Genetic susceptibility to multiple sclerosis.

Genetic Susceptibility to Multiple Sclerosis
David G. Haegert,MD,X and Maria G. Marrosu, M D t
Twin studies have established that susceptibility to multiple sclerosis (MS)
is partly genetic. Attempts to identify the
relevant genetic loci have involved population-basedstudies, to detect associationsbetween a genetic marker and MS,
and family studies, to detect linkage between a putative marker and MS. Most of this genetic work is driven by the
view that MS is an autoimmune disease. Thus, the focus has been mainly on genes known to be important in the
immune response: human leukocyte antigen (HLA) genes, T-cell receptor genes, and immunoglobulin genes. To date,
only the particular HLA-DR2 haplotype that is common in Caucasians can be concluded to be important in MS
susceptibility in most populations. Studies of other genetic loci have been few, the data obtained often have been
conflicting or controversial, and further studies are needed to clarify the biological significance of these loci in MS.
Recommendatioas for further studies are provided in order to overcome some of the problems that have plagued
earlier work in MS such as nonreproducibility of results.
Haegen DG, Marrosu MG. Genetic susceptibility to multiple sclerosis.
Ann Neurol 1994;36(S2):S204-S2 10
Epidemiology, Molecular Genetics, and
Multiple Sclerosis
To understand the geographical distribution of multiple sclerosis (MS) we need accurate data on the genetic
composition of the different populations in which MS
occus and data on the distribution of putative susceptibility alleles within the constituent ethnic groups. Epidemiological and molecular genetic studies must be
performed together if we are to unravel the complexities of genetic susceptibility to MS. This principle is
exemplified by brief consideration of two earlier findings: first, that the HLA-DR2 serological specificity
lowers the threshold for susceptibility to MS 111; and
second, that there is a north-south gradient in Europe
of HLA-DR2 12). Do these two observations explain
in part ethnic differences in MS susceptibility within
Europe? HLA serological analysis indicates that both
Scandinavians f3) and Hungarian Gypsies [4] have a
hgh population frequency of DR2, yet only the former
population has a hgh prevalence of MS 14, 5 ) . The
development and application of molecular genetic or
DNA-based typing methods have shown clearly that
the DR2 serological specificity is expressed on the
products of at least five different DR2 alleles 161. One
DR2 haplotype is common in Scandinavians CS] and is
MS associated in various population-based studies in
Scandinavia [ S , 7-10) and elsewhere [lo), whereas a
different DR2 haplotype is present in the Hungarian
Gypsy population 111). This example illustrates two
points: First, genetic studies in MS should be molecular
in type and should avoid surrogate markers such as
serological specificities, which often do not provide an
adequate measure of the polymorphisms present at a
particular locus and may lead to incorrect conclusions
concerning genetic associations with MS. Second, molecular genetics can complement epidemiology and
provide precise data on the distribution and diversity
of susceptibility alleles within and between populations
and facilitate identification of genetic clines that could
explain regional differences in MS prevalence within a
population.
From the ‘Discipline of Pathology, Memorial University of Newfoundland, St. John’s, Newfoundlaad, Canada, and the tISUNt0 di
Nempsichiatria Infantile, Universiddegli Studi di Cagliari, Cagliari,
Italy.
Address correspondence to Dr Haegert, Discipline of Pathology,
Health Science Center, Memorial University of Newfoundland,
St. John’s, Newfoundland, Canada AIB 3V6.
Genetic Susceptibility to Multiple Sclerosis:
An Introduction
Twin studies [12-17) provide evidence that two or
more genes play a role in susceptibility to MS. It remains controversial, however, as to whether we have
identified the major genetic locus in this disease [lS,
191. Also unresolved are questions of additive and hierarchical effects of different genes (see E201 for review) and whether there are both susceptibility and
protective MS genes fS, 21, 22).
Most of the current work on genetic susceptibility
to MS is driven by the hypothesis that MS is an autoimmune disease 123). Therefore, investigators have focused on genetic loci recognized to play a major role in
the immune response, that is, human leukocyte antigen
(HLA) genes IS, 7-10), T-cell receptor (TCR)genes
(for review see [24]), and immunoglobulin (Ig) genes
S204 Copyright 0 1994 by the American Neurological Association
[25,26}. Since myelin basic protein (MBP) genes may
encode the target molecules of the autoimmune process in MS 1271, some investigators [28, 271 have also
focused on MBP gene polymorphisms.
The contribution of candidate disease marker loci to
MS susceptibility has been evaluated by two complehave been dementary approaches. First, a~~ociations
tected between particular genetic markers and MS in
PoPulation-based studies. Most Co-onlY
this is done
using the case-control method, but an important caveat
is that small differences in ethnicity or regional genetic
differences can influence sgdica&Iy the-experimental
findings with this method. An alternative method that
controls for ethnicity and regional genetic differences
is to use parental haplotypes not transmitted to patients
as control haplotypes {30, 31). Second, and an important complement to association studies, is the identification of linkage between a particular marker and MS.
This is essential since a population association between
a genetic marker and MS could reflect linkage or population stratification [311. The standard linkage method
in MS is to search for haplotype sharing in identity by
descent analysis of affected sibling pairs [e.g., 32-34}.
When a given locus makes a minor contribution to
disease susceptibility, however, identity by descent
analysis may not detect linkage and a more powerful
linkage test, the transmission test for linkage disequilibrium (TDT) [31), may be used. This latter test has
the advantage of applicability to simplex as well as multiplex families. The TDT may be used to detect linkage
only when a population association with a marker allele
or haplotype has already been identified. For the future
we anticipate that the significance of positive associaill be
tions of marker alleles or haplotypes with MS w
tested by the TDT.
HLA and Multiple Sclerosis
HLA Background
Early HLA-MS case-control studies demonstrated a
positive association with =-A3
and -B7 serological
specificities. Subsequent work established that these
associations were secondary to a stronger disease association with HLA-DR2 (for review see {lo]).
HLA-DR2 is a serologically defined specificity that
maps to the HLA-D or class I1 region (Fig). This MS
association has been confirmed in many population
studies 15, 7-10, 351. Because of this finding and the
evidence that experimental allergic encephalomyelitis
(EAE) 1361 and many other autoimmune diseases are
associated with MHC class I1 1371, most HLA studies
over the past 10 years in MS have focused on the class
I1 region. It is necessary to have a basic knowledge
of this region in order to understand the investigative
strategies and the significance of the experimental
of the H u gene compk.x
A simp[z$ed
shows the transcriptionally active HLA c h s I1 or D region
genes; psetldogenes r e f w d to in the text are in parentheses and
the location o f c h s III and I genes rekztive to claSs II genes is
shoton.
findings. As shown in the Figure, the HLA class I1
region is divided into three major subregions termed
DP, DQ, and DR. Each subregion contains at least one
pair of expressed genes that encodes a pair of functional a and P chains. For example, the DQ region
contains DQAl and DQB1 genes, which code for
DQa and P chains respectively. The a and P chains
typically are expressed on various cells involved in the
immune response. Both a and P chains contain two
external domains; the membrane distal domain is the
major site of variation between class I1 molecules.
Whereas DRa chains (encoded by DRA) are essentially invariant, DRP, DQa, DQP, DPa, and DPP
show sequence variants that are responsible for the
polymorphisms characteristic of these molecules. This
polymorphism is determined by the germline genetic polymorphism of the class I1 region with multiple alleles at most class I1 loci (reviewed in E371).
The nomenclature of the HLA alleles follows that of
the Human Gene Mapping Nomenclature Committee
1383. However, the terminology is confusing. This is
because each allele is now defined on the basis of its
nucleotide sequence [39], but its original name was
based on its serological specificity 1401. Thus, originally, D R alleles were termed DR1-DR10, with some
alleles being given an International HLA Workshop or
w designation (e.g., DRw6). Later, as new serological
reagents became available, some of the serologically
defined specificities were split into several specificities.
DRw5 was therefore split into D R w l l and DRwl2
and DRw6 was split into DRwl3 and DRwl4. The
next specificity to be split, DR2, was divided into
DRwl5 and w l 6 specificities. When DR2 16, 37-41]
alleles were officially named for the first time by
the World Health Organization (WHO) nomenclature
committee [391, the first two letters were derived from
the serological specificities. Thus, DR2 alleles are now
termed DRB1*1501, '1502, *1503, '1601, and *1602
163. DRBl refers to the gene locus encoding DRP
chains, and the allelic numbers that follow the asterisk
are named for the respective splits; for example, '1501,
*1502,and *1503 are named for the DRw/ 15 split. Simi-
Haegert and Marrosu: Genetic Susceptibility to MS
S205
larly, DQ specificitieswere initially termed DQwl, w2,
and w3, and DQw4 was added later. DQwl was subsequently split into w5 and w6 and so DQwb-derived
DQBl alleles are termed DQB1'0601, "0602, and so
on {6, 39-4 11. In summary, there is a clear logic to the
HLA class I1 nomenclature but it is not immediately
obvious on first exposure except to those who have
monitored closely the changes in the nomenclature of
the HLA system over the years.
This nomenclature was emphasized at the workshop
for several reasons. First, to establish the genetic basis
of susceptibility to MS we must identify the precise
disease-associated HLA class I1 alleles rather than identify surrogate serological markers. Second, since a serological specificity such as DR2 is shared by the products of a number of different alleles, we need to be
able to distinguish between these different alleles to
identdy the specific DR2 allele associated with MS.
Third, definition of the alleles by nucleotide sequence
means that precise detection of MS-associated alleles
must use DNA-based typing methods. We anticipate
that future HLA MS studies will only use methods
designed to distinguish between DNA sequences, for
example, polymerase chain reaction amplification and
sequence-specific oligonucleotide typing (PCR-SSO
typing).
An important feature of the HLA class I1 region is
that certain DR and DQ alleles are in strong linkage
disequilibrium, that is, inherited together on a chromosome as a single genetic unit termed a haplotype (haploid genotype) 137). This means that detection of a
positive disease association with a given allele may occur because the particular allele detected is in linkage
disequiltbrium with one or more other alleles that are
primarily responsible for the disease association. Moreover, certain alleles are present on more than one haplotype; for example, DQA1'0102 is present on an MSassociated haplotype and a second haplotype that is
not MS associated [351. Therefore, the current strategy
[37) in studying HLA class I1 associations with autoimmune disease involves first, definition of the diseaseassociated haplotype; second, identification of the disease susceptibility allele; and third, determination of
the structural basis of the allelic association.
HLA Class I1 Associations with Multiple Sclerosis
Population-based case-control studies 15, 7-10, 351
have used serological or DNA-based HLA class I1 typing to detect MS associations. Methodology is mentioned before discussing findings because the reliability
of the data obtained is partly dependent on the methodology used. Thus, early HLA MS studies were serological, but interpretation of the results is difficult. This
is because serological typing has a significant error rate
1421 and is unable to idenufy products of all alleles
S206
(e.g., DR2 subtypes). Any reports based on serology
and, therefore, virtually all HLA studies before 1985,
are of uncertain significance. At present we believe that
serology should be used only in preliminary studies
and must be followed by molecular typing.
Numerous case-control studies show HLA class I1
associations with MS in most populations. There is
considerable evidence 15, 7-10, 351 that the common
Caucasian DR2 haplotype DRB l"150 1-DQAl"0102DQB l"0602 is the dominant MS-associated haplotype worldwide. Particularly strong associations with
this haplotype are observed in northern European MS
patients and in patients of northern European descent
[lo}, but an MS association with this haplotype also is
observed in other populations f43-451. At the present
time we cannot be certain as to whether all populations
show HLA-MS associations, as studies detecting no associations have often been serological or have used
small numbers of patients and control subjects.
Heterogeneity in HLA associations with MS between ethnic groups is also controversial 143,461. On
the one hand a collaborative French CanadianSardinian study 1351 showed a positive association in
French Canadian MS patients with the common Caucasian DR2 haplotype, but completely different associations in Sardinian MS patients, specifically MS associations with DR4 and DQA1'0301, DQBl"0302, and
"0201 (see also 1471). Additional evidence for heterogeneity is the MS association with particular DPBl
alleles in Caucasoid and Chinese MS patients who lack
the DR2 haplotype f48}. Conversely, Olerup and Hillert [ 5 , 431 argue that the only haplotype repeatedly
shown to be MS associated is the DR2 haplotype and
that other associations are doubdul. Indeed, a common
feature of MS case-control studies is that initial findings
of HLA-MS associations cannot be duplicated when
the studies are repeated in the same or different ethnic
groups. For example, weak associations of MS with
DR4 have been reported by two groups 149, SO}, but
importantly not by a third group 151 that directly addressed the questions of a DR4 association with MS.
Furthermore, a Swedish study 181 reported an association of a DR4 haplotype with primary chronicprogressive MS, but a second Swedish study 151 was
nonconfirmatory. Similarly, evidence that a DR3 haplotype is associated with relapsing-remitting MS, was
obtained [ 5 , 81, although this findmg has been challenged by others f5 I}. Hierarchical associations of two
DQBl alleles ('0302 and "0201) with MS have been
documented in Sardinia C353, but not tested for in s u b
sequent studies. Very recently C22) three haplotypes
were reported to be negatively associated with MS.
In view of the difficulties in repeatability in HLA MS
studies, it would be prudent to perform additional casecontrol studies to address the possibility that apart
from the wellestablished DR2-haplotype association,
Annals of Neurology Supplement 2 to Volume 36, 1994
there are very few positive or negative HLA associations with MS.
HLA Class II Associations with Multiple Sclerosis:
Theoretical Modeh
The finding of class I1 associations with MS has led to
various models, some of which have been vigorously
analyzed. The models are as follows: (1) The MSassociated DR2 haplotype (DRB1*1501-DQA1*0102DQB l"0602) includes the disease susceptibility gene
(DSG). (2) The DSG is a non-HLA gene linked to the
DR2 haplotype. Candidate genes include TAP (transporter associated with antigen processing) and LMP
(large multifunctional protease) genes, but evidence
against a role for both sets of genes has been obtained
[52] (D.A.S. Compston, unpublished observations,
1994). In this context of non-HLA genes, Hillert and
Olerup analyzed polymorphsms centromeric and telomeric to DR-DQ loci in MS patients. They reported
that any non-HLA gene associated with MS must be
close to the DR and DQ loci, as the DR2 haplotype
does not extend to DP or DQA2 loci or to HLA class
I11 loci [9, 531 (see Fig). (3) HLA class I1 associations
with MS in different ethnic groups are not primarily
due to sharing of particular HLA class I1 haplotypes or
alleles, but are due to sharing of polymorphic HLA
class I1 sequences or codons. Evidence has been obtained in support of a primary association of MS with
a shared DQBl sequence 17, 541 and a DQAl codon
1547. However, formal two-locus linkage analysis 15,
551 establishes that these DQBl and DQAl associations with MS are secondary to an MS association with
the DR2 haplotype. Two possible exceptions exist:
First, Haegert and Francis [561 reported a primary MS
association in French Canadians with DQBl codons
for leucine at residue 26 and second, Allen and coauthors [221 reported a primary MS association in Sweden of DRB1 codons for glycine at residue 86. In view
of the history of MS case-control studies, confirmatory
investigations are needed if we are to accept either
codon as having an important role in MS susceptibility.
(4) DQa-p heterodimers encoded in cis or in trans may
determine MS susceptibility 1541. When this model
was formally tested by excluding the DR2 haplotype
from the analysis [5J, little evidence was obtained to
support the model. (5) MS is associated with mutated
HLA class I1 alleles. In keeping with the observations
in other autoimmune diseases [37,57], DNA sequencing [58J indicates that the HLA class I1 alleles in MS
patients are normal.
In summary, two plausible explanations persist for
HLA associations with MS. Either the HLA class I1
region contains a DSG or the DSG is a non-HLA gene
linked to the DR2 haplotype and close to the DR-DQ
loci. Unfortunately, the evidence at present does not
distinguish between these possibilities.
HLA-Multiple Sclerosis Linkage Studiu
The population-based case-control studies suggest that
tight linkage will be detected between the DR2 haplotype and MS.Although a recent study of nine Swedish
multiplex families suggests linkage 1591, two larger
studies of affected sibling pairs using identity by descent analysis [32} (D.A.S. Compston, unpublished
data, 1994) found no evidence of linkage. An explanation for these negative findings is difficult in view of
the consistent demonstration of MS associations with
this haplotype. One possibility is that linkage WIU be
detected only in populations where the MS-DR2 haplotype associations are particularly strong. A second
possibility is that insufficient sibling pairs were used in
any one study to detect linkage, and in that regard the
more powerful TDT may be appropriate, especially
since the prerequisite population associations for TDT
analysis have been already identified.
T-cell Receptor Gene Polymorphism and
Multiple Sclerosis
In the majority of T cells the antigen receptor consists
of a heterodimer encoded by TCR a and p genes. In
the context of the autoimmune hypothesis, the TCR
gene region may play a role in genetic susceptibility to
MS. Interest in this concept derives in part from studies of EAE in which there is strong evidence that MBP
recognition is restricted to very few TCR gene families
[36}. Therefore, it may be postulated that in MS, inheritance of certain TCR alleles or haplotypes could
predispose to myelin damage. The approach has involved identification of various TCR a and p polymorphsms by restriction fragment length polymorphism
(RFLP) analysis. Similar to the HLA studies, the TCR
studies include analysis of population associations and
of linkages in families, and both constant region and
variable region TCR polymorphisms have been the focus of study.
T C R a and p Restriction Fragment Length
Polymorphism and Multiple Sclerosis: The Findings
First, associations with TCR p RFLPs have been detected by some investigators 160-621 but not by others
163,641; the finding of an association between a TCR
a RFLP and MS [651 in one study has been refuted in
a second study 1661 in which the particular RFLP was
interpreted as a restriction enzyme artifact. Second,
linkage studies give conflicting findings. Thus, some
studies [67-691 reported that neither TCR a nor TCR
p RFLPs segregate with MS in families, but another
study [ 3 3 } of affected sibling pairs found good evidence for TCR p haplotype sharing in MS.
The reason for the discrepant findings is not entirely
evident but deserves discussion. First, and an important
consideration, polymorphic restriction sites that define
the RFLPs are most commonly located in noncoding
Haegert and Marrosu: Genetic Susceptibility to MS
S207
regions and may not be in strong linkage disequilib
rium with disease-susceptibility TCR alleles. Second,
there is a high rate of recombination among TCR
genes, and thus there may be a little conservation of
haplotypes from one generation to the next; that is,
linkage could be present but go undetected. Third, if
there is more than one HLA susceptibility allele or
haplotype, TCR genes involved in MS susceptibility
could vary because TCR gene products recognize antigen in the context of particular HLA molecules (reviewed in {70J).
In summary, to date there is no evidence that TCR a
gene polymorphisms play a role in genetic susceptibility to MS, but at least some evidence supports a role
of TCR f3 genes in MS development. Clearly further
linkage studies are indicated.
Minor Areas of Study
The studies of Ig gene polymorphisms in MS are few
and the data are conflicting. Thus, a population-based
study 125J found no association with Ig constant region
gene polymorphisms, but a family study reported that
Ig variable region gene polymorphisms and MS are
linked [26}. In view of all the claims of genetic predisposition to MS that have been subsequently refuted,
further Ig gene studies are needed to establish the precise role of these particular genes in MS susceptibility.
MBP flanking region gene polymorphisms were initially reported to be associated with and linked to MS
in a Finnish study [28], but no evidence of linkage was
obtained in a study by a second group of investigators
[29]. As studies are few in this area, further data are
needed before any definite conclusions can be reached.
Conclusions and Recommendations for
Fume Studies
Twin studies indicate that two or more genes are involved in the susceptibility to MS, but at present we
can only be confident that the HLA gene region, and
in particular the common Caucasian HLA-DR2haplotype, plays a role in MS susceptibility in most populations. Even with the application of molecular methods,
the biological significance of other gene regions in MS
remains unclear or controversial, with the possible exception of the TCR f3 gene region. Several recommendations that will affect further studies in this particularly difficult area of investigation can be made:
1. Nonreproducibility of HLA population-based association studies is a recurrent issue and probably reflects inadequate patient-control matching. Future
studies demand particular care to control for regional genetic differences; affected family-based
control subjects are particularly important.
HLA typing must be done by DNA-based methods, and serological findings remain unproved unless confirmed by molecular typing. Since the terminology for HLA alleles is now based on the
nucleotide sequence of the alleles, HLA MS studies
should use t h ~ saccepted terminology.
More powerful methods are indicated to detect
weak associations and linkage including the relative
predispositional effect (WE)method for associations C71) and the TDT 1311. The RPE method
is particularly useful in detecting weak associations
because it determines the predispositional, neutral,
or protective effects of alleles relative to one another, even when one allele is strongly disease associated.
A study reporting an MS association must be repeated in a second case-control study. Large casecontrol numbers are needed to detect weak associations, and cases need to be stratified according to
disease subtype. In linkage studies (e.g., for TCR)
stratification according to DR haplotype and subtype of MS may be important.
Studies in populations with a low MS prevalence
and where HLA class I1 linkages are non-Caucasian
in type may provide new insights into genetic susceptibility in high-prevalence areas. Thus, the precise alleles in an HLA haplotype responsible for MS
associations may someday be identified.
References
1. Compston DAS. Genetic susceptibility to multiple sclerosis.
In: Matthews WB, Compston A, Allen IV, Martyn CN, eds.
McAlpine’s multiple sclerosis. 2nd ed. Edinburgh: ChurchillLivingstone, 1991:301-3 19
2. Bodmer JG, Kennedy LJ, Lindsay J, Wasik AM. Applications
of serology and the ethnic distribution of three locus HLA haplotypes. Br Med Bull 1987;43:94-121
3. Lindblom B, Svejgaard A. HLA genes and haplotypes in the
Scandinavian populations. In: Tsuji K, Aizawa M, Sasazuki T,
eds. HLA 1991, vol 1. Oxford: Oxford University Press, 1992:
65 1-655
4. Gy6di E, Benczur M, P M y G, et al. Association between HLA
B7, DR2 and dysfunction of natural- and antibody-mediated
cytotoxicity without connection with deficient interferon production in multiple sclerosis. Hum Immunol 1982;4:209-217
5. Olerup 0,Hillen J. HLA class 11-associated genetic susceptibility in multiple sclerosis: a critical evaluation. Tissue Antigens
1991;38:1-15
6. Bodmer JG, Marsh SGE, Albert ED, et al. Nomenclature for
factors of the HLA system, 1991. In: Tsuji K, Aitawa M, Sasazuki T, eds. HLA 1991,vol 1. Oxford: Oxford University Press,
1992:17-3 1
7. Vandal F, Sollid LM, Vandvik B, et al. Patients with multiple
sclerosis carry DQB 1 genes which encode shared polymorphic
amino acid sequences. Hum Immunol 1989;25:103-110
8. Olerup 0, Hillen J, Fredriksen S, et al. Primarily chronic progressive and relapsing/remittingmultiple sclerosis:two immunogenetically distinct entities. Proc Natl Acad Sci USA 1989;86
7113-71 17
9. Olerup 0, Hillert J, Fredrikson S. HLA-D region-associated
S208 Annals of Neurology Supplement 2 to Volume 36, 1994
MS-susceptibility genes may be located telomeric to the
HLA-DP subregion. Tissue Antigens 1990;36:37-39
10. TiwariJL, Terasaki PI. Multiple sclerosis. In: TiwariJL, Terasaki
PI, eds. HLA and disease associations. New York Springer,
1985~152-167
11. KAlmAn B, Takacs K, Gy6di E, et al. Sclerosis multiplex in
Gypsies. Acta Neurol Scand 1991;84:181-185
12. Ebers GC, Bulman DE, Sadovnick AD, et al. A populationbased twin study in multiple sclerosis. N Engi J Med 1986315:
1638-1642
13. Heltberg A, Holm NV. Concordance in twins and recurrence
in sibships in multiple sclerosis. Lancet 1982;1:1068
14. Kinnunen E, Juntunen J, Ketonen L,et al. Genetic susceptibility
to multiple sclerosis: a co-twin study of a nationwide series.
Arch Neurol 1988;45:1108-1111
15. Bobowick AR, Kurtzke JF, Brody A, et al. Twin study of multiple sclerosis: an epidemiologic inquiry. Neurology 1978;28:
978-987
16. Sadovnick AD, Armstrong H, Rice GPA, et al. A populationbased study of multiple sclerosis in twins: update. Ann Neurol
1993;33:281-285
17. Mumford CJ, Wood NW, Keb-Wood H, et al. The British
Isles survey of multiple sclerosis in twins. Neurology 1994;44:
11-15
18. Moller E, Bohme J, Valugerdi MA, et al. Speculationson mechanisms of HLA associations with autoimmune diseases and the
specificity of “autoreacave”T lymphocytes. Immunol Rev 1930;
118:5-19
19. Ebers GC. Infections and demyelinating disease: editorial overview. Curr Opin Neurol Neurosurg 1992;5:173-174
20. Compston A, Sadovnick AD. Epidemiology and genetics of
multiple sclerosis. Curr Opin Neurol Neurosurg 1992;5:175181
21. Haegert DG, Michaud M, Francis GS. Multiple sclerosis in
French Canadians: evidence for HLA class I1 susceptibility and
resistance genes. Can J N e w 1 Sci 1990;17:382-386
22. Allen M, Sandberg-Wollheim M, Sjogren K, et al. Association
of susceptibility to multiple sclerosis in Sweden with HLA
class I1 DRBl and DQBl alleles. Hum Immunol 1994;39:
41-48
23. Fauchet R, Semana G, Middleton D, et al. Multiple sclerosis
report. In: Tsuji K, Aizawa M, Sasazuki T, eds. HLA 1991, vol.
1. Oxford: Oxford University Press, 1992:734-740
24. Ransohoff RM. T-cell receptor germline genes and multiple
sclerosis susceptibility: an unfinished tale. Neurology 1992;42:
714-718
25. Hillert J. Immunoglobulin gamma constant gene region polymorphisms in multiple sclerosis. J Neuroimmunol 1993;43:
9-14
26. Walter MA, Gibson WT, Ebers GC, Cox DW. Susceptibilityto
multiple sclerosis is associated with the proximal immunoglobulin heavy chain variable region. J Clin Invest 1991;87:12661273
27. Allegretta M. N i c k JA, Sriram S, Albertini RJ. T cells responsive to myelin basic protein in patients with multiple sclerosis.
Science 1990;247:718-72 1
28. Tienari PJ, Wikstrom J, Sajantila A, et al. Genetic Susceptibility
to multiple sclerosis linked to myelin basic protein gene. Lancet
1992i340987-991
29. Rose J, Gerken S, Lynch S , et al. Genetic susceptibilityin familial multiple sclerosis not linked to the myelin basic protein gene.
Lancet 19!93;341:1179- 1181
30. Thomson G. HLA disease associations:models for insulin dependent diabetes mellitus and the study of complex human genetic disorders. Annu Rev Genet 1988;22:31-50
31. Spielman RS, McGinnis RE, Ewens WJ. Transmission test for
linkage disequilibrium: the insulin gene region and insulin-
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
dependent diabetes mefitus (IDDM). Am J Hum Genet 1993;
52:506-516
Ebers GC, Paty DW. Stiller CR, et al. HLA-typing in multiple
sclerosis sibling pairs. Lancet 1982;2:88-90
Seboun E, Robinson MA, Doolittle TH, et al. A susceptibility
locus for multiple sclerosis is linked to the T cell receptor beta
chain complex. Cell 1989;57:1095-1100
Cox NJ, Spielman RS. The insulin gene and susceptibility to
IDDM. Genet Epidemiol 1989;6:65-69
Haegert DG, Muntoni F, Murru MR, et al. HLA-DQA1 and
-DQBl associations with multiple sclerosis in Sardinia and
French Canada: evidence for itnmunogeneticdy distinct patient
groups. Neurology 1993;43:548-552
Steinman L. Multiple sclerosis and its animal models: the role
of the major histocompatibility complex and the T cell receptor
repertoire. Springer Semin Immunopathol 1992;14:79-93
Nepom GT, Erlich H. MHC class-I1 molecules and autoimmunity. Annu Rev Immunol 1991;9:493-525
Shows TB, Alper CA, Bootsma D, et al. International system
for human gene nomenclature (1979). Cytogenet Cell Genet
1979;25:96-116
Bodmer WF, Albert E, Bodmer JG, et al. Nomenclature for
factors of the HLA system, 1987. In: Dupont B, ed. Immunobiology of HLA, vol 1. New York Springer, 1989:
72-79
Nomenclature for factors of the HLA system 1980. In: Terasaki
PI, ed. Histocompatibility testing 1980. Los Angeles: UCLA
Tissue Typing Laboratory, 1980:18-20
Bodmer WF, Albert E, Bodmer JG. et al. Nomenclature for
factors of the HLA system 1984. In: Albert ED, Baur MP, Mayr
WR, eds. Histocompatibility testing 1984. Berlin: Springer,
19844-8
Juji T, A k a T, Tokunaga K, et al. The serology studies of the
Eleventh International Histocompatibility Workshop: an overview. In: Tsuji K, Aizawa M, Sasazuki T, eds. HLA 1991,
vol 1. Oxford: Oxford University Press, 1992:83-108
Hillert J, Olerup 0. HLA and MS. Neurology 1993;43:2426
(Correspondence)
Hao Q, Saida T, Kawakami H, et al. HLAs and genes in Japanese patients with multiple sclerosis: evidence for increased frequencies of HLA-Cw3, HLA-DR2 and HLA-DQB1*0602.
HW Immun01 1992;35:116-124
Serjeantson SW, Gao X, Hawkins BR, et al. Novel HLADR2-related haplotypes in Hong Kong Chinese implicate the
DQB1*0602 allele in susceptibility to multiple sclerosis. Eur J
Immunogenet 1992;19:11-19
Haegert DG, Muntoni F, Murru MR, et al. HLA and MS. Neurology 1993;43:2426 (Correspondence)
Marrosu MG, Muntoni F, Munu MR, et al. Role of predisposing HLA-DQA and HLA-DQP alleles in Sardinian multiple
sclerosis. Arch Neurol 1993;50256-260
Dekker JW,
Easteal S, Jakobsen IB, et al. HLA-DPBI alleles
correlate with risk for multiple sclerosis in Caucasoid and Cantonese patients lacking the h& risk DQB 1’0602 allele. Tissue
Antigens 1993;41:31-36
Kurdi A, Ayesh I, Abdallat A, et al. Different B lymphocyte
alloantigens associated with multiple sclerosis in Arabs and
North Europeans. Lancet 1977;1:1123-1125
Marrosu MG, Muntoni F, Murru MR, et al. Sardinian multiple
sclerosis is associated with HLA-DR4: a serologic and molecular
analysis. Neurology 1988;38:1749-1753
Francis DA, Thompson AJ, Brookes P, et al. Multiple sclerosis
and HLA: is the susceptibility gene really HLA-DR or -DQ?
Hum Immunol 1991;32:119-124
Middleton D, Megaw G, Savage D. Tap1 and Tap2 polymorphisms in multiple sclerosis patients. Hum Immunol 1333;
37(suppl 1):62
Haegert and Marrosu: Genetic Susceptibility to MS S209
53. Wen J, Olerup 0. Multiple sclerosis is associated with genes
within or close to the HLA-DR-DQsubregion on a normal
DR15, DQ6, Dw2 haplotype. Neurology 1993;43:163-168
54. Spurkland A, [email protected], Vandvik B, et al. HLA-DQA1
and HLA-DQBI genes may jointly determine susceptibility to
develop multiple sclerosis. Hum Immunol 1991;30:69-75
55. Haegert DG, Francis GS. HLA-DQpolymorphisms do not explain HLA class I1 associations with multiple sclerosis in two
Canadian patient groups. Neurology 1993;43:1207-1210
56. Haegert DG, Francis GS. Contribution of a single DQP chain
residue to multiple sclerosis in French Canadians. Hum Immuno1 1992;34:85-90
57. Todd JA, Bell JI, McDevitt HO. HLA-DQP gene contributes
to susceptibility and resistance to insulindependent diabetes
mellitus. Nature 1987;329:599-604
58. Cowan EP, Pierce ML,McFarland HF, McFarlin DE. HLA-DR
and -DQ allelic sequences in multiple sclerosis patients are identical to those found in the general population. Hum Immunol
1991;32:203-2 10
59. Hillert J, Kill T, Olerup 0. HLA-Dw2 in multiple sclerosis: a
world-wide association and a close segregation with disease in
multiplex families. Hum Immunol 1993;3650
60. Beall SS, Concannon P, Charmley P, et al. The germline repertoire of T cell receptor beta chain genes in patients with chronic
progressive multiple sclerosis.J Neuroimmunol 1989;21:59-66
61. Beall SS, Biddison WE, McFarlin DE, et al. Susceptibility for
multiple sclerosis is determined, in part, by inheritance of a
175-Kb region of the TCR V beta chain locus and HLA class
I1 genes. J Neuroimmunol 1993;45:53-60
62. Martinez-Naves E, Victoria-Gutierrez M, Uria DF, et al. The
germline repertoire of T cell receptor beta-chain genes in mula-
ple sclerosis patients from Spain. J Neuroimmunol 1993;47:
9-13
63. Fugger L, Sandberg-Wollheim M, Morling N, et al. The germline repertoire of T-cell receptor chain genes in patients with
relapsinglremittingmultiple sclerosis or optic neuritis. Immunogenetics 1990;31:278-280
64. Hillert J, Leng C, Olerup 0. No association with germline T
cell receptor beta-chain gene alleles or haplotypes in Swedish
patients with multiple sclerosis.J Neuroimmunol1991;32:141147
65. OksenbergJR,Sherrin M, Begovich AB, et al. T-cell receptor
Va and Ca alleles associated with multiple sclerosis and myasthenia gravis. Proc Natl Acad Sci USA 1989;86988-992
66. W e r t J, Leng C, Olerup 0. T-cell receptor a chain germline
gene polymorphism in multiple sclerosis. Neurology 1992;42:
80-84
67. Lynch SG, Rose JW,Petajan JH, et al. Discordance of T-cell
receptor P-chain genes in familial multiple sclerosis. Ann N e w 1
1991;30402-410
68. Lynch SG, RoseJW, Petajan H, Leppert M. Discordance of the
T-cell receptor alpha-chain gene in familial multiple sclerosis.
Neurology 1992;42:839-844
69. Hashimoto L, Mak TW,Ebers GC.T-cell receptor alpha chain
polymorphisms in multiple sclerosis.J Neuroimmunol1992;40
1-8
70. Robinson MA, Kindt TJ.Linkage between T cell receptor genes
and susceptibility to multiple sclerosis: a complex issue. Reg
IIUINIIO~ 1992;4:274-283
7 1. Payami H, Joe S. Farid N, et al. Relative predispositional effects
(RPEs) of marker alleles with disease: HLA-DR alleles and
Graves disease. Am J Hum Genet 1989;45:541-546
S210 Annals of Neurology Supplement 2 to Volume 36, 1994