Cutting Edge: T Regulatory Cell Depletion Reactivates Latent Simian Immunodeficiency Virus (SIV) in Controller Macaques While Boosting SIV-Specific T Lymphocytes

The burden of the HIV epidemic, which spreads unabated such that for every HIV-infected person who starts antiretroviral therapy (ART) two new people become infected, calls for a cure (1). The “Berlin patient” (2) demonstrated that a cure for HIV infection is feasible, but the mechanisms responsible for achieving success in this case are complex and poorly understood. Furthermore, virus relapse in the “Mississippi baby” (3, 4) and nonhuman primate studies documenting seeding of the viral reservoir prior to detectable viremia (5) suggest that virus eradication strategies should involve therapeutic approaches that go beyond ART.

Multiple strategies for a HIV cure are currently being investigated. One of the most advanced strategies is the shock and kill approach, consisting of induction of viral expression in latently infected cells that could trigger immune-mediated clearance of the infected cells through CTLs, NK cells, or immunotoxins (6, 7). The most critical limitation of this strategy is the lack of effective latency-reversing agents (LRAs). Available LRAs activate only a minor fraction of resting cells, the most critical component of the reservoirs (1, 8). In contrast, even if such effective compounds become available, reactivated virus cannot be effectively cleared due to HIV/SIV-specific CTL impairment and exhaustion (4). Therefore, there is intense research aimed at both improving LRA efficacy as well as restoring the impairment of SIV-specific T cells.

There is strong evidence that T regulatory cells (Tregs) can become latently infected with HIV and may represent a potentially important HIV reservoir: 1) Tregs expand in blood and tissues in chronically HIV-infected patients and SIV-infected macaques (9); 2) the fraction of Tregs containing HIV/SIV DNA is higher than that of non-Tregs in HIV-infected patients on ART (10) and in SIV-infected rhesus macaques (RMs) (11); and 3) Tregs are less susceptible to cell death than conventional T cells (9). As such, therapeutic interventions aiming at Treg depletion may directly contribute to the reduction of the size of virus reservoir.

Furthermore, through their regulatory function, Tregs can indirectly shape the reservoir. During the acute HIV/SIV infection, Tregs may decisively contribute to the rapid establishment of the HIV reservoir by reversing the immune activation status of CD4⁺ T cells (9). During chronic HIV/SIV infection, Tregs contribute to the impairment of CTL responses, as suggested by the following observations: Treg expansion correlates with loss of CTL function (12–14); ex vivo Treg depletion from blood and lymph nodes enhances T cell responses to HIV or SIV Ags (9); HIV nonprogressors have a high perforin/FOXP3 ratio; and HLA B27⁺ and B57⁺ HIV-specific CD8⁺ T cells from elite controllers are able to evade Treg suppression (15, 16). Inhibition of cellular immune responses by Tregs was also reported in other infectious and noninfectious conditions, such as hepatitis C and several types of cancer (9). Furthermore, Treg depletion resulted in a significant improvement of cellular immune responses and prolonged survival in cancer patients with T lymphomas (9). Altogether, these observations support a major involvement of Tregs in suppressing the protective effector immune responses against HIV. It is thus conceivable that Treg depletion in HIV/SIV subjects may improve cell-mediated immunity

Two RMs (Macacca mulatta) received plasma equivalent to 300 50% tissue culture-infective doses of the SIVsab92018 (18). Animals were housed and handled in accordance with guidelines for the care and use of laboratory animals from the Public Health Service, the American Association for Accreditation of Laboratory Animal Care, and the Animal Welfare Act (19). The University of Pittsburgh Institutional Animal Care and Use Committee approved all protocols and procedures.

SIVsab infection follow-up was performed, as described (17, 18). After 1.5 y postinfection, when the animals had completely controlled the virus, they were treated twice with Ontak (Ligand Pharmaceutical, La Jolla, CA) i.v. (15 μg/kg) for consecutive 5 d at 21-d interval, as reported (20). Blood samples were collected prior to Ontak administration and then at 1, 3, 7, 10, 14, 17, and 21 d posttreatment (dpt) initiation. Plasma and mononuclear cells were isolated as described (17, 18) for viral load quantification and flow cytometry.

Plasma viral loads were quantified with an ultrasensitive quantitative PCR assay, as described (17). This assay has a sensitivity of one viral RNA copy/ml; however, due to frequent sampling, which limited the volumes of plasma, assay sensitivity was five copies/ml.

Whole blood was stained for flow cytometry with multiple combinations of the following mAbs: CD3-FITC (SP34), CD20-PE (3G8), CD8-PerCP (SK1), CD4-allophycocyanin (L200), CD25-FITC (2A3), HLA-DR-PerCP (l243), and Ki-67-FITC (B56) (all from BD Biosciences), and FOXP3-allophycocyanin (PCH101) (eBioscience, San Diego, CA), as described (17, 18). Data were acquired with a FACSCalibur flow cytometer (BD Immunocytometry Systems) and analyzed with CellQuest software (BD Biosciences). CD4⁺ and CD8⁺ T cell percentages were obtained by gating first on lymphocytes and then on CD3⁺ T cells. Immune activation and proliferation markers were determined by gating on lymphocytes, then on CD3⁺ T cells, and finally on CD4⁺CD3⁺ or CD8⁺CD3⁺ T cells.

Env- and Gag SIVsab-specific CD4 and CD8 responses (IFN-γ, TNF-α, IL-2, MIP-1β, and CD107a) were measured by intracellular staining and assessed by flow cytometry, as described (8), using SIVsab-specific peptide pools: Env (52 peptides) and Gag (pool 1: 1–68; pool 2: 69–136 peptides). SIV-specific cells were acquired the same day on a custom four-laser BD LSR-II instrument (BD Biosciences). Only singlet events were gated, and a minimum of 250,000 live CD3 cells was acquired with FACSDIVA 8.0. Populations were analyzed using FlowJo software version 7.6.5 (Tree Star, Ashland, OR), and the graphs were generated with GraphPad Prism 6.04.

The measurements were performed using anti-granzyme B PE (eBioscience) and anti-human perforin Abs from BD Biosciences (δG9). Standard intracellular cytokine stimulation and staining procedures were performed, as described (21). Monensin (1 μg/ml) and concanamycin A (0.2 μM; Sigma-Aldrich) were used for inhibition of cytotoxic granule acidification. For each tube, between 500,000 and 1,000,000 total events were acquired with a LSRII flow cytometer (BD Immunocytometry Systems, San Jose, CA). Ab capture beads (BD Biosciences) were used to prepare individual compensation tubes for each Ab. Data analysis was performed using FlowJo (version 8.5.2; Tree Star). Reported data have been corrected for background, when appropriate.

The limited number of included RMs did not permit us to perform statistical calculations. However, the various parameters were compared within the same animal prior and after reactivation with Ontak, to assess the biological effects of this treatment. For the increases in perforin expression by CD4⁺ and CD8⁺ T cells, results of both treatments were grouped together and paired t test was performed using Prism GraphPad.

Two RMs were experimentally infected with SIVsab and followed for 500 d. Similar to our previous results (17, 18), after very active virus replication during the acute infection (Supplemental Fig. 1A), with significant depletion of CD4⁺ T cells (Supplemental Fig. 1B), SIVsab was completely controlled in RMs during chronic infection (Supplemental Fig. 1A). This allowed for control of chronic immune activation (Supplemental Fig. 1C, 1D) and restoration of the CD4⁺ T cells in the gut (Supplemental Fig. 1B). As suggested by previous results, control was most likely due to effective cellular immune responses (18).

Ontak administration to SIVsab-infected controller RMs after 500 d of complete control of viral replication resulted in significant depletion of the CD25⁺CD4⁺ T cells (Fig. 1A). Depletion was rapid and massive, with up to >50% of the CD25⁺CD4⁺ T cells being depleted by 14 dpt. However, after the second administration of Ontak, there was a very robust increase in the number of CD25⁺CD4⁺ T cells (Fig. 1A). The number of CD8⁺ T cells expressing CD25 showed a similar pattern of initial depletion, followed by a robust rebound (Fig. 1A).

We further monitored the dynamics of FOXP3⁺ CD25⁺CD4⁺ T cells and report that Ontak administration induces a rapid and massive (75–85%) depletion of the FOXP3⁺ CD25⁺CD4⁺ T cells (Fig. 2B). A trend of FOXP3⁺ CD25⁺CD4⁺ T cell rebound was observed prior to the second administration of Ontak (Fig. 1A). The impact on the CD8⁺ T cells was only moderate, as the frequency of CD8⁺ T cells expressing FOXP3 is very low (Fig. 1B).

A major increase of CD4⁺ T cell activation was observed after administration of Ontak. The frequency of the circulating CD4⁺ and CD8⁺ T cells expressing Ki-67 dramatically increased (on an average of 8- to 10-fold) after each Ontak administration (Fig. 1C). These robust increases of CD4⁺ and CD8⁺ T cell activation persisted beyond the 5-d course of each Ontak administration.

One may argue that these Ontak-induced increases in the levels of immune activation may have inadvertently boosted the size of the CD25⁺CD4⁺ T cell fraction, which is associated with cell activation. This is not necessary a weakness of our approach. Ontak targets the cells that express CD25 and depletes them through the immunotoxin component of the drug. As such, increases in the pool of CD25-expressing CD4⁺ T cells will only result in an increase in the pool of the Ontak-depleted CD4⁺ cells, thus improving the drug ability to directly target the reservoir.

Prior to Ontak administration, the levels of SIVsab in plasma were below five viral RNA copies/ml (Supplemental Fig. 1). After treatment, SIVsab rebounded up to 10³ copies/ml (Fig. 1D). This virus rebound, albeit relatively low, is not negligible and strongly suggests that Ontak could be used either alone or in combination with other conventional LRAs for virus reactivation. Unlike LRAs currently tested, which require the presence of effective CTLs to clear the cells expressing the reactivated virus, Ontak has the advantage that the cells targeted by this drug are destroyed by the diphtheria toxin that is coupled to the IL-2. Furthermore, through Treg depletion, Ontak also has the ability to boost the cell-mediated immune responses necessary for effective clearance of the reactivated virus.

One of the most notable effects of Treg depletion with Ontak was a significant boost of SIV-specific CD8⁺ T cells (Fig. 2A). This increase occurred in every test performed, with every tested peptide pool, and thus, the absolute numbers (per milliliter) of the aggregated SIV-specific CD4⁺ and CD8⁺ T cells at different time points following different rounds of Ontak administration, as determined by intracellular cytokine staining and flow cytometry, were significantly boosted (p = 0.0001 and p = 0.0029 for CD4⁺ and CD8⁺ T cells, respectively) in Ontak-treated RM (Fig. 2B). SIV-specific CD4⁺ T cells increased in RM1-10 from 6,430/ml prior to Ontak administration to 23,437/ml at 7 dpt, whereas in RM2-10 they increased from 7,456/ml prior to Ontak administration to 19,864/ml at 7 dpt. Similarly, SIV-specific CD8⁺ T cells increased in RM1-10 from 10,346/ml prior to Ontak administration to 47,899/ml at 7 dpt, whereas in RM2-10 they increased from 5,678/ml prior to Ontak administration to 45,673/ml at 7 dpt. Furthermore, measurements of granzyme and perforin as surrogate markers of cytotoxic responses revealed a massive boost of their expression on both CD4⁺ and CD8⁺ T cells. Thus, perforin expression increased 4- to 8-fold on CD4⁺ T cells (p = 0.0342) and 2- to 7-fold on the CD8⁺ T cells (p = 0.0313) at 7 dpt (Fig. 2C).

Altogether, our results demonstrate the feasibility of Treg depletion with Ontak in chronically SIV-infected RMs, also pointing to the efficacy of the proposed approach on both reactivating the latent virus and boosting CTL responses.

Other approaches targeting Tregs have also been reported to have potential impact for cure research. Treg blockade with the anti–CTLA-4 drug ipilimumab in a HIV-infected patient on ART had significant effects on the total number and phenotype of CD4⁺ T cells. Furthermore, the drug induced a profound increase in cell-associated unspliced HIV RNA, resulting in a subsequent decline in plasma HIV RNA (22). Similarly, Treg blockade with the anti–CTLA-4 human Ab MDX-010 in SIVmac-infected RMs receiving ART (23) resulted in decreased expression of IDO and TGF-β in tissues and was associated with decreased viral RNA levels in the lymph nodes and increased effector function of both SIV-specific CD4⁺ and CD8⁺ T cells. Dampening Treg function in SIV-infected RMs did not have detrimental virologic effects and was shown to provide a valuable approach to complement ART and therapeutic vaccination during treatment of HIV-1 infection (23). These results further validate our approach as a promising cure strategy.

As such, our results provide strong proof-of-concept data that support approaches aimed at altering the number and function of Tregs as a strategy for cure research. Treg depletion has a dual effect, leading to both reactivation of the latent virus and improved clearing of the reactivated virus. As none of the LRAs in development has been reported to have such a dual activity, Treg depletion, alone or in combination with other LRAs, holds great promises for cure research and has a real potential to provide both improved shock (through reactivation by LRA and Treg-depleting agent) and kill (through CTL boosting after Treg depletion) effects.

We thank Dr. Claire Chougnet, Dr. Nicholas Chomont, and Dr. Ruy M. Ribeiro for very helpful discussion.

The authors have no financial conflicts of interest.