Gold drug auranofin restricts the viral reservoir in the monkey AIDS model and induces containment of viral load following ART suspension : AIDS

Secondary Logo

Journal Logo


Gold drug auranofin restricts the viral reservoir in the monkey AIDS model and induces containment of viral load following ART suspension

Lewis, Mark G.a,*; DaFonseca, Sandrinab,*; Chomont, Nicolasb; Palamara, Anna T.c,d; Tardugno, Mariae; Mai, Antonelloe; Collins, Matta; Wagner, Wendeline L.a; Yalley-Ogunro, Jakea; Greenhouse, Jacka; Chirullo, Barbaraf; Norelli, Sandrof; Garaci, Enricof; Savarino, Andreaf

Author Information
doi: 10.1097/QAD.0b013e328347bd77



Despite the potency of antiretroviral drugs to inhibit viral replication, current antiretroviral treatments do not eradicate HIV from the body [1,2]. Memory CD4+ T cells harboring a transcriptionally silent but replication-competent provirus play a major role in HIV persistence [3–5]. HIV primarily persists in long-lived central memory (TCM) and transitional memory (TTM) CD4+ T cells through T-cell survival and continuous low-level proliferation [6]. As opposed to short-lived effector CD4+ T cells (TEM), TCM and TTM cells express CD27 along with CD28 [7–9]. Both receptors share the capacity to promote survival of the dividing cells [10]. CD27 is particularly important for maintenance of T-cell memory [10–12] and differentiation into a Th1 (IFN-γ+) phenotype [13], although it does not seem to play a major role in maintenance of naive T cells (TN) and T-cell generation from thymic precursors [11].

Substitution of the CD4+CD27+CD28+ TCM and TTM cells by transplantation of HIV-resistant stem cells [14], removal or destruction of latently infected cells using the so-called ‘shock and kill’ strategies [15,16], and other approaches targeting self-renewal, or ‘stem cell-ness’ of memory T cells in association with antiretroviral therapy (ART) [6] have been proposed as potential strategies to eradicate HIV.

Here, we propose that compounds able to promote differentiation to short-lived phenotypes or death of long-lived latently infected TCM cells could be used to decrease the lifespan of the latently infected cells thus restricting the viral reservoir. This approach might also counteract the re-seeding of the HIV reservoir through cryptic viral replication by limiting the persistence of the newly infected cell pools.

The gold-based compound auranofin is an orally administrable drug adopted for treatment of rheumatoid arthritis [17], a disease characterized, among other aspects, by increased TCM cell pools in peripheral blood [18]. Auranofin has also been used in other immune-mediated diseases such as discoid lupus erythematosus [19]. It has a well known toxicity profile allowing prolonged treatment of regularly monitored individuals. Side-effects include diarrhea, proteinuria, and rarely thrombocytopenia/bone-marrow suppression [20]. The mechanism of action of auranofin is only partly understood: auranofin impairs the proliferative capacity of T lymphocytes in vitro[20,21]; decreases production of pro-inflammatory cytokines in macrophages and T cells [22,23]; induces apoptosis in the Jurkat T-cell line [24,25]. That auranofin affects T-cell proliferation/survival in vivo is shown by its successful use in Jessner's syndrome, a benign lymphoproliferative disorder [26]. Instead, T-cell regeneration is likely to be left unaffected, as shown in an HIV-1-infected individual with psoriatic arthritis treated with auranofin and displaying no decrease in absolute CD4 cell counts [27]. Finally, auranofin contributes to apoptosis and differentiation of leukemic cells in vitro in combination with anticancer therapies [28]. If similar effects could be translated to lentivirally infected cells, this drug might become a valuable candidate to target the viral reservoir. Here, we demonstrate that in-vivo exposure to auranofin reduces viral reservoir cells in SIVmac251-infected macaques treated with ART, and delays and attenuates the viral load rebound following therapy suspension.


Cell separation and isolation of CD4+ T-cell subsets by cell sorting, cell culture and flow cytometry

The protocol is described in Supplemental Digital Content 1,

Animal treatment

The Indian Rhesus macaques used in this study were housed at BIOQUAL, Inc., according to standards and guidelines as set forth in the Animal Welfare Act, The Guide for the Care and Use of Laboratory Animals, and the Association for the Assessment and Accreditation of Laboratory Animal Care (AAALAC), following approval by the Institutional Animal Care and Use Committee (IACUC). A schematic representation of the nonhuman primate studies is presented in the Figure 1 of Supplemental Digital Content 2, SIVmac251-infected nonhuman primates that had been kept under a regimen consisting of tenofovir/emtricitabine and raltegravir [29] and had been stably nonviremic for 8 weeks [i.e. <40 copies of viral RNA (vRNA)/ml], were administered auranofin (Prometheus Laboratories, San Diego, California, USA) by the oral route twice daily with food (a starting dose of 1.5 mg/kg per day in the first week followed by 2 mg/kg per day). ART was not stopped during the entire treatment period. Historical data demonstrated that the same macaques had had either progressively decreasing, or consistently subnormal CD4 cell counts before ART, clearly showing that they should not be regarded as long-term nonprogressors [29]. Tenofovir (PMPA) and emtricitabine (FTC) were kindly provided by Gilead Sciences (Foster City, California, USA). Raltegravir, darunavir and ritonavir were bought from the manufacturers. Suberoylanilide hydroxamic acid (SAHA) was synthesized following a previously published protocol [30]. The preparation obtained with this method is reagent-grade. Animals were dosed subcutaneously with tenofovir (20 mg/kg per day) and emtricitabine (50 mg/kg per day), and orally with raltegravir (100 mg b.i.d.), darunavir (375 mg b.i.d.) and ritonavir (50 mg b.i.d.).

Clinical laboratory analyses

Plasma vRNA levels, and cell-associated viral DNA (vDNA) were measured by a quantitative TaqMan PCR assay, following reverse transcription in case of vRNA measurement (Applied Biosystems, Foster City, California, USA) [29].

Hematological analyses were performed by IDEXX (IDEXX Preclinical Research, West Sacramento, California, USA). Calculation of CD4+ and CD8+ T-cell numbers is described in [29]. Samples were run on a FACSCalibur (BD Biosciences, San Jose, California, USA). Proportions of rhesus TN, TCM and TEM cells were determined by 6-color flow cytometry following a validated protocol described by Pitcher et al.[31].

Statistical analyses

Statistical analyses are detailed in the text of Supplemental Digital Content 3,


Auranofin induces differentiation of long-lived cells towards short-lived phenotypes ex vivo

We first evaluated the effect of auranofin on CD4+ T-cell viability at therapeutic concentrations [32] by measuring the percentage of CD4+ T cells becoming stainable with the cell death markers Annexin V and Vivid after exposure to auranofin for 48 h. Auranofin induced a dose-dependent increase in the frequency of Annexin V+ and Vivid+ cells (Fig. 1a and b). Since auranofin has been shown to induce both cell death and differentiation [28,33], we hypothesized that the induction of cell death was associated with the pro-differentiating effect of the drug. We found that auranofin down-modulated CD27 at the cell surface (Fig. 1c), and this down-modulation was accompanied by an increase in the frequency of Annexin V+ cells (Fig. 1d). Importantly, no such down-modulation was observed when measuring differentiation stage-independent antigens such CD3 or CD4 (data not shown). Altogether, our findings indicate that auranofin induces CD4+ T-cell differentiation to short-lived phenotypes, and that this differentiation is associated with cell death.

Fig. 1:
Auranofin induces cell death and CD27 down-modulation in primary human CD4+ T cells.(a, b) Effect of auranofin on human CD4+ T-cell viability at day 2 of incubation (gated on total CD4+ T cells). Cell death was estimated measured by stainability with AnnexinV and Vivid. (c) Dose-dependent down-modulation of CD27 by auranofin in total CD4+ T cells. In (a–c), the data points represent the means from four independent experiments. The dotted lines mark the 95% confidence limits of the regression lines. The table shows the P values for the correlations between the different parameters analyzed, as obtained by Spearman's multiple correlation analysis. (d) Dot plot displaying CD27 expression (%) and frequency of Annexin V+ cells in CD4+ T cells after a 48-h incubation in the presence or absence of auranofin (gated on live and Annexin V+ cells). One experiment out of three showing similar results. The P value shown below the panel refers to the dose-dependent increase in CD27low/−Annexin V+ cells, as obtained by linear regression.

To confirm the pro-differentiating effect of auranofin, we sorted the CD4+ T-naive (TN; i.e. CD45RA+CD27+CCR7+), TCM (CD45RA-CD27+CCR7+), TTM (CD45RACD27+CCR7) and TEM (CD45RACD27CCR7) subpopulations from enriched human CD4+ T-cell fractions, and treated them with auranofin (10–500 nmol/l). Phenotype analysis conducted after 2 days of incubation showed that the auranofin-treated cell subpopulations abandoned more promptly their original phenotype when compared to matched untreated controls (see Fig. 2, Supplemental Digital Content 2,, for one representative experiment). A proportion of the TCM and TTM cells progressed toward a TEM phenotype, but another consistent proportion, though displaying CD27 down-modulation, also expressed the CCR7 antigen (see Fig. 2, Supplemental Digital Content 2, CD27low/−CCR7+ cells represent an as yet poorly understood phenotype that was reported in cell cultures after ex-vivo T-cell activation and in ileum biopsies from HIV-1-infected individuals [6,34]. The prevalence of this phenotype in the resulting phenotypes of auranofin-treated TN and TCM cells supports the idea that auranofin specifically interferes with a pathway involving CD27 down-modulation. Induction of the CD27low/− phenotypes preceded an increased trend to cell death, especially in the memory compartment, as shown by time-course experiments (see Fig. 3, Supplemental Digital Content 2, In latently infected primary CD4+ T cells isolated from successfully ART-treated HIV-1-infected individuals, auranofin reproduced the pro-differentiating effect (data not shown), but did not induce viral reactivation. We conclude that auranofin decreases the lifespan of long-lived cells ex vivo and that this decreased lifespan is marked by partial or total CD27 down-modulation.

Pilot study of auranofin and antiretroviral therapy in SIVmac251-infected macaques

Because auranofin induced the differentiation and death of CD4+ T-cell subpopulations that encompass the viral reservoir in humans, we conducted a pilot study to evaluate the effect of this drug on the lentiviral reservoir in a recently published animal model, that is SIVmac251-infected rhesus macaques with viral loads stably suppressed by three-drug ART (tenofovir + emtricitabine + raltegravir) [29] (see Fig. 1 of Supplemental Digital Content 2, for a schematic description of the animal study design). Auranofin was well tolerated, and serum chemistry (kidney and liver enzymes) and hematology values remained within normal limits. One month of auranofin treatment induced a significant reduction in the frequency of the long-lived TCM/TTM CD4+ T-cell subpopulation in peripheral blood (Fig. 2a), accompanied by a relative increase in the TEM subset (Fig. 2b), whereas the frequency of TN cells remained unchanged (34.2 ± 7.0% at baseline and 42.9 ± 9.3% at day 30; P = 0.1077; one-tailed Student's t-test). Similar effects were observed in the CD8+ T-cell subpopulations (data not shown). Total CD4+ T-cell counts were maintained stable (1643 ± 653 at time 0 vs. 1550 ± 615 at day 30; P = 0.52, two-tailed paired t-test), in agreement with the unaltered CD4 cell counts upon treatment of psoriatic arthritis using auranofin in humans [27]. Since TN cells did not increase significantly, the relative decrease of TCM/TTM cells was likely to be a specific effect of the drug and not the result of a late rise in TN cells as observed by Autran et al.[35] in humans under ART. Reduction in the frequency of TCM/TTM cells was paralleled by a decrease in cell-associated vDNA in peripheral blood mononuclear cells (PBMCs), which fell below the limit of detection (2 copies/5 × 105 cells) in all study subjects after 4 weeks of auranofin treatment (Fig. 2c). The macaques maintained an undetectable viral load during the whole study period, apart from a transient blip in monkey P044 (showing 720 vRNA copies/ml at 8 weeks). Despite maintenance of viral suppression in the majority of the macaques, we observed a rebound in vDNA in all animals after 8 weeks of treatment with auranofin [21.83 ± 11.20 vDNA copies/5 × 105 cells (mean ± SD)]. This rebound in vDNA in PBMCs suggested cryptic viral replication in anatomical reservoirs of macaques treated with ART, as recently demonstrated in a similar animal model by Bourry et al.[36]. This hypothesis is supported by cell culture data using HIV-1 and showing an increased capacity of TEM cells to sustain viral replication, as compared to their parent TCM/TTM lineage [37]. The results of this pilot study provided proof of concept that, at least transiently, auranofin might decrease the vDNA reservoir.

Fig. 2:
Changes in the memory CD4+ T-cell subsets and frequency of cells harboring viral DNA in PBMCs of rhesus macaques infected with SIVmac251 and treated with ART and auranofin for 4 weeks.(a) Proportions of TCM/TTM cells; (b) proportions of TEM cells; (c) viral DNA (vDNA) copy numbers in PBMCs. In (a and b), a Logit transformation, that is Log[proportion/(1-proportion)], has been adopted in the y-axis in order to restore normality. The P values are shown, as obtained by paired t-test analysis. In (b), a one-tailed t-test was used, due to the expectancy for no-effect or increase, resulting from the significant relative decrease in TCM/TTM cells documented in panel a by a two-tailed test.

Auranofin and antiretroviral therapy intensification decreases the size of the vDNA reservoir in SIVmac251-infected macaques

The pilot study prompted us to evaluate the long-term effects of auranofin on vDNA in a controlled study. As controls, we enrolled two rhesus macaques with stably suppressed viral loads by the same ART adopted in the auranofin group. To prevent a possible viral reservoir replenishment through ongoing viral replication in anatomical compartment(s), ART was intensified with ritonavir-boosted darunavir in all animals. Intensified ART (iART) reinforced the decreasing trend of vDNA in the auranofin-treated macaques (P = 0.0066; t-test for regression). This trend was not observed in the iART-only controls (monkeys 4416 and 4423) (P = 0.6393; t-test for regression). The between-group difference in response was significant when measured by both the F test for slope (Fig. 3) and repeated-measures two-way ANOVA (P = 0.013). We concluded that auranofin and iART significantly reduced the CD4+ T-cell-associated vDNA in SIVmac251-infected macaques. Virus could not be cultivated from peripheral blood CD4+ T cells of monkeys after 11 weeks of iART and auranofin, suggesting that the majority of the vDNA copies, still detectable in the PBMCs of some of the monkeys at this stage, were defective.

Fig. 3:
Effect of auranofin and intensified ART (iART) on viral reservoirs.(a) Averaged viral DNA (± SEM) values in PBMCs of macaques treated with iART and auranofin (n = 6) and iART alone (n = 2). (b) Averaged viral loads (± SEM) in plasma following administration of HDAC inhibitor, SAHA to the aforementioned monkeys treated with iART and auranofin (n = 6) and two historical controls treated with iART alone. The lines parallel to the y-axis delimit the periods in which SAHA cycles were administered. The first SAHA cycle (dashed lines) was administered to all monkeys; the subsequent cycle (dotted lines) was given only to the monkeys under iART and auranofin. In each panel, the P values in black show the significant differences between the trend curves.

Auranofin restricts a replication-competent SIVmac251 reservoir

RT real-time PCR analysis failed to show any detectable vRNA in lymph node biopsies from the study subjects, apart from one iART-alone control, that is 4416, showing 20 copies of vRNA/5 × 105 cells. In order to show the impact of auranofin and iART on replication-competent viral reservoirs, we thus performed a dynamic test using the potent histone deacetylases (HDAC) inhibitor, vorinostat (SAHA), endowed with antilatency/pro-replicative activity [16]. Monkeys at week 10 of iART/auranofin treatment were treated with an oral dose of SAHA of 178.5 mg/m2 of body surface twice daily for 3 days, and response was compared with that of two historical controls (i.e. subjects 4388 and 4398) that had received the same dosage of SAHA while being treated with the same iART protocol in the absence of auranofin. Despite two administration cycles of SAHA, viral load remained undetectable in all six auranofin-treated monkeys. Instead, viremia rebounded following only one cycle of SAHA in the two controls (Fig. 3b). This observation shows that auranofin and iART restricted a viral reservoir harboring a replication-competent and SAHA-sensitive virus.

Follow-up after suspension of iART and auranofin

To test the capacity of auranofin to induce a natural long-term control of SIVmac251 replication in the absence of iART, all treatments were suspended after observing an undetectable vDNA in PBMCs on, at least, three subsequent occasions. In the iART-only group, macaque 4423 was addressed to treatment suspension following the same protocol adopted for the auranofin group. The other iART-only control, 4416, showing an increasing vDNA (103 copies/5 × 105 PBMCs at week 14 of iART), was addressed to treatment suspension in the subsequent week, that is after the median period of exposure to iART in the auranofin group. Monkey P252 met the characteristics for enrolment in another study and was therefore excluded from the treatment suspension protocol. The auranofin group, when compared to the iART-alone group, showed significant delays in the viral load rebounds (P = 0.0495; Gehan-Breslow-Wilcoxon test; Fig. 4; see Fig. 4 of Supplemental Digital Content 2, for survival curves). Noteworthy, in two monkeys treated with auranofin (M970 and P249), viral loads rebounded at weeks 7 and 8 following therapy suspension, a time remarkably longer than that observed in iART-alone controls (mean 1.5 weeks). Although early peaks reminiscent of new acute infections were observed, viral loads of auranofin-treated monkeys dropped to set points significantly lower than either the pre-ART viral loads from the same animals or the viral set points of iART-only controls (Fig. 4a and b), which were comparable to pretherapy values (Fig. 4c) and consistent with trends previously described in both humans and nonhuman primates following suspension of ART [38,39]. Macaque M970 was subjected to an extended follow-up (Fig. 4b). At 7 months following therapy suspension, this animal remains a viremic controller meeting a recently published definition [40], that is viral load consistently 5000 copies/ml or less (undetectable on, at least, two occasions), and high CD4 cell counts. Despite therapy suspension, CD4 cell counts continued to increase in this monkey (+15.23 ± 5.408 cells/μl/week; P = 0.0088; t-test for regression; Fig. 4b). No significant CD4 consumption was found in two of the other monkeys from the auranofin-and-iART group (M974 and P249; P = 0.2559 and 0.8737, respectively; t-test for regression; Fig. 4b). Of note, all macaques addressed to the auranofin-and-iART group had shown either continuously decreasing or constantly subnormal CD4 cell counts before the onset of ART [29].

Fig. 4:
Viro-immunological follow-up of SIVmac251-infected rhesus macaques following suspension of intensified ART (iART) alone (panel a) or iART and auranofin (panel b).(a, b) Viral loads (VLs) are shown in red (units on the left y-axis). CD4 (black) and CD8 (green) cell counts are also shown (units on the right y-axis). Black lines refer to the CD4 slope [solid lines: significant decreases (P < 0.05, t-test for slope), dashed lines: nonsignificant decreases]. Note that monkey M974 had displayed low CD4 cell counts also before the infection [29]. Monkey M970 was selected for extended follow-up. (c) Comparison between the VL set points following suspension of iART (post-ART) and pre-ART VLs. Blue: iART and auranofin; red: iART alone. The P values are reported for the differences between pre and post-ART VLs in the iART and auranofin group (horizontal quote) and for the difference between post-ART VLs in the iART and auranofin and iART-alone groups (vertical quote).

Because the CD4/CD8 ratio is predictive of disease progression [41] and viral reservoir magnitude [4,6], we monitored this parameter over time after treatment suspension. Monkeys that had received auranofin showed, concomitantly to the viral load peak, a transiently decreased CD4/CD8 ratio (see Fig. 5a of Supplemental Digital Content 2,, for the general trends), which was sustained by an increase in the absolute numbers of CD8+ T lymphocytes which was statistically evident at day 4 (P < 0.05; Student Newman Keuls test), that is in temporal vicinity to the viral load peak (see Fig. 4b; and Fig. 5b in Supplemental Digital Content 2, for the general trends). The CD4/CD8 ratio then stabilized in these monkeys due to a decrease of CD8 cell counts from peak (P < 0.05; Student Newman Keuls test) (Fig. 4b; Fig. 5, Supplemental Digital Content 2, Instead, the monkeys which had received iART alone showed either a CD4/CD8 ratio which remained constantly subnormal, or CD8 cell counts decreasing in parallel with CD4+ T cells (Fig. 4a and Fig. 5 in Supplemental Digital Content 2,

In addition to the CD4/CD8 ratio, monkeys maintaining stable CD4 cell counts showed other characteristics that resembled those of rare HIV-1-infected humans who exhibit prolonged viral control after interruption of treatment initiated early during acute infection [42]: they showed undetectable levels of cell-associated vDNA or vRNA in rectal biopsies (i.e. <2 copies/5 × 105 cells) and maintained low vDNA levels in PBMCs (see Fig. 6 in Supplemental Digital Content 2, As expected, the latter parameter significantly correlated with the slope of the CD4 cell counts (see Fig. 6 in Supplemental Digital Content 2,, in agreement with observations in humans showing that absolute CD4 cell counts identify the HIV reservoir [6].


Our data suggest that reduction of the viral reservoir can be achieved by pharmacological strategies, and that this reduction is associated with partial control of viral replication after treatment suspension. Apart from the Berlin patient case [14], reduction of viral load following therapy suspension has so far been reported only when treating during early infection but not in the chronic phase [42–45]. The results of the present study are consistent with a model in which acceleration of lymphocyte turnover drives the infected T cells to progress to shorter-lived phenotypes and die, with the associated virus following their fate. This model is supported by the significant correlation existing between the change in CD4+ TCM/TTM cells and the half-life of the cell-associated vDNA during treatment with iART/auranofin (Table 1). If this conclusion is correct, strategies targeting the long-lived TCM/TTM cells might be adopted to decrease the magnitude of the viral reservoir.

Table 1:
Multiple correlation of selected viro-immunological parameters in the group of monkeys treated with auranofin and intensified ART.

According to multiple correlation analysis, the eventual re-establishment of the vDNA set point following therapy suspension significantly correlated to the TCM/TTM decrease during auranofin treatment, as well as to other variables such as the aviremic period following therapy suspension and the viral load peak upon re-appearance of vRNA in plasma (Table 1). One unexpected finding of the present study was the novel acute-infection-like viral load increase/decrease that occurred following therapy suspension in the auranofin/iART-treated macaques. During acute infection, the long-term impact of the amplitude and duration of the peak on the eventual reservoir is supported by studies conducted in humans and monkeys that had received ART at the early stages of the disease and showing that blunting the initial viremia induces long-term control of the viral reservoir [42–44]. Partial correlation analysis suggested that similar phenomena occurred also in the monkeys that had received auranofin. After removing the effects of auranofin on vDNA and TCM/TTM cells during iART, the viral load peak was the only parameter that still correlated with the eventual vDNA reservoir in the absence of therapy (P = 0.04). Why the early events are particularly important for the resulting viral reservoir is at present unknown.

Immune-mediated mechanisms which merit further investigation may also have played an important role in circumscribing the viral reservoir upon viral load re-appearance. It is well known that the immune response is important for re-establishing control of viremia in the initial phases of the disease [46]. Macaques that had been treated with auranofin showed, in temporal vicinity to the viral load re-appearance, a peak in CD8 cell counts which did not occur in the iART-only group. It is possible to hypothesize that this CD8 peak is linked to subsequent control of viral replication. In this context, the initial balance between viral load and immune response could be pivotal for determining the eventual reservoir. The occurrence of the CD8 peak in the auranofin group and its lack in iART-only controls raise the hypothesis that reduction of the TCM/TTM pools induced by auranofin not only did not compromise the immune response against the virus, but, rather, facilitated it in some ways. This hypothesis is supported by results showing that the pool of TCM cells is a correlate of anergy towards the viral antigens in Macaca mulatta but not in Cercocebus atys, which is naturally resistant to CD4+ T-cell loss and full-blown AIDS [47]. Be that as it may, one important consequence of this phenomenon could be the re-opening, during the chronic phase of the infection, of a window of opportunity for treatments and therapeutic vaccines [42,48] during or before the reoccurrence of the acute infection.


Auranofin and iART decreased the TCM/TTM pool and restricted the magnitude of the viral reservoir in rhesus macaques. These events were followed by eventual containment of viremia when therapy was interrupted. Future investigation of these phenomena may pave the way to discovery of a cure for HIV-1/AIDS.


A.S. is thankful to Drs Marco Sgarbanti and Silvia Vendetti, Rome, Italy, for illuminating scientific discussion. Funded by Fondazione Roma, Italy. M.G.L., assisted by M.C. and W.L.W., conducted the in-vivo experiments. J.Y.O. and J.G., respectively, conducted the ex-vivo analyses. S.D.F. and N.C. conducted the in-vitro analyses and participated in the experimental design, data analysis and interpretation and manuscript drafting. A.T.P. participated in data analysis and interpretation. E.G. suggested the use of auranofin. B.C. and S.N. conducted preparatory work on which this study is based and participated in data analysis and interpretation. A.S. conceived and coordinated the study, did the experimental design, participated in the ex-vivo data generation, conducted the statistical analyses and drafted the manuscript.


1. Colin L, Van Lint C. Molecular control of HIV-1 postintegration latency: implications for the development of new therapeutic strategies. Retrovirology 2009; 6:111.
2. Trono D, Van Lint C, Rouzioux C, Verdin E, Barré-Sinoussi F, Chun TW, Chomont N. HIV persistence and the prospect of long-term drug-free remissions for HIV-infected individuals. Science 2010; 329:174–180.
3. Chun TW, Stuyver L, Mizell SB, Ehler LA, Mican JA, Baseler M, et al. Presence of an inducible HIV-1 latent reservoir during highly active antiretroviral therapy. Proc Natl Acad Sci U S A 1997; 94:13193–13197.
4. Chun TW, Justement JS, Pandya P, Hallahan CW, McLaughlin M, Liu S, et al. Relationship between the size of the human immunodeficiency virus type 1 (HIV-1) reservoir in peripheral blood CD4+ T cells and CD4+:CD8+ T cell ratios in aviremic HIV-1-infected individuals receiving long-term highly active antiretroviral therapy. J Infect Dis 2002; 185:1672–1676.
5. Finzi D, Hermankova M, Pierson T, Carruth LM, Buck C, Chaisson RE, et al. Identification of a reservoir for HIV-1 in patients on highly active antiretroviral therapy. Science 1997; 278:1295–1300.
6. Chomont N, El-Far M, Ancuta P, Trautmann L, Procopio FA, Yassine-Diab B, et al. HIV reservoir size and persistence are driven by T-cell survival and homeostatic proliferation. Nat Med 2009; 15:893–900.
7. Fritsch RD, Shen X, Sims GP, Hathcock KS, Hodes RJ, Lipsky PE. Stepwise differentiation of CD4 memory T cells defined by expression of CCR7 and CD27. J Immunol 2005; 175:6489–6497.
8. Zielinski CE, Corti D, Mele F, Pinto D, Lanzavecchia A, Sallusto F. Dissecting the human immunologic memory for pathogens. Immunol Rev 2011; 240:40–51.
9. Okada R, Kondo T, Matsuki F, Takata H, Takiguchi M. Phenotypic classification of human CD4+ T cell subsets and their differentiation. Int Immunol 2008; 20:1189–1199.
10. Hendriks J, Xiao Y, Borst J. CD27 promotes survival of activated T cells and complements CD28 in generation and establishment of the effector T cell pool. J Exp Med 2003; 198:1369–1380.
11. Hendriks J, Gravestein LA, Tesselaar K, van Lier RA, Schumacher TN, Borst J. CD27 is required for generation and long-term maintenance of T cell immunity. Nat Immunol 2000; 1:433–440.
12. Riou C, Yassine-Diab B, Van Grevenynghe J, Somogyi R, Greller LD, Gagnon D, et al. Convergence of TCR and cytokine signaling leads to FOXO3a phosphorylation and drives the survival of CD4+ central memory T cells. J Exp Med 2007; 204:79–91.
13. van Oosterwijk MF, Juwana H, Arens R, Tesselaar K, van Oers MH, Eldering E, et al. CD27-CD70 interactions sensitise naive CD4+ T cells for IL-12-induced Th1 cell development. Int Immunol 2007; 19:713–718.
14. Hütter G, Nowak D, Mossner M, Ganepola S, Müssig A, Allers K, et al. Long-term control of HIV by CCR5 Delta32/Delta32 stem-cell transplantation. N Engl J Med 2009; 360:692–698.
15. Hamer DH. Can HIV be cured? Mechanisms of HIV persistence and strategies to combat it. Curr HIV Res 2004; 2:99–111.
16. Savarino A, Mai A, Norelli S, El Daker S, Valente S, Rotili D, et al. Shock and kill’ effects of class I-selective histone deacetylase inhibitors in combination with the glutathione synthesis inhibitor buthionine sulfoximine in cell line models for HIV-1 quiescence. Retrovirology 2009; 6:52.
17. Eisler R. Chrysotherapy: a synoptic review. Inflamm Res 2003; 52:487–501.
18. Maldonado A, Mueller YM, Thomas P, Bojczuk P, O’Connors C, Katsikis PD. Decreased effector memory CD45RA+ CD62L- CD8+ T cells and increased central memory CD45RA- CD62L+ CD8+ T cells in peripheral blood of rheumatoid arthritis patients. Arthritis Res Ther 2003; 5:R91–R96.
19. Dalziel K, Going G, Cartwright PH, Marks R, Beveridge GW, Rowell NR. Treatment of chronic discoid lupus erythematosus with an oral gold compound (auranofin). Br J Dermatol 1986; 115:211–216.
20. Kean WF, Hart L, Buchanan WW. Auranofin. Br J Rheumatol 1997; 36:560–572.
21. Hashimoto K, Whitehurst CE, Lipsky PE. Synergistic inhibition of T cell proliferation by gold sodium thiomalate and auranofin. J Rheumatol 1994; 21:1020–1026.
22. Han S, Kim K, Kim H, Kwon J, Lee YH, Lee CK, et al. Auranofin inhibits overproduction of pro-inflammatory cytokines, cyclooxygenase expression and PGE2 production in macrophages. Arch Pharm Res 2008; 31:67–74.
23. Kim TS, Kang BY, Lee MH, Choe YK, Hwang SY. Inhibition of interleukin-12 production by auranofin, an antirheumatic gold compound, deviates CD4(+) T cells from the Th1 to the Th2 pathway. Br J Pharmacol 2001; 134:571–578.
24. Rigobello MP, Folda A, Dani B, Menabò R, Scutari G, Bindoli A. Gold(I) complexes determine apoptosis with limited oxidative stress in Jurkat T cells. Eur J Pharmacol 2008; 582:26–34.
25. Cox AG, Brown KK, Arner ES, Hampton MB. The thioredoxin reductase inhibitor auranofin triggers apoptosis through a Bax/Bak-dependent process that involves peroxiredoxin 3 oxidation. Biochem Pharmacol 2008; 76:1097–1109.
26. Hafejee A, Winhoven S, Coulson IH. Jessner's lymphocytic infiltrate responding to oral auranofin. J Dermatolog Treat 2004; 15:331–332.
27. Shapiro DL, Masci JR. Treatment of HIV associated psoriatic arthritis with oral gold. J Rheumatol 1996; 10:1818–1820.
28. Park SJ, Kim M, Kim NH, Oh MK, Cho JK, Jin JY, Kim IS. Auranofin promotes retinoic acid- or dihydroxyvitamin D3-mediated cell differentiation of promyelocytic leukaemia cells by increasing histone acetylation. Br J Pharmacol 2008; 154:1196–1205.
29. Lewis MG, Norelli S, Collins M, Barreca ML, Iraci N, Chirullo B, et al. Response of a simian immunodeficiency virus (SIVmac251) to raltegravir: a basis for a new treatment for simian AIDS and an animal model for studying lentiviral persistence during antiretroviral therapy. Retrovirology 2010; 7:21.
30. Mai A, Esposito M, Sbardella G, Massa S. A new facile and expeditious synthesis of /N/-hydroxy-/N(/-phenyloctanediamide, a potent inducer of terminal cytodifferentiation. Org Prep Proced Int 2001; 33:391–394.
31. Pitcher CJ, Hagen SI, Walker JM, Lum R, Mitchell BL, Maino VC, et al. Development and homeostasis of T cell memory in rhesus macaque. J Immunol 2002; 168:29–43.
32. Lewis D, Capell HA, McNeil CJ, Iqbal MS, Brown DH, Smith WE. Gold levels produced by treatment with auranofin and sodium aurothiomalate. Ann Rheum Dis 1983; 42:566–570.
33. Kim IS, Jin JY, Lee IH, Park SJ. Auranofin induces apoptosis and when combined with retinoic acid enhances differentiation of acute promyelocytic leukaemia cells in vitro. Br J Pharmacol 2004; 142:749–755.
34. Yukl S, Sinclair E, Epling L, Li Q, Shergill A, McQuaid K, et al. CD4+ T cell reconstitution, T cell activation, and memory T cell subset composition in blood and gut of HIV-negative and ART-suppressed HIV-positive patients: implications for HIV persistence in the gut. J Int AIDS Soc 2010; 13 (Suppl 3):O1.
35. Autran B, Carcelain G, Li TS, Blanc C, Mathez D, Tubiana R, et al. Positive effects of combined antiretroviral therapy on CD4+ T cell homeostasis and function in advanced HIV disease. Science 1997; 277:112–116.
36. Bourry O, Mannioui A, Sellier P, Roucairol C, Durand-Gasselin L, Dereuddre-Bosquet N, et al. Effect of a short-term HAART on SIV load in macaque tissues is dependent on time of initiation and antiviral diffusion. Retrovirology 2010; 7:78.
37. Groot F, van Capel TM, Schuitemaker J, Berkhout B, de Jong EC. Differential susceptibility of naïve, central memory and effector memory T cells to dendritic cell-mediated HIV-1 transmission. Retrovirology 2006; 3:52.
38. Angel JB, Routy JP, Tremblay C, Ayers D, Woods R, Singer J, et al.A randomized controlled trial of HIV therapeutic vaccination using ALVAC with or without Remune.AIDS 2011 [Epub ahead of print].
39. Cervasi B, Paiardini M, Serafini S, Fraternale A, Menotta M, Engram J, et al. Administration of fludarabine-loaded autologous red blood cells in simian immunodeficiency virus-infected sooty mangabeys depletes pSTAT-1-expressing macrophages and delays the rebound of viremia after suspension of antiretroviral therapy. J Virol 2006; 80:10335–10345.
40. Cafaro A, Bellino S, Titti F, Maggiorella MT, Sernicola L, Wiseman RW, et al. Impact of viral dose and major histocompatibility complex class IB haplotype on viral outcome in mauritian cynomolgus monkeys vaccinated with Tat upon challenge with simian/human immunodeficiency virus SHIV89.6P. J Virol 2010; 84:8953–8958.
41. Margolick JB, Gange SJ, Detels R, O’Gorman MR, Rinaldo CR Jr, Lai S. Impact of inversion of the CD4/CD8 ratio on the natural history of HIV-1 infection. J Acquir Immune Defic Syndr 2006; 42:620–626.
42. Hocqueloux L, Prazuck T, Avettand-Fenoel V, Lafeuillade A, Cardon B, Viard JP, Rouzioux C. Long-term immunovirologic control following antiretroviral therapy interruption in patients treated at the time of primary HIV-1 infection. AIDS 2010; 24:1598–1601.
43. Lori F, Lewis MG, Xu J, Varga G, Zinn DE Jr, Crabbs C, et al. Control of SIV rebound through structured treatment interruptions during early infection. Science 2000; 290:1591–1593.
44. Benlhassan-Chahour K, Penit C, Dioszeghy V, Vasseur F, Janvier G, Rivière Y, et al. Kinetics of lymphocyte proliferation during primary immune response in macaques infected with pathogenic simian immunodeficiency virus SIVmac251: preliminary report of the effect of early antiviral therapy. J Virol 2003; 77:12479–12493.
45. Trkola A, Kuster H, Rusert P, Joos B, Fischer M, Leemann C, et al. Delay of HIV-1 rebound after cessation of antiretroviral therapy through passive transfer of human neutralizing antibodies. Nat Med 2005; 11:615–622.
46. Borrow P, Shattock RJ, Vyakarnam A. EUROPRISE Working Group. Innate immunity against HIV: a priority target for HIV prevention research. Retrovirology 2010; 7:84.
47. Bostik P, Noble ES, Mayne AE, Gargano L, Villinger F, Ansari AA. Central memory CD4 T cells are the predominant cell subset resistant to anergy in SIV disease resistant sooty mangabeys. AIDS 2006; 20:181–188.
48. Autran B, Carcelain G, Combadiere B, Debre P. Therapeutic vaccines for chronic infections. Science 2004; 305:205–208.

central memory; eradication; remission; reservoir; SIVmac251; therapy suspension; viral DNA

Supplemental Digital Content

© 2011 Lippincott Williams & Wilkins, Inc.