Immune Control of Clade C HIV Infection

1
Immune Control of Clade C HIV Infection

• Bruce D Walker, MD
• University of KwaZulu Natal
• Ragon Institute
• of MGH, Harvard and MIT

2
Question

• Why do children do worse than adults when they
  become HIV infected?
• Can answering this question shed light on
  important immune mechanisms?

3
HIV Prevalence in Antenatal Clinics 2001
4
Philip Goulder
Photini Kiepiela
Not shown H. Coovadia S. Abdool Karim
5
Doris Duke Medical Research Institute UKZN
6
DDMRI
7
Mark Schwartz
8
Kaye Ajao
Krista Dong
9
Thumbi Ndungu
10
Can we find signals indicating what is involved
in immune control?
11
Host Immune Responses
Virus
Host Genetics
12
Host Genetics
13
HLA Class I
817 B alleles
263 C alleles
486 A alleles
14
Experimental Design

• Subjects
• HIV-infected Zulu/Xhosa (n706)
• Methods
• High resolution HLA typing
• Viral load determination

15
HLA B-association with control of HIV
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
B57
p lt 0.0001
p 0.0014
p 0.0006
B8101
B5801
B5802
B1801
p lt 0.0001
Kiepiela et al, Nature, 2004
p 0.0008
16
HLA-association with control of HIV
2.0
2.5
3.0
3.5
4.5
5.0
5.5
6.0
4.0
p 0.0006
B5801
B5802
p lt 0.0001
Kiepiela et al, Nature, 2004
17
Marked differences in HLA associations B5801
and B5802
plt 0.0001
10000000
1000000
100000
Viral load
(RNA Copies/ml)
10000
Median
Median
Median
1000
75, 200
33, 000
14, 650
100
10
B5801
B5802
Neither
-
n 93
n 259
n 722
Ngumbela et al, ARHR 2008
18
Marked differences in HLA associations B5801
and B5802
plt 0.0001
plt 0.0001
10000000
1000000
100000
Viral load
(RNA Copies/ml)
10000
Median
Median
Median
1000
75, 200
33, 000
14, 650
100
10
B5801
B5802
Neither
n 93
n 259
n 722
Ngumbela et al, ARHR 2008
19
HLA B5802 is deleterious?
20
Conclusions I

• HLA B alleles influence viral load
• Certain alleles are associated with protection,
  and others with progression

21
How do HLA alleles influence viral load?
22
B Pocket
F Pocket
HLA Class I
Infected cell
23
Viral Peptide
B Pocket
F Pocket
HLA Class I
Infected cell
24
T Cell Receptor
B Pocket
F Pocket
HLA Class I
Infected cell
25
Experimental Design

• Subjects
• HIV infected treatment naïve Zulu/Xhosa
• Methods
• Determine the dominant protein targeted by CD8 T
  cells by ICS using pooled peptides (61 subjects)
• Determine breadth of CD8 T cell responses by
  Elispot, using individual overlapping peptides
  spanning all HIV proteins (578 subjects)
• Correlate to viral load

26
A dominant CD8 Gag-specific response is
associated with lower viremia
Ramduth, Chetty, Mngquandaniso et al, JID 2005
27
The breadth of the Gag-specific CD8 response is
associated with lower viral load
94,000
21,000
Kiepiela, Leslie, Honeyborne et al, Nature
Medicine 2007
28
The breadth of the Env-specific CD8 response is
associated with higher viral load
220,000
29,000
Kiepiela, Leslie, Honeyborne et al, Nature
Medicine 2007
29
HLA-association with control of HIV
2.0
2.5
3.0
3.5
4.5
5.0
5.5
6.0
4.0
p 0.0006
B5801
B5802
p lt 0.0001
Kiepiela et al, Nature, 2004
30
Gag-specific CD8 responses are minimal to
undetectable in the first 6 months after vertical
transmission
Thobakgale et al, JV 2007
31
Are there functional differences in HIV-specific
immune responses?
32
Experimental Design

• Subjects selected from a cohort of 288 persons
  with chronic HIV infection
• 13 high Gag-specific CD8 T cell responses
• Mean breadth 82.6
• Mean magnitude 33801304 SFC/106 PBMC
• 13 low Gag-specific CD8 T cell responses
• Mean breadth 0.640.6
• Mean magnitude 808.81038 SFC/106 PBMC

33
Subjects matched for CD4 count and overall
Elispot responses
No difference in terms of CD4 cell counts or
total elispot responses in high vs low Gag
responders
Julg, Williams, Reddy et al, unpublished
34
Experimental Design

• Methods
• Measure the ability of freshly isolated CD8 T
  cells to
• Inhibit a heterologous strain of virus in
  autologous CD4 cells
• Proliferate in response to HIV peptides
• Secrete cytokines in response to HIV peptides

35
High Gag responders have significantly lower
viral loads
Julg, Williams, Reddy et al, unpublished
36
1000000
100000
10000
CD4 uninfected
CD4 infected with
p24 pg/ml
1000
HIV-1 (X4) MOI 0.1
100
10
1
0
1
2
3
4
5
6
7
8
9
10
11
12
13
Day
37
1000000
100000
10000
CD4 uninfected
CD4 infected with
p24 pg/ml
1000
HIV-1 (X4) MOI 0.1
100
CD8 Cells added
10
1
0
1
2
3
4
5
6
7
8
9
10
11
12
13
Day
Julg, Williams, Reddy et al, unpublished
38
High Gag-responders inhibit virus replication
better
Julg, Williams, Reddy et al, unpublished
39
Conclusions II

• Broader Gag-specific CD8 T cell responses are
  associated with lower viral loads
• Broader Env-specific CD8 T cell responses are
  associated with higher viral loads
• Broad Gag-specific CD8 T cell responses are
  associated with enhanced ability to neutralize
  infectious HIV and greater functionality

40
Virus
41
Viral Epitope
42
W
S
HLA B5801
43
Rapid selection for mutation in the
B5801- restricted TW10 epitope
TSTLQEQIAW
Wild type
T242N
--N-------
44
Wild-type virus outgrows mutant T242N
viral variant
viral variant
days post infection
Martinez-Picado, J Virol, 2006
45
Number of HLA-B-associated Gag mutations correlate
d with median viral load for each allele
B4901
B51
B41
B4202
100000
B0705
B1801
r -0.56
50000
B1402
B5802
p 0.0034
B1510
B0801
B1503
Viral Load
B3501
B4501
B0702
B4201
B4403
B3910
B1302
B1401
B5301
B5801
B5702
B1516
B8101
10000
B5703
Matthews, Prendergast et al, J Virol, 2008
0
1
2
3
4
5
6
Number of HLA-B Gag mutations
46
Conclusions III

• Gag-specific CD8 T cell responses select for
  fitness mutants that cripple HIV

47
What is this epitope-specific immune pressure
doing at a population level?
48
Global CTL Escape Collaboration

• Subjects
• 2800 persons with chronic HIV infection
• 9 cohorts, 5 continents
• Methods
• HLA typing
• Virus sequencing in Gag
• Analyze the relationship between HLA prevalence
  and detection of escape mutations

Y Kawashima et al, Nature 2009
49
Percent escape mutant in the entire
population
Y Kawashima et al, Nature in press
50
Percent escape mutant in persons NOT expressing
HLA B51
Y Kawashima et al, Nature in press
51
Conclusions IV

• HIV is being shaped by the immune response on a
  global level, and some protective epitopes are
  being lost

52
Do host genetics affect transmission?
53
Impact of Protective HLA alleles on HIV
Transmission
P 0.05
Percent mothers expressing protective HLA
alleles
N61
N236
Thobakgale et al, JV in press
54
Impact of Protective HLA alleles on disease
progression in infants
P 0.047
Percent mother or child expressing protective
HLA alleles
N44
N17
Thobakgale et al, JV in press
55
Conclusions V

• HLA is impacting transmission
• Protective HLA alleles should increase in the
  population as the epidemic progresses

56
What else can international partnerships and
local leadership achieve?
57
KwaZulu Natal Research Institute for TB and HIV
(K-RITH)
58
Given where we are now, what should the
priorities be moving forward?
59
By age 22, 66 of mothers are HIV
infected (Durban, March 2009)
75
66
Female (15-49)
50
Male (15-49)
RSA antenatal clinics 2006

PMMH March 2009
25
0
0
8
16
24
32
40
48
54
60
Age
60
A vaccine for HIV is possible, but still a long
way off More efforts must turn to prevention

• Whizzkidsunited.com

61
(No Transcript)
62
Zenele Mncube Nompumelelo Mkhwanazi Cheryl
Day Janet Giddy Henry Sunpath Helga
Holst Thumbi Ndungu Jerry Coovadia Salim Abdool
Karim William Carr Victoria Kasprowicz
Marcus Altfeld Marylyn Addo Mark Brockman Philip
Goulder Jill Conley Marcus McGilvray
Christina Thobakgale Kholiswa Ngumbela Danni
Ramduth Kirebnashe Nair Boris Julg Katie
Williams Isobel Honeyborne Mary Van der
Stock Eshia Moodley Karen Bishop Sharon
Reddy Fundi Chonco Wendy Mphatswe Sister Kesia