TYPE Original Research PUBLISHED 05 September 2022 DOI 10.3389/fimmu.2022.1001068 IRF7 expression correlates OPEN ACCESS with HIV latency reversal EDITED BY Guido Poli, Vita-Salute San Raffaele University, upon specific blockade Italy REVIEWED BY of immune activation Fabio Romerio, Johns Hopkins Medicine, United States Massimo Pizzato, Ifeanyi Jude Ezeonwumelu 1, Edurne Garcı́a-Vidal 1, University of Trento, Italy Eudald Felip 1,2, Maria C. Puertas 1,3, Bruna Oriol-Tordera 1, *CORRESPONDENCE Eva Riveira-Muñoz Lucı́a Gutiérrez-Chamorro 1, André Gohr 4, Marta Ruiz-Riol 1,3, eriveira@irsicaixa.es Marta Massanella 1,3, Bonaventura Clotet 1,3,5, Ester Ballana eballana@irsicaixa.es Javier Martinez-Picado 1,3,5,6, Roger Badia 1, SPECIALTY SECTION Eva Riveira-Muñoz 1* and Ester Ballana 1* This article was submitted to Viral Immunology, 1 IrsiCaixa AIDS Research Institute – IrsiCaixa and Health Research Institute Germans Trias i Pujol a section of the journal (IGTP), Hospital Germans Trias i Pujol, Universitat Autònoma de Barcelona, Badalona, Spain, Frontiers in Immunology 2 Medical Oncology Department, Catalan Institute of Oncology (ICO)-Badalona, B-ARGO (Badalona Applied Research Group in Oncology) and IGTP (Health Research Institute Germans Trias i Pujol), RECEIVED 22 July 2022 Universitat Autònoma de Barcelona, Badalona, Spain, 3 Consorcio Centro de Investigación ACCEPTED 15 August 2022 Biomédica en Red de Enfermedades Infecciosas (CIBERINFEC), Instituto de Salud Carlos III, PUBLISHED 05 September 2022 Madrid, Spain, 4 Scientific Computing Facility, Max Planck Institute of Molecular Cell Biology and Genetics, Dresden, Germany, 5 Centre for Health and Social Care Research (CESS), Faculty of CITATION Medicine, University of Vic – Central University of Catalonia (UVic – UCC), Vic, Spain, 6 Catalan Ezeonwumelu IJ, Garcı´a-Vidal E, Institution for Research and Advanced Studies (ICREA), Barcelona, Spain Felip E, Puertas MC, Oriol-Tordera B, Gutiérrez-Chamorro L, Gohr A, Ruiz-Riol M, Massanella M, Clotet B, Martinez-Picado J, Badia R, The persistence of latent HIV reservoirs allows for viral rebound upon antiretroviral Riveira-Muñoz E and Ballana E (2022) therapy interruption, hindering effective HIV-1 cure. Emerging evidence suggests that IRF7 expression correlates with HIV latency reversal upon specific modulation of innate immune stimulation could impact viral latency and contribute to blockade of immune activation. the clearing of HIV reservoir. Here, the latency reactivation capacity of a subclass of Front. Immunol. 13:1001068. selective JAK2 inhibitors was characterized as a potential novel therapeutic strategy for doi: 10.3389/fimmu.2022.1001068 HIV-1 cure. Notably, JAK2 inhibitors reversed HIV-1 latency in non-clonal lymphoid COPYRIGHT © 2022 Ezeonwumelu, Garcı´a-Vidal, and myeloid in vitro models of HIV-1 latency and also ex vivo in CD4+ T cells from Felip, Puertas, Oriol-Tordera, Gutiérrez- ART+ PWH, albeit its function was not dependent on JAK2 expression. Chamorro, Gohr, Ruiz-Riol, Massanella, Immunophenotypic characterization and whole transcriptomic profiling supported Clotet, Martinez-Picado, Badia, Riveira- Muñoz and Ballana. This is an open- reactivation data, showing common gene expression signatures between latency access article distributed under the reactivating agents (LRA; JAK2i fedratinib and PMA) in contrast to other JAK inhibitors, terms of the Creative Commons Attribution License (CC BY). The use, but with significantly fewer affected gene sets in the pathway analysis. In depth distribution or reproduction in other evaluation of differentially expressed genes, identified a significant upregulation of IRF7 forums is permitted, provided the expression despite the blockade of the JAK-STAT pathway and downregulation of original author(s) and the copyright owner(s) are credited and that the proinflammatory cytokines and chemokines. Moreover, IRF7 expression levels original publication in this journal is positively correlated with HIV latency reactivation capacity of JAK2 inhibitors and cited, in accordance with accepted academic practice. No use, also other common LRAs. Collectively, these results represent a promising step distribution or reproduction is towards HIV eradication by demonstrating the potential of innate immune permitted which does not comply with modulation for reducing the viral reservoir through a novel pathway driven by IRF7. these terms. KEYWORDS IRF7, JAK-STAT signalling, HIV-1 cure strategies, latency reversal agents, innate immunity Frontiers in Immunology 01 frontiersin.org Ezeonwumelu et al. 10.3389/fimmu.2022.1001068 Introduction Thus, targeting the JAK-STAT pathway has been proposed as an attractive cellular target to achieve HIV suppression and to HIV-1 infection remains incurable due to the persistence of reduce inflammation concomitantly, due to its specific anti- HIV in long-lived reservoirs that support viral rebound upon inflammatory properties in various cell types (17, 18). Indeed, treatment interruption. Therefore, the current therapeutic several Janus kinase inhibitors (JAKi) have been developed and option for people with HIV (PWH) is a life-long suppressive approved for the treatment of inflammatory diseases as antiretroviral therapy (ART), albeit long-term side effects and myelofibrosis and rheumatoid arthritis (19). Recently, JAKi development of treatment resistance hampering treatment received emergency use approval to treat COVID-19, due to efficacy, and immune dysfunction persists indefinitely (1, 2). their potent efficacy in reducing type I IFN-driven inflammation One strategy for eliminating latently infected cells is the “shock in vivo, paving the way towards their use as treatment strategies and kill”, which aims to activate virus transcription, protein against other viral infections, including HIV-1 (20). JAKi expression and virion production using latency reversal agents treatment leads to the inhibition of various proinflammatory (LRAs). Then, reactivated cells will be subsequently eliminated cytokines, including interleukin-6 (IL-6), tumor necrosis factor by either viral cytopathic effect or host cytolytic effector alpha (TNF-a), and IL-1 among others (21). In clear contrast to mechanisms (3). Although some of these LRAs have shown their proposed antiviral role, a commonly reported effect of JAKi promising potential in reversing HIV-1 latency in vitro, reducing treatment in humans is the increased risk of reactivation of the size of the viral reservoirs has remained unsuccessful in chronic viral infections [reviewed in (22)]. Thus, selective JAKi clinical trials due to the complexities of mechanisms governing might also induce reactivation of latent HIV while preventing latency reactivation (4, 5). Conversely, the permanent silencing chronic inflammation, representing therefore an interesting new of the latent reservoir and HIV transcription have also been strategy for HIV cure and prompting an in-depth evaluation of explored in recent years, the so called “block and lock” strategy, their putative role in chronic HIV infection. which aims to establish a state of deep latency (6). Here, we show that a specific subclass of selective JAK2i Latently infected cells can escape the viral immune response reverses HIV latency in vitro and ex vivo in CD4+ T cells from and persist for long periods. Latency is primarily established in ART-suppressed PWH. Evaluation of the underlying host infected activated CD4+ T cells as they transition to a resting cellular factors involved, identified interferon regulatory factor memory state (7). HIV-1 persists in all subsets of memory CD4+T 7 (IRF7) as a novel protein putatively relevant for HIV latency cells, including memory stem cells (TSCM), central memory cells reversal, an effect that was independent of JAK-STAT signalling (TCM), transitional memory cells (TTM) and effector memory cells blockade. Overall, these data demonstrate that targeted (TEM) (8–10). These subsets of memory CD4+ T cells show inhibition of the JAK-STAT pathway may provide a selective, distinct phenotypes in their transcriptional activity, HIV effective, and novel mechanism for purging HIV-1 reservoir in proviral inducibility, and contribution to the latent reservoir lymphocytes and macrophages, having at the same time a (11–14). Besides resting memory CD4+ T cells, cells of the significant impact on HIV-1 induced chronic inflammation. myeloid lineage, especially macrophages, are believed to be an important sanctuary for HIV-1, due to their ability to spread HIV- 1 infection in immune-privileged sites (15). Monocytes and Results macrophages are key players in the innate immune response to pathogens and are recruited to sites of infection and Identification of specific JAK2 inhibitors inflammation. While the establishment of a viral set-point as latency reversing or latency coincides with the downregulation of acute innate immune promoting agents in non-clonal cellular responses, systemic inflammation persist during chronic HIV models of HIV-1 latency infection and, although reduced in PWH on ART, immune activation persists, closely associated with T-cell exhaustion and To identify new targets for HIV latency reversal or compromising CD4+ T-cell recovery and immune function (16). blockade, we developed two non-clonal in vitro models of The Janus kinase–signal transducer and activator of HIV-1 latency based on lymphoid (Jurkat) and myeloid (HL- transcription (JAK-STAT) pathway is an essential mediator of 60) cell lines infected with a GFP-expressing HIV-1 virus (J- host immune response to viral infections, including HIV, HIG and HL-HIG) (Figure 1A). Using these models, we regulating cytokine secretion and induction of antiviral assessed the latency reversal capacities of a panel of JAKi interferon (IFN) stimulatory genes. Cytokines and IFNs are with different selectivity indexes for the JAK family among the causative orchestrators of chronic inflammation (Figure 1B). The histone deacetylase inhibitors (HDACi) associated with viral reservoir formation and maintenance. vorinostat (VOR) and panobinostat (PNB) were used as Frontiers in Immunology 02 frontiersin.org Ezeonwumelu et al. 10.3389/fimmu.2022.1001068 positive controls for latency reactivation. A specific subclass of control of immune activation, prior to immunophenotypic selective JAK2i, fedratinib and its structural analogue characterization by flow cytometry (Figure S2). Ex vivo, TG101209, showed potent latency reactivation capacity in CD4+T cell immune activation was not induced upon JAKi both models of HIV-1 latency (2-fold increase, p<0.01), treatment and significant reductions in the frequency of although two other JAKi, AZD-1480 and AZ-960, were able activation markers HLA-DR and CD25 was observed to reverse latent HIV-1 only in the lymphocyte model (Figure 2B), a condition which has been linked to increased (Figure 1C). Conversely, the JAK2i pacritinib significantly susceptibility of CD4+T cells to productive HIV-1 infection (27). blocked HIV latency reversal in both models (30% decrease, However, in contrast to the other JAKi, treatment with p<0.01), and the pan-JAKi ruxolitinib blocked HIV-1 latency fedratinib slightly induced the early activation marker CD69 reversal only in myeloid cells, suggesting a putative role of but markedly less than the immune activation observed upon pacritinib as a latency promoting agent (LPA). Expanded dose- treatment with PMA+ionomycin (Fold change FC=2, p=0.002 vs response testing of selective JAK2i confirmed screening results, FC=89, p<0.001). showing significant viral latency reactivation or blockade in a Then, we characterized the relative abundance of the distinct dose-dependent manner (Figure 1D). Comparatively similar CD4+ T cell subsets treated with JAKi, as distinct CD4+ T cell tendencies in latency reactivation were also observed in the populations differentially contribute to HIV-1 reservoir (28). well-established latency cell model ACH2, and in an additional Phenotypic characterization of the various CD4+ T-cell myeloid latency model U-HIG (Figure S1). Results in the J- populations showed no significant changes in the quiescent HIG and HL-HIG models were further confirmed by naïve (TN) and TCM CD4+ T cell phenotype upon treatment measuring the production of p24 capsid viral protein with all JAKi relative to the untreated control (Figures 2C, D). (CAp24) through viral protein spot (VIP-SPOT) assay, an However, we observed significantly higher enrichment of TEM enzyme-linked ImmunoSpot (ELISpot) approach (23) CD4+ T cell phenotype in fedratinib treated lymphocytes (Figure 1E and Figure S1), further suggesting that the use of (FC=2.3 vs. untreated control, p<0.01) compared to either specific subclass of JAK2i may represent a novel strategy for pacritinib or ruxolitinib (FC=1.4 vs. untreated control, purging the HIV reservoir. p<0.01), recapitulating previous observation that T EM To determine the contribution of JAK2 as the target cellular phenotype potentiates HIV-1 latency reversal in CD4+ T cells factor affecting viral latency, we evaluated the latency reversal (14). Notably, fedratinib favoured TEM enrichment over TTM capacity of fedratinib in JAK2 depleted cells. Knockdown of CD4+ T cell enrichment, with a concomitant reduction in the JAK2 by siRNA did not affect HIV latency reversal capacity TTM phenotype showing a similar trend than PMA+ionomycin, neither in fedratinib nor in phorbol 12-myristate 13-acetate although to a much lesser extent. (PMA) treated cells (Figure 1F), suggesting that JAK2 function is not affecting HIV-1 latency and thus, pointing to a distinct mechanism of action. Gene expression changes following JAK2i treatment are comparable to common LRAs Fedratinib induces HIV-1 reactivation in the absence of immune activation in To delineate the mechanism driving JAK2i-induced effects on CD4+ T lymphocytes ex vivo HIV-1 latency, whole transcriptome profiling was performed on our myeloid latency in vitro model, treated with the LRA To evaluate the LRA capacity of JAK2i ex vivo, purified fedratinib, the LPA pacritinib, the pan-JAKi ruxolitinib and resting CD4+ T cells from infected individuals on ART were PMA as a positive control of latency reactivation and immune treated with fedratinib, the best performing JAK2i and viral activation. Hierarchical clustering of all samples using the union RNA production in culture supernatant was quantified using an of all differentially expressed genes (DEGs) comparing each ultra-sensitive nested qPCR (24). Characteristics of study condition to the control revealed distinct genetic signatures participants are summarized in Table S1. In concordance with amongst the JAKi (29) (Figure 3A). Interestingly, within the in vitro data, fedratinib was able to ex vivo reverse HIV-1 latency JAKi cluster, the LRA fedratinib showed the highest similarity to a similar extent than in non-clonal in vitro cell line with PMA, the condition with the highest genetic perturbations in models (Figure 2A). contrast to the LPA pacritinib that was identified as the condition Global T-cell activation is generally detrimental for in vivo with the lowest gene expression changes, being similar to the use, generating interest in LRAs that do not activate T cells (25, untreated control (Figure 3B). A closer study of overlapping DEGs 26). To evaluate JAK2i impact on T-cell activation, primary between the LRAs fedratinib and PMA revealed co-upregulation CD4+T cells from uninfected donors were treated ex vivo of several host genes associated with HIV-1 replication cycle, with the selective JAK2i, fedratinib and pacritinib, the including TNF-a, TNF-associated immediate early response gene pan-JAKi ruxolitinib or PMA+ionomycin as a positive 3 (IER3) and CDKN1A (encoding the cyclin-dependent kinase Frontiers in Immunology 03 frontiersin.org Ezeonwumelu et al. 10.3389/fimmu.2022.1001068 A B C D E F FIGURE 1 Identification of specific JAK2 inhibitors as latency reversing or latency promoting agents in non-clonal cellular models of HIV-1 latency. (A) Generation of non-clonal models of HIV-1 latency for identifying new latency reversal agents (LRA) by flow cytometry. Latently infected models of HIV-1 latency were generated by infecting Jurkat (J-HIG) and HL-60 (HL-HIG) cell lines with the envelope-deficient HIV-1 NL4-3 clone (HIG) encoding GFP. Viral reactivation was measured as the proportion of GFP+ cells 20 hours post incubation with compounds. (B) Janus kinase (JAK) selectivity by JAK inhibitors (JAKi) included in the LRA screening. Selectivity refers to 50% inhibitory concentrations (IC50) of corresponding enzymatic activities measured in cell free assays. (C) HIV-1 latency reactivation capacity of JAKi in J-HIG (left panel) and HL-HIG (right panel) models. Bar plots represent relative HIV-1 reactivation and red dots represent cell viability of treatment conditions normalized to the untreated control (DMSO). (D) Expanded screening of JAK2 specific inhibitors in J-HIG (left panel) and HL-HIG (right panel) models. Cells were treated with compounds at the indicated concentrations, as in (C). (E) Viral protein production in the J-HIG (left panel) and HL-HIG (right panel) models measured by quantifying HIV-1 CAp24 antigen production 48 h post treatment with indicated compounds using VIP-SPOT assay. (F) Effect of JAK2 siRNA knockdown on latency reactivation capacity of LRAs. JAK2 knockdown was confirmed by western blot. Bar plots represent relative HIV-1 reactivation of siRNA JAK2 (siJAK2) and non-targeting (siNT) cells, treated or not with fedratinib at 5 µM or PMA at 100 ng/ml. Data are expressed as mean ± SD of at least three independent experiments. ND, no drug; VOR, vorinostat. ∗p<0.05; ∗∗p<0.01; ∗∗∗p<0.001. not significant (ns) for p>0.05. p21) (Figure 3C and Figure S3) (30), an effect that was further (31). Enriched gene-sets for each treatment condition relative confirmed in uninfected HL-60 and Jurkat cell lines, linking its to the untreated control were annotated and hierarchically expression changes to fedratinib treatment itself (Figure S3). clustered with respect to their normalized enrichment scores. Next, we performed gene-set enrichment analysis (GSEA) Overall, JAKi induced global downregulation of signalling using the hallmark gene-sets to elucidate signalling pathways processes, including several pathways involved in cellular involved in HIV-1 latency modulation by these compounds metabolism (Figure 3D). Unexpectedly, the pathways Frontiers in Immunology 04 frontiersin.org Ezeonwumelu et al. 10.3389/fimmu.2022.1001068 A B C D FIGURE 2 Fedratinib induces HIV-1 reactivation in the absence of immune activation in primary CD4+ T lymphocytes ex vivo. (A) Quantification of HIV-1 RNA in the cell culture supernatant of isolated CD4+ T cell from ART suppressed HIV-1+ individuals treated with fedratinib. PMA+Iono was used as a positive control for HIV-1 latency reactivation. Absolute HIV-1 copy number (copies/ml) for each HIV-1+ individual are represented as bar plots, and fold changes (FC) relative to the corresponding untreated replicate (ND) are indicated above treatment bars. Undetermined ct values (UD), and FC values if not available (na) are highlighted in the chart. (B) Immune cell activation of JAKi-treated CD4+ T cells from healthy donors (n = 5). CD4+ T cells were stained with early (CD69) and late activation (HLA-DR and CD25) markers for flow cytometry 72 h post treatment with compounds. (C) Frequency of major CD4+ T cell subsets from JAKi-treated CD4+ T cells from healthy donors. TN, naïve T cells; TCM, central memory T cells; TTM, transitional memory T cells; TEM, effector memory T cells. (D) Corresponding contribution of individual CD4+ T cell subsets to the major CD4+ T cell pool analysed in (C). Numbers inside sections indicate percentages. Data are pooled from the five healthy donors from (C). Bars show mean ± SD ∗p<0.05; ∗∗p<0.01; ∗∗∗p<0.001. positively enriched upon fedratinib and PMA treatment (the as PMA (Figure 3E, top panel). Moreover, out of all the majority of the genes in these pathways are upregulated), were treatment conditions, only fedratinib significantly enriched mainly downregulated in the other JAKi tested. These include the Reactome HIV budding and maturation gene-set (Figure gene-sets up-regulated in response to ultraviolet light S4), confirming its LRA capacity to induce viral protein (UV_RESPONSE_UP), apoptosis (APOPTOSIS) and a group production (Figure 1E). of pathways related to innate immune signalling as TNF-alpha and TGF-beta signalling and IL6-JAK-STAT3 pathway, being contrary to its reported function. Reactome GSEA revealed Fedratinib upregulates IRF7 expression similar gene-set enrichment patterns between PMA and despite blockade of JAK-STAT signalling fedratinib as the hallmark GSEA, further demonstrating the and cytokine production difference between fedratinib and the other JAKi tested (Figure 3E). Notably, fedratinib treatment resulted in the Taking into account the differences observed between least modulation of signalling pathways compared to other fedratinib and other JAKi, and the distinct observed effects on JAKi and PMA. In addition, the gene-set enrichment profile of JAK-STAT signalling pathway, we conducted a more detailed fedratinib mirrored that observed upon PMA treatment with evaluation of the pathway. Examination of leading-edge genes of twice as much tendency for up-regulating signalling pathways the JAK-STAT hallmark gene-set indicated that none of the Frontiers in Immunology 05 frontiersin.org Ezeonwumelu et al. 10.3389/fimmu.2022.1001068 A B C D E FIGURE 3 Gene expression changes following fedratinib treatment are comparable to common LRAs. (A) Heat map representation of gene expression changes after treatment of HL-HIG cells with JAKi and PMA and untreated control (ND). Drugs were used at concentrations previously shown to be effective at reversing or promoting latency in our model. Heat map for the different treatment conditions was generated by unsupervised hierarchical clustering of significantly expressed genes (p < 0.05). (B) Volcano plots of differentially expressed genes (DEG) for each treatment condition relative to the untreated control (ND). DEGs were selected based on Log2 gene expression (Log2FC) and p < 0.05. (C) Venn diagram of overlapping DEGs among treatment conditions shown in (B). (D) Gene set enrichment analysis (GSEA) for the different treatment conditions using Hallmark genesets. Heatmap visualization of the transcriptional signatures in the distinct treatment conditions was generated using an unsupervised hierarchical clustering of normalized enrichment scores (NES). (E) Reactome gene set enrichment map of significantly enriched pathways. The enrichment patterns for the Reactome gene set are highlighted and nodes manually laid out for clarity (full annotation of gene set enrichment map for fedratinib treatment condition is shown in Figure S4). Node size represents number of genes, node colour represents significance (NES), and edge thickness represents number of shared genes. Significantly down- or up-regulated genes (A, B) and gene sets (D, E) are highlighted in blue or red, respectively. Transcriptomic data represents RNA sequencing analysis from three independent experiments. JAKs was enriched, but instead a significant upregulation of pJAK2, pSTAT1 (Figure 4B) and IFN-stimulated genes as IL6, STAT3 and IRF1 was observed (Figure 4A). To confirm the CXCL10 and IL8 in multiple cell lines (Figure 4C, Figure S5A), RNA-Seq data, the modulation of JAK-STAT signalling pathway indicating efficient blockade of the JAK-STAT signalling was validated in cells treated with either fedratinib or ruxolitinib. pathway. Similar results were obtained in A549 dual cells, As expected, all JAKi induced a potent downregulation of where fedratinib blocked IFN-sensitive response element Frontiers in Immunology 06 frontiersin.org Ezeonwumelu et al. 10.3389/fimmu.2022.1001068 (ISRE) and NF-kB induced transcription, measured by According to the classical model of JAK-STAT signalling corresponding reporter genes, despite enrichment of the TNFa pathway, transcriptional activity of the IRFs is induced signalling pathway (Figure S5B-D). However, fedratinib downstream of crucial immune sensors for transcription of upregulated the expression of the interferon regulatory factors type I IFN, that subsequently lead to cytokine and IFN- IRF1 and IRF7 (Figure 4B), although only IRF7 expression was induced phosphorylation of the JAKs and STATs, leading to differentially modulated, depending on the JAKi used, ranging the transcription of ISGs and other target genes. Hence, we between no change to 8 FC from pacritinib to fedratinib, further evaluated the dependency of fedratinib-induced IRF7 respectively (Figure 4D). We also observed that the expression expression and classical JAK-STAT signalling over time, using of IRF7 positively correlated with other transcriptional lipopolysaccharide (LPS) as a positive control for activation of activating IRFs, i. e., IRF1, IRF3, IRF5 and IRF9, but not with the JAK-STAT signalling pathway. Interestingly, LPS failed to the transcriptional repressor IRF8 (Figure 4E) (32). induce IRF7 expression despite clearly inducing activation of the A B C F D E G H I FIGURE 4 Fedratinib upregulates IRF7 expression despite blockade of JAK-STAT signalling and cytokine production. (A) Gene expression of leading-edge genes from the Hallmark IL6-JAK-STAT3 signalling gene set. Fedratinib treatment condition was used as a reference set for leading-edge genes selection and ranking. (B) Evaluation of JAK-STAT signalling pathway at protein level in JAKi-treated HL60 cells using western blot analysis. GAPDH was used as a loading control. (C) Gene expression of IL6 and CXCL10 following treatment with JAKi. Relative mRNA expression was measured by quantitative RT-PCR and normalized to GAPDH. (D) Heatmap showing all differentially expressed interferon regulatory factors (IRF) from whole transcriptome analysis. (E) Spearman correlation matrix for differentially expressed IRF. Negatively and positively correlated gene expression of IRF in the transcriptomic data are shown in blue and red, respectively. (F) Western blot analysis showing time dependence of IRF7 induction by fedratinib despite consistent blockade of JAK2 phosphorylation. LPS treatment was used as a positive control for the activation of the JAK-STAT pathway (induction of pJAK2), and HSP90 as a loading control. Induction of IRF7 gene expression was confirmed by qRT-PCR. (G) IRF7 activation measured by protein expression of phosphorylated IRF7 (pIRF7) in HL60 cells treated with either fedratinib or LPS for 24 h A representative experiment is shown. Bar plots represent the mean quantification of pIRF7 bands obtained by densitometry analysis of three independent experiments. Values were normalized to that of Hsp90 used as a loading control and relativized to the untreated condition (ND). (H, I) Dose-response induction of IRF7 expression upon 24 h fedratinib treatment in Jurkat cells by Western blot analysis (H) and qRT-PCR (I). Data are expressed as mean ± SEM of at least three independent experiments. ∗p<0.05; ∗∗p<0.01; ∗∗∗p<0.001. #UD - undetermined cycle threshold (below limit of detection). Frontiers in Immunology 07 frontiersin.org Ezeonwumelu et al. 10.3389/fimmu.2022.1001068 JAK-STAT pathway (phospho-JAK2 induction). On the Induction of IRF7 expression contrary, fedratinib treatment significantly upregulated IRF7 positively correlates with latency expression despite blocking JAK2 activation in a time- reversal modulation dependent manner (Figure 4F). Interestingly, fedratinib also upregulated IRF7 phosphorylation, indicative of increased To further investigate IRF7 role in latency reversal, IRF7 IRF7 activation which allows translocation to the nucleus and expression was knocked-down by siRNA in myeloid HIV-1 at its turn that might impact viral transcription (Figure 4G). latently infected cells. IRF7 knockdown significantly impaired Significant induction of IRF7 expression was also observed in a the HIV-1 latency reversal capacity of fedratinib (p = 0.005, dose-response treatment of fedratinib in the lymphoid Jurkat cell Figure 5A), confirming the key role of IRF7 expression. Then, line (Figures 4H, I). Overall, our results suggest that fedratinib modulation of IRF7 expression was evaluated in the extended induces HIV-1 reactivation through an IFN-independent panel of JAK2i, showing significant upregulation only in cells regulation of IRF7 transcription and activation. treated with the other JAK2i acting as LRAs, in contrast to the A B C D E F G FIGURE 5 Modulation of IRF7 expression correlates with latency reversal capacity. (A) Effect of IRF7 siRNA knockdown on latency reactivation capacity of LRAs. IRF7 knockdown was confirmed by western blot (pictograph). Bar plots represent relative HIV-1 reactivation of IRF7 knockdown (siIRF7) and non-targeting (siNT) control cells, treated or not with fedratinib at 5 µM or PMA at 0.1µg/ml. (B) IRF7 protein expression in HL60 cells treated with the distinct JAK2i. A representative experiment is shown. Bar plots represent the mean quantification of bands obtained by densitometry analysis of 3 independent experiments. Values were normalized to that of Hsp90 used as a loading control and relativized to the untreated condition (ND). (C-E) Gene expression of IRF7 and CXCL10 in JAK2i -treated HL-60 cells. Relative mRNA expressions are expressed as fold-change differences in expression of gene-of-interest relative to GAPDH gene. (D, F) Correlation plots of IRF7 and CXCL10 gene expression (C, D) versus HIV-1 latency reversal capacity of JAK2i-treated cells. (G) Fedratinib induces Tat-dependent viral expression of HIV-1 LTR promoter and correlated with IRF7 expression. HeLa TZM-bl cells were transfected with Tat-expressing plasmid for 24h. HIV-1 transactivation was measured by a luciferase-based assay 20 hours post treatment with compounds. Left panel, relative Tat-induced HIV-1 transactivation upon JAK2i treatment. Right panel, correlation between HIV-1 transactivation and IRF7 expression (see also Figure S5F). Values were normalized to the untreated control. Data are expressed as mean ± SD of at least three independent experiments. ∗p<0.05; ∗∗p<0.01; ∗∗∗p<0.001. not significant (ns) for p>0.05. Frontiers in Immunology 08 frontiersin.org Ezeonwumelu et al. 10.3389/fimmu.2022.1001068 significant downregulation of IRF7 by the LPA Pacritinib, at model, being negatively correlated with its latency reactivation both protein (Figure 5B) and mRNA expression level capacity. Importantly, all common LRAs significantly induced (Figure 5C). Indeed, the latency-reversing capacity of the CXCL10 production with a strong positive correlation, in distinct JAK2i tested, directly correlated with IRF7 expression contrast to the JAK2i-LRA (Figure 6C). Altogether, these data (rho=0.86, p=0.0056, Figure 5D). In contrast, treatment with suggest that although IRF7 role in latency reversal is not JAK2i-LRAs significantly blocked CXCL10 production common to all LRAs, it is not limited to JAK2i, indicating a (Figure 5E), providing additional proof of the independent novel but yet unrecognized mechanism that might be relevant role of anti-inflammatory signalling and IRF7 induction and for the development of novel therapeutic strategies directed indicating an inverse correlation with latency reactivation in our towards HIV cure. model (rho=-0.715, p=0.046, Figure 5F). Overall, these results suggest a direct effect of IRF7 expression on HIV-1 latency reversal, putatively through LTR transcriptional induction. Fedratinib is a potent antiviral agent in Regarding the viral proteins studied till date, HIV-1 Tat has acute HIV infection attracted more attention in viral latency because it potently plays a crucial role in viral transcription regulation. Thus, to further Several JAKi, including ruxolitinib and tofacitinib, are potent investigate the role of IRF7 in HIV-1 viral transcription, an HIV- inhibitors of HIV-1 replication (17, 18, 33). This observation raises 1 Tat transactivation assay was performed in HeLa TZM-bl cells. the possibility that fedratinib, besides reversing HIV-1 latency, TZM-bl cells harbour an integrated copy of HIV-1 LTR, might also block new HIV-1 infections, a highly desired effect in controlling luciferase reporter gene expression, which mimics a clinical setting. Thus, we evaluated the antiviral activity of selective a latent integrated provirus. First, TZM-bl cells were validated as JAK2i fedratinib and AZD1480, and the pan-JAKi ruxolitinib, in an appropriate model for measuring Tat-dependent viral acute HIV-1 infection in primary CD4+T cells from healthy donors reactivation by dose-dependent transfection of a Tat (n=3). Fedratinib and other JAKi potently blocked HIV-1 expression plasmid in the presence of PMA, as a positive replication in acute HIV-1 infection (Figure 7A). Interestingly, control of transcription induction (Figure S5E). Then, Tat fedratinib was the most potent antiviral agent of all the JAKi expressing TZM-bl cells were treated with JAK2i to determine tested (EC50 = 0.285µM, Figure 7B). the role of fedratinib-induced IRF7 expression on Tat-dependent viral reactivation. IRF7-dependent latency reactivation was observed upon treatment with JAK2i, with significant Discussion induction in Tat-mediated HIV-1 transcription by fedratinib in contrast to pacritinib (p=0.003 vs p=0.079, Figure 5G and Current ART can achieve long-term viral suppression, S5F). More importantly, IRF7 expression levels significantly although with several limitations, including the inability to correlated with Tat-mediated transactivation (Figure 5G, right cure HIV infection due to the presence of latent reservoirs and panel), further demonstrating the role of IRF7 in Tat-mediated failure to address HIV-driven inflammation. Over the last years, HIV-1 transcription. various LRAs have been explored as putative cure strategies, with several PKC activators and HDAC inhibitors undergoing clinical trials for the elimination or reduction of latent reservoirs, but IRF7 is a general regulator of with mostly unsatisfactory curative effects (3, 34, 35). Hence, the HIV-1 latency identification of new LRAs capable of reversing HIV-1 latency without causing global T-cell activation and reducing To investigate whether IRF7 plays a more general role in inflammatory signalling as the specific subclass of JAK2 HIV-1 latency, we screened a panel of common LRAs for IRF7 inhibitors here characterized, are of key importance to achieve dependency, including the HDACi vorinostat (VOR) and a functional cure. Moreover, functional characterization of Panobinostat (PNB), the BET bromodomain inhibitor JQ1, latency reversal activity of JAK2i allowed the identification of TNF-a, and PMA. As expected, all tested LRAs showed IRF7 as a new common cellular factor controlling HIV-1 latency significant latency reactivation capacity in the HL-HIG model in both lymphoid and myeloid cell lines. (p<0.001, Figure 6A). JQ1 and PMA treatment significantly Current LRAs are mainly designed to reactivate the HIV-1 upregulated IRF7 expression in a dose-dependent manner and provirus in CD4+ T cells, but their ability to abolish viral latency in the case of JQ1, positively correlated with its latency in myeloid-derived cells is largely unknown. Moreover, the reactivation capacity (rho=0.892, p=0.001 for JQ1, Figure 6B). resistance of macrophages to HIV-1 mediated killing and the Conversely, no effect was observed for TNF-a or PNB, and the presence of infected macrophages in immune-privileged regions, HDACi VOR significantly blocked IRF7 induction in the HL60 including the central nervous system, may pose a barrier to Frontiers in Immunology 09 frontiersin.org Ezeonwumelu et al. 10.3389/fimmu.2022.1001068 A B C FIGURE 6 Latency reactivation by common LRAs positively correlates with the induction of ISG. (A) Dose-response latency reversal activity by distinct well-described LRAs in a model of HIV-1 latency (HL-HIG). Bar plots represent relative HIV-1 reactivation normalized to the untreated control (ND). (B) Top panels, IRF7 gene expression upon treatment with the indicated LRAs. Relative mRNA expressions are expressed as fold-change differences in expression of gene-of-interest relative to GAPDH gene. Bottom panels, Pearson correlation plots of HIV latency reversal activity versus IRF7 gene expression. (C) Top panels, CXCL10 gene expression upon treatment with the indicated LRAs. Relative mRNA expressions are expressed as fold-change differences in expression of gene-of-interest relative to GAPDH gene. Bottom panels, pearson correlation plots of HIV latency reversal activity versus CXCL10 gene expression. All statistical comparisons were performed with Student’s t tests. *p<0.05; **p<0.01; ***p<0.001. Data are expressed as mean ± SD of at least three independent experiments. eliminating infected cells by current “shock and kill” strategies inhibitors with broad and potent latency reactivation activity. (28). Recent advances have highlighted the importance of JAKi are FDA-approved drugs for the treatment of myelofibrosis myeloid reservoirs as sanctuaries of HIV persistence (28); and rheumatoid arthritis. The distinct JAKi differ in their however, current techniques are primarily based in T cells and selectivity for the JAKs, JAK1-3 and TYK2, with demonstrable have not been able to comprehensively and specifically map the phenotypic differences in their different modes of action (29). myeloid HIV reservoir, due to its much smaller size and Interestingly, reactivation potential was not common to all JAK the more widely dispersion throughout the body, highlighting inhibitors tested, putatively due to their distinct selectivities and the need of novel in vitro and ex vivo experimental models. Thus, more relevant, to the effects exerted over other host signalling our initial in vitro screening was performed in parallel in non- pathways (29, 36). Previous studies have identified JAK1/2i as clonal myeloid and lymphoid models of HIV-1 latency, which antiviral agents against HIV infection, blocking both latent and allowed the identification of a specific subclass of JAK2 productive infection, in line with our findings for the pan-JAK Frontiers in Immunology 10 frontiersin.org Ezeonwumelu et al. 10.3389/fimmu.2022.1001068 A B FIGURE 7 Fedratinib is a potent antiviral agent in acute HIV infection, specifically targeting multispliced HIV RNA transcripts. (A) Antiviral activity of JAKi in acute HIV infection of activated CD4+T cells from healthy donors (n = 3). HIV-1 infection was measured as proportion of GFP+ CD4+ T cells after infection with GFP-encoding NL4-3 env-deficient HIV-1 clone, 48 h post treatment with compounds at the indicated concentrations. Bar plots represent relative HIV infection and dot plots cell viability of corresponding treatment conditions normalized to the untreated control (DMSO). (B) EC50 and CC50 values of JAKi in acute HIV-1 infection model. Values were calculated from (A). EC50, effective concentration required to block HIV-1 replication by 50% and CC50, concentration required to induce 50% of cell death in cell culture. (Values are expressed as mean ± SD. **p<0.01; ***p<0.001. inhibitor ruxolitinib and JAK2i pacritinib (17, 33). In contrast, preclinical and clinical toxic events, thus, the fewer off-targets, the the JAK2i fedratinib and its analogue TG101209 significantly better for its clinical applications. induced HIV-1 reactivation both in vitro and ex vivo, an effect The mechanisms for the establishment, maintenance and that was independent of their inhibitory activity on the JAK- reactivation of HIV-1 latency mainly operate at the STAT pathway and subsequent anti-inflammatory properties. transcriptional level by both viral (38–41) and host machinery, Indeed, we found that in vitro treatment of primary lymphocytes usually through chromatin modification and epigenetic with fedratinib did not induce major changes in immune cell regulations (40, 42–44). Thus, based on pathway analyses, in activation or CD4+ T cell populations, except for enrichment in depth evaluation of specific candidate genes responsible for the proportion of CD4+ T effector memory cells. Emerging fedratinib latency reactivation activity focused on pathways evidence suggests that agents impairing differentiation into T controlling host cell transcription, leading to the identification effector memory phenotype in vivo block latency reactivation of IRFs as putative targets, and among them IRF7, showing the (37), highlighting the importance of effector memory cells as an most significant change in gene expression, both in lymphoid important latent reservoir for eradication strategies within the and myeloid cells. IRFs are a family of transcription factors that context of “shock and kill”. play pivotal roles in many aspects of the immune response, The identification of FDA-approved compounds such as including immune cell development and differentiation and fedratinib that can purge HIV-1 reservoir in distinct cell types regulating responses to pathogens. Among them, IRF3, IRF5, but also harbour potent immunomodulatory properties, and IRF7 are critical to the production of type I interferons specifically counteracting hyperinflammatory responses and downstream of pathogen recognition receptors that detect viral restoring normal immune signalling, represents a valuable asset RNA and DNA. IRF7 is a multifunctional transcription factor for HIV cure strategies (2). These desirable attributes prompted us (45) that exhibits broader DNA binding specificity than other to characterize its mode of action to identify the cellular pathways IRFs (46), and its expression and activation is also more tightly underlying its latency reactivation potential. Indeed, whole regulated (45). transcriptome profiling and pathway analysis confirmed the At first glance, the upregulation of IRF7 expression and singularity of fedratinib compared to other JAKi. Interestingly, activation despite the effective blockade of the IFN-induced fedratinib and PMA, but not the JAKi non-LRAs ruxolitinib and JAK-STAT signalling pathway may seem counterintuitive. pacritinib, similarly upregulated pathways associated with However, our data clearly pointed towards a new, as yet apoptosis and several immune-related signatures. More unrecognized role of IRF7, not directly linked to its canonical importantly, fedratinib was the inhibitor presenting the lowest and well-described role as a master regulator of IFN expression number of altered gene sets in the pathway analysis, suggesting the (47). In agreement with our data, an IFN-independent existence of fewer undesired off-target interactions. Small regulation of IRF7 transcription has already been described in molecule drugs interact with unintended, often unknown, the literature (47), representing an alternative mechanism for biological targets, and these off-target effects may lead to both IRF7 activation, albeit significantly less studied than the IFN- Frontiers in Immunology 11 frontiersin.org Ezeonwumelu et al. 10.3389/fimmu.2022.1001068 mediated classical signalling pathway. We provide several regulations and to the ethical principles suggested in the compelling evidences demonstrating IRF7 role in lymphoid Declaration of Helsinki. and myeloid-derived experimental models of HIV latency, Peripheral blood mononuclear cells (PBMC) from buffy including (i) specific impairment of fedratinib latency coats of healthy donors and PWH study participants were reactivation activity upon IRF7 knockdown, (ii) direct obtained by Ficoll-Paque density gradient centrifugation and correlation of IRF7 expression levels and HIV latency used for fresh purification of CD4+ T lymphocytes (StemCell reactivation and (iii) correlation between Tat-mediated HIV-1 Technologies). Purity of the CD4+ T lymphocyte population was transcription induced by JAK2i and IRF7 upregulation. Indeed, confirmed by flow cytometry. CD4+ T lymphocytes were kept in upregulation of IRF7 expression has been previously identified complete RPMI culture medium (RPMI 1640 medium in cells treated with BRD4 inhibitors acting as LRA in J-Lat supplemented with 10% heat-inactivated fetal bovine serum latently infected HIV-1 model (48). In line with this data, we also (FBS; Gibco), 100 U/ml penicillin, 100 mg/ml streptomycin observed that IRF7 expression directly correlated with HIV-1 (Gibco, Life Technologies)), hIL-2 (6.5 IU/ml, Roche) and reactivation in response to distinct LRAs, including PMA, the stimulated (donor) or not (PWH study participants) for 3 days BRD4 inhibitor JQ1 and the HDACi Panobinostat, in addition with PHA (4 µg/ml; Sigma-Aldrich). to specific JAK2i. Moreover, upregulation of IRF7 expression The human cell lines HL60, Jurkat, U937, MT-4, HEK293T, was independent of IFN-related gene expression as all LRAs ACH2 and TZM-bl cells were obtained from the AIDS Reagent except those blocking JAK function significantly up-regulated Program, National Institutes of Health (Germantown, MD). ISG expression, providing further proof of the independent role HL60, Jurkat, U937, MT-4 and ACH2 cells were cultured in of IRF7 in HIV-1 latency reactivation. Furthermore, through complete RPMI culture medium (RPMI 1640 medium genome-wide quantitative trait locus analysis in chronic HIV-1 supplemented with 10% FBS, 100 U/ml penicillin and 100 mg/ infected individuals, IRF7 has been identified as a host genetic ml streptomycin). HEK293T and TZM-bl cells were cultured in factor affecting the size and transcriptional activity of HIV-1 Dulbecco’s modified Eagle’s medium (DMEM), supplemented reservoirs, further supporting the role of IRF7 in HIV-1 with 10% FBS, 100 U/ml penicillin and 100 mg/ml streptomycin. latency (49). A549-Dual™ hACE2-TMPRSS2 cells (In vivoGen) were Overall, we have identified and characterized the JAK2i, cultured in complete DMEM as described above in addition to fedratinib, as a novel LRA. Several attributes distinguish the following antibiotics: 10 µg/ml of Blasticidin, 100 µg/ml of fedratinib as an excellent candidate for HIV cure strategy, Hygromycin, 0.5 µg/ml of Puromycin, and 100 µg/ml of Zeocin, including the demonstrated capacity in reversing HIV-1 (In vivogen). A549-Dual™ cells are adherent epithelial cells latency without causing global T-cell activation, reducing derived from the human A549 lung carcinoma cell line with a inflammatory signalling, and high specificity in the stable integration of two inducible reporter constructs for modulation of a limited number of signalling pathways. studying the expression of NFkB and interferon regulatory Moreover, functional characterization of fedratinib latency factor (IRF) pathways. reactivation capacity revealed new molecular determinants of All cell cultures were maintained at 37°C in a 5% HIV-1 latency maintenance and reactivation, specifically IRF7. CO2 incubator. Therefore, mechanistic insights into the molecular basis of the divergent roles played by IRF7 in host protection against viruses warrant further exploration in future studies. Patients and samples This study sampled a cohort of chronically infected HIV-1 Materials and methods positive individuals attending Hospital Germans Trias i Pujol, Badalona, Spain. Study participants were included if the Primary cultures and cell lines individuals were older than 18 years old and had been on suppressive ART with undetectable plasma HIV-1 RNA levels Buffy coats from healthy donors were purchased from the (<40 copies/ml) for a minimum of two years. All participants Catalan Banc de Sang i Teixits (http://www.bancsang.net/en/ were males (n=4). The median age of the participants was 45y index.html). The buffy coats received were totally anonymous (range 38-56y). Frozen PBMCs (isolated as described below and and untraceable and the only information given was whether or cryopreserved) from PWH visiting our clinic were used. not they have been tested for diseases. All donors provided Immunological and virological characteristics from all informed consent at the time of blood collection. All methods participants are found in Table S1. All participants in the study were carried out in accordance with relevant guidelines and provided informed consent, and the work was approved by the Frontiers in Immunology 12 frontiersin.org Ezeonwumelu et al. 10.3389/fimmu.2022.1001068 Scientific Committee of Fundació IrsiCaixa and the Ethics Aldrich) was used at 10 µM, raltegravir at 5 µM and efavirenz at Committee of Hospital Germans Trias i Pujol (Ref. CEI PI-18–021). 0.32 µM were obtained from the NIH AIDS Research and Reference Reagent Program. TNFa (Merck) was used at a concentration of 0.08-2 ng/ml and LPS at 100 ng/ml (Merck). Viral strains Fedratinib, TG101209, AZD1480, AZ960, gandolitinib, WP1066, XL109, ruxolitinib, momelotinib, JQ1 were used at a The envelope-deficient HIV-1 NL4-3 clone (HIG) encoding concentration of 0.2-5 µM, pacritinib at 40 nM-1 µM and Q- internal ribosome entry site (IRES)-green fluorescent protein VD-Oph at 10 µM (all from Selleckchem). Vorinostat (SAHA) (GFP) (NL4-3-GFP) (50) was pseudotyped with vesicular was used at a concentration of 0.2-5 µM (Prochifar srl, Italy) and stomatitis virus G protein (VSV-G) by cotransfection of panobinostat at 3.2-400 nM (LC Laboratories). HEK293T cells using polyethylenimine (Polysciences) as previously described (51, 52). HIV-1 stock of the fully replicative NL4-3 clone (NL4-3) was grown in lymphoid MT-4 VIP-SPOT assay cells. Three days after transfection, supernatants were harvested, filtered and stored at -80°C. Viral stocks were concentrated using ELISpot plates (Immobilon-P polyvinylidene difluoride Lenti-X concentrator (Clontech). Viruses were titrated by membrane, Cat#MSIPS4W10; Millipore) were coated in infection of TZM-bl cells followed by GFP quantification by advance with 10mg/ml capture antibody (mouse monoclonal flow cytometry or qPCR for HIG or NL4-3, respectively. antibody to HIV p24; clone 39/5.4A, Cat#ab9071; Abcam). On the day of cell culture, the wells were washed with sterile PBS and incubated with 100ml of a blocking solution containing 5% Generation of non-clonal HIV bovine serum albumin (BSA) (MACS BSA stock solution, latently infected cells and Cat#130-091-376; Miltenyi) in PBS, for at least 30 min at viral reactivation assays room temperature. This was followed by additional washing of wells and replacement with culture medium containing HL-HIG Latently infected cells (J-HIG, HL-HIG and U-HIG) were or J-HIG cells at 9 x 103 cells per well and compounds. After 48 h generated by infecting lymphoid (Jurkat) and myeloid (HL60 of incubation at 37°C in a 5% CO2-humified atmosphere, the and U937) cells with HIG virus and maintained in culture for 10 plates were developed as described elsewhere (23). After drying, days to allow for the attrition of productively infected cells as the spots were counted using an automated ELISpot reader unit previously described (53). HIV-1 reactivation was measured as (Cellular Technology Limited, Shaker Heights, OH). the percentage of GFP + cells by flow cytometry 20 h post incubation with the compounds relative to untreated control (DMSO). The capacity of known and putative LRAs to induce HIV-1 expression was confirmed in latently infected ACH2 cells Ex vivo reactivation of primary CD4+ T by measuring intracellular HIV CAp24 by flow cytometry 48 h cells from PWH after treatment. For evaluation of cell death, cells were stained for 30 min Purified CD4+ T lymphocytes from HIV-1+ participants with LIVE/DEAD™ Fixable Near-IR Dead Cell Stain Kit were kept in complete RPMI culture medium and preincubated (Invitrogen, Thermo Fischer Scientific) in PBS according to with the pan-caspase inhibitor Q-VD-Oph (10 µM, Sigma- manufacturer’s instructions. Alternatively, viable cells were Aldrich) for 2 h. To evaluate the latency reactivation capacity, identified according to forward and side laser light scatter flow half a million CD4+ T lymphocytes were cultured with PMA + cytometry analysis. Cells were washed and fixed in 1% ionomycin or fedratinib at the indicated concentrations (5 µM formaldehyde before the analysis. Flow cytometry assays were fedratinib, and 50 ng/ml PMA + 1 µg/ml ionomycin). Cells were performed in a FACS LSR II or a FACSCanto II flow cytometer cultured in the presence of antiretrovirals (efavirenz, zidovudine (BD Biosciences). The data was analyzed using the FlowJo and raltegravir) and maintained in 10 µM Q-VD-Oph for 72 h. software (BD Biosciences). Freshly collected cell-culture supernatant were centrifuged for 1 hour at 25 000 g to pellet HIV particles. HIV-1 reactivation was determined by quantification of viral RNA in the supernatant as previously described (24). Briefly, viral RNAs Compounds and drug treatments were extracted using Viral RNA/DNA Mini kit (Invitrogen) and quantified using a nested real-time reverse transcription- PMA (Sigma-Aldrich) was used at 0.4-100 ng/ml and polymerase chain reaction (RT-PCR). A Superscript III One- ionomycin (Sigma-Aldrich) was used at a concentration of 100 Step RT-PCR system (Invitrogen) was used to generate and pre- ng/ml. 3-azido-3-deoxythymidine (zidovudine; AZT) (Sigma- amplify viral RNA with the following primers: ULF1 (forward) Frontiers in Immunology 13 frontiersin.org Ezeonwumelu et al. 10.3389/fimmu.2022.1001068 5’- ATG CCA CGT AAG CGA AAC TCT GGG TCT CTC TDG Pool (D-001810-10-05), human JAK2 siRNA-SMARTpool (L- TTA GAC - 3’; UR1 (reverse) 5’- CCA TCT CTC TCC TTC 003146-00- 0005), and human IRF7 siRNA-SMARTpool (L- TAG C -3’. The following cycling conditions were used: reverse 011810-00-0005) all from Dharmacon, Waltham, USA. transcription at 50°C for 30 min, denaturation at 94°C for 2 min, 16 cycles of amplification (94°C 15 s, 55°C 30 s, 68°C 1 min) and final elongation at 68°C for 5 min. Preamplified products were subjected to nested real-time PCR with the following primers Quantitative RT-polymerase chain and probe: LambdaT (forward) 5’- ATG CCA CGT AAG CGA reaction (qRT-PCR) AAC T - 3’; UR2 (reverse) 5’ - CTG AGG GAT CTC TAG TTA Total RNA was extracted using NucleoSpin RNA II kit CC - 3’; UHIV Taqman 5’- 56-FAM/CAC TCA AGG/ZEN/CAA (740955, Macherey-Nagel) or the Maxwell® HT simplyRNA GCT TTA TTG AGG C/3IbkFQ/- 3’ on a QuantStudio 5 PCR Kit (REF: AX2420, LOT: 0000410025, Promega) and Total DNA system (Applied Biosystems). Serial dilutions of the HIV-1 NL43 was extracted using Magmax DNA multi-sample ultra 2.0 kit strain were run in parallel with each experiment for the (A45721, appliedbiosystems) on a KingFisher ™ Flex quantification of viral RNA. Purification System (Thermofisher Scientific) as recommended by the manufacturer. Reverse transcription was performed using the PrimeScript™ RT-PCR Kit (RR036A, Takara) following Immunophenotypic characterization of manufacturer instructions. mRNA levels of all genes were PBMCs by flow cytometry measured by two-step quantitative RT-PCR and normalized to GAPDH mRNA expression using the DDCt method. Primers Purified CD4+ T lymphocytes from freshly processed and DNA probes were TaqMan Gene expression assays from PBMCs isolated from healthy donors were stained with Thermofisher Scientific (Cat#433182: Hs00266705_g1-GAPDH, CD4+T cell phenotype defining markers for flow cytometry 72 Hs01014809_g1-IRF7, Hs00171042_m1-CXCL10, h post treatment with compounds: CD4-BV786, CD45RA-PE, Hs00174131_m1-IL6, Hs00174103_m1-IL8, Hs00355782_m1- CCR7-BV510 (Biolegend); CD3-FITC, CD27-400 (BD CDKN1A, Hs00174128_m1-TNFa, and Cat#4351372: Biosciences). Immune activation was determined using CD69- Hs00931699_m1-CIITA). BV650, CD25-APC (Biolegend) and HLA-DR-PeCy7 (BD Biosciences). Immunophenotyping of CD4+ T cell population identification was performed based on the following gating strategy (Figure S2): CD4 + T cell populations were RNA-sequencing and library preparation defined based on the following expression combinations gated on the live singlet CD3+CD4+ lymphocytes: T naïve TN Cellular RNA was extracted from the latent model of HIV-1 (CD45RA+CCR7+CD27+), T central memory TCM (CD45RA- infection, HL-HIG for RNA sequencing using Maxwell® HT CCR7+CD27+), T transitional memory TTM (CD45RA-CCR7- simplyRNA Kit (Promega) on a KingFisher™ Flex Purification CD27+) and T effector memory TEM (CD45RA-CCR7+CD27-) System (Thermofisher Scientific). RNA-sequencing samples (13). Immune activation markers HLA-DR+, CD25+ and were prepared in biological triplicates. Determination of RNA CD69+ CD4+ lymphocytes were also gated on the live singlet integrity (RIN) was performed using Agilent RNA 6000 Nano CD3+CD4+ lymphocytes. Cells were washed and fixed in 1% Kit (Cat#5067-1511) and Agilent 2100 Bioanalyzer System. After formaldehyde before the analysis. Flow cytometry assays were quality control check, RNA library was constructed using performed in a FACS LSR II or a FACSCanto II flow cytometer Illumina TruSeq Stranded mRNA LT Sample Prep Kit and (BD Biosciences). The data was analyzed using the FlowJo sequencing was performed using NovaSeq 6000 System with software (BD Biosciences). 150 bp paired-ends reads. RNA interference Transcriptomic analysis Cells were transfected using Amaxa Cell line Nucleofector Transcriptomic analysis was performed as implemented in Kit V (VCA-1003, Lot F-13862, Lonza) in a 6-well plate the computational workflow for the detection of differentially following manufacturer instructions for RNAi transfections. expressed genes and pathways from RNA-seq data (54). Reads Briefly, 20 pmol of the corresponding siRNA was nucleofected were aligned to the human GRCh37/hg19 (annotation in 3 x 106 of the corresponding cell line. RNA and protein lysates NCBI_105.20190906) using HISAT2. Low-expression genes were collected 24 h post-transfection. siRNAs used for with at least one zero counts were filtered out and the transfection were ON-TARGETplus Non-targeting siRNA remaining reads normalized with Relative Log Expression Frontiers in Immunology 14 frontiersin.org Ezeonwumelu et al. 10.3389/fimmu.2022.1001068 (RLE) method as implemented in DESeq2 R library. Differential Tat-dependent viral expression of HIV-1 gene expression between the control and treatment groups was LTR promoter estimated with DESeq2 Wald test (55). Sequencing files can be accessed on gene expression omnibus repository (GSE195855). HeLa TZM-bl cells harboring an integrated copy of HIV-1 Gene Set Enrichment Analysis (GSEA) was performed on a LTR, controlling luciferase reporter gene expression was used for pre-ranked GSEA list based on Log 2FC values of differentially Tat-dependent viral transactivation assay. Cells were transfected expressed genes (DEGs: Log2FC > 1, p-value < 0.05), against or not with 10 – 50 ng of Tat-expressing plasmid (60) using Molecular Signatures Database (MsigDB v7.4) “Hallmark” and Lipofectamine 3000 reagent (Invitrogen, Cat#L3000015) in a 24- “Reactome” gene-sets. Weighted enrichment statistics were well plate following manufacturer instructions. Cells were based on 1000 permutations. Significantly enriched gene-sets treated with compounds 24 h post transfection. HIV-1 Tat with FDR adjusted q-value <0.1 were selected for Enrichment expression was then measured by a luciferase-based assay 20 h map visualization as previously described (56, 57). Briefly, post drug treatment. Enrichment files were inputted into the Enrichment Map app within the Cytoscape program for visualization (58). Parameters were set at default values (node cutoff FDR Q value 0.1; Jaccard Viral infection and antiviral activity Overlap combined coefficient cutoff 0.375, k-constant 0.5). Nodes were manually laid out and combined into common Activated CD4 + T cells from healthy donors were infected biological process for clarity using the AutoAnnotate app. with the envelope-deficient HIV-1 NL4-3 clone (HIG) encoding internal ribosome entry site (IRES)-green fluorescent protein (GFP) (NL4-3-GFP) by spinoculation (1200×g, 1 h 30 min at Immunoblot 37°C). The anti-HIV activity of the different compounds was determined by the infection of cells in the presence of different Treated cells were rinsed, lysed, subjected to SDS-PAGE and concentrations of drugs, and 50% effective concentrations (EC50) transferred to a polyvinylidene difluoride (PVDF) membrane as were calculated, as previously described (61). Viral replication was previously described (59). The following antibodies were used measured 48 h later by quantification of GFP + expression by for immunoblotting: antirabbit and anti-mouse horseradish flow cytometry. peroxidase-conjugated secondary antibodies (1:5000; Pierce); anti-GAPDH (1:2500; ab9485; Abcam); anti-human Hsp90 (1:1000, 610418, BD Biosciences); anti-pJAK2 (1:1000; 3774), Statistical analysis anti-STAT3 (1:1000; 3743), anti-phosphoSTAT1 (1:1000; 9167), anti-IRF7 (1:1000; 4920), anti-IRF3 (1:1000; 11904) anti-IRF1 Statistical significance for in vitro and ex vivo experiments (1:1000; 8478), anti-pIRF7 (1:1000; 12390) all from Cell was calculated using appropriate t test in Graphpad Prism Signalling and anti-JAK2 (1:5000; 108596; Abcam). Blots were (v9.3.0). All experiments were performed in at least three immersed in chemiluminescent substrate (SuperSignal West independent replicates and n values are provided in the figure Pico Plus or Femto, Thermo Fisher Scientific), and signal was legends. Plots were drawn using GraphPad Prism and visualized using ChemiDoc MP imaging system (BIORAD). R software. Interferon signalling Data availability statement To study the modulation of the NF-kB and interferon The datasets presented in this study can be found in online regulatory factor (IRF) signalling pathways by the JAK repositories. The names of the repository/repositories and inhibitors, 5 x 104 A549-Dual™ hACE2-TMPRSS2 cells (In accession number(s) can be found below: gene expression vivoGen, Toulouse, France) were seeded per well in flat 96-well omnibus (GSE195855). plates and incubated for 24 h with the compounds. The activities of the NF-kB and IRF signalling pathways were assessed by incubating the cell supernatants with QUANTI-Blue™ Solution Ethics statement and QUANTI-Luc ™ respectively (In vivoGen, Toulouse, France) according to manufacturer’s instructions. SEAP and The studies involving human participants were luciferase were read in an EnSight™ multimode plate reader reviewed and approved by Ethics Committee of Hospital (PerkinElmer, Waltham, MA, USA). Germans Trias i Pujol (Ref. CEI PI-18–021). The patients/ Frontiers in Immunology 15 frontiersin.org Ezeonwumelu et al. 10.3389/fimmu.2022.1001068 participants provided their written informed consent to Supplementary material participate in this study. The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/ Author contributions fimmu.2022.1001068/full#supplementary-material Study conception and design: IE, BC, ER-M, and EB; Data SUPPLEMENTARY FIGURE 1 Determination of HIV-1 reactivation capacity in in vitro models of HIV-1 collection: IE, EG-V, MP, BO-T, and LG-C; analysis and latency. HIV reactivation in in vitro lymphoid ACH2 model of latency (A) interpretation of results: MR-R, MM, JM-P, AG, RB, ER-M, and in in vitro non-clonal myeloid latency model U-HIG (B). HIV and EB; Draft manuscript preparation: IE and EB. All authors reactivation was determined by the intracellular expression of HIV-1 p24 antigen in ACH2 cells cultured for 48 h or by GFP+ expression in reviewed the results and approved the final version of U-HIG cultured for 20 h as measured by flow cytometry in the presence the manuscript. of subtoxic concentrations of indicated compounds. HDCAi panobinostat (PNB) and vorinostat (VOR) were used as controls for HIV-1 reactivation. Bar plots represent relative HIV-1 reactivation and red dots represent cell viability of treatment conditions normalized to the untreated control Funding (DMSO). Representative ELISPOT pictographs of HIV-1 CAp24 antigen capture in non-clonal models of HIV-1 latency J-HIG (C) and HL-HIG (D). Pearson correlation plots of CAp24 antigen capture (C, D) versus relative This work has been funded by Instituto de Salud Carlos III HIV-1 reactivation capacity as measured by GFP detection (flow (ISCIII) through the project PI19/00194 (Co-funded by cytometry) of indicated compounds () in J-HIG (E) and HL-HIG (F) European Regional Development Fund/European Social Fund) models of HIV latency. All statistical comparisons were performed with Student’s t tests. ∗p<0.05; ∗∗p<0.01; ∗∗∗p<0.001. Data are expressed as “Investing in your future”) and in part by Grifols. IE is a research mean ± SD of at least three independent experiments. fellow from la Caixa (LCF/BQ/IN18/11660017) cofunded by the European Union’s Horizon 2020 research and innovation SUPPLEMENTARY FIGURE 2 program under the Marie Sklodowska-Curie grant agreement Gating strategy for the immunophenotypic characterization of PBMCs by flow cytometry. CD4 + T cell populations were gated on the live singlet No. 713673. EG-V is a research fellow from PERIS, supported by CD3+CD4+ lymphocytes: T naïve TN (CD45RA+CCR7+CD27+), T central Ministry of Health of the Government of Catalonia (SLT017/20/ memory TCM (CD45RA-CCR7+CD27+), T transitional memory T TM 000090). LG-C is a research fellow supported by the Secretariat (CD45RA-CCR7-CD27+) and T effector memory T EM (CD45RA- CCR7+CD27-). Immune activation markers HLA-DR+, CD25+ and of Universities and Research of the Government of Catalonia CD69+ CD4+ lymphocytes were also gated on the live singlet and European Social Fund (2019 FI-B00420). JM-P is supported CD3+CD4+ lymphocytes. by grants PID2019-109870RB-I00 from the Spanish Ministry of SUPPLEMENTARY FIGURE 3 Science and Innovation and NIH/NIAID (RID-HIV: 1 UM1 Top 30 downregulated and upregulated genes following JAKi treatment AI164561-01). EF, RB, and EB are fellows from ISCIII (CM20/ in a model of HIV-1 latency (HL-HIG). Bar plots of top 30 differentially 00027; CP19/00011 and CPII19/00012 respectively). We thank enriched genes (DEG) for each treatment condition (A, PMA; B, Fedratinib; C, Ruxolitinib and D, pacritinib) relative to the untreated Foundation Dormeur for financial support for the acquisition of control (ND), based on Log2 gene expression (Log2FC) and p < 0.05. the QuantStudio-5 real time PCR system. Significantly down- or up-regulated DEG are highlighted in blue or red, respectively. (E, F) Confirmation of significantly top upregulated and downregulated genes in additional samples in myeloid HL60 and lymphoid Jurkat cell lines. Relative mRNA expression was measured Conflict of interest by quantitative RT-PCR and normalized to GAPDH. All statistical comparisons were performed with Student’s t tests. * p<0.05; **p<0.01; ***p<0.001. Data are expressed as mean ± SD of at least The authors declare that the research was conducted in the three independent experiments. absence of any commercial or financial relationships that could be construed as a potential conflict of interest. SUPPLEMENTARY FIGURE 4 Reactome HIV budding, and maturation signalling is significantly enriched by fedratinib treatment in latently infected model. (A) Reactome Gene set enrichment map of significantly enriched pathways for fedratinib Publisher’s note treatment condition. Reactome Gene set clusters are annotated, and nodes manually laid out for clarity (see also ). Single gene sets are highlighted and annotated in-text, underneath the cytoscape All claims expressed in this article are solely those of the enrichment map. Node size represents number of genes, node colour authors and do not necessarily represent those of their affiliated represents significance (NES), and edge thickness represents number of organizations, or those of the publisher, the editors and the shared genes. (B) Gene expression of leading-edge genes from the Reactome HIV Budding and Maturation signalling gene set. Fedratinib reviewers. Any product that may be evaluated in this article, or treatment condition was used as a reference set for leading-edge genes claim that may be made by its manufacturer, is not guaranteed selection and ranking. Significantly down- or up-regulated gene sets (A) or endorsed by the publisher. and genes (B) are highlighted in blue or red, respectively. Frontiers in Immunology 16 frontiersin.org Ezeonwumelu et al. 10.3389/fimmu.2022.1001068 SUPPLEMENTARY FIGURE 5 HIV-1 Tat transfected HeLa TZM-bl cells. HeLa TZM-bl cells were Fedratinib potently blocks cytokine and NFkB signalling pathways in vitro. transfected or not with Tat-expressing plasmid for 24h and HIV-1 Tat (A, B) IL8 and TNFa gene expression of JAKi-treated HL-60 (left panel) expression was measured by a luciferase-based assay. (F) Relative IRF7 and Jurkat (right panel) cells. Relative mRNA expression was measured by mRNA expression of non-transfected TZM-bl cells treated with JAK2i for quantitative RT-PCR and normalized to GAPDH. (C) Gene expression of 20h. ∗p<0.05; ∗∗p<0.01; ∗∗∗p<0.001. Data are expressed as mean ± SD of leading-edge genes from the Hallmark TNFa via NFkB signalling gene set. at least three independent experiments. Fedratinib treatment condition was used as a reference set for leading- edge genes selection and ranking. Significantly down- or up-regulated SUPPLEMENTARY TABLE 1 genes are highlighted in blue or red, respectively. (D) Induction of Immunological and virological characteristics of study participants at the promoters for interferon stimulatory response element (ISRE) and NFkB time of cell sample collection. * ABC, abacavir; COBI, cobicistat; DRV, by JAKi in A549-Dual™ hACE2-TMPRSS2 cells. All statistical comparisons darunavir; DTG, dolutegravir; FTC, emtricitabine; RAL, raltegravir; TAF, were performed with Student’s t tests. (E) Luciferase activity of mock and tenofovir alafenamide; 3TC, lamivudine. References 1. Solomon DA, Sax PE. Current state and limitations of daily oral therapy for and virus reactivation in vitro. Antimicrob Agents Chemother (2014) 58:1977–86. treatment. Curr Opin HIV AIDS (2015) 10:219–25. doi: 10.1097/ doi: 10.1128/AAC.02496-13 COH.0000000000000165 18. Gavegnano C, Brehm JH, Dupuy FP, Talla A, Ribeiro SP, Kulpa DA, et al. 2. Deeks SG, Archin N, Cannon P, Collins S, Jones RB, de Jong MAWP, et al. Novel mechanisms to inhibit HIV reservoir seeding using jak inhibitors. PloS Research priorities for an HIV cure: International AIDS society global scientific Pathog (2017) 13:e1006740. doi: 10.1371/journal.ppat.1006740 strategy 2021. Nat Med (2021) 27:2085–98. doi: 10.1038/s41591-021-01590-5 19. Spinelli FR, Meylan F, O’Shea JJ, Gadina M. JAK inhibitors: Ten years after. 3. Kim Y, Anderson JL, Lewin SR. Getting the “Kill” into “Shock and kill”: Eur J Immunol (2021) 51:1615–27. doi: 10.1002/eji.202048922 Strategies to eliminate latent HIV. Cell Host Microbe (2018) 23:14–26. doi: 10.1016/ 20. Marconi VC, Moser C, Gavegnano C, Deeks SG, Lederman MM, Overton ET, j.chom.2017.12.004 et al. Randomized trial of ruxolitinib in antiretroviral-treated adults with human 4. Ait-Ammar A, Kula A, Darcis G, Verdikt R, De Wit S, Gautier V, et al. immunodeficiency virus. Clin Infect Dis (2022) 74:95–104. doi: 10.1093/cid/ciab212 Current status of latency reversing agents facing the heterogeneity of HIV-1 cellular 21. Wijaya I, Andhika R, Huang I, Purwiga A, Budiman KY, Bashari MH, et al. and tissue reservoirs. Front Microbiol (2020) 10:3060. doi: 10.3389/ The use of janus kinase inhibitors in hospitalized patients with COVID-19: fmicb.2019.03060 Systematic review and meta-analysis. Clin Epidemiol Glob Heal (2021) 5. Sengupta S, Siliciano RF. Targeting the latent reservoir for HIV-1. Immunity 11:100755. doi: 10.1016/j.cegh.2021.100755 (2018) 48:872–95. doi: 10.1016/j.immuni.2018.04.030 22. Ezeonwumelu IJ, Garcia-Vidal E, Ballana E. JAK-STAT pathway: A novel 6. Moranguinho I, Valente ST. Block-and-lock: New horizons for a cure for hiv- target to tackle viral infections. Viruses (2021) 13:2379. doi: 10.3390/v13122379 1. Viruses (2020) 12:1443. doi: 10.3390/v12121443 23. Puertas MC, Bayó n-Gil Á , Garcia-Guerrero MC, Salgado M, Urrea V, 7. Siliciano JD, Kajdas J, Finzi D, Quinn TC, Chadwick K, Margolick JB, et al. Moró n-Ló pez S, et al. VIP-SPOT: an innovative assay to quantify the productive Long-term follow-up studies confirm the stability of the latent reservoir for HIV-1 HIV-1 reservoir in the monitoring of cure strategies. MBio (2021) 12:e0056021. in resting CD4+ T cells. Nat Med (2003) 9:727–8. doi: 10.1038/nm880 doi: 10.1128/mBio.00560-21 8. Chomont N, El-Far M, Ancuta P, Trautmann L, Procopio FA, Yassine-Diab 24. Vandergeeten C, Fromentin R, DaFonseca S, Lawani MB, Sereti I, Lederman B, et al. HIV Reservoir size and persistence are driven by T cell survival and MM, et al. Interleukin-7 promotes HIV persistence during antiretroviral therapy. homeostatic proliferation. Nat Med (2009) 15:893–900. doi: 10.1038/nm.1972 Blood (2013) 121:4321–29. doi: 10.1182/blood-2012-11-465625 9. Buzon MJ, Sun H, Li C, Shaw A, Seiss K, Ouyang Z, et al. HIV-1 persistence 25. Shan L, Deng K, Shroff NS, Durand CM, Rabi SA, Yang HC, et al. in CD4+ T cells with stem cell-like properties. Nat Med (2014) 20:139–42. Stimulation of HIV-1-Specific cytolytic T lymphocytes facilitates elimination of doi: 10.1038/nm.3445 latent viral reservoir after virus reactivation. Immunity (2012) 36:491–501. doi: 10.1016/j.immuni.2012.01.014 10. Cristina G, Judith G-E, Victor U, Bonaventura C, Vicenç F, José BM, et al. Atlas of the HIV-1 reservoir in peripheral CD4 T cells of individuals on successful 26. Bullen CK, Laird GM, Durand CM, Siliciano JD, Siliciano RF. New ex vivo antiretroviral therapy. MBio (2022) 12:e03078–21. doi: 10.1128/mBio.03078-21 approaches distinguish effective and ineffective single agents for reversing HIV-1 latency in vivo. Nat Med (2014) 20:425–29. doi: 10.1038/nm.3489 11. Grau-Expó sito J, Luque-Ballesteros L, Navarro J, Curran A, Burgos J, Ribera E, et al. Latency reversal agents affect differently the latent reservoir present in 27. Biancotto A, Iglehart SJ, Vanpouille C, Condack CE, Lisco A, Ruecker E, distinct CD4+ T subpopulations. PloS Pathog (2019) 15:e1007991. doi: 10.1371/ et al. HIV-1-induced activation of CD4+ T cells creates new targets for HIV-1 journal.ppat.1007991 infection in human lymphoid tissue ex vivo. Blood (2008) 111:699–704. doi: 10.1182/blood-2007-05-088435 12. Pardons M, Fromentin R, Pagliuzza A, Routy J-P, Chomont N. Latency- reversing agents induce differential responses in distinct memory CD4 T cell 28. Khanal S, Schank M, Gazzar M. El; moorman, J.P.; yao, Z.Q. HIV-1 latency subsets in individuals on antiretroviral therapy. Cell Rep (2019) 29:2783–2795.e5. and viral reservoirs: Existing reversal approaches and potential technologies, doi: 10.1016/j.celrep.2019.10.101 targets, and pathways involved in hiv latency studies. Cells (2021) 10:475. doi: 10.3390/cells10020475 13. Kwon KJ, Timmons AE, Sengupta S, Simonetti FR, Zhang H, Hoh R, et al. Different human resting memory CD4+ T cell subsets show similar low inducibility 29. Singer JW, Al-Fayoumi S, Taylor J, Velichko S, O’Mahony A. Comparative of latent HIV-1 proviruses. Sci Transl Med (2020) 12:eaax6795. doi: 10.1126/ phenotypic profiling of the JAK2 inhibitors ruxolitinib, fedratinib, momelotinib, scitranslmed.aax6795 and pacritinib reveals distinct mechanistic signatures. PloS One (2019) 14: e0222944. doi: 10.1371/journal.pone.0222944 14. Kulpa DA, Talla A, Brehm JH, Ribeiro SP, Yuan S, Bebin-Blackwell A-G, et al. Differentiation into an effector memory phenotype potentiates HIV-1 latency 30. Dı́ez-Fuertes F, de la Torre-Tarazona HE, Calonge E, Pernas M, Alonso- reversal in CD4 + T cells. J Virol (2019) 93:e00969–19. doi: 10.1128/jvi.00969-19 Socas M. Del m.; capa, l.; garcı́a-pé rez, j.; sakuntabhai, a.; alcamı́, j. transcriptome sequencing of peripheral blood mononuclear cells from elite controller-long term 15. Balcom EF, Roda WC, Cohen EA, Li MY, Power C. HIV-1 persistence in the non progressors. Sci Rep (2019) 9:14265. doi: 10.1038/s41598-019-50642-x central nervous system: viral and host determinants during antiretroviral therapy. Curr Opin Virol (2019) 38:54–62. doi: 10.1016/j.coviro.2019.06.004 31. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, et al. Gene set enrichment analysis: A knowledge-based approach for 16. Rajasuriar R, Khoury G, Kamarulzaman A, French MA, Cameron PU, interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A (2005) Lewin SR. Persistent immune activation in chronic HIV infection: Do any 102:15545–50. doi: 10.1073/pnas.0506580102 interventions work? AIDS (2013) 27:1199–1208. doi: 10.1097/QAD. 32. Antonczyk A, Krist B, Sajek M, Michalska A, Piaszyk-Borychowska A, 0b013e32835ecb8b Plens-Galaska M, et al. Direct inhibition of IRF-dependent transcriptional 17. Gavegnano C, Detorio M, Montero C, Bosque A, Planelles V, Schinazi RF. regulatory mechanisms associated with disease. Front Immunol (2019) 10:1176. Ruxolitinib and tofacitinib are potent and selective inhibitors of HIV-1 replication doi: 10.3389/fimmu.2019.01176 Frontiers in Immunology 17 frontiersin.org Ezeonwumelu et al. 10.3389/fimmu.2022.1001068 33. de Armas LR, Gavegnano C, Pallikkuth S, Rinaldi S, Pan L, Battivelli E, et al. for reactivation from other “Shock” drugs. J Virol (2018) 92:e02056–17. The effect of JAK1/2 inhibitors on HIV reservoir using primary lymphoid cell doi: 10.1128/JVI.02056-17 model of HIV latency. Front Immunol (2021) 12:720697. doi: 10.3389/ 49. Zhang Z, Trypsteen W, Blaauw M, Chu X, Rutsaert S, Vandekerckhove fimmu.2021.720697 L, et al. IRF7 and RNH1 are modifying factors of HIV-1 reservoirs: a genome- 34. Archin NM, Liberty AL, Kashuba AD, Choudhary SK, Kuruc JD, Crooks wide association analysis. BMC Med (2021) 19:282. doi: 10.1186/s12916-021- AM, et al. Administration of vorinostat disrupts HIV-1 latency in patients on 02156-5 antiretroviral therapy. Nature (2012) 487:482–5. doi: 10.1038/nature11286 50. Dapp M, Clouser CL, Patterson S, Mansky LM. 5-azacytidine can induce 35. Rasmussen TA, Tolstrup M, Brinkmann CR, Olesen R, Erikstrup C, lethal mutagenesis in human immunodeficiency virus type 1. J Virol (2009) Solomon A, et al. Panobinostat, a histone deacetylase inhibitor, for latent-virus 83:11950–8. doi: 10.1128/JVI.01406-09 reactivation in HIV-infected patients on suppressive antiretroviral therapy: a phase 1/2, single group, clinical trial. Lancet HIV (2014) 1:e13–21. doi: 10.1016/S2352- 51. Badia R, Pujantell M, Riveira-Muñoz E, Puig T, Torres-Torronteras J, Martı́ 3018(14)70014-1 R, et al. The G1/S specific cyclin D2 is a regulator of HIV-1 restriction in non- proliferating cells. PloS Pathog (2016) 12:e1005829. doi: 10.1371/ 36. Hantschel O. Unexpected off-targets and paradoxical pathway activation by journal.ppat.1005829 kinase inhibitors. ACS Chem Biol (2015) 10:234–45. doi: 10.1021/cb500886n 52. Pujantell M, Franco S, Galvá n-Femenı́a I, Badia R, Castellvı́ M, Garcia-Vidal 37. Vigó n L, Martı́nez-Romá n P, Rodrı́guez-Mora S, Torres M, Puertas MC, E, et al. ADAR1 affects HCV infection by modulating innate immune response. Mateos E, et al. Provirus reactivation is impaired in HIV-1 infected individuals on Antiviral Res (2018) 156:116–27. doi: 10.1016/j.antiviral.2018.05.012 treatment with dasatinib and antiretroviral therapy. Biochem Pharmacol (2021) 192:114666. doi: 10.1016/j.bcp.2021.114666 53. Li P, Kaiser P, Lampiris HW, Kim P, Yukl SA, Havlir DV, et al. Stimulating the RIG-I pathway to kill cells in the latent HIV reservoir following viral 38. Yukl S, Pillai S, Li P, Chang K, Pasutti W, Ahlgren C, et al. Latently-infected reactivation. Nat Med (2016) 22:807–11. doi: 10.1038/nm.4124 CD4+ T cells are enriched for HIV-1 tat variants with impaired transactivation activity. Virology (2009) 387:98–108. doi: 10.1016/j.virol.2009.01.013 54. Chen Y, Lun ATL, Smyth GK. From reads to genes to pathways: differential expression analysis of RNA-seq experiments using rsubread and the edgeR quasi- 39. Donahue DA, Kuhl BD, Sloan RD, Wainberg MA. The viral protein tat can likelihood pipeline. F1000Research (2016) 5:1438. doi: 10.12688/ inhibit the establishment of HIV-1 latency. J Virol (2012) 86:3253–63. doi: 10.1128/ f1000research.8987.2 JVI.06648-11 55. Love MI, Huber W, Anders S. Moderated estimation of fold change and 40. Donahue DA, Wainberg MA. Cellular and molecular mechanisms involved dispersion for RNA-seq data with DESeq2. Genome Biol (2014) 15:550. in the establishment of HIV-1 latency. Retrovirology (2013) 10:11. doi: 10.1186/ doi: 10.1186/s13059-014-0550-8 1742-4690-10-11 56. Lin GL, Wilson KM, Ceribelli M, Stanton BZ, Woo PJ, Kreimer S, et al. 41. Ranasinghe S, Cutler S, Davis I, Lu R, Soghoian DZ, Qi Y, et al. Association Therapeutic strategies for diffuse midline glioma from high-throughput of HLA-DRB1–restricted CD4+ T cell responses with HIV immune control. Nat combination drug screening. Sci Transl Med (2019) 11:eaaw0064. doi: 10.1126/ Med (2013) 19:930–3. doi: 10.1038/nm.3229 scitranslmed.aaw0064 42. Coiras M, Ló pez-Huertas MR, Pé rez-Olmeda M, Alcamı́ J. Understanding 57. Reimand J, Isserlin R, Voisin V, Kucera M, Tannus-Lopes C, Rostamianfar HIV-1 latency provides clues for the eradication of long-term reservoirs. Nat Rev A, et al. Pathway enrichment analysis and visualization of omics data using g: Microbiol (2009) 7:798–812. doi: 10.1038/nrmicro2223 Profiler, GSEA, cytoscape and EnrichmentMap. Nat Protoc (2019) 14:482–517. 43. Archin NM, Bateson R, Tripathy MK, Crooks AM, Yang K-H, Dahl NP, doi: 10.1038/s41596-018-0103-9 et al. HIV-1 expression within resting CD4+ T cells after multiple doses of 58. Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, et al. vorinostat. J Infect Dis (2014) 210:728–35. doi: 10.1093/infdis/jiu155 Cytoscape: A software environment for integrated models of biomolecular 44. Cary DC, Fujinaga K, Peterlin BM. Molecular mechanisms of HIV latency. J interaction networks. Genome Res (2003) 13:2498–504. doi: 10.1101/gr.1239303 Clin Invest (2016) 126:448–54. doi: 10.1172/JCI80565 59. Garcia-Vidal E, Castellvı́ M, Pujantell M, Badia R, Jou A, Gomez L, et al. 45. Ning S, Pagano JS, Barber GN. IRF7: activation, regulation, modification Evaluation of the innate immune modulator acitretin as a strategy to clear the HIV and function. Genes Immun (2011) 12:399–414. doi: 10.1038/gene.2011.21 reservoir. Antimicrob Agents Chemother (2017) 61:e01368–17. doi: 10.1128/ 46. Mogensen TH. IRF and STAT transcription factors - from basic biology to AAC.01368-17 roles in infection, protective immunity, and primary immunodeficiencies. Front 60. Ruiz A, Pauls E, Badia R, Riveira-Muñoz E, Clotet B, Ballana E, et al. Immunol (2019) 9. doi: 10.3389/fimmu.2018.03047 Characterization of the influence of mediator complex in HIV-1 transcription. J 47. Ning S, Huye LE, Pagano JS. Regulation of the transcriptional activity of the Biol Chem (2014) 289:27665–76. doi: 10.1074/jbc.M114.570341 IRF7 promoter by a pathway independent of interferon signalling. J Biol Chem 61. Garcia-Vidal E, Badia R, Pujantell M, Castellvı́ M, Felip E, Clotet B, et al. (2005) 280:12262–70. doi: 10.1074/jbc.M404260200 Dual effect of the broad spectrum kinase inhibitor midostaurin in acute and latent 48. Abner E, Stoszko M, Zeng L, Chen H-C, Izquierdo-Bouldstridge A, Konuma HIV-1 infection. Antiviral Res (2019) 168:18–27. doi: 10.1016/j.antiviral. T, et al. A new quinoline BRD4 inhibitor targets a distinct latent HIV-1 reservoir 2019.05.003 Frontiers in Immunology 18 frontiersin.org