bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1 Hydrogen sulfide blocks HIV rebound by maintaining mitochondrial 2 bioenergetics and redox homeostasis 3 4 Virender Kumar Pal1,2, Ragini Agrawal1,2, Srabanti Rakshit2, Pooja Shekar3, Diwakar 5 Tumkur Narasimha Murthy4, Annapurna Vyakarnam2, and Amit Singh*1,2. 6 7 1Department of Microbiology and Cell Biology, Indian Institute of Science, Bangalore,

8 India

9 2Centre for Infectious Disease Research (CIDR), Indian Institute of Science, Bangalore,

10 India

11 3Bangalore Medical College and Research Institute, Bangalore, India

12 4Department of Internal Medicine, Bangalore Medical College and Research Institute,

13 Bangalore, India

14

15 * Correspondence:

16 Amit Singh ([email protected])

17

18

19

20

21

22

23

24

25

26

27

28

1 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

29 Abstract

30 A fundamental challenge in HIV eradication is to understand how the virus establishes

31 latency, maintains stable cellular reservoirs, and promotes rebound upon interruption

32 of antiretroviral treatment (ART). Here, we discovered an unexpected role of the

33 ubiquitous gasotransmitter hydrogen sulfide (H2S) in HIV latency and reactivation. We

34 show that reactivation of HIV-1 is associated with down-regulation of the key H2S

35 producing enzyme cystathionine-g-lyase (CTH) and reduction in endogenous H2S.

36 Genetic silencing of CTH disrupts redox homeostasis, impairs mitochondrial function,

37 and remodels the transcriptome of latent cells to trigger HIV reactivation. Chemical

38 complementation of CTH activity using a slow-releasing H2S donor, GYY4137,

39 suppressed HIV reactivation and diminished virus replication. Mechanistically,

40 GYY4137 blocked HIV reactivation by inducing the Keap1-Nrf2 pathway, inhibiting

41 NF-kB, and recruiting the epigenetic silencer, YY1, to the HIV promoter. In latently

42 infected CD4+ T cells from ART-suppressed human subjects, GYY4137 in

43 combination with ART prevented viral rebound and improved mitochondrial

44 bioenergetics. Moreover, prolonged exposure to GYY4137 exhibited no adverse

45 influence on proviral content or CD4+ T cell subsets, indicating that diminished viral

46 rebound is due to a loss of transcription rather than a selective loss of infected cells. In

47 summary, this work provides mechanistic insight into H2S-mediated suppression of

48 viral rebound and suggests the inclusion of an H2S donor in the current ART regimen

49 to achieve a functional HIV-1 cure.

50

51

52

53

2 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

54 Introduction

55 Human Immunodeficiency Virus (HIV), the causative agent of Acquired Immuno-

56 Deficiency Syndrome (AIDS), is responsible for 0.6 million deaths and 1.7 million new

57 infections in 2019 (https://www.who.int/gho/hiv/epidemic_status/deaths_text/en/). Despite

58 advances in antiretroviral therapy (ART), the persistence of latent but replication-

59 competent HIV in cellular reservoirs is a major barrier to cure (1). Understanding host

60 factors and signaling pathways underlying HIV latency and rebound upon cessation of

61 ART is of the highest importance in the search for an HIV cure (2). Host-generated

62 gaseous signaling molecules such as nitric oxide (NO), carbon monoxide (CO), and

63 hydrogen sulfide (H2S) modulate antimicrobial activity, inflammatory response, and

64 metabolism of immune cells to influence bacterial and viral infections (3-8).

65 Surprisingly, while circumstantial evidence links NO and CO with HIV-1 infection (9-

66 11), the role of H2S remained completely unexplored.

67 H2S is primarily synthesized in mammals, including humans, via the reverse

68 transsulfuration pathway using methionine and cysteine metabolizing enzymes

69 cystathionine-gamma-lyase (CSE/CTH), cystathionine-beta-synthase (CBS), and

70 cysteine-aminotransferase (CAT)-3-mercaptopyruvate sulfurtransferase (MPST),

71 respectively (4). Because H2S is lipophilic, it rapidly permeates through biological

72 membranes without membrane channels (lipid bilayer permeability, PM ³ 0.5 ± 0.4

– 2– – 73 cm/s), dissociates into HS and S in aqueous solution, and maintains an HS : H2S ratio

74 of 3:1 at physiological pH (4). Interestingly, expression and enzymatic activity of CTH

75 was diminished in human CD8+ T cells and hepatic tissues derived from AIDS patients,

76 respectively (12, 13). Additionally, the levels of H2S precursors L-cysteine and

77 methionine were severely limited in AIDS patients (14). Also, HIV infection reduces

78 the major cellular antioxidant glutathione (GSH) (15), whose production is dependent

3 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

79 on L-cysteine and H2S (16, 17). Consequently, supplementation with N-acetyl cysteine

80 (NAC), which is an exogenous source of L-cysteine, has been proposed for AIDS

81 treatment (18). Since functional CBS and CTH enzymes are known to promote T cell

82 activation and proliferation (19), endogenous levels of H2S can potentiate T cell-based

83 immunity against HIV. Therefore, we think that the endogenous levels and biochemical

84 activity of H2S, which critically depends upon the cysteine and transsulfuration

85 pathway (20), are likely to be important determinants of HIV infection.

86 Calibrated production of H2S improves the functioning of various immune-cells

87 by protecting from deleterious oxidants, stimulating mitochondrial oxidative

88 phosphorylation (OXPHOS), and counteracting systemic inflammation (4, 21, 22). In

89 contrast, supraphysiological concentrations of H2S exert cytotoxicity by inhibiting

90 OXPHOS, promoting inflammation, and inducing redox imbalance (4, 21, 22). These

91 H2S-specific physiological changes are also crucial for HIV infection. First, chronic

92 inflammation contributes to HIV-1 pathogenesis by affecting innate and adaptive

93 immune responses (23), impairing recovery of CD4+ T cells post-ART (24), increasing

94 the risk of comorbidities (25, 26), and worsening organ function (27). Second, HIV-1

95 preferentially infects CD4+ T cells with high OXPHOS (28), relies on glycolysis for

96 continuous replication (29), and reactivates in response to altered mitochondrial

97 bioenergetics (30). Third, a higher capacity to resist oxidative stress promotes virus

98 latency, whereas increased oxidative stress levels induce reactivation (31-33). Fourth,

99 H2S directly (via S-linked sulfhydration) or indirectly modulates the activity of

100 transcription factors NF-kB and AP-1 (34, 35) that are well known to trigger HIV-1

101 reactivation from latency (36-38). Lastly, the thioredoxin/thioredoxin reductase

102 (Trx/TrxR) system crucial for maintaining HIV-1 reservoirs (39) has been shown to

103 regulate H2S- mediated S-persulfidation (40). Altogether, many physiological

4 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

104 processes vital for HIV pathogenesis overlap with the underlying mechanism of H2S

105 mediated-signaling.

106 Based on the role of H2S in regulating redox balance, mitochondrial

107 bioenergetics, and inflammation, we hypothesize that intracellular levels of H2S

108 modulate the HIV-1 latency and reactivation program. To test this hypothesis, we

109 utilized cell line based models of HIV-1 latency and CD4+ T cells of HIV-infected

110 patients on suppressive ART, which harbors the latent but replication-competent virus.

111 Biochemical and genetic approaches were exploited to investigate the link between H2S

112 and HIV-1 latency. Finally, we used NanoString gene expression technology, oxidant-

113 specific fluorescent dyes, and real-time extracellular flux analysis to examine the role

114 of H2S in mediating HIV-1 persistence by regulating gene-expression, redox signaling,

115 and mitochondrial bioenergetics.

116

117 Results

118 Diminished biogenesis of endogenous H2S during HIV-1 reactivation

119 To investigate the link between HIV-1 latency and H2S, we measured changes

120 in the expression of genes encoding H2S-generating enzymes CBS, CTH, and MPST in

121 monocytic (U1) and lymphocytic (J1.1) models of HIV-1 latency (Figure 1A). The U1

122 and J1.1 cell lines are well-studied models of post-integration latency and were derived

123 from chronically infected clones of a promonocytic (U937) and a T-lymphoid (Jurkat)

124 cell line, respectively (41, 42). Both U1 and J1.1 show very low basal expression of the

125 HIV-1 genome, which can be induced by several latency reversal agents (LRAs) such

126 as PMA, TNF-a, and prostratin (Figure 1B) (41, 42). First, we confirmed virus

127 reactivation by measuring HIV-1 gag transcript in U1 with a low concentration of PMA

128 (5 ng/ml). Treatment of PMA induces detectable expression of gag at 12 h, which

5 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

129 continued to increase for the entire 36 h duration of the experiment (Figure 1C). Next,

130 we assessed the mRNA and protein levels of CBS, CTH, and MPST in U1 during virus

131 latency (untreated) and reactivation (PMA-treated). The mRNA and protein expression

132 levels of CTH showed a significant reduction at 24 and 36 h post-PMA treatment as

133 compared to the untreated control, whereas the expression of CBS was not detected and

134 MPST remained unaffected (Figure 1D-E). As an uninfected control, we measured the

135 expression of CBS, CTH, and MPST in U937 cells. In contrast to U1, PMA treatment

136 stimulated the mRNA and protein levels of CTH in U937 (Figure 1F-G), while CBS

137 was barely detectable. Similar to U1 monocytic cells, PMA treatment reactivated HIV-

138 1 and reduced the expression of CTH in J1.1 lymphocytic cells but not in the uninfected

139 Jurkat cells in a time-dependent manner (Figure 1–figure supplement 1A-B). The

140 expression of CBS was reduced in both J1.1 and Jurkat upon PMA treatment, indicating

141 that only CTH specifically down-regulates in response to HIV-1 reactivation (Figure

142 1–figure supplement 1A-B). Finally, we directly measured endogenous H2S levels using

143 a fluorescent probe 7-azido-4-methylcoumarin (AzMc) that quantitatively detects H2S

144 in living cells (43). Consistent with the expression data, reactivation of HIV-1 by PMA

145 reduced H2S generation in U1, whereas H2S levels were significantly increased by

146 PMA in the uninfected U937 (Figure 1H-I). Taken together, these data indicate that

147 HIV-1 reactivation is associated with diminished biogenesis of endogenous H2S.

148

149 CTH-mediated reactivation of HIV-1 from latency

150 Our results suggest that H2S biogenesis via CTH is associated with HIV-1

151 latency. Therefore, we next asked whether CTH-derived H2S regulates reactivation of

152 HIV-1 from latency. To test this idea, we depleted endogenous CTH levels in U1 using

153 RNA interference (RNAi). The short hairpin RNA specific for CTH (U1-shCTH)

6 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

154 silenced the expression of CTH mRNA and protein by 90 % as compared to non-

155 targeting shRNA (U1-shNT) (Figure 2A-B). Moreover, using AzMC probe, we

156 confirmed reduction in endogenous H2S levels in U1-shCTH as compared to U1-shNT

157 (Figure 2–figure supplement 1A-B). Next, we investigated the effect of H2S depletion

158 via CTH suppression on HIV-1 latency by measuring viral transcription (gag

159 transcript), translation (HIV-1 p24 capsid protein), and release (HIV-1 p24 abundance

160 in the cell supernatant). We found that the low basal expression of HIV-1 gag in U1-

161 shCTH was stimulated by 16-fold upon depletion of CTH (p = 0.0018) (Figure 2C).

162 Furthermore, while PMA induced expression of gag transcript by 73-fold in U1-shNT,

163 a further enhancement to 200-fold was observed in U1-shCTH (Figure 2C). Consistent

164 with this, levels of p24 capsid protein inside cells or released in the supernatant were

165 significantly elevated in U1-shCTH as compared to U1-shNT with (p = 0.028) or

166 without PMA-treatment (p = 0.015) (Figure 2D-E). We also depleted CTH levels in

167 J1.1 cells using RNAi and monitored HIV-1 reactivation by assessing intracellular p24

168 levels under basal conditions. The depletion of CTH triggered HIV-1 reactivation from

169 latency as evident from a 7-fold increase in p24 levels as compared to shNT (Figure 2–

170 figure supplement 1C). A minor increase in p24 levels (3.6-fold) was also apparent

171 upon the depletion of CBS in J1.1 (Figure 2–figure supplement 1C). These data indicate

172 that CTH and likely CTH-derived H2S supports latency and impedes the reactivation

173 of HIV-1.

174

175 Altered expression of genes involved in HIV-1 reactivation upon CTH depletion

176 The above results indicate that CTH is a target controlling HIV-1 reactivation

177 from latency, prompting us to investigate the mechanism. Several pathways that induce

178 HIV-1 reactivation from latency are also influenced by H2S. These include redox

7 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

179 signaling (44), NF-kB pathway (45), inflammatory response (46), and mitochondrial

180 bioenergetics (47). Hence, we conducted a focused expression profiling of 185 human

181 genes intrinsically linked to HIV-1 reactivation using the NanoString nCounter

182 technology (see supplementary Table S2 for the gene list) to measure absolute amounts

183 of multiple transcripts without reverse transcription. Because depletion of CTH

184 promoted reactivation of the HIV-1 in U1, we compared the expression profile of U1-

185 shCTH with U1-shNT to further understand the link between H2S biogenesis and HIV-

186 1 latency. Examination of 84 genes related to oxidative stress response revealed

187 differential regulation of 22 genes in U1-shCTH compared with U1-shNT (Figure 2F

188 and supplementary Table S3). Interestingly, more than 90 % of genes showing altered

189 expression were down-regulated in U1-shCTH. Of these, genes encoding key cellular

190 antioxidant enzymes and buffers such as catalase (CAT), superoxide dismutase

191 (SOD1), peroxiredoxin family, thioredoxin family (TXN, TXNRD1 and TXNRD2),

192 sulfiredoxin (SRXN1), glutathione metabolism (GCLM, GSS, and GSR), and sulfur

193 metabolism (MPST, AHCY, and MAT2A) were significantly less expressed in U1-

194 shCTH as compared to U1-shNT (Figure 2F). Since oxidative stress elicits HIV-1

195 reactivation (31), these findings indicate that HIV-1 reactivation through CTH

196 depletion could be a consequence of an altered redox balance. Further, pathways

197 involved in promoting HIV-1 reactivation such as NF-kB signaling (36) and apoptosis

198 (48) were also induced in U1-shCTH as compared to U1-shNT (Figure 2F). Notably,

199 multiple HIV-1 restriction factors important for viral latency including type I interferon

200 signaling (IRF2, STAT-1) (49, 50), APOBEC3G (51), CDK9-CCNT1 (52, 53), SLP-1

201 (54), and chromatin remodelers (SMARCB1, BANF1, YY1) (55, 56) were down-

202 regulated in U1-shCTH. HIV-1 proteins are known to target mitochondria to induce

203 mitochondrial depolarization, elevate mitochondrial ROS, and apoptosis for replication

8 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

204 (57). Several genes involved in sustaining mitochondrial function and membrane

205 potential (e.g., BAX, UCP2, LHPP, MPV17) were repressed in U1-shCTH (Figure 2F),

206 highlighting a potential association between unrestricted virus replication,

207 mitochondrial dysfunction, and CTH depletion.

208 Since PMA is known to reactivate HIV-1 (58), we examined the gene

209 expression in U1-shNT upon reactivation of HIV-1 by PMA. We found that 80 % of

210 genes affected by the depletion of CTH were similarly perturbed in response to PMA.

211 Signature of transcripts associated with oxidative stress response, sulfur metabolism,

212 anti-viral factors, and mitochondrial function was comparable in U1-shCTH and PMA-

213 treated U1-shNT (Figure 2F). Based on these similarities, we hypothesized that

214 combining PMA with CTH depletion would have an additive effect on the expression

215 of genes linked to HIV-1 reactivation. Indeed, exposure of U1-shCTH to PMA induced

216 gene expression changes which surpassed those produced by PMA treatment or CTH-

217 depletion alone (Figure 2F). Notably, significant expression changes in U1-shCTH

218 upon PMA treatment are consistent with our data showing maximal HIV-1 activation

219 under these conditions (see Figure 2C). Overall, these results indicate that CTH

220 contributes to HIV-1 latency by modulating multiple pathways coordinating cellular

221 homeostasis (e.g., redox balance, mitochondrial function) and the anti-viral response.

222

223 CTH is required to maintain redox homeostasis and mitochondrial function

224 Physiological levels of H2S support redox balance and mitochondrial function

225 by maintaining GSH balance (16), protecting against ROS (16), and acting as a

226 substrate for the electron transport chain (21, 22). On this basis, we reasoned that the

227 depletion of CTH could contribute to redox imbalance and mitochondrial dysfunction

228 to promote HIV-1 reactivation. We first measured total glutathione content

9 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

229 (GSH+GSSG) and GSSG concentration in U1-shNT and U1-shCTH using chemical-

230 enzymatic analysis of whole-cell extract (31). Whole-cell glutathione content was not

231 significantly different in U1-shNT and U1-shCTH (p = 0.2801) (Figure 3A). However,

232 the GSSG concentration was elevated in U1-shCTH resulting in a concomitant decrease

233 in the GSH/GSSG ratio of U1-shCTH as compared to U1-shNT (p = 0.0032) (Figure

234 3A). The increased GSSG pool and reduced GSH/GSSG poise confirm that cells are

235 experiencing oxidative stress upon CTH depletion. We measured total ROS using a

236 fluorescent probe, 5,6-chloromethyl-2¢,7¢-dicholrodihydrofluorescein diacetate (CM-

237 H2DCFDA), which non-specifically responds to any type of ROS within the cells. Both

238 U1-shNT and U1-shCTH showed comparable levels of cytoplasmic ROS (Figure 3–

239 figure supplement 1A), which remained unchanged after stimulation with PMA. In

240 addition to cytoplasmic ROS, we also measured mitochondrial ROS (mitoROS) using

241 the red fluorescent dye MitoSOX, which selectively stains mitoROS. Lowering the

242 levels of CTH severely increased mitoROS in U1, which was further accentuated after

243 PMA stimulation (Figure 3B). We then studied the effect of CTH depletion on

244 mitochondrial functions using a Seahorse XF Extracellular Flux Analyzer (Agilent) as

245 described previously by us (Figure 3C) (30, 47). Both basal and ATP-coupled

246 respiration was significantly decreased in U1-shCTH as compared to U1-shNT (Figure

247 3D-E), consistent with the role of endogenous H2S in reducing cytochrome C oxidase

248 for respiration (47). The maximal respiratory capacity, attained by the dissipation of the

249 mitochondrial proton gradient with the uncoupler FCCP, was markedly diminished in

250 U1-shCTH (Figure 3E). The maximal respiration also facilitated the estimation of the

251 spare respiratory capacity, which was nearly exhausted in U1-shCTH. Additional

252 hallmarks of HIV-1 reactivation and dysfunctional mitochondria such as coupling

253 efficiency and non-mitochondrial oxygen consumption rate (nmOCR) (30) were also

10 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

254 adversely affected upon depletion of CTH (Figure 3E). These results indicate that CTH

255 depletion decelerates respiration, diminishes the capacity of macrophages to maximally

256 respire, and promotes nmOCR. All of these parameters are important features of

257 mitochondrial health and are likely to be crucial for maintaining HIV-1 latency.

258

259 A small-molecule H2S donor diminished HIV-1 reactivation and viral replication

260 One of the aims of the HIV-1 functional cure is to identify small molecules that

261 can be combined with ART to delay or halt virus replication, reactivation, and

262 replenishment of the latent reservoir (59, 60). Having shown that diminished levels of

263 endogenous H2S is associated with HIV-1 reactivation, we next examined if elevating

264 H2S levels using a small-molecule H2S donor- GYY4137 sustains HIV-1 latency. The

265 GYY4137 is a widely used H2S donor that releases a low amount of H2S over a

266 prolonged period to mimic physiological production (61). Because H2S discharge is

267 unusually sluggish by GYY4137, the final concentration of H2S released is likely to be

268 significantly lower than the initial concentration of GYY4137. We confirmed this by

269 measuring H2S release in U1 cells treated with low (0.3 mM) or high (5 mM)

270 concentrations of GYY4137 using methylene blue colorimetric assay. As a control, we

271 treated U1 cells with 0.3 mM NaHS, which rapidly releases a high amount of H2S. As

272 expected, NaHS treatment rapidly released 90 % of H2S within 5 min of treatment

273 (Figure 4A). In contrast, GYY4137 uniformly released a low amount of H2S for the

274 entire 120 h duration of the experiment (Figure 4A).

275 We systematically tested the effect of GYY4137 on HIV-1 reactivation using

276 multiple models of HIV-1 latency and replication. As a control, we used N-acetyl

277 cysteine (NAC) that is known to block HIV-1 reactivation (62). Pretreatment of U1

278 with a non-toxic dose of GYY4137 (5 mM) (Figure 4–figure supplement 1A)

11 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

279 diminished the expression of gag transcript by 2-fold at 24 h and 10-fold at 48 h post-

280 PMA treatment (Figure 4B). The effect of GYY4137 on p24 levels in the supernatant

281 was even more striking as it completely abolished the time-dependent increase in p24

282 concentration post-PMA treatment (Figure 4C). As expected, pretreatment with NAC

283 similarly prevented PMA-triggered reactivation of HIV-1 in U1 (Figure 4C).

284 Pretreatment of U1 with spent GYY4137, which comprises the decomposed backbone,

285 showed no effect on PMA induced gag transcript (Figure 4–figure supplement 1B).

286 Because TNF-a is a physiologically relevant cytokine that reactivates HIV-1 from

287 latency (41), we tested the effect of GYY4137 on TNF-a-mediated virus reactivation.

288 Treatment of U1 with TNF-a stimulated the expression of gag transcript by 42- and

289 169-fold at 24 and 48 h, post-treatment, respectively (Figure 4D). The addition of

290 GYY4137 or NAC nearly abolished the reactivation of HIV-1 in response to TNF-a

291 treatment (Figure 4D). Earlier, we showed that depletion of CTH stimulated HIV-1

292 reactivation in U1 (see Figure 2C). Therefore, we tested if GYY4137 could

293 complement this genetic deficiency and subvert HIV-1 reactivation. The U1-shCTH

294 cells were pre-treated with 5 mM GYY4137 and p24 levels in the supernatant were

295 measured. The elevated levels of p24 in U1-shCTH were reduced by 2-fold under basal

296 conditions and 3.2-fold upon PMA stimulation in response to GYY4137 (Figure 4E).

297 Both RNAi and chemical complementation data provide evidence that H2S is one of

298 the factors regulating HIV-1 latency and reactivation in U1.

299 Similar to U1, we next examined whether GYY4137 subverts HIV-1

300 reactivation in the J1.1 T-cell line model. Treatment of J1.1 with PMA for 12 h resulted

301 in a 38-fold increase in gag transcript as compared to untreated J1.1 (Figure 4F),

302 indicating efficient reactivation of HIV-1. Importantly, pretreatment with various non-

303 toxic concentrations of GYY4137 (Figure 4–figure supplement 1C) reduced the

12 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

304 stimulation of gag transcription by 2-fold upon subsequent exposure to PMA (Figure

305 4F). The inhibitory effect of GYY4137 on HIV-1 reactivation was relatively greater

306 compared to NAC in J1.1 (Figure 4F). We used another well-established T cell-based

307 model of HIV-1 latency (J-Lat) to examine the influence of H2S. In J-Lat, the integrated

308 HIV-1 genome encodes green-fluorescent protein (GFP), which allows precise

309 quantification of HIV-1 reactivation from latency in response to LRAs such as PMA,

310 TNF-a, and prostratin (63). Consistent with this, treatment of J-Lat with 5 µM

311 prostratin for 24 h induced significant HIV-1 reactivation, which was translated as 100

312 % increase in GFP+ cells (Figure 4–figure supplement 1D). Pretreatment with

313 GYY4137 significantly reduced HIV-1 reactivation in a dose-dependent manner

314 (Figure 4–figure supplement 1D).

315 We also examined if an endogenous increase in H2S by GYY4137 halts the

316 replication of HIV-1. To this end, we infected CD4+ T cell (Jurkat) with CXCR4-using

317 HIV-1 (pNL4.3) and monitored HIV-1 replication by measuring the expression of gag

318 transcript and the levels of p24 in the supernatant. A progressive increase in gag

319 transcription and p24 release was detected for the entire 96 h duration of the

320 experiment, indicating efficient viral replication (Figure 4G-H). Pretreatment with a

321 non-toxic concentration of GYY4137 (0.3 mM) (Figure 4–figure supplement 1E)

322 resulted in a 3 to 3.4-fold inhibition of HIV-1 replication at 72 and 96 h post-infection

323 (Figure 4G-H). Overall, these data establish that elevated levels of endogenous H2S

324 efficiently suppress HIV-1 reactivation and replication.

325

326 GYY4137 reduced the expression of host genes involved in HIV-1 reactivation

327 To dissect the mechanism of GYY4137-mediated inhibition on HIV-1

328 reactivation, we examined the expression of 185 genes associated with HIV-1

13 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

329 reactivation using the NanoString nCounter technology as described above (Figure 5A

330 and supplementary Table S4). Expression was analyzed for viral latency (unstimulated

331 U1), reactivation (PMA-stimulated treated U1; PMA), and H2S-mediated suppression

332 (U1-GYY+PMA). Consistent with the role of PMA-mediated oxidative stress in

333 preceding HIV-1 reactivation (31), expression of genes encoding ROS, and RNS

334 generating enzymes (e.g., NADPH oxidase [NCF1, CYBB] and Nitric oxide synthase

335 [NOS2]) were up-regulated (Figure 5A). Also, the expression of major antioxidant

336 enzymes (e.g., GPXs, PRDXs, and CAT) and redox buffers (GSH and TRX pathways)

337 remained repressed in U1-PMA, indicative of elevated oxidative stress. Additionally,

338 pro-inflammatory signatures (e.g., TGFB1 and SERPINA1) and trans-activators (e.g.,

339 FOS) were induced, whereas host factors involved in HIV-1 restriction (e.g., IRF1 and

340 YY1) were repressed in U1-PMA as compared to U1. In agreement with the reduction

341 in endogenous H2S levels during HIV-1 reactivation, expression of CTH involved in

342 H2S anabolism was down-regulated and H2S catabolism (SQRDL) was up-regulated by

343 PMA.

344 We noticed that the treatment with GYY4137 reversed the effect of PMA on

345 the expression of genes associated with oxidative stress, inflammation, anti-viral

346 response, apoptosis, and trans-activators (Figure 5A). For example, GYY4137 elicited

347 a robust induction of genes regulated by nuclear factor erythroid 2-related factor 2

348 (Nrf2) in U1-GYY+PMA compared to U1 or U1-PMA (Figure 5A). Nrf2 acts as a

349 master regulator of redox metabolism (64) by binding to the antioxidant response

350 element (ARE) and initiating transcription of major antioxidant genes (e.g., GSH

351 pathway, TXNRD1, HMOX1, CTH, GPX4, and SRXN1). The Nrf2 activity has been

352 shown to pause HIV-1 infection by inhibiting the insertion of reverse-transcribed viral

353 cDNA into the host chromosome (65). Furthermore, sustained activation of Nrf2-

14 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

354 dependent antioxidant response is essential for the establishment of viral latency (33).

355 A few genes encoding H2O2 detoxifying enzymes (e.g., CAT, GPX1, and PRDXs) were

356 repressed in U1-GYY+PMA, indicating that the expression of these enzymes was likely

357 counterbalanced by the elevated expression of other antioxidant systems by H2S. In line

358 with the antagonistic effect of GYY4137 on HIV-1 transcription, the expression of

359 HIV-1 trans-activator (FOS) was down-regulated, and an inhibitor of NF-kB signaling

360 (NFKBIA) was induced in U1-GYY+PMA as compared to U1 or U1-PMA. Lastly,

361 GYY4137 stimulated the expression of several anti-HIV (YY1, STAT1, STAT3, and

362 IRF1) and pro-survival factors (CDKN1A and LTBR) that were repressed by PMA

363 (Figure 5A). Altogether, H2S supplementation induces a major realignment of redox

364 metabolism and immune pathways associated with HIV-1 reactivation.

365

366 GYY4137-mediated modulation of the Keap1-Nrf2 axis, NF-kB signaling, and

367 activity of epigenetic factor YY1

368 Our expression data indicate activation of the Nrf2 pathway and modulation of

369 transcription factors such as NF-kB and YY1 upon treatment with GYY4137. We

370 tested if the mechanism of H2S-mediated subversion of HIV-1 reactivation involves

371 these pathways. H2S has recently been shown to prevent cellular senescence by

372 activation of Nrf2 via S-persulfidation of its negative regulator Keap-1 (66). Under

373 unstimulated conditions, Nrf2 binds to Keap1, and the latter promotes Nrf2 degradation

374 via the proteasomal machinery (66). Nrf2 disassociates from the S-persulfidated form

375 of Keap1, accumulates in the cytoplasm, and translocates to nuclei where it induces

376 transcription of antioxidant genes upon oxidative stress (66). We first examined if

377 GYY4137 treatment accumulates Nrf2 in the cytoplasm. As expected, Nrf2 was not

378 detected in U1 owing to its association with Keap1 under unstimulated conditions.

15 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

379 However, a noticeable accumulation of Nrf2 was observed in U1-GYY+PMA

380 compared to U1 or U1-PMA (Figure 5B). As an additional verification, we quantified

381 the levels of an Nrf2-dependent protein HMOX-1. Similar to Nrf2, the levels of

382 HMOX-1 were also induced in U1-GYY+PMA (Figure 5B). These findings are

383 consistent with our NanoString data showing activation of Nrf2-specific oxidative

384 stress responsive genes in GYY4137 treated U1.

385 Next, we determined if GYY4137 treatment targets NF-kB, which is a major

386 regulator of HIV-1 reactivation (36). The cellular level of phosphorylated serine-536 in

387 p65, a major subunit of NF-kB, is commonly measured to assess NF-kB activation (67).

388 Since PMA reactivates HIV-1, we observed an increase in p65 ser-536 phosphorylation

389 in U1-PMA (Figure 5C). In contrast, pretreatment with GYY4137 significantly

390 decreased PMA-induced p65 ser-536 phosphorylation (Figure 5C). These findings

391 suggest that GYY4137 is likely to affect the DNA binding and transcriptional activity

392 of NF-kB. Consistent with this idea, a two-step chromatin immunoprecipitation and

393 genomic qPCR (ChIP-qPCR) assay showed that GYY4137-treatment significantly

394 reduced the occupancy of p65 at its binding site (ER, enhancer region) on the HIV-1

395 LTR (Figure 5D-E).

396 GYY4137 induces the expression of another transcription factor YY1, which

397 binds to HIV-1 LTR at RBEI and RBEIII and recruits histone deacetylase (HDACs) to

398 facilitate repressive chromatin modifications (56). Overexpression of YY1 is known to

399 promote HIV-1 latency (68). We tested if GYY4137 promotes the binding of YY1 at

400 RBEI and RBEIII sites on HIV-1 LTR by ChIP-qPCR. The occupancy of YY1 at RBEI

401 and RBEIII was significantly enriched in the case of U1-GYY+PMA as compared to

402 U1-PMA or U1 (Figure 5F). Taken together, these results indicate that increasing

403 endogenous H2S levels by using a slow-releasing donor effectively modulate the

16 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

404 Keap1-Nrf2 axis and activity of transcription factors to maintain redox balance and

405 control HIV-1 reactivation.

406

407 GYY4137 blocks HIV-1 rebound from latent CD4+ T cells isolated from infected

408 individuals on suppressive ART

409 Having shown that H2S suppresses HIV-1 reactivation in multiple cell line

410 models of latency, we next studied the ability of GYY4137 to limit virus transcription

411 in primary CD4+ T cells derived from virally suppressed patients. We used a previously

412 established methodology of maintaining CD4+ T cells from infected individuals on

413 ART for a few weeks without any loss of phenotypic characteristics associated with

414 HIV reservoirs (60). We isolated CD4+ T cells from peripheral blood mononuclear cells

415 (PBMCs) of five HIV-infected subjects on suppressive ART. The CD4+ T cells were

416 expanded in the presence of interleukin-2 (IL-2), phytohemagglutinin (PHA), and

417 feeder cells (Figure 6A). The expanded cells were cultured for 4 weeks in a medium

418 containing IL-2 with ART (100 nM efavirenz, 180 nM zidovudine, and 200 nM

419 raltegravir) either in the presence or absence of GYY4137 (100 µM) (Figure 6A). In

420 this model, the virus reactivates by day 14 followed by progressive suppression at day

421 21 and latency establishment by day 35. The virus can be reactivated from this latent

422 phase using well-known LRAs (60). We quantified the temporal expression of viral

423 RNA using RT-qPCR periodically at 7 days interval. The levels of viral RNA increased

424 initially (day 7 to 14), followed by low to undetectable levels by day 28 (Figure 6B).

425 By day 28, virus RNA was uniformly untraceable in cells derived from GYY4137

426 treated and untreated groups, indicating the establishment of latency (Figure 6B).

427 We next assessed the ability of GYY4137 to efficiently block viral reactivation.

428 On day 28, we stimulated the CD4+ T cells with the protein kinase C (PKC) activator,

17 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

429 prostratin, in the absence of any treatment. The activation of viral transcription was

430 measured 24 h later by RT-qPCR. The removal of ART uniformly resulted in a viral

431 reactivation by prostratin in CD4+ T cells of HIV-patients (Figure 6C). In contrast,

432 upon ART+GYY4137 removal followed by prostratin stimulation, viral reactivation

433 was attenuated by 90 % (N = 5, mean) for all 5 patient samples, and individual

434 inhibition ranged from 78.9 % to 100 % (Figure 6C). Overall, pretreatment with

435 ART+GYY4137 significantly reduced prostratin-mediated HIV-reactivation when

436 compared to ART alone (p < 0.01) (Figure 6D).

437 We also examined the immune-phenotype of CD4+ T cells treated with ART or

438 ART+GYY4137 by monitoring the expression of activation (CD38) and quiescence

439 (CD127) marker. As expected, CD38 expression increased, and CD127 expression

440 decreased at day 7 and 14 during the activation phase, followed by a gradual reversal

441 of the pattern during the resting phase at day 21 and 28 (Figure 6E-F and Figure 6–

442 figure supplement 1A). Interestingly, GYY4137 treatment did not alter the temporal

443 changes in the expression of activation and quiescence markers on CD4+ T cells (Figure

444 6E-F). Since memory CD4+ T cells are preferentially targeted by HIV (69, 70), we

445 further analyzed if the status of naive (TN), central memory (TCM), transitional memory

446 (TTM), and effector memory (TEM) is affected by the GYY4137 in our ex vivo expansion

447 model. We observed that the CD4+ T cells that responded to ex vivo activation were

448 mainly composed of TTM and TEM in our patient samples (Figure 6–figure supplement

449 1B-E). This is consistent with the study reporting the presence of translation-competent

450 genomes mainly in the TTM and TEM of ART-suppressed individuals (71). Interestingly,

451 the fraction of TTM shows a progressive decline, whereas TEM displays gradual increase

452 during transition from activation (7-14 days) to quiescence phase (21-28 days) (Figure

453 6–figure supplement 1D-E). The addition of GYY4137 did not affect dynamic changes

18 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

454 in the frequency of TTM and TEM subpopulations (Figure 6–figure supplement 1D-E).

455 Finally, we tested if the reduction in viral RNA upon GYY4137 treatment is

456 due to the loss of cells with ability to reactivate virus or selection of cells subset that is

457 non-responsive to reactivating stimuli. We did not find significant differences in total

458 HIV DNA content between the CD4+ T cells immediately isolated from the patient’s

459 PBMCs (ex vivo) and the expanded cells treated with ART+GYY4137 or ART alone

460 for the entire duration of the experiment (Figure 6G). The viability of cells remained

461 comparable between ART+GYY4137 and ART treatment groups overtime (Figure 6–

462 figure supplement 1F). Altogether, using a range of cellular and immunological assays,

463 we confirmed that the characteristics of an individual’s viral reservoir remain

464 preserved, and the suppression of viral RNA upon GYY4137 treatment is the result of

465 H2S-mediated inhibition of HIV-1 transcription rather than a reduction in proviral

466 content or an altered CD4+ T cell subsets. In sum, prolonged exposure to GYY4137

467 results in potent inhibition of viral reactivation, suggesting a new H2S-based

468 mechanism to neutralize bursts of virus reactivation under suppressive ART in vivo.

469

470 GYY4137 prevents mitochondrial dysfunction in CD4+ T cells of HIV-patients

471 during viral rebound

472 Because virus reactivation upon depletion of endogenous H2S resulted in

473 mitochondrial dysfunction and redox imbalance in U1, we tested if the elevation of H2S

474 levels by GYY4137 improves mitochondrial health of primary CD4+ T cells during

475 virus reactivation ex vivo. As described earlier, CD4+ T cells harboring latent virus upon

476 prolonged (28 days) treatment of ART and ART+GYY4137 were stimulated by

477 prostratin or left unstimulated and subjected to mitochondrial flux analysis. The

478 unstimulated cells from both ART and ART+GYY4147 cultures did not show any

19 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

479 difference in OCR (Figure 7A). In contrast, several features reflecting efficient

480 mitochondrial activity such as basal respiration and ATP-coupled respiration were

481 significantly higher in prostratin stimulated CD4+ T cells in case of ART+GYY4137

482 treatment than ART alone (Figure 7B-C). Consistent with this, mitoROS generation

483 upon stimulation with prostratin or other LRAs such as PMA/ionomycin was reduced

484 in ART+GYY4137 treated CD4+ T cells than ART alone (p = 0.005 and p = 0.014),

485 and was nearly comparable to unstimulated cells (Figure 7D-E). The reduction in

486 mitoROS could be a consequence of GYY4137-mediated increase in the expression of

487 Nrf2-dependent antioxidant systems. We directly tested this by RT-qPCR analysis of a

488 selected set of Nrf2-dependent genes on CD4+ T cells treated with GYY4137 for 28

489 days. Consistent with our findings in U1, treatment with ART+GYY4137 uniformly

490 induced the expression of antioxidant genes in the latently infected CD4+ T cells

491 compared to cell treated with ART alone (Figure 7F).

492 Altogether, these data suggest that H2S not only prevents virus reactivation but

493 also improves mitochondrial bioenergetics and maintains redox homeostasis, which

494 could be important for in vivo suppression of viral rebound and replenishment of the

495 reservoir.

496

497 Discussion

498 The major conclusion of our study is that HIV-1 reactivation is coupled to

499 depletion of endogenous H2S, which is associated with dysfunctional mitochondrial

500 bioenergetics, in particular suppressed OXPHOS, GSH/GSSG imbalance, and elevated

501 mitoROS. Decreased H2S also impaired the expression of genes involved in

502 maintenance of redox balance, mitochondrial function, inflammation, and HIV-1

503 restriction, which correlates with reactivation of HIV-1. This conclusion is supported

20 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

504 in multiple cell line models of HIV-1 latency. Chemical complementation with

505 GYY4137 identified H2S as an effector molecule. Finally, we confirmed clinical

506 relevance of our finding in primary CD4+ T cells isolated from ART-suppressed

507 patients to recapitulate features of HIV-1 latency and reactivation (60). Overall, our

508 data show that while H2S deficiency reactivates HIV-1, the H2S donor GYY4137 can

509 potently inhibit residual levels of HIV-1 transcription during suppressive ART and

510 block virus reactivation upon stimulation. Hence, our findings highlight that H2S based

511 therapies (72) can be exploited to lock HIV in a state of persistent latency by impairing

512 the ability to reactivate.

513

514 How does H2S promote HIV-1 latency and suppress reactivation? Several

515 studies revealed that HIV-1 reactivation is associated with loss of mitochondrial

516 functions, oxidative stress, inflammation, and apoptosis (30, 31, 48, 57). Several of

517 these biological dysfunctions are corrected by H2S. For example, localized delivery of

518 H2S in physiological concentrations improves mitochondrial respiration, mitigates

519 oxidative stress, and exerts anti-inflammatory functions (5, 16, 22, 44-47, 72). Our

520 NanoString findings, XF flux analysis, and redox measurement supports a mechanism

521 whereby H2S sufficiency promotes sustenance of mitochondrial health and redox

522 homeostasis to control HIV-1 reactivation. Conversely, genetic depletion of

523 endogenous H2S promotes HIV-1 reactivation by decelerating OXPHOS, GSH/GSSH

524 imbalance, and elevated mitoROS. It is widely known that reduced OXPHOS leads to

525 accumulation of NADH, which traps flavin mononucleotide (FMN) in the reduced state

526 on respiratory complex I leading to mitoROS generation by incomplete reduction of O2

527 (73). Also, H2S-depletion increased nmOCR, which indicates elevated activities of

528 other ROS producing enzymes such as NADPH oxidase and

21 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

529 cyclooxygenases/lipoxygenase (74, 75). All of these metabolic changes are likely

530 contributors of overwhelming ROS and disruption of GSH/GSSG poise in U1-shCTH,

531 resulting in HIV-1 reactivation. In fact, consistent with our findings, oxidative shift in

532 GSH/GSSG poise, deregulated mitochondrial bioenergetics, and ROI have been

533 reported to promote HIV-1 reactivation (30, 31). The addition of GYY4137 stimulated

534 OXPHOS and blocked HIV reactivation, which is in line with the known ability of H2S

535 in sustaining respiration by acting as a substrate of cytochrome C oxidase (CcO) (47).

536 Also, H2S can prevent mitochondria-mediated apoptosis (76), which is crucial for HIV-

537 1 reactivation and infection (48). Finally, activation of Nrf2-specific antioxidant

538 pathways upon exposure to GYY4137 agrees with the potential of H2S in augmenting

539 the cellular antioxidative defense machinery (4, 16, 44). Importantly, cell lines (U1 and

540 J1.1) modeling HIV-1 latency and PBMCs of individuals that naturally suppress HIV-

541 1 reactivation (long-term non-progressors [LTNPs]) display robust capacities to resist

542 oxidative stress and apoptosis (31, 77).

543 Our transcription profiling showed a positive correlation between GYY4137-

544 induced H2S sufficiency and HIV-1 restriction factors (e.g., YY1, APOBEC3, IRF1,

545 Nrf2) and negative correlation with HIV-1 trans-activators (e.g., NFKB, FOS),

546 consistent with an interrelationship between cell metabolism and HIV latency (28, 78).

547 Interestingly, a sustained induction of Nrf2-driven cellular antioxidant response is

548 essential for the successful transition between productive and latent HIV-1 infection

549 (31, 33). Moreover, inhibition of Nrf2 promoted viral transcription and increased ROS

550 generation (33), whereas its activation reduced HIV-1 infection (65). On this basis, we

551 think that H2S-mediated inhibition of HIV is partly related to activation of Nrf2-

552 dependent antioxidant genes. In HIV infection, the suppression of viral reactivation and

553 pro-inflammatory mediator expression was paralleled by the inhibition of NF-kB.

22 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

554 Interestingly, H2S inhibited activation of NF-kB in several models of inflammatory

555 diseases including hemorrhagic shock, lung injury, and paramyxovirus infections (79,

556 80). Similarly to these findings, we found that exposure of U1 with GYY4137 inhibited

557 the NF-kB p65 subunit and reduced its binding to HIV-1 LTR. Currently, no

558 antiretroviral drugs inhibit basal transcription of provirus and blocks reactivation upon

559 therapy (ART) interruption. It has been suggested that epigenetic silencing will be an

560 important pre-requisite for persistent latency (81, 82). Our expression and ChIP-qPCR

561 data suggest that H2S could exert its influence on HIV replication machinery via

562 activating YY1- an epigenetic silencer of HIV-1 LTR (56, 68). Interestingly, another

563 gasotransmitter NO modulates YY1 function via S-nitrosylation of its cysteine residues

564 (83), indicating that YY1 is amenable to thiol-based-modifications and raises the

565 possibility of YY1 S-persulfidation by H2S as a mechanism to epigenetically

566 reconfigure HIV-1/promoter LTR.

567 The inefficiency of current ART in preventing virus rebound after therapy

568 cessation poses a major hurdle to HIV cure. Clinical studies revealed that reservoir sizes

569 have larger influence on viral rebound after ART cessation (84). For example, smaller

570 reservoir size maintained due to sustained immunological control in case of a

571 Mississippi baby (85), the VISCONTI cohort (86), and a French teenager (87) delayed

572 viral rebound after treatment interruption. Using a primary cell system that has been

573 successfully used to recapitulate latency and reactivation seen in patients (60), we

574 confirmed that pretreatment with GYY4137 attenuated viral rebound in primary human

575 CD4+ T cells upon stimulation with prostratin. Collectively, our results suggest that

576 H2S reduces HIV-1 transcriptional activity, promotes silencing of its promoter, and

577 reduces its potential for reactivation. The virus-suppressing effects of H2S were

578 observed alongside its beneficial consequences on cellular physiology (e.g.,

23 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

579 mitochondrial function and redox balance). Both mitochondrial dysfunction and

580 oxidative stress are major complications associated with long-term exposure to ART

581 and contributes to inflammation and organ damage (88-90). Future studies will

582 investigate if the prolonged treatment with H2S donors results in a better outcome due

583 to recruitment of specific repressor and/or epigenetic silencer of HIV-1 LTR without

584 detrimental consequences on cellular physiology. It’s tempting to speculate that over

585 time (H2S in combination with ART) can push transcriptional repression beyond a

586 certain threshold where it will be impossible to reactivate HIV, thereby blocking and

587 locking virus in a state of persistent latency. Our results provide a proof-of-concept of

588 a gasotransmitter H2S that can be explored for a functional cure of HIV. Blocking HIV

589 rebound by inhibitors of viral factors (e.g., Tat) invariably results in the emergence of

590 resistant mutants (91). In this context, blocking HIV-1 rebound via H2S-directed

591 modulation of host pathways could help in overcoming the problem of evolution of

592 escape variants. Finally, therapy non-compliance and frequent blips contributes to

593 continuous replenishment of latent reservoir in vivo (92). Combining H2S donors with

594 ART regiments could potentially prevent reservoir replenishment during infection.

595 In conclusion, we identify H2S as a central factor in HIV-1 reactivation. Our

596 systematic mechanistic dissection of the role of H2S in cellular bioenergetics, redox

597 metabolism, and latency unifies many previous phenomena associated with viral

598 persistence. Lastly, our results provide a rationale for including pharmacological

599 donors of H2S in strategies to eradicate HIV in vivo.

600

601 Material and methods

602 Study design

24 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

603 The primary objective of this study was to understand the role H2S gas in

604 modulating HIV latency and reactivation. First, we examined the differential

605 expression of enzymes involved in H2S biogenesis and levels of H2S upon HIV-1

606 latency and reactivation. Next, we genetically silenced the expression of the main H2S

607 producing gene CTH and showed its importance in maintaining HIV-1 latency. We

608 performed detailed mechanistic studies on understanding the role of CTH in regulating

609 cellular antioxidant response, mitochondrial respiration, ROS generation to maintain

610 HIV-1 latency. Lastly, pharmacological donor of H2S (GYY4137) was used to reliably

611 increase endogenous H2S levels and to study its consequence on HIV latency and

612 rebound. Finally, primary CD4+ T cells derived from ART treated HIV infected patients

613 were used to assess the effect of GYY4137 in modulating HIV-1 rebound by improving

614 mitochondrial bioenergetics and mitigating redox stress.

615

616 Subject samples

617 Peripheral blood mononuclear cells (PBMCs) were collected from five HIV-1

618 seropositive subjects on stable suppressive ART (Table S5). All subject signed

619 informed consent forms approved by Indian Institute of Science, Bangalore and

620 Bangalore Medical College and Research Institute (BMCRI) review boards (Institute

621 human ethics committee [IHEC] No-3-14012020).

622

623 Mammalian cell lines and culture conditions

+ 624 The human pro-monocytic cell line U937, CD4 T lymphocytic cell line Jurkat,

625 HEK293T were procured from ATCC, Manassas, VA. The chronically infected ACH-

626 2, J-Lat 6.3, U1, J1.1 and TZM-bl cell lines were obtained through AIDS Research and

627 Reference Reagent Program, NIH, USA. The cell lines were cultured in RPMI1640

25 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

628 (Cell Clone) supplemented with L-glutamine (2 mM), 10% fetal bovine serum (MP

629 biomedicals), penicillin (100 units/mL), streptomycin (100 µg/mL) at 37°C and 5%

630 CO2. HEK293T and TZM-bl cells were cultured in DMEM (Cell Clone) supplemented

631 with 10% FBS.

632

633 Chemical reagents

634 Sodium hydrosulfide (NaHS), morpholin-4-ium 4-methoxphenyl(morpholino)

635 phosphinodithioate dichloromethane complex (GYY4137), Phorbol-12-myristate-13-

636 acetate (PMA), N-acetyl cysteine (NAC), prostratin and L-Cysteine were purchased

637 from Sigma-Aldrich. Recombinant human TNF-a was purchased from InvivoGen. The

638 antiretroviral drugs efavirenz, zidovudine, raltegravir, and lamivudine were obtained

639 through the NIH AIDS Reagent Program.

640

641 Latent viral reactivation

642 Latently infected U1 and J1.1 (2 ´ 105 cells/ml) were stimulated with PMA (5

643 ng/ml) or TNF-a (100 ng/ml) for the time indicated in the figure legends. HIV-1

644 reactivation was determined by intracellular gag RT-qPCR or p24 estimation in

645 supernatant by HIV-1 p24 ELISA (J. Mitra and Co. Pvt. Ltd., India). J-Lat 6.3 cells

646 were stimulated with Prostratin (2.5 µM) for 24 h and HIV-1 reactivation was assessed

647 by estimating GFP+ cells (excitation: 488 nm; emission: 510 nm) using BD

648 FACSVerse flow cytometer (BD Biosciences). The data were analysed using FACSuite

649 software (BD Biosciences).

650

651 Reverse transcription quantitative PCR (RT-qPCR)

26 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

652 Total cellular RNA was isolated by RNeasy mini kit (Qiagen), according to the

653 manufacturer’s protocol. RNA (500 ng) was reverse transcribed to cDNA (iScriptä

654 cDNA synthesis kit, Bio-Rad), subjected to quantitative real-time PCR (iQä SYBR®

655 Green Supermix, Bio-Rad), and performed using the Bio-Rad C1000ä real time PCR

656 system. HIV-1 reactivation was assessed using gene-specific primers (Table S6). The

657 expression level of each gene is normalized to human β-actin as an internal reference

658 gene.

659

660 Western blot analysis

661 Total cell lysates of PMA-treated and untreated U1, J1.1, U937, and Jurkat cell

662 lines were prepared using radioimmunoprecipitation (RIPA) lysis buffer (50 mM Tris

663 [pH 8.0], 150 mM NaCl, 1% Triton X-100, 1 % sodium deoxycholate, 0.1% SDS

664 (sodium dodecyl sulfate), 1´ protease inhibitor cocktail (Sigma-Aldrich), 1´

665 phosphatase inhibitor cocktail (Sigma-Aldrich). After incubation on ice for 20 min, the

666 lysates were centrifuged at 12,000 rpm, 4°C for 15 mins. Clarified supernatant was

667 taken and total protein concentration was determined by Bicinchoninic acid assay

668 (Pierceä, Thermo Fisher Scientific). Total protein extracts were separated by SDS-

669 PAGE and transferred onto polyvinylidene difluoride membranes. Membranes were

670 probed with an anti-CBS (EPR8579), CTH (ab151769), MPST (ab154514), and anti-

671 HIV-1 p24 (ab9071) from abcam; Nrf2 (CST-12721), Keap1 (CST-4678), NF-kB p65

672 (CST-6956), phospho-NF-kB p65 (Ser536) (CST-3033), YY1 (CST-63227) and

673 GAPDH (CST-97166) from Cell Signaling Technologies, Inc. Anti-rabbit IgG (CST-

674 7074) and anti-mouse IgG (CST-7076) were used a secondary antibodies. Proteins were

675 detected by ECL and visualized by chemiluminescence (Perkin Elmer, Waltham, MA)

676 using the Bio-Rad Chemidoc Imaging system (Hercules, CA). For membrane

27 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

677 reprobing, stripping buffer was used (2% SDS [w/v], 62 mM Tris-Cl buffer (0.5 M, pH

678 6.7) and 100 mM b-mercaptoethanol) for 20 min at 55°C. After extensive washing with

679 PBS containing 0.1 % Tween 20 (Sigma-Aldrich), membrane was blocked and

680 reincubated with desired antibodies.

681

682 H2S detection assays

683 Endogenous H2S levels of U1 and U937 cells were detected using H2S specific

684 fluorescent probe as described (43). Briefly, cells were treated with PMA (5 ng/ml) for

685 24 h, washed with 1´ PBS, and stained with 200 µM 7-Azido-4-Methylcoumarin

686 (AzMC) (Sigma). The stained cells were mounted on a glass slide and visualized using

687 Leica TCS SP5 confocal microscope (excitation: 405 nm; emission: 450 nm). Images

688 obtained were analyzed using LAS AF Lite software (Leica Microsystems) and semi-

689 quantification of 50 cells was performed using ImageJ software.

690 H2S generation was also measured using methylene blue assay. The supernatant

691 of U1 cells treated with NaHS or GYY4137 was incubated with Zinc acetate (1%) and

692 NaOH (3%) (1:1 ratio) to trap H2S for 30 min. The reaction was terminated using 10%

693 trichloroacetic acid solution. Following this, reactants were incubated with 20 mM N,N-

694 dimethylphenylendiamine (NNDPD) in 7.2 M HCl and 30 mM FeCl3 in 1.2 M HCl for

695 30 mins and absorbance was measured at 670 nm. The concentration of H2S was

696 determined by plotting absorbance on a standard curve generated using NaHS (0-400

697 μM; R2=0.9982).

698

699 Stable cell line generation

700 For generating CBS and CTH knockdown in U1 and J1.1 cells, we used

701 validated pooled gene specific shRNAs from the RNAi Consortium (TRC) library

28 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

702 (Sigma Aldrich, USA; shRNA sequences given in Table S1). The lentiviral particles

703 were generated in HEK293T cells using the packaging vectors, psPAX2 and pMD2.G.

704 The pLKO.1-puro vector encoding a non-mammalian targeting shRNA (shNT) was

705 used as a control. The U1 cells were transduced with lentiviral particles in opti-MEM

706 containing polybrene (10 µg/ml) for 6 hours. Cells were washed and stable clones were

707 selected in culture medium containing 250 ng/ml of puromycin. Total RNA or cell

708 lysates were prepared to validate knockdown of CBS and CTH.

709

710 Intracellular HIV-1 p24 staining

711 For intracellular p24 staining, U1-shCTH and U1-shNT cells were stimulated

712 with PMA (5 ng/ml), washed with PBS followed by fixation and permeabilization using

713 a fixation and permeabilization kit (eBiosciences). Permeabilized cells were then

714 incubated with 50 µl of 1:100 dilution of phycoerythrin (PE)-conjugated mouse anti-

715 p24 monoclonal antibody (KC57-RD1; Beckman Coulter, Inc.) for 30 min at room

716 temperature. After incubation the cells were washed twice and the fluorescence of

717 stained samples were acquired using BD FACSVerse flow cytometer (BD

718 Biosciences). The data were analysed using FACSuite software (BD Biosciences).

719

720 NanoString nCounter assay

721 Total RNA was isolated using an RNeasy mini kit (Qiagen) according to

722 manufacturer’s instructions. RNA concentration and purity were measured using a

723 Nanodrop spectrophotometer (Thermo Fisher Scientific, Waltham, MA), bioanalyzers

724 systems (Agilent Technologies, Inc.), and Qubit Assays (Thermo Fisher Scientific). An

725 nCounter gene expression assay was performed according to the manufacturer’s

726 protocol. The assay utilized a custom-made NanoString codeset designed to measure

29 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

727 185 genes, including 6 house-keeping genes (Table S2). This custom-made panel

728 included genes associated with oxidative stress and HIV-1 infection (Table S2) All

729 genes were assayed simultaneously in multiplexed reactions and analyzed by fully

730 automated nCounter Prep Station and digital analyzer (NanoString Technologies). The

731 data were normalized to B2M, used as a housekeeping gene due to its minimum % CV

732 across the samples, and analysis was done using nSolver 4.0 software.

733

734 Measurement of Oxygen Consumption Rates

735 Oxygen consumption rates (OCR) were measured using a Seahorse XFp

736 extracellular flux analyzer (Agilent Technologies) as per manufacturer’s instructions.

737 Briefly, cells (U1 or Primary CD4+ T cells) were seeded at a density of 104-105 per well

738 in a Seahorse flux analyzer plate precoated with Cell-Tak (Corning). Cells were

739 incubated for 1 h in a non-CO2 incubator at 37°C before loading the plate in the seahorse

740 analyzer. To assess mitochondrial respiration, three OCR measurements were

741 performed without any inhibitor in XF assay media to measure basal respiration,

742 followed by sequential addition of oligomycin (1 µM), an ATP synthase inhibitor

743 (complex V) and three OCR measurements to determine ATP-linked OCR and proton

744 leakage. Next, cyanide-4-(trifluoromethoxy)phenylhydrazone (FCCP; 0.25 µM), was

745 injected to determine the maximal respiration rate and the spare respiratory capacity

746 (SRC). Finally, rotenone (0.5 µM) and antimycin A (0.5 µM), inhibitors of NADH

747 dehydrogenase (complex I) and cytochrome c - oxidoreductase (complex III),

748 respectively, were injected to completely shut down the electron transport chain (ETC)

749 to analyze non-mitochondrial oxygen consumption rate (nmOCR). Seahorse data were

750 normalized to total amount of protein (µg) and mitochondrial respiration parameters

751 were analyzed using Wave Desktop 2.6 software (Agilent Technologies).

30 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

752

753 Estimation of intracellular glutathione content and ROS

754 Total cell lysate was prepared from 107 cells using sonication in MES (2-(N-

755 morpholino)ethanesulfonic acid) buffer. Lysates were clarified by centrifugation and

756 total protein concentration was estimated using BCA assay. Total glutathione, reduced

757 glutathione (GSH), and oxidized glutathione (GSSG) were measured using the

758 glutathione assay kit (Cayman Chemical, Ann Arbor, MI, USA) according to the

759 manufacturer’s instructions. To measure ROS, cells were loaded with 10 µM of CM-

760 H2DCFDA (Excitation: 492 nm; Emission; 517 nm) or 5 µM of MitoSOXÔ Red

761 (Excitation: 510 nm; Emission; 580 nm) for 30 min at 37°C and exposed to H2O2 (100

762 µM) or antimycin A (2 µM) or PMA (5 ng/ml) or left untreated at 37°C and 5% CO2.

763 The fluorescence of stained samples was acquired using BD FACSVerse flow

764 cytometer (BD Biosciences). Data were analyzed using FlowJo software (BD

765 Biosciences).

766

767 Virus Production

768 HIV-1 particle production was carried out using Lipofectamine 2000

769 transfection reagent (Invitrogen, Life Technologies), according to the manufacturer’s

770 protocol, in HEK293T cells using HIV-1 NL4-3 DNA (NIH AIDS Reagent Program,

771 Division of AIDS, NIAID, NIH). The medium was replaced with fresh medium at 6 h

772 post-transfection, and supernatants were collected after 60 h, centrifuged (10 min, 200

773 ´g, room temperature), and filtered through a 0.45 µm-pore-size membrane filter (MDI;

774 Membrane Technologies) to clear cell debris. Virus was concentrated using 5´ PEG-

775 itÔ (System Biosciences) as per manufacturer’s protocol and virus pellet obtained was

31 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

776 aliquoted in opti-MEM and stored at -80°C. Viral titration was done using HIV-1

777 reporter cell line, TZM-bl (NIH AIDS reagent program) as described earlier (93).

778

779 Assessing the effect of GYY4137 on HIV-1 replication

780 Briefly, 0.5´106 Jurkat cells were infected with HIV-1 NL4.3 at 0.2 M.O.I for

781 4 h at 37°C. Post-infection cells were washed with 1´ PBS and cultured in presence or

782 absence of 300 μM GYY4137. Cells and culture supernatants were harvested to isolate

783 total RNA for gag RT-qPCR and p24 estimation, respectively. The p24 concentration

784 was determined in the culture supernatant by sandwich HIV-1 p24 ELISA (J. Mitra and

785 Co. Pvt. Ltd., India) according to the manufacturer’s instruction.

786

787 Chromatin immunoprecipitation and quantitative genomic PCR

788 The chromatin immunoprecipitation (ChIP) assays was performed using

789 SimpleChIPÒ Enzymatic Chromatin IP Kit (Magnetic Beads) (Cell Signaling

790 Technologies, Inc., MA, US) according to the manufacturer’s instructions. Briefly, 107

791 U1 cells were pre-treated with GYY4137 (5 mM) for 6 h or left untreated and then

792 stimulated with PMA (30 ng/ml) for 4 h. Cells were fixed using 1% formaldehyde,

793 neutralized with glycine, and harvested in ice-cold PBS. Cells were lysed and nuclei

794 were sheared by sonication using BioruptorÒ Pico (Diagenode Inc.) to obtain DNA

795 fragments of 200 to 500 nucleotides. The clarified lysates were immunoprecipitated

796 using ChIP-grade anti-NF-kB p65 (CST-6956) or anti-YY1 (CST-63227) or normal

797 rabbit IgG (as a negative control) for overnight at 4°C followed by incubation with

798 ChIP-grade protein G magnetic beads at 4°C for 4 h. Chromatin was eluted from protein

799 G beads, reverse-cross linked, and DNA was column purified. Quantitative genomic

800 PCR was done by SYBR green-based real-time PCR using primers spanning the ER,

32 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

801 RBE I, and RBE III regions on HIV-1 LTR. The GAPDH gene was used as the

802 reference gene to see non-specific binding. Total input (10 %) was used to normalize

803 equal amount of chromatin taken across the samples. The relative proportion of co-

804 immunoprecipitated DNA fragments were determined with the help of threshold cycle

[CT(input-3.32)- CT(IP)] 805 (CT) values for each qPCR product using the equation (100 ´ 2 ). The

806 data obtained were represented as fold enrichment normalized to IgG background for

807 each IP reaction.

808

809 Generation of expanded primary CD4+ T cells from aviremic subjects

810 The PBMCs isolated from blood samples using Histopaque-1077 (Sigma-

811 Aldrich) density gradient centrifugation were used for CD4+ T cells purification. The

812 CD4+ T cells were purified from 50 × 106 PBMCs using EasySepä Human CD4+ T cell

813 isolation kit (STEMCELL Technologies). Primary CD4+ T cells were cultured at 37°C

814 in a 5 % (v/v) CO2 humidified atmosphere in Gibcoâ RPMI 1640 medium (Life

815 Technologies) with GlutaMAX, HEPES, 100 U/ml interleukin-2 (IL-2; Peprotech,

816 London, United Kingdom), 1 µg/mL phytohemagglutinin (PHA) (Thermo Fisher

817 Scientific), gamma-irradiated feeder PBMCs (healthy control) and either ART alone

818 (100 nM efavirenz, 180 nM zidovudine, and 200 nM raltegravir) or ART + 100 µM

819 GYY4137. After 7 days, CD4+ T cells were cultured only with ART alone or ART +

820 GYY4137 and 100 U/ml IL-2. For stimulation experiments on day 28, ART and ART

821 + GYY4137 were washed off and 1 × 106 cells were treated with 1 µM prostratin for

822 24 h.

823

824 HIV-1 RNA and DNA isolation from primary CD4+ T cells

33 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

825 CD4+ T cells (1 × 106) from ART or ART + GYY4137 treated groups were

826 harvested to isolate total RNA using Qiagen RNAeasy isolation kit and 200 ng of total

827 RNA was reverse transcribed (iScriptä cDNA synthesis kit, Bio-Rad). Reverse

828 transcribed cDNA was diluted 10-fold and amplified using primers against HIV LTRs

829 and seminested – PCR was performed using primers and probe listed in Table S6.

830 Serially diluted pNL4.3 plasmid was used to obtain the standard curve. Isolation and

831 RT-qPCR of total HIV DNA was performed as described earlier (60). Briefly, 1 × 106

832 cells were lysed (10 mM Tris-HCl, 50 nM KCl, 400 mg/ml proteinase K) at 55°C for

833 16 h followed by inactivation at 95°C for 5 min. Digested product was used as a

834 template to set up first PCR with Taq polymerase (NEB), 1X Taq buffer, dNTPs, HIV

835 and CD3 primers for 12 cycles. The second round amplification was done using

836 seminested PCR strategy wherein 10 fold dilution of first round PCR product was used

837 as a template, HIV and CD3 primers/probes, Taqman™ Fast Advance master mix

838 (Applied Biosystemsä) using SetupOnePlusä Real-time PCR system (Applied

839 Biosystemsä). DNA isolated from ACH2 cells that contain single copy of HIV per cell

840 was used to obtain standard curve.

841

842 Surface marker analysis of primary CD4+ T cells derived from HIV patients

843 CD4+ T cells derived from HIV-1 infected patients were stained for surface

844 markers using monoclonal antibodies: CD4-BUV395 (SK3), CD45RA-APC-H7

845 (HI100), CD27-BV785, CCR7-Alexa 647 (G043H7), CD38-PE-Cy5 and CD127-

846 PerCP-Cy5.5. Additionally, cells were stained with Live/Dead fixable Aqua dead cell

847 stain or AviD (Invitrogen) to exclude dead cells from the analysis as per manufacturer’s

848 instructions. Stained samples were run on BD FACSAriaÔ Fusion flow cytometer (BD

34 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

849 Biosciences, San Jose, CA) and data were analyzed with FlowJo version 9.9.6 software

850 (Treestar, Ashland, OR).

851

852 Statistical analysis

853 All statistical analyses were performed using GraphPad Prism software for

854 Macintosh (version 9.0.0). The data values are indicated as mean ± S.D. For statistical

855 analysis Student t-test (in which two groups are compared) and one-way or two-way

856 ANOVA (for analysis involving multiple groups) were used. Analysis of NanoString

857 data was performed using the nSolver platform. Differences in P values <0.05 and fold

858 change >1.5 were considered significant.

859

860 Acknowledgements:

861 We are grateful to Prof. C. Grundner at the Seattle Children’s Research Institute for

862 critical reading of the manuscript and valuable input. We acknowledge Dr. D. K. Saini

863 at Department of Molecular Reproduction, Development and Genetics, IISc for

864 providing shRNA constructs used in this study. We are grateful to Ms. S. Kumar and

865 Dr. N. R. Sundaresan at Department of Microbiology and Cell Biology, IISc for helping

866 with ChIP experiments carried out in this study. We gratefully acknowledge the

867 NanoString services provided by TheraCUES Innovations Pvt Ltd, Bangalore

868

869 Funding:

870 This work was supported by Wellcome Trust-Department of Biotechnology (DBT)

871 India Alliance grant IA/S/16/2/502700 (A.S.) and in part by DBT grants

872 BT/PR13522/COE/34/27/2015, BT/PR29098/Med/29/1324/2018, and

873 BT/HRD/NBA/39/07/2018-19 (A.S.), DBT-IISc Partnership Program grant 22-0905-

35 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

874 0006-05-987 436, and the Infosys Foundation. A.S. is a senior fellow of Wellcome

875 Trust-DBT India Alliance. VKP is grateful to Indian Institute of Science for fellowship.

876

877 Authors contributions:

878 VKP and AS participated in the design of the study. VKP, RA, and SR carried out the

879 experiment. VKP, PS, SR, and DTNM contributed in recruiting and isolating PBMCs

880 from ART treated HIV-1 infected subjects. VKP, SR, AV, and AS contributed to

881 reagents and analyzed the data. VKP and AS conceived the study, supervised the

882 project, and drafted the manuscript. All authors read and approved the final manuscript.

883

884 Competing interests:

885 The authors declare that they have no conflict of interests.

886

887 Data Availability:

888 This study includes no data deposited in external repositories.

889

890 This article includes one Supplementary file with the following contents:

891 Figure 1-figure supplement 1. HIV-1 reactivation decreases levels of H2S

892 metabolizing enzymes in a T cell line model of latency.

893 Figure 2-figure supplement 1. Genetic silencing of CTH reduces endogenous H2S

894 levels and reactivates HIV-1 from latency.

895 Figure 3-figure supplement 1. Effect of CTH knockdown on cytosolic ROS

896 generation.

897 Figure 4-figure supplement 1. Effect of GYY4137 on HIV-1 reactivation and cellular

898 viability.

36 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

899 Figure 6-figure Supplement 1. Phenotypic features of CD4+ T cells were preserved

900 upon prolong treatment with GYY4137.

901 Table S1. Sequences of shRNA clones.

902 Table S2. List of genes used for NanoString nCounter Gene Expression Analysis

903 Table S3. List of differentially expressed genes from NanoString nCounter Gene

904 Expression analysis for U1-shNT, U1-shCTH, U1-shNT+PMA and U1-shCTH+PMA

905 samples.

906 Table S4. List of differentially expressed genes from NanoString nCounter Gene

907 Expression analysis for U1 untreated, GYY4137, PMA and GYY4137+PMA treated

908 samples.

909 Table S5. Characteristic of ART treated HIV-1 infected study participants

910 Table S6. List of primers and probes used in this study.

911

912 Additional data files

913 Figure 2-source data file 1. This file contains RAW values associated with the

914 NanoString analysis of host genes affected upon depletion of CTH in U1.

915 Figure 5-source data file 1. This file contains RAW values associated with the

916 NanoString analysis of host genes affected upon PMA induced HIV reactivation in

917 presence or absence of GYY4137 treatment.

918

919 References

920 1. Finzi D, Hermankova M, Pierson T, Carruth LM, Buck C, Chaisson RE, et al. 921 Identification of a reservoir for HIV-1 in patients on highly active antiretroviral 922 therapy. Science. 1997;278(5341):1295-300. 923 2. Deeks SG, Lewin SR, Ross AL, Ananworanich J, Benkirane M, Cannon P, et al. 924 International AIDS Society global scientific strategy: towards an HIV cure 2016. 925 Nat Med. 2016;22(8):839-50.

37 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

926 3. Chinta KC, Saini V, Glasgow JN, Mazorodze JH, Rahman MA, Reddy D, et al. 927 The emerging role of gasotransmitters in the pathogenesis of tuberculosis. Nitric 928 Oxide. 2016;59:28-41. 929 4. Pal VK, Bandyopadhyay P, Singh A. Hydrogen sulfide in physiology and 930 pathogenesis of bacteria and viruses. IUBMB Life. 2018;70(5):393-410. 931 5. Tinajero-Trejo M, Jesse HE, Poole RK. Gasotransmitters, poisons, and 932 antimicrobials: it's a gas, gas, gas! F1000Prime Rep. 2013;5:28. 933 6. Rahman MA, Cumming BM, Addicott KW, Pacl HT, Russell SL, Nargan K, et 934 al. Hydrogen sulfide dysregulates the immune response by suppressing central 935 carbon metabolism to promote tuberculosis. Proc Natl Acad Sci U S A. 936 2020;117(12):6663-74. 937 7. Shiloh MU, Manzanillo P, Cox JS. Mycobacterium tuberculosis senses host- 938 derived carbon monoxide during macrophage infection. Cell Host Microbe. 939 2008;3(5):323-30. 940 8. Braverman J, Stanley SA. Nitric Oxide Modulates Macrophage Responses to 941 Mycobacterium tuberculosis Infection through Activation of HIF-1alpha and 942 Repression of NF-kappaB. J Immunol. 2017;199(5):1805-16. 943 9. Blond D, Raoul H, Le Grand R, Dormont D. Nitric oxide synthesis enhances 944 human immunodeficiency virus replication in primary human macrophages. J 945 Virol. 2000;74(19):8904-12. 946 10. Devadas K, Dhawan S. Hemin activation ameliorates HIV-1 infection via heme 947 oxygenase-1 induction. J Immunol. 2006;176(7):4252-7. 948 11. Liu XM, Durante ZE, Peyton KJ, Durante W. Heme oxygenase-1-derived 949 bilirubin counteracts HIV protease inhibitor-mediated endothelial cell 950 dysfunction. Free Radic Biol Med. 2016;94:218-29. 951 12. Hyrcza MD, Kovacs C, Loutfy M, Halpenny R, Heisler L, Yang S, et al. Distinct 952 transcriptional profiles in ex vivo CD4+ and CD8+ T cells are established early 953 in human immunodeficiency virus type 1 infection and are characterized by a 954 chronic interferon response as well as extensive transcriptional changes in CD8+ 955 T cells. J Virol. 2007;81(7):3477-86. 956 13. Martin JA, Sastre J, de la Asuncion JG, Pallardo FV, Vina J. Hepatic gamma- 957 cystathionase deficiency in patients with AIDS. JAMA. 2001;285(11):1444-5. 958 14. Hortin GL, Landt M, Powderly WG. Changes in plasma amino acid 959 concentrations in response to HIV-1 infection. Clin Chem. 1994;40(5):785-9. 960 15. Staal FJ, Ela SW, Roederer M, Anderson MT, Herzenberg LA, Herzenberg LA. 961 Glutathione deficiency and human immunodeficiency virus infection. Lancet. 962 1992;339(8798):909-12. 963 16. Kimura Y, Goto Y, Kimura H. Hydrogen sulfide increases glutathione production 964 and suppresses oxidative stress in mitochondria. Antioxid Redox Signal. 965 2010;12(1):1-13. 966 17. McBean GJ. The transsulfuration pathway: a source of cysteine for glutathione 967 in astrocytes. Amino Acids. 2012;42(1):199-205. 968 18. Roederer M, Staal FJ, Ela SW, Herzenberg LA, Herzenberg LA. N- 969 acetylcysteine: potential for AIDS therapy. Pharmacology. 1993;46(3):121-9. 970 19. Miller TW, Wang EA, Gould S, Stein EV, Kaur S, Lim L, et al. Hydrogen sulfide 971 is an endogenous potentiator of T cell activation. J Biol Chem. 2012;287(6):4211- 972 21. 973 20. Mathai JC, Missner A, Kugler P, Saparov SM, Zeidel ML, Lee JK, et al. No 974 facilitator required for membrane transport of hydrogen sulfide. Proc Natl Acad 975 Sci U S A. 2009;106(39):16633-8.

38 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

976 21. Modis K, Bos EM, Calzia E, van Goor H, Coletta C, Papapetropoulos A, et al. 977 Regulation of mitochondrial bioenergetic function by hydrogen sulfide. Part II. 978 Pathophysiological and therapeutic aspects. Br J Pharmacol. 2014;171(8):2123- 979 46. 980 22. Fu M, Zhang W, Wu L, Yang G, Li H, Wang R. Hydrogen sulfide (H2S) 981 metabolism in mitochondria and its regulatory role in energy production. Proc 982 Natl Acad Sci U S A. 2012;109(8):2943-8. 983 23. Deeks SG, Tracy R, Douek DC. Systemic effects of inflammation on health 984 during chronic HIV infection. Immunity. 2013;39(4):633-45. 985 24. Nakanjako D, Ssewanyana I, Mayanja-Kizza H, Kiragga A, Colebunders R, 986 Manabe YC, et al. High T-cell immune activation and immune exhaustion among 987 individuals with suboptimal CD4 recovery after 4 years of antiretroviral therapy 988 in an African cohort. BMC Infect Dis. 2011;11:43. 989 25. Leng SX, Margolick JB. Understanding frailty, aging, and inflammation in HIV 990 infection. Curr HIV/AIDS Rep. 2015;12(1):25-32. 991 26. Alzahrani J, Hussain T, Simar D, Palchaudhuri R, Abdel-Mohsen M, Crowe SM, 992 et al. Inflammatory and immunometabolic consequences of gut dysfunction in 993 HIV: Parallels with IBD and implications for reservoir persistence and non-AIDS 994 comorbidities. EBioMedicine. 2019;46:522-31. 995 27. Tattevin P, Camus C, Arvieux C, Ruffault A, Michelet C. Multiple organ failure 996 during primary HIV infection. Clin Infect Dis. 2007;44(3):e28-9. 997 28. Valle-Casuso JC, Angin M, Volant S, Passaes C, Monceaux V, Mikhailova A, et 998 al. Cellular Metabolism Is a Major Determinant of HIV-1 Reservoir Seeding in 999 CD4(+) T Cells and Offers an Opportunity to Tackle Infection. Cell Metab. 1000 2019;29(3):611-26 e5. 1001 29. Hegedus A, Kavanagh Williamson M, Huthoff H. HIV-1 pathogenicity and 1002 virion production are dependent on the metabolic phenotype of activated CD4+ 1003 T cells. Retrovirology. 2014;11:98. 1004 30. Tyagi P, Pal VK, Agrawal R, Singh S, Srinivasan S, Singh A. Mycobacterium 1005 tuberculosis Reactivates HIV-1 via Exosome-Mediated Resetting of Cellular 1006 Redox Potential and Bioenergetics. mBio. 2020;11(2). 1007 31. Bhaskar A, Munshi M, Khan SZ, Fatima S, Arya R, Jameel S, et al. Measuring 1008 glutathione redox potential of HIV-1-infected macrophages. J Biol Chem. 1009 2015;290(2):1020-38. 1010 32. Savarino A, Mai A, Norelli S, El Daker S, Valente S, Rotili D, et al. "Shock and 1011 kill" effects of class I-selective histone deacetylase inhibitors in combination with 1012 the glutathione synthesis inhibitor buthionine sulfoximine in cell line models for 1013 HIV-1 quiescence. Retrovirology. 2009;6:52. 1014 33. Shytaj IL, Lucic B, Forcato M, Penzo C, Billingsley J, Laketa V, et al. Alterations 1015 of redox and iron metabolism accompany the development of HIV latency. 1016 EMBO J. 2020;39(9):e102209. 1017 34. Sen N, Paul BD, Gadalla MM, Mustafa AK, Sen T, Xu R, et al. Hydrogen sulfide- 1018 linked sulfhydration of NF-kappaB mediates its antiapoptotic actions. Mol Cell. 1019 2012;45(1):13-24. 1020 35. Xie L, Feng H, Li S, Meng G, Liu S, Tang X, et al. SIRT3 Mediates the 1021 Antioxidant Effect of Hydrogen Sulfide in Endothelial Cells. Antioxid Redox 1022 Signal. 2016;24(6):329-43. 1023 36. Williams SA, Kwon H, Chen LF, Greene WC. Sustained induction of NF-kappa 1024 B is required for efficient expression of latent human immunodeficiency virus 1025 type 1. J Virol. 2007;81(11):6043-56.

39 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1026 37. Brooks DG, Arlen PA, Gao L, Kitchen CM, Zack JA. Identification of T cell- 1027 signaling pathways that stimulate latent HIV in primary cells. Proc Natl Acad Sci 1028 U S A. 2003;100(22):12955-60. 1029 38. Duverger A, Wolschendorf F, Zhang M, Wagner F, Hatcher B, Jones J, et al. An 1030 AP-1 binding site in the enhancer/core element of the HIV-1 promoter controls 1031 the ability of HIV-1 to establish latent infection. J Virol. 2013;87(4):2264-77. 1032 39. Chirullo B, Sgarbanti R, Limongi D, Shytaj IL, Alvarez D, Das B, et al. A 1033 candidate anti-HIV reservoir compound, auranofin, exerts a selective 'anti- 1034 memory' effect by exploiting the baseline oxidative status of lymphocytes. Cell 1035 Death Dis. 2013;4:e944. 1036 40. Doka E, Pader I, Biro A, Johansson K, Cheng Q, Ballago K, et al. A novel 1037 persulfide detection method reveals protein persulfide- and polysulfide-reducing 1038 functions of thioredoxin and glutathione systems. Sci Adv. 2016;2(1):e1500968. 1039 41. Folks TM, Justement J, Kinter A, Dinarello CA, Fauci AS. Cytokine-induced 1040 expression of HIV-1 in a chronically infected promonocyte cell line. Science. 1041 1987;238(4828):800-2. 1042 42. Perez VL, Rowe T, Justement JS, Butera ST, June CH, Folks TM. An HIV-1- 1043 infected T cell clone defective in IL-2 production and Ca2+ mobilization after 1044 CD3 stimulation. J Immunol. 1991;147(9):3145-8. 1045 43. Chen B, Li W, Lv C, Zhao M, Jin H, Jin H, et al. Fluorescent probe for highly 1046 selective and sensitive detection of hydrogen sulfide in living cells and cardiac 1047 tissues. Analyst. 2013;138(3):946-51. 1048 44. Banerjee R. Hydrogen sulfide: redox metabolism and signaling. Antioxid Redox 1049 Signal. 2011;15(2):339-41. 1050 45. Gao C, Xu DQ, Gao CJ, Ding Q, Yao LN, Li ZC, et al. An exogenous hydrogen 1051 sulphide donor, NaHS, inhibits the nuclear factor kappaB inhibitor kinase/nuclear 1052 factor kappab inhibitor/nuclear factor-kappaB signaling pathway and exerts 1053 cardioprotective effects in a rat hemorrhagic shock model. Biol Pharm Bull. 1054 2012;35(7):1029-34. 1055 46. Bhatia M. Role of hydrogen sulfide in the pathology of inflammation. Scientifica 1056 (Cairo). 2012;2012:159680. 1057 47. Szabo C, Ransy C, Modis K, Andriamihaja M, Murghes B, Coletta C, et al. 1058 Regulation of mitochondrial bioenergetic function by hydrogen sulfide. Part I. 1059 Biochemical and physiological mechanisms. Br J Pharmacol. 2014;171(8):2099- 1060 122. 1061 48. Khan SZ, Hand N, Zeichner SL. Apoptosis-induced activation of HIV-1 in 1062 latently infected cell lines. Retrovirology. 2015;12:42. 1063 49. Sgarbanti M, Borsetti A, Moscufo N, Bellocchi MC, Ridolfi B, Nappi F, et al. 1064 Modulation of human immunodeficiency virus 1 replication by interferon 1065 regulatory factors. J Exp Med. 2002;195(10):1359-70. 1066 50. Nguyen NV, Tran JT, Sanchez DJ. HIV blocks Type I IFN signaling through 1067 disruption of STAT1 phosphorylation. Innate Immun. 2018;24(8):490-500. 1068 51. Gillick K, Pollpeter D, Phalora P, Kim EY, Wolinsky SM, Malim MH. 1069 Suppression of HIV-1 infection by APOBEC3 proteins in primary human CD4(+) 1070 T cells is associated with inhibition of processive reverse transcription as well as 1071 excessive cytidine deamination. J Virol. 2013;87(3):1508-17. 1072 52. Amini S, Clavo A, Nadraga Y, Giordano A, Khalili K, Sawaya BE. Interplay 1073 between cdk9 and NF-kappaB factors determines the level of HIV-1 gene 1074 transcription in astrocytic cells. Oncogene. 2002;21(37):5797-803.

40 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1075 53. Budhiraja S, Famiglietti M, Bosque A, Planelles V, Rice AP. Cyclin T1 and 1076 CDK9 T-loop phosphorylation are downregulated during establishment of HIV- 1077 1 latency in primary resting memory CD4+ T cells. J Virol. 2013;87(2):1211-20. 1078 54. McNeely TB, Dealy M, Dripps DJ, Orenstein JM, Eisenberg SP, Wahl SM. 1079 Secretory leukocyte protease inhibitor: a human saliva protein exhibiting anti- 1080 human immunodeficiency virus 1 activity in vitro. J Clin Invest. 1995;96(1):456- 1081 64. 1082 55. Rafati H, Parra M, Hakre S, Moshkin Y, Verdin E, Mahmoudi T. Repressive LTR 1083 nucleosome positioning by the BAF complex is required for HIV latency. PLoS 1084 Biol. 2011;9(11):e1001206. 1085 56. Coull JJ, Romerio F, Sun JM, Volker JL, Galvin KM, Davie JR, et al. The human 1086 factors YY1 and LSF repress the human immunodeficiency virus type 1 long 1087 terminal repeat via recruitment of histone deacetylase 1. J Virol. 1088 2000;74(15):6790-9. 1089 57. Badley AD, Roumier T, Lum JJ, Kroemer G. Mitochondrion-mediated apoptosis 1090 in HIV-1 infection. Trends Pharmacol Sci. 2003;24(6):298-305. 1091 58. Pardons M, Fromentin R, Pagliuzza A, Routy JP, Chomont N. Latency-Reversing 1092 Agents Induce Differential Responses in Distinct Memory CD4 T Cell Subsets in 1093 Individuals on Antiretroviral Therapy. Cell Rep. 2019;29(9):2783-95 e5. 1094 59. Bailon L, Mothe B, Berman L, Brander C. Novel Approaches Towards a 1095 Functional Cure of HIV/AIDS. Drugs. 2020;80(9):859-68. 1096 60. Kessing CF, Nixon CC, Li C, Tsai P, Takata H, Mousseau G, et al. In Vivo 1097 Suppression of HIV Rebound by Didehydro-Cortistatin A, a "Block-and-Lock" 1098 Strategy for HIV-1 Treatment. Cell Rep. 2017;21(3):600-11. 1099 61. Li L, Whiteman M, Guan YY, Neo KL, Cheng Y, Lee SW, et al. Characterization 1100 of a novel, water-soluble hydrogen sulfide-releasing molecule (GYY4137): new 1101 insights into the biology of hydrogen sulfide. Circulation. 2008;117(18):2351-60. 1102 62. Roederer M, Staal FJ, Raju PA, Ela SW, Herzenberg LA, Herzenberg LA. 1103 Cytokine-stimulated human immunodeficiency virus replication is inhibited by 1104 N-acetyl-L-cysteine. Proc Natl Acad Sci U S A. 1990;87(12):4884-8. 1105 63. Jordan A, Bisgrove D, Verdin E. HIV reproducibly establishes a latent infection 1106 after acute infection of T cells in vitro. EMBO J. 2003;22(8):1868-77. 1107 64. Pall ML, Levine S. Nrf2, a master regulator of detoxification and also antioxidant, 1108 anti-inflammatory and other cytoprotective mechanisms, is raised by health 1109 promoting factors. Sheng Li Xue Bao. 2015;67(1):1-18. 1110 65. Furuya AK, Sharifi HJ, Jellinger RM, Cristofano P, Shi B, de Noronha CM. 1111 Sulforaphane Inhibits HIV Infection of Macrophages through Nrf2. PLoS Pathog. 1112 2016;12(4):e1005581. 1113 66. Yang G, Zhao K, Ju Y, Mani S, Cao Q, Puukila S, et al. Hydrogen sulfide protects 1114 against cellular senescence via S-sulfhydration of Keap1 and activation of Nrf2. 1115 Antioxid Redox Signal. 2013;18(15):1906-19. 1116 67. Zhong H, Voll RE, Ghosh S. Phosphorylation of NF-kappa B p65 by PKA 1117 stimulates transcriptional activity by promoting a novel bivalent interaction with 1118 the coactivator CBP/p300. Mol Cell. 1998;1(5):661-71. 1119 68. Bernhard W, Barreto K, Raithatha S, Sadowski I. An upstream YY1 binding site 1120 on the HIV-1 LTR contributes to latent infection. PLoS One. 2013;8(10):e77052. 1121 69. Chomont N, El-Far M, Ancuta P, Trautmann L, Procopio FA, Yassine-Diab B, et 1122 al. HIV reservoir size and persistence are driven by T cell survival and 1123 homeostatic proliferation. Nat Med. 2009;15(8):893-900.

41 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1124 70. Brenchley JM, Hill BJ, Ambrozak DR, Price DA, Guenaga FJ, Casazza JP, et al. 1125 T-cell subsets that harbor human immunodeficiency virus (HIV) in vivo: 1126 implications for HIV pathogenesis. J Virol. 2004;78(3):1160-8. 1127 71. Pardons M, Baxter AE, Massanella M, Pagliuzza A, Fromentin R, Dufour C, et 1128 al. Single-cell characterization and quantification of translation-competent viral 1129 reservoirs in treated and untreated HIV infection. PLoS Pathog. 1130 2019;15(2):e1007619. 1131 72. Szabo C. Hydrogen sulphide and its therapeutic potential. Nat Rev Drug Discov. 1132 2007;6(11):917-35. 1133 73. Murphy MP. How mitochondria produce reactive oxygen species. Biochem J. 1134 2009;417(1):1-13. 1135 74. Kim C, Kim JY, Kim JH. Cytosolic phospholipase A(2), lipoxygenase 1136 metabolites, and reactive oxygen species. BMB Rep. 2008;41(8):555-9. 1137 75. Chacko BK, Kramer PA, Ravi S, Benavides GA, Mitchell T, Dranka BP, et al. 1138 The Bioenergetic Health Index: a new concept in mitochondrial translational 1139 research. Clin Sci (Lond). 2014;127(6):367-73. 1140 76. Hu LF, Lu M, Wu ZY, Wong PT, Bian JS. Hydrogen sulfide inhibits rotenone- 1141 induced apoptosis via preservation of mitochondrial function. Mol Pharmacol. 1142 2009;75(1):27-34. 1143 77. Cohen I, Boya P, Zhao L, Metivier D, Andreau K, Perfettini JL, et al. Anti- 1144 apoptotic activity of the glutathione peroxidase homologue encoded by HIV-1. 1145 Apoptosis. 2004;9(2):181-92. 1146 78. Castellano P, Prevedel L, Valdebenito S, Eugenin EA. HIV infection and latency 1147 induce a unique metabolic signature in human macrophages. Sci Rep. 1148 2019;9(1):3941. 1149 79. Cao H, Zhou X, Zhang J, Huang X, Zhai Y, Zhang X, et al. Hydrogen sulfide 1150 protects against bleomycin-induced pulmonary fibrosis in rats by inhibiting NF- 1151 kappaB expression and regulating Th1/Th2 balance. Toxicol Lett. 1152 2014;224(3):387-94. 1153 80. Li H, Ma Y, Escaffre O, Ivanciuc T, Komaravelli N, Kelley JP, et al. Role of 1154 hydrogen sulfide in paramyxovirus infections. J Virol. 2015;89(10):5557-68. 1155 81. Lange UC, Verdikt R, Ait-Ammar A, Van Lint C. Epigenetic crosstalk in chronic 1156 infection with HIV-1. Semin Immunopathol. 2020;42(2):187-200. 1157 82. Kumar A, Darcis G, Van Lint C, Herbein G. Epigenetic control of HIV-1 post 1158 integration latency: implications for therapy. Clin Epigenetics. 2015;7:103. 1159 83. Bonavida B, Baritaki S. Inhibition of Epithelial-to-Mesenchymal Transition 1160 (EMT) in Cancer by Nitric Oxide: Pivotal Roles of Nitrosylation of NF-kappaB, 1161 YY1 and Snail. For Immunopathol Dis Therap. 2012;3(2):125-33. 1162 84. Hatzakis AE, Touloumi G, Pantazis N, Anastassopoulou CG, Katsarou O, 1163 Karafoulidou A, et al. Cellular HIV-1 DNA load predicts HIV-RNA rebound and 1164 the outcome of highly active antiretroviral therapy. AIDS. 2004;18(17):2261-7. 1165 85. Persaud D, Gay H, Ziemniak C, Chen YH, Piatak M, Jr., Chun TW, et al. Absence 1166 of detectable HIV-1 viremia after treatment cessation in an infant. N Engl J Med. 1167 2013;369(19):1828-35. 1168 86. Saez-Cirion A, Bacchus C, Hocqueloux L, Avettand-Fenoel V, Girault I, 1169 Lecuroux C, et al. Post-treatment HIV-1 controllers with a long-term virological 1170 remission after the interruption of early initiated antiretroviral therapy ANRS 1171 VISCONTI Study. PLoS Pathog. 2013;9(3):e1003211. 1172 87. Frange P, Faye A, Avettand-Fenoel V, Bellaton E, Descamps D, Angin M, et al. 1173 HIV-1 virological remission lasting more than 12 years after interruption of early

42 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1174 antiretroviral therapy in a perinatally infected teenager enrolled in the French 1175 ANRS EPF-CO10 paediatric cohort: a case report. Lancet HIV. 2016;3(1):e49- 1176 54. 1177 88. Ibeh BO, Emeka-Nwabunnia IK. Increased oxidative stress condition found in 1178 different stages of HIV disease in patients undergoing antiretroviral therapy in 1179 Umuahia (Nigeria). Immunopharmacol Immunotoxicol. 2012;34(6):1060-6. 1180 89. Hulgan T, Gerschenson M. HIV and mitochondria: more than just drug toxicity. 1181 J Infect Dis. 2012;205(12):1769-71. 1182 90. Reyskens KM, Fisher TL, Schisler JC, O'Connor WG, Rogers AB, Willis MS, et 1183 al. Cardio-metabolic effectsof HIV protease inhibitors (lopinavir/ritonavir). PLoS 1184 One. 2013;8(9):e73347. 1185 91. Mousseau G, Aneja R, Clementz MA, Mediouni S, Lima NS, Haregot A, et al. 1186 Resistance to the Tat Inhibitor Didehydro-Cortistatin A Is Mediated by 1187 Heightened Basal HIV-1 Transcription. mBio. 2019;10(4). 1188 92. Jones LE, Perelson AS. Transient viremia, plasma viral load, and reservoir 1189 replenishment in HIV-infected patients on antiretroviral therapy. J Acquir 1190 Immune Defic Syndr. 2007;45(5):483-93. 1191 93. Cherne MD, Hall J, Kellner A, Chong CF, Cole AL, Cole AM. Avirulins, a Novel 1192 Class of HIV-1 Reverse Transcriptase Inhibitors Effective in the Female 1193 Reproductive Tract Mucosa. Viruses. 2019;11(5). 1194

1195

1196

1197

1198

1199

1200

1201

1202

1203

1204

1205

1206

1207

1208

43 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1209 Figures 1210

A. B. Latent HIV-1 Active HIV-1 Methionine PMA gag transcript H2S pathway cycle (RT-qPCR) Methionine Homocysteine H2S levels Cysteine (AzMC dye) Gag Vif Vpu Env Nef CBS LTR LTR H S Pol Vpr 2 + Lanthionine Tat Cystathionine Homocystine Rev CTH C. 600 pathway gag e ** H2S + Homolanthionine Transulfuration Cysteine 400

Mercaptopyruvate fold chang

200 MPST * H S gag 2 0 Cysteine catabolism Pyruvate 6 12 24 36 Time (h) D. U1 E. U1 8 6 h Time (h) : 6 h 12 h 24 h 36 h ) 6 12 h PMA : - + - + - + - + ed

t 24 h 4 43 CTH rea e ns 36 h t 2 1 0.9 1.2 1.1 2.5 1.7 3.6 2.3 un / 0 54 CBS ed t -2 rea t Fold chang - ns 33 MPST

A -4

M ** 1 1.4 1.7 1.5 2.4 2.4 2.5 2.0 P

( -6 GAPDH -8 *** mpst 33 cbs cth (kDa) F. U937 G. U937 8 Time (h) : 6 h 12 h 24 h 36 h *** 6 h ) 6 12 h PMA : - + - + - + - + ed t 4 24 h 43 CTH rea e 36 h t 2 ns 1 1.1 1.3 1.7 2.0 2.4 2.1 4.0 un / CBS 0 54 ed t 1 1.4 1.0 0.5 0.9 0.8 1.1 0.4 -2 ns rea t

Fold chang MPST - 33

A -4 1 1.0 1.4 1.3 1.4 1.9 0.9 1.9 M P

( -6 37 GAPDH -8 cbs cth mpst (kDa)

H. untreated PMA I. 80 ***

U1 60 y t ensi t

in 40 ***

Mean 20 U937

0 1211 U1 U1+PMA U937 U937+PMA

1212 Figure 1. HIV reactivation diminishes expression of H2S metabolic enzymes and

1213 endogenous H2S levels.

1214 (A) Schematic showing H2S producing enzymes in mammalian cells.

1215 (B) Experimental strategy for measuring HIV-1 reactivation and H2S production in U1.

44 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1216 (C) U1 cells were stimulated with 5 ng/ml PMA and the expression of gag transcript

1217 was measured at indicated time points.

1218 (D and E) Time-dependent changes in the expression of CTH, CBS and MPST at

1219 mRNA and protein level during HIV-1 latency (-PMA) and reactivation (+PMA [5

1220 ng/ml]) in U1 cells.

1221 (F and G) Time-dependent changes in the expression of CTH, CBS and MPST at

1222 mRNA and protein level in U937 cells with or without PMA treatment. Results were

1223 quantified by densitometric analysis for CTH, CBS and MPST band intensities and

1224 normalized to GAPDH, using ImageJ software.

1225 (H and I) U1 and U937 cells treated with 5 ng/ml PMA for 24 h or left untreated,

1226 stained with AzMC for 30 min at 37°C, and images were acquired using Leica TCS

1227 SP5 confocal microscope (H). Scale bar represents 40 µm. Average fluorescence

1228 intensity was quantified by ImageJ software (I). Results are expressed as mean ±

1229 standard deviation and are representative of data from three independent experiments.

1230 *, P<0.05; **, P<0.01; ***, P<0.001, by two-way ANOVA with Tukey’s multiple

1231 comparison test.

1232 Figure 1 includes the following figure supplement:

1233 Figure supplement 1. HIV-1 reactivation decreases levels of H2S metabolizing

1234 enzymes in a T cell line model of latency.

1235

1236

1237

1238

1239

1240

45 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1241

A. B. C. D. E. Viral RNA Intracellular p24 Viral release cth 2.0 250 Untreated * 60 14 12 PMA 50 ) U1 shNT shCTH E8 200 e 1.5 s 10 e 40 43 CTH 150 * 8 *** 6 *** 1 0.9 0.1 100 30 1.0 4 20 fold chang

MPST 33 50 -1 p24 (ng/ml 2 Fold chang 0.5 20 ** 4 V 0.6 gag % p24 +ve cell HI GAPDH 10 2 0.3 ** 33 ** 0.0 0 0 0.0 (kDa) 6 12 24 36 12 24 36 shNT shNT shNT Time (h) Time (h) shCTH shCTH shCTHshCTH E8 E9 shNT shCTH shNT+PMA shCTH+PMA F. shNT shCTH shNT shCTH shNT shCTH +PMA +PMA shNT shCTH +PMA +PMA GCLM CD69 CCS GGT1 IRF2 signaling CD44 pathway GSR GSTP1 CDK9 CAT CCNT1 VPS4A GSS Oxidative stress YY1 response COPS6 Factors HTATSF1 involved in HIV infection NUDT1 STAT1 SOD1 GCLC SIRT2 GADD45A Pro- CASP3 apoptotic CD74 MPST MAP3K5 AHCY Sulfur GOT2 FHL2 Anti-apoptotic metabolism TST DHCR24 MAT2A LHPP MSRA MPV17 Mitochondrial UCP2 function -1 0 1 1242 Row Z-Score

1243 Figure 2. Genetic silencing of CTH reactivates HIV-1 and alters gene expression

1244 associated with redox stress, apoptosis, and mitochondrial function.

1245 (A) Total RNA was isolated from shCTH E8, shCTH E9, and non-targeting shRNA

1246 (shNT) lentiviral vectors transduced U1 cells and change in CTH mRNA was examined

1247 by RT-qPCR.

1248 (B) Cell lysates of U1, shNT, and shCTH E8 were assessed for CTH abundance using

1249 immuno-blotting. The CTH band intensities were quantified by densitometric analysis

1250 and normalized to GAPDH.

1251 (C) shCTH and shNT were treated with 5 ng/ml PMA or left untreated for 24 h and

1252 HIV-1 reactivation was determined by gag RT-qPCR.

46 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1253 (D and E) shCTH and shNT were treated with 5 ng/ml PMA or left untreated. At the

1254 indicated time points, HIV-1 reactivation was measured by flow-cytometry using

1255 fluorescently tagged (PE-labeled) antibody specific to intracellular p24 (Gag) antigen

1256 and p24 ELISA in the supernatant. Results are expressed as mean ± standard deviation

1257 and are representative of data from two independent experiments. *, P<0.05; **,

1258 P<0.01; ***, P<0.001, by two-way ANOVA with Tukey’s multiple comparison test.

1259 (F) Total RNA isolated from untreated or PMA (5 ng/ml; 24 h)-treated U1-shNT and

1260 U1-shCTH were examined by NanoString Technology to assess the expression of genes

1261 associated with HIV infection and oxidative stress response. Heatmap showing

1262 functional categories of significantly differential expressed genes (DEGs). Gene

1263 expression data obtained were normalized to internal control β2 microglobulin (B2M),

1264 and fold changes were calculated using the nSolver 4.0 software. Genes with fold

1265 changes values of >1.5 and P <0.05 were considered as significantly altered.

1266 Figure 2 includes the following figure supplement:

1267 Figure supplement 1. Genetic silencing of CTH reduces endogenous H2S levels and

1268 reactivates HIV-1 from latency.

1269 Source data file 1. This file contains RAW values associated with the NanoString

1270 analysis of host genes affected upon depletion of CTH in U1.

1271

1272

1273

1274

1275

1276

1277

47 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1278

A. ns B.. 60 10 * 25 ** 2.0 shNT ** shNT+PMA ) ) shCTH

) 1.8 8 o 20

ein ** ein

t shCTH+PMA 40 t

hione

t 6 1.6

pro 15 pro a f f t o o *

glu GSSG 4 10 1.4 mg 20 mg al / / t o M M RFU (MitoSOX T 2 5 1.2 GSH/GSSG rati ( (

0 0 0 1.0 shNT shCTH shNT shCTH shNT shCTH 0 30 60 240 Time (minutes)

C. Rotenone/ D. Rotenone/ 250 Oligomycin FCCP antimycin A 100 Oligomycin FCCP antimycin A shNT ) ) shCTH 200 80 ein t Spare pro respiratory

60 g protein

150 g

capacity / Maximal / respiration OCR OCR min min

100 / 40

s/ ATP Basal production 50 respiration 20 (pmole Proton leak (pmoles Non-mitochondrial respiration 0 0 0 20 40 60 80 0 20 40 60 80 Time (minutes) Time (minutes) E. 100 shNT

) shCTH ***

ein 80 t

pro ***

g 60

/ *** OCR min 40 * s/ ** *** 20 ns (pmole 0

SRC nmOCR fficiency production P T H+ Proton Aleak Basal Respiration Coupling E 1279 Maximal Respiration

1280 Figure 3. CTH maintains redox homeostasis and mitochondrial bioenergetics to

1281 promote HIV-1 latency.

1282 (A) Total and oxidized cellular glutathione (GSSG) content was assessed in U1-shCTH

1283 and U1-shNT cell lysates using glutathione assay kit.

1284 (B) U1-shNT and U1-shCTH were stained with MitoSOXÔ Red dye for 30 mins at

1285 37°C and analyzed by flow cytometry (Ex-510 nm, Em-580 nm).

1286 (C) Schematic representation of Agilent Seahorse XF Cell Mito Stress test profile to

1287 assess key parameters related to mitochondrial respiration.

1288 (D) U1-shNT and U1-shCTH (5×104) were seeded in triplicate wells of XF microplate

1289 and incubated for 1 h at 37°C in a non-CO2 incubator. Oxygen consumption was

48 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1290 measured without adding any drug (basal respiration), followed by measurement of

1291 OCR change upon sequential addition of 1 μM oligomycin (ATP synthase inhibitor)

1292 and 0.25 μM carbonyl cyanide 4-(trifluoromethoxy) phenylhydrazone (FCCP), which

1293 uncouples mitochondrial respiration and maximizes OCR. Lastly, rotenone (0.5 μM)

1294 and antimycin A (0.5 μM) were injected to completely inhibit respiration by blocking

1295 complex I and complex III, respectively.

1296 (E) Various respiratory parameters derived from OCR measurement were determined

1297 by Wave desktop software. nmOCR; non-mitochondrial oxygen consumption rate and

1298 SRC; spare respiratory capacity. Error bar represent standard deviations from mean.

1299 Results are representative of data from three independent experiments. *, P<0.05; **,

1300 P<0.01; ***,P<0.001; ns, nonsignificant, by two-way ANOVA with Bonferroni’s

1301 multiple comparison test.

1302 Figure 3 includes the following figure supplement:

1303 Figure supplement 1. Effect of CTH knockdown on cytosolic ROS generation.

1304

1305

1306

1307

1308

1309

1310

1311

1312

1313

1314

49 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1315

A. B. ** 500 ** 300 µM NaHS 800 untreated ** 300 µM GYY4137 PMA 400 5 mM GYY4137 600 GYY4137+PMA e NAC+PMA

) 400

M 300

( 200

S ** 2 200 fold chang ** H g **

ga 40 100 20 0 0 0.08 0.25 0.5 1 24 48 72 96 120 24 48 Time (h) Time (h) C. D. ** 8 untreated 200 untreated ** PMA TNF- **

) GYY4137+PMA 6 150 GYY4137+TNF- NAC+PMA e NAC+TNF-

4 100 ** ** -1 p24 (ng/ml fold chang g

V ***

HI 2 40 *** ga 20 *** 0 0 0 24 48 24 48 Time (h) Time (h) E. F. 6 shNT 200 untreated shCTH 5 ** PMA **

) shNT+GYY4137 ** 4 shCTH+GYY4137 e 150 *** 3 ** * * 2 ** *** 100 6 fold chang ns

-1 p24 (ng/ml ns

g ns V 1.0 4 ga ** * HI ** 0.5 ** 2 0.0 0 untreated PMA treated NAC - + ------+ - - - - - GYY4137 - - 5 2.51.2 0.6 0.3 - - 5 2.51.2 0.6 0.3 (mM) G. H. 120 uninfected 4 uninfected 100 HIV-NL4.3 HIV-NL4.3 ) e 3 HIV-NL4.3 80 HIV-NL4.3 + GYY4137 + GYY4137 60 * 2 * fold chang g 40 -1 p24 (ng/ml

V 1 ga

20 HI ** ** 0 0 24 48 72 96 24 48 72 96 1316 Time (h) Time (h)

1317 Figure 4. H2S donor (GYY4137) suppresses HIV-1 reactivation and replication.

1318 (A) U1 cells were treated with NaHS or GYY4137 and media supernatant was

1319 harvested to assess H2S production by methylene blue assay over time.

1320 (B and C) U1 cells were pre-treated with 5 mM GYY4137 or 5 mM NAC for 24 h and

1321 then stimulated with 5 ng/ml PMA for 24 h and 48 h. Total RNA was isolated and HIV-

1322 1 reactivation assessed by gag RT-qPCR (B). Culture supernatant was harvested to

1323 monitor HIV-1 release by p24 ELISA (C).

50 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1324 (D) U1 cells were pretreated with 5 mM GYY417, 5 mM NAC for 24 h or left untreated

1325 and then stimulated with 100 ng/ml TNF-ɑ for 24 h and 48 h. HIV-1 reactivation was

1326 assessed by gag RT-qPCR.

1327 (E) U1-shCTH and U1-shNT were pretreated with 5 mM GYY4137 for 24 h and

1328 stimulated with 5 ng/ml PMA for 24 h. Culture supernatant was harvested to determine

1329 HIV-1 reactivation by HIV-1 p24 ELISA.

1330 (F) J1.1 cells were pretreated with indicated concentrations of GYY4137 or 5 mM NAC

1331 for 24 h and then stimulated with 5 ng/ml PMA for 12 h. Cells were harvested to isolate

1332 total RNA and HIV-1 reactivation was assessed by gag RT-qPCR.

1333 (G and H) Jurkat cells were infected with HIV-NL4.3 at 0.2 MOI for 4 h, cells were

1334 washed, seeded in fresh media, and cultured in presence or absence of 300 μM

1335 GYY4137. HIV-1 replication was monitored by measuring intracellular gag transcript

1336 levels and p24 protein levels in culture supernatant by ELISA. Results are expressed as

1337 mean ± standard deviation and data are representative of three independent

1338 experiments. *, P<0.05; **, P<0.01; ***, P<0.001, by two-way ANOVA with Tukey’s

1339 multiple comparison test.

1340 Figure 4 includes the following figure supplement:

1341 Figure supplement 1. Effect of GYY4137 on HIV-1 reactivation and cellular viability.

1342

1343

1344

1345

1346

1347

1348

51 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1349

A. U1-GYY U1-GYY U1 U1-GYY U1-PMA +PMA U1 U1-GYY U1-PMA +PMA CCS BCL2 GCLC FHL2 GSR CDKN1A Apoptosis HMOX1 LTBR SRXN1 SQSTM1 VIMP CTH SLC7A11 ETHE1 PRNP SQRDL Suflur TXNRD1 GOT1 metabolism MSRA GGT1 DUSP1 TST FTH1 CD74 NOS2 Oxidative- SERPINA1 Innate GPX4 stress TGFB1 immune GCLM response XPO1 response NCF1 APOBEC3G CYBB CBX5 Factors NQO1 PPIA involved in GSTP1 CD4 HIV infection PRDX5 CXCR4 ATOX1 BTRC PRDX3 PTK2B UCP2 FOS CAT NFKBIA GPX1 STAT1 Transcription DHCR24 STAT3 factors PRDX6 IRF1 PRDX1 YY1 -1 0 1 Row Z-Score B. C. U1-GYY U1-GYY U1 U1-GYY U1-PMA +PMA U1 U1-PMA +PMA Nrf2 71 54 (Ser536) 1 2.8 4.1 7.1 1 1.5 1.1 HMOX1 29 54 1 1.7 3.9 7.8 1 1.1 1.2 GAPDH YY1 33 54 (kDa) 1 1.1 1.1 GAPDH 33 (kDa) D. E. F. NF- B ChIP-qPCR YY1 ChIP-qPCR 5!-LTR 10 *** 8 U1 U3 R U5 ** **** t t U1-PMA 8 6 *** Nuc-0 Nuc-1 U1-GYY 6 ** p65p50 YY1 4 +PMA YY1 4 LSF * ns +! ns 2

Fold enrichmen 2 RBEIII ER RBEI Fold enrichmen 0 0 1350 ER GAPDH RBEI RBEIII GAPDH

1351 Figure 5. GYY4137 modulates Nrf2, NF-kB, and YY1 pathways.

1352 (A) U1 cells pre-treated with 5 mM GYY4137 for 24 h or left untreated and then

1353 stimulated with 5 ng/ml PMA for 24 h or left unstimulated. Total RNA was isolated

1354 and expression of genes associated with HIV infection and oxidative stress response

1355 was assessed by nCounter NanoString technology. Heatmap showing functional

1356 categories of significant DEGs in all four conditions: untreated (U1), GYY4137 alone

1357 (U1-GYY) or PMA- alone (U1-PMA) and GYY4137 +PMA (U1-GYY+PMA). Gene

52 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1358 expression data obtained were normalized to internal control β2 microglobulin (B2M),

1359 and fold changes were calculated using the nSolver 4.0 software. Genes with fold

1360 changes values of >1.5 and P <0.05 were considered as significantly altered.

1361 (B) Total cell lysates were used to analyze the expression levels of Nrf2 and HMOX1

1362 by immuno-blotting. Results were quantified by densitometric analysis of Nrf2 and

1363 HMOX1 band intensities and normalized to GAPDH.

1364 (C) U1 cells were pretreated with 5 mM GYY4137 for 6 h and then stimulated with 30

1365 ng/ml PMA for 4 h or left unstimulated. Cells were harvested to prepare total cell lysate.

1366 Levels of phosphorylated NF-κB p65 (Ser536), NF-κB p65, and YY1 were determined

1367 by immuno-blotting. Results were quantified by densitometric analysis for each blot

1368 and were normalized to GAPDH.

1369 (D) Schematic depiction of the binding sites for NF-κB (p65-p50 heterodimer) and YY1

1370 on HIV-1 5´-LTR. Highlighted arrow in red indicates the regions targeted for genomic

1371 qPCR; ER site for NF-κB, and RBEI and RBEIII sites for YY1 enrichments,

1372 respectively.

1373 (E and F) U1 cells were pretreated with 5 mM GYY4137 for 6 h, stimulated with PMA

1374 (30 ng/ml) for 4 h, fixed with formaldehyde, and lysed. Lysates were subjected to

1375 immunoprecipitation for p65 and YY1 and protein-DNA complexes were purified

1376 using protein-G magnetic beads. The enrichment of NF-κB p65 and YY1 on HIV-1

1377 LTR was assessed by qPCR for designated regions using purified DNA as a template.

1378 Results are expressed as mean ± standard deviation and data are representative of three

1379 independent experiments *, P<0.05; **, P<0.01; ***, P<0.001; ****, P<0.0001; ns,

1380 nonsignificant, by two-way ANOVA with Tukey’s multiple comparison test.

1381 Figure 5 includes the following figure supplement:

53 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1382 Source data file 1. This file contains RAW values associated with the NanoString

1383 analysis of host genes affected upon PMA induced HIV reactivation in presence or

1384 absence of GYY4137 treatment.

1385

1386

1387

1388

1389

1390

1391

1392

1393

1394

1395

1396

1397

1398

1399

1400

1401

1402

1403

1404

1405

1406

54 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1407

A. Sample collection CD4+ T cells isolation Ex vivo activation Culture in IL-2 medium Reactivation

H S HIV+ 2 Blood withdrawal PHA, IL-2 1. ART +Prostratin and feeders 2. ART+GYY4137 HIV-1

B. C. Subject A Subject B Subject C Subject D Subject E 103 103 )

A 2 cells 10 2 6 10 0 1 / -1 RN

V 83.5% 1

HI 10 78.9% 1 93.4% 94.2%

(copies 10 Limit of detection ND ND ND ND ND ND ND ND ND 100% 100 7 14 21 28 Pros Pros Pros Pros Pros Pros Pros Pros Pros Pros Days in culture ART ART+ ART ART+ ART ART+ ART ART+ ART ART+ ART ART+GYY4137 GYY4137 GYY4137 GYY4137 GYY4137 GYY4137 D. E. F. G. 4 ns 103 100 100 10 ns ** ** )

80 80 s 3 A

) 10 ) A 2 cells 10 cell DN 6

60 60 6 0

ns 0 2 1 IV

1 10 -1 RN s/ H s/

V 40 40

1 al t HI

10 CD38+ (% opie o 1 CD127+ (% opie c

T 10 c ( 20 20

0 100 0 0 10 Pros Pros 0 7 14 21 28 0 7 14 21 28 ex vivo 14 28 ART ART+ Days Days Days 1408 GYY4137 ART ART+GYY4137 ART ART+GYY4137 ART ART+GYY4137

1409 Figure 6. GYY4137 subverts HIV-1 reactivation in latent CD4+ T cells derived

1410 from HIV-1 infected patients.

1411 (A) Schematic representation of PBMCs extraction from blood samples of ART-treated

1412 HIV-1 infected subjects. CD4+ T cells were sorted and activated with PHA (1 μg/ml),

1413 IL-2 (100 U/ml), feeder PBMCs (gamma-irradiated) from healthy donor. CD4+ T cells

1414 were activated in presence of ART or ART in combination with 100 µM GYY4137.

1415 Post-activation cells were culture with ART alone or ART plus GYY4137 treatment in

1416 IL-2 containing medium.

1417 (B) Total RNA was isolated from five patients CD4+ T cells expanded in ART or ART

1418 plus GYY4137. HIV-1 RNA levels were measured every 7 days by ultrasensitive semi-

1419 nested RT-qPCR with detection limit of three viral RNA copies per million cells.

55 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1420 (C) On day 28, cells were washed of any treatment and both ART or ART+GYY4137

1421 treatment groups were stimulated with 1 μM prostratin for 24 h. HIV-1 RNA copies

1422 were assessed by RT-qPCR. Reduction in viral stimulation in GYY4137 treated

1423 samples are represented as percentage values. ND - non-determined.

1424 (D) Aggregate plot for 5 patients from data shown in C.

1425 (E-F) Primary CD4+ T cells expanded and cultured in presence of ART or

1426 ART+GYY4137 were analyzed overtime for the expression of activation (CD38) and

1427 quiescence markers (CD127) by flow cytometry.

1428 (G) Total HIV-1 DNA content was determined up to 28 days in ART or

1429 ART+GYY4137 treated groups. Results are expressed as mean ± standard deviation.

1430 **, P<0.01; ns, nonsignificant, by two-way ANOVA with Tukey’s multiple comparison

1431 test.

1432 Figure 6 includes the following figure supplement:

1433 Figure supplement 1. Phenotypic features of CD4+ T cells were preserved upon

1434 prolong treatment with GYY4137.

1435

1436

1437

1438

1439

1440

1441

1442

1443

1444

56 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1445

A. Rotenone/ B. Rotenone/ C. Oligomycin FCCP antimycin A Oligomycin FCCP antimycin A Pros Pros+GYY4137 80 150 120 ns ns ) ) ) 100 ein ein ein t t t 60 ***

pro 100 80 pro pro

g

g g ** / / / 40 60 * OCR OCR OCR min min min /

/ ns s/ 50 40 20 20 ns (pmole (pmoles (pmoles 0 0 0 0 20 40 60 80 0 20 40 60 80 P T Time (minutes) Time (minutes) Basal A SRC Maximal nmOCR leak Couplingficiency ART ART+GYY4137 Pros Pros+GYY4137 H+ Proton production f RespirationRespiration E D. E. F. 500 1000 20 ** ** * 15 ns ART 400 10 800 ** ART+GYY4137 e 4 * ** SOX) SOX) * ART+Pros

o 300 o 600 t t 3 ns Mi Mi ns ART+GYY4137+Pros ( 200 ( 400 U U 2 F F Fold chang R R

100 200 1

0 0 0 hmox1 txnrd1 gclc prdx1 Pros

GYY4137+Pros PMA/IonoGYY4137 1446 unstimulated unstimulated +PMA/Iono

1447 Figure 7. Effect of H2S on mitochondrial respiration and ROS generation in latent

1448 CD4+ T cells derived from HIV-1 patients.

1449 (A) Primary human CD4+ T cells from HIV infected subjects were activated and

1450 cultured ex vivo with ART or ART+GYY4137. On day 28, cells from ART or

1451 ART+GYY4137 treatment groups were harvested to assess mitochondrial respiration

1452 by using Seahorse XF mito-stress test as described in materials and methods.

1453 (B) Cells from ART and ART+GYY4137 treated groups were stimulated with 1 µM

1454 prostratin 6 h. Post-stimulation mitochondrial respiration profile was determined by

1455 Seahorse XF mito-stress test.

1456 (C) Various mitochondrial respiratory parameters derived from OCR measurement

1457 were determined by Wave desktop software. nmOCR; non-mitochondrial oxygen

1458 consumption rate and SRC; spare respiratory capacity.

1459 (D) On day 28, Cells from both ART and ART+GYY4137 treatment groups were

1460 stimulated with 1 μM Prostratin for 6 h. Cells were harvested and stained with 5 μM

57 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1461 MitoSOX-Red dye for 30 min followed by washing. Samples were analyzed by flow

1462 cytometry. Unstimulated- cells cultured under ART alone.

1463 (E) Both ART and ART+GYY4137 treated cells at day 14 post-activation were

1464 stimulated with 1 μg/ml PMA and 100 μg/ml ionomycin (Iono) for 6 h. Cells were

1465 harvested post-stimulation and stained with 5 μM MitoSOX-Red dye. Samples were

1466 analyzed by flow cytometry to assess mitoROS generation. Unstimulated- cells

1467 cultured under ART alone.

1468 (F) CD4+ T cells from ART and ART+GYY4137 treated groups were stimulated with

1469 1 µM prostratin on 28th day for 24 h or left unstimulated. Cells were harvested to isolate

1470 total RNA and expression of hmox1, txnrd1, gclc and prdx1 were determined by RT-

1471 qPCR. Data obtained were normalized to internal control β2 microglobulin (B2M).

1472 Error bar represent standard deviations from mean. Results are representative of data

1473 from three patients samples. *, P<0.05; **, P<0.01, by two-way ANOVA with Tukey’s

1474 multiple comparison test.

1475

1476

1477

1478

1479

1480

1481

1482

1483

1484

1485

58 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1 Supplementary Materials for 2 Hydrogen sulfide blocks HIV rebound by maintaining mitochondrial bioenergetics and 3 redox homeostasis 4 5 Virender Kumar Pal1,2, Ragini Agrawal1,2, Srabanti Rakshit2, Pooja Shekar3, Diwakar Tumkur 6 Narasimha Murthy4, Annapurna Vyakarnam2, and Amit Singh1,2. 7 8 The file includes:

9 Figure 1-figure supplement 1. HIV-1 reactivation decreases levels of H2S metabolizing

10 enzymes in a T cell line model of latency.

11 Figure 2-figure supplement 1. Genetic silencing of CTH reduces endogenous H2S levels and

12 reactivates HIV-1 from latency.

13 Figure 3- figure supplement 1. Effect of CTH knockdown on cytosolic ROS generation.

14 Figure 4-figure supplement 1. Effect of GYY4137 on HIV-1 reactivation and cellular

15 viability.

16 Figure 6-figure Supplement 1.. Phenotypic features of CD4+ T cells were preserved upon

17 prolong treatment with GYY4137.

18 Table S1. Sequences of shRNA clones.

19 Table S2. List of genes used for NanoString nCounter Gene Expression Analysis

20 Table S3. List of differentially expressed genes from NanoString nCounter Gene Expression

21 analysis for U1-shNT, U1-shCTH, U1-shNT+PMA and U1-shCTH+PMA samples.

22 Table S4. List of differentially expressed genes from NanoString nCounter Gene Expression

23 analysis for U1 untreated, GYY4137, PMA and GYY4137+PMA treated samples.

24 Table S5. Characteristic of ART treated HIV-1 infected study participants

25 Table S6. List of primers and probes used in this study.

26 27 28 29 30

1 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

31

A. J1.1 B. Jurkat Time (h): 6 h 12 h 24 h 36 h Time (h): 6 h 12 h 24 h 36 h PMA: - + - + - + - + PMA: - + - + - + - + 43 CTH 43 CTH 1 1 1 0.9 0.9 1.0 1.2 1.2 1 0.9 1.0 0.8 1.3 0.8 1.2 0.8 CBS CBS 54 54 1 0.9 1 0.8 1 0.6 1 0.5 1 1.1 1.2 1.1 1.3 0.9 1.0 0.7 33 MPST 33 MPST 71 GAPDH p55 33 (kDa) 43 p41 HIV-1 p24

16 p24 GAPDH 33 32 (kDa)

33 Figure 1-figure supplement 1. HIV-1 reactivation decreases levels of H2S metabolizing

34 enzymes in a T cell line model of latency.

35 (A-B) J1.1 and Jurkat cells were stimulated with 5 ng/ml PMA for 6 h, 12 h, 24 h, and 36 h and

36 total cell lysates were prepared to analyze CTH, CBS, and MPST protein levels. HIV-1

37 reactivation was assessed by immunoblotting for intracellular viral protein, p24. The results

38 are representative of data from three independent experiments. Results were quantified by

39 densitometric analysis for CTH, CBS, and MPST band intensities and normalized to GAPDH.

40

41

42

43

44

45

46

47

48

2 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

49

A. B. C. 40 J1.1 shNT shCBS shCTH *** CBS shNT shCTH 54 1 0.8 0.01 0.7 30 43 CTH y t i s 1 0.8 0.6 0.1 en t

in 20 MPST 33 1 1.4 1.7 1.8

Mean HIV-1 p24 10 16 1 1.6 3.8 7.0 GAPDH 0 33 50 shNT shCTH (kDa)

51 Figure 2-figure supplement 1. Genetic silencing of CTH reduces endogenous H2S levels

52 and reactivates HIV-1 from latency.

53 (A and B) U1 cells stably transduced with non-targeting (shNT) and shCTH lentiviral vectors

54 were stained with AzMC for 30 min at 37°C, and images were acquired using ZEISS LSM 880

55 confocal microscope (A). Scale bar represents 20 µm. Average fluorescence intensity was

56 quantified by ImageJ software (B). Results are expressed as mean ± standard deviation and are

57 representative of data from two independent experiments. ***, P<0.001, by two-tailed

58 student’s t-test.

59 (C) Total cell lysates were prepared from un-transduced (J1.1), non-targeting (shNT), shCBS,

60 and shCTH knockdown J1.1 cells. Expression of CBS, CTH, MPST and HIV-1 p24 levels was

61 assessed by immuno-blotting. Results are expressed as mean ± standard deviation and are

62 representative of data from two independent experiments.

63

64

65

66

67

68

69

3 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

70

A. 1.2 ns

1.0 ) 0.8

0.6

0.4 RFU (DCFDA 0.2

0.0

shNT shNT shCTH shCTH 71 +PMA +PMA

72 Figure 3- figure supplement 1. Effect of CTH knockdown on cytosolic ROS generation.

73 (A) U1-shNT and U1- shCTH cells were left untreated or treated with 5 ng/ml PMA for 6 h,

74 stained with CM-H2DCFDA dye (5 μM) analyzed using flow cytometry. Results are expressed

75 as mean ± standard deviation and data are representative of two independent experiment. ns,

76 nonsignificant, by two-way ANOVA with Tukey’s multiple comparison test.

77

78

79

80

81

82

83

84

85

86

87

88

89

90

4 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

91

A. B. 120 untreated 60 ns 100 PMA ns ns GYY4137+PMA s 80 e NAC+PMA 40 60

40 fold chang % viable cell g 20

20 ga

0 24 48 0 Time (h) UT PMA PMA+ Spent GYY4137 C. D. 100 untreated 120 PMA ns 80 100 n s 80 60 * * 60 40 40

% viable cell ** % HIV activatio 20 20

0 0 NAC - + ------+ - - - - - Prostratin: - + + + + + GYY4137 - - 5 2.51.2 0.6 0.3 - - 5 2.51.2 0.6 0.3 Treatments: - - NAC 1.25 2.5 5 (mM) GYY4137 (mM) E. ns 120 ns **** **** 24 h 100 **** 48 h

s 80 72 h 96 h 60

40 % viable cell 20

0 0 0.3 0.6 1.2 2.5 5 GYY4137 (mM) 92

93 Figure 4-figure supplement 1. Effect of GYY4137 on HIV-1 reactivation and cellular

94 viability.

95 (A) U1 cells were pretreated with 5 mM GYY4137 or 5 mM NAC for 24 and then stimulated

96 with 5 ng/ml PMA for 24 h or 48 h. Cells were stained with 3 µM propidium iodide (PI) for 15

97 min in dark, washed, and analyzed using flow cytometry.

98 (B) Decomposed GYY4137 (spent GYY4137) which was aerated for at least 180 days at left

99 at room temperature was used to pretreat U1 cells. Cells pretreated with 5 mM spent GYY4137

100 for 24 h or left untreated were stimulated with 5 ng/ml PMA for 24 h and HIV-1 reactivation

101 was assessed by gag RT-qPCR.

5 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

102 (C) J1.1 cells were pretreated with GYY4137 for 24 h and then stimulated with PMA for 12 h.

103 Cells were then harvested, stained with PI, and subjected to flow cytometry to assess viability.

104 (D) J-Lat cells were pretreated with 5 mM NAC or indicated concentrations of GYY4137 for

105 24 h and then stimulated with 2.5 µM prostratin for 24 h. HIV-1 reactivation was determined

106 by estimating GFP expressing cells using flow cytometry.

107 (E) Jurkat cells were treated with 5 mM, 2.5 mM, 1.2 mM, 0.6 mM and 0.3 mM of GYY4137

108 for 24 h, 48 h, 72 h, and 96 h. Samples were harvested and stained with PI to determine cells

109 viability by flow cytometry. Error bar represent standard deviations from mean. Results are

110 representative of data from two independent experiments. *, P<0.05; **,P<0.01; ***,P<0.001;

111 ****, P<0.0001; ns, nonsignificant, by two-way ANOVA with Tukey’s multiple comparison

112 test.

113

114

115

116

117

118

119

120

121

122

123

124

125

126

6 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

127

A. 250K 250K 105 105 200K 200K 104 104 150K 150K

3 SSC-A FSC-H FSC-A 100K 100K 103 SSC-A 10

50K 50K 0 0 -103 -103

0 50K 100K 150K 200K 250K -103 0 103 104 105 0 50K 100K 150K 200K 250K -103 0 103 104 105 FSC-A AviD FSC-A CD4

105 105 105 104 104 104 4 3 10 TCM 10

103 3 10 CD38 CCR7 103 CD127

CD45RA 0 0 0 0 T TEM TM -103 -103 -103 -103

0 103 104 105 -103 0 103 104 105 0 103 104 105 0 103 104 105 CD27 CD27 CD45RA CD45RA

B. ART ART+GYY4137 C. ART ART+GYY4137 D. ART ART+GYY4137 100 100 100

80 80 ) 80 ) ) 60 60 60 cells (% 40 cells (% cells (% 40 40 M M N C T T Frequency of T T Frequency of Frequency of 20 20 20 0 0 0 0 7 14 21 28 0 7 14 21 28 0 7 14 21 28 Days Days Days E. ART ART+GYY4137 F. ART ART+GYY4137 100 120 100

80 ) ) 80 (% 60 y t 60

cells (% 40 iabili M 40 E V T Frequency of 20 20 0 0 0 7 14 21 28 0 7 14 21 28 128 Days Days

129 Figure 6-figure Supplement 1. Phenotypic features of CD4+ T cells were preserved upon

130 prolong treatment with GYY4137.

131 (A) Flow cytometry-gating strategy used to analyze the expression of activation (CD38),

132 quiescence (CD127), and frequency of different memory subsets of CD4+ T cells. The figure

133 represents expression of different markers at Day 7 post activation of CD4+ T cells from a

134 single patient sample.

135 (B-E) Primary human CD4+ T cells from HIV infected patients cultured with ART or ART

136 with GYY4137 were analyzed by flow cytometry overtime to determine frequency of different

7 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

+ 137 subsets of CD4 T cells- TN (naive [B]) and TCM (central memory [C]), TTM (transition memory

138 [D]), TEM (effector memory [E]).

139 (F) Cell viability was assessed by Live-Dead staining overtime. Results obtained suggest no

140 difference in viability between ART alone and ART with GYY4137 treated group overtime.

141 Error bar represent standard deviations from mean. Results are representative of data from four

142 patients samples.

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

8 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

160 Table S1. Sequences of shRNA clones. 161

Gene Accession number shRNA cat. No. Sequence code name CBS NM_000071.2 NM_000071.1- CCGGCTTGCCAGATATTCTGAAGAACTCGAGTTCTTCAGAATATCTGGCA C9 409s1c1 AGTTTTTG NM_000071.1- CCGGCCGTCAGACCAAGTTGGCAAACTCGAGTTTGCCAACTTGGTCTGAC C10 1604s1c1 GGTTTTTG CTH NM_001902.4,NM_1537 NM_001902.4- CCGGGCACCTCATTATCTTTCATAACTCGAGTTATGAAAGATAATGAGGT E8 42.3 1472s1c1 GCTTTTTG NM_001902.4- CCGGCCTTCATAATAGACTTCGTTTCTCGAGAAACGAAGTCTATTATGAA E9 837s1c1 GGTTTTTG 162

163

164

165

166

167

168

169

170

171

9 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

172 Table S2. List of genes used for NanoString nCounter Gene Expression Analysis.

Sl. Gene Accession Position Target Sequence

no. name

1 ABCF1 NM_001090.2 851-950 GATGTCCTCCCGCCAAGCCATGTTAGAAAATGCATCTGACATCAAGCTGGA

GAAGTTCAGCATCTCCGCTCATGGCAAGGAGCTGTTCGTCAATGCAGAC

2 ACTB NM_001101.2 1011-1110 TGCAGAAGGAGATCACTGCCCTGGCACCCAGCACAATGAAGATCAAGATCA

TTGCTCCTCCTGAGCGCAAGTACTCCGTGTGGATCGGCGGCTCCATCCT

3 AHCY NM_001161766.1 1147-1246 ATCCTTGGCCGGCACTTTGAGCAGATGAAGGATGATGCCATTGTGTGTAAC

ATTGGACACTTTGACGTGGAGATCGATGTCAAGTGGCTCAACGAGAACG

4 AKR1C2 NM_001135241.2 256-355 CTGCAGAGGTTCCTAAAAGTAAAGCTCTAGAGGCCGTCAAATTGGCAATAG

AAGCCGGGTTCCACCATATTGATTCTGCACATGTTTACAATAATGAGGA

5 ALB NM_000477.5 856-955 TCTTACCAAAGTCCACACGGAATGCTGCCATGGAGATCTGCTTGAATGTGCT

GATGACAGGGCGGACCTTGCCAAGTATATCTGTGAAAATCAAGATTCG

6 ALOX12 NM_000697.1 1946-2045 TTACAGCCCGGAATGAGCAACTTGACTGGCCCTATGAATATCTGAAGCCCA

GCTGCATAGAGAACAGTGTCACCATCTGAGCCCTAGAGTGACTCTACCT

10 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

7 AOX1 NM_001159.3 1416-1515 GCGATAGTCAATTCAGGAATGAGAGTCTTTTTTGGAGAAGGGGATGGCATT

ATTAGAGAGTTATGCATCTCATATGGAGGCGTTGGTCCAGCCACCATCT

8 APOBEC NM_021822.2 762-861 GAAAAGAGACGGTCCGCGTGCCACCATGAAGATCATGAATTATGACGAATT

3G TCAGCACTGTTGGAGCAAGTTCGTGTACAGCCAAAGAGAGCTATTTGAG

9 APOE NM_000041.2 97-196 GGGCTGCGTTGCTGGTCACATTCCTGGCAGGATGCCAGGCCAAGGTGGAGC

AAGCGGTGGAGACAGAGCCGGAGCCCGAGCTGCGCCAGCAGACCGAGTG

10 ATOX1 NM_004045.3 144-243 GCTGAAGCTGTCTCTCGGGTCCTCAATAAGCTTGGAGGAGTTAAGTATGAC

ATTGACCTGCCCAACAAGAAGGTCTGCATTGAATCTGAGCACAGCATGG

11 B2M NM_004048.2 26-125 CGGGCATTCCTGAAGCTGACAGCATTCGGGCCGAGATGTCTCGCTCCGTGG

CCTTAGCTGTGCTCGCGCTACTCTCTCTTTCTGGCCTGGAGGCTATCCA

12 BAD NM_004322.3 964-1063 AGGGAGGGCTGACCCAGATTCCCTTCCGGTGCGTGTGAAGCCACGGAAGGC

TTGGTCCCATCGGAAGTTTTGGGTTTTCCGCCCACAGCCGCCGGAAGTG

13 BAG2 NM_004282.3 786-885 AAGAGTCATTTAATGTCGCTCTACAGTGCATGTTCATCTGAGGTGCCACATG

GGCCAGTTGATCAGAAGTTTCAATCCATAGTAATTGGCTGTGCTCTTG

14 BANF1 NM_001143985.1 553-652 GGAGGAAAGGGGTTTTGACAAGGCCTATGTTGTCCTTGGCCAGTTTCTGGTG

CTAAAGAAAGATGAAGACCTCTTCCGGGAATGGCTGAAAGACACTTGT

11 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

15 BAX NM_138761.3 343-442 TTTTTCCGAGTGGCAGCTGACATGTTTTCTGACGGCAACTTCAACTGGGGCC

GGGTTGTCGCCCTTTTCTACTTTGCCAGCAAACTGGTGCTCAAGGCCC

16 BCL2 NM_000657.2 6-105 GTGAAGCAGAAGTCTGGGAATCGATCTGGAAATCCTCCTAATTTTTACTCCC

TCTCCCCGCGACTCCTGATTCATTGGGAAGTTTCAAATCAGCTATAAC

17 BHMT NM_001713.2 41-140 CGCAGCGGGAAGGCTCGCCTAGTCGGTCCGCATCCGTGTCGACCACCTGTC

TGGACACCACGAAGATGCCACCCGTTGGGGGCAAAAAGGCCAAGAAGGG

18 BNIP3 NM_004052.2 585-684 AGGGGGCATATTCTCTGCAGAATTTCTGAAAGTTTTCCTTCCATCTCTGCTG

CTCTCTCATTTGCTGGCCATCGGATTGGGGATCTATATTGGAAGGCGT

19 BTRC NM_033637.2 1912-2011 ACACCTACATCTCCAGATAAATAACCATACACTGACCTCATACTTGCCCAGG

ACCCATTAAAGTTGCGGTATTTAACGTATCTGCCAATACCAGGATGAG

20 CASP3 NM_032991.2 686-785 ACTCCACAGCACCTGGTTATTATTCTTGGCGAAATTCAAAGGATGGCTCCTG

GTTCATCCAGTCGCTTTGTGCCATGCTGAAACAGTATGCCGACAAGCT

21 CASP8 NM_001228.4 302-401 AGATGGACTTCAGCAGAAATCTTTATGATATTGGGGAACAACTGGACAGTG

AAGATCTGGCCTCCCTCAAGTTCCTGAGCCTGGACTACATTCCGCAAAG

22 CAT NM_001752.2 1131-1230 ATGCTTCAGGGCCGCCTTTTTGCCTATCCTGACACTCACCGCCATCGCCTGG

GACCCAATTATCTTCATATACCTGTGAACTGTCCCTACCGTGCTCGAG

12 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

23 CBS NM_000071.2 589-688 GCGGGAGCGTGAAGGACCGCATCAGCCTGCGGATGATTGAGGATGCTGAGC

GCGACGGGACGCTGAAGCCCGGGGACACGATTATCGAGCCGACATCCGG

24 CBX5 NM_012117.1 636-735 TATGAAGAGAGACTGACATGGCATGCATATCCTGAGGATGCGGAAAACAA

AGAGAAAGAAACAGCAAAGAGCTAAAGGAGGGGATGGTCTCTGTCATTTC

25 CCL2 NM_002982.3 1-100 GAGGAACCGAGAGGCTGAGACTAACCCAGAAACATCCAATTCTCAAACTGA

AGCTCGCACTCTCGCCTCCAGCATGAAAGTCTCTGCCGCCCTTCTGTGC

26 CCL3 NM_002983.2 682-781 CTGTGTAGGCAGTCATGGCACCAAAGCCACCAGACTGACAAATGTGTATCG

GATGCTTTTGTTCAGGGCTGTGATCGGCCTGGGGAAATAATAAAGATGC

27 CCL4 NM_002984.2 202-301 GAAGCTTCCTCGCAACTTTGTGGTAGATTACTATGAGACCAGCAGCCTCTGC

TCCCAGCCAGCTGTGGTATTCCAAACCAAAAGAAGCAAGCAAGTCTGT

28 CCL5 NM_002985.2 281-380 AGTGTGTGCCAACCCAGAGAAGAAATGGGTTCGGGAGTACATCAACTCTTT

GGAGATGAGCTAGGATGGAGAGTCCTTGAACCTGAACTTACACAAATTT

29 CCL8 NM_005623.2 690-789 AAGGAGAGATGGGTCAGGGATTCCATGAAGCATCTGGACCAAATATTTCAA

AATCTGAAGCCATGAGCCTTCATACATGGACTGAGAGTCAGAGCTTGAA

30 CCNT1 NM_001240.2 1850-1949 GCACAAGACTCACCCATCTAATCATCATCATCATCATAATCACCACTCACAC

AAGCACTCTCATTCCCAACTTCCAGTTGGTACTGGGAACAAACGTCCT

13 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

31 CCR2 NM_001123041.2 744-843 TCTGATCTGCTTTTTCTTATTACTCTCCCATTGTGGGCTCACTCTGCTGCAAA

TGAGTGGGTCTTTGGGAATGCAATGTGCAAATTATTCACAGGGCTGT

32 CCR3 NM_001837.2 981-1080 CAGTGCTCTTTACCCAGAGGATACAGTATATAGCTGGAGGCATTTCCACACT

CTGAGAATGACCATCTTCTGTCTCGTTCTCCCTCTGCTCGTTATGGCC

33 CCR4 NM_005508.4 673-772 TCCCTTCCTGGCTTTCTGTTCAGCACTTGTTATACTGAGCGCAACCATACCTA

CTGCAAAACCAAGTACTCTCTCAACTCCACGACGTGGAAGGTTCTCA

34 CCR5 NM_000579.1 2731-2830 TAGGAACATACTTCAGCTCACACATGAGATCTAGGTGAGGATTGATTACCT

AGTAGTCATTTCATGGGTTGTTGGGAGGATTCTATGAGGCAACCACAGG

35 CCS NM_005125.1 35-134 GGGTCCAGAATGGCTTCGGATTCGGGGAACCAGGGGACCCTCTGCACGTTG

GAGTTCGCGGTGCAGATGACCTGTCAGAGCTGTGTGGACGCGGTGCGCA

36 CD209 NM_001144899.1 951-1050 TAGAGCTTGTTTTTCTGGCCCATCCTTGGAGCTTTATGAGTGAGCTGGTGTG

GGATGCCTTTGGGGGTGGACTTGTGTTCCAAGAATCCACTCTCTCTTC

37 CD247 NM_198053.1 1491-1590 TGGCAGGACAGGAAAAACCCGTCAATGTACTAGGATACTGCTGCGTCATTA

CAGGGCACAGGCCATGGATGGAAAACGCTCTCTGCTCTGCTTTTTTTCT

38 CD4 NM_000616.4 976-1075 TGGCAGGCGGAGAGGGCTTCCTCCTCCAAGTCTTGGATCACCTTTGACCTGA

AGAACAAGGAAGTGTCTGTAAAACGGGTTACCCAGGACCCTAAGCTCC

14 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

39 CD44 NM_001001392.1 430-529 ACACCATGGACAAGTTTTGGTGGCACGCAGCCTGGGGACTCTGCCTCGTGC

CGCTGAGCCTGGCGCAGATCGATTTGAATATAACCTGCCGCTTTGCAGG

40 CD69 NM_001781.1 461-560 AGGACATGAACTTTCTAAAACGATACGCAGGTAGAGAGGAACACTGGGTTG

GACTGAAAAAGGAACCTGGTCACCCATGGAAGTGGTCAAATGGCAAAGA

41 CD74 NM_001025159.1 965-1064 TTCAGCCCCCAGCCCCTCCCCCATCTCCCACCCTGTACCTCATCCCATGAGA

CCCTGGTGCCTGGCTCTTTCGTCACCCTTGGACAAGACAAACCAAGTC

42 CDK7 NM_001799.2 786-885 CTGAGGAACAGTGGCCGGACATGTGTAGTCTTCCAGATTATGTGACATTTA

AGAGTTTCCCTGGAATACCTTTGCATCACATCTTCAGTGCAGCAGGAGA

43 CDK9 NM_001261.2 401-500 GGTGTTCGACTTCTGCGAGCATGACCTTGCTGGGCTGTTGAGCAATGTTTTG

GTCAAGTTCACGCTGTCTGAGATCAAGAGGGTGATGCAGATGCTGCTT

44 CDKN1A NM_000389.2 1976-2075 CATGTGTCCTGGTTCCCGTTTCTCCACCTAGACTGTAAACCTCTCGAGGGCA

GGGACCACACCCTGTACTGTTCTGTGTCTTTCACAGCTCCTCCCACAA

45 CDO1 NM_001801.2 1423-1522 TGAATAATAGAGGAACTTACGGCACTCTGCCATTTGTTAATGAAAGGAAGT

GCAGAGGATTTAGAAAAGTACATGATCCCCAGACCACAACAAACCAAAA

46 CEBPB NM_005194.2 1421-1520 CAACCGCACATGCAGATGGGGCTCCCGCCCGTGGTGTTATTTAAAGAAGAA

ACGTCTATGTGTACAGATGAATGATAAACTCTCTGCTTCTCCCTCTGCC

15 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

47 COPS6 NM_006833.4 861-960 CAGATTTTTATGATCAATGCAACGACGTGGGGCTCATGGCCTACCTCGGCAC

CATCACCAAAACGTGCAACACCATGAACCAGTTTGTGAACAAGTTCAA

48 CR2 NM_001006658.1 486-585 GGTGTCAAGCAAATAATATGTGGGGGCCGACACGACTACCAACCTGTGTAA

GTGTTTTCCCTCTCGAGTGTCCAGCACTTCCTATGATCCACAATGGACA

49 CSAD NM_001244706.1 825-924 CTGCACTGACCTGTGCTGATATGGACTTCCTCCTCAACGAGCTGGAGCGGCT

AGGCCAGGACCTGTGAGCCTTCTCTGTCTTGCTGCCGGCCTTGATACC

50 CTH NM_001190463.1 861-960 CCTCTGCAATCGAGGTCTGAAGACTCTACATGTCCGAATGGAAAAGCATTT

CAAAAACGGAATGGCAGTTGCCCAGTTCCTGGAATCTAATCCTTGGGTA

51 CX3CL1 NM_002996.3 1217-1316 CCCCGGAGCTGTGGTAGTAATTCATATGTCCTGGTGCCCGTGTGAACTCCTC

TGGCCTGTGTCTAGTTGTTTGATTCAGACAGCTGCCTGGGATCCCTCA

52 CXCL12 NM_199168.3 70-169 CCGCCCGCCCGCCCGCCCGCGCCATGAACGCCAAGGTCGTGGTCGTGCTGG

TCCTCGTGCTGACCGCGCTCTGCCTCAGCGACGGGAAGCCCGTCAGCCT

53 CXCL8 NM_000584.2 26-125 ACAGCAGAGCACACAAGCTTCTAGGACAAGAGCCAGGAAGAAACCACCGG

AAGGAACCATCTCACTGTGTGTAAACATGACTTCCAAGCTGGCCGTGGCT

54 CXCR4 NM_003467.2 1336-1435 ATTGATGTGTGTCTAGGCAGGACCTGTGGCCAAGTTCTTAGTTGCTGTATGT

CTCGTGGTAGGACTGTAGAAAAGGGAACTGAACATTCCAGAGCGTGTA

16 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

55 CYBB NM_000397.3 2687-2786 TTTGAAGCATGAAAAAAGAGGGTTGGAGGTGGAGAATTAACCTCCTGCCAT

GACTCTGGCTCATCTAGTCCTGCTCCTTGTGCTATAAAATAAATGCAGA

56 CYGB NM_134268.4 709-808 AAGCACAAGGTGGAACCGGTGTACTTCAAGATCCTCTCTGGGGTCATTCTG

GAGGTGGTCGCCGAGGAATTTGCCAGTGACTTCCCACCTGAGACGCAGA

57 DHCR24 NM_014762.2 976-1075 CAGAGCCCAGCAAGCTGAATAGCATTGGCAATTACTACAAGCCGTGGTTCT

TTAAGCATGTGGAGAACTATCTGAAGACAAACCGAGAGGGCCTGGAGTA

58 DUOX1 NM_175940.1 4856-4955 TGAGCCCACACCTCACCTCTGTTCTTCCTATTTCTGGCTGCCTCAGCCTTCTC

TGATTTCCCACCTCCCAACCTTGTTCCAGGTGGCCATAGTCAGTCAC

59 DUOX2 NM_014080.4 3321-3420 AGGCACATCGTGTGTGTGGCAATCTTCTCGGCCATCTGTGTTGGCGTGTTTG

CAGATCGTGCTTACTACTATGGCTTTGCCTCGCCACCCTCGGACATTG

60 DUSP1 NM_004417.2 988-1087 TCAAGAATGCTGGAGGAAGGGTGTTTGTCCACTGCCAGGCAGGCATTTCCC

GGTCAGCCACCATCTGCCTTGCTTACCTTATGAGGACTAATCGAGTCAA

61 ELANE NM_001972.2 196-295 ACTTCTGCGGCGCCACCCTGATTGCGCCCAACTTCGTCATGTCGGCCGCGCA

CTGCGTGGCGAATGTAAACGTCCGCGCGGTGCGGGTGGTCCTGGGAGC

62 EP300 NM_001429.2 716-815 CCAGCCAGGCCCAACAGAGCAGTCCTGGATTAGGTTTGATAAATAGCATGG

TCAAAAGCCCAATGACACAGGCAGGCTTGACTTCTCCCAACATGGGGAT

17 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

63 EPHX2 NM_001979.5 421-520 AGGAAGTGCTCCGAGACCGCTAAAGTCTGCCTCCCCAAGAATTTCTCCATA

AAAGAAATCTTTGACAAGGCGATTTCAGCCAGAAAGATCAACCGCCCCA

64 EPX NM_000502.4 2061-2160 GACAGGTTCTGGTGGCAGAAACGAGGTGTTTTCACCAAAAGACAGCGCAAG

GCCCTGAGCAGAATTTCCTTGTCTCGAATTATATGTGACAATACCGGTA

65 ETHE1 NM_014297.3 691-790 GAGGACTCTGAACCCTCGGCTCACCCTCAGCTGTGAGGAGTTTGTCAAAAT

CATGGGCAACCTGAACTTGCCTAAACCTCAGCAGATAGACTTTGCTGTT

66 FCAR NM_002000.2 1416-1515 TGCTGAGATTATAGGCATGAGCCACCACGCCTGGCCAGATGCATGTTCAAA

CCAATCAAATGGTGTTTTCTTATGCAGGACTGATCGATTTGCACCCACC

67 FHL2 NM_001039492.2 249-348 CATTGCAACGAATCTCTCTTTGGCAAGAAGTACATCCTGCGGGAGGAGAGC

CCCTACTGCGTGGTGTGCTTTGAGACCCTGTTCGCCAACACCTGCGAGG

68 FOS NM_005252.2 1476-1575 ACTCAAGTCCTTACCTCTTCCGGAGATGTAGCAAAACGCATGGAGTGTGTAT

TGTTCCCAGTGACACTTCAGAGAGCTGGTAGTTAGTAGCATGTTGAGC

69 FOXM1 NM_202002.1 1001-1100 CAATTCGCCATCAACAGCACTGAGAGGAAGCGCATGACTTTGAAAGACATC

TATACGTGGATTGAGGACCACTTTCCCTACTTTAAGCACATTGCCAAGC

70 FTH1 NM_002032.2 393-492 CCAAATACTTTCTTCACCAATCTCATGAGGAGAGGGAACATGCTGAGAAAC

TGATGAAGCTGCAGAACCAACGAGGTGGCCGAATCTTCCTTCAGGATAT

18 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

71 GADD45 NM_001924.2 866-965 GTTACTCCCTACACTGATGCAAGGATTACAGAAACTGATGCCAAGGGGCTG

A AGTGAGTTCAACTACATGTTCTGGGGGCCCGGAGATAGATGACTTTGCA

72 GCLC NM_001498.2 521-620 CTCAAGTGGGGCGATGAGGTGGAATACATGTTGGTATCTTTTGATCATGAA

AATAAAAAAGTCCGGTTGGTCCTGTCTGGGGAGAAAGTTCTTGAAACTC

73 GCLM NM_002061.2 600-699 TCATCATCAACTAGAAGTGCAGTTGACATGGCCTGTTCAGTCCTTGGAGTTG

CACAGCTGGATTCTGTGATCATTGCTTCACCTCCTATTGAAGATGGAG

74 GGT1 NM_001032364.2 713-812 CATGAATGCCCACAGCATGGGCATCGGGGGTGGCCTCTTCCTCACCATCTAC

AACAGCACCACACGAAAAGCTGAGGTCATCAACGCCCGCGAGGTGGCC

75 GLA NM_000169.2 981-1080 GCTGCTCCTTTATTCATGTCTAATGACCTCCGACACATCAGCCCTCAAGCCA

AAGCTCTCCTTCAGGATAAGGACGTAATTGCCATCAATCAGGACCCCT

76 GNMT NM_018960.4 716-815 GCTTGAGTAAGTTCCGGCTCTCCTACTACCCACACTGTCTGGCATCCTTCAC

GGAGCTGCTCCAAGCAGCCTTCGGAGGTAAGTGCCAGCACAGCGTCCT

77 GOT1 NM_002079.2 616-715 GTCCTCACCAACCTGGGAGAATCACAATGCTGTGTTTTCCGCTGCTGGTTTT

AAAGACATTCGGTCCTATCGCTACTGGGATGCAGAGAAGAGAGGATTG

78 GOT2 NM_002080.2 2146-2245 GGAGAGTAGGAAACTGTACTTTATCTCGGCATCCTCTTGAATGATAGTGCA

AGTTTCTCCAGTTGGGATGTTGTCTCTGCCCGGTTGGACCTCCTCCCTT

19 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

79 GPX1 NM_000581.2 746-845 GGTTTTCATCTATGAGGGTGTTTCCTCTAAACCTACGAGGGAGGAACACCTG

ATCTTACAGAAAATACCACCTCGAGATGGGTGCTGGTCCTGTTGATCC

80 GPX2 NM_002083.2 831-930 CCTCTGGTTGGTGATTCAACTTGGGCTCCAAGACTTGGGTAAGCTCTGGGCC

TTCACAGAATGATGGCACCTTCCTAAACCCTCATGGGTGGTGTCTGAG

81 GPX3 NM_002084.3 1297-1396 GTCCAATTGTTCTGCTCTAACTGATACCTCAACCTTGGGGCCAGCATCTCCC

ACTGCCTCCAAATATTAGTAACTATGACTGACGTCCCCAGAAGTTTCT

82 GPX4 NM_001039847.1 436-535 CAGGGAGTAACGAAGAGATCAAAGAGTTCGCCGCGGGCTACAACGTCAAA

TTCGATATGTTCAGCAAGATCTGCGTGAACGGGGACGACGCCCACCCGCT

83 GPX5 NM_003996.3 886-985 CATCCTGAAGAACATTCCTAGACATTCTGACTCTTCCATCTCTCTCTACCCTG

GAGGTGTAGAAATAGCAATGGGGTCACAGTCACACAATTTAGGTTCC

84 GSR NM_000637.2 1526-1625 ATGCAGGGACTTGGGTGTGATGAAATGCTGCAGGGTTTTGCTGTTGCAGTG

AAGATGGGAGCAACGAAGGCAGACTTTGACAACACAGTCGCCATTCACC

85 GSS NM_000178.2 1511-1610 CAACCAGGCCACGGGACCTTCTATCCTCTGTATTTGTCATTCCTCTCCTAGC

CCTCCTGAGGGGTATCCTCCTAAAGACCTCCAAAGTTTTTATGGAAGG

86 GSTP1 NM_000852.2 416-515 TTTTGAGACCCTGCTGTCCCAGAACCAGGGAGGCAAGACCTTCATTGTGGG

AGACCAGATCTCCTTCGCTGACTACAACCTGCTGGACTTGCTGCTGATC

20 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

87 GSTZ1 NM_145871.2 1096-1195 CGTAACACCACGGGGAAGGCTGTGTGCCTTTTCTCATCCGCTTTTGTTGTGT

GTGACTCCAAAGAATGCCCGCGCTGAAATTTGGCGTGAATTAAACTGA

88 GUSB NM_000181.3 1900-1999 CCGATTTCATGACTGAACAGTCACCGACGAGAGTGCTGGGGAATAAAAAGG

GGATCTTCACTCGGCAGAGACAACCAAAAAGTGCAGCGTTCCTTTTGCG

89 HMOX1 NM_002133.1 1431-1530 TGTTGTTTTTATAGCAGGGTTGGGGTGGTTTTTGAGCCATGCGTGGGTGGGG

AGGGAGGTGTTTAACGGCACTGTGGCCTTGGTCTAACTTTTGTGTGAA

90 HPRT1 NM_000194.1 241-340 TGTGATGAAGGAGATGGGAGGCCATCACATTGTAGCCCTCTGTGTGCTCAA

GGGGGGCTATAAATTCTTTGCTGACCTGCTGGATTACATCAAAGCACTG

91 HSP90AA NM_001017963.2 1656-1755 CTTGACCAATGACTGGGAAGATCACTTGGCAGTGAAGCATTTTTCAGTTGA

1 AGGACAGTTGGAATTCAGAGCCCTTCTATTTGTCCCACGACGTGCTCCT

92 HSP90AB NM_007355.2 1881-1980 AGCCAATATGGAGCGGATCATGAAAGCCCAGGCACTTCGGGACAACTCCAC

1 CATGGGCTATATGATGGCCAAAAAGCACCTGGAGATCAACCCTGACCAC

93 HSPA1A NM_005345.5 99-198 GCTTCCCAGAGCGAACCTGTGCGGCTGCAGGCACCGGCGCGTCGAGTTTCC

GGCGTCCGGAAGGACCGAGCTCTTCTCGCGGATCCAGTGTTCCGTTTCC

94 HTATSF1 NM_014500.3 1091-1190 CTTTGATAGGCACCCAGATGGTGTGGCCTCTGTGTCCTTTCGGGATCCAGAG

GAAGCTGATTATTGTATTCAGACTCTCGATGGAAGATGGTTTGGTGGC

21 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

95 IFNA1 NM_024013.1 586-685 ATCCCTCTCTTTATCAACAAACTTGCAAGAAAGATTAAGGAGGAAGGAATA

ACATCTGGTCCAACATGAAAACAATTCTTATTGACTCATACACCAGGTC

96 IFNB1 NM_002176.2 611-710 ACAGACTTACAGGTTACCTCCGAAACTGAAGATCTCCTAGCCTGTGCCTCTG

GGACTGGACAATTGCTTCAAGCATTCTTCAACCAGCAGATGCTGTTTA

97 IFNG NM_000619.2 971-1070 ATACTATCCAGTTACTGCCGGTTTGAAAATATGCCTGCAATCTGAGCCAGTG

CTTTAATGGCATGTCAGACAGAACTTGAATGTGTCAGGTGACCCTGAT

98 IL10 NM_000572.2 231-330 AAGGATCAGCTGGACAACTTGTTGTTAAAGGAGTCCTTGCTGGAGGACTTT

AAGGGTTACCTGGGTTGCCAAGCCTTGTCTGAGATGATCCAGTTTTACC

99 IL12B NM_002187.2 1436-1535 GCAAGGCTGCAAGTACATCAGTTTTATGACAATCAGGAAGAATGCAGTGTT

CTGATACCAGTGCCATCATACACTTGTGATGGATGGGAACGCAAGAGAT

100 IL16 NM_001172128.1 3281-3380 TCCTGTGAGACGAAGCTACTTGACGAAAAGACCAGCAAACTCTATTCTATC

AGCAGCCAAGTGTCATCGGCTGTCATGAAATCCTTGCTGTGCCTTCCAT

101 IL1B NM_000576.2 841-940 GGGACCAAAGGCGGCCAGGATATAACTGACTTCACCATGCAATTTGTGTCT

TCCTAAAGAGAGCTGTACCCAGAGAGTCCTGTGCTGAATGTGGACTCAA

102 IL2 NM_000586.2 301-400 AGGATGCAACTCCTGTCTTGCATTGCACTAAGTCTTGCACTTGTCACAAACA

GTGCACCTACTTCAAGTTCTACAAAGAAAACACAGCTACAACTGGAGC

22 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

103 IRF1 NM_002198.1 511-610 CTGTGCGAGTGTACCGGATGCTTCCACCTCTCACCAAGAACCAGAGAAAAG

AAAGAAAGTCGAAGTCCAGCCGAGATGCTAAGAGCAAGGCCAAGAGGAA

104 IRF2 NM_002199.3 1625-1724 AGGCAGCACTAGCGACATTGCAGTCTGCTTCTGCACCTTATCTTAAAGCACT

TACAGATAGGCCTTCTTGTGATCTTGCTCTATCTCACAGCACACTCAG

105 KLRD1 NM_002262.3 597-696 CAATTTTACTGGATTGGACTCTCTTACAGTGAGGAGCACACCGCCTGGTTGT

GGGAGAATGGCTCTGCACTCTCCCAGTATCTATTTCCATCATTTGAAA

106 KRT1 NM_006121.2 691-790 TGCAGCAGGTAGATACCTCCACTAGAACCCATAATTTAGAGCCCTACTTTGA

GTCATTCATCAACAATCTCCGAAGGAGAGTGGACCAACTGAAGAGTGA

107 LDHA NM_001165414.1 1691-1790 AACTTCCTGGCTCCTTCACTGAACATGCCTAGTCCAACATTTTTTCCCAGTG

AGTCACATCCTGGGATCCAGTGTATAAATCCAATATCATGTCTTGTGC

108 LHPP NM_022126.3 382-481 ACCATACCTGCTCATCCATGACGGAGTCCGCTCAGAATTTGATCAGATCGAC

ACATCCAACCCAAACTGTGTGGTAATTGCAGACGCAGGAGAAAGCTTT

109 LPO NM_001160102.1 1613-1712 CTGCTGGCCAAGAAATCCAAGCTGATGAAACAGAATAAAATGATGACTGGA

GAGCTGCGCAACAAGCTTTTCCAGCCAACTCACAGGATCCATGGCTTTG

110 LTBR XM_006718983.2 681-780 GTGCTGCCTGGGCCCTCGAGTGTACACACTGCGAGCTACTTTCTGACTGCCC

GCCTGGCACTGAAGCCGAGCTCAAAGATGAAGTTGGGAAGGGTAACAA

23 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

111 MAP3K5 NM_005923.3 2416-2515 TTTAGGAAAAGGCACTTATGGGATAGTCTACGCAGGTCGGGACTTGAGCAA

CCAAGTCAGAATTGCTATTAAGGAAATCCCAGAGAGAGACAGCAGATAC

112 MAT1A NM_000429.2 2276-2375 GGGGATTAACCCACTAGCTCTTGTTGAGCCGTGGGCAATTGTCTGAAAAGT

GAAGACAGAACCACAGGGCTATTTTGTTTGCTTCATGTGTCCCAGAAGA

113 MAT2A NM_005911.4 806-905 AAGCAGTTGTGCCTGCGAAATACCTTGATGAGGATACAATCTACCACCTAC

AGCCAAGTGGCAGATTTGTTATTGGTGGGCCTCAGGGTGATGCTGGTTT

114 MB NM_005368.2 587-686 CAAGAGAGAGCGGGGTCTGATCTCGTGTAGCCATATAGAGTTTGCTTCTGA

GTGTCTGCTTTGTTTAGTAGAGGTGGGCAGGAGGAGCTGAGGGGCTGGG

115 MBL2 NM_000242.2 1757-1856 TGGACTTGTCTTTTGGTGGACATGGTGCACTAATTTCACTACCTATCCAGGA

GTGGAACTGGTAGAGGATGAGGAAAGCATGTATTCAGCTTTAGTAGAT

116 MPO NM_000250.1 546-645 AGACCTGCTGGAGAGGAAGCTGCGGTCCCTGTGGCGAAGGCCATTCAATGT

CACTGATGTGCTGACGCCCGCCCAGCTGAATGTGTTGTCCAAGTCAAGC

117 MPST NM_001013436.1 265-364 CGCGACGCGCGACGCGAGTTCGAGGAGCGCCACATCCCGGGCGCCGCTTTC

TTCGACATCGACCAGTGCAGCGACCGCACCTCGCCCTACGACCACATGC

118 MPV17 NM_002437.4 261-360 TGGTACAAGGTTTTGGATCGGTTCATCCCTGGCACCACCAAAGTGGATGCA

CTGAAGAAGATGTTGTTGGATCAGGGGGGCTTTGCCCCGTGTTTTCTAG

24 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

119 MSRA NM_001135670.2 473-572 TGCAGAAGTCGTCCGAGTGGTGTACCAGCCAGAACACATGAGTTTTGAGGA

ACTGCTCAAGGTCTTCTGGGAGAATCACGACCCGACCCAAGGTATGCGC

120 MT3 NM_005954.2 166-265 CACTAGCCAAGCCGCGCGTCCAGTTGCTTGGAGAAGCCCGTTCACCGCCTC

CAGCTGCTGCTCTCCTCGACATGGACCCTGAGACCTGCCCCTGCCCTTC

121 MTR NM_000254.2 6817-6916 TGAGTGAACTACAGCTATACCTCACAATAAGAATGAATCTCAGAAAATATT

AAGGAAAAAAGCAAGTTTGAAGAGACCACATGGGGCGTACTATTTTTAT

122 NCF1 NM_000265.4 375-474 TGAGCCTGCCCACCAAGATCTCCCGCTGTCCCCACCTCCTCGACTTCTTCAA

GGTGCGCCCTGATGACCTCAAGCTCCCCACGGACAACCAGACAAAAAA

123 NCF2 NM_000433.3 1005-1104 GCTCACCGTGTGCTATTTGGGTTTGTGCCTGAGACAAAAGAAGAGCTCCAG

GTCATGCCAGGGAACATTGTCTTTGTCTTGAAGAAGGGCAATGATAACT

124 NCOA7 NM_181782.3 3076-3175 TCAGGATCTGGAGGTGTGGGCATTTGATTGAAATTCAGACTGCCTTAAAAT

ATAACATTAAAAAGACTGGGTTCGATCAGCCCTCCTAAAGCTGGCTGGA

125 NFATC1 NM_172389.1 1985-2084 CGAATTCTCTGGTGGTTGAGATCCCGCCATTTCGGAATCAGAGGATAACCA

GCCCCGTTCACGTCAGTTTCTACGTCTGCAACGGGAAGAGAAAGCGAAG

126 NFKBIA NM_020529.1 946-1045 GGATGAGGAGAGCTATGACACAGAGTCAGAGTTCACGGAGTTCACAGAGG

ACGAGCTGCCCTATGATGACTGTGTGTTTGGAGGCCAGCGTCTGACGTTA

25 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

127 NOS2 NM_000625.4 1006-1105 CCCCAGCGGAGTGATGGCAAGCACGACTTCCGGGTGTGGAATGCTCAGCTC

ATCCGCTATGCTGGCTACCAGATGCCAGATGGCAGCATCAGAGGGGACC

128 NOX4 NM_001143836.2 1796-1895 ATAGCAAAATATAACAGAGGAAAAACAGTTGGTGTTTTCTGTTGTGGACCC

AATTCACTATCCAAGACTCTTCATAAACTGAGTAACCAGAACAACTCAT

129 NOX5 NM_024505.2 2136-2235 AGCTCCATAAGGTGGACTTTATCTGGATCAACAGAGACCAGCGGTCTTTCG

AGTGGTTTGTGAGCCTGCTGACTAAACTGGAGATGGACCAGGCCGAGGA

130 NQO1 NM_000903.2 791-890 CCGAATTCAAATCCTGGAAGGATGGAAGAAACGCCTGGAGAATATTTGGGA

TGAGACACCACTGTATTTTGCTCCAAGCAGCCTCTTTGACCTAAACTTC

131 NUDT1 NM_002452.3 394-493 TGTGGCCCGACGACAGCTACTGGTTTCCACTCCTGCTTCAGAAGAAGAAATT

CCACGGGTACTTCAAGTTCCAGGGTCAGGACACCATCCTGGACTACAC

132 PDLIM1 NM_020992.2 796-895 GGATCCCAACAAGCCCTCAGGATTCAGAAGTGTTAAAGCTCCTGTCACTAA

AGTGGCTGCGTCGATTGGAAATGCTCAGAAGTTGCCTATGTGTGACAAA

133 POLR1B NM_019014.3 3321-3420 GGAGAACTCGGCCTTAGAATACTTTGGTGAGATGTTAAAGGCTGCTGGCTA

CAATTTCTATGGCACCGAGAGGTTATATAGTGGCATCAGTGGGCTAGAA

134 PPIA NM_021130.2 926-1025 GGAATATTGAAAATGTAGGCAGCAACTGGGCATGGTGGCTCACTGTCTGTA

ATGTATTACCTGAGGCAGAAGACCACCT

26 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

135 PRDX1 NM_002574.2 633-732 GACCCATGAACATTCCTTTGGTATCAGACCCGAAGCGCACCATTGCTCAGG

ATTATGGGGTCTTAAAGGCTGATGAAGGCATCTCGTTCAGGGGCCTTTT

136 PRDX2 NM_181738.1 240-339 GTCGGACTACAAAGGGAAGTACGTGGTCCTCTTTTTCTACCCTCTGGACTTC

ACTTTTGTGTGCCCCACCGAGATCATCGCGTTCAGCAACCGTGCAGAG

137 PRDX3 NM_014098.3 964-1063 CATGGTTAGTTGCTAGTACAAGGAATCCTTTATTGGTAACATCTTGGTGGCT

GGCTAGCTAGTTTCTACAGAACATAATTTGCCTCTATAGAAGGCTATT

138 PRDX4 NM_006406.1 541-640 AGGAGGACTTGGGCCAATAAGGATTCCACTTCTTTCAGATTTGACCCATCAG

ATCTCAAAGGACTATGGTGTATACCTAGAGGACTCAGGCCACACTCTT

139 PRDX5 NM_012094.4 601-700 GGAAGGAGACAGACTTATTACTAGATGATTCGCTGGTGTCCATCTTTGGGA

ATCGACGTCTCAAGAGGTTCTCCATGGTGGTACAGGATGGCATAGTGAA

140 PRDX6 NM_004905.2 1391-1490 GTGATAAGTTTCTATCAAAATGGGGAGATTGCAGAAAAGGCTTCCCTTGGC

TCCCAAGGAGGTGTAGCAGGTGTGAGCAATATTAGTGCCATGTGCCTTT

141 PRNP NM_000311.3 1681-1780 GCATGAGCTCTGTGTGTACCGAGAACTGGGGTGATGTTTTACTTTTCACAGT

ATGGGCTACACAGCAGCTGTTCAACAAGAGTAAATATTGTCACAACAC

142 PTGR1 NM_001146109.1 908-1007 TGATAATGTAGGTGGAGAGTTTTCAAACACTGTTATCGGCCAGATGAAGAA

ATTTGGAAGGATTGCCATATGTGGAGCCATCTCTACATATAACAGAACC

27 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

143 PTGS1 NM_000962.2 701-800 ACCCCCAAGGCACCAACCTCATGTTTGCCTTCTTTGCACAACACTTCACCCA

CCAGTTCTTCAAAACTTCTGGCAAGATGGGTCCTGGCTTCACCAAGGC

144 PTGS2 NM_000963.1 496-595 GCTACAAAAGCTGGGAAGCCTTCTCTAACCTCTCCTATTATACTAGAGCCCT

TCCTCCTGTGCCTGATGATTGCCCGACTCCCTTGGGTGTCAAAGGTAA

145 PTK2B NM_004103.3 736-835 CCAGTAGATGTGGAAAAGGAGGACGTGCGTATCCTCAAGGTCTGCTTCTAT

AGCAACAGCTTCAATCCTGGGAAAAACTTCAAACTGGTCAAATGCACTG

146 RBL2 NM_005611.3 1311-1410 GAATCTCCACACCACTAACTGGTGTTAGGTACATTAAGGAGAATAGCCCTT

GTGTGACTCCAGTTTCTACAGCTACGCATAGCTTGAGTCGTCTTCACAC

147 RNF7 NM_014245.2 506-605 TTCAGAGCCCTGGTGGATCTTGTAATCCAGTGCCCTACAAAGGCTAGAACA

CTACAGGGGATGAATTCTTCAAATAGGAGCCGATGGATCTGTGGTCCTT

148 RPLP0 NM_001002.3 251-350 CGAAATGTTTCATTGTGGGAGCAGACAATGTGGGCTCCAAGCAGATGCAGC

AGATCCGCATGTCCCTTCGCGGGAAGGCTGTGGTGCTGATGGGCAAGAA

149 SEPP1 NM_005410.2 201-300 GGAGCATAAGAGATCAAGATCCAATGCTAAACTCCAATGGTTCAGTGACTG

TGGTTGCTCTTCTTCAAGCCAGCTGATACCTGTGCATACTGCAGGCATC

150 VIMP NM_203472.1 467-566 GAAATGCAAAGAAGCCCCAGGAGGAAGACAGTCCTGGGCCTTCCACTTCAT

CTGTCCTGAAACGGAAATCGGACAGAAAGCCTTTGCGGGGAGGAGGTTA

28 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

151 SELL NR_029467.1 1586-1685 CTAACTCCAGTGAAGTAATGGGGTCCTGCTCAAGTTGAAAGAGTCCTATTTG

CACTGTAGCCTCGCCGTCTGTGAATTGGACCATCCTATTTAACTGGCT

152 SERPINA NM_000295.4 761-860 TCACTGTCAACTTCGGGGACACCGAAGAGGCCAAGAAACAGATCAACGATT

1 ACGTGGAGAAGGGTACTCAAGGGAAAATTGTGGATTTGGTCAAGGAGCT

153 SERPINC NM_000488.3 1420-1519 TCAAGGCCAACAGGCCTTTCCTGGTTTTTATAAGAGAAGTTCCTCTGAACAC

1 TATTATCTTCATGGGCAGAGTAGCCAACCCTTGTGTTAAGTAAAATGT

154 SFTPD NM_003019.4 696-795 AAAGTGGGCTTCCAGATGTTGCTTCTCTGAGGCAGCAGGTTGAGGCCTTAC

AGGGACAAGTACAGCACCTCCAGGCTGCTTTCTCTCAGTATAAGAAAGT

155 SIRT2 NM_012237.3 611-710 CTATGACAACCTAGAGAAGTACCATCTTCCCTACCCAGAGGCCATCTTTGAG

ATCAGCTATTTCAAGAAACATCCGGAACCCTTCTTCGCCCTCGCCAAG

156 SLC7A11 NM_014331.3 637-736 TCCATTACCAGCTTTTGTACGAGTCTGGGTGGAACTCCTCATAATACGCCCT

GCAGCTACTGCTGTGATATCCCTGGCATTTGGACGCTACATTCTGGAA

157 SLPI NM_003064.2 331-430 TTTCTGTGAGATGGATGGCCAGTGCAAGCGTGACTTGAAGTGTTGCATGGG

CATGTGTGGGAAATCCTGCGTTTCCCCTGTGAAAGCTTGATTCCTGCCA

158 SMARCB NM_003073.3 1061-1160 AGAAGGAGAACTCACCAGAGAAGTTTGCCCTGAAGCTGTGCTCGGAGCTGG

1 GGTTGGGCGGGGAGTTTGTCACCACCATCGCATACAGCATCCGGGGACA

29 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

159 SOD1 NM_000454.4 36-135 GCCTATAAAGTAGTCGCGGAGACGGGGTGCTGGTTTGCGTCGTAGTCTCCT

GCAGCGTCTGGGGTTTCCGTTGCAGTCCTCGGAACCAGGACCTCGGCGT

160 SOD2 NM_000636.2 202-301 TTTGGGGTATCTGGGCTCCAGGCAGAAGCACAGCCTCCCCGACCTGCCCTA

CGACTACGGCGCCCTGGAACCTCACATCAACGCGCAGATCATGCAGCTG

161 SOD3 NM_003102.2 140-239 GGTGCAGCTCTCTTTTCAGGAGAGAAAGCTCTCTTGGAGGAGCTGGAAAGG

TGCCCGACTCCAGCCATGCTGGCGCTACTGTGTTCCTGCCTGCTCCTGG

162 SPINK1 NM_003122.2 66-165 GACGTGGTAAGTGCGGTGCAGTTTTCAACTGACCTCTGGACGCAGAACTTC

AGCCATGAAGGTAACAGGCATCTTTCTTCTCAGTGCCTTGGCCCTGTTG

163 SQRDL NM_021199.2 1521-1620 AAAGAGTTCCTTGATGGGTAATGGTGACCAAATGCCTCCCTTTTCAGTACCT

TTGAACAGCAACCATGTGGGCTACTCATGATGGGCTTGATTCTTTGGG

164 SQSTM1 NM_003900.3 1446-1545 GGGCCAGTTTCTCTGCCTTCTTCCAGGATCAGGGGTTAGGGTGCAAGAAGC

CATTTAGGGCAGCAAAACAAGTGACATGAAGGGAGGGTCCCTGTGTGTG

165 SRXN1 NM_080725.1 1656-1755 AAGAGAGTGAGAGTAGAAGCTGAAAGACTTCTTGAGTTCTTGGCCTGGAAC

TGGGACTAGGACAGTGTCACTTCTGCTAAGTTCTTTTGGTCAGAGCAAA

166 STAT1 NM_139266.1 456-555 ACAGTGGTTAGAAAAGCAAGACTGGGAGCACGCTGCCAATGATGTTTCATT

TGCCACCATCCGTTTTCATGACCTCCTGTCACAGCTGGATGATCAATAT

30 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

167 STAT3 NM_003150.3 2061-2160 AAAGAAGGAGGCGTCACTTTCACTTGGGTGGAGAAGGACATCAGCGGTAA

GACCCAGATCCAGTCCGTGGAACCATACACAAAGCAGCAGCTGAACAACA

168 SUOX NM_000456.2 1516-1615 GGCGGGATTACAAAGGCTTCTCTCCATCTGTGGACTGGGAGACTGTAGATTT

TGACTCTGCTCCATCCATTCAGGAACTTCCTGTCCAGTCGGCCATCAC

169 TGFB1 NM_000660.3 1261-1360 TATATGTTCTTCAACACATCAGAGCTCCGAGAAGCGGTACCTGAACCCGTGT

TGCTCTCCCGGGCAGAGCTGCGTCTGCTGAGGCTCAAGTTAAAAGTGG

170 TNF NM_000594.2 1011-1110 AGCAACAAGACCACCACTTCGAAACCTGGGATTCAGGAATGTGTGGCCTGC

ACAGTGAAGTGCTGGCAACCACTAAGAATTCAAACTGGGGCCTCCAGAA

171 TNFRSF1 NM_001066.2 836-935 CCCAGCTGAAGGGAGCACTGGCGACTTCGCTCTTCCAGTTGGACTGATTGTG

B GGTGTGACAGCCTTGGGTCTACTAATAATAGGAGTGGTGAACTGTGTC

172 TNFSF10 NM_003810.2 116-215 GGGGGGACCCAGCCTGGGACAGACCTGCGTGCTGATCGTGATCTTCACAGT

GCTCCTGCAGTCTCTCTGTGTGGCTGTAACTTACGTGTACTTTACCAAC

173 TPO NM_175719.3 297-396 CTCCAGCTCAGCTTCTGTCTTTTTCCAAACTTCCTGAGCCAACAAGCGGAGT

GATTGCCCGAGCAGCAGAGATAATGGAAACATCAATACAAGCGATGAA

174 TRAPPC6 NM_001270891.1 403-502 CCTGCGCGGCGCCCTCTATACCCTGGGCATTGAGAGCGTGGTCACCGCCTCC

A GTGGCAGCCCTGCCCGTCTGTAAGTTCCAGGTGGTGATTCCGAAATCC

31 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

175 TRIM5 NM_033092.2 801-900 GAAAGCTTCCTGGAAGACTCAAATACAGTATGACAAAACCAACGTCTTGGC

AGATTTTGAGCAACTGAGAGACATCCTGGACTGGGAGGAGAGCAATGAG

176 TST NM_001270483.1 1007-1106 CCTCTTAGGTGAAGGAGATGACATGTTTTTAGAATTGCTGTGCAAGGCTCAC

CCTCTCTCTGTCAACACTGGAATAAACTTTGCCTTTTCTGAGTAGTGA

177 TTN NM_001256850.1 34831- GTGCCTGAGGCTCCCAAAGAAGTTGTTCCTGAAAAGAAAGTGCCAGTGCCT

34930 CCTCCTAAAAAGCCTGAAGTGCCACCCACAAAAGTCCCAGAGGTGCCAA

178 TXN NM_003329.2 56-155 CAGCCAAGATGGTGAAGCAGATCGAGAGCAAGACTGCTTTTCAGGAAGCCT

TGGACGCTGCAGGTGATAAACTTGTAGTAGTTGACTTCTCAGCCACGTG

179 TXNRD1 NM_001093771.1 1010-1109 AGTGATGATCTTTTCTCCTTGCCTTACTGCCCGGGTAAGACCCTGGTTGTTG

GAGCATCCTATGTCGCTTTGGAGTGCGCTGGATTTCTTGCTGGTATTG

180 TXNRD2 NM_006440.3 496-595 TTTAACATCAAAGCCAGCTTTGTTGACGAGCACACGGTTTGCGGCGTTGCCA

AAGGTGGGAAAGAGATTCTGCTGTCAGCCGATCACATCATCATTGCTA

181 UCP2 NM_003355.2 101-200 CTGCCGTGTTCTCCCTGCGGCTCGGACACATAGTATGACCATTAGGTGTTTC

GTCTCCCACCCATTTTCTATGGAAAACCAAGGGGATCGGGCCATGATA

182 VPS4A NM_013245.2 1941-2040 CAAGCTCTGCCTCAAAGACCGAGTGACATAAGCCATTCCCACCCTCCTAGG

TTCACATCCAGGGCTGTGTCTTCCTTGGGGGAGGAGATGGTGTCGTTTA

32 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

183 XCL1 NM_002995.2 267-366 CCTCACTACCCAGCGACTGCCGGTTAGCAGAATCAAGACCTACACCATCAC

GGAAGGCTCCTTGAGAGCAGTAATTTTTATTACCAAACGTGGCCTAAAA

184 XPO1 NM_003400.3 3641-3740 AATCTTTCTTCAGGAATATGTGGCTAATCTCCTTAAGTCGGCCTTCCCTCAC

CTACAAGATGCTCAAGTAAAGCTCTTTGTGACAGGGCTTTTCAGCTTA

185 YY1 NM_003403.3 1119-1218 GACCCTGGAGGGCGAGTTCTCGGTCACCATGTGGTCCTCAGATGAAAAAAA

AGATATTGACCATGAGACAGTGGTTGAAGAACAGATCATTGGAGAGAAC

173

174

175

176

177

178

179

180

181

182

33 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

183 Table S3. List of differentially expressed genes from NanoString nCounter Gene Expression analysis for U1-shNT, shCTH, shNT+PMA 184 and shCTH+PMA samples. 185

Gene Name shNT shNT shCTH shCTH shNT+PMA shNT+PMA shCTH+PMA shCTH+PMA

AHCY 6800.61 6798.82 3007.09 3079.2 1541.84 1618.61 804.5 860.85

APOBEC3G 234.38 201 108.63 90.2 198.53 187.99 104.49 113

ATOX1 2007 1951.48 1489.12 1382.74 895.87 941 675.16 735.96

BAG2 699.43 740.18 499.9 483.13 178.78 223.11 152.72 167.26

BANF1 3394.76 3101.1 2070.74 1902.73 1205.03 1173.42 780.38 832.61

BAX 3677.24 3480.76 1997.68 1939.6 2444.62 2646.38 1357.63 1471.93

BNIP3 1022.62 969.89 7734.99 7048.59 522.51 467.92 5217.17 4695.3

BTRC 462.58 458.36 280.71 292.55 327.93 341.9 189.98 197

CASP3 1773.86 1800.47 1415.1 1273.72 1021.31 1054.63 827.15 832.61

CAT 11705.25 11386.66 6447.75 6002.32 1934.96 1907.83 1090.93 1128.48

CBX5 2396.81 2510.87 1287.24 1253.33 974.89 1005.05 558.98 523.35

CCNT1 1375.42 1209.17 823.87 795.29 802.03 790.2 430.38 470.57

CCS 414.48 363.71 283.6 267.45 277.55 319.18 189.25 180.65

34 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

CD44 2578.14 2314.13 1581.41 1450.97 2638.22 2492.47 1390.51 1492.74

CD69 165.3 135.06 325.9 338.82 233.1 249.97 210.44 219.3

CD74 1794.83 1626.06 1086.32 967.84 906.73 887.29 507.83 544.91

CDK9 689.56 611.5 461.45 457.25 407.93 426.6 251.36 266.14

COPS6 10065.84 9760.6 7308.15 6926.23 5343.6 5275.2 3573.83 3540.06

CXCL8 592.11 505.15 8366.59 8854.85 12612.29 12636.94 9282.76 8908.13

DHCR24 527.96 496.64 1047.87 1106.66 396.08 379.09 645.2 637.09

FHL2 176.4 178.66 129.78 138.04 166.93 171.47 74.53 83.26

GADD45A 189.97 173.35 330.7 330.98 73.09 71.27 65.76 73.6

GCLC 1615.96 1501.63 825.8 839.21 1225.77 1263.28 938.21 918.1

GCLM 1486.44 1451.65 1045.94 1017.25 1027.24 1013.31 453.76 481.72

GGT1 378.7 352.01 191.31 164.7 179.77 174.57 94.99 86.98

GOT2 2367.2 2204.59 1317.04 1279.21 918.59 892.46 421.61 457.93

GPX4 8558.43 7968.64 4669.26 4564.68 7274.61 7450.57 4099.21 4332.52

GSR 3815.4 3477.57 2787.9 2797.63 2851.57 2673.24 2004.3 2227.22

GSS 948.61 847.59 610.45 583.53 405.96 448.29 319.31 356.83

35 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

GSTP1 33364.08 32080.89 18260.76 16725.39 8306.79 8218.04 4798.48 5133.9

HTATSF1 1079.36 1024.13 738.31 683.13 435.59 497.88 273.28 315.2

IRF2 843.75 776.34 472.98 465.1 904.76 854.24 554.6 556.06

LHPP 794.41 762.51 367.23 370.98 353.61 365.66 199.48 206.66

LTBR 767.27 713.59 462.41 417.25 559.05 569.15 283.51 321.89

MAP3K5 181.33 142.51 116.32 135.69 106.67 92.96 79.65 70.62

MAT2A 4873.79 3889.14 1902.5 1924.69 2268.81 2273.49 851.26 882.41

MPST 1873.78 1978.07 1003.64 1087.84 805.99 819.12 576.52 527.07

MPV17 1560.45 1525.03 1034.41 1008.62 915.62 913.12 740.2 734.48

MSRA 1390.22 1337.86 893.09 896.47 1120.08 1184.78 685.39 695.82

NUDT1 1611.03 1535.66 817.14 742.74 530.41 505.11 279.86 300.33

PRDX1 7352.01 7163.59 5674.82 5528.6 5215.2 5379.53 4369.56 4820.93

PRDX3 11113.14 10357.22 6808.25 6555.26 5538.19 5225.62 3347.32 3353.47

PRDX4 3350.35 3272.32 2406.25 2286.26 964.02 1014.34 831.53 872.01

PRDX5 4445.75 4316.66 3067.65 2908.22 2390.3 2459.42 1750.75 1813.89

PRDX6 3735.22 3587.11 2431.24 2299.59 1469.74 1525.65 1392.71 1487.54

36 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

SIRT2 1845.4 1779.2 1095.93 1036.07 1374.92 1362.44 959.4 893.56

SMARCB1 1532.08 1426.12 1213.22 1103.52 768.45 774.7 512.22 579.11

SOD1 321.96 363.71 226.88 214.9 186.68 198.32 151.25 171.72

SRXN1 967.11 944.37 543.16 519.21 926.49 859.4 439.15 459.42

STAT1 5331.44 4973.89 3300.3 3220.37 4678.86 4914.71 2753.99 2807.81

TST 3956.02 3622.21 1776.57 1690.97 1624.81 1631.01 930.91 950.06

TXN 21556.45 20985.62 12849.35 12049.34 13859.79 14205.97 8202.79 8715.59

TXNRD1 5689.18 5314.2 3246.46 3265.86 5325.82 5596.45 2686.04 2700.02

TXNRD2 891.86 893.32 574.88 563.92 466.21 471.02 303.97 288.44

UCP2 1350.75 1216.62 719.09 676.86 469.17 481.35 320.04 354.6

VPS4A 1090.47 1042.21 682.56 669.8 616.34 693.1 415.77 413.33

YY1 3268.93 3107.48 2516.8 2403.12 1877.67 1889.24 1402.21 1405.02

186

187

188

37 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

189 Table S4. List of differentially expressed genes from NanoString nCounter Gene Expression analysis for U1 untreated, GYY4137, PMA

190 and GYY4137+PMA treated samples.

Gene name Untreated Untreated GYY4137 GYY4137 PMA PMA GYY4137+PMA GYY4137+PMA

U1 U1

SLC7A11 155.21 164.96 1603.4 1227.34 791.24 593.89 5890.68 6221.89

SRXN1 480.44 511.08 1121.46 1220.36 718.73 614.32 2618.63 2663.93

VIMP 35.98 32.38 46.05 43.04 36.26 34.52 105.07 104.38

TXNRD1 3171.21 3258.42 4192.18 3911.2 3123.37 2865.17 7959.31 7667.31

GCLM 1189.47 1172.48 1850 1640.33 670.74 630.52 1773.11 1565.68

GCLC 737.95 696.86 974.12 823.65 933.07 806.64 1310.13 1316.08

FTH1 21411.12 20353.01 127359.45 102750.92 238445.8 221289 643242.69 631401.5

HMOX1 395.78 336.09 4481.75 3622.68 2674.43 2214.22 10581.22 11129.97

GSR 2162.35 2208.51 3238.52 3362.09 3398.49 3150.49 5964.56 6169.7

GPX4 5594.59 5173.24 5606.28 4895.39 4513.91 4165.66 7481.55 7810.26

DUSP1 47.97 40.86 77.77 51.19 246.33 176.83 577.9 599.04

38 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

PRDX1 6725.5 6387.34 3107.55 3285.31 7922 7614.85 4109.35 4440.64

PRDX3 6118.77 5924.83 3687.72 3491.23 3234.27 3194.87 1937.29 1765.36

PRDX5 2971.55 2990.94 2560.12 2715.27 1629.4 1596.38 1553.12 1477.19

PRDX6 2177.87 2149.16 1154.2 1151.72 1869.33 1651.33 981.78 1066.48

GPX1 1494.24 1239.54 1951.3 1676.39 1647.53 1423.07 620.59 710.23

PRNP 1113.27 1190.98 1645.35 1504.22 1259.37 1144.8 3058.62 2820.5

CCS 219.41 227.4 304.92 261.75 287.92 226.85 469.55 494.67

MSRA 816.26 825.59 862.58 795.73 893.61 778.46 1318.34 1347.85

UCP2 557.34 656.77 514.68 492.1 556.64 447.35 392.38 347.17

ATOX1 1116.8 1091.54 690.68 665.44 635.55 554.43 508.95 467.44

DHCR24 620.84 665.25 336.64 401.36 693.13 660.81 305.37 258.68

NOS2 12.16 12.48 25.58 17.45 36.26 33.11 59.1 52.19

CYBB 294.9 336.09 400.08 375.76 1066.36 929.22 630.44 739.73

NQO1 721.02 559.64 960.81 894.62 1336.15 1306.83 655.07 614.93

NCF1 222.94 285.99 168.83 235 9386.11 7471.84 1494.01 1275.24

39 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

CAT 6398.15 7127.37 8670.86 9332.43 1282.83 1146.91 1666.4 1747.21

GSTP1 16617.27 17221.78 10642.62 11077.46 6391.77 5419.66 3776.07 3982.28

BCL2 249.75 277.51 311.06 241.98 132.23 133.85 88.66 95.3

FHL2 132.63 137.98 35.81 36.06 121.57 90.17 57.46 56.73

CDKN1A 227.88 194.26 539.24 415.32 1884.26 1800.68 3524.88 3526.19

LTBR 506.55 504.14 718.31 631.7 545.98 414.24 751.93 782.84

SQSTM1 2445.96 2482.17 8485.65 7189.53 7390.95 6768.05 19671.71 20338

CTH 326.64 295.24 814.49 685.22 146.09 130.33 2426.54 2568.63

ETHE1 600.38 526.5 435.9 408.34 988.52 925 812.68 844.11

SQRDL 3696.1 4049.33 4393.75 4257.88 6131.58 5533.78 6852.75 7215.76

GOT1 152.39 152.63 181.11 141.93 122.63 115.54 366.11 310.87

GGT1 194.72 201.97 201.58 184.97 91.71 78.2 55.82 56.73

TST 1945.05 1955.67 855.42 844.6 772.05 712.95 467.9 510.55

CD74 745.71 813.26 473.76 437.42 426.54 419.17 147.76 201.95

SERPINA1 83.95 80.94 64.46 58.17 148.22 159.22 50.89 38.57

40 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

TGFB1 2386.7 2275.58 1671.96 1548.42 6333.12 5824.74 4152.04 4073.05

XPO1 7199.59 7061.85 6369.61 5794.67 4144.95 3745.78 6219.03 6423.84

APOBEC3G 84.66 98.67 79.81 60.49 85.31 100.04 50.89 59

CBX5 1276.24 1385.23 1079.51 952.79 704.86 643.2 778.2 839.57

PPIA 788.75 764.69 268.09 295.49 294.32 322.66 49.25 72.61

CD4 1326.33 1436.11 1016.07 1086.57 523.58 420.58 246.27 235.99

CXCR4 13544.12 13203.29 15015.91 13298.3 689.94 672.79 466.26 621.74

BTRC 255.39 255.15 316.18 273.39 255.93 225.44 495.81 458.36

PTK2B 1329.86 1517.82 1342.48 1134.27 1674.19 1270.9 3061.9 3331.05

FOS 349.93 386.2 463.52 369.95 1903.45 1568.2 1164.02 1064.21

NFKBIA 1055.42 901.91 1296.43 1012.12 2064.48 1982.44 2700.71 2843.19

STAT1 3422.36 3465.79 4003.9 3601.74 3464.61 3107.51 5342.32 5284.75

STAT3 565.1 560.41 985.37 882.99 621.69 555.14 1313.42 1377.35

IRF1 50.8 53.96 62.42 41.88 33.06 44.38 113.28 108.92

YY1 2330.96 2374.25 2490.54 2216.19 1533.43 1330.78 2190.12 2536.86

191

41 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

192 Table S5. Characteristic of ART treated HIV-1 infected study participants.

Sl. no. Patient ID Age Sex CD4 cell count Viral load (copies/ml) Date of Date of start Data of sample collection

(cells/mm3) Diagnosis of ART

1 Subject A 32 F 555 5540.00 02/07/15 23-03-2015 16-08-2016

2 Subject B 31 M 484 <34 04-05-2012 23/06/15 05-10-2016

3 Subject C 34 F Target not detected 23-06-2011 08/01/15 10-08-2016

4 Subject D 25 F 642 Target not detected 26/09/06 16/10/06 01/10/19

5 Subject E 40 M 351 Target not detected 02/01/08 01/12/08 18/11/19

193

194

195

196

197

198

199

200

42 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

201 Table S6. List of primers and probes used in this study. 202

Host genes RT-qPCR primers

Primer Sequence

FP CTH TATTTACTCTGGCCGAGAGC

RP CTH TCCTCTAAGCCCACAGAAAG

FP CBS GCCTGAAGTGTGAGCTCTTG

RP CBS CACGATGATGCAGCGATAGC

FP MPST GCCGCTTTCTTCGACATC

RP MPST TGGCGTCGTAGATCACG

FP p24 ATAATCCACCTATCCCAGTAGGAGAAAT

RP p24 TTGGTTCCTTGTCTTATGTCCAGAATGC

FP actin ATGTGGCCGAGGACTTTGATT

RP actin AGTGGGGTGGCTTTTAGGATG

FP B2M GCCCAAGATAGTTAAGTGGGATCG

RP B2M TCATCCAATCCAAATGCGGC

43 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

FP GCLC GGCACAAGGACGTTCTCAAGT

RP GCLC CAAGGGTAGGATGGTTTGGG

FP TXN CCCTCTCTGAAAAGTATTCCAACG

RP TXN TGGCTCCAGAAAATTCACCCA

FP TXNRD1 ATGGGCAATTTATTGGTCCTCAC

RP TXNRD1 CCCAAGTAACGTGGTCTTTCAC

FP HO1 TAGAAGAGGCCAAGACTGCG

RP HO1 GGGCAGAATCTTGCACTTTGTT

ChIP-qPCR

ER P1 TTTCCGCTGGGGACTTTC

ER P2 CCAGTACAGGCAAAAAGCAG

RBEI P1 GAGCCCTCAGATGCTAC

RBEI P2 CAGTGGGTTCCCTAGTTAGC

RBEIII P1 GACAGCCTCCTAGCATTTCG

RBEIII P2 CAGCGGAAAGTCCCTTGTAG

GAPDH FP GGACCTGACCTGCCGTCTAGAA

44 bioRxiv preprint doi: https://doi.org/10.1101/2021.04.21.440760; this version posted April 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

GAPDH RP GGTGTCGCTGTTGAAGTCAGAG

Total HIV DNA and RNA in primary cells

HIV L1 ATGCCACGTAAGCGAAACTCTGGGTCTCTCTAGTTAGAC

HIV R1 CCATCTCTCTCCTTCTAGC

HIV L2 ATGCCACGTAAGCGAAACT

HIV R2 CTGAGGGATCTCTAGTTACC

HIV Taq probe FAM-CACTCAAGGCAAGCTTTATTGAGGC-NFQ

CD3OUT5 ACTGACATGGAACAGGGGAAG

CD3OUT3 CCAGCTCTGAAGTAGGGAACATAT

CD3IN5 GGCTATCATTCTTCTTCAAGGT

CD3IN3 CCTCTCTTCAGCCATTTAAGTA

CD3 Taq probe FAM-AGCAGAGAACAGTTAAGAGCCTCCAT-NFQ

203

45