bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

1 Functional proteomic atlas of HIV-infection in

2 primary human CD4+ T cells

3

4 Adi Naamati1, James C Williamson1,2, Edward JD Greenwood1,2, Sara Marelli1, Paul J

5 Lehner1,2, Nicholas J Matheson1,*

6

7 1. Department of Medicine, University of Cambridge, Cambridge Biomedical Campus,

8 Cambridge, CB2 0XY, UK.

9 2. Cambridge Institute for Medical Research, University of Cambridge, Cambridge, CB2

10 8BX

11

12 *Correspondence: [email protected].

13

14 Viruses manipulate host cells to enhance their replication, and the identification of

15 host factors targeted by viruses has led to key insights in both viral pathogenesis and

16 cellular physiology. We previously described global changes in cellular protein levels

17 during human immunodeficiency virus (HIV) infection using transformed CEM-T4 T

18 cells as a model. In this study, we develop an HIV reporter virus displaying a

19 streptavidin-binding affinity tag at the surface of infected cells, allowing facile one-

20 step selection with streptavidin-conjugated magnetic beads. We use this system to

21 obtain pure populations of HIV-infected primary human CD4+ T cells for detailed

22 proteomic analysis, including quantitation of >9,000 proteins across 4 different

23 donors, and temporal profiling during T cell activation. Remarkably, amongst 650

24 cellular proteins significantly perturbed during HIV infection of primary T cells bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

25 (q<0.05), almost 50% are regulated directly or indirectly by the viral accessory

26 proteins Vpr, Vif, Nef and Vpu. The remainder have not been previously characterised,

27 but include novel Vif-dependent targets FMR1 and DPH7, and 192 targets not

28 identified and/or regulated in T cell lines, such as AIRD5A and PTPN22. We therefore

29 provide a high-coverage functional proteomic atlas of HIV infection, and a

30 mechanistic account of HIV-dependent changes in its natural target cell.

31 Remodelling of the host proteome during viral infection may reflect direct effects of viral

32 proteins, secondary effects or cytopathicity accompanying viral replication, or host

33 countermeasures such as the interferon (IFN) response. By defining time-dependent

34 changes in protein levels in infected cells, and correlating temporal profiles of cellular and

35 viral proteins, we have shown that it is possible to differentiate these phenomena, and

36 identify direct cellular targets of human cytomegalovirus (HCMV) and HIV1-3. To enable time

37 course analysis and minimise confounding effects from uninfected bystander cells, pure

38 populations of synchronously infected cells must be sampled sequentially as they progress

39 through a single round of viral replication. In the case of HIV, we previously satisfied these

40 conditions by spinoculating the highly permissive CEM-T4 lymphoblastoid T cell line4-6 with

41 Env-deficient NL4-3-ΔEnv-EGFP virus7 at a high multiplicity of infection (MOI)3.

42 The utility of cancer cell line models (such as CEM-T4) is, however, limited by the extent to

43 which they retain the characteristics of the primary cells from which they were derived, and

44 cancer-specific and in vitro culture-dependent reprogramming are well described8. For

45 example, the HIV accessory proteins Vif, Nef and Vpu are required for viral replication in

46 primary T cells, but not in many T cell lines9-12, and HIV is restricted by type I interferon (IFN)

47 in primary T cells, but not CEM-derived T cells13. In addition, whilst ensuring a high %

48 infection, dysregulation of the cellular proteome at high MOIs may not be indicative of protein

49 changes when a single transcriptionally active provirus is present per cell. In this study, we

50 therefore sought to apply our temporal proteomic approach to HIV infection of primary

51 human CD4+ T lymphocytes, the principle cell type infected in vivo, at an MOI <1. To this bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

52 end, we have developed a reporter virus for antibody-free magnetic cell sorting (AFMACS)14

53 of HIV-infected cells (Fig. 1a). AFMACS allows facile, one-step affinity purification of infected

54 cells from mixed cultures, bypassing the need for high MOIs or fluorescence-activated cell

55 sorting (FACS). We use this system to generate a detailed atlas of cellular protein dynamics

56 in HIV-infected primary human CD4+ T cells, show how this resource can be used to identify

57 novel cellular proteins regulated by HIV, and assign causality to individual HIV accessory

58 proteins.

59 AFMACS-based magnetic selection requires the high-affinity 38 amino acid streptavidin-

60 binding peptide (SBP)15 to be displayed at the cell surface by fusion to the N-terminus of the

61 truncated Low-affinity Nerve Growth Factor Receptor (SBP-ΔLNGFR)16. Cells expressing

62 this marker may be selected directly with streptavidin-conjugated magnetic beads, washed

63 to remove unbound cells, then released by incubation with the naturally occurring vitamin

64 biotin14. To engineer a single round HIV reporter virus encoding SBP-ΔLNGFR, we

65 considered 3 settings in the proviral construct: fused to the endogenous Env signal peptide

66 (as a direct replacement for Env); or as an additional cistron, downstream of nef and either a

67 P2A peptide or IRES. We used Env-deficient pNL4-3-ΔEnv-EGFP (HIV-1) as a backbone

68 and, since increased size of lentiviral genome is known to reduce packaging efficiency17,

69 tested each approach in constructs from which EGFP was removed and/or the 3’ long

70 terminal repeat (LTR) truncated. Further details relating to construct design are described in

71 the Methods and Supplementary Methods.

72 For initial screening, VSVg-pseudotyped viruses were made in 293T cells under standard

73 conditions, and used to spinoculate CEM-T4 T cells (CEM-T4s). Infected cells were

74 identified by expression of EGFP and/or cell surface LNGFR, combined with Nef/Vpu-

75 mediated downregulation of CD418,19. Whilst infection is not truly “productive” (because Env

76 is deleted), Gag alone is sufficient for assembly and release of virions20, and other structural

77 and non-structural viral proteins are expressed in accordance with full length viral infection3. bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

78 As expected, all viruses tested expressed SBP-ΔLNGFR at the cell surface of infected cells

79 (Supplementary Fig. 1a), but the larger constructs resulted in lower infectious viral titres

80 (Supplementary Fig. 1a,b). We therefore selected pNL4-3-ΔEnv-SBP-ΔLNGFR, pNL4-3-

81 ΔENV-Nef-P2A-SBP-ΔLNGFR and pNL4-3-ΔEnv-Nef-IRES-SBP-ΔLNGFR-Δ3 for further

82 evaluation (Supplementary Fig. 2a). Viruses generated from these constructs expressed

83 high levels of SBP-ΔLNGFR 48 hrs post-infection, and depleted CD4 and tetherin to a

84 similar extent. However, only the pNL4-3-ΔENV-Nef-P2A-SBP-ΔLNGFR virus (Fig. 1b)

85 expressed high levels of LNGFR 24 hrs post-infection in both CEM-T4s (Supplementary

86 Fig. 2a) and primary human CD4+ T cells (Fig. 1c). This is consistent with Nef-P2A-SBP-

87 ΔLNGFR expression from completely spliced transcripts early in HIV infection21, with the

88 P2A peptide ensuring that translation of Nef and SBP-ΔLNGFR follow similar kinetics.

89 Since analysis of cells at early as well as late time points is essential for the generation of

90 time course data, we focussed on pNL4-3-ΔEnv-Nef-P2A-SBP-ΔLNGFR (now termed HIV-

91 AFMACS). To confirm that HIV-AFMACS virus could be used for cell selection (Fig. 1a),

92 infected primary T cells were captured by streptavidin-conjugated magnetic beads, released

93 by incubation with excess biotin, then analysed by flow cytometry. Compared with

94 unselected cells (input) or cells released during washing (flow-through), selected cells were

95 markedly enriched for SBP-ΔLNGFR expression and CD4 downregulation (Fig. 1d). In fact,

96 from mixed populations containing approximately 20-40% infected cells (corresponding to an

97 effective MOI/=90% were routinely achieved by AFMACS of both CEM-

98 T4s and primary human CD4+ T cells, with

99 (Fig. 1e).

100 To gain a comprehensive, unbiased overview of viral and cellular protein dynamics during

101 HIV-infection of its natural target cell, we used the HIV-AFMACS virus to spinoculate

102 activated, primary human CD4+ T cells, sorted infected (SBP-ΔLNGFR positive) and

103 uninfected (SBP-ΔLNGFR negative) cells by AFMACS 24 hrs and 48 hrs post-infection, and

104 analysed whole cell lysates using tandem mass tag (TMT)-based quantitative proteomics bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

105 (Fig. 2a,b and Supplementary Fig. S3a)1,3. Interpretation of HIV-dependent proteomic

106 remodelling in primary T cells is complicated by concurrent changes in relative protein

107 abundance resulting from T cell activation22. We therefore exploited multiplex TMT labelling

108 to measure parallel protein abundances in resting and activated (uninfected) T cells from the

109 same donor, as well as control (mock-infected) T cells obtained at each time point.

110 In total, we quantitated 9,076 proteins across 10 different conditions. This and other datasets

111 described in this study will be deposited to the ProteomeXchange consortium (accessible at

112 http://proteomecentral.proteomexchange.org) and are summarised in an interactive

113 spreadsheet which allows generation of temporal profiles for any quantitated proteins of

114 interest (Supplementary Table 1). For example, the restriction factor tetherin (targeted by

115 HIV-1 Vpu10) is upregulated by T cell activation, then progressively depleted in HIV-infected

116 (red, SBP-ΔLNGFR positive) but not uninfected (blue, SBP-ΔLNGFR negative) cells (Fig.

117 2c, left panel). Conversely, the restriction factor SAMHD1 (targeted by some HIV-2/SIV Vpx

118 and Vpr variants, but not HIV-123-25) is depleted by T cell activation, independent of HIV

119 infection (Fig. 2c, right panel). In these graphical representations, relative protein

120 abundances for each condition are depicted by bars, and ratios of protein abundances in

121 paired experimental/control cells from each condition/time point are depicted by lines (grey,

122 resting/activated; red, SBP-ΔLNGFR positive, infected; blue, SBP-ΔLNGFR negative,

123 uninfected).

124 Aside from tetherin, levels of many other reported Vpu (CD4, SNAT1)2,19, Nef (CD4,

125 SERINC5)11,12,18, Vif (APOBEC3 and PPP2R5 families)3,9 and Vpr (UNG, HLTF, ZGPAT,

126 VPRBP, MUS81, EME1, MCM10, TET2)26-33 substrates were all reduced by HIV infection in

127 primary T cells (Fig. 2d, and Supplementary Fig. 3b). Conversely, and consistent with our

128 previous observations in CEM-T4s, APOBEC3B and SERINC1 were not depleted

129 (Supplementary Fig. 3b)2,3. In the absence of donor haplotyping, polymorphisms at the

130 MHC-I locus make routine proteomic quantification problematic. Nonetheless, our data are bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

131 consistent with depletion of HLA-A and HLA-B, but not HLA-C (Supplementary Fig. 3b), as

132 previously reported for Nef/Vpu variants from NL4-3 HIV34-36.

133 Together with cellular proteins, we identified gene products from 7 viral open reading frames

134 (ORFs; Supplementary Fig. 3c). As expected37, viral regulatory and accessory proteins

135 expressed from fully spliced, Rev-independent transcripts (Tat, Rev, Nef-P2A and SBP-

136 ΔLNGFR) were expressed early in infection, peaking at 24 hrs. Conversely, viral structural

137 proteins expressed from unspliced, Rev-dependent transcripts (Gag and Gagpol) were

138 expressed late in infection, increasing progressively from 24 to 48 hrs. Viral accessory

139 proteins expressed from partially spliced transcripts were either not detected (Vpr and Vpu)

140 or incompletely quantitated (Vif).

141 Inter-individual variability is known to affect gene expression during T cell activation38.

142 Accordingly, to identify reproducible HIV targets, we analysed primary human CD4+ T cells

143 from 3 further donors. In each case, mock-infected cells were compared with HIV-infected

144 cells selected using AFMACS 48 hrs post-infection (Fig. 3a,b and Supplementary Fig. 4a).

145 Aside from APOBEC3 proteins, we recently discovered the PPP2R5A-E (B56) family of

146 PP2A phosphatase regulatory subunits to be degraded by diverse Vif variants, spanning

147 primate and ruminant lentiviruses3. To formally document Vif-dependent changes in primary

148 T cells, both wildtype (WT) and Vif-deficient (ΔVif) viruses were therefore included. Whilst

149 some donor-dependent differences were apparent, most sample-sample variability was

150 accounted for by HIV infection (Supplementary Fig. 4b), and all accessory protein

151 substrates from Fig. 2d and Supplementary Fig. 3b were significantly depleted by WT HIV

152 (Fig. 3c – left panel). In total, we quantitated 8,789 cellular proteins, of which 650 were

153 significantly regulated by HIV infection (q<0.05).

154 Compared with a previous, similar experiment using CEM-T4s3, we observed greater

155 variability in protein abundances between replicates (Supplementary Fig. 5a), but a high

156 degree of correlation in HIV-dependent changes between cell types (Supplementary Fig.

157 5b). As well as “canonical” accessory protein targets, we have recently discovered that most bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

158 protein-level changes in HIV-infected CEM-T4s may be explained by primary and secondary

159 effects of Vpr, including degradation of at least 34 additional substrates39. These changes

160 were recapitulated in primary T cells (Fig. 3c – middle panel), with 33 newly described Vpr

161 substrates quantitated, and 32 decreased in abundance. Several other cell surface proteins

162 reported to be downregulated by Nef and/or Vpu were also depleted, but the magnitude of

163 effect was typically modest, and many were unchanged (Supplementary Fig. 5c – left

164 panel). Likewise, we did not see evidence of HIV/Vif-dependent transcriptional regulation of

165 RUNX1 target gene products such as T-bet/TBX21 (Supplementary Fig. 5c – middle

166 panel)40. Nonetheless, taken together, known accessory protein-dependent changes,

167 characterised in transformed T cell lines, are able to account for 297/650 (46%) of proteins

168 regulated by HIV in primary T cells (Supplementary Fig. 5d), including 175/299 (59%) of

169 proteins decreased in abundance.

170 As with individual proteins, pathways and processes downregulated by HIV infection of

171 primary T cells are dominated by the effects of accessory proteins (Fig. 3d and

172 Supplementary Fig. 6a). These include the DNA damage response and cell cycle (Vpr)39,41-

173 47, cytidine deamination and PP2A activity (Vif)3,9,48 and amino acid transport (Vpu/Nef)2.

174 Proteins upregulated by HIV are more diverse, with fewer dominant functional clusters.

175 Nonetheless, we saw marked increases in proteins associated with lipid and sterol

176 metabolism (Supplementary Fig. 6b). A similar effect has been reported in T cell lines at

177 the transcriptional level, and attributed to the expression of Nef49,50. Similarly, several

178 proteins in these pathways are indirectly regulated by Vpr (Supplementary Fig. 6b)39.

179 Despite the overall agreement with cell line data, 1,252/8,789 (14%) of cellular proteins

180 quantitated in this experiment were not identified in a previous, similar experiment using

181 CEM-T4s3. Furthermore, having excluded known accessory-protein dependent changes,

182 192/650 (30%) proteins regulated by HIV in primary T cells are either not detected, or not

183 significantly/concordantly regulated, in CEM-T4s (Fig. 3c – right panel and Supplementary

184 Fig. 5d). These proteins may represent accessory protein substrates expressed in primary T bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

185 cells but not T cell lines, or proteins regulated by alternative, cell type-specific mechanisms,

186 such as the interferon response (Supplementary Fig. 5c – right panel)51. To validate our

187 proteomic data, we focused on 2 such novel HIV targets with commercially available

188 antibodies: ARID5A and PTPN22. These proteins were readily identified in proteomic

189 datasets from primary T cells (9-14 unique peptides) but not CEM-T4s, and consistently

190 depleted across all donors with a fold change >2 (Supplementary Fig. 7a). As expected,

191 depletion was also seen by immunoblot (Fig. 3e).

192 We previously showed that substrates of different HIV accessory proteins could be

193 distinguished by their characteristic patterns of temporal regulation in HIV-infected CEM-

194 T4s3, and similar clustering was observed in primary T cells (Supplementary Fig. 7b). Vpr

195 is packaged stoichiometrically in virions52-54 and, since the number of fusogenic HIV particles

196 exceeds the infectious MOI by at least several fold55, all cells in our time course experiment

197 were necessarily exposed to incoming Vpr. Accordingly, depletion of known Vpr substrates

198 was near-maximal by 24 hrs in infected (red, SBP-ΔLNGFR positive) cells, with partial

199 depletion also seen in uninfected (blue, SBP-ΔLNGFR negative) cells (Fig. 2e – upper

200 panel). In contrast, since de novo viral protein synthesis is absolutely required, depletion of

201 known Vif, Nef and Vpu substrates increased progressively from 24 to 48 hrs, and was only

202 seen in HIV-infected (red, SBP-ΔLNGFR positive) cells (Fig. 2e – lower panel). Based on

203 their patterns of temporal regulation, ARID5A and PTPN22 are therefore very likely to

204 represent novel Vpr substrates, specific for primary T cells (Supplementary Fig. 7c).

205 Consistent with this, another member of the ARID5 subfamily of AT-rich interaction domain

206 (ARID)-containing proteins, ARID5B, is a widely conserved target of Vpr variants from

207 primate lentivuses39, and shares a similar temporal profile (Supplementary Fig. 7d).

208 In respect of Vif targets, as predicted, both APOBEC3 and PPP2R5 family proteins were

209 depleted in primary CD4+ T cells infected with WT, but not ΔVif viruses (Fig. 4a,b). Vif-

210 dependent depletion of PPP2R5A-E causes a marked increase in protein phosphorylation in

211 HIV-infected CEM T4 T cells, particularly substrates of the aurora kinases (AURKA/B)3. bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

212 AURKB activity is enhanced by “activation loop” auto-phosphorylation at threonine 232

213 (T232), antagonised by PP2A-B5656,57. Accordingly, a marked increase in AURKB T232

214 phosphorylation is seen in CEM-T4s transduced with Vif as a single transgene (Fig. 4c). We

215 therefore confirmed depletion of PPP2R5D by immunoblot of AFMACS-selected primary T

216 cells and, as a functional correlate, observed increased AURKB phosphorylation (Fig. 4d).

217 Besides these known substrates, we also noted differential regulation of several other

218 proteins in primary T cells infected with WT vs ΔVif viruses (Fig. 4a). Modest changes in

219 PPP2R1A, PPP2R1B and PPP2CA (catalytic/structural subunits of the trimeric PP2A

220 holoenzyme) and PPFIA1 and SGO1 (known PP2A interactors)58-60 are likely to be

221 secondary to destabilisation of PP2A by PPP2R5 depletion, or reflect proximity of the

222 holoenzyme to the Vif-cullin E3 ligase complex (Fig. 4b). Conversely, DPH7 and FMR1 are

223 not known to interact physically with PP2A, and show more profound and consistent

224 depletion (Fig. 4e). We therefore suspected these proteins to be novel Vif substrates.

225 To confirm these findings, we first re-examined our proteomic data from CEM-T4s3. Unlike

226 ARID5A and PTPN22, DPH7 and FMR1 are expressed in CEM-T4 as well as primary T

227 cells, and only decreased in abundance in HIV-infected cells in the presence of Vif (Fig. 4e).

228 Next, we confirmed Vif-dependent depletion of both proteins by immunoblot, in cells infected

229 with WT (but not ΔVif) viruses (Fig. 4f). Finally, we repeated these observations in cells

230 transduced with Vif as a single transgene (Fig. 4g). Vif is therefore both necessary and

231 sufficient for depletion of DPH7 and FMR1 and, taken together with APOBEC3 and PPP2R5

232 family proteins/interactors, we can account for all significant Vif-dependent changes in the

233 natural target cell of HIV infection.

234 Compared with FACS, bead-based magnetic sorting is fast, simple and scalable for

235 simultaneous processing of multiple samples and large cell numbers61. In conventional,

236 antibody-based immunomagnetic selection, cells remain coated with beads and antibody-

237 antigen complexes, risking alteration of their behaviour or viability through cross-linking of

238 cell-surface receptors or internalisation of ferrous beads61-63. Conversely, AFMACS-based bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

239 selection is antibody free, and selected cells are released from the beads by incubation with

240 biotin, suitable for a full range of downstream applications14. The HIV-AFMACS virus

241 described in this study allows routine isolation of HIV-infected cells subjected to an MOI<1,

242 avoiding artefacts associated with high MOIs and facilitating experiments in primary cells,

243 where high levels of infection are difficult to achieve in practice.

244 To demonstrate the utility of this system and provide a resource for the community, we have

245 generated the first high coverage proteomic atlas of HIV-infected primary human CD4+ T

246 cells. As well as identifying HIV-dependent changes in cells from multiple donors, viral

247 regulation may be assessed against a background of endogenous regulation triggered by T

248 cell activation. Unlike T cell lines, primary T cells express a full range of proteins relevant to

249 HIV infection in vivo, and are not confounded by the genetic and epigenetic effects of

250 transformation. Furthermore, proteins unique to primary T cells were significantly more likely

251 to be regulated by HIV infection than proteins detected in both primary T cells and CEM-T4s

252 (131/1,252=10.5% vs 519/7,537=6.9%; p<0.0001, two-tailed Fisher’s exact test). Our data

253 validate many, but not all, previously reported HIV accessory protein targets. For some

254 Vpu/Nef substrates, such as NTB-A and CCR7, downregulation from the plasma membrane

255 may occur in the absence of protein degradation64-66. For others, such as ICAM-1/3,

256 accessory protein expression may prevent upregulation, without reducing levels below

257 baseline67. Alternatively, and particularly where targets have been discovered using model

258 cell line or overexpression systems, regulation may not be quantitatively significant at the

259 protein level in the context and/or natural cell of HIV infection.

260 Previous temporal proteomic analyses of HIV-infected primary human CD4+ T cells were

261 hampered by limited coverage of the cellular proteome (<2,000 proteins), and did not detect

262 regulation of known or novel HIV accessory protein targets68,69. A more recent study

263 quantitated 7,761 proteins in FACS-sorted T cells at a single time point 96 hrs post-infection

264 with an R5-tropic, GFP-expressing Nef-deficient virus70. Depletion of several accessory

265 protein targets (including APOBEC3 and PPP2R5 families) was confirmed, and many bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

266 proteins differentially regulated at 96 hrs were also regulated at 48 hrs in our study

267 (Supplementary Fig. 8a). In keeping with the late time point, changes were dominated by

268 pathways involved in cell death and survival, and factors maintaining viability of HIV-infected

269 T cells (such as BIRC5) were enriched. Conversely, the full dataset is not available, effect

270 sizes were generally compressed (Supplementary Fig. 8a), and 413/650 (64%) of the HIV-

271 dependent changes identified in our study were obscured, including depletion of ARID5A,

272 PTPN22, DPH7 and 19/51 known accessory protein substrates (Supplementary Fig. 8b).

273 To maximise viral titers, enable synchronous single round infections and avoid co-receptor-

274 dependent targeting of T cell lineages with pre-existing proteomic differences, we used an

275 Env-deficient proviral backbone and pseudotyped viruses with VSVg. Different envelope

276 proteins may alter the route of viral entry, and/or Env-dependent signalling early in

277 infection71,72. To limit the impact of these effects, we focussed on cellular proteins

278 progressively regulated over 48 hrs infection, and included SBP-ΔLNGFR negative cells

279 (flow-through, exposed to incoming virions but not infected) as a control in our time course

280 analysis. Temporal profiling is particularly well suited to identifying and characterising host

281 factors regulated directly by viral proteins. In fact, as we show here, HIV accessory proteins

282 account for much or most of the proteomic remodelling in infected cells. The abundance of

283 direct accessory protein targets likely explains why proteins and processes/pathways

284 downregulated by HIV in primary T cells correlate so well with changes seen in T cell lines,

285 and are robust to inter-individual variation. In comparison, upregulated proteins concord less

286 well with changes in T cell lines, and functional effects are less homogeneous. This may be

287 because upregulated proteins reflect indirect effects (for example, secondary changes in

288 transcription), which are more likely to be cell-type specific.

289 Amongst the HIV accessory proteins, Vif was thought until recently to exclusively degrade

290 APOBEC3 family cytidine deaminases. As well as confirming equivalently-potent Vif-

291 dependent depletion of PPP2R5 family phosphatase subunits in primary T cells, our data

292 revealed unexpected Vif-dependent depletion of DPH7 and FMR1. Further work will be bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

293 required to confirm that these proteins are recruited directly for degradation by the ubiquitin-

294 proteasome system, determine whether (like APOBEC3 and PPP2R5 proteins) they are

295 antagonised by Vif variants from diverse primate and non-primate lentiviral lineages, and

296 identify relevant in vitro virological phenotypes associated with target depletion.

297 Nonetheless, FMR1 is known to reduce HIV virus infectivity when over-expressed in

298 producer cells73, and the other novel targets highlighted in this study also impact processes

299 relevant for HIV replication, such as inflammatory cytokine signalling (ARID5A)74, T cell

300 activation (PTPN22)75 and translational fidelity (DPH7)76,77. The diversity of these targets

301 underscores the benefit of an unbiased, systems-level approach to viral infection, and the

302 capacity of the resources presented in this study to reveal unsuspected aspects of the host-

303 virus interaction in the natural target cell of HIV infection.

304 bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

305 Methods

306 General cell culture

307 CEM-T4 T cells (CEM-T4s)5 were obtained from the AIDS Reagent Program, Division of

308 AIDS, NIAD, NIH: Dr J.P. Jacobs and cultured in RPMI supplemented with 10 % FCS,

309 100units/ml penicillin and 0.1 mg/ml streptomycin at 37 °C in 5 % CO2. HEK 293T cells

310 (Lehner laboratory stocks) were cultured in DMEM supplemented with 10 % FCS,

311 100units/ml penicillin and 0.1 mg/ml streptomycin at 37 °C in 5 % CO2. All cells were

312 confirmed to be mycoplasma negative (Lonza MycoAlert). Cell line authentication was not

313 undertaken.

314 Primary cell isolation and culture

315 Primary human CD4+ T cells were isolated from peripheral blood by density gradient

316 centrifugation over Lympholyte-H (Cedarlane Laboratories) and negative selection using the

317 Dynabeads Untouched Human CD4 T Cells kit (Invitrogen) according to the manufacturer’s

318 instructions. Purity was assessed by flow cytometry for CD3 and CD4 and routinely found to

319 be ≥95%. Cells were activated using Dynabeads Human T-Activator CD3/CD28 beads

320 (Invitrogen) according to the manufacturer’s instructions and cultured in RPMI supplemented

321 with 10% FCS, 30U/ml recombinant human IL-2 (PeproTech), 100units/ml penicillin and

322 0.1mg/ml streptomycin at 37°C in 5% CO2.

323 Ethics statement

324 Ethical permission for this study was granted by the University of Cambridge Human Biology

325 Research Ethics Committee (HBREC.2017.20). Written informed consent was obtained from

326 all volunteers prior to providing blood samples.

327 HIV-1 molecular clones

328 pNL4-3-ΔEnv-EGFP7 was obtained from the AIDS Reagent Program, Division of AIDS,

329 NIAD, NIH: Drs Haili Zhang, Yan Zhou, and Robert Siliciano and the complete proviral bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

330 sequence verified by Sanger sequencing (Source BioScience). Derived from the HIV-1

331 molecular clone pNL4-3, it encodes EGFP in the env ORF), resulting in a large, critical env

332 deletion and expression of a truncated Env-EGFP fusion protein retained in the endoplasmic

333 reticulum (ER) by a 3’ KDEL ER-retention signal.

334 The SBP-ΔLNGFR selection marker comprises the high-affinity 38 amino acid SBP fused to

335 the N-terminus of a truncated (non-functional) member of the Tumour Necrosis Factor

336 Receptor superfamily (LNGFR)14. As a type I transmembrane glycoprotein, expression at the

337 cell surface requires a 5’ signal peptide.

338 To replace EGFP with SBP-ΔLNGFR (generating pNL4-3-ΔEnv-SBP-ΔLNGFR) a synthetic

339 gene fragment (gBlock; Integrated DNA Technologies, IDT) was incorporated into pNL4-3-

340 ΔEnv-EGFP by Gibson assembly between SalI/BsaBI sites (gBlock #1; Supplementary

341 Methods). In this construct, SBP-ΔLNGFR is fused with the endogenous Env signal peptide.

342 To express SBP-ΔLNGFR downstream of nef and a “self-cleaving” Porcine teschovirus-1 2A

343 (P2A) peptide (generating pNL4-3-ΔEnv-EGFP-Nef-P2A-SBP-ΔLNGFR) a gBlock (IDT) was

344 incorporated into pNL4-3-ΔEnv-EGFP by Gibson assembly between HpaI/XhoI sites

345 (gBlock #2; Supplementary Methods). In this construct, SBP-ΔLNGFR is co-translated

346 with codon-optimised Nef (Nef-hu) and includes an exogenous murine immunoglobulin (Ig)

347 signal peptide. SBP-ΔLNGFR was located downstream (rather than upstream) of Nef-hu to

348 avoid disruption of Nef myristoylation by addition of a 5’ proline residue following P2A

349 “cleavage”.

350 To express SBP-ΔLNGFR downstream of nef and an Encephalomyocarditis virus (EMCV)

351 internal ribosome entry site (IRES; generating pNL4-3-ΔEnv-EGFP-Nef-IRES-SBP-

352 ΔLNGFR) a gBlock (IDT) was incorporated into pNL4-3-ΔEnv-EGFP by Gibson assembly

353 between HpaI/XhoI sites (gBlock #3; Supplementary Methods). In this construct, SBP-

354 ΔLNGFR is translated independently of Nef-hu and includes an exogenous murine Ig signal bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

355 peptide. A widely-used replication-competent HIV EGFP reporter virus was previously

356 generated using a similar approach78,79.

357 In all constructs, Nef or Nef-hu expression is mediated by the WT HIV LTR promoter and

358 naturally occurring splice sites. In constructs with a P2A peptide or IRES, the use of codon-

359 optimised Nef-hu minimises homology with the U3 region of the 3’ LTR (overlapped by the

360 endogenous nef sequence) and reduces the risk of recombination.

361 To remove EGFP from constructs with a P2A peptide or IRES, a gBlock (IDT) was

362 incorporated by Gibson assembly between SalI/BsaBI sites (gBlock #4; Supplementary

363 Methods). To avoid generating a truncated protein product fused to the Env signal peptide,

364 the env start codon and other potential out of frame start codons in the vpu ORF were

365 disrupted with point mutations, whilst maintaining the Vpu protein sequence. Redundant env

366 sequence was minimised, without disrupting the Rev response element (RRE).

367 To truncate the U3 region of the 3’ LTR in constructs with a P2A peptide or IRES (with or

368 without EGFP), a gBlock (IDT) was incorporated by Gibson assembly between XhoI/NaeI

369 sites (gBlock #5; Supplementary Methods). Previous studies have shown that, in the

370 presence of an intact nef ORF, the overlap between nef and the U3 region is dispensable for

371 HIV gene expression and replication80.

372 To generate a Vif-deficient HIV-AFMACS molecular clone (pNL4-3-ΔVif-ΔEnv-Nef-P2A-SBP-

373 ΔLNGFR), an AgeI/PflMI restriction fragment was subcloned from pNL4-3-ΔVif-ΔEnv-EGFP3

374 into pNL4-3-ΔEnv-Nef-P2A-SBP-ΔLNGFR (HIV-AFMACS).

375 Where appropriate, additional unique restriction sites were included to facilitate future

376 cloning. All sequences were verified by Sanger sequencing (Source BioScience). The

377 complete HIV-AFMACS (pNL4-3-ΔEnv-Nef-P2A-SBP-ΔLNGFR) sequence is available in the

378 Supplementary Methods.

379 Lentivectors for transgene expression bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

380 For over-expression of codon optimised NL4-3 Vif (Vif-hu) in CEM-T4s, a gBlock (IDT) was

381 incorporated into pHRSIN-SE-P2A-SBP-ΔLNGFR-W14 in place of SBP-ΔLNGFR by Gibson

382 assembly between KpnI/XhoI sites to generate pHRSIN-SE-P2A-Vif-hu-W (in which Vif-hu

383 and EGFP expression are coupled by a P2A peptide).

384 Viral stocks

385 VSVg-pseudotyped NL4-3-ΔEnv-based viral stocks were generated by co-transfection of

386 293T cells with pNL4-3-ΔEnv molecular clones and pMD.G at a ratio of 9:1 (µg) DNA and a

387 DNA:FuGENE 6 ratio of 1 µg:6 µl. Media was changed the next day and viral supernatants

388 harvested and filtered (0.45 µm) at 48 h prior to concentration with LentiX Concentrator

389 (Clontech) and storage at -80 °C.

390 VSVg-pseudotyped pHRSIN-based lentivector stocks were generated by co-transfection of

391 293Ts with lentivector, p8.91 and pMD.G at a ratio of 2:1:1 (μg) DNA and a DNA:FuGENE 6

392 ratio of 1ug:3μl. Viral supernatants were harvested, filtered, concentrated and stored as per

393 NL4-3-ΔEnv-based viral stocks.

394 All viruses were titered by infection/transduction of known numbers of relevant target cells

395 under standard experimental conditions followed by flow cytometry for LNGFR and/or

396 CD4/EGFP at 48 h to identify % infected cells.

397 T cell infections

398 CEM-T4s were infected/transduced by spinoculation at 800 g for 2 h in a non-refrigerated

399 benchtop centrifuge in complete media supplemented with 10mM HEPES. Primary human

400 CD4+ T cells were infected using the same protocol 48 h after activation with CD3/CD28

401 Dynabeads.

402 Antibody-Free Magnetic Cell Sorting (AFMACS)

403 AFMACS-based selection of CEM-T4 or primary human CD4+ T cells using the streptavidin-

404 binding SBP-ΔLNGFR affinity tag was carried out essentially as previously described14. For bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

405 primary T cells, CD3/CD28 Dynabeads were first removed using a DynaMag-2 magnet

406 (Invitrogen). 24 or 48 h post-infection, washed cells were resuspended in incubation buffer

407 (IB; Hank’s balanced salt solution, 2% dialysed FCS, 1x RPMI Amino Acids Solution

408 (Sigma), 2 mM L-glutamine, 2mM EDTA and 10mM HEPES) at 10e7 cells/ml and incubated

409 with Biotin Binder Dynabeads (Invitrogen) at a bead-to-total cell ratio of 4:1 for 30 mins at

410 4oC. Bead-bound cells expressing SBP-ΔLNGFR were selected using a DynaMag-2

411 (Invitrogen), washed to remove uninfected cells, then released from the beads by incubation

412 in complete RPMI with 2 mM biotin for 15 mins at room temperature (RT). Enrichment was

413 routinely assessed by flow cytometry pre- and post-selection.

414 Proteomic analysis

415 Sample preparation

416 For TMT-based whole cell proteomic analysis of primary human CD4+ T cells, resting or

417 activated cells were washed with ice-cold PBS with Ca/Mg pH 7.4 (Sigma) and frozen at -80

418 °C. Samples were lysed, reduced, alkylated, digested and labelled with TMT reagents using

419 the iST-NHS Sample Preparation Kit (PreOmics GmbH) according to the manufacturer’s

420 instructions. Typically, 5e6 resting or 1e6 activated cells were used for each condition.

421 Where indicated, cells were spinoculated with VSVg-pseudotyped viruses at an effective

422 multiplicity of infection (MOI) of approximately 0.5 and subjected to AFMACS-based

423 selection 24 or 48 h post-infection.

424 Off-line high pH reversed-phase (HpRP) peptide fractionation

425 HpRP fractionation was conducted on an Ultimate 3000 UHPLC system (Thermo Scientific)

426 equipped with a 2.1 mm × 15 cm, 1.7 µm Acquity BEH C18 column (Waters, UK). Solvent A

427 was 3% ACN, solvent B was 100% ACN, and solvent C was 200 mM ammonium formate

428 (pH 10). Throughout the analysis C was kept at a constant 10%. The flow rate was 400

429 µL/min and UV was monitored at 280 nm. Samples were loaded in 90% A for 10 min before

430 a gradient elution of 0–10% B over 10 min (curve 3), 10-34% B over 21 min (curve 5), 34- bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

431 50% B over 5 min (curve 5) followed by a 10 min wash with 90% B. 15 s (100 µL) fractions

432 were collected throughout the run. Peptide-containing fractions were orthogonally

433 recombined into 24 fractions (e.g. fractions 1, 25, 49, 73 and 97) and dried in a vacuum

434 centrifuge. Fractions were stored at −80 °C prior to analysis.

435 Mass spectrometry

436 Data were acquired on an Orbitrap Fusion mass spectrometer (Thermo Scientific) coupled to

437 an Ultimate 3000 RSLC nano UHPLC (Thermo Scientific). HpRP fractions were

438 resuspended in 20 µl 5% DMSO 0.5% TFA and 10 uL injected. Fractions were loaded at 10

439 μl/min for 5 min on to an Acclaim PepMap C18 cartridge trap column (300 um × 5 mm, 5 um

440 particle size) in 0.1% TFA. Solvent A was 0.1% FA and solvent B was ACN/0.1% FA. After

441 loading, a linear gradient of 3–32% B over 3 h was used for sample separation over a

442 column of the same stationary phase (75 µm × 50 cm, 2 µm particle size) before washing at

443 90% B and re-equilibration.

444 An SPS/MS3 acquisition was used for all samples and was run as follows. MS1: Quadrupole

445 isolation, 120’000 resolution, 5e5 AGC target, 50 ms maximum injection time, ions injected

446 for all parallelisable time. MS2: Quadrupole isolation at an isolation width of m/z 0.7, CID

447 fragmentation (NCE 35) with the ion trap scanning out in rapid mode from m/z 120, 8e3 AGC

448 target, 70 ms maximum injection time, ions accumulated for all parallelisable time. In

449 synchronous precursor selection mode the top 10 MS2 ions were selected for HCD

450 fragmentation (65NCE) and scanned out in the orbitrap at 50’000 resolution with an AGC

451 target of 2e4 and a maximum accumulation time of 120 ms, ions were not accumulated for

452 all parallelisable time. The entire MS/MS/MS cycle had a target time of 3 s. Dynamic

453 exclusion was set to +/−10 ppm for 90 s, MS2 fragmentation was trigged on precursor ions

454 5e3 counts and above.

455 Data processing and analysis bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

456 Spectra were searched by Mascot within Proteome Discoverer 2.1 in two rounds. The first

457 search was against the UniProt Human reference proteome (26/09/17), the HIV-AFMACS

458 proteome and compendium of common contaminants (GPM). The second search took all

459 unmatched spectra from the first search and searched against the human trEMBL database

460 (Uniprot, 26/09/17). The following search parameters were used. MS1 Tol: 10 ppm. MS2 Tol:

461 0.6 Da. Fixed Mods: Ist-alkylation (+113.084064 Da) (C) and TMT (N-term, K). Var Mods:

462 Oxidation (M). Enzyme: Trypsin (/P). MS3 spectra were used for reporter ion-based

463 quantitation with a most confident centroid tolerance of 20 ppm. PSM FDR was calculated

464 using Mascot percolator and was controlled at 0.01% for ‘high’ confidence PSMs and 0.05%

465 for ‘medium’ confidence PSMs. Normalisation was automated and based on total s/n in each

466 channel. Proteins/peptides satisfying at least a ‘medium’ FDR confidence were taken forth to

467 statistical analysis in R. This consisted of a moderated T-test (Limma) with Benjamini-

468 Hochberg correction for multiple hypotheses to provide a q value for each comparison81.

469 Further data manipulation and general statistical analysis was conducted using Excel,

470 XLSTAT and GraphPad Prism 7.

471 For functional analysis of proteins significantly downregulated or upregulated by WT HIV

472 (q<0.05) in the single time point proteomic experiment (Fig. 3a), enrichment of Gene

473 Ontology (GO) Biological Process (GOTERM_BP_FAT) and Molecular Function

474 (GOTERM_MF_FAT) terms against a background of all proteins quantitated was determined

475 using the Database for Annotation, Visualization and Integrated Discovery (DAVID) 6.8

476 (accessed on 22/7/2018 at https://david.ncifcrf.gov/) with default settings82,83. Human

477 proteins annotated to GO:0016126 (sterol biosynthetic process) were retrieved from AmiGO

478 2 (accessed on 27/7/2018 at http://amigo.geneontology.org/amigo)84. To account for

479 redundancy between annotations, enriched GO terms were visualised using the Enrichment

480 Map 3.1.0 plugin85 for Cytoscape 3.6.1. (downloaded from http://cytoscape.org/)86 with

481 default settings (q value cut-off of 0.1) and sparse-intermediate connectivity. Clusters were bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

482 manually labelled to highlight the prevalent biological functions amongst each set of related

483 annotations.

484 For clustering according to profiles of temporal expression, known accessory protein

485 substrates from Fig. 2c,d and Supplementary Fig. 3b and additional Vpr substrates shown

486 in Fig. 3c were analysed using Cluster 3.0 (downloaded from

487 http://bonsai.hgc.jp/~mdehoon/software/cluster/software.htm)87 and visualised using Java

488 TreeView 1.1.6r4 (downloaded from http://jtreeview.sourceforge.net)88. Only proteins

489 significantly downregulated by WT HIV (q<0.05) in the single time point proteomic

490 experiment (Fig. 3a) were included. Where more than one isoform was quantitated, only the

491 canonical isoform was used (PPP2R5C, Q13362; ZGPAT, Q8N5A5; NUSAP1, Q9BXS6).

492 Data from the time course proteomic experiment (Fig. 2a) were expressed as log2(ratio)s of

493 abundances in experimental (Expt):control (Ctrl) cells for each condition/time point, and

494 range-scaled to highlight patterns of temporal expression relative to the biological response

495 range (minimum-maximum) for each protein. Agglomerative hierarchical clustering was

496 performed using uncentered Pearson correlation and centroid linkage89.

497 Comparison with CEM-T4 T cells

498 To compare results between primary and transformed T cells at a similar depth of proteomic

499 coverage, we re-analysed TMT-labelled peptide eluates from a previous study3 conducted in

500 CEM-T4s spinoculated in triplicate with VSVg-pseudotyped NL4-3-ΔEnv-EGFP WT or ΔVif

501 viruses at an MOI of 1.5. This extended analysis consisted of reinjection of HpRP fractions

502 on longer (3 h) gradients using a higher performance (75 as opposed to 50 cm) analytical

503 column and the MS parameters employed in this study. In total, the new CEM-T4 dataset

504 covered 8,065 proteins, comparable with the datasets from primary T cells described here.

505 Comparisons with other published datasets

506 We have recently characterised 34 new Vpr substrates, together with further, extensive Vpr-

507 dependent changes (downregulated and upregulated proteins) in HIV-infected CEM-T4s39. bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

508 For comparison with this study, a list of 1,388 proteins concordantly regulated by Vpr

509 (q<0.05) in the context of both viral infection and Vpr-bearing virus-like particles was

510 compiled from published datasets. RUNX1 target genes were previously reported to be

511 regulated by Vif at a transcriptional level because of competition for CBFβ binding40. For

512 comparison with this study, a published list of 155 genes with RUNX1-associated regulatory

513 domains exhibiting Vif-dependent differential gene expression in Jurkat T cells after 4 or 6

514 hrs of PMA and PHA treatment was used. Curated lists of ISGs have been previously

515 described90,91. For comparison with this study, a list of 377 ISGs was compiled from these

516 studies. A recent study quantitated 7,761 proteins in FACS-sorted T cells at a single time

517 point 96 hrs post-infection with an R5-tropic, GFP-expressing Nef-deficient virus70. The

518 comparator is GFP negative rather than mock-infected cells (equivalent to SBP-ΔLNGFR

519 negative cells in this study), and the full dataset is not available. For comparison with this

520 study, a published list of 1,551 differentially expressed proteins (q<0.05) was therefore used.

521 Flow cytometry

522 For primary T cells, CD3/CD28 Dynabeads were first removed using a DynaMag-2 magnet

523 (Invitrogen). Typically 2e5 washed cells were incubated for 15 mins in 100 µL PBS with the

524 indicated fluorochrome-conjugated antibody. All steps were performed on ice or at 4°C and

525 stained cells were fixed in PBS/1% paraformaldehyde.

526 Immunoblotting

527 Cells were lysed in PBS/2% SDS supplemented with Halt Protease Inhibitor Cocktail

528 (Thermo Scientific) and Halt Phosphatase Inhibitor Cocktail (Thermo Scientific) for 10 mins

529 at RT. Benzonase (Sigma) was included to reduce lysate viscosity. Post-nuclear

530 supernatants were heated in Laemelli Loading Buffer for 25 mins at 50°C, separated by

531 SDS-PAGE and transferred to Immobilon-P membrane (Millipore). Membranes were blocked

532 in PBS/5% non-fat dried milk (Marvel)/0.2% Tween and probed with the indicated primary

533 antibody overnight at 4°C. Reactive bands were visualised using HRP-conjugated secondary bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

534 antibodies and SuperSignal West Pico or Dura chemiluminescent substrates (Thermo

535 Scientific). Typically 10-20μg total protein was loaded per lane (Pierce BCA Protein Assay

536 kit).

537 Antibodies used in this study

Antibody Source* Catalogue number Application†

BV421-conjugated anti-CD4 BioLegend 317434 FC

PE-conjugated anti-CD4 BioLegend 561843 FC

PE-conjugated anti-tetherin BioLegend 348405 FC

AF647-conjugated anti-LNGFR BioLegend 345114 FC

FITC-conjugated anti-LNGFR BioLegend 345103 FC

Anti-PPP2R5D Abcam ab188323 IB

Anti-Vif NIH ARP #6459 IB

Anti-p24 Abcam ab9071 IB

Anti-Nef NIH ARP 3689 IB

Anti-PTPN22 (D6D1H) CST 14693 IB

Anti-ARID5A GeneTex GTX361940 IB

Anti-FMRP (FMR1) CST 4317 IB

Anti-DPH7 Atlas Antibodies R10023 IB

Anti-tubulin CST 3873 IB bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Anti-β-actin Sigma A5316 IB

Anti-total AURKB CST 3094 IB

Anti-phospho-AURK CST 2914 IB

538 *ARP, AIDS Reagent Program; CST, Cell Signalling Technology.

539 †FC, flow cytometry; IB, immunoblot.

540 bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

541 Acknowledgments

542 This work was supported by the MRC (CSF MR/P008801/1 to NJM), NHSBT (WPA15-02 to

543 NJM), the Wellcome Trust (PRF 210688/Z/18/Z to PJL), the NIHR Cambridge BRC, and a

544 Wellcome Trust Strategic Award to CIMR. The authors thank Dr Reiner Schulte and the

545 CIMR Flow Cytometry Core Facility team, and members of the Lehner laboratory for critical

546 discussion.

547

548 Competing interests

549 The authors declare no competing interests. bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

550 Figures

551 Figure 1: Antibody-free magnetic selection of HIV-infected primary T cells.

552 a, Workflow for AFMACS-based magnetic selection of HIV-infected primary T cells.

553 b, Schematic of HIV-AFMACS provirus (pNL4-3-ΔEnv-Nef-P2A-SBP-ΔLNGFR). For

554 simplicity, reading frames are drawn to match the HXB2 HIV-1 reference genome. Length is

555 indicated in base pairs (bp). The complete sequence is available in the Supplementary

556 Methods. Nef-hu, codon-optimised Nef; RRE, Rev response element; SP, signal peptide.

557 c, Expression of cell surface SBP-∆LNGFR and CD4 on primary T cells 24 or 48 hrs post-

558 infection with HIV-AFMACS. Cells were stained with anti-LNGFR and anti-CD4 antibodies at

559 the indicated time points and analysed by flow cytometry.

560 d,e, Magnetic sorting of HIV-infected (red, LNGFR+, CD4 low) and uninfected (blue,

561 LNGFR-, CD4 high) cells. Cells were separated using AFMACS 48 hrs post-infection with

562 HIV-AFMACS and analysed as in c. Representative (d) and summary (e) data from 6

563 independent experiments in CEM-T4s (triangles) and primary T cells (circles) are shown,

564 with means and 95% confidence intervals (CIs).

565

566 Figure 2: Temporal proteomic analysis of HIV infection in primary T cells.

567 a, Overview of time course proteomic experiment. Control (pale grey,

568 resting/activated/mock) and experimental (dark grey, resting/activated; red, LNGFR+, HIV-

569 infected; blue, LNGFR-, uninfected) cells are indicated for each condition/time point.

570 b, Magnetic sorting of HIV-infected (LNGFR+, selected) cells used for a. Corresponding

571 uninfected (LNGFR-, flow-through) cells are shown in Supplementary Fig. 3a. Cells were

572 separated using AFMACS at the indicated time points post-infection with HIV-AFMACS,

573 stained with anti-LNGFR and anti-CD4 antibodies and analysed by flow cytometry. bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

574 c, Expression profiles of illustrative restriction factors regulated by T cell activation and HIV

575 infection (tetherin) or T cell activation alone (SAMHD1) in cells from a,b. Relative

576 abundances (bars, fraction of maximum) and log2(ratio)s of abundances (lines) in

577 experimental (Expt):control (Ctrl) cells are shown for each condition/time point and coloured

578 as in a (summarised in the legend).

579 d, Expression profiles of illustrative accessory protein targets (CD4, Nef/Vpu; SERINC5, Nef;

580 SNAT1, Vpu; APOBEC3G, Vif; PPP2R5D, Vif; UNG, Vpr) in cells from a,b. Axes, scales and

581 colours are as in c. Expression profiles of other accessory protein targets are shown in

582 Supplementary Fig. 3b.

583 e, Patterns of temporal regulation of Vpr vs other accessory protein (Vif/Nef/Vpu) targets in

584 cells from a,b. Log2(ratio)s of abundances in experimental (Expt):control (Ctrl) cells are

585 shown for 45 accessory protein targets (as in Supplementary Fig. 7b). Colours are as in c,

586 and average profiles (mean, black lozenges/dotted lines) are highlighted for each group of

587 targets.

588

589 Figure 3: Proteins and pathways regulated by HIV in primary T cells.

590 a, Overview of single time point proteomic experiment. HIV-infected (LNGFR+) primary T

591 cells were isolated using AFMACS 48 hrs post-infection with WT (red) or ΔVif (blue) HIV-

592 AFMACS.

593 b, AFMACS-based enrichment of WT (red circles) and ΔVif (blue triangles) HIV-infected

594 (LNGFR+) cells used for a, with means and 95% CIs. Corresponding cells pre-selection are

595 included for each donor/virus (WT, grey circles; ΔVif, grey triangles). Cells were stained with

596 anti-LNGFR and anti-CD4 antibodies and analysed by flow cytometry, with representative

597 data in Supplementary Fig. 4a. bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

598 c, Protein abundances in WT HIV-infected vs mock-infected cells from a. Volcano plots show

599 statistical significance (y axis) vs fold change (x axis) for 8,795 cellular and viral proteins

600 quantitated in cells from all 3 donors (no missing values). Proteins with Benjamini-Hochberg

601 false discovery rate (FDR)-adjusted p values (q values) <0.05 or >0.05 are indicated (FDR

602 threshold of 5%) . Proteins highlighted in each plot are summarised in the legend.

603 Vpr/Vif/Nef/Vpu substrates (green circles) comprise proteins from Fig. 2c,d and

604 Supplementary Fig. 3b, excluding negative controls (SAMHD1, APOBEC3B, SERINC1,

605 HLA-C) and HLA-A/B alleles (different in each donor) but including SMUG1 (not identified in

606 time course proteomic experiment)26 and both quantitated isoforms of PP2R5C (Q13362 and

607 Q13362-4) and ZGPAT (Q8N5A5 and Q8N5A5-2). Additional Vpr substrates (gold circles)

608 and Vpr-dependent changes (gold crosses) comprise recently described direct and indirect

609 Vpr targets39. HIV-dependent changes only identified in primary T cells (red circles and

610 crosses) comprise proteins with q<0.05 either not identified or not concordantly regulated by

611 HIV in CEM-T4s3 (and exclude known accessory protein-dependent changes). Further

612 details on comparator datasets used in this figure are provided in the Methods.

613 d, Enrichment Map85 network-based visualisation of Gene Ontology (GO) functional

614 annotation terms enriched amongst upregulated (red) or downregulated (blue) proteins with

615 q<0.05 in WT HIV-infected vs mock-infected cells from a. Each node represents a GO term,

616 with node size indicating number of annotated proteins, edge thickness representing degree

617 of overlap, and similar GO terms placed close together. Degree of enrichment is mapped to

618 node colour (left side, enriched amongst upregulated proteins; right side, enriched amongst

619 downregulated proteins) as a gradient from white (no enrichment) to red (high enrichment).

620 Highlighted nodes (arrow heads) represent GO terms enriched amongst both upregulated

621 and downregulated proteins.

622 e, Depletion of ARID5A and PTPN22 by HIV. AFMACS-selected (LNGFR+) HIV-infected

623 cells from a (donor A) were lysed in 2% SDS and analysed by immunoblot with anti-ARID5A,

624 anti-PTPN22, anti-Nef and anti-α-tubulin antibodies. Same lysates as Fig. 4d. bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

625

626 Figure 4: Vif-dependent cellular targets in primary T cells.

627 a, Protein abundances in WT HIV-infected vs ΔVif HIV-infected cells from single time point

628 proteomic experiment (Fig. 3a). Statistical significance (y axis) vs fold change (x axis) is

629 shown for 8,795 cellular and viral proteins quantitated in cells from all 3 donors (no missing

630 values). Proteins with Benjamini-Hochberg false discovery rate (FDR)-adjusted p values (q

631 values) <0.05 or >0.05 are indicated (FDR threshold of 5%) . Highlighted groups of

632 differentially regulated proteins are summarised in the legend, including 2 quantitated

633 isoforms of PP2R5C (Q13362 and Q13362-4) and FMR1 (Q06787 and Q06787-2).

634 b, Abundances of Vif-dependent PPP2R5 family and related proteins highlighted in a in

635 mock-infected (grey), WT HIV-infected (red) and ΔVif HIV-infected (green) cells from single

636 time point proteomic experiment (Fig. 3a). Mean abundances (fraction of maximum) with

637 95% CIs are shown. Only the canonical isoform of PPP2R5C (Q13362) is shown.

638 c, Activation of AURKB by Vif. CEM-T4s were mock-transduced or transduced with control

639 (pCtrl) or Vif-expressing (pVif) lentiviruses for 48 hrs (62-78% GFP+), lysed in 2% SDS and

640 analysed by immunoblot with anti-phospho-AURK (T232), anti-total AURKB, anti-Vif and

641 anti-α-tubulin antibodies.

642 d, Vif-dependent activation of AURKB. AFMACS-selected (LNGFR+) HIV-infected cells from

643 Fig. 3a (donor A) were lysed in 2% SDS and analysed by immunoblot with anti-PPP2R5D,

644 anti-phospho-AURK (T232), anti-total AURKB and anti-Vif antibodies. Same lysates as Fig.

645 3e

646 e, Abundances of DPH7 and FMR1 in mock-infected (grey), WT HIV-infected (red) and ΔVif

647 HIV-infected (green) primary T cells from single time point proteomic experiment (Fig. 3a)

648 and CEM-T4s3. Mean abundances (fraction of maximum) with 95% CIs intervals are shown.

649 Only the canonical isoform of FMR1 (Q06787) is shown. bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

650 f, Vif-dependent depletion of DPH7 and FMR1. CEM-T4s were mock-infected or infected

651 with WT or ΔVif HIV-AFMACS viruses for 48 hrs (77-82% LNGFR+ cells), lysed in 2% SDS,

652 and analysed by immunoblot with anti-DPH7, ant-FMR1, anti-Vif and anti-α-tubulin

653 antibodies.

654 g, Depletion of DPH7, FMR1 and PPP2R5D by Vif. CEM-T4s were mock-transduced or

655 transduced with control (pCtrl) or Vif-expressing (pVif) lentiviruses for 48 hrs (86-88% GFP+

656 cells), lysed in 2% SDS and analysed by immunoblot with anti-phospho-AURK, anti-total

657 AURKB, anti-Vif and anti- α-tubulin (loading control) antibodies.

658 bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

659 References

660 1. Weekes, M.P., et al. Quantitative temporal viromics: an approach to investigate host-

661 pathogen interaction. Cell 157, 1460-1472 (2014).

662 2. Matheson, N.J., et al. Cell Surface Proteomic Map of HIV Infection Reveals

663 Antagonism of Amino Acid Metabolism by Vpu and Nef. Cell Host Microbe 18, 409-

664 423 (2015).

665 3. Greenwood, E.J., et al. Temporal proteomic analysis of HIV infection reveals

666 remodelling of the host phosphoproteome by lentiviral Vif variants. Elife 5(2016).

667 4. O'Doherty, U., Swiggard, W.J. & Malim, M.H. Human immunodeficiency virus type 1

668 spinoculation enhances infection through virus binding. J Virol 74, 10074-10080

669 (2000).

670 5. Foley, G.E., et al. Loss of neoplastic properties in vitro. II. Observations on KB

671 sublines. Cancer Res 25, 1254-1261 (1965).

672 6. Popovic, M., Read-Connole, E. & Gallo, R.C. T4 positive human neoplastic cell lines

673 susceptible to and permissive for HTLV-III. Lancet 2, 1472-1473 (1984).

674 7. Zhang, H., et al. Novel single-cell-level phenotypic assay for residual drug

675 susceptibility and reduced replication capacity of drug-resistant human

676 immunodeficiency virus type 1. J Virol 78, 1718-1729 (2004).

677 8. Gillet, J.P., Varma, S. & Gottesman, M.M. The clinical relevance of cancer cell lines.

678 J Natl Cancer Inst 105, 452-458 (2013).

679 9. Sheehy, A.M., Gaddis, N.C., Choi, J.D. & Malim, M.H. Isolation of a human gene that

680 inhibits HIV-1 infection and is suppressed by the viral Vif protein. Nature 418, 646-

681 650 (2002).

682 10. Neil, S.J., Zang, T. & Bieniasz, P.D. Tetherin inhibits retrovirus release and is

683 antagonized by HIV-1 Vpu. Nature 451, 425-430 (2008).

684 11. Usami, Y., Wu, Y. & Gottlinger, H.G. SERINC3 and SERINC5 restrict HIV-1 infectivity

685 and are counteracted by Nef. Nature 526, 218-223 (2015). bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

686 12. Rosa, A., et al. HIV-1 Nef promotes infection by excluding SERINC5 from virion

687 incorporation. Nature 526, 212-217 (2015).

688 13. Goujon, C., et al. Human MX2 is an interferon-induced post-entry inhibitor of HIV-1

689 infection. Nature 502, 559-562 (2013).

690 14. Matheson, N.J., Peden, A.A. & Lehner, P.J. Antibody-free magnetic cell sorting of

691 genetically modified primary human CD4+ T cells by one-step streptavidin affinity

692 purification. PLoS One 9, e111437 (2014).

693 15. Keefe, A.D., Wilson, D.S., Seelig, B. & Szostak, J.W. One-step purification of

694 recombinant proteins using a nanomolar-affinity streptavidin-binding peptide, the

695 SBP-Tag. Protein Expr Purif 23, 440-446 (2001).

696 16. Ruggieri, L., et al. Cell-surface marking of CD(34+)-restricted phenotypes of human

697 hematopoietic progenitor cells by retrovirus-mediated gene transfer. Hum Gene Ther

698 8, 1611-1623 (1997).

699 17. Kumar, M., Keller, B., Makalou, N. & Sutton, R.E. Systematic determination of the

700 packaging limit of lentiviral vectors. Hum Gene Ther 12, 1893-1905 (2001).

701 18. Guy, B., et al. HIV F/3' orf encodes a phosphorylated GTP-binding protein

702 resembling an oncogene product. Nature 330, 266-269 (1987).

703 19. Willey, R.L., Maldarelli, F., Martin, M.A. & Strebel, K. Human immunodeficiency virus

704 type 1 Vpu protein induces rapid degradation of CD4. J Virol 66, 7193-7200 (1992).

705 20. Gheysen, D., et al. Assembly and release of HIV-1 precursor Pr55gag virus-like

706 particles from recombinant baculovirus-infected insect cells. Cell 59, 103-112 (1989).

707 21. Klotman, M.E., et al. Kinetics of expression of multiply spliced RNA in early human

708 immunodeficiency virus type 1 infection of lymphocytes and monocytes. Proc Natl

709 Acad Sci U S A 88, 5011-5015 (1991).

710 22. Geiger, R., et al. L-Arginine Modulates T Cell Metabolism and Enhances Survival and

711 Anti-tumor Activity. Cell 167, 829-842 e813 (2016).

712 23. Hrecka, K., et al. Vpx relieves inhibition of HIV-1 infection of macrophages mediated

713 by the SAMHD1 protein. Nature 474, 658-661 (2011). bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

714 24. Laguette, N., et al. SAMHD1 is the dendritic- and myeloid-cell-specific HIV-1

715 restriction factor counteracted by Vpx. Nature 474, 654-657 (2011).

716 25. Lim, E.S., et al. The ability of primate lentiviruses to degrade the monocyte restriction

717 factor SAMHD1 preceded the birth of the viral accessory protein Vpx. Cell Host

718 Microbe 11, 194-204 (2012).

719 26. Schrofelbauer, B., Yu, Q., Zeitlin, S.G. & Landau, N.R. Human immunodeficiency

720 virus type 1 Vpr induces the degradation of the UNG and SMUG uracil-DNA

721 glycosylases. J Virol 79, 10978-10987 (2005).

722 27. Lahouassa, H., et al. HIV-1 Vpr degrades the HLTF DNA translocase in T cells and

723 macrophages. Proc Natl Acad Sci U S A (2016).

724 28. Hrecka, K., et al. HIV-1 and HIV-2 exhibit divergent interactions with HLTF and

725 UNG2 DNA repair proteins. Proc Natl Acad Sci U S A 113, E3921-3930 (2016).

726 29. Maudet, C., et al. HIV-1 Vpr induces the degradation of ZIP and sZIP, adaptors of the

727 NuRD chromatin remodeling complex, by hijacking DCAF1/VprBP. PLoS One 8,

728 e77320 (2013).

729 30. Lapek, J.D., Jr., Lewinski, M.K., Wozniak, J.M., Guatelli, J. & Gonzalez, D.J.

730 Quantitative Temporal Viromics of an Inducible HIV-1 Model Yields Insight to Global

731 Host Targets and Phospho-Dynamics Associated with Protein Vpr. Mol Cell

732 Proteomics 16, 1447-1461 (2017).

733 31. Zhou, X., DeLucia, M. & Ahn, J. SLX4-SLX1 Protein-independent Down-regulation of

734 MUS81-EME1 Protein by HIV-1 Viral Protein R (Vpr). J Biol Chem 291, 16936-16947

735 (2016).

736 32. Romani, B., Shaykh Baygloo, N., Aghasadeghi, M.R. & Allahbakhshi, E. HIV-1 Vpr

737 Protein Enhances Proteasomal Degradation of MCM10 DNA Replication Factor

738 through the Cul4-DDB1[VprBP] E3 Ubiquitin Ligase to Induce G2/M Cell Cycle

739 Arrest. J Biol Chem 290, 17380-17389 (2015).

740 33. Lv, L., et al. Vpr Targets TET2 for Degradation by CRL4(VprBP) E3 Ligase to Sustain

741 IL-6 Expression and Enhance HIV-1 Replication. Mol Cell 70, 961-970 e965 (2018). bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

742 34. Schwartz, O., Marechal, V., Le Gall, S., Lemonnier, F. & Heard, J.M. Endocytosis of

743 major histocompatibility complex class I molecules is induced by the HIV-1 Nef

744 protein. Nat Med 2, 338-342 (1996).

745 35. Cohen, G.B., et al. The selective downregulation of class I major histocompatibility

746 complex proteins by HIV-1 protects HIV-infected cells from NK cells. Immunity 10,

747 661-671 (1999).

748 36. Apps, R., et al. HIV-1 Vpu Mediates HLA-C Downregulation. Cell Host Microbe 19,

749 686-695 (2016).

750 37. Karn, J. & Stoltzfus, C.M. Transcriptional and posttranscriptional regulation of HIV-1

751 gene expression. Cold Spring Harb Perspect Med 2, a006916 (2012).

752 38. Ye, C.J., et al. Intersection of population variation and autoimmunity genetics in

753 human T cell activation. Science 345, 1254665 (2014).

754 39. Greenwood, E.J., et al. Vpr drives massive cellular proteome remodelling in HIV-1

755 infection. bioRxiv (2018).

756 40. Kim, D.Y., et al. CBFbeta stabilizes HIV Vif to counteract APOBEC3 at the expense

757 of RUNX1 target gene expression. Mol Cell 49, 632-644 (2013).

758 41. Jowett, J.B., et al. The human immunodeficiency virus type 1 vpr gene arrests

759 infected T cells in the G2 + M phase of the cell cycle. J Virol 69, 6304-6313 (1995).

760 42. He, J., et al. Human immunodeficiency virus type 1 viral protein R (Vpr) arrests cells

761 in the G2 phase of the cell cycle by inhibiting p34cdc2 activity. J Virol 69, 6705-6711

762 (1995).

763 43. Re, F., Braaten, D., Franke, E.K. & Luban, J. Human immunodeficiency virus type 1

764 Vpr arrests the cell cycle in G2 by inhibiting the activation of p34cdc2-cyclin B. J Virol

765 69, 6859-6864 (1995).

766 44. Rogel, M.E., Wu, L.I. & Emerman, M. The human immunodeficiency virus type 1 vpr

767 gene prevents cell proliferation during chronic infection. J Virol 69, 882-888 (1995). bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

768 45. Poon, B., et al. Human immunodeficiency virus type 1 vpr gene induces phenotypic

769 effects similar to those of the DNA alkylating agent, nitrogen mustard. J Virol 71,

770 3961-3971 (1997).

771 46. Roshal, M., Kim, B., Zhu, Y., Nghiem, P. & Planelles, V. Activation of the ATR-

772 mediated DNA damage response by the HIV-1 viral protein R. J Biol Chem 278,

773 25879-25886 (2003).

774 47. Laguette, N., et al. Premature activation of the SLX4 complex by Vpr promotes G2/M

775 arrest and escape from innate immune sensing. Cell 156, 134-145 (2014).

776 48. Harris, R.S., et al. DNA deamination mediates innate immunity to retroviral infection.

777 Cell 113, 803-809 (2003).

778 49. van 't Wout, A.B., et al. Nef induces multiple genes involved in cholesterol synthesis

779 and uptake in human immunodeficiency virus type 1-infected T cells. J Virol 79,

780 10053-10058 (2005).

781 50. Shrivastava, S., Trivedi, J. & Mitra, D. Gene expression profiling reveals Nef induced

782 deregulation of lipid metabolism in HIV-1 infected T cells. Biochem Biophys Res

783 Commun 472, 169-174 (2016).

784 51. Vermeire, J., et al. HIV Triggers a cGAS-Dependent, Vpu- and Vpr-Regulated Type I

785 Interferon Response in CD4(+) T Cells. Cell Rep 17, 413-424 (2016).

786 52. Cohen, E.A., Dehni, G., Sodroski, J.G. & Haseltine, W.A. Human immunodeficiency

787 virus vpr product is a virion-associated regulatory protein. J Virol 64, 3097-3099

788 (1990).

789 53. Yu, X.F., Matsuda, M., Essex, M. & Lee, T.H. Open reading frame vpr of simian

790 immunodeficiency virus encodes a virion-associated protein. J Virol 64, 5688-5693

791 (1990).

792 54. Yuan, X., Matsuda, Z., Matsuda, M., Essex, M. & Lee, T.H. Human

793 immunodeficiency virus vpr gene encodes a virion-associated protein. AIDS Res

794 Hum Retroviruses 6, 1265-1271 (1990). bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

795 55. Thomas, J.A., Ott, D.E. & Gorelick, R.J. Efficiency of human immunodeficiency virus

796 type 1 postentry infection processes: evidence against disproportionate numbers of

797 defective virions. J Virol 81, 4367-4370 (2007).

798 56. Yasui, Y., et al. Autophosphorylation of a newly identified site of Aurora-B is

799 indispensable for cytokinesis. J Biol Chem 279, 12997-13003 (2004).

800 57. Meppelink, A., Kabeche, L., Vromans, M.J., Compton, D.A. & Lens, S.M. Shugoshin-

801 1 balances Aurora B kinase activity via PP2A to promote chromosome bi-orientation.

802 Cell Rep 11, 508-515 (2015).

803 58. Liu, Y.C., et al. The PPFIA1-PP2A protein complex promotes trafficking of Kif7 to the

804 ciliary tip and Hedgehog signaling. Science signaling 7, ra117 (2014).

805 59. Tang, Z., et al. PP2A is required for centromeric localization of Sgo1 and proper

806 chromosome segregation. Dev Cell 10, 575-585 (2006).

807 60. Xu, Z., et al. Structure and function of the PP2A-shugoshin interaction. Mol Cell 35,

808 426-441 (2009).

809 61. Plouffe, B.D., Murthy, S.K. & Lewis, L.H. Fundamentals and application of magnetic

810 particles in cell isolation and enrichment: a review. Rep Prog Phys 78, 016601

811 (2015).

812 62. Stanciu, L.A., Shute, J., Holgate, S.T. & Djukanovic, R. Production of IL-8 and IL-4 by

813 positively and negatively selected CD4+ and CD8+ human T cells following a four-

814 step cell separation method including magnetic cell sorting (MACS). J Immunol

815 Methods 189, 107-115 (1996).

816 63. Bernard, F., et al. Ex vivo isolation protocols differentially affect the phenotype of

817 human CD4+ T cells. J Immunol Methods 271, 99-106 (2002).

818 64. Shah, A.H., et al. Degranulation of natural killer cells following interaction with HIV-1-

819 infected cells is hindered by downmodulation of NTB-A by Vpu. Cell Host Microbe 8,

820 397-409 (2010).

821 65. Bolduan, S., et al. HIV-1 Vpu affects the anterograde transport and the glycosylation

822 pattern of NTB-A. Virology 440, 190-203 (2013). bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

823 66. Ramirez, P.W., et al. Downmodulation of CCR7 by HIV-1 Vpu results in impaired

824 migration and chemotactic signaling within CD4(+) T cells. Cell Rep 7, 2019-2030

825 (2014).

826 67. Sugden, S.M., Pham, T.N. & Cohen, E.A. HIV-1 Vpu Downmodulates ICAM-1

827 Expression, Resulting in Decreased Killing of Infected CD4(+) T Cells by NK Cells. J

828 Virol 91(2017).

829 68. Chan, E.Y., et al. Dynamic host energetics and cytoskeletal proteomes in human

830 immunodeficiency virus type 1-infected human primary CD4 cells: analysis by

831 multiplexed label-free mass spectrometry. J Virol 83, 9283-9295 (2009).

832 69. Nemeth, J., et al. In Vivo and in Vitro Proteome Analysis of Human

833 Immunodeficiency Virus (HIV)-1-infected, Human CD4(+) T Cells. Mol Cell

834 Proteomics 16, S108-S123 (2017).

835 70. Kuo, H.H., et al. Anti-apoptotic Protein BIRC5 Maintains Survival of HIV-1-Infected

836 CD4(+) T Cells. Immunity 48, 1183-1194 e1185 (2018).

837 71. Melikyan, G.B. HIV entry: a game of hide-and-fuse? Curr Opin Virol 4, 1-7 (2014).

838 72. Arthos, J., et al. HIV-1 envelope protein binds to and signals through integrin

839 alpha4beta7, the gut mucosal homing receptor for peripheral T cells. Nat Immunol 9,

840 301-309 (2008).

841 73. Pan, Q., Rong, L., Zhao, X. & Liang, C. Fragile X mental retardation protein restricts

842 replication of human immunodeficiency virus type 1. Virology 387, 127-135 (2009).

843 74. Higa, M., et al. Regulation of inflammatory responses by dynamic subcellular

844 localization of RNA-binding protein Arid5a. Proc Natl Acad Sci U S A 115, E1214-

845 E1220 (2018).

846 75. Hasegawa, K., et al. PEST domain-enriched tyrosine phosphatase (PEP) regulation

847 of effector/memory T cells. Science 303, 685-689 (2004).

848 76. Carette, J.E., et al. Haploid genetic screens in human cells identify host factors used

849 by pathogens. Science 326, 1231-1235 (2009). bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

850 77. Ortiz, P.A., Ulloque, R., Kihara, G.K., Zheng, H. & Kinzy, T.G. Translation elongation

851 factor 2 anticodon mimicry domain mutants affect fidelity and diphtheria toxin

852 resistance. J Biol Chem 281, 32639-32648 (2006).

853 78. Schindler, M., et al. Down-modulation of mature major histocompatibility complex

854 class II and up-regulation of invariant chain cell surface expression are well-

855 conserved functions of human and simian immunodeficiency virus nef alleles. J Virol

856 77, 10548-10556 (2003).

857 79. Schindler, M., et al. Nef-mediated suppression of T cell activation was lost in a

858 lentiviral lineage that gave rise to HIV-1. Cell 125, 1055-1067 (2006).

859 80. Munch, J., et al. The role of upstream U3 sequences in HIV-1 replication and CD4+ T

860 cell depletion in human lymphoid tissue ex vivo. Virology 341, 313-320 (2005).

861 81. Schwammle, V., Leon, I.R. & Jensen, O.N. Assessment and improvement of

862 statistical tools for comparative proteomics analysis of sparse data sets with few

863 experimental replicates. J Proteome Res 12, 3874-3883 (2013).

864 82. Huang da, W., Sherman, B.T. & Lempicki, R.A. Systematic and integrative analysis

865 of large gene lists using DAVID bioinformatics resources. Nat Protoc 4, 44-57 (2009).

866 83. Huang da, W., Sherman, B.T. & Lempicki, R.A. Bioinformatics enrichment tools:

867 paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids

868 Res 37, 1-13 (2009).

869 84. Carbon, S., et al. AmiGO: online access to ontology and annotation data.

870 Bioinformatics 25, 288-289 (2009).

871 85. Merico, D., Isserlin, R., Stueker, O., Emili, A. & Bader, G.D. Enrichment map: a

872 network-based method for gene-set enrichment visualization and interpretation.

873 PLoS One 5, e13984 (2010).

874 86. Shannon, P., et al. Cytoscape: a software environment for integrated models of

875 biomolecular interaction networks. Genome Res 13, 2498-2504 (2003).

876 87. de Hoon, M.J., Imoto, S., Nolan, J. & Miyano, S. Open source clustering software.

877 Bioinformatics 20, 1453-1454 (2004). bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

878 88. Saldanha, A.J. Java Treeview--extensible visualization of microarray data.

879 Bioinformatics 20, 3246-3248 (2004).

880 89. Eisen, M.B., Spellman, P.T., Brown, P.O. & Botstein, D. Cluster analysis and display

881 of genome-wide expression patterns. Proc Natl Acad Sci U S A 95, 14863-14868

882 (1998).

883 90. Schoggins, J.W., et al. Pan-viral specificity of IFN-induced genes reveals new roles

884 for cGAS in innate immunity. Nature 505, 691-695 (2014).

885 91. Schoggins, J.W., et al. A diverse range of gene products are effectors of the type I

886 interferon antiviral response. Nature 472, 481-485 (2011).

887 bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Fig. 1.

CD4 From 24 hrs post-infection a Bead

Primary human Release selected CD4+ T-cells SBP-ΔLNGFR cells with biotin Infected

SBP-ΔLNGFR + CD4 low

Activate with CD3/CD28

Uninfected

SBP-ΔLNGFR − CD4 high

Infect with Incubate with Wash to remove HIV-AFMACS magnetic beads unbound cells b c Primary T cells

HIV-AFMACS (pNL4-3-ΔEnv-Nef-P2A-SBP-ΔLNGFR) 24h 48h

0 2000 4000 6000 8000 9657 bp

vif tat LNGFR+ LNGFR+ tat vpu RRE 29.4% 38.4% 5’ LTR gag rev 3’ LTR pol vpr rev SP P2A CD4 Nef-hu SBP―ΔLNGFR LNGFR

d Primary T cells e CEM-T4s Primary T cells

100 Input Selected Flow-through 80 LNGFR+ LNGFR+ LNGFR+ 31.1% 96.5% 5.2% 60

40 LNGFR+ % 20

CD4 0

LNGFR Input Selected

Flow-through bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Fig. 2. a b Input Selected LNGFR- LNGFR+ 8.48% 91.5%

24h infection - infection infection - infection infection with CD4+ cell T – AFMACS – AFMACS LNGFR+

AFMACS virus - AFMACS 19.1% Day 5 Day 48h post Day 3 – Day HIV Day 4 Day 24h post Day 1 – Day isolation and activation

LNGFR- LNGFR+ 1.46% 98.5%

48h

LNGFR+ 41.1% CD4 24h_Neg 24h_Pos 48h_Mock 48h_Pos 24h_Mock Activated_2 48h_Neg Activated_1 Resting_2 Resting_1

LNGFR LNGFR

Digest proteins and label with TMT reporters

Ctrl Mix peptides, fractionate and analyse by LC/MS3 Expt_Resting/Activated Expt_LNGFR_Pos Expt_LNGFR_Neg c Tetherin SAMHD1

1 0 1 0 /Ctrl) /Ctrl) Expt Expt Abundance Abundance Log2( Log2( (fraction max.) (fraction 0 -2.5 max.) (fraction 0 -2.5 - - 24h 48h 24h 48h Resting Activated (time post- Resting Activated (time post- infection) infection) d e CD4 SNAT1 Vpr targets 0 /Ctrl) Expt

-2.5 SERINC5 PPP2R5D Log2( Vif, Nef and Vpu targets

0 /Ctrl) Expt APOBEC3G UNG -2.5 Log2( /Ctrl Expt Abundance Resting 24h_Pos 48h_Pos 24h_Neg 48h_Neg Activated Time course bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Fig. 3. Donor A Donor B Donor C a b AFMACS 48h post-infection

100

80 V if_Pos V if_Pos V if_Pos Δ Δ Δ WT_Pos WT_Pos Mock Mock WT_Pos Mock 60 WT 40 ΔVif LNGFR+ % Digest proteins and label with TMT reporters 20

0

Input Mix peptides, fractionate and analyse by LC/MS3 Selected

c Accessory protein targets Other Vpr-dependent changes Changes not seen in T cell lines

Decreased Increased Decreased Increased Decreased Increased 30 30 30 Nef-P2A Gagpol SBP-ΔLNGFR Gag 20 20 Rev 20 ARID5A

PTPN22

10 10 10 FDR 5% Vif FDR 5% FDR 5% og10(p)

-L 0 0 0 -6 -4 -2 0 2 4 6 -6 -4 -2 0 2 4 6 -6 -4 -2 0 2 4 6 Log2(WT HIV/Mock)

Vpr/Vif/Nef/Vpu substrates Additional Vpr substrates HIV-dependent changes only q<0.05 HIV proteins Other Vpr-dependent changes identified in primary T cells q>0.05

d Upregulated by WT HIV Downregulated by WT HIV e

Lipid metabolism Cell cycle regulation Primary T cells

Histone modification and chromatin organisation Vif WT Δ Mock Organic acid Nucleic acid catabolism binding and ARID5A DNA damage and repair regulation PTPN22 RNA processing Nef Lymphocyte activation and α-Tubulin DNA recombination Cytidine deamination DNA modification and viral replication

Amino acid PP2A Helicase transport activity activity bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Fig. 4. a Vif-dependent changes b PPP2R5A PPP2R5C PPP2R5D Decreased Increased 30 1 1 1

0 0 0 Vif Vif Vif WT WT WT Δ Δ Δ Mock Mock 20 Mock PPP2R5E PPP2R1A PPP2R1B FDR 5% 1 1 1 Vif 10 0 0 0 Vif Vif Vif WT WT WT Δ Δ Δ Mock Mock Mock

SGO1 og10(p) PPP2CA PPFIA1

-L 0 1 1 1 -3 -2 -1 0 1 2 3 Log2(WT/ΔVif) Abundance (fraction (fraction Abundance max.) 0 0 0 Vif Vif Vif WT WT WT Δ Δ Δ Mock Mock

APOBEC3 family PPP2R5 family DPH7, FMR1 Mock PPP2R1A/B, PPP2CA, PPFIA1, SGO1 q>0.05 c e CEM-T4s DPH7 FMR1 1 1

pVif Primary T cells pCtrl Mock 0 0 Vif Vif WT p-AURKB WT Δ Δ Mock Mock

AURKB DPH7 FMR 1 1 Vif CEM-T4s Abundance (fraction (fraction Abundance max.) 0 0 α-Tubulin Vif Vif WT WT Δ Δ Mock Mock d f g Primary T cells CEM-T4s CEM-T4s Vif Vif pVif pCtrl Mock WT WT Δ Mock Δ Mock

DPH7 DPH7 PPP2R5D

FMR1 FMR1 p-AURKB

AURKB Vif PPP2R5D

Vif α-Tubulin Vif

β-Actin bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

1 Supplementary Figures

2 Supplementary Figure 1: Initial screen of SBP-ΔLNGFR-expressing HIV viruses.

3 a,b, Expression of GFP (pNL4-3-ΔEnv-EGFP only) or cell surface SBP-∆LNGFR (all other

4 constructs) and CD4 on CEM-T4s 48 hrs post-infection with indicated pNL4-3-ΔEnv-based

5 viruses (a). Cells were stained with anti-LNGFR and anti-CD4 antibodies and analysed by

6 flow cytometry. gRNA length is shown for each construct, and compared with functional viral

7 titre derived from the % LNGFR+ cells (b). The 3 viruses selected for further testing are

8 named/highlighted (bold text). pNL4-3-ΔEnv-EGFP was included as a control.

9

10 Supplementary Figure 2: Time course evaluation of selected SBP-ΔLNGFR-

11 expressing HIV viruses.

12 a, Expression of GFP (pNL4-3-ΔEnv-EGFP only) or cell surface SBP-∆LNGFR (all other

13 constructs) and CD4 or tetherin in CEM-T4s 24 or 48 hrs post-infection with HIV-AFMACS.

14 Cells were stained with anti-LNGFR and anti-CD4 or anti-tetherin antibodies at the indicated

15 time points and analysed by flow cytometry. Schematics indicate the location/setting of SBP-

16 ΔLNGFR within each provirus, with ORFs and non-coding features coloured as in Fig. 1b

17 (but with Nef in black and EGFP in green). For simplicity, reading frames are drawn to match

18 the HXB2 HIV-1 reference genome. The final HIV-AFMACS virus is highlighted (pNL4-3-

19 ΔEnv-Nef-P2A-SBP-ΔLNGFR, bold text). pNL4-3-ΔEnv-EGFP was included as a control.

20

21 Supplementary Figure 3: Additional controls for time course proteomic experiment.

22 a, Uninfected (LNGFR-, flow-through) cells used for time course proteomic experiment (Fig.

23 2a). Corresponding HIV-infected (LNGFR+, selected) cells are shown in Fig. 2b. Cells were

24 separated using AFMACS at the indicated time points post-infection with HIV-AFMACS,

25 stained with anti-LNGFR and anti-CD4 antibodies and analysed by flow cytometry. bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

26 b,c, Expression profiles of accessory protein targets (b; APOBEC3 and PPP2R5 family

27 proteins, Vif; HLA-A/B alleles, Nef; other downregulated proteins, Vpr) and viral proteins (c)

28 from time course proteomic experiment (Fig. 2a). Axes, scales and colours are as in Fig. 2c

29 with the exception of APOBEC3C (range of log2(Expt/Ctrl) 0 to -4 rather than 0 to -2.5).

30 SERINC1, APOBEC3B and HLA-C alleles are included as controls. Only HLA alleles

31 quantitated by >1 unique peptide are shown. Since viral proteins are not expressed in

32 control (Ctrl) cells, log2(ratio)s of abundances in Expt:Ctrl cells are not shown. Only

33 canonical isoforms of PPP2R5C (Q13362) and ZGPAT (Q8N5A5) are shown.

34

35 Supplementary Figure 4: Additional controls for single time point proteomic

36 experiment.

37 a, AFMACS-based enrichment of HIV-infected (LNGFR+) cells used for single time point

38 proteomic experiment (Fig. 3a). Cells were stained with anti-LNGFR and anti-CD4

39 antibodies and analysed by flow cytometry. Representative data is shown from donor B, with

40 summary data in Fig. 3b.

41 b, Principle component analysis of mock-infected (grey), WT (red) and ΔVif (blue) HIV-

42 infected samples from single time point proteomic experiment (Fig. 3a). The correlation

43 matrix was analysed for 8,795 cellular and viral proteins quantitated in cells from all 3 donors

44 (no missing values). Visually-identical results were obtained with/without viral proteins.

45

46 Supplementary Figure 5: Comparisons of HIV-dependent changes in primary T cells

47 with other datasets.

48 a, Between-sample coefficients of variation (%) for all protein abundances from primary T

49 cells (single time point proteomic experiment, Fig. 3a) or CEM-T4s1. Boxplots show median, bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

50 interquartile range and Tukey whiskers for mock-infected (grey) and WT HIV-infected (red)

51 cells.

52 b, Protein abundances in WT HIV-infected vs mock-infected cells from primary T cells (x

53 axis; single time point proteomic experiment, Fig. 3a) or CEM-T4s (y axis)1. Fold changes

54 are compared for proteins with q<0.05 in both datasets.

55 c, Protein abundances in WT HIV-infected vs mock-infected cells from single time point

56 proteomic experiment (Fig. 3a), with details for volcano plots as in Fig. 3c. Proteins

57 highlighted in each plot are summarised in the legend. Additional Nef/Vpu targets (green

58 circles) comprise (alphabetical order, grouped by study): CCR72,

59 CD37/CD53/CD63/CD81/CD823,4, CD99 (2 quantitated gene products, H7C2F2 and

60 P14209)/PLP2/UBE2L65, ICAM1/36, NTB-A7, PVR8,9, SELL10. RUNX1 target genes (blue

61 circles) were previously reported to be regulated by Vif at a transcriptional level because of

62 competition for CBFβ binding11. ISGs (purple circles) were previously curated from published

63 microarray data sets from IFN-treated cells12,13.

64 d, Mechanisms underlying significant HIV-dependent proteins changes (q<0.05) in single

65 time point proteomic experiment (Fig. 3a). Established accessory protein targets from Fig.

66 3c (left panel, 22 proteins) and c (left panel, 3 proteins) and additional Vpr substrates and

67 Vpr-dependent changes from Fig. 3c (middle panel) are shown. Amongst the 32 additional

68 Vpr substrates depleted in primary T cells, 26 had q<0.05. Remaining proteins are

69 categorised based on the identification plus/minus HIV-dependent regulation in a previous,

70 similar experiment using CEM-T4s14. Of 131 proteins not identified in the comparator CEM-

71 T4 dataset, one is known Vpr target SMUG115 and 4 other proteins were found to be

72 regulated by Vpr in other experiments using CEM-T4s (ZNF512B, ATXN7, CLUH and

73 SLC39A3)14.

74 Further details on comparator datasets used in this figure are provided in the Methods.

75 bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

76 Supplementary Figure 6: Pathways regulated by HIV in primary T cells.

77 a, Gene Ontology (GO) functional annotations enriched amongst downregulated (blue) or

78 upregulated (red) proteins with q<0.05 in WT HIV-infected vs mock-infected cells from single

79 time point proteomic experiment (Fig. 3a). The 10 most enriched terms (ranked by p value)

80 are shown in each case, with a Benjamini-Hochberg FDR threshold of 5% indicated (dashed

81 line).

82 b, Protein abundances in WT HIV-infected vs mock-infected cells from single time point

83 proteomic experiment (Fig. 3a), with details for volcano plot as in Fig. 3c. 57 proteins

84 annotated with the GO term “sterol biosynthetic process” (GO:0016126) are highlighted in

85 red. Amongst these, 15 proteins are regulated by Vpr in CEM-T4s (circles)14.

86

87 Supplementary Figure 7: Temporal clustering of HIV accessory protein targets.

88 a, Abundances of ARID5A and PTPN22 in mock-infected (grey), WT HIV-infected (red) and

89 ΔVif HIV-infected (green) primary T cells from single time point proteomic experiment (Fig.

90 3a). Relative abundances (fraction of maximum) with 95% confidence intervals are shown.

91 b, Hierarchical cluster analysis of 45 accessory protein targets according to profiles of

92 temporal expression from time course proteomic experiment (Fig. 2a). Vpr (gold) vs other

93 accessory protein (Vif/Nef/Vpu; green) targets are highlighted. The heatmap shows range-

94 scaled log2(ratio)s of abundances in experimental (Expt):control (Ctrl) cells for each

95 condition/time point. Unscaled data for the same proteins are shown in Fig. 2e.

96 c, d, Expression profiles of ARID5A, PTPN22 and AIRD5B from time course proteomic

97 experiment (Fig. 2a). Axes, scales and colours are as in Fig. 2c.

98

99 Supplementary Figure 8: Comparison of this study with Kuo et al., 201816. bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

100 a, Protein abundances in WT HIV-infected vs mock-infected cells from this study (x axis;

101 single time point proteomic experiment, Fig. 3a) or Kuo et al., 201816 (y axis). Fold changes

102 are compared for proteins with q<0.05 in both datasets.

103 b, Protein abundances in WT HIV-infected vs mock-infected cells from single time point

104 proteomic experiment (Fig. 3a), with details for volcano plot as in Fig. 3c. Proteins with

105 q<0.05 not reported by Kuo et al., 201816 are highlighted in red.

106 bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

107 Supplementary Tables

108 Supplementary Table 1: Functional proteomic atlas of HIV-infection in primary human

109 CD4+ T cells.

110 Interactive spreadsheet enabling generation of temporal profiles of protein abundance for

111 any quantitated genes of interest (“Gene search and plots” worksheet). Time course data

112 (cells from Fig. 2a) are presented as in Fig. 2c, with relative protein abundances (fraction of

113 maximum) for each condition depicted by bars, and log2(ratio)s of protein abundances in

114 paired experimental/control cells from each condition/time point depicted by lines (grey,

115 resting/activated; red, LNGFR+, infected; blue, LNGFR-, uninfected). Single time point data

116 (cells from Fig. 3a) are presented as in Fig. 4b, with relative protein abundances (fraction of

117 maximum, mean plus 95% CIs) for each condition depicted by bars (grey, mock; red, WT

118 HIV; green, ΔVif HIV). The number of unique peptides is shown for each protein/experiment,

119 with most confidence reserved for proteins with values >1. For the single time point

120 experiment, p values (unadjusted) and q values (Benjamini-Hochberg FDR-adjusted) are

121 shown (highlighted in gold if <0.05). The complete (unfiltered) proteomic datasets (“Time

122 course dataset” and “Single time point dataset” worksheets) are also included.

123 bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

124 Supplementary Methods

125 gBlock sequences

126 gBlock #1

127 tgtttatccatttcagaattgggtgtcgacatagcagaataggcgttactcgacagaggagagcaagaaatggagccagtaga

128 tcctagactagagccctggaagcatccaggaagtcagcctaaaactgcttgtaccaattgctattgtaaaaagtgttgctttc

129 attgccaagtttgtttcatgacaaaagccttaggcatctcctatggcaggaagaagcggagacagcgacgaagagctcatcag

130 aacagtcagactcatcaagcttctctatcaaagcagtaagtagtacatgtaatgcaacctataatagtagcaatagtagcatt

131 agtagtagcaataataatagcaatagttgtgtggtccatagtaatcatagaatataggaaaatattaagacaaagaaaaatag

132 acaggttaattgatagactaatagaaagagcagaagacagtggcaatgagagtgaaggagaagtatcagcacttgtggagatg

133 ggggtggaaatggggcaccatgctccttgggatattgatgatctgtagtgctacagaaaaattgtgggtcacagtctattatg

134 gggtacctgtcgcggccgcagggtcagggatggacgaaaagacaaccggatggcgaggaggacacgtggtcgagggactggca

135 ggagagctggaacagctgcgggctagactggaacaccatcctcagggacagcgagagcccggaagtggaaaggaagcctgccc

136 cacagggctgtacactcattctggagaatgctgtaaagcttgtaacctgggagagggagtggcacagccatgcggagccaatc

137 agactgtgtgcgagccttgtctggactccgtcacattctctgatgtggtcagtgccacagaaccttgcaagccatgtactgag

138 tgcgtgggcctgcagtctatgagtgctccttgtgtggaggctgacgatgcagtctgccggtgtgcatacggatactatcagga

139 cgagactaccggcagatgtgaagcttgcagggtgtgtgaggcaggctcagggctggtctttagctgccaggataaacagaaca

140 ccgtgtgcgaggaatgtcctgacgggacatatagcgatgaggccaatcacgtggacccctgcctgccttgtactgtgtgcgag

141 gataccgaaaggcagctgcgcgaatgtaccagatgggcagacgccgagtgcgaggaaatcccagggcgatggattactcggtc

142 caccccccctgaaggatcagacagcaccgcaccatctacacaggagccagaagcaccaccagagcaggatctgatcgcctcca

143 ccgtggctggcgtggtcacaactgtcatggggagctctcagccagtggtcacccggggcaccacagataacctgattcccgtg

144 tattgctccatcctggcagccgtcgtcgtcggactggtggcatacatcgccttcaagcggtgagctagccgtaacgatgttca

145 tcaaatattactgggctgctat

146 gBlock #2

147 tcaggaactaaagaatagtgctgttaacttgctcaatgccacagccatagcagtagctgaggggacagatagggttatagaag

148 tattacaagcagcttatagagctattcgccacatacctagaagaataagacagggcttggaaaggattttgctataagatggg

149 ggggaaatggtctaagtcttccgtgataggctggcccgctgttcgagaaaggatgcgccgcgccgaaccagcggccgatggcg

150 taggtgccgtatctagggaccttgagaagcacggtgccataacgtcttctaacaccgcagccaacaatgcggcatgtgcgtgg

151 ttggaggcacaggaagaggaggaggtaggattcccggtaacgccacaagtacccttgcgccctatgacctacaaggccgccgt

152 ggacctgagccatttcctgaaggagaaaggtggcttggaagggcttatccactcccaaagacgccaggatattcttgaccttt

153 ggatctaccacacacagggttactttcccgactggcaaaattatacgcccggtcccggtgtacggtatcctcttacttttggg

154 tggtgctataaactcgtgccggtggagccagataaggtagaggaggccaataaaggtgagaacacaagtttgcttcaccccgt bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

155 ttcccttcatggtatggacgacccagaacgggaagttcttgaatggcggttcgatagcaggttggcatttcatcatgtcgcac

156 gggagcttcaccccgagtactttaaaaattgtgcggccgcaggaagcggagctactaacttcagcctgctgaagcaggctgga

157 gacgtggaggagaaccctggacctatgggctggtcatgtatcattctgtttctggtcgcaaccgcaactggagtgcattcaca

158 ggtgcagctcagcgctgggtcagggatggacgaaaagacaaccggatggcgaggaggacacgtggtcgagggactggcaggag

159 agctggaacagctgcgggctagactggaacaccatcctcagggacagcgagagcccggaagtggagcgatcgcgaaggaagcc

160 tgccccacagggctgtacactcattctggagaatgctgtaaagcttgtaacctgggagagggagtggcacagccatgcggagc

161 caatcagactgtgtgcgagccttgtctggactccgtcacattctctgatgtggtcagtgccacagaaccttgcaagccatgta

162 ctgagtgcgtgggcctgcagtctatgagtgctccttgtgtggaggctgacgatgcagtctgccggtgtgcatacggatactat

163 caggacgagactaccggcagatgtgaagcttgcagggtgtgtgaggcaggctcagggctggtctttagctgccaggataaaca

164 gaacaccgtgtgcgaggaatgtcctgacgggacatatagcgatgaggccaatcacgtggacccctgcctgccttgtactgtgt

165 gcgaggataccgaaaggcagctgcgcgaatgtaccagatgggcagacgccgagtgcgaggaaatcccagggcgatggattact

166 cggtccaccccccctgaaggatcagacagcaccgcaccatctacacaggagccagaagcaccaccagagcaggatctgatcgc

167 ctccaccgtggctggcgtggtcacaactgtcatggggagctctcagccagtggtcacccggggcaccacagataacctgattc

168 ccgtgtattgctccatcctggcagccgtcgtcgtcggactggtggcatacatcgccttcaagcggtgactcgagacctagaaa

169 aacatggagcaatca

170 gBlock #3

171 ggagtcaggaactaaagaatagtgctgttaacttgctcaatgccacagccatagcagtagctgaggggacagatagggttata

172 gaagtattacaagcagcttatagagctattcgccacatacctagaagaataagacagggcttggaaaggattttgctataaga

173 tgggggggaaatggtctaagtcttccgtgataggctggcccgctgttcgagaaaggatgcgccgcgccgaaccagcggccgat

174 ggcgtaggtgccgtatctagggaccttgagaagcacggtgccataacgtcttctaacaccgcagccaacaatgcggcatgtgc

175 gtggttggaggcacaggaagaggaggaggtaggattcccggtaacgccacaagtacccttgcgccctatgacctacaaggccg

176 ccgtggacctgagccatttcctgaaggagaaaggtggcttggaagggcttatccactcccaaagacgccaggatattcttgac

177 ctttggatctaccacacacagggttactttcccgactggcaaaattatacgcccggtcccggtgtacggtatcctcttacttt

178 tgggtggtgctataaactcgtgccggtggagccagataaggtagaggaggccaataaaggtgagaacacaagtttgcttcacc

179 ccgtttcccttcatggtatggacgacccagaacgggaagttcttgaatggcggttcgatagcaggttggcatttcatcatgtc

180 gcacgggagcttcaccccgagtactttaaaaattgttgatctagacgcccccccctaacgttactggccgaagccgcttggaa

181 taaggccggtgtgcgtttgtctatatgttattttccaccatattgccgtcttttggcaatgtgagggcccggaaacctggccc

182 tgtcttcttgacgagcattcctaggggtctttcccctctcgccaaaggaatgcaaggtctgttgaatgtcgtgaaggaagcag

183 ttcctctggaagcttcttgaagacaaacaacgtctgtagcgaccctttgcaggcagcggaaccccccacctggcgacaggtgc

184 ctctgcggccaaaagccacgtgtataagatacacctgcaaaggcggcacaaccccagtgccacgttgtgagttggatagttgt

185 ggaaagagtcaaatggctctcctcaagcgtattcaacaaggggctgaaggatgcccagaaggtaccccattgtatgggatctg

186 atctggggcctcggtgcacatgctttacatgtgtttagtcgaggttaaaaaacgtctaggccccccgaaccacggggacgtgg

187 ttttcctttgaaaaacacgatgataatatgggctggtcatgtatcattctgtttctggtcgcaaccgcaactggagtgcattc bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

188 acaggtgcagctcagcgctgggtcagggatggacgaaaagacaaccggatggcgaggaggacacgtggtcgagggactggcag

189 gagagctggaacagctgcgggctagactggaacaccatcctcagggacagcgagagcccggaagtggagcgatcgcgaaggaa

190 gcctgccccacagggctgtacactcattctggagaatgctgtaaagcttgtaacctgggagagggagtggcacagccatgcgg

191 agccaatcagactgtgtgcgagccttgtctggactccgtcacattctctgatgtggtcagtgccacagaaccttgcaagccat

192 gtactgagtgcgtgggcctgcagtctatgagtgctccttgtgtggaggctgacgatgcagtctgccggtgtgcatacggatac

193 tatcaggacgagactaccggcagatgtgaagcttgcagggtgtgtgaggcaggctcagggctggtctttagctgccaggataa

194 acagaacaccgtgtgcgaggaatgtcctgacgggacatatagcgatgaggccaatcacgtggacccctgcctgccttgtactg

195 tgtgcgaggataccgaaaggcagctgcgcgaatgtaccagatgggcagacgccgagtgcgaggaaatcccagggcgatggatt

196 actcggtccaccccccctgaaggatcagacagcaccgcaccatctacacaggagccagaagcaccaccagagcaggatctgat

197 cgcctccaccgtggctggcgtggtcacaactgtcatggggagctctcagccagtggtcacccggggcaccacagataacctga

198 ttcccgtgtattgctccatcctggcagccgtcgtcgtcggactggtggcatacatcgccttcaagcggtgactcgagacctag

199 aaaaacatggagcaatcacaag

200 gBlock #4

201 tgtttatccatttcagaattgggtgtcgacatagcagaataggcgttactcgacagaggagagcaagaaatggagccagtaga

202 tcctagactagagccctggaagcatccaggaagtcagcctaaaactgcttgtaccaattgctattgtaaaaagtgttgctttc

203 attgccaagtttgtttcatgacaaaagccttaggcatctcctatggcaggaagaagcggagacagcgacgaagagctcatcag

204 aacagtcagactcatcaagcttctctatcaaagcagtaagtagtacatgtaatgcaacctataatagtagcaatagtagcatt

205 agtagtagcaataataatagcaatagttgtgtggtccatagtaatcatagaatataggaaaatattaagacaaagaaaaatag

206 acaggttaattgatagactaatagaaagagcagaagacagtggcaacgagagtgaaggagaagtatcagcacttgtggagatg

207 ggggtggaaatggggcaccacgctccttgggatattgacgatctgtaggctagccgtaacgatgttcatcaaatattactggg

208 ctgctat

209 gBlock #5

210 atacatcgccttcaagcggtgactcgagtagccactttttaaaagaaaaggggggactggaagggctaattcactcccaaaga

211 agacaagatatcccatccggagtacttcaagaactgctgacatcgagcttgctacaagggactttccgctggggactttccag

212 ggaggcgtggcctgggcgggactggggagtggcgagccctcagatgctgcatataagcagctgctttttgcctgtactgggtc

213 tctctggttagaccagatctgagcctgggagctctctggctaactagggaacccactgcttaagcctcaataaagcttgcctt

214 gagtgcttcaagtagtgtgtgcccgtctgttgtgtgactctggtaactagagatccctcagacccttttagtcagtgtggaaa

215 atctctagcacccaggaggtagaggttgcagtgagccaagatcgcgccactgcattccagcctgggcaagaaaacaagactgt

216 ctaaaataataataataagttaagggtattaaatatatttatacatggaggtcataaaaatatatatatttgggctgggcgca

217 gtggctcacacctgcgcccggccctttgggaggccgaggcaggtggatcacctgagtttgggagttccagaccagcctgacca

218 acatggagaaaccccttctctgtgtatttttagtagattttattttatgtgtattttattcacaggtatttctggaaaactga

219 aactgtttttcctctactctgataccacaagaatcatcagcacagaggaagacttctgtgatcaaatgtggtgggagagggag bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

220 gttttcaccagcacatgagcagtcagttctgccgcagactcggcgggtgtccttcggttcagttccaacaccgcctgcctgga

221 gagaggtcagaccacagggtgagggctcagtccccaagacataaacacccaagacataaacaccccaggtccaccccgcctgc

222 tgcccaggcagagccgattcaccaagacgggaattaggatagagaaagagtaagtcacacagagccggctgtgcgggagaacg

223 gagttcta

224 HIV-AFMACS (pNL4-3-ΔEnv-Nef-P2A-SBP-ΔLNGFR) complete sequence

225 tggaagggctaatttggtcccaaaaaagacaagagatccttgatctgtggatctaccacacacaaggctacttccctgattgg

226 cagaactacacaccagggccagggatcagatatccactgacctttggatggtgcttcaagttagtaccagttgaaccagagca

227 agtagaagaggccaatgaaggagagaacaacagcttgttacaccctatgagccagcatgggatggaggacccggagggagaag

228 tattagtgtggaagtttgacagcctcctagcatttcgtcacatggcccgagagctgcatccggagtactacaaagactgctga

229 catcgagctttctacaagggactttccgctggggactttccagggaggtgtggcctgggcgggactggggagtggcgagccct

230 cagatgctacatataagcagctgctttttgcctgtactgggtctctctggttagaccagatctgagcctgggagctctctggc

231 taactagggaacccactgcttaagcctcaataaagcttgccttgagtgctcaaagtagtgtgtgcccgtctgttgtgtgactc

232 tggtaactagagatccctcagacccttttagtcagtgtggaaaatctctagcagtggcgcccgaacagggacttgaaagcgaa

233 agtaaagccagaggagatctctcgacgcaggactcggcttgctgaagcgcgcacggcaagaggcgaggggcggcgactggtga

234 gtacgccaaaaattttgactagcggaggctagaaggagagagatgggtgcgagagcgtcggtattaagcgggggagaattaga

235 taaatgggaaaaaattcggttaaggccagggggaaagaaacaatataaactaaaacatatagtatgggcaagcagggagctag

236 aacgattcgcagttaatcctggccttttagagacatcagaaggctgtagacaaatactgggacagctacaaccatcccttcag

237 acaggatcagaagaacttagatcattatataatacaatagcagtcctctattgtgtgcatcaaaggatagatgtaaaagacac

238 caaggaagccttagataagatagaggaagagcaaaacaaaagtaagaaaaaggcacagcaagcagcagctgacacaggaaaca

239 acagccaggtcagccaaaattaccctatagtgcagaacctccaggggcaaatggtacatcaggccatatcacctagaacttta

240 aatgcatgggtaaaagtagtagaagagaaggctttcagcccagaagtaatacccatgttttcagcattatcagaaggagccac

241 cccacaagatttaaataccatgctaaacacagtggggggacatcaagcagccatgcaaatgttaaaagagaccatcaatgagg

242 aagctgcagaatgggatagattgcatccagtgcatgcagggcctattgcaccaggccagatgagagaaccaaggggaagtgac

243 atagcaggaactactagtacccttcaggaacaaataggatggatgacacataatccacctatcccagtaggagaaatctataa

244 aagatggataatcctgggattaaataaaatagtaagaatgtatagccctaccagcattctggacataagacaaggaccaaagg

245 aaccctttagagactatgtagaccgattctataaaactctaagagccgagcaagcttcacaagaggtaaaaaattggatgaca

246 gaaaccttgttggtccaaaatgcgaacccagattgtaagactattttaaaagcattgggaccaggagcgacactagaagaaat

247 gatgacagcatgtcagggagtggggggacccggccataaagcaagagttttggctgaagcaatgagccaagtaacaaatccag

248 ctaccataatgatacagaaaggcaattttaggaaccaaagaaagactgttaagtgtttcaattgtggcaaagaagggcacata

249 gccaaaaattgcagggcccctaggaaaaagggctgttggaaatgtggaaaggaaggacaccaaatgaaagattgtactgagag

250 acaggctaattttttagggaagatctggccttcccacaagggaaggccagggaattttcttcagagcagaccagagccaacag

251 ccccaccagaagagagcttcaggtttggggaagagacaacaactccctctcagaagcaggagccgatagacaaggaactgtat

252 cctttagcttccctcagatcactctttggcagcgacccctcgtcacaataaagataggggggcaattaaaggaagctctatta bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

253 gatacaggagcagatgatacagtattagaagaaatgaatttgccaggaagatggaaaccaaaaatgatagggggaattggagg

254 ttttatcaaagtaagacagtatgatcagatactcatagaaatctgcggacataaagctataggtacagtattagtaggaccta

255 cacctgtcaacataattggaagaaatctgttgactcagattggctgcactttaaattttcccattagtcctattgagactgta

256 ccagtaaaattaaagccaggaatggatggcccaaaagttaaacaatggccattgacagaagaaaaaataaaagcattagtaga

257 aatttgtacagaaatggaaaaggaaggaaaaatttcaaaaattgggcctgaaaatccatacaatactccagtatttgccataa

258 agaaaaaagacagtactaaatggagaaaattagtagatttcagagaacttaataagagaactcaagatttctgggaagttcaa

259 ttaggaataccacatcctgcagggttaaaacagaaaaaatcagtaacagtactggatgtgggcgatgcatatttttcagttcc

260 cttagataaagacttcaggaagtatactgcatttaccatacctagtataaacaatgagacaccagggattagatatcagtaca

261 atgtgcttccacagggatggaaaggatcaccagcaatattccagtgtagcatgacaaaaatcttagagccttttagaaaacaa

262 aatccagacatagtcatctatcaatacatggatgatttgtatgtaggatctgacttagaaatagggcagcatagaacaaaaat

263 agaggaactgagacaacatctgttgaggtggggatttaccacaccagacaaaaaacatcagaaagaacctccattcctttgga

264 tgggttatgaactccatcctgataaatggacagtacagcctatagtgctgccagaaaaggacagctggactgtcaatgacata

265 cagaaattagtgggaaaattgaattgggcaagtcagatttatgcagggattaaagtaaggcaattatgtaaacttcttagggg

266 aaccaaagcactaacagaagtagtaccactaacagaagaagcagagctagaactggcagaaaacagggagattctaaaagaac

267 cggtacatggagtgtattatgacccatcaaaagacttaatagcagaaatacagaagcaggggcaaggccaatggacatatcaa

268 atttatcaagagccatttaaaaatctgaaaacaggaaagtatgcaagaatgaagggtgcccacactaatgatgtgaaacaatt

269 aacagaggcagtacaaaaaatagccacagaaagcatagtaatatggggaaagactcctaaatttaaattacccatacaaaagg

270 aaacatgggaagcatggtggacagagtattggcaagccacctggattcctgagtgggagtttgtcaatacccctcccttagtg

271 aagttatggtaccagttagagaaagaacccataataggagcagaaactttctatgtagatggggcagccaatagggaaactaa

272 attaggaaaagcaggatatgtaactgacagaggaagacaaaaagttgtccccctaacggacacaacaaatcagaagactgagt

273 tacaagcaattcatctagctttgcaggattcgggattagaagtaaacatagtgacagactcacaatatgcattgggaatcatt

274 caagcacaaccagataagagtgaatcagagttagtcagtcaaataatagagcagttaataaaaaaggaaaaagtctacctggc

275 atgggtaccagcacacaaaggaattggaggaaatgaacaagtagataaattggtcagtgctggaatcaggaaagtactatttt

276 tagatggaatagataaggcccaagaagaacatgagaaatatcacagtaattggagagcaatggctagtgattttaacctacca

277 cctgtagtagcaaaagaaatagtagccagctgtgataaatgtcagctaaaaggggaagccatgcatggacaagtagactgtag

278 cccaggaatatggcagctagattgtacacatttagaaggaaaagttatcttggtagcagttcatgtagccagtggatatatag

279 aagcagaagtaattccagcagagacagggcaagaaacagcatacttcctcttaaaattagcaggaagatggccagtaaaaaca

280 gtacatacagacaatggcagcaatttcaccagtactacagttaaggccgcctgttggtgggcggggatcaagcaggaatttgg

281 cattccctacaatccccaaagtcaaggagtaatagaatctatgaataaagaattaaagaaaattataggacaggtaagagatc

282 aggctgaacatcttaagacagcagtacaaatggcagtattcatccacaattttaaaagaaaaggggggattggggggtacagt

283 gcaggggaaagaatagtagacataatagcaacagacatacaaactaaagaattacaaaaacaaattacaaaaattcaaaattt

284 tcgggtttattacagggacagcagagatccagtttggaaaggaccagcaaagctcctctggaaaggtgaaggggcagtagtaa

285 tacaagataatagtgacataaaagtagtgccaagaagaaaagcaaagatcatcagggattatggaaaacagatggcaggtgat

286 gattgtgtggcaagtagacaggatgaggattaacacatggaaaagattagtaaaacaccatatgtatatttcaaggaaagcta bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

287 aggactggttttatagacatcactatgaaagtactaatccaaaaataagttcagaagtacacatcccactaggggatgctaaa

288 ttagtaataacaacatattggggtctgcatacaggagaaagagactggcatttgggtcagggagtctccatagaatggaggaa

289 aaagagatatagcacacaagtagaccctgacctagcagaccaactaattcatctgcactattttgattgtttttcagaatctg

290 ctataagaaataccatattaggacgtatagttagtcctaggtgtgaatatcaagcaggacataacaaggtaggatctctacag

291 tacttggcactagcagcattaataaaaccaaaacagataaagccacctttgcctagtgttaggaaactgacagaggacagatg

292 gaacaagccccagaagaccaagggccacagagggagccatacaatgaatggacactagagcttttagaggaacttaagagtga

293 agctgttagacattttcctaggatatggctccataacttaggacaacatatctatgaaacttacggggatacttgggcaggag

294 tggaagccataataagaattctgcaacaactgctgtttatccatttcagaattgggtgtcgacatagcagaataggcgttact

295 cgacagaggagagcaagaaatggagccagtagatcctagactagagccctggaagcatccaggaagtcagcctaaaactgctt

296 gtaccaattgctattgtaaaaagtgttgctttcattgccaagtttgtttcatgacaaaagccttaggcatctcctatggcagg

297 aagaagcggagacagcgacgaagagctcatcagaacagtcagactcatcaagcttctctatcaaagcagtaagtagtacatgt

298 aatgcaacctataatagtagcaatagtagcattagtagtagcaataataatagcaatagttgtgtggtccatagtaatcatag

299 aatataggaaaatattaagacaaagaaaaatagacaggttaattgatagactaatagaaagagcagaagacagtggcaacgag

300 agtgaaggagaagtatcagcacttgtggagatgggggtggaaatggggcaccacgctccttgggatattgacgatctgtaggc

301 tagccgtaacgatgttcatcaaatattactgggctgctattaacaagagatggtggtaataacaacaatgggtccgagatctt

302 cagacctggaggaggcgatatgagggacaattggagaagtgaattatataaatataaagtagtaaaaattgaaccattaggag

303 tagcacccaccaaggcaaagagaagagtggtgcagagagaaaaaagagcagtgggaataggagctttgttccttgggttcttg

304 ggagcagcaggaagcactatgggcgcagcgtcaatgacgctgacggtacaggccagacaattattgtctgatatagtgcagca

305 gcagaacaatttgctgagggctattgaggcgcaacagcatctgttgcaactcacagtctggggcatcaaacagctccaggcaa

306 gaatcctggctgtggaaagatacctaaaggatcaacagctcctggggatttggggttgctctggaaaactcatttgcaccact

307 gctgtgccttggaatgctagttggagtaataaatctctggaacagatttggaataacatgacctggatggagtgggacagaga

308 aattaacaattacacaagcttaatacactccttaattgaagaatcgcaaaaccagcaagaaaagaatgaacaagaattattgg

309 aattagataaatgggcaagtttgtggaattggtttaacataacaaattggctgtggtatataaaattattcataatgatagta

310 ggaggcttggtaggtttaagaatagtttttgctgtactttctatagtgaatagagttaggcagggatattcaccattatcgtt

311 tcagacccacctcccaatcccgaggggacccgacaggcccgaaggaatagaagaagaaggtggagagagagacagagacagat

312 ccattcgattagtgaacggatccttagcacttatctgggacgatctgcggagcctgtgcctcttcagctaccaccgcttgaga

313 gacttactcttgattgtaacgaggattgtggaacttctgggacgcagggggtgggaagccctcaaatattggtggaatctcct

314 acagtattggagtcaggaactaaagaatagtgctgttaacttgctcaatgccacagccatagcagtagctgaggggacagata

315 gggttatagaagtattacaagcagcttatagagctattcgccacatacctagaagaataagacagggcttggaaaggattttg

316 ctataagatgggggggaaatggtctaagtcttccgtgataggctggcccgctgttcgagaaaggatgcgccgcgccgaaccag

317 cggccgatggcgtaggtgccgtatctagggaccttgagaagcacggtgccataacgtcttctaacaccgcagccaacaatgcg

318 gcatgtgcgtggttggaggcacaggaagaggaggaggtaggattcccggtaacgccacaagtacccttgcgccctatgaccta

319 caaggccgccgtggacctgagccatttcctgaaggagaaaggtggcttggaagggcttatccactcccaaagacgccaggata

320 ttcttgacctttggatctaccacacacagggttactttcccgactggcaaaattatacgcccggtcccggtgtacggtatcct bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

321 cttacttttgggtggtgctataaactcgtgccggtggagccagataaggtagaggaggccaataaaggtgagaacacaagttt

322 gcttcaccccgtttcccttcatggtatggacgacccagaacgggaagttcttgaatggcggttcgatagcaggttggcatttc

323 atcatgtcgcacgggagcttcaccccgagtactttaaaaattgtgcggccgcaggaagcggagctactaacttcagcctgctg

324 aagcaggctggagacgtggaggagaaccctggacctatgggctggtcatgtatcattctgtttctggtcgcaaccgcaactgg

325 agtgcattcacaggtgcagctcagcgctgggtcagggatggacgaaaagacaaccggatggcgaggaggacacgtggtcgagg

326 gactggcaggagagctggaacagctgcgggctagactggaacaccatcctcagggacagcgagagcccggaagtggagcgatc

327 gcgaaggaagcctgccccacagggctgtacactcattctggagaatgctgtaaagcttgtaacctgggagagggagtggcaca

328 gccatgcggagccaatcagactgtgtgcgagccttgtctggactccgtcacattctctgatgtggtcagtgccacagaacctt

329 gcaagccatgtactgagtgcgtgggcctgcagtctatgagtgctccttgtgtggaggctgacgatgcagtctgccggtgtgca

330 tacggatactatcaggacgagactaccggcagatgtgaagcttgcagggtgtgtgaggcaggctcagggctggtctttagctg

331 ccaggataaacagaacaccgtgtgcgaggaatgtcctgacgggacatatagcgatgaggccaatcacgtggacccctgcctgc

332 cttgtactgtgtgcgaggataccgaaaggcagctgcgcgaatgtaccagatgggcagacgccgagtgcgaggaaatcccaggg

333 cgatggattactcggtccaccccccctgaaggatcagacagcaccgcaccatctacacaggagccagaagcaccaccagagca

334 ggatctgatcgcctccaccgtggctggcgtggtcacaactgtcatggggagctctcagccagtggtcacccggggcaccacag

335 ataacctgattcccgtgtattgctccatcctggcagccgtcgtcgtcggactggtggcatacatcgccttcaagcggtgactc

336 gagacctagaaaaacatggagcaatcacaagtagcaatacagcagctaacaatgctgcttgtgcctggctagaagcacaagag

337 gaggaagaggtgggttttccagtcacacctcaggtacctttaagaccaatgacttacaaggcagctgtagatcttagccactt

338 tttaaaagaaaaggggggactggaagggctaattcactcccaaagaagacaagatatccttgatctgtggatctaccacacac

339 aaggctacttccctgattggcagaactacacaccagggccaggggtcagatatccactgacctttggatggtgctacaagcta

340 gtaccagttgagccagataaggtagaagaggccaataaaggagagaacaccagcttgttacaccctgtgagcctgcatggaat

341 ggatgaccctgagagagaagtgttagagtggaggtttgacagccgcctagcatttcatcacgtggcccgagagctgcatccgg

342 agtacttcaagaactgctgacatcgagcttgctacaagggactttccgctggggactttccagggaggcgtggcctgggcggg

343 actggggagtggcgagccctcagatgctgcatataagcagctgctttttgcctgtactgggtctctctggttagaccagatct

344 gagcctgggagctctctggctaactagggaacccactgcttaagcctcaataaagcttgccttgagtgcttcaagtagtgtgt

345 gcccgtctgttgtgtgactctggtaactagagatccctcagacccttttagtcagtgtggaaaatctctagcacccaggaggt

346 agaggttgcagtgagccaagatcgcgccactgcattccagcctgggcaagaaaacaagactgtctaaaataataataataagt

347 taagggtattaaatatatttatacatggaggtcataaaaatatatatatttgggctgggcgcagtggctcacacctgcgcccg

348 gccctttgggaggccgaggcaggtggatcacctgagtttgggagttccagaccagcctgaccaacatggagaaaccccttctc

349 tgtgtatttttagtagattttattttatgtgtattttattcacaggtatttctggaaaactgaaactgtttttcctctactct

350 gataccacaagaatcatcagcacagaggaagacttctgtgatcaaatgtggtgggagagggaggttttcaccagcacatgagc

351 agtcagttctgccgcagactcggcgggtgtccttcggttcagttccaacaccgcctgcctggagagaggtcagaccacagggt

352 gagggctcagtccccaagacataaacacccaagacataaacacccaacaggtccaccccgcctgctgcccaggcagagccgat

353 tcaccaagacgggaattaggatagagaaagagtaagtcacacagagccggctgtgcgggagaacggagttctattatgactca

354 aatcagtctccccaagcattcggggatcagagtttttaaggataacttagtgtgtagggggccagtgagttggagatgaaagc bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

355 gtagggagtcgaaggtgtccttttgcgccgagtcagttcctgggtgggggccacaagatcggatgagccagtttatcaatccg

356 ggggtgccagctgatccatggagtgcagggtctgcaaaatatctcaagcactgattgatcttaggttttacaatagtgatgtt

357 accccaggaacaatttggggaaggtcagaatcttgtagcctgtagctgcatgactcctaaaccataatttcttttttgttttt

358 ttttttttatttttgagacagggtctcactctgtcacctaggctggagtgcagtggtgcaatcacagctcactgcagcctcaa

359 cgtcgtaagctcaagcgatcctcccacctcagcctgcctggtagctgagactacaagcgacgccccagttaatttttgtattt

360 ttggtagaggcagcgttttgccgtgtggccctggctggtctcgaactcctgggctcaagtgatccagcctcagcctcccaaag

361 tgctgggacaaccggggccagtcactgcacctggccctaaaccataatttctaatcttttggctaatttgttagtcctacaaa

362 ggcagtctagtccccaggcaaaaagggggtttgtttcgggaaagggctgttactgtctttgtttcaaactataaactaagttc

363 ctcctaaacttagttcggcctacacccaggaatgaacaaggagagcttggaggttagaagcacgatggaattggttaggtcag

364 atctctttcactgtctgagttataattttgcaatggtggttcaaagactgcccgcttctgacaccagtcgctgcattaatgaa

365 tcggccaacgcgcggggagaggcggtttgcgtattgggcgctcttccgcttcctcgctcactgactcgctgcgctcggtcgtt

366 cggctgcggcgagcggtatcagctcactcaaaggcggtaatacggttatccacagaatcaggggataacgcaggaaagaacat

367 gtgagcaaaaggccagcaaaaggccaggaaccgtaaaaaggccgcgttgctggcgtttttccataggctccgcccccctgacg

368 agcatcacaaaaatcgacgctcaagtcagaggtggcgaaacccgacaggactataaagataccaggcgtttccccctggaagc

369 tccctcgtgcgctctcctgttccgaccctgccgcttaccggatacctgtccgcctttctcccttcgggaagcgtggcgctttc

370 tcatagctcacgctgtaggtatctcagttcggtgtaggtcgttcgctccaagctgggctgtgtgcacgaaccccccgttcagc

371 ccgaccgctgcgccttatccggtaactatcgtcttgagtccaacccggtaagacacgacttatcgccactggcagcagccact

372 ggtaacaggattagcagagcgaggtatgtaggcggtgctacagagttcttgaagtggtggcctaactacggctacactagaag

373 aacagtatttggtatctgcgctctgctgaagccagttaccttcggaaaaagagttggtagctcttgatccggcaaacaaacca

374 ccgctggtagcggtggtttttttgtttgcaagcagcagattacgcgcagaaaaaaaggatctcaagaagatcctttgatcttt

375 tctacggggtctgacgctcagtggaacgaaaactcacgttaagggattttggtcatgagattatcaaaaaggatcttcaccta

376 gatccttttaaattaaaaatgaagttttaaatcaatctaaagtatatatgagtaaacttggtctgacagttaccaatgcttaa

377 tcagtgaggcacctatctcagcgatctgtctatttcgttcatccatagttgcctgactccccgtcgtgtagataactacgata

378 cgggagggcttaccatctggccccagtgctgcaatgataccgcgagacccacgctcaccggctccagatttatcagcaataaa

379 ccagccagccggaagggccgagcgcagaagtggtcctgcaactttatccgcctccatccagtctattaattgttgccgggaag

380 ctagagtaagtagttcgccagttaatagtttgcgcaacgttgttgccattgctacaggcatcgtggtgtcacgctcgtcgttt

381 ggtatggcttcattcagctccggttcccaacgatcaaggcgagttacatgatcccccatgttgtgcaaaaaagcggttagctc

382 cttcggtcctccgatcgttgtcagaagtaagttggccgcagtgttatcactcatggttatggcagcactgcataattctctta

383 ctgtcatgccatccgtaagatgcttttctgtgactggtgagtactcaaccaagtcattctgagaatagtgtatgcggcgaccg

384 agttgctcttgcccggcgtcaatacgggataataccgcgccacatagcagaactttaaaagtgctcatcattggaaaacgttc

385 ttcggggcgaaaactctcaaggatcttaccgctgttgagatccagttcgatgtaacccactcgtgcacccaactgatcttcag

386 catcttttactttcaccagcgtttctgggtgagcaaaaacaggaaggcaaaatgccgcaaaaaagggaataagggcgacacgg

387 aaatgttgaatactcatactcttcctttttcaatattattgaagcatttatcagggttattgtctcatgagcggatacatatt

388 tgaatgtatttagaaaaataaacaaataggggttccgcgcacatttccccgaaaagtgccacctgacgtctaagaaaccatta bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

389 ttatcatgacattaacctataaaaataggcgtatcacgaggccctttcgtctcgcgcgtttcggtgatgacggtgaaaacctc

390 tgacacatgcagctcccggagacggtcacagcttgtctgtaagcggatgccgggagcagacaagcccgtcagggcgcgtcagc

391 gggtgttggcgggtgtcggggctggcttaactatgcggcatcagagcagattgtactgagagtgcaccatatgcggtgtgaaa

392 taccgcacagatgcgtaaggagaaaataccgcatcaggcgccattcgccattcaggctgcgcaactgttgggaagggcgatcg

393 gtgcgggcctcttcgctattacgccaggggaggcagagattgcagtaagctgagatcgcagcactgcactccagcctgggcga

394 cagagtaagactctgtctcaaaaataaaataaataaatcaatcagatattccaatcttttcctttatttatttatttattttc

395 tattttggaaacacagtccttccttattccagaattacacatatattctatttttctttatatgctccagttttttttagacc

396 ttcacctgaaatgtgtgtatacaaaatctaggccagtccagcagagcctaaaggtaaaaaataaaataataaaaaataaataa

397 aatctagctcactccttcacatcaaaatggagatacagctgttagcattaaataccaaataacccatcttgtcctcaataatt

398 ttaagcgcctctctccaccacatctaactcctgtcaaaggcatgtgccccttccgggcgctctgctgtgctgccaaccaactg

399 gcatgtggactctgcagggtccctaactgccaagccccacagtgtgccctgaggctgccccttccttctagcggctgccccca

400 ctcggctttgctttccctagtttcagttacttgcgttcagccaaggtctgaaactaggtgcgcacagagcggtaagactgcga

401 gagaaagagaccagctttacagggggtttatcacagtgcaccctgacagtcgtcagcctcacagggggtttatcacattgcac

402 cctgacagtcgtcagcctcacagggggtttatcacagtgcacccttacaatcattccatttgattcacaatttttttagtctc

403 tactgtgcctaacttgtaagttaaatttgatcagaggtgtgttcccagaggggaaaacagtatatacagggttcagtactatc

404 gcatttcaggcctccacctgggtcttggaatgtgtcccccgaggggtgatgactacctcagttggatctccacaggtcacagt

405 gacacaagataaccaagacacctcccaaggctaccacaatgggccgccctccacgtgcacatggccggaggaactgccatgtc

406 ggaggtgcaagcacacctgcgcatcagagtccttggtgtggagggagggaccagcgcagcttccagccatccacctgatgaac

407 agaacctagggaaagccccagttctacttacaccaggaaaggc

408 bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

409 Supplementary References

410 1. Greenwood, E.J., et al. Temporal proteomic analysis of HIV infection reveals

411 remodelling of the host phosphoproteome by lentiviral Vif variants. Elife 5(2016).

412 2. Ramirez, P.W., et al. Downmodulation of CCR7 by HIV-1 Vpu results in impaired

413 migration and chemotactic signaling within CD4(+) T cells. Cell Rep 7, 2019-2030

414 (2014).

415 3. Haller, C., et al. HIV-1 Nef and Vpu are functionally redundant broad-spectrum

416 modulators of cell surface receptors, including tetraspanins. J Virol 88, 14241-14257

417 (2014).

418 4. Lambele, M., et al. Vpu Is the Main Determinant for Tetraspanin Downregulation in

419 HIV-1-Infected Cells. J Virol 89, 3247-3255 (2015).

420 5. Jain, P., et al. Large-Scale Arrayed Analysis of Protein Degradation Reveals Cellular

421 Targets for HIV-1 Vpu. Cell Rep 22, 2493-2503 (2018).

422 6. Sugden, S.M., Pham, T.N. & Cohen, E.A. HIV-1 Vpu Downmodulates ICAM-1

423 Expression, Resulting in Decreased Killing of Infected CD4(+) T Cells by NK Cells. J

424 Virol 91(2017).

425 7. Shah, A.H., et al. Degranulation of natural killer cells following interaction with HIV-1-

426 infected cells is hindered by downmodulation of NTB-A by Vpu. Cell Host Microbe 8,

427 397-409 (2010).

428 8. Matusali, G., Potesta, M., Santoni, A., Cerboni, C. & Doria, M. The human

429 immunodeficiency virus type 1 Nef and Vpu proteins downregulate the natural killer

430 cell-activating ligand PVR. J Virol 86, 4496-4504 (2012).

431 9. Bolduan, S., Reif, T., Schindler, M. & Schubert, U. HIV-1 Vpu mediated

432 downregulation of CD155 requires alanine residues 10, 14 and 18 of the

433 transmembrane domain. Virology 464-465, 375-384 (2014). bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

434 10. Vassena, L., Giuliani, E., Matusali, G., Cohen, E.A. & Doria, M. The human

435 immunodeficiency virus type 1 Vpr protein upregulates PVR via activation of the

436 ATR-mediated DNA damage response pathway. J Gen Virol 94, 2664-2669 (2013).

437 11. Kim, D.Y., et al. CBFbeta stabilizes HIV Vif to counteract APOBEC3 at the expense

438 of RUNX1 target gene expression. Mol Cell 49, 632-644 (2013).

439 12. Schoggins, J.W., et al. A diverse range of gene products are effectors of the type I

440 interferon antiviral response. Nature 472, 481-485 (2011).

441 13. Schoggins, J.W., et al. Pan-viral specificity of IFN-induced genes reveals new roles

442 for cGAS in innate immunity. Nature 505, 691-695 (2014).

443 14. Greenwood, E.J., et al. Vpr drives massive cellular proteome remodelling in HIV-1

444 infection. bioRxiv (2018).

445 15. Schrofelbauer, B., Yu, Q., Zeitlin, S.G. & Landau, N.R. Human immunodeficiency

446 virus type 1 Vpr induces the degradation of the UNG and SMUG uracil-DNA

447 glycosylases. J Virol 79, 10978-10987 (2005).

448 16. Kuo, H.H., et al. Anti-apoptotic Protein BIRC5 Maintains Survival of HIV-1-Infected

449 CD4(+) T Cells. Immunity 48, 1183-1194 e1185 (2018).

450 bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Supplementary Fig. 1. a EGFP retained EGFP removed Location of SBP-ΔLNGFR Full length Truncated Full length Truncated 3’ LTR 3’ LTR 3’ LTR 3’ LTR

9.0kbp 8.9kbp

GFP+ LNGFR+ 83.7% 74.2% Fused to Env Not Not signal peptide tested tested

pNL4-3-ΔEnv- pNL4-3-ΔEnv- EGFP (control, no SBP-ΔLNGFR SBP-ΔLNGFR)

10.6kbp 10.1kbp 9.5kbp 9.1kbp

LNGFR+ LNGFR+ LNGFR+ LNGFR+ 40.8% 43.8% 85.8% 89.2% Downstream of Nef-P2A

pNL4-3-ΔEnv- Nef-P2A-SBP- ΔLNGFR

11.1kbp 10.6kbp 10.0kbp 9.6kbp

LNGFR+ LNGFR+ LNGFR+ LNGFR+ 0.15% 7.72% 49.9% 73.3% Downstream of Nef-IRES

pNL4-3-ΔEnv- Nef-IRES-SBP- ΔLNGFR-Δ3 CD4

GFP (pNL4-3-ΔEnv-EGFP) or LNGFR (all other constructs) b 1.E+09 2 pNL4-3-ΔEnv-EGFP 1.E+08 1 3 1. pNL4-3-ΔEnv-SBP-ΔLNGFR 1.E+07 2. pNL4-3-ΔEnv-Nef-P2A-SBPΔLNGFR 3. pNL4-3-ΔEnv-Nef-IRES-SBP-ΔLNGFR-Δ3 1.E+06

Viral titre (IU/ml) Other constructs 1.E+05 8 9 10 11 12

gRNA length (kbp) bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Supplementary Fig. 2. a EGFP Nef SBP-ΔLNGFR 24h 48h 48h

GFP+ GFP+ pNL4-3-ΔEnv-EGFP (control, no SBP-ΔLNGFR) 50.2% 85.7%

Original construct with EGFP fused to Env signal peptide (substituted for Env)

GFP

pNL4-3-ΔEnv-SBP-ΔLNGFR LNGFR+ LNGFR+ 19.4% 77.4%

SBP-ΔLNGFR fused to Env signal peptide (substituted for EGFP)

LNGFR+ LNGFR+ pNL4-3-ΔEnv-Nef-P2A-SBP-ΔLNGFR 35.1% 88.9%

SBP-ΔLNGFR downstream of Nef-P2A with EGFP removed

LNGFR+ LNGFR+ pNL4-3-ΔEnv-Nef-IRES-SBP-ΔLNGFR-Δ3 10.5% 71.8%

SBP-ΔLNGFR downstream of Nef-IRES with EGFP removed and trunc. 3’ LTR Tetherin CD4

LNGFR bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Supplementary Fig. 3. a Flow-through c LNGFR- LNGFR+ Tat Nef-P2A Gag 92.7% 7.32%

24h

Rev SBP-ΔLNGFR Gagpol LNGFR- LNGFR+ 89.1% 10.9%

48h Abundance

Time course

LNGFR b APOBEC3C APOBEC3D APOBEC3F APOBEC3B

PPP2R5A PPP2R5B PPP2R5C PPP2R5E SERINC1

HLTF ZGPAT VPRBP MCM10

MUS81 EME1 TET2

HLA-A1 HLA-A31 HLA-B8 HLA-Cw7 HLA-Cw15 /Ctrl Abundance Expt

Time course bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Supplementary Fig. 4. a b Mock WT ΔVif Input Selected 100 LNGFR- LNGFR+ 5.53% 94.7%

Donor A WT Donor A Donor B LNGFR+ ) 41.4% % Donor A Donor B 0 21.83 21.83 LNGFR- LNGFR+ Donor B

4.02% 96% F2 ( Donor C Donor C ΔVif

Donor C LNGFR+ 33.1% CD4 -100 -100 0 100 LNGFR LNGFR F1 (29.68 %) bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Supplementary Fig. 5. a b 4 R² = 0.6929 469 proteins - T4 - T4 with q<0.05 in 2 CEM 40% Primary T cell both datasets

Primary T cell 30% 0 ↓ ↑ 20% ↑ 11 210 -2 - T4 10%

CEM ↓ 231 17 Log2(WT HIV/Mock)_CEM Log2(WT Coefficient of of Coefficient variation 0% -4 -4 -2 0 2 4 Mock Mock Log2(WT HIV/Mock)_Primary T cell WT HIV WT HIV

c Additional Nef/Vpu targets RUNX1 target genes Interferon-stimulated genes (ISGs)

Decreased Increased Decreased Increased Decreased Increased 30 30 30

20 20 20

10 10 10 FDR 5% FDR 5% FDR 5% og10(p)

-L 0 0 0 -6 -4 -2 0 2 4 6 -6 -4 -2 0 2 4 6 -6 -4 -2 0 2 4 6 Log2(WT HIV/Mock)

Additional Nef/Vpu targets RUNX1-dependent ISGs q<0.05 q>0.05

d Only identified in primary T cells 25 26 126

Established accessory protein targets Only regulated in Additional Vpr targets

dependent 66 - primary T cells 246 Other Vpr dependent changes HIV Regulated in CEM-T4s but uncharacterised

650 161 changes in primary T T cellsin primary changes bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Supplementary Fig. 6. a b Downregulated Upregulated Sterol biosynthetic process by WT HIV by WT HIV

Decreased Increased 30

20

10 FDR 5%

containing compound metabolic process metabolic compound - containing 30

20 og10(p) 0 -L -6 -4 -2 0 2 4 6

10 Log2(WT HIV/Mock) RNA RNA processmetabolic processmetabolic of compound Regulation nitrogen ofRegulation biosynthetic process cellular macromolecule process biosynthetic of Regulation macromolecule of Regulation RNA processmetabolic Nucleic acidNucleic binding binding DNA compound binding Heterocyclic Organic cyclic binding compound of Regulation nucleobase

0 0 - Log10(p) Vpr-dependent Sterol biosynthetic Other process annotation q<0.05

Log10(p) q>0.05 -10

-20 Lipid processmetabolic Sterol process metabolic Steroid metabolic processSteroid metabolic Sterol biosynthetic processSterol biosynthetic Steroid processbiosynthetic Organic acid processcatabolic Cholesterol biosynthetic biosynthetic processCholesterol Carboxylic acidCarboxylic processcatabolic -30 Secondary alcohol metabolic process metabolic alcohol Secondary Secondary alcohol biosynthetic process biosynthetic alcohol Secondary bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Supplementary Fig. 7. b a ARID5A PTPN22

1 1 targets /Vpu targets Nef Resting Activated 24h_Pos 24h_Neg 48h_Pos 48h_Neg Vpr Vif/ Abundance

(fraction max.) (fraction 0 0 +1 DNTTIP2 Vif Vif WT WT Δ Δ UTP14A Mock Mock CCNT1 APOBEC3D PPP2R5A c ARID5A Tetherin -1 APOBEC3G 1 0 SERINC5 PPP2R5C APOBEC3F APOBEC3C PPP2R5E PPP2R5D 0 -2.5 VPRBP

- CD4 24h 48h ZGPAT Resting Activated (time post- HLTF infection) SLC38A1 CDCA2 GSG2 PTPN22 CWC25 1 0 DNAJB6 ZNF267

/Ctrl) KIF18B ECT2

Expt NUSAP1 CDCA5 BBX Log2( Abundance (fraction (fraction Abundance max.) 0 -2.5 ZNF512B -

24h 48h ESCO2 Resting Activated (time post- KIF18A infection) ARHGAP11A MCM10 GNL3L CCDC137 d GEMIN4 MSH6 ARID5B GNL2 1 0 DDX20 UNG

/Ctrl) NEPRO EME1 Expt PINX1 MUS81 Abundance (fraction max.) (fraction 0 -2.5 Log2( NOL7 - 24h 48h Resting Activated (time post- infection) bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Supplementary Fig. 8. a b Changes only identified in this study

Decreased Increased 4 30 R² = 0.7058 237 proteins with q<0.05 in

et al., 2018 al., et 2 both datasets 20 This study 0 ↓ ↑

↑ 8 84 -2 10

et al., 2018 al., et ↓ 139 6 FDR 5% Kuo

Kuo HIV/Mock)_ Log2(WT -4 -4 -2 0 2 4 og10(p)

Log2(WT HIV/Mock)_this study -L 0 -6 -4 -2 0 2 4 6 Log2(WT HIV/Mock)

Only identified in this study q<0.05 Reported in Kuo et al. 2018 q>0.05 bioRxiv preprint doi: https://doi.org/10.1101/389346; this version posted August 10, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Supplementary Table 1. (screenshot from Excel spreadsheet) Neg Pos 48 hrs ΔVif Mock ). 0.00 0.986 1.000 ΔVif/Mock Neg Pos 24 hrs https://www.genecards.org/ -1.28 0.000 0.000 ΔVif/Mock WT Mock Protein abundance (fraction of (fraction maximum)Protein abundance -1.29 0.000 0.000 Activated WT/Mock Protein abundance (fraction of (fraction maximum)Protein abundance "APOBEC3" (check for aliases at at aliases for (check "APOBEC3" between proteins. proteins. between not " under the number of unique of peptides. number the under " Mock ) proteomic experiments. proteomic ) Multiple Fig. 3a Resting p value q value Log2(ratio) 1 2 1 2 1.00 0.88 0.96 0.37 0.36 0.44 0.35 0.40 0.42 0.29 0.25 0.86 0.86 0.83 0.23 0.63 0.97 0.23 1.00 varies automatically varies Donor A Donor B Donor C Donor A Donor B Donor C Donor A Donor B Donor C "ZIP"; "APOBEC3G", "ZIP"; "APOBEC3G", not time time Time point point Single Single Single course ) and) single ( point time using the "Time course dataset" or "Single dataset" worksheets. point time dataset" or course using "Time the "tetherin"; "ZGPAT" Fig. 2a Vif Δ 14 WT e.g. "BST2", not Single time point time Single Mock 1 0

0.5 Abundance (fraction max.) (fraction Abundance 11 must match exactly

Time course Log2(Expt/Ctrl) of protein abundances cells protein of in (lines) paired experimental/control 0.5 0 -0.5 -1 -1.5 -2 -2.5

Unique

peptides 48h

(top left) to retreive protein abundances from time course ( course time abundances protein from retreive to left) (top 24h

scale for log2(ratio)s for scale - Activated (timeActivated post-infection) ARID5A (not protein names) may be entered, and these names) be entered, protein may (not Resting 1 0

0.5 Abundance (fraction max.) (fraction Abundance names gene Gene Gene name Enter gene name of interest here interest of name gene Enter Where a gene is represented by >1 protein (fragments, isoforms or variants with different accession numbers) the search returns " returns accession search the different numbers) with variants isoforms or (fragments, protein a >1 geneWhere is by represented unique most with peptides manually protein the isIn be cases, such checked must data but shown for - For time course data, note that the details. further See for table Information legend in Supplementary Enter gene in name interest gold box of Enter Onl y