Evaluating a New Class of AKT/mTOR Activators for HIV Latency Reversing Activity Ex Vivo and In Vivo - PubMed (original) (raw)

. 2021 Mar 25;95(8):e02393-20.

doi: 10.1128/JVI.02393-20. Epub 2021 Feb 3.

Roland Schwarzer 1 2, William Brantley 3, Benjamin Varco-Merth 3, Hannah S Sperber 4 5, Philip A Hull 1, Mauricio Montano 1, Stephen A Migueles 6, Danielle Rosenthal 6, Louise E Hogan 2, Jeffrey R Johnson 1 7, Thomas A Packard 1, Zachary W Grimmett 1, Eytan Herzig 1 2, Emilie Besnard 1, Michael Nekorchuk 3, Feng Hsiao 1 8, Steven G Deeks 2, Michael Snape 9, Bernard Kiernan 9, Nadia R Roan 1 8, Jeffrey D Lifson 10, Jacob D Estes 3, Louis J Picker 3, Eric Verdin 1 2 11, Nevan J Krogan 1 7, Timothy J Henrich 2, Mark Connors 6, Melanie Ott 1 2 11, Satish K Pillai 12 4, Afam A Okoye 13, Warner C Greene 14 2 11

Affiliations

Evaluating a New Class of AKT/mTOR Activators for HIV Latency Reversing Activity Ex Vivo and In Vivo

Andrea Gramatica et al. J Virol. 2021.

Abstract

An ability to activate latent HIV-1 expression could benefit many HIV cure strategies, but the first generation of latency reversing agents (LRAs) has proven disappointing. We evaluated AKT/mTOR activators as a potential new class of LRAs. Two glycogen synthase kinase-3 inhibitors (GSK-3i's), SB-216763 and tideglusib (the latter already in phase II clinical trials) that activate AKT/mTOR signaling were tested. These GSK-3i's reactivated latent HIV-1 present in blood samples from aviremic individuals on antiretroviral therapy (ART) in the absence of T cell activation, release of inflammatory cytokines, cell toxicity, or impaired effector function of cytotoxic T lymphocytes or NK cells. However, when administered in vivo to SIV-infected rhesus macaques on suppressive ART, tideglusib exhibited poor pharmacodynamic properties and resulted in no clear evidence of significant SIV latency reversal. Whether alternative pharmacological formulations or combinations of this drug with other classes of LRAs will lead to an effective in vivo latency-reversing strategy remains to be determined.IMPORTANCE If combined with immune therapeutics, latency reversing agents (LRAs) have the potential to reduce the size of the reservoir sufficiently that an engineered immune response can control the virus in the absence of antiretroviral therapy. We have identified a new class of LRAs that do not induce T-cell activation and that are able to potentiate, rather than inhibit, CD8+ T and NK cell cytotoxic effector functions. This new class of LRAs corresponds to inhibitors of glycogen synthase kinase-3. In this work, we have also studied the effects of one member of this drug class, tideglusib, in SIV-infected rhesus monkeys. When tested in vivo, however, tideglusib showed unfavorable pharmacokinetic properties, which resulted in lack of SIV latency reversal. The disconnect between our ex vivo and in vivo results highlights the importance of developing next generation LRAs with pharmacological properties that allow systemic drug delivery in relevant anatomical compartments harboring latent reservoirs.

Copyright © 2021 American Society for Microbiology.

PubMed Disclaimer

Figures

FIG 1

FIG 1

Inhibition of GSK-3 reverses latency in blood and gut tissue from HIV-positive individuals on ART studied ex vivo. (A) Schematic representation of the workflow followed for the analysis of latency reversion in blood and GALT cultures measured as HIV-1 virion levels (i.e., numbers of HIV-1 RNA copies per microliter) in the culture supernatant of blood or GALT samples. (B) Blood CD4+ T cells treated for 48 h with a single LRA or anti-CD3/CD28-coated beads. (C) Normalized percentage of live CD4+ T cells compared to untreated controls. (D) GALT samples were treated for 24 h with 1 μM tideglusib or anti-CD3/CD28 coated beads. (E and F) LRA-stimulated CD4+ T cells used for panels B and C were tested for surface expression of CD69 and CD25 activation markers. Lymphocytes were gated on single cells > live > CD3+ and CD69+ or CD25+. (G) After virus purification, culture supernatants of treated CD4+ T cells were tested for cytokine release using a proinflammatory panel including tumor necrosis factor alpha (TNF-α), IFN-γ, IL-2, IL-4, IL-6, IL-10, IL-13, and IL-17. Latency-reversal data are presented as fold induction relative to the untreated control (DMSO). Unless specified in the figure, treatments were done at the following concentrations/volumes: DMSO control, 0.01% (vol/vol) final volume; anti-CD3/CD28 coated beads, 125 μl (25 μl per 1 × 106 cells); bryostatin, 10 nM; panobinostat, 50 nM. Statistical significance was calculated from the HIV-1 RNA copy number values using a ratio paired t test compared with the DMSO control (*, P < 0.05; **, P < 0.001; ***, P < 0.0001; ****, P < 0.00001; ns, not significant). Error bars represent standard errors of the means.

FIG 2

FIG 2

Tideglusib activates the AKT/mTOR pathway in CD4+ T as detected by NanoString and whole-cell phosphoproteomics analysis. Tideglusib influences activation status of specific components of the AKT/mTOR pathway. (A) NanoString analysis of CD4+ T cells isolated from five HIV-infected individuals on suppressive ART treated with 1 μM tideglusib for up to 6 h. The differential expression of phospho-GSK3β (dark blue), phospho-AKT (orange), phospho-p70 S6 kinase (light blue), and phospho-TSC2 (red) is represented as linear fold change versus the untreated control (DMSO). Stars indicate points with P values of ≤0.05. Each data point is derived from 5 biological replicates (n = 5). (B) MS/MS spectra identifying GSK-3β S9 (TTS*FAESCKPVQQPSAFGSMK), TSC2 S939 (STS*LNERPK), and Rictor T1295 (SNSVSLVPPGSSHT*LPR). Phosphorylation was annotated by the Skyline spectrum builder, and y fragment ions are indicated in purple and blue, respectively. Fragment ion charge states are 1+ unless indicated otherwise. (C) Schematic representation of the phosphorylation sites detected by NanoString and phosphoproteomics on the respective protein/protein complexes.

FIG 3

FIG 3

Pathway regulation as induced by treatment with tideglusib in CD4+ T cells from HIV-positive donors on ART. (A) Heat map displaying global significance statistics and (B) pathway regulation obtained with the NanoString software following analysis of 700 RNA transcripts isolated from CD4+ T cells treated with 1 μM tideglusib for 0.25 to 6 h. A detailed list of the transcripts associated with each pathway is reported in the manufacturer’s protocol of the solid tumor assay (NanoString Technologies). The “global significance” of the differential expression of the analyzed RNA transcripts compared to the untreated control for each time point after treatment is reported with a color-coded score: a high score (orange) indicates transcripts that exhibit highly significant differential expression, while a low score (blue) indicates less significant differential expression. Regulation of selected pathways is reported with a color-coded score: light blue to red indicates pathway activation; light blue to white indicates pathway inhibition. (C) Tideglusib activates genes involved in splicing and positive regulation of transcription. A set of 142 gene names was extracted from the differential analysis of 1 μM tideglusib versus untreated phosphoproteomics results using cutoffs of log2 fold change of >1 or <−1 and a P value of <0.05. The gene set was then submitted to Enrichr for enrichment analysis. Tabular enrichment results obtained were downloaded into WikiPathways and Gene Ontology:Biological Process, and selected terms were plotted using R based on relevance, uniqueness, and combined score.

FIG 4

FIG 4

AKT and mTOR modulate tideglusib-mediated NF-κB activation in CD4+ T cells. (A) CD4+ T cells were purified from blood of three HIV-negative individuals and treated for the indicated times with either 1 μΜ tideglusib, anti-CD3/CD28, or a control DMSO solution. (B) Blood CD4+ T cells treated for 1 h with either 1 μΜ tideglusib, anti-CD3/CD28 coated beads, and increasing concentrations of the mTOR inhibitor PP242 or the AKT inhibitor API-2. Nuclear extracts were prepared from these cells (A and B), and p65 binding to target dsDNA was quantitated by enzyme-linked immunosorbent assay (ELISA). Extracts were also immunoblotted for TATA-binding protein (TBP) to test equivalence of protein loading. (C) CellTiter-Blue assays were used to assess potential changes in cell viability in CD4+ T cells treated as for panel B. (D) Fold induction of virion RNA in three HIV-infected individuals on ART treated with tideglusib (1 μM) and DMSO, PP242, or IKK-V, an IKK inhibitor. (E) Fold induction of virion RNA in five HIV-infected individuals on ART treated with SC79 (0.01 to 1 μM) or anti-CD3/CD28-coated beads or left untreated (DMSO control). (F) Normalized percentage of live CD4+ T cells compared to untreated controls. (G) LRA-stimulated CD4+ T cells were tested for surface expression of CD69 and CD25 activation markers. Lymphocytes were gated on single cells > live > CD3+ and CD69+ or CD25+. Unless otherwise specified in the figure, treatments were done at the following concentrations/volumes: DMSO control, 0.01% (vol/vol) final volume; anti-CD3/CD28-coated beads, 125 μl (25 μl per 1 × 106 cells). Data are means and SEM; statistical analysis employed paired two-tailed Student’s t test. *, P < 0.05; **, P < 0.01; ns, not significant.

FIG 5

FIG 5

Schematic summary of the pathway induced by the GSK-3i tideglusib, leading to activation of AKT/mTOR signaling. All phosphorylation events reported here were experimentally confirmed by NanoString analysis or whole-cell phosphoproteomics. Solid arrows/lines indicate interaction previously reported in the literature; dotted arrows/lines represent proposed interactions. Green arrows/lines indicate activation; red arrows/lines indicate inhibition (note that in many circumstances, a negative regulatory phosphorylation blocks the inhibitory effect of a protein, leading to activation). The indicated phosphorylation sites represent activating (green) or inhibitory (red) phosphorylation events.

FIG 6

FIG 6

Inhibition of GSK-3 does not increase glycolytic metabolism in CD4+ T cells. Flow-cytometric assessment of surface expression of GLUT1 analysis in CD4+ T cells isolated from the blood of HIV-negative individuals after either 2 h (A) or 24 h (B) treatment with DMSO control, 1 μM tideglusib, or anti-CD3/CD28-coated beads, in the presence or absence of 250 nM PP242. (C) Extracts from cells used in panels A and B probed for phosphoS9-GSK-3β by immunoblotting. (D) Assessment of CD4+ T cell glycolytic profile using a Seahorse XF analyzer to determine OCR (oxygen consumption rate [data not shown]) and ECAR (extracellular acidification rate) measurement, 24 h after treatment as for panels A and B. (E) Measurement of compensatory glycolysis as derived from panel D. (F) Measurement of basal glycolysis as calculated from panel D. (G) Quantitation of

l

-lactate production measured in the supernatant of CD4+ T cell cultures, 24 h after treatment with the indicated GSK-3 inhibitors or anti-CD3/CD28, with or without 2-DG. All data represent 3 biological replicates except for data in panels D, E, and F, which represent 6 biological replicates. Data are means and SEM; statistical analysis was performed using a paired two-tailed Student's t test. *, P < 0.05; **, P < 0.01; ns, not significant. AA/Rot, antimycin A/rotenone; mpH/min, milli-pH units per minute.

FIG 7

FIG 7

Treatment with tideglusib does not affect cytotoxic effector functions of CD8+ T, NK cells, or HIV-specific CTLs from HIV-positive individuals on ART. PBMCs isolated from five independent HIV-negative donors were cultured in the presence of the indicated LRAs orleft untreated (DMSO control) for 48 h. Treated PBMCs were then either left unstimulated or activated in the presence of PMA/ionomycin (for CD8+ T cells in panels A, B, and C) or K562 cells (for NK cells in panels D, E, and F). (A) Assessment of cell death in treated or treated and stimulated cells. Percent viability determined by viable cells relative to DMSO or DMSO with PMA/ionomycin is presented. (B and E) Fold change in IFN-γ-producing cells following treatment with tideglusib or bryostatin in the presence of PMA/ionomycin stimulation for CTLs (B) and NK cells (E). (C and F) Fold change in CD107a+ cells following treatment with tideglusib or bryostatin in the presence of PMA/ionomycin stimulation; data are for CTLs (C) and NK cells (F). The gating strategy is presented in Fig. S1 to S4. (D) Assessment of cell death in treated or treated and stimulated cells. Percent viability determined by viable cells relative to DMSO or DMSO plus K562 cells is presented. (G) Representative flow plots depicting infected CD4+ T-cell elimination (ICE) of gated HIVSF162-infected CD4+ T cells after a 1-h incubation in fresh medium alone (left column) or with negatively selected CD8+ T cells that had been initially stimulated for 6 days with HIVSF162-infected autologous CD4+ T-cell targets, without (middle) or with (right) tideglusib (1 μM). Data for a representative LTNP/EC (top row) and three participants (progressors) are shown (clinical data for the three study participants are reported in Table 4). Total percentages of HIV-infected (p24+) cells in each plot were determined as the sum of the percentages of the upper quadrants. ICE values (red) were calculated as follows: [(percent p24 expression of infected targets only − percent p24 expression of infected targets mixed with day 6 cells)/percent p24 expression of infected targets only] × 100. (H) Summary data of HIV-specific CD8+ T cell cytotoxic responses, measured by net GrB substrate fluorescence in infected targets (background fluorescence in cultures of effectors coincubated with uninfected targets has been subtracted) and ICE for 3 LTNP/EC (red circles) and 3 participants (progressors), which include CD8+ T cells derived from untreated (blue circles) or tideglusib-treated (red circles) cultures. Horizontal lines designate median values. Statistical significance was calculated using a ratio paired t test compared with each stimulated/DMSO control (*, P < 0.05; **, P < 0.005; ***, P < 0.0005; ns, not significant). Values are means and standard deviations.

FIG 8

FIG 8

Effect of oral tideglusib on SIV viral reactivation in vivo. (A) Schematic showing timeline of rhesus macaque (RM) tideglusib treatment, including SIVmac239X infection, initiation of ART at 12 days postinfection (dpi), and oral tideglusib treatment phases (8.75 mg/kg for 22 days, starting at 672 dpi; 8.75 mg/kg for 22 days, starting at 708 dpi; 20 mg/kg for 21 days, starting at 852 dpi; 40 mg/kg for 21 days, starting at 873 dpi; and 80 mg/kg for 28 days, starting at 915 dpi). (B) Mean (and SEM) SIVmac239 plasma viral load profiles during initial SIV infection and ART suppression (left) and log area under the curve of SIV plasma viral loads for 0 to 84 dpi (right) for tideglusib-treated RM (red, n = 5) and control RM (gray, n = 5). (C) Mean (and SEM) change from baseline in percent CD169 on CD16− CD14+ classical monocytes in blood following SIV infection and ART suppression for tideglusib-treated RM (red, n = 5) and control RM (gray, n = 5). (D) Mean (and SEM) change from baseline in percent HLA-DR and CD69 (activation) and percent Ki67 (proliferation) on CD4+ memory T cells in blood during treatment periods (pale gray) of tideglusib-treated RM (red, n = 5) and control RM (gray, n = 5). (E) Individual SIV plasma viral load profiles of tideglusib-treated RM (left) and control RM (right) during treatment phases. (F) Mean (and SEM) change from baseline in percent CD169 on CD16− CD14+ classical monocytes in blood of tideglusib-treated RM (red, n = 5) and control RM (gray, n = 5) during treatment periods. (G) Cell-associated SIV RNA in peripheral blood mononuclear cells, lymph node cells, and colons of tideglusib-treated RM (red) and control RM (gray) at 7 days before or 35 days after the start of the 20-mg/kg-treatment period. Measurements below the limits of detection (dotted line) are depicted with unfilled circles. (H) Quantification of SIV RNA-positive cells by RNAscope in the colon of tideglusib-treated RM (red) and control RM (gray) on day 14 of the 40-mg/kg-treatment period. *, treatment with clinical product AMO-02. ns, not significant.

FIG 9

FIG 9

Treatment with tideglusib reverses latency in blood CD4+ T cells from SIV-positive RM on ART studied ex vivo. Blood CD4+ T cells, isolated from three RM on suppressive ART, were either treated for 48 h with 1 μM tideglusib, treated with 10 nM PMA plus 500 nM ionomycin (PMA/I), or left untreated (DMSO control). Data are means and SEM.

FIG 10

FIG 10

Effect of intravenous tideglusib on SIV viral reactivation in vivo. (A) Mean (and SEM) SIV plasma viral load profiles of RM following intravenous administration of tideglusib or vehicle control. Tideglusib was administered biweekly at 5 mg/kg at 1,027, 1,041, 1,055, and 1,069 dpi followed by a 10-mg/kg dose at 1,097 dpi. (B) Cell-associated SIV RNA (in log copies/106 cell equivalents) and (C) ratio of cell-associated SIV RNA to SIV DNA in PBMCs at 0, 1, and 2 days following a 5-mg/kg intravenous infusion of tideglusib or vehicle control at 1,027 dpi. (D) Mean (and SEM) tideglusib drug levels in plasma following a 5-mg/kg intravenous infusion of tideglusib or vehicle control at 1,027 dpi. Limit of detection, 5 ng/ml.

References

    1. Deeks SG. 2012. HIV: shock and kill. Nature 487:439–440. 10.1038/487439a. -DOI -PubMed
    1. Rasmussen TA, Tolstrup M, Sogaard OS. 2016. Reversal of latency as part of a cure for HIV-1. Trends Microbiol 24:90–97. 10.1016/j.tim.2015.11.003. -DOI -PubMed
    1. Battivelli E, Dahabieh MS, Abdel-Mohsen M, Svensson JP, Tojal Da Silva I, Cohn LB, Gramatica A, Deeks S, Greene WC, Pillai SK, Verdin E. 2018. Distinct chromatin functional states correlate with HIV latency reactivation in infected primary CD4(+) T cells. Elife 7:e34655. 10.7554/eLife.34655. -DOI -PMC -PubMed
    1. Barouch DH, Deeks SG. 2014. Immunologic strategies for HIV-1 remission and eradication. Science 345:169–174. 10.1126/science.1255512. -DOI -PMC -PubMed
    1. Schwarzer R, Gramatica A, Greene WC. 2020. Reduce and control: a combinatorial strategy for achieving sustained HIV remissions in the absence of antiretroviral therapy. Viruses 12:188. 10.3390/v12020188. -DOI -PMC -PubMed

Grants and funding

LinkOut - more resources