Endoplasmic reticulum chaperones are involved in the morphogenesis of rotavirus infectious particles

Abstract

The final assembly of rotavirus particles takes place in the endoplasmic reticulum (ER). In this work, we evaluated by RNA interference the relevance to rotavirus assembly and infectivity of grp78, protein disulfide isomerase (PDI), grp94, calnexin, calreticulin, and ERp57, members of the two ER folding systems described herein. Silencing the expression of grp94 and Erp57 had no effect on rotavirus infectivity, while knocking down the expression of any of the other four chaperons caused a reduction in the yield of infectious virus of about 50%. In grp78-silenced cells, the maturation of the oligosaccharide chains of NSP4 was retarded. In cells with reduced levels of calnexin, the oxidative folding of VP7 was impaired and the trimming of NSP4 was accelerated, and in calreticulin-silenced cells, the formation of disulfide bonds of VP7 was also accelerated. The knockdown of PDI impaired the formation and/or rearrangement of the VP7 disulfide bonds. All these conditions also affected the correct assembly of virus particles, since compared with virions from control cells, they showed an altered susceptibility to EGTA and heat treatments, a decreased specific infectivity, and a diminished reactivity to VP7 with monoclonal antibody M60, which recognizes only this protein when its disulfide bonds have been correctly formed. In the case of grp78-silenced cells, the virus produced bound less efficiently to MA104 cells than virus obtained from control cells. All these results suggest that these chaperones are involved in the quality control of rotavirus morphogenesis. The complexity of the steps of rotavirus assembly that occur in the ER provide a useful model for studying the organization and operation of the complex network of chaperones involved in maintaining the quality control of this organelle.

FIG. 1.

The luminal molecular chaperones change their distribution in rotavirus-infected cells. (A) MA104 cells grown on 10-mm coverslips were infected or mock infected with RRV at an MOI of 3, fixed at 8 hpi, and stained with the following antibodies: rat MAb to grp94; mouse MAbs to calnexin (Cnx), calreticulin (Crt), or PDI; rabbit hyperimmune sera to NSP4 (C-239) and NSP5 (C6); and MAb M60 to VP7. They were then stained with the corresponding secondary antibodies: goat anti-rat IgG coupled to Alexa 488, goat anti-mouse IgG coupled to Alexa 568, or goat anti-rabbit IgG coupled to Alexa 488. (B) Cells were grown as described above, infected with RRV at an MOI of 3, and then fixed at different times postinfection (4 h, 6 h, and 8 h) and stained with a MAb to PDI and rabbit hyperimmune sera to NSP4 or NSP5 as primary antibodies and then with the secondary antibodies goat anti-mouse IgG coupled to Alexa 488 and goat anti-rabbit IgG coupled to Alexa 568.

FIG. 2.

siRNAs directed to molecular chaperones decrease the amount of the cognate cellular mRNA and protein. MA104 cells were transfected with siRNAs (Cnx, calnexin; Crt, calreticulin) to the various chaperones as described under Materials and Methods, and the cells were harvested 72 h posttransfection. (A) The proteins were resolved by SDS-10% PAGE, transferred to a nitrocellulose membrane, and stained with the indicated chaperone antibodies and with an antibody to vimentin. The levels of protein were determined by densitometric analysis of the corresponding bands and were normalized to the band corresponding to vimentin (the lower band in each box in the panel), which was used as a loading control. Numbers indicate the increase or decrease in the level of protein, as related to the value obtained for cells transfected with the irrelevant siRNA control (Irr), which was taken as 1, after normalization to vimentin. Only the values with a statistically significant difference (P < 0.01) are shown. The arithmetic means and standard deviations from at least six independent experiments are shown. (B) Total RNA was extracted with TRIzol, and the levels of chaperone mRNAs were determined by real-time PCR as indicated under Materials and Methods. The results are expressed as increases or decreases with respect to the value obtained for the cells transfected with the irrelevant, control siRNA, which was taken as 1. The arithmetic means and standard deviation from at least three independent experiments are shown. The values with a statistically significant difference (P < 0.05) are shown in bold.

FIG. 3.

The yield of rotavirus progeny decreases when grp78, PDI, calnexin, and calreticulin are knocked down. MA104 cells were transfected with the indicated siRNA (Irr, irrelevant; Cnx, calnexin; Crt, calreticulin), and 72 h posttransfection, the cells were infected with RRV at an MOI of 3. Twelve hours postinfection, the cells were harvested, and the level of infectious progeny virus produced was determined by an immunoperoxidase assay as described in Materials and Methods. The virus yield is expressed as a percentage of the yield obtained from cells transfected with the irrelevant siRNA. One-hundred percent infection equals 3.1 × 10⁷ FFU/ml. The reductions in infectivity that were statistically significantly different (P < 0.01 [**] or P < 0.05 [*]) are indicated. The arithmetic means and standard deviations from seven independent experiments are shown.

FIG. 4.

Silencing the expression of calnexin and grp78 decreases the specific infectivity of rotavirus TLPs. Cells were transfected with the indicated siRNA (Irr, irrelevant; Cnx, calnexin), and 72 h posttransfection, the cells were infected with RRV at an MOI of 3. At 5 hpi, the cells were starved for 1 h in methionine-free medium and then metabolically labeled with 50 μCi/ml of ³⁵S-labeled Easy Tag Express for 6 h. The viral particles were purified by isopycnic centrifugation in CsCl gradients as described in Materials and Methods. (A) The amount of ³⁵S incorporated into virus particles (cpm) was determined by liquid scintillation counting of trichloroacetic acid-precipitable counts. Data are expressed as the percentages of DLPs or TLPs obtained in cells transfected with siRNAs to grp94, grp78, and calnexin, relative to the corresponding particles (DLPs or TLPs) purified from cells transfected with the control, irrelevant siRNA. The relative proportions of DLPs versus TLPs are also shown for the particles obtained from cells transfected with each of the siRNAs employed. (B) Specific infectivities of the purified viral particles. The infectivity of the TLPs isolated from cells in which the chaperones were knocked down was titrated by an immunoperoxidase assay, and the numbers of FFU/ml obtained from this assay were divided by the amount of ³⁵S (cpm/ml) obtained from the respective purified TLP preparation. The specific infectivities (FFU/cpm) of the viruses are expressed in units relative to the level of TLPs obtained from cells transfected with the siRNA control. (C) Monolayers of MA104 cells were incubated with different dilutions of purified TLPs for 1 h at 4°C. The cells were then washed with ice-cold PBS and harvested as described in Materials and Methods. The virus bound to cells was determined by an ELISA. Data are expressed as percentages of virus bound to the cells relative to the amount of virus obtained from control cells. The arithmetic means and standard deviations from at least three (B and C) or four (A) independent experiments are shown. The values showing a statistically significant difference (P < 0.01 [**] or P < 0.05 [*]) are indicated.

FIG. 5.

RRV TLPs assembled in cells with reduced amounts of grp78 and calnexin have altered susceptibilities to EGTA and temperature treatments. (A) Purified TLPs were incubated at 37°C for 10 min with the indicated concentrations of EGTA. After a 10-fold dilution in MEM, the viruses were absorbed to cells for 1 h at 37°C. The infection was allowed to proceed for 14 h at 37°C and the infectious virus foci were detected by an immunoperoxidase assay as described in Materials and Methods. Data are expressed as percentages of the infectivity obtained with mock-treated virus. (B) Purified particles were incubated for the indicated times at 45°C. The virus was then absorbed to cells for 1 h at 37°C and the infection left to proceed as described above. Data are expressed as percentages of the virus infectivity obtained when the virus was preincubated for 15 min at 37°C as a control. The arithmetic means and standard deviations from three independent experiments performed in duplicate are shown. The values showing a statistically significant difference (P < 0.01 [**] or P < 0.05 [*]) are indicated. Irr, irrelevant; Cnx, calnexin.

FIG. 6.

Calnexin and grp78 participate in the folding of viral glycoproteins. Cells were transfected with the indicated siRNAs (Irr, irrelevant; 78, grp78; 94, grp94; Cnx, calnexin) and 72 h later were infected with RRV at an MOI of 3. At 7 hpi, the cells were starved for 1 h in methionine-free medium and then metabolically labeled with a pulse of 5 min of [³⁵S]methionine (150 μCi/ml). After the pulse, the cells were chased with Eagle's MEM supplemented with 1 mM cycloheximide and 10 mM methionine for the indicated times. At the end of chase, the cells were incubated for 2 min with ice-cold PBS containing 40 mM N-ethylmaleimide, and the cells were harvested in lysis buffer. Representative gels from at least four independent experiments are shown in panels A and B. (A) Cell lysates were immunoprecipitated (ipp) either with MAb M60 to VP7 or with polyclonal antibody C-239 to NSP4, and the samples were resolved by SDS-12.5% PAGE and detected by autoradiography. The immunoprecipitated proteins were detected by autoradiography as described in Materials and Methods. The dot in the ipp NSP4 blot at 0 min indicates the presence of a faster-migrating form of NSP4. The asterisk indicates a less defined form of NSP4. (B) Total cell lysates were mixed with sample buffer, boiled, resolved by SDS-10% PAGE, and detected by autoradiography. The lower panel shows the VP7 protein immunoprecipitated with MAb M60 from these samples, as described above. (C) Relative amounts of total and immunoprecipitated VP7, as determined by densitometry of the cells shown in the autoradiograph in panel B. The percentage of immunoprecipitated VP7 is normalized to the amount of total protein contained in each sample. The arithmetic means and standard deviations from four independent experiments are shown.

FIG. 7.

PDI participates in the folding of VP7. Cells were transfected with the indicated siRNAs (Irr, irrelevant; 57, ERp57; Crt, calreticulin) and 72 h later were infected with RRV at an MOI of 3. The cells were then metabolically labeled for 5 min and chased as described in the legend for Fig. 6. A representative gel of at least four independent experiments is shown. (A) Cell lysates were immunoprecipitated with MAb M60 to VP7, and the samples were resolved by SDS-12.5% PAGE and detected by autoradiography. (B) Relative amounts of immunoprecipitated VP7, as determined by densitometry of the cells shown in the autoradiograph in panel A. The percentage of immunoprecipitated VP7 in chaperone-deficient cells is normalized to the amount of VP7 immunoprecipitated from cells transfected with the irrelevant siRNA. The arithmetic means and standard deviations from four independent experiments are shown. The values showing a statistically significant difference (P < 0.05 [*]) are indicated.

FIG. 8.

The VP7 protein incorporated into TLPs produced in chaperone-silenced cells has an altered immunoreactivity. TLPs produced in cells transfected with the indicated siRNAs (Irr, irrelevant; Cnx, calnexin; Crt; calreticulin) were metabolically labeled and purified by CsCl density centrifugation as described in Materials and Methods. Purified TLPs, treated or not treated with 3 mM EGTA for 15 min at 37°C, were immunoprecipitated with MAb M60 to VP7. The immunoprecipitated proteins were resolved by SDS-11% PAGE and detected by autoradiography. Nonimmunoprecipitated total (Tot) TLPs, and TLPs immunoprecipitated (ipp) before (−EGTA) or after (+EGTA) treatment with EGTA are indicated. The amount of TLPs loaded in the gel after immunoprecipitation was four times larger than that of nonimmunoprecipitated TLPs. The relative amounts of VP7 and VP6 were determined by densitometry and normalized to the corresponding VP6 or VP7 protein present in total nonimmunoprecipitated TLPs (lanes labeled Tot). These results are shown in the two lowest panels. The arithmetic means and standard deviations from four independent experiments are shown. The values showing a statistically significant difference (P < 0.001 [**] or P < 0.01 [*]) are indicated.