HSP70 and its cochaperone CPIP promote potyvirus infection in Nicotiana benthamiana by regulating viral coat protein functions

Abstract

This study demonstrates that heat shock protein 70 (HSP70) together with its cochaperone CPIP regulates the function of a potyviral coat protein (CP), which in turn can interfere with viral gene expression. HSP70 was copurified as a component of a membrane-associated viral ribonucleoprotein complex from Potato virus A-infected plants. Downregulation of HSP70 caused a CP-mediated defect associated with replication. When PVA CP was expressed in trans, it interfered with viral gene expression and replication-associated translation (RAT). However, CP produced in cis interfered specifically with RAT. CPIP binds to potyviral CP, and overexpression of CPIP was sufficient to restore RAT inhibited by expression of CP in trans. Restoration of RAT was dependent on the ability of CPIP to interact with HSP70 since expression of a J-domain mutant, CPIP(Delta66), had only a minor effect on RAT. CPIP-mediated delivery of CP to HSP70 promoted CP degradation by increasing its ubiquitination when assayed in the absence of virus infection. In conclusion, CPIP and HSP70 are crucial components of a distinct translation activity that is associated with potyvirus replication.

Figure 1.

HSP70 as a Component of a Purified Viral Protein Complex. (A) Diagram of the reporter- and SIII-tagged PVA. The position of the SIII-tag in the sequence of wild-type PVA VPg and NIb is shown. (B) Affinity purification of membrane-associated viral RNP complexes. A silver-stained SDS-PAGE gel of column eluates from strep-tactin sepharose is shown. Numbers indicate the positions of proteins identified by LC-MS/MS analysis as follows: (1) HSP70-3, (2) NIb, and (3) NIa. The masses of the molecular mass markers are indicated on the left. (C) The specific presence of LC-MS/MS identified proteins in the column eluates of NIb^SIII and VPg^SIII samples was verified by immunoblot analysis. Protein species are indicated on the right and molecular mass markers on the left. (D) The presence of viral RNA in the column eluates of NIb^SIII and VPg^SIII samples was analyzed by RT-PCR with the expected product being 350 bp. RNA samples from SIII purifications were incubated prior to PCR with (+) or without (−) reverse transcriptase (Moloney murine leukemia virus [M-MLV]-RT) either in the presence of a (−)-strand or a (+)-strand-specific primer. PVA icDNA was used as a positive control and water as a negative control in the PCR reactions.

Figure 2.

Inhibition of PVA^wt Infection by Downregulation of HSP70. (A) Immunoblot analysis of HSP70 protein levels in TRV:00 and TRV:HSP70 infected N. benthamiana plant leaves at 7 DAI. A protein band on the Ponceau S–stained membrane at the approximate position of ribulose-1,5-bisphosphate carboxylase/oxygenase was used as a loading control. (B) Inhibition of PVA systemic accumulation in TRV:HSP70 plants. Viral gene expression was analyzed from upper, noninoculated leaves. The average RLUC value presented for PVA^wt in TRV:HSP70 plants was calculated from those three plants that developed systemic infections. No PVA accumulation was detected in 8 of 11 plants at 5 and 6 DAI. The asterisk indicates the low level of RLUC for PVA^wt in the upper leaves of TRV:HSP70 plants at 5 DAI. (C) Inhibition of virus-independent FLUC expression in TRV:HSP70 plants. FLUC values used for normalizing viral gene expression are shown. The FLUC value is the mean from the three different viral assays (PVA^wt, PVA^CPmut, and PVA^ΔGDD). (D) and (E) Transient delay of PVA^wt infection in TRV:HSP70 plants. RLUC values were normalized using FLUC expression in either TRV:00 or TRV:HSP70 plants, respectively. An arrow is used to indicate the delay in PVA^wt gene expression. (F) Inhibition of PVA^wt gene expression in TRV:HSP70 plants at 2 DAI. At this time point, viral gene expression was detected in all samples except in PVA^wt infected TRV:HSP70 plants. Rluc values were normalized using FLUC values. (G) and (H) Inhibition of viral expression and, in particular, PVA^wt infection by quercetin. Gene expression of PVA^wt, PVA^CPmut, and PVA^ΔGDD infected leaves were assayed at 2 DAI (G) and 4 DAI (H). Values are presented for PVA^wt, PVA^CPmut, and PVA^ΔGDD in the presence or absence of quercetin. Note the differing y axis scales in (G) and the subpanels of (H). (I) FLUC expression in the quercetin assay. Shown are the FLUC values used for normalizing viral gene expression. Each FLUC value is the mean from three different viral assays (PVA^wt, PVA^CPmut, and PVA^ΔGDD) at 2 DAI. The error bars indicate the sd.

Figure 3.

Inhibition of Viral Gene Expression by the Viral CP. (A) Time course of PVA^wt gene expression in control plants (pRD400) and when coexpressed with CP^wt or CP^mut. (B) Time course of PVA^CPmut gene expression in control plants (pRD400) and when coexpressed with CP^wt or CP^mut. (C) Time course of PVA^ΔGDD gene expression in control plants (pRD400) and when coexpressed with CP^wt or CP^mut. (D) Expression of CP^wt and CP^mut was detected by immunoblot analysis using anti-CP. These samples did not contain any virus. Total protein from PVA infected N. benthamiana leaves was used as a positive (+) and pRD400 infiltration as a negative (−) control. (E) CP-mediated inhibition of gene expression is virus specific. Expression of virus-independent FLUC in pRD400 control plants and when coexpressed with CP^wt or CP^mut. The FLUC values are derived from the infection assays above ([A] to [C]). (F) CP^wt coexpression inhibits PVA^CPmut gene expression less than that of PVA^wt or PVA^ΔGDD. The viruses used are indicated below their respective columns. The error bars indicate the sd.

Figure 4.

CPIP Can Regulate CP-Mediated Inhibition of Viral Gene Expression. (A) Gene expression of PVA^wt in pRD400 control plants and when coexpressed with CPIP and CPIP^Δ66 at 2 and 3 DAI. (B) Expression of CPIP and CPIP^Δ66 was detected by immunoblot analysis using anti-c-myc. Plant total protein was used as negative control (−). (C) General inhibition of potyvirus infection by constitutive CPIP^Δ66 expression. PVA, TEV, and TVMV CP levels in systemic leaves of control and transgenic plants expressing CPIP^Δ66 (line CPIP-39; Hofius et al., 2007) at 13 (PVA) and 14 (TEV and TVMV) DAI. Values represent means (n = 20) ± se and are given as percentage of wild-type level. Plants had developed six to eight leaves prior to virus infection and were sampled in leaves 5 to 7 above the inoculated leaf. (D) and (E) Coexpression of CPIP or CPIP^Δ66 antagonizes CP-mediated inhibition of viral gene expression. Viral gene expression was analyzed in plants expressing CP^wt alone or together with CPIP or CPIP^Δ66 at 3 (D) and 4 (E) DAI. Asterisks indicate the low viral gene expression during CP^wt coexpression only. The error bars indicate the sd.

Figure 5.

CP Interacts with CPIP and HSP70. (A) Analysis of HSP70 coprecipitation in CP immunoprecipitation. The proteins expressed in each sample are indicated at the top. The left-most sample (in [A] and [B]) is plant total protein (+ control). (B) Sensitivity of HSP70 coprecipitation with CP to Mg-ATP. CP was expressed and immunoprecipitated either in the absence (−) or presence (+) of Mg-ATP. Precipitation of both CP and HSP70 was detected by immunoblot analysis. (C) Analysis of HSP70 and CP coprecipitation in CPIP immunoprecipitation. The proteins expressed in each sample are indicated at the top. The left-most sample (+ control) is virion CP for anti-CP, and total protein from plants expressing CPIP and CPIP^Δ66 for anti-HSP70 and anti-CPIP. (D) The presence of endogenous CPIP in membrane-derived and purified NIb^SIII and VPg^SIII samples detected by immunoblot analysis. The right-most sample (+ control) is total protein from plants expressing CPIP.

Figure 6.

CPIP-Mediated HSP70 Delivery of CP Promotes Modification by Ubiquitin and CP Degradation. (A) CP amounts detected by ELISA when expressed alone or together with CPIP or CPIP^Δ66 given as ng of CP/mL at 3 and 4 DAI. (B) Ubiquitination level of CP when expressed alone or together with CPIP or CPIP^Δ66 at 4 DAI given as the relative amount of ubiquitin per CP. (C) Presence of ubiquitinated high molecular weight CP in CP immunoprecipitates. The proteins expressed in each sample are indicated at the top. IgG heavy chain detected with anti-mouse Ab was used as loading control. The masses of the molecular mass markers are indicated on the left ([C] and [D]). (D) Presence of high molecular weight ubiquitinated proteins and CP in the column eluates of NIb^SIII and VPg^SIII samples.

Figure 7.

Proposed Model for the Role of CPIP and HSP70 in Regulating CP during PVA Replication. During formation and within RCs, viral genomes and proteins are continuously synthesized through replication and RAT. The increasing amount of CP may either inhibit RAT directly or indirectly by inhibiting replication. CPIP functions within the RCs by binding CP and mediating HSP70-dependent regulation, a process required for sustainable RAT. At a later infection stage, CP is abundant, causing depletion of CPIP. Viral multiplication is ceased, and infection can proceed to assembly phase.