The next frontier

1
The next frontier
Aspectos genòmicos y factores de
protecctiòn Amalio Telenti, University of Lausanne
2

• 1 Exploiting differences among pathogens
• 2 Exploiting extreme phenotypes
• 3 Exploiting technological breakthroughs

3
1 - Exploiting differences amongpathogens
4

• A 32 year-old person is found at the time of a
  blood donation to be HIV/HCV.
• His HIV viremia is 4.9 log copies/ml, CD4 T cells
  are 168 cell/ul.
• His HCV viremia is undetectable, liver tests are
  normal.

5
Genome analyses How do we do it?

• DNA from a large number of individuals
• Large scale genotyping of common human variation
  (500000 1 moi polymorphisms).
• Association analysis with correction for the
  large number of tests (significative p-values
  should be lt10-7 to 10-8

6
Genome-wide genotyping
500.000 to 1.000.000 SNPs/individual
7
Chromosomal location of locus of susceptibility
to HIV-1 and to Hepatitis C
HIV Chr. 6
N478 HIV
N1350 HepC
Fellay et al Science 2007 Rauch el al. Submitted
8
Survival/Progression
rs2395029 (HLA-B5701) rs9264942 (HLA-C
-35) rs9261174 (ZNRD1) rs333 (CCR5?32)
Fellay et al,
Years
9
Genetic score and progression
10
Chromosomal location of locus of susceptibility
to HIV-1 and to Hepatitis C
HIV Chr. 6
N478 HIV
N1350 HepC
Fellay et al Science 2007 Rauch el al. Submitted
11
HCV - Genetic determinants of spontaneous
clearance and treatment success
12
HCV - Viral and genetic determinants of
treatment success
13
GWAS Results

• We have reached experimental power conditions to
  identify most common human (Caucasian) variation
  influencing susceptibility to HIV-1
• We can know explain 22 of population variance by
  genetics, population effects, gender and age.
• Clear and profoundly different signals for
  various pathogens (n2).

14
2 - Exploiting extreme phenotypes
15
A rapid progressor
16
Rapid progression a genetic extreme
17
Integrating host and viral parameters
Casado et al.
18
 Sooty-like  patterns of evolution
19
(No Transcript)
20
A genomic, tanscriptomic and immunogenetic study
of rapid progression
Red associated with an AIDS event/death
lt350 CD4 T cells
CD4 evolution of Rapid Progressors (n73) during
3 years after seroconversion
21
CD4 T cell analysis
22
Interferon-stimulated genes
CD4 T cell analysis
23
Interferon stimulated genes Rapid progressors
versus Sooty-like profiles
More expressed in Rapid progressors
More expressed in Sooty-like
CD4 T cell analysis
24
PM
AGM
Gene expression changes in African green monkeys
(natural host model) and Asian pigtailed macaques
(pathogenic model) between day 10 and day 45 post
infection.
Lederer et al. PLoS Pathogens 2009
25
Interferon-stimulated genes
T cell receptor signalling
CD4 T cell analysis
26
(No Transcript)
27
(No Transcript)
28
Transcriptome Results

• The analysis of extreme phenotypes (beyond elite
  controllers) remains of major interest.
• New profiles, such as the rare sooty-like can
  be very informative and directly link to some of
  the non-pathogenic primate models.
• However.what to do with the long lists of
  candidate genes??

29
3 - Exploiting technological breakthroughs
30

• A severe hemophilia patient received multiple
  blood transfusions through the early 1980ies.
• Today this patient remains HIV negative, while
  HCV positive

31

• Characterization of high-risk HIV-1 seronegative
  hemophiliacs. Salkowitz et al.
• Among hemophiliacs from the MACS who remained
  HIV-1 seronegative despite a high (94) risk for
  acquisition of HIV-1 infection, 7/43 (16) were
  homozygous for the protective CCR5 Delta32
  polymorphism. Among the remainder, neither CCR5
  density nor beta-chemokine production, nor in
  vitro susceptibility to infection with the HIV-1
  isolate JR-FL could distinguish HRSN hemophiliacs
  from healthy controls.
• Clin Immunol. 2001 Feb98(2)200-11.

32
Genetic frequency in a population
ltltltltltlt1
gt5
1
Primary immuno-deficiencies
Common trait disease
?????????
Severe
?????
Mild
Disease manifestation / risk
33
Genetic frequency in a population
ltltltltltlt1
gt5
1
Primary immuno-deficiencies
Common trait disease
Severe
?????
Mild
Disease manifestation / risk
34
Genetic frequency in a population
ltltltltltlt1
gt5
1
RARE AND PRIVATE MUTATIONS
Primary immuno-deficiencies
Common trait disease
Severe
?????
Mild
Disease manifestation / risk
35
James WATSON
Whole Genome Sequencing
some 11,000 of Watsons SNPs (15 novel) are
predicted to change the amino-acid sequence and
so, perhaps, the function of a protein.
36
Where could be the genes of interest?
37
(No Transcript)
38
Ortiz et al Mol Biol Evol 2009
39
siRNA/shRNA screens for genes needed for HIV
replication in human cells
Brass et al. Science 2008. Konig et al. Cell
2008. Zhou et al. Cell Host Microbe 2008 Jeung et
al. J Biol Chem 2009.

• gt1000 gene candidates
• Only 3 genes common to at least three studies.
• 38 genes common to 2 or more studies.
• No restriction factors identified

40
Predicted interaction networks of genes
identified as HIV dependency factors in silencing
screens and differentially expressed during HIV-1
infection.
NF-kappa-B
Proteasome
Mediator complex
Protein kinases
Nuclear pore
41
(No Transcript)
42
Putting it to work
Analysis
Law
Ethics
Equipment
Materials
43
(No Transcript)
44
(No Transcript)
45
Haas
46
Genotyping
Imaging
Proteomics
Transcriptomics
47
Mathematics, statistics and computer sciences
"Scientists have learned to expect everything
from mathematicians short of actual help" John
HAMMERSLEY, Bull Inst Math Appl 10, 235, 1974
48
The role of the physician

• Identification of study phenotyes
• Avoiding low-power, limited scope studies.
• Bringing the best predictors to clinical use.

49
VIRAL LOAD GENETICS - Effect estimates in
Genome-wide studies Recruiting seroprevalent
versus seroconverter individuals
ltStronger in seroprevalent
Stronger in seroconvertersgt
50
Final Conclusions

• The genetic basis of human susceptibility to
  HIV-1 susceptibility to infection includes
  common variants (probably known by now), and a
  undefined number of rare variants.
• Technological breakthroughs are not adequately
  supported by clinical cohorts.

51
Will I ever use this knowledge?
52
Duke University J. Fellay K. Dang D. Goldstein

University of Lausanne A Ciuffi M. Rotger M.
Ortiz S. Colombo University Hospital Bern A.
Rauch Genomics Platf. Geneva P. Descombes Ragon
Institute P. McLaren P. De Bakker B. Walker

Pasteur Institute L. Quintana-Murci
Carlos III C. Lopez Galindez C. Casado

IrsiCaixa J. Dalmau J. Martinez-Picado
53
(No Transcript)