Role of cell-type-specific endoplasmic reticulum-associated degradation in polyomavirus trafficking

Abstract

BK polyomavirus (BKPyV) is a widespread human pathogen that establishes a lifelong persistent infection and can cause severe disease in immunosuppressed patients. BKPyV is a nonenveloped DNA virus that must traffic through the endoplasmic reticulum (ER) for productive infection to occur; however, it is unknown how BKPyV exits the ER before nuclear entry. In this study, we elucidated the role of the ER-associated degradation (ERAD) pathway during BKPyV intracellular trafficking in renal proximal tubule epithelial (RPTE) cells, a natural host cell. Using proteasome and ERAD inhibitors, we showed that ERAD is required for productive entry. Altered trafficking and accumulation of uncoated viral intermediates were detected by fluorescence in situ hybridization and indirect immunofluorescence in the presence of an inhibitor. Additionally, we detected a change in localization of partially uncoated virus within the ER during proteasome inhibition, from a BiP-rich area to a calnexin-rich subregion, indicating that BKPyV accumulated in an ER subcompartment. Furthermore, inhibiting ERAD did not prevent entry of capsid protein VP1 into the cytosol from the ER. By comparing the cytosolic entry of the related polyomavirus simian virus 40 (SV40), we found that dependence on the ERAD pathway for cytosolic entry varied between the polyomaviruses and between different cell types, namely, immortalized CV-1 cells and primary RPTE cells.

Fig 1

Time course of proteasome and ERAD pathway involvement during infection. RPTE cells were infected at 5 IU/cell at 4°C for 1 h, moved to 37°C, and treated at the indicated time points with either 10 μM epoxomicin (Epox) or 10 μM Eeyarestatin I (EerI). Whole-cell lysates were harvested at 24 hpi, resolved by reducing SDS-PAGE, and probed for TAg and GAPDH. Similar results were obtained for at least 3 independent experiments. M, mock infected; UT, untreated.

Fig 2

BKPyV trafficking is altered when ERAD is inhibited. RPTE cells were infected at 5 IU/cell at 4°C for 1 h and then moved to 37°C. Cells were treated with 10 μM epoxomicin (epox) or DMSO at 6 hpi and fixed at 24 hpi. (A) BKPyV was stained by FISH against the viral genome, and nuclei were stained with DAPI and then imaged by confocal microscopy. The relative size of fluorescent areas was measured using ImageJ software on >260 puncta from each condition collected from three independent experiments. (B) Fixed cells were stained for VP2/3 and DAPI and imaged by fluorescence microscopy. Quantitation of relative fluorescence was performed on VP2/3 staining alone per cell using ImageJ software on a total of 200 cells from three independent experiments. Values are corrected total cell fluorescence to normalize for cell size. Statistical analysis for both panels was performed by independent-sample t test using GraphPad Prism software. Scatter plots show each data point value along with the mean and standard deviation.

Fig 3

BKPyV behaves like an ERAD substrate. RPTE cells were infected at an MOI of 5 IU/cell and fixed at 24 hpi. Fixed cells were stained for VP2/3 and the ER marker (A) BiP or (B) calnexin (cnx), and images were taken by confocal microscopy. Arrowheads point to enlarged colocalized areas on the right. Colocalization of the individual VP2/3 puncta with calnexin or BiP was measured using MetaMorph software at the Center for Live Cell Imaging at the University of Michigan. Statistical analysis was performed as for Fig. 2. The line represents the median colocalization for each condition. (C) RPTE cells were transfected with CD3δ-YFP and the medium was changed at 12 h posttransfection. Transfected cells were infected at 5 IU/cell at 24 h posttransfection, treated with 10 μM epoxomicin at 6 hpi, and fixed at 24 hpi (48 h posttransfection). Fixed cells were then stained for CD3δ-YFP with anti-GFP and capsid proteins with anti-VP2/3 and imaged by confocal microscopy.

Fig 4

BKPyV enters the cytosol. (A) RPTE cells were infected at 5 IU/cell, harvested at the indicated time points under alkylating conditions, and fractionated into pellet and supernatant fractions. Fractions were separated by nonreducing SDS-PAGE and probed for VP1, ER marker PDI, and cytosolic marker GAPDH. One-third of the protein for the supernatant was loaded for the pellet fractions, and the film was exposed for a shorter time. VP1 bands here include monomers and higher-molecular-weight species that have entered the nonreducing gel. (B) RPTE cells were infected at 5 IU/cell and treated at 6 hpi for 2 h with 1.25 μg/ml BFA or left untreated, harvested at 16 hpi under alkylating conditions, and then separated into pellet and supernatant fractions. Fractions were resolved by nonreducing SDS-PAGE and probed as described above but with BiP as the ER marker. “VP1” here represents monomers that have entered the nonreducing gel. (C) RPTE cells were infected with biotinylated BKPyV at 5 IU/cell, harvested at 16 hpi under alkylating conditions, and assayed as described for panel A. The Western blot was first probed with streptavidin-HRP to show the presence of biotinylated VP1 monomer (biotin-VP1) and then probed for VP1, PDI, and GAPDH.

Fig 5

ERAD inhibition does not prevent cytosolic entry of BKPyV. (A) RPTE cells were infected with BKPyV at 5 IU/cell at 4°C for 1 h and treated with 10 μM epoxomicin (Epox), 10 μM Eeyarestatin I (EerI), or DMSO at 6 hpi, harvested at 16 hpi under alkylating conditions, then separated into pellet and cytosolic fractions, and assayed as for Fig. 4. (B) BKPyV genomic DNA isolated from treated and untreated supernatant fractions was measured by qPCR and normalized to untreated levels. Averages from three independent experiments are shown, with error bars representing standard deviations. A one-tailed t test was performed; the significance for BFA treatment was a P value of 0.0008, and that for epoxomicin treatment was 0.03. Eeyarestatin I treatment did not cause a significant decrease in DNA.

Fig 6

Cell-type-specific requirements for polyomavirus trafficking. (A) CV-1 cells were infected with BKPyV and treated with BFA, epoxomicin (Epox), or DMSO as for Fig. 5A. Pellet and supernatant fractions were separated and assayed as for Fig. 4. (B) BKPyV genomic DNA from the CV-1 supernatant fractions was quantified by qPCR and represented as in Fig. 4B. A t test determined significance of the decrease in BFA treatment to be a P value of 0.02 and for epoxomicin treatment to be a P value of 0.02. (C) RPTE or CV-1 cells were inoculated with 5 IU/cell SV40 at 4°C for 1 h, then moved to 37°C, and incubated in the presence of 10 μM epoxomicin, 10 μM Eeyarestatin I (EerI), or DMSO. RPTE cells were incubated for 24 h with 3.2 μM ganglioside GM1 prior to infection. Whole-cell lysates were harvested at 24 hpi, resolved by reducing SDS-PAGE, blotted, and probed for TAg. (D) As in panel C, RPTE cells were infected with 5 IU/cell SV40 in the presence of 1.25 μg/ml BFA, 10 μM epoxomicin, 10 μM Eeyarestatin I, or DMSO. Proteins were harvested under alkylating conditions at 10 hpi and fractions analyzed as for Fig. 4. The Western blot was probed for SV40 VP1 monomers, PDI, and GAPDH. (E) SV40 genomic DNA was isolated and quantified by qPCR. Values for each condition were normalized to untreated levels; the averages from three independent experiments are represented, with error bars showing the standard deviations. A t test was performed for each condition; the significance in CV-1 cells was a P value of 0.008 for BFA treatment, with no significant decrease for epoxomicin treatment. In RPTE cells, a P value of 0.01 for BFA treatment and a P value of 0.01 for epoxomicin treatment were considered significant.