1

1 Temporal proteomic analysis of HIV infection reveals

2 remodelling of the host phosphoproteome

3 by lentiviral Vif variants

4

5 Edward JD Greenwood 1,2,*, Nicholas J Matheson1,2,*, Kim Wals1, Dick JH van den Boomen1,

6 Robin Antrobus1, James C Williamson1, Paul J Lehner1,*

7 1. Cambridge Institute for Medical Research, Department of Medicine, University of

8 Cambridge, Cambridge, CB2 0XY, UK.

9 2. These authors contributed equally to this work.

10 *Correspondence: [email protected]; [email protected]; [email protected]

11

12 Abstract

13 Viruses manipulate host factors to enhance their replication and evade cellular restriction.

14 We used multiplex tandem mass tag (TMT)-based whole cell proteomics to perform a

15 comprehensive time course analysis of >6,500 viral and cellular proteins during HIV

16 infection. To enable specific functional predictions, we categorized cellular proteins regulated

17 by HIV according to their patterns of temporal expression. We focussed on proteins depleted

18 with similar kinetics to APOBEC3C, and found the viral accessory protein Vif to be

19 necessary and sufficient for CUL5-dependent proteasomal degradation of all members of the

20 B56 family of regulatory subunits of the key cellular phosphatase PP2A (PPP2R5A-E).

21 Quantitative phosphoproteomic analysis of HIV-infected cells confirmed Vif-dependent

22 hyperphosphorylation of >200 cellular proteins, particularly substrates of the aurora kinases.

23 The ability of Vif to target PPP2R5 subunits is found in primate and non-primate lentiviral 2

24 lineages, and remodeling of the cellular phosphoproteome is therefore a second ancient and

25 conserved Vif function.

26

27 Introduction

28 Viruses hijack host proteins and processes to optimize the cellular environment for viral

29 replication and/or persistence. Manipulation by viruses signposts critical pathways in viral

30 pathogenesis and cell biology, and evolutionary pressure has led to conflict between cellular

31 restriction factors (limiting viral replication) and viral countermeasures (overcoming

32 restriction in vivo). We previously used multiplex whole cell proteomic analysis of Human

33 Cytomegalovirus (HCMV)-infected fibroblasts to define expression time courses of viral and

34 cellular proteins and identify novel proteins involved in the host-HCMV interaction, a

35 technique we termed Quantitative Temporal Viromics (QTV) (Weekes et al., 2014). Here, we

36 provide a comprehensive temporal proteomic analysis of HIV infection.

37 The HIV-1 “accessory proteins” Vif, Vpr, Nef and Vpu share a common ability to target

38 cellular proteins for degradation (Simon et al., 2015; Sugden et al., 2016). Whilst dispensible

39 for viral replication in vitro, they are essential for pathogenesis in vivo. Nef and Vpu are

40 multifunctional adaptors which co-opt endolysosomal and proteasomal machinery to

41 downregulate numerous plasma membrane proteins, including their canonical substrates

42 CD4, tetherin and MHC class I. In contrast, although Vif and Vpr are known to target

43 cytoplasmic and nuclear proteins for proteasomal degradation, relatively few cellular

44 substrates have been reported.

45 The only known Vif targets are members of the APOBEC family of cytosine deaminases,

46 which are otherwise incorporated into viral particles and act as dominant restriction factors

47 causing hyper-mutation of the HIV genome (Desimmie et al., 2014; Malim, 2009). Whilst Nef,

48 Vpr and Vpu are found exclusively in primate lentiviruses, Vif is found in four of the five 3

49 extant lentiviral lineages, infecting primate, feline, bovine and small ruminant hosts (Gifford,

50 2012), and the ability to target cognate host APOBEC proteins is conserved across Vif

51 variants from all these diverse lineages (Larue et al., 2010).

52 Cellular proteins regulated by HIV have generally been identified using non-systematic,

53 candidate approaches. We recently used a different, unbiased plasma membrane proteomic

54 approach to reveal >100 previously unsuspected cell surface proteins depleted by HIV-1,

55 including novel Nef (SERINC3/5) and Vpu (SNAT1) targets (Matheson et al., 2015). Whole

56 cell proteomic studies of HIV-infected cells have been variably hampered by limited

57 proteome coverage, asynchronous infections and confounding by the presence of bystander

58 (uninfected) cells (Supplementary file 1). Consequently, it has been difficult to attribute

59 changes in protein levels to expression of specific viral genes, and intracellular proteins

60 targeted by HIV accessory proteins have not been discovered in this fashion.

61 In this study, we extend our tandem mass tag (TMT)-based temporal proteomic approach to

62 describe global changes in HIV-infected T cells, comprising expression time courses of

63 >6,500 proteins. We cluster proteins according to their patterns of temporal expression, and

64 identify >100 cellular proteins regulated by HIV, including candidate resistance/restriction

65 factors and HIV accessory protein targets. To test the utility of our approach, we focus on

66 proteins depleted with similar kinetics to APOBEC3C, and confirm the B56 family of

67 regulatory subunits of the key cellular phosphatase PP2A (PPP2R5A-E) to be novel Vif

68 targets. We use large-scale quantitative phosphoproteomics to demonstrate Vif-dependent

69 remodelling of the cellular phosphoproteome during HIV infection, and show that, along with

70 APOBEC proteins, antagonism of PP2A-B56 is an ancient and conserved Vif function.

71

72 4

73 Results

74 Systematic time course analysis of protein dynamics during HIV infection

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

76 HIV infection, we analysed total proteomes of CEM-T4 T cells infected with HIV. As

77 previously described (Matheson et al., 2015), cells were spinoculated with Env-deficient,

78 VSVg-pseudotyped virus at an MOI sufficient to achieve a synchronous single round

79 infection with <10% uninfected bystander cells. We exploited 6-plex TMT labelling to

80 quantitate 6,538 proteins in whole cell lysates of uninfected cells (0 h), at four timepoints

81 following HIV-1 infection (6, 24, 48, and 72 h), and in cells infected for 72 h in the presence

82 of reverse transcriptase inhibitors (RTi) (Figure 1A). The complete dataset has been

83 deposited to the ProteomeXchange consortium with the dataset identifier PXD004187

84 (accessible at http://proteomecentral.proteomexchange.org) and is summarised in an

85 interactive spreadsheet (Figure 1 – source data 1), which allows generation of temporal

86 profiles for any quantitated genes of interest.

87 We observed a tight correlation between levels of Env-GFP expression determined by mass

88 spectrometry and flow cytometry (r2 = 0.97) (Figure 1B). As expected, the well characterised

89 HIV cell surface targets downregulated in our plasma membrane proteomic analysis were

90 also depleted in our whole cell proteomic analysis (Figure 1 – figure supplement 1A). The

91 magnitude of effect was generally greater in the plasma membrane proteomic analysis

92 (Figure1 – figure supplements 1A-B), suggesting that regulation of cell surface proteins by

93 redistribution or sequestration is an important feature of this system.

94 We detected gene products from 7/9 HIV-1 open reading frames (ORFs; Figures 1B-C). As

95 previously reported, expression of regulatory proteins (Tat and Rev) from Rev-independent

96 completely spliced mRNA transcripts occurred earlier in viral replication than expression of

97 structural proteins from Rev-dependent unspliced (Gag and Gagpol) and partially spliced 5

98 (Env) mRNA transcripts (Karn and Stoltzfus, 2012; Pollard and Malim, 1998), with Rev

99 expression lagging Tat in our experiment. Vif and Nef showed intermediate temporal profiles

100 (Figure 1C), with progressively increasing Nef expression from 24-48 h inversely correlating

101 with downregulation of CD4 and HLA-A (Figure 1 – figure supplement 1A). Finally, we saw

102 an increase in plasma membrane VSVg levels immediately after infection (reflecting fusion

103 of incoming virions), followed by a rapid decline (Figure 1 – figure supplement 1C).

104 Compared with numerous cell surface targets (Haller et al., 2014; Matheson et al., 2015),

105 relatively few intracellular proteins depleted by HIV accessory proteins have been described.

106 Nonetheless, we confirmed downregulation of the Vif target APOBEC3C (Smith and Pathak,

107 2010) and the Vpr target UNG (Schrofelbauer et al., 2005) (Figure 1D). The temporal

108 pattern of UNG depletion was distinct from that of other accessory protein substrates,

109 including APOBEC3C, with degradation seen as early as 6 h post-infection, and preserved in

110 the presence of reverse transcriptase inhibitors. This is likely to reflect the high abundance of

111 Vpr packaged within incoming viral particles (Lu et al., 1993; Paxton et al., 1993), abrogating

112 the need for de novo protein synthesis. As well as recruiting substrates for degradation, Vpu

113 increases β-catenin levels by sequestering the ß-TrCP substrate-recognition unit of the

114 SCFß-TrCP E3 ubiquitin ligase complex (Besnard-Guerin et al., 2004). In addition, HIV infection

115 causes cell cycle arrest at G2/M (Jowett et al., 1995), a point in the cell cycle associated with

116 upregulation of cyclin B1 (Norbury and Nurse, 1992). Accordingly, we observed progressive

117 accumulation of both β-catenin and cyclin B1 (Figure 1D).

118 Temporal clustering of cellular proteins modulated by HIV

119 Gene Set Enrichment Analysis (GSEA) revealed time-dependent perturbation of multiple

120 cellular processes and pathways during HIV infection (Figure 1 – figure supplements 2A-

121 B), with protein-level changes generally supporting earlier transcriptome-level data. For

122 example, genes associated with lipid metabolism (Figure 1 – figure supplements 2A and 6

123 2C) are induced by expression of Nef (Shrivastava et al., 2016; van 't Wout et al., 2005),

124 whereas genes associated with RNA processing (Figure 1 – figure supplements 2B and

125 2D) are suppressed in HIV-infected cells (Chang et al., 2011; Sherrill-Mix et al., 2015).

126 To facilitate data mining and identify specific host factors regulated by HIV infection, we

127 classified cellular proteins according to their patterns of temporal expression (Figure 2A).

128 We observed 4 main clusters: (#1) 29 proteins downregulated late in infection, rescued in

129 the presence of reverse transcriptase inhibitors; (#2) 59 proteins downregulated earlier in

130 infection, incompletely rescued in the presence of reverse transcriptase inhibitors; (#3) 29

131 proteins progressively upregulated during infection, abolished in the presence of reverse

132 transcriptase inhibitors; and (#4) 49 proteins progressively upregulated during infection, even

133 in the presence of reverse transcriptase inhibitors (Figure 2B). We validated protein

134 downregulation (clusters #1 and #2) and upregulation (clusters #3 and #4) in an independent

135 infection time course experiment, using Stable Isotope Labelling with Amino Acids in Culture

136 (SILAC) as an alternative quantitative proteomic approach (Figure 2C and Figure 2 – figure

137 supplement 1A). Details of all proteins in clusters #1-4, including validation time course

138 data, are available in Figure 2 – source data 1.

139 Distinct patterns of temporal regulation imply different mechanisms and biological

140 significance. The Nef, Vpu and Vif accessory protein targets CD4, SNAT1, APOBEC3C are

141 found in cluster #1, where progressive downregulation and rescue by reverse transcriptase

142 inhibitors suggest dependence on de novo viral protein synthesis (compare Figure 2B with

143 Figure 1D, top left panel, and Figure 1 – figure supplement 1A). Upregulation of proteins

144 in cluster #3 is also likely to require de novo viral protein synthesis, and the indirect Vpu

145 target β-catenin is found in this cluster (compare Figure 2B with Figure 1D, middle left

146 panel). Conversely, reverse transcriptase inhibitor-independent regulation of proteins in

147 clusters #2 and #4 implies a cellular response to HIV infection, or a direct effect of viral

148 proteins in incoming virions, and the Vpr target UNG is found in cluster #2 (compare Figure 7

149 2B with Figure 1D, top right panel). As well as mechanistic differences, analysis using the

150 Database for Annotation, Visualisation and Integrated Discovery (DAVID) revealed that

151 clusters #1-4 contained proteins associated with distinct biological functions and processes

152 (Figure 2 – figure supplement 1B). Whilst accumulation of some proteins in cluster #3 may

153 be secondary to G2/M cell cycle arrest, other changes are unlikely to reflect the interferon

154 (IFN) or unfolded protein responses, because we did not see accumulation of either the

155 highly IFN-inducible protein ISG15 (Figure 1 – figure supplement 1D) or proteins

156 associated with ER stress (Figure 1 – figure supplement 1E).

157 Regulation of resistance/restriction factors and candidate accessory protein targets

158 Restriction factors are cellular proteins whose primary biological activity is antiviral, and

159 which are induced by IFN or viral infection, antagonized by viral proteins, and show genetic

160 evidence of positive selection (Duggal and Emerman, 2012). Proteins which reduce

161 permissivity for viral infection, but fail to meet strict criteria for restriction factors, may be

162 classified as resistance factors (Goujon et al., 2013). Restriction and resistance factors are

163 characteristically increased (cellular response) or decreased (viral antagonism) during viral

164 infection. Our temporal proteomic approach therefore identifies unsuspected viral regulation

165 of known resistance/restriction factors, and provides a strategy for the discovery of novel

166 host antiviral factors. Accordingly, we found the actin regulatory proteins gelsolin and CAPG

167 (both cluster #4) to be strongly induced during HIV infection (Figure 1 – figures

168 supplement 1F). Actin cytoskeletal remodeling is required for virological synapse formation

169 and cell-cell transmission of HIV (Jolly et al., 2004), and gelsolin levels have been reported

170 to control early HIV infection in macrophages (Garcia-Exposito et al., 2013). In contrast, we

171 discovered marked depletion of FMR1 (cluster #1, Figure 2D, upper panel) and TFAP4

172 (cluster #2, Figure 2D, lower panel) during HIV infection. Both these proteins reduce

173 production of infectious HIV virus (Figure 2E) (Imai and Okamoto, 2006; Pan et al., 2009),

174 but their distinct patterns of temporal expression suggest different mechanisms of viral 8

175 regulation. We predict that other regulated proteins in clusters #1-4, without known roles in

176 HIV infection, will also represent novel cellular resistance/restriction factors.

177 To test the utility of our approach, we focussed on proteins in cluster #1 highlighted in our

178 functional analysis (Figures 3A-B and Figure 2 – figure supplement 1B). First, three

179 members of the deoxynucleotide triphosphate (dNTP) biosynthetic pathway were depleted

180 during HIV infection: thymidylate synthetase (TYMS), which catalyses the methylation of

181 deoxyuridylate (dUMP) to deoxythymidylate (dTMP); and two subunits of ribonucleotide

182 reductase (RNR), RRM1 and RRM2, which catalyses the formation of deoxyribonucleotides

183 from ribonucleotides (Figure 3B, left panels). HIV replication is tightly regulated by dNTP

184 availability, and SAMHD1 (which tends to oppose the effects of RNR) is a well-described

185 HIV restriction factor (Ayinde et al., 2012; Baldauf et al., 2012; Hrecka et al., 2011; Laguette

186 et al., 2011; Lahouassa et al., 2012; Taylor et al., 2015). Second, and most strikingly, all

187 detected members of the B56 family of protein phosphatase 2A (PP2A) regulatory subunits

188 (PPP2R5A, C, D and E) were profoundly depleted by HIV (Figure 3B, right panels). These

189 subunits determine the specificity and localisation of PP2A holoenzymes (PP2A-B56), a

190 ubiquitous family of heterotrimeric serine-threonine phosphatases with critical roles in many

191 aspects of cellular physiology (McCright et al., 1996). We confirmed downregulation of

192 RRM2, PPP2R5A and PPP2R5D by immunoblot (Figure 3 – figure supplement 1).

193 Because cluster #1 also contained APOBEC3C, and intracellular proteins in this cluster

194 (including PPP2R5 subunits) show near-identical patterns of temporal expression, we

195 hypothesised that some or all of these proteins might be novel Vif targets.

196 Systematic multiplex proteomic analysis of Vif targets

197 To examine this hypothesis and systematically identify novel Vif targets, we performed a 3-

198 way proteomic comparison of mock-infected cells and cells infected with wildtype (WT) or

199 Vif-deficient (∆Vif) HIV viruses (Figure 4 – figure supplement 1A). To complement our high 9

200 MOI time course experiment, we used an MOI of 1.5, resulting in approximately 75%

201 productive infection and (typically) one or two copies of the viral genome per cell (Figure 4 –

202 figure supplement 1B). We exploited 10-plex TMT labelling to analyse samples in triplicate

203 at a single timepoint 48 h post-infection, with the resulting statistical power compensating for

204 the reduced magnitude of changes due to the presence of bystander (uninfected) cells. As

205 expected, Vif protein was only detected in WT infection, but levels of other viral proteins

206 were equivalent (Figure 4A).

207 Vif-independent HIV targets such as tetherin (Vpu substrate), SNAT1 (Vpu substrate), CD4

208 (Nef/Vpu substrate) and UNG (Vpr substrate) were depleted in cells infected with both WT

209 (Figure 4A, left panel) and ∆Vif (Figure 4A, middle panel) viruses, with no difference in

210 abundance in the presence or absence of Vif (Figure 4A, right panel). Conversely, known

211 Vif targets APOBEC3C, APOBEC3G and APOBEC3D were all decreased by WT but not

212 ΔVif HIV infection (Figure 4A), and APOBEC3B, which is resistant to Vif (Doehle et al.,

213 2005), was unchanged across all conditions. In addition to APOBEC family members, all five

214 PP2A-B56 regulatory subunits PPP2R5A-E were depleted in the presence of Vif (Figure

215 4A). Vif-dependent degradation of PPP2R5 subunits was confirmed by immunoblot of HIV-

216 infected CEM-T4 T cells (PPP2R5A and PPP2R5D; Figure 4B,) and intracellular flow

217 cytometry of HIV-infected CEM-T4 and primary human CD4+ T cells (PPP2R5D; Figure 4 –

218 figure supplement 2A-B).

219 CUL5-dependent proteasomal degradation of PPP2R5 subunits

220 To test whether degradation of PPP2R5 subunits by Vif was post-translational, we

221 expressed HA-tagged PPP2R5A in 293T cells. Transfection of Vif caused a marked loss of

222 intracellular HA staining (Figure 4C, middle panels), and the same effect was also seen in

223 cells expressing all other PPP2R5 subunits or APOBEC3G (Figure 4D). Degradation of

224 APOBEC family members by Vif is mediated by recruitment of a cullin-5 (CUL5) E3 ubiquitin 10

225 ligase complex, resulting in substrate-specific ubiquitination and proteasomal degradation

226 (Malim and Bieniasz, 2012). We therefore predicted that depletion of PPP2R5 subunits

227 would exploit the same pathway.

228 Accordingly, we found that the Vif C114S mutant, which is unable to recruit CUL5 (Bergeron

229 et al., 2010), was defective for PPP2R5A degradation (Figure 4C, right panels), and

230 PPP2R5A degradation by wildtype Vif was rescued in the presence of the proteasome

231 inhibitor bortezomib (Figure 5A). A similar rescue of PPP2R5B was seen when Vif was co-

232 transfected with dominant negative, but not wildtype, CUL5 (Figure 5B), and knockdown of

233 other cellular components of the CUL5 E3 ligase complex recruited by Vif (EloB, EloC and

234 CBFβ) (Jager et al., 2012; Malim and Bieniasz, 2012) rescued both PPP2R5B and

235 APOBEC3G from degradation, with the magnitude of rescue similar for both substrates

236 (Figure 5C).

237 Consistent with a protein-level interaction between Vif and PPP2R5 subunits, we observed

238 co-immunoprecipitation of FLAG-tagged Vif with HA-tagged and endogenous PPP2R5D in

239 293T cells (Figure 5 – figure supplement 1A), and co-immunoprecipitation of untagged Vif

240 with endogenous PPP2R5D in CEM-T4 T cells infected with HIV (Figure 5 – figure

241 supplement 1B). As in 293T cells transfected with Vif, PPP2R5 subunit depletion in CEM-

242 T4 T cells infected with HIV was abolished in the presence of bortezomib (Figure 5 – figure

243 supplement 2A-B). Finally, we confirmed using cycloheximide chase (Figure 5 – figure

244 supplement 2) and [35S]methionine/[35S]cysteine metabolic labelling/pulse-chase (Figure 5

245 – figure supplement 3) analyses that degradation of PPP2R5D was accelerated in the

246 presence of Vif in HIV-infected CEM-T4 T cells. Vif is therefore both necessary and sufficient

247 for degradation of PPP2R5 subunits, and employs the same cellular machinery required for

248 degradation of APOBEC family members.

249 Remodelling of the cellular phosphoproteome by HIV infection 11

250 The substrate specificity of the PP2A phosphatase holoenzyme is determined by binding of

251 its regulatory subunits (Yang and Phiel, 2010). To identify the phenotypic consequences of

252 Vif-mediated PPP2R5A-E subunit depletion, we used titanium dioxide-based

253 phosphopeptide enrichment and 10-plex TMT labelling to analyse total phosphoproteomes

254 of the mock-, WT and ∆Vif virus-infected cells described in Figure 4A and Figure 4 – figure

255 supplements 1A-B. In total, we quantitated 8631 phosphopeptides from 2767 proteins

256 (Figure 6 – source data 1). Phosphopeptide abundance was normalized to total protein

257 abundance determined from the whole cell proteomic analysis, allowing differential

258 phosphorylation to be distinguished from altered protein expression (Wu et al., 2011). HIV

259 infection resulted in marked remodelling of the cellular phosphoproteome (Figure 6 – figure

260 supplement 1A), and analysis using the Database for Annotation, Visualisation and

261 Integrated Discovery (DAVID) revealed enhanced phosphorylation of proteins associated

262 with cell cycle regulation and activation of the DNA damage response (Figure 6 – figure

263 supplement 1B). To isolate those changes which specifically resulted from Vif-mediated

264 PPP2R5A-E subunit depletion, we focused on differences between cells infected with WT

265 and ∆Vif viruses.

266 Remarkably, compared with the small number of protein-level changes in the presence or

267 absence of Vif (specifically, APOBEC and PPP2R5 family members; Figure 4A, right panel

268 and Figure 6A left panel), we saw striking Vif-dependent changes in the phosphoproteome

269 (Figure 6A, right panel). Furthermore, as predicted for antagonism of a phosphatase,

270 almost all changes represented increases in phosphopeptide abundance, indicating

271 increased protein phosphorylation (with a total of 238 peptides from 192 proteins showing

272 abundance changes of >2 fold with a q value of <0.01). To confirm that the observed

273 changes resulted from PP2A antagonism, we compared our Vif-dependent changes in

274 protein phosphorylation with published alterations to the phosphoproteome of HeLa cells

275 following treatment with the PP2A inhibitor okadaic acid (Kauko et al., 2015). Despite the 12

276 different cell types and treatments, a highly significant correlation was found between our

277 observed Vif-dependent changes in HIV-infected cells and the published changes resulting

278 from okadaic acid treatment (Figure 6B and Figure 6 – figure supplement 1D).

279 To identify individual kinases with enhanced activity in the presence of Vif-dependent

280 PPP2R5A-E subunit depletion, we interrogated our data using PhosFate

281 (http://phosfate.com/), which infers kinase activity from quantitative phosphoproteomic data

282 by examining the coordinated regulation of known phosphosites. We found marked

283 activation of aurora kinase A (AURKA) and B (AURKB) in cells infected with WT but not ∆Vif

284 viruses (Figure 6 – figure supplement 1C), and confirmed this observation by comparing

285 phosphorylation of sites listed on the PhosphoSite kinase-substrate database

286 (http://www.phosphosite.org/) between WT and ∆Vif virus infections (Figure 6C). Next, we

287 compared Vif-dependent changes in protein phosphorylation with published alterations to the

288 phosphoproteome of HeLa cells following treatment with the aurora kinase inhibitors

289 MLN8054 (Figures 6D, upper panel and Figure 6 – figure supplement 1E) and

290 AZD1152/ZM447439 (AZDZM; Figure 6 – figure supplement 1E) (Kettenbach et al., 2011).

291 As expected, we found a significant inverse correlation between our Vif-dependent changes

292 in HIV-infected cells and the published changes resulting from aurora kinase inhibition,

293 whereas no such correlation was seen for control datasets from the same study employing

294 DMSO or the PLK1-3 inhibitor BI2536 (Figure 6D, lower panel, and Figure 6 – figure

295 supplement 1E).

296 Finally, to fully characterize the behavior of these kinases in our dataset, we manually

297 curated the literature for substrates of aurora kinases, including PLK1 as a negative control

298 for Vif-specific effects (Figure 6 – source data 2). PLK1 protein abundance was

299 upregulated in both WT and ∆Vif infections, with enhanced phosphorylation of kinase-

300 specific phosphosites, but no difference between WT and ∆Vif viruses (Figure 6E, left

301 panels, and Figure 1 – figure supplement 1G). By contrast, whilst the aurora kinases were 13

302 also upregulated equally in WT and ∆Vif infections, increased phosphorylation of kinase-

303 specific phosphosites was only seen in the presence of Vif (Figures 6E, middle and right

304 panels, and Figure 1 – figure supplement 1G). Depletion of PPP2R5A-E subunits by Vif is

305 therefore responsible for the selective amplification of aurora kinase activity in HIV-infected T

306 cells.

307 Depletion of PPP2R5A-E subunits by phylogenetically diverse lentiviral Vif variants

308 The vif gene is found in all primate lentiviral lineages, and in most of the extant non-primate

309 lineages. We therefore assembled a panel of vif genes from diverse primate and non-

310 primate lentiviruses (Figure 7A and Figure 7 – figure supplement 1), including 14 vif

311 variants from HIV-1 clades A-F and 6 vif variants from SIVcpz and SIVgor of chimpanzees

312 and gorillas, the most closely related viruses to HIV. Multiple vif variants from two other

313 primate lentiviral lineages were also represented: SIVsmm of sooty mangabeys, and the

314 viruses that resulted from cross species transmission of SIVsmm, HIV-2 and SIVmac; and

315 SIVagm of African green monkeys. Finally, a non-primate lentivirus vif variant was included,

316 from a small ruminant lentivirus (SRLV, or maedi-visna virus) isolated from sheep (Sargan et

317 al., 1991).

318 Vif variants were tested by transfection of 293T cells stably expressing HA-tagged PPP2R5

319 subunits, with the proportion of HA-tagged protein degraded in transfected cells quantitated

320 by intracellular flow cytometry. All HIV-1 variants tested were able to degrade HA-PPP2R5A,

321 but the magnitude of effect was variable (Figure 7 – figure supplement 2A). We therefore

322 screened a diverse selection of Vif variants for degradation of different PPP2R5 subunits

323 (Figure 7 – figure supplement 2B). The ability to deplete PPP2R5 subunits was conserved

324 across all PPP2R5A-E/Vif combinations, but most marked for PPP2R5B. We therefore

325 tested our entire panel of Vif variants for depletion of PPP2R5B, and found strong and

326 consistent degradation (Figure 7B and Figure 7 – figure supplement 2C). 14

327 Finally, we focused specifically on the distantly related SRLV and NL4-3 (HIV-1) Vif variants.

328 Vif-dependent antagonism of APOBEC proteins shows lineage-specificity, and SRLV Vif is

329 unable to antagonize human APOBEC3G (Larue et al., 2010). Nonetheless, despite only

330 sharing 15% amino acid identity with NL4-3 Vif (Figure 7 – figure supplement 1), SRLV Vif

331 was still able to associate with (Figure 7 – figure supplement 3A) and efficiently degrade

332 human PPP2R5 subunits (Figure 7C). Whilst Vif variants from primate lentiviruses (including

333 HIV-1) require CBFβ to enable proper protein folding, stability and interaction with the CUL5

334 E3 ligase complex (Fribourgh et al., 2014; Kim et al., 2013; Miyagi et al., 2014; Salter et al.,

335 2012) and mediate APOBEC depletion (Hultquist et al., 2012; Jager et al., 2012; Zhang et

336 al., 2012), Vif variants from non-primate lentiviruses (including SRLV) neither interact with

337 CBFβ (Ai et al., 2014; Kane et al., 2015; Yoshikawa et al., 2016; Zhang et al., 2014) nor

338 require CBFβ to antagonize their cognate APOBEC proteins (Ai et al., 2014; Kane et al.,

339 2015). As with APOBEC proteins, we found CBFβ but not EloB to be dispensable for

340 degradation of HA-PPP2R5E by SRLV Vif (Figure 7 – figure supplement 3B).

341

342 Discussion

343 In this study, we provide a comprehensive description of temporal changes in >6,500 viral

344 and cellular proteins during HIV infection. Our data confirm known HIV targets, and identify

345 many more proteins regulated by infection. Compared with other studies (Supplementary

346 file 1), we achieve a step-change in depth of proteomic coverage, and by utilising multiplex

347 TMT-based quantitation, we facilitate high-resolution time-based analysis. To generate a cell

348 surface proteomic map of HIV infection, we previously employed selective aminooxy-

349 biotinylation of sialylated glycoproteins (Plasma Membrane Profiling; PMP) to quantitate 804

350 plasma membrane proteins (Matheson et al., 2015). Although 1,030 proteins quantitated in

351 our whole cell proteomic analysis also had Gene Ontology Cellular Component annotations 15

352 suggesting localisation to the plasma membrane, there was limited overlap with our PMP

353 dataset (Figure 1 – figure supplement 1H, upper panel). The techniques are therefore

354 non-redundant, and this is likely to reflect differential enrichment of scarce or poorly

355 soluble/aggregate-prone glycoproteins using PMP, and intrinsic transmembrane proteins

356 lacking significant extracellular domains or glycosylation sites using whole cell proteomics

357 (Figure 1 – figure supplement 1H, lower panel).

358 In our earlier temporal proteomic study of HCMV infection, we utilised temporal classification

359 of cellular protein expression to predict novel immunoreceptors (Weekes et al., 2014). Here,

360 we develop and extend this methodology to predict cellular targets of specific HIV proteins.

361 Whereas HCMV encodes >150 canonical ORFs (Wilkinson et al., 2015), HIV-1 encodes only

362 9 genes and 15 proteins. Amongst these, the accessory proteins Vif, Vpr, Vpu and Nef have

363 distinct patterns of temporal expression, and HIV-1 is therefore ideally suited to this

364 approach. Vpu is translated from the same transcripts as Env (Schwartz et al., 1990), and

365 therefore expressed late in the viral replication cycle (Figure 1B). Accordingly, cell surface

366 proteins targeted specifically by Vpu (tetherin and SNAT1) are depleted late in the time

367 course of infection (Figure 1 – figure supplement 1A), and intracellular proteins known to

368 be targeted by Vif (APOBEC3C) and Vpr (UNG) show distinct temporal profiles (Figure 1D).

369 Based on similarity with the temporal profile of APOBE3C, we predicted that other proteins in

370 cluster #1 (Figure 3) would be candidate Vif targets. We validated this prediction by

371 comparing changes in protein expression during WT and ΔVif virus infections, and

372 demonstrated that as with APOBEC3C degradation, Vif was necessary for depletion of the

373 PP2A regulatory subunits PPP2R5A-E. Conversely, like known Vpr-target UNG, proteins in

374 cluster #2 are downregulated early in viral infection, in the absence of Vif, and in the

375 presence of reverse transcriptase inhibitors. Vpr is reported to antagonize DNA repair

376 pathways and inhibit innate immune sensing of viral nucleic acids (Laguette et al., 2014;

377 Schrofelbauer et al., 2005). Interestingly, cluster #2 is markedly enriched for nucleic acid 16

378 binding proteins, including proteins from families with known roles in DNA damage repair

379 and nucleic acid sensing (Figure 2 – figure supplement 1B and Figure 2 – source data

380 1). Whilst this manuscript was in preparation, downregulation of a second protein in cluster

381 #2, helicase-like transcription factor (HLTF), was also attributed to Vpr in incoming viral

382 particles (Hrecka et al., 2016; Lahouassa et al., 2016). Remarkably, as for Vif targets

383 APOBEC3C and PPP2R5A-E, temporal profiles of Vpr targets UNG and HLTF cluster very

384 tightly (Figure 2 – figure supplements 1C-D). Other proteins in cluster #2 with similar

385 temporal profiles are therefore very strong candidates for novel Vpr targets, and

386 downregulation of these proteins by Vpr may antagonize DNA repair pathways or inhibit viral

387 nucleic acid sensing.

388 Reversible serine/threonine phosphorylation is the most commonly observed post-

389 translational modification (Khoury et al., 2011), and PP1 and PP2A, are the major cellular

390 serine/threonine phosphatases. The core PP2A enzyme consists of one catalytic subunit

391 encoded by PPP2CA or PPP2CB and one structural subunit encoded by PPP2R1A or

392 PPP2R1B. Specificity of the holoenzyme is determined by binding of an additional regulatory

393 subunit, encoded by a total of 15 genes, split into four families (Yang and Phiel, 2010). We

394 found Vif-dependent proteasomal degradation of all five members of the B56 family

395 (PPP2R5A-E; also known as the B’, PR61 or PPP2R5 family). Since each PP2A

396 holoenzyme contains a single regulatory subunit, it is unlikely that depletion of individual

397 regulatory subunits destabilizes other B56 family members. Given the high sequence

398 similarity between B56 subunits, and the ability of Vif to deplete individual subunits

399 expressed non-stoichiometrically in 293T cells, it is much more likely that degradation is

400 mediated by a conserved Vif interaction domain in all five family members.

401 PP2A makes up 0.2-1% of total eukaryotic cellular protein (Lin et al., 1998; Ruediger et al.,

402 1991), and whilst in many cases the relevant regulatory subunits have not been

403 characterized, published targets of PP2A-B56 holoenzymes are nonetheless implicated in a 17

404 multitude of cellular processes (Yang and Phiel, 2010). In order to confirm functional PP2A-

405 B56 antagonism and identify relevant PP2A-B56 substrates in HIV-infected cells, we

406 therefore carried out a comprehensive, unbiased analysis of cellular protein phosphorylation

407 during productive HIV infection, and provide a multiplex TMT-based replicated analysis of

408 >8,500 cellular phosphopeptides. As expected, we found enhanced phosphorylation of

409 proteins associated with Vpr-mediated activation of the DNA damage response and G2/M

410 cell cycle arrest (Figure 6 – figure supplement 1B), reflecting increased activity of the

411 mammalian checkpoint kinases ATR/ATM (Figure 6 – figure supplement 1C) (Lai et al.,

412 2005; Nakai-Murakami et al., 2007; Roshal et al., 2003; Vassena et al., 2013). Conversely,

413 PP2A-B56 antagonism by Vif resulted in hyperphosphorylation of a more limited subset of

414 host phosphoproteins, mirroring previously reported changes seen with PP2A inhibition

415 using okadaic acid.

416 Our unbiased analysis of Vif-dependent kinase pathways in HIV-infected cells identified a

417 striking increase in activity of the aurora kinases (Figure 6C and Figure 6 – figure

418 supplement 1C). Aurora kinase activity and abundance peak in late G2 and mitosis

419 (Bischoff et al., 1998; Ly et al., 2014), and PP2A-B56 holoenzymes antagonize aurora

420 kinase functions in other systems (Bastos et al., 2014; Espert et al., 2014; Kruse et al., 2013;

421 Xu et al., 2013). Conversely, aurora kinase activity is typically inhibited by the DNA damage

422 response (Bensimon et al., 2011), and reduced activity would therefore be expected in HIV-

423 infected cells. Instead, we propose that Vif-mediated antagonism of PP2A-B56 sustains

424 aurora kinase activity in the presence of the DNA damage response. Interestingly, PLK1 is

425 also inhibited by the DNA damage response in other systems (Bensimon et al., 2011), but

426 kinase-active PLK1 is recruited to the SLX4 complex by Vpr in HIV-infected cells (Laguette

427 et al., 2014), consistent with the results of this study (Figure 6E). Manipulation of mitotic

428 kinases is therefore a shared feature of the HIV accessory proteins Vpr and Vif, and 18

429 pharmacological inhibitors targeting these cellular kinases may represent a viable antiviral

430 strategy.

431 Replication of WT and ΔVif viruses in vitro is equivalent in permissive cell lines lacking

432 APOBEC3G expression (Sheehy et al., 2002). Aurora kinase activity controls Lck kinase

433 location and phosphorylation at the immunological synapse (Blas-Rus et al., 2016), and

434 kinase-active Lck is also recruited to the virological synapse during cell-cell transmission of

435 HIV (Vasiliver-Shamis et al., 2009). It is therefore possible that Vif-mediated PP2A-B56

436 antagonism directly enhances cell-cell spread in vivo or in vitro in primary T cells or

437 macrophages, but it is not currently practicable to compare replication of WT and ΔVif

438 viruses on an APOBEC family-negative background in primary cells. Alternatively, PP2A-

439 B56 antagonism may enhance HIV replication or persistence in vivo indirectly, by modulating

440 T cell activation or macrophage polarization (Blas-Rus et al., 2016; Ding et al., 2015). It is

441 also possible that in other cell types or systems, such as terminally differentiated (non-

442 cycling) macrophages, signalling through alternative kinases may be differentially amplified

443 by Vif-mediated PP2A-B56 depletion. Nonetheless, since PP2A-B56 antagonism spans

444 lineages of lentiviruses which are primarily tropic for both lymphocytes (primate lentiviruses)

445 and myeloid cells (non-primate lentiviruses), it is likely that modulation of key kinases is

446 conserved across cell types.

447 The significance of host factors targeted by HIV is proven in vivo by evolutionary

448 conservation of antagonism across a range of HIV and SIV viruses, and by the existence of

449 similar mechanisms in other viruses. For example, MHC class I proteins are targeted by Nef

450 variants of all primate lentiviruses (Specht et al., 2008), and manipulation of MHC class I is a

451 common attribute of many virus families (Randow and Lehner, 2009). Here, we show that

452 degradation of PP2A-B56 subunits is conserved across Vif variants from diverse HIV and

453 SIV lentiviruses of primates, as well as a small ruminant lentivirus of sheep (SRLV). The

454 lentiviral genus is ancient (Gifford et al., 2008; Katzourakis et al., 2007; Keckesova et al., 19

455 2009; Worobey et al., 2010), and species-specific lineages have developed through virus-

456 host co-evolution. Accordingly, the most recent common ancestor of the primate and small

457 ruminant lentiviruses is likely to have existed in the common ancestor of primates and

458 ruminants, approximately 100 million years ago (Hedges et al., 2015). We therefore propose

459 that degradation of PP2A-B56 subunits is a primordial feature of Vif, present in the common

460 ancestor of primate lentiviral and SRLV Vif variants. Alternatively, Vif variants from these

461 lineages may have independently acquired this ability. Either possibility strongly suggests a

462 critical selective advantage for lentiviral replication or persistence in vivo.

463

464 Materials and methods

465 General cell culture

466 CEM-T4 T cells (AIDS Reagent Program, Division of AIDS, NIAD, NIH: Dr J.P. Jacobs)

467 (Foley et al., 1965) were cultured in RPMI supplemented with 10 % FCS, 100units/ml

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

469 (Lehner laboratory stocks) (Matheson et al., 2015) were cultured in DMEM supplemented

470 with 10 % FCS, 100units/ml penicillin and 0.1 mg/ml streptomycin at 37 °C in 5 % CO2. All

471 cells were confirmed to be mycoplasma negative (Lonza MycoAlert). Cell line authentication

472 was not undertaken.

473 Stable Isotope Labelling with Amino Acids in Cell Culture (SILAC)

474 For SILAC labelling, CEM-T4 T cells were grown for at least 7 cell divisions in SILAC RPMI

475 lacking lysine and arginine (Thermo Scientific) supplemented with 10 % dialysed FCS

476 (Gibco), 100units/ml penicillin and 0.1 mg/ml streptomycin, 280 mg/L proline (Sigma) and

477 light (K0, R0; Sigma), medium (K4, R6; Cambridge Isotope Laboratories) or heavy (K8, R10;

478 Cambridge Isotope Laboratories) 13C/15N-containing lysine (K) and arginine (R) at 50mg/L. 20

479 Primary cell isolation and culture

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

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

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

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

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

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

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

487 0.1mg/ml streptomycin at 37°C in 5% CO2.

488 HIV molecular clones

489 pNL4-3-dE-EGFP (derived from the HIV-1 molecular clone pNL4-3 but encoding Enhanced

490 Green Fluorescent Protein (EGFP) in the env open reading frame (ORF), rendering Env

491 non-functional) was obtained through the AIDS Reagent Program, Division of AIDS, NIAD,

492 NIH: Drs Haili Zhang, Yan Zhou, and Robert Siliciano (Zhang et al., 2004) and the complete

493 sequence verified by Sanger sequencing (Source BioScience).

494 To generate a Vif-deficient clone, overlapping PCR mutagenesis was used to introduce a

495 stop codon early in the Vif ORF, after the final in-frame start codon, as shown below.

496 Wild type sequence from start of Vif ORF (boxes indicate in-frame start codons)

497 ATGGAAAACAGATGGCAGGTGATGATTGTGTGGCAAGTAGACAGGATGAGGATTAACA

498 CATGGAAAAGATTAGTAAAACACCATATGTATATT

499 Mutagenised sequence (underlined region indicates introduced stop codons)

500 ATGGAAAACAGATGGCAGGTGATGATTGTGTGGCAAGTAGACAGGATGAGGATTAACA

501 CATGGAAAAGATTAGTAAAACACCATATGTAATAA 21

502 Restriction fragments were subcloned back into pNL4-3-dE-EGFP and mutations verified by

503 Sanger sequencing (Source BioScience) and immunoblot of infected CEM-T4s for Vif

504 protein.

505 Vectors for transgene expression

506 For lentiviral transgene expression in 293T cells, N-terminal 4xHA tagged PP2R5 genes

507 were subcloned from pCEP-4xHA-PPP2R5A-E (a kind gift from Dr David Virshup, Addgene

508 plasmids #14532-14537 (Seeling et al., 1999)) into pHRSIN-PGK-puro (van den Boomen et

509 al., 2014). APOBEC3G-HA was subcloned from pcDNA3.1-APOBEC3G-HA (AIDS Reagent

510 Program, Division of AIDS, NIAD, NIH: Dr. Warner C. Greene (Sheehy et al., 2002; Stopak

511 et al., 2003)).

512 PcVif, Pc∆Vif, and PcVif C114S expression vectors were a kind gift from Prof Michael Malim,

513 and have been previously described (Huthoff and Malim, 2007). pCRV1 Vif expression

514 vectors for HIV-1 and primate lentiviral Vif variants were a kind gift from Prof Viviana Simon

515 and have been previously described (Binka et al., 2012; Letko et al., 2013). SRLV Vif was

516 subcloned into pCRV1 by PCR from an SRLV EV1 Vif expression cloning vector, a kind gift

517 from Dr Barbara Blacklaws, University of Cambridge (Wu et al., 2008). The dominant

518 negative (DN) CUL5 expression plasmid pcDNA3-DN-hCUL5-FLAG was a kind gift from Prof

519 Wade Harper (Addgene plasmid #15823 (Jin et al., 2005)).

520 Lentivectors for shRNA expression

521 For lentiviral shRNA-mediated knockdown of EloB (TCEB2), EloC (TCEB1) and CBFβ

522 (CBFB) in 293T cells, hairpins were cloned into pHRSIREN-PGK-hygro (related to

523 pHRSIREN-PGK-SBP-ΔLNGFR-W, but expressing hygromycin resistance (Matheson et al.,

524 2014).The following oligonucleotides were inserted using BamHI-EcoRI (only top

525 oligonucleotides are shown). Gene specific target sequences are underlined and the source

526 of the target sequence design shown in parentheses. 22

527 EloB (Broad institute GPP Web Portal at https://www.broadinstitute.org/)

528 GATCCGCCACAAGACCACCATCTTTATTCAAGAGATAAAGATGGTGGTCTTGTGGCTTTTTTG

529 EloC (Takara Clontech RNAi Design Tool at http://bioinfo.clontech.com/rnaidesigner)

530 GATCCGCACCGAGATTCCTGAATTCTTCAAGAGAGAATTCAGGAATCTCGGTGTTTTTTACGCGTG

531 CBFβ (Broad institute GPP Web Portal at https://www.broadinstitute.org/)

532 GATCCGAAGATAGAGACAGGTCTCATTTCAAGAGAATGAGACCTGTCTCTATCTTCTTTTTTG

533 Viral stocks

534 VSVg-pseudotyped NL4-3-dE-EGFP HIV viral stocks were generated by co-transfection of

535 293T cells with pNL4-3-dE-EGFP molecular clones and pMD.G at a ratio of 9:1 (µg) DNA

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

537 supernatants harvested and filtered (0.45 µm) at 48 h prior to concentration with LentiX

538 Concentrator (Clontech) and storage at -80 °C. VSVg-pseudotyped pHRSIN and pHRSIREN

539 lentivector stocks were generated by co-transfection of 293T cells with lentivector, p8.91 and

540 pMD.G at a ratio of 2:1:1 (µg) DNA and a DNA:FuGENE 6 ratio of 1 µg:3 µl. Viral

541 supernatants were harvested, filtered, concentrated if required and stored at -80 °C. NL4-3-

542 dE-EGFP HIV viral stocks were titred by infection/transduction of known numbers of relevant

543 target cells under standard experimental conditions followed by flow cytometry for GFP and

544 CD4 at 48-72 h to identify % infected cells.

545 CEM-T4 T cell infections

546 CEM-T4 T cells were infected with concentrated HIV viral stocks by spinoculation at 800 x g

547 for 2 h in a non-refrigerated benchtop centrifuge in complete media supplemented with

548 10mM HEPES. In experiments with reverse transcriptase inhibitors, cells were incubated

549 with zidovudine (10 μM) and efavirenz (100 nM) (AIDS Reagent Program, Division of AIDS, 23

550 NIAD, NIH) for 1 h prior to spinoculation, and inhibitors maintained at these concentrations

551 during subsequent cell culture.

552 Tandem Mass Tag (TMT)-based whole cell proteomic time course analysis

553 Sample preparation

554 For the TMT-based HIV infection time course, CEM-T4 T cells were spinoculated with VSVg-

555 pseudotyped NL4-3-dE-EGFP HIV at a multiplicity of infection (MOI) of 10 in the presence or

556 absence of reverse transcriptase inhibitors. Aliquots of cells were harvested sequentially at

557 the indicated timepoints, and dead cells removed using the Miltenyi Dead Cell Removal kit.

558 Cells were analysed by flow cytometry for CD4 and GFP expression (confirming 95 %

559 productive infection) and subjected to Plasma Membrane Profiling (PMP; plasma membrane

560 proteomic analysis) and this data has been previously published (Matheson et al., 2015).

561 For whole cell proteomic analysis, 2x106 viable cells per timepoint were then washed with

562 ice-cold PBS with Ca/Mg pH 7.4 (Sigma) and frozen at -80 °C prior to Filter Aided Sample

563 Preparation (FASP) essentially as previously described (Wisniewski et al., 2009)

564 Cell pellets were thawed on ice, lysed in 4% SDS/100 mM HEPES (Sigma) supplemented

565 with Complete Protease Inhibitor Cocktail (without EDTA; Roche), and sonicated at 4 °C

566 using a Diagenode Bioruptor. Protein concentrations were determined using the Pierce BCA

567 Protein Assay kit (Thermo Scientific) and 100 μg protein per timepoint subjected to

568 downstream processing. Lysates were transferred to Microcon-30 kDa Centrifugal Filter

569 Units, reduced (100 mM DTT) and alkylated (50 mM iodoacetamide) at room temperature,

570 washed with a total of 5 column volumes of 8 M urea/100 mM HEPES and 100 mM HEPES

571 pH 8.5, then incubated with 1 μg/50 μl modified sequencing grade trypsin (Promega) in 100

572 mM HEPES pH 8.5 at 37 °C for 8 h. After digestion, peptide eluates were collected by

573 centrifugation and stored at +4 °C prior to TMT labelling the next day. 24

574 For TMT labelling, TMT 6-plex reagents (Thermo Scientific) were dissolved in anhydrous

575 acetonitrile (0.8 mg/40 µl) according to the manufacturer’s instructions. Peptide

576 concentrations were determined using the Pierce Micro BCA Protein Assay kit (Thermo

577 Scientific) and 25 μg peptide per sample labelled with 20 μl reconstituted TMT 6-plex

578 reagent at a final acetonitrile concentration of 30% (v/v). Samples were labelled as follows:

579 uninfected cells, 0h (TMT 126); 6h (TMT 127); 24 h (TMT 128); 48 h (TMT 129); 72 h (TMT

580 130); 72 h plus reverse transcriptase inhibitors (TMT 130). Following incubation at room

581 temperature for 1 h, reactions were quenched with hydroxylamine to a final concentration of

582 0.3 % (v/v). Samples were mixed at a ratio of 1:1:1:1:1:1 and dried down to remove

583 acetonitrile prior to off-line peptide fractionation.

584 Off-line High pH Reversed-Phase (HpRP) peptide fractionation

585 TMT-labelled tryptic peptides were subjected to HpRP-HPLC fractionation using a Dionex

586 Ultimate 3000 powered by an ICS-3000 SP pump with an Agilent ZORBAX Extend-C18

587 column (4.6 mm × 250 mm, 5 μm particle size). Mobile phases (H20, 0.1 % NH4°H or MeCN,

588 0.1 % NH4°H) were adjusted to pH 10.5 with the addition of formic acid and peptides were

589 resolved using a linear 40 min 0.1-40 % MeCN gradient over 40 min at a 400 μl/min flow rate

590 and a column temperature of 15 °C. Eluting peptides were collected in 15 s fractions. One

591 hundred and twenty fractions covering the peptide-rich region were re-combined to give 10

592 samples for analysis. To preserve orthogonality, fractions were combined across the

593 gradient, with each of the concatenated samples comprising 12 fractions which were 10

594 fractions apart. Re-combined fractions were dried down using an Eppendorf Concentrator

595 and resuspended in 15 µl MS solvent (3 % MeCN, 0. 1% TFA)

596 Mass spectrometry

597 Data for TMT labelled samples were generated using an Orbitrap Fusion Tribrid mass

598 spectrometer (Thermo Scientific). Peptides were fractionated using an RSLCnano 3000 25

599 (Thermo Scientific) with solvent A comprising 0.1 % formic acid and solvent B comprising 80

600 % MeCN, 20 % H2O, 0.1 % formic acid. Peptides were loaded onto a 50 cm Acclaim

601 PepMap C18 column (Thermo Scientific) and eluted using a gradient rising from 10 to 25 %

602 solvent B by 90 min and 40% solvent B by 115 min at a flow rate of 250 nl/min. MS data was

603 acquired in the Orbitrap at 120,000 fwhm between 400-1600 m/z. Spectra were acquired in

604 profile with AGC 4x105. Ions with a charge state between 2+ and 6+ were isolated for

605 fragmentation in top speed mode using the quadrupole with a 1.5 m/z isolation window. CID

606 fragmentation was performed at 35% collision energy with fragments detected in the ion trap

607 between 400-1200 m/z. AGC was set to 5x103 and MS2 spectra were acquired in centroid

608 mode. TMT reporter ions were isolated for quantitation in MS3 using synchronous precursor

609 selection. Ten fragment ions were selected for MS3 using HCD at 53% collision energy.

610 Fragments were scanned in the Orbitrap at 60,000 fwhm between 100-500 m/z with AGC set

611 to 2x105. MS3 spectra were acquired in profile mode with injection parallelisation enabled.

612 Data processing and analysis

613 Raw MS files were processed using Proteome Discoverer 1.4.0.288 (Thermo Scientific).

614 Data were searched against a concatenated human (UniProt, downloaded on 04/11/13) and

615 HIV-1 (based on pNL4-3, Genbank: AF324493.2) database as previously described

616 (Matheson et al., 2015). VSVg (UniProt: P03522) and dEnv-EGFP-KDEL (inferred from the

617 pNL4-3-dE-EGFP sequence) were substituted for Env. Precursor mass tolerance and

618 fragment mass tolerance were set to 10 ppm and 0.6 Da, respectively, with a maximum of 2

619 missed tryptic cleavage sites. Percolator was used for post-processing of search results with

620 a peptide false discovery rate of 0.01. Observed reporter ion intensities were adjusted for lot-

621 specific isotopic impurities and missing quan values replaced with the minimum detected ion

622 intensity. Protein abundances were calculated using unique peptides and normalised

623 according to median protein ratios. 26

624 The complete HIV-1 infection time course mass spectrometry proteomics dataset has been

625 deposited to the ProteomeXchange Consortium (Vizcaino et al., 2013) via the PRIDE

626 Proteomics Identifications (Vizcaino et al., 2013) partner repository with the dataset identifier

627 PXD004187 (accessible at http://proteomecentral.proteomexchange.org).

628 For Gene Set Enrichment Analysis (GSEA), all proteins identified by >1 unique peptide were

629 analysed using GSEA v2.2.2 (downloaded from

630 http://software.broadinstitute.org/gsea/index.jsp) and KEGG Pathway

631 (c2.cp.kegg.v5.1.symbols.gmt) and Gene Ontology Biological Process

632 (c5.bp.v5.1.symbols.gmt) gene sets from the Molecular Signatures Database (MSigDB) v5.1

633 (Mootha et al., 2003; Subramanian et al., 2005). Pairwise comparisons were conducted

634 between uninfected cells and infected cells at each timepoint. Genes were ranked by

635 log2_Ratio_of_Classes and FDR q-values were calculated using 1,000 gene_set

636 permutations. For charting, nominal FDR q values of 0 were replaced with minimum values

637 for each pairwise comparison.

638 For clustering according to profiles of temporal expression, proteins identified by >1 unique

639 peptide and with a minimum fold change from baseline (0 h) of >2 were analysed using

640 Cluster 3.0 (downloaded from http://bonsai.hgc.jp/~mdehoon/software/cluster/software.htm)

641 (de Hoon et al., 2004; Eisen et al., 1998) and visualised using Java TreeView 1.1.6r4

642 (downloaded from http://jtreeview.sourceforge.net) (Saldanha, 2004). Data were expressed

643 as log2 (fold change in protein abundance compared with uninfected cells) and

644 agglomerative hierarchical clustering performed using uncentered Pearson correlation and

645 centroid linkage (Eisen et al., 1998; Weekes et al., 2014)

646 For functional analysis of proteins in clusters #1-4, enrichment of Gene Ontology Biological

647 Process and Molecular Function terms against a background of all proteins quantitated was

648 determined using the Database for Annotation, Visualization and Integrated Discovery 27

649 (DAVID) 6.7 (accessed on 29/4/16 at http://david.abcc.ncifcrf.gov) with default settings

650 (Huang da et al., 2009a, b). Annotation clusters with enrichment scores >1.3 (equivalent to a

651 geometric means of all included enrichment p values <0.05) were considered significant.

652 For the interactive spreadsheet of all TMT data, gene name aliases were added using

653 GeneALaCart (accessed on 4/5/16 at https://genealacart.genecards.org) (Rebhan et al.,

654 1997).

655 For comparison with PMP, Gene Ontology Cellular Component (GOCC) terms were

656 imported using Perseus 1.4.1.3 (downloaded from http://maxquant.org). The number of

657 plasma membrane proteins quantitated was estimated by counting proteins with GOCC

658 terms “plasma membrane”, “cell surface” and “extracellular” or with short, membrane-specific

659 GOCC terms but no subcellular assignment (Matheson et al., 2015). Glycosylation sites

660 were identified from the UniProt Knowledgebase (accessed on 4/12/15 at

661 http://www.uniprot.org). Experimentally identified N- and O-linked glycosylation sites and

662 predicted sites of N- and mucin-type O-linked glycosylation (using the NegNGlyc and

663 NetOGlyc tools) were included.

664 Stable Isotope Labelling with Amino Acids in Culture (SILAC)-based proteomic

665 validation time course

666 Sample preparation

667 For the SILAC-based validation time course, CEM-T4 T cells pre-labelled with heavy lysine

668 and arginine were spinoculated with VSVg-pseudotyped NL4-3-dE-EGFP HIV at an MOI of

669 10, and cells pre-labelled with medium lysine and arginine were mock-spinoculated without

670 virus. Aliquots of HIV-1-infected (heavy) and mock (medium) cells were harvested

671 sequentially at the indicated timepoints, dead cells removed using the Miltenyi Dead Cell

672 Removal kit, and equal cell numbers mixed prior to whole cell proteomic analysis. Cell lysis,

673 protein extraction and digestion and off-line peptide fractionation were carried out essentially 28

674 as for TMT-based whole cell proteomics, except that 100 mM Tris/HCl pH 7.4 was

675 substituted for 100 mM HEPES in lysis and wash buffers cells, 50mM ammonium

676 bicarbonate was substituted for 100 mM HEPES in digest buffer, and peptide eluates were

677 not subjected to TMT labelling. Cells were also subjected to PMP and this data has been

678 previously published (Matheson et al., 2015).

679 Mass spectrometry

680 Data for SILAC labelled samples were generated using a Q Exactive Orbitrap mass

681 spectrometer (Thermo Scientific). Peptides were fractionated using an RSLCnano 3000

682 (Thermo Scientific) with solvent A comprising 0.1 % formic acid and solvent B comprising 80

683 % MeCN, 20 % H20, 0.1 % formic acid. Peptides were loaded onto a 50 cm C18 EASYspray

684 column (Thermo Scientific) and eluted using a gradient rising from 10% to 36% B by 75 min

685 and 55% B by 100 min. MS data was acquired at 70,000 fwhm between 400-1650 m/z with

686 AGC of 1x106 and 250 ms injection time. MS2 data was acquired at 17,500 fwhm with AGC

687 of 5x104, 200 ms injection time and a loop count of 10. HCD fragmentation was performed at

688 NCE of 28 % and an underfill ratio of 20 %.

689 Data processing and analysis

690 Raw MS files were processed using MaxQuant 1.3.0.5. Data were searched against a

691 concatenated human (UniProt, downloaded 04/11/13) and HIV-1 (based on pNL4-3,

692 Genbank: AF324493.2) database as previously described (Matheson et al., 2015). VSVg

693 (UniProt: P03522) and dEnv-EGFP-KDEL (inferred from the pNL4-3-dE-EGFP sequence)

694 were substituted for Env. Fragment ion tolerance was set to 0.5 Da with a maximum of 2

695 missed tryptic cleavage sites. Carbamidomethyl (C) was defined as a fixed modification,

696 oxidation (M), acetylation (protein N-terminal) and deamidation (NQ) were selected as

697 variable modifications. A reversed decoy database was used with the false discovery rate for 29

698 both peptides and proteins set at 0.01. Peptide re-quantify was enabled and quantitation

699 utilized razor and unique peptides. Normalized protein ratios are reported.

700 For validation of downregulation or upregulation of proteins in clusters #1-4 at the indicated

701 timepoints, mean log2 (H/M protein abundance) for proteins in each cluster were compared

702 with 0 (no regulation) using 2-tailed 1-sample T-tests conducted using XLSTAT.

703 Tandem Mass Tag (TMT)-based whole cell proteomic/phosphoproteomic single

704 timepoint analysis

705 Sample preparation

706 For the TMT-based single timepoint analyses, CEM-T4 T cells were mock-spinoculated or

707 spinoculated in triplicate with VSVg-pseudotyped NL4-3-dE-EGFP wildtype and Vif-deficient

708 HIV at an MOI of 1.5. Cells were harvested 48 h after infection, and dead cells removed

709 using the Miltenyi Dead Cell Removal kit. 2x106 viable cells per condition were washed with

710 ice-cold PBS with Ca/Mg pH 7.4 (Sigma), lysed in 8M urea, 50 mM TEAB (triethylammonium

711 bicarbonate) pH 8.5 including phosphatase inhibitors (phosSTOP, Roche) and subjected to

712 10 rounds of sonication (30 s on/off) in a Diagenode Bioruptor sonicator at 4 °C. Lysate

713 protein concentrations were quantified using the Pierce BCA Protein Assay kit (Thermo

714 Scientific). 800 µg lysate/replicate was reduced with 10 mM TCEP for 20 min at room

715 temperature and alkylated with 20 mM iodoacetamide (IAM) for 20 min at room temperature

716 in the dark before quenching excess IAM with 15 mM DTT. Digestion was performed by first

717 adding LysC at a 1:100 enzyme:protein ratio and incubating at 30 °C for 3 h. This digest was

718 then diluted 4x with 50mM TEAB and trypsin was added at a 1:50 enzyme:protein ratio and

719 digested overnight at 37 °C with shaking on a Themomixer (Eppendorf). Digests were

720 subsequently acidified with TFA and cleaned up by SPE using SepPak C18 cartridges

721 (Waters). SPE eluates were divided into 50 µg and 750 µg-equivalent aliquots before drying

722 in a vacuum centrifuge. The 50 µg aliquots were resuspended in 100 mM TEAB prior to TMT 30

723 labelling essentially as per the manufacturer’s instructions. TMT-labelled samples were then

724 pooled and dried under vacuum prior to HpRP fractionation, and the sample pool used for

725 whole cell proteome analysis. The 750 µg aliquots were subjected to phosphopeptide

726 enrichment.

727 Phosphopeptide enrichment

728 750 µg-equivalent aliquots were resuspended in a loading solution of final concentration 4%

729 TFA, 1M glycolic acid and 80 % acetonitrile, added to 4.5 mg of 10 µm titanium dioxide resin

730 (Titansphere, GL Sciences) and shaken vigorously for 30 min. Beads were then washed for

731 5 min with vigorous shaking with 100 µL of the following solutions: loading solution, 1% TFA

732 80% ACN and 0.1% TFA 10% ACN. Enriched peptides were eluted with 100 µL 5 %

733 ammonium hydroxide for 20 min with vigorous shaking before being acidified with TFA and

734 formic acid. Acidified peptide pools were cleaned up by SPE using SepPak C18 cartridges

735 (Waters) before TMT labelling and HpRP fractionation.

736 Off-line High pH Reversed-Phase (HpRP) peptide fractionation

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

738 equipped with a 2.1 mm x 25 cm, 1.7 µ, Kinetex-Evo C18 column (Phenomenex). Solvent A

739 was 3% ACN, Solvent B was 100% ACN, solvent C was 200 mM ammonium formate (pH

740 10). Throughout the analysis solvent C was kept at a constant 10%. The flow rate was 400

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

742 a gradient elution of 0-50% B over 43 min followed by a 10 min wash with 90% B. 15 s

743 (100µL) fractions were collected throughout the run. Peptide containing fractions were

744 orthogonally recombined into 24 fractions (i.e. fractions 1, 25, 49, 73, 97 combined) and

745 dried in a vacuum centrifuge. Fractions were stored at -80 °C prior to analysis.

746 Mass spectrometry 31

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

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

749 resuspended 20 µl 5% DMSO 0.5% TFA. Samples were analysed using a nanoLC-MS

750 platform consisting of an Ultimate 3000 RSLC nano UHPLC coupled to an Orbitrap Fusion

751 instrument (Thermo Scientific). 50% of whole cell and 80% of phosphoproteome fractions

752 were loaded at 10 μl/min for 5 min on to an Acclaim PepMap C18 cartridge trap column (300

753 um x 5 mm, 5 um particle size) in 0.1% TFA. After loading a linear gradient of 3-32% solvent

754 B over 2h was used for sample separation over a column of the same stationary phase (75

755 µm x 50 cm, 2 µm particle size) before washing at 90% B and re-equilibration. Solvents were

756 A: 0.1% FA and B:ACN/0.1% FA. For whole cell proteome samples electrospray ionisation

757 was achieved by applying 2.1 kV directly to a stainless steel emitter (Thermo Scientific). For

758 phosphopeptide samples a distal coated silica emitter was used (New Objective).

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

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

761 for all parallelisable time. MS2: Quadrupole isolation at an isolation width of m/z 1.6, CID

762 fragmentation (NCE 35) with the ion trap scanning out in rapid mode from m/z 120, 5e3 AGC

763 target, 70 ms maximum injection time (150 ms for phosphopeptides), ions accumulated for

764 all parallelisable time in centroid mode. For phosphopeptides multistage activation was

765 enabled and set to trigger upon neutral loss of 79.9663 Da. MS3: In synchronous precursor

766 selection mode the top 10 MS2 ions were selected for HCD fragmentation (65NCE) and

767 scanned out in the orbitrap at 60’000 resolution with an AGC target of 2e4 and a maximum

768 accumulation time of 148 ms, ions were not accumulated for all parallelisable time. The

769 entire MS/MS/MS cycle had a target time of 2 s. Dynamic exclusion was set to +/- 10 ppm

770 for 60 s, MS2 fragmentation was trigged on precursor ions 5e3 counts and above.

771 Data processing and analysis 32

772 Spectra were searched by Mascot within Proteome Discoverer 2.1 against the UniProt

773 Human database (21/03/16). The database included forward and randomised reversed

774 Human database, the HIV proteome previously mentioned as well as a compendium of

775 common contaminants (GPM). The following search parameters were used. MS1 Tol: 10

776 ppm, MS2 Tol: 0.6Da, Fixed mods: Carbamidomethyl (C) and TMT (N-term, K), Var mods:

777 Oxidation (M), Enzyme: Trypsin (/P). Phosphopeptide samples also included variable

778 modification of Phosphorylation (STY). MS3 spectra were used for reporter ion based

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

780 using Mascot percolator and was controlled at 0.01 % for “high” confidence PSMs and

781 0.05% for “medium” confidence PSMs. Phosphopeptide site confidence was assessed using

782 the ptmRS node (the successor to phosphoRS (Taus et al., 2011). Reporter signal to noise

783 (s/n) with a cut-off of 10 was used for quantitation. Normalisation was automated and based

784 on total s/n in each channel. Protein/peptide abundance was calculated and output in terms

785 of “scaled” values, where the total s/n across all reporter channels is calculated and a

786 normalised contribution of each channel is output. Proteins/peptides satisfying at least a

787 “medium” FDR confidence were taken forth to statistical analysis in R. This consisted of a

788 moderated T-test (Limma) with Benjamini-Hochberg correction for multiple hypotheses to

789 provide a q value for each comparison (Schwammle et al., 2013).

790 For kinase activity profiling, pairwise comparisons in phosphopeptide abundance between

791 mock and WT or ΔVif HIV-infected cells were conducted using PhosFate (accessed at

792 http://phosfate.com/), including all kinases represented by >1 target phosphosite. For direct

793 comparison of WT and ΔVif HIV-infected cells, phosphopeptides spanning phosphorylation

794 sites annotated in the PhospPhosite database (accessed at http://www.phosphosite.org/)

795 were identified. Mean log2 (fold change in phosphopeptide abundance) was calculated for

796 each kinase represented by >4 target phosphosites. Further details of manually curated

797 AURKA, AURKB and PLK1 substrates are shown in Figure 6 – source data 2. 33

798 For functional analysis of proteins hyperphosphorylated in the presence of HIV infection,

799 enrichment of Gene Ontology Biological Process and Molecular Function terms against a

800 background of all identified phosphoproteins was determined using the Database for

801 Annotation, Visualization and Integrated Discovery (DAVID) 6.7 (accessed on 29/4/16 at

802 http://david.abcc.ncifcrf.gov) with default settings (Huang da et al., 2009a, b). Proteins

803 containing phosphopeptides significantly upregulated (q-values <0.01) in cells infected with

804 WT HIV-1 compared with mock infected cells were analysed. Annotation clusters with

805 enrichment scores >1.3 (equivalent to a geometric means of all included enrichment p-

806 values <0.05) were considered significant.

807 Stable cell lines

808 Stable 293T cell lines were generated by transduction with pHRSIN-PGK-puro-4xHA-

809 PPP2R5A-E or pHRSIN-PGK-puro-APOBEC3G-HA and selection with puromycin at 1

810 µg/ml. For knockdown experiments, 293T cells transduced with HA-PPP2R5B or

811 APOEC3G-HA were subsequently transduced with pHRSIREN (control or gene-specific

812 shRNA expression) and selected with hygromycin at 200 µg/ml.

813 Flow cytometry

814 Regulation of exogenous PPP2R5 subunits by Vif expression in stable 293T cell lines

815 293T cells stably expressing HA-PPP2R5A-E or APOBEC3G-HA were transfected in 24 well

816 plates using Fugene 6, with 200 ng Vif expression vector and 20 ng pMAXGFP (Lonza). 24 h

817 later, media was changed, and 12 h subsequently (36 h post-transfection) cells were

818 harvested for flow cytometry. Briefly, cells were dissociated using PBS/EDTA and fixed and

819 permeabilised using a commercial kit (Cytofix/Cytoperm, BD). Permeabilised cells were

820 stained with a fluorescently-conjugated anti-HA antibody, washed, and acquired on a BD

821 FACScalibur or BD LSRFortessa (BD Biosciences). Where indicated, DMSO or bortezomib

822 (10 nM) were added when the media was changed 24 h post-transfection. For CUL5 34

823 WT/DN co-transfection, 100 ng of Vif expression vector was used, with 100 ng of CUL5 WT

824 or DN expression vector. Where 293T cells expressing GFP-tagged PPP2R5B were used,

825 pCMV-SPORT6-mCherry was substituted for pMAXGFP, and cells were analysed without

826 permeabilisation.

827 Regulation of endogenous PPP2R5D by HIV infection of T cells

828 CEM-T4 T cells or activated primary human CD4+ T cells were analysed as described for

829 HA-PPP2R5A-E, but cells were infected with NL4-3-dE-EGFP WT or ΔVif viruses, and

830 stained with unconjugated anti-PPPR5D followed by an AF647-conjugated secondary

831 antibody. For time course analyses in CEM-T4 T cells, bortezomib (20 nM) or cycloheximide

832 (50 µg/ml) were added 24 h post-infection, and median fluorescence intensity (MFI) values

833 for GFP positive (infected) cells compared at four timepoints (0, 4, 8 and 12 h). Where

834 indicated, cells were stained for CD4 or ICAM3 without permeabilisation.

835 Antibodies

836 The following primary antibodies were used for immunoblot (alphabetical order): anti-

837 APOBEC3G (AIDS Reagent Program, Division of AIDS, NIAID, NIH from

838 Immunodiagnostics, 10069), anti-ß-catenin (ab32572, Abcam), anti-calreticulin (PA3-900,

839 Thermo), anti-FLAG (M2, Sigma), anti-HA (16B12, Biolegend), anti-p24 (ab9071, Abcam),

840 anti-PPP2R5A (ab89621, Abcam), anti-PPP2R5D (ab88075, Abcam), anti-PPP2R5D

841 (EPR15617, ab188323, Abcam), anti-RRM2 (ab57653, Abcam), anti-Vif (#319, AIDS

842 Reagent Program, Division of AIDS, NIAID, NIH: Dr. Michael Malim #6459 (Fouchier et al.,

843 1996; Simon et al., 1997).

844 The following primary antibodies were used for flow cytometry: anti-CD4-AF647 (clone

845 OKT4; BioLegend), anti-HA-DyLight 650 (16B12, ab117515, Abcam), anti-PPP2R5D

846 (EPR15617, ab188323, Abcam), anti-ICAM3 (TU41, 555957, BD). 35

847 The following primary antibodies were used for immunoprecipitation anti-PPP2R5D

848 (EPR15617, ab188323, Abcam).

849 The following secondary antibodies were used: goat anti-mouse-AF647 and donkey anti-

850 rabbit-AF647 (flow cytometry, Molecular Probes); goat anti-mouse-HRP and anti-rabbit-HRP

851 (immunoblot, Jackson ImmunoResearch).

852 Immunoblotting

853 CEM-T4 T cells or 293T cells were typically lysed in TBS/2 % SDS supplemented with

854 Benzonase (Sigma) to reduce lysate viscosity. Lysates were heated in Laemlli Loading

855 Buffer for 15 min at 95 °C, separated by SDS-PAGE and transferred to Immobilon-P

856 membrane (Millipore). Membranes were blocked in PBS/5 % non-fat dried milk (Marvel)/0.2

857 % Tween and probed with the indicated primary antibody overnight at 4°C. Reactive bands

858 were visualised using HRP-conjugated secondary antibodies and SuperSignal West Pico or

859 Dura chemiluminescent substrates (Thermo Scientific).

860 Immunoprecipitation

861 CEM-T4 T cells or 293T cells were lysed in 1 % NP-40. Lysates were pre-cleared with

862 Protein A-Sepharose (Sigma) or IgG-Sepharose (GE Healthcare) and incubated for 16 h at 4

863 °C with anti-HA coupled to agarose beads (Sigma EZview Red Anti-HA Affinity Gel) or anti-

864 PPP2R5D/Protein A-Sepharose (Sigma). After washing in 0.5% NP-40, samples were eluted

865 with 0.5 mg/ml HA peptide at 37 ºC for 1 h (anti-HA immunoprecipitation) or in Laemlli

866 Loading Buffer without DTT at 70 °C for 10 minutes (anti-PPP2R5D immunoprecipitation).

867 Samples were separated by SDS-PAGE, and immunoblotted as described.

868 Pulse-chase

869 CEM-T4 T cells were starved for 20 min in methionine-free, cysteine-free RPMI/5% dialysed

870 FCS (Invitrogen), labeled with [35S]methionine/[35S]cysteine (EasyTag EXPRESS,

871 PerkinElmer) for 15 min, then chased in RPMI/10% FCS at 37 °C. Cells were lysed in 1% 36

872 Triton X-100 at the indicated timepoints, and subjected to immunoprecipitation with anti-

873 PPP2R5D as described. Samples were separated by SDS-PAGE and processed for

874 autoradiography using the Packard Cyclone Storage Phosphor System.

875 Infectious viral release

876 pCMV-SPORT6 expression vectors encoding APOBEC3G, tetherin, TFAP4 and FMR1 were

877 obtained from the MGC/IMAGE clone collection (Dharmacon) with the following identifiers:

878 APOBEC3G (IMAGE:3905631), BST-2 (IMAGE:5217945), TFAP4 (IMAGE:4181538) and

879 FMR1 (IMAGE:30347992). As a control, mCherry was subcloned into pCMV-SPORT6. 293T

880 cells were transfected in 24 well plates using Fugene 6. Each well received a transfection

881 mix containing 135 ng NL4-3-dE-EGFP and 15 ng pMD.G. Shortly after, each well received

882 a second transfection mix of 150 ng pCMV-SPORT6 mCherry, APOBEC3G, tetherin, TFAP4

883 or FMR1. The media was changed 24 h post-infection, and 48 h post-infection, cell

884 supernatants were harvested. Contaminating 293T cells in the supernatants were removed

885 by centrifugation, and a small proportion used to infect a fixed number of HeLa cells. After 48

886 h, the HeLa cells were analysed by flow cytometry to determine the proportion that had

887 become GFP positive (infected). The MOI in each well was calculated and normalized to the

888 MOI resulting from the supernatants of 293T cells receiving mCherry.

889

890 37

891 Acknowledgments

892 The authors thank Dr Viviana Simon (Icahn School of Medicine at Mount Sinai), Dr Barbara

893 Blacklaws (University of Cambridge), and Prof Michael Malim (King’s College London) for

894 providing reagents, Dr Jenny Ho (Thermo) for help with proteomics, Dr Yagnesh Umrania

895 (University of Cambridge) for help with bioinformatics, Dr Reiner Schulte and the CIMR Flow

896 Cytometry Core Facility team, and the Lehner laboratory for critical discussion. This work

897 was supported by a Wellcome Trust PRF (101835/Z/13/Z) to PJL and RTF to NJM

898 (093964/Z/10/Z), the NIHR Cambridge BRC, a Wellcome Trust Strategic Award to CIMR,

899 and the Addenbrooke’s Charitable Trust. NJM is a Raymond and Beverly Sackler student.

900

901 Competing interests

902 The authors declare that no competing interests exist. 38

903 Figures

904 Figure 1. TMT-based proteomic time course analysis of HIV-infected cells

905 (A) Workflow of 6-plex TMT-based whole cell proteomic time course experiment. CEM-T4 T

906 cells were infected with NL4-3-dE-EGFP HIV at an MOI of 10. In subsequent figures

907 timepoints 1-5 show protein abundance 0, 6, 24, 48 and 72h after HIV infection (where 0h =

908 uninfected cells) and timepoint 6 shows protein abundance 72h after HIV infection in the

909 presence of reverse transcriptase inhibitors (RTi).

910 (B) Comparison of temporal profiles of Env-GFP obtained by proteomic (TMT) versus flow

911 cytometric quantitation. Cells from (A) were analysed by flow cytometry. Relative abundance

912 is expressed as fraction of maximum TMT reporter ion or fluorescence intensity. For linear

913 regression, log2 (fold change in protein abundance compared with uninfected cells) is

914 shown.

915 (C-D) Temporal profiles of viral proteins (C) and previously reported HIV targets (D). GAPDH

916 and β-actin are included as controls. Relative abundance is expressed as fraction of

917 maximum TMT reporter ion intensity.

918 See also Figure 1 – figure supplement 1 (Additional temporal profiles and comparison with

919 Plasma Membrane Profiling), Figure 1 – figure supplement 2 (Gene Set Enrichment

920 Analysis of HIV infection) and Figure 1 – source data 1 (Interactive spreadsheet of TMT

921 time course data).

922

923 Figure 2. Identification and SILAC-based proteomic validation of novel HIV targets

924 (A) Hierarchical cluster analysis of temporal profiles of proteins regulated by HIV. The

925 heatmap shows log2 (fold change in protein abundance compared with uninfected cells) and

926 clusters #1-4 are indicated. 39

927 (B) Average temporal profiles of proteins in clusters #1-4. Relative abundance is expressed

928 as fraction of maximum TMT reporter ion intensity.

929 (C) SILAC-based validation of novel HIV targets. CEM-T4 T cells pre-labelled with heavy

930 amino acids were infected with NL4-3-dE-EGFP HIV at an MOI of 10, and control cells pre-

931 labelled with medium amino acids were mock-infected without virus. Aliquots of HIV-infected

932 (heavy; H) and mock (medium; M) cells were harvested sequentially at the indicated

933 timepoints and subjected to SILAC-based whole cell proteomic analysis (Figure 2 – figure

934 supplement 1A). Log2 (H/M protein abundance) at 24, 48 and 72 h is shown for proteins

935 from clusters #1-4. *** p value <0.001.

936 (D) Temporal profiles of novel HIV-1 targets FMR1 and TFAP4. Relative abundance is

937 expressed as fraction of maximum TMT reporter ion intensity.

938 (E) Antagonism of HIV production by FMR1 and TFAP4. 293T cells were co-transfected with

939 pNL4-3-dE-EGFP/pMD.G and either mCherry or the indicated cellular protein. 48 h culture

940 supernatants were assayed for infectious virus by infection of HeLa cells. Well-characterised

941 restriction factors APOBEC3G and tetherin were included as controls, and infectious virus

942 release normalised compared with mCherry. Mean values and standard errors are shown

943 from at least 4 replicates. All four proteins significantly reduced viral release compared with

944 mCherry in an ANOVA analysis with Bonferroni post-test, p values <0.001.

945 See also Figure 2 – figure supplement 1 (workflow of SILAC-based proteomic time course

946 experiment, functional analysis of clusters #1-4 and prediction of novel Vpr targets) and

947 Figure 2 – source data 1 (clusters #1-4 summary proteomic data)

948

949 Figure 3. Cellular proteins progressively downregulated by HIV infection

950 (A) Enlargement of cluster #1 from hierarchical cluster analysis (Figure 2A). The heatmap

951 shows log2 (fold change in protein abundance compared with uninfected cells). Enzymes 40

952 associated with deoxynucleotide metabolism (blue) and B56 family regulatory subunits of

953 serine/threonine protein phosphatase PP2A (red) are highlighted, along with known Vif

954 target APOBEC3C (boxed) and other proteins of interest (bold).

955 (B) Temporal profiles of enzymes associated with deoxynucleotide metabolism (blue) and

956 B56 family regulatory subunits of serine/threonine protein phosphatase PP2A (red). Relative

957 abundance is expressed as fraction of maximum TMT reporter ion intensity, and the

958 temporal profile of APOBEC3C is shown for comparison.

959 See also Figure 3 – figure supplement 1 (immunoblot validation of novel HIV targets)

960

961 Figure 4. Vif-mediated depletion of PP2A-B56 family members PPP2R5A-E

962 (A) Proteomic analysis of CEM-T4 T cells infected with WT and ΔVif HIV. Cells were infected

963 with NL4-3-ΔE-EGFP viruses at an MOI of 1.5, and harvested 48 h post-infection.

964 Scatterplots display pairwise comparisons between WT, ΔVif and mock-infected cells. Each

965 point represents a single protein, plotted by its log2 (fold change in abundance) versus the

966 statistical significance of that change. q values were determined using Limma with

967 Benjamini-Hochberg adjustment for multiple testing, with increasing -log2 (q value) indicating

968 increasing significance. Points above the dotted line change with a q value <0.01. HIV

969 proteins and host proteins of interest are highlighted with different symbols (see key).

970 (B) Depletion of PPP2R5A and PPP2R5D during HIV infection. CEM-T4 T-cells were

971 infected with NL4-3-dE-EGFP WT and ΔVif viruses at an MOI of 1 or 10 and analysed by

972 immunoblot (IB) 48 h post-infection. p24 (capsid), Vif and calreticulin are included as

973 controls.

974 (C) Depletion of exogenous PPP2R5A by Vif. 293T cells stably expressing HA-PPP2R5A

975 were co-transfected with GFP plus empty vector, NL4-3 Vif or NL4-3 Vif with a single amino

976 acid mutation C114S and analysed by intracellular flow cytometry for HA 36 h post- 41

977 transfection. Histograms show GFP positive (transfected, red shading) and negative

978 (untransfected, blue line) cells. Median fluorescence intensity (MFI) values are shown for

979 GFP positive (red) and negative (blue) cells.

980 (D) Depletion of PPP2R5A-E family members by Vif. 293T cells stably expressing HA-

981 tagged PPP2R5A-E or APOBEC3G were co-transfected with GFP plus NL4-3 Vif expression

982 vectors, and intracellular HA staining quantitated by flow cytometry 36 h post transfection.

983 Histograms show GFP positive (transfected, red shading) and negative (untransfected, blue

984 line) cells. MFI values are shown for GFP positive (red) and negative (blue) cells.

985 See also Figure 4 – figure supplement 1 (workflow and controls for single timepoint

986 proteomic/phosphoproteomic experiment)

987

988 Figure 5. Mechanism of Vif-mediated degradation of PPP2R5A-E subunits

989 (A) Proteasomal degradation. 293T cells stably expressing HA-PPP2R5A were transfected

990 with NL4-3 Vif in the presence of DMSO (control) or the proteasome inhibitor bortezomib and

991 analysed by intracellular flow cytometry for HA.

992 (B) CUL5-dependent degradation. 293T cells stably expressing GFP-PPP2R5B were co-

993 transfected with NL4-3 Vif plus empty vector, wildtype cullin-5 (CUL5 WT) or a dominant

994 negative cullin-5 mutant (CUL5 DN) and analysed by flow cytometry for GFP.

995 (C) CUL5 complex-dependent degradation. 293T cells stably expressing HA-PPP2R5B

996 (upper panels) or HA-APOBEC3G (lower panels) were transduced with the indicated

997 shRNA. Cells were then transfected with NL4-3 Vif and analysed by intracellular flow

998 cytometry for HA. Green/red shading shows Vif-transfected cells in the indicated shRNA

999 background. Red lines showing HA staining in cells transduced with control shRNA are

1000 included in each panel for reference. 42

1001 In all experiments, cells were analysed 36 h post-transfection, and transfected cells

1002 determined by co-transfection with GFP (A and C) or mCherry (B). MFI values are shown for

1003 transfected (red/green) and untransfected (blue) cells.

1004

1005 Figure 6. Global phosphoproteomic analysis of cells infected with WT or ∆Vif HIV

1006 (A) Vif-dependent changes in peptide and phosphopeptide abundance. CEM-T4 T cells from

1007 Figure 4A and Figure 4 – figure supplement 1A were subjected to TMT-based

1008 phosphoproteomic analysis. Scatterplots display differences in protein (left panel, as in

1009 Figure 4A, right panel) and phosphopeptide abundance (right panel) between WT and

1010 ΔVif-infected cells. Each point represents a single protein or phosphopeptide, plotted by its

1011 log2 (fold change in abundance) versus the statistical significance of that change. q values

1012 were determined using Limma with Benjamini-Hochberg adjustment for multiple testing, with

1013 increasing -log2 (q value) indicating increasing significance. Proteins and phosphopeptides

1014 downregulated (red) or upregulated (green) with a fold change >2 and q value <0.01 are

1015 highlighted.

1016 (B) Comparison of changes in phosphopeptide abundance between WT and ΔVif-infected

1017 CEM-T4 T cells with previously published data for okadaic acid-treated HeLa cells (Kauko et

1018 al., 2015). Lines show linear correlation with associated 95% confidence areas, r2 values

1019 and p values of a non-zero correlation.

1020 (C) Analysis of changes in phosphopeptide abundance between WT and ΔVif-infected cells

1021 CEM-T4 T cells using the PhosphoSite kinase-substrate database. Bars show log2 (fold

1022 change in phosphopeptide abundance) for peptides spanning known kinase substrate sites.

1023 Error bars show the standard error of the mean.

1024 (D) Comparison of changes in phosphopeptide abundance between WT and ΔVif-infected

1025 CEM-T4 T cells with previously published data for kinase inhibitor-treated HeLa cells 43

1026 (Kettenbach et al., 2011). At low concentrations, MLN8054 is a selective AURKA inhibitor,

1027 but at 5 μM (as shown) reduced activity of AURKB and PLK1 is also observed. Lines show

1028 linear correlation with associated 95% confidence areas, r2 values and p values of a non-

1029 zero correlation.

1030 (E) Vif-specific hyperphosphorylation of aurora kinase substrates. Protein abundances of

1031 PLK1, AURKA and AURKB were compared with normalised abundances of manually

1032 curated phosphopeptides targeted by the respective kinases. Abundances of kinase proteins

1033 were compared using Limma with Benjamini-Hochberg adjustment for multiple testing.

1034 Abundances of target phosphopeptides were compared by Repeated Measures ANOVA with

1035 Bonferroni post-test. N.S., p value>0.05; *p value<0.05; **p value<0.01; ***p value<0.001.

1036 See also Figure 6 – figure supplement 1 (further phosphoproteomic analysis), Figure 6 –

1037 source data 1 (single timepoint phosphoproteomic data) and Figure 6 – source data 2

1038 (previously reported AURAKA, AURAKB and PLK1 targets)

1039

1040 Figure 7. Phylogenetic conservation of PPP2R5A-E subunit degradation

1041 (A) Phylogenetic tree based on amino acid alignment of Vif variants used in this study.

1042 (B) Conservation of PPP2R5B subunit degradation by phylogenetically diverse lentiviral Vif

1043 variants. 293T cells stably expressing HA-PPP2R5B were transfected with a panel of

1044 lentiviral Vif variants and analysed by intracellular flow cytometry for HA 36 h post-

1045 transfection. Median fluorescence intensity of the transfected population is shown as a

1046 proportion of median fluorescence intensity of the untransfected population for each

1047 condition, normalized to the empty vector control. Datapoints represent mean values for

1048 different Vif variants obtained from up to four independent experiments.

1049 (C) Depletion of PPP2R5A-E subunits by small ruminant lentivirus Vif. 293T cells stably

1050 expressing HA-PPP2R5A-E or HA-APOBEC3G were transfected with NL4-3 (HIV-1) or 44

1051 SRLV Vif variants. Histograms show GFP positive (transfected, red shading) and negative

1052 (untransfected, blue line) cells. MFI values are shown for GFP positive (red) and negative

1053 (blue) cells.

1054 See also Figure 7 – figure supplement 1 (additional data on phylogenetic conservation of

1055 PPP2R5A-E subunit degradation) and Figure 7 – figure supplement 2 (identity/similarity

1056 matrix of lentiviral Vif variants)

1057

1058 45

1059 Figure supplements

1060 Figure 1 – figure supplement 1. Additional temporal profiles and comparison with

1061 Plasma Membrane Profiling

1062 (A) Comparison of temporal profiles of cell surface Nef- (CD4 and HLA-A) and Vpu- (CD4,

1063 tetherin and SNAT1) targets obtained by whole cell or plasma membrane proteomics.

1064 Expression levels from our whole cell proteomic analysis (WCP, red) and our previous

1065 Plasma Membrane Profiling analysis ((Matheson et al., 2015); PMP, blue) are shown.

1066 Relative abundance is expressed as fraction of maximum TMT reporter ion intensity.

1067 (B) Dynamic range of protein regulation observed by whole cell or plasma membrane

1068 proteomics. Histograms show frequencies of log2 (fold change compared with

1069 mock/uninfected cells) values for proteins quantitated in our whole cell proteomic analysis

1070 (WCP, red) and our previous Plasma Membrane Profiling analysis ((Matheson et al., 2015);

1071 PMP, blue) at 24, 48 and 72 h. All proteins identified by >1 unique peptide in both WCP and

1072 PMP experiments are shown.

1073 (C-G) Temporal profiles of selected control proteins (C-E), actin regulatory proteins (F) and

1074 mitotic kinases (G). Relative abundance is expressed as fraction of maximum TMT reporter

1075 ion intensity.

1076 (H) Quantitation of plasma membrane proteins by whole cell or plasma membrane

1077 proteomics. The pie chart shows overlap of proteins with Gene Ontology Cellular

1078 Compartment annotations indicative of plasma membrane localisation quantitated in our

1079 whole cell proteomic analysis (WCP, red) and our previous Plasma Membrane Profiling

1080 analysis ((Matheson et al., 2015); PMP, blue). Protein numbers are indicated. The bar chart

1081 details glycosylation status of proteins quantitated in WCP (red), PMP (blue) or both (tan)

1082 experiments. Glycosylation sites were identified from the UniProt Knowledgebase (accessed

1083 on 4/12/15 at http://www.uniprot.org). Protein numbers (glycosylated/total) are indicated. 46

1084

1085 Figure 1 – figure supplement 2. Gene Set Enrichment Analysis of HIV infection

1086 (A) Pathways and processes upregulated by HIV infection. KEGG Pathway and Gene

1087 Ontology Biological Process gene sets enriched in infected cells at the indicated timepoints

1088 compared with uninfected cells were determined using GSEA. Indicative false discovery rate

1089 (FDR) thresholds for gene set enrichment are shown and gene sets related to lipid

1090 metabolism are highlighted (green).

1091 (B) Pathways and processes downregulated by HIV infection. KEGG Pathway and Gene

1092 Ontology Biological Process gene sets enriched in uninfected cells compared with infected

1093 cells at the indicated timepoints were determined using GSEA. Indicative FDR thresholds for

1094 gene set enrichment are shown and gene sets related to RNA processing are highlighted

1095 (green).

1096 (C-D) Expression levels of all quantitated proteins in Lipid_Metabolic_Process (C) and

1097 RNA_Processing (D) gene sets. Log2 (fold change in protein abundance compared with

1098 uninfected cells) is shown at the indicated timepoints.

1099

1100 Figure 2 – figure supplement 1. Workflow of SILAC-based proteomic time course

1101 experiment, functional analysis of clusters #1-4 and prediction of novel Vpr targets

1102 (A) Workflow of SILAC-based proteomic validation time course experiment. CEM-T4 T cells

1103 pre-labelled with heavy amino acids were infected with NL4-3-dE-EGFP HIV at an MOI of

1104 10, and control cells pre-labelled with medium amino acids were mock-infected without virus.

1105 Aliquots of HIV-infected (heavy) and mock (medium) cells were harvested sequentially at the

1106 indicated timepoints and subjected to SILAC-based whole cell proteomic analysis. 47

1107 (B) Functional annotation clusters enriched amongst proteins from clusters #1-4. Enrichment

1108 of Gene Ontology Molecular Function and Biological Process terms against a background of

1109 all quantitated proteins was determined using DAVID. Functional annotation clusters with

1110 enrichment scores >1.3 (equivalent to a geometric mean of all included enrichment p values

1111 <0.05) were considered significant. Representative Gene Ontology terms are indicated.

1112 (C) Temporal profiles of Vpr targets UNG and HLTF. Relative abundance is expressed as

1113 fraction of maximum TMT reporter ion intensity.

1114 (D) Enlargement of part of cluster #2 from hierarchical cluster analysis (Figure 2A). The

1115 heatmap shows log2 (fold change in protein abundance compared with uninfected cells).

1116 Previously characterised Vpr targets UNG and HLTF are highlighted (red).

1117

1118 Figure 3 – figure supplement 1. Immunoblot validation of novel HIV targets

1119 Depletion of proteins in cluster #1 by HIV-1 infection. CEM-T4s were infected with NL4-3-dE-

1120 EGFP HIV at an MOI of 10 and analysed by immunoblot (IB) 48 h post-infection. Depletion

1121 of positive control (CD4 and APOBEC3G) and novel (RRM2 and PPP2R5A/D) HIV targets

1122 was confirmed. ImageJ (Schneider et al., 2012) was used to determine band intensities,

1123 which were normalised to calreticulin intensity and compared with TMT-based proteomic

1124 quantitation at 48 h. p24 (capsid) and Vif are included as additional controls.

1125

1126 Figure 4 – figure supplement 1. Workflow and controls for single timepoint

1127 proteomic/phosphoproteomic experiment

1128 (A) Workflow of TMT-based single timepoint whole cell proteomic and phosphoproteomic

1129 experiment. CEM-T4 T cells were mock-infected or infected with NL4-3-dE-EGFP WT or

1130 ΔVif HIV at an MOI of 1.5. Cells were harvested for proteomic analysis 48 h post-infection. 48

1131 Samples were subjected to (1) whole cell proteome and (2) phosphopeptide analysis. Each

1132 condition was carried out in triplicate.

1133 (B) Quantitation of infected cells from experiment described in (A). Cells were analysed by

1134 flow cytometry for CD4 and GFP expression, with infected cells losing CD4 and gaining

1135 GFP. Example flow cytometric analysis of one replicate for each condition is shown,

1136 demonstrating the % infected cells. Across all replicates, cells infected with WT or ΔVif

1137 viruses were 74-78 % infected.

1138

1139 Figure 4 – figure supplement 2. Depletion of endogenous PPP2R5D during HIV

1140 infection of primary cells

1141 (A-B) CEM-T4 T cells (A) or activated primary human CD4+ T cells (B) were infected with

1142 NL4-3-dE-EGFP WT and ΔVif viruses at an MOI of 1 and analysed by intracellular flow

1143 cytometry for PPP2R5D 48 h post-infection. ICAM3 is included as a control. MFI values are

1144 shown for ΔVif (blue) and WT (red) viruses.

1145

1146 Figure 5 – figure supplement 1. Co-immunoprecipitation of Vif and PPP2R5D

1147 (A) Co-immunoprecipitation in 293Ts. WT 293T cells or 293T cells stably expressing HA-

1148 tagged PPP2R5D were transfected with FLAG-tagged NL4-3 Vif, pre-treated with

1149 bortezomib (10 nM) for 16 h, and analysed by immunoblot (IB) for HA-PPP2R5D and FLAG-

1150 Vif 48 h post-infection (left panels). Lysates were subjected to immunoprecipitation (IP) with

1151 anti-HA (middle panels) or anti-PPP2R5D (right panels, WT 293Ts only) and re-analysed by

1152 immunoblot. 293T cells transfected with empty vector or FLAG-tagged K5 protein of Kaposi's

1153 sarcoma-associated herpesvirus (KSHV) were included as controls. 49

1154 (B) Co-immunoprecipitation during HIV infection of T cells. CEM-T4 T cells were infected

1155 with NL4-3-dE-EGFP HIV at an MOI of 1.5, pre-treated with bortezomib (10 nM) for 16 h,

1156 and analysed by immunoblot (IB) for PPP2R5D and Vif 48 h post-infection (left panels).

1157 Lysates were subjected to immunoprecipitation (IP) with anti-PPP2R5D (right panels) and

1158 re-analysed by immunoblot. Uninfected CEM-T4 T cells and infected CEM-T4 T cells without

1159 bortezomib pre-treatment were included as controls.

1160

1161 Figure 5 – figure supplement 2. Time course analysis of endogenous PPP2R5D during

1162 HIV infection of T cells

1163 (A-B) Rescue by proteasome inhibition. CEM-T4 T cells were infected with NL4-3-dE-EGFP

1164 HIV at an MOI of 1.5 and analysed by flow cytometry (A) at the indicated timepoints in the

1165 presence (bottom panels) or absence (top panels) of bortezomib (BZB) added 24 h post-

1166 infection (+0 h). MFI values (PPP2R5D staining) for GFP positive (infected) cells are shown,

1167 normalized to values at +0 h (B).

1168 (C-D) Cycloheximide chase. CEM-T4 T cells were infected with NL4-3-dE-EGFP WT (top

1169 panels) and ΔVif (bottom panels) viruses at an MOI of 1.5 and analysed by flow cytometry

1170 (C) at the indicated timepoints in the presence of cycloheximide (CHX) added 24 h post-

1171 infection (+0 h). MFI values (PPP2R5D staining) for GFP positive (infected) cells are shown,

1172 normalized to values at +0 h (D). As predicted, the presence of CHX inhibited the production

1173 of new Env-EGFP protein.

1174

1175 Figure 5 – figure supplement 3. Pulse-chase analysis of endogenous PPP2R5D during

1176 HIV infection of T cells

1177 CEM-T4 T cells were infected with NL4-3-dE-EGFP WT or ΔVif viruses at an MOI of 1.5 and

1178 pulsed with [35S]methionine/[35S]cysteine 48 h post-infection. Cells were chased until the 50

1179 indicated timepoints, subjected to immunoprecipitation (IP) with anti-PPP2R5D and analysed

1180 by autoradiography.

1181

1182 Figure 6 – figure supplement 1. Further phosphoproteomic analysis

1183 (A) Remodelling of the cellular phosphoproteome by HIV infection. CEM-T4 T cells from

1184 Figure 4A and Figure 4 – figure supplement 1A were subjected to TMT-based

1185 phosphoproteomic analysis. The scatterplot displays differences in phosphopeptide

1186 abundance between WT HIV and mock-infected cells. Each point represents a single protein

1187 or phosphopeptide, plotted by its log2 (fold change in abundance) versus the statistical

1188 significance of that change. q values were determined using Limma with Benjamini-

1189 Hochberg adjustment for multiple testing, with increasing -log2 (q value) indicating

1190 increasing significance. Proteins and phosphopeptides downregulated (red) or upregulated

1191 (green) with a fold change >2 and q value <0.01 are highlighted.

1192 (B) Functional annotation clusters enriched amongst proteins hyperphosphorylated in the

1193 presence of HIV infection. Proteins containing phosphopeptides significantly upregulated (q

1194 values <0.01) in cells infected with WT HIV compared with mock-infected cells were

1195 analysed. Enrichment of Gene Ontology Molecular Function and Biological Process terms

1196 against a background of all identified phosphoproteins was determined using DAVID.

1197 Functional annotation clusters with enrichment scores >1.3 (equivalent to a geometric mean

1198 of all included enrichment p-values <0.05) were considered significant. Representative Gene

1199 Ontology terms are indicated.

1200 (C) PhosFate analysis of kinase activity in HIV-infected versus mock-infected cells. Data are

1201 shown for WT HIV (upper panel) and ΔVif HIV (lower panel). A positive activity score

1202 indicates enhanced phosphorylation of kinase-specific phosphosites in infected cells. Aurora

1203 kinases A and B (AurA/AurB; red) and other control mitotic/checkpoint kinases (PLK1, ATR 51

1204 and ATM; blue) are highlighted. Kinases represented in the dataset by a single target

1205 phosphosite were excluded.

1206 (D-E) Comparison of phosphoproteomic dataset with previously published data for (D)

1207 okadaic acid-treated (Kauko et al., 2015) and (E) kinase inhibitor-treated (Kettenbach et al.,

1208 2011) HeLa cells. Each row shows a different pairwise comparison: top row, HIV WT versus

1209 mock; middle row, HIV ΔVif vs mock; bottom row, HIV WT vs HIV ΔVif. Each column shows

1210 a different inhibitor treatment. At low concentrations, MLN8054 is a selective AURKA

1211 inhibitor, but at 5 μM (as shown) reduced activity of AURKB and PLK1 is also observed.

1212 AZDZM indicates a combined analysis of selective AURKB inhibitors AZD1152 and

1213 ZM447439. BI2536 is a selective PLK1-3 inhibitor. Each scatterplot compares log2 (fold

1214 change) in WT/ ΔVif/mock-infected cells (y axis) with log2 (fold change) in inhibitor treated

1215 cells (x axis). Lines indicate linear correlation with 95% confidence areas, r2 values and p

1216 values of a non-zero correlation. For each column, the most significant correlation is

1217 highlighted (red).

1218

1219 Figure 7 – figure supplement 1. Identity/similarity matrix of lentiviral Vif variants

1220 Upper-right half of matrix shows pairwise similarity between Vif variants, lower-right half

1221 shows pairwise identity. For pairwise comparisons with NL4-3 Vif, relevant similarity values

1222 (black lines) and identity values (white lines) are highlighted. Matrix is based on a Vif amino

1223 acid alignment carried out with the PSI-Coffee variant of the T-Coffee alignment algorithm

1224 (Notredame et al., 2000) and the SIAS tool available at

1225 http://imed.med.ucm.es/Tools/sias.html using default settings.

1226

1227 Figure 7 – figure supplement 2. Additional data on phylogenetic conservation of

1228 PPP2R5A-E subunit degradation 52

1229 (A) Conservation of PPP2R5A subunit degradation by HIV-1 Vif variants. 293T cells stably

1230 expressing HA-PPP2R5A were transfected with the indicated Vif variants and analysed by

1231 intracellular flow cytometry for HA 36 h post-transfection. Median fluorescence intensity of

1232 the transfected population is shown as a proportion of median fluorescence intensity of the

1233 untransfected population for each condition, normalized to the empty vector control. Mean

1234 values and standard errors are shown.

1235 (B) Conservation of PPP2R5A-E subunit degradation by phylogenetically diverse lentiviral

1236 Vif variants. 293T cells stably expressing different HA-tagged PPP2R5A-E subunits were

1237 transfected with the indicated Vif variants and analysed by intracellular flow cytometry for HA

1238 36 h post-transfection. Median fluorescence intensity of the transfected population is shown

1239 as a proportion of median fluorescence intensity of the untransfected population for each

1240 condition, normalized to the empty vector control. Each datapoint represents a different Vif

1241 variant.

1242 (C) Representative data for primate lentiviral Vif degradation of PPP2R5B. Data shown in

1243 Figure 7B was acquired in several experiments, one example is shown here. Histograms

1244 show GFP positive (transfected, red shading) and negative (untransfected, blue line) cells.

1245 MFI values are shown for GFP positive (red) and negative (blue) cells.

1246

1247 Figure 7 – figure supplement 3. Mechanism of PPP2R5E degradation by SRLV Vif

1248 (A) Co-immunoprecipitation of SRLV Vif with PPP2R5E. WT 293T cells or 293T cells stably

1249 expressing HA-tagged PPP2R5E were transfected with FLAG-tagged SRLVE Vif, pre-

1250 treated with bortezomib (10 nM) for 16 h, and analysed by immunoblot (IB) for HA-PPP2R5E

1251 and FLAG-Vif 48 h post-infection (left panels). Lysates were subjected to

1252 immunoprecipitation (IP) with anti-HA and re-analysed by immunoblot (right panels). 293T

1253 cells transfected with empty vector were included as controls. 53

1254 (B) CBFβ-independent degradation of PPP2R5E by SRLV Vif. 293T cells stably expressing

1255 HA-PPP2R5E were transduced with the indicated shRNA. Cells were then transfected with

1256 NL4-3 Vif (upper panels) or SRLV Vif (lower panels) and analysed by intracellular flow

1257 cytometry for HA. Red/green shading shows Vif-transfected cells in the indicated shRNA

1258 background. Red lines showing HA staining in cells transduced with control shRNA are

1259 included in each panel for reference. Cells were analysed 36 h post-transfection, and

1260 transfected cells determined by co-transfection with GFP. MFI values are shown for

1261 transfected (red/green) and untransfected (blue) cells.

1262 54

1263 Source data

1264 Figure 1 – source data 1. Interactive spreadsheet of TMT time course data

1265 Interactive spreadsheet enabling generation of temporal profiles of protein abundance for

1266 any quantitated genes of interest (“Gene search and plots” worksheet). Detailed instructions

1267 are incorporated into the spreadsheet. The complete (unfiltered) TMT-based proteomic time

1268 course dataset (“Complete TMT time course data” worksheet) and a database of gene name

1269 aliases (“Gene name aliases” worksheet) are also included. Protein abundance is depicted

1270 on a colour scale (red = downregulated; green = upregulated). The number of unique

1271 peptides, peptides and peptide spectral matches are specified for each protein, along with

1272 ratio counts and variability for each TMT condition.

1273

1274 Figure 2 – source data 1. Clusters #1-4 summary proteomic data

1275 Spreadsheet of all proteomic data for proteins in clusters #1-4. Each cluster is represented

1276 by a single worksheet, with proteins ranked according to the hierarchical cluster analysis

1277 shown (Figure 2A) and protein abundance depicted on a colour scale (red = downregulated;

1278 green = upregulated). As well as data from the TMT-based proteomic time course

1279 experiment (Figure 1 and Figure 1 – source data 1), additional data from the SILAC-based

1280 proteomic validation (Figure 2C and Figure 2 – figure supplement 1A) and TMT-based

1281 single timepoint experiment (Figure 4A and Figure 4 – figure supplement 1A) are also

1282 included. The number of unique peptides is shown for each protein (TMT experiments) and

1283 each timepoint (SILAC experiment). q values for the TMT-based single timepoint experiment

1284 were determined using Limma with Benjamini-Hochberg adjustment for multiple testing, with

1285 q values <0.01 highlighted in gold.

1286 55

1287 Figure 6 – source data 1. Single timepoint phosphoproteomic data

1288 Spreadsheet of cellular phosphopeptides identified in mock, WT and ΔVif HIV-infected cells

1289 in the TMT-based single timepoint phosphoproteomic experiment (Figure 6, Figure 4 –

1290 figure supplement 1A and Figure 6 – figure supplement 1). Peptide sequence, details of

1291 the cognate protein and position of the peptide within the protein are shown. The column

1292 “Phosphosite Probabilities” indicates the probability that each serine, threonine or tyrosine

1293 within the peptide is phosphorylated. The amino acid is stated (S, serine; T, threonine; Y,

1294 tyrosine) with the position in the peptide in parentheses, followed by the probability (%).

1295 Each potential phosphosite is separated by a semicolon. Phosphosites with a probability of

1296 over 75 % are listed in the column “Modifications in Master Proteins”, which shows a

1297 summary of the phosphorylated amino acids identified and their position in the protein. Log2

1298 (fold change) compares phosphopeptide abundance normalized to total protein abundance,

1299 with abundance depicted on a colour scale (red = downregulated; green = upregulated). q

1300 values were determined using Limma with Benjamini-Hochberg adjustment for multiple

1301 testing, with q values <0.01 highlighted in gold.

1302

1303 Figure 6 – source data 2. Previously reported AURKA, AURKB and PLK1 targets

1304 AURKA, AURKB and PLK1 targets were manually curated from the literature. Peptides listed

1305 overlap reported sites of phosphorylation by AURKA, AURAKB or PLK1, or are the only

1306 phosphopeptide identified from a protein known to be phosphorylated by one of these

1307 kinases, but at an unknown site. Peptides where the identified phosphorylation site explicitly

1308 matches the one reported are plotted in Figure 6E (blue shading; excluded peptides

1309 highlighted in red text). Studies cited: (Asano et al., 2013; Dephoure et al., 2008; Hengeveld

1310 et al., 2012; Kettenbach et al., 2011; Santamaria et al., 2011; Welburn et al., 2010; Yu et al.,

1311 2005). 56

1312 References

1313 Ai, Y., Zhu, D., Wang, C., Su, C., Ma, J., Ma, J., and Wang, X. (2014). Core-binding factor

1314 subunit beta is not required for non-primate lentiviral Vif-mediated APOBEC3 degradation.

1315 Journal of virology 88, 12112-12122.

1316 Arainga, M., Guo, D., Wiederin, J., Ciborowski, P., McMillan, J., and Gendelman, H.E.

1317 (2015). Opposing regulation of endolysosomal pathways by long-acting nanoformulated

1318 antiretroviral therapy and HIV-1 in human macrophages. Retrovirology 12, 5.

1319 Asano, E., Hasegawa, H., Hyodo, T., Ito, S., Maeda, M., Takahashi, M., Hamaguchi, M., and

1320 Senga, T. (2013). The Aurora-B-mediated phosphorylation of SHCBP1 regulates cytokinetic

1321 furrow ingression. J Cell Sci 126, 3263-3270.

1322 Ayinde, D., Casartelli, N., and Schwartz, O. (2012). Restricting HIV the SAMHD1 way:

1323 through nucleotide starvation. Nat Rev Microbiol 10, 675-680.

1324 Baldauf, H.M., Pan, X., Erikson, E., Schmidt, S., Daddacha, W., Burggraf, M., Schenkova,

1325 K., Ambiel, I., Wabnitz, G., Gramberg, T., et al. (2012). SAMHD1 restricts HIV-1 infection in

1326 resting CD4(+) T cells. Nat Med 18, 1682-1687.

1327 Bastos, R.N., Cundell, M.J., and Barr, F.A. (2014). KIF4A and PP2A-B56 form a spatially

1328 restricted feedback loop opposing Aurora B at the anaphase central spindle. J Cell Biol 207,

1329 683-693.

1330 Bensimon, A., Aebersold, R., and Shiloh, Y. (2011). Beyond ATM: The protein kinase

1331 landscape of the DNA damage response. Febs Letters 585, 1625-1639.

1332 Bergeron, J.R., Huthoff, H., Veselkov, D.A., Beavil, R.L., Simpson, P.J., Matthews, S.J.,

1333 Malim, M.H., and Sanderson, M.R. (2010). The SOCS-box of HIV-1 Vif interacts with

1334 ElonginBC by induced-folding to recruit its Cul5-containing ubiquitin ligase complex. PLoS

1335 Pathog 6, e1000925. 57

1336 Besnard-Guerin, C., Belaidouni, N., Lassot, I., Segeral, E., Jobart, A., Marchal, C., and

1337 Benarous, R. (2004). HIV-1 Vpu sequesters beta-transducin repeat-containing protein

1338 (betaTrCP) in the cytoplasm and provokes the accumulation of beta-catenin and other

1339 SCFbetaTrCP substrates. J Biol Chem 279, 788-795.

1340 Binka, M., Ooms, M., Steward, M., and Simon, V. (2012). The activity spectrum of Vif from

1341 multiple HIV-1 subtypes against APOBEC3G, APOBEC3F, and APOBEC3H. J Virol 86, 49-

1342 59.

1343 Bischoff, J.R., Anderson, L., Zhu, Y., Mossie, K., Ng, L., Souza, B., Schryver, B., Flanagan,

1344 P., Clairvoyant, F., Ginther, C., et al. (1998). A homologue of Drosophila aurora kinase is

1345 oncogenic and amplified in human colorectal cancers. EMBO J 17, 3052-3065.

1346 Blas-Rus, N., Bustos-Moran, E., Perez de Castro, I., de Carcer, G., Borroto, A., Camafeita,

1347 E., Jorge, I., Vazquez, J., Alarcon, B., Malumbres, M., et al. (2016). Aurora A drives early

1348 signalling and vesicle dynamics during T-cell activation. Nat Commun 7, 11389.

1349 Chan, E.Y., Qian, W.J., Diamond, D.L., Liu, T., Gritsenko, M.A., Monroe, M.E., Camp, D.G.,

1350 2nd, Smith, R.D., and Katze, M.G. (2007). Quantitative analysis of human immunodeficiency

1351 virus type 1-infected CD4+ cell proteome: dysregulated cell cycle progression and nuclear

1352 transport coincide with robust virus production. J Virol 81, 7571-7583.

1353 Chan, E.Y., Sutton, J.N., Jacobs, J.M., Bondarenko, A., Smith, R.D., and Katze, M.G.

1354 (2009). Dynamic host energetics and cytoskeletal proteomes in human immunodeficiency

1355 virus type 1-infected human primary CD4 cells: analysis by multiplexed label-free mass

1356 spectrometry. J Virol 83, 9283-9295.

1357 Chang, S.T., Sova, P., Peng, X., Weiss, J., Law, G.L., Palermo, R.E., and Katze, M.G.

1358 (2011). Next-generation sequencing reveals HIV-1-mediated suppression of T cell activation

1359 and RNA processing and regulation of noncoding RNA expression in a CD4+ T cell line.

1360 MBio 2. 58

1361 de Hoon, M.J., Imoto, S., Nolan, J., and Miyano, S. (2004). Open source clustering software.

1362 Bioinformatics 20, 1453-1454.

1363 Dephoure, N., Zhou, C., Villen, J., Beausoleil, S.A., Bakalarski, C.E., Elledge, S.J., and Gygi,

1364 S.P. (2008). A quantitative atlas of mitotic phosphorylation. Proc Natl Acad Sci U S A 105,

1365 10762-10767.

1366 Desimmie, B.A., Delviks-Frankenberrry, K.A., Burdick, R.C., Qi, D., Izumi, T., and Pathak,

1367 V.K. (2014). Multiple APOBEC3 restriction factors for HIV-1 and one Vif to rule them all. J

1368 Mol Biol 426, 1220-1245.

1369 Ding, L., Gu, H., Gao, X., Xiong, S., and Zheng, B. (2015). Aurora kinase a regulates m1

1370 macrophage polarization and plays a role in experimental autoimmune encephalomyelitis.

1371 Inflammation 38, 800-811.

1372 Doehle, B.P., Schafer, A., and Cullen, B.R. (2005). Human APOBEC3B is a potent inhibitor

1373 of HIV-1 infectivity and is resistant to HIV-1 Vif. Virology 339, 281-288.

1374 Duggal, N.K., and Emerman, M. (2012). Evolutionary conflicts between viruses and

1375 restriction factors shape immunity. Nat Rev Immunol 12, 687-695.

1376 Eisen, M.B., Spellman, P.T., Brown, P.O., and Botstein, D. (1998). Cluster analysis and

1377 display of genome-wide expression patterns. Proc Natl Acad Sci U S A 95, 14863-14868.

1378 Espert, A., Uluocak, P., Bastos, R.N., Mangat, D., Graab, P., and Gruneberg, U. (2014).

1379 PP2A-B56 opposes Mps1 phosphorylation of Knl1 and thereby promotes spindle assembly

1380 checkpoint silencing. J Cell Biol 206, 833-842.

1381 Foley, G.E., Handler, A.H., Lynch, P.M., Wolman, S.R., Stulberg, C.S., and Eagle, H. (1965).

1382 Loss of neoplastic properties in vitro. II. Observations on KB sublines. Cancer Res 25, 1254-

1383 1261. 59

1384 Fouchier, R.A., Simon, J.H., Jaffe, A.B., and Malim, M.H. (1996). Human immunodeficiency

1385 virus type 1 Vif does not influence expression or virion incorporation of gag-, pol-, and env-

1386 encoded proteins. J Virol 70, 8263-8269.

1387 Fribourgh, J.L., Nguyen, H.C., Wolfe, L.S., Dewitt, D.C., Zhang, W., Yu, X.F., Rhoades, E.,

1388 and Xiong, Y. (2014). Core binding factor beta plays a critical role by facilitating the

1389 assembly of the Vif-cullin 5 E3 ubiquitin ligase. Journal of virology 88, 3309-3319.

1390 Garcia-Exposito, L., Ziglio, S., Barroso-Gonzalez, J., de Armas-Rillo, L., Valera, M.S.,

1391 Zipeto, D., Machado, J.D., and Valenzuela-Fernandez, A. (2013). Gelsolin activity controls

1392 efficient early HIV-1 infection. Retrovirology 10.

1393 Gifford, R.J. (2012). Viral evolution in deep time: lentiviruses and mammals. Trends Genet

1394 28, 89-100.

1395 Gifford, R.J., Katzourakis, A., Tristem, M., Pybus, O.G., Winters, M., and Shafer, R.W.

1396 (2008). A transitional endogenous lentivirus from the genome of a basal primate and

1397 implications for lentivirus evolution. Proc Natl Acad Sci U S A 105, 20362-20367.

1398 Goujon, C., Moncorge, O., Bauby, H., Doyle, T., Ward, C.C., Schaller, T., Hue, S., Barclay,

1399 W.S., Schulz, R., and Malim, M.H. (2013). Human MX2 is an interferon-induced post-entry

1400 inhibitor of HIV-1 infection. Nature 502, 559-562.

1401 Haller, C., Muller, B., Fritz, J.V., Lamas-Murua, M., Stolp, B., Pujol, F.M., Keppler, O.T., and

1402 Fackler, O.T. (2014). HIV-1 Nef and Vpu are functionally redundant broad-spectrum

1403 modulators of cell surface receptors, including tetraspanins. Journal of virology 88, 14241-

1404 14257.

1405 Haverland, N.A., Fox, H.S., and Ciborowski, P. (2014). Quantitative proteomics by SWATH-

1406 MS reveals altered expression of nucleic acid binding and regulatory proteins in HIV-1-

1407 infected macrophages. J Proteome Res 13, 2109-2119. 60

1408 Hedges, S.B., Marin, J., Suleski, M., Paymer, M., and Kumar, S. (2015). Tree of life reveals

1409 clock-like speciation and diversification. Molecular biology and evolution 32, 835-845.

1410 Hengeveld, R.C., Hertz, N.T., Vromans, M.J., Zhang, C., Burlingame, A.L., Shokat, K.M.,

1411 and Lens, S.M. (2012). Development of a chemical genetic approach for human aurora B

1412 kinase identifies novel substrates of the chromosomal passenger complex. Mol Cell

1413 Proteomics 11, 47-59.

1414 Hrecka, K., Hao, C., Gierszewska, M., Swanson, S.K., Kesik-Brodacka, M., Srivastava, S.,

1415 Florens, L., Washburn, M.P., and Skowronski, J. (2011). Vpx relieves inhibition of HIV-1

1416 infection of macrophages mediated by the SAMHD1 protein. Nature 474, 658-661.

1417 Hrecka, K., Hao, C., Shun, M.C., Kaur, S., Swanson, S.K., Florens, L., Washburn, M.P., and

1418 Skowronski, J. (2016). HIV-1 and HIV-2 exhibit divergent interactions with HLTF and UNG2

1419 DNA repair proteins. Proc Natl Acad Sci U S A 113, E3921-3930.

1420 Huang da, W., Sherman, B.T., and Lempicki, R.A. (2009a). Bioinformatics enrichment tools:

1421 paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res

1422 37, 1-13.

1423 Huang da, W., Sherman, B.T., and Lempicki, R.A. (2009b). Systematic and integrative

1424 analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc 4, 44-57.

1425 Hultquist, J.F., Binka, M., LaRue, R.S., Simon, V., and Harris, R.S. (2012). Vif proteins of

1426 human and simian immunodeficiency viruses require cellular CBFbeta to degrade APOBEC3

1427 restriction factors. Journal of virology 86, 2874-2877.

1428 Huthoff, H., and Malim, M.H. (2007). Identification of amino acid residues in APOBEC3G

1429 required for regulation by human immunodeficiency virus type 1 Vif and Virion

1430 encapsidation. J Virol 81, 3807-3815. 61

1431 Imai, K., and Okamoto, T. (2006). Transcriptional repression of human immunodeficiency

1432 virus type 1 by AP-4. J Biol Chem 281, 12495-12505.

1433 Jager, S., Kim, D.Y., Hultquist, J.F., Shindo, K., LaRue, R.S., Kwon, E., Li, M., Anderson,

1434 B.D., Yen, L., Stanley, D., et al. (2012). Vif hijacks CBF-beta to degrade APOBEC3G and

1435 promote HIV-1 infection. Nature 481, 371-375.

1436 Jin, J., Ang, X.L., Shirogane, T., and Wade Harper, J. (2005). Identification of substrates for

1437 F-box proteins. Methods Enzymol 399, 287-309.

1438 Jolly, C., Kashefi, K., Hollinshead, M., and Sattentau, Q.J. (2004). HIV-1 cell to cell transfer

1439 across an Env-induced, actin-dependent synapse. J Exp Med 199, 283-293.

1440 Jowett, J.B., Planelles, V., Poon, B., Shah, N.P., Chen, M.L., and Chen, I.S. (1995). The

1441 human immunodeficiency virus type 1 vpr gene arrests infected T cells in the G2 + M phase

1442 of the cell cycle. J Virol 69, 6304-6313.

1443 Kane, J.R., Stanley, D.J., Hultquist, J.F., Johnson, J.R., Mietrach, N., Binning, J.M.,

1444 Jonsson, S.R., Barelier, S., Newton, B.W., Johnson, T.L., et al. (2015). Lineage-Specific

1445 Viral Hijacking of Non-canonical E3 Ubiquitin Ligase Cofactors in the Evolution of Vif Anti-

1446 APOBEC3 Activity. Cell reports 11, 1236-1250.

1447 Karn, J., and Stoltzfus, C.M. (2012). Transcriptional and posttranscriptional regulation of

1448 HIV-1 gene expression. Cold Spring Harb Perspect Med 2, a006916.

1449 Katzourakis, A., Tristem, M., Pybus, O.G., and Gifford, R.J. (2007). Discovery and analysis

1450 of the first endogenous lentivirus. Proc Natl Acad Sci U S A 104, 6261-6265.

1451 Kauko, O., Laajala, T.D., Jumppanen, M., Hintsanen, P., Suni, V., Haapaniemi, P., Corthals,

1452 G., Aittokallio, T., Westermarck, J., and Imanishi, S.Y. (2015). Label-free quantitative

1453 phosphoproteomics with novel pairwise abundance normalization reveals synergistic RAS

1454 and CIP2A signaling. Sci Rep 5, 13099. 62

1455 Keckesova, Z., Ylinen, L.M., Towers, G.J., Gifford, R.J., and Katzourakis, A. (2009).

1456 Identification of a RELIK orthologue in the European hare (Lepus europaeus) reveals a

1457 minimum age of 12 million years for the lagomorph lentiviruses. Virology 384, 7-11.

1458 Kettenbach, A.N., Schweppe, D.K., Faherty, B.K., Pechenick, D., Pletnev, A.A., and Gerber,

1459 S.A. (2011). Quantitative phosphoproteomics identifies substrates and functional modules of

1460 Aurora and Polo-like kinase activities in mitotic cells. Science signaling 4, rs5.

1461 Khoury, G.A., Baliban, R.C., and Floudas, C.A. (2011). Proteome-wide post-translational

1462 modification statistics: frequency analysis and curation of the swiss-prot database. Sci Rep

1463 1.

1464 Kim, D.Y., Kwon, E., Hartley, P.D., Crosby, D.C., Mann, S., Krogan, N.J., and Gross, J.D.

1465 (2013). CBFbeta stabilizes HIV Vif to counteract APOBEC3 at the expense of RUNX1 target

1466 gene expression. Molecular cell 49, 632-644.

1467 Kraft-Terry, S.D., Engebretsen, I.L., Bastola, D.K., Fox, H.S., Ciborowski, P., and

1468 Gendelman, H.E. (2011). Pulsed stable isotope labeling of amino acids in cell culture

1469 uncovers the dynamic interactions between HIV-1 and the monocyte-derived macrophage. J

1470 Proteome Res 10, 2852-2862.

1471 Kruse, T., Zhang, G., Larsen, M.S., Lischetti, T., Streicher, W., Kragh Nielsen, T., Bjorn,

1472 S.P., and Nilsson, J. (2013). Direct binding between BubR1 and B56-PP2A phosphatase

1473 complexes regulate mitotic progression. J Cell Sci 126, 1086-1092.

1474 Laguette, N., Bregnard, C., Hue, P., Basbous, J., Yatim, A., Larroque, M., Kirchhoff, F.,

1475 Constantinou, A., Sobhian, B., and Benkirane, M. (2014). Premature activation of the SLX4

1476 complex by Vpr promotes G2/M arrest and escape from innate immune sensing. Cell 156,

1477 134-145. 63

1478 Laguette, N., Sobhian, B., Casartelli, N., Ringeard, M., Chable-Bessia, C., Segeral, E.,

1479 Yatim, A., Emiliani, S., Schwartz, O., and Benkirane, M. (2011). SAMHD1 is the dendritic-

1480 and myeloid-cell-specific HIV-1 restriction factor counteracted by Vpx. Nature 474, 654-657.

1481 Lahouassa, H., Blondot, M.L., Chauveau, L., Chougui, G., Morel, M., Leduc, M.,

1482 Guillonneau, F., Ramirez, B.C., Schwartz, O., and Margottin-Goguet, F. (2016). HIV-1 Vpr

1483 degrades the HLTF DNA translocase in T cells and macrophages. Proc Natl Acad Sci U S A.

1484 Lahouassa, H., Daddacha, W., Hofmann, H., Ayinde, D., Logue, E.C., Dragin, L., Bloch, N.,

1485 Maudet, C., Bertrand, M., Gramberg, T., et al. (2012). SAMHD1 restricts the replication of

1486 human immunodeficiency virus type 1 by depleting the intracellular pool of deoxynucleoside

1487 triphosphates. Nat Immunol 13, 223-228.

1488 Lai, M.Y., Zimmerman, E.S., Planelles, V., and Chen, J.J. (2005). Activation of the ATR

1489 pathway by human immunodeficiency virus type 1 Vpr involves its direct binding to

1490 chromatin in vivo. J Virol 79, 15443-15451.

1491 Larue, R.S., Lengyel, J., Jonsson, S.R., Andresdottir, V., and Harris, R.S. (2010). Lentiviral

1492 Vif degrades the APOBEC3Z3/APOBEC3H protein of its mammalian host and is capable of

1493 cross-species activity. J Virol 84, 8193-8201.

1494 Letko, M., Silvestri, G., Hahn, B.H., Bibollet-Ruche, F., Gokcumen, O., Simon, V., and

1495 Ooms, M. (2013). Vif proteins from diverse primate lentiviral lineages use the same binding

1496 site in APOBEC3G. J Virol 87, 11861-11871.

1497 Lin, X.H., Walter, J., Scheidtmann, K., Ohst, K., Newport, J., and Walter, G. (1998). Protein

1498 phosphatase 2A is required for the initiation of chromosomal DNA replication. Proc Natl

1499 Acad Sci U S A 95, 14693-14698.

1500 Lu, Y.L., Spearman, P., and Ratner, L. (1993). Human immunodeficiency virus type 1 viral

1501 protein R localization in infected cells and virions. J Virol 67, 6542-6550. 64

1502 Ly, T., Ahmad, Y., Shlien, A., Soroka, D., Mills, A., Emanuele, M.J., Stratton, M.R., and

1503 Lamond, A.I. (2014). A proteomic chronology of gene expression through the cell cycle in

1504 human myeloid leukemia cells. Elife 3, e01630.

1505 Malim, M.H. (2009). APOBEC proteins and intrinsic resistance to HIV-1 infection. Philos

1506 Trans R Soc Lond B Biol Sci 364, 675-687.

1507 Malim, M.H., and Bieniasz, P.D. (2012). HIV Restriction Factors and Mechanisms of

1508 Evasion. Cold Spring Harb Perspect Med 2, a006940.

1509 Matheson, N.J., Peden, A.A., and Lehner, P.J. (2014). Antibody-free magnetic cell sorting of

1510 genetically modified primary human CD4+ T cells by one-step streptavidin affinity

1511 purification. PLoS One 9, e111437.

1512 Matheson, N.J., Sumner, J., Wals, K., Rapiteanu, R., Weekes, M.P., Vigan, R., Weinelt, J.,

1513 Schindler, M., Antrobus, R., Costa, A.S., et al. (2015). Cell Surface Proteomic Map of HIV

1514 Infection Reveals Antagonism of Amino Acid Metabolism by Vpu and Nef. Cell host &

1515 microbe 18, 409-423.

1516 McCright, B., Rivers, A.M., Audlin, S., and Virshup, D.M. (1996). The B56 family of protein

1517 phosphatase 2A (PP2A) regulatory subunits encodes differentiation-induced

1518 phosphoproteins that target PP2A to both nucleus and cytoplasm. J Biol Chem 271, 22081-

1519 22089.

1520 Miyagi, E., Kao, S., Yedavalli, V., and Strebel, K. (2014). CBFbeta enhances de novo protein

1521 biosynthesis of its binding partners HIV-1 Vif and RUNX1 and potentiates the Vif-induced

1522 degradation of APOBEC3G. Journal of virology 88, 4839-4852.

1523 Mootha, V.K., Bunkenborg, J., Olsen, J.V., Hjerrild, M., Wisniewski, J.R., Stahl, E., Bolouri,

1524 M.S., Ray, H.N., Sihag, S., Kamal, M., et al. (2003). Integrated analysis of protein

1525 composition, tissue diversity, and gene regulation in mouse mitochondria. Cell 115, 629-640. 65

1526 Nakai-Murakami, C., Shimura, M., Kinomoto, M., Takizawa, Y., Tokunaga, K., Taguchi, T.,

1527 Hoshino, S., Miyagawa, K., Sata, T., Kurumizaka, H., et al. (2007). HIV-1 Vpr induces ATM-

1528 dependent cellular signal with enhanced homologous recombination. Oncogene 26, 477-

1529 486.

1530 Navare, A.T., Sova, P., Purdy, D.E., Weiss, J.M., Wolf-Yadlin, A., Korth, M.J., Chang, S.T.,

1531 Proll, S.C., Jahan, T.A., Krasnoselsky, A.L., et al. (2012). Quantitative proteomic analysis of

1532 HIV-1 infected CD4+ T cells reveals an early host response in important biological pathways:

1533 protein synthesis, cell proliferation, and T-cell activation. Virology 429, 37-46.

1534 Norbury, C., and Nurse, P. (1992). Animal cell cycles and their control. Annu Rev Biochem

1535 61, 441-470.

1536 Notredame, C., Higgins, D.G., and Heringa, J. (2000). T-Coffee: A novel method for fast and

1537 accurate multiple sequence alignment. J Mol Biol 302, 205-217.

1538 Pan, Q., Rong, L., Zhao, X., and Liang, C. (2009). Fragile X mental retardation protein

1539 restricts replication of human immunodeficiency virus type 1. Virology 387, 127-135.

1540 Pathak, S., De Souza, G.A., Salte, T., Wiker, H.G., and Asjo, B. (2009). HIV induces both a

1541 down-regulation of IRAK-4 that impairs TLR signalling and an up-regulation of the antibiotic

1542 peptide dermcidin in monocytic cells. Scand J Immunol 70, 264-276.

1543 Paxton, W., Connor, R.I., and Landau, N.R. (1993). Incorporation of Vpr into human

1544 immunodeficiency virus type 1 virions: requirement for the p6 region of gag and mutational

1545 analysis. J Virol 67, 7229-7237.

1546 Pollard, V.W., and Malim, M.H. (1998). The HIV-1 Rev protein. Annu Rev Microbiol 52, 491-

1547 532.

1548 Randow, F., and Lehner, P.J. (2009). Viral avoidance and exploitation of the ubiquitin

1549 system. Nature cell biology 11, 527-534. 66

1550 Rasheed, S., Yan, J.S., Hussain, A., and Lai, B. (2009). Proteomic characterization of HIV-

1551 modulated membrane receptors, kinases and signaling proteins involved in novel angiogenic

1552 pathways. J Transl Med 7, 75.

1553 Rebhan, M., Chalifa-Caspi, V., Prilusky, J., and Lancet, D. (1997). GeneCards: integrating

1554 information about genes, proteins and diseases. Trends Genet 13, 163.

1555 Ringrose, J.H., Jeeninga, R.E., Berkhout, B., and Speijer, D. (2008). Proteomic studies

1556 reveal coordinated changes in T-cell expression patterns upon infection with human

1557 immunodeficiency virus type 1. J Virol 82, 4320-4330.

1558 Roshal, M., Kim, B., Zhu, Y.H., Nghiem, P., and Planelles, V. (2003). Activation of the ATR-

1559 mediated DNA damage response by the HIV-1 viral protein R. Journal of Biological

1560 Chemistry 278, 25879-25886.

1561 Ruediger, R., Van Wart Hood, J.E., Mumby, M., and Walter, G. (1991). Constant expression

1562 and activity of protein phosphatase 2A in synchronized cells. Mol Cell Biol 11, 4282-4285.

1563 Saldanha, A.J. (2004). Java Treeview--extensible visualization of microarray data.

1564 Bioinformatics 20, 3246-3248.

1565 Salter, J.D., Lippa, G.M., Belashov, I.A., and Wedekind, J.E. (2012). Core-binding factor

1566 beta increases the affinity between human Cullin 5 and HIV-1 Vif within an E3 ligase

1567 complex. Biochemistry 51, 8702-8704.

1568 Santamaria, A., Wang, B., Elowe, S., Malik, R., Zhang, F., Bauer, M., Schmidt, A., Sillje,

1569 H.H., Korner, R., and Nigg, E.A. (2011). The Plk1-dependent phosphoproteome of the early

1570 mitotic spindle. Mol Cell Proteomics 10, M110 004457.

1571 Sargan, D.R., Bennet, I.D., Cousens, C., Roy, D.J., Blacklaws, B.A., Dalziel, R.G., Watt,

1572 N.J., and McConnell, I. (1991). Nucleotide sequence of EV1, a British isolate of maedi-visna

1573 virus. J Gen Virol 72 ( Pt 8), 1893-1903. 67

1574 Schneider, C.A., Rasband, W.S., and Eliceiri, K.W. (2012). NIH Image to ImageJ: 25 years

1575 of image analysis. Nat Methods 9, 671-675.

1576 Schrofelbauer, B., Yu, Q., Zeitlin, S.G., and Landau, N.R. (2005). Human immunodeficiency

1577 virus type 1 Vpr induces the degradation of the UNG and SMUG uracil-DNA glycosylases. J

1578 Virol 79, 10978-10987.

1579 Schwammle, V., Leon, I.R., and Jensen, O.N. (2013). Assessment and improvement of

1580 statistical tools for comparative proteomics analysis of sparse data sets with few

1581 experimental replicates. J Proteome Res 12, 3874-3883.

1582 Schwartz, S., Felber, B.K., Fenyo, E.M., and Pavlakis, G.N. (1990). Env and Vpu proteins of

1583 human immunodeficiency virus type 1 are produced from multiple bicistronic mRNAs. J Virol

1584 64, 5448-5456.

1585 Seeling, J.M., Miller, J.R., Gil, R., Moon, R.T., White, R., and Virshup, D.M. (1999).

1586 Regulation of beta-catenin signaling by the B56 subunit of protein phosphatase 2A. Science

1587 283, 2089-2091.

1588 Sheehy, A.M., Gaddis, N.C., Choi, J.D., and Malim, M.H. (2002). Isolation of a human gene

1589 that inhibits HIV-1 infection and is suppressed by the viral Vif protein. Nature 418, 646-650.

1590 Sherrill-Mix, S., Ocwieja, K.E., and Bushman, F.D. (2015). Gene activity in primary T cells

1591 infected with HIV89.6: intron retention and induction of genomic repeats. Retrovirology 12,

1592 79.

1593 Shrivastava, S., Trivedi, J., and Mitra, D. (2016). Gene expression profiling reveals Nef

1594 induced deregulation of lipid metabolism in HIV-1 infected T cells. Biochem Biophys Res

1595 Commun 472, 169-174. 68

1596 Simon, J.H., Fouchier, R.A., Southerling, T.E., Guerra, C.B., Grant, C.K., and Malim, M.H.

1597 (1997). The Vif and Gag proteins of human immunodeficiency virus type 1 colocalize in

1598 infected human T cells. J Virol 71, 5259-5267.

1599 Simon, V., Bloch, N., and Landau, N.R. (2015). Intrinsic host restrictions to HIV-1 and

1600 mechanisms of viral escape. Nat Immunol 16, 546-553.

1601 Smith, J.L., and Pathak, V.K. (2010). Identification of specific determinants of human

1602 APOBEC3F, APOBEC3C, and APOBEC3DE and African green monkey APOBEC3F that

1603 interact with HIV-1 Vif. J Virol 84, 12599-12608.

1604 Specht, A., DeGottardi, M.Q., Schindler, M., Hahn, B., Evans, D.T., and Kirchhoff, F. (2008).

1605 Selective downmodulation of HLA-A and -B by Nef alleles from different groups of primate

1606 lentiviruses. Virology 373, 229-237.

1607 Stopak, K., de Noronha, C., Yonemoto, W., and Greene, W.C. (2003). HIV-1 Vif blocks the

1608 antiviral activity of APOBEC3G by impairing both its translation and intracellular stability. Mol

1609 Cell 12, 591-601.

1610 Subramanian, A., Tamayo, P., Mootha, V.K., Mukherjee, S., Ebert, B.L., Gillette, M.A.,

1611 Paulovich, A., Pomeroy, S.L., Golub, T.R., Lander, E.S., et al. (2005). Gene set enrichment

1612 analysis: a knowledge-based approach for interpreting genome-wide expression profiles.

1613 Proc Natl Acad Sci U S A 102, 15545-15550.

1614 Sugden, S.M., Bego, M.G., Pham, T.N., and Cohen, E.A. (2016). Remodeling of the Host

1615 Cell Plasma Membrane by HIV-1 Nef and Vpu: A Strategy to Ensure Viral Fitness and

1616 Persistence. Viruses 8.

1617 Taus, T., Kocher, T., Pichler, P., Paschke, C., Schmidt, A., Henrich, C., and Mechtler, K.

1618 (2011). Universal and confident phosphorylation site localization using phosphoRS. J

1619 Proteome Res 10, 5354-5362. 69

1620 Taylor, H.E., Simmons, G.E., Jr., Mathews, T.P., Khatua, A.K., Popik, W., Lindsley, C.W.,

1621 D'Aquila, R.T., and Brown, H.A. (2015). Phospholipase D1 Couples CD4+ T Cell Activation

1622 to c-Myc-Dependent Deoxyribonucleotide Pool Expansion and HIV-1 Replication. PLoS

1623 Pathog 11, e1004864.

1624 van 't Wout, A.B., Swain, J.V., Schindler, M., Rao, U., Pathmajeyan, M.S., Mullins, J.I., and

1625 Kirchhoff, F. (2005). Nef induces multiple genes involved in cholesterol synthesis and uptake

1626 in human immunodeficiency virus type 1-infected T cells. J Virol 79, 10053-10058.

1627 van den Boomen, D.J., Timms, R.T., Grice, G.L., Stagg, H.R., Skodt, K., Dougan, G.,

1628 Nathan, J.A., and Lehner, P.J. (2014). TMEM129 is a Derlin-1 associated ERAD E3 ligase

1629 essential for virus-induced degradation of MHC-I. Proc Natl Acad Sci U S A 111, 11425-

1630 11430.

1631 Vasiliver-Shamis, G., Cho, M.W., Hioe, C.E., and Dustin, M.L. (2009). Human

1632 immunodeficiency virus type 1 envelope gp120-induced partial T-cell receptor signaling

1633 creates an F-actin-depleted zone in the virological synapse. J Virol 83, 11341-11355.

1634 Vassena, L., Giuliani, E., Matusali, G., Cohen, E.A., and Doria, M. (2013). The human

1635 immunodeficiency virus type 1 Vpr protein upregulates PVR via activation of the AIR-

1636 mediated DNA damage response pathway. Journal of General Virology 94, 2664-2669.

1637 Vizcaino, J.A., Cote, R.G., Csordas, A., Dianes, J.A., Fabregat, A., Foster, J.M., Griss, J.,

1638 Alpi, E., Birim, M., Contell, J., et al. (2013). The PRoteomics IDEntifications (PRIDE)

1639 database and associated tools: status in 2013. Nucleic Acids Res 41, D1063-1069.

1640 Weekes, M.P., Tomasec, P., Huttlin, E.L., Fielding, C.A., Nusinow, D., Stanton, R.J., Wang,

1641 E.C., Aicheler, R., Murrell, I., Wilkinson, G.W., et al. (2014). Quantitative temporal viromics:

1642 an approach to investigate host-pathogen interaction. Cell 157, 1460-1472. 70

1643 Welburn, J.P., Vleugel, M., Liu, D., Yates, J.R., 3rd, Lampson, M.A., Fukagawa, T., and

1644 Cheeseman, I.M. (2010). Aurora B phosphorylates spatially distinct targets to differentially

1645 regulate the kinetochore-microtubule interface. Mol Cell 38, 383-392.

1646 Wilkinson, G.W., Davison, A.J., Tomasec, P., Fielding, C.A., Aicheler, R., Murrell, I.,

1647 Seirafian, S., Wang, E.C., Weekes, M., Lehner, P.J., et al. (2015). Human cytomegalovirus:

1648 taking the strain. Medical microbiology and immunology 204, 273-284.

1649 Wisniewski, J.R., Zougman, A., Nagaraj, N., and Mann, M. (2009). Universal sample

1650 preparation method for proteome analysis. Nat Methods 6, 359-362.

1651 Worobey, M., Telfer, P., Souquiere, S., Hunter, M., Coleman, C.A., Metzger, M.J., Reed, P.,

1652 Makuwa, M., Hearn, G., Honarvar, S., et al. (2010). Island biogeography reveals the deep

1653 history of SIV. Science 329, 1487.

1654 Wu, C., Barbezange, C., McConnell, I., and Blacklaws, B.A. (2008). Mapping and

1655 characterization of visna/maedi virus cytotoxic T-lymphocyte epitopes. J Gen Virol 89, 2586-

1656 2596.

1657 Wu, R., Dephoure, N., Haas, W., Huttlin, E.L., Zhai, B., Sowa, M.E., and Gygi, S.P. (2011).

1658 Correct interpretation of comprehensive phosphorylation dynamics requires normalization by

1659 protein expression changes. Mol Cell Proteomics 10, M111 009654.

1660 Xu, P., Raetz, E.A., Kitagawa, M., Virshup, D.M., and Lee, S.H. (2013). BUBR1 recruits

1661 PP2A via the B56 family of targeting subunits to promote chromosome congression. Biology

1662 open 2, 479-486.

1663 Yang, J., and Phiel, C. (2010). Functions of B56-containing PP2As in major developmental

1664 and cancer signaling pathways. Life sciences 87, 659-666.

1665 Yoshikawa, R., Izumi, T., Nakano, Y., Yamada, E., Moriwaki, M., Misawa, N., Ren, F.,

1666 Kobayashi, T., Koyanagi, Y., and Sato, K. (2016). Small ruminant lentiviral Vif proteins 71

1667 commonly utilize cyclophilin A, an evolutionarily and structurally conserved protein, to

1668 degrade ovine and caprine APOBEC3 proteins. Microbiology and immunology 60, 427-436.

1669 Yu, C.T.R., Hsu, J.M., Lee, Y.C.G., Tsou, A.P., Chou, C.K., and Huang, C.Y.F. (2005).

1670 Phosphorylation and stabilization of HURP by aurora-A: Implication of HURP as a

1671 transforming target of aurora-A. Mol Cell Biol 25, 5789-5800.

1672 Zhang, H., Zhou, Y., Alcock, C., Kiefer, T., Monie, D., Siliciano, J., Li, Q., Pham, P.,

1673 Cofrancesco, J., Persaud, D., et al. (2004). Novel single-cell-level phenotypic assay for

1674 residual drug susceptibility and reduced replication capacity of drug-resistant human

1675 immunodeficiency virus type 1. J Virol 78, 1718-1729.

1676 Zhang, J., Wu, J., Wang, W., Wu, H., Yu, B., Wang, J., Lv, M., Wang, X., Zhang, H., Kong,

1677 W., et al. (2014). Role of cullin-elonginB-elonginC E3 complex in bovine immunodeficiency

1678 virus and maedi-visna virus Vif-mediated degradation of host A3Z2-Z3 proteins.

1679 Retrovirology 11, 77.

1680 Zhang, W., Du, J., Evans, S.L., Yu, Y., and Yu, X.F. (2012). T-cell differentiation factor CBF-

1681 beta regulates HIV-1 Vif-mediated evasion of host restriction. Nature 481, 376-379. Figure 1 Figure 1 A B Figure 1

0 h 6 h 24 h 48 h 72 h 72 h Proteomics (TMT) Flow cytometry

+ HIV + RTi Env-GFP 6 ↓ ↓ ↓ ↓ ↓ ↓ 1 r2=0.97 Flow)

( 0

Digest proteins and label peptides with TMT reporters 2 abundance -6 0 6

0 Log -6 ↓ 0 h 6 h RTi Log (TMT) 24 h 48 h 72 h Relative Relative 2 Mix peptides, fractionate and analyse by LC/MS3

C D Tat Rev APOBEC3C UNG 1 1 1 1

0 0 0 0 0 h 6 h 0 h 6 h 0 h 6 h 0 h 6 h RTi RTi RTi RTi 24 h 48 h 72 h 24 h 48 h 72 h 24 h 48 h 72 h 24 h 48 h 72 h Nef Vif β-catenin Cyclin B1 1 1 1 1

0 0 0 0 0 h 6 h 0 h 6 h 0 h 6 h 0 h 6 h RTi RTi RTi RTi 24 h 48 h 72 h 24 h 48 h 72 h 24 h 48 h 72 h 24 h 48 h 72 h Gag Gagpol GAPDH β-actin 1 1 1 1 Relative abundance Relative abundance 0 0 0 0 0 h 6 h 0 h 6 h 0 h 6 h 0 h 6 h RTi RTi RTi RTi 72 h 24 h 48 h 24 h 48 h 72 h 24 h 48 h 72 h 24 h 48 h 72 h Figure 2 A B C FigureFigure 2 2

Cluster average profile SILAC-based validation 0 h 6 h 24 h 48 h 72 h RTi Cluster 1 *** *** *** 1 3 0 ① ① 0 -3 24 h 48 h 72 h 0 h 6 h RTi 24 h 48 h 72 h Cluster 2 *** *** *** 1 3 0 ② ② 0 -3 24 h 48 h 72 h 0 h 6 h RTi 24 h 48 h 72 h

Cluster 3 *** *** *** 1 3 0 ③ ③

0 -3 24 h 48 h 72 h

0 h 6 h RTi 24 h 48 h 72 h

+2 Cluster 4 3 *** *** *** 1 (SILAC H/M)

0 0 ④ ④ 2 Relative abundance 0 Log -3 -2 24 h 48 h 72 h 0 h 6 h RTi 24 h 48 h 72 h

D E FMR1 1

1

0 ① 0 h 6 h RTi 24 h 48 h 72 h

TFAP4 0.5 1

Relative abundance 0 0 Normalised infectious virus release

0 h 6 h ② RTi  24 h 48 h 72 h   TFAP4 FMR1 mCherry Tetherin APOBEC3G Figure 3

A B FigureFigure 3 3

0 h 6 h 24 h 48 h 72 h RTi APOBEC3C PPP2R5A IGSF8 1 1 ① PPP2R1B PPFIA1 IAA0101 0 0 0 h 6 h

TYMS 0 h 6 h RTi RTi 72 h 24 h 48 h RRM2 24 h 48 h 72 h ATA2 TMS PPP2R5C ARL4C 1 1 CD4 PPP2R5E FMR1 0 0 PPP2R5D 0 h 6 h 0 h 6 h RTi PPP2R5C RTi 72 h 24 h 48 h 72 h 24 h 48 h PPP2R5A AAGAB RRM1 PPP2R5 APOBEC3C 1 1 SPN SLC38A1 TCF12 0 0 REPIN1 0 h 6 h 0 h 6 h RTi RTi 72 h 24 h 48 h PRL1 24 h 48 h 72 h RRM1 PRO1 RRM2 PPP2R5E CCP110 1 1 +2 PRSS21 ORC6 0 SLC2A1 0 0 21 0 h 6 h 0 h 6 h RTi RTi 24 h 48 h 72 h -2 PHACTR4 24 h 48 h 72 h Figure 4

A HIV WT vs Mock HIV ΔVif vs Mock HIV WT vs HIV ΔVif

Depleted in HIV Increased in HIV Vif not necessary Vif necessary for infection infection for depletion depletion

14 14 14

Vif

q<0.01 7 7 7 q>0.01 (q value) 2 APOBEC3B

-3 -2 -1 0 1 2 3 -3 -2 -1 0 1 2 3 -3 -2 -1 0 1 2 3 -Log PPP2R5 Family HIV Protein β-catenin CD4 Log2 (fold change) APOBEC3 Family Tetherin SNAT1 UNG

B MOI 1 MOI 10 C HA-PPP2R5A 293T cells

Mw I.B. Mock HIV WT (KDa) HIV WT ΔVif HIV Mock ΔVif HIV 50 PPP2R5A HA-PPP2R5A

75 PPP2R5D GFP GFP GFP + Empty vector +Vif +Vif C114S

91 112 114 p24 25 86 26 112

25 Vif

50 Calreticulin HA-PPP2R5A Untransfected Untransfected Untransfected Transfected: Transfected: Transfected: empty vector Vif Vif C114S D 293T cells expressing: HA-PPP2R5A HA-PPP2R5B HA-PPP2R5C HA-PPP2R5D HA-PPP2R5E HA-APOBEC3G 159 445 231 279 372 1468 38 32 48 16 63 60

Untransfected cells Cells transfected with Vif Figure 5 Figure 5 Figure 5 A B 293T HA-PPP2R5A cells 293T GFP-PPP2R5B cells DMSO Bortezomib Vif + Empty Vector Vif + Cul5 WT Vif + Cul5 DN 96 136 3236 3297 3231 15 50 154 164 1482

HA-PPP2R5A GFP-PPP2R5B Untransfected cells Untransfected cells Untransfected cells Vif + Cul5 WT/DN + DMSO + Bortezomib transfected cells Vif transfected cells Vif transfected cells Vif + empty vector + DMSO + Bortezomib transfected cells

C 293T HA-PPP2R5B cells Sh-Cntrl Sh-elob Sh-eloc Sh-cbfb 75 94 105 84 Untransfected cells 21 45 63 56 +shRNA Vif + sh-Cntrl

Vif + sh-target

HA-PPP2R5B 293T HA-APOBEC3G cells Sh-Cntrl Sh-elob Sh-eloc Sh-cbfb 801 764 888 865 38 163 748 518

HA-APOBEC3G Figure 6

A B HIV WT vs HIV ΔVif Okadaic Acid Protein Phosphopeptide PP2A inhibitor 

 (fold change)

2     HIV WT vs ΔVif ΔVif vs HIV WT

infected CEM-T4 2 log r =0.4  P<0.0001

Okadaic acid treated HeLa

log2 (fold change) vs untreated      D  MLN8054 AURKA, Aurora AURKB, kinase PLK1 inhibitor inhibitor  (q value) 2 r2=0.05     P<0.0001

-Log -3 -2 -1 0 1 -2-1012345  Log2 (fold change)  C 

HIV WT vs HIV ΔVif (fold change)  2     HIV WT vs ΔVif ΔVif vs HIV WT infected CEM-T4 log    MLN8054 treated HeLa     log2 (fold change) vs untreated     

BI2536   PLK1-3 inhibitor  (fold change)

2 2 r=0.0003  P=0.5

Log    (fold change) 2    HIV WT vs ΔVif ΔVif vs HIV WT infected CEM-T4 log

   

                                BI2536 treated HeLa  

log2 (fold change) vs untreated 

E  PLK1  AURKA  AURKB *** *** *    *** N.S. *** N.S. ** N.S.   

      Protein Relative

abundance      

  

Mock Mock Mock HIV WT HIV ΔVif HIV WT HIV ΔVif HIV WT HIV ΔVif

*** N.S. N.S. N.S.  ***  *** ***  *** ***

  

  

   Relative abundance    phosphopeptide

  

   Mock Mock Mock HIV WT HIV ΔVif HIV WT HIV ΔVif HIV WT HIV ΔVif Figure 7 Figure 7 Figure 7 A B HIV-1 293T HA-PPP2R5B cells

100%



50%

transfected cells  SIVcpz/SIVgor

Normalised PPP2R5B in Vif Normalised PPP2R5B in Vif 0%

HIV-1 SIVagm Controls  SIVcpz/gor Non-primate 

HIV-2/SIVsmmm/SIVmac SIVagm

Non-primate

HIV-2/SIVsmm/SIVmac

C 293T cells expressing: HA-APOBEC3G HA-PPP2R5A HA-PPP2R5B HA-PPP2R5C HA-PPP2R5D HA-PPP2R5E 1219 110 130 115 160 189 Empty 1208 108 136 113 191 188 vector

858 89 136 136 141 200 35 24 22 41 50 40 NL4-3

843 88 115 129 81 225 562 24 41 40 30 40 SRLV

Untransfected cells Transfected cells Figure 1 – figure supplement 1 A Figure S1 WCP PMP

CD4 Tetherin HLA-A SNAT1 1 1 1 1

0 0 0 0 0 h 6 h 0 h 6 h 0 h 6 h 0 h 6 h RTi RTi RTi RTi 48 h 72 h 24 h 24 h 48 h 72 h 24 h 48 h 72 h 24 h 48 h 72 h Relative abundance

B C WCP PMP

24 h 48 h 72 h 400 400 400 VSVg 1 200 200 200

Frequency 0 0 0 0 -4 0 4 -4 0 4 -4 0 4 0 h 6 h RTi 24 h 48 h 72 h Relative abundance Log2 (ratio) Log2 (ratio) Log2 (ratio)

D E

ISG15 BiP PDI 1 1 1

0 0 0 0 h 6 h 0 h 6 h 0 h 6 h RTi RTi RTi Relative abundance Relative abundance 24 h 48 h 72 h 24 h 48 h 72 h 24 h 48 h 72 h

F H

PMP WCP Gelsolin CAPG 1 1 403 401 629

0 0 0 h 6 h 0 h 6 h RTi RTi 24 h 48 h 72 h 24 h 48 h 72 h Relative abundance

G 245/403

60%

161/401 AURKA AURKB PLK1 40% 1 1 1

20% 71/629

0 0 0 % glycosylated 0% 0 h 6 h 0 h 6 h 0 h 6 h RTi RTi RTi 24 h 48 h 72 h 24 h 48 h 72 h 24 h 48 h 72 h Relative abundance PMP WCP PMP & WCP Figure 1 – figure supplement 2 A Figure S2 -log10 (FDR q value) 0 1 2 3 4 KEGG_GLYCOLYSIS_GLUCONEOGENESIS KEGG_OXIDATIVE_PHOSPHORYLATION KEGG_PYRUVATE_METABOLISM 6 h REGULATION_OF_BINDING T_CELL_ACTIVATION KEGG_PROGESTERONE_MEDIATED_OOCYTE_MATURATION NEGATIVE_REGULATION_OF_CELLULAR_PROTEIN_METABOLIC_… NEGATIVE_REGULATION_OF_PROTEIN_METABOLIC_PROCESS 24 h KEGG_APOPTOSIS KEGG_LEUKOCYTE_TRANSENDOTHELIAL_MIGRATION LIPID_METABOLIC_PROCESS KEGG_LYSOSOME CELLULAR_LIPID_METABOLIC_PROCESS 48 h MICROTUBULE_BASED_PROCESS KEGG_VALINE_LEUCINE_AND_ISOLEUCINE_DEGRADATION LIPID_METABOLIC_PROCESS CELLULAR_LIPID_METABOLIC_PROCESS KEGG_FOCAL_ADHESION 72 h KEGG_LYSOSOME

KEGG_PEROXISOME Upregulated by HIV infection KEGG_T_CELL_RECEPTOR_SIGNALING_PATHWAY KEGG_LYSOSOME KEGG_P53_SIGNALING_PATHWAY RTi KEGG_REGULATION_OF_ACTIN_CYTOSKELETON KEGG_FOCAL_ADHESION

5% 25% FDR B FDR log10 (FDR q value) -4 -3 -2 -1 0 NEGATIVE_REGULATION_OF_TRANSCRIPTION_FROM_RNA_POL… NEGATIVE_REGULATION_OF_RNA_METABOLIC_PROCESS NEGATIVE_REGULATION_OF_TRANSCRIPTION_DNA_DEPENDENT 6 h INFLAMMATORY_RESPONSE

G1_S_TRANSITION_OF_MITOTIC_CELL_CYCLE KEGG_GLYCOLYSIS_GLUCONEOGENESIS KEGG_FRUCTOSE_AND_MANNOSE_METABOLISM KEGG_RIBOSOME KEGG_ARGININE_AND_PROLINE_METABOLISM 24 h KEGG_PENTOSE_PHOSPHATE_PATHWAY RNA_METABOLIC_PROCESS RNA_PROCESSING KEGG_RIBOSOME KEGG_RNA_POLYMERASE 48 h RIBONUCLEOPROTEIN_COMPLEX_BIOGENESIS_AND_ASSEMBLY RNA_PROCESSING RIBONUCLEOPROTEIN_COMPLEX_BIOGENESIS_AND_ASSEMBLY RNA_METABOLIC_PROCESS MRNA_PROCESSING_GO_0006397 72 h KEGG_RIBOSOME Downregulated Downregulated HIVby infection KEGG_RIBOSOME KEGG_AMINOACYL_TRNA_BIOSYNTHESIS RNA_SPLICINGVIA_TRANSESTERIFICATION_REACTIONS RTi KEGG_GLYCOLYSIS_GLUCONEOGENESIS RIBONUCLEOPROTEIN_COMPLEX_BIOGENESIS_AND_ASSEMBLY

5% 25% FDR FDR

C D RNA_PROCESSING 2 LIPID_METABOLIC_PROCESS 2

1 1

0 0 (infected/0 h) (infected/0 h)

2 2 -1 -1 Log Log -2 -2 6 h 24 h 48 h 72 h RTi 6 h 24 h 48 h 72 h RTi Figure S5 A A Figure S5 Figure 2 – figure supplement 1 Mock 48 h + HIV WT 48 h + HIV ΔVIf 48 h A Figure S3

↓ ↓ ↓ ↓ ↓ ↓ ↓ ↓ ↓ 0 h 24 h 48 h 72 h Digest proteins and either label peptides with TMT reporters or enrich phosphopeptides with titanium dioxide then label with MEDIUMTMT reporters Mock① ②

HEAVY

+ HIV ↓ 3 Mix peptides or phosphopeptidesMix aliquots, fractionate of MEDIUM and and analyse by LC/MS HEAVY cells at 3x timepoints ① ② ↓ ↓ ↓

Digest proteins, fractionate peptides and analyse by LC/MS2 A

B

A B Mock C HIV WT HIV ΔVif Figure S4

0 h 6 h 24 h 48 h 72 h RTi Infected A Infected Enrichment score Infected Figure S1 WCP KIF18BPMP UNG 0.8% 77% ② 77% A A 1 0 1 Figure2 3 S3 4 5 6 NKX2Figure-5 S3 CD4 Tetherin DDX31HLA -A SNAT1 Protein phosphatase regulatory activity CCDC137 1 1 1 1 Deoxyribonucleotide metabolic process+2 FAM60A 0 DUSP11 Cluster 1 Cluster Nucleotide binding 0 h 24 h 48 h 72 h 0 h 0 h 6 h 24 h 48 h 72 RTi h 0 UNG 24 h Ribonucleoprotein48 h 72 h complex biogenesis0 0 0 0 MEDIUM GNL3L MEDIUM DNA binding 0 h 6 h 0 h 6 h 0 h 6 h 0 h 6 h RTi RTi RTi HLTF RTi HLTF 48 h 72 h 24 h 24 h 48 h 72 h 24 h 48 h 72 h

-2 24 h 48 h 72 h Mock Relative abundance Mock 1 2 Cluster Transcription factor activity CDCA5 CD4 HEAVY Cell cyle HEAVY RRP36

Microtubule cytoskeleton organisation TAF1C + HIV + BHIV C 0 Env-GFP ZNF593 Cluster 3 Cluster Steroid biosynthetic process WCP PMP Mix aliquots of and SMN1

0 h 6 h MEDIUM Mix aliquotsRegulation of MEDIUM of RTi andprotein complex dissasembly 24 h 48 h 72 h

RRP8 HEAVY cells at 3x timepoints HEAVY cells at 3x timepoints Peptidase activiy 24 h 48 h 72 h 400 400 400 VSVg

Cluster 4 Proteolysis↓ ↓ ↓ ↓ ↓ ↓ 1 Digest proteins, fractionate peptides Digest proteins,200 fractionatep=0.05200 peptides 200 2 and analyse by LC/MS2 and analyse by LC/MS Frequency 0 0 0 0 -4 0 4 -4 0 4 -4 0 4 0 h 6 h RTi

B 24 h 48 h 72 h Relative abundance Log2 (ratio) Log2 (ratio) Log2 (ratio)

B

B CB DC E

48 h post-infection 0 h 6 h 24 h 48 h 72 h RTi 0 h 6 h 24 h 48 h 72 h RTi

UNG UNG KIF18B KIF18B ② ② NKX2-5 1 1 ISG15NKX2-5 BiP PDI I.B DDX31 DDX31 1 1 1

Mock HIV NormalisedCCDC137Proteomic I.B intensity analysis ratio ratio CCDC137 +2 +2 FAM60A FAM60A 0 0 CD4 0.1DUSP11 0.1 DUSP11

0 h 6 h 0 0

RTi 0 0 h 6 h

RTi 0 UNG 0 24 h 48 h 72 h UNG 24 h 48 h 72 h 0 h 6 h 0 h 6 h 0 h 6 h RTi RTi GNL3L RTi Relative abundance Relative abundance GNL3L 24 h 48 h 72 h 24 h 48 h 72 h HLTF 24 h 48 h 72 h HLTF -2 -2 HLTF HLTF 1 1 APOBEC3G 0.3CDCA5 N.D. CDCA5 RRP36 RRP36 F TAF1C TAF1C H 0 0 0.5ZNF593 0.5 ZNF593 RRM2 SMN1 PMP SMN1 WCP 0 h 6 h RTi 0 h 6 h RTi 24 h 48 h 72 h 24 h 48 h 72 h GelsolinRRP8 CAPG RRP8 1 1 PPP2R5A 0.0 0.1 403 401 629

0 0 0 h 6 h 0 h 6 h RTi RTi 24 h 48 h 72 h 24 h 48 h 72 h Relative abundance PPP2R5D 0.1 0.1

G 245/403

60%

HIV-1 p24 161/401 AURKA AURKB PLK1 40% 1 1 1

HIV-1 Vif 20% 71/629

0 0 0 % glycosylated 0% 0 h 6 h 0 h 6 h 0 h 6 h RTi RTi RTi 24 h 48 h 72 h 24 h 48 h 72 h 24 h 48 h 72 h Relative abundance

Calreticulin PMP WCP PMP & WCP A

A Figure S4

Enrichment score 0 1 2 3 4 5 6 Protein phosphatase regulatory activity Deoxyribonucleotide metabolic process

Cluster 1 Cluster Nucleotide binding Ribonucleoprotein complex biogenesis DNA binding

Cluster 2 Cluster Transcription factor activity Cell cyle Microtubule cytoskeleton organisation

Cluster 3 Cluster Steroid biosynthetic process Regulation of protein complex dissasembly Peptidase activiy

Cluster 4 Proteolysis p=0.05 Figure 3 – figure supplement 1 A B Figure S1 WCP PMP

B CD4 Tetherin48 h post-infectionHLA-A SNAT1 1 1 1 1

0 0 I.B 0 0

Mock HIV NormalisedProteomic I.B intensity analysis ratio ratio 0 h 6 h 0 h 6 h 0 h 6 h 0 h 6 h RTi RTi RTi RTi 48 h 72 h 24 h 24 h 48 h 72 h 24 h 48 h 72 h 24 h 48 h 72 h Relative abundance CD4 0.1 0.1 B C WCP PMP APOBEC3G 0.3 N.D.

24 h 48 h 72 h 400 400 400 VSVg 0.51 0.5 200 200 RRM2200

Frequency 0 0 0 0 -4 0 4 -4 0 4 -4 0 4 0 h 6 h RTi

PPP2R5A 24 h 48 h 72 h 0.0Relative abundance 0.1 Log2 (ratio) Log2 (ratio) Log2 (ratio)

D E PPP2R5D 0.1 0.1

ISG15 BiP PDI 1 1 1 HIV-1 p24 0 0 0 0 h 6 h 0 h 6 h 0 h 6 h RTi RTi RTi Relative abundance Relative abundance 24 h 48 h 72 h 24 h 48 h 72 h HIV-1 Vif 24 h 48 h 72 h

F H Calreticulin PMP WCP Gelsolin CAPG 1 1 403 401 629

0 0 0 h 6 h 0 h 6 h RTi RTi 24 h 48 h 72 h 24 h 48 h 72 h Relative abundance

G 245/403

60%

161/401 AURKA AURKB PLK1 40% 1 1 1

20% 71/629

0 0 0 % glycosylated 0% 0 h 6 h 0 h 6 h 0 h 6 h RTi RTi RTi 24 h 48 h 72 h 24 h 48 h 72 h 24 h 48 h 72 h Relative abundance PMP WCP PMP & WCP Figure 4 – figure supplement 1 Figure S5 A A Figure S5

Mock 48 h + HIV WT 48 h + HIV ΔVifΔVIf 48 h

↓ ↓ ↓ ↓ ↓ ↓ ↓ ↓ ↓

Digest proteins and either label peptides with TMT reporters or enrich phosphopeptides with titanium dioxide then label with TMT reporters ① ②

↓

Mix peptides or phosphopeptides, fractionate and analyse by LC/MS3

① ②

B Mock HIV WT HIV ΔVif Infected Infected Infected 0.8% 77% 77% CD4

Env-GFP Figure 4 – figure supplement 2

A CEM-T4 T cells

Secondary only

WT ∆Vif GFP negative GFP positive 2409

2148 2207 592 PPP2R5D

GFP PPP2R5D PPP2R5D

B Primary human CD4+ T cells Secondary only

WT ∆Vif GFP negative GFP positive 53.5 70.7 50.8 31.6 PPP2R5D

GFP PPP2R5D PPP2R5D

1662 2124

1666 2334 ICAM3

GFP ICAM3 ICAM3 Figure 5 - figure supplement 1

A 293T 293T 293T WT HA-PPP2R5D 293T WT HA-PPP2R5D 293T WT

Lysate IP: IP: Mw HA PPP2R5D Vif-FLAG K5-FLAG Empty Vector Vif-FLAG K5-FLAG Vif-FLAG K5-FLAG

(KDa) Empty Vector Empty Vector Vif-FLAG K5-FLAG Vif-FLAG K5-FLAG Empty Vector Empty Vector 75 IB: 75 IB: IB: 75 HA HA PPP2R5D

37 37 IB: 37 IB: IB: FLAG FLAG 25 FLAG 25 25

B CEM-T4 + BZB - BZB + BZB - BZB

Lysate IP: PPP2R5D Mock HIV WT HIV WT Mock HIV WT HIV WT IB: 75 IB: 75 PPP2R5D PPP2R5D

25 IB: IB: 25 Vif Vif Figure 5 - figure supplement 2 A +0 h +4 h +8 h +12 h

HIV WT  PPP2R5D GFP 

HIV WT + BZB      PPP2R5D GFP B GFP+ cells HIV WT + BZB  W T BZB W T NI

W T BZB 1.0  HIVW WT T NI population

1.0

0.5Normalised MFI of GFP+

   Time (h)   0.5 0.0 C +0 h +8 h +12 h 0 4 8 12 +4 h

0.0 HIV WT + CHX 0 4 8 12 PPP2R5D GFP

HIV ΔVif + CHX PPP2R5D

GFP D GFP+ cells 

1.0 HIVdVif ΔVif CHX + CHX  WT CHX 1.0

population dVif CHX HIVWT WT CHX + CHX 0.5 Normalised MFI of GFP+

 0.5   Time (h)  

0.0 0 4 8 12

0.0  0 4 8 12



     Figure 5 - figure supplement 3

Pulse 15 min HIV WT HIV ΔVif Mw Chase (h) 0 1 2 4 0 1 2 4 (KDa) IP: 75 PPP2R5D

S5 Figure S5 Figure

↓

Figure SX Figure

77% h 3 Infected enrich

↓ HIV ΔVif HIV ② + HIV ΔVIf 48 48 ΔVIf HIV +

AURKBAurB MAPKAPK2 AurB   

↓ AURKAAurA Chk2 AurA analyse LC/MS by Chk2 ATR HIPK2 = 0.008 = 0.01 = 0.0003 2 2 2 reporters or r CDK1 PKCB P38A p=0.0004 r p<0.0001 r p=0.5

and DNAPK Chk1

PKCB         

PAK4 Ret Chk2      

CDK1

CDK2 PLK1

↓ ATM PKCB

PLK1 BI2536

 PKCD   h ATM ERK1 HIPK2 PAK4 p70S6K

77% PKCE

fractionate ATR CDK2 Infected

71/629 WCP 71/629 , ASK1 CDK2 WCP

Chk1   

↓ ↓ P38A CDC7 MEK1

peptides with TMT TMT peptides with

TTK MAPKAPK2 MKK4 WCP & PMP Figure S1 161/401 Figure S1 Figure WCP & PMP

161/401 629 HIV WT HIV 629    -5 CDK5 P38D Enrichment score Ret

ERK1 PKG1 iso2 MAPKAPK2

+ HIV WT 48 WCP   

WCP PMP label label 245/403 PMP

KIS 245/403 DNAPK

TTK

PLK1 RTi = 0.02 = 0.01 = 1x10 RTi RTi p=0.05 RTi RTi NEK2 PKCD ↓ RTi

2 2 2    ①

r p<0.0001 p<0.0001 r

DNAPK PAK1 PKCA p=0.9 r

71/629 72 h 72 WCP 72 h 72 72 h 72 72 h 72 72 h 72 401 ASK1 h 72 Akt1 401 BRAF

phosphopeptides

ERK1

P38D 0123456 MKK4          48 h 48 48 h 48 48 h 48       48 h 48 48 h 48 0% 48 h 48 0%

Ret

MEK1 60% 40% 20% MEK1 60% 40% 20%

② Figure S1 WCP & PMP

629 161/401

AZDZM    glycosylated % 24 h 24 glycosylated % 24 h 24 PRKD1 24 h 24 24 h 24 h 24

P38D h 24 CDK7

↓

PMP

PAK1 CDK1 PMP PKCT

PDI PDI and either 6 h 6

6 h 6

6 h 6 6 h 6

6 h 6 6 h 6 WCP VSVg    VSVg PAK1 PMP

PKCD JNK1 245/403

403

SNAT1

403

SNAT1

PKG1 iso2 BUB1 ERK2 0 h 0 RTi 0 h 0 RTi 0 h 0 0 h 0 RTi 0 h 0 h h 0



p90RSK HIPK2 PKCI 

DNA repair

72 h 72 72 h 72 Cot h 72 Akt1 AURKBAurB 401

peptides or Cot NEK2 

↓

0.8% Cot proteins 1 0 1 0 1 0 1 0 1 0 h 48

1 h 0 48

DNA packaging

h 48

0%

Infected PAK4

p70S6K NEK2 Meiotic cycle cell Meiotic 

60% 40% 20%

Relative abundance Relative 

abundance Relative ①

H

H ARAF RTi

P38A glycosylated % RTi 24 h 24 phosphopeptides reporters titaniumwith dioxide thenTMT label with CDK5 RTi  RTi 24 h 24 RTi RTi h 24 Mock Mock 48

PKACA PKACA PKCH PMP

PDI 72 h 72  72 h 72 6 h 6 72 h 72  h 72 6 h 6 Mix h 72 72 h 72 Akt2 BUB1 h 6 VSVg Digest BUB1  = 0.05 = 0.03 = 0.0006

403

SNAT1

2 2 2

GTF2F1 PKACA

ERK2

DNA damage checkpoint

r p<0.0001 p<0.0001 r p=0.3 r

48 h 48 48 h 48 0 h 0 48 h 48 48 h 48 h 0 48 h 48 0 h 0 ↓ h 48

A A

KIS DYRK2 CAMK2A

C

C

Chromosome condensation

  

M phase of mitotic cell cycle

     

 

24 h 24   p90RSK  24 h 24 CAMK2A h 24 PKCA h 24

24 h 24

5 h 24

- PMP

PMP CAMK2A Chk1 RAF1

Actin cytoskeleton organisation

BiP

BiP



1 0 Figure S3 Figure h 6 1 0 6 h 6 1 0 Env-GFP h 6  6 h 6 6 h 6  h 6

PLK1 CD4 DYRK1A GTF2F1 p90RSK PLK1 MLN8054

HLA- HLA-

GSK3B AURKAAurA abundance Relative CK1E H 0 h 0 RTi 0 h 0 0 h 0 RTi 0 h 0 RTi 0 h 0 0 h 0  Response to DNA damagestimulus DNAResponse to

KIF18B NKX2

DDX31 Akt2CCDC137 FAM60A DUSP11 UNG GNL3L HLTF CDCA5 RRP36 TAF1C ZNF593 SMN1 RRP8 ERK2 PLK3



 RTi

72 h 72 72 h 72 DYRK2 GSK3B Src h 72 Regulation of cytoskeleton organisation



Regulation of myeloid differentiation ofmyeloid cell Regulation

GSK3B JNK1 TTK

72 h 72

1 0

1 0 1 0 48 h 48 1 0 1 0 1 0

48 h 48

A mTOR ATM h 48

CDK7 B

(fold change)

4

A

4 C abundance Relative abundance Relative

PKG1 2 RTi HIV WT vs Mock vs HIV WT RTi RTi AMPKA1 PKCH RTi h 48 RTi 24 h 24 RTi h 24 HIV ΔVif vs Mock ΔVif HIV h 24 WCP

WCP PKG1 iso2

PMP CK2A1

 

DYRK1A 

BiP

72 h 72

72 h 72 72 h 72 72 h 72 72 h 72 6 h 6 72 h 72 6 h 6 6 h 6

24 h 24 Src DYRK1A ATR PLK1 (ratio)

(ratio)

HLA-

0

0

PKCH CDK5

2

CK1E

2

48 h 48

48 h 48 48 h 48 72 h h 48

48 h 48

0 h 0 72 h h 48

A h 0   h 0 

6 h 6 PMP PMP CDC7 NuaK1 Akt1 E

E

= 0.0007 = 0.0004 = 0.001

Log h 24 24 h 24 24 h 24

Log p70S6K CK2A1 h 24 24 h 24 NuaK1

24 h 24 2 2 2

r r p=0.5 p=0.4 p=0.1 r

0 h 0 -4

-4 BRAF CK1E PKR -2

1 0 h 6 6 h 6 6 h 6 6 h 6 6 h 6 1 0 6 h 6 1 0

0

0 Src       RSK2    NuaK1 CAPG

      CAPG 4 DMSO A

4 Taxol

4

Relative abundance Relative AURKB Tetherin

AURKB

Tetherin

MARK2 RSK2 Akt2 0 h 0

400 200 RTi 0 h 0 0 h 0 400 200 RTi Figure 6 – figure supplement 1 supplement figure – 6 Figure 5 RTi 0 h 0 0 h 0

WCP MARK2 h 0 ASK1

PKCT

  

72 h 72 72 h 72 PKG1 P38B MKK4 h 72 (ratio) 4 (ratio) 0 0 (ratio)

0

2

CDK3

2 PRKD1 DYRK2

2

1 0 48 h 1 0 48 h 1 0 1 0 1 0 h 48 1 0 48 h 48 72 h h 48

HEAVY WCP WCP PKR PKG1 GTF2F1 MEDIUM    PMP 3

E

Log

Log CDK9 RTi MARK2 RTi AMPKA1RTi 24 h 24 RTi RTi RTi RTi Log 24 h 24 RTi h 24

-4

-4 PKCA ULK1

PKCE -4

2

2 h   

and

72 h 72 h 72 h 72 6 h 6 72 h 72 72 h 72 72 h 72 72 h 72 6 h 6 72 h 72 0 0 PLK3 h 6 ↓

mTOR P38B 0

4

4 CAPG

4

72

AURKB

Tetherin CDK3 PKCT CDK4

400 200 1    400 200

48 h 48 48 h 48 48 h 48 h 0 48 h 48 400 200 h 48 48 h 48 h 48 0 h 0 48 h 48

② RAF1 h 0 P38B CDK3

timepoints

24 h 24 CDK14 h 24 24 h 24 24 h 24 CK2A1 h 24 0 24 h 24    CDK7 24 h 24 24 h 24 (ratio) (ratio) 0 (ratio) 0

0

2

2 RAF1 BRAF CDK6 by LC/MS by

2

h 24 h

MEDIUM

24 h

CD4 1 0 CD4 48 h h 6 1 0 h 6 6 h 6 1 0 6 h 6 h 6 6 h 6 6 h 6 6 h 6 = 0.4 = 0.03 = 0.3    ↓ PKR ULK1 WCP CDK14 -1 ISG15 ISG15

2 2 2

fractionate peptides

AURKA Gelsolin Log

AURKA 48 Gelsolin r p=<0.0001 r p=0.0002 p<0.0001 r

Log CDK9 CDK6

Log JNK1 0 h 0 RTi 0 h 0 RTi 0 h 0 RTi 0 h 0 0 h 0 RTi 0 h 0 0 h 0 0 h 0 -4 -4

PLK3 PRKD1-4 RSK2 -2

        

     

0 cells at 3x 0 72 h 72 72 h 72 72 h 72 PKCI ARAF h 72

0 mTOR

analyse

4

2

  ULK1  0 - +2 400 200

CDK4 -3 CDC7 400 200

(fold change) in h

400 200 HIV WT vs Mock WT HIV Okadaic Acid

1 0 1 0 1 0

1 0 1 0 1 0

1 0 h 48

2 h 48 1 h 0 48

h 48

↓ CDK6 CDK4 CDK9

proteins,

Relative abundance Relative

Relative abundance Relative

Relative abundance Relative

Relative abundance Relative

Relative abundance Relative

Relative abundance Relative

Frequency Relative abundance Relative

Relative abundance Relative

and Frequency

24  

PKCE 

-4 CDK14 AMPKA1

24 h 24 24 h 24

24 h 24

(ratio) h 24 Log

0

2 2 Mix aliquotsMix of 2

ARAF 2 Inhibitor treated vs untreated HeLa cells log (fold change) (fold log (fold change) (fold HEAVY PKCI log

log

KIS change) (fold 2

24 h

5 CD4

6 h 6 h 6

6 h 6

- h 6

f phosphopeptide abundance

3 3 2 2 1 1 0 0

1 1 2 2 3 3

3 2 1 0 3 2 1 0

3 1 2 3 1 2 3 3 2 2 1 1 0 0 1 1 2 2 3 3

ISG15

‐ ‐ ‐ ‐ ‐ ‐

2 ‐ ‐ ‐ ‐ ‐ ‐ HIV WT / Mock / WT HIV HIV WT HIV HIV ΔVif HIV /

‐ ‐ ‐ ‐ ‐ ‐

Digest

Mock / ΔVif HIV Activity score score Activity

-Log q value (signifcance) value q AURKA

Gelsolin

Log

0 h 0 0 h 0 0 h 0 h 0

G

B D F G -4

A B D F A C

CEM-T4 cells CEM-T4

0 h cells CEM-T4 Mock cells CEM-T4 0 + HIV

200 400

0 1 0 1 0 1 0 1

Relative abundance Relative

Relative abundance Relative

Relative abundance Relative

Relative abundance Relative

Frequency

RTi

RTi G B

D F A

72 h 72 72 h 72

48 h 48 48 h 48

24 h 24 24 h 24

6 h 6 UNG h 6 HLTF

0 h 0 0 h 0 1 0 1 0

B A Figure 7 – figure supplement 1

% Similarity

SIVcpz HIV-2/SIVsmm/SIVmac SIVagm SIVcp SIVcp SIVcp SIVcp SIVcp HIV-2 SivSmm SIVSmm SIVmac SIVagm SIVagm A1 A2 AE B1 B2 B3 LAI NL4-3 C1 C2 C3 D1 D2 F2 SIVgor EV1 MB897 LB7 EK505 Tan1 Tan21 ROD pBJ Pgm 239 Sab Tan1 A1 x 96.8 94.9 95.4 94.0 94.2 95.8 94.9 94.5 94.9 95.2 94.2 94.2 92.8 92.1 93.1 90.5 87.5 84.3 85.5 63.3 64.2 63.7 63.7 59.8 62.1 49.9 A2 89.1 x 94.9 94.2 93.8 93.8 94.9 94.7 93.8 94.5 93.8 94.0 93.3 92.8 92.1 92.6 90.3 87.3 83.6 84.8 63.7 64.4 64.0 63.7 59.1 62.4 49.2

AE 82.8 82.8 x 93.8 93.5 94.2 94.9 94.2 93.8 94.0 94.2 93.5 93.5 92.8 93.1 94.5 91.0 88.5 84.3 84.8 64.0 65.4 64.9 64.4 59.6 62.6 49.9

B1 80.7 79.2 77.6 x 96.3 96.8 97.9 96.5 94.0 95.2 95.6 95.8 95.6 94.5 91.7 93.1 91.7 86.8 84.8 85.2 63.3 64.0 63.5 63.7 60.3 62.4 49.2

B2 81.8 80.7 82.3 86.5 x 96.1 96.5 95.8 93.5 94.0 94.5 94.9 94.5 94.0 90.8 92.1 90.8 87.5 84.1 85.0 64.2 64.2 63.7 63.7 60.5 62.6 48.7

B3 81.8 79.7 81.8 88.0 88.0 x 97.5 96.5 94.0 94.7 94.7 95.6 95.2 94.2 91.2 92.6 91.5 86.8 83.8 85.0 63.0 64.0 63.5 63.0 58.9 62.4 49.4

LAI 84.4 81.3 84.4 88.5 88.5 92.7 x 97.2 94.7 95.4 96.1 96.5 96.1 94.5 92.6 93.8 92.1 88.0 84.1 85.0 63.5 64.4 64.0 63.5 59.6 62.6 49.7

NL4-3 82.3 81.8 81.8 89.6 88.0 90.1 89.6 x 94.2 94.7 94.5 94.9 94.7 94.9 92.1 93.5 91.2 87.1 84.1 85.0 62.8 64.2 63.7 63.0 59.4 62.8 49.7

C1 78.6 77.1 78.6 78.1 79.7 80.2 81.8 80.2 x 97.7 97.7 94.0 95.2 93.1 92.1 93.3 90.5 87.1 84.3 85.0 63.7 65.4 64.9 64.2 58.9 62.1 50.3

C2 79.7 79.2 79.7 83.3 82.3 82.3 82.8 82.8 91.7 x 98.2 94.5 94.7 94.0 92.6 93.8 91.0 86.8 84.8 85.0 64.2 65.1 65.1 64.7 59.6 61.9 49.7

C3 81.3 77.1 80.2 81.3 81.3 82.3 83.9 80.2 91.7 91.7 x 94.7 95.6 94.2 92.8 94.0 91.5 87.5 84.5 85.2 63.5 65.1 64.7 64.2 59.8 62.1 49.7

D1 80.2 79.2 77.6 84.9 82.3 83.9 84.4 82.3 78.6 79.2 80.7 x 96.5 94.7 91.7 92.1 91.0 87.8 83.6 84.3 63.3 64.0 63.5 63.0 59.4 62.4 49.9

D2 78.6 77.1 77.6 84.9 81.8 85.4 87.5 82.8 82.3 80.7 83.3 89.1 x 94.5 91.0 92.1 89.8 88.0 83.1 83.6 64.0 64.7 64.7 64.2 59.6 62.8 49.9

F2 78.1 77.1 78.6 80.7 81.3 83.3 82.8 83.9 77.6 79.7 80.2 81.8 82.8 x 90.1 91.0 90.5 87.5 83.4 84.5 63.7 64.4 64.4 64.0 60.7 62.8 49.2

SIVcpMB897 74.7 75.3 78.5 73.1 73.7 74.2 77.4 77.4 72.6 73.7 73.7 72.0 72.6 72.6 x 95.4 91.5 88.5 83.8 83.6 61.7 63.5 63.3 62.4 57.3 61.2 49.2

SIVcpLB7 76.0 75.0 80.2 75.5 76.6 75.5 78.1 78.1 73.4 75.5 76.0 73.4 74.0 72.9 87.1 x 92.6 89.4 85.7 85.5 62.8 64.2 64.0 63.0 58.4 62.4 49.4

SIVcpEK505 71.4 71.4 72.9 72.4 71.9 72.4 75.5 72.4 69.3 70.8 69.8 73.4 70.3 71.4 73.7 75.6 x 88.5 86.1 88.0 63.5 63.7 63.3 63.5 59.4 62.8 48.7

SIVgor 59.9 58.3 60.4 55.7 56.3 56.3 57.3 56.3 55.2 54.7 57.8 58.9 59.9 59.9 62.9 62.7 61.1 x 82.7 83.6 64.9 64.4 64.9 64.4 60.7 61.0 48.3

SIVcpTan1 55.7 55.2 55.2 55.7 56.8 55.2 55.2 55.2 55.2 55.7 56.3 57.3 54.7 54.7 57.0 58.0 61.7 51.8 x 95.8 62.8 62.8 63.0 63.7 59.4 59.8 47.8

SIVcpTan21 56.3 55.2 54.7 55.7 56.3 56.3 56.3 55.2 55.7 55.2 56.3 56.8 54.7 54.7 56.5 56.5 63.2 54.9 84.9 x 63.3 63.3 63.0 63.7 59.4 60.3 47.3

HIV-2 ROD 32.3 33.3 33.9 31.8 33.3 31.8 31.3 31.3 31.3 31.8 31.3 31.3 33.3 32.8 26.9 30.1 32.6 36.3 31.2 30.2 x 89.6 90.1 89.4 69.3 67.9 36.0

SivSmm pBJ 30.7 31.8 32.3 30.7 30.7 30.2 29.7 30.2 31.8 30.7 30.7 31.8 32.8 30.2 28.0 28.5 30.6 33.7 29.1 28.1 72.6 x 97.7 92.6 69.3 70.9 37.6

SIVSmm Pgm 28.6 29.7 30.7 28.6 28.6 28.1 27.6 28.1 29.7 29.2 28.6 29.7 31.8 29.7 26.3 26.9 29.0 34.2 30.2 28.1 71.6 93.0 x 92.8 69.3 70.4 37.4

SIVmac239 29.7 30.7 30.2 29.2 29.2 28.6 28.6 28.1 29.2 28.6 28.6 29.2 31.8 30.2 27.4 26.9 30.1 33.2 30.7 29.1 72.6 81.9 80.9 x 70.2 69.3 37.2

SIVagmSab 33.3 32.3 32.3 32.3 32.3 31.8 32.8 31.3 31.3 32.3 33.3 32.8 33.9 35.4 29.0 30.6 32.6 37.3 32.2 32.2 37.0 36.7 37.2 38.6 x 65.8 32.1

SIVagmTan1 37.0 36.5 34.4 35.9 37.0 35.9 35.4 37.0 38.0 38.0 37.0 36.5 38.5 34.9 31.7 33.2 35.2 32.1 30.7 31.7 38.0 39.1 38.1 37.2 40.6 x 34.6

EV1 15.1 14.6 15.6 16.7 15.1 14.6 15.6 16.1 15.1 15.6 13.5 16.1 16.7 16.1 15.1 15.5 16.1 13.5 14.1 13.6 10.2 12.1 13.0 11.2 10.0 11.3 x % Identity Figure 7 – figure supplement 2 A 293T HA-PPP2R5A cells B Figure S7 100% 100% NL4-3 A2  B2 C3 SIVcpz 50% 50% HIV-2

 transfected cells transfected cells Normalised HA-signal in Normalised HA-signal in 0% 0%

           

  293T cells expressing: C 293T HA-PPP2R5B cells

Empty vector NL4-3 C133S NL4-3 LAI B1 274 274 300 253 215 289 235 40 59 60

A1 C2 C3 D1 D2 257 292 227 227 274 84 38 30 30 64

SIVcpzMB897 SIVcpzEK505 SIVcpzTAN1 SIVcpzTAN2 SIVgor 213 237 284 269 286 87 112 87 67 74

SIVsmmPGm SIVsmmPBj HIV-2ROD SIVagmSab SIVagmTan 286 279 257 352 274 89 96 151 99 91 Figure 7 - figure supplement 3 A 293T 293T 293T WT HA-PPP2R5E 293T WT HA-PPP2R5E

Mw Lysate IP : HA SRLV SRLV Vif-FLAG Empty Vector SRLV Vif-FLAG Empty Vector SRLV Vif-FLAG Empty Vector SRLV Vif-FLAG (KDa) Empty Vector

75 IB: HA 75 IB:HA

37 37 IB: FLAG IB: FLAG

25 25 B

NL4-3 Vif Sh-Cntrl Sh-elob Sh-cbfb

293T-PPP2R5E cells 313 334 313 55 248 209 Untransfected cells +shRNA Vif + sh-Cntrl

Vif + sh-target

HA-PPP2R5E

SRLV Vif Sh-Cntrl Sh-elob Sh-cbfb

305 337 316 49 167 49

HA-PPP2R5E