Intercellular adhesion molecule-1: a protective haplotype ... - Nature

Genes and Immunity (2003) 4, 518–523 & 2003 Nature Publishing Group All rights reserved 1466-4879/03 $25.00 www.nature.com/gene

BRIEF COMMUNICATION

Intercellular adhesion molecule-1: a protective haplotype against multiple sclerosis I Cournu-Rebeix1, E Geńin2, G Lesca1, A Azoulay-Cayla3, N Tubridy1, E Noe´1, M Clanet4, G Edan5, F Clerget-Darpoux2, G Se´mana6 and B Fontaine1,3 Laboratoire des affections de la mye´line et des canaux ioniques musculaires-INSERM U546, Faculte´ de Me´decine Pitie´-Salpeˆtrie`re, Paris, France; 2INSERM U535, Centre Hospitalier Universitaire, Le Kremlin Biceˆtre, France; 3Fe´de´ration de Neurologie, Groupe Hospitalier Pitie´-Salpeˆtrie`re, Paris, France; 4Service de Neurologie, Centre Re´gional Hospitalier Universitaire Purpan, Toulouse, France; 5Service de Neurologie, Centre Re´gional Hospitalier Universitaire Pontchaillou, Rennes, France; 6Laboratoire d’Immunogeńe´tique, Etablissement Français du Sang Bretagne et Laboratoire d’Immunologie (UPRES EA 1257), Centre Hospitalier Re´gional Universitaire et Faculte´ de Me´decine, Rennes, France 1

Intercellular adhesion molecule-1 (ICAM-1) and its receptors are adhesion molecules that play a key role in the transmigration of inflammatory cells through the blood–brain barrier, one of the earliest events in multiple sclerosis (MS), which leads to demyelination in the central nervous system. To investigate the role of genes encoding ICAM-1 and its receptors, we used a strategy of genetic linkage and association in 439 case–parent MS families of French origin, well characterized according to HLA status and severity. We demonstrate that the genes encoding ICAM-1 receptors do not influence MS susceptibility or severity. ICAM-1 had a modest, but significant effect on MS genetic susceptibility, independent of HLA and disease severity. We observed a rare, and an as yet unreported, ICAM-1 gene haplotype defined by amino acids K469 and R241 that was never transmitted to patients suggesting a protective effect against MS in our population. Genes and Immunity (2003) 4, 518–523. doi:10.1038/sj.gene.6364009 Keywords: multiple sclerosis; intercellular adhesion molecule-1; haplotypes; association; transmission disequilibrium test; receptors

Multiple sclerosis (MS) is the most common neurological disease in young adults and is characterized by a complex and multifactorial pathogenesis. Numerous studies have shown that both genetic and environmental factors contribute to an increased risk of developing MS.1,2 Little is known, however, about the exact nature of the involved genetic factors, except for the role of HLA Class II with, in most populations, an association found with the DR2 allele. The role of environmental exposures, hypothetically viral infection, which causes inflammation and demyelination in the central nervous system (CNS) of MS patients has also been suggested but remains unclear. The infiltration of inflammatory cells into the CNS is considered to be one of the earliest events in the formation of a plaque, the fundamental pathological lesion in MS. Inflammatory cells, activated by poorly understood mechanisms, enter the brain by crossing the blood–brain barrier (BBB) through a complex cascade of molecular events in which adhesion molecules are keyplayers.3,4 Adhesion molecules, induced at the surface of endothelial cells by cytokines released by T lymphocytes and macrophages, allow transmigration of inflammatory cells across the BBB, leading to demyelination.5 Molecules that contribute to this mechanism are well

Correspondence: Dr B Fontaine, INSERM U546, 105 bd de l’hoˆpital, 75013 Paris, France. E-mail: [email protected]

characterized.6,7 Several reports have shown that intercellular adhesion molecule 1 (ICAM-1), which is expressed at the surface of endothelial cells, has a prominent role in facilitating the entrance of inflammatory cells into the CNS.8–10 ICAM-1 is a ligand of lymphocyte function-associated antigen-1 (LFA-1 or CD11a/CD18 or aLb2) and Mac-1 (CD11b/CD18 or aMb2), two receptors localized to the membrane of inflammatory cells.11,12 Studies of the interaction between ICAM-1 and its receptors have demonstrated that both LFA-1 and Mac-1 contribute equally to adhesion efficiency during transendothelial migration of neutrophils.13,14 ICAM-1 has also been described as the major receptor for rhinovirus,15,16 strengthening the interest in this gene, according to the viral hypothesis of MS. Because of the key role of ICAM-1 in BBB physiology, genetic variations in its gene or in the genes coding for its receptors may be involved in MS susceptibility and/or severity. Conflicting results on the role of ICAM-1 in MS susceptibility have been reported. An association was first reported with the allele K469 of an intragenic SNP of ICAM-1 on a Polish sample of cases and controls.17 This association was not replicated in subsequent studies using different samples of cases and controls18,19 and in a meta-analysis combining the sample of Luomala and co-workers and the initial sample.20 Moreover, no association was found with another intragenic SNP (SNP G/R 241).

Intercellular adhesion molecule-1 I Cournu-Rebeix et al

Table 1

519

Primers for amplification by polymerase chain reaction of adhesion molecule genes

Genes

Base change

AA change

Primers for amplification

ICAM1

CA repeat

None

50 -(6-Fam)GCAAGCAGTGGGGAAGGG-30 50 -TGGGTGACAGAGCGAGAG-30 50 -CGTCCATCCCTGTCTGCT-30 50 -GTCCTCTGCGGTCACACT-30 50 -GCCCCCGACTGGACGAGA-30 50 -GGATACAACAGGCGGTGA-30 50 -CTCTGCATCGTCTCCTCT-30 50 -ACCCCACCCTTGTCTCCT-30 50 -GGAGCACTTGGTGAAGACAA-30 50 -GGGGAATACAGCGGACACA-30 50 -AGCACATGGATGCTTTTCTC-30 50 -CAGTGTCCTCCGCCCTGT-30 50 -CCCTGATGTCCTTCTCTCC A-30 50 -GGAACAACCCCCGTCTTC-30 50 -CAATCTCGGCTCACTGCAAC-30 50 -GAGGACAGGAGTTCAAGACCA-30 50 -GCCTCCTCCACCGAAGTGT-30

ICAM1

C/T12957

G/R241

ICAM1

T/C13858

K/E469

CD18

C/G10385

G/G273

CD18

C/A23837

P/Q819

Cd11a

C/T21383

None

Cd11a

T/C26442

None

Cd11a

C/T35579

None

CD11b

C/T2573a

A/V858

0

Primer or probes for genotyping

0

TM (1C)

None 0

5 -CCGAGACTGGGAACAGCC-30

55

50 -GCACATTCACGGTCACCT-30

55

0

0

5 -GTCAGGATGGCGCCCAGCTT-3 50 -CCCAGAAGCACCCGGCC-30

55 58

0

0

5 -CTGAGGAAAGTGTGTGGGGC-3

55

50 -CCAAGGTCAGAGCTCTCCTC-30

55

0

0

5 -AATTAGCTGGGTGTGGTGGC-3

55

50 -(6-fam)CTCTTCAAGGCCC-30 50 -(Vic)CTCTTCAAAGCCC-30

60

5 -CTCTGAGTTTTCCGGGAAGATG-3

Position in mRNA. DNA samples were amplified by polymerase chain reaction (PCR) according to standard protocols. Genotyping of the CA repeat was performed by electrophoresis of a PCR fragment amplified with fluorescent primers on an ABI PRISM 377 sequencer (Applied Biosystems, USA). Genotypes of the SNP were assessed using either a single labeled base extension method for an ABI PRISM 377 sequencer (SnaPshot) (ICAM1, CD18 and CD11a) or a 50 nuclease assay on an ABI 7000 sequence detection system (CD11b) (Applied Biosystems, USA).

a

In this study, we investigate the association between ICAM-1 and MS in a large population of French MS case–parent trios, using two single nucleotide polymorphisms and one microsatellite (Table 1). We also study the role of genes coding for ICAM-1 receptors LFA-1 and MAC-1 that have never been studied before in relation to MS susceptibility or severity (six SNPs, Table 1). Using a strategy of multilocus genetic analysis, we report the association of a protective ICAM-1 haplotype against MS in this population. To test for genetic association with the disease, the transmission disequilibrium test (TDT) was performed at each locus.21 This test focuses on heterozygous parents and compares alleles transmitted and nontransmitted to affected offspring. For ICAM-1, each marker was tested individually but also in relation with the other markers by looking at the haplotype distribution. It has been shown that haplotype tests may be more powerful at detecting the role of a given genomic region in disease susceptibility than single-locus methods.22–25 Indeed, disease susceptibility may be due to the combined effects of multiple sequence variants. To control for the numerous tests performed, we adopted a strategy in two steps. All the markers were tested in a first subset of 150 families referred to as batch 1. For such a sample size, the probability to detect a locus with a genotype relative risk of 2 (assuming a multiplicative model) for a nominal type I error of 5% is greater than 80%.26 When for a marker, the uncorrected P-value was below 5% in batch 1, the remaining 289 case–parent families (batch 2) were tested. A total of 439 case–parent MS families, all comprising one affected child with definite MS27,28 and his two parents, were recruited throughout France by the French Multiple Sclerosis Genetic Group as previously described.29,30 Informed written consent was obtained from each individual participating in the study in accordance with the Helsinki Convention and the French law

relating to biomedical research. Geographic and ethnic origin, sex, age at onset, duration of the disease, disease course and disease severity as measured by the Expanded Disability Status Scale (EDSS)31 were recorded for each patient. Since we have previously demonstrated that familial factors contribute to MS severity,32 we stratified our sample based on severity. We used the progression index (PI) defined as the ratio of the EDSS vs disease duration (in years) for patients with a disease duration of 5 years or more.32 Patients with a PI less than 0.4 and over 0.4 were classified as mild and severe, respectively. Among the 430 patients with information on PI, 250 were classified as mild and 180 as severe. HLA Class II DR alleles were determined for 429 out of the 439 patients. Since HLA-DR2 allele is associated with MS,33 we have also stratified our sample according to whether or not the proband was a carrier of the HLA Class II DR2 allele (among the 429 patients genotyped for HLA, 223 (52%) were carriers). The demographic characteristics of our population are similar to published series. There was a female preponderance (68%) and the mean age at onset was 26 years (standard deviation (s.d.) ¼ 7). The mean duration of the disease was 9 years (s.d. ¼ 7) and the mean EDSS was 3.5 (s.d. ¼ 2). Parental transmitted and nontransmitted allele frequencies of CD18, CD11a, CD11b and ICAM-1 gene polymorphisms are presented in Table 2. With regard to the frequencies of the ICAM-1 microsatellite alleles, there were 7, 85 and 8% of alleles 1, 2 and 3 respectively, considering the nontransmitted parental alleles. There were 5, 86 and 9 of alleles 1, 2 and 3, respectively, considering patient alleles. Performing single-locus TDT on batch 1 for CD18, CD11a and CD11b genes, no association was found with MS susceptibility with or without stratification on HLA status and on disease severity (data not shown). For the ICAM-1 gene, there was a difference in batch 1 in the distribution between transmitted and nontransmitted Genes and Immunity


520

Table 2 Allele frequencies of CD18, Cd11a and CD11b genes (in %) Gene

Polymorphism

Number of chromosomes

Rarest allele a

Frequency of the rarest allele

292 280 292 250 292 292 702 442

G273 Q819 T21383 C26442 T35579 A858 R241 E469

24 21 46 44 17 14 11 46

292 280 292 250 292 292 702 442

G273 Q819 T21383 C26442 T35579 A858 R241 E469

27 18 40 43 13 14 9 48

(a) Nontransmitted parental alleles CD18 G/G273 P/Q819 CD11a Intron 11 SNP Intron 16 SNP Intron 21 SNP CD11b A/V858 ICAM-1 G/R241 K/E469 (b) Patients CD18

G/G273 P/Q819 Intron 11 SNP Intron 16 SNP Intron 21 SNP A/V858 G/R241 K/E469

CD11a CD11b ICAM-1

Amino-acid position for CD18 and CD11b and ICAM-1 polymorphisms, base position for CD11a SNPs. Allele frequencies for each polymorphism were calculated using, as controls, parental alleles not transmitted to the affected child in each case–parent family. These frequencies provide unbiased estimates of alleles frequencies in the general population.41

a

a G/R241

K/E469 *

CA repeat *

5’...CCCTGGACGGGCTGTTC...3’

5’...TCACCCGCGAGGTGACC...3’

5’... CCCTGGACTGGCTGTTC

5’...TCACCCGCTAGGTGACC...3’

… 3’

TTCCCAGTAGGCG(CA)nTACACATCTGTT

ICAM-1 mRNA 5′

3′

120 bp

ICAM-1 protein

b

N C

Rhinoviruses

Ig-1

Ig-2

Ig-3

Ig-4

Ig-5

K/E469 p=0.0156*

8

G/R241

Mac-1

7

TDT value

LFA-1

6 5 4 3 2 1 0

469K/241R

469K/241G

469E/241R

469E/241G

Haplotypes ICAM-1

Figure 1 Genetic analysis of ICAM-1 gene in MS. (a) Analysis of a microsatellite and two single nucleotide polymorphisms in the human ICAM-1 gene: The exon distribution and the protein structural domains (indicated by black boxes) of ICAM-1 are illustrated. Locations of DNA polymorphisms are indicated by stars. Locations of binding sites of LFA-1, Mac-1 and human rhinoviruses are shown in a diagram of an ICAM-1 molecule. (b) TDT results for ICAM-1 haplotypes. *Statistically significant: To test for transmission of haplotypes in the ICAM-1 gene, the multilocus TDT implemented in Genehunter was used (Genehunter version 2.0 beta).

Genes and Immunity


Table 3 Analysis of ICAM-1 gene in batch 1 and the entire population (batch 1+batch 2) with and without stratification for severity and

521

HLA-DR2 status: number of allele transmissions and nontransmissions from parents to patients Families Batch 1 G/R241 K/E469 CA repeat

Entire population G/R241 K/E469 CA repeat

HLA DR2+ G/R241 K/E469 CA repeat

HLA DR2 G/R241 K/E469 CA repeat Mild group G/R241 K/E469 CA repeat Severe group G/R241 K/E469 CA repeat

Allele

Heterozygous parents

Transmitted alleles

Nontransmitted alleles

w2a

P

1 1 1 2 3

38 130 35 82 51

21 61 11 52 21

17 69 24 30 30

0.421 0.492 4.829 5.902 1.588

>0.05 >0.05 0.03 0.01 >0.05

1 1 1 2 3

145 202 96 207 123

81 101 40 110 63

64 101 56 97 60

1.993 0.000 2.667 0.816 0.073

>0.05 >0.05 >0.05 >0.05 >0.05

1 1 1 2 3

68 96 46 94 52

39 53 16 52 28

29 43 30 42 24

1.471 1.042 4.261 1.064 0.308

>0.05 >0.05 0.04 >0.05 >0.05

1 1 1 2 3

71 102 35 70 39

39 46 18 35 19

32 56 17 35 20

0.690 0.980 0.029 0.000 0.026

>0.05 >0.05 >0.05 >0.05 >0.05

1 1 1 2 3

43 68 23 49 30

24 37 8 28 15

19 31 15 21 15

0.581 0.529 2.130 1.000 0.000

>0.05 >0.05 >0.05 >0.05 >0.05

1 1 1 2 3

48 72 24 55 35

26 37 9 31 17

22 35 15 24 18

0.333 0.056 1.500 0.891 0.029

>0.05 >0.05 >0.05 >0.05 >0.05

Results are shown only for allele 1 of each tested SNP.

a

alleles at the microsatellite marker. One of the alleles of the microsatellite (allele 1 that has never been reported before) was significantly less transmitted and, conversely, one of the alleles (allele 2) was more transmitted to affected offspring (Table 3). According to our decision criteria for the two-step analysis, only the microsatellite should then be tested on batch 2 (289 families). However, we decided to also genotype the different individuals at the two ICAM-1 intragenic SNPs, in order to obtain some haplotypic information. The results are presented in Table 3. When the loci were studied individually, no difference was found on the allele transmissions from heterozy gous parents to affected offspring in batch 2 and in the entire population (batch 1+batch 2) at any of the tested markers. However, after stratification on HLA status, a lack of transmission of the CA-repeat allele 1 could be observed in DR2+families (uncorrected P ¼ 0.04) but there was no significant difference in the transmissions of this allele to DR2+ and DR2 patients (w2 tests). For the frequent three-locus haplotypes, no significant difference was observed between the number of trans-

missions and nontransmissions from heterozygous parents to affected offspring (data not shown). For rarer haplotypes, however, we observed a lack of transmission of one haplotype, R241/K469/2. This particular haplotype was indeed found in four families and never transmitted to affected offspring. This same combination R241 and K469 was also observed in three other families in which it was never transmitted to the affected child. In one family, it was associated to allele 3 at the CA repeat and in the two other families, no typing was available for the CA repeat. Thus, we show that the ICAM-1 haplotype defined by R241 and K469 is found in seven informative families and is never transmitted from parents to affected children (TDT ¼ 7, binomial exact test P ¼ 0.01) (Figure 1). Linkage disequilibrium between the different ICAM-1 markers was evaluated on the transmitted parental haplotypes. Linkage disequilibrium is strong between ICAM-1 G/R241 and ICAM-1 K/E469 SNPs (D ¼ 0.0359, Po1.5 105) as well as between the microsatellite and ICAM-1 K/E469 SNP (D ¼ 0.028, P44.2 107). However, there is no evidence for linkage disequilibrium between the microsatellite and ICAM-1 G/R241 alleles (D ¼ 0.0001, P40.99). Genes and Immunity


522

In this article, we report a genetic study of ICAM-1 and its receptors as candidates for MS genes. We took advantage of a large cohort of case–parent MS families of French origin to look for both linkage and association between markers of these candidate genes and MS. No association between MS and genes encoding ICAM-1 receptors was detected. However, we found that one haplotype of the ICAM-1 gene formed by the combination of allele R at position 241 in exon 4 and allele K at position 469 in exon 6 was never transmitted to affected offspring, suggesting that this haplotype may be protective against MS. ICAM-1 gene is located on chromosome 19 p13, a region identified as of possible significance in two independent screenings in MS families.34,35 Previous studies have addressed the role of ICAM-1 in MS genetics with conflicting results. One study, performed on case–controls of Polish origin, found a genetic association with ICAM-1 K469.17 This result was not confirmed in three subsequent studies.18–20 However, except for the study of Marrosu et al,19 no multilocus analysis of the two SNPs was performed. In the study by Marrosu et al,19 haplotypes were studied but the particular haplotype we evidenced here was not observed probably because of the small sample size.19 The ICAM-1 molecule consists of five immunoglobulin-like (Ig) domains (Ig-1 to Ig-5), each of them encoded by its own exon. Exon 4 polymorphism (G/R241) changes an arginine into a lysine in domain Ig-3. Exon 6 polymorphism (K/E469) substitutes a glutamic acid to a lysine in Ig-like domain 5. To explain the protective effect of the R241/K469 haplotype in MS, we propose two hypotheses. The first postulates a decreased binding efficacy of ICAM-1 to its ligands Mac-1 and LFA-1. Accordingly, the binding site of Mac-1 to ICAM-1 is located within Ig-3 domain. It has also been shown that, although domain Ig-1 contains the major binding site for LFA-1 to ICAM-1, Ig-3, Ig-4 and Ig-5 were also important for the accessibility of LFA-1 binding site, decreasing the efficacy of the interaction several fold.36 The second hypothesis relates to the role of ICAM-1 as a viral receptor. ICAM-1 has been shown to be an important receptor for the majority of human rhinoviruses, an interaction that initiates entry of the virus into the host cell.15,37–39 Mutational and antigenic analyses localized the rhinovirus binding site to domain Ig-1 of ICAM-1 and demonstrated that Ig-3, Ig-4 and Ig-5 domains are also important for the accessibility of the rhinovirus binding site.36,40 Therefore, a combination of allele K469 and R241 in domain Ig-5 and Ig-3, respectively, of ICAM1 molecule may influence susceptibility to viral infection, an often suspected environmental factor in MS pathophysiology. In conclusion, in a large population of MS case–parent trios, we found that a particular haplotype of ICAM-1 may have a protective role in MS. Although this haplotype is a rare haplotype, the association is strong enough to be detected in the French population.

Acknowledgements We thank patients, their families and physicians for their help and support. The French Multiple Sclerosis Genetics Genes and Immunity

Group, Genethon, the CIC Pitie´-Salpeˆtrie`re, INSERM and the French Ministry of Research (CRB) provided help for the DNA bank. This work was supported by ARSEP, INSERM and FRM (Action 2000).

References 1 Alperovitch A. Epidemiologie de la scle`rose en plaques. Med Therap 1995; 1: 545–548. 2 Poser CM. The epidemiology of multiple sclerosis: a general overview. Ann Neurol 1994; 36: 180–193. 3 Cannella B, Cross AH, Raine CS. Upregulation and coexpression of adhesion molecules correlate with relapsing autoimmune demyelination in the central nervous system. J Exp Med 1990; 172: 1521–1524. 4 Springer TA. Adhesion receptors of the immune system. Nature 1990; 346: 425–434. 5 Raine CS, Cannella B. Adhesion molecules and central nervous system inflammation. Semin Neurosci 1992; 4: 201–219. 6 Osborn L. Leukocyte adhesion to endothelium in inflammation. Cell 1990; 62: 3–6. 7 Springer TA. Traffic signals for lymphocyte recirculation and leukocyte emigration: the multistep paradigm. Cell 1994; 76: 301–314. 8 Pryce G, Male D, Campbell I, Greenwood J. Factors controlling T-cell migration across rat cerebral endothelium in vitro. J Neuroimmunol 1997; 75: 84–94. 9 Male D, Rahman J, Pryce G, Tamatani T, Miyasaka M. Lymphocyte migration into the CNS modelled in vitro: roles of LFA-1, ICAM-1 and VLA-4. Immunology 1994; 81: 366–372. 10 Greenwood J, Wang Y, Calder VL. Lymphocyte adhesion and transendothelial migration in the central nervous system: the role of LFA-1, ICAM-1, VLA-4 and VCAM-1. off. Immunology 1995; 86: 408–415. 11 Marlin SD, Springer TA. Purified intercellular adhesion molecule-1 (ICAM-1) is a ligand for lymphocyte functionassociated antigen 1 (LFA-1). Cell 1987; 51: 813–819. 12 Diamond MS, Staunton DE, Marlin SD, Springer TA. Binding of the integrin Mac-1 (CD11b/CD18) to the third immunoglobulin-like domain of ICAM-1 (CD54) and its regulation by glycosylation. Cell 1991; 65: 961–971. 13 Smith CW, Marlin SD, Rothlein R, Toman C, Anderson DC. Cooperative interactions of LFA-1 and Mac-1 with intercellular adhesion molecule-1 in facilitating adherence and transendothelial migration of human neutrophils in vitro. J Clin Invest 1989; 83: 2008–2017. 14 Neelamegham S, Taylor AD, Burns AR, Smith CW, Simon SI. Hydrodynamic shear shows distinct roles for LFA-1 and Mac1 in neutrophil adhesion to intercellular adhesion molecule-1. Blood 1998; 92: 1626–1638. 15 Greve JM, Davis G, Meyer AM et al. The major human rhinovirus receptor is ICAM-1. Cell 1989; 56: 839–847. 16 Olson NH, Kolatkar PR, Oliveira MA et al. Structure of a human rhinovirus complexed with its receptor molecule. Proc Natl Acad Sci 1993; 90: 507–511. 17 Mycko MP, Kwinkowski M, Tronczynska E, Szymanska B, Selmaj KW. Multiple sclerosis: the increased frequency of the ICAM-1 exon 6 gene point mutation genetic type K469. Ann Neurol 1998; 44: 70–75. 18 Luomala M, Elovaara I, Koivula T, Lehtimaki T. Intercellular adhesion molecule-1 K/E 469 polymorphism and multiple sclerosis. Ann Neurol 1999; 45: 546–547. 19 Marrosu MG, Schirru L, Fadda E et al. ICAM-1 gene is not associated with multiple sclerosis in sardinian patients. J Neurol 2000; 247: 677–680. 20 Killestein J, Schrijver HM, Crusius JB et al. Intracellular adhesion molecule-1 polymorphisms and genetic


523 21

22 23

24

25

26 27 28 29

30

susceptibility to multiple sclerosis: additional data and metaanalysis. Ann Neurol 2000; 47: 277–279. Spielman RS, McGinnis RE, Ewens WJ. Transmission test for linkage disequilibrium: the insulin gene region and insulindependent diabetes mellitus (IDDM). Am J Hum Genet 1993; 52: 506–516. Akey J, Jin L, Xiong M. Haplotypes vs single marker linkage disequilibrium tests: what do we gain? Eur J Hum Genet 2001; 9: 291–300. Douglas JA, Boehnke M, Gillanders E, Trent JM, Gruber SB. Experimentally-derived haplotypes substantially increase the efficiency of linkage disequilibrium studies. Nat Genet 2001; 28: 361–364. Fallin D, Cohen A, Essioux L et al. Genetic analysis of case/ control data using estimated haplotype frequencies: application to APOE locus variation and Alzheimer’s disease. Genome Res 2001; 11: 143–151. Zaykin DV, Westfall PH, Young SS et al. Testing association of statistically inferred haplotypes with discrete and continuous traits in samples of unrelated individuals. Hum Hered 2002; 53: 79–91. Risch N. The future of genetic studies of complex human diseases. Science 1996; 273: 1516–1517. Poser CM, Paty DW, Scheinberg L et al. New diagnostic criteria for multiple sclerosis: guidelines for research protocols. Ann Neurol 1983; 13: 227–231. Goodkin DE, Doolittle TH, Hauser SS et al. Diagnostic criteria for multiple sclerosis research involving multiply affected families. Arch Neurol 1991; 48: 805–807. Mertens C, Brassat D, Reboul J et al. A systematic study of oligodendrocyte growth factors as candidates for genetic susceptibility to MS. French Multiple Sclerosis Genetics Group. Neurology 1998; 51: 748–753. Reboul J, Mertens C, Levillayer F et al. Cytokines in genetic susceptibility to multiple sclerosis: a candidate gene approach. French Multiple Sclerosis Genetics Group. J Neuroimmunol 2000; 102: 107–112.

31 Kurtzke JF. Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology 1983; 33: 1444–1452. 32 Brassat D, Azais-Vuillemin C, Yaouanq J et al. Familial factors influence disability in MS multiplex families. French Multiple Sclerosis Genetics Group. Neurology 1999; 52: 1632–1636. 33 Yaouanq J, Semana G, Eichenbaum S et al. Evidence for linkage disequilibrium between HLA-DRB1 gene and multiple sclerosis. The French Research Group on Genetic Susceptibility to MS. Science 1997; 276: 664–665. 34 Sawcer S, Jones HB, Feakes R et al. A genome screen in multiple sclerosis reveals susceptibility loci on chromosome 6p21 and 17q22. Nat Genet 1996; 13: 464–468. 35 The Multiple Sclerosis Genetics Group, Haines JL, TerMinassian M et al. A complete genomic screen for multiple sclerosis underscores a role for the major histocompatability complex. The Multiple Sclerosis Genetics Group. Nat Genet 1996; 13: 469–471. 36 Staunton DE, Dustin ML, Erickson HP, Springer TA. The arrangement of the immunoglobulin-like domains of ICAM-1 and the binding sites for LFA-1 and rhinovirus. Cell 1990; 61: 243–254. 37 Staunton DE, Merluzzi VJ, Rothlein R et al. A cell adhesion molecule, ICAM-1, is the major surface receptor for rhinoviruses. Cell 1989; 56: 849–853. 38 Hoover-Litty H, Greve JM. Formation of rhinovirus-soluble ICAM-1 complexes and conformational changes in the virion. J Virol 1993; 67: 390–397. 39 Rossmann MG. Viral cell recognition and entry. Protein Sci 1994; 3: 1712–1725. 40 Ockenhouse CF, Tegoshi T, Maeno Y et al. Human vascular endothelial cell adhesion receptors for Plasmodium falciparuminfected erythrocytes: roles for endothelial leukocyte adhesion molecule 1 and vascular cell adhesion molecule 1. J Exp Med 1992; 176: 1183–1189. 41 Thomson G. Mapping disease genes: family-based association studies. Am J Hum Genet 1995; 57: 487–498.

Genes and Immunity