolipid were generous gifts of Dr. Ishizuka (Teikyo University). Heparin-Sepharose and octyl-Sepharose CL6B were obtained from Pharmacia (Uppsala, Sweden)
nol / w at er (55/45/10, by volume) including 0.02% CaC12 • 2H 20 (solvent system 1) was used as a developing solvent system.
2.6. Enzyme-linked immunosorbent assay (ELISA) of anti-glycosphingolipid antibody Binding of human IgG or IgM to the sulfated GL was determined by an ELISA procedure in the following manner. 150 pmol of sulfatide or 50 pmol of SGGL in methanol was added to each well and evaporated to dryness. Unreactive sites were saturated with 1% bovine serum albumin (BSA) in phosphate-buffered saline (PBS), at pH 7.4, Sera were tested at various dilutions, beginning at 1:100, and incubated for 16 h at 4°C. After four washes, peroxidase-conjugated goat anti-human IgG or IgM was added to the wells. After overnight incubation and washing, 50 txl of substrate solution (TM-Blue, Transbenic Sciences Inc., MA) was added to each well. The reaction was stopped by adding 2 N HCl and the absorbance at 450 nm was measured. Absorbance values were corrected by subtracting values obtained in wells without antigens.
2.3. HCMV antibodies titers 2. 7. Affinity column chromatography CMV complement fixing antibody (CF) titers were expressed as the reciprocal of serum dilutions which were determined in the clinical virology laboratory of the Teikyo University Hospital. Anti-HCMV IgM antibody and anti-HCMV IgG antibody titers were measured by enzyme-linked immunosorbent assay (ELISA) with HCMV-specific antigen-coated plates (Enzygnost, Behringwerke, Marburg, Germany).
2.4. Purification of acidic and neutral GL Purification of acidic and neutral GL from human root nerves was carried out as described previously (Ogawa-Goto et al., 1990, 1993). Briefly, an acidic lipid fraction eluted from a DEAE-Sephadex (A-25) column with 0.25 M ammonium acetate in methanol contained mainly gangliosides and sulfatide. The subsequent acidic fraction, eluted with 0.5 M ammonium acetate in methanol, contained sulfoglucuronyl glycosphingolipids (SGGL) including SGPG and SGLPG, which were monoclonal antibody HNK-l-positive GL. SGLPG was further purified fi'om SGGL fraction by HPLC with Aquasil column (Asahikasei, Tokyo, Japan).
2.5. TLC-immunostaining GL antigens were detected on aluminium-backed TLC plates (Silicagel 60, Merck) as described previously (Ogawa-Goto et al., 1993). Chloroform/metha-
Sera from HCMV-infected patients were absorbed with glycolipid-conjugated octyl-Sepharose CL-4B beads (Pharmacia, Uppsala, Sweden). Conjugates of sulfatide and octyl-Sepharose CL-4B were prepared as reported previously (Hirabayashi et al., 1983). The sulfatide-coated gel thus obtained was packed in a small column for affinity chromatography. 50 /xl of patient's serum diluted with 140 ~1 of PBS was applied on the affinity column. The column was first washed with 2.5 ml of PBS. A pass-through fraction was collected and designated as the CSE-I fraction. The antibody retained in the column was then eluted with 1.25 ml of 1.0 M NaI in PBS and then with 1.25 ml of 3.0 M NaSCN in PBS designated as CSE-II fraction. As a control experiment, a Sepharose CL-6B column which had not been conjugated sulfatide was also used for absorption of sera. The pass-through and eluted fractions obtained by the same procedure as the CSE column were designated as Cont-I and Cont-II, respectively. For antibody analyses, each fraction was deionized and concentrated by ultrafiltration by using Centricon 30 (Amicon, Japan). Heparin-Sepharose was also used for absorption of antibodies with the same elution procedure as the CSE-Sepharose column. A pass-through fraction was designated as Hep-I and an eluted fraction by 1.0 M NaI and 3.0 M NaSCN in PBS was designated as Hep-II.
K. Ogawa-Goto et aL /Journal of Neuroimmunology 55 (1994) 55-60
57
3. Results
Table 1 Antibody titers against H C M V and S G G L
3.1. TLC-immunostaining and ELISA assay of sera from HCMV-infected patients
Patients
a n t i - H C M V Ab
No.
IgG 1)
IgM 1)
CF
T L C 2)
ELISA 3)
1 2 3 4 5 6 7 8 lm * 2m * m e a n ± SD
+ + + + + + + nd + +
+ + + nd nd + -
256 32 64 4> 32 8 16 16 128 256
+ + + + + + + + + +
0.57 1.25 1.33 0.74 0.94 0.61 1.33 1.39 1.11 0.52 0.98 ± 0.33
The TLC-immunostaining assay was carried out to determine whether sera from HCMV-infected patients contain antibodies against GL of human peripheral nerves (Fig. 1). Acidic and neutral GL purified from human root nerves were used as antigens together with authentic ones. The IgM antibody from a newborn infant with congenital CMV infection bound to an acidic glycolipid in SGGL fraction (Fig. 1B, lane 3) and a high amount of sulfatide (40 nmol) in lane 2, but not to a small amount of sulfatide in lane 1 (0.16 nmol). The positive antigen in SGGL fraction seemed likely to be SGLPG, one of the higher SGGL. Other glycolipids were not stained by IgM from the patient, including authentic gangliosides of bovine brain (lane 1), gangliosides purified from motor nerves (lane 2), neutral glycolipids of motor nerves (lane 4) or those from erythrocyte membrane (lane 5). Similar staining patterns were observed from a mother of another newborn with congenital CMV infection (Fig. 1C) and also from eight HCMV-infected patients tested, whose serum IgG against HCMV were positive (Table 1). None of the control sera in the present study exhibited anti-SGGL antibody binding by TLC-immunostaining assays. The ELISA absorbance values for anti-SGGL antibodies in HCMV infected patients were significantly higher than those in the control group (Table 1). These results suggested that the CMV-infected patients' sera contained elevated titers of antibodies against SGGL which also cross-reacted with sulfatide with a low affinity.
A.
Controls (n = 28)
0.21 ± 0.14
1) determined by ELISA, 2) determined by TLC-immunostaining, 3) ELISA absorbance values for IgM with 200 fold diluted sera. * mothers of patients No. 1 and No. 2; nd, not determined.
In order to characterize the SGLPG-like antigen, SGLPG was purified by HPLC and used as an antigen. Fig. 2B shows a binding pattern of serum IgG of the patient with congenital CMV infection. SGLPG (lanes 1 and 2) was strongly positive. SGPG was also stained (lane 1), but it was observed that the binding to SGLPG was much stronger than that of SGPG. A similar tendency was observed in all SGGL-positive sera tested by TLC immunostaining. In the case of a monoclonal antibody HNK-1 which reacted with SGGL, the higher reactivity to SGLPG was not observed (Fig. 2C). Fig. 3 shows the binding pattern of the serum IgM from the patient with congenital CMV infection to a
B,
~iiii~iiiil if! ii! iii ¸/ii! ¸¸~
1 2 3 4 5
anti-SGGL Ab
C.
i?iii! ¸ i i¸i¸¸
1 2 3 4 5
1 2 3 4 5
Fig. 1. TLC-immunostaining pattern with sera from H C M V patients. Lane 1, standard gangliosides as 4.2 nmol sialic acids and 0.16 nmol sulfatide; lane 2, acidic glycolipid fraction purified from h u m a n cauda equina containing gangliosides (1.2 nmol sialic acids) and sulfatide (about 40 nmol); lane 3, S G G L fraction purified from h u m a n cauda equina containing 0.085 nmol S G P G and a trace amount of S G L P G (less than 0.04 nmol); lane 4, the higher neutral glycosphingolipids purified from 30 mg h u m a n cauda equina; lane 5, standard neutral glycosphingolipids containing Gb4Cer, Gb3Cer and LacCer (about 0.8 nmol each). (A) Orcinol spray; (B, C) immunostained with IgG and l g M from patient's serum, respectively. The plate was developed with solvent system 1.
K. Ogawa-Goto et al. /Journal of Neuroimmunology 55 (1994) 55-60
58
A
B
C
(A)
Anti-CSE
antibody
Cont-II
[] •
anti-CSE IgG Ab anti-CSE IgM Ab
Cont-I
CSE-II ~ CSE-I i (_)
~, i
SGPG
00
02
,.
!
i
0,4 Absorbance at 450nm
0.6
0.8
SGLP( (B) A n t i - S G G L
1 2
1 2
1 2
Fig. 2. Binding of serum IgG from a HCMV patient and HNK-1 monoclonal antibody to SGLPG purified by HPLC. Lane 1, SGGL fraction before purification containing 0.34 nmol SGPG and 0.08 nmol SGLPG; lane 2, 0.16 nmol SGLPG purified by HPLC. (A) Orcinol spray; (B) immunostained with IgG from a patient's serum; (C) immunostained with a monoclonal antibody HNK-1. The plate was developed with solvent system 1.
antibody
[] •
Cont-II ~ Cont-, CSE-II
CSE-I ~11 (_) i
0.0
v a r i e t y o f s u l f a t e d glycolipids. S e m i n o l i p i d ( l a n e 3), L a c C e r - s u l f a t e ( l a n e 2) w a s p o s i t i v e as w e l l as s u l f a t i d e ( l a n e 4), b u t c h o l e s t e r o l s u l f a t e w a s n e g a t i v e ( l a n e 1), suggesting that a sulfated sugar chain may be important for the low affinity binding.
anti-SGGL IgG Ab anti-SGGL IgM Ab
(C) A n t i - C M V
0'.2
0.4 Absorbance at 450nm
l
0.6
0.8
0.6
0.8
IgG a n t i b o d y
Cont-II Cont-I
3.2. Absorption of the patient's serum by affinity column chromatography S e r a f r o m H C M V - i n f e c t e d p a t i e n t s w e r e a p p l i e d to C S E - c o n j u g a t e d S e p h a r o s e C L - 4 B c o l u m n . I n a passt h r o u g h f r a c t i o n ( C S E - I f r a c t i o n ) (Fig. 4 A ) , I g G a n d I g M a n t i b o d y activity a g a i n s t C S E w a s d e c r e a s e d to
A. Orcinol
1234
B. Immunostaining
CSE-II
CSE-I (-) O0
02
0.4 Absorbance at 410nm
Fig. 4. Absorption of serum antibodies from the CMV patient with sulfatide-conjugated Sepharose CL-4B column. (A) IgM and IgG titers against sulfatide, (B) IgM and IgG titers against SGGL, (C) CMV-specific IgG antibody titer. 50 /xl of patient's serum diluted with 140 #1 of PBS was applied to a sulfatide-conjugated Sepharose CL-4B column. The columns were first washed with 2,5 ml of PBS and pass-through fractions were collected and designated as CSE-I. The antibody retained in the columns were then eluted with 1 M NaI in PBS and then with 3 M NaSCN in PBS designated as CSE-II fractions. As a control study, Sepharose CL-6B column which had not been conjugated by sulfatide were also used with the same procedure. Pass-through and eluted fractions obtained by the same procedure as the CSE column were designated as Cont-I and Cont-II, respectively. Anti-sulfatide and anti-SGGL antibody titers and antiCMV IgG titers were determined by ELISA as described in Materials and Methods.
1234
Fig. 3. Binding of serum IgG from a HCMV patient to various sulfated glycolipids. Lane 1, 20 nmol cholesterol sulfate; lane 2, 0.4 nmol LacCer-sulfate; lane 3, 0.4 nmol sulfatide; lane 4, 0.5 nmol seminolipids. (A) Orcinol spray; (B) immunostained with IgG from patient's serum. The plate was developed with solvent system 1.
less t h a n 2 0 % o f t h a t w i t h o u t t r e a t m e n t , a n d it w a s r e c o v e r e d in C S E - I I f r a c t i o n e l u t e d w i t h N a I a n d N a S C N . I n this a f f i n i t y c h r o m a t o g r a p h y , a n t i - S G G L a n t i b o d i e s w e r e also a d s o r b e d to t h e C S E - S e p h a r o s e
K. Ogawa-Gotoet al./Journal of Neuroirnmunology 55 (1994) 55-60
CL-4B column (Fig. 4B), suggesting that anti-CSE antibody and anti-SGGL antibody might be the same. In contrast, CMV-specific IgG antibody was detected in the CSE-I fraction, but not in the CSE-II fraction which contained high titers of anti-CSE antibodies under this condition (Fig. 4C). An assay of CMV CF titers in these fractions showed the same results as anti-HCMV antibodies (IgG) (data not shown). For absorption of the patient's serum, another affinity column chromatography was also carried out by using a heparin-Sepharose column. Anti-CSE antibody and anti-SGGL antibody activities in the patient's serum were obtained in an eluted fraction (Hep-II), whereas anti-HCMV IgG antibody was detected in a pass-through fraction (Hep-I) and not in the Hep-II fraction (data not shown). 4. Discussion
Increased activity of antibodies against sulfated GL has recently been demonstrated in patients with GBS or several neuropathies by various independent groups (McGinniss et al., 1988 Fredman et al., 1991; Ilyas et al., 1991; Pestronk et al., 1991). However, there have been no reports about associations between serum anti-sulfated GL antibodies and clinical or serological evidence of recent infectious episodes. In the present paper we showed that the CMV-infected patients' sera contained elevated titers of antibodies against SGGL, which also bound to sulfatide with a low affinity. This is the first report which demonstrates a possible association of CMV infection and antibodies against sulfated GL. SGPG contains sulfated glucuronic acid on its nonreducing end of nLc4Cer (Chou et al., 1985; Ariga et al., 1987), and SGLPG has a similar structure but with an additional unit of lactosamine (sulfated glucuronylnLc6Cer) (Chou et al., 1986; Ariga et al., 1987). Both were monoclonal antibody HNK-l-reactive GL. The carbohydrate determinant recognized by HNK-1 was also shared with myelin-associated glycoprotein (MAG), neural cell adhesive molecule (NCAM) and PO glycoprotein. It was of particular interest that the antibodies in CMV-infected patients' sera showed a higher reactivity to SGLPG than SGPG, as demonstrated by our TLC immunostaining assays. The similar specificities were previously observed in some patients with GBS (Ilyas et al., 1991). However, neither HNK-1 nor a monoclonal IgM against MAG from a polyneuropathy patient with paraproteinemia showed the higher reactivity to SGLPG than SGPG. A repeating lactosamine structure of SGLPG, besides sulfated glucuronic acid, could be an important epitope of the antibodies in the HCMV-infected patients, whereas the reason for the higher reactivity to SGLPG than
59
SGPG remains to be determined. Thus, a fine specificity of the antibodies detected in the HCMV-infected patients was very unique, and differed from that of HNK-1 and other related antibodies, although their levels were lower than monoclonal IgM against MAG from paraproteinemic polyneuropathy patients. There are three possible mechanisms for production of anti-sulfated GL antibodies by CMV infection. First, CMV contains a shared sulfated carbohydrate epitope on its envelope. However, our results showed that the anti-sulfated GL antibodies, which were purified by affinity column chromatography, did not contain antiCMV activity, suggesting that they are unlikely to be antibodies against the virus itself. Second, the antibodies may be produced against altered glycoconjugate patterns of host cells as a result of virus infection (Andrews et al., 1989). Last, they may be anti-idiotypic antibodies, which are produced against antibodies to viral attachment proteins and recognize the host cell viral receptor. Many examples of anti-receptor and anti-idiotypic antibodies induced by experimental viral infections have been reported, such as rheovirus and measles virus (Kauffman et al., 1983; Krah and Choppin, 1988). Recently, several independent groups reported that the initial interaction of CMV with host cells is mediated with glycosaminoglycans such as heparan sulfate (Kari and Gehrz, 1992; Neyz et al., 1992; Compton et al, 1993), indicating that cell surface glycoconjugates may act as initial receptors for CMV. On the basis of their findings and our present results, the anti-sulfated GL antibody might be anti-receptor and anti-idiotypic antibodies. However, the molecule which acts as immunogen remains to be determined. Low levels of antibodies against SGGL have been found in sera from patients with several neurological and other diseases and also with healthy individuals (McGinniss et al., 1988; Ilyas et al., 1991). Because infection with HCMV is a relatively common occurrence and the illness is frequently mild or subclinical, healthy individuals sometimes contains high titer antibodies against HCMV, indicating recent inapparent HCMV infection. Therefore, a further study on antiHCMV antibody analysis is necessary both in the diseased patients and healthy individuals reported as having increased anti-SGGL antibodies, which may be informative to prove an association between HCMV infection and anti-SGGL antibodies. Such a study is being performed at present. In our preliminary study we have already found that a healthy case who showed increased anti-SGGL antibody by TLC-immunostaining assay had high titer antibodies against HCMV (data not shown). Although the biological significance of anti-SGGL antibodies is uncertain, the present study suggests one of potential mechanisms for production of anti-SGGL antibodies.
60
K. Ogawa-Got o et al. / Journal of Neuroimmunology 55 (1994) 55--60
Acknowledgements The authors gratefully acknowledge Dr. D.M. Marcus ( B a y o r C o l l e g e o f M e d i c i n e ) for his c r i t i c a l r e a d i n g o f this m a n u s c r i p t . W e also t h a n k Ms. F. T a k e d a ( t h e clinical virology laboratory of the Teikyo University h o s p i t a l ) for t e s t i n g C M V C F titers. T h i s s t u d y was s u p p o r t e d in p a r t by g r a n t s f r o m t h e M i n i s t r y o f H e a l t h and Welfare of Japan, and from the Ministry of Culture, Science and Education of Japan.
References Andrews, P.W., Gonczol, E., Fenderson, B.A., Holmes, E.H., O'Malley, G., Hakomori, S. and Plotkin, S. (1989) Human cytomegalovirus induces stage-specific embryonic antigen 1 in differentiating human teratocarcinoma cells and fibroblasts. J. Exp. Med. 169, 1349-1359. Ariga, T., Kohriyama, T., Freddo, L., Latov, N., Saito, M., Kon, K., Ando, S., Suzuki, M., Hemling, M.E., Rinehart, K.L. Jr., Kusunoki, S. and Yu, R.K. (1987) Characterization of sulfated glucuronic acid containing glycolipids reacting with IgM M-proteins in patients with neuropathy. J. Biol. Chem. 262, 848-853. Chou, D.K.H., Ilyas, A.A., Evans, J.E., Quarles, R.H. and Jungalwara, F.B. (1985) Structure of a glycolipid reacting with monoclonal IgM in neuropathy and with HNK-1. Biochern. Biophys. Res. Commun. 128, 383-388. Chou, D.K.H., llyas, A.A., Evans, J.E., Costello, C, Quarles, R.H. and Jungalwara, F.B. (1986) Structure of sulfated glucuronyl glycolipids in the nervous system reacting with HNK-1 antibody and some lgM paraproteins in neuropathy. J. BioL Chem. 261, 11717-11725. Compton, T., Nowlin, D.M. and Cooper, N.R. (1993) Initiation of human cytomegalovirus infection requires initial interaction with cell surface heparin sulfate. Virology 193, 834-841. Fredman, P., Vedeler, C.A., Nyland, H., Aarli, J.A. and Svennerholm, L. (1991) Antibodies in sera from patients with inflammatory demyelinating polyradicuroneuropathy react with ganglioside LM1 and sulphatide of peripheral nerve myelin. J. Neurol. 238, 75-79. Hirabayashi, Y., Suzuki, T., Suzuki, Y., Taki, T., Matsumoto, M., Higashi, H. and Kato, S. (1983) A new method for purification of
anti-glycosphingolipid antibody. Avian anti-hematoside (NeuGc) antibody. J. Biochem. 94, 327-330. Ilyas, A.A., Mithen, F.A., Dalakas, M.C., Wargo, M., Chen, Z.-W., Bielory, L. and Cook, S.D. (1991) Antibodies to sulfated glycolipids in Guillain-Barr6 syndrome. Z NeuroL Sci. 105, 108-117. Kari, B. and R. Gehrz (1992) A human cytomegalovirus glycoprotein complex designated gC-II is a major heparin-binding component of the envelope. J. Virol. 66, 1761-1764. Kauffman, R.S., Noseworthy, J.H., Nepom, J.T., Finberg, R., Fields, B.N. and Green, M.I. (1983) Cell receptors for the mammalian rheovirus. 11. Monoclonal anti-idiotypic antibody blocks viral binding to cells. J. ImmunoL 131, 2539-2541. Krah, D.L. and Choppin, P.W. (1988) Mice immunized with measles virus develop antibodies to a cell surface receptor for binding virus. J. Virol. 62, 1565-1572. McGinnis, S., Kohriyama, T., Yu, R.K., Pesce, M.A. and Latov, N. (1988) Antibodies to sulfated glucuronic acid containing glycosphingolipids in neuropathy associated with anti-MAG antibodies and in normal subjects. Z NeuroimmunoL 17, 119-126. Neyts, J., Snoeck, R., Schols, D., Balazarini, J., Esko, J.D., Schepdael, A.V. and Clercq, E.D. (1992) Sulfated polymers inhibit the interaction of human cytomegalovirus with cell surface heparan sulfate. Virology 189, 48-58. Norton, W.T., Abe, T., Poduslo, S.E. and DeVries, G.H. (1975) The lipid composition of isolated brain cells and axons. J. Neurosci. Res. 1, 57-75. Ogawa-Goto, K., Funamoto, N., Abe, T. and Nagashima, K. (1990) Different ceramide compositions of gangliosides between human motor and sensory nerves. J. Neurochem. 55, 1486-1493. Ogawa-Goto, K., Ohta, Y., Kubota, K., Funamoto, N., Abe, T., Taki, T. and Nagashima, K. (1993) Glycosphingolipids of human peripheral nervous system myelins isolated from cauda equina. J. Neurochem. 61, 1398-1403. Pestronk, P., Li, F., Griffin, J., Feldman, E.L., Cornblath, D., Trotter, J., Zhu, S., Yee, W.C., Phillips, D., Peeples, D.M. and Winslow, B. (1991) Polyneuropathy syndromes associated with serum antibodies to sulfatide and myelin-associated glycoprotein. Neurology 41, 357-362. Ropper, A.H., Wijdicks, E.F.M. and Truax, B.T. (1991) Guillain-Barrd syndrome. F.A. Davis Company, Philadelphia, PA. Winer, J.B., Hughes, R.A.C., Anderson, M.J., Jones, D.M., Kangro, H. and Watkins, R.P.F. (1988) A prospective study of acute idiopathic neuropathy. II. Antecedent events. J. Neurol. Neurosurg. Psychiatr. 51, 613-618. Yamakawa, T. and Nagai, Y. (1978) Glycolipids at the cell surface and their biological functions. T1BS 3, 128-131.
Journal of Neuroimmunology55 (1994) 55-60
Antibodies against sulfated glycosphingolipids of peripheral nerve myelins detected in patients with human cytomegalovirus infection K. Ogawa-Goto *, K. Kubota, A. Kurotani, T. Abe Department of Pediatrics, Teikyo University School of Medicine, 2-11-1 Kaga, Itabashi-ku, Tokyo 173, Japan
Received 21 March 1994; revision received and accepted 12 July 1994
Abstract
In this paper we studied whether cytomegalovirus (CMV) infection could induce a production of antibodies against PNS glycosphingolipids (GL). Sera from patients with congenital CMV infection were tested for IgM and IgG antibodies against acidic and neutral GL purified from human PNS. TLC-immunostaining assay revealed that the CMV-infected patients' sera contained antibodies against sulfoglucuronyl glycosphingolipids (SGGL), which also bound to sulfatide with a low affinity. No reactivity was observed to PNS gangliosides or neutral GL. The antibody also bound to other sulfated glycolipids including seminolipid and LacCer-sulfate, but not to cholesterol-sulfate, suggesting that a sulfated sugar chain may be important for their low-affinity binding. Furthermore, both anti-sulfatide and anti-SGGL antibodies were absorbed with sulfatide-conjugated octyl-Sepharose and heparin-Sepharose columns, whereas CMV-specific IgG titer was not decreased by the absorption of anti-sulfated GL antibody. These results suggest that CMV infection might specifically induce production of antibodies against sulfated GL, whereas these antibodies differed from CMV-specific antibody. Keywords: Cytomegalovirus; Sulfated glycosphingolipids; Antibody, HNK-1
1. Introduction
Glycosphingolipids (GL) are components of plasma m e m b r a n e s and are considered to play important roles in m e m b r a n e function (Yamakawa and Nagai, 1978), especially in the nervous system by their enrichment and localization in the neural m e m b r a n e (Norton et al., 1975, Ogawa-Goto et al., 1993). Over the last few years, much interest has been directed to the potential immunological role of G L antigens in the pathogenesis of peripheral neuropathy including Guillain-Barr6 syn-
* Corresponding author. Abbreviations: PNS, peripheral nervous system; HPLC, high-performance liquid chromatography; GL, glycosphingolipids; ELISA, enzyme-linked immunosorbent assay; CMV, cytomegalovirus; GBS, Guillain-Barr6 syndrome; SGGL, sulfated glucuronyl glycosphingolipids; SGPG, SO3-V3-GIcUfll-3nLc4Cer; SGLPG, SO3-VII 3GlcU/~l-3nl.,c6Cer; nLc4Cer, Gall31-4GlcNAc[31-3Gal~I-4Glcl31l'Cer; nLc6Cer, Galfll-4GlcNAcfll-3Galfll-4GlcNAcfll-3Gal/314Glcfll-l'Cer; LacCer, Gal/31-4Glcl-l'Cer; Gb3Cer, Galal-4Galfll4Glcl-YCer; Gb4Cer, GalNAc/31-3Galal-4Gal/31-4Glcl-l'Cer; sulfatide (CSE), GalCer-Ia-sulfate; lacotosylsulfatide, LacCer-II 3sulfate; 0165-5728/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0165-5728(94)00102-2
drome (GBS). GBS is characterized as an acute inflammatory demyelinating polyneuropathy, in which myelin is thought to be the target of immune attack. The disease often occurs between 1 and 6 weeks after an infection such as respiratory infection and gastroenteritis. From these evidences, it was suggested that autoimmune mechanisms may be involved in this disease. Antibodies which are produced against infective agents may cross-react with peripheral nerve myelin antigens. Cytomegalovirus (CMV) was reported to be one of the frequent antecedent infections associated with autoimmune inflammatory demyelinating neuropathies (Winer et al., 1988; R o p p e r et al., 1991), while no reports have so far been published about shared antigens between C M V and peripheral nervous system (PNS) myelin. The purpose of this study was to find out whether CMV infection could induce antibody productions against PNS GL. Sera from patients with congenital C M V infection were tested for IgM and I g G antibodies against acidic and neutral G L purified from human PNS, demonstrating that the antibodies bound to sulfated glycolipids which were considered to be myelinassociated.
56
K. Ogawa-Goto et al. /Journal of Neuroimmunologs, 55 (1994) 55- 60
2. Materials and methods
2.1. Patients Sera were obtained from infants with symptomatic congenital HCMV infection over a period of months. Sera were also obtained from mothers of the HCMVinfected infants, and normal adults (laboratory workers), and were frozen at -75°C prior to analysis.
2.2. Materials Human root nerves without neurological disease were dissected at autopsy and stocked at -80°C until use. Standard GL (LacCer, Gb3Cer, Gb4Cer) were prepared from human erythrocyte membrane. Monoclonal antibody HNK-1 hybridoma clone was obtained from the American Type Culture Collection (Rockville, MD). Cholesterol sulfate was purchased from Sigma (St. Louis, MO). LacCer-sulfate and seminolipid were generous gifts of Dr. Ishizuka (Teikyo University). Heparin-Sepharose and octyl-Sepharose CL6B were obtained from Pharmacia (Uppsala, Sweden)
nol / w at er (55/45/10, by volume) including 0.02% CaC12 • 2H 20 (solvent system 1) was used as a developing solvent system.
2.6. Enzyme-linked immunosorbent assay (ELISA) of anti-glycosphingolipid antibody Binding of human IgG or IgM to the sulfated GL was determined by an ELISA procedure in the following manner. 150 pmol of sulfatide or 50 pmol of SGGL in methanol was added to each well and evaporated to dryness. Unreactive sites were saturated with 1% bovine serum albumin (BSA) in phosphate-buffered saline (PBS), at pH 7.4, Sera were tested at various dilutions, beginning at 1:100, and incubated for 16 h at 4°C. After four washes, peroxidase-conjugated goat anti-human IgG or IgM was added to the wells. After overnight incubation and washing, 50 txl of substrate solution (TM-Blue, Transbenic Sciences Inc., MA) was added to each well. The reaction was stopped by adding 2 N HCl and the absorbance at 450 nm was measured. Absorbance values were corrected by subtracting values obtained in wells without antigens.
2.3. HCMV antibodies titers 2. 7. Affinity column chromatography CMV complement fixing antibody (CF) titers were expressed as the reciprocal of serum dilutions which were determined in the clinical virology laboratory of the Teikyo University Hospital. Anti-HCMV IgM antibody and anti-HCMV IgG antibody titers were measured by enzyme-linked immunosorbent assay (ELISA) with HCMV-specific antigen-coated plates (Enzygnost, Behringwerke, Marburg, Germany).
2.4. Purification of acidic and neutral GL Purification of acidic and neutral GL from human root nerves was carried out as described previously (Ogawa-Goto et al., 1990, 1993). Briefly, an acidic lipid fraction eluted from a DEAE-Sephadex (A-25) column with 0.25 M ammonium acetate in methanol contained mainly gangliosides and sulfatide. The subsequent acidic fraction, eluted with 0.5 M ammonium acetate in methanol, contained sulfoglucuronyl glycosphingolipids (SGGL) including SGPG and SGLPG, which were monoclonal antibody HNK-l-positive GL. SGLPG was further purified fi'om SGGL fraction by HPLC with Aquasil column (Asahikasei, Tokyo, Japan).
2.5. TLC-immunostaining GL antigens were detected on aluminium-backed TLC plates (Silicagel 60, Merck) as described previously (Ogawa-Goto et al., 1993). Chloroform/metha-
Sera from HCMV-infected patients were absorbed with glycolipid-conjugated octyl-Sepharose CL-4B beads (Pharmacia, Uppsala, Sweden). Conjugates of sulfatide and octyl-Sepharose CL-4B were prepared as reported previously (Hirabayashi et al., 1983). The sulfatide-coated gel thus obtained was packed in a small column for affinity chromatography. 50 /xl of patient's serum diluted with 140 ~1 of PBS was applied on the affinity column. The column was first washed with 2.5 ml of PBS. A pass-through fraction was collected and designated as the CSE-I fraction. The antibody retained in the column was then eluted with 1.25 ml of 1.0 M NaI in PBS and then with 1.25 ml of 3.0 M NaSCN in PBS designated as CSE-II fraction. As a control experiment, a Sepharose CL-6B column which had not been conjugated sulfatide was also used for absorption of sera. The pass-through and eluted fractions obtained by the same procedure as the CSE column were designated as Cont-I and Cont-II, respectively. For antibody analyses, each fraction was deionized and concentrated by ultrafiltration by using Centricon 30 (Amicon, Japan). Heparin-Sepharose was also used for absorption of antibodies with the same elution procedure as the CSE-Sepharose column. A pass-through fraction was designated as Hep-I and an eluted fraction by 1.0 M NaI and 3.0 M NaSCN in PBS was designated as Hep-II.
K. Ogawa-Goto et aL /Journal of Neuroimmunology 55 (1994) 55-60
57
3. Results
Table 1 Antibody titers against H C M V and S G G L
3.1. TLC-immunostaining and ELISA assay of sera from HCMV-infected patients
Patients
a n t i - H C M V Ab
No.
IgG 1)
IgM 1)
CF
T L C 2)
ELISA 3)
1 2 3 4 5 6 7 8 lm * 2m * m e a n ± SD
+ + + + + + + nd + +
+ + + nd nd + -
256 32 64 4> 32 8 16 16 128 256
+ + + + + + + + + +
0.57 1.25 1.33 0.74 0.94 0.61 1.33 1.39 1.11 0.52 0.98 ± 0.33
The TLC-immunostaining assay was carried out to determine whether sera from HCMV-infected patients contain antibodies against GL of human peripheral nerves (Fig. 1). Acidic and neutral GL purified from human root nerves were used as antigens together with authentic ones. The IgM antibody from a newborn infant with congenital CMV infection bound to an acidic glycolipid in SGGL fraction (Fig. 1B, lane 3) and a high amount of sulfatide (40 nmol) in lane 2, but not to a small amount of sulfatide in lane 1 (0.16 nmol). The positive antigen in SGGL fraction seemed likely to be SGLPG, one of the higher SGGL. Other glycolipids were not stained by IgM from the patient, including authentic gangliosides of bovine brain (lane 1), gangliosides purified from motor nerves (lane 2), neutral glycolipids of motor nerves (lane 4) or those from erythrocyte membrane (lane 5). Similar staining patterns were observed from a mother of another newborn with congenital CMV infection (Fig. 1C) and also from eight HCMV-infected patients tested, whose serum IgG against HCMV were positive (Table 1). None of the control sera in the present study exhibited anti-SGGL antibody binding by TLC-immunostaining assays. The ELISA absorbance values for anti-SGGL antibodies in HCMV infected patients were significantly higher than those in the control group (Table 1). These results suggested that the CMV-infected patients' sera contained elevated titers of antibodies against SGGL which also cross-reacted with sulfatide with a low affinity.
A.
Controls (n = 28)
0.21 ± 0.14
1) determined by ELISA, 2) determined by TLC-immunostaining, 3) ELISA absorbance values for IgM with 200 fold diluted sera. * mothers of patients No. 1 and No. 2; nd, not determined.
In order to characterize the SGLPG-like antigen, SGLPG was purified by HPLC and used as an antigen. Fig. 2B shows a binding pattern of serum IgG of the patient with congenital CMV infection. SGLPG (lanes 1 and 2) was strongly positive. SGPG was also stained (lane 1), but it was observed that the binding to SGLPG was much stronger than that of SGPG. A similar tendency was observed in all SGGL-positive sera tested by TLC immunostaining. In the case of a monoclonal antibody HNK-1 which reacted with SGGL, the higher reactivity to SGLPG was not observed (Fig. 2C). Fig. 3 shows the binding pattern of the serum IgM from the patient with congenital CMV infection to a
B,
~iiii~iiiil if! ii! iii ¸/ii! ¸¸~
1 2 3 4 5
anti-SGGL Ab
C.
i?iii! ¸ i i¸i¸¸
1 2 3 4 5
1 2 3 4 5
Fig. 1. TLC-immunostaining pattern with sera from H C M V patients. Lane 1, standard gangliosides as 4.2 nmol sialic acids and 0.16 nmol sulfatide; lane 2, acidic glycolipid fraction purified from h u m a n cauda equina containing gangliosides (1.2 nmol sialic acids) and sulfatide (about 40 nmol); lane 3, S G G L fraction purified from h u m a n cauda equina containing 0.085 nmol S G P G and a trace amount of S G L P G (less than 0.04 nmol); lane 4, the higher neutral glycosphingolipids purified from 30 mg h u m a n cauda equina; lane 5, standard neutral glycosphingolipids containing Gb4Cer, Gb3Cer and LacCer (about 0.8 nmol each). (A) Orcinol spray; (B, C) immunostained with IgG and l g M from patient's serum, respectively. The plate was developed with solvent system 1.
K. Ogawa-Goto et al. /Journal of Neuroimmunology 55 (1994) 55-60
58
A
B
C
(A)
Anti-CSE
antibody
Cont-II
[] •
anti-CSE IgG Ab anti-CSE IgM Ab
Cont-I
CSE-II ~ CSE-I i (_)
~, i
SGPG
00
02
,.
!
i
0,4 Absorbance at 450nm
0.6
0.8
SGLP( (B) A n t i - S G G L
1 2
1 2
1 2
Fig. 2. Binding of serum IgG from a HCMV patient and HNK-1 monoclonal antibody to SGLPG purified by HPLC. Lane 1, SGGL fraction before purification containing 0.34 nmol SGPG and 0.08 nmol SGLPG; lane 2, 0.16 nmol SGLPG purified by HPLC. (A) Orcinol spray; (B) immunostained with IgG from a patient's serum; (C) immunostained with a monoclonal antibody HNK-1. The plate was developed with solvent system 1.
antibody
[] •
Cont-II ~ Cont-, CSE-II
CSE-I ~11 (_) i
0.0
v a r i e t y o f s u l f a t e d glycolipids. S e m i n o l i p i d ( l a n e 3), L a c C e r - s u l f a t e ( l a n e 2) w a s p o s i t i v e as w e l l as s u l f a t i d e ( l a n e 4), b u t c h o l e s t e r o l s u l f a t e w a s n e g a t i v e ( l a n e 1), suggesting that a sulfated sugar chain may be important for the low affinity binding.
anti-SGGL IgG Ab anti-SGGL IgM Ab
(C) A n t i - C M V
0'.2
0.4 Absorbance at 450nm
l
0.6
0.8
0.6
0.8
IgG a n t i b o d y
Cont-II Cont-I
3.2. Absorption of the patient's serum by affinity column chromatography S e r a f r o m H C M V - i n f e c t e d p a t i e n t s w e r e a p p l i e d to C S E - c o n j u g a t e d S e p h a r o s e C L - 4 B c o l u m n . I n a passt h r o u g h f r a c t i o n ( C S E - I f r a c t i o n ) (Fig. 4 A ) , I g G a n d I g M a n t i b o d y activity a g a i n s t C S E w a s d e c r e a s e d to
A. Orcinol
1234
B. Immunostaining
CSE-II
CSE-I (-) O0
02
0.4 Absorbance at 410nm
Fig. 4. Absorption of serum antibodies from the CMV patient with sulfatide-conjugated Sepharose CL-4B column. (A) IgM and IgG titers against sulfatide, (B) IgM and IgG titers against SGGL, (C) CMV-specific IgG antibody titer. 50 /xl of patient's serum diluted with 140 #1 of PBS was applied to a sulfatide-conjugated Sepharose CL-4B column. The columns were first washed with 2,5 ml of PBS and pass-through fractions were collected and designated as CSE-I. The antibody retained in the columns were then eluted with 1 M NaI in PBS and then with 3 M NaSCN in PBS designated as CSE-II fractions. As a control study, Sepharose CL-6B column which had not been conjugated by sulfatide were also used with the same procedure. Pass-through and eluted fractions obtained by the same procedure as the CSE column were designated as Cont-I and Cont-II, respectively. Anti-sulfatide and anti-SGGL antibody titers and antiCMV IgG titers were determined by ELISA as described in Materials and Methods.
1234
Fig. 3. Binding of serum IgG from a HCMV patient to various sulfated glycolipids. Lane 1, 20 nmol cholesterol sulfate; lane 2, 0.4 nmol LacCer-sulfate; lane 3, 0.4 nmol sulfatide; lane 4, 0.5 nmol seminolipids. (A) Orcinol spray; (B) immunostained with IgG from patient's serum. The plate was developed with solvent system 1.
less t h a n 2 0 % o f t h a t w i t h o u t t r e a t m e n t , a n d it w a s r e c o v e r e d in C S E - I I f r a c t i o n e l u t e d w i t h N a I a n d N a S C N . I n this a f f i n i t y c h r o m a t o g r a p h y , a n t i - S G G L a n t i b o d i e s w e r e also a d s o r b e d to t h e C S E - S e p h a r o s e
K. Ogawa-Gotoet al./Journal of Neuroirnmunology 55 (1994) 55-60
CL-4B column (Fig. 4B), suggesting that anti-CSE antibody and anti-SGGL antibody might be the same. In contrast, CMV-specific IgG antibody was detected in the CSE-I fraction, but not in the CSE-II fraction which contained high titers of anti-CSE antibodies under this condition (Fig. 4C). An assay of CMV CF titers in these fractions showed the same results as anti-HCMV antibodies (IgG) (data not shown). For absorption of the patient's serum, another affinity column chromatography was also carried out by using a heparin-Sepharose column. Anti-CSE antibody and anti-SGGL antibody activities in the patient's serum were obtained in an eluted fraction (Hep-II), whereas anti-HCMV IgG antibody was detected in a pass-through fraction (Hep-I) and not in the Hep-II fraction (data not shown). 4. Discussion
Increased activity of antibodies against sulfated GL has recently been demonstrated in patients with GBS or several neuropathies by various independent groups (McGinniss et al., 1988 Fredman et al., 1991; Ilyas et al., 1991; Pestronk et al., 1991). However, there have been no reports about associations between serum anti-sulfated GL antibodies and clinical or serological evidence of recent infectious episodes. In the present paper we showed that the CMV-infected patients' sera contained elevated titers of antibodies against SGGL, which also bound to sulfatide with a low affinity. This is the first report which demonstrates a possible association of CMV infection and antibodies against sulfated GL. SGPG contains sulfated glucuronic acid on its nonreducing end of nLc4Cer (Chou et al., 1985; Ariga et al., 1987), and SGLPG has a similar structure but with an additional unit of lactosamine (sulfated glucuronylnLc6Cer) (Chou et al., 1986; Ariga et al., 1987). Both were monoclonal antibody HNK-l-reactive GL. The carbohydrate determinant recognized by HNK-1 was also shared with myelin-associated glycoprotein (MAG), neural cell adhesive molecule (NCAM) and PO glycoprotein. It was of particular interest that the antibodies in CMV-infected patients' sera showed a higher reactivity to SGLPG than SGPG, as demonstrated by our TLC immunostaining assays. The similar specificities were previously observed in some patients with GBS (Ilyas et al., 1991). However, neither HNK-1 nor a monoclonal IgM against MAG from a polyneuropathy patient with paraproteinemia showed the higher reactivity to SGLPG than SGPG. A repeating lactosamine structure of SGLPG, besides sulfated glucuronic acid, could be an important epitope of the antibodies in the HCMV-infected patients, whereas the reason for the higher reactivity to SGLPG than
59
SGPG remains to be determined. Thus, a fine specificity of the antibodies detected in the HCMV-infected patients was very unique, and differed from that of HNK-1 and other related antibodies, although their levels were lower than monoclonal IgM against MAG from paraproteinemic polyneuropathy patients. There are three possible mechanisms for production of anti-sulfated GL antibodies by CMV infection. First, CMV contains a shared sulfated carbohydrate epitope on its envelope. However, our results showed that the anti-sulfated GL antibodies, which were purified by affinity column chromatography, did not contain antiCMV activity, suggesting that they are unlikely to be antibodies against the virus itself. Second, the antibodies may be produced against altered glycoconjugate patterns of host cells as a result of virus infection (Andrews et al., 1989). Last, they may be anti-idiotypic antibodies, which are produced against antibodies to viral attachment proteins and recognize the host cell viral receptor. Many examples of anti-receptor and anti-idiotypic antibodies induced by experimental viral infections have been reported, such as rheovirus and measles virus (Kauffman et al., 1983; Krah and Choppin, 1988). Recently, several independent groups reported that the initial interaction of CMV with host cells is mediated with glycosaminoglycans such as heparan sulfate (Kari and Gehrz, 1992; Neyz et al., 1992; Compton et al, 1993), indicating that cell surface glycoconjugates may act as initial receptors for CMV. On the basis of their findings and our present results, the anti-sulfated GL antibody might be anti-receptor and anti-idiotypic antibodies. However, the molecule which acts as immunogen remains to be determined. Low levels of antibodies against SGGL have been found in sera from patients with several neurological and other diseases and also with healthy individuals (McGinniss et al., 1988; Ilyas et al., 1991). Because infection with HCMV is a relatively common occurrence and the illness is frequently mild or subclinical, healthy individuals sometimes contains high titer antibodies against HCMV, indicating recent inapparent HCMV infection. Therefore, a further study on antiHCMV antibody analysis is necessary both in the diseased patients and healthy individuals reported as having increased anti-SGGL antibodies, which may be informative to prove an association between HCMV infection and anti-SGGL antibodies. Such a study is being performed at present. In our preliminary study we have already found that a healthy case who showed increased anti-SGGL antibody by TLC-immunostaining assay had high titer antibodies against HCMV (data not shown). Although the biological significance of anti-SGGL antibodies is uncertain, the present study suggests one of potential mechanisms for production of anti-SGGL antibodies.
60
K. Ogawa-Got o et al. / Journal of Neuroimmunology 55 (1994) 55--60
Acknowledgements The authors gratefully acknowledge Dr. D.M. Marcus ( B a y o r C o l l e g e o f M e d i c i n e ) for his c r i t i c a l r e a d i n g o f this m a n u s c r i p t . W e also t h a n k Ms. F. T a k e d a ( t h e clinical virology laboratory of the Teikyo University h o s p i t a l ) for t e s t i n g C M V C F titers. T h i s s t u d y was s u p p o r t e d in p a r t by g r a n t s f r o m t h e M i n i s t r y o f H e a l t h and Welfare of Japan, and from the Ministry of Culture, Science and Education of Japan.
References Andrews, P.W., Gonczol, E., Fenderson, B.A., Holmes, E.H., O'Malley, G., Hakomori, S. and Plotkin, S. (1989) Human cytomegalovirus induces stage-specific embryonic antigen 1 in differentiating human teratocarcinoma cells and fibroblasts. J. Exp. Med. 169, 1349-1359. Ariga, T., Kohriyama, T., Freddo, L., Latov, N., Saito, M., Kon, K., Ando, S., Suzuki, M., Hemling, M.E., Rinehart, K.L. Jr., Kusunoki, S. and Yu, R.K. (1987) Characterization of sulfated glucuronic acid containing glycolipids reacting with IgM M-proteins in patients with neuropathy. J. Biol. Chem. 262, 848-853. Chou, D.K.H., Ilyas, A.A., Evans, J.E., Quarles, R.H. and Jungalwara, F.B. (1985) Structure of a glycolipid reacting with monoclonal IgM in neuropathy and with HNK-1. Biochern. Biophys. Res. Commun. 128, 383-388. Chou, D.K.H., llyas, A.A., Evans, J.E., Costello, C, Quarles, R.H. and Jungalwara, F.B. (1986) Structure of sulfated glucuronyl glycolipids in the nervous system reacting with HNK-1 antibody and some lgM paraproteins in neuropathy. J. BioL Chem. 261, 11717-11725. Compton, T., Nowlin, D.M. and Cooper, N.R. (1993) Initiation of human cytomegalovirus infection requires initial interaction with cell surface heparin sulfate. Virology 193, 834-841. Fredman, P., Vedeler, C.A., Nyland, H., Aarli, J.A. and Svennerholm, L. (1991) Antibodies in sera from patients with inflammatory demyelinating polyradicuroneuropathy react with ganglioside LM1 and sulphatide of peripheral nerve myelin. J. Neurol. 238, 75-79. Hirabayashi, Y., Suzuki, T., Suzuki, Y., Taki, T., Matsumoto, M., Higashi, H. and Kato, S. (1983) A new method for purification of
anti-glycosphingolipid antibody. Avian anti-hematoside (NeuGc) antibody. J. Biochem. 94, 327-330. Ilyas, A.A., Mithen, F.A., Dalakas, M.C., Wargo, M., Chen, Z.-W., Bielory, L. and Cook, S.D. (1991) Antibodies to sulfated glycolipids in Guillain-Barr6 syndrome. Z NeuroL Sci. 105, 108-117. Kari, B. and R. Gehrz (1992) A human cytomegalovirus glycoprotein complex designated gC-II is a major heparin-binding component of the envelope. J. Virol. 66, 1761-1764. Kauffman, R.S., Noseworthy, J.H., Nepom, J.T., Finberg, R., Fields, B.N. and Green, M.I. (1983) Cell receptors for the mammalian rheovirus. 11. Monoclonal anti-idiotypic antibody blocks viral binding to cells. J. ImmunoL 131, 2539-2541. Krah, D.L. and Choppin, P.W. (1988) Mice immunized with measles virus develop antibodies to a cell surface receptor for binding virus. J. Virol. 62, 1565-1572. McGinnis, S., Kohriyama, T., Yu, R.K., Pesce, M.A. and Latov, N. (1988) Antibodies to sulfated glucuronic acid containing glycosphingolipids in neuropathy associated with anti-MAG antibodies and in normal subjects. Z NeuroimmunoL 17, 119-126. Neyts, J., Snoeck, R., Schols, D., Balazarini, J., Esko, J.D., Schepdael, A.V. and Clercq, E.D. (1992) Sulfated polymers inhibit the interaction of human cytomegalovirus with cell surface heparan sulfate. Virology 189, 48-58. Norton, W.T., Abe, T., Poduslo, S.E. and DeVries, G.H. (1975) The lipid composition of isolated brain cells and axons. J. Neurosci. Res. 1, 57-75. Ogawa-Goto, K., Funamoto, N., Abe, T. and Nagashima, K. (1990) Different ceramide compositions of gangliosides between human motor and sensory nerves. J. Neurochem. 55, 1486-1493. Ogawa-Goto, K., Ohta, Y., Kubota, K., Funamoto, N., Abe, T., Taki, T. and Nagashima, K. (1993) Glycosphingolipids of human peripheral nervous system myelins isolated from cauda equina. J. Neurochem. 61, 1398-1403. Pestronk, P., Li, F., Griffin, J., Feldman, E.L., Cornblath, D., Trotter, J., Zhu, S., Yee, W.C., Phillips, D., Peeples, D.M. and Winslow, B. (1991) Polyneuropathy syndromes associated with serum antibodies to sulfatide and myelin-associated glycoprotein. Neurology 41, 357-362. Ropper, A.H., Wijdicks, E.F.M. and Truax, B.T. (1991) Guillain-Barrd syndrome. F.A. Davis Company, Philadelphia, PA. Winer, J.B., Hughes, R.A.C., Anderson, M.J., Jones, D.M., Kangro, H. and Watkins, R.P.F. (1988) A prospective study of acute idiopathic neuropathy. II. Antecedent events. J. Neurol. Neurosurg. Psychiatr. 51, 613-618. Yamakawa, T. and Nagai, Y. (1978) Glycolipids at the cell surface and their biological functions. T1BS 3, 128-131.