Seneca Valley Virus Enters PK-15 Cells via Caveolae-Mediated Endocytosis and Macropinocytosis Dependent on Low-pH, Dynamin, Rab5, and Rab7

ABSTRACT

Seneca Valley virus (SVV), a new pathogen resulting in porcine vesicular disease, is prevalent in pig herds worldwide. Although an understanding of SVV biology pathogenesis is crucial for preventing and controlling this disease, the molecular mechanisms for the entry and post-internalization of SVV, which represent crucial steps in viral infection, are not well characterized. In this study, specific inhibitors, Western blotting, and immunofluorescence detection revealed that SVV entry into PK-15 cells depends on low-pH conditions and dynamin. Furthermore, results showed that caveolae-mediated endocytosis (CavME) contributes crucially to the internalization of SVV, as evidenced by cholesterol depletion, downregulation of caveolin-1 expression by small interfering RNA knockdown, and overexpression of a caveolin-1 dominant negative (caveolin-1-DN) in SVV-infected PK-15 cells. However, SVV entry into PK-15 cells did not depend on clathrin-mediated endocytosis (CME). Furthermore, treatment with specific inhibitors demonstrated that SVV entry into PK-15 cells via macropinocytosis depended on the Na+/H+ exchanger (NHE), p21-activated kinase 1 (Pak1), and actin rearrangement, but not phosphatidylinositol 3-kinase (PI3K). Electron microscopy showed that SVV particles or proteins were localized in CavME and macropinocytosis. Finally, knockdown of GTPase Rab5 and Rab7 by siRNA significantly inhibited SVV replication, as determined by measuring viral genome copy numbers, viral protein expression, and viral titers. In this study, our results demonstrated that SVV utilizes caveolae-mediated endocytosis and macropinocytosis to enter PK-15 cells, dependent on low pH, dynamin, Rab5, and Rab7.

IMPORTANCE Entry of virus into cells represents the initiation of a successful infection. As an emerging pathogen of porcine vesicular disease, clarification of the process of SVV entry into cells enables us to better understand the viral life cycle and pathogenesis. In this study, patterns of SVV internalization and key factors required were explored. We demonstrated for the first time that SVV entry into PK-15 cells via caveolae-mediated endocytosis and macropinocytosis requires Rab5 and Rab7 and is independent of clathrin-mediated endocytosis, and that low-pH conditions and dynamin are involved in the process of SVV internalization. This information increases our understanding of the patterns in which all members of the family Picornaviridae enter host cells, and provides new insights for preventing and controlling SVV infection.

INTRODUCTION

Seneca Valley virus (SVV), also referred to as Senecavirus A, is a non-enveloped, positive-sense, single-stranded RNA virus belonging to the Senecavirus genus in the family Picornaviridae (1). SVV-001, the first recognized SVV strain, which was identified and isolated as a PER. C6 cell culture contaminant in the United States in 2002, was initially used as a source of oncolytic virus to treat cancer patients (2, 3). In recent years, vesicular disease outbreaks caused by SVV have been confirmed in many countries (4–6). Presently, multiple SVV strains are prevalent in the pig herds in China. Moreover, the virulence of SVV seems to increase as it passages through pigs, and certain emerging SVV strains lead to acute death in newborn piglets (7, 8). It has been reported that newly-isolated SVV strains are different from other Chinese strains by genomic analysis, indicating that novel SVV strains have emerged in China. Studies have shown that SVV can infect and proliferate in a variety of cell lines, including PK-15 (9), BHK-21 (9), 293T (10), and SK6 cells (11). These studies indicate that SVV causes a broad spectrum of infections, which has further increased our interest in investigating this feature.

Viral entry into cells represents the initiation of a successful infection. After binding to cell surface receptors or adherence molecules, viruses are internalized through various endocytic pathways (12, 13). These endocytic patterns include clathrin-mediated endocytosis (CME), caveolae-mediated endocytosis (CavME), macropinocytosis, and clathrin- and caveolae-independent endocytosis (14). Among these endocytic patterns, CME is the major internalization route and is utilized by many viruses such as human immunodeficiency virus-1 (HIV-1) (15), vesicular stomatitis virus (VSV) (16), and African swine fever virus (ASFV) (17). With CME, clathrins are recruited and reorganized into a matrix to facilitate endocytosis of virus-receptor complexes with small vesicles and then delivered into endosomes (18). Caveolin-1, the key protein in caveolae, plays a crucial role in endosomal membrane trafficking and is involved in the CavME process (19). CavME is another critical endocytic pathway that includes the participation of lipid rafts, dynamin, and a complex signaling pathway activated by tyrosine kinases and phosphatases (13). Simian virus 40 (SV40) (20), respiratory syncytial virus (RSV) (21), and classical swine fever virus (CSFV) (22) have been reported to enter cells via this endocytic pathway. In addition to small endocytic vesicles mediated by clathrin or caveolin-1 entry into cells, macropinocytosis mainly recognizes large endocytic vesicles (23). Macropinosome morphogenesis occurs spontaneously or responds to receptor stimulation via the plasma membrane, and is not mediated by ligand distribution (23). Vaccinia and Ebola viruses enter host cells via macropinocytosis (24, 25), and certain bacterial pathogens also utilize this pathway for cell entry (26, 27).

Various endocytic pathways are involved in viral entry by the Picornaviridae family. Both coxsackievirus B3 (CBV3) and foot-and-mouth disease virus (FMDV) enter HeLa and IBRS-2 cells via the CME route, respectively (28, 29). Echovirus 1 (EV1) and coxsackievirus A9 internalization depends on dynamin. Interestingly, the same virus infects different cells via various internalization patterns. For example, poliovirus entry into brain microvascular endothelial cells is caveolae-dependent, whereas internalization into HeLa cells is clathrin- and caveolae-independent (30–32). However, as a member of the picornavirus family, the entry route and mechanism by which SVV enters PK-15 cells remain unknown.

In this study, we explored the internalization pattern exhibited by SVV and found that SVV entry into PK-15 cells was CavME- and macropinocytosis-dependent and CME-independent, requiring dynamin and a low pH in this process. Additionally, Rab5 and Rab7 facilitate the transport of SVV from early to late endosomes during viral infection. These results clarify the mechanism of endocytic pathways by SVV infection in cultured PK-15 cells and provide new targets for the prevention and control of SVV infection.

RESULTS

SVV entry into PK-15 cells is low-pH dependent.

The internalization of some viruses is associated with the acidic environment of endosomes (33, 34). To determine whether low pH plays a role in the internalization of SVV in PK-15 cells, we explored the effects of ammonium chloride (NH₄Cl) and chloroquine (CQ), 2 lysosome-associated acidification inhibitors, on SVV binding, entry, and replication in PK-15 cells. Cell viability was first detected using MTT assays in NH₄Cl- or CQ-treated PK-15 cells to analyze the cytotoxicity of inhibitors, and the cell viability was not affected at concentrations up to 1,500 μM NH₄Cl or 100 μM CQ (Fig. S1A and B). Subsequently, PK-15 cells pretreated with different concentrations of NH₄Cl for 1 h at 37°C were incubated with SVV for 1 h at 4°C (virus binding) to allow viral adsorption in the presence of inhibitors, followed by removing unbound viruses and further incubation for 1 h at 37°C (virus entry). Viral genome copy numbers were detected via reverse transcriptase quantitative PCR (RT-qPCR) at different stages, and the results showed that the differences in copy numbers occurred only in SVV entry (Fig. 1A) rather than SVV binding (Fig. S2A) in PK-15 cells treated with NH₄Cl. Viral replication was evaluated 6 h postinfection (hpi). As shown in Fig. 1B to D, viral genome copy numbers, viral VP1 protein expression, and viral titers were significantly decreased in PK-15 cells treated with increasing concentrations of NH₄Cl, indicating that NH₄Cl treatment affected viral infection. Additionally, the infection was further determined using indirect immunofluorescence assay (IFA). Similar to the previous results of virus replication, the number of SVV-infected cells decreased gradually with increasing inhibitor concentration (Fig. 1E). CQ, another acidification inhibitor, was used to assess the effects of low pH on viral binding, entry, and replication (34, 35). Consistent with the results of NH₄Cl treatment, CQ treatment inhibited viral entry and infection (Fig. 1F to J), as revealed by detecting viral copy numbers, VP1 expression, viral titers, and the number of SVV-infected cells. Taken together, these results indicated that a low-pH environment is necessary for SVV entry into PK-15 cells.

FIG 1.

SVV entry into PK-15 cells requires acidic endosomal pH. (A to B) NH₄Cl treatment inhibited SVV internalization. PK-15 cells pretreated with various concentrations of NH₄Cl for 1 h at 37°C were infected with SVV (MOI = 1) for 1 h at 4°C in the presence of inhibitors, followed by removal of unbound viruses, then incubated at 37°C for 1 h (virus entry) (A) and 6 h (virus replication) (B) and SVV-infected cells were processed to measure viral genome copy numbers by RT-qPCR. (C to E) NH₄Cl treatment reduced SVV replication. PK-15 cells were treated with different concentrations of NH₄Cl, as described in the corresponding panel B, VP1 expression, viral titers, and the number of SVV-infected cells were determined by Western blotting (C), viral titer assay (D), and IFA (E). (F to G) CQ treatment inhibited SVV internalization. PK-15 cells were treated with various concentrations of CQ and subsequently inoculated with SVV at the indicated time points, as described in A to B. SVV genome copy numbers for SVV entry (F) and SVV replication (G) were detected by RT-qPCR. (H to J) CQ treatment reduced SVV replication. PK-15 cells were treated with various concentrations of CQ as described in the corresponding panel B. VP1 expression, viral titers, and the numbers of SVV-infected cells were analyzed by Western blotting (H), viral titers assay (I), and IFA (J). Scale bars, 100 μm. Data are expressed by means ± standard deviations (SD) from 3 independent experiments (not significant [ns], P > 0.05; *, P < 0.05; **; P < 0.01; ***, P < 0.001; ****, P < 0.0001).

Dynamin is necessary for SVV internalization.

Dynamin, a large GTPase, participates in the scission of newly-formed vesicles from the cell membrane during endocytosis, and is involved in both CME and CavME (36). To investigate the potential role of dynamin in SVV entry and infection, dynasore, a dynamin inhibitor (37), was selected, and cells treated with different concentrations of dynasore were analyzed using a cell viability assay. No significant differences were observed from exposure to 10 μM to 50 μM dynasore (Fig. S1C). The uptake of fluorescently labeled transferrin (Tfn) was used to evaluate the effect of dynasore on the CME and CavME, which is a well characterized cargo and extensively used as a positive control in these 2 endocytic pathways (38). PK-15 cells pretreated with dynasore (50 μM) were incubated with Tfn and then fixed and observed by confocal fluorescence microscopy. The results of confocal fluorescence microscopy showed that the dynasore significantly reduced the red fluorescent signals of Tfn in PK-15 cells, indicating the inhibition of Tfn uptake by dynasore (Fig. 2A). PK-15 cells were pretreated with the different concentrations of dynasore for 1 h at 37°C, followed by SVV binding for 1 h at 4°C and SVV entry for 1 h at 37°C, and SVV replication for 6 h at 37°C. Finally, viral genome copy numbers were measured via RT-qPCR. As displayed in Fig. 2B and 2C, SVV entry and replication were inhibited by dynasore (Fig. 2B and C) but not by SVV binding (Fig. S2C). Moreover, VP1 expression, viral titers, and the number of SVV-infected PK-15 cells treated with dynasore were measured. The results showed that dynasore significantly reduced VP1 expression, viral titers, and the number of SVV-infected cells in a dose-dependent fashion (Fig. 2D to F), indicating that dynasore impeded SVV endocytosis. To confirm the effect of dynamin on the uptake of Tfn, plasmids of dynamin- dominant negative (DN) and dynamin-wild type (WT) were transfected into PK-15 cells to examine this uptake effect. The results of confocal fluorescence microscopy showed that the red fluorescent signals of Tfn were observed in PK-15 cells expressing dynamin-WT but no in PK-15 cells expressing dynamin-DN (Fig. 2G), indicated that dynamin-DN inhibited Tfn uptake. Subsequently, to further confirm the effect of dynamin on SVV internalization, plasmids of dynamin-DN and dynamin-WT were transfected into PK-15 cells to examine the effect of these plasmids on SVV entry and replication. As shown in Fig. 2H to K, dynamin-DN suppressed SVV entry and replication, VP1 expression, and viral titers in comparison with dynamin-WT. IFA results showed that VP1 proteins labeled with red fluorescent signals were observed in PK-15 cells expressing dynamin-WT, whereas VP1 signals were barely detected in PK-15 cells expressing dynamin-DN (Fig. 2L). Additionally, the decrease in dynamin expression by siRNA treatment was used to evaluate the effect of dynamin on SVV endocytosis. siRNA targeting dynamin significantly reduced SVV entry and infection, similar to the decrease in viral RNA copy numbers, VP1 expression, and viral titers (Fig. 2M to P). These results suggested that dynamin is involved in SVV entry into PK-15 cells.

FIG 2.

SVV entry into PK-15 cells depends on dynamin. (A) Dynasore inhibited Tfn uptake. PK-15 cells pretreated with 50 μM dynasore or DMSO for 1 h at 37°C were incubated with 10 μg/mL Tfn for 1 h at 4°C, and then shifted to 37°C for 1 h; after wash with PBS, the cells were fixed and stained with DAPI. (B to C) Dynasore treatment suppressed SVV internalization. PK-15 cells were treated with various concentrations of dynasore and infected with SVV at the indicated time points, as described in Fig. 1A and B. Viral genome copy numbers during SVV entry (B) and replication (C) were measured by RT-qPCR. (D to F) Dynasore treatment suppressed SVV replication. PK-15 cells were treated with various concentrations of dynasore, as described in Fig. 1B, followed by Western blotting (D), viral titers (E), and IFA (F) determinations. Scale bars, 100 μm. (G) Effect of dynamin on Tfn uptake. PK-15 cells transfected with dynamin-DN or dynamin-WT plasmids were incubated with 10 μg/mL Tfn for 1 h at 4°C, and then shifted to 37°C for 1 h; after wash with PBS, the cells were fixed and stained with DAPI. (H and L) The effect of dynamin on SVV internalization and replication. PK-15 cells were transfected with plasmids harboring dynamin-DN and dynamin-WT for 18 h, followed by infection with SVV, as described in Fig. 1A and B. SVV genome copy numbers during SVV entry (H), and SVV replication (I) were detected by RT-qPCR, and VP1 expression, viral titers, and the red fluorescence of internalized virus were detected by Western blotting (J), viral titers assays (K), and confocal fluorescence microscopy (L). Scale bars, 20 μm. (M to P) Dynamin knockdown inhibited SVV internalization and replication. PK-15 cells were transfected with siRNA targeting dynamin for 48 h, followed by infection with SVV, as described in Fig. 1A and B. SVV genome copy numbers during SVV entry (M) and SVV replication (N) were detected by RT-qPCR, and VP1 expression (O) and viral titers (P) were detected by Western blotting and viral titers assays, respectively. Data are expressed by means ± standard deviations (SD) from 3 independent experiments (*, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001).

CavME is involved in SVV entry into PK-15 cells.

The formation and stability of caveolae are dependent on caveolin-1, which can be disrupted by M-β-CD (39). SVV entry into PK-15 cells is dynamin-dependent (Fig. 2) and dynamin is an important regulator of CavME (36) prompting us to explore the role of CavME in SVV internalization. Suitable concentrations of M-β-CD were determined using a cell viability assay. The results showed that 3 concentrations (100, 500, and 1,000 μM) of M-β-CD did not affect cell viability (Fig. S1D), which were then used in subsequent experiments. Previous studies have confirmed that the cholera toxin subunit B (CTB) is internalized through the CavME and served as a positive marker for this endocytic pathway (40, 41). PK-15 cells pretreated with M-β-CD (1,000 μM) were incubated with CTB and then a CTB uptake assay was performed by confocal fluorescence microscopy. The results showed that the red fluorescent signals of CTB significantly reduced in M-β-CD-treated cells compared to the M-β-CD-untreated cells (Fig. 3A), indicating the inhibition of CTB uptake by M-β-CD. Subsequently, the effect of M-β-CD on SVV internalization was evaluated. PK-15 cells were pretreated with the different concentrations of M-β-CD for 1 h, followed by SVV incubation (viral binding) for 1 h at 4°C, SVV internalization (viral entry) for 1 h at 37°C, and SVV infection (viral replication) for 6 h at 37°C. Viral genome copy numbers were detected by RT-qPCR. As shown in Fig. S2D, 3B, and C, no differences in SVV binding were observed between the M-β-CD-treated and -untreated groups (Fig. S2D), while SVV entry and replication were disrupted by M-β-CD exposure (Fig. 3B and C). In addition to measuring the levels of SVV mRNA, we detected viral replication using Western blotting, viral titer assays, and IFA. The results showed that VP1 expression, SVV titers, and the amounts of fluorescent signals of the VP1 protein gradually decreased with an increase in M-β-CD concentration (Fig. 3D to F). Additionally, to confirm the effect of caveolin-1 on the uptake of CTB in CavME, the CTB uptake was observed as a positive control. The caveolin-1-DN plasmids were used to specifically disturb caveolae formation and stability, and plasmids of caveolin-1-DN and caveolin-1-WT were transfected into PK-15 cells to examine the effect of CTB uptake. As shown in Fig. 3G, the results of confocal fluorescence microscopy showed that the red fluorescent signals of CTB were observed in cells expressing caveolin-1-WT, whereas CTB signals were barely detected in PK-15 cells expressing caveolin-1-DN, indicating that the inhibition of CTB uptake by caveolin-1-DN. Next, to further determine whether caveolin-1 is essential for SVV internalization, plasmids of caveolin-1-DN and caveolin-1-WT were transfected into PK-15 cells to determine the role of caveolin-1 in SVV internalization. Compared with caveolin-1-WT plasmids transfected cells, caveolin-1-DN expression not only significantly affected viral genome copy numbers in SVV entry and replication (Fig. 3H and I), but also reduced VP1 expression and viral titers (Fig. 3J and K). Moreover, decreased expression of caveolin-1 by siRNA was used to further confirm the role of caveolin-1 in SVV internalization. The results showed that siRNA targeting caveolin-1 had a strong inhibitory effect on SVV mRNA, VP1 expression, and viral titers compared with the control siRNA (Fig. 3M to O). In IFA experiments, SVV successfully infected caveolin-1-WT-expressing cells, which exhibited a red fluorescent signal (Fig. 3L). Transmission electron microscopy (TEM) was exploited to observe the morphology of virus internalization (42, 43). As shown in Fig. 3P and Q, SVV infection induced caveolae-like vesicles in PK-15 cells (Fig. 3P), and virus particles or proteins were located in this vesicle by immunoelectron microscopy (IEM) detection (Fig. 3Q). Taken together, these results indicated that SVV internalization is caveolin-1-dependent in PK-15 cells.

FIG 3.

SVV entry into PK-15 cells depends on CavME. (A) M-β-CD inhibited CTB uptake. PK-15 cells pretreated with 1,000 μM M-β-CD or DMSO for 1 h at 37°C were incubated with 10 μg/mL CTB for 1 h at 4°C, and then shifted to 37°C for 1 h; after wash with PBS, the cells were fixed and stained with DAPI. (B to C) M-β-CD treatment suppressed SVV entry and replication but did not SVV binding. PK-15 cells were treated with various concentrations of M-β-CD then infected with SVV at the indicated time points, as described in Fig. 1A and B. The viral genome copy numbers during SVV entry (B) and replication (C) were measured by RT-qPCR. (D to F) M-β-CD treatment reduced SVV replication. PK-15 cells were treated with various concentrations of M-β-CD, as described in Fig. 1B, followed by Western blotting (D), viral titers (E), and IFA (F) determinations. Scale bars, 100 μm. (G) Effect of caveolin-1 on CTB uptake. PK-15 cells transfected with caveolin-1-DN or caveolin-1-WT plasmids were incubated with 10 μg/mL CTB for 1 h at 4°C, and then shifted to 37°C for 1 h; after wash with PBS, the cells were fixed and stained with DAPI. (H to L) The effect of caveolin-1 on SVV internalization and replication. PK-15 cells were transfected with plasmids harboring caveolin-1-DN and caveolin-1-WT for 18 h, followed by SVV infection, as described in Fig. 1A and B. SVV genome copy numbers during SVV entry (H), and SVV replication (I) were detected by RT-qPCR, and VP1 expression (J), viral titers (K), and internalized viruses (L) were detected by Western blotting, viral titers assays, and confocal fluorescence microscopy, respectively. Scale bars, 20 μm. (M to O) Caveolin-1 knockdown inhibited SVV replication. PK-15 cells were transfected with siRNA targeting cavrolin-1 for 48 h, followed by SVV infection, as described in Fig. 1B. SVV genome copy numbers during SVV replication were detected by RT-qPCR (M), and VP1 expression (N) and viral titers (O) were detected by Western blotting and viral titers assays, respectively. (P to Q) Ultrastructural analysis on CavME of SVV. PK-15 cells were infected with SVV for 30 min at 37°C,followed by fixing, processing, and incubation with primary and secondary antibodies conjugated to colloidal gold particles. Magnifications of dashed areas are shown, and the arrowheads indicate colloidal gold particles, representing SVV particles or proteins. Scale bars, 200 nm. Data are expressed by means ± standard deviations (SD) from 3 independent experiments (not significant [ns], P > 0.05; *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001).

SVV internalization into PK-15 cells is CME-independent.

CME is another viral endocytosis pathway regulated by dynamin (36). To determine whether CME is involved in SVV infection, chlorpromazine (CPZ), an inhibitor of the CME pathway (22), was used to evaluate SVV internalization in PK-15 cells. Nontoxic concentrations of CPZ for PK-15 cells were screened using a cell viability assay, and the cell activity was not influenced by CPZ treatment (up to 50 μM) (Fig. S1E). The assay of Tfn uptake was used to evaluate the effect of CPZ on CME. PK-15 cells pretreated with CPZ were incubated with Tfn, and then a Tfn uptake assay was performed by confocal fluorescence microscopy. The results showed that a significant reduction of the red fluorescence signals of Tfn was observed in CPZ (50 μM)-treated PK-15 cells (Fig. 4A), indicating the inhibition of Tfn uptake by CPZ. Subsquently, SVV entry and replication in CPZ-treated PK-15 cells were analyzed using RT-qPCR for SVV mRNA, Western blotting for VP1 expression, and IFA for the number of SVV-infected cells. As shown in Fig. S2E, 4B, and C, in addition to the upregulation of viral genome copy numbers at the SVV replication stage, CPZ treatment did not alter the level of SVV mRNA in terms of virus binding and entry compared with the untreated group. The data from Western blotting (Fig. 4D), IFA (Fig. 4E), and viral titers (Fig. 4F) also confirmed the above results, as evidenced by VP1 expression, the number of SVV-infected cells, and SVV titers. Epidermal growth factor receptor pathway substrate 15 (EPS15) is a major component of the clathrin adaptor complex and participates in the CME pathway (44). The effect of CME on SVV internalization was analyzed by overexpressing EPS15-WT or EPS15-DN. No significant differences in SVV genome copy numbers, VP1 expression, or viral titers were observed between EPS15-WT- and EPS15-DN-expressed PK-15 cells (Fig. 4G to J). Moreover, siRNA targeting the clathrin heavy chain (siCHC) was performed to evaluate the effect of CME on SVV infection. As shown in Fig. 4K to M, siCHC downregulated the expression of clathrin heavy chain, but did not affect SVV mRNA, VP1 expression, or viral titers. Taken together, CME-mediated endocytosis was not involved in SVV endocytosis.

FIG 4.

SVV entry into PK-15 cells is CME-independent. (A) CPZ inhibited Tfn uptake. PK-15 cells pretreated with 50 μM CPZ or DMSO for 1 h at 37°C were incubated with 10 μg/mL Tfn for 1 h at 4°C, and then shifted to 37°C for 1 h; after wash with PBS, the cells were fixed and stained with DAPI. (B to C) CPZ treatment had no effect on SVV entry and replication. PK-15 cells were treated with various concentrations of CPZ and infected with SVV at the indicated time points, as described in Fig. 1A and B. Viral genome copy numbers during SVV entry (B) and replication (C) were measured by RT-qPCR. (D to F) CPZ treatment did not reduce SVV replication. PK-15 cells were treated with various concentrations of CPZ, as described in Fig. 1B, followed by Western blotting (D), and IFA (E), viral titers (F). Scale bars, 100 μm. (G to J) The effect of ESP15 on SVV internalization and replication. PK-15 cells were transfected with plasmids harboring EPS15-DN and EPS-WT for 18 h, followed by SVV infection, as described in Fig. 1A and B. SVV genome copy numbers during SVV entry (G), and SVV replication (H) were detected by RT-qPCR, and VP1 expression (I) and viral titers (J) were analyzed by Western blotting and viral titers assays. (K to M) Clathrin heavy chain (CHC) knockdown did not affect SVV replication. PK-15 cells were transfected with siCHC for 48 h, followed by SVV infection, as described in Fig. 1B. SVV genome copy numbers during SVV replication were detected by RT-qPCR (K), and VP1 expression (L) and viral titers (M) were determined by Western blotting and viral titers assays, respectively. Data are expressed by means ± standard deviations (SD) from 3 independent experiments (not significant [ns], P > 0.05; *, P < 0.05).

Entry of SVV into PK-15 cells depends on macropinocytosis.

In addition to CavME and CME, macropinocytosis has also been reported to be a vital endocytic pathway for picornavirus entry and infection (32, 45). To determine the involvement of macropinocytosis in SVV internalization in PK-15 cells, 5-(N-ethyl-N-isopropyl) amiloride (EIPA), a macropinocytosis inhibitor of Na+/H+ exchange, was treated the cells. The results indicated that the addition of EIPA up to 40 μM had no effect on cell viability (Fig. S1F). Dextran, a specific fluid-phase marker of macropinocytosis (38), was used to evaluate the effect of EIPA on macropinocytosis in PK-15 cells. PK-15 cells pretreated with EIPA were incubated with Dextran, and then a Dextran uptake assay was performed by confocal fluorescence microscopy. The results showed that the red fluorescence signals of Dextran significantly reduced in EIPA (40 μM)-treated PK-15 cells (Fig. 5A), indicating the inhibition of Dextran uptake by EIPA. Subsequently, the effect of EIPA on SVV internalization was measured using RT-qPCR for SVV mRNA. The results showed that EIPA significantly affected SVV entry and infection (Fig. 5B and C) but did not inhibit SVV binding (Fig. S2F) according to measures of viral genome copies. The effect of EIPA on SVV replication was further analyzed using Western blotting, viral titers, and IFA. EIPA treatment strongly reduced VP1 expression, viral titers, and the number of SVV-infected cells (Fig. 5D to F). These results suggested that EIPA treatment inhibits SVV entry and infection. The p21-activated kinase 1 (Pak1), a serine/threonine kinase, is responsible for cytoskeletal dynamics and motility (46) and is necessary at all stages of macropinocytosis (33). To further determine SVV internalization depends on macropinocytosis, 1, 1′-dithiobis-2-naphthalenol (IPA-3) (47), a Pak1 inhibitor, was used to treat PK-15 cells. The results showed that 7.5 and 15 μM concentrations of IPA-3 did not affect cell viability (Fig. S1G), which were then used in subsequent experiments. SVV internalization and replication in IPA-3-treated PK-15 cells were measured using RT-qPCR for SVV mRNA. As shown in Fig. S2G, 5G, and H, SVV internalization and replication were significantly inhibited by IPA-3 (Fig. 5G and H) but not by SVV binding (Fig. S2G). Next, VP1 expression, viral titers, and the number of SVV-infected PK-15 cells treated with IPA-3 were measured. The results showed that IPA-3 significantly reduced VP1 expression, viral titers, and the number of SVV-infected cells in a dose-dependent fashion (Fig. 5I to K), indicating that IPA-3 impeded SVV endocytosis through affecting macropinocytosis. To extend the studies with chemical reagent (IPA-3), we further examined the effect of reducing the intracellular Pak1 expression on SVV replication, using siRNA targeting Pak1. PK-15 cells were transfected with 3 Pak1 siRNAs (siPak1-1, siPak1-2, and siPak1-3) to silence the Pak1 expression, showing that the PK-15 cells transfected with siPak1-1 exhibited obvious silence effect of Pak1 expression, which was then used in subsequent experiments (Fig. 5L). To further explore the effects of Pak1 reduction on SVV replication in PK-15 cells, the expression of Pak1 was knocked down by siPak1-1. As shown in Fig. 5M, the RT-qPCR results after SVV infection for 6 h showed that SVV replication in PK-15 cells was significantly inhibited by siPak1 compared with the control group (siCon), indicating that Pak1 play an important role in SVV endocytosis. Subsequently, the results of VP1 expression (Fig. 5N) and viral titers (Fig. 5O) also confirmed the above results, siPak1 had significant inhibitory effect on SVV replication compared with the control group. Taken together, these results indicated that SVV internalization in PK-15 cells depends on macropinocytosis.

FIG 5.

SVV entry into PK-15 cells depends on macropinocytosis. (A) EIPA inhibited Dextran uptake. PK-15 cells pretreated with 40 μM Dextran or DMSO for 1 h at 37°C were incubated with 10 μg/mL Dextran for 1 h at 4°C, and then shifted to 37°C for 1 h; after wash with PBS, the cells were fixed and stained with DAPI. (B to C) EIPA treatment suppressed SVV entry and replication but did not inhibit SVV binding. PK-15 cells were treated with various concentrations of EIPA, then infected with SVV at the indicated time points, as described in Fig. 1A and B. Viral genome copy numbers during SVV entry (B) and replication (C) were measured by RT-qPCR. (D to F) EIPA treatment inhibited SVV replication. PK-15 cells were treated with various concentrations of EIPA, as described in Fig. 1B, followed by Western blotting (D), viral titers (E), and IFA (F) determinations. Scale bars, 100 μm. (G to H) IPA-3 treatment suppressed SVV entry and replication but did not inhibit SVV binding. PK-15 cells were treated with various concentrations of IPA-3, then infected with SVV at the indicated time points, as described in Fig. 1A and B. Viral genome copy numbers during SVV entry (G) and replication (H) were measured by RT-qPCR. (I to K) IPA-3 treatment inhibited SVV replication. PK-15 cells were treated with various concentrations of IPA-3, as described in Fig. 1B, followed by Western blotting (I), viral titers (J), and IFA (K) determinations. (L) The effect of silencing Pak1 with siRNA targeting Pak1. PK-15 cells were transfected with siPak1 or siCon for 48 h and then detected by Western blotting with anti-Pak1 antibody. (M to O) Pak1 knockdown inhibited SVV replication. PK-15 cells were transfected with siPak1 for 48 h, followed by SVV infection, as described in Fig. 2N to 2P. SVV genome copy numbers during SVV replication were detected by RT-qPCR (M), and VP1 expression (N) and viral titers (O) were determined by Western blotting and viral titers assays, respectively. Data are expressed by means ± standard deviations (SD) from 3 independent experiments (not significant [ns], P > 0.05; *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001).

Entry of SVV into PK-15 cells requires actin cytoskeleton rearrangements but is independent on phosphatidylinositide 3-kinase.

Phosphatidylinositide 3-kinase (PI3K) is involved in macropinocytosis (48), prompting us to explore the effect of PI3K on SVV entry and infection. Wortmannin, a specific inhibitor of PI3K, was added to cells to inhibit macropinocytosis (33). Viability of PK-15 cells treated with various concentrations of wortmannin was measured by cell viability assays. 20 μM wortmannin showed no obvious change in terms of cell viability compared to untreated cells (Fig. S1H). SVV entry and infection in wortmannin-treated PK-15 cells were determined by RT-qPCR, which was different from EIPA treatment. SVV binding, entry, and replication were not reduced in wortmannin-treated PK-15 cells (Fig. S2H, 6A, and B). Subsequent detection of VP1 expression, viral titers, and the number of SVV-infected cells also confirmed that wortmannin did not inhibit SVV infection (Fig. 6C to E). These results indicated that the entry of SVV into PK-15 cells does not depend on PI3K. It has been reported that actin polymerization and cytoskeleton rearrangement are associated with macropinocytosis (49). To determine whether SVV internalization depends on actin, jasplakinolide (50), an inhibitor of actin rearrangement, was used in this experiment. Several concentrations (50–150 nM) of jasplakinolide had no obvious effect on cell viability (Fig. S1I) and were used in subsequent experiments. RT-qPCR results showed that SVV entry and infection were clearly inhibited and that SVV binding did not change with increases in jasplakinolide (Fig. S2I, 6F, and G). Consistent with the above RT-qPCR results, the decreases in VP1 expression and viral titers confirmed that SVV replication was inhibited by jasplakinolide treatment (Fig. 6H and I). Moreover, in IFA experiments, the number of green fluorescence signals corresponding to VP1 proteins decreased gradually in jasplakinolide-treated PK-15 cells (Fig. 6J). These findings suggested that rearrangement of the actin cytoskeleton is involved in SVV endocytosis and reconfirms that macropinocytosis is another endocytosis pathway for SVV internalization. Additionally, TEM results showed macropinocytosis was mediated by SVV infection of PK-15 cells (Fig. 6K), which contained virus particles or proteins using IEM observations (Fig. 6L).

FIG 6.

SVV entry into PK-15 cells depends on actin cytoskeleton rearrangement but not PI3K. (A to B) Wortmannin treatment did not inhibit SVV internalization. PK-15 cells were pretreated with various concentrations of wortmannin and subsequently inoculated with SVV at the indicated time points, as described in Fig. 1A and B. SVV genome copy numbers during SVV entry (A), and replication (B) were detected by RT-qPCR. (C to E) Wortmannin treatment did not affect SVV replication. Wortmannin treatment PK-15 cells were pretreated with various concentrations of wortmannin, as described in the corresponding panel B, and VP1 expression (C), viral titers (D), and the numbers of SVV-infected cells (E) were detected by Western blotting, viral titers assays, and IFA determinations. Scale bars, 100 μm. (F to G) Jasplakinolide treatment inhibited SVV entry and replication but did not impede SVV binding. PK-15 cells were pretreated with various concentrations of jasplakinolide and subsequently infected with SVV at the indicated time points, as described in Fig. 1A and B. SVV genome copy numbers during SVV entry (F) and replication (G) were detected by RT-qPCR. (H to J) Jasplakinolide treatment inhibited SVV replication. PK-15 cells were pretreated with various concentrations of jasplakinolide, as described in the corresponding panel B. VP1 expression (H), viral titers (I), and the numbers of SVV-infected cells (J) were detected by Western blotting, viral titers assays, and IFA determinations. Scale bars, 100 μm. (K to L) Ultrastructural analysis on macropinocytosis of SVV. PK-15 cells were processed as described in Fig. 3P and Q. Magnifications of dashed areas are shown, and arrowheads indicate colloidal gold particles, representing SVV particles or proteins. Scale bars, 500 or 200 nm. Data are expressed by means ± standard deviations (SD) from 3 independent experiments (not significant [ns], P > 0.05; *, P < 0.05; **, P < 0.01; ****, P < 0.0001).

Rab5- and Rab7-dependent intracellular trafficking are required by SVV infection.

Upon internalization, Rab proteins participate in membrane trafficking in eukaryotic cells by regulating effector proteins (51). Among these, Rab5 and Rab7 regulate early endosomes by fuzing the plasma membrane (52) and transferring early endosomes to late endosomes and lysosomes (53), respectively. To explore the effects of Rab5 and Rab7 on SVV entry and replication in PK-15 cells, the expression of Rab5 and Rab7 was knocked down by siRNA targeting Rab5 (siRab5) and Rab7 (siRab7). As shown in Fig. 7A and D, the RT-qPCR results after SVV infection for 6 h showed that SVV infection in PK-15 cells was significantly inhibited by siRab5 or siRab7 compared with the control (siCon), indicating that Rab5 and Rab7 play important roles in SVV endocytosis. Subsequently, the results of VP1 expression (Fig. 7B and E) and viral titers (Fig. 7C and F) also confirmed the above results, siRab5 or siRab7 had greater suppression in terms of SVV replication than the control group. Additionally, to rule out any effect of siRNA on cell viability, an MTT assay was performed, and the results showed no differences in cell viability between the siRab5, siRab7, and siCon groups (Fig. 7G). Collectively, these findings demonstrated that Rab5 and Rab7 play important roles in SVV endocytosis and subsequent intracellular trafficking to early and late endosomes.

FIG 7.

Rab5 and Rab7 are required for SVV endocytosis. (A to C) Rab5 knockdown inhibited SVV replication. PK-15 cells were transfected with siRNA targeting Rab5 for 48 h, followed by SVV infection, as described in Fig. 1B. SVV genome copy numbers during SVV replication were detected by RT-qPCR (A), and VP1 expression (B) and viral titers (C) were analyzed by Western blotting and viral titers assays, respectively. (D to F) Rab7 knockdown inhibited SVV replication. PK-15 cells were transfected with siRNA targeting Rab7 for 48 h, followed by SVV infection, as described in Fig. 1B. SVV genome copy numbers during SVV replication were detected by RT-qPCR (D), and VP1 expression (E) and viral titers (F) were determined by Western blotting and viral titers assays, respectively. (G) Cell viability after delivers siRab5 or siRab7 was evaluated by MTT assays. Data are expressed by means ± standard deviations (SD) from 3 independent experiments (not significant [ns], P > 0.05; *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001).

DISCUSSION

The mechanisms by which viruses enter host cells are unique and variable across various viral families (14). Among these entry mechanisms, endocytic pathways are the most important and widely reported pathways for viral entry into cells (13). The endocytosis of viruses is diverse and cell type-dependent, and viruses can utilize different endocytosis pathways to uptake virion components and deliver them into the cytoplasm, which successfully completes infection and replication (54). Clarifying how viruses enter cells is not only crucial for deepening the understanding of viral infection, but also helpful for development of antiviral drugs. In our study, we screened the endocytosis pathways for SVV entry into PK-15 cells and systematically analyzed each stage. Our results showed that SVV entry into PK-15 cells requires low-pH conditions and dynamin, is dependent on CavME and macropinocytosis, and is independent of CME. Moreover, actin cytoskeleton rearrangements, Rab5, and Rab7, are involved in SVV infection.

Some viruses enter cells via directly fuzing with plasma membranes and then releasing viral genomes into the cytoplasm, in which a change in the pH environment is not essential for this process, whereas a low-pH environment in endosomes is required for virus uptake by endocytic pathways (13). An increasing number of reports have shown that picornaviruses utilize endosomal pathways to uptake viral components and successfully infect host cells, such as the encephalomyocarditis virus (EMCV) (47), enterovirus 71 (EV71) (55), and porcine sapelovirus (PSV) (56). Furthermore, the low-pH environment of endosomes facilitates viral entry into the cytoplasm (14, 57). In this study, NH₄Cl and CQ, two inhibitors of endosomal acidification (34), were used to specifically change the pH environment of endosomes and lysosomes, which resulted in a significant inhibitory effect on SVV entry and replication (Fig. 1), indicating the importance of low-pH-dependent uptake in endosomes for the SVV life cycle.

Dynamin, a large GTPase, is essential for splitting endocytic membranes after vesicle formation (58) and is required for endocytic pathways (36). As a noncompetitive dynamin GTPase inhibitor, dynasore selectively inhibits dynamin activity and impedes endocytosis (59). Experimental results showed that dynasore treatment reduced SVV entry and replication, as revealed by measuring viral genome copy numbers, VP1 expression, viral titers, and the number of SVV-infected cells (Fig. 2B to F). The inhibitory effect of dynamin-DN plasmids or siRNA targeting dynamin further confirmed that dynamin contributes crucially to the endocytic process of SVV in PK-15 cells (Fig. 2G to P). These findings indicated that SVV internalization is dependent on dynamin.

Since GTPase dynamin is a crucial regulator of CavME and CME (60), the above results also indirectly suggested that CavME or CME may be involved in SVV endocytosis, which prompted us to explore the specific pathway involved in endocytosis. CavME is related to ligands, and endocytic vesicles are mainly composed of cholesterol, lipid rafts, dynamin II, tyrosine, and other kinases (61). Caveolae are subdomains of lipid rafts (61) and are directly involved in the internalization of several non-enveloped viruses (39), in which cholesterol participates in the formation of caveolae and is essential for lipid raft assembly. The sensitivity of CavME mainly originates from cholesterol depletion by reagents such as M-β-CD, which is a characteristic of CavME (13, 61). Therefore, M-β-CD treatment was used to disrupt CavME, which obviously inhibited SVV internalization and replication as measured by viral genome copy numbers, viral protein expression, viral titers, and the numbers of SVV-infected cells (Fig. 3B to F). These results were consistent with a previous report in which SVV infection was dependent on CavME in PAM-Tang cells (62). Caveolin-1 plays a crucial role in the stabilization and formation of caveolae and serves as a scaffolding protein responsible for key signaling molecules in CavME (63). Thus, transfection of caveolin-1-DN plasmids or sicaveolin-1 disrupted CavME and further confirmed the inhibitory effect of these 2 factors on SVV internalization (Fig. 3G to O). First, the transfection of caveolin-1-DN plasmids or sicaveolin-1 significantly reduced the internalization of SVV compared to the control group, as analyzed by RT-qPCR, Western blotting, and viral titer assays. Second, the number of SVV-positive cells was increased in PK-15 cells expressing caveolin-1-WT, and SVV localized in PK-15 cells expressing caveolin-1-WT. Third, caveolae-like vesicles were induced by SVV infection and contained viral proteins or particles identified by TEM or IEM detection. These data suggested that SVV internalization depends on the CavME pathway, and that cholesterol plays a crucial role in this endocytic process. In addition, the effect of the CME pathway, another endocytic pathway, on SVV infection was studied. In response to receptor-related endocytic signals, several components of the plasma membrane are transported by the CME pathway, including clathrin light and clathrin heavy chains (64). In the CME pathway, GTPase dynamin promotes vesicle scission at the necks of clathrin-coated pits (65). Our results showed that CPZ, a specific inhibitor of the CME pathway (13, 66), did not affect SVV entry and replication, as measured by SVV genome copy numbers, viral protein expression, the numbers of SVV-infected cells, or viral titers (Fig. 4B to F). Furthermore, no differences in SVV entry and replication were observed between EPS15-DN- and EPS15-WT- expressing PK-15 cells (Fig. 4G to J), and a decrease in clathrin expression by siRNA targeting clathrin heavy chain did not affect SVV infection (Fig. 4K to M). These data indicated that SVV does not hijack the CME pathway to enter PK-15 cells.

Macropinocytosis is also a crucial endocytosis pathway that accompanies cell membrane ruffling (67), is independent of receptors (68), and is sensitive to the Na+/H+ exchanger (34) involved in viral entry and replication. In this study, the genome copy numbers, VP1 expression, viral titers, and cell numbers in SVV infection were obviously reduced by treatment with EIPA, an inhibitor of the Na+/H+ exchanger (Fig. 5B to F). Pak1, a serine/threonine kinase, is responsible for cytoskeletal dynamics and motility (46) and is required at all stages of macropinocytosis (33). Therefore, the effect of Pak1 on SVV internalization and replication was evaluated. IPA-3, a Pak1 inhibitor, or specific siRNA targeting Pak1 exhibited a significant inhibitory role in the endocytic process of SVV in PK-15 cells (Fig. 5G to O). These findings indicated that SVV internalization depends on macropinocytosis. Moreover, macropinocytosis was induced by SVV infection, and some virus particles or proteins were localized in the macropinosome (Fig. 6K and L). Thus, our results demonstrated that SVV entry into PK-15 cells depended on CavME and macropinocytosis but not CME. However, dependence on macropinocytosis is different between PK-15 and PAM-Tang cells, and CME and macropinocytosis were not involved in the entry of SVV into PAM-Tang cells, and only CavME played an internalizing role (62). These results indicated that the same virus infects diverse cell types through different endocytic pathways. PI3K is essential for ruffle formation and macropinosome fusion (33), which prompted us to explore the effects of PI3K on SVV internalization by using a specific inhibitor (wortmannin). These results showed that wortmannin treatment had no obvious effect on SVV entry (Fig. 6A–E). A similar result was observed for the internalization of newcastle disease virus (NDV) in DF-1 cells and porcine deltacoronavirus (PDCoV) in IPI-2I cells (34, 69). Furthermore, actin- and microfilament-induced cell membrane ruffling and blebbing activation are related to macropinocytosis (33). After disruption of the dynamics of actin filaments with jasplakinolide, PK-15 cells significantly affected SVV entry and replication, but not binding (Fig. 6). Jasplakinolide treatment affects the recruitment of host proteins and lipids for virus replication by disrupting actin filament function (70). These results indicated that the internalization of SVV in PK-15 cells requires the rearrangement of the actin cytoskeleton but does not depend on PI3K. However, the specific mechanism involving interactions between SVV entry and actin filaments needs to be explored.

Rab GTPases are hijacked by various viruses to participate in endocytic trafficking (13, 53, 71). The GTPases Rab5 and Rab7 are 2 crucial regulators for the transportation of early and late endosomes, respectively, and are required for virus trafficking (22, 72). In this study, the downregulation of Rab5 and Rab7 expression with siRNA targeting the corresponding mRNAs was further evaluated for the effects of Rab5 and Rab7 on SVV internalization in PK-15 cells. The results indicated that SVV entry was dependent on Rab5 and Rab7 in PK-15 cells (Fig. 7), which is similar to the process of PDCoV entry (34). This process of SVV internalization indicated that internalized SVV are fused with Rab5-mediated early endosomes and guide Rab7-mediated late endosomes, which can effectively avoid the degradation of viral components by cytoplasmic enzymes. However, the specific mechanism requires further investigation.

In conclusion, our results showed, for the first time, that internalization of SVV in cultured PK-15 cells is dependent on low pH, dynamin, CavME, and macropinocytosis, and SVV is finally transported to lysosomes through Rab5-dependent early endosomes and Rab7-dependent late endosomes before the release of viral genomes. Clarification of the SVV life cycle will provide important insights for the development of antiviral drugs.

MATERIALS AND METHODS

Cells, viruses, and antibodies.

PK-15 cells were originally obtained from the American Type Culture Collection (ATCC) and cultured in Dulbecco’s modified Eagle’s medium (DMEM) (Gibco) containing 5% fetal bovine serum (FBS) (Gibco, Life Technologies), streptomycin, and penicillin at 37°C in a 5% CO₂ incubator. SVV CHhb17 strain preserved in our laboratory was used in this study (73). An anti-SVV VP1 monoclonal antibody was obtained from our laboratory and other antibodies were purchased commercially: rabbit anti-GFP antibody (SY0243; Huabio), mouse anti-β-actin antibody (A5441; Sigma), rabbit anti-dynamin antibody (2342S; CST), rabbit anti-caveolin-1 antibody (A19006; ABclonal), rabbit anti-clathrin heavy chain antibody (A12423; ABclonal), rabbit anti-Pak1 antibody (A0809; ABclonal), rabbit anti-Rab5a antibody (A12304; ABclonal), rabbit anti-Rab7 antibody (A12308; ABclonal), horseradish peroxidase (HRP)-conjugated anti-rabbit and -mouse secondary antibodies (A0545 and A9044; Sigma), tetramethylrhodamine isothiocyanate (TRITC)-conjugated rabbit anti-mouse (T2402; Sigma), and fluorescein isothiocyanate (FITC)-conjugated rabbit anti-mouse (F9137; Sigma).

Chemical reagents.

NH₄Cl (A9434; Sigma), CQ (S6999; Selleck), dynasore (ab120192; Abcam), M-β-CD (C4767; Sigma), CPZ (C0982; Sigma), EIPA (sc-202458; Santa Cruz), IPA-3 (HY-15663; MCE), wortmannin (S2758; Selleck), jasplakinolide (B7189; Apexbio), and 4’,6-diamidino-2-phenylindole (DAPI) (D21490; Invitrogen) were obtained from Sigma-Aldrich. Chemical reagents (chemical inhibitors) dissolved in dimethyl sulfoxide (DMSO) or ultrapure water in accordance with the manufacturer’s recommendations were used in this study.

Viral infection and TCID₅₀ assay.

For SVV binding, entry, and replication in chemical reagent-treated cells, PK-15 cells cultured to a confluence of approximately 90% were pretreated with various concentrations of chemical reagents (NH₄Cl, CQ, dynasore, M-β-CD, CPZ, EIPA, wortmannin, and jasplakinolide) for 1 h at 37°C, then infected with SVV (multiplicity of infection (MOI) = 1) for 1 h at 4°C (SVV binding) in the presence of inhibitors, followed by removing unbound viruses and incubation for 1 h at 37°C (SVV entry) or for 6 h at 37°C (SVV replication). Cells treated with chemical reagents were processed according to different experiments. The whole cell culture medium was collected at the indicated times in SVV-infected cells and was assayed for viral titers by serial dilution using Spearman and Karber’s method and represented as 50% tissue culture infectious dose (TCID₅₀).

RNA extraction and RT-qPCR.

Total RNA was extracted from PK-15 cells using TRIzol reagent (15596018; Invitrogen), according to the manufacturer's protocol, and cDNA synthesis was conducted using the Vazyme cDNA Synthesis Kit (R323-01; Vazyme) with the corresponding primers. cDNA was detected at least three times using the Taq Pro Universal SYBR qPCR Master Mix Kit (Q712-02; Vazyme) (Roche LightCycler real-time PCR detection system). The 2^−ΔΔCT method was utilized to evaluate the relative expression accumulations of the mRNAs of different genes. The primer sequences (5′–3′) for SVV-3D and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) in RT-qPCR are as follows: qSVV-3D-F (CCAACAAGGGTTCCGTCTTC), qSVV-3D-R (TTGGACGAATTTGCGTTTTAGA), qGAPDH-F (TCGGAGTGAACGGATTTGGC), and qGAPDH-R (TGACAAGCTTCCCGTTCTCC).

For the binding assays, PK-15 cells pretreated with the indicated concentration of inhibitors for 1 h at 37°C were washed three times with cold phosphate-buffered saline (PBS), and then were inoculated with SVV in the presence of the inhibitors at 4°C for 1 h. After the incubation period, the inoculum was removed, and the cells were washed three times with cold PBS and then harvested by three freeze-thaw cycles and subjected to viral RNA extraction and quantification of viral load via real-time RT-PCR, as described by other viruses (34, 41). For the entry assays, after the binding period, cells treated with inhibitors or transfected with siRNAs or various plasmids were incubated with medium for 1 h at 37°C and washed with cold PBS, and then harvested and subjected to viral RNA extraction and quantification of viral load via real-time PCR (RT-PCR). For the replication experiments, after the entry period, the cells were washed with PBS, followed by incubation in medium containing 2% FBS at 37°C. At 6 hpi, the cells were washed with cold PBS and then harvested and subjected to viral RNA extraction and quantification of viral load via RT-PCR.

Cell viability assay.

The effect of various inhibitors on cell viability was detected using the MTT Cell Proliferation and Cytotoxicity assay kit (C0009M; Beyotime). Briefly, PK-15 cells were pretreated with various concentrations of inhibitors, followed by processing with the corresponding reagents and measuring cell viability at the indicated time points in accordance with the manufacturer's protocols.

Plasmid and siRNA transfections.

PK-15 cells cultured to approximately 70% to 80% monolayers were transfected with plasmids carrying dynamin-DN, dynamin-WT, caveolin-1-DN, caveolin-1-WT, EPS15-DN, or EPS15-WT using Lipofectamine 2000 (11668019, Invitrogen) in accordance with the manufacturer's protocol for 18 h, followed by infection with SVV and analysis at the indicated time points. siRNAs targeting the dynamin gene (sidynamin, sc-43736; Santa Cruz), caveolin-1 gene (sicaveolin-1, sc-29241; Santa Cruz), clathrin heavy chain (siCHC, sc-35067; Santa Cruz), Rab5 (siRab5, sc-36344; Santa Cruz), Rab7 (siRab7, sc-29460; Santa Cruz), or Pak1 (sense, 5′-GCUUGCUUCAGACGUCAAATT-3′; antisense, 5′-UUUGACGUCUGAAGCAAGCTT-3′) were designed by GenePharma, and these siRNAs were transfected using Lipofectamine RNAiMAX (13778; Invitrogen) according to the manufacturer’s protocol for 36 to 48 h.

Western blotting.

Proteins were extracted from PK-15 cells using RIPA lysis buffer (P0013B; Beyotime) and quantified using a bicinchoninic acid protein assay kit (23225; Thermo). Twenty micrograms of total proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to nitrocellulose membranes (66485; Pall). The membranes were blocked in PBS with 5% nonfat milk for 2 h at room temperature, incubated with primary antibody or HRP-conjugated secondary antibody for 2 h, and then exposed using SuperSignal West Pico PLUS Chemiluminescent Substrate Kit (34580; Thermo) in a chemiluminescence apparatus (ProteinSimple).

Indirect immunofluorescence and confocal microscopy.

PK-15 cells grown to a confluence of approximately 80% to 90% were pretreated with various inhibitors or transfected with the indicated plasmids for 18 h in 24-well culture plates, followed by SVV infection for 6 h, as described for viral infection or incubation. The cells were fixed with 4% paraformaldehyde (16005; Sigma), permeabilized with 0.1% Triton X-100 (T8787; Sigma), and blocked with 5% bovine serum albumin, followed by incubation with anti-VP1 monoclonal antibody, secondary FITC- or TRITC-conjugated anti-mouse antibodies, and DAPI. Fluorescent images were obtained using an immunofluorescence microscope (Olympus IX73) or a confocal immunofluorescence microscope (Leica TCS SP8 STED).

Uptake assay.

Alexa Fluor 568-labeled human transferrin (Tfn) (T23365, Thermo), Alexa Fluor 594-labeled cholera toxin subunit B (CTB) (C34777, Thermo), and Alexa Fluor 594-labeled dextran (D22913, Thermo) were used for the uptake assays, as previously described (22, 41, 74). PK-15 cells were pretreated with the indicated inhibitors for 1 h at 37°C or transfected with various plasmids (WT or DN plasmids) for 18 h. These cells were incubated with 10 μg/mL of Tfn, CTB, or dextran for 1 h at 4°C and then washed and incubated at 37°C for 1 h. After a wash with PBS, the cells were fixed with 4% paraformaldehyde, stained with DAPI, and then observed using a confocal immunofluorescence microscope (Leica TCS SP8 STED).

Transmission electron microscopy.

PK-15 cells were incubated with SVV for 1 h at 4°C, and then transferred to 37°C. After incubation for 30 min, the cells were fixed with glutaraldehyde at room temperature and processed for ultrathin cryosectioning. Ultrathin sections of cells were examined using a Hitachi H-7500 transmission electron microscope (Hitachi Ltd.). The localization of SVV particles or proteins has previously been observed using immunoelectron microscopy (IEM) (43). Ultrathin cryosections were incubated with anti-VP1 antibody (primary antibody) and secondary antibody conjugated to colloidal gold particles, followed by processing with a silver enhancement kit. Images were acquired using a Hitachi H-7500 transmission electron microscope (Hitachi Ltd.).

Statistical analysis.

Statistical differences for all data were determined by one-way analysis of variance (ANOVA) or Student's t test using GraphPad Prism 8.0 software (GraphPad Software). P value < 0.05 was considered to be statistically significant.