320 likes | 484 Views
Leprosy as a human model for the genetic study of common infectious diseases. Erwin Schurr Centre for the Study of Host Resistance Departments of Human Genetics and Medicine McGill University. Spectrum of genetic predisposition. Casanova, Abel EMBO J 2007.
E N D
Leprosy as a human model for the genetic study of common infectious diseases Erwin Schurr Centre for the Study of Host Resistance Departments of Human Genetics and Medicine McGill University
Spectrum of genetic predisposition Casanova, Abel EMBO J 2007
Complexity: strength of genetic effect Relative Risk: P(affected/DD)/P(affected/dd) Number of genes Modest Effect Major Effect….. Mendelian Effect RR 1 2 5 10 100 polygenic “major genes” PIDs Two issues: (i) Functional validation/causality (ii) Population effect One big question: What proportion of the host genetics component of disease susceptibility is explained by these different effect classes
Tracking of major genetic effects by modelling gene-environment interactions or by understanding genetic heterogeneityExample: NRAMP1 candidate gene in tuberculosis Major effects
Major effects Tracking of major genetic effects by linkage based genome scanning in tuberculosis and leprosy
Leprosy • Chronic human infectious disease • Mycobacterium leprae • Infection rate unknown • Primary tropism • Macrophages • Schwann cells
Paucibacillary (Cell mediated) No disease Leprosy per se Single lesion Multibacillary (Antibody mediated) Clinical spectrum Infection
Chromosome 6 6p21 LOD = 2.7 Chromosome 10 Chromosome 13 10p13 LOD = 2.0 (Paucibacillary) 13q13 LOD = 1.4 6q25 – q27 LOD = 4.3 Genome scan, 2003 • 86 Vietnamese multiplex families • 388 markers tested for linkage with leprosy per se Mira et al., 2003
rs1040079 High Resolution Association Mapping Gene Map: 2 Genes SNP Map: 81 SNPs Association Plot PARK2_e01(-2599) Association Map 0.84/0.16 0.66/0.34 0.51/0.49 Mira et al Nature 2004
SNP1 SNP2
Chromosome region 6p21 • Objective: Identify susceptibility variants located in the linked chromosomal region on 6p21 following the same strategy used for positional cloning of PARK2/PACRG variants, i.e. systematic association scanning of markers located in the linkage peak confidence interval POSTER # 64 Andrea Alter et al.
Family based transmission disequilibrium test Case – control Study design II A1/A2 A2/A2 A1/A2 Test statistic = Σall cells (observed -expected)2 / expected Χ2 distribution Logistic regression a = # of A1 transmissions/not A2 b = # of A2 transmissions/not A1 Test statistic = (a - b)2/(a + b) Χ2 distribution Conditional logistic regression
Low resolution association scan 418 SNPs - 55 failed 363 SNPs - 44 non informatifs: f(a)<0.05 319 SNPs - 12 non Hardy Weinberg 307 SNPs over 10 Mb
Low resolution association scan Linkage peak 108 genes HLA III II a 116 genes b c Non-HLA d e f P = 0.01 -log10P 31.0 33.0 35.0 37.0 39.0 41.0 [Mb] Chromosome 6 Position Alcais et al Nat Genet 2007
a c -log10P PPIAP9 MCCD1 ATP6V1G2 LTA TNF LTB LST1 10 genes BAT1 NFKBIL1 NCR3 100 bp 5’UTR Exon 3 Exon 2 LTA LTA+252 LTA+368 LTA-294 LTA-293 LTA+10 LTA+80 rs1041981 Linkage disequilibrium fine mapping of 90kb interval overlapping LTA Alcais et al Nat Genet 2007
ns 1.97 (1.30-2.99) 0.0009 ns 1.74 (1.16-2.60) 0.007 ns 1.63 (1.09-2.43) 0.02 Replication in the Indian sample MCCD1-NS rs2259435 LTA-294 rs2844482 LTA-293 rs2071590 LTA+10 rs1800683 LTA+80 rs2239704 LTA+252 rs909253 LTA+368 rs746868 Vietnamese Bin structure MAF§ 0.16 0.26 0.50 0.30 0.49 0.30 0.16 Univariate OR (95% CI) P ns Multivariate OR (95% CI) P ns ns 1.97 (1.30-2.99) 0.0009 ns ns ns ns Indian Bin structure MAF§ 0.23 0.22 0.26 0.28 0.42 0.28 0.41 Univariate OR (95% CI) P 1.87 (1.37-2.57) 0.00009 1.78 (1.29-2.45) 0.0004 ns ns ns ns ns Multivariate OR (95% CI) P 1.82 (1.38-2.41) 0.00003 ns* ns 1.60 (1.10-2.33) 0.01 ns ns * ns Any SNP in the bin is sufficient to explain the observed association with leprosy § MAF (Minor Allele Frequency) corresponds to the frequency of the risk allele for each of the associated SNPs
Additive Dominant Recessive P [OR(95% CI)] P [OR(95% CI)] P [OR(95% CI)] LTA+80 MAF:0.414 0.79 [0.96 (0.762-1.209) 0.95[1.01(0.72-1.43)] 0.48[0.86(0.57-1.30)] Replication in Brazilian sample No replication in the Brazilian sample: WHY? Alcais et al Nat Genet 2007
The effect of LTA+80 is strongly age-dependent Age [years] 0 - 15 16 - 25 26 - 35 >35 All ages 20 10 5 Odds ratio [95%CI] 2 1 0 10 20 Proportion of cases (%) 30 40 50 Alcais et al Nat Genet 2007
Classical HLA I genes and HLA-DRB1 DRB1 HLA II HLA III 123 SNPs, 94 samples, >95% non- single SNP bins LTA HLA I HLA-C HLA-B
HLA class II and leprosy in Vietnam Vanderborght et al Genes Immun 2007 April, epub
Effect of HLA class II in Vietnamese Similar results were obtained for 5 other LTA markers India: LTA is not in LD with HLA – DRB1, HLA – DQA1 and HLA – DQB1 Brazil: association of LTA is also independent of HLA – DRB1
Linkage disequilibrium between LTA+80 and class I alleles R2 was not calculated for f(a)<0.05 Class I alleles were determined for 37 Vietnamese individuals Alcais et al Nat Genet 2007
Functional role for LTA +80 • Knight et al. 2002 Nat Genet • ABF – 1 • Transcriptional repressor • Lymphoid tissue specific • Binds only if ‘A’ allele at LTA+80 • Consistent with ‘A’ allele as risk factor for leprosy E47 ABF – 1 Luciferase reporter TACCGCCCAGCAGTGTCCTG E2 – box: CAGCTG Luciferase reporter TACCGCCCCGCAGTGTCCTG
Take home lesson • The unit of replication in genetic association studies is the “bin” not the “SNP” (recommended even if functional SNPs are being tested) • Multivariate analysis is obligatory • Proper consideration of non-genetic covariates can be essential for successful replication
Detection of Granuloma-Forming Capacity in Leprosy “Mitsuda’s Test” (1919) Antigen: Lepromin Heat-killed M. leprae from human or armadillo (1.6 x 108 bacilli/ml) 0.1 ml intradermal injection DTH (24-48 hrs) granuloma (3-4 wks) Mitsuda reaction Fernandez reaction < Specificity < + = CMI, good prognosis _ = ?? >10mm = pos., CMI, good prognosis <3mm = neg., anergy, poor prognosis
Mitsuda distribution Patients Controls POSTER # 105 Ranque et al JID 2007 in press
Mitsuda – Why the fuss? • The Mitsuda test is an in-vivo assay for the innate capacity of a person to form an infectious granuloma • Efficient formation of infectious granuloma is an important defense against numerous infectious diseases • In tuberculosis, inability to maintain granuloma underlies reactivational tuberculosis, the disease type that mainly contributes to transmission of the disease
Genome scan in 20 multiplex families POSTER # 105 Ranque et al JID 2007, in press
Acknowledgments Université de Paris René Descartes, INSERM U550 France B Ranque G Antoni A Alcaïs L Abel McGill Centre for Study of Host Resistance, Canada A Alter L Simkin L de Léséleuc A Verville M Girard M Mira McGill University and Genome Quebec Innovation Centre, Canada TJ Hudson A Montpetit P Lepage A Bélisle All India Institute of Medical Sciences + JALMA, India NK Mehra M Singh K Katoch Oswaldo Cruz Institute, Brazil MO Moraes P Vanderborght Hospital for Dermato-Venereology, Vietnam N Van Thuc V Hong Thai N Thu Huong N Ngoc Ba SUPPORT: CIHR, HHMI, ANR
Related Posters • # 64 Alter et al. • # 73 Gallant et al. • # 105 Ranque et al. • #188 Di Pietrantonio et al.