Genetic variations and humoral immune responses to myelin

oligodendroglia glycoprotein in adult phenotypes of

X-linked adrenoleukodystrophy

Stephan Schmidta,*, Giovanna Maria Marrosub, Heike Kolschc, Claus G. Haased,Stanislav Ferenczike, Piotr Sokolowskif, Wolfgang Kohlerf, Martina Schmidta,

Andreas Papassotiropoulosg, Reinhard Heunc, Hans Grosse-Wildee, Thomas Klockgethera

aDepartment of Neurology, University of Bonn, Sigmund-Freud-Str. 20, 53105 Bonn, GermanybMultiple Sclerosis Center, University of Cagliari, Cagliari, Italy

cDepartment of Psychiatry and Psychotherapy, University of Bonn, Bonn, GermanydDepartment of Neurology, University of Essen, Essen, Germanye Institute of Immunology, University of Essen, Essen, Germany

fNeurological Department of Sachsisches Krankenhaus, Wermsdorf, GermanygDepartment of Psychiatry, University of Zurich, Zurich, Switzerland

Received 28 August 2002; received in revised form 15 November 2002; accepted 3 December 2002

Abstract

The lack of phenotype/genotype association in X-linked adrenoleukodystrophy (X-ALD) has prompted the search for disease modifying

factors. We previously demonstrated increased serum antibody responses against myelin oligodendrocyte glycoprotein (MOG) in various

clinical phenotypes of X-ALD allowing speculations that myelin specific humoral immune responses might be involved in phenotype

generation of X-ALD. In the present study, we investigated the possible association of (1) a naturally occurring variable number tandem

repeat (vntr) polymorphism (C allele) in the 3V flanking region of the interleukin-6 gene (IL-6), previously demonstrated to modify the course

of Alzheimer’s disease, systemic lupus erythematodes and Multiple Sclerosis (MS), (2) a tetranucleotide repeat polymorphism (TAAA)n in

the 3V flanking region of the MOG gene and (3) HLA class II alleles with adult clinical phenotypes and serum antibody responses to MOG in

70 adult X-ALD patients. HLA class II alleles, (TAAA)n MOG gene polymorphisms, and IL-6 C allele were not associated with clinical

phenotypes. Anti-MOG antibodies were detectable in 17/54 X-ALD patients (31.5%). Anti-MOG antibodies were associated with the 226 bp

(TAAA)n MOG gene polymorphism but not with distinct clinical phenotypes.

D 2002 Elsevier Science B.V. All rights reserved.

Keywords: Adrenoleukodystrophy; Adrenomyeloneuropathy; Myelin oligodendrocyte glycoprotein; HLA; Interleukin-6

1. Introduction

X-linked adrenoleukodystrophy (X-ALD) is an inherited

disorder of peroxisomal metabolism characterized by the

accumulation of very long chain fatty acids (VLCFA) in the

nervous system, the adrenal glands and testes (Moser, 1997).

Following the cloning of the X-ALD gene, more than 300

mostly ‘‘kindred-specific’’ mutations have been identified

(Moser, 1997). Notably, there is virtually no correlation

between different kinds of mutations and a particular clinical

phenotype, not even within the same kindred. Most strik-

ingly, the two most common phenotypes, cerebral ALD

(CALD) and adrenomyeloneuropathy (AMN), are not only

distinct with regard to the age of onset (CALD predom-

inantly in childhood, AMN almost exclusively in adulthood)

but also with regard to the pattern of nervous system

involvement (confluent cerebral inflammatory demyelinat-

ing lesions in CALD, neuronopathy of long tracts in the

myelon in association with a peripheral neuropathy with

little or no signs of inflammation in AMN). These discrep-

ancies have prompted the search for modifying genes,

however, as yet without conclusive results (Moser, 1997).

Unlike Multiple Sclerosis (MS), the most common human

demyelinating disease of the central nervous system (CNS),

0165-5728/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved.

doi:10.1016/S0165-5728(02)00445-9

* Corresponding author. Tel.: +49-228-287-5712; fax: +49-228-287-

5024.

E-mail address: Stephan.Schmidt@ukb.uni-bonn.de (S. Schmidt).

www.elsevier.com/locate/jneuroim

Journal of Neuroimmunology 135 (2003) 148–153

no firm association of X-ALD with HLA class II alleles has

been established (McGuiness et al., 1997; Moser, 1997;

Berger et al., 1995). Although several lines of evidence

suggest that excessive production of tumor necrosis factor-

alpha (TNFa) is pathogenetically relevant in the cerebral

form of X-ALD, the exact molecular mechanisms of CNS

demyelination have remained enigmatic (Aubourg et al.,

2000). Nonetheless, the hitherto unexplained contribution

of the immune system to the phenotypic differences in X-

ALD patients suggests that modifying genes in ‘‘immunor-

elevant’’ regions such as cytokine and myelin protein genes

might be involved in phenotype modification of X-ALD.

As one example of a disease-modifying genetic varation

of a cytokine gene, the so-called C allele of a variable

number tandem repeat (vntr) polymorphism located in the

3V flanking region of the interleukin-6 (IL-6) gene (IL-6)

has been shown to be associated with delayed onset of

Alzheimer’s disease (Papassotiropoulos et al., 1999) and a

benign course of MS in Sardinian patients (Vandenbroeck et

al., 2000). In accordance with the latter finding, a significant

underrepresentation of the IL-6 C allele was observed in

patients with progressive forms of MS (Schmidt et al.,

2000). Here, we tested the hypothesis if the IL-6 C allele

is also associated with distinct adult phenotypes of X-ALD.

The demonstration of myelin oligodendrocyte glycopro-

tein (MOG)-specific antibodies and antibody-dependent

immune effector mechanisms in lesion formation of exper-

imental autoimmune encephalomyelitis (EAE) (Stefferl et

al., 2000) and within inflammatory lesions of MS patients

(Raine et al., 1999) has highlighted the possible pathoge-

netic contribution of humoral immune responses in human

CNS demyelinating disease. We recently demonstrated

elevated levels of autoantibodies against a recombinant

preparation of MOG in various phenotypes of X-ALD

(Schmidt et al., 2001) confirming previous studies indicat-

ing that MOG-specific antibody responses are not confined

to MS (Reindl et al., 1999). Since cellular and humoral

MOG-specific immune responses in EAE are strongly

dependent on the genetic background of the experimental

animals (Stefferl et al., 2000), we tested the hypothesis if

anti-MOG autoantibody responses in X-ALD patients are

associated with HLA class II alleles and a naturally occur-

ring tetranucleotide (TAAA)n repeat polymorphism in the 3Vflanking region of the MOG gene.

2. Patients and methods

2.1. X-ALD patients

Serum and peripheral blood samples were obtained from

72 X-ALD patients recruited from Sachsisches Krankenhaus

Hubertusburg, Wermsdorf. X-ALD was diagnosed by deter-

mination of increased serum levels of very long chain fatty

acids (C22, C24, C26) and their respectice ratios (C26:C22,

C24:C22). The assignment of clinical phenotypes of X-ALD

patients was based on clinical history, neurological and

electrophysiological examination as well as magnetic reso-

nance imaging (MRI) (Moser, 1997). Cranial MRI was in

particular used to further differentiate between three AMN

subtypes (Kumar et al., 1995): Normal brain MRI and

thoracic cord atrophy (pure AMN), pyramidal or cortico-

spinal tract hyperintense lesions with otherwise normal brain

MRI and thoracic cord atrophy (AMN with tract involve-

ment), and white matter lesions in various brain regions

combined with pyramidal and corticospinal tract involve-

ment and thoracic cord atrophy (ALMN). A definite clinical

phenotype characterization was available in 70 X-ALD

patients who were selected for further statistical analysis.

16 patients represented pairs of heterozygous symptomatic

mothers and their affected sons. All but three patients were of

German Caucasian origin. For statistical analysis, AMN

patients were divided into two groups according to their

clinical subtypes, namely pure AMN and ALMN/AMN with

tract involvement. The clinical phenotypes and demographic

data of the 70 X-ALD patients included in the study are

given in Table 1. For the frequency distribution of IL-6 C

allele, a sample of 73 healthy individuals [32 males (44%),

41 females (56%)] with a mean age of 38.2F 6.7 years

(range 19–68 years) selected randomly from a large pop-

ulation-based sample participating in a family study of

dementia and depression was included (Heun et al., 2001).

All subjects were of German ancestry living in the federal

state of Nordrhein–Westfalen. All patients and control sub-

jects gave informed consent to participate in the study, and

the study was approved by the local ethics committees.

3. Genetic analysis

3.1. Analysis of IL-6 vntr polymorphism

Leukocyte DNA was isolated using standard protocols.

The vntr polymorphism, situated in the 3Vflank of IL-6 was

Table 1

Demographic and clinical data of X-ALD patients

n Age

(years)

Age of

onset (years)

All patients 70 42.2F 11.9

(19–68)

30.9F 13.2

(8–65)

Pure AMN 31 39.6F 8.8

(21–59)

27.0F 8.2

(13–41)

ALMN/AMN with

tract involvement

15 43.9F 12.2

(30–66)

29.3F 14.0

(8–58)

Female

heterozygotes

16 52.2F 10.9

(33–68)

44.1F11.5

(20–65)

CALD 4 35.5F 5.9

(28–42)

13.3F 7.5

(8–22)

Addison 2 24.5F 7.8

(19–30)

n.k.

Asymptomatic 2 22.0F 1.4

(21–23)

n.a.

n.a.: not applicable; n.k.: not known.

S. Schmidt et al. / Journal of Neuroimmunology 135 (2003) 148–153 149

studied as previously described (Papassotiropoulos et al.,

1999). The following oligonucleotide primers were used:

5V-GCA ACT TTG AGT GTG TCA CGT GAA-3V(for-(forward) and 5V-TGA CGT GAT GGA TGC AAC AC-3V(reverse). After polymerase chain reaction (35 cycles of 1

min at 94 jC, 1 min at 60 jC, and 1.5 min at 72 jC),amplification products were separated by 7% polyacryla-

mide gel electrophoresis and visualized by ethidium bromide

staining. Alleles were designated as previously described

(Bowcock et al., 1989). Four alleles (A–D) coding for

variants of four different lengths were identified. Allele A

yielded a product with the approximate size of 760 base pairs

(bp); the approximate sizes of alleles B, C, and D were 680,

640 and 610 bp, respectively. Only the presence or absence

of the IL-6 C allele was considered for statistical analysis.

3.2. Analysis of MOG gene (TAAA)n polymorphism

The polymorphic tetranucleotide repeats in exon 8 of the

MOG gene were analyzed as previously described (Malfroy

et al., 1995). Briefly, the oligonucleotide primers MOG50

(5V-ATCTTTCCTTCCTCTCATCC-3V) and MOG52 (5V-GGCTGGAGTAGAGGGAG-3V) were used for the tetranu-

cleotide repeat. Twenty-five cycles of PCR were performed

at the following conditions: 30 s at 94 jC, 30 s of 54 jC,and 30 s at 72 jC. In addition, an additional 12-cycle PCR

was performed using the oligonucleotide primer MOG51

(5V-CCAGGAGGCAGAGGTTG-3V) under the same PCR

conditions. Amplification products were loaded onto 6%

denaturing polyacrylamide gels with the GS-350 internal

lane standard (Applied Biosystems) and analyzed on ABI

373 DNA sequencing system. Seven alleles (A–G) coding

for variants of seven different lengths were identified. Allele

A yielded a product with the approximate size of 230 bp,

the approximate sizes of alleles B, C, D, E, F and G were

226, 222, 218, 214, 210 and 206 bp, respectively.

3.3. Analysis of HLA-DRB1*, -DQB1* alleles

HLA class II alleles (HLA-DRB1*, -DQB1*) were

determined according to established protocols (Olerup and

Zetterquist 1992; Olerup et al., 1993). For subtyping of

DRB1* genes allele specific primer mixes were used

(Dynal, Oslo, Norway). HLA alleles were classified accord-

ing to the nomenclature proposed by the World Health

Organization Nomenclature Commitee (Bodmer et al.,

1994). The frequency distribution of HLA-DRB1*, -DQB1*

alleles was compared to a previously published cohort of

174 healthy Caucasian subjects (Ferencik and Grosse-

Wilde, 1997).

3.4. Determination of anti-MOG antibodies

Recombinant human MOG corresponding to the extrac-

ellular IgG-like domain (rhMOGIgD) was expressed and

purified as previously outlined (Reindl et al., 1999; Haase

et al., 2001; Schmidt et al., 2001) followed by ion exchange

chromatography on a strong cationic membrane system

(Sartobind, Sartorius, Gottingen, Germany). Serum anti-

bodies directed against rhMOGIgD were determined by

immunoblotting as previously described (Reindl et al.,

1999; Haase et al., 2001; Schmidt et al., 2001).

3.5. Statistical analysis

Statistical analysis was performed with the SPSS software

package (SPSS, Chicago, IL, USA). Group comparisons

were calculated by one-way analysis of variance (ANOVA).

The presence of allelic polymorphisms was compared by

Pearson’s v2 test. Logistic regression was performed to test

for interactions between genotypes, phenotypes, HLA-

DRB1* alleles, MOG gene (TAAA)n polymorphism, IL-6

C allele, and anti-MOG-antibody responses.

All investigators performing the genetic analyses were

blinded with regard to the presence of anti-MOG antibodies

and the clinical phenotype of X-ALD patients.

4. Results

4.1. IL-6 vntr polymorphism

Genotyping of IL-6 vntr polymorphism was performed in

68 patients. The frequency distribution of IL-6 C allele in X-

ALD patients and healthy donors is given in Table 2. The

IL-6 C allele was equally distributed between X-ALD

patients and healthy controls. However, stratification of X-

ALD patients according to their clinical phenotypes

revealed that IL-6 C allele was more frequent in patients

with pure AMN as compared to other X-ALD phenotypes

and healthy controls. This difference, however, did not

reach statistical significance ( p = 0.081).

4.2. MOG gene (TAAA)n polymorphism

Genotyping of MOG gene (TAAA)n tetranucleotide

polymorphisms was performed in 68 patients. The fre-

quency distribution of MOG gene (TAAA)n alleles in X-

ALD patients is given in Table 3. The allelic distribution of

Table 2

Frequency and distribution of IL-6 C allele

C allele present C allele absent

All patients 26/68 (38.2%) 42/68 (61.8%)

Pure AMN 15/29 (51.7%) 14/29 (48.3%)

ALMN/AMN with

tract involvement

5/15 (33.3%) 10/15 (66.7%)

Female heterozygotes 5/16 (31.3%) 11/16 (68.7%)

CALD 1/4 (25%) 3/4 (75%)

Addison 0/2 2/2 (100%)

Asymptomatic 0/2 2/2 (100%)

Healthy donors 23/73 (31.5%) 50/73 (68.5%)

S. Schmidt et al. / Journal of Neuroimmunology 135 (2003) 148–153150

MOG gene (TAAA)n alleles was not significantly different

as compared to the previously published allelic distribu-

tions in healthy Caucasian controls (Malfroy et al., 1995)

with the 226, 222 and 218 bp alleles being the most

common allelic polymorphisms. Two patients were homo-

zygous for the 230 bp, two for the 226 bp, and two for the

218 bp allele. One patient was homozygous for the 222 bp,

and one for the 210 bp allele. There was no association of

MOG gene (TAAA)n alleles with distinct clinical pheno-

types.

4.3. HLA-DRB1*, -DQB1* alleles

The frequency distribution of HLA-DRB1* alleles is

given in Table 4. HLA-DRB1*04 alleles (32.9%) were most

common in the whole group of X-ALD patients. HLA-

DRB1*1501 (13.6%), -DRB1*0401 (12.9%), -DRB1*0101

(11.4%), and -DRB1*0301 (10.7%) as well as -DRB1*1301

(10.7%) were the most common single alleles. This

distribution was not statistically significantly different

from that of healthy Caucasian controls as previously

published (Ferencik and Grosse-Wilde, 1997). After exclu-

sion of heterozygote mothers, HLA-DRB1* frequency

distribution remained essentially the same with HLA-

DRB1*04 alleles occurring in 24.4% of patients. Notably,

HLA-DRB1*1401 alleles were underrepresented in X-ALD

patients ( p < 0.05).

As for the DQ-locus HLA-DQB1*0301 (23.7%), -DQB1*

0501 (12.6%), -DQB1*0602 (14.1%), and -DQB1*0603

(11.1%) were the most common single alleles.

There was no association of HLA-class II alleles with

clinical phenotypes.

4.4. Anti-MOG antibodies and their association with MOG

gene (TAAA)n polymorphism

The presence of anti-MOG antibodies was analyzed in 54

X-ALD patients. Anti-MOG antibodies were detectable in

17/54 sera (31.5%). Nine patients with anti-MOG antibodies

had AMN, two ALMN/AMN with tract involvement, four

were female heterozygotes, one had Addison’s disease and

one was asymptomatic.

While the presence of anti-MOG antibodies was neither

associated with clinical phenotypes, age of onset, duration

of disease, HLA-class II alleles or IL-6 C allele, a statisti-

cally significant association was established with the 226 bp

(TAAA)n MOG gene polymorphisms ( p < 0.05).

5. Discussion

The present study demonstrates no apparent association

of HLA class II alleles, IL-6 C allele and (TAAA)n MOG

gene polymorphisms with adult clinical phenotypes of X-

ALD. While no association of clinical phenotypes with

HLA alleles and (TAAA)n MOG gene polymorphisms

was detectable, the presence of serum anti-MOG anti-

bodies was associated with the 226 bp (TAAA)n MOG

gene polymorphism but not with distinct clinical pheno-

types.

5.1. Clinical considerations

The phenotypic heterogeneity of X-ALD makes the

existence of disease modifying factors or genes plausible.

Table 4

Frequency and distribution of HLA-DRB1* alleles in X-ALD patients

HLA- All patients Pure AMN ALMN/AMN tract Female heterozyg CALD Addison Asymptomatic

DRB1*01 17/70 (24.3%) 7/31 (22.6%) 4/15 (26.7%) 2/16 (12.5%) 3/4 (75%) 0/2 1/2 (50%)

DRB1*15 18/70 (25.7%) 11/31 (35.5%) 4/15 (26.7%) 2/16 (12.5%) 0/4 1/2 (50%) 0/2

DRB1*0301 13/70 (18.6%) 3/31 (9.7%) 3/15 (20%) 6/16 (37.5%) 1/4 (25%) 0/2 (%) 0/2

DRB1*04 23/70 (32.9%) 8/31 (34.8%) 7/15 (46.7%) 5/16 (31.3%) 1/4 (25%) 1/2 (50%) 1/2 (50%)

DRB1*0701 11/70 (15.7%) 6/31 (19.4%) 0/15 3/16 (18.8%) 1/4 (25%) 1/2 (50%) 0/2

DRB1*08 4/70 (5.7%) 2/31 (6.5%) 1/15 (6.7%) 1/16 (6.3%) 0/4 0/2 0/2

DRB1*0901 0/70 0/31 0/15 0/16 0/4 0/2 0/2

DRB1*11 14/70 (20%) 4/31 (12.9%) 6/15 (40%) 3/16 (18.8%) 1/4 (25%) 0/2 0/2

DRB1*01201 5/70 (7.1%) 4/31 (12.9%) 0/15 1/16 (6.3%) 0/4 0/2 0/2

DRB1*13 20/50 (28.6%) 11/31 (35.5%) 1/15 (6.7%) 6/16 (37.5%) 1/4 (25%) 1/2 (50%) 0/2

Table 3

Frequency and distribution of MOG gene (TAAA)n polymorphism

230 bp 226 bp 222 bp 218 bp 214 bp 210 bp

All patients 25/68 (36.8%) 29/68 (42.6%) 26/68 (38.2%) 33/68 (48.5%) 6/68 (8.8%) 9/68 (5.7%)

Pure AMN 12/30 (40%) 11/30 (36.7%) 12/30 (40%) 13/30 (43.3%) 1/30 (3.3%) 5/30 (16.7%)

ALMN/AMN tract 5/15 (33.3%) 8/15 (53.3%) 5/15 (33.3%) 8/15 (53.3%) 2/15 (13.3%) 1/15 (6.7%)

Female heterozygotes 5/16 (31.3%) 6/16 (37.5%) 8/16 (50%) 8/16 (50%) 2/16 (12.5%) 2/16 (12.5%)

CALD 2/3 (66.7%) 2/3 (66.7%) 0 1/3 (33.3%) 0 1/3 (33.3%)

Addison 1/2 (50%) 1/2 (50%) 0 2/2 (100%) 0 0

Asymptomatic 0 1/2 (50%) 1/2 (50%) 1/2 (50%) 1/2 (50%) 0

S. Schmidt et al. / Journal of Neuroimmunology 135 (2003) 148–153 151

Paradigmatically, CALD and AMN emerge as distinct

clinical phenotypes with widely different histopathological

features and prognostic implications in the same kindred

harboring identical mutations in the X-ALD gene (Moser,

1997). The complexity of clinical phenotype generation in

X-ALD is further highlighted by the existence of distinct

subtypes even within the AMN phenotype. Almost half of

AMN patients exhibit signs of cerebral demyelination on

MRI (Kumar et al., 1995). This finding is of prognostic

relevance since patients with ALMN progress more rapidly,

and may even convert into CALD (Moser, 1997; van Geel et

al., 2001). The question if AMN patients with tract involve-

ment represent a distinct subgroup of AMN patients or

merely a more advanced variant of pure AMN is currently

unsolved (Kumar et al., 1995; Moser, 1997).

5.2. IL-6 vntr polymorphism

Recently, several independent studies reported disease

modifying effects of genetic variations of IL-6. First, IL-6 C

allele is associated with delayed onset of Alzheimer’s

disease (Papassotiropoulos et al., 1999). Second, the same

allelic polymorphism is associated with a benign course of

MS in Sardinian patients (Vandenbroeck et al., 2000), and is

partially protective in patients with systemic lupus eryth-

ematodes (Linker-Israeli et al., 1999). This disease-modify-

ing effect of IL-6 C allele was further confirmed by a

previous study of our group demonstrating that IL-6 C allele

is underrepresented in MS patients with both primary and

secondary progressive MS (Schmidt et al., 2000). Here, we

demonstrated that IL-6 C allele is more frequent in X-ALD

patients with pure AMN although not reaching statistical

significance. However, while an overrepresentation of IL-6

allele would nicely fit into the context of the aforementioned

studies, a protective effect of IL-6 C allele on the course of

X-ALD is unlikely, since the frequency distribution of IL-6

C allele is almost identical in patients with ALMN/AMN

with tract involvement (i.e. patients with a less favorable

prognosis) and healthy donors. Moreover, in contrast to SLE

and MS, the pathogenetic role of IL-6 in X-ALD has not

been established.

5.3. HLA-DRB1*, -DQB1* alleles

MS, the most common demyelinating disease of the

CNS, is associated with the HLA alleles DRB1*1501 and

DQB1*0602 in northern Europeans (Olerup and Hillert,

1991). The histopathological similarities of demyelinating

lesions in MS and cerebral forms of X-ALD (Moser, 1997)

gave rise to speculations that X-ALD might also be asso-

ciated with certain HLA class II alleles. However, previous

studies investigating this association yielded conflicting

results. Whilst an association of HLA-DRB1*16 alleles

was reported in a group of 29 X-ALD patients (Berger et

al., 1995), this finding was not confirmed in a larger study

of 83 X-ALD (McGuiness et al., 1997). Similarly, the

present study encompassing 70 X-ALD patients failed to

demonstrate an association of certain HLA-DR and -DQ

alleles with X-ALD in general and clinical phenotypes in

particular. Moreover, no association between HLA alleles

and serum anti-MOG antibody responses was detectable.

5.4. Anti-MOG antibodies and their association with MOG

gene (TAAA)n polymorphism

The human MOG gene is located within the MHC region

on chromosome 6p21.3–p22 (Pham-Dinh et al., 1995).

Naturally occurring polymorphisms of the MOG gene were

first identified in exon 1 as a trinucleotide (CTC)n and in the

3V flanking region (exon 8) as a tetranucleotide (TAAA)nrepeat motif (Roth et al., 1995b). Most recently, four novel

sequence variations of the MOG gene have been identified

(Gomez-Lira et al., 2000). The external localization of

MOG in the myelin sheath, its restriction to the CNS, and

the accumulated experimental evidence that cellular and

humoral immune responses directed against MOG contrib-

ute to myelin breakdown in MS (Raine et al., 1999; Stefferl

et al., 2000) make this protein an attractive candidate as an

antigenic target not only in MS but also in cerebral forms of

X-ALD. Based on this assumption, we previously inves-

tigated anti-MOG antibody responses in serum of X-ALD

patients (Schmidt et al., 2001). In the present study, MOG-

specific autoantibodies were detectable in 17/54 patients

(31.5%) confirming our previous findings that MOG-spe-

cific antibodies in X-ALD patients are detectable in a

frequency comparable to that in MS patients (Reindl et

al., 1999; Haase et al., 2001; Schmidt et al., 2001). In the

latter study, none of the 13 AMN patients, i.e. those patients

without evidence of CNS demyelination, had anti-MOG

antibodies suggesting that the presence of these antibodies

might be indicative of CNS demyelination (Schmidt et al.,

2001). However, in the larger group of 54 patients analyzed

here, anti-MOG antibodies were also detectable in patients

with pure AMN. Moreover, no association of the anti-MOG

antibody responses with any of the clinical phenotypes was

observed arguing against a discriminatory function of anti-

MOG antibodies in phenotype generation.

Since previous investigations indicated that human MOG

exists in alternatively spliced isoforms (Pham-Dinh et al.,

1995), it is tempting to speculate that minor differences in

MOG composition caused by naturally occurring polymor-

phisms might also influence the antigenicity and immune

responses against MOG. However, previous studies in MS

patients investigating a possible role of allelic variations of

the MOG gene with regard to disease susceptibility for MS

yielded negative results (Roth et al., 1995a). A recent

genetic association study revealed equal distribution of six

(including four novel) MOG gene polymorphisms in both

X-ALD patients and healthy controls (Gomez-Lira et al.,

2000). Here, we investigated the possible association of a

naturally occurring tetranucleotide (TAAA)n polymorphism

in the 3V flanking region of the MOG gene not included in

S. Schmidt et al. / Journal of Neuroimmunology 135 (2003) 148–153152

the aforementioned study with anti-MOG antibody

responses in serum. While we found a very similar fre-

quency distribution of MOG gene (TAAA)n alleles in X-

ALD patients as compared to previously published healthy

Caucasian controls (Malfroy et al., 1995), the 226 bp MOG

gene (TAAA)n polymorphism was associated with the

presence of anti-MOG antibodies in serum. However, due

to the lack of association with clinical phenotypes in X-

ALD, the clinical relevance of this finding remains to be

established.

References

Aubourg, P., Dubois-Dalcq, M., 2000. X-linked adrenoleukodystrophy

enigma: how does the ALD peroxisomal transporter mutation affect

CNS glia? Glia 29, 186–190.

Berger, J., Bernheimer, H., Fae, I., Braun, A., Roscher, A., Molzer, B.,

Fischer, G., 1995. Association of X-linked adrenoleukodystrophy

with HLA DRB1 alleles. Biochem. Biophys. Res. Commun. 216,

447–451.

Bodmer, J.G., Marsh, S.G., Albert, E.D., et. al., 1994. Nomenclature of the

factors of the HLA. Eur. J. Immunogenet. 21, 485–517.

Bowcock, A.M., Ray, A., Ehrlich, H., Sehgal, P.B., 1989. Rapid detection

and sequencing of alleles in the 3Vflanking region of the interleukin-6

gene. Nucleic Acids Res. 17, 6855–6864.

Ferencik, S., Grosse-Wilde, H., 1997. Caucasian German normal. In: Ter-

asaki, P.I., Gjertson, D.W. (Eds.), HLA-1997. UCLA Tissue Typing

Laboratory, CA, USA, pp. 222–223.

Gomez-Lira, M., Marzari, M.G., Uziel, G., Pignatti, P., Rizzuto, N.,

Salviati, A., 2000. Myelin oligodendrocyte glycoprotein (MOG)

polymorphisms and adrenoleukodystrophy. J. Neuroimmunol. 111,

245–247.

Haase, C.G., Guggenmoos, J., Brehm, U., Andersson, M., Olssom, T.,

Reindl, M., Schneidewind, J.M., Zettl, U.K., Heidenreich, F., Berger,

T., Wekerle, H., Hohlfeld, R., Linington, C., 2001. The fine specificity

of the myelin oligodendrocyte glycoprotein autoantibody response in

patients with multiple sclerosis and healthy controls. J. Neuroimmunol.

114, 220–225.

Heun, R., Papassotiropoulos, A., Jessen, F., Maier, W., Breitner, J.C., 2001.

A family study of Alzheimer disease and early- and late-onset depres-

sion in elderly patients. Arch. Gen. Psychiatry 58, 190–196.

Kumar, A.J., Kohler, W., Kruse, B., Naidu, S., Bergin, A., Edwin, A.,

Moser, H.W., 1995. MR findings in adult-onset adrenoleukodystrophy.

Am. J. Neuroradiol. 16, 1227–1237.

Linker-Israeli, M., Wallace, D.J., Prehn, J., Michael, D., Honda, M., Taylor,

K.D., Paul-Labrador, M., Fischel-Ghodsian, N., Fraser, P.A., Klinen-

berg, J.R., 1999. Association of IL-6 gene alleles with systemic lupus

erythematodes (SLE) and with elevated IL-6 expression. Genes Immun.

1, 45–52.

Malfroy, L., Viratelle, C., Coppin, H., Borot, N., Roth, M.P., 1995.

Polymorphic tri- and tetranucleotide repeats in exons 1 and 8 of the

myelin oligodendrocyte glycoprotein (MOG) gene. Hum. Genet. 96,

737–738.

McGuiness, M.C., Powers, J.M., Bias, W.B., Schmeckpeper, B.J., Se-

gal, A.H., Gowda, V.C., Wesselingh, S.L., Berger, J., Griffin, D.E.,

Smith, K.D., 1997. Human leukocyte antigens and cytokine expres-

sion in cerebral inflammatory demyelinative lesions of X-linked

adrenoleukodystrophy and multiple sclerosis. J. Neuroimmunol. 75,

174–182.

Moser, H.W., 1997. Adrenoleukodystrophy: phenotype, genetics, patho-

genesis and therapy. Brain 120, 1485–1508.

Olerup, O., Hillert, J., 1991. HLA class II-associated susceptibility in multi-

ple sclerosis: a critical review. Tissue Antigens 38, 1–15.

Olerup, O., Zetterquist, H., 1992. HLA-DR typing by PCR amplification

with sequence-specific primers in 2 hours: an alternative to serological

DR typing in clinical practice including donor– recipient matching in

cadaveric transplantation. Tissue Antigens 39, 225–235.

Olerup, O., Aldener, A., Fogdell, A., 1993. HLA-DQB1 and HLA-DQA1

typing by PCR amplification with sequence specific primers in 2 hours.

Tissue Antigens 41, 119–134.

Papassotiropoulos, A., Bagli, M., Jessen, F., Bayer, A.T., Maier, W., Rao,

M.L., Heun, R., 1999. A genetic variation of the inflammatory cytokine

interleukin-6 delays the initial onset and reduces the risk for sporadic

Alzheimer’s disease. Ann. Neurol. 45, 666–668.

Pham-Dinh, D., Della Gaspera, B., Kerlero de Rosbo, N., Dautigny,

N., 1995. Structure of the human myelin/oligodendrocyte glycopro-

tein gene and multiple alternative spliced isoforms. Genomics 29,

345–352.

Raine, C.S., Cannela, B., Hauser, S.L., Genain, C.P., 1999. Demyelination

in primate autoimmune encephalomyelitis and acute multiple sclerosis

lesions: a case for antigen-specific antibody mediation. Ann. Neurol.

46, 144–160.

Reindl, M., Linington, C., Brehm, U., Egg, R., Dilitz, E., Deisenhammer,

F., Poewe, W., Berger, T., 1999. Antibodies against the myelin oligo-

dendrocyte glycoprotein and the myelin basic protein in multiple scle-

rosis and other neurological diseases: a comparative study. Brain 122,

2047–2056.

Roth, M., Dolbois, L., Borot, N., Pontarotti, P., Clanet, M., Coppin, H.,

1995a. Myelin oligodendrocyte glycoprotein (MOG) gene polymor-

phisms and multiple sclerosis: no evidence of disease association with

MOG. J. Neuroimmunol. 61, 117–122.

Roth, M.P., Malfroy, L., Offer, C., Sevin, J., Enault, G., Borot, N., Pontar-

otti, P., Coppin, H., 1995b. The human myelin oligodendrocyte (MOG)

gene: complete nucleotide sequence and structural characterization. Ge-

nomics 28, 241–250.

Schmidt, S., Papassotiropoulos, A., Bagli, M., Harzheim, M., Heun, R.,

Klockgether, T., 2000. No association of serum levels of interleukin-6

and its soluble receptor components with a genetic variation in the

3Vflanking region of the interleukin-6 gene in patients with multiple

sclerosis. Neurosci. Lett. 294, 139–142.

Schmidt, S., Haase, C.G., Bezman, L., Moser, H., Schmidt, M., Kohler, W.,

Linington, C., Klockgether, T., 2001. Serum autoantibody responses to

myelin oligodendrocyte glycoprotein and myelin basic protein in X-

linked adrenoleukodystrophy and multiple sclerosis. J. Neuroimmunol.

119, 88–94.

Stefferl, A., Brehm, U., Linington, C., 2000. The myelin oligodendrocyte

glycoprotein (MOG): a model for antibody-mediated demyelination in

experimental autoimmune encephalomyelitis and multiple sclerosis.

J. Neural Transm., Suppl. 58, 123–133.

Vandenbroeck, K., Fiten, P., Ronsse, I., Goris, A., Porru, I., Melis, C.,

Rolesu, M., Billiau, A., Marrosu, M.G., Odenakker, G., 2000. High-

resolution analysis of IL-6 minisatellite polymorphism in Sardinian

multiple sclerosis: effect on course and onset of disease. Genes Immun.

1, 460–463.

Van Geel, B.M., Bezman, L., Loes, D.J., Moser, H.W., Raymond, G.V.,

2001. Evolution of phenotypes in adult male patients with X-linked

adrenoleukodystrophy. Ann. Neurol. 49, 186–194.

S. Schmidt et al. / Journal of Neuroimmunology 135 (2003) 148–153 153