Histone HVI

Histone Deacetylase Inhibitors Impair the Elimination of
HIV-Infected Cells by Cytotoxic T-Lymphocytes

Richard Brad Jones1,2, Rachel O’Connor1, Stefanie Mueller1,2, Maria Foley2,3, Gregory L. Szeto2,3,
Dan Karel1, Mathias Lichterfeld4, Colin Kovacs5,6, Mario A. Ostrowski5,6,7, Alicja Trocha1,
Darrell J. Irvine1,2,3,8, Bruce D. Walker1,4,8*
1 The Ragon Institute of Massachusetts General Hospital, Massachusetts Institute of Technology, and Harvard University, Boston, Massachusetts, United States of America,
2 Koch Institute for Integrative Cancer Research, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States of America, 3 Department of Biological
Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States of America, 4 Massachusetts General Hospital, Boston, Massachusetts,
United States of America, 5 The Maple Leaf Medical Clinic, Toronto, Ontario, Canada, 6 Department of Medicine, University of Toronto, Toronto, Ontario, Canada, 7 Li Ka
Shing Medical Institute, St. Michael’s Hospital, Toronto, Ontario, Canada, 8 Howard Hughes Medical Institute, Chevy Chase, Maryland, United States of America
Abstract
Resting memory CD4+ T-cells harboring latent HIV proviruses represent a critical barrier to viral eradication. Histone
deacetylase inhibitors (HDACis), such as suberanilohydroxamic acid (SAHA), romidepsin, and panobinostat have been
shown to induce HIV expression in these resting cells. Recently, it has been demonstrated that the low levels of viral gene
expression induced by a candidate HDACi may be insufficient to cause the death of infected cells by viral cytopathic effects,
necessitating their elimination by immune effectors, such as cytotoxic T-lymphocytes (CTL). Here, we study the impact of
three HDACis in clinical development on T-cell effector functions. We report two modes of HDACi-induced functional
impairment: i) the rapid suppression of cytokine production from viable T-cells induced by all three HDACis ii) the selective
death of activated T-cells occurring at later time-points following transient exposures to romidepsin or, to a lesser extent,
panobinostat. As a net result of these factors, HDACis impaired CTL-mediated IFN-c production, as well as the elimination of
HIV-infected or peptide-pulsed target cells, both in liquid culture and in collagen matrices. Romidepsin exerted greater
inhibition of antiviral function than SAHA or panobinostat over the dose ranges tested. These data suggest that treatment
with HDACis to mobilize the latent reservoir could have unintended negative impacts on the effector functions of CTL. This
could influence the effectiveness of HDACi-based eradication strategies, by impairing elimination of infected cells, and is a
critical consideration for trials where therapeutic interruptions are being contemplated, given the importance of CTL in
containing rebound viremia.
Citation: Jones RB, O’Connor R, Mueller S, Foley M, Szeto GL, et al. (2014) Histone Deacetylase Inhibitors Impair the Elimination of HIV-Infected Cells by Cytotoxic
T-Lymphocytes. PLoS Pathog 10(8): e1004287. doi:10.1371/journal.ppat.1004287
Editor: Guido Silvestri, Emory University, United States of America
Received January 12, 2014; Accepted June 18, 2014; Published August 14, 2014
Copyright: ß 2014 Jones et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported in part by the Ragon Institute of MGH, MIT, and Harvard. RBJ is a Banting Fellow of the Canadian Institute of Health Research
and an Ontario HIV Treatment Network Junior Investigator. GLS is supported by a Ruth L. Kirschstein National Research Service Award from the NIH. BDW and DJI
are investigators of the Howard Hughes Medical Institute. The funders had no role in study design, data collection and analysis, decision to publish, or preparation
of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* Email: [email protected]
Introduction homeostatic proliferation, it is unlikely that current ART regimens

could cure an individual within a lifetime [7,8].
Antiretroviral therapy (ART) is capable of durably suppressing Such theoretical and experimental analyses have led to the
viremia in HIV-infected subjects, but is unable to cure infection. consensus that the eradication of HIV from an infected individual
The financial and psychological burden of lifelong therapy, as well will require a means for actively depleting the resting CD4+ T-cell
as a growing appreciation for co-morbidities that occur in HIV- reservoir, most likely to be achieved by inducing viral expression
infected individuals on long-term therapy, such as cardiovascular that could trigger immune-mediated clearance of infected cells.
disease and neurocognitive disorders, have led to the prioritization While a variety of compounds have been shown to reactivate virus
of HIV cure research [1,2]. The best understood, and perhaps from CD4+ T-cells, a class of drugs known as histone deacetylase
most obstinate, barrier to eradicating infection is the existence of a inhibitors (HDACis) has emerged as the front-runner and a
pool of infected resting memory CD4+ T-cells [3–5]. By virtue of number of these, including vorinostat (suberoylanilide hydroxamic
their quiescent state, these cells are not thought to express HIV acid or SAHA), romidepsin, and panobinostat, have entered into
antigens, rendering them invisible to the immune system. These HIV clinical trials aimed at testing their abilities to reduce or
cells are very long-lived, with an estimated half-life of 44 months, eradicate viral reservoirs in the context of ART [9–11] (reviewed
suggesting that 60 years of uninterrupted ART would be required in [12]).
for full decay of the reservoir [6]. As the reservoir almost certainly It was initially thought that the reactivation of latent HIV by
replenishes itself through ongoing rounds of re-infection and HDACis would be sufficient to eliminate infected cells through
PLOS Pathogens | www.plospathogens.org 1 August 2014 | Volume 10 | Issue 8 | e1004287

HDACis Impair CTL Killing
Author Summary elevated serum Th1 cytokines and accelerated progression in a

murine model of graft-versus-host disease [33]. Even within an
The advent of antiretroviral therapy has greatly improved individual, the effects of a given HDACi have been reported to be
the prognosis for HIV-infected individuals with access to divergent depending upon the particular facet of the immune
care. However, current therapies are unable to cure response being studied. For example, in a murine model of
infection, committing treated individuals to a lifetime of allogeneic bone marrow transplantation it has been shown that
medication with significant economic burden. Further- SAHA decreased levels of serum cytokines and GVHD while
more, it has become clear that antiretroviral therapy does having no effect on donor T-cell proliferation or killing of host cells
not completely restore health, leaving treated HIV-infected [20]. In a second example, co-administration of the HDACi MS-
individuals at increased risk of cardiovascular disease, 275 with a viral-vectored vaccine served to suppress the immune
neurological disorders, and other health issues. Thus, there response to the vector, while enhancing the response to the viral
is a need to develop therapies capable of curing HIV
vector insert and suppressing autoimmune pathology [34].
infection. It is thought that, to be successful, curative
strategies will need to combine a means to flush the virus This diversity of effects of HDACis on immune cells is likely
out of the latently-infected cells in which it hides, with a rooted in two sources. First, there are 18 different HDAC enzymes
means to kill these unmasked targets. A front-running in humans, divided into four different families. Different HDACi
approach proposes to use a class of drugs called histone drugs interact with different subsets of these enzymes, depending
deacetylase inhibitors (HDACis) as flushing agents, with upon the dose being used [35]. Second, in addition to altering
cytotoxic T-lymphocytes (CTL, or killer T-cells) to purge histone acetylation status, and thus chromatin structure, HDACis
viral reservoirs. Here, we uncover an unexpected negative can interact with multiple transcription factors, including NF-kB,
interaction between these two agents, whereby HDACis AP-1, and others either by interfering with co-repressor HDAC
suppress the ability of CTL to kill HIV-infected cells. This enzymes that are recruited to transcription factor binding sites
interaction has the potential to limit the effectiveness of resulting in enhanced transcription, or by directly blocking
combining CTL with HDACis in flush and kill approaches to deacetylation of the transcription factor itself [36–40]. Foxp3, for
HIV eradication, and should be considered in the prioriti- example, requires acetylation of lysine residues for maximal
zation and optimization of potential curative strategies. activation. HDACis, by preventing deacetylation of Foxp3,
enhance its activity and thus boost the numbers and function of
viral cytopathic effects. Recent data, showing that in vitro Tregs in vivo [41]. Other nonhistone proteins that serve as direct
treatment of patient PBMCs with 500 nM SAHA failed to lead HDAC substrates include p53, GATA-1, STAT3, and STAT5
to a reduction in inducible viral reservoirs, suggests that this may [42,43]. This high degree of complexity, both in terms of
not be the case, and that immune effectors, such as HIV-specific immunological outcomes and underlying mechanisms, necessitates
cytotoxic T-lymphocytes (CTL), natural killer (NK) cells, or that HDACis be studied in a context that is matched to their
immunotoxins will likely be needed to recognize and eliminate intended utility. Thus, there is a need to understand the impact of
these exposed target cells in so called ‘flush-and-kill’ strategies [13]. HDACis being taken forwards in flush-and-kill eradication
Notably, in this same study, CD8+ T-cells freshly isolated from strategies on the abilities of HIV-specific CTL to eliminate
ART-treated HIV-infected patients could eliminate infected cells infected target cells.
in a primary cell model of latency only if pre-stimulated with Here, we report a series of experiments designed to assess the
peptides and IL-2 ex vivo [13], highlighting that, in the case of effects of the HDACis currently being tested in HIV eradication
CTL-based flush-and-kill strategies, the functional state of virus- clinical trials on the in vitro function of virus-specific T-cells. We
specific CD8+ T-cells will be important for reservoir elimination. show that, while the HDACis tested did not exhibit detectable
In evaluating strategies predicated upon coordinating CTL with toxicity to ex vivo bulk CD8+ T-cells over the doses and time-
latency-reversing drugs it is critical to consider potential side effects courses tested, romidepsin and, to a lesser extent, panobinostat
of these drugs on CTL function. This is particularly true in the exhibited delay toxicity to activated CD8+ T-cells and to CTL
case of HDACis, which are known to exert potent and diverse clones. Each of the HDACis tested rapidly suppressed the
effects on both the innate and adaptive immune system (reviewed production of IFN-c from PMA/ionomycin by ex vivo stimulated
in [14] and [15]). A number of HDACis, including SAHA, have CD8+ and CD4+ T-cells at an early time-point not associated with
been shown to suppress the production of inflammatory cytokines losses in viability. The production of IFN-c in response to peptides
by both T-cells and innate immune cells, in vitro and in vivo [16– representing viral epitopes was rapidly and durably suppressed by
21]. SAHA, romidepsin, and other HDACis have also been shown treatment of both CTL clones and ex vivo CD8+ T-cells with
to interfere with the differentiation of monocytes into dendritic HDACis. Treatment with romidepsin abrogated the proliferation
cells (DCs), as well as to block the ability of DCs to upregulate of HIV-Gag- and CMV-pp65-specific CD8+ and CD4+ T-cells,
CD1a, CD80, CD83 and other co-stimulatory molecules, resulting while panobinostat and SAHA only significantly impaired
in impaired priming of T-cells [22–26]. These immunosuppressive proliferation of CMV-pp65-specific CD8+ T-cell responses. Of
activities of HDACis have been associated with therapeutic principal importance, we observed that each of the HDACis tested
benefits in murine models of graft-versus-host disease (GVHD) exhibited significant impairment of the abilities of CTL clones to
and autoimmune/inflammatory disorders such as autoimmune eliminate HIV-infected target cells. These results indicate that
lymphoproliferative syndrome, experimental autoimmune enceph- HDACis can exert negative effects on the CTL that may be
alomyelitis (EAE, a model of multiple sclerosis), and diabetes needed for flush-and-kill approaches to eradication.
mellitus [27–32].
While it is tempting, based on this body of evidence, to generally Results
characterize HDACis as immunosuppressive agents, this would
not be an accurate assessment. The HDACi panobinostat Effects of HDAC Inhibitors on T-Cell Viability over a
(LBH589), which is also in clinical trials for HIV eradication 21 Hour Time-Course
(CLEAR trial, ClinicalTrials.gov Identifier NCT00256139), has As a prerequisite to assessing the effects of HDACis on T-cell
been reported to enhance T-cell activation in vivo, resulting in function, we first measured the impact of HDACi treatment on

T-cell viability. Freshly isolated PBMC from an HIV-uninfected strikingly higher levels of cell death in stimulated CD8+ and CD4+
donor were treated separately with serial dilutions of romidepsin, T-cells (Fig. 2B, right panel). As with CTL clones, romidepsin
panobinostat or SAHA at pharmacologically relevant concentra- exhibited substantially greater toxicity to activated cells than
tions (see Methods for justification of concentration ranges panobinostat, particularly at the lower concentrations. In parallel
selected). At 4 and 21 hour time-points cells were stained with to the viability experiments, to confirm the activity of our drugs,
7-Aminoactinomycin D (7-AAD, stains DNA of dead cells) and we tested these same stocks of romidepsin and panobinostat for
fluorochrome-conjugated annexin-V (stains phosphotidyl serine, their ability to reverse HIV latency in the ACH2 cell line model
an early apoptosis marker) and analyzed by flow cytometry. At [44], and observed very similar dose response curves (Fig. 2C).
4 hours of treatment we did not observe significant loss of cell Thus, romidepsin and panobinostat are disproportionately toxic to
viability under any of the treatment conditions tested (Fig. 1A–C). activated T-cells, including those that have received short-term
At 21 hours of treatment we continued to observe a lack of impact (48 hour) TCR stimulation. Romidepsin was substantially more
of HDACi treatment on CD8+ T-cell viability, but did observe a toxic to CTL and activated T-cells than panobinostat, despite
significant induction of CD4+ T-cell death at doses of 50 and similar potencies of latency-reversal by these two drugs as
100 nM, reaching 21.460.9% dead cells with 100 nM romidep- measured against a cell line model.
sin, (p = 0.0038 compared to no treatment)) (Fig. 1A–C). Neither
SAHA nor panobinostat was associated with loss of T-cell viability HDAC Inhibitors Impair IFN-c Production from T-Cells
at 21 hours.
before Losses in Viability Occur
The effects of HDACis on cell viability were also assessed using
Next, we measured the effects of exposure to HDACis on IFN-c
pulse-wash experimental setup with cryopreserved PBMC from
production from ex vivo T-cells. PBMC from an ARV-treated
two HIV-infected ARV-treated subjects. Cells were treated with
HIV-infected subject were exposed to the indicated concentrations
drugs for 6 hours, washed, plated in fresh medium, and then
of HDACis for 4 hours, and then stimulated with PMA/
viability was assessed at 16, 21, and 28 hours by 7-AAD staining
ionomycin for 5 hours. We observed that romidepsin, panobino-
and flow cytometry. Beginning at 21 hours, significant losses in
stat, and SAHA each significantly suppressed the production of
viability were only observed in CD4+ T-cells upon treatment with
IFN-c by both CD8+ and CD4+ T-cells at these pharmacologically
higher concentrations of romidepsin or panobinostat, and these
relevant concentrations (Fig. 3A), without any significant losses in
increased by 28 hours. A significant loss in CD8+ T-cell viability
cell viability (Fig. 3B). Thus, while in previous experiments
was only observed with 100 nM romidepsin in cells from one of
treatment with romidepsin and, to a lesser extent panobinostat,
the two subjects at the 28 hour time-point (Supporting Fig. S1).
resulted in the death of activated T-cells at relatively late time-
No significant losses in viability were observed in association with
points post-exposure, treatment with all three HDACis was also
treatment with up to 1,000 nM of SAHA in this experiment (data
found to suppress IFN-c production at early time-points, before
not shown). Thus, while relatively high doses of romidepsin and
any losses in viability were detectable.
panobinostat negatively impacted the viability of CD4+ T-cells, no
effects on bulk CD8+ T-cell viability were observed out to
21 hours. HDAC Inhibitors Exert Rapid and Durable Suppression of
IFN-c Production from HIV-Specific Antigen-Stimulated
Romidepsin and Panobinostat Are Disproportionately CD8+ T-Cells
Toxic to Activated T-Cells, Including CTL Clones We next measured the effect of exposure to HDACis on
We also tested the effects of HDACis on CTL clones. Following cytokine production from virus-specific CD8+ T-cells using IFN-c
4 hours of exposure we did not observe losses in cell viability at the ELISPOT assays. CMV-pp65- and HIV-Nef-RW8-specific CD8+
doses tested (Fig. 1D). However, 21 hours of sustained exposure T-cell clones isolated from two ARV-treated HIV-infected subjects
led to significant losses in CTL viability with as little as 25 nM were treated with the indicated pharmacologically relevant
romidepsin, but not with panobinostat or SAHA (Fig. 1D). To concentrations of romidepsin, panobinostat, or SAHA. As a point
further explore this, we tested the effects of a 4 hour pulse/wash of contrast we also included conditions where cells were pre-
exposure to romidepsin, panobinostat, and SAHA on the viability treated with ALT-803, an IL-15 superagonist expected to increase
of the same clone, and on an additional clone, specific for HIV- IFN-c ELISPOT responses (see Methods). Following 2 hours of
Env-VPVWKEATTTL. We observed high levels of CTL death treatment with the designated concentrations of HDACis or ALT-
with romidepsin treatment that were significant with as little as 803, CTL clones were cultured with autologous BLCL targets in
6 nM of drug (Fig. 2A). Substantially less cell death was observed ELISPOT plates, and stimulated for 12 hours with cognate
with equivalent concentrations of panobinostat, however signifi- peptides (without washing to remove HDACis or ALT-803). For
cant effects were still detected at 25 and 50 nM of drug for the both CTL clones, we observed that treatment with romidepsin,
Env-specific CTL clone, and at 50 nM for the Gag-specific CTL panobinostat, or SAHA resulted in reduced frequencies of T-cell
clone. No significant losses in viability were observed with up to responses, measured in spot-forming cells (SFCs) as compared to
1,000 nM of SAHA (data not shown). We posited that the untreated controls (Fig. 4A). Inhibition of cytokine production
differential level of toxicity of romidepsin and panobinostat was dose dependent, and particularly marked for romidepsin,
towards CTL, versus ex vivo T-cells, may be attributable to the where IFN-c production was essentially abolished. In contrast, the
greater activation state of the former. To test this, PBMC from an IL-15 superagonist ALT-803 enhanced frequencies of IFN-c
ARV-treated HIV-infected subject were thawed and stimulated producing cells.
with anti-CD3 of anti-CD28 antibodies for 48 hours. These cells We then tested the effects of these drugs on IFN-c production
were pulsed with the indicated doses of HDACis for 4 hours, drugs from CD8+ T-cells directly ex vivo from ARV-treated HIV-
were washed out, cells were cultured for an additional 17 hours, infected subjects. PBMC were treated with HDACis or ALT-803
and viability was assessed. Parallel treatments were performed on for 2 hours then peptides representing optimal CD8+ T cell
PBMC directly after thawing without stimulation. As in previous epitopes that been previously mapped in these subjects were added
experiments, we observed only low levels of cell death in HDACi- (without washing to remove HDACis). We observed that, as with
treated unstimulated T-cells (Fig. 2B, left panel). We observed CTL clones, all three of the HDACis tested suppressed


Figure 1. Effects of HDACis on the viability of PBMC CD8+ and CD4+ T-cells. A–C. PBMC were treated with HDACi drugs at the indicated
concentrations for either 4 or 21 hours (drugs left in) then stained with Annexin-V-Fitc (stains phosphatidyl serine on apoptotic cells), 7-AAD (stains
DNA of dead cells), CD4 pacific blue, and CD8 alexa-fluor 700 and analyzed by flow cytometry. A. Shown are representative flow cytometry data
indicating gating on dead (Annexin-V+7-AAD+) and dying (Annexin-vbright7-AADdim/2) cells. B, C. Shown are summary data for treatments PBMC
gated on CD8+ cells (left panel) or CD4+ cells (right panel) with romidepsin and panobinostat (B) or with SAHA (C) at the indicated doses for the
indicated times. P values for comparisons between the multiple doses of romidepsin tested and the untreated condition were calculated by Kruskal-
Wallis test and found to be significant for ex vivo CD4+ T-cells (p = 0.0080). Post-hoc Dunn’s multiple comparison tests were performed for each of
these and the indicated p values are adjusted for multiple testing. Other drug treatment conditions and cell types were not significant by Kruskal-
Wallis tests. D. The effects of HDACi treatment on the viability of an HIV-Gag-SLYNTVATL-specific T-cell clone were determined in the same manner as
for PBMC, including use of the same statistical tests. Error bars represent SD. * p,0.05, ** p,0.01, *** p,0.001, **** p,0.0001.
doi:10.1371/journal.ppat.1004287.g001
HIV-specific IFN-c production from ex vivo samples in a dose- ALT-803 for 4 hours prior to addition of specific peptides. Cells
dependent manner (Fig. 4B). In contrast, treatment with IL-15 were then co-cultured with HDACis and peptide for 48 hours,
superagonist resulted in a significant increase in IFN-c production. washed extensively, and resuspended in fresh medium with 20 U/
The above ELISPOT assays were based on 16 hour peptide ml IL-2 for an additional 5 days. We observed the proliferation of
stimulation periods, with the final readout representing the CD8+ T-cells in response to CMV-pp65 peptide pools (Fig. 5A),
cumulative IFN-c production over this period. Based on the data and a lesser degree of proliferation in response to HIV-Gag
presented in Figs. 2 and 3 we postulated that two factors may peptide pools (data not shown). Romidepsin, at the pharmacolog-
have contributed to this net loss in IFN-c production: i) the rapid ically relevant dose of 25 nM, abrogated proliferation of CD8+ T-
suppression of IFN-c production from viable cells at early time- cells in response to both peptide pools (Fig. 5A, B). CD8+ T-cells
points post-stimulation and ii) particularly in the case of treated with 25 nM panobinostat or 500 nM SAHA (also
romidepsin, the death of HDACi exposed stimulated cells at later pharmacologically relevant concentrations, see Methods) exhibited
time-points. The involvement of both of these factors would result significantly reduced proliferation in response to CMV-pp65 than
in the HDACi suppressive effect being both rapid in onset and untreated controls (panobinostat – p = 0.042, SAHA – p = 0.024,
durable, as HDACi wash-out periods would not reverse losses in Fig. 5C and D). Statistically significant effects of panobinostat or
IFN-c production caused by the death of antigen-specific cells. We SAHA on the proliferation of CD8+ T-cells in response to HIV-
tested this by probing the kinetics of this suppressive effect. To test Gag were not observed. However, our sensitivity to detect such an
the contributions of the former, we performed short-term effect may have been limited by the relatively low levels of baseline
treatments, with 2 hour exposures of ex vivo PBMC to either proliferation in response to Gag. As a point of contrast, the IL-
romidepsin or SAHA, followed by 6 hour stimulations with the 15SA ALT-803 significantly enhanced the proliferation of CD8+
HIV-Gag optimal epitope AW11. To test the latter, we exposed T-cells in response to both CMV-pp65 and HIV-Gag peptide
PBMCs with romidepsin or SAHA for 2 hours, washing to remove pools.
the drug, culturing for 24 hours in the absence of drug, and then In these experiments we also examined the effects of HDACis
stimulating with peptide for 12 hours in ELISPOT assays (24 hour on CD4+ T-cell proliferation (Fig. 5B, C, D). As with CD8+ T-
washout). We observed the dose-dependent impairment of IFN-c cells, we observed that romidepsin abrogated proliferation of
production in both the 6 hour stimulation and 24 hour wash-out CD4+ T-cells in response to either HIV-Gag or CMV-pp65
assays following treatment with romidepsin (Fig. 2C). With peptide pools. A consistent effect on CD4+ T-cell proliferation was
SAHA treatment, rapid impairment was observed in 6 hour not observed for either panobinostat or SAHA in response to
ELISPOT assays, but this was largely normalized following a either HIV-Gag or CMV-pp65.
24 hour washout period. Panobinostat was not tested in this assay. To further test the durability of the inhibition of T-cell
Thus each of the HDACis studied suppress IFN-c production proliferation by romidepsin we performed a variation on the
from HIV-specific CD8+ T-cells (both ex vivo and CTL clones). above experiment where, after washing out drugs and peptides at
This effect was rapid in onset for both romidepsin and SAHA, and 48 hours, additional peptides were added back along with the
in the case of romidepsin was sustained following removal of drug fresh medium. We observed significant inhibition of proliferation
(for at least 24 hours). While we were unable to directly measure of CD8+ T-cells (p = 0.005), and a trend towards reduced
the viability of the antigen-specific T-cells at the end of these proliferation of CD4+ T-cells (p = 0.067) in response to HIV-
ELISPOT assays, our data with anti-CD3 stimulated cells Gag peptides, despite 5 days of antigenic stimulation in the
(Fig. 2B) support the idea that the inability of these T-cells to absence of drug (Fig. 5F). As with the IFN-c ELISPOT results,
recover over a 24 hour period may reflect the selective depletion of the selective depletion of antigen-specific cells by stimulation in the
these cells by the combination of antigenic stimulation and presence of romidepsin may explain the lack of recovery following
romidepsin treatment. a romidepsin wash-out period. Thus, HDACis differentially affect
the proliferation of T-cells in response to viral peptides, with
HDAC Inhibitors Differentially Impair Proliferation of romidepsin abrogating proliferation of CD8+ and CD4+ T-cells in
Virus-Specific CD8+ T-Cells response to either CMV-pp65 or HIV-Gag peptides, while
Next we tested the effects of HDACis on the proliferation of panobinostat and SAHA only exhibited significant suppression
CD8+ T-cells in response to HIV-Gag and CMV-pp65 peptide of CMV-pp65-specific CD8+ T-cells.
pools using a standard CFSE proliferation assay. We reasoned that
in vivo there would be a delay between the time of HDACi HDAC Inhibitors Impair CTL Killing of HIV-Infected CD4+
administration and the expression of HIV antigens, during which Target Cells
antigen-specific T-cells would be exposed to drugs. For a single Of greatest relevance to flush and kill approaches to HIV
round of HDACi administration, cells that recognized induced eradication is whether HDACis impact the ability of CTL to kill
HIV antigens would have a chance to recover from any effects of HIV-infected target cells. We tested this using several variations of
HDACis upon drug clearance, and potentially proliferate. To an infected-cell elimination assay. HLA-A02+ primary CD4+ T-
model this scenario, PBMCs were exposed to HDACis or cells were infected with HIV JR-CSF and then co-cultured with

Figure 2. Romidepsin and panobinostat are disproportionately toxic to activated CD4+ and CD8+ T-cells. A. HIV-Env-VL11 and HIV-Gag
SL9 specific CD8+ CTL clones were isolated from subjects OM292 and OM9 respectively, and exposed to HDACis at the indicated concentrations for
4 hours. Cells were then washed to remove drugs and replated in fresh R10 medium supplemented with 50 U/ml IL-2 for an additional 17 hours. Cell
viability was measured by 7-AAD staining and flow cytometry. B. PBMC from subject OM292 either directly ex vivo (left panel) or following 48 hours
of stimulation with anti-CD3 and anti-CD28 (right panel) were exposed to HDACis at the indicated concentrations for 4 hours. Cells were then washed
to remove drugs and replated in fresh R10 medium. Cell viability was then measured by 7-AAD staining and flow cytometry. For A and B, shown are
graphs of mean values taken from triplicate wells 6 SEM following subtraction of background (% 7-AAD+ in no drug controls). P values were
calculated by two-way ANOVA with Dunnett’s multiple comparison test (comparing to the no drug control) * p,0.05, ** p,0.01, *** p,0.001, ****
p,0.0001. C. In parallel to the viability assays shown in A and B, ACH2 cells were treated with HDACis for 4 hours, then washed and replated in fresh
medium. Shown are mean 6 SEM values for % HIV-Gag+ cells as measured in triplicate by intracellular staining and flow cytometry.
HLA-A02-restricted HIV-specific CTL clone of defined epitope this to co-cultures of untreated CTLs and infected CD4+ cells in
specificity that had been pre-treated for 2 hours with HDACi the absence of HDACi. Representative control data using an
drugs at various effector:target ratios. Following a 16 hour co- untreated HIV-Gag-SLYNTVATL-specific CTL clone are pre-
culture in the continued presence of HDACis, we determined the sented in Fig. 6A,B. As shown in Fig. 6C, co-cultures treated
frequency of HIV-infected cells by flow cytometry and compared with pharmacologically relevant doses of panobinostat and

Figure 3. HDAC inhibitors impair IFN-c production from PMA/ionomycin stimulated CD4+ and CD8+ T-cells. A, B. PBMC from subject
OM292 were exposed to HDACis for 4 hours. Cells were then washed and cultured in fresh medium with PMA/ionomycin for an additional 5 hours in
the presence of brefeldin A. A. IFN-c production in CD4+ and CD8+ T-cells was measured by intracellular cytokine staining flow cytometry. Shown are
mean 6 SEM values. P values were calculated by two-way ANOVA with Dunnett’s multiple comparison test (comparing to the no drug control) * p,
0.05, ** p,0.01, *** p,0.001, **** p,0.0001. B. Viability in CD4+ and CD8+ T-cells was measured by Annexin-V and 7-AAD staining flow cytometry.
Shown are mean 6 SEM values.
romidepsin showed significantly reduced elimination of infected pronounced effect at low effector to target ratios. To control for
cells relative to untreated cultures. Treatment with SAHA was non antigen-specific elimination of infected cells, we co-cultured
associated with reduced elimination of infected cells at clone:target infected target cells with a CMV-pp65-specific CTL clone in
ratios of 1:100 and 1:35 (mean 6 SD of proportion killed, no parallel to the HIV-Gag-specific CTL clone over the range of
treatment - 41.461.9 vs SAHA - 12.9614.4 at 1:100, and no effector:target ratios tested. We observed a lack of elimination of
treatment – 63.664.7 vs SAHA 55.0611.0), but this was not infected target cells with this CMV-pp65-specific clone (data-
statistically significant after correction for multiple testing in this points are along the x-axis, Fig. 4E).
experiment (p = 0.25). To further corroborate our results, we performed an additional
In order to isolate the effects of HDACis on CTL from potential experiment with an HLA-A02-restricted HIV-specific CTL
effects on target cells, we performed variations of the above targeting a different Gag epitope (FLGKIWPSHK). We again
experiment where we pre-treated CTL with drugs at the above observed significant impairment of infected-cell elimination by
concentrations for 4 hours and then washed to remove the drugs CTL that had been pre-treated with romidepsin and SAHA
prior to co-culturing with targets. Using an HIV-Gag-SLYNT- (Fig. 6F). The effect of panobinostat was not significant in this
VATL-specific CTL clone we tested two different ranges of assay, though a lesser proportion of infected targets were killed at
clone:target ratios (Figs. 6D (high range), 6E (low range)). In both each of the clone:target ratios. Thus, even transient exposures to
of these experiments we observed significantly impaired elimina- pharmacologically relevant doses of HDACis impair the abilities of
tion of HIV-infected target cells by CTL that had been pre-treated CD8+ T cells to eliminate HIV-infected target cells. As with other
with SAHA, panobinostat, or romidepsin, with the most measures of T-cell function reported above, the impairment of

Figure 4. HDAC inhibitors impair IFN-c production from antigen-stimulated CD8+ T-cells. A, B. PBMC or CTL clones from chronically HIV-
infected ARV-treated subjects (OM265 and OM292) were cultured for 2 hours with the indicated drugs and then plated for IFN-c ELISPOT assays with
12 hour peptide stimulation periods. A. Treated CTL clones were stimulated with overlapping 15mer peptides spanning the CMV-pp65 protein (left
panel) or the MHC-I-A24 restricted HIV-Nef peptide ‘RW8’ (right panel). Each treatment condition was tested in duplicate. Shown are mean SFC/104
CTL clone cells for each condition with error bars representing SEM. B. Experiments analogous to those depicted in A were performed for 8 different
optimal HIV CD8+ T-cell epitopes (5 for OM292 and 1 for OM265). Shown are summary data for all 8 responses depicting the mean SFC/106 PBMC (of
triplicate wells) under different treatment conditions. For SAHA, panobinostat, and romidepsin conditions statistical significance for each drug was
evaluated using two-way ANOVA tests, and p values were calculated using Dunnett’s multiple comparison test to account for the use of three
different drug concentrations. For IL-15SA p values were calculated using the Wilcoxon matched pair test. C. PBMC from subjects OM265 and OM292
were treated with SAHA or romidepsin at the indicated concentrations for 2 hours and then either: stimulated with peptides for 6 a hour ELISPOT
assay, or washed, cultured for 24 hours in the absence of drugs, and stimulated with peptides for a 12 hours ELISPOT assay. P values were calculated
by two-way ANOVA with Dunnett’s multiple comparison test (comparing to the no drug control) * p,0.05, ** p,0.01, *** p,0.001, **** p,0.0001.
CTL elimination of infected cells was particularly severe following (Fig. 7A). At the high clone:target ratio of 1:10 used in this assay
treatment with romidepsin at the doses tested. we observed relatively modest impairment of viral elimination in
In an in vivo setting, one could envision a scenario where the 4 hour wash-out experiment by romidepsin and panobinostat
exposure to HDACis induces expression of latent HIV, but then (mean 6 SD % infected targets killed, 73.061.3% - no treatment
clearance of the drug occurs, potentially allowing CTL to recover versus 53.563.0%–50 ng/ml romidepsin, p = 0.0013;
functionality and eliminate these unmasked target cells. To model 60.266.1%–50 ng/ml panobinostat, p = 0.016). We did not
this, we performed an additional series of viral elimination assays observe significant impairment of viral elimination by SAHA in
where we pre-treated CTL with HDACis for 5 hours, then washed this experiment; 55.463.1 – 1,000 ng/ml SAHA, p = 0.07)
to remove these drugs, and cultured CTL in fresh medium for 4 or (Fig. 7B). In the 14 hour wash-out experiment we observed
14 hours prior to a 10 hour co-culture with infected target cells greater impairment of target cell killing in the romidepsin and


Figure 5. HDAC inhibitors impair proliferation of virus-specific T-cells. PBMC from 11 chronically HIV-infected subjects on suppressive ARV
therapy were labeled with CFSE and treated with the indicated drugs for 4 hours. HIV-Gag or CMV-pp65 peptide pools were then added to final
concentrations of 1 mg/ml/peptide. 48 hours later, cells were washed thoroughly, re-suspended in medium with 20 U/ml IL-2, and cultured for an
additional 5 days. Cells were then stained with a viability dye and antibodies to CD3, CD8, and analyzed by flow cytometry. A. Shown are
representative flow cytometry data of a CMV-pp65 response gated on viable, CD3+CD8+ lymphocytes depicting CFSE (diminution indicates
proliferation) – x-axis by CD8 – y-axis. B–E. Summary data depicting frequencies of CFSEdim cells within viable, CD3+CD8+ or CD3+CD82 (CD42 T-cell)
lymphocyte populations. F. Shown are data analogous to B–E, but from a separate experiment where peptides were added back with fresh medium
following HDACi wash-out at 48 hours. P values were calculated by the Wilcoxon matched-pairs signed rank test.
panobinostat treated samples, as compared to the 4 hour wash-out target cell marked with a white arrow. In total, three CTL:target
(mean 6 SD % infected targets killed, 64.860.4% no treatment cell engagements and kills are observed in this field of view within
versus 060%–50 ng/ml romidepsin, 27.262.3% 50 ng/ml pa- 1 hour 40 minutes. Killed target cells appear green due to the
nobinostat). In contrast, we observed a lack of inhibition in ability of sytox dye to pass through their compromised membranes
SAHA-treated 14 hour wash-out samples (62.860.3% 1,000 ng/ and bind to nuclear DNA. For the CTL treated with romidepsin
ml SAHA). The differences in elimination of infected cells between or SAHA we observed less efficient killing of target cells. The
the 4 hour and 14 hour wash-out experiments were significant images in the middle panel of Fig. 8A correspond to Supporting
across the dose ranges tested, by two-way ANOVA, for romidepsin Movie S2 and follow a SAHA-treated CTL (yellow arrow) that
(p = 0.0187) and for panobinostat (p = 0.0231). These differences migrates and engages with a target cell (white arrow). The CTL
for SAHA were not significant (p = 0.065), and, in contrast to fails to kill this target and instead is dragged through the collagen
romidepsin and panobinostat, the trend for SAHA was towards until the target cell pulls away and escapes out of frame. In
improved elimination of infected cells following the rest period. Supporting Movie S2 one can see that this CTL subsequently
Based on our data from Figs. 1 and 2, we postulated that the engages with a second target cell that it also fails to kill. Two other
exacerbation, rather than recovery, of CTL impairment over the CTL are observed in the two right panels and, in contrast to the
14 hour wash-out period was due to the progressive loss of viability no treatment condition, neither of these is associated with a killed
of cells following romidepsin or panobinostat treatment, even target cell. The lower panel corresponds to Supporting Movie
following HDACi removal. In order to test this, we had S3 and shows a similar event observed in the romidepsin condition
maintained CTL in culture in parallel to these elimination assays. where a CTL (yellow arrow) engages with a target cell (white
Contemporaneously with the termination of elimination assays, we arrow) but fails to kill it. The target cell pulls the CTL along and
assessed the viability of CTL in these cultures by assessing the then escapes Fig. 8A. In-line with previous experiments, we did
frequencies of apoptotic (Annexin-V+) and dead (7-AAD+) cells by also observe an increased frequency of dead/apoptotic CTL (non-
flow cytometry (Fig. 7C). We observed the dose dependent motile, small condensed cytoplasm) in association with romidepsin
induction of apoptosis (Annexin-V+) in CTL that had been treatment (Fig. 8B, middle panel, pink arrow). The total numbers
exposed to romidepsin and, to a lesser extent, panobinostat. Thus, of killed target cells following 20 minutes of imaging were counted
in-line with our previous results, rather than allowing for recovery in four fields of view (FOV) each for the three treatment
of CTL function, a wash-out period exacerbated the impairment conditions. Overall, we observed significantly fewer killing events
of infected-cell elimination by romidepsin and panobinostat. In in the wells containing CTL that had been treated with
contrast, exposure to SAHA at these doses was not associated with romidepsin (p = 0.011), but no significant effect of SAHA
the induction of apoptosis. treatment (p = 0.181) Fig. 8B. Thus, treatment with romidepsin
impairs the ability of CTL to kill peptide-pulsed target cells in a
3D model of the extracellular environment. The two modes of
Visualizing the Effects of HDAC Inhibitors on CTL Killing
romidepsin-induced impairment supported by the data presented
in 3D Collagen Matrices by Time-Lapse Microscopy in previous figures – functional impairment of viable cells, and
To effectively eliminate HIV-infected target cells in vivo, CTL delayed onset apoptosis – were directly visualized in this
need to migrate through the extracellular matrix to catch and kill experiment with both an apoptotic and a viable but ineffective
moving targets. We visualized the effects of pharmacologically CTL present in the romidepsin FOV presented in Fig. 8B.
relevant doses of romidepsin and SAHA on the killing activity of
an HIV-Gag-SLYNTVATL-specific CTL in a 3D collagen matrix
Discussion
designed to approximate this in vivo environment as has been
previously described [45]. CTL were membrane-labeled with In this study we explored the potential for HDACi latency-
Alexa-Fluor555-conjugated cholera toxin B and pre-treated with reversing drugs to impact upon multiple functions of virus-specific
the indicated drugs for 24 hours, or maintained as no-treatment CTL. We observed that romidepsin, panobinostat, and SAHA all
controls. These effectors were then mixed with peptide-pulsed rapidly suppressed IFN-c production from virus-specific CD8+ T-
BLCL expressing the restricting class I allele in a collagen matrix cells. The proliferation of both CD8+ and CD4+ T-cells in
and visualized by time-lapse brightfield and fluorescence micros- response to either CMV-pp65 or HIV-Gag peptide pools was
copy. CTL are identified in these images by fluorescence from the abrogated by treatment with romidepsin, while the effects of
Alexa-Fluor555 label (red in images), and dead cells acquire SAHA and panobinostat on proliferation were only significant for
fluorescence by permitting a sytox green dye in the medium to the CD8+ T-cell response to CMV-pp65. Critically, each of these
pass through their disrupted membranes and bind to genomic HDACis also impaired the ability of HIV-specific CTL to
DNA (green in images). Untreated CTLs migrated through the eliminate infected CD4+ and peptide pulsed BLCL as measured
collagen matrix and efficiently killed target cells. Images from a in liquid culture and collagen matrices respectively.
representative field of view are shown in the top panels of Fig. 8A The overall data support the idea that there are mechanisms
and the corresponding movie is given in Supporting Movie S1. both involving and independent of losses in T-cell viability at play,
The yellow arrow follows a migrating CTL that engages and kills a and that these make differential contributions depending on the

Figure 6. HDAC inhibitors impair CTL killing of HIV-infected primary CD4+ cells. CD4+ T-cells were enriched from HIV-uninfected A02+
donors and either infected with HIV JR-CSF or maintained as mock infected controls. 24 hours post-infection, these target cells were co-cultured with
an HIV-Gag- or CMV-pp65-specific CTL clones at the indicated effector (clone):target (CD4+ cell) ratios for 16 hours. Cells were then stained with
fluorochrome conjugated antibodies to CD8, CD4, and HIV-Gag (intracellular staining) and analyzed by flow cytometry. A. Shown is the gating
strategy utilized for killing assays, with the placement of Gag+ and Gag+CD4dim gates determined based on a mock-infected control (right panel). B.

Data from a representative experiment with indicated effector:target ratios of an HIV-Gag-SLYNTVATL-specific T-cell clone. Panels C–F utilize total
Gag+ cells and calculate a % Infected Cells Killed as (%Gag+ No Effectors - %Gag+ at Given E:T ratio)/%Gag+ No Effectors*100, ex. in panel B %Infected
Cells Killed at 1:36 ratio = (22.126.7)/22.1*100 = 69.7%. C–F. Shown are the results from 4 independent experiments. In C CTL were treated with
romidepsin or panobinostat at 50 nM or with SAHA at 500 nM for 2 hours, and then co-cultured with target cells without a washing step. In D–F CTL
were treated with romidepsin or panobinostat at 25 nM or with SAHA at 500 nM for 6 hours, then washed thoroughly before initiating a 16 hour co-
culture with target cells. D and E both utilize an HIV-Gag-SLYNTVATL-specific CTL clone at high and low E:T ratios respectively. E utilizes an HIV-Gag-
FLGKIWPSHK-specific CTL clone. C–F. Shown are means 6 SEM of data from triplicate wells. P values were calculated by two-way ANOVA with
Tukey’s multiple comparison test.
functional assay being tested, drug concentrations, exposure/rest clearance of drugs in vivo may serve to mitigate some of the
times, and the type of effector cell (ex vivo T-cells, or in vitro effects of HDACis on HIV-specific CTL, although we attempted
expanded CTL clone). The data presented in Figs. 1 and 2 to account for this by washing out drug in most assays. On the
clearly demonstrate that romidepsin and, to a lesser extent, other hand, our in vitro experiments incorporated only single
panobinostat are disproportionately toxic to activated T-cells as doses of HDACis whereas the SAHA and panobinostat clinical
compared to their resting counterparts. However, data presented trials have incorporated repeated dosing. This could potentially
in Fig. 3 demonstrate that all 3 HDACis suppress IFN-c exacerbate any effects, particularly if these drugs cause lasting
production from viable T-cells. In the case of romidepsin, our changes to the T-cell functional profile, or compromise the
data support the idea that these two modes of suppression are survival of T-cells. In the CLEAR trial, for example, patients
linked, with early suppression of functionality followed by later received panobinostat on days 1, 3, and 5 every other week for 8
onset of apoptosis. However, in the case of SAHA, functional weeks. An additional challenge to predicting whether a therapeutic
suppression appears to occur in the absence of any impact on cell window between latency reversal and CTL inhibition exists is that
viability. Treatment with panobinostat appears to be intermediate infected cell elimination assays, which utilized highly functional
between these two situations, depending also upon the exposure CTL clones, at relatively high effector:target ratios in most
concentration. experiments, and activated target cells expressing high levels of
We propose that the net effects of HDACi treatment on HIV- HIV-Gag, can be interpreted as an idealized environment for
specific T-cells in the majority of our functional assays involved CTL killing of target cells. Thus, it is reasonable to speculate that
contributions from both of these modes of suppression, particularly even a subtle difference in killing efficiency observed in our in vitro
in the case of romidepsin. For example, the impaired abilities of assays may manifest as a critical difference in vivo where CTL are
romidepsin-treated CTL to eliminate HIV-infected target cells likely to be exhausted or otherwise functionally impaired, and
over 16 hour co-cultures likely resulted both from rapid suppres- where CTL encounters with reactivated target cells are rare.
sion of CTL function (CTL are capable of killing target cells within Following from this, while we did not observe significant
2–10 minutes [46–48]) as well as apoptosis at later time-points impairment of infected cell killing at 1 nM or 5 nM of
post-exposure. This is supported by the observed exacerbation of panobinostat, we cannot rule out that more subtle impairments
this impairment following a 14 hour romidepsin wash-out period in function may lead to reductions in abilities of CTL to eliminate
(Fig. 7). This is also directly visualized in Fig. 8A, and exposed natural reservoir cells in vivo. Based on the magnitudes of
Supporting Movie S3. Here, following romidepsin treatment, the effects observed in vitro, along with consideration of the
apoptotic CTL are visible, but a CTL also remains viable pharmacokinetic and pharmacodynamic properties of these drugs,
(excludes sytox dye) throughout the 5:34 (h:mm) imaging, while at the dosing regimens being taken forwards into HIV clinical
failing to kill a target cell that it engages from 1:03 to 2:26. These trials we propose that HDACis are differentially likely to impact
two modes of HDACi-induced CTL impairment – inhibition of upon relevant T-cell functions in vivo, with the following
function in viable T-cells and reduction in T-cell viability (with the hierarchy: romidepsin.panobinostat.SAHA. Ultimately, howev-
former leading to the latter in some cases) – have the potential to er, we hope that the primary outcome of our study will be to
have different impacts in a therapeutic setting. It may be possible motivate the incorporation of assays measuring ex vivo T-cell
to mitigate the impact of transient CTL impairment on flush-and- function into ongoing and planned HDACi clinical trials, and that
kill eradication strategies by designing dosing schedules that target immunosuppression will be considered as a potential factor
a temporal therapeutic window, whereby either HIV antigen limiting the effectiveness of any observed outcomes.
expression occurs before CTL are substantially impaired, or We must also highlight, in a more general sense, the potential
persists while CTL are given time to functionally recover. A better risk of treating HIV-infected subjects, whose immune systems do
understanding of the kinetics and durability of HIV antigen not fully recover even with ART, with HDACis. In addition to the
presentation from reactivated latently-infected cells is required to ex vivo and in vitro data presented in the current study, a number
evaluate the plausibility of this approach. On the other hand, of studies have observed in vivo immunosuppressive activities of
losses in viability of activated CTL, as we observed in vitro with HDACis, including romidepsin and SAHA [16–20]. While
romidepsin and, to a lesser extent, panobinostat, could result in an opportunistic infections have not been reported in HIV-infected
irreversible impairment of virus-specific cellular immune respons- subjects receiving SAHA and panobinostat, this is a greater
es, particularly if HIV-specific T-cell responses are primed first, for concern for romidepsin. We observed greater impairment of CTL
example by therapeutic vaccination. Thus, our data suggest that function, and toxicity with romidepsin treatment as compared to
the potential for a given HDACi to impair CTL function both in panobinostat and SAHA at the doses tested. Lymphopenia is also
the first few hours of treatment when viability is intact, and over known to be a common side-effect of romidepsin treatment, and
longer time-lines by the gradual induction of cell death should be serious and sometimes fatal infections, including pneumonia and
considered in designing flush-and-kill approaches. sepsis, have been reported in clinical trials in oncology settings
Our findings are limited to ex vivo and in vitro experiments, and [49]. Notably though, these side-effects were observed at higher
the extent to which HDACis impact CTL function in HIV- doses than those planned for HIV eradication trials. The
infected patients is presently unknown. On one hand, the incorporation of immunological end-points as important safety

Figure 7. Impairment of CTL killing is sustained for at least 14 hours after removal of romidepsin or panobinostat. An HIV-Gag-
specific CTL clone was treated with panobinostat, romidepsin, or SAHA at the indicated doses for 5 hours, or maintained as an untreated control.
Cells were then washed two times, resuspended in R10–50, and cultured for 4 or 14 hours. These CTL were then co-cultured with autologous HIV-
infected CD4+ T-cells at a clone:target ratio of 1:10 for 8 hours, and levels of infection were measured as described above (see Fig. 4 legend). A.
Shown are flow cytometry plots gated on the viable (FSC/SSC) CD82 population, and depicting HIV-Gag staining (x-axis) by CD4 staining (y-axis). For

the plots show, CTL that had been treated with romidepsin were given a 14 hour wash-out period prior to co-culture with infected cells. B. Shown are
summary flow cytometry for 4 hour wash-outs (left panels) and 14 hour wash-outs (right panels). Each condition was tested in duplicate. Error bars
represent SEM. C. Portions of the CTL used in the above elimination assays were maintained in culture. At the conclusions of the 14 hour wash-out
elimination assay, 29 hours in total since addition of drugs, the frequencies of dead (7-AAD+) and apoptotic (Annexin-V+) CTL were measured by flow
cytometry. Shown are % Annexin-V+ following treatment with the indicated doses of drugs. Error bars represent SEM. P values were calculated by
two-way ANOVA with Dunnett’s multiple comparison test (comparing to the no drug control) * p,0.05, ** p,0.01, *** p,0.001, **** p,0.0001.
parameters in the early stages of testing romidepsin in HIV- as detected by IFN-c ELISPOTS through a mechanism involving
infected subjects would help to address concerns regarding a STAT1, STAT3, STAT4, and STAT5 [52,53], and thus allowed
potential impact on the immune system in general. The possibility for a control for cell responsiveness to soluble factors that impact
that romidepsin or other HDACis could impair HIV-specific T- cellular functions. ALT-803 was obtained from Altor Bioscience
cell responses in vivo should also be ruled out before any therapy Corporation at 1.38 mg/ml in PBS, and was used at a working
interruptions are contemplated as part of eradication trials, given stock of 50 mg/ml. The concentrations of HDACis used in
the importance of CTL in containing any rebound viremia. experiments were selected to be pharmacologically relevant to
The HIV-specific cellular immune response, in addition to current clinical trials. The mean clinical Cmax of panobinostat at
playing a vital role in the ongoing health of HIV-infected the 20 mg p.o. dosing regimen of the ongoing CLEAR trial is
individuals, stands to make a powerful contribution to HIV 40 nM [54], the steady-state plasma concentration is 15 to 22 nM
eradication strategies. Thus we would make a case that a critical (Novartis Investigator’s brochure), and the EC50 values for in vitro
stage in evaluating potential latency reversing drugs to be used in HIV activation in cell line models of latency has been reported to
CTL-based flush-and-kill strategies should be to test the effects of range from 10–16 nM [11,55]. The mean Cmax of SAHA given at
these drugs on the functions of HIV-specific T-cells. This 400 mg p.o. as in the two latency reactivation trials that have been
information can both be used to help prioritize classes of conducted is 1 mM [56], and the reported EC50 values for in vitro
candidates, and to select optimal drug candidates from within a HIV activation in cell line and primary cell models of latency
class. We would also suggest that this should be extended to flush- range from 885–3,950 nM [11,55,57]. The mean clinical Cmax
and-kill strategies focused on other immune effectors. Notably, values for of romidepsin in the intermediate and high dosing arms
romidepsin has also been reported to both suppress the cytotoxic of the upcoming dose escalation study in HIV-infected subjects
activity and to drive apoptosis of NK cells, which are another (ClinicalTrials.gov Identifier: NCT01933594) are: 2 mg/m2, Cmax
potential key immune effector in flush and kill strategies [24]. 69.5632.6 nM, and (5 mg/m2, 178.1632.6 nM), and the report-
Information regarding how a drug influences CTL and other ed EC50 values for in vitro HIV activation in a cell line and
immune effectors should be generated along with other param- primary cell models of HIV range from 3–4.5 nM [55,57].
eters such as EC50 and toxicity in order to prioritize candidates.
Through screening efforts, or rational design based on an
T-Cell Viability Assay
improved understanding of the mechanisms by which HDACis
PBMC were obtained from HIV-negative buffy coats, or from
exert their immunosuppressive effects, it may also be possible to
cryopreserved leukophersis samples from HIV-infected subjects (as
identify HDACis that retain the ability to induce the expression of
indicated). CTL clones were taken from long-term in vitro cultures
latent HIV without impairing the ability of CTL or other immune
effectors to eradicate these unmasked targets. (restimulated bi-weekly). Cells were plated at 200,000 cells/well
(PBMC) or 50,000 cells/well (CTL clones) in 200 ul of RPMI-
1640+10% FBS (R10) medium (PBMC) or R10+50 U/ml IL-2
Materials and Methods (CTL clones) in 96-well round-bottom plates. Where indicated,
Ethics Statement cells were stimulated with 1 mg/ml each of anti-CD3 (OKT3
HIV-infected individuals were recruited from the Maple Leaf clone) and anti-CD28 (CD28.2 clone), both from eBioscience, for
Medical Clinic in Toronto, Canada through a protocol approved 48 hours prior to HDACi addition. HDACis were added at the
by the University of Toronto Institutional Review Board and from indicated concentrations with all conditions tested in triplicate.
the Boston area (United States) under a protocol approved by the Cells were cultured at 37uC, 5% CO2 for 4 the indicated periods
Institution Review Board at the Massachusetts General Hospital. of time. Where indicated, HDACis were washed out with
Secondary use of the samples from Toronto was approved through 26250 ml medium and cells were replated in fresh medium. At
the Massachusetts General Hospital Institutional Review Board. the time of harvest, cells were stained with Annexin-V-Fitc, 7-
All subjects were adults, and gave written informed consent. AAD, anti-CD4 pacific blue (BioLegend), and anti-CD8 alexa-
Clinical data for study subjects are given in Table 1. fluor 700 (BioLegend) using the FITC Annexin V Apoptosis
Detection Kit with 7-AAD (BioLegend), washed, fixed in 4%
Latency Reversing Drugs paraformaldehyde and then analyzed immediately on an LSR-II
Romidepsin was purchased from Selleckchem. An 18.5 mM flow cytometer.
stock was prepared in 100% DMSO (hybrimax, Sigma). This was
diluted 3,7006 in PBS to give a working stock of 5 mM in 0.03% PMA/Ionomycin Intracellular Cytokine Staining Assays
DMSO. SAHA was purchased from Sigma and dissolved to PBMC were plated at 100,000 cells/well in 96-well round
10 mM in 100% DMSO. This was diluted to a working stock of bottom plates in R10 medium and HDACis were added at the
1 mM in PBS (10% DMSO in working stock). Panobinostat was indicated concentrations. Following 4 hours of exposure, cells
obtained from Selleckchem as a 10 mM stock in DMSO. This was were washed with 26250 ml medium and then cultured with
diluted 5006 in in PBS to give a working stock of 20 mM in 0.2% 200 ml fresh R10 medium+1/500 dilution of the 5006 PMA/
DMSO. ALT-803 is an IL-15 superagonist (IL-15SA) produced by ionomycin cell stimulation cocktail (eBioscience) in the presence of
Altor Bioscience Corporation [50,51]. IL-15 has previously been 1 mg/ml Brefeldin A (BD). Following a 5 hour stimulation period,
shown to increase the magnitude of HIV-specific T-cell responses cells were split and stained with either: i) anti-CD4 pacific blue


Figure 8. Time-lapse microscopy of CTL killing of peptide pulsed BLCL target cells in 3D collagen matrices. An HIV-Gag-SLYNTVATL-
specific CTL clone was labeled with Alexa-Fluor555 conjugated cholera toxin subunit B either cultured with 500 nM SAHA or 25 nM romidepsin for
20 hours, or maintained as an untreated control. These effector cells were combined with SLYNTVATL peptide pulsed target cells, matched on the
restricting allele, in a collagen matrix medium containing sytox green viability dye. These mixtures were then plated in three separate wells of an 8-
well cover slip and imaged by time-lapse brightfield and fluorescent microscopy. A. Shown are representative fields of view from the no treatment
(upper panel), 500 nM SAHA (middle panel), and 25 nM romidepsin (lower panel) conditions advancing in time from left to right. Time stamps are
given in hh:mm format. Clones described in the results are indicated with yellow arrows and killed target cells are indicated with white arrows in the
upper right panel. B. The number of killed (sytox green positive) BLCL at T = 20 minutes were counted in each of the four fields of view acquired for
each condition. Each field of view is plotted as a single point on the graph along with means and SD. P values were calculated by the Kruskal-Wallis
test. The reported values are corrected for multiple testing using Dunn’s multiple comparison test.
(BioLegend), and anti-CD8 alexa-fluor 700 (BioLegend), permea- background number of spots in the negative control (DMSO only)
bilized (cytofix/cytoperm, cytoperm/wash, BD) and stained control wells from the number of spots in the experimental well.
intracellularly with anti-IFN-c FITC (to measure IFN-c produc- For assays performed on CTL clones values are reported as SFC
tion) ii) Annexin-V-Fitc, 7-AAD, anti-CD4 pacific blue (BioLe- per thousand CTL clone cells. For assays performed on PBMC
gend), and anti-CD8 alexa-fluor 700 (BioLegend) using the FITC values are reported as SFC per million PBMC.
Annexin V Apoptosis Detection Kit with 7-AAD (BioLegend) (to
measure viability). Cells were fixed % paraformaldehyde and then Proliferation Assays
analyzed immediately on an LSR-II flow cytometer. PBMC from 12 ARV-treated HIV-infected subjects were
thawed and labeled with 0.5 mM CFSE in PBS. Cells were plated
Generation of T-Cell Clones at 100,000 cells/well in 96-well round bottom plates in R10.
PBMCs were stimulated with optimal CD8 T-cell epitopes for Triplicate wells were prepared for each experimental antigen/
6-hours, enriched for antigen-specific cells using the IFN-c drug combination, and outside wells of the plate were filled with
secretion Detection and Enrichment Kit (Miltenyi), and cloned PBS. HDACis or ALT-803 were added in 50 ml of R10 to final
at limiting dilution on irradiated feeder cells as has been previously concentrations of: ALT-803 72 ng/ml, romidepsin 25 nM,
described [58]. Clones were selected from 96-well plates at panobinostat 25 nM, SAHA 500 nM. These concentrations were
dilutions where no more than 1 in 5 wells displayed growth and selected to fall below the Cmax concentrations in ongoing and
screened for specificity by IFN-c ELISPOT. Specific clones were planned clinical trials (see results). No treatment controls were
expanded on irradiated feeder cells. prepared in parallel with 50 ml of R10. Cells were incubated at
37uC, 5% CO2 for 4 hours. Pools of overlapping 15mer peptides
IFN-c ELISPOT spanning CMV-pp65 (Positive Control Pool Pepmix
ELISPOT assays were performed as previously described [59]. HCMV(pp65), JPT Peptide Technologies) and 18mer peptides
CD8+ T-cell clones were plated at 10,000 cells/well with 50,000 overlapping by 10 amino acids spanning HIV-Gag were obtained
autologous EBV-transformed B cell lymphoblastoid cell lines from the MGH peptide core facility and added to final
(BLCL) as target cells. Peripheral blood mononuclear cells concentrations of 1 mg/ml/peptide. After 48 hours of further
(PBMC) were plated at 100,000 cells/well. Peptides representing incubation, cells were washed 26 with 200 ml of R10 and
optimal HIV CD8+ T-cell epitopes were used at final concentra- resuspended in 200 ml of R10 supplemented with 20 U/ml IL-2
tions of 1 mg/ml/peptide, while the pool of overlapping CMV- (NIH AIDS reagent program). Where indicated, the Gag peptide
pp65 15mer peptides (JPT peptide technologies) was used at 1 mg/ pool was re-added along with this fresh medium at a concentration
ml/peptide. Cells were incubated for 16 hours with peptide in of 1 mg/ml/peptide. Following an additional 5 days of culture (7
ELISPOT plates before beginning development. The number of days total) cells were stained with anti-CD4 pacific blue, anti-CD8
specific spot-forming cells (SFC) was calculated by subtracting the alexafluor700, and anti-CD3 PE-CY7 (all antibodies from
Table 1. Clinical characteristics of study subjects.
ID Viral Load CD4 count ART Regimen Treatment Duration
653116 20 624 Emtricitabine, Tenofovir, efavirenz 81 months

514023 20 1258 Emtricitabine, Tenofovir, atazanavir, ritonavir 92 months
128450 40 635 Emtricitabine, Tenofovir, Raltegravir 23 months
198605 50 1234 Combivir, Tenofavir 66 months
296260 20 671 Emtricitabine, Tenofovir, efavirenz 62 months
379080 48 643 Atripla 53 months
409231 50 795 Tenofavir, Kaletra, Abacavir 19 months
498553 48 1128 Ritonavir, Atazanavir, Truvada 20 months
557014 20 1426 Truvada, Raltegravir 35 months
doi:10.1371/journal.ppat.1004287.t001

BioLegend) and analyzed by flow cytometry on an LSRII Time-Lapse Microscopy of CTL Killing in 3D Collagen
instrument. Data were analyzed using Flowjo software (Treestar). Matrices
Following confirmation of functional activity (see above), an
Production of HIV HIV-Gag-specific CTL clone was labeled with 5 mg/ml cholera
The HIV molecular clone JR-CSF was obtained from the NIH toxin subunit B AlexaFluor 555 (Life Technologies) in 0.5 ml
AIDS Reagent Program (www.aidsreagent.org). Viral stocks were R10+50 U/ml IL-2 for 10 minutes at 37uC. Cells were washed,
produced by transfection of 293T cells using Fugene HD split into equal aliquots in R10+50 U/ml IL-2 and treated with
(Promega) following the manufacturer’s instructions. Superna- HDACi drugs at the indicated concentrations for 14 or 24 hours.
tants were harvested 48 hours post-transfection, centrifuged at HLA-A02+ BLCL were then pulsed with 100 ng/ml SLYNT-
1,0006g for 5 minutes to pellet cell debris, and filtered through a VATL Gag peptide for 30 minutes, and washed. CTL and BLCL
400 mM membrane (Steriflip, Millipore). Viral titers were were each resuspended to final concentrations of 2.76106 cells/ml
determined by titration on primary cells by the Ragon Institute in R10+50 U/ml IL-2. 3D collagen time-lapse microscopy was
Virology Core. performed as previously described [45].
ACH2 Latency Reversal Assays Statistical Analysis

The chronically HIV-infected ACH2 cell line was maintained in Statistical analyses were performed using Prism software
R10 medium. Cells were plated at 100,000 cells/well in 96-well (Graphpad). The statistical tests used to calculate p values are
round-bottom plates. Latency reversal drugs were added at the indicated in the corresponding figure legends.
indicated concentrations and incubated for 4 hours. Cells were
then washed two times to remove drugs, replated in fresh R10 Supporting Information
medium, and cultured for an additional 20 hours. Cells were fixed Figure S1 Effects of HDACis on the viability of PBMC
and permeabilized using cytofix/cytoperm and perm/wash CD8+ and CD4+ T-cells from HIV-infected subjects, and
reagents (BD) and then stained intracellularly with anti-HIV- on HIV-specific CTL clones. A. Cryopreserved PBMC from
Gag KC57-RD1 (Beckman Coulter) diluted 1/100 in perm/wash the ARV-treated HIV-infected subjects OM292 and OM265 were
buffer. After washing to remove unbound antibodies, cells were thawed and then treated with HDACi drugs at the indicated
analyzed on a FACSCalibur instrument (BD). concentrations for 6 hours. Cells were harvested at the indicated
time-points 21 hours and stained with 7-AAD (stains DNA of dead
HIV Elimination Assays cells), CD4, CD3, and CD8. Conditions were tested in triplicate.
HIV-Gag SLYNTVATL and FLGKIWPSHK specific CD8+ Shown are summary data depicting means 6 SEM. P values were
T-cell clones were obtained from two different HIV-infected elite calculated by two-way ANOVA with Dunnett’s multiple compar-
controller subjects and maintained as previously described ison test (comparing to the no drug control) * p,0.05, ** p,0.01,
[58,60]. The specificities and cytotoxic potential of these clones *** p,0.001, **** p,0.0001.
were confirmed two days prior to performing each elimination (EPS)
assay experiment by measuring degranulation in response to Movie S1 Time-lapse microscopy of untreated CTL
cognate peptide stimulation by a CD107a-staining flow cytometry with peptide pulsed BLCL target cells. Shown is a time-
method [61]. On this same day, HLA-A02+ PBMC from HIV- lapse microscopy movie corresponding to the images depicted in
uninfected subjects were thawed and enriched for CD4+ T-cells the upper panel of Fig. 7A.
by negative selection (Easysep, Stemcell Technologies). These (MOV)
CD4+ T-cells were stimulated with 1 mg/ml each anti-CD3
OKT3 and anti-CD28.2 antibodies (eBioscience) in R10 supple- Movie S2 Time-lapse microscopy of SAHA-treated with
mented with 50 U/ml IL-2 (NIH AIDS Reagent Program) for peptide pulsed BLCL target cells. Shown is a time-lapse
48 hours. These activated target cells were infected with HIV JR- microscopy movie corresponding to the images depicted in the
CSF using a previously described magnetofection method [62]. middle panel of Fig. 7A.
(MOV)
Infection levels were monitored by flow cytometry staining for
CD4 (BioLegend) and intracellularly for HIV-Gag (KC57 clone, Movie S3 Time-lapse microscopy of romidepsin-treat-
Beckman Coulter). Targets were determined to be ready for use ed with peptide pulsed BLCL target cells. Shown is a time-
in the elimination assay when they were 20% Gag+ by flow lapse microscopy movie corresponding to the images depicted in
cytometry. At this time-point, CTL clones were split into equal the lower panel of Fig. 7A.
aliquots and cultured in R10+50 U/ml IL-2 supplemented with (MOV)
HDACis and ALT-803 as indicated. Following 4 hours of
incubation in 5% CO2 at 37uC, CTL were washed with 36 Acknowledgments
with 1 ml R10. Target CD4+ T-cells were plated at 50,000 cells/
We thank Musie Ghebremichael for assistance with statistical analyses, and
well in 96-well round-bottom plates, and CTL were added at the the NIH AIDS Research and Reference Reagent Program for the
indicated effector:target ratios. Each condition was set up in provision of IL-2. We also thank Altor Bioscience Corporation for the
triplicate. Co-culture was allowed to proceed for 16 hours at provision of ALT-803.
37uC, 5% CO2. Cells were then surface stained with anti-CD8-
Fitc and anti-CD4-APC (BioLegend), permeabilized (cytofix/ Author Contributions
cytoperm, BD) and stained intracellularly with anti-HIV-Gag-PE
Conceived and designed the experiments: RBJ GLS MF ML DJI BDW.
at 1/100 dilution (KC57 clone, Beckman Coulter). Cells were Performed the experiments: RBJ RO SM GLS DK ML AT. Analyzed the
then fixed with 4% formalin in PBS and analyzed on an LSR-II data: RBJ RO SM GLS. Contributed reagents/materials/analysis tools:
flow cytometer instrument (BD Biosciences). Data were analyzed ML. Wrote the paper: RBJ. Contributed patient samples: MAO CK.
using Flowjo software (Treestar). Supervised the project: DJI BDW.

References
1. Deeks SG (2011) HIV infection, inflammation, immunosenescence, and aging. 28. Camelo S, Iglesias AH, Hwang D, Due B, Ryu H, et al. (2005) Transcriptional
Annu Rev Med 62: 141–155. therapy with the histone deacetylase inhibitor trichostatin A ameliorates
2. Deeks SG, Lewin SR, Havlir DV (2013) The end of AIDS: HIV infection as a experimental autoimmune encephalomyelitis. J Neuroimmunol 164: 10–21.
chronic disease. Lancet 382: 1525–1533. 29. Dowdell KC, Pesnicak L, Hoffmann V, Steadman K, Remaley AT, et al. (2009)
3. Chun TW, Stuyver L, Mizell SB, Ehler LA, Mican JA, et al. (1997) Presence of Valproic acid (VPA), a histone deacetylase (HDAC) inhibitor, diminishes
an inducible HIV-1 latent reservoir during highly active antiretroviral therapy. lymphoproliferation in the Fas -deficient MRL/lpr(2/2) murine model of
Proc Natl Acad Sci U S A 94: 13193–13197. autoimmune lymphoproliferative syndrome (ALPS). Exp Hematol 37: 487–494.
4. Finzi D, Hermankova M, Pierson T, Carruth LM, Buck C, et al. (1997) 30. Garcia BA, Busby SA, Shabanowitz J, Hunt DF, Mishra N (2005) Resetting the
Identification of a reservoir for HIV-1 in patients on highly active antiretroviral epigenetic histone code in the MRL-lpr/lpr mouse model of lupus by histone
therapy. Science 278: 1295–1300. deacetylase inhibition. J Proteome Res 4: 2032–2042.
5. Wong JK, Hezareh M, Gunthard HF, Havlir DV, Ignacio CC, et al. (1997) 31. Larsen L, Tonnesen M, Ronn SG, Storling J, Jorgensen S, et al. (2007)
Recovery of replication-competent HIV despite prolonged suppression of Inhibition of histone deacetylases prevents cytokine-induced toxicity in beta cells.
plasma viremia. Science 278: 1291–1295. Diabetologia 50: 779–789.
6. Finzi D, Blankson J, Siliciano JD, Margolick JB, Chadwick K, et al. (1999) 32. Reddy P, Sun Y, Toubai T, Duran-Struuck R, Clouthier SG, et al. (2008)
Latent infection of CD4+ T cells provides a mechanism for lifelong persistence of Histone deacetylase inhibition modulates indoleamine 2,3-dioxygenase-depen-
HIV-1, even in patients on effective combination therapy. Nat Med 5: 512–517. dent DC functions and regulates experimental graft-versus-host disease in mice.
7. Chomont N, El-Far M, Ancuta P, Trautmann L, Procopio FA, et al. (2009) HIV J Clin Invest 118: 2562–2573.
reservoir size and persistence are driven by T cell survival and homeostatic 33. Wang D, Iclozan C, Liu C, Xia C, Anasetti C, et al. (2012) LBH589 enhances T
proliferation. Nat Med 15: 893–900. cell activation in vivo and accelerates graft-versus-host disease in mice. Biol
8. Buzon MJ, Massanella M, Llibre JM, Esteve A, Dahl V, et al. (2010) HIV-1 Blood Marrow Transplant 18: 1182–1190 e1181.
replication and immune dynamics are affected by raltegravir intensification of 34. Bridle BW, Chen L, Lemay CG, Diallo JS, Pol J, et al. (2013) HDAC inhibition
HAART-suppressed subjects. Nat Med 16: 460–465. suppresses primary immune responses, enhances secondary immune responses,
9. Archin NM, Liberty AL, Kashuba AD, Choudhary SK, Kuruc JD, et al. (2012) and abrogates autoimmunity during tumor immunotherapy. Mol Ther 21: 887–
Administration of vorinostat disrupts HIV-1 latency in patients on antiretroviral 894.
therapy. Nature 487: 482–485. 35. Khan N, Jeffers M, Kumar S, Hackett C, Boldog F, et al. (2008) Determination
10. Archin NM, Espeseth A, Parker D, Cheema M, Hazuda D, et al. (2009) of the class and isoform selectivity of small-molecule histone deacetylase
Expression of latent HIV induced by the potent HDAC inhibitor suberoylanilide inhibitors. Biochem J 409: 581–589.
hydroxamic acid. AIDS Res Hum Retroviruses 25: 207–212. 36. Liu T, Kuljaca S, Tee A, Marshall GM (2006) Histone deacetylase inhibitors:
11. Rasmussen TA, Schmeltz Sogaard O, Brinkmann C, Wightman F, Lewin SR, et multifunctional anticancer agents. Cancer Treat Rev 32: 157–165.
al. (2013) Comparison of HDAC inhibitors in clinical development: effect on 37. Marks PA, Richon VM, Miller T, Kelly WK (2004) Histone deacetylase
HIV production in latently infected cells and T-cell activation. Hum Vaccin inhibitors. Adv Cancer Res 91: 137–168.
Immunother 9: 993–1001. 38. Lindemann RK, Gabrielli B, Johnstone RW (2004) Histone-deacetylase
inhibitors for the treatment of cancer. Cell Cycle 3: 779–788.
12. Wightman F, Ellenberg P, Churchill M, Lewin SR (2012) HDAC inhibitors in
HIV. Immunol Cell Biol 90: 47–54. 39. Mayo MW, Denlinger CE, Broad RM, Yeung F, Reilly ET, et al. (2003)
Ineffectiveness of histone deacetylase inhibitors to induce apoptosis involves the
13. Shan L, Deng K, Shroff NS, Durand CM, Rabi SA, et al. (2012) Stimulation of
transcriptional activation of NF-kappa B through the Akt pathway. J Biol Chem
HIV-1-specific cytolytic T lymphocytes facilitates elimination of latent viral
278: 18980–18989.
reservoir after virus reactivation. Immunity 36: 491–501.
40. Johnstone RW (2002) Histone-deacetylase inhibitors: novel drugs for the
14. Licciardi PV, Karagiannis TC (2012) Regulation of immune responses by
treatment of cancer. Nat Rev Drug Discov 1: 287–299.
histone deacetylase inhibitors. ISRN Hematol 2012: 690901.
41. Tao R, de Zoeten EF, Ozkaynak E, Chen C, Wang L, et al. (2007) Deacetylase
15. Akimova T, Beier UH, Liu Y, Wang L, Hancock WW (2012) Histone/protein
inhibition promotes the generation and function of regulatory T cells. Nat Med
deacetylases and T-cell immune responses. Blood 119: 2443–2451.
13: 1299–1307.
16. Bosisio D, Vulcano M, Del Prete A, Sironi M, Salvi V, et al. (2008) Blocking 42. Cocchi F, DeVico AL, Garzino-Demo A, Arya SK, Gallo RC, et al. (1995)
TH17-polarizing cytokines by histone deacetylase inhibitors in vitro and in vivo. Identification of RANTES, MIP-1 alpha, and MIP-1 beta as the major HIV-
J Leukoc Biol 84: 1540–1548. suppressive factors produced by CD8+ T cells. Science 270: 1811–1815.
17. Glauben R, Batra A, Fedke I, Zeitz M, Lehr HA, et al. (2006) Histone 43. Beier UH, Akimova T, Liu Y, Wang L, Hancock WW (2011) Histone/protein
hyperacetylation is associated with amelioration of experimental colitis in mice. deacetylases control Foxp3 expression and the heat shock response of T-
J Immunol 176: 5015–5022. regulatory cells. Curr Opin Immunol 23: 670–678.
18. Leoni F, Zaliani A, Bertolini G, Porro G, Pagani P, et al. (2002) The antitumor 44. Clouse KA, Powell D, Washington I, Poli G, Strebel K, et al. (1989) Monokine
histone deacetylase inhibitor suberoylanilide hydroxamic acid exhibits antiin- regulation of human immunodeficiency virus-1 expression in a chronically
flammatory properties via suppression of cytokines. Proc Natl Acad Sci U S A infected human T cell clone. J Immunol 142: 431–438.
99: 2995–3000. 45. Foley MH, Forcier T, McAndrew E, Gonzalez M, Chen H, et al. (2014) High
19. Mishra N, Reilly CM, Brown DR, Ruiz P, Gilkeson GS (2003) Histone avidity CD8+ T cells efficiently eliminate motile hiv-infected targets and execute
deacetylase inhibitors modulate renal disease in the MRL-lpr/lpr mouse. J Clin a locally focused program of anti-viral function. PLoS One 9: e87873.
Invest 111: 539–552. 46. Rothstein TL, Mage M, Jones G, McHugh LL (1978) Cytotoxic T lymphocyte
20. Reddy P, Maeda Y, Hotary K, Liu C, Reznikov LL, et al. (2004) Histone sequential killing of immobilized allogeneic tumor target cells measured by time-
deacetylase inhibitor suberoylanilide hydroxamic acid reduces acute graft- lapse microcinematography. J Immunol 121: 1652–1656.
versus-host disease and preserves graft-versus-leukemia effect. Proc Natl Acad 47. Sanderson CJ (1976) The mechanism of T cell mediated cytotoxicity. II.
Sci U S A 101: 3921–3926. Morphological studies of cell death by time-lapse microcinematography.
21. Shi ZJ, Ouyang DY, Zhu JS, Xu LH, He XH (2012) Histone deacetylase Proc R Soc Lond B Biol Sci 192: 241–255.
inhibitor suberoylanilide hydroxamic acid exhibits anti-inflammatory activities 48. Matter A (1979) Microcinematographic and electron microscopic analysis of
through induction of mitochondrial damage and apoptosis in activated target cell lysis induced by cytotoxic T lymphocytes. Immunology 36: 179–190.
lymphocytes. Int Immunopharmacol 12: 580–587. 49. Corporation C (2013) Istodax Package Insert.
22. Brogdon JL, Xu Y, Szabo SJ, An S, Buxton F, et al. (2007) Histone deacetylase 50. Zhu X, Marcus WD, Xu W, Lee HI, Han K, et al. (2009) Novel human
activities are required for innate immune cell control of Th1 but not Th2 effector interleukin-15 agonists. J Immunol 183: 3598–3607.
cell function. Blood 109: 1123–1130. 51. Han KP, Zhu X, Liu B, Jeng E, Kong L, et al. (2011) IL-15:IL-15 receptor alpha
23. Ge Z, Da Y, Xue Z, Zhang K, Zhuang H, et al. (2013) Vorinostat, a histone superagonist complex: high-level co-expression in recombinant mammalian
deacetylase inhibitor, suppresses dendritic cell function and ameliorates cells, purification and characterization. Cytokine 56: 804–810.
experimental autoimmune encephalomyelitis. Exp Neurol 241: 56–66. 52. Jennes W, Kestens L, Nixon DF, Shacklett BL (2002) Enhanced ELISPOT
24. Kelly-Sell MJ, Kim YH, Straus S, Benoit B, Harrison C, et al. (2012) The detection of antigen-specific T cell responses from cryopreserved specimens with
histone deacetylase inhibitor, romidepsin, suppresses cellular immune functions addition of both IL-7 and IL-15–the Amplispot assay. J Immunol Methods 270:
of cutaneous T-cell lymphoma patients. Am J Hematol 87: 354–360. 99–108.
25. Moreira JM, Scheipers P, Sorensen P (2003) The histone deacetylase inhibitor 53. Strengell M, Matikainen S, Siren J, Lehtonen A, Foster D, et al. (2003) IL-21 in
Trichostatin A modulates CD4+ T cell responses. BMC Cancer 3: 30. synergy with IL-15 or IL-18 enhances IFN-gamma production in human NK
26. Wang B, Morinobu A, Horiuchi M, Liu J, Kumagai S (2008) Butyrate inhibits and T cells. J Immunol 170: 5464–5469.
functional differentiation of human monocyte-derived dendritic cells. Cell 54. Shapiro GI, Frank R, Dandamudi UB, Hengelage T, Zhao L, et al. (2012) The
Immunol 253: 54–58. effect of food on the bioavailability of panobinostat, an orally active pan-histone
27. Bohmig GA, Krieger PM, Saemann MD, Wenhardt C, Pohanka E, et al. (1997) deacetylase inhibitor, in patients with advanced cancer. Cancer Chemother
n-butyrate downregulates the stimulatory function of peripheral blood-derived Pharmacol 69: 555–562.
antigen-presenting cells: a potential mechanism for modulating T-cell responses 55. Wei DG, Chiang V, Fyne E, Balakrishnan M, Barnes T, et al. (2014) Histone
by short-chain fatty acids. Immunology 92: 234–243. Deacetylase Inhibitor Romidepsin Induces HIV Expression in CD4 T Cells

from Patients on Suppressive Antiretroviral Therapy at Concentrations cell responses without high-frequency fixation of mutations. J Virol 83: 8722–
Achieved by Clinical Dosing. PLoS Pathog 10: e1004071. 8732.
56. Rubin EH, Agrawal NG, Friedman EJ, Scott P, Mazina KE, et al. (2006) A 59. Streeck H, Frahm N, Walker BD (2009) The role of IFN-gamma Elispot assay in
study to determine the effects of food and multiple dosing on the HIV vaccine research. Nat Protoc 4: 461–469.
pharmacokinetics of vorinostat given orally to patients with advanced cancer. 60. Walker BD, Flexner C, Birch-Limberger K, Fisher L, Paradis TJ, et al. (1989)
Clin Cancer Res 12: 7039–7045. Long-term culture and fine specificity of human cytotoxic T-lymphocyte clones
57. Mejia EJ, Loveridge ST, Stepan G, Tsai A, Jones GS, et al. (2014) Study of reactive with human immunodeficiency virus type 1. Proc Natl Acad Sci U S A
Marine Natural Products Including Resorcyclic Acid Lactones from Humicola 86: 9514–9518.
fuscoatra That Reactivate Latent HIV-1 Expression in an in Vitro Model of 61. Jones RB, Garrison KE, Mujib S, Mihajlovic V, Aidarus N, et al. (2012) HERV-
Central Memory CD4+ T Cells. J Nat Prod 77: 618–24 K-specific T cells eliminate diverse HIV-1/2 and SIV primary isolates. J Clin
58. Jones RB, Yue FY, Gu XX, Hunter DV, Mujib S, et al. (2009) Human Invest 122: 4473–4489.
immunodeficiency virus type 1 escapes from interleukin-2-producing CD4+ T- 62. Sacha JB, Watkins DI (2010) Synchronous infection of SIV and HIV in vitro for
virology, immunology and vaccine-related studies. Nat Protoc 5: 239–246.

Copyright of PLoS Pathogens is the property of Public Library of Science and its content may
not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.

Histone HVI

Uploaded by

Copyright:

Available Formats

Histone HVI

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Histone HVI

Uploaded by

Copyright:

Available Formats

Histone Deacetylase Inhibitors Impair the Elimination of

HIV-Infected Cells by Cytotoxic T-Lymphocytes

Introduction homeostatic proliferation, it is unlikely that current ART regimens

PLOS Pathogens | www.plospathogens.org 1 August 2014 | Volume 10 | Issue 8 | e1004287

Author Summary elevated serum Th1 cytokines and accelerated progression in a

PLOS Pathogens | www.plospathogens.org 2 August 2014 | Volume 10 | Issue 8 | e1004287

PLOS Pathogens | www.plospathogens.org 3 August 2014 | Volume 10 | Issue 8 | e1004287

PLOS Pathogens | www.plospathogens.org 4 August 2014 | Volume 10 | Issue 8 | e1004287

PLOS Pathogens | www.plospathogens.org 5 August 2014 | Volume 10 | Issue 8 | e1004287

PLOS Pathogens | www.plospathogens.org 6 August 2014 | Volume 10 | Issue 8 | e1004287

PLOS Pathogens | www.plospathogens.org 7 August 2014 | Volume 10 | Issue 8 | e1004287

PLOS Pathogens | www.plospathogens.org 8 August 2014 | Volume 10 | Issue 8 | e1004287

PLOS Pathogens | www.plospathogens.org 9 August 2014 | Volume 10 | Issue 8 | e1004287

PLOS Pathogens | www.plospathogens.org 10 August 2014 | Volume 10 | Issue 8 | e1004287

PLOS Pathogens | www.plospathogens.org 11 August 2014 | Volume 10 | Issue 8 | e1004287

PLOS Pathogens | www.plospathogens.org 12 August 2014 | Volume 10 | Issue 8 | e1004287

PLOS Pathogens | www.plospathogens.org 13 August 2014 | Volume 10 | Issue 8 | e1004287

PLOS Pathogens | www.plospathogens.org 14 August 2014 | Volume 10 | Issue 8 | e1004287

PLOS Pathogens | www.plospathogens.org 15 August 2014 | Volume 10 | Issue 8 | e1004287

Table 1. Clinical characteristics of study subjects.

ID Viral Load CD4 count ART Regimen Treatment Duration

653116 20 624 Emtricitabine, Tenofovir, efavirenz 81 months

PLOS Pathogens | www.plospathogens.org 16 August 2014 | Volume 10 | Issue 8 | e1004287

ACH2 Latency Reversal Assays Statistical Analysis

PLOS Pathogens | www.plospathogens.org 17 August 2014 | Volume 10 | Issue 8 | e1004287

PLOS Pathogens | www.plospathogens.org 18 August 2014 | Volume 10 | Issue 8 | e1004287

PLOS Pathogens | www.plospathogens.org 19 August 2014 | Volume 10 | Issue 8 | e1004287

You might also like