1 JOSE CARRERAS INTERNATIONAL LEUKEMIA FOUNDATION FORM 1 Jaebok Choi, PhD Project Title Epigenetic modulation of GvHD and GvL by In Vivo Administration of Azacitidine Applicant Name Jaebok Choi Address 660 South Euclid Avenue Campus Box 8007 St. Louis, MO 63110-1010, USA e-mail [email protected]Telephone (314)362-9335 Fax (314)362-9333 Sponsor Name John F. DiPersio, MD PhD Address 660 South Euclid Avenue Campus Box 8007 St. Louis, MO 63110-1010, USA e-mail [email protected]Telephone (314)362-9337 Fax (314)362-9333 Institutional Financial Officer Name Joe Gindhart Address 700 Rosedale Avenue Campus Box 1034 St. Louis, MO 63122-1408 e-mail [email protected]Telephone (314)935-7089 Fax (314)935-4309 Project will involve: [X] Biohazards [_] Human Subjects [_] Adults [_] Children [_] Fetal Material [X] Laboratory Animals
43
Embed
JOSE CARRERAS INTERNATIONAL LEUKAEMIA FOUNDATION · relapsed/refractory leukemia, and marrow failure states such as myelodysplasia and aplastic anemia. However, allogeneic BMT is
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
JOSE CARRERAS INTERNATIONAL LEUKEMIA FOUNDATION
FORM 1 Jaebok Choi, PhD
Project Title Epigenetic modulation of GvHD and GvL by In Vivo Administration of
Aim 3. Determine the molecular mechanisms underlying the suppressor function of Tregs and azacTs.
1) Determine whether the expression of the three candidate genes is regulated via DNA methylation
using bisulfite sequencing.
2) Determine the roles of three candidate genes in Tregs and azacTs by retrovirus-mediated
overexpression and viral-based RNAi.
3) Validate the differential expression of these candidate genes in human Tregs and azacTs.
B. Background and Significance: Allogeneic BMT is the only curative treatment for patients with
relapsed/refractory leukemia, and marrow failure states such as myelodysplasia and aplastic anemia.
However, allogeneic BMT is complicated by GvHD which is mediated by alloreactive donor T cells and can
be life threatening especially in recipients of unrelated or HLA-mismatched stem cell products. These same
alloreactive donor T cells can mediate GvL. Managing the threat of GvHD while maximizing the beneficial
GvL effect would broaden the scope and usefulness of allogeneic BMT. In animal models Tregs have been
shown to prevent GvHD by suppressing alloreactive donor T cells without sacrificing GvL, thereby
providing a promising treatment option2. However, several limitations have prevented the routine clinical use
of Tregs7-10
: 1) the low circulating numbers of Tregs in peripheral blood, 2) loss of suppressor activity
following in vitro expansion and 3) the lack of Treg-specific surface markers necessary to purify in vitro
expanded Tregs. We have recently reported that in vivo AzaC treatment of mice transplanted with allogeneic
T cells increases donor Tregs in vivo, thereby mitigating GvHD while maintaining GvL6. These data suggest
that epigenetic immunomodulation using AzaC may be a simple alternative to cellular Treg therapy. Our
study, if successful, will provide insights into cellular (Aim 1) and molecular (Aim 3) mechanisms
underlying AzaC-mediated regulation of GvHD and GvL and the foundation for future clinical trials aimed
at mitigating GvHD and overcoming HLA barriers while maintaining a potent GvL effect in humans
undergoing allogeneic BMT (Aim 2).
B.1. Immunomodulatory effect of AzaC. FOXP3 is a forkhead box transcription factor exclusively
expressed in CD4+CD25+ Tregs11-13
. It has been shown that mutations in Foxp3 lead to autoimmune
diseases due to the loss of functional Tregs and that forced expression of Foxp3 in non-Tregs, such as
CD4+CD25- T cells, results in conversion of these T cells to functional Tregs11-16
. This suggests that Foxp3
is necessary and sufficient for the development of Tregs. Interestingly,
the Foxp3 locus in both humans and mice is hypomethylated in Tregs
but are heavily methylated in CD4+CD25- T cells3-5
. These findings led
us to examine whether the FDA-approved hypomethylating agent,
AzaC, could be used to enhance the expression of Foxp3 via epigenetic
modulation and convert CD4+CD25- alloreactive T cells into functional
Tregs in an effort to mitigate GvHD after allogeneic BMT. We have
4
Fig 2. Effect of AzaC on GvHD and GvL. a. *p<0.0001. Y axis (b) indicates photon flux in log scale measured over the entire body of each mouse (BLI). Figure from ref. 6.
Fig 3. Relative mRNA expression of three candidate genes in MoFlo sorted cells.
recently reported that AzaC markedly induced FOXP3 expression6 (Fig. 1) and that these AzaC-induced
Tregs (azacTs) were able to function as suppressors in vitro6. More
importantly, in vivo AzaC treatment of mice transplanted with allogeneic
conventional pan T cells (Tconv) (B6 to Balb/c) resulted in the absence of
GvHD-related symptoms and significantly improved survival compared to
mice that received PBS after transplant (Fig. 2a)6. In addition, while AzaC
treatment had no effect on the growth of luciferase-expressing (for BLI6,17
)
murine A20 leukemia cells (A20+PBS-Tconv vs. A20+AzaC-Tconv), the
AzaC group (A20+AzaC+Tconv) had a significantly lower leukemic
burden and percentage of mice with leukemia than the control group
(A20+AzaC-Tconv) (Fig. 2b) and a significantly improved survival rate
than the PBS group (A20+PBS+Tconv)6. These data strongly suggest that
AzaC preferentially suppresses GvHD while maintaining GvL6. In this
proposal, we will test four hypotheses mentioned above in Section A to
elucidate the cellular mechanisms underlying the AzaC-mediated
regulation of GvHD and GvL. We will also test the hypothesis that AzaC
induces similar immunomodulatory effects in human T cells using
xenograft GvHD/leukemia models.
B.2. Molecular mechanisms of the suppressor function of Tregs and azacTs. We found that the
suppressor function of azacTs was independent of granzymes A and B but partially dependent on perforin 16.
In addition, their suppressor function was cell-contact dependent6. Most importantly, the suppressor
function of azacTs was Foxp3 independent, based on the observations that azacTs generated from Foxp3 -/-
T cells were as suppressive as azacTs from wild-type littermate controls in vitro and that in vivo
administration of AzaC significantly prolonged survival of Foxp3 -/- mice by suppressing autoimmunity6.
Considering the global hypomethylating effect of AzaC, it is likely that AzaC not only upregulates the
expression of Foxp3 but also other genes critical for suppressor function and that these genes are
regulated in a similar fashion to Foxp3, via DNA methylation. Our genome-wide RNA profiling analyses
identified three genes in addition to Foxp3 (84 fold) consistently upregulated in naïve Tregs (nTregs), anti-
CD3/CD28 bead-activated Tregs (aTregs), and azacTs compared to PBS-
treated T cells (pbsTs)6. 2700079J08Rik (11 fold) may encode a short non-
coding RNA with no known function. Tmem176a (9 fold) has recently been
shown to be highly expressed in a model of allograft tolerance and Ms4a4c
(7 fold) has no known function but its homolog Ms4a4b is critically
involved in Th1 function and/or differentiation18-20
. We validated that these
three genes were overexpressed in nTregs, expanded Tregs (expTregs) and
azacTs compared to pbsTs using real-time RT-PCR (Fig. 3). We
hypothesize that overexpression of one or more of these candidate genes is
responsible for the suppressor function of Tregs and azacTs.
C. Materials and Methods:
C.1. Aim 1. Determine the efficacy of AzaC to mitigate GvHD while maintaining GvL.
C.1.1. Determine effect of AzaC on trafficking and expansion of donor T cells using BLI. Allogeneic
BMT will be performed as follows6,17
. The recipient mice (Balb/c; H-2d) will be conditioned with lethal
irradiation (925 cGy) (day -1) and injected with T cell depleted bone marrow (TCD BM) (5x106 cells)
obtained from donor mice (B6, CD45.2+; H-2b) (day 0). We will transduce Tconv with a murine retrovirus
containing the click beetle red luciferase/GFP fusion (CBRluc/gfp) gene and inject mice with these
transduced Tconv (2x106) (day 11). Mice will be subcutaneously injected with either PBS or AzaC (2 mg/kg,
day 15, 17, 19 and 21) and BLI will be performed immediately prior to the first dose of AzaC, three days
after the last injection of AzaC, then weekly until the last day of experiment (day 52) to compare the
trafficking and expansion of the donor Tconv in collaboration with Dr. David Piwnica-Worms at Molecular
Imaging Center at WUSM.
C.1.2. Determine effect of AzaC on myeloid and lymphoid leukemia trafficking and expansion.
Allogeneic BMT will be performed as follows. The recipient mice (Balb/c; H-2d) will be conditioned (925
cGy) (day -1) and injected with TCD BM (5x106 cells) (B6, CD45.2+; H-2b) along with 1x10
4 A20-
CBRluc/gfp leukemia cells or murine APL-CBRluc/gfp cells21
(day 0; both H-2b), followed by the injection
of Tconv (2x106) (day 11) and AzaC (C.1.1). BLI will be performed one day prior to donor T cell infusion,
immediately prior to the first dose of AzaC, three days after the last injection of AzaC, then weekly until the
Choi, Jaebok
5
Fig 4. Effect of AzaC on human FOXP3.
last day of experiment (day 52) to compare the trafficking and expansion of the leukemia cells.
C.1.3 Determine whether AzaC-mediated prevention of GvHD requires Tregs using Foxp3DTR
KI mice.
Recently, Perez-Simon and colleagues proposed that AzaC might directly suppress the proliferation of
alloreactive T cells and thus GvHD, based on in vitro cell culture data
22. In contrast, our data demonstrated
that AzaC prevents GvHD by increasing donor T cell-derived FOXP3+ Tregs6. To address the requirement
for Tregs in the AzaC-mediated inhibition of GvHD, we obtained Foxp3DTR
KI mice that express the human
diphtheria toxin receptor (DTR) under the control of the Foxp3 promoter23
. BMT and AzaC treatment will be
performed as described in C.1.1 with one modification in which 2x106 T cells obtained from Foxp3
DTR mice
(CD45.2+) will be injected (day 11) in place of transduced Tconv. Diphtheria toxin (50 μg/kg body weight;
Sigma) will be injected intraperitoneally after each AzaC treatment (days 16, 18, 20, and 22) to deplete
FOXP3+ Tregs. We hypothesize that in vivo depletion of FOXP3+ Tregs after administration of diphtheria
toxin will lead to the loss of alloreactive T cell suppression and an exacerbation of lethal GvHD. If true, the
data would support our belief that AzaC prevents GvHD by increasing the peripheral conversion of
CD4+CD25
- alloreactive T cells into functionally suppressive FOXP3
+ Tregs. If our hypothesis is incorrect
and AzaC continues to prevent GvHD in the absence of Tregs, the data would support the theory proposed
by the Perez-Simon group22
that AzaC directly suppresses the proliferation of alloreactive T cells.
C.1.4. GvHD assessment. Throughout this proposal, mice will be weighed and monitored for survival and
signs of GvHD as describe by Cooke et al.24
. All surviving animals will be bled four days after the last
treatment of AzaC to determine complete blood counts and to assess donor chimerism and engraftment using
flow cytometry. Mice will be sacrificed at the last day of experiment (day 100 and day 50 for xenograft
models below section C.2) or at the time of >20% weight loss from GvHD and peripheral blood and
splenocytes will be analyzed using flow cytometry to determine donor chimerism and engraftment. Portions
of colon, kidney, liver, lung, small intestine, skin, and spleen will be saved for assessment of donor T cell
subset infiltration and GvHD by histology, immunohistochemistry, and fluorescence microscopy.
C.1.5. Statistics. Throughout this proposal, all animal studies will be statistically analyzed using Kaplan-
Meier survival curves and log rank test to determine statistical significance between groups. 10 mice each
group will be used in multiple independent experiments. Dr. William Shannon at Washington University
School of Medicine (WUSM) will be our statistics consultant.
C.2. Aim 2. Determine the role of AzaC on GvHD and GvL in a xenograft GvHD/leukemia model.
C.2.1. Determine effect of AzaC on xenogeneic GvHD using a NOD/SCID/γcnull
(NSG) mouse model.
The effect of AzaC on human T cells in vitro is comparable to that in murine T cells (Fig. 4). Our lab has
developed a unique and informative xenotransplant model to assess the GvHD potential of human T cells in
immunodeficient mice25
. NSG immunodeficient mice will be conditioned with sublethal irradiation (250
cGy). Naïve human T cells isolated from human PBMCs using the AutoMACS will be administered one day
after sublethal irradiation by retro-orbital injection (3x106 cells), followed by AzaC injection. Various doses
of AzaC (0.1, 0.5, 1 or 5 mg/kg) will be injected subcutaneously every other day into the mice at various
time points (starting at day 5, 7, or 10 post T cell infusion; up to 4 doses). GvHD assessment will be
performed as described above (C.1.4) to determine optimal timing and dosing of AzaC treatment (C.1.5).
C.2.2. Determine effect of AzaC on GvL using NSG mice engrafted with luciferase-expressing G2
human B cell-derived ALL cells. Methods to evaluate GvL will be identical to the GvHD studies described
above with the following modifications. Luciferase-expressing G2 cells were generated in our lab by
transducing the G2 human pre-B ALL cell line26,27
with retrovirus containing the CBRluc/gfp gene. NSG
mice will be injected with 1x106 G2-CBRluc/gfp leukemia cells on day -15. These mice will be sublethally
irradiated (250 cGy; no effect on G2 expansion in vivo) on day -1, followed by the injection of human T cells
(3x106) on day 0, followed by AzaC treatment as determined in C.2.1. GvL assessment will be performed
using flow cytometry of peripheral blood and BLI at day 4 post human T cell infusion, and then weekly until
day 46 post human T cell infusion.
C.3. Aim 3. Determine the molecular mechanisms underlying
the suppressor function of Tregs and azacTs.
C.3.1. Determine whether the expression of the three candidate
genes is regulated via DNA methylation using bisulfite
sequencing. Using the CpG Island Searcher28
, we identified
putative CpG islands in their upstream regulatory sequences.
Therefore, we will perform bisulfite sequencing to further confirm that these three genes are regulated by
DNA methylation. First, we will isolate/generate nTregs, GFP+ expanded Tregs, GFP- expanded Tregs,
GFP+ azacTs, GFP- azacTs, naïve CD4+CD25- T cells, and pbsTs from Foxp3-ires-GFP KI mice. Next,
Choi, Jaebok
6
genomic DNA isolation and bisulfite conversion will be performed with the ZR genomic DNA II kit and EZ
DNA Methylation-Gold (Zymo Research). Nested PCR primers for each CpG target have been designed
using MethPrimer29
. PCR products will be cloned using a TOPO TA cloning kit (Invitrogen). Plasmid DNA
from at least 15 colonies will be sequenced (see Section D).
C.3.2. Determine the roles of three candidate genes in Tregs and azacTs. Since antibodies against these
three candidate genes are not commercially available, our study will be focused on their roles in the
suppressor function of Tregs and azacTs.
C.3.2.1. Retrovirus-mediated over expression of three candidate genes in non-Tregs. We will first
overexpress these genes via retroviral transduction in CD4+CD25- T cells. The retroviral construct will
include GFP as a cell-sorting marker. GFP+ cells will be sorted using the MoFlo cell sorter. Next, we will
perform mixed lymphocyte reaction (MLR) to determine if the overexpression of any of these candidate
genes can convert non-Tregs into suppressive Treg-like cells. nTregs and GFP+ transductants with empty
vector will be used as positive and negative controls, respectively. To further validate the role of each
candidate gene we will also overexpress them in bulk CD4+ T cells from Foxp3 -/- mice to determine if
forced expression can rescue the loss of suppressor function seen in Foxp3 -/- T cells using MLR. If one of
these candidate genes is indeed sufficient for the generation and suppressor function of Tregs and azacTs in
vitro, we will attempt to validate their suppressor function in vivo by performing allogeneic BMT. BMT will
be performed as follows. 5x106 TCD BM (B6, CD45.2+) and 5x10
5 naive Tconv (B6, CD45.1+) along with
5x105 non-Tregs overexpressing one of these three candidate genes will be injected into lethally irradiated
(925cGy) recipient mice (Balb/c; H-2d) on day 0. nTregs and GFP+ transductants with empty vector will
function as positive and negative controls respectively. All surviving animals will be bled at days 30 and 60
post BMT to determine complete blood counts, donor chimerism and engraftment using flow cytometry.
Mice will be bled and sacrificed at the last day of experiment (day 100 post BMT) and peripheral blood,
splenocytes and other GvHD target organs will be analyzed as described above (C.1.4).
C.3.2.2. viral-based RNAi-mediated down regulation of three candidate genes in Tregs and azacTs. To
determine whether these genes are necessary for the suppressor function of Tregs and azacTs, we will knock
down the expression of these genes in Tregs and azacTs using viral-based RNAi (Section D). Gene knock
down will be confirmed by quantitative RT-PCR. Their suppressor function will be determined by
performing MLR in which these cells will function as suppressors, CFSE-labeled CD4+CD25- Teff as
responders (both from B6), and irradiated whole splenocytes from Balb/c as stimulators. We will next
validate their suppressor function in vivo by performing allogeneic BMT as described above (C.3.2.1). To
test whether the viral-based RNAi-transduced Tregs and azacTs lose their suppressor properties, we will
inject them on day 0 in a 1:1 ratio with Tconv and both survival and GvHD will be assessed (C.1.4 and
C.3.2.1). Positive (Tregs and azacTs transduced with scramble shRNA) and negative (pbsTs transduced
either with scramble shRNA or with the RNAi construct) controls will be tested. More than one candidate
genes may be necessary to induce suppressor function. Therefore, we will also overexpress combinations of
genes using IRES or 2A peptide base retroviral vectors30
currently in use in our lab.
C.3.3. Validate the differential expression of the three candidate genes in human Tregs and azacTs. We
will examine mRNA expression of the three candidate genes in human Tregs (naïve and anti-CD3/CD28
bead-activated) and azacTs using real-time RT-PCR (QuantiTect RT-PCR kit (Qiagen)). Since two of the
candidate genes, Tmem176a (human ortholog is TMEM176A) and Ms4a4c (a.k.a. Ms4a9 and the human
ortholog is MS4A10), are predicted to produce transmembrane proteins, they may provide new cell surface
markers for the identification and purification of nTregs or in vitro expanded Tregs and azacTs. Successful
results will lead us to examine the function of these genes in human Tregs and AzaC-induced Tregs as
described in mouse models. However, this is outside the scope for this proposal.
D. Institutional Environment and Support. Besides the support provided by Drs. Piwnica-Worms and
Shannon for BLI and Biostatistics, respectively, our lab has closely interacted with Drs. Tim Ley (Foxp3-
ires-GFP KI) and Chyi-Song Hsieh (Foxp3 -/- and Foxp3DTR
). The resources and services provided include
Siteman Cancer Center Core Facilities (High Speed Cell Sorter Core; MoFlo and FACScan, Biostatistics
Core, Clinical Trials Core, Good Manufacturing Practice (GMP) Facility, and Multiplexed Gene Analysis
Core) as well as ICTS cores including the Research Design and Biostatistics and Biomedical Informatics. All
viral-based RNAi constructs will be purchased at the Genome Center at WUSM. DNA sequencing will be
performed at the Genome Center (WUSM). Our lab (AutoMACS) is juxtaposed to many of the Siteman
Cancer Center Core facilities and all other members of the Section of Stem Biology in which the faculty
have had a long track record of mentoring and training future independent scientists and physician-
investigators who will enhance the success in the project described in this proposal.
Choi, Jaebok
7
E. Literature Cited:
1. Haribhai D, Lin W, Relland LM, Truong N, Williams CB, Chatila TA. Regulatory T cells
dynamically control the primary immune response to foreign antigen. J Immunol. 2007;178:2961-
2972.
2. Edinger M, Hoffmann P, Ermann J, et al. CD4+CD25+ regulatory T cells preserve graft-
versus-tumor activity while inhibiting graft-versus-host disease after bone marrow transplantation.
Nat Med. 2003;9:1144-1150.
3. Floess S, Freyer J, Siewert C, et al. Epigenetic control of the foxp3 locus in regulatory T
cells. PLoS Biol. 2007;5:e38.
4. Baron U, Floess S, Wieczorek G, et al. DNA demethylation in the human FOXP3 locus
discriminates regulatory T cells from activated FOXP3(+) conventional T cells. Eur J Immunol.
2007;37:2378-2389.
5. Kim HP, Leonard WJ. CREB/ATF-dependent T cell receptor-induced FoxP3 gene
expression: a role for DNA methylation. J Exp Med. 2007;204:1543-1551.
6. Choi J, Ritchey J, Prior JL, et al. In vivo administration of hypomethylating agents mitigate
graft-versus-host disease without sacrificing graft-versus-leukemia. Blood. 2010;116:129-139.
7. Yamazaki S, Patel M, Harper A, et al. Effective expansion of alloantigen-specific Foxp3+
CD25+ CD4+ regulatory T cells by dendritic cells during the mixed leukocyte reaction. Proc Natl
Acad Sci U S A. 2006;103:2758-2763.
8. Allan SE, Broady R, Gregori S, et al. CD4+ T-regulatory cells: toward therapy for human
diseases. Immunol Rev. 2008;223:391-421.
9. De Rosa V, Procaccini C, Cali G, et al. A key role of leptin in the control of regulatory T
cell proliferation. Immunity. 2007;26:241-255.
10. Karakhanova S, Munder M, Schneider M, Bonyhadi M, Ho AD, Goerner M. Highly
efficient expansion of human CD4+CD25+ regulatory T cells for cellular immunotherapy in
patients with graft-versus-host disease. J Immunother. 2006;29:336-349.
11. Hori S, Nomura T, Sakaguchi S. Control of regulatory T cell development by the
Piwnica-Worms and John F. DiPersio Jaebok Choi, Julie Ritchey, Julie L. Prior, Matthew Holt, William D. Shannon, Elena Deych, David R.
graft-versus-host disease without sacrificing graft-versus-leukemiaIn vivo administration of hypomethylating agents mitigate
http://bloodjournal.hematologylibrary.org/cgi/content/full/116/1/129Updated information and services can be found at:
(1542 articles)Transplantation � collections: BloodArticles on similar topics may be found in the following
http://bloodjournal.hematologylibrary.org/misc/rights.dtl#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:
http://bloodjournal.hematologylibrary.org/misc/rights.dtl#reprintsInformation about ordering reprints may be found online at:
http://bloodjournal.hematologylibrary.org/subscriptions/index.dtlInformation about subscriptions and ASH membership may be found online at:
. Hematology; all rights reservedCopyright 2007 by The American Society of 200, Washington DC 20036.semimonthly by the American Society of Hematology, 1900 M St, NW, Suite Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published
only.For personal use at WASHINGTON UNIV SCH MEDICINE on October 12, 2010. www.bloodjournal.orgFrom
In vivo administration of hypomethylating agents mitigate graft-versus-hostdisease without sacrificing graft-versus-leukemiaJaebok Choi,1 Julie Ritchey,1 Julie L. Prior,2 Matthew Holt,1 William D. Shannon,3 Elena Deych,3 David R. Piwnica-Worms,2
and John F. DiPersio1
1Division of Oncology, Department of Medicine, 2Molecular Imaging Center, Mallinckrodt Institute of Radiology and Department of Developmental Biology, and3Division of General Medical Sciences, Department of Medicine, Washington University, St Louis, MO
Regulatory T cells (Tregs) suppress graft-versus-host disease (GVHD) while preserv-ing a beneficial graft-versus-leukemia (GVL)effect. Thus, their use in allogeneic stemcell transplantation (SCT) provides apromising strategy to treat GVHD. How-ever, 3 obstacles prevent their routineuse in human clinical trials: (1) low circu-lating number of Tregs in peripheral blood,(2) loss of suppressor function after invitro expansion, and (3) lack of Treg-specific surface markers necessary forefficient purification. FOXP3 is exclu-sively expressed in Tregs and forced ex-
pression in CD4�CD25� T cells can con-vert these non-Tregs into Tregs withfunctional suppressor function. Here, weshow that the FDA-approved hypomethy-lating agents, decitabine (Dec) and azaci-tidine (AzaC), induce FOXP3 expressionin CD4�CD25� T cells both in vitro and invivo. Their suppressor function is depen-dent on direct contact, partially depen-dent on perforin 1 (Prf1), but independentof granzyme B (GzmB), and surprisingly,Foxp3. Independence of Foxp3 suggeststhat genes responsible for the suppres-sor function are also regulated by DNA
methylation. We have identified 48 candi-date genes for future studies. Finally,AzaC treatment of mice that received atransplant of major histocompatibilitycomplex mismatched allogeneic bonemarrow and T cells mitigates GVHD whilepreserving GVL by peripheral conversionof alloreactive effector T cells into FOXP3�
Tregs and epigenetic modulation of genesdownstream of Foxp3 required for thesuppressor function of Tregs. (Blood.2010;116(1):129-139)
Introduction
Allogeneic stem cell transplantation (SCT) represents the mosteffective treatment for patients with marrow failure states and otherhematologic malignancies such as acute and chronic leukemias.One of the major complications of allogeneic SCT is graft-versus-host disease (GVHD), caused by donor T cells reacting against hostantigens.1 This acute inflammatory reaction can be mild, moderate,or life-threatening especially in recipients of unrelated or humanleukocyte antigen–mismatched stem cell products.2 However, thesesame alloreactive donor T cells provide a beneficial graft-versus-leukemia (GVL) effect, reducing the risk of leukemia relapse.3,4
Therefore, the current clinical goal in treatment of GVHD is topreferentially suppress GVHD while maintaining GVL.
Regulatory T cells (Tregs) are known to contribute to themaintenance of self-tolerance by regulating inflammatory re-sponses and to suppression of autoimmunity and GVHD in mousemodels.5-9 The major population of Tregs is naturally occurringTregs or nTregs. They are generated in the thymus and defined byCD4�CD25�FOXP3�.5-8 Small number of Tregs can also begenerated in the periphery from naive CD4�CD25� T cells byT cell–receptor stimulation along with retinoic acid, TGF-�, andIL-10.10,11 Because Tregs can also mitigate GVHD by suppressingalloreactive donor T cells without sacrificing GVL in animalmodels, their use in the allogeneic transplantation setting providesa promising strategy to treat or mitigate GVHD.9 However,circulating numbers of Tregs in peripheral blood are limited
(5%–10% of CD4� T cells), and despite significant improvementsin methodologies for in vitro purification of Tregs, the currentprotocols for in vitro Treg expansion are inefficient, costly,and time-consuming.12-15 Furthermore, the lack of Treg-specificcell surface markers makes it impossible to purify Tregs expanded invitro, and expanded Tregs often fail to maintain their suppressorfunction,13,16 possibly due to the loss of expression of FOXP3 and/orchemokine receptors, such as CXCR3,17 CCR6,18 and CCR819 thatfacilitate trafficking of Tregs to sites of inflammation.
FOXP3 is a forkhead box transcription factor exclusivelyexpressed in nTregs.5-8 Its mutations lead to autoimmunediseases due to the loss of functional nTregs and forcedexpression of FOXP3 in CD4�CD25� T cells induces regulatoryproperties.5,7,8,20-22 These data suggest that Foxp3 is necessaryand sufficient for functional nTregs. Recent reports demon-strated that the Foxp3 locus in both humans and mice isunmethylated in Tregs while heavily methylated and silenced inCD4�CD25� T cells.23-25
Dec and AzaC, analogues of 2�-deoxycytidine and cytidine,respectively, are hypomethylating agents that the FDA approvedfor the treatment of myelodysplastic syndromes. Dec can incorpo-rate into replicating DNA, while AzaC incorporates primarily intoRNA with some integration into DNA after 5-aza-ribonucleotidesare converted into 5-aza-deoxyribonucleotides by ribonucleotidereductase.26-29 Once incorporated into DNA, they can trap DNA
Submitted December 1, 2009; accepted April 21, 2010. Prepublished online asBlood First Edition paper; April 27, 2010; DOI 10.1182/blood-2009-12-257253.
Presented in abstract form at the 50th annual meeting of the American Societyof Hematology, San Francisco, CA, December 8, 2008.
The online version of this article contains a data supplement.
The publication costs of this article were defrayed in part by page chargepayment. Therefore, and solely to indicate this fact, this article is herebymarked ‘‘advertisement’’ in accordance with 18 USC section 1734.
Based on these reports, we hypothesized that Dec and AzaCcould be used to induce the expression of FOXP3 in CD4�CD25�
T cells via epigenetic modification and convert these non-Tregsinto Tregs. In this study, we report that these drugs induce theexpression of Foxp3 in activated CD4�CD25� T cells generatingfunctional Tregs with suppressor properties. We further demon-strate that in vivo treatment of mice with AzaC after allogeneicSCT dramatically mitigates GVHD while preserving GVL at leastin part by increasing the peripheral conversion of CD4�CD25�
alloreactive T effector cells (Teffs) into functionally suppressiveFOXP3� Tregs. In addition, the suppressor function of theseAzaC-induced Tregs is independent of Foxp3, suggesting thatAzaC modifies the expression of not only Foxp3 but also othergenes that are necessary for Treg suppressor function. Thus, ourstudy suggests that epigenetic modulation of events distal to Foxp3is also a critical mechanism by which in vivo administration ofAzaC controls GVHD. Our study provides a solid foundation for apharmacologic therapy to limit GVHD without sacrificing GVL.
Methods
Mice
Balb/c (H-2Kd, CD45.2�) and C57BL/6 (B6; H-2Kb, CD45.2�) mice wereobtained from Taconic Farms. Congenic B6 mice expressing the CD45.1gene were purchased from The Jackson Laboratory. Animal care andeuthanasia were approved by the Washington University School of Medi-cine Animal Studies Committee. Six- to 12-week-old mice were usedexcept for Foxp3 KO experiments in which 3-week-old Foxp3 KO (B6background) mice5 and littermate control mice were used.
Cell culture
Human peripheral blood mononuclear cells (PBMCs) were harvested byficoll gradient centrifugation. CD4�CD25� T cells and CD4�CD25�
T cells were isolated from the human PBMCs using Miltenyi microbeadsand AutoMACS (Miltenyi Biotec).31 The isolated CD4�CD25� T cellswere activated for 2 to 3 days in the presence of anti-CD3/CD28 beads(bead:cell � 1:1; Invitrogen) and Stemline T-cell expansion medium (Sigma-Aldrich) supplemented with L-glutamine (4mM), penicillin (100 U/mL),streptomycin (100 �g/mL), and human recombinant IL-2 (hIL-2; 50 U/mL).The activated T cells were incubated in the presence of Dec (0.1-10�M;Sigma-Aldrich) or phosphate-buffered saline (PBS) for an additional 1 to4 days. Mouse CD4�CD25�T cells and CD4�CD25� T cells were isolatedfrom mouse spleens using Miltenyi microbeads and AutoMACS. Theisolated CD4�CD25� T cells were activated for 2 days (2 or 4 days forFoxp3 KO and littermate controls) in the presence of beads (bead:cell � 1:1) and Dulbecco modified Eagle medium supplemented with 10% FCS,L-glutamine (2mM), penicillin (100 U/mL), streptomycin (100 �g/mL),MEM nonessential amino acids (1�), sodium pyruvate (1mM), HEPESbuffer (20mM), 2-mercaptoethanol (50�M) and hIL-2 (10 U/mL), calledXcyte media. The activated T cells were incubated in the presence of Dec(0.1-10�M), AzaC (0.5-4�M), or PBS for an additional 2 days.
Real-time RT-PCR
Human CD4�CD25� T cells that were activated with beads for 3 days(defined as day 0) and incubated in the presence of Dec or PBS foradditional 1 to 4 days (day 1 through day 4) as described in the previousparagraph. Total RNA was isolated from these cells every day (day 0through day 4) using the RNeasy Plus Mini Kit (QIAGEN). The 7300 RealTime PCR system (Applied Biosystems), QuantiTect Primer Assaysprimers (GAPDH as an internal control), and QuantiTect SYBR GreenRT-PCR kits (QIAGEN) were used.
RNA profiling analysis
Total RNA was isolated from the following cells (all B6, CD45.2�) usingTRIzol Reagent (Invitrogen): CD4�CD25� naive Tregs, CD4�CD25�
Tregs that were incubated in Xcyte media (100 U/mL hIL-2) for 4 days inthe presence of the beads (bead:cell � 1:1), PBS-treated T cells (pbsTs),and Dec-induced Tregs (dcTs). Target preparation, gene chip hybridization,and analysis were performed by the Washington University Siteman CancerCenter Gene Chip Facility. Labeled target was made from non-amplifiedtotal RNA and was hybridized to Mouse Genome 430 2.0 array (Affymetrix).
Flow cytometric analysis
The following antibodies were used. For human T cells: CD4-APC,CD25-FITC (BD Pharmingen), and FOXP3-PE (clone PCH101; eBio-science), for mouse T cells: CD4-FITC, CD25-PE (BD Pharmingen),Foxp3-PE and Foxp3-PECy5 (clone FJK-16s; eBioscience), and for SCT:H-2Kb-FITC, CD4-PE, CD3-PECy7, B220-APC and CD45.2-biotin/streptavidin APC-Cy7 (BD Pharmingen). All cells were analyzed on aFACScan cytometer (BD Biosciences).
CFSE-based proliferation assays and MLR
CD4�CD25� T cells (B6, CD45.1) were labeled with carboxyfluoresceindiacetate, succinimidyl ester (CFSE) at a final concentration of 300nM. TheCFSE-labeled cells were incubated in 200 �L of Xcyte media withsuppressors, such as CD4�CD25� nTregs, dcTs, azacTs, or pbsTs, (all B6,CD45.2), and �-irradiated (20 Gy) splenocytes/antigen presenting cells(APCs; Balb/c, CD45.2) for mixed lymphocyte reaction (MLR) or beads forproliferation assays. 1:1:1 � Teff:bead/APC:suppressor ratio was usedunless otherwise indicated. DcTs and azacTs are extensively washed withPBS (2�) before coculture with T effectors, thus T effectors are notexposed to any detectable levels of decitabine or azacitidine.
SCT
SCT was performed as previously described31 with the following modifica-tions. Total-body irradiation (TBI; 900 cGy) using a Mark I cesiumirradiator (J. L. Shepherd and Associates) was used to condition recipientmice (Balb/c). B6 mice (CD45.1 or CD45.2) T cell–depleted (TCD) bonemarrow (BM; 5 � 106 cells) was used as a stem cell source. To induceGVHD, 5 � 105 B6 conventional CD4�/CD8� T cells (Tconv; CD45.1)were infused along with donor (B6) TCD BM. To test suppressor functionof dcTs, 5 � 105 B6 CD4�CD25� nTregs (CD45.2; for control), pbsTs ordcTs (CD45.2) generated from CD4�CD25� T cells were injected withTCD BM and Tconv. In some cases, 10 � 106 pbsTs or dcTs (CD45.2)generated from Tconv were given in the place of Tconv and pbsTs or dcTsthat were derived from CD4�CD25� T cells. For delayed donor lympho-cyte infusions, 2 � 106 or 10 � 106 Tconv were given on day 11 after SCT.For examination of GVL effect, 1 � 104 A20-luc/egfp leukemic cells weregiven along with TCD BM. Animals losing 20% of their starting bodyweights were killed.
BLI
GVL effect assessment and bioluminescence imaging (BLI) of animalswere done as previously described.31,32 Briefly, mice were injected intra-peritoneally with 150 �g/g D-luciferin (Biosynth) in PBS and imaged10 minutes later. Imaging was done using a charge-coupled device camera(IVIS 100; Caliper Corporation; exposure time, 1-60 seconds; binning, 8;field of view, 15; f/stop 1, open filter) at the Molecular Imaging Center(Washington University). Mice were anesthetized using isoflourane (2.5%vaporized in O2). For analysis, total photon flux (photons per second) wasmeasured from a fixed region of interest over the entire abdomen and thoraxusing Living Image 2.50 and IgorPro software (Wavemetrics). BLI wasperformed 1 day before donor T-cell infusion and the first injection of AzaCand 3 days after the last injection of AzaC, then once every week untilday 52 after SCT.
130 CHOI et al BLOOD, 8 JULY 2010 � VOLUME 116, NUMBER 1 only.For personal use at WASHINGTON UNIV SCH MEDICINE on October 12, 2010. www.bloodjournal.orgFrom
AzaC and Dec were prepared in cold PBS and were injected 4 timessubcutaneously or intraperitoneally every other day starting on day 4after donor lymphocyte infusion (day 15 after SCT). Foxp3 KO micewere also injected 4 times subcutaneously every other day starting at16 to 26 days of age.
Statistical analysis
The significance of differences in survival of treatment groups wasanalyzed using the log-rank test. For proliferation assays and MLR,unpaired t test was used. P values less than .05 were considered significant.For RNA profiling analyses, the probes were identified using GA-Mantelmethod.33 This method uses genetic algorithm strategy to select probes thatoptimize the Mantel correlation among the samples. GA-Mantel was run100 times and the probes that appeared in at least 30 of 100 solutions werepicked. All analyses were performed using SAS Version 9.1 (SAS Institute).
Results
The effect of hypomethylating agents on FOXP3 expression inhuman and mouse CD4�CD25�FOXP3� T cells
Because the hypomethylating effect of Dec and AzaC can onlyoccur in cells actively undergoing cellular division,29 we stimulatedhuman CD4�CD25� T cells using anti-CD3/CD28 antibody-coated beads and hIL-2 (50 U/mL) before treatment with variousconcentrations of Dec and AzaC. Approximately 60% of Dec-treated CD4�CD25� T cells (dcTs) and only approximately 10% ofpbsTs expressed FOXP3 (Figure 1A). Maximum FOXP3 expres-sion was achieved after exposure to 5 to 10�M Dec for
4 days. However, cell counts significantly declined when T cellswere exposed to concentrations of Dec in excess of 5�M (cellcounts were 69.7% 14.1% of the 1�M Dec cell counts [n � 4]).Treating activated CD4�CD25� T cells or conventional T cells(Tconv; isolated using Miltenyi pan T-cell isolation kit) with 1�MDec for 4 days optimized both cell expansion and FOXP3expression. Real-time reverse transcription–polymerase chain reac-tion (RT-PCR) demonstrated levels of mRNA for FOXP3 that werecomparable to those seen in bead-activated nTregs (10- to 12-foldincrease above baseline; Figure 1B). Dec also induced FOXP3expression in approximately 80% of bead-activated murineCD4�CD25� T cells (Figure 1A). Consistent with these results,both Dec and AzaC markedly increased GFP expression inactivated CD4�CD25� T cells obtained from Foxp3-ires-GFPknock-in (KI) mice34 (Figure 1C). To determine the duration ofFOXP3 expression in dcTs, we stained for intracellular FOXP3daily for 7 days and found that approximately 50% of dcTsremained FOXP3� at day 7 (Figure 1D). These results suggest thatboth hypomethylating agents induce both FOXP3 mRNA andprotein expression in CD4�CD25� non-Tregs.
Suppressor function of hypomethylating agent-induced Tregsin vitro
We examined whether dcTs and AzaC-induced Tregs (azacTs) aresuppressive in proliferation assays in which anti-CD3/CD28 beadswere used as stimulators (Figure 2, supplemental Figures 1-2,available on the Blood Web site; see the Supplemental Materialslink at the top of the online article). Both dcTs and azacTs were ableto function as suppressors, while pbsTs had no effect (Figure 2A,
Figure 1. Effect of Dec on FOXP3 expression in anti-CD3/CD28 antibody coated bead-activated CD4�CD25� T cells. (A) Both human and mouse CD4�CD25� T cellsexpress FOXP3 after Dec treatment in the presence of anti-CD3/CD28 antibody coated beads. (B) Real-time RT-PCR for human FOXP3 mRNA was performed in triplicate atvarious times after Dec treatment of activated T cells. Levels of mRNA for FOXP3 increase each day and are comparable to that seen in bead-activated Tregs (brown) by day 4.(C) CD4�CD25� T cells obtained from Foxp3-ires-GFP KI mice confirm induction of FOXP3 expression after Dec and AzaC treatment (indirectly measured as GFPexpression). (D) DcT FOXP3 expression (blue) persists for at least 7 days, suggesting prolonged and stable expression of FOXP3 in vitro. Treg (red) is shown for comparison.A pool of 2 independent experiments.
HYPOMETHYLATING AGENTS AND GVHD 131BLOOD, 8 JULY 2010 � VOLUME 116, NUMBER 1 only.For personal use at WASHINGTON UNIV SCH MEDICINE on October 12, 2010. www.bloodjournal.orgFrom
supplemental Figures 1-2). Using Foxp3-ires-GFP KI mice, wesorted FOXP3� azacTs and FOXP3� azacTs based on GFPexpression. Only GFP� azacTs suppressed proliferation of Teffs inbead based activation assays, suggesting that the suppressoractivity was derived from FOXP3� (GFP�) cells (Figure 2B).Surprisingly, naive nTregs were less active than dcTs in suppress-ing the proliferation of CD4�CD25� Teffs at 1:1 nTreg:Teff ratiosbut could suppress proliferation of bead activated Teffs at higherTreg:Teff (2:1 or 4:1) ratios (Figure 2A). This unexpected resultcould be explained by the potent stimulatory effects of anti-CD3/CD28 beads on Teffs while naive nTregs are relatively slow toproliferate in response to beads and lack suppressive activity untilthey are activated. Thus, Teffs start to proliferate before nTregswhile dcTs and azacTs are already activated before coculture withthe Teffs. Supporting this idea, nTregs as well as dcTs and azacTswere able to function as suppressors in an in vitro MLR in thepresence of stimulatory allogeneic antigen presenting cells (APCs;Figure 2C-D). During the 6 days of an MLR, nTregs becomeactivated and suppress in these MLRs. In addition, when Tregswere activated with beads before the proliferation assay, weobserved suppression although activated nTregs were still lesssuppressive than dcTs (supplemental Figure 2).
The strong suppressor activity after Dec or AzaC treatmentmay be due to their effects on altering both specific gene andglobal gene expression profiles. It is, therefore, likely that Decand AzaC induce not only Foxp3, but other genes that arecritically involved in the suppressor function of nTregs, perhapseven more dramatically than in resting or activated nTregs,thereby making dcTs more potent than nTregs in these in vitrobead based activation assays. nTregs appear to have multiplemechanisms of suppressor function.35-44 In some cases nTregsuse CD25 or IL-35 as mediators of suppression while, in others,they use galectin-1, PRF1/GZMB, or CTLA-4.35,36,40-44 Theyeven secrete immunosuppressive cytokines, such as TGF-betaand IL-10, especially in the case of adaptive Tregs.37-39 Bycomparing the gene expression profiles of dcTs and nTregs, we
found that many of the candidate genes associated with Tregfunction (IL-35, CD25, galectin-1, Prf1, GzmB, TGF-beta, andIL-10) with the exception of CTLA-4 were expressed at higherlevels in dcTs than in naive or activated nTregs (supplementalTable 1). These data suggest that hypomethylating agentsmodulate genes that have been shown to be associated with thesuppressive phenotype of Tregs.
In vivo suppressor function of hypomethylating agent-inducedTregs generated in vitro
We next examined whether dcTs and azacTs are able to suppressGVHD in an allogeneic SCT model. Lethally irradiated Balb/cmice underwent a transplantation with TCD BM (CD45.2) andTconv (CD45.1) isolated from B6 mice along with dcTs, azacTs, orpbsTs generated from B6 (CD45.2; Figure 3A). Mice that under-went a transplantation with dcTs had complete donor chimerism,improved survival, and significantly less weight loss comparedwith mice that received pbsTs (Figure 3B-C and supplementalFigure 3). These results are consistent with less clinical GVHD inthe dcT group. Similar results were observed when azacTs weretested, resulting in improved survival compared with pbsTs (supple-mental Figure 4). These data suggest that both dcTs and azacTs canfunction as suppressor cells in vivo. The mean percentages ofdonor-derived B cells in the peripheral blood were also higher inthe dcT group than in the pbsT group (Figure 3D). Despite thereduced signs of GVHD seen in the dcT group compared with thepbsT group, the overall survival of the dcT group was inferior tothat of the nTreg group. One explanation is that the percentage ofFOXP3� cells in dcTs is only 60% to 80% compared with the over90% Treg group. Therefore, dcTs generated in vitro containsignificant FOXP3� CD4�CD25� activated Teffs (ie, FOXP3�
dcTs) which, like Tconv, may both promote engraftment andGVHD in murine transplant recipients. We therefore tested whetherdcTs generated from Tconv, as opposed to purified CD4�CD25�
T cells, could promote engraftment (by FOXP3� dcTs) and
Figure 2. DcTs suppress the proliferation of Teffs. CFSE-basedproliferation assays and MLR were performed with each popula-tion 2.5 � 104 per well in 96-well round-bottom plates. Fortranswell plate experiments, each population (1 � 105 per well) wasincubated in 96-well flat-bottom transwell plates in the presence ofbeads as stimulators. The proliferation of CFSE-labeled CD4�CD25�
T cells was analyzed after 3 days (cells with beads) or 6 days(cells with �-irradiated APCs) on a FACScan cytometer(BD Biosciences). All cultures were evaluated in triplicate. Hypo-methylating agent–treated cells suppress the proliferation of anti-CD3/CD28 antibody coated bead-activated Teffs (A-B) and allogeneicAPC-activated Teffs (C-D). Treg (1�), (2�), and (4�) indicate theratios of Treg:Teff � 1:1, 2:1, and 4:1, respectively. azacT (a) and (b)indicate that these azacTs were generated in the presence of AzaC1�M and 2�M, respectivley. dcTs, azacTs, and pbsTs (right) weregenerated in the presence of hIL-2 (500 �/mL; panel A). Only FACSpurified GFP� (thus, FOXP3�) and not GFP� CD4� cells obtainedfrom Foxp3-ires-GFP KI mice are suppressive (panel B). CD45.1was used to gate on CFSE-labeled Teffs. dcT: Dec-treatedT cells, pbsT: PBS-treated T cells, azacT: AzaC-treated T cells. Neg:negative control, CFSE-labeled Teffs alone; pos: CFSE-labeled Teffswith stimulators, anti-CD3/CD28 antibody coated beads or alloge-neic APC; all others contain both CFSE-labeled Teffs and stimulatorsplus indicated cells such as nTregs, dcTs, pbsTs, or azacTs. GFP�
azacT: MoFlo sorted GFP� cells after treatment of AzaC; GFP�
azacT: MoFlo sorted GFP� cells after treatment of AzaC.
132 CHOI et al BLOOD, 8 JULY 2010 � VOLUME 116, NUMBER 1 only.For personal use at WASHINGTON UNIV SCH MEDICINE on October 12, 2010. www.bloodjournal.orgFrom
mitigate GVHD (by FOXP3� dcTs) in an allogeneic SCT (Figure3E). We found that mice that received these dcTs demonstratedimproved survival compared with mice that received pbsTs (Figure3F). In addition, mice that received Tconv-generated dcTs demon-strated complete donor chimerism and significantly higher numbersof donor B cells than mice that received pbsTs (Figure 3G andsupplemental Figure 5). These data are consistent with less clinicalGVHD in the dcT group.
The effect of in vivo treatment of mice after allogeneic SCT withhypomethylating agents: impact on engraftment, GVHD, andimmune reconstitution
We examined whether in vivo treatment of mice with hypomethyl-ating drugs after transplantation could mitigate GVHD by increas-ing FOXP3� Tregs in murine allogeneic transplantation models(Figure 4A). To minimize the death related to myelosuppression ofthese hypomethylating agents when given immediately after trans-plantation we first infused T cell–depleted allogeneic BM (5 � 106;CD45.2) after TBI (900 cGy) on day 0. After recovery of bloodcounts mice were then infused with 2 � 106 major histocompatibil-ity complex (MHC) mismatched Tconv on day �11. Mice weretreated with PBS, Dec (1 mg/kg) or AzaC (2 mg/kg) subcutane-ously on days �15, �17, �19, and �21. Despite delay intreatment with Dec, we observed that the survival rate of micetreated in vivo with Dec after allogeneic SCT was not different
from mice treated after transplantation with PBS due to theexcessive myelosuppressive effects of Dec (all deaths were associ-ated with pancytopenia; Figure 4B). In sharp contrast, AzaCtreatment of mice that underwent a transplantation with TCD BMplus 2 � 106 MHC mismatched Tconv cells resulted in no detect-able GVHD-related symptoms, such as weight loss, ruffled fur, ordiarrhea (Figure 4B-C). Most importantly, 100% of mice (23/23)that received posttransplantation AzaC and donor T cells in vivosurvived with donor engraftment (20/23 showed � 85% donorcells) and no GVHD similar to those mice that received only TCDBM without donor T cells and dramatically different from thosemice that received donor T cells without AzaC (3/23 survived;Figure 4B-D). We also found that the AzaC group had highpercentages of donor CD3� T cells and B220� B cells aftertransplantation, which is consistent with less clinical GVHD(Figure 4E-F). Significant increase of FOXP3� Tregs in theperipheral blood was also noted in mice treated in vivo with AzaCafter transplantation, suggesting that the suppression of GVHDmay be mediated, at least in part, by AzaC-induced peripheralconversion of FOXP3� donor T cells to FOXP3� Tregs in vivo(Figure 4G). Although it is possible that AzaC promoted theexpansion of infused donor nTregs resulting in reduced GVHD,mice that received Treg-depleted Tconv ( 98% FOXP3� Tconv)and were treated with AzaC in vivo also had dramatically reducedGVHD (supplemental Figure 6).
Figure 3. DcTs mitigate GVHD. (A) Schema of the experiments. 9 Gy total-body irradiation was used to condition recipient mice (Balb/c). B6 mice (CD45.2) TCD BM(5 � 106 cells) were used as a stem cell source. To induce GVHD, 5 � 105 B6 Tconv (CD45.1) was infused along with donor TCD BM. To test suppressor function of dcTs,5 � 105 B6 CD4�CD25� nTregs (for control), pbsTs or dcTs (all CD45.2) generated from CD4�CD25� T cells were injected with TCD BM and Tconv. (B-D) Mice that underwenta transplantation with dcTs show significantly higher survival rate (B) and less weight loss (C) than mice infused with pbsTs and a trend toward increase of B cells(D) analyzed 1 month after transplantation. A pool of 2 independent experiments. (E) Schema of the experiments. B6 mice (CD45.1) TCD BM (5 � 106 cells) were used as astem cell source. A total of 10 � 106 pbsTs/dcTs generated from Tconv (B6, CD45.2) were given. (F-G) Mice that underwent a transplantation with dcTs show significantlyhigher survival rate (F) and more B cells (G) than mice that received pbsTs. A pool of 3 independent experiments analyzed 1 month after transplantation.
HYPOMETHYLATING AGENTS AND GVHD 133BLOOD, 8 JULY 2010 � VOLUME 116, NUMBER 1 only.For personal use at WASHINGTON UNIV SCH MEDICINE on October 12, 2010. www.bloodjournal.orgFrom
Recently, Perez-Simon and colleagues proposed that AzaCmight directly suppress the proliferation of Tconv and thus GVHDbased on in vitro cell culture data in which more than 99% of Tconvwere blocked at G0/G1 phase after CD3/CD28 bead activation in
the presence of AzaC.45 To determine the impact of AzaC in vivo,Tconv were first labeled with CFSE and then infused on day �11after transplantation. Mice were treated in vivo with either PBS,or AzaC as described above. Peripheral blood was obtained on
Figure 5. AzaC increase FOXP3� Tregs that may inhibit the proliferation of allogeneic Tconv in vivo. In vivo AzaC treatment (blue) attenuates the proliferation ofallogeneic Tconv compared with the PBS control (red, left panel; peripheral blood on day 19 after SCT, gated on CD45.1� donor T cells). This inhibited proliferation of Tconv islikely to be mediated by FOXP3� Tregs induced by AzaC (middle and right; splenocytes on day 19 after SCT, gated on CD45.1� donor T cells). One representative from eachgroup with identical results is shown. AzaC (n � 4), PBS (n � 2).
Figure 4. AzaC treatment of mice that underwent a transplantation with delayed allogeneic T cells mitigates GVHD. (A) Schema of the experiments. B6 mice (CD45.2) TCD BM(5 � 106 cells) were used as a stem cell source. To induce GVHD, 2 � 106 Tconv (B6, CD45.1) were given on day 11 after SCT followed by the treatment withAzaC/Dec/PBS (every otherday; 4 doses) starting on day 15 after SCT. (B-F) Mice treated withAzaC (2 mg/kg) show significantly higher survival rate (B), less weight loss (C), and more B cells (E) and T cells (F) withdonor engraftment (D) than mice infused with pbsTs. (G) AzaC group also show increased Treg population in peripheral blood. T rec: T cells from recipient; T BM: T cells from donor BM;T donT: T cells from donor T cells; B BM: B cells from donor BM. (D-G)Analyzed 1 month after transplantation.Apool of 4 independent experiments.
134 CHOI et al BLOOD, 8 JULY 2010 � VOLUME 116, NUMBER 1 only.For personal use at WASHINGTON UNIV SCH MEDICINE on October 12, 2010. www.bloodjournal.orgFrom
day �19 and FACS gated on donor Tconv (H-2Kb; CD45.1) wasperformed. Surprisingly, we observed that Tconv proliferatedrobustly in vivo after AzaC treatment (Figure 5). Moreover, all ofthe FOXP3� Tregs were derived from donor Tconv that haddivided the most in vivo (Figure 5). These results are consistentwith the critical role of cellular proliferation on the biologicactivities of both Dec and AzaC. Therefore, the modest reduction ofproliferation of donor Tconv in transplant recipients treated withAzaC could be explained by a direct effect on Tconv proliferationor more likely on the peripheral conversion of FOXP3� Tconv toFOXP3� Tregs. These in vivo generated “azacTs” could thensuppress the proliferation of Tconv, thus limiting clinical GVHDwhile maintaining a potent GVL effect (see next paragraph).
The effect of in vivo treatment of mice after allogeneic SCT withhypomethylating agents: impact on GVL
It has been shown that nTregs preferentially suppress GVHD whilemaintaining GVL. Thus, we examined whether AzaC injectionafter allogeneic SCT results in similar effects. Using a well-established GVL model involving luciferase-expressing murineA20 leukemia cells, we tested the clearance of leukemia cells after
allogeneic SCT (PBS vs AzaC treatment after transplantation invivo) using in vivo bioluminescent imaging (BLI)31 (Figure 6A).We tested 2 different doses of Tconv: 2 � 106 and 10 � 106. Whileboth doses preserved GVL activity after AzaC treatment,10 � 106 Tconv induced a more robust GVHD only in those micetreated with PBS (supplemental Figure 7). Thus, we used10 � 106 Tconv for all the following GVL experiments. WhileAzaC treatment had no effect on the growth of A20 leukemia cellsin vivo (A20�PBS-Tconv vs A20�AzaC�Tconv; Figure 6B-C), theAzaC group (A20�AzaC�Tconv) had a significantly lower leuke-mic burden and percentage of mice with leukemia than the controlgroup (A20�AzaC�Tconv) and comparable to the PBS group and asignificantly better survival rate than the PBS group (A20�PBS�Tconv),suggesting that AzaC treatment of mice after allogeneic SCTpreferentially suppresses GVHD while maintaining GVL (Figure6B-C, supplemental Figures 7-8).
The mechanisms of hypomethylating agent-mediated inductionof Tregs
Two appealing models of how Tregs suppress the proliferation ofTeffs are (1) a cell-contact (or close proximity) dependent
Figure 6. AzaC treatment of mice that underwent a transplantation with delayed allogeneic T cells mitigates GVHD while preserving GVL. (A) Schema of theexperiments. B6 mice (CD45.2) TCD BM (5 � 106 cells) were used as a stem cell source. To induce GVHD, 10 � 106 Tconv (B6, CD45.1) were given on day 11 after SCTfollowed by the treatment with AzaC or PBS (every other day; 4 doses) starting on day 15 after SCT. For examination of GVL effect, 1 � 104 A20-luc/egfp leukemic cells weregiven along with TCD BM. (B-C) Mice treated with AzaC show significantly higher survival rate (B) and lower leukemic burden (C). Y axis in top panels indicates photon flux(photons/sec) in log scale measured from the dorsal and the ventral view with a region of interest drawn over the entire body of each mouse. Actual images of 1 representativemouse from each group are shown in bottom panels (scale: photons/sec/cm2/sr). A pool of 3 independent experiments.
HYPOMETHYLATING AGENTS AND GVHD 135BLOOD, 8 JULY 2010 � VOLUME 116, NUMBER 1 only.For personal use at WASHINGTON UNIV SCH MEDICINE on October 12, 2010. www.bloodjournal.orgFrom
mechanism via perforin 1 (PRF1) and granzyme B (GZMB), IL-35,cell surface-bound TGF-� or Galectin-1, and CD25, and (2) acell-contact independent mechanism mediated by secreted immuno-suppressive cytokines, such as TGF-� and IL-10.35-44 To determinewhether or not dcTs function in a cell-contact dependent fashion,we performed proliferation assays using coculture (direct contact)and transwell plates (cell contact-independent). As shown insupplemental Figure 9, coculture of dcTs and Teffs was necessaryfor suppression, suggesting that dcTs function in a cell-contact orclose proximity dependent manner. Next, we determined whethersuppressor function of dcTs requires PRF1 and GZMB using dcTsgenerated from Prf1 and GzmB deficient (KO) mice.46,47 DcTs fromPrf1 KO and GzmB KO mice were highly suppressive, suggestingthat these proteins are not necessary for the suppressor function ofdcTs, although PRF1 might be partially required (Figure 7A).
Because Dec and AzaC treatment leads to global hypomethyla-tion, it raised the question of whether dcTs and azacTs function in aFOXP3-dependent manner. We observed that the dcTs from theFoxp3 KO mice5 (Foxp3�/Y) were as suppressive as dcTs from theFoxp3 wild-type (WT) littermate controls (Foxp3�/Y) and WT B6mice (Figure 7B). We also observed similar results with azacTsgenerated from Foxp3 KO mice (Figure 7C). These results suggest
that the suppressor function of dcTs and azacTs does not requireFOXP3 expression and that the treatment with Dec and AzaCinduces the expression of not only FOXP3, but also modulates theexpression of other candidate genes that are necessary for thesuppressor function of azacTs and dcTs. These genes may beindependent or downstream targets of FOXP3 and regulated, likeFoxp3, by DNA methylation. Therefore, we performed genome-wide RNA profiling analyses using non-amplified total RNAs in aneffort to identify candidate genes that are regulated by hypomethy-lating agents and potentially mediate the observed suppressivefunction of nTregs, dcTs, and azacTs. Using GA-Mantel analysis,33
we identified 49 probes/genes that were expressed in a similarfashion within the 3 Treg groups (naive nTregs, activated nTregs,and dcTs) but differently in the pbsT control group (Figure 7D andsupplemental Table 2). As we expected, Foxp3 was one of thesegenes. Among these 48 candidate genes, 46 were up-regulated and2 down-regulated in the Treg groups.
Effect of AzaC on Foxp3 KO mice
Because the suppressor function of these hypomethylating agent-induced Tregs is independent of FOXP3 and since AzaC injection
Figure 7. Mechanism of hypomethylating agent-induced generation of Tregs. (A) Perforin (Prf1) is partially required for the suppressor function of dcT. DcTs from both WTand granzyme B (GzmB)–deficient mice were equally suppressive while those from Prf1-deficient mice were significantly less suppressive. A pool of 2 independentexperiments. (B-C) The impact of dcTs (B) and azacTs (C) from WT, Foxp3 heterozygote (Foxp3�/Y) and Foxp3 deficient (Foxp3�/Y) mice on Teffs were compared inproliferation assays. They are equally suppressive regardless of their origins. Neg: negative control, CFSE-labeled Teffs alone; pos: CFSE-labeled Teffs with stimulators,anti-CD3/CD28 antibody coated beads or allogeneic APC; all others contain both CFSE-labeled Teffs and stimulators plus indicated cells. (D) GA-Mantel33 analysis was run100 times. Forty-nine probes appeared in at least 30 of 100 solutions (Mantel correlation � 0.97). Foxp3 is the most frequently selected probe. Heat map shows these49 probes that separate the 3 Treg groups from the pbsT group. nnTreg: naive nTregs (n � 2), anTreg: activated nTregs (n � 3), dcT (n � 3), pbsT (n � 3). (E) AzaC treatmentprolongs survival of Foxp3 deficient mice. A treatment of 2 mg/kg AzaC was given every other day (4 doses) starting at 16 to 26 days of age. A pool of 4 independentexperiments. (F) Shown are 41-day-old Foxp3 KO mice treated with PBS (left) and AzaC (right); 25% (n � 8) had WT-looking ears.
136 CHOI et al BLOOD, 8 JULY 2010 � VOLUME 116, NUMBER 1 only.For personal use at WASHINGTON UNIV SCH MEDICINE on October 12, 2010. www.bloodjournal.orgFrom
in vivo postallogeneic SCT mitigates GVHD by increasing Tregs,we hypothesized that injection of AzaC might suppress autoimmu-ity and increase survival of Foxp3 KO mice by convertingauto-reactive T cells into Foxp3-independent Tregs. We observedthat injections of AzaC shortly after birth (days 16-26) significantlyimproved survival of these mice compared with Foxp3 KO miceinjected with PBS control (Figure 7E). In addition, AzaC sup-pressed autoimmunity, based on the observation that the AzaCgroup had normal looking ears while the PBS group had small,scaly, and thick ears, which is one of the features of Foxp3 KOmice (Figure 7F).
Discussion
In this report, we describe a novel and simple approach to limitGVHD while maintaining GVL by generating Tregs from alloreac-tive donor T cells in vivo using AzaC. This approach may help toovercome the current obstacles for the routine use of nTregs in thehuman clinical trials including limiting numbers of nTregs inperipheral blood, loss of suppressor properties after in vitroexpansion, and lack of cell surface markers necessary for efficientaffinity purification.5,7-9,20-22
The Foxp3-independence of dcTs and azacTs suggests thatFoxp3 and the genes responsible for the suppressor function ofTregs are probably regulated by the same mechanism, that is, DNAmethylation. Thus, although Dec and AzaC can directly induceFOXP3 expression, they can also modulate the expression ofFOXP3 target genes required for the suppressor function of Tregs.This suggests that these genes as well as Foxp3 remain hypomethy-lated in Tregs, but methylated in non-Tregs. A recent reportdemonstrates that Foxp3 is not required for the development ofnTregs, but for the maintenance of nTreg suppressor properties bymaintaining the expression of Foxp3 target genes.48 This observa-tion suggests that Foxp3 might be involved in maintaining hypom-ethylation status of these Foxp3 target genes or is essential for theincreased or continued expression of these candidate genes. Thesedata may suggest why AzaC treatment of Foxp3 KO mice did notpermanently rescue them in the absence of continued expression ofFoxp3. In an elegant system in which the Foxp3 gene has beenreplaced with GFP, the same group showed that GFP� (FOXP3�)cells contain gene expression profiles similar to nTregs.48 Togetherwith our data this observation strongly suggests that there is acommon regulatory mechanism that controls the DNA methylationstatus of both Foxp3 and its target genes.
In an effort to identify these target genes, we found 48 candidategenes that are differentially expressed in the Treg groups compar-ing to the pbsT group. Since their functions in Tregs are currentlyunknown (with the exception of Foxp3), it will be of interest todetermine whether one or more of these genes are responsible forthe suppressor function of Tregs and if their expression can bemodulated by hypomethylating agents in CD4� cells from both WTand Foxp3 KO mice. Finally, it would be of interest to determinewhether any of these gene products are expressed on the cellsurface and could potentially be used for purification of nTregs, exvivo–expanded nTregs, or hypomethylating agent–induced Tregs.
Although the suppressor function of dcTs and azacTs areFoxp3-independent, FOXP3 expression is currently the only surro-gate marker of the hypomethylation status of those genes critical inthe generation of Tregs. As shown in Figure 2B, only GFP� azacTswere suppressive. It is possible that GFP� azacTs might benonproliferating cells limiting the ability of AzaC to reduce the
promoter methylation status of critical Treg-dependent genes inthese cells since AzaC can incorporate only into replicatingDNA.26-29 In addition to the direct hypomethylation of target genesby Dec and AzaC, it is also possible that Dec and AzaC might altermiRNA expression, thus altering expression of Foxp3 target genesand their suppressor functions.49 In addition, Dec and AzaC mighthave other unknown functions. It has been recently suggested thatAzaC may have indirect effects on gene expression via DNAdamage/stress response signaling pathways.50,51
It is possible that AzaC inhibits activation and proliferation ofalloreactive donor T cells, thereby decreasing GVHD as proposedby Perez-Simon and colleagues based on their in vitro experi-ments.45 According to our CFSE-based in vivo experiments (Figure5) and GVL experiments (Figure 6), however, it is not likely thatAzaC simply eliminates or blocks the proliferation of allogeneicTconv. Rather, it is more likely that AzaC treatment suppressesGVHD while maintaining a potent GVL effect by peripheralconversion of donor T cells into FOXP3� Tregs. Nonetheless, wedo not exclude the possibility that AzaC indirectly attenuates theproliferation of Tconv by decreasing proinflammatory cytokines.Indeed, we observed reduced levels of serum IFN-�, IL-6, andIL-10 in the AzaC group (supplemental Figure 10; high serumIL-10 levels and high dose of IL-10 administration are associatedwith increased risks of GVHD in humans52,53 and mice,54 respec-tively55). It is not clear at present whether AzaC mitigates GVHDvia direct effects on Tconv or via reducing the production ofinflammatory cytokines, or both. Therefore, future studies willfocus on studies determining the molecular mechanisms by whichAzaC reduces reduction of GVHD while maintaining a potent GVLresponse.
While Noelle and colleagues suggest that the suppressorfunction of mouse nTregs is independent of Prf1,41 our data areconsistent with previous reports on both human and mousenTregs.42,56,57 Our data suggest that Prf1 may play at least a partialrole in suppressive function of both nTregs and dcTs. Ourexperiments further demonstrate that dcTs, like nTregs, requiredirect contact or close proximity for their suppressive effects. Inaddition, although GzmB appears to be necessary for a fullsuppressor function of mouse nTregs when they regulate NK cellsand B cells,56,57 our data demonstrating that the suppressor functionof dcTs is independent of GzmB are consistent with a recent reporton mouse nTregs in the allogeneic transplantation setting.58 Finally,we have demonstrated that although dcTs appear to be equallysuppressive to nTregs in vitro, they are significantly less suppres-sive than nTregs when generated in vitro and infused in vivo(Figure 3B). Similar results are seen when purified nTregs havebeen activated/expanded in vitro and tested in vivo.13,16 Thereduction of in vivo suppressive activity of dcTs might be due toreduced survival in vivo after strong ex vivo activation/expansionwith low concentrations of IL-2 (10 U/mL). Alternatively, exvivo–activated dcTs (like ex vivo–expanded nTregs) may havesignificant alterations in trafficking in vivo, thus limiting optimalinteractions and direct contact with Teffs at sites of Teff expansionin vivo.
In conclusion, AzaC treatment of allo-transplanted mice miti-gates deleterious GVHD while preserving beneficial GVL. Thebeneficial effect of AzaC on GVHD in these preclinical studies hasonly been tested when given within days after the infusion ofalloreactive donor T cells. It is conceivable that delaying theinfusions of AzaC may have an impact on both GVHD and GVL.Thus the effect of altering the timing of AzaC treatment will need tobe explored in the future. Our studies have also demonstrated that
HYPOMETHYLATING AGENTS AND GVHD 137BLOOD, 8 JULY 2010 � VOLUME 116, NUMBER 1 only.For personal use at WASHINGTON UNIV SCH MEDICINE on October 12, 2010. www.bloodjournal.orgFrom
AzaC treatment in vivo also suppresses autoimmunity in Foxp3KO mice and prolongs their survival. Thus, the administration ofhypomethylating agents such as AzaC after transplantation inhumans undergoing allogeneic SCT may provide a simple andrelatively nontoxic approach to limiting GVHD while preservingthe GVL and engraftment potential of donor T cells. These studiesalso provide insights into alternative pathways and candidate genes(in addition to Foxp3), which both may mediate the suppressivefunction of regulatory T cells and, like Foxp3, are regulated byDNA methylation.
Acknowledgments
We thank Tim Ley and Sheng Cai for providing Foxp3-GFP KI,Prf1 KO, and GzmB KO mice and Chyi Hsieh for Foxp3 KO mice.We thank Le’erin Voss, Sandeep Sodhi, and Mark Needles forassisting our research as summer students, and Mark Schroeder andLinda Eissenberg for critical reading the manuscript. J.F.D. issupported by the National Cancer Institute (R01 CA83845; and
R21 grants CA110489, CA132269, CA141523 P01 CA101937,and P50 CA94056). J.C. is supported by the Bryan ThomasCampbell Foundation. D.R.P.-W. is supported by P50 CA94056.
Authorship
Contribution: J.C. and J.F.D. designed and analyzed the experi-ments and wrote the paper; J.C. and J.R. performed the animalstudy; J.C., J.L.P., and D.R.P.-W. carried out BLI; J.C. and M.H.conducted real-time RT-PCR; W.D.S. and E.D. analyzed RNAprofiles; and all authors discussed the results and commented on themanuscript.
Conflict-of-interest disclosure: The authors declare no compet-ing financial interests.
Correspondence: John F. DiPersio, MD, PhD, Division ofOncology, Washington University School of Medicine, 660 SEuclid Ave, Campus Box 8007, St Louis, MO 63110; e-mail:[email protected].
References
1. Kernan NA, Collins NH, Juliano L, Cartagena T,Dupont B, O’Reilly RJ. Clonable T lymphocytes inT cell-depleted bone marrow transplants correlatewith development of graft-v-host disease. Blood.1986;68(3):770-773.
3. Butturini A, Bortin MM, Gale RP. Graft-versus-leukemia following bone marrow transplantation.Bone Marrow Transplant. 1987;2(3):233-242.
4. Butturini A, Gale RP. The role of T-cells in pre-venting relapse in chronic myelogenous leuke-mia. Bone Marrow Transplant. 1987;2(4):351-354.
5. Fontenot JD, Gavin MA, Rudensky AY. Foxp3programs the development and function ofCD4�CD25� regulatory T cells. Nat Immunol.2003;4(4):330-336.
6. Fontenot JD, Rasmussen JP, Williams LM,Dooley JL, Farr AG, Rudensky AY. RegulatoryT cell lineage specification by the forkhead tran-scription factor foxp3. Immunity. 2005;22(3):329-341.
7. Hori S, Nomura T, Sakaguchi S. Control of regu-latory T cell development by the transcription fac-tor Foxp3. Science. 2003;299(5609):1057-1061.
8. Khattri R, Cox T, Yasayko SA, Ramsdell F. An es-sential role for Scurfin in CD4�CD25� T regula-tory cells. Nat Immunol. 2003;4(4):337-342.
9. Edinger M, Hoffmann P, Ermann J, et al.CD4�CD25� regulatory T cells preserve graft-versus-tumor activity while inhibiting graft-versus-host disease after bone marrow transplantation.Nat Med. 2003;9(9):1144-1150.
10. Lee YK, Mukasa R, Hatton RD, Weaver CT. De-velopmental plasticity of Th17 and Treg cells.Curr Opin Immunol. 2009;21(3):274-280.
11. Roncarolo MG, Gregori S, Battaglia M, BacchettaR, Fleischhauer K, Levings MK. Interleukin-10-secreting type 1 regulatory T cells in rodents andhumans. Immunol Rev. 2006;212:28-50.
12. Yamazaki S, Patel M, Harper A, et al. Effectiveexpansion of alloantigen-specific Foxp3�CD25� CD4� regulatory T cells by dendritic cellsduring the mixed leukocyte reaction. Proc NatlAcad Sci U S A. 2006;103(8):2758-2763.
13. Allan SE, Broady R, Gregori S, et al. CD4�T-regulatory cells: toward therapy for human dis-eases. Immunol Rev. 2008;223:391-421.
14. De Rosa V, Procaccini C, Cali G, et al. A key role
of leptin in the control of regulatory T cell prolif-eration. Immunity. 2007;26(2):241-255.
15. Karakhanova S, Munder M, Schneider M,Bonyhadi M, Ho AD, Goerner M. Highly efficientexpansion of human CD4�CD25� regulatoryT cells for cellular immunotherapy in patients withgraft-versus-host disease. J Immunother. 2006;29(3):336-349.
16. Oliveira V, Sawitzki B, Chapman S, et al. Anti-CD4-mediated selection of Treg in vitro - in vitrosuppression does not predict in vivo capacity toprevent graft rejection. Eur J Immunol. 2008;38(6):1677-1688.
17. Koch MA, Tucker-Heard G, Perdue NR, KillebrewJR, Urdahl KB, Campbell DJ. The transcriptionfactor T-bet controls regulatory T cell homeosta-sis and function during type 1 inflammation. NatImmunol. 2009;10(6):595-602.
18. Zheng Y, Chaudhry A, Kas A, et al. RegulatoryT-cell suppressor program co-opts transcriptionfactor IRF4 to control T(H)2 responses. Nature.2009;458(7236):351-356.
19. Chaudhry A, Rudra D, Treuting P, et al. CD4�regulatory T cells control TH17 responses in aStat3-dependent manner. Science. 2009;326(5955):986-991.
20. Bennett CL, Brunkow ME, Ramsdell F, et al. Arare polyadenylation signal mutation of theFOXP3 gene (AAUAAA–AAUGAA) leads to theIPEX syndrome. Immunogenetics. 2001;53(6):435-439.
21. Bennett CL, Christie J, Ramsdell F, et al. The im-mune dysregulation, polyendocrinopathy, enter-opathy, X-linked syndrome (IPEX) is caused bymutations of FOXP3. Nat Genet. 2001;27(1):20-21.
22. Chae WJ, Henegariu O, Lee SK, Bothwell AL.The mutant leucine-zipper domain impairs bothdimerization and suppressive function of Foxp3 inT cells. Proc Natl Acad Sci U S A. 2006;103(25):9631-9636.
23. Floess S, Freyer J, Siewert C, et al. Epigeneticcontrol of the foxp3 locus in regulatory T cells.PLoS Biol. 2007;5(2):e38.
24. Baron U, Floess S, Wieczorek G, et al. DNA de-methylation in the human FOXP3 locus discrimi-nates regulatory T cells from activated FOXP3(�)conventional T cells. Eur J Immunol. 2007;37(9):2378-2389.
25. Kim HP, Leonard WJ. CREB/ATF-dependentT cell receptor-induced FoxP3 gene expression:
a role for DNA methylation. J Exp Med. 2007;204(7):1543-1551.
26. Stresemann C, Lyko F. Modes of action of theDNA methyltransferase inhibitors azacytidine anddecitabine. Int J Cancer. 2008;123(1):8-13.
27. Baylin SB. DNA methylation and gene silencing incancer. Nat Clin Pract Oncol. 2005;2 suppl 1:S4-11.
28. Jones PA, Taylor SM. Cellular differentiation, cyti-dine analogs and DNA methylation. Cell. 1980;20(1):85-93.
29. Robertson KD. DNA methylation, methyltrans-ferases, and cancer. Oncogene. 2001;20(24):3139-3155.
30. Robert MF, Morin S, Beaulieu N, et al. DNMT1 isrequired to maintain CpG methylation and aber-rant gene silencing in human cancer cells. NatGenet. 2003;33(1):61-65.
31. Rettig MP, Ritchey JK, Prior JL, Haug JS, Piwnica-Worms D, DiPersio JF. Kinetics of in vivo eliminationof suicide gene-expressing T cells affects engraft-ment, graft-versus-host disease, and graft-versus-leukemia after allogeneic bone marrow transplanta-tion. J Immunol. 2004;173(6):3620-3630.
32. Gross S, Piwnica-Worms D. Real-time imaging ofligand-induced IKK activation in intact cells and inliving mice. Nat Methods. 2005;2(8):607-614.
33. Deych E, Culverhouse R, Shannon WD. Microar-ray dimension reduction based on maximizingMantel correlation coefficients using a geneticalgorithm search strategy [plenary lecture]. Pro-ceedings of the International Federation of Clas-sification Societies. 2009; in press.
34. Haribhai D, Lin W, Relland LM, Truong N, WilliamsCB, Chatila TA. Regulatory T cells dynamically con-trol the primary immune response to foreign antigen.J Immunol. 2007;178(5):2961-2972.
35. Garín MI, Chu CC, Golshayan D, Cernuda-Morollon E, Wait R, Lechler RI. Galectin-1: a keyeffector of regulation mediated by CD4�CD25�T cells. Blood. 2007;109(5):2058-2065.
36. Takahashi T, Tagami T, Yamazaki S, et al. Immu-nologic self-tolerance maintained by CD25(�)CD4(�) regulatory T cells constitutively express-ing cytotoxic T lymphocyte-associated antigen 4.J Exp Med. 2000;192(2):303-310.
37. Nakamura K, Kitani A, Strober W. Cell contact-dependent immunosuppression by CD4(�)CD25(�) regulatory T cells is mediated by cellsurface-bound transforming growth factor beta.J Exp Med. 2001;194(5):629-644.
138 CHOI et al BLOOD, 8 JULY 2010 � VOLUME 116, NUMBER 1 only.For personal use at WASHINGTON UNIV SCH MEDICINE on October 12, 2010. www.bloodjournal.orgFrom
38. Asseman C, Mauze S, Leach MW, Coffman RL,Powrie F. An essential role for interleukin 10 inthe function of regulatory T cells that inhibit intes-tinal inflammation. J Exp Med. 1999;190(7):995-1004.
39. von Boehmer H. Mechanisms of suppression bysuppressor T cells. Nat Immunol. 2005;6(4):338-344.
40. Pandiyan P, Zheng L, Ishihara S, Reed J,Lenardo MJ. CD4�CD25�Foxp3� regulatoryT cells induce cytokine deprivation-mediated apo-ptosis of effector CD4� T cells. Nat Immunol.2007;8(12):1353-1362.
41. Gondek DC, Lu LF, Quezada SA, Sakaguchi S,Noelle RJ. Cutting edge: contact-mediated sup-pression by CD4�CD25� regulatory cells in-volves a granzyme B-dependent, perforin-independent mechanism. J Immunol. 2005;174(4):1783-1786.
42. Grossman WJ, Verbsky JW, Barchet W, ColonnaM, Atkinson JP, Ley TJ. Human T regulatory cellscan use the perforin pathway to cause autolo-gous target cell death. Immunity. 2004;21(4):589-601.
43. Vignali DA, Collison LW, Workman CJ. How regu-latory T cells work. Nat Rev Immunol. 2008;8(7):523-532.
44. Collison LW, Workman CJ, Kuo TT, et al. The in-hibitory cytokine IL-35 contributes to regulatoryT-cell function. Nature. 2007;450(7169):566-569.
45. Sanchez-Abarca LI, Gutierrez-Cosio S,Santamaria C, et al. Immunomodulatory effect of5-azacytidine (5-azaC): potential role in the trans-plantation setting. Blood. 2010;115(1):107-121.
46. Kagi D, Ledermann B, Burki K, et al. Cytotoxicitymediated by T cells and natural killer cells isgreatly impaired in perforin-deficient mice. Na-ture. 1994;369(6475):31-37.
47. Heusel JW, Wesselschmidt RL, Shresta S,Russell JH, Ley TJ. Cytotoxic lymphocytes re-quire granzyme B for the rapid induction of DNAfragmentation and apoptosis in allogeneic targetcells. Cell. 1994;76(6):977-987.
48. Lin W, Haribhai D, Relland LM, et al. RegulatoryT cell development in the absence of functionalFoxp3. Nat Immunol. 2007;8(4):359-368.
49. Liston A, Lu LF, O’Carroll D, Tarakhovsky A,Rudensky AY. Dicer-dependent microRNA path-way safeguards regulatory T cell function. J ExpMed. 2008;205(9):1993-2004.
50. Mabaera R, Greene MR, Richardson CA, ConineSJ, Kozul CD, Lowrey CH. Neither DNA hypom-ethylation nor changes in the kinetics of erythroiddifferentiation explain 5-azacytidine’s ability toinduce human fetal hemoglobin. Blood. 2008;111(1):411-420.
51. Lavelle D, Saunthararajah Y, Desimone J. DNAmethylation and mechanism of action of5-azacytidine. Blood. 2008;111(4):2485; authorreply, Blood. 2008;111(4)2486.
52. Hempel L, Korholz D, Nussbaum P, Bonig H,Burdach S, Zintl F. High interleukin-10 serum lev-els are associated with fatal outcome in patientsafter bone marrow transplantation. Bone MarrowTransplant. 1997;20(5):365-368.
53. Takatsuka H, Takemoto Y, Okamoto T, et al. Pre-dicting the severity of graft-versus-host diseasefrom interleukin-10 levels after bone marrowtransplantation. Bone Marrow Transplant. 1999;24(9):1005-1007.
54. Blazar BR, Taylor PA, Smith S, Vallera DA.Interleukin-10 administration decreases survivalin murine recipients of major histocompatibilitycomplex disparate donor bone marrow grafts.Blood. 1995;85(3):842-851.
55. Welniak LA, Blazar BR, Murphy WJ. Immunobiol-ogy of allogeneic hematopoietic stem cell trans-plantation. Annu Rev Immunol. 2007;25:139-170.
56. Zhao DM, Thornton AM, DiPaolo RJ, ShevachEM. Activated CD4�CD25� T cells selectivelykill B lymphocytes. Blood. 2006;107(10):3925-3932.
57. Cao X, Cai SF, Fehniger TA, et al. Granzyme Band perforin are important for regulatory T cell-mediated suppression of tumor clearance. Immu-nity. 2007;27(4):635-646.
58. Cai SF, Cao X, Hassan A, Fehniger TA, Ley TJ.Granzyme B is not required for regulatory T cell-mediated suppression of graft-versus-host dis-ease. Blood. 2010;115(9):1669-1677.
HYPOMETHYLATING AGENTS AND GVHD 139BLOOD, 8 JULY 2010 � VOLUME 116, NUMBER 1 only.For personal use at WASHINGTON UNIV SCH MEDICINE on October 12, 2010. www.bloodjournal.orgFrom
W Vhshington University in St. LouisScHoor or MnucrNE
Department of Internal Medicine
October 13,2010
Re; Jaebok Choi PhD; E.D. Thomas Post Doctoral
Timothy|. Ley, M.D.Lewis T. and Rosalind B. Apple Chair in OncologyProfessor of Medicine and GeneticsDirector, Section of Stem Cell BiologyAssociate Director, Cancer Genomics, The Genome Center
Fellowship 2010
I am delighted to write this letter of support for Jaebok Choi's application for the E.D.Thomas Post Doctoral Fellowship. I have had the opportpnity to.work with Jaebok andwatched as he has developed into an outstanding young scientist in the Stem CellBiology Section of the Oncology Division.
Jaebok's application represents an innovative and highly relevant translational researchproject. lt explores a novel way of mitigating graft vs. host disease (GvHD) whilemaintaining engraftment and a potent graft vs. leukemia (GvL) effect using epigeneticapproaches. Jaebok first repeated some work initially published by Warren Leonardand colleagues, who first demonstrated that the FoxP3 locus is unmethylated in Tregsand highly methylated in "conventional" T cells. These studies suggested to Jaebok thatpharmacologic intervention with DNA methyltransferase inhibitors might reduce themethylation of the FoxP3 promoter and activate this gene. He has shown that FoxP3 isindeed transcriptionally activated by azacitidine and decitabine, and that these Tregsdemonstrate a potent suppressive phenotype in vitro and in vivo. His work has beenrecently published in Blood (1 16:129-139, 2010).
Since coming to the DiPersio lab, Jaebok has developed into an outstanding T cellimmunologist and transplantation biologist. His recent work will provide the foundationfor future clinical trials in humans (starting this fall under the supervision of JohnDiPersio); further, he has developed several new directions for studying Treg biology,and new insights into pathways and genes that mediate the suppressive function of bothnatural Tregs and pharmacologically induced Treg-like cells.
Because of his outstanding work as a post-doc, Jaebok was promoted to the rank ofInstructor in July 2009. The Section of Stem Cell Biology and the Division of Oncologyare clearly committed to Jaebok's career development. He works in a rich and diverseenvironment of scientists and translational investigators focusing on the biology of Tregfunction and transplantation biology, enhancing his productivity and potentialfor long-term successes.
ln conclusion, I strongly support both Jaebok Choi's application as well as his futuredevelopment as an independent investigator in our Division. I believe that his E.D.Thomas Post Doctoral Fellowship application is highly innovative and exciting, and likelyto provide a new approach for overcoming HLA barriers for allogeneic stem celltransplantation.
Sincerelv vours.n( t lt -11---
Timothy tef, U.o.
Washington University School of Medicine, Campus Box 8007, 660 South Euclid Avenue, St. Louis, Missouri 63 1 l0(314) 362-9337, Fax: (314) 362-9333, [email protected]
i1:a
510 South Kingshighway Boulevard, Campus Box 8225, St. Louis, Missouri 63110 314/362-9359, FAX: 314/362-0152
SCHOOL OF MEDICINE Molecular Imaging Center Division of Radiological Sciences October 11, 2010 Review Committee E.D. Thomas Post Doctoral Fellowship 2010 Re: Jaebok Choi, PhD Dear Committee Members: It is with great pleasure that I support the application of Dr. Jaebok Choi for an E.D. Thomas Post Doctoral Fellowship and his proposal, “Epigenetic Modulation of GvHD and GvL by In Vivo Administration of Azacitidine”. This proposal builds well on his prior work, published in Blood 2010. I have had the opportunity to work closely with Jaebok and watched as he has developed into a first rate scientist in the Section of Stem Cell Biology, Department of Medicine, Washington University School of Medicine. Jaebok has the skills and training to develop into an outstanding independent investigator in stem cell biology. I am Director of the Molecular Imaging Center, Mallinckrodt Institute of Radiology, Washington University School of Medicine, Professor of Radiology and Professor of Developmental Biology. My laboratory focuses on studies of gene expression and protein function in vivo using non-invasive imaging techniques. I have had the opportunity to work closely with Dr. Choi and feel that he has the skills and training to develop into an outstanding independent investigator. I would be delighted to serve as a collaborator on his proposed research project. His application represents an innovative and highly relevant translational research project. It combines solid basic science with innovative translational preclinical studies which define a novel way of mitigating graft vs. host disease (GvHD) while maintaining engraftment and a potent graft vs. leukemia (GvL) effect using epigenetics. The project was conceived and driven forward by Jaebok. He first repeated some work initially published by Warren Leonard demonstrating that the FoxP3 locus was unmethylated in Treg cells and highly methylated in “conventional” T cells. These studies suggested to Jaebok that pharmacologic intervention with FDA-approved hypomethylating agents might epigenetically reduce the methylation state of the FoxP3 promoter, thus transcriptionally activating this locus. He has shown this and demonstrated that not only is FoxP3 transcriptionally activated by these hypomethylating agents, but that azacitidine- and decitabine-induced Tregs demonstrate a potent suppressive phenotype in vitro and most importantly in vivo. In conclusion, I am delighted to strongly support both Dr. Choi’s application as well as his future development as an independent investigator. I believe that this E.D. Thomas Post Doctoral Fellowship application is innovative, highly relevant and likely to provide a novel way of overcoming barriers for future recipients of allogeneic stem cell transplantation.
David Piwnica-Worms, M.D., Ph.D. Professor of Radiology Professor of Developmental Biology Co-Director, BRIGHT Institute Director, Molecular Imaging Center Email: [email protected]
2
510 South Kingshighway Boulevard, Campus Box 8225, St. Louis, Missouri 63110 314/362-9359, FAX: 314/362-0152
Sincerely yours,
David Piwnica-Worms Professor
October 13, 2010 Jaebok Choi, PhD Research Instructor Internal Medicine/Division of Oncology Washington University School of Medicine 660 South Euclid Avenue, Campus Box 8007 St. Louis, MO 63110 Dear Jaebok: I am excited about participating on your “Epigenetic Modulation of GvHD and GvL by In Vivo Administration of Azacitidine” proposal and am happy to offer my services and support in any way needed. As Director of the Department of Medicine's Biostatistics Consulting Center, I have extensive experience in experimental design and data analysis of biomedical research in many different clinical and research areas. I am honored to be involved with this important project, and am happy to commit the resources and staff of the consulting core to help you with this project. I cannot state strongly enough my enthusiasm for this work. I wish you success in this endeavor. Sincerely,
William D. Shannon, Ph.D. Associate Professor of Biostatistics in Medicine Director of the Dept. of Medicine's Biostatistics Consulting Center