Identification of Genes involved in Multifactorial Diseases and

Identification of Genes involved in Multifactorial Diseases and Traits Elon Pras Institute of Human Genetics Sheba Medical Center Israel

Multifactorial (complex) inheritance • Inheritance controlled by many genes with small additive effects (polygenic) plus the effects of the environment • The genes and environmental factors causing a particular multifactorial trait may vary from person to person

The contributions of genetic and environmental factors to diseases Familial PAN Rheumatoid Arthritis Polyglandular Deficiency Syndrome Systemic. Lupus Rheumatic Fever Erythematosus GENETIC ENVIRONMENTAL Tuberculosis Familial SLE Rare Genetics simple Unifactorial High recurrence rate Ankylosing Spondylitis Common Genetics complex Multifactorial Low recurrence rate

MYCOBACTERIUM TUBERCULOSIS SUSCEPTIBILITY LOCI • Gene map locus 17 q 11. 2 -q 12, 12 q 14, • 11 q 23 -q 24, 6 q 23 -q 24, 5 q 33. 1, 3 p 21. 3, • 2 q 37. 1, 2 q 35, 20 q 13. 2, 19 p 13. 3

Multifactorial Diseases (Complex Diseases) Diabetes mellitus Hypercholesterolemia Ischemic heart disease Hypertension Asthma Crohns Rheumatoid arthritis Multiple Sclerosis SLE Glaucoma Manic depression Schizophrenia

Multifactorial Traits Height Weight IQ Eye color Skin Color

What evidence do we have for genetic factors in autoimmune diseases • Familial risks (what is the incidence of a disorder in relatives compared with the incidence in the general population? ) • Twin studies (what is the incidence in monozygotic compared with dizygotic twins? ) • Adoption studies (what is the incidence in adopted children of the disorders which their parent had? ) • Population and Migration studies (what is the incidence in people from a particular ancestry group when they move to a different geographical area? )

Multiple Sclerosis (MS) Population prevalence of 0. 1% Risk in siblings of affected individuals is 2 -5% Over 20 fold increase in risk Risk in monozygotic twins is 25 -30% Risk in dizygotic twins is 4%

Determining the incidence of a disease in twins helps delineate whethere are genetic and environmental components Disease Concordance Identical Non-identical (DZ) (MZ) Manic depressive psychosis 67% 5% Rheumatoid arthritis 34% 7% Asthma 47% 24% Coronary artery disease 19% 9% Diabetes mellitus 56% 11% Both genetic and environmental factors important

It is possible to estimate what proportion of the etiology can be related to genetic factors as opposed to environmental factors. This is referred to as heritability Estimating Heritability from Twin Studies: variance in DZ pairs - variance in MZ pairs h 2 = variance in DZ pairs

Estimates of Heritability Disease Heritability Coronary artery disease 0. 65 Asthma 0. 80 Peptic Ulcer 0. 37 Ankylosing spondylitis 0. 70 Hypertension 0. 65 Schizophrenia 0. 85 Both genetic and environmental factors important

Mapping and Identification of Genes involved in Multi-factorial Diseases Candidate gene approach Conventional positional cloning studies Sib pair analysis Genome wide association studies Whole Exome/Genome studies

Candidate Gene Association Studies Gene Y SNP allele G/A 80/20 in patients 50/50 in controls p<0. 001 Gene X Allele A/T 50/50 in patients 50/50 in controls p=1. 00

Ankylosing Spondylitis HLA B 27 Patients 90% Controls 5%

Variation at HLA-DRB 1 is associated with resistance to enteric fever The minor allele of rs 7765379 was under-represented in enteric fever cases (frequency = 1. 04%) Compared to the control subjects (frequency = 5. 5%; OR = 0. 18, P = 4. 5 × 10 -10 ) Imputation-based fine-mapping across the extended MHC region showed that the classical HLA-DRB 1*04: 05 allele (OR = 0. 14, P = 2. 60 × 10) could entirely explain the association at rs 7765379, thus implicating HLA-DRB 1 as a major contributor to resistance against enteric fever, presumably through antigen presentation The minor allele of rs 7765379 that we observed to confer resistance to enteric fever has previously been shown to strongly associate with increased susceptibility to Crohn's disease and rheumatoid arthritis Nature Genetics 46, 1333– 1336 (2014)

Recombination hotspots are widespread and account for LD structure 7 q 21

Only 30% of the candidate gene association studies prove to be right over time

The candidate gene approach

Genome Wide Association Study (GWAS)

Recombination hotspots are widespread and account for LD structure 7 q 21

Whole Genome Association Approach to Common Disease: The View from 2015 • • • Identify an optimum set of 300, 000 tag SNP’s Collect 500 cases and 500 controls Genotype all DNA’s for all SNP’s That adds up to 300 million genotypes Genotyping just dropped to $0. 0005

Variation in the Human IL 1 A Gene IL 1 A in Europeans • 18. 5 kb • 50 SNPs • 46 common SNPs (> 10%MAF) Carlson et al. (2004) Am J Hum Genet. 74: 106 -120. Homozygote common Heterozygote Homozygote alternative allele Missing Data

Replication A Must Replication Hirschhorn & Daly Nat. Genet. Rev. 6: 95, 2005 NCI-NHGRI Working Group on Replication Nature 447: 655, 2007

Manhattan Plot

GWAS for RA in the Japanese Population

SLE GWAS Studies

GWAS MS Study results

GWAS FOR BECHET

Functional Studies

Currently Mapped Type 1 Non-HLA Diabetes Loci and Candidate Causal Genes where Known Current type 1 diabetes loci, their chromosome location, the most disease-associated SNP

The most relevant alleles associated with susceptibility in rheumatoid arthritis according to genome wide association studies Gene candidate SNP OR PTPN 22 Rs 2476601 1. 23 -1. 75 PADI 4 Rs 2240340 a 1. 4 STAT 4 T/C Rs 1188934 1. 22 (0. 98 -1. 53) FCGR 2 A Rs 12746613 1. 1 CTLA 4 Rs 3087243 0. 75 -1. 136 CCL 21 Rs 2812378 1. 1 TRAF 1 Rs 3761847 1. 1 (0. 97 -1. 32) IRF 5 Rs 10488631 1. 16 (0. 72 -1. 87) CCR 6 Rs 3093023 0. 79 (0. 64 -0. 98) CD 40 Rs 4810485 0. 91 -1. 02 IL 2 RA Rs 2104286 0. 92 HLA-DRB World J Orthopv. 5(4); 2014

PTPN 22 - protein tyrosine phosphatase, non-receptor type 22 • • • autoimmune Addison disease Graves disease hashimoto idiopathic inflammatory myopathy juvenile idiopathic arthritis rheumatoid arthritis systemic lupus erythematosus systemic scleroderma type 1 diabetes Alopecia areata Masthenia gravis

The most important non-HLA genetic risk factor in autoimmune disorders PTPN 22 encodes a tyrosine phosphatase, expressed by haematopoietic cells and functions as a key regulator of immune homeostasis Inhibiting T-cell receptor signaling. Promotes the expansion of pathogenic autoinmmune T cells. Alters B cell repertoire. Promotes autoantibody production. Impairs type I interferon production by myeloid cells.

PTPN 22 (R 620 W) The mutation may be conserved in human evolution because it may provide a hyper-immune response to infectious disease A recent study suggests that the R 620 W mutation does not, on a population basis, reduce life span.

Polymorphisms in TNFAIP 3 (A 20) and Human Diseases

Loss-of-function mutations in TNFAIP 3 leading to A 20 • haploinsufficiency cause an early-onset autoinflammatory disease Qing Zhou, Hongying Wang, Daniella M Schwartz, Monique Stoffels, Yong Hwan Park, Yuan Zhang, Dan • Yang, Erkan Demirkaya, Masaki Takeuchi, Wanxia Li Tsai, Jonathan J Lyons, Xiaomin Yu, Claudia Ouyang, Celeste Chen, David T Chin, Kristien Zaal, Settara C Chandrasekharappa, Eric P Hanson, Zhen Yu, James C Mullikin, Sarfaraz A Hasni, Ingrid E Wertz, Amanda K Ombrello, Deborah L Stone, Patrycja Hoffmann, Anne Jones, Beverly K Barham, Helen L Leavis, Annet van Royen-Kerkof, Cailin Sibley, Ezgi D Batu, Ahmet Gül, Richard M Siegel, Manfred Boehm, Joshua D Milner, Seza Ozen, Massimo Gadina, Jae. Jin Chae, Ronald M Laxer, Daniel L Kastner & Ivona Aksentijevich Nature Genetics 2016 •

Increased rate of MEFV heterozygotes Polyarteritis nodosa Ankylosing spondylitis Multiple Sclerosis Ulcerative colitis

Drug Targets

Familial SLE Nat Genet. 2011 Oct 23; 43(12): 1186 -8

Loss-of-function variant in DNASE 1 L 3 causes a familial form of systemic lupus erythematosus Subject ID Age (years) Age of onset (years) Gender SLE activity index Nephritis Serology profile −, NEGATIVE; +, POSITIVE; ANA, ANTI-NUCLEAR ANTIBODY; ANTI-DSDNA, ANTIBODY RECOGNIZING DOUBLE-STRANDED DNA; ANCA, ANTINEUTROPHIL CYTOPLASMIC ANTIBODY. F 1 -A 3. 5 2 F 13 − ANA, anti-ds. DNA, ANCA, low c 3 and c 4 F 1 -B 9. 5 4 F 8 − ANA, anti-ds. DNA, ANCA, low c 3 and c 4 F 1 -C 11 5 M 14 + ANA, anti-ds. DNA, ANCA, low c 3 and c 4 F 2 -A 11 2. 5 F 11 + ANA, anti-ds. DNA, ANCA, low c 3 and c 4, anti-cardiolipin F 2 -B 5 4. 5 M 18 + ANA, anti-ds. DNA, ANCA, low c 3 and c 4, anti-cardiolipin F 2 -C 10 6 F 9 + ANA, anti-ds. DNA, ANCA, low c 3 and c 4, anti-cardiolipin F 2 -D 6 4. 5 M 22 + ANA, anti-ds. DNA, ANCA, low c 3 and c 4, anti-cardiolipin F 3 -A 4. 5 3 M 18 + ANA, anti-ds. DNA, ANCA, low c 3 and c 4 F 3 -B 15 2 M 8 + ANA, anti-ds. DNA, ANCA, low c 3 and c 4 F 4 -A 12 4. 5 F 13 − ANA, anti-ds. DNA, ANCA, low c 3 and c 4 F 4 -B 9 8 F 14 + ANA, anti-ds. DNA, ANCA, low c 3 and c 4 F 5 -A 13 3 M 14 + ANA, anti-ds. DNA, low C 3 and c 4 F 5 -B 10 5 M 11 + ANA, low c 3 and c 4 F 5 -C 22 12 M 12 + ANA, anti-ds. DNA, low C 3 and C 4 F 5 -D 24 11 M 10 – ANA, anti-ds. DNA, low C 3 and C 4, anti-cardiolipin F 6 -A 14 6 M 10 − ANA, anti-ds. DNA, low C 3 and C 4 F 6 -B 12 11 M 9 − ANA, anti-ds. DNA, low C 3 and C 4

Rare variants in the CYP 27 B 1 gene are associated with multiple sclerosis. Ramagopalan SV, Dyment DA, Cader MZ, Morrison KM, Disanto G, Morahan JM, Berlanga-Taylor AJ, Handel A, De Luca GC, Sadovnick AD, Lepage P, Montpetit A, Ebers GC. Ann Neurol. 2011 Abstract OBJECTIVE: Multiple sclerosis (MS) is a complex neurological disease. Genetic linkage analysis and genotyping of candidate genes in families with 4 or more affected individuals more heavily loaded for susceptibility genes has not fully explained familial disease clustering. METHODS: We performed whole exome sequencing to further understand the heightened prevalence of MS in these families. RESULTS: Forty-three individuals with MS (1 from each family) were sequenced to find rare variants in candidate MS susceptibility genes. On average, >58, 000 variants were identified in each individual. A rare variant in the CYP 27 B 1 gene causing complete loss of gene function was identified in 1 individual. Homozygosity for this mutation results in vitamin Ddependent rickets I (VDDR 1), whereas heterozygosity results in lower calcitriol levels. This variant showed significant heterozygous association in 3, 046 parent-affected child trios (p = 1 × 10(-5)). Further genotyping in >12, 500 individuals showed that other rare loss of function CYP 27 B 1 variants also conferred significant risk of MS, Peto odds ratio = 4. 7 (95% confidence interval, 2. 3 -9. 4; p = 5 × 10(-7)). Four known VDDR 1 mutations were identified, all overtransmitted. Heterozygous parents transmitted these alleles to MS offspring 35 of 35× (p = 3 × 10(-9)). INTERPRETATION: A causative role for CYP 27 B 1 in MS is supported; the mutations identified are known to alter function having been shown in vivo to result in rickets when 2 copies are present. CYP 27 B 1 encodes the vitamin D-activating 1 alpha hydroxylase enzyme, and thus a role for vitamin D in MS pathogenesis is strongly implicated.

Missing Heritability

DM Type-1 Celiac Islet cell AB

Approximated genetic risk in Diabetes Mellitus Type I General population 1 in 300 Sib of isolated case 1 in 14 Sib, no shared HLA 1: 100 Sib, two or more shared HLA haplotypes 1: 6 Harper P: Practical Medical Genetic counselling

DM Type-1 Celiac Islet cell AB

Links between Crohns, IBD & UC

Gene X Allele A/T 50/50 in controls 50/50 in patients Gene Y allele G/T 80/20 in patients 50/50 in controls

DRB 1 DQB 1 DRA

Multifactorial Diseases (Complex Diseases) Diabetes mellitus Hypercholesterolemia Ischemic heart disease Hypertension Asthma Crohns Rheumatoid arthritis Multiple Sclerosis SLE Glaucoma Manic depression Schizophrenia