Genetic association between polymorphisms in the ADAMTS14 ...

408 views
368 views

Published on

0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
408
On SlideShare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
2
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Genetic association between polymorphisms in the ADAMTS14 ...

  1. 1. Journal of Neuroimmunology 164 (2005) 140 – 147 www.elsevier.com/locate/jneuroim Genetic association between polymorphisms in the ADAMTS14 gene and multiple sclerosis Robert Goertschesa, Manuel Comabellaa,*, Arcadi Navarrob, Hector Perkala, Xavier Montalbana a Unitat de Neuroimmunologia Clınica, Hospital Universitari Vall d’Hebron (HUVH), Escuela de Enfermeria 2a planta, ´ Psg Vall d’Hebron 119-129, 08035 Barcelona, Spain b Unitat de Biologıa Evolutiva, Universitat Pompeu Fabra, Barcelona, Spain ´ Received 28 January 2005; accepted 4 April 2005 Abstract ADAMTS14 is a novel member of the ADAMTS (a disintegrin-like and metalloproteinase domain with thrombospondin type 1 modules) metalloproteinase family which processes extracellular matrix proteins. In the present study we performed a comprehensive investigation of the ADAMTS14 as a candidate gene for susceptibility to multiple sclerosis (MS). Eight single nucleotide polymorphisms (SNPs) were analyzed in a case – control study of 287 patients with MS [192 with relapsing – remitting MS (RRMS) and 95 with primary – progressive MS (PPMS)], and 285 age- and sex-matched controls. Allele and genotype frequencies were compared between controls and the MS subgroups, and gene-based haplotypes were reconstructed by computational procedures. Pairwise linkage disequilibrium values (D¶) suggested that three locus pairs (SNPs 3 through 5) had alleles in strong disequilibrium and constituted a haplotype block spanning 14 kb. Overall comparisons of allele and genotype frequencies showed association for SNPs 3 and 6 with MS. Stratification of MS patients according to major clinical forms revealed an increased frequency of both allele C ( p = 0.006) and CC homozygosity ( p = 0.008) at SNP6 in RRMS patients compared with controls. PPMS was associated with allele A at SNP2 compared with RRMS ( p = 0.003) and controls ( p = 0.009), and with CG heterozygosity at SNP3 compared with controls ( p = 0.005). Haplotype frequency comparisons showed significant association between PPMS and the AGGGC haplotype compared with controls ( p = 0.0004), and negative association between RRMS and the GGAGT haplotype compared with controls ( p = 0.0026). No association was detected between different genotypes and disease severity measured by the Multiple Sclerosis Severity Score (MSSS). These findings suggest a potentially important role for the ADAMTS14 gene in predisposition to MS. D 2005 Elsevier B.V. All rights reserved. Keywords: Polymorphism; ADAMTS14 gene; Multiple sclerosis 1. Introduction with undefined environmental exposures, resulting in an autoimmune response against myelin proteins and progres- Multiple sclerosis (MS) is considered to be a chronic sive neurological dysfunctions (Hauser and Goodkin, autoimmune disease of the central nervous system (CNS) 2001; Oksenberg et al., 2001). The genetic component of white matter, characterized by inflammation, demyelination the disease is mainly indicated by both an increased and axonal injury (Compston and Coles, 2002). A large relative risk in siblings compared with the general body of research supports a multifactorial etiology, with an population, and a higher concordance rate in monozygotic underlying genetic susceptibility likely acting in concert compared with dizygotic twins (Sadovnick et al., 1993; Dyment et al., 2004). The strongest and most consistent evidence for a susceptibility gene in MS lies within the * Corresponding author. Tel.: +34 9327 46176; fax: +34 9327 46480. major histocompatibility complex (MHC) on chromosome E-mail address: mcomabel@vhebron.net (M. Comabella). 6p21.3. Association with the HLA-DR2 haplotype 0165-5728/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jneuroim.2005.04.005
  2. 2. R. Goertsches et al. / Journal of Neuroimmunology 164 (2005) 140 – 147 141 Table 1 2. Patients and methods Demographic and baseline clinical characteristics of MS patients and healthy controls involved in the study 2.1. Patients and controls Characteristics HC MS RRMS PPMS (n = 285) (n = 287) (n = 192) (n = 95) A total number of 287 unrelated affected individuals were Female/male 169/116 169/118 120/72 49/46 included in the study. All subjects were of Spanish origin and (% women) (59.3) (58.9) (62.5) (51.6) Age (years)a 40.3 43.0 37.8 53.5 satisfied Poser’s criteria for clinically definite MS (Poser et (11.4) (12.3) (10.1) (9.5) al., 1983). There were 192 patients with relapsing –remitting Duration of disease – 11.4 10.4 13.4 MS (RRMS) and 95 with primary – progressive MS (PPMS). (years)a (7.1) (6.4) (8.0) Disability was measured by the Multiple Sclerosis Severity Age at disease onset – 31.6 27.4 40.3 Score (MSSS), which relates scores on the Expanded (years)a (10.8) (8.8) (9.3) EDSSb – 3.5 2.0 6.0 Disability Status Scale (EDSS) to the distribution of disability (4.0) (2.0) (3.0) in patients with comparable disease durations. The control EDSS: Expanded Disability Status Scale; RRMS: relapsing – remitting population comprised of 285 unrelated individuals recruited multiple sclerosis; PPMS: primary – progressive multiple sclerosis; HC: at our hospital transfusion centre, which serves the geograph- healthy controls. ic area from where the patients were enrolled. The study was a Data are expressed as mean (S.D.). approved by the Ethics Committee of Vall d¶Hebron b Data are expressed as median (interquartile range). University Hospital and all the subjects involved in the study gave written informed consent. A summary of demographic and baseline clinical characteristics of MS patients and (DRB1*1501 – DQB1*0602) has been repeatedly demon- healthy controls is shown in Table 1. strated in multiple populations, primarily those of Northern European descent (Allen et al., 1994; Weinshenker et al., 2.2. Single nucleotide polymorphisms 1998; Villoslada et al., 2002; Barcellos et al., 2003). Although the MHC region contributes significantly to MS SNPs with publicly stated minor allele frequencies > 0.10 risk, much of the genetic effect in MS remains to be were selected from Celera Discovery Systemi SNP explained. database, available from Applied Biosystems Web store As part of the collaborative Genetic Analysis in and displayed in the SNPbrowser software version 2.0 Multiple Sclerosis in EuropeanS (GAMES) study (Sawcer (http://www.allsnps.com/snpbrowser). Genotype assay and Compston, 2003), we performed a whole genome details and SNP positions are shown in Table 2. association screen in MS using a panel of 5546 microsatellite markers and pooled DNA methodology. 2.3. ADAMTS14 genotyping After obtaining accurate data for 5131 markers and applying a sliding-window method to the genomic regions Genomic DNA from peripheral blood samples was displaying evidence of association with MS in the Spanish obtained using standard methods. Genotyping of the population (Goertsches et al., 2003), we pinpointed a selected SNPs was performed applying the 5¶ nuclease region of interest at 10q22.1 positioned between two assay technology for allelic discrimination using fluorogenic contiguous markers (D10S537, D10S1685) and harbour- TaqMan probes commercially available from Applied ing the genes encoding for perforin and ADAMTS14 with Biosystems. Briefly, polymerase amplification was per- the latter containing D10S1685 in its genomic sequence. formed in 5 Al reactions using 2.5 Al TaqMan Universal An initial screen of the ADAMTS14 with single nucleotide polymorphisms (SNPs) in DNA samples of 192 relaps- Table 2 ing – remitting MS and 191 control individuals revealed Characteristics of single nucleotide polymorphisms (SNPs) association between gene polymorphisms and susceptibil- SNP Physical positiona Intron Variation Celera ID MAF ity for MS. These findings being in accordance with previous observations on the important role that metallo- 1 71.786.645 2 C/T hCV1229664 0.19 2 71.794.022 2 A/G hCV1229671 0.30 proteinases may play in MS pathogenesis (Rosenberg, 3 71.801.641 2 C/G hCV1229684 0.29 2002; Fiotti et al., 2004) prompted us to extend the study 4 71.808.365 3 A/G hCV11453368 0.21 of the ADAMTS14 as a candidate gene for MS. Therefore, 5 71.815.810 4 A/G hCV1229703 0.26 we performed a population case – control study with an 6 71.821.387 4 C/T hCV11453336 0.13 increased sample size (287 MS cases and 285 controls) 7 71.853.444 15 A/G hCV1229765 0.50 8 71.861.562 19 A/T hCV1229794 0.35 and SNP marker density comparing allele and genotype MAF: Minor Allele Frequency from individuals of Caucasian origin frequencies in the ADAMTS14 gene in MS patients and (Applied Biosystems). Microsatellite D10S1685 is located 2.1 kb down- healthy controls. We also reconstructed gene-based hap- stream of SNP6 at 71.823.494. lotypes in the population by computational procedures and a SNPbrowser version 2.0 (Applied Biosystems) based on NCBI build 34 demonstrated genetic association with MS. genome.
  3. 3. 142 R. Goertsches et al. / Journal of Neuroimmunology 164 (2005) 140 – 147 between pairs of loci was calculated with the Haploview software (Barrett et al., 2005). Haplotypes and assignment of haplotype pairs to each individual were obtained from Table 3 Allele and genotype frequency distribution in MS patients and controls SNP Analysis MS, n (%) Controls, n (%) OR (95% CI) p-Value 1 Allele C 152 (26.6) 127 (22.3) 1.3 (1.0 – 1.7) 0.091 T 420 (73.4) 443 (77.7) Genotype CC 19 (57.6) 14 (42.4) 1.4 (0.7 – 2.8) 0.375 CT 114 (53.5) 99 (46.5) 1.3 (0.9 – 1.8) 0.206 TT 153 (47.1) 172 (52.9) 0.8 (0.5 – 1.1) 0.098 2a Allele A 206 (36.5) 194 (34.5) 1.1 (0.9 – 1.4) 0.482 G 358 (63.5) 368 (65.5) Genotype AA 61 (53.0) 54 (47.0) 1.2 (0.8 – 1.8) 0.477 Fig. 1. Diagram of Haploview generated block structure of ADAMTS14 at AG 84 (49.4) 86 (50.6) 1.0 (0.7 – 1.4) 0.833 chromosomal location 10q22.1. Upper bars: Complete gene length and GG 137 (49.3) 141 (50.7) 0.9 (0.7 – 1.3) 0.705 distinction exon/intron (vertical bars/horizontal lines), distribution of SNP 3 Allele markers 1 to 8 spanning 74.9 kb. Below: Linkage disequilibrium plot of C 208 (36.2) 218 (38.3) 0.9 (0.7 – 1.2) 0.482 computed pairwise LD statistics for all markers, 14 kb haplotype block G 366 (63.8) 352 (61.8) identification due to strong LD constituted by consecutive markers 3, 4, and Genotype 5 (D¶; pair 3/4 = 0.94, pair 3/5 = 1.0, pair 4/5 = 1.0); D¶ values of 1.0 are not CC 28 (35.9) 50 (64.1) 0.5 (0.3 – 0.8) 0.007 shown (empty box). CG 152 (56.3) 118 (43.7) 1.6 (1.2 – 2.2) 0.006 GG 107 (47.8) 117 (52.2) 0.9 (0.6 – 1.2) 0.356 PCR MasterMix, No AmpErase UNG, 0.25 Al of 4 Allele A 69 (12.1) 92 (16.2) 0.7 (0.5 – 1.0) 0.048 TaqMan probe, 20 ng of genomic DNA template and G 501 (87.9) 476 (83.8) 1.25 Al of MilliQ water. Thermal cycling and end-point PCR Genotype analysis were performed on an ABI PRISM 7900HT AA 4 (33.3) 8 (66.7) 0.5 (0.2 – 1.7) 0.241 Sequence Detection System under specified conditions: 95 AG 61 (44.5) 76 (55.5) 0.8 (0.5 – 1.1) 0.135 GG 220 (52.4) 200 (47.6) 1.4 (1.0 – 2.1) 0.066 -C for 10 min, and 40 cycles each of 95 -C for 15 s and 60 5 Allele -C for 1 min. SNP variation was assessed by means of the A 148 (25.8) 134 (23.6) 1.1 (0.9 – 1.5) 0.390 allelic discrimination assay employing the Applied Biosys- G 426 (74.2) 434 (76.4) tems software package SDS 2.1. Litigious results from SNP Genotype genotyping were tested twice. AA 17 (53.1) 15 (46.9) 1.1 (0.6 – 2.3) 0.739 AG 114 (52.3) 104 (47.7) 1.1 (0.8 – 1.6) 0.446 GG 156 (48.6) 165 (51.4) 0.9 (0.6 – 1.2) 0.367 2.4. Statistical methods 6 Allele C 530 (92.7) 500 (88.0) 1.7 (1.2 – 2.6) 0.008 Differences in allele, genotype, and haplotype frequen- T 42 (7.3) 68 (12.0) cies between cases and controls were compared using a chi- Genotype square test. To avoid problems caused by multiple testing, CC 246 (52.7) 221 (47.3) 1.8 (1.1 – 2.7) 0.011 CT 38 (39.6) 58 (60.4) 0.6 (0.4 – 0.9) 0.023 we applied the sequential Bonferroni correction (Rice, TT 2 (28.6) 5 (71.4) 0.4 (0.1 – 2.0) 0.250 1989) to determine the significance of each individual p- 7 Allele value in this study. This is a conservative procedure, given A 295 (51.9) 303 (53.5) 0.9 (0.7 – 1.2) 0.590 that the LD structure of the region under study makes tests G 273 (48.1) 263 (46.5) Genotype non-independent (Becker and Knapp, 2004) and that the AA 71 (47.3) 79 (52.7) 0.9 (0.6 – 1.3) 0.431 tests for alleles and genotypes are also non-independent. AG 153 (51.3) 145 (48.7) 1.1 (0.8 – 1.6) 0.530 Calculations were performed with the SPSS 11.5 package GG 60 (50.4) 59 (49.6) 1.0 (0.7 – 1.5) 0.935 (SPSS Inc, Chicago, IL) for MS Windows and InStat 8 Allele (GraphPad software, San Diego). Comparisons between A 229 (40.2) 214 (37.5) 1.1 (0.9 – 1.4) 0.362 T 341 (59.8) 356 (62.5) median MSSS and genotypes were performed with the non- Genotype parametric Kruskal – Wallis test implemented in the MSSS AA 40 (47.1) 45 (52.9) 0.9 (0.6 – 1.4) 0.557 test program (http://wwwgene.cimr.cam.ac.uk/MSgenetics/ AT 149 (54.6) 124 (45.4) 1.4 (1.0 – 2.0) 0.036 GAMES/MSSS/). Detection of significant departure from TT 96 (45.3) 116 (54.7) 0.7 (0.5 – 1.0) 0.083 Hardy– Weinberg equilibrium was computed with Arlequin a Deviates significantly from Hardy – Weinberg proportions at a = 0.05 2.000 (Schneider et al., 1995) and linkage disequilibrium level; OR = odds ratio; 95% CI = 95% confidence interval.
  4. 4. R. Goertsches et al. / Journal of Neuroimmunology 164 (2005) 140 – 147 143 Table 4 Table 5 Allele and genotype frequency distribution in MS subgroups and controls Median multiple sclerosis severity score (MSSS) in the different genotype SNP Analysis Group 1, n (%) Group 2, n (%) OR (95% CI) p-Value groups 6 Allele RRMS Controls SNP Homozygous Heterozygous Homozygous p-Value C 357 (93.5) 500 (88.0) 1.9 (1.2 – 3.1) 0.006 allele 1 allele 2 T 25 (6.5) 68 (12.0) 1 4.48 4.64 4.79 0.78 Genotype 2 4.88 4.83 4.55 0.71 CC 167 (43.0) 221 (57.0) 2.0 (1.2 – 3.3) 0.008 3 4.72 4.84 4.53 0.79 CT 23 (28.4) 58 (71.6) 0.5 (0.3 – 0.9) 0.017 4 4.32 4.85 4.71 0.92 TT 1 (16.7) 5 (83.3) 0.3 (0.1 – 2.5) 0.237 5 4.84 4.66 4.74 0.98 2a Allele PPMS Controls 6 4.61 5.32 8.24 0.18 A 84 (45.2) 194 (34.5) 1.6 (1.1 – 2.2) 0.009 7 5.19 4.37 4.82 0.12 G 102 (54.8) 368 (65.5) 8 4.87 4.87 4.48 0.72 Genotype AA 27 (33.3) 54 (66.7) 1.7 (1.0 – 2.9) 0.046 AG 30 (25.9) 86 (74.1) 1.1 (0.7 – 1.8) 0.765 significant deviation for SNP2. Haplotype block identifica- GG 36 (20.3) 141 (79.7) 0.6 (0.4 – 1.0) 0.055 tion with reduced recombination was inferred from D¶ 3 Allele values for each marker pair and generated a block of 14.169 C 73 (38.4) 218 (38.3) 1.0 (0.7 – 1.4) 0.966 kb, starting at SNP3 and ending with SNP5 (Fig. 1). G 117 (61.6) 352 (61.7) Genotype CC 9 (15.3) 50 (84.7) 0.5 (0.2 – 1.0) 0.060 3.2. Distribution of ADAMTS14 alleles and genotypes and CG 55 (31.8) 118 (68.2) 2.0 (1.2 – 3.1) 0.005 susceptibility to MS GG 31 (20.9) 117 (79.1) 0.7 (0.4 – 1.1) 0.145 2a Allele PPMS RRMS Eight SNPs spanning 74.9 kb were genotyped in 287 MS A 84 (45.2) 122 (32.3) 1.7 (1.2 – 2.5) 0.003 patients and 285 healthy controls (Fig. 1). The allele G 102 (54.8) 256 (67.7) Genotype frequencies obtained for each marker were in accordance AA 27 (44.3) 34 (55.7) 1.9 (1.0 – 3.3) 0.034 with the publicly stated minor allele frequencies (MAFs) AG 30 (35.7) 54 (64.3) 1.2 (0.7 – 2.0) 0.525 represented in Table 2. We first compared the overall allele GG 36 (26.3) 101 (73.7) 0.6 (0.3 – 0.9) 0.020 and genotype frequencies for the 8 SNPs between MS a Deviates significantly from Hardy – Weinberg proportions at a = 0.05 patients and controls. As shown in Table 3, comparisons of level; OR = odds ratio; 95% CI = 95% confidence interval. the allele frequencies at SNP6 revealed a significant association of allele C with the MS group (OR = 1.7; 95% unphased genotypes by using version 2.0.2 of the PHASE CI = 1.2 to 2.6; p = 0.008). At the genotype level, MS was program (Stephens et al., 2001, 2003). Possible uncertainty significantly associated with CC homozygosities at SNP6 in the reconstruction stage was compensated with exclusion (OR = 1.8; 95% CI = 1.1 to 2.7; p = 0.011), and CG hetero- of haplotype pairs not exceeding the threshold set at 90%, zygosities at SNP3 (OR = 1.6; 95% CI = 1.2 to 2.2; p = 0.006). leaving for analyses 534 control haplotypes and 546 MS Conversely, MS was negatively associated with CC homo- haplotypes, the latter comprising of 368 RRMS and 178 zygosity at SNP3 (OR = 0.5; 95% CI = 0.3 to 0.8; p = 0.007). PPMS haplotypes. Table 6 2.5. Bioinformatic functional validation of implicated Reconstructed haplotypes from five variable sites and their relative genetic variation frequency H 23456 Frequencies, n (%) OR (95% CI) p SNPs 2, 3, and 6 have been stated in the public database MS patients Controls dbSNP as rs4747075, rs7081273, and rs4746060 in the 1 ACGGC 147 (26.9) 152 (28.5) 0.9 (0.7 – 1.2) 0.553 respective order. These IDs were introduced into PupaSNP 2 GGGAC 146 (26.7) 127 (23.8) 1.2 (0.9 – 1.6) 0.249 (Conde et al., 2004) software that retrieves comprehensive 3 GGGGC 78 (14.3) 79 (14.8) 1.0 (0.7 – 1.4) 0.832 information from the Ensembl genome assembly version 4 GGAGC 45 (8.2) 45 (8.4) 1.0 (0.6 – 1.5) 0.927 5 GCGGC 41 (7.5) 42 (7.9) 1.0 (0.6 – 1.5) 0.840 19.34b (NCBI build 34 genome). 6 GGAGTa 17 (3.1) 39 (7.3) 0.4 (0.2 – 0.7) 0.002 7 AGGGCa 55 (10.1) 29 (5.4) 2.0 (1.2 – 3.1) 0.004 8 GCGGT 8 (1.5) 12 (2.3) 0.7 (0.3 – 1.6) 0.375 3. Results 9 GGGGT 7 (1.3) 3 (0.6) 2.3 (0.6 – 9.0) 0.342 10 AGGAC 1 (0.2) 4 (0.8) 0.2 (0.0 – 2.2) 0.214 11 AGGGT 1 (0.2) 2 (0.4) 0.5 (0.0 – 5.4) 0.621 3.1. Hardy –Weinberg equilibrium tests and linkage dis- a After stratifying the MS group, comparisons between RRMS patients equilibrium (LD) strength between SNPs and controls (H6), and between PPMS patients and controls (H7) were statistically significant ( p = 0.0026 and p = 0.0004 respectively). H: Test of Hardy –Weinberg (HW) equilibrium was carried haplotype number; numbers 2 through 6 represent SNP position; OR = odds out for all SNPs in the control population and showed ratio; 95% CI = 95% confidence interval.
  5. 5. 144 R. Goertsches et al. / Journal of Neuroimmunology 164 (2005) 140 – 147 We next analysed allele and genotype frequencies in CI = 0.3 to 0.9; p = 0.02) when contrasted to RRMS. controls and MS patients split by the clinical form (Table Finally, CG heterozygosity was associated with PPMS 4). Comparisons between RRMS patients and controls compared with controls at SNP3 (OR = 2.0; 95% CI = 1.2 revealed significant association of both allele C (OR = 1.9; to 3.1; p = 0.005). 95% CI = 1.2 to 3.1; p = 0.006) and CC homozygosity (OR = 2.0; 95% CI = 1.2 to 3.3; p = 0.008) with RRMS at 3.3. ADAMTS14 genotypes and severity of MS SNP6. PPMS was associated with allele A at SNP2 compared with RRMS (OR = 1.7; 95% CI = 1.2 to 2.5; As described in Patients and Methods, disability was p = 0.003) and controls (OR = 1.6; 95% CI = 1.1 to 2.2; assessed by the MSSS. Overall comparisons of median p = 0.009). On the genotype level PPMS was negatively scores between genotypes in MS patients did not reach associated with SNP2 GG homozygosity (OR = 0.6; 95% statistical significance at any of the 8 SNP loci (Table 5). Control 24 Haplotype pairs (%) 20 16 12 8 4 0 1/1 1/2 1/3 1/4 1/5 1/6 1/7 1/8 1/10 1/11 2/2 2/3 2/4 2/5 2/6 2/7 2/8 2/9 2/10 3/3 3/4 3/5 3/6 3/7 4/4 4/5 4/6 5/5 5/8 6/6 6/8 7/7 7/11 8/9 MS 24 Haplotype pairs (%) 20 16 12 8 4 0 1/1 1/2 1/3 1/4 1/5 1/6 1/7 1/8 1/10 1/11 2/2 2/3 2/4 2/5 2/6 2/7 2/8 2/9 2/10 3/3 3/4 3/5 3/6 3/7 4/4 4/5 4/6 5/5 5/8 6/6 6/8 7/7 7/11 8/9 RRMS 24 Haplotype pairs (%) 20 16 12 8 4 0 1/10 1/11 2/10 7/11 1/1 1/2 1/3 1/4 1/5 1/6 1/7 1/8 2/2 2/3 2/4 2/5 2/6 2/7 2/8 2/9 3/3 3/4 3/5 3/6 3/7 4/4 4/5 4/6 5/5 5/8 6/6 6/8 7/7 8/9 PPMS 24 Haplotype pairs (%) 20 16 12 8 4 0 1/10 1/11 2/10 7/11 1/1 1/2 1/3 1/4 1/5 1/6 1/7 1/8 2/2 2/3 2/4 2/5 2/6 2/7 2/8 2/9 3/3 3/4 3/5 3/6 3/7 4/4 4/5 4/6 5/5 5/8 6/6 6/8 7/7 8/9 Fig. 2. Distribution of haplotype pairs in controls and MS subgroups. Haplotypes were computationally reconstructed and assembled as pairs. The x-axis shows haplotype combinations and the y-axis (bars) represents frequencies of haplotype pairs in the controls and MS subgroups. Solid bars indicate haplotype pairs displaying statistically significant differences when compared between groups: MS vs. controls (for 1/1, p = 0.026; 1/7, p = 0.01; and 3/6, p = 0.026); RRMS vs. controls (3/6, p = 0.004); PPMS vs. controls (1/7, p = 0.0009).
  6. 6. R. Goertsches et al. / Journal of Neuroimmunology 164 (2005) 140 – 147 145 3.4. Distribution of ADAMTS14 haplotypes in MS patients association in their genotype and allele frequencies with and healthy controls MS (SNPs 2, 3, and 6). Furthermore, by means of haplotype inference we were able to ascertain haplotype Haplotypes were computationally reconstructed consid- pairs that conferred disease risk (pair 1/7 = ACGGC/ ering variations at five sites (SNPs 2 through 6) in the AGGGC) or protection (pair 1/1 = ACGGC/ACGGC; pair ADAMTS14 gene, including those variants that revealed 3/6 = GGGGC/GGAGT) for the disease. association in preceding allele and genotype analysis. From Genes encoding metalloproteinases are attractive can- 32 possible combinations of these SNPs, the most common didates to study MS due to their role in brain extracellular 11 haplotype states were included for further frequency and matrix (ECM) cleavage and regulation of inflammation distribution studies. and acquired immunity (Romanic and Madri, 1994; Parks Comparisons of haplotype frequencies between the et al., 2004). Normal ECM is in a state of dynamic whole MS group and controls are represented in Table 6. equilibrium, accounting for a stability between synthesis MS was significantly associated with the AGGGC haplo- and degradation. For the degradative process there is a type (H7) (OR = 2.0; 95% CI = 1.2 to 3.1; p = 0.004), and balance between disintegrating proteinases and the negatively associated with the GGAGT haplotype (H6) corresponding inhibitors. It is believed that in MS a (OR = 0.4; 95% CI = 0.2 to 0.7; p = 0.002). disruption of this balance in favor of proteolysis leads to When haplotype frequencies were compared between pathologic brain ECM destruction and MRI lesions controls and MS subgroups, a significant association was (Waubant et al., 1999). Even so, roles of metalloprotei- found between PPMS and H7 compared with controls nases regulating axon elongation or facilitating process (OR = 2.7; 95% CI = 1.5 to 4.8; p = 0.0004). RRMS was outgrowth and remyelination by oligodendrocytes suggest negatively associated with H6 compared with controls that metalloproteinase activity may also act beneficially in (OR = 0.4; 95% CI = 0.2 to 0.7; p = 0.003). the disease (Larsen et al., 2003; Hayashita-Kinoh et al., Fig. 2 represents the distribution of haplotypes assembled 2002; Uhm et al., 1998). as pairs. MS was significantly associated with pair 1/7 To date, metalloproteinases (MMP-2 and -9), their (OR = 2.0; 95% CI = 1.2 to 3.5; p = 0.01) and negatively inhibitors (TIMPs), and members of the A Disintegrin associated with pairs 1/1 (OR = 0.5; 95% CI = 0.2 to 0.9; And Metalloproteinase family (ADAM-10 and -17) have p = 0.026) and 3/6 (OR = 0.1; 95% CI = 0.0 to 0.7; p = 0.003) been reported to play a role in MS pathogenesis (Leppert when compared with the control population. Finally, et al., 1998; Kouwenhoven et al., 2001; Galboiz et al., distribution of haplotypes in MS subgroups showed strong 2001; Avolio et al., 2003; Seifert et al., 2002; Kieseier et association between pair 1/7 and PPMS (OR = 3.3; 95% al., 2003). Nevertheless, association studies between gene CI = 1.7 to 6.4; p = 0.0009), and negative association metalloproteinase polymorphisms and MS are scanty and between pair 3/6 and RRMS (OR = 0.1; 95% CI = 0.0 to nearly confined to MMP-9 with discordant results (Chat- 1.0; p = 0.004) when compared with controls. away et al., 1999; Nelissen et al., 2000a,b; Fiotti et al., 2004). 3.5. Bioinformatic functional validation of implicated A presumably complementing role in controlling brain genetic variations matrix structure and organization has been ascribed to the novel protease family A Disintegrin And Metalloproteinase The program PupaSNP did not assign functional Domain with Thrombospondin Motifs (ADAMTS) (Kuno et properties for marker 2, 3, or 6. al., 1997; Kuno and Matsushima, 1998; Kuno et al., 1999). ADAMTS-14 (OMIM*607506) (Tang, 2001; Bolz et al., 2001; Colige et al., 2002; Cal et al., 2002) is a member of a 4. Discussion structurally and functionally distinct subfamily of ADAMTS proteases (ADAMTS-2 and -3) and has been shown to be This report indicates a possible genetic role for synthesized as a latent enzyme that requires cell type- ADAMTS14 in the MS pathogenesis. An initial genome regulated activation to display aminoprocollagen peptidase wide screen disclosed genomic regions containing clusters activity (Colige et al., 2002). of microsatellite markers displaying evidence of associa- The physiological function of the ADAMTS proteases in tion with MS (Goertsches et al., 2003). Individual SNP the nervous system and their possible implication in genotyping focused on the 10q22.1 region was performed pathological processes such as MS has yet to be elucidated. in 192 MS patients and 191 healthy controls, and showed Nevertheless, evidence for a potential role of these proteases significant association of the ADAMTS14 gene with MS. in the biology and pathology of the CNS comes from their Further analysis with an increased sample size of 287 MS decomposing activity on several proteoglycans enriched in patients and 285 controls resulted in a more robust the nervous system, and from the expression of various confirmation of the previous findings and pointed the members of this family (ADAMTS13 , ADAMTS14, ADAMTS14 as a candidate gene for MS susceptibility. In ADAMTS16, and ADAMTS18) in the human brain (Cal et the present study, 3 out of 8 SNPs showed significant al., 2002).
  7. 7. 146 R. Goertsches et al. / Journal of Neuroimmunology 164 (2005) 140 – 147 Haplotype studies are becoming essential to association References analysis of candidate genes. In order to perform haplotype Allen, M., Sandberg-Wollheim, M., Sjogren, K., Erlich, H.A., Petterson, U., analysis in a population-based case – control study, haplo- Gyllensten, U., 1994. Association of susceptibility to multiple sclerosis types must be determined by estimation in the absence of in Sweden with HLA class II DRB1 and DQB1 alleles. Hum. Immunol. family information or laboratory methods for establishing 39, 41 – 48. phase. We used a computer-based algorithm implementing Avolio, C., Ruggieri, M., Giuliani, F., Liuzzi, G.M., Leante, R., Riccio, P., a Bayesian approach to infer phase information (Stephens Livrea, P., Trojano, M., 2003. Serum MMP-2 and MMP-9 are elevated in different multiple sclerosis subtypes. J. Neuroimmunol. 136, 46 – 53. et al., 2001, 2003). This approach is not based on HW Barcellos, L.F., Oksenberg, J.R., Begovich, A.B., Martin, E.R., Schmidt, S., equilibrium, which was violated in our study by the Vittinghoff, E., Goodin, D.S., Pelletier, D., Lincoln, R.R., Bucher, P., finding of significant departure in SNP2. The overall Swerdlin, A., Pericak-Vance, M.A., Haines, J.L., Hauser, S.L., 2003. haplotype distribution was different between MS patients Multiple sclerosis genetics group. HLA-DR2 dose effect on suscepti- and controls. Two of the 11 haplotypes analysed (H6, H7) bility to multiple sclerosis and influence on disease course. Am. J. Hum. Genet. 72, 710 – 716. were particularly different in their frequencies among these Barrett, J.C., Fry, B., Maller, J., Daly, M.J., 2005. Haploview: analysis two groups and strengthened the contribution of SNPs 2, and visualization of LD and haplotype maps. Bioinformatics 21, 3, and 6 to the association between the over- and 263 – 265. undertransmitted haplotypes and MS. Of note, H7 was a Becker, T., Knapp, M., 2004. A powerful strategy to account for multiple risk haplotype for MS in PPMS patients, and H6 was testing in the context of haplotype analysis. Am. J. Hum. Genet. 75, 561 – 570. associated with reduced risk in RRMS patients. Further- Bolz, H., Ramirez, A., von Brederlow, B., Kubisch, C., 2001. Character- more, when haplotypes were assembled as pairs, combi- ization of ADAMTS14, a novel member of the ADAMTS metallopro- nation of H7 and H1 was strongly associated with PPMS, teinase family. Biochim. Biophys. Acta 1522, 221 – 225. whereas combination of H6 and H3 was unseen in RRMS Cal, S., Obaya, A.J., Llamazares, M., Garabaya, C., Quesada, V., Lopez- (Fig. 2). The presence of different haplotypes conferring Otin, C., 2002. Cloning, expression analysis, and structural charac- terization of seven novel human ADAMTSs, a family of metallopro- risk for MS in patients with PPMS and RRMS may teinases with disintegrin and thrombospondin-1 domains. Gene 283, contribute to the heterogeneity found between these two 49 – 62. groups as there exists general agreement that patients with Chataway, J., Sawcer, S., Feakes, R., Coraddu, F., Broadley, S., Jones, H.B., PPMS differ significantly according to epidemiology, CNS Clayton, D., Gray, J., Goodfellow, P.N., Compston, A., 1999. A screen histopathology, neuroimaging findings, and response to of candidates from peaks of linkage: evidence for the involvement of myeloperoxidase in multiple sclerosis. J. Neuroimmunol. 98, 208 – 213. treatment in comparison with RRMS patients (Thompson Colige, A., Vandenberghe, I., Thiry, M., Lambert, C.A., Van Beeumen, J., et al., 1997). It is yet to be explored whether these Li, S.W., Prockop, D.J., Lapiere, C.M., Nusgens, B.V., 2002. Cloning haplotypes are related to different levels of gene expres- and characterization of ADAMTS-14, a novel ADAMTS displaying sion or functional changes in the encoded protein. high homology with ADAMTS-2 and ADAMTS-3. J. Biol. Chem. 277, SNPs in introns have been demonstrated in various 5756 – 5766. Conde, L., Vaquerizas, J.M., Santoyo, J., Al-Shahrour, F., Ruiz-Llorente, S., studies to display important regulatory features with Robledo, M., Dopazo, J., 2004. PupaSNP Finder: a web tool for finding phenotypic effect, such as an alteration in a binding site SNPs with putative effect at transcriptional level. Nucleic Acids Res. for a transcription factor in autoimmune diseases (Prokunina 32, 242 – 248. et al., 2002; Tokuhiro et al., 2003; Helms et al., 2003). Compston, A., Coles, A., 2002. Multiple sclerosis. Lancet 359, 1221 – 1231. Furthermore, significant variation at intron –exon border Dyment, D.A., Ebers, G.C., Sadovnick, A.D., 2004. Genetics of multiple sclerosis. Lancet Neurol. 3, 104 – 110. consensus sequences can lead to incorrect processing of a Fiotti, N., Zivadinov, R., Altamura, N., Nasuelli, D., Bratina, A., Tommasi, gene. Unfortunately, the operation of the bioinformatic M.A., Bosco, A., Locatelli, L., Grop, A., Cazzato, G., Guarnieri, G., programme PupaSNP did not generate such information for Giansante, C., Zorzon, M., 2004. MMP-9 microsatellite polymorphism associated SNPs 2, 3, and 6. and multiple sclerosis. J. Neuroimmunol. 152, 147 – 153. In summary, the observations reported here support the Galboiz, Y., Shapiro, S., Lahat, N., Rawashdeh, H., Miller, A., 2001. Matrix metalloproteinases and their tissue inhibitors as markers of disease hypothesis that polymorphisms within the ADAMTS14 gene subtype and response to interferon-beta therapy in relapsing and may influence genetic predisposition for MS. Further secondary progressive multiple sclerosis patients. Ann. Neurol. 50, studies in other MS cohorts will be required to confirm 443 – 451. the association of ADAMTS14 haplotypes with MS. Finally, Goertsches, R., Villoslada, P., Comabella, M., Montalban, X., Navarro, functional in vitro and in vivo studies are needed to A., de la Concha, E.G., Arroyo, R., Lopez de Munain, A., Otaegui, D., Palacios, R., Perez-Tur, J., Jonasdottir, A., Benediktsson, K., investigate differential gene expression patterns related to Fossdal, R., Sawcer, S., Setakis, E., Compston, A., Spanish MS risk or protective haplotypes in PPMS and RRMS patients. Genetics Group, 2003. A genomic screen of Spanish multiple sclerosis patients reveals multiple loci associated with the disease. J. Neuroim- munol. 143, 124 – 128. Acknowledgement Hauser, S.L., Goodkin, S.L., 2001. Multiple sclerosis and other demyelin- ating diseases. In: Braunwald, E., Fauci, A.D., Kasper, D.L., Hauser, S.L. (Eds.), Harrison’s Principle of Internal Medicine, 15th edR McGraw This work represents part of the dissertation of Robert Hill, New York. Goertsches at the Heinrich Heine Universitat Dusseldorf, ¨ ¨ Hayashita-Kinoh, H., Kinoh, H., Okada, A., Komori, K., Itoh, Y., Chiba, T., Germany. Kajita, M., Yana, I., Seiki, M., 2002. Membrane-type 5 matrix
  8. 8. R. Goertsches et al. / Journal of Neuroimmunology 164 (2005) 140 – 147 147 metalloproteinase is expressed in differentiated neurons and regulates Romanic, A.M., Madri, J.A., 1994. Extracellular matrix-degrading protei- axonal growth. Cell Growth Differ. 12, 573 – 580. nases in the nervous system. Brain Pathol. 4, 145 – 156. Helms, C., Cao, L., Krueger, J.G., Wijsman, E.M., Chamian, F., Gordon, Rosenberg, G.A., 2002. Matrix metalloproteinases and neuroinflammation D., Heffernan, M., Daw, J.A., Robarge, J., Ott, J., Kwok, P.Y., Menter, in multiple sclerosis. Neuroscientist 8, 586 – 595. A., Bowcock, A.M., 2003. A putative RUNX1 binding site variant Sadovnick, A.D., Armstrong, H., Rice, G.P., Bulman, D., Hashimoto, between SLC9A3R1 and NAT9 is associated with susceptibility to L., Paty, D.W., Hashimoto, S.A., Warren, S., Hader, W., Murray, psoriasis. Nat. Genet. 35, 349 – 356. T.J., Seland, T.P., Metz, L., Bell, R., Duquette, P., Gray, T., Nelson, Kieseier, B.C., Pischel, H., Neuen-Jacob, E., Tourtellotte, W.W., Hartung, R., Weinshenker, B., Brunt, D., Ebers, G.C., 1993. A population- H.P., 2003. ADAM-10 and ADAM-17 in the inflamed human CNS. based study of multiple sclerosis in twins: update. Ann. Neurol. 33, Glia 42, 398 – 405. 281 – 285. Kouwenhoven, M., Ozenci, V., Gomes, A., Yarilin, D., Giedraitis, V., Press, Sawcer, S., Compston, A., 2003. The Genetic Analysis of Multiple R., Link, H., 2001. Multiple sclerosis: elevated expression of matrix sclerosis in EuropeanS: concepts and design. J. Neuroimmunol. 143, metalloproteinases in blood monocytes. J. Autoimmun. 16, 463 – 470. 13 – 16. Kuno, K., Matsushima, K., 1998. ADAMTS-1 protein anchors at the Schneider, S. Roessli, D., Excoffier, L., 1995. Arlequin ver. 2000: a extracellular matrix through the thrombospondin type I motifs and its software for population genetics data analysis. University of spacing region. J. Biol. Chem. 273, 13912 – 13917. Geneva, Switzerland, Genetics and Biometry Laboratory, 2000. 12, Kuno, K., Kanada, N., Nakashima, E., Fujiki, F., Ichimura, F., Mat- 921 – 927. sushima, K., 1997. Molecular cloning of a gene encoding a new Seifert, T., Kieseier, B.C., Ropele, S., Strasser-Fuchs, S., Quehenberger, F., type of metalloproteinase-disintegrin family protein with thrombos- Fazekas, F., Hartung, H.P., 2002. TACE mRNA expression in peripheral pondin motifs as an inflammation associated gene. J. Biol. Chem. mononudear cells precedes new lesions on MRI in multiple sclerosis. 272, 556 – 562. Mult. Scler. 8, 447 – 451. Kuno, K., Terashima, Y., Matsushima, K., 1999. ADAMTS-1 is an active Stephens, M., Donnelly, P., 2003. A comparison of bayesian methods for metalloproteinase associated with the extracellular matrix. J. Biol. haplotype reconstruction from population genotype data. Am. J. Hum. Chem. 274, 18821 – 18826. Genet. 73, 1162 – 1169. Larsen, P.H., Wells, J.E., Stallcup, W.B., Opdenakker, G., Yong, V.W., 2003. Stephens, M., Smith, N.J., Donnelly, P., 2001. A new statistical method for Matrix metalloproteinase-9 facilitates remyelination in part by proces- haplotype reconstruction from population data. Am. J. Hum. Genet. 68, sing the inhibitory NG2 proteoglycan. J. Neurosci. 23, 11127 – 11135. 978 – 989. Leppert, D., Ford, J., Stabler, G., Grygar, C., Lienert, C., Huber, S., Miller, Tang, B.L., 2001. ADAMTS: a novel family of extracellular matrix K.M., Hauser, S.L., Kappos, L., 1998. Matrix metalloproteinase-9 proteases. Int. J. Biochem. Cell Biol. 33, 33 – 44. (gelatinase B) is selectively elevated in CSF during relapses and stable Thompson, A.J., Polman, C.H., Miller, D.H., McDonald, W.I., Brochet, B., phases of multiple sclerosis. Brain 121, 2327 – 2334. Filippi, M., Montalban, X., De Sa, J., 1997. Primary progressive Nelissen, I., Vandenbroeck, K., Fiten, P., Hillert, J., Olsson, T., Marrosu, multiple sclerosis. Brain 120, 1085 – 1096. M.G., Opdenakker, G., 2000a. Polymorphism analysis suggests that the Tokuhiro, S., Yamada, R., Chang, X., Suzuki, A., Kochi, Y., Sawada, T., gelatinase B gene is not a susceptibility factor in multiple sclerosis. J. Suzuki, M., Nagasaki, M., Ohtsuki, M., Ono, M., Furukawa, H., Neuroimmunol. 105, 58 – 63. Nagashima, M., Yoshino, S., Mabuchi, A., Sekine, A., Saito, S., Nelissen, I., Fiten, P., Vandenbroeck, K., Hillert, J., Olsson, T., Marrosu, Takahashi, A., Tsunoda, T., Nakamura, Y., Yamamoto, K., 2003. An M.G., Opdenakker, G., 2000b. PECAM1, MPO and PRKAR1A at intronic SNP in a RUNX1 binding site of SLC22A4, encoding an chromosome 17q21 – q24 and susceptibility for multiple sclerosis in organic cation transporter, is associated with rheumatoid arthritis. Nat. Sweden and Sardinia. J. Neuroimmunol. 108, 153 – 159. Genet. 35, 341 – 348. Oksenberg, J.R., Baranzini, S.E., Barcellos, L.F., Hauser, S.L., 2001. Uhm, J.H., Dooley, N.P., Oh, L.Y., Yong, V.W., 1998. Oligodendrocytes Multiple sclerosis: genomic rewards. J. Neuroimmunol. 113, 171 – 184. utilize a matrix metalloproteinase, MMP-9, to extend processes along an Parks, W.C., Wilson, C.L., Lopez-Boado, Y.S., 2004. Matrix metallopro- ´ astrocyte extracellular matrix. Glia 22, 53 – 63. teinases as modulators of inflammation and innate immunity. Nat. Rev., Villoslada, P., Barcellos, L.F., Rio, J., Begovich, A.B., Tintore, M., Sastre- Immunol. 4, 617 – 629. Garriga, J., Baranzini, S.E., Casquero, P., Hauser, S.L., Montalban, X., Poser, C.M., Paty, D.W., Scheinberg, L., McDonald, W.I., Davis, F.A., Oksenberg, J.R., 2002. The HLA locus and multiple sclerosis in Spain. Ebers, G.C., Johnson, K.P., Sibley, W.A., Silberberg, D.H., Tourtellotte, Role in disease susceptibility, clinical course and response to interferon- W.W., 1983. New diagnostic criteria for multiple sclerosis: guidelines beta. J. Neuroimmunol. 130, 194 – 201. for research protocols. Ann. Neurol. 13, 227 – 231. Waubant, E., Goodkin, D.E., Gee, L., Bacchetti, P., Sloan, R., Stewart, T., Prokunina, L., Castillejo-Lopez, C., Oberg, F., Gunnarsson, I., Berg, L., Andersson, P.B., Stabler, G., Miller, K., 1999. Serum MMP-9 and Magnusson, V., Brookes, A.J., Tentler, D., Kristjansdottir, H., Grondal, TIMP-1 levels are related to MRI activity in relapsing multiple G., Bolstad, A.I., Svenungsson, E., Lundberg, I., Sturfelt, G., Jonssen, sclerosis. Neurology 53, 1397 – 1401. A., Truedsson, L., Lima, G., Alcocer-Varela, J., Jonsson, R., Gyllensten, Weinshenker, B.G., Santrach, P., Bissonet, A.S., McDonnell, S.K., Schaid, U.B., Harley, J.B., Alarcon-Segovia, D., Steinsson, K., Alarcon- D., Moore, S.B., Rodriguez, M., 1998. Major histocompatibility Riquelme, M.E., 2002. A regulatory polymorphism in PDCD1 is complex class II alleles and the course and outcome of MS: a associated with susceptibility to systemic lupus erythematosus in population-based study [see comments]. Neurology 51, 742 – 747. humans. Nat. Genet. 32, 666 – 669. Rice, W.R., 1989. Analyzing tables of statistical tests. Evolution 43, 223 – 225.

×