Side-stepping Regulatory T Cell-Mediated Evasion in

Side-stepping Regulatory T Cell-Mediated Evasion
in Chronic Viral Infection to Build Better Vaccines
Frances Terry1, Andres Gutiérrez2, Rui Liu2, Ryan Tassone2, Steve Gregory2, Phyllis Losikoff3, Chris Bailey-Kellogg4,
Leonard Moise1,2, William Martin1, Anne S. De Groot1,2
1EpiVax,
Inc., Providence, RI, USA, 2Institute for Immunology and Informatics, University of Rhode Island, Providence, RI, USA
3Rhode Island Hospital and the Warren Alpert Medical School at Brown University, Providence, RI, USA 4Dartmouth College, Hanover, NH, USA
Abstract
Methods and Results
Janus Score Predicts Cytokine Release
In Roman mythology, Janus is the god of beginnings and
transitions, of gates, doors, doorways, endings, and time. He
is usually portrayed as a two-faced god, looking both to the
future and the past.
T cell
receptor
face
Avg Janus Human Score vs +/-­‐ Cytokine Release Average Janus Human Score Fig 1. (left) JanusMatrix separates the amino acid sequence of T cell
epitopes into TCR-facing residues (epitope) and HLA binding cleft-facing
residues (agretope), then compares the TCR face to other putative T cell
epitopes.
Cross-reactive peptides:
•  Are predicted to bind the same MHC allele.
•  Share same/similar T cell-facing residues.
MHCbinding
face
IEDB Posi2ve 2.5 IEDB Nega2ve p=0.000
p=0.008
2 1.5 1 0.5 0 IL-­‐10 IL-­‐4 A Cytokines B A) Peptides documented as IL-10-positive in IEDB have
significantly higher potential for Human Genome (HG,
above) and Human Microbiome (HM, below) cross-reactivity
as measured by Janus Matrix. B) Peptides documented as
IL-4-positive in IEDB have significantly lower potential for HG
and HM cross-reactivity as measured by Janus Matrix.
TCR cross-reactivity prediction:
•  Given a protein or peptide, T cell epitopes are
identified based on MHC contacts (P1, P4, P6, P9)
using EpiMatrix.
•  JanusMatrix searches for potentially cross-reactive
TCR by screening TCR-facing residues (P2, P3, P5,
P7, P8) against a preloaded, EpiMatrix-processed
reference databases.
Avg Janus Microbiome Score vs +/-­‐ Cytokine Release Reference databases available include:
•  Human genome (HG)
•  Human gut microbiome (HM)
•  Human pathogens (HP)
(bacteria and viruses)
Background
IEDB Posi2ve 30 IEDB Nega2ve p=0.000
p=0.004
25 20 15 10 5 0 IL-­‐10 IL-­‐4 A Cytokines B Predicted Treg-activating HIV and HCV sequences possess TCR faces shared by numerous human proteins
Cross-Reactivity
Epitope networks are shown, illustrating the abundance of TCR faces one HCV and one HIV peptide share with the human genome
as determined by JanusMatrix analysis. The HIV and HCV source peptides are represented by green diamonds, their constituent 9mer epitopes by gray squares, their cross-conserved partners in the human genome by blue triangles, and the source human
proteins by light purple circles. In the HIV peptide (right) a single cross-conserved epitope can JHEPAT
be found
in 32 different HLA class I
5306
alleles; several additional JHEPAT
9-mer
epitopes are cross-conserved with 12 other HLA sequences
protein orange highlight).
5306
No. of(source
Pages
20 9October 2014
•  Intrinsic characteristic of TCR, i.e., each single TCR can potentially interact with different
peptide-MHC complexes.
•  Critical to many aspects of T cell biology, including positive and negative selection.
•  Involved in heterologous immunity. Immune response can be “preset” based on prior exposure
to cross-reactive epitopes.
•  Can have positive or negative (e.g., leading to pathology) effects.
Source peptide
Research Article
Source 9-mer epitope
A
JOURNAL OF HEPATOLOGY
20 October 2014
8
*
2
* Tregs
HCV Treg Epitope0 ExpandsJOURNAL
OF
Ab VL
Ab VL
Ab VL
in Chronic HCV Infection
100
A
Medium
Medium alone B
10
**
%CD3+CD4+FoxP3+ cells
%CD3+CD4+FoxP3+ cells
+
A
HCV_G1_p7_794
HCV_G1_NS4b_1941
Medium
Anti-CD3
HCV_G1_p7_794
Anti-CD3 + HCV_G1_p7_794
Source human protein
80
TCR-conserved
human
No. of5306
Pages 9
JHEPAT
9-mer epitope 60
20 October 2014
6
4
100
+
10
-
-
+
-
proliferation
105
10
4
105
1.61%
10
103
CD304
JHEPAT 5306
Medium alone
HCV_G1_p7_794
HCV_G1_NS4b_1941
**
10
A
% of Max
A
%CD3+CD4+FoxP3+ cells
Can identify TCRs that recognize epitopes
from pathogens and also homologs from
Self and Commensals
20 October 2014
FoxP3
Methods and Results
We designed an immunoinformatic algorithm, JanusMatrix, to
identify such epitopes. For a pathogen-derived epitope, the
algorithm searches the human genome for the same T cell
receptor (TCR) face in 9mers that can bind human leukocyte
antigens (HLA). Using JanusMatrix, we screened a wide
range of human-host viruses for TCR-face similarity to self
and discovered that chronic viruses generally appear more
human-like than viruses that cause acute infection. We also
discovered a promiscuous class II epitope located within nonstructural hepatitis C virus (HCV) protein p7 that exhibits
homology with hundreds of human sequences. The epitope
induces an increase in CD4+CD25+FoxP3+ Treg number and
function in peripheral blood leukocyte cultures derived from an
HLA-diverse cohort of HCV-infected patients, but not in
cultures derived from patients who spontaneously cleared
HCV or from non-infected individuals. Similar patterns have
been observed in HIV.
JanusMatrix
Average Janus Microbiome Score Background
Vaccines against many pathogens that cause chronic infection
are unavailable, due largely to effective immunoevasive
mechanisms. A novel escape mechanism observed in chronic
viral infection is suppression of viral-specific effector CD4+
and CD8+ T cells by stimulating regulatory T cells (Tregs)
educated on host sequences during tolerance induction. A
significant epitope property that is beginning to gain wider
attention is homology with host sequences. Viral epitopes with
substantial homology to self may activate Tregs that suppress
protective inflammatory responses and thereby enable viral
persistence.
0
4
69.20%
103
No. of Pages
0
40
+)
HCV Treg Epitope Expands0 iTregs
(CD304
10 10 10
0
10 10 10
HEPATOLOGY
Research Article20
in Chronic HCV Infection
CD4
FoxP3
The majority
of individuals in our patient population we
B
medium
only
HCV_G1_p7_794
0
A
3
Anti-CD3
105
HCV_G1_p7_794
Anti-CD3
+ HCV_G1_p7_794
Medium
alone
104
4
5
3
4
5
the
derivation of HCV_G1_p7_79
5
105infected with HCV genotype 1,10
010 69.20%
103 104 105
Comp-FITC-A::-CFSE
104
5
3
References / Acknowledgments
3
CD304
FoxP3
104
Count
CD304
CD304
H-Thymidine incorporation (cpm x103)
FoxP3
FoxP3
1.61%
Conclusions
Count
80
% of Max
H-Thymidine incorporation (cpm x103)
%CD3+CD4+FoxP3+ cells
9.22% of HCV_G1_p7_794 to22.86%
The addition
PBMC cultures derived from
4
10
8
single HCV genotype 3-infected patient, however, also resulted
8
p7_794 analog
3
3
10
10
10+3 FoxP3+CD25+ cells (Suppleme
103a marked increase in CD3+CD4
T cell
Medium
B
6
6
60
HCV_G1_p7_794
0
0 in this regard, that HCV gen
0tary Fig. 2). It is pertinent to note,
proliferation 0
Anti-CD3
4
4
type 3 encodes a peptide sequence: LALLVLLLPQRAYAW, whi
Anti-DC3 + HCV_G1_p7_794
*
**
40
IL-2
exhibits HCV_G1_p7_794 homology.
2
2
IL-2 + HCV_G1_p7_794
10
Conceivably,
other viral pathogens 0that10cause
chronic disea
3
3
104 105
*
0
103 104 105
0
10 104 105
0
103 104 105
e.g., herpes simplex, Epstein-Barr, human immunodeficiency a
0
0
20
+
+
+
CD4
CD4
FoxP3
Ab VL
Ab VL
Ab VL
cytomegalovirus, also avoid or reduce TFoxP3
eff cell responses
Ab+VL+
Ab-VLIsotypeexploiting
controls similarity to self and activating nTreg cells [29,3
B
0individuals
Fig. 2.inHCV_G1_p7_794
fails to elicit an increase
in CD3+CD4+FoxP3+ cells
expands Tregs
chronic HCV-infected
•  CD3+CD4+FoxP3+ cells5 from chronic HCV-infected
individuals,
cultured
inthat
theEpstein-Barr
absence of
B•  HCV_G1_p7_794
105
Indeed,
recent
analyses
indicated
virus a
10
1000
3
4
5
among PBMCs derived from non-infected individuals. (A)
PBMCs obtained
10 not in those who clear infection
10
10
0
10
but
or
are
not
infected.
HCV_G1_p7_794
express
CD304
(neuropilin),
indicative
of
nTreg
cells.
9.22%
800 cytomegalovirus contained fewer T1000
22.86%
5
cell epitopes and exhibit
*
from infected patients (Ab+VL+, n = 4), patients who cleared infection (Ab+VL!, 104
4
800
10
+CD4+FoxP3+ cells, cultured in the
! !
Comp-FITC-A::-CFSE
alone
800
•  Its 8human analog expandsMedium
Tregs
in
uninfected
and
chronic
infected
• 
The
vast
majority
of
CD3
presence
of
HCV_G1_p7_794,
n = 6) and non-infected controls (Ab VL , n = 4) were cultured with medium
600 higher cross-reactivity with the human genome than did eith
600
3
3
alone,
HCV_G1_p7_794 or HCV_G1_NS4b_1941.
Cells
werefor
collected
after
5 days 10 were CD304 characteristic
p7_794 analog
10
600
subjects, while a control peptide
HCV_G1_NS4b_194,
does
not
any
group.
of
iTreg
cells.
Ebola
or
Marburg
virus
[14].
Ebola
and
Marburg viruses, on t
⁄
⁄
400
and were analysed by flow cytometry.
Significantly different: p = 0.014;
400
Medium
400
B
6
⁄⁄
+
! !
0
0
other
hand,
were
composed
of
significantly
fewer pepti
p <0.001. (B) PBMCs obtained from Ab+VL
patients
(n
=
4)
and
Ab
VL
controls
HCV_G1_p7_794
200
200
200
(n = 4) were cultured in the presence or absence
of the human p7_794 analogue.
sequences
that
were cross-reactive with human and express
Anti-CD3
The Context of Immune
4
0
Significantly more CD3+CD4+FoxP3+ cellsAnti-DC3
were recovered
from
PBMCs
cultured
0 a 1 larger
0 predicted
+ HCV_G1_p7_794
1
2
3 T4 cell
5
2
3 number
4
5
of
epitopes.
th
101 102Thus,
103 104viruses
105
⁄
⁄⁄
10
10
10
10
10
10
10
10
10
10
Response can tip the balance
p
=
0.001;
p
=
0.048.
** with the analogue than from medium alone:
IL-2
5
4
5
acute disease
and viruses
such as HCV,
which adapt
0
103 104 10
Homology
with the human genome representsIL-2a+ novel
means by which viruses
seek #261
to establish
chroniccause
infections
escape
human
immunity
0
103 10#229
10that
#265
2
from inflammation to
HCV_G1_p7_794
FoxP3
CD4
CD304
10
humans
and cause chronic
infection,
may
differ
substantially
Patient IDTFoxP3
and ensure their survival.
Better
classification
of
viral
epitopes
as
either
effector
or
regulatory
cell
activating
will
improve
the
design
of
vaccines
regulation
CD4
265
lating in the peripheral blood of chronically-infected patients,
0
Fig. 4. Fewer
HCV-G1_p7_794
responsive
Treg cells
express CD304 (neuropilin).
terms
of their Treg
cell epitope
content.
Isotype
controls
Fig.
3.
HCV_G1_p7_794
suppresses
the
proliferation
of
PBMCs
derived
from
against chronic
viruses.
266-VL- confirmed in the current study (Fig. 2), provided an initial indicaPBMCs obtained
from an HCV-infected
patient (representative
4 patients)
Although
immunosuppression
by of
Treg
cells were
is read
Ab+VL+
Ab
HCV-infected patients. CSFE-labelled, Ab+VL+ PBMCs were cultured with
267
incubated
in
medium
alone
(A)
or
medium
that
contained
HCV_G1_p7_794
(B).
tion of the role of Treg cells in the pathogenesis of chronic hepamedium alone, or 1000
medium containing anti-CD3, HCV_G1_p7_794, or anti-CD3 demonstrated in mice, demonstrating the suppressor activity
+
+
+
1000
800
The cells were collected on day 5, stained and analysed
by flow cytometry. Panels
Fig. 2. HCV_G1_p7_794 fails to elicit an increase in268
CD3 CD4titis
FoxP3
cells
C [2,3,5,11].
It remained unclear until recently, however,
+
and HCV_G1_p7_794.
800 Cells were collected after 5 days; proliferation was human CD3+CD4+CD25+FoxP3
T
+
+reg cells has proven problema
on the
right indicate the percentage of CD4 FoxP3 cells in each population that
800
among PBMCs derived from non-infected individuals.
obtained
269(A) PBMCs
estimated by flow cytometry as a loss in fluorescence
intensity. Data
were
whether
this increase represented the HCV epitope-specific600
!
[31]. Recent studies indicate that the nature of the responde
express
from infected patients (Ab+VL+, n = 4), patients who cleared
infection
(Ab+VL
,T cells or 5
obtained in a single 600
experiment representative 600
of two HCV-infected patients
(A).CD304.
270
response
of
the
nonspecific
consequence
of
chronic
reg
+
! will circumvent
+
low
1)  Moise L, Gutierrez AH, Bailey-Kellogg C, Terry F, Leng Q, Abdel Hady KM, VerBerkmoes NC, Sztein MB, Losikoff PT, Martin WD, Rothman AL, De
Groot AS. 4)  Moise L, Terry F, Gutierrez
AH, Tassone R, Losikoff P, Gregory SH, Bailey-Kellogg
C, Martin
WD and De Groot
AS. Smarter vaccine design
400
PBMCs obtained from
the three patients (#229, #261,
n = 6) and non-infected controls (Ab!VL!, n = 4) were271
culturedinflammation
with medium and liver disease [7].
400
400 #265) were cultured 5 days cells (CD4 CD25 vs. CD4 CD25 ) and the ratio of Treg cells
The two-faced T cell epitope: examining the host-microbe interface with JanusMatrix. Hum Vaccin Immunother. 2013 Jul;9(7):1577-86..
regulatory T cell-mediated evasion in chronic HIV and HCV infection. Front. Microbiol.,
06 October
in the presence
of anti-CD3 2014
or 20 ng/ml IL-2 with or without HCV_G1_p7_794; responder cells exerts significant effects on the outcome of t
alone, HCV_G1_p7_794 or HCV_G1_NS4b_1941. Cells were collected after 5 days
200
200
272
The
ability
of
HCV-derived
epitopes
to
stimulate
T
cell
200
309 duringthat
2)  He L, De Groot AS, Gutierrez AH, Martin WD, Moise L and Bailey-Kellog C. Integrated assessment of predicted MHC binding and cross-conservation
with by
5)  flow
Losikoff
PT,⁄Significantly
Mishra S,
Terry⁄p F,
Gutierrez A, Ardito MT, Fast L, NevolaregM, Martincell
WD,
Bailey-Kellogg
De Groot
AS, Gregory
epitope,
homologous
tonormally function to
elicitHCV
the activity
of nT
proliferation
was
estimated byC,
[3H]-thymidine
incorporation
theSH.
last
reg cells, which
and were analysed
cytometry.
different:
= 0.014;
suppression
assays
[32,33].
Nonetheless,
the HCV_G1_p7_7
0
273 sequences,
responses
is
well
documented;
epitopes
associated
with
both
⁄⁄
+ human
+
! !
18
h
of
incubation
(B).
0
self reveals patterns of viral camouflage. BMC Bioinf., 2014.
multiple
protein
induces
a
regulatory
T
cell
response
in
infected
patients.
J
Hepatol.
2015
Jan;62(1):48-55.
0
310
autoimmune reactivity to self-antigens (proteins). In
p <0.001. (B) PBMCs obtained from Ab VL patients (n = 4) and Ab VL controls
5
101 102 103 104 10suppress
101 102 103 104 105
101 102 103 104 105
responsive Treg cells suppressed the mitogenic response of ce
274
structural
and
non-structural
HCV
proteins
have
been
reported
(n =Fc-derived
4) were cultured in6) 
theCousens
presence or absence
of the human
analogue.
3)  De Groot AS, Moise L, McMurry JA, Wambre E, Van Overtvelt L, Moingeon P, Scott DW, Martin W. Activation of natural regulatory T cells by IgG
L, Najafian
N,p7_794
Martin
WD, De Groot AS. Tregitope: Immunomodulation powerhouse. Hum Immunol. 2014 Dec;75(12):1139-46.
311
this regard, it is pertinent to remark that HCV_G1_p7_794 is com+
+
+
derived from HCV-infected patients in the experiments report
275
[20,26].
Using
HLA
class
II-peptide
tetramer
complexes,
other
FoxP3
CD4
CD304
Significantly more CD3 CD4 FoxP3 cells were recovered from PBMCs cultured
peptide "Tregitopes". Blood. 2008 Oct 15;112(8):3303-11.
+
312
prised
of
epitopes
that are homologous to those found in hun292
⁄⁄
expressed CD39, a marker that distinguishes FoxP3 Treg cells here
despite the fact that the CD3+CD4+FoxP3! responder cells
investigators quantified and characterized the response of Treg
with the analogue than from medium alone: ⁄p = 0.001;276
p = 0.048.
313 [6,19].
dreds of human
This suggests that the autoimmune
Fig. 4. Fewer
HCV-G1_p7_794
responsive
Tregtransiently
cells expressexpress
CD304 (neuropilin).
Discussions leading to JanusMatrix development were supported by NIH U19 grant AI082642 (to ADG) We thank
at HCV
iCubed
and
Dartmouth
for contributions
to thisTeff
work.
293 proteins.
from
activated
cells that
FoxP3
277 our
cellcolleagues
specific for single
epitopes
[20,26].
The study#261
described
#229
#265
outnumbered
CD3+CD4+CD25+FoxP3+ suppressor cells amo
PBMCs obtainedInfrom
an HCV-infected
patient (representative
of 4 patients)
314 were
response to294
a large number of proteins is inhibited by a single
contrast
to HCV_G1_p7_794,
the human
peptide
analogue
278
herein is the first, however, to identify a promiscuous
HCV ID
peptotal
PBMCs by a greater than 10:1 ratio.
Patient
+
+
+
incubated in
medium
alone
(A)
or
medium
that
contained
HCV_G1_p7_794
(B).
295
315
elicited a significant increase
or cells
limited number of nTreg cell
For ques)ons regarding immunogenicity predic)on services and deimmuniza)on 265
op)ons, lease contact: elsey 279Hazelwood t 401-­‐272-­‐2123, xt. 149; or at i(p7_794)
nfo@epivax.com in CD3
CD4
FoxP3
www.epivax.com clones, responsive to a common
tidepatients,
sequence a
(HCV_G1_p7_794)
that exhibitseextensive
human
lating inpthe
peripheral
blood of K
chronically-infected
*