Detecting Allelic Imbalance in Inflammatory Bowel Disease
Download
Report
Transcript Detecting Allelic Imbalance in Inflammatory Bowel Disease
Understanding the Next Steps in Determining the
Role of the Genome in IBD
Judy H. Cho, M.D.
Ward-Coleman Professor of Translational Genetics and Medicine, Icahn
School of Medicine at Mount Sinai
December 6th, 2014
Questions
What naturally occurring genetic polymorphisms are
associated with IBD?
What naturally occurring genetic
polymorphisms are associated with IBD?
Assessment
Value: molecular insight profoundly shaped the
landscape of IBD research
What’s worked and what hasn’t
Family-based: linkage NOD2. Otherwise, family-based studies have not
worked—re-vist with microbiome-based studies & genetic counseling
Case-control: 163 loci and counting
Surprises & advances
Magnitude of sample sizes required unexpected: NOD2 (Nature 2001)
involved 416 CD samples - > 30,000 cases
Number of significant loci larger than expected: 163 and counting
Advances: autophagy, IL-23 pathway, M1-M2 macrophage subsets
Challenges: long journey from genes biology drug targeting new
drugs
Questions
What naturally occurring genetic polymorphisms are
associated with IBD?
What are the functional consequences of IBDassociated polymorphisms?
What are the functional consequences of IBDassociated polymorphisms?
Altered cytokine responses to microbial stimulation :
NOD2, XIAP (direct); many indirect effects
Genotype-dependent variable bacterial clearance:
autophagy, NADPH oxidase
Direct ex vivo analyses
IL23R and Th17/Tc17 cells (Sarin et al., PNAS 2011)
NOD2 & Paneth cell morphologies (Van Dussen
Gastroenterology 2014)
Altered regulation of gene expression
What are the functional consequences of IBDassociated polymorphisms?
Altered cytokine responses to microbial stimulation :
NOD2, XIAP
Genotype-dependent variable bacterial clearance:
autophagy, NADPH oxidase
Direct ex vivo analyses
IL23R and Th17/Tc17 cells (Sarin et al., PNAS 2011)
NOD2 & Paneth cell morphologies (Van Dussen
Gastroenterology 2014)
Altered regulation of gene expression
Expression quantitative trait loci (eQTL)
mapping
mRNA expression as a continous trait
Heritable
Mappable to specific SNPs
Cell lines, tissues and contextspecificity
Presently defined eQTLs likely only a
subset of genuine eQTLs
LPS- & IFNg stimulated monocytes
define more eQTLs
80% of transcripts with eQTLs
Morley, et al., Nature 2004; 430: 743
Dixon, et al., Nature Genetics 3007; 39: 1202
Fairfax et al., Science 2014
Fine-mapping in autoimmunity
Fine-mapped autoimmune loci
90% are non-coding
60% map to immune enhancers
Histone marks: greatest enrichment seen for
H3K27ac—active/stimulated enhancers
Disease-associated SNPs in enhancers are near, but not
within consensus transcription factor binding sites
Farh et al., Nature 2014
Enrichment of CD loci genes in open chromatin
regions of Th17 cells, but not monocytes
Category (N)
Th17 cell
DNase I HS
site
Monocyte
DNase I HS
site
Yes (14,267)
No (10,195)
Yes (16,317)
No (8145)
CD loci
non CD loci
log OR
(#gene=1328) (#gene=23,134)
(95% CI)
945 (71.2%) 13,322 (57.6%)
0.6
(0.48, 0.72)
383 (28.8%) 9812 (42.4%)
862 (64.9%) 15,455 (66.8%)
-0.08
466 (35.1%) 7679 (33.2%) (-0.20, 0.032)
p-value
5.90E-22
Apples to oranges: Th17 cells vs. monocytes
Pending: tissue-specific enhancer landscape of
organ- and context-specific tissue macrophages
0.15
From co-expression to (genetic) causal
networks
aka NRAMP
NOD2
SLC11A1
IL10
VDR
HCK
CARD9
DOK3
Highly correlated RNA expression
between NOD2, IL10 & HCK
(hematopoietic cell kinase)
HCK: key for differentiation of
M2 macrophages
LGALS9
Gene in IBDassociated locus
Cis eQTL in HCK: variable HCK
expression is driving variable
NOD2 & IL10 expression
Eric Schadt
Jostins et al, Nature 2012
Primary value of direct ex-vivo analyses:
pathogenic & protective cells
Defining innate cell hierarchies by high-dimensional Cytof analysis
Uninflamed
CD14
HLADR
BDCA3
Inflamed
High dimensional analyses
needs to be matched to
deep clinical information
Basic science questions
Cellular plasticity and
diffentiation
DNA-RNA-protein
What are the functional consequences of IBDassociated polymorphisms?
Altered cytokine responses to microbial stimulation :
NOD2, XIAP
Genotype-dependent variable bacterial clearance:
autophagy, NADPH oxidase
Direct ex vivo analyses
IL23R and Th17/Tc17 cells (Sarin et al., PNAS 2011)
NOD2 & Paneth cell morphologies (Van Dussen
Gastroenterology 2014)
Altered regulation of gene expression
Acceleration of in vitro & in vivo studies with CRISPR/CAS9 technologies:
refined genetic & molecular definition will improve modeling
Questions
What naturally occurring genetic polymorphisms are
associated with IBD?
What are the functional consequences of IBDassociated polymorphisms?
Why do Ashkenazi Jewish populations have a higher
IBD prevalence?
Rare variants
Common variants
Rare variants are “less rare” in AJs
AJ
Flemish
128 whole genome
sequenced in AJs
Rare variants are
“less rare” in AJs
compared to Flemish
Profound population
bottleneck in AJs
Derived allele frequency
Carmi et al., Nature Commun 2014
Ashkenazi Jews with a higher composite burden
score than non-Jewish European ancestry
IBD Genetic Burden Score
Developing integrative models
of common variant risk?
Striking overlap between IBD & mycobacterial
susceptibility
NOD2
RIPK2
**TNFSF15
163 IBD loci
LRRK2
IL23R
7/9
6/7
C13orf31
7 multigenic
leprosy
GWAS loci
IL12B
STAT1
IRF8
TYK2
STAT3
IFNGR2
IFNGR1
9 single gene
mycobacterial (Tb) genes
Anti-TNF treatment of IBD associated with re-activation
of latent mycobacterial disease
NEJM 2001; 345: 1098
Epidemiologic support for the Jewish-Tb hypothesis
NYC, 6 years before 1890 per 100,000
Deaths from tuberculosis, London 1894-1900
Population
Deaths per 100,000
Mussulman Arabs
1130
Europeans
513
Jews
75
Jacobs J. The Jewish Encyclopedia; a guide to
its contents, an aid to its use. New York, London:
Funk & Wagnalls company; 1906.
Polygenic adaptation: positive
selection at scores to hundreds of
genes that confer selective advantage
Population
NY
Brooklyn
African-American
774.21
531.35
Ireland
645.73
452.79
Bohemia
499.13
347.22
Russia and Poland
98.21
(mostly Jews)
76.72
Scotland
384.12
269.24
Scandinavia
357.00
218.92
Canada
352.32
266.27
Germany
328.80
295.61
France
394.98
252.82
England and Wales 322.50
233.78
Italy
233.85
123.00
United States
(White)
205.14
180.79
Hungary (mostly
Jews)
155.05
120.77
Questions
What naturally occurring genetic polymorphisms are
associated with IBD?
What are the functional consequences of IBDassociated polymorphisms?
Why do Ashkenazi Jewish populations have a higher
IBD prevalence?
What factors are genetic and what are nongenetic/stochastic/developmental?
Better modeling of genetic contributions
More frequent modeling of environmental/stochastic
factors
Systematic analysis of rare variants
via enteroids
Biopsy
Add growth factors
Stem cells
Stem cells
> Ascertaining by genotypes
> CRISPR-CAS9
Bank of rare variants
broadly amenable for study
IgA-coating enriches for colitogenic bacteria
Stool-based collections: IgA enrichment may identify
functionally important bacteria—reducing microbial
complexity
Frequent sampling
Palm et al., Cell 2014
Questions
What naturally occurring genetic polymorphisms are
associated with IBD?
What are the functional consequences of IBDassociated polymorphisms?
Why do Ashkenazi Jewish populations have a higher
IBD prevalence?
What factors are genetic and what are nongenetic/stochastic/developmental?
UC: limited vs. extensive disease
CD: ileal post-op—modeling first steps of disease
recurrence
Age-dependent effects: disease severity & age of onset
Acknowledgements
Cho lab
NIDDK IBD Genetics Consortium
Steve Brant
Richard Duerr
Dermot McGovern
John Rioux
Mark Silverberg
Mark Daly
Phil Schumm
Thad Stappenbeck-Wash U: enteroids
Yale University
Clara Abraham
Richard Flavell
RISK Pediatric Consortium
Subra Kugathasan
Ted Denson
Ken Hui
Monica Bowen
Kyle Gettler
Kaida Ning
Nai-Yun Hsu
Felix Chuang
Yashoda Sharma
Mount Sinai collaborators
Inga Peter
Eric Schadt
Miriam Merad
Bruce Sands
Jean-Fred Colombel