A B cell follicle sanctuary permits persistent productive SIV infection in elite controllers

Immune control restricts productive SIV infection to T_FH

Although CD4⁺ T_FH have been shown to constitute a major, and often disproportionate, component of productively HIV- and SIV-infected cell populations during chronic phase infection, it is unclear whether preferential viral targeting of this subset is based on enhanced susceptibility to infection, a longer intrinsic lifespan of infected cells, location in a particularly conducive environment for productive infection, the relative expansion of CD4⁺ T_FH vs. non-T_FH in progressive infection, or preferential shielding of productively infected CD4⁺ T_FH from immune elimination^29–33 . To initially address this question, we quantified replication-competent virus by co-culture analysis of sorted CD4⁺ T_FH vs. non-T_FH memory T cells from lymph nodes (LNs) of four rhesus monkeys over the course of controlled LAV (nef-deleted SIVmac239) vs. progressive wildtype (WT) SIVmac239 infection. CD4⁺ T_FH, delineated by high co-expression of PD1 and CD200 on CD4⁺, CD95^high memory cells^21,34 , were compared to non-T_FH CD4⁺ memory T cells that lacked PD1 and CD200 expression completely (PD-1^neg/CD200^neg) or expressed these markers at low level (PD-1^dim/CD200^dim; Suppl. Figs. 1 and 2). During acute LAV infection, and all phases of progressive WT SIV infection, replication-competent virus was present at similar levels in both CD4⁺ T_FH and non-T_FH memory T cells (Fig. 1a–c; Suppl. Fig. 3). However, after the onset of immunologic (primarily CD8⁺ T cell-mediated) control of LAV replication, replication-competent virus could only be isolated from T_FH (Fig. 1c). These data suggest that the preferential localization of replication-competent LAV within CD4⁺ T_FH noted in our previous report²¹ was related to the development of effective CD8⁺ T cell responses, which appear to preferentially restrict LAV replication in the T cell zone of the LN paracortex relative to the B cell follicles.

To further explore this possibility and ascertain its applicability to immune control of pathogenic WT SIV infection, we compared the frequency of cells harboring replication-competent SIV⁺ and levels of SIV RNA and DNA in CD4⁺ T_FH vs. non-T_FH compartments of LNs from 27 rhesus monkeys chronically infected with WT SIVmac239 or SIVmac251, including ten EC (plasma viral set point ≤600 copies/ml), seven semi-controllers (plasma viral set point between 1,900 and 35,000 copies/ml), and ten typical progressors (plasma viral set points ≥85,000 copies/ml) (Suppl. Table 1). The frequency of T_FH within the overall LN CD4⁺ memory T cell populations of these monkeys reflected their virologic status with EC monkeys manifesting median T_FH frequencies of 5% (only slightly higher than uninfected monkeys), semi-controllers 10%, and non-controllers 27% (Fig. 2a,b). Strikingly, and in keeping with the findings in LAV-infected monkeys, monkeys with elite control of WT SIV showed exquisite restriction of replication-competent virus to their very minor population of LN CD4⁺ T_FH (Fig. 2c,d). Of note, levels of replication-competent SIV were similarly low in both the PD-1⁻/CD200⁻ vs. PD-1^dim/CD200^dim CD4⁺ memory T cell subsets, suggesting that the ability to host productive SIV infection in EC LN is associated with full T_FH differentiation (e.g., a PD-1^high/CD200^high phenotype) and physical residence in the B cell follicle (see below). Similar restricted localization of replication-competent SIV was observed in splenic CD4⁺ T_FH vs. non-T_FH subsets (Suppl. Fig. 4), consistent with restricted viral replication within CD4⁺ T_FH in all secondary lymphoid tissues in these EC monkeys. In contrast, replication-competent SIV was identified at similar levels in CD4⁺ T_FH vs. non-T_FH memory T cell subsets in both LNs and spleen from monkeys with progressive infection, whereas semi-controller monkeys were intermediate – showing a variable degree of preferential, but not exclusive, localization of replication-competent SIV in CD4⁺ T_FH vs. non-T_FH in LNs and spleen (Fig. 2c,d; Suppl. Fig. 4). Quantification of cell-associated SIV RNA in the CD4⁺ T_FH vs. non-T_FH memory T cell subsets from these same monkeys confirmed these results, showing significantly higher SIV RNA content in CD4⁺ T_FH vs. non-T_FH in EC and in semi-controllers, but no difference in the SIV RNA content in CD4⁺ T_FH vs. non-T_FH in typical progressors (Fig. 2e). These data demonstrate restriction of productive SIV to CD4⁺ T_FH in the presence of immune control, but not its absence. Importantly, levels of cell-associated SIV DNA were only very modestly enriched in CD4⁺ T_FH vs. non-T_FH in both EC and semi-controllers (Fig. 2f), suggesting that the mechanism maintaining preferential T_FH localization of replication-competent SIV operates primarily, if not exclusively, on productively SIV-infected (SIV antigen-expressing) CD4⁺ T cells, as would be expected for a CD8⁺ T cell-mediated process.

Although the phenotypic criteria used to classify and sort CD4⁺ T_FH vs. non-T_FH have been anatomically validated (e.g., PD-1^high/CD200^high CD4⁺ T cells are exquisitely restricted to B cell follicles; Suppl. Fig. 2), we confirmed the anatomic restriction of productive SIV infection to B cell follicles structures in EC LNs using RNAscope in situ hybridization and combined, conventional in situ hybridization/immune fluorescence analysis (Fig. 3; Suppl. Figs. 5 and 6). Productively SIV-infected T_FH were differentiated from follicular dendritic cell-bound virions by well-described morphologic criteria^31,35 . Consistent with the cell sorting-based assays, we found that a median of 95% of productively SIV-infected cells within the LNs of EC monkeys were localized in B cell follicles (from which CD8⁺ T cells were largely excluded), whereas in progressors, over half of productively SIV-infected cells were found in the CD8⁺ T cell-rich paracortex (the latter observation in agreement with our previous study³¹ ). Thus, using 3 different assays of productive SIV infection – i) frequencies of cells with replication-competent SIV within sorted cell populations, ii) levels of SIV RNA vs. DNA within sorted cell populations, and iii) anatomic distribution of SIV RNA⁺ cells by in situ analysis – we have shown a striking association between immune control of infection (as reflected by plasma viral set points) and restriction of productive SIV infection to CD4⁺ T_FH.

CD8⁺ T cells restrict productive SIV infection to T_FH in EC

Taken together, these data suggest that the potent SIV-specific CD8⁺ T cell responses responsible for elite SIV control in rhesus monkeys effectively clear productive SIV infection from the T cell zones of secondary lymphoid tissues, but are unable to completely suppress SIV replication in the CD4⁺ T_FH within B cell follicles. To further test this hypothesis, we monitored the distribution of both replication-competent SIV and SIV RNA and DNA in seven EC monkeys (Suppl. Table 1) following transient in vivo CD8⁺ lymphocyte depletion with the anti-CD8α mAb M-T807R1 (Fig. 4). As expected, mAb M-T807R1 administration completely depleted CD8⁺ T cells, including SIV-specific CD8⁺ T cells, from blood and LN by day 10 post-treatment, with partial recovery by day 21 and near maximal recovery by day 35 (Figs. 4a,b). SIV replication dramatically increased and then decreased in close inverse association with CD8⁺ T cell numbers: plasma viral loads rebounded by nearly 4 logs at the CD8⁺ T cell nadir (day 10) before being brought back under nearly complete control by day 35 (Fig. 4a). Although NK cells are also depleted by mAb M-T807R1 treatment, these cells are very few in number in LN (Fig. 4c), and in separate studies, we have noted that their efficient depletion by IL-15 blockade is not associated with enhanced SIV replication (Okoye and Picker, unpublished observation), observations that together strongly suggest that the rebound in SIV replication is related to loss of SIV-specific CD8⁺ T cells rather than NK cells.

Prior to CD8⁺ lymphocyte depletion, replication-competent SIV was, as described above, highly restricted to the T_FH CD4⁺ subset and CD4⁺ T cell-associated SIV RNA (but not SIV DNA) was significantly higher in CD4⁺ T_FH vs. non-T_FH in EC secondary lymphoid tissues (Fig. 4d–f; Suppl. Fig. 4). However, at day 10 post-CD8⁺ lymphocyte depletion, replication-competent SIV and cell-associated SIV RNA levels increased dramatically in CD4⁺ non-T_FH (~2 logs for cell-associated SIV RNA) whereas cell-associated SIV DNA levels increased only modestly (~1 log), findings indicating a marked enhancement of productive infection within these cells. CD8⁺ T cell recovery was associated with a progressive decline in the levels of both replication-competent SIV and cell-associated SIV RNA in the non-T_FH CD4⁺ memory T cells, with pre-depletion levels re-established in most monkeys by day 35 post-depletion and all monkeys by day 84–136 post-depletion (Fig. 4d–f; Suppl. Fig. 7). Although CD8⁺ lymphocyte depletion also transiently increased levels of cell-associated SIV RNA in CD4⁺ T_FH, the net effect of CD8⁺ T cell removal was elimination or marked reduction of the pre-depletion disparity in levels of productive SIV infection between CD4⁺ T_FH and non-T_FH – the loss of CD8⁺ T cells having the effect of transiently "converting" EC to progressors in terms of the distribution of productive infection between T_FH vs. non-T_FH memory CD4⁺ T cells. The transient equalization of productive SIV infection between CD4⁺ T_FH and non-T_FH after the CD8⁺ lymphocyte depletion was also observed within the spleen (Suppl. Fig. 7), suggesting this was a general effect in all secondary lymphoid tissues.

CD8⁺ lymphocyte depletion has secondary effects on CD4⁺ memory T cell homeostasis, in particular the enhancement of CD4⁺ memory T cell proliferation³⁶ . To determine whether induction of homeostatic proliferation of non-T_FH may have played a role in the equalization of replication-competent SIV levels in CD4⁺ T_FH and non-T_FH, we determined the effect of artificially inducing non-T_FH CD4⁺ memory T cell proliferation using a non-CD8⁺ lymphocyte-depleting mechanism – IL-7 administration³⁷ – on the distribution of replication-competent SIV within LN CD4⁺ T cell compartments of EC monkeys. Administration of two doses of recombinant IL-7 (30 μg/kg) induced levels of proliferation (%Ki-67⁺) among non-T_FH CD4⁺ T cells comparable to those induced by CD8⁺ lymphocyte depletion (Fig. 5a). However, this treatment had no effect on the plasma viral load or, most importantly, on the highly restricted localization of replication-competent SIV in T_FH in these EC monkeys (Fig. 5b,c). Taken together, these data support a CD8⁺ T cell-mediated mechanism for the transient equalization of productive SIV infection between CD4⁺ T_FH and non-T_FH after CD8⁺ lymphocyte depletion of EC monkeys, and in these monkeys, strongly point to a differential ability of their highly effective SIV-specific CD8⁺ T cells to clear productive SIV infection of CD4⁺ memory T cells within the T cell zones vs. B follicles of their secondary lymphoid tissues.

T_FH preferentially harbor productive SIV infection after cART

The apparent ability of B follicular structures to substantially shield productively SIV-infected CD4⁺ T_FH from CD8⁺ T cell-mediated clearance raises the question of whether this mechanism might counter the ability of spontaneous or therapeutic vaccine-induced CD8⁺ T cell responses to eradicate or functionally cure HIV/SIV infection in the setting of effective pharmacologic viral suppression⁶ . Given the data described above, we would predict that if viral reactivation or residual viral replication occurs in CD4⁺ T_FH within the secondary lymphoid tissues of individuals on cART, it would be relatively protected from clearance by such CD8⁺ T cell responses. Moreover, it is possible that even the conventional (less effective than EC) virus-specific CD8⁺ T cell responses present in typical individuals on cART might be able to generate sufficient CD8⁺ T cell-mediated immune pressure to restrict productive HIV/SIV infection arising from reactivation or residual replication in these individuals to the protected environment of the B cell follicle. To determine whether SIV reactivation or residual replication within CD4⁺ memory T cells was differentially distributed between the extra-follicular T cell zone and B cell follicles in the setting of pharmacologically suppressed SIV infection, we determined the levels of SIV RNA and DNA within CD4⁺ T_FH and non-T_FH cells in two different cohorts of monkeys with progressive or semi-controlled SIV infection that were treated with cART for at least four months, all with plasma viral loads at the time of testing ≤60 copies/ml (Suppl. Table 2; note that the level of productive SIV infection in these cART-suppressed monkeys was below the level of detection of our co-culture assay). In Cohort #1, LN and splenic CD4⁺ T_FH and non-T_FH were sorted on the basis of PD-1 and CD200 expression, as described above; in Cohort #2, LN CD4⁺ T_FH were sorted on the basis of PD-1 and CXCR5 expression, and the CD4⁺ non-T_FH were separated into the traditional central memory (T_CM) and transitional effector memory (T_Tr/EM) subsets on the basis of CD28, CCR5, and CCR7 expression³ (Suppl. Fig. 8). As shown in Fig. 6a–c, cell-associated SIV RNA, but not SIV DNA, was significantly higher in CD4⁺ T_FH, compared to non-T_FH in both cohorts, consistent with preferential localization of residual productive SIV infection within the B cell follicles of secondary lymphoid tissues in the majority of cART-treated monkeys. Although T_FH constitute only a median of 12% of total CD4⁺ memory T cells within LNs of these cART-suppressed monkeys (Fig. 2a,b), these data indicate that a substantial fraction of viral reactivation or residual viral replication in effectively cART-treated monkeys occurs with the protected environment of the B cell follicle.

Animals and viruses

This study used 50 purpose-bred adult rhesus macaques (Macaca mulatta) of Indian genetic background (39 males/11 females; age range: 4–17 years), all of which were free of cercopithicine herpesvirus 1, D type simian retrovirus, and simian T-lymphotrophic virus type 1 at the start of the study. We studied 46 monkeys at the Oregon National Primate Research Center with the approval of the Oregon National Primate Research Center Institutional Animal Care and Use Committee, under the standards of the NIH Guide for the Care and Use of Laboratory Animals. We also studied four rhesus macaques (SIV-infected progressors) that were housed at the National Institutes of Health in accordance with the same standards under the approval by the Institutional Animal Care and Use Committee of the National Cancer Institute. All rhesus macaques were infected with SIVmac239, SIVmac251, or with LAV (SIVmac239Δnef) (see Suppl. Table 1). We CD8⁺ lymphocyte-depleted seven EC monkeys in chronic phase of WT SIV infection (191–967 days post-infection) with anti-CD8 monoclonal antibody (mAb) M-T807R1 (National Institutes of Health Nonhuman Primate Reagent Resource Program, Boston, MA.)³⁶ . M-T807R1 was injected subcutaneously (10 mg/kg) at day 0, and intravenously (5 mg/kg) at days 3, 7, and 10. After the recovery of CD8⁺ lymphocyte depletion (5–6 months after the antibody treatment), we administered two subcutaneous injections of recombinant simian IL-7 (30 μg/kg each, Cytheris) at days 0 and 7 as described³⁷ to three of the previously CD8⁺ cell-depleted monkeys (Rh25610, Rh25687, Rh26623). We used the cART regimen described below to suppress SIV replication in 17 WT SIV-infected monkeys (see Suppl. Table 2).

Cell processing

Mononuclear cell preparations were obtained from blood, LN, and spleen, as previously described^21,45 . CD4⁺ memory T cell subsets were isolated from LN or spleen by sorting on a FACS Aria II (BD Biosciences) on the basis of cell surface staining patterns (lineage markers CD3 and CD4, memory markers CD95 and CD28, and T_FH markers PD-1 and CD200) after excluding dead cells (LiVEDEAD® Fixable Aqua Dead Cell Stain Kits, Invitrogen), as shown in Suppl. Fig. 1. In some experiments, CD4⁺ memory T cell subset (T_CM, T_Tr/EM, and T_FH) cells were sorted as shown in Suppl. Fig. 8. To determine the phenotype of T cells and to sort CD4⁺ memory T cell subsets, these antibodies were used; CD3 (SP34-2, BD), CD4 (L200, BD), CD8 (DK25, Dako), CD8a (SK1, BD), CD200 (OX104, BioLegend), CD95 (DX2, eBioscience), CD28 (28.2, BD), PD-1 (J105, eBioscience), CXCR5 (MU5UBEE, eBioscience), CCR7 (15053, R&D Systems), and CCR5 (3A9, BD).

Viral quantification

Replication-competent SIV was rescued from graded numbers of sorted CD4⁺ memory T cell subsets (10³, 10⁴, or 10⁵ cells) by the co-cultivation of these sorted cells with CEMx174 cells (10⁵ cells, NIH AIDS Reagent Program) in 24-well plates, followed by flow cytometric analysis of intracellular SIV-Gag p27 expression [with anti-CD3 (SP34-2) and anti-CD4 (L200), anti-Gag p27 (55-2F12, NIH AIDS Research and Reference Reagent Program) antibodies were used to detect viral replication in CEMx174], as described²¹ . Co-cultured cells were harvested and analyzed at days 13 to 36, with quantitative comparisons of the extent of infection in co-cultures containing different CD4⁺ memory T cell populations performed at the earliest time point of maximum expression of Gag p27 in the co-cultures of any of the three CD4⁺ memory T cell subsets. For cell-associated viral load determination, sorted CD4⁺ memory T cell subsets from LN were pelleted and frozen immediately and stored at −80 °C until use for quantitative PCR/RT-PCR. Quantitative PCR and RT-PCR assays targeting a highly conserved nucleic acid sequence in gag were used for standard measurement of plasma SIV RNA and cell-associated SIV RNA/DNA in sorted CD4⁺ memory T cell subsets isolated from LN, as previously described (with cell-associated SIV RNA/DNA normalized to co-determined cell equivalents based on quantitation of 2 copies of CCR5 sequence per cell)^46,47 .

SIV-specific T cell response assays

SIV-specific CD8⁺ T cells were quantified in mononuclear cells isolated from blood and LN by flow cytometric intracellular cytokine analysis, as previously described in detail⁴⁵ . Briefly, mixes of sequential (11 amino acid overlapping) 15-mer peptides (AnaSpec) spanning the SIVmac239 Gag, Env, Pol, Nef, Rev/Tat, Vif/Vpr/Vpx proteins were used as antigens in conjunction with anti-CD28 (BD Biosciences) and anti-CD49d co-stimulatory mAb (BD Biosciences). Cells were incubated at 37 °C with peptide mixes/pools and co-stimulatory antibodies for 1 hr, followed by an additional 8 hr. incubation in the presence of Brefeldin A (5 μg/ml, Sigma-Aldrich). Co-stimulation in the absence of peptides served as background control. Cells were stained with fluorochrome-conjugated monoclonal antibodies [CD3 (SP34-2), CD4 (L200), CD8a (SK1), IFN-γ (B27)(BD Biosciences), TNF (MAB11, eBioscience), CD69 (FN50, BD Biosciences)], and data were collected on an LSRII (BD Biosciences) and analyzed using the FlowJo software program (version 8.8.7; Tree Star, Inc.). Response frequencies to each peptide mix were defined by the intracellular expression of TNF and/or IFN-γ by CD3⁺, CD4⁻, CD8⁺ T cells after subtraction of background. Response frequencies to each peptide mix were added to obtain total SIV-specific response frequencies.

Immunofluorescence confocal microscopy

Tissue sections (5 μm) were mounted on Superfrost Plus Microscope Slides (Fisher Scientific), heated at 60°C for 1h, dewaxed in xylenes and rehydrated in graded ethanols and double-distilled H₂O. Heat-induced epitope retrieval was performed by heating sections in 0.01% citraconic anhydride containing 0.05% Tween-20 in a pressure cooker set at 122 °C for 30 sec. Tissues were blocked in blocking buffer [TBS containing 0.05% Tween-20 (TBS-Tw) and 0.25% casein] for 10 min at room temperature and then incubated with mouse anti-CD4 (clone 1F6; 1:25; Vector Labs), goat anti-PD1 (1:200; R&D Systems; AF1086) and rabbit anti-CD200 (1:200; Sigma; HPA031149) diluted in blocking buffer overnight at room temperature. Slides were washed in TBS-Tw and incubated with donkey anti-mouse-Alexa488, donkey anti-goat–Alexa647, and donkey anti-rabbit–Alexa594 (all Invitrogen; 1:400) for 1h at room temperature in the dark and washed in TBS-Tw. Slides were incubated with 0.1% Sudan Black B in 70% ethanol (Cat. No. 4410; ENG Scientific) for 30 min at room temperature to quench autofluorescence then incubated with 300 nM DAPI for 10 min. Slides were washed, mounted with Prolong® Gold (Invitrogen) and imaged on an Olympus FV10i confocal microscope using a 10x phase contrast objective (NA 0.4) with 2x Optical zoom using sequential mode to separately capture the fluorescence from the different fluorochromes at an image resolution of 1024x1024 pixels.

Fluorescent immunohistochemistry (FIHC) and fluorescent in situ hybridization (FISH)

SIV RNA⁺, CD8⁺, and CD20⁺ cells were detected in 4% PFA-fixed, paraffin-embedded LN sections by a combination of FIHC and FISH assays with a previously described modified method⁴⁸ . Briefly, the sections were treated with methanol-hydrogen peroxide then hybridized overnight at 45 °C with antisense SIVmac239 digoxigenin-UTP-labeled riboprobe. The hybridized sections were blocked with 3% normal sheep (Jackson ImmunoResearch Laboratories) and horse serum (Sigma-Aldrich) in 0.1 M Tris, pH7.4, and then incubated with sheep anti-digoxigenin-horseradish peroxidase (SAD-POD; Roche Applied Science). SAD-POD was detected by a fluorescent tyramide signal amplification technique (TSA Plus Fluorescein Kit; PerkinElmer). After completion of the FISH assay, the sections were incubated in mouse anti-human CD20 monoclonal antibody (Clone L26; Thermo Scientific), followed by incubation with goat anti-mouse IgG-Biotinylated antibody (Vector Laboratories) and then stained with Streptavidin Alexa-647 conjugate (Life Technologies). The double-stained sections were then incubated with rabbit anti-human CD8 monoclonal antibody (Clone SP16; Thermo Scientific), followed by Alexa 594-conjugated goat anti-rabbit IgG antibody (Life Technologies). The triple-stained sections were mounted in ProLong Gold antifade reagent (Life Technologies). All confocal images were acquired using a Leica TCS SP8 confocal microscope (Leica Microsystems) equipped with a 20x/NA 0.75 oil immersion objective. Large field images of the tissue sections were generated by assembly of numerous juxtaposed individual sub-images into a montage, assisted by Fiji software (http://fiji.sc/wiki/index.php/Fiji).

RNAscope analysis

We also utilized a novel next-generation, ultrasensitive RNA in situ hybridization technology, RNAscope, for SIV in situ hybridization, as previously described⁴⁹ . Quantification of the number of productively infected SIV RNA⁺ cells either within or outside follicles was performed by manual counting of high magnification whole Aperio scanned tissue sections using the Aperio elliptical annotation tool independently by three blinded individuals. Productively infected SIV RNA⁺ cells were classified based on an intense centric SIV RNA signal within a single cell (both nuclear and cytoplasmic) that diffused toward the edges of the cell, which is distinguished from the "lattice-like" filamentous pattern of extracellular FDC-bound virus that is seen within the follicles³¹ . The average fraction of productively infected SIV RNA⁺ cells within the B cell follicles is reported for EC vs. progressors (total productively infected SIV RNA⁺ cells counted: 245 for EC; 1,252 for progressors).

Combination antiretroviral therapy (cART)

For Cohort 1 (Supplementary Table 2), cART consisted of subcutaneously injected 20mg/kg^–1 d^–1 tenofovir (TFV) and 50 mg/kg^–1 d^–1 emtricitabine (FTC), and orally administered (given in food) 240 mg/kg^–1 d^–1 raltegravir and 600 mg darunavir twice daily boosted with 100 mg ritonavir twice daily. All Cohort 1 monkeys had been chronically infected with SIVmac251 at the time of cART initiation and were treated with the cART for a minimum of 18 weeks. For Cohort 2, cART consisted of 20 mg/kg^–1 d^–1 TFV, 50 mg/kg^–1 d^–1 FTC, 2.5 mg/kg^–1 d^–1 dolutegravir (DTG; all given as a single subcutaneous injection) and 600?mg darunavir twice daily boosted with 100?mg ritonavir twice daily (given orally in food). Daily darunavir and ritonavir (Oregon Health & Science University pharmacy) were stopped after 96 days. All Cohort #2 monkeys had been infected with SIVmac239 prior to ART initiation at day 42 post-infection, and treated with cART for 26 weeks at the time of biopsy. The injectable component of the cART regimen was composed of 20 mg/mL TFV, 50 mg/mL FTC, and 2.5 mg/mL DTG in a solvent containing 25% (v/v) polyethylene glycol 400 (PEG-400), 15% (w/v) captisol and 0.075 N sodium hydroxide (NaOH) in water. The injectable cART formulation was prepared by mixing DTG stock solution (10 mg/mL in PEG-400), TFV stock solution (80 mg/mL in 0.3 N NaOH), FTC stock solution (200 mg/mL in 0.3 N NaOH), and 30% (w/w) captisol solution at a 1:1:2 (v:v:v) ratio. The final solution had a pH ~6, was sterile-filtered, aliquoted into sterile glass vials and frozen at −20°C until used. This injectable component of the cART regimen was administered subcutaneously once daily at 1 mL/kg body weight.

Statistical analysis

We conducted all statistical analyses without multiplicity correction, using nonparametric, two-sided tests⁵⁰ . Study animals were selected for inclusion in the study based on pre-specified criteria and these selected animals contributed data to all analyses. The only exceptions to this were instances when insufficient sorted cells could be recovered from available tissue samples, resulting in some variability in the group size across the various analyses. Because the animal groups were defined by controller status or ART status, no randomization was used, and no blinding was performed. To compare cell-associated viral DNA, RNA, or percentage of SIV-Gag p27 in CEMx174, within-animal differences across CD4⁺ memory subsets based on PD-1 and CD200 expression (PD-1⁻/CD200⁻, PD-1^dim/CD200^{− to dim}, PD-1^high/CD200^high; Suppl. Fig. 1) or other differentiation markers (T_CM, T_Tr/EM, T_FH; Suppl. Fig. 8), we conducted Friedman tests (non-parametric randomized block ANOVA) at the 0.05 significance level. For pairwise comparisons between categories of within-animal differences, we applied the Wilcoxon signed-rank test (at the 0.05 level). For comparisons across groups of animals we applied 0.05-level Kruskal-Wallis and Wilcoxon rank-sum tests (for >2 and 2 groups, respectively). We used the R statistical computing language for all statistical analyses⁵¹ .

• 1.Kirchhoff F. Immune evasion and counteraction of restriction factors by HIV-1 and other primate lentiviruses. Cell Host Microbe. 2010;8:55–67. doi: 10.1016/j.chom.201006004. [DOI] [PubMed] [Google Scholar]
• 2.Picker LJ, Hansen SG, Lifson JD. New paradigms for HIV/AIDS vaccine development. Annu Rev Med. 2012;63:95–111. doi: 10.1146/annurev-med-042010-085643. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 3.Okoye AA, Picker LJ. CD4(+) T-cell depletion in HIV infection: mechanisms of immunological failure. Immunol Rev. 2013;254:54–64. doi: 10.1111/imr.12066. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 4.Deeks SG, Walker BD. Human immunodeficiency virus controllers: mechanisms of durable virus control in the absence of antiretroviral therapy. Immunity. 2007;27:406–416. doi: 10.1016/j.immuni.2007年08月01日0. [DOI] [PubMed] [Google Scholar]
• 5.Mudd PA, Watkins DI. Understanding animal models of elite control: windows on effective immune responses against immunodeficiency viruses. Curr Opin HIV AIDS. 2011;6:197–201. doi: 10.1097/COH.0b013e3283453e16. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 6.Katlama C, et al. Barriers to a cure for HIV: new ways to target and eradicate HIV-1 reservoirs. Lancet. 2013;381:2109–2117. doi: 10.1016/S0140-6736(13)60104-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 7.Hatano H, et al. Evidence for persistent low-level viremia in individuals who control human immunodeficiency virus in the absence of antiretroviral therapy. J Virol. 2009;83:329–335. doi: 10.1128/JVI.01763-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 8.Pereyra F, et al. Persistent low-level viremia in HIV-1 elite controllers and relationship to immunologic parameters. J Infect Dis. 2009;200:984–990. doi: 10.1086/605446. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 9.Bailey JR, Williams TM, Siliciano RF, Blankson JN. Maintenance of viral suppression in HIV-1-infected HLA-B*57+ elite suppressors despite CTL escape mutations. J Exp Med. 2006;203:1357–1369. doi: 10.1084/jem.20052319. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 10.Chun TW, et al. Effect of antiretroviral therapy on HIV reservoirs in elite controllers. J Infect Dis. 2013;208:1443–1447. doi: 10.1093/infdis/jit306. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 11.Mens H, et al. HIV-1 continues to replicate and evolve in patients with natural control of HIV infection. J Virol. 2010;84:12971–12981. doi: 10.1128/JVI.00387-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 12.O’Connell KA, et al. Control of HIV-1 in elite suppressors despite ongoing replication and evolution in plasma virus. J Virol. 2010;84:7018–7028. doi: 10.1128/JVI.00548-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 13.Hunt PW, et al. Relationship between T cell activation and CD4+ T cell count in HIV-seropositive individuals with undetectable plasma HIV RNA levels in the absence of therapy. J Infect Dis. 2008;197:126–133. doi: 10.1086/524143. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 14.Hatano H, et al. Prospective antiretroviral treatment of asymptomatic, HIV-1 infected controllers. PLoS Pathog. 2013;9:e1003691. doi: 10.1371/journal.ppat.1003691. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 15.Walker BD, Yu XG. Unravelling the mechanisms of durable control of HIV-1. Nat Rev Immunol. 2013;13:487–498. doi: 10.1038/nri3478. [DOI] [PubMed] [Google Scholar]
• 16.Saez-Cirion A, Pancino G. HIV controllers: a genetically determined or inducible phenotype? Immunol Rev. 2013;254:281–294. doi: 10.1111/imr.12076. [DOI] [PubMed] [Google Scholar]
• 17.Mendoza D, et al. Cytotoxic capacity of SIV-specific CD8(+) T cells against primary autologous targets correlates with immune control in SIV-infected rhesus macaques. PLoS Pathog. 2013;9:e1003195. doi: 10.1371/journal.ppat.1003195. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 18.Mudd PA, et al. Reduction of CD4+ T cells in vivo does not affect virus load in macaque elite controllers. J Virol. 2011;85:7454–7459. doi: 10.1128/JVI.00738-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 19.Friedrich TC, et al. Subdominant CD8+ T-cell responses are involved in durable control of AIDS virus replication. J Virol. 2007;81:3465–3476. doi: 10.1128/JVI.02392-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 20.Blankson JN, et al. Isolation and characterization of replication-competent human immunodeficiency virus type 1 from a subset of elite suppressors. J Virol. 2007;81:2508–2518. doi: 10.1128/JVI.02165-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 21.Fukazawa Y, et al. Lymph node T cell responses predict the efficacy of live attenuated SIV vaccines. Nat Med. 2012;18:1673–1681. doi: 10.1038/nm.2934. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 22.Crotty S. Follicular helper CD4 T cells (TFH) Ann Rev Immunol. 2011;29:621–663. doi: 10.1146/annurev-immunol-031210-101400. [DOI] [PubMed] [Google Scholar]
• 23.Vinuesa CG, Cyster JG. How T cells earn the follicular rite of passage. Immunity. 2011;35:671–680. doi: 10.1016/j.immuni.201111001. [DOI] [PubMed] [Google Scholar]
• 24.Connick E, et al. CTL fail to accumulate at sites of HIV-1 replication in lymphoid tissue. J Immunol. 2007;178:6975–6983. doi: 10.4049/jimmunol.178.11.6975. [DOI] [PubMed] [Google Scholar]
• 25.Tjernlund A, et al. In situ detection of Gag-specific CD8+ cells in the GI tract of SIV infected Rhesus macaques. Retrovirology. 2010;7:12. doi: 10.1186/1742-4690年7月12日. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 26.Sasikala-Appukuttan AK, et al. Location and dynamics of the immunodominant CD8 T cell response to SIVDeltanef immunization and SIVmac251 vaginal challenge. PloS one. 2013;8:e81623. doi: 10.1371/journal.pone.0081623. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 27.Robinson HL, Amara RR. Protective immunity from a germinal center sanctuary. Nat Med. 2012;18:1614–1616. doi: 10.1038/nm.2986. [DOI] [PubMed] [Google Scholar]
• 28.Skinner PJ, Connick E. Overcoming the immune priviledge of B-cell follicles to cure HIV-1 infection. J Hum Virol Retrovirol. 2014;1(1):00001. doi: 10.15406/jhvrv.20140100001. [DOI] [Google Scholar]
• 29.Perreau M, et al. Follicular helper T cells serve as the major CD4 T cell compartment for HIV-1 infection, replication, and production. J Exp Med. 2013;210:143–156. doi: 10.1084/jem.20121932. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 30.Lindqvist M, et al. Expansion of HIV-specific T follicular helper cells in chronic HIV infection. J Clin Invest. 2012;122:3271–3280. doi: 10.1172/JCI64314. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 31.Brenchley JM, et al. Differential infection patterns of CD4+ T cells and lymphoid tissue viral burden distinguish progressive and nonprogressive lentiviral infections. Blood. 2012;120:4172–4181. doi: 10.1182/blood-2012-06-437608. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 32.Petrovas C, et al. CD4 T follicular helper cell dynamics during SIV infection. J Clin Invest. 2012;122:3281–3294. doi: 10.1172/JCI63039. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 33.Klatt NR, et al. SIV infection of rhesus macaques results in dysfunctional T- and B-cell responses to neo and recall Leishmania major vaccination. Blood. 2011;118:5803–5812. doi: 10.1182/blood-2011-07-365874. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 34.Chtanova T, et al. T follicular helper cells express a distinctive transcriptional profile, reflecting their role as non-Th1/Th2 effector cells that provide help for B cells. J Immunol. 2004;173:68–78. doi: 10.4049/jimmunol.173.1.68. [DOI] [PubMed] [Google Scholar]
• 35.Estes JD. Pathobiology of HIV/SIV-associated changes in secondary lymphoid tissues. Immunol Rev. 2013;254:65–77. doi: 10.1111/imr.12070. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 36.Okoye A, et al. Profound CD4+/CCR5+ T cell expansion is induced by CD8+ lymphocyte depletion but does not account for accelerated SIV pathogenesis. J Exp Med. 2009;206:1575–1588. doi: 10.1084/jem.20090356. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 37.Leone A, et al. Increased CD4+ T cell levels during IL-7 administration of antiretroviral therapy-treated simian immunodeficiency virus-positive macaques are not dependent on strong proliferative responses. J Immunol. 2010;185:1650–1659. doi: 10.4049/jimmunol.0902626. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 38.Quigley MF, Gonzalez VD, Granath A, Andersson J, Sandberg JK. CXCR5+ CCR7− CD8 T cells are early effector memory cells that infiltrate tonsil B cell follicles. Eur J Immunol. 2007;37:3352–3362. doi: 10.1002/eji.200636746. [DOI] [PubMed] [Google Scholar]
• 39.Wood GS, Garcia CF, Dorfman RF, Warnke RA. The immunohistology of follicle lysis in lymph node biopsies from homosexual men. Blood. 1985;66:1092–1097. [PubMed] [Google Scholar]
• 40.Tenner-Racz K, et al. Monoclonal antibodies to human immunodeficiency virus: their relation to the patterns of lymph node changes in persistent generalized lymphadenopathy and AIDS. AIDS. 1987;1:95–104. [PubMed] [Google Scholar]
• 41.Tenner-Racz K, et al. Cytotoxic effector cell granules recognized by the monoclonal antibody TIA-1 are present in CD8+ lymphocytes in lymph nodes of human immunodeficiency virus-1-infected patients. Am J Pathol. 1993;142:1750–1758. [PMC free article] [PubMed] [Google Scholar]
• 42.Petrovas C, et al. A population of CD8+ T cells located in germinal centers that is functionally capable of mediating bispecific antibody killing of HIV-infected T cells. 20th International AIDS Conference; Melbourne, Australia. 2014. http://pag.aids2014.org/Abstracts.aspx?SID=1147&AID=5387. [Google Scholar]
• 43.Shan L, et al. Stimulation of HIV-1-specific cytolytic T lymphocytes facilitates elimination of latent viral reservoir after virus reactivation. Immunity. 2012;36:491–501. doi: 10.1016/j.immuni.2012年01月01日4. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 44.Anolik JH, Aringer M. New treatments for SLE: cell-depleting and anti-cytokine therapies. Best Prac & Res Clin Rheum. 2005;19:859–878. doi: 10.1016/j.berh.200505006. [DOI] [PubMed] [Google Scholar]
• 45.Hansen SG, et al. Immune clearance of highly pathogenic SIV infection. Nature. 2013;502:100–104. doi: 10.1038/nature12519. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 46.Cline AN, Bess JW, Piatak M, Jr, Lifson JD. Highly sensitive SIV plasma viral load assay: practical considerations, realistic performance expectations, and application to reverse engineering of vaccines for AIDS. J Med Primatol. 2005;34:303–312. doi: 10.1111/j.1600-0684.2005.00128.x. [DOI] [PubMed] [Google Scholar]
• 47.Venneti S, et al. Longitudinal in vivo positron emission tomography imaging of infected and activated brain macrophages in a macaque model of human immunodeficiency virus encephalitis correlates with central and peripheral markers of encephalitis and areas of synaptic degeneration. Am J Pathol. 2008;172:1603–1616. doi: 10.2353/ajpath.2008.070967. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 48.Brown CR, et al. Unique pathology in simian immunodeficiency virus-infected rapid progressor macaques is consistent with a pathogenesis distinct from that of classical AIDS. J Virol. 2007;81:5594–5606. doi: 10.1128/JVI.00202-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 49.Smedley J, et al. Tracking the luminal exposure and lymphatic drainage pathways of intravaginal and intrarectal inocula used in nonhuman primate models of HIV transmission. PloS one. 2014;9:e92830. doi: 10.1371/journal.pone.0092830. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 50.Wolfe MHDA. Nonparametric Statistical Methods. John Wiley & Sons; New York: 1999. [Google Scholar]
• 51.R Core Team. R: a language and environment for statistical computing. R Foundation for Statistical Computing; Vienna, Austria: 2014. < http:/www.r-project.org/>. [Google Scholar]

Supplementary Materials