SQSTM1 downregulates avian metapneumovirus subgroup C replication via mediating selective autophagic degradation of viral M2-2 protein

ABSTRACT

Avian metapneumovirus subgroup C (aMPV/C), an important pathogen causing acute respiratory infection in chickens and turkeys, contributes to substantial economic losses in the poultry industry worldwide. aMPV/C has been reported to induce autophagy, which is beneficial to virus replication. Sequestosome 1 (SQSTM1/P62), a selective autophagic receptor, plays a crucial role in viral replication by clearing ubiquitinated proteins. However, the relationship between SQSTM1-mediated selective autophagy and aMPV/C replication is unclear. In this study, we found that the expression of SQSTM1 negatively regulates aMPV/C replication by reducing viral protein expression and viral titers. Further studies revealed that the interaction between SQSTM1 and aMPV/C M2-2 protein is mediated via the Phox and Bem1 (PB1) domain of the former, which recognizes a ubiquitinated lysine at position 67 of the M2-2 protein, and finally degrades M2-2 via SQSTM1-mediated selective autophagy. Collectively, our results reveal that SQSTM1 degrades M2-2 via a process of selective autophagy to suppress aMPV/C replication, thereby providing novel insights for the prevention and control of aMPV/C infection.

IMPORTANCE

The selective autophagy plays an important role in virus replication. As an emerging pathogen of avian respiratory virus, clarification of the effect of SQSTM1, a selective autophagic receptor, on aMPV/C replication in host cells enables us to better understand the viral pathogenesis. Previous study showed that aMPV/C infection reduced the SQSTM1 expression accompanied by virus proliferation, but the specific regulatory mechanism between them was still unclear. In this study, we demonstrated for the first time that SQSTM1 recognizes the 67th amino acid of M2-2 protein by the interaction between them, followed by M2-2 degradation via the SQSTM1-mediated selective autophagy, and finally inhibits aMPV/C replication. This information supplies the mechanism by which SQSTM1 negatively regulates viral replication, and provides new insights for preventing and controlling aMPV/C infection.

INTRODUCTION

Avian metapneumovirus (aMPV), an emerging pathogen, causes a reduction in egg production and acute upper respiratory tract infection in poultry, and accordingly represents a significant threat to the global poultry industry (1, 2). On the basis of genetic differences and antigenic divergence, aMPV can be divided into four subgroups (A, B, C, and D) (1, 3), among which, subgroup aMPV/C is considered to be relatively close to human metapneumovirus (hMPV) in terms of genetics and antigenicity (1, 4–6). aMPVs are single-stranded negative RNA viruses within the family Paramyxoviridae (6, 7), which comprise the following eight types of proteins: nucleocapsid protein (N), phosphoprotein (P), matrix protein (M), fusion protein (F), second matrix protein (M2), small hydrophobic protein (SH), surface glycoprotein (G), and RNA-dependent RNA polymerase (L) (1, 8). Interestingly, the M2 gene of aMPV/C further encodes two putative proteins, M2-1 and M2-2, on the basis of two overlapping open reading frames (9, 10). The M2-1 protein is responsible for mRNA elongation, whereas the M2-2 protein is involved in viral replication, pathogenicity, and host immunogenicity (9, 11, 12).

Autophagy, a highly conserved physiological process in cells, is induced by diverse stresses associated with changes in energy and oxygen levels, drugs, and protein accumulation (13, 14). Autophagy is generally a non-selective process that involves the essential indiscriminate engulfment of cytosolic contents (15). However, autophagy mediated by different cargo receptors selectively promotes the degradation of aggregated proteins or damaged organelles and finally maintains an appropriate level of cellular quality (16–18). The selective receptors involved in autophagy include SQSTM1, optineurin (OPTN), nuclear dot protein 52 (NDP52/CALCOO52), and neighbor to BRCA1 (NBR1), which simultaneously combine with substrate proteins and microtubule-associated protein 1 light chain 3 (LC3) prior to subsequent lysosome-mediated degradation (16). Among these receptors, SQSTM1 is a classical selective autophagic receptor containing PB1, ZZ-type zinc finger motif (ZZ), and ubiquitin-binding domain (UBA), along with an LC3-interacting region (LIR) (19). The interaction between the LIR of SQSTM1 and LC3 induces the recruitment and stretching of an autophagy membrane, followed by encapsulation of the substrate–SQSTM1 complex within double-membrane-bound vesicles for subsequent degradation (20). Importantly, the ubiquitination of substrate proteins has been proven to be a marker of selective autophagy recognition and degradation, thereby indicating that a ubiquitin-dependent sensor system is responsible for determining substrate specificity (18, 21–23). Previous studies have demonstrated that SQSTM1 inhibits the replication of chikungunya virus and coxsackievirus B3 by targeting the ubiquitinated capsid proteins (24, 25). In addition, more specific interactions have also been reported. For example, SQSTM1 has been demonstrated to specifically interact with non-ubiquitinated proteins such as Sindbis virus capsid protein, followed by its subsequent delivery to autolysosomes for degradation (26). However, the regulatory mechanisms underlying the interactions between SQSTM1 and aMPV/C or viral proteins have yet to be sufficiently determined.

In this study, we investigated the mechanisms underlying the SQSTM1-mediated regulation of aMPV/C replication, and accordingly found that SQSTM1 inhibits aMPV/C replication via selective autophagy-associated aMPV/C M2-2 degradation. Collectively, our results reveal a novel mechanism of selective degradation of aMPV/C M2-2 by SQSTM1, which could provide an important insight for the prevention and control of aMPV infection.

RESULTS

SQSTM1 negatively regulates aMPV/C replication by autophagy

SQSTM1 has been reported to be involved in the replication of numerous viruses (27), and our previous study has confirmed that aMPV/C infection reduces the expression of SQSTM1 in Vero or DF-1 cells (28). To clarify the relationship between aMPV/C replication and autophagy, Vero cells treated or untreated with siRNA targeting ATG7 (siATG7 or siCon) were infected with aMPV/C for 0, 48, or 72 h and then determined by Western blotting and a 50% tissue culture infection dose (TCID₅₀). These results showed that aMPV/C infection induced autophagy, accompanied with LC3-II increase and SQSTM1 reduction, whereas the knockdown of ATG7 attenuated above changes and inhibited aMPV/C replication (Fig. 1A and B). These data are similar to our previous results (28). Moreover, aMPV/C infection-mediated SQSTM1 reduction motivated us to explore the effects of SQSTM1 on aMPV/C replication. Vero cells transfected with FLAG-SQSTM1 plasmids were infected with aMPV/C for 48 or 72 h, and the expression levels of aMPV/C N protein and viral titers at different time points were analyzed via Western blotting and a TCID₅₀ assay, respectively. Compared to the control group, the overexpression of SQSTM1 caused a significant reduction in the expression of aMPV/C N protein (Fig. 1C) and viral titers (Fig. 1D). Vero cells transfected with siSQSTM1 or siCon were infected with aMPV/C for 48 or 72 h, SQSTM1 silencing contributed to a clear increase in the expression of aMPV/C N protein (Fig. 1E) and aMPV/C titers (Fig. 1F). Furthermore, Vero cells transfected with siSQSTM1 or siCon were transfected with pFLAG or pFLAG-SQSTM1 plasmids and then infected with aMPV/C for 72 h. The results showed that SQSTM1 indeed inhibited the expression of aMPV/C N protein (Fig. 1G). We also explored the effects of SQSTM1 expression on aMPV/C replication in the presence of siATG7. Vero cells transfected with siATG7 or siCon were transfected with pFLAG-N or pFLAG-SQSTM1 plasmids and then infected with aMPV/C for 72 h. As shown in Figure 1H and I, the knockdown of ATG7 attenuated the inhibitory effect of SQSTM1 on aMPV/C replication compared to the control group. Collectively, these results indicated that the autophagy was involved in the inhibitory role of SQSTM1 in aMPV/C replication.

Fig 1.

SQSTM1 inhibits aMPV/C replication. (A and B) Vero cells were transfected with 40 pM siATG7 or siCon for 24 h, followed by infection with aMPV/C for 0, 48, or 72 h, and then detected for viral N protein, ATG7, SQSTM1, LC3, and ACTB using Western blotting (A) or aMPV/C titers using TCID₅₀ (B). (C and D) Vero cells were transfected with the plasmids pFLAG-SQSTM1 or pFLAG for 24 h, followed by infection with aMPV/C (MOI = 0.1), and cells or whole-cell culture media were collected at 48 or 72 hours postinfection (hpi) and were analyzed for viral N protein, FLAG, and ACTB using Western blotting (C) or aMPV/C titers using TCID₅₀ (D). (E and F) Vero cells were transfected with 40 pM siSQSTM1 or siCon for 36 h, followed by infection with aMPV/C, and cells or whole-cell culture media were collected at 48 or 72 hpi and were analyzed for viral N protein, SQSTM1, and ACTB using Western blotting (E) or aMPV/C titers using TCID₅₀ (F). (G) Vero cells treated with siSQSTM1 or siCon were transfected with the plasmids pFLAG-SQSTM1 or pFLAG for 24 h, followed by infection with aMPV/C, and the proteins were collected and analyzed for viral N protein, SQSTM1, and ACTB. (H and I) Vero cells transfected with 40 pM siATG7 or siCon were transfected with the plasmids pFLAG-SQSTM1 or pFLAG for 12 h, followed by infection with aMPV/C, and the proteins or whole-cell culture supernatants were collected at 72 hpi and were analyzed for viral N protein, ATG7, FLAG, and ACTB using Western blotting (H) or aMPV/C titers using TCID₅₀ (I). Data are shown as the means ± standard deviations (SD) of three independent experiments (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

aMPV/C M2-2 interacts with SQSTM1 and induces autophagy

SQSTM1 mediates the degradation of viral proteins via autophagy process that eventually leads to an inhibition of viral replication (24, 29, 30). It is assumed that it is the interaction between SQSTM1 and viral proteins that underlies the degradation of the viral protein. To identify those aMPV/C proteins that interact with SQSTM1, we examined the colocalization of viral proteins and SQSTM1 using confocal imaging. As shown in Figure 2A, SQSTM1 localized significantly to aMPV/C N, M2-2, or the third or fourth domains of L (L3 or L4) (31) (Fig. 2A). M2-2 plays an important role in aMPV/C replication and immunogenicity (9), thus, we decided to focus on M2-2 in further analyses. To confirm the interaction between SQSTM1 and M2-2, we cotransfected HEK-293T cells with the expression plasmids pFLAG-SQSTM1 and pEGFP-M2-2 or pFLAG-SQSTM1 and pEGFP-C1 (control plasmids), and performed a coimmunoprecipitation (co-IP) assays using anti-GFP and anti-FLAG antibodies. A specific band appeared between FLAG-SQSTM1 and GFP-M2-2, whereas no interaction was observed between FLAG-SQSTM1 and GFP (Fig. 2B). The specific interaction between SQSTM1 and M2-2 was subsequently confirmed by a reverse co-IP detection (Fig. 2C). In pEGFP-M2-2 alone-transfected HEK-293T cells, M2-2 was co-precipitated with endogenous SQSTM1 (Fig. 2D). These results confirmed the interaction between SQSTM1 and aMPV/C M2-2.

Fig 2.

SQSTM1 interacts with M2-2. (A) Vero cells were transfected with plasmids encoding different genes of aMPV/C (GFP tag) and mCherry-SQSTM1 for 24 h, respectively. The cells were fixed and stained with DAPI (blue), and the colocalization between viral proteins (green) and mCherry-SQSTM1 (red) was examined by confocal fluorescence microscopy. (B) HEK-293T cells were cotransfected with pFLAG-SQSTM1 and pEGFP-M2-2 or pEGFP-C1 plasmids for 24 h. The cells were processed, followed by immunoprecipitation with anti-GFP, and analyzed for FLAG, GFP, and ACTB. (C) HEK-293T cells transfected with plasmids encoding GFP-M2-2 and FLAG-SQSTM1 or FLAG for 24 h were processed and immunoprecipitated with anti-FLAG, and analyzed for FLAG, GFP, and ACTB. (D) HEK-293T cells transfected with plasmids encoding GFP-M2-2 or GFP for 24 h were processed and immunoprecipitated with anti-GFP, and analyzed for SQSTM1, GFP, and ACTB.

Autophagy plays an important role in M2-2 degradation

The SQSTM1 inhibits viral replication by degrading viral proteins via selective autophagy (24, 29, 30). We continued to explore whether the autophagy was involved in aMPV/C M2-2 degradation. Rapamycin (Rap), an autophagy inducer (29), and chloroquine (CQ), an inhibitor that increases the pH in lysosomes to prevent autophagosome–lysosome fusion (28), were used to regulate autophagy for evaluating M2-2 expression. Rap treatment induced degradation of aMPV/C M2-2 accompanied with the increase of LC3-II, whereas CQ treatment restored the Rap-mediated aMPV/C M2-2 reduction (Fig. 3A), indicating that the autophagy was responsible for M2-2 degradation. ATG5 and ATG7 proteins, two autophagy-related proteins, play important roles in autophagy process, and their deficiency seriously affects the activation of autophagy (32, 33). To exclude the likelihood of a non-specific effect of pharmacological treatment, we further evaluated the effects of knockdown of ATG5 or ATG7 on aMPV/C M2-2 degradation. As shown in Figure 3B through E, silencing ATG5 or ATG7 restored aMPV/C M2-2 reduction by Rap, which reconfirmed that the autophagic process was involved in the degradation of aMPV/C M2-2. We also investigated whether the interaction between M2-2 protein and SQSTM1 was affected by Rap in the co-IP assay, and found that Rap treatment promoted the interaction between endogenous SQSTM1 and M2-2 protein (Fig. 3F). Overall, these results indicate that SQSTM1 may be related to M2-2 degradation.

Fig 3.

M2-2 is degraded by autophagy. (A) HEK-293T cells transfected with plasmids encoding GFP-M2-2 for 24 h were treated with 20 µM CQ and/or 500 nM Rap for 12 h, followed by detection with anti-GFP, LC3, or ACTB. (B) HEK-293T cells were transfected with 40 pM of siATG5 (1#, 2#, or 3#) or siCon for 36 h, and the extracted proteins were analyzed for ATG5 or ACTB. (C) HEK-293T cells treated with siATG5 or siCon were transfected with pEGFP-M2-2 for 36 h in the presence or absence of 500 nM Rap and then analyzed for GFP, ATG5, LC3, or ACTB proteins. (D) HEK-293T cells were transfected with 40 pM of siATG7 or siCon for 36 h, and the extracted proteins were analyzed for ATG7 or ACTB. (E) HEK-293T cells treated with siATG7 or siCon were transfected with pEGFP-M2-2 for 36 h in the presence or absence of 500 nM Rap and then analyzed for GFP, ATG7, LC3, or ACTB proteins. (F) HEK-293T cells were transfected with pEGFP-M2-2 or pEGFP-C1 plasmids for 24 h and then treated with 500 nM Rap for 12 h, followed by immunoprecipitation with anti-GFP, and analyzed for SQSTM1, GFP, and ACTB.

SQSTM1 is involved in the autophagic degradation of M2-2

SQSTM1, a selective autophagy receptor, interacts with viral proteins and delivers them to autophagosomes for degradation (27, 29). We continued to investigate whether SQSTM1 may be involved in the autophagy-mediated aMPV/C M2-2 degradation. HEK-293T cells were cotransfected with different concentrations of pFLAG-SQSTM1 plasmids and pEGFP-M2-2 plasmids, followed by Western blotting. As shown in Figure 4A, SQSTM1 overexpression contributed to a dose-dependent reduction in M2-2 expression, indicating that SQSTM1 degraded aMPV/C M2-2. The SQSTM1-knockdown cells were cotransfected with pFLAG-SQSTM1 and pEGFP-M2-2 plasmids. The knockdown SQSTM1 enhanced the expression of M2-2 compared to the control group. However, SQSTM1 overexpression attenuated this increase (Fig. 4B). To determine whether SQSTM1-mediated M2-2 degradation is related to autophagy, ATG7 knockout (ATG7-KO) 293T cells and ATG7 knockdown (siATG7) Vero cells were cotransfected with pFLAG-SQSTM1 and pEGFP-M2-2 plasmids. The knockout or knockdown of ATG7 blocked SQSTM1-induced M2-2 degradation compared to the control groups (Fig. 4C and D), indicating that SQSTM1 induced M2-2 degradation through autophagy pathway. To further verify the role of SQSTM1 in the autophagic degradation of M2-2, Vero cells transfected with siSQSTM1 or siCon were transfected with the plasmids pEGFP-M2-2 and were treated with Rap. The knockdown of SQSTM1 attenuated Rap-mediated M2-2 degradation compared to the control group (Fig. 4E), indicating that SQSTM1 played an important role in Rap-mediated M2-2 degradation.

Fig 4.

SQSTM1 is involved in M2-2 degradation via autophagy. (A) HEK-293T cells cotransfected with different concentrations of plasmids encoding FLAG-SQSTM1 and GFP-M2-2 for 24 h were processed and analyzed for GFP, FLAG, or ACTB proteins. (B) Vero cells treated with siSQSTM1 or siCon were transfected with the plasmids pFLAG-SQSTM1 or pFLAG and GFP-M2-2 for 24 h, and cells were collected for detection with anti-GFP, anti-FLAG, and anti-ACTB antibodies. (C and D) ATG7-KO or siATG7 HEK 293T cells were transfected with plasmids expressing pFLAG-SQSTM1 and GFP-M2-2 for 24 h, followed by detection with anti-GFP, anti-FLAG, anti-ATG7, or ACTB antibodies. (E) Vero cells transfected with 40 pM siSQSTM1 or siCon were transfected with plasmids encoding GFP-M2-2 in the presence or absence of 500 nM Rap, followed by detection with anti-GFP, SQSTM1, LC3, or ACTB antibodies. (F) Vero cells treated with siSQSTM1 were cotransfected with plasmids encoding FLAG-SQSTM1 or FLAG, GFP-M2-2, and mCherry-LC3 for 24 h. The cells were incubated with anti-FLAG antibodies, followed by treatment with the indicated secondary antibodies, and the fluorescence signals of M2-2 (green), FLAG-SQSTM1 (purple), mCherry-LC3 (red), and DAPI (blue) were observed under a confocal fluorescence microscope. (G) Vero cells treated with siSQSTM1 were cotransfected with plasmids encoding FLAG-SQSTM1 or FLAG and GFP-M2-2 for 24 h. The cells were incubated with anti-FLAG and anti-LC3 antibodies, followed by treatment with the indicated secondary antibodies, and the fluorescence signals of M2-2 (green), FLAG-SQSTM1 (purple), LC3 (red), and DAPI (blue) were observed under a confocal fluorescence microscope. (H) HEK-293T (SQSTM1-KO) cells were cotransfected with plasmids encoding FLAG-SQSTM1 or FLAG, GFP-M2-2, and mCherry-LC3 for 24 h, followed by treatment with 20 µM CQ for 12 h, and precipitated with anti-GFP antibodies. The extracted proteins were detected with anti-Cherry, anti-GFP, anti-FLAG, and anti-ACTB antibodies. (I) HEK-293T (SQSTM1-KO) cells were cotransfected with plasmids encoding FLAG-SQSTM1 or FLAG and GFP-M2-2 for 24 h, followed by treatment with 20 µM CQ for 12 h, and precipitated with anti-GFP antibodies. The extracted proteins were detected with anti-LC3, anti-GFP, anti-FLAG, and anti-ACTB antibodies.

Given that SQSTM1 serves as a cargo receptor that promotes the localization of viral proteins in autophagosomes (29), we subsequently examined the colocalization between M2-2 and the autophagosome marker LC3. Vero cells treated with siSQSTM1 were transfected with the plasmids pEGFP-M2-2 and pFLAG-SQSTM1 in the presence or absence of mCherry-LC3 expression, and the colocalization between M2-2 and LC3 was observed using confocal imaging. As shown in Figure 4F and G, there was a weak colocalization between M2-2 and exogenous LC3 (mCherry-LC3) or endogenous LC3 in cells in the absence of SQSTM1 overexpression, whereas the overexpression of SQSTM1 promoted a significant colocalization between M2-2 and exogenous LC3 (mCherry-LC3) and endogenous LC3 in these cells. We cotransfected SQSTM1 knockout (SQSTM1-KO) HEK-293T cells with the plasmids pFLAG-SQSTM1 or pFLAG, pEGFP-M2-2, and pmCherry-LC3 plasmids, and performed a co-IP assay using anti-GFP, anti-FLAG, and anti-Cherry antibodies. Compared to the control cells, the exogenous expression of SQSTM1 promoted an increase in the mCherry-LC3 precipitated by GFP-M2-2 in the (SQSTM1-KO) HEK-293T cells, further indicating that SQSTM1 enhances the interaction between M2-2 and exogenous LC3 (Fig. 4H). We transfected (SQSTM1-KO) HEK-293T cells with pEGFP-M2-2 and pFLAG-SQSTM1 or pFLAG plasmids, and found that SQSTM1 overexpression also promoted the interaction between M2-2 and endogenous LC3 (Fig. 4I). These results indicate that SQSTM1 is involved in the selective autophagic degradation of M2-2 by enhancing the interaction between aMPV/C M2-2 and LC3.

PB1 domain of SQSTM1 is required for SQSTM1-mediated M2-2 degradation

SQSTM1 contains three main functional domains (the PB1, ZZ, and UBA domains) and an LIR motif (Fig. 5A). Among these, the PB1 and UBA domains and LIR motif are believed to play important roles in the autophagic degradation of substrates (20). Thus, we continued to screen and identify the specific interacting domains. A series of plasmids harboring mutated SQSTM1 (pFLAG-SQSTM1ΔPB1, pFLAG-SQSTM1ΔZZ, pFLAG-SQSTM1ΔUBA, or pFLAG-SQSTM1ΔLIR) were constructed and cotransfected with pEGFP-M2-2 into cells, and the colocalization between the different SQSTM1 mutants and M2-2 was examined. Deletion of the PB1 domain of SQSTM1 had the altered distribution of the colocalization between SQSTM1 and M2-2, as manifested by a dispersive distribution and less colocalization (Fig. 5B). No appreciable changes were observed in the distribution of colocalizations between the other mutants and M2-2 (Fig. 5B), indicating that the PB1 domain of SQSTM1 might contribute to the colocalization of SQSTM1 to M2-2. HEK-293T cells were also cotransfected with the different mutants (pFLAG-SQSTM1ΔPB1, pFLAG-SQSTM1ΔZZ, pFLAG-SQSTM1ΔUBA, or pFLAG-SQSTM1ΔLIR) and pEGFP-M2-2 for the co-IP assay using anti-GFP and anti-FLAG antibodies. A specific band in cells co-expressing SQSTM1ΔPB1 and GFP-M2-2 disappeared (Fig. 5C), further indicating that the PB1 domain of SQSTM1 is indeed required for the interaction between SQSTM1 and M2-2.

Fig 5.

PB1 domain of SQSTM1 is required for SQSTM1-mediated M2-2 degradation. (A) A schematic representation of SQSTM1 and SQSTM1 mutants. (B) Vero cells were coexpressed with GFP-M2-2 and FLAG-SQSTM1 or FLAG-SQSTM1 mutants for 24 h, respectively. The cells were fixed and stained with DAPI (blue), and the colocalization between M2-2 (green) and FLAG-SQSTM1 or its mutants (red) was examined by confocal fluorescence microscopy. (C) HEK-293T cells were transfected with plasmids encoding GFP-M2-2 and FLAG-SQSTM1 or FLAG-SQSTM1 mutants, followed by precipitation with anti-GFP antibodies, and analyzed for GFP, FLAG, or ACTB proteins. (D–G) HEK-293T cells or SQSTM1-KO HEK-293T cells cotransfected with different concentrations of plasmids encoding FLAG-SQSTM1ΔPB1 (D and E) or FLAG-SQSTM1 (F and G) and GFP-M2-2 for 24 h were processed, and the extracted proteins were detected with anti-GFP, anti-FLAG, and anti-ACTB antibodies. (H and I) Vero cells were transfected with FLAG, FLAG-SQSTM1, or FLAG-SQSTM1ΔPB1 for 24 h, followed by infection with aMPV/C, and cells or whole-cell culture media were collected at 72 hours postinfection (hpi) and were analyzed for viral N protein, FLAG, or ACTB (H) or aMPV/C titers (I). (J and K) siSQSTM1-treated Vero cells were transfected with FLAG, FLAG-SQSTM1, or FLAG-SQSTM1ΔPB1 for 24 h, followed by infection with aMPV/C, and cells were collected and analyzed for viral N protein, FLAG, or ACTB (J) and aMPV/C titers using TCID₅₀ (K). Data are shown as the means ± standard deviations (SD) of three independent experiments (**, P ＜0.01).

We continued to examine whether the PB1 domain of SQSTM1 plays a role in M2-2 degradation. To this end, HEK-293T cells or SQSTM1-knockout (SQSTM1-KO) HEK-293T cells were cotransfected with different concentrations of the plasmids pEGFP-M2-2, pFLAG, and pFLAG-SQSTM1 or pFLAG-SQSTM1-ΔPB1. As shown in Figure 5D through G, cells expressing the wild-type SQSTM1 instead of those expressing SQSTM1 lacking the PB1 domain were characterized by a dose-dependent degradation of M2-2, indicating that the PB1 domain of SQSTM1 plays an important role in M2-2 degradation. Subsequently, we analyzed the effects of SQSTM1 PB1 domain deletion on aMPV/C replication. Vero cells transfected with the plasmids pFLAG, pFLAG-SQSTM1, or pFLAG-SQSTM1-ΔPB1 were infected with aMPV/C for 72 h, and aMPV/C replication was analyzed using Western blotting and a TCID₅₀ assay. As shown in Figure 5H and I, SQSTM1 lacking the PB1 domain restored the expression of viral proteins and viral titers. Further results showed inhibition of virus replication by SQSTM1 overexpression, but SQSTM1-ΔPB1 has no inhibitory effect in Vero cells treated with siSQSTM1 (Fig. 5J and K). Collectively, these results provide compelling evidence that the PB1 domain of SQSTM1 is necessary for the SQSTM1-mediated degradation of M2-2 and the inhibition of aMPV/C replication.

Ubiquitination of the 67th amino acid of M2-2 is essential for the SQSTM1-mediated autophagic degradation of M2-2

SQSTM1 recognizes ubiquitinated substrates and delivers them to autophagosomes for subsequent lysosomal degradation (27, 29). We continued to examine whether M2-2 bound to SQSTM1 undergoes ubiquitination. HEK-293T cells were cotransfected with the plasmids pEGFP-M2-2 or pEGFP-C1 and HA-Ub for 36 h. M2-2 was found to be indeed ubiquitinated by a ubiquitination analysis (Fig. 6A). Cells coexpressing SQSTM1 and Ub served as a positive control for ubiquitination. HEK-293T cells transfected with pEGFP-M2-2 and HA-Ub, HA-Ub-K48, or HA-Ub-K63 were further used for examining the type of M2-2 ubiquitination. Similar to the positive group, ubiquitination of M2-2 increased in cells expressing Ub-K63, but the ubiquitination of M2-2 in cells expressing Ub-K48 was barely detectable, indicating that M2-2 undergoes ubiquitination in a K63-linked ubiquitination-dependent manner (Fig. 6B). To further screen and identify the ubiquitinated sites of M2-2, these sites were predicted online (http://bdmpub.biocuckoo.org), and there are three putative ubiquitination sites (Fig. 6C). Each of these three ubiquitination sites was subsequently mutated from lysine (K) to arginine (R), and the key ubiquitination site(s) of M2-2 was further identified based on a ubiquitination assay. In the cells expressing the K67R mutant of M2-2, M2-2 remained unubiquitinated, but the ubiquitination in cells expressing the two other mutants was unaffected (Fig. 6D). To investigate the role of the 67th amino acid of M2-2 in SQSTM1-mediated M2-2 degradation, we cotransfected HEK-293T cells with different concentrations of the plasmids pFLAG-SQSTM1 and pEGFP-M2-2 (K67R) or pEGFP-M2-2, and analyzed the expression levels of GFP-M2-2 (K67R) or GFP-M2-2. SQSTM1 overexpression induced a significant dose-dependent reduction in GFP-M2-2 expression, but it is unable to mediate the degradation of GFP-M2-2 (K67R) (Fig. 6E and F), indicating that the 67th amino acid of M2-2 may be a key site for the SQSTM1-mediated degradation of M2-2. We further performed a co-IP assay to investigate whether mutation of the 67th amino acid influenced the interaction between SQSTM1 and M2-2 using HEK-293T cells expressing GFP, GFP-M2-2, or GFP-M2-2 (K67R). Interestingly, endogenous SQSTM1 specific bands were observed in GFP-M2-2-expressing cells, whereas these bands were virtually undetectable in the presence of GFP or GFP-M2-2 (K67R) (Fig. 6G). Collectively, these results indicate that mutation (K67R) of the 67th amino acid of M2-2 disrupted the interaction between M2-2 and SQSTM1, thereby suppressing the SQSTM1-mediated autophagic degradation of M2-2.

Fig 6.

Ubiquitination of the 67th amino acid of M2-2 is required for the SQSTM1-mediated autophagic degradation of M2-2. (A) HEK-293T cells were coexpressed with HA-Ub and GFP-M2-2 or GFP for 24 h, followed by precipitation with anti-GFP antibodies, and the extracted proteins were detected with anti-HA, GFP, and ACTB antibodies. Cells were coexpressed with HA-Ub and GFP-SQSTM1 as a positive control. (B) HEK-293T cells were coexpressed with GFP-M2-2 and HA-Ub, HA-Ub-K48, or HA-Ub-K63 for 24 h. Protein extraction and Western blotting were performed as described for A. (C) Potential ubiquitination sites of GFP-M2-2 were predicted in http://bdmpub.biocuckoo.org. (D) HEK-293T cells were coexpressed with GFP or GFP-M2-2 or its mutants (K8R, K25R, and K67R) and HA-Ub for 24 h. Protein extraction and Western blotting were performed as described for A. (E and F) HEK-293T cells were transfected with different concentrations of plasmids encoding FLAG-SQSTM1 or FLAG and plasmids encoding GFP-M2-2 (K67R) (E) or GFP-M2-2 (F) for 24 h, and the extracted proteins were detected with anti-FLAG, GFP, and ACTB antibodies. (G) HEK-293T cells were expressed with GFP-M2-2 (K67R), GFP-M2-2, or GFP for 24 h, followed by precipitation with anti-GFP antibodies, and the extracted proteins were detected with anti-SQSTM1, GFP, and ACTB antibodies.

SQSTM1 mediates the degradation of hMPV M2-2

Due to close genetic and antigenic relatedness between aMPV/C and hMPV (6), we constructed a phylogenetic tree of the M2-2 gene sequences of aMPV subgroups and hMPV using the neighbor-joining clustering method, and found that only aMPV/C and hMPV isolates belong to the separate subclusters within the same cluster (Fig. 7A). Similarly, the co-IP assay showed that hMPV M2-2 (hM2-2) was observed to interact with SQSTM1 by detecting exogenous or endogenous SQSTM1 (Fig. 7B through D). We further found the colocalization between hMPV M2-2 and SQSTM1 in cells (Fig. 7E). We transfected HEK-293T cells with different concentrations of the pFLAG-SQSTM1 and pEGFP-hMPV M2-2 plasmids and found that exogenous SQSTM1 expression increased, accompanied by the reduction of hMPV M2-2 expression (Fig. 7F). To demonstrate that hMPV M2-2 reduction is associated with SQSTM1-mediated selective autophagy, we analyzed the effect of ATG7 silencing on the SQSTM1-mediated degradation of hMPV M2-2. As shown in Figure 7G, the knockdown of ATG7 restored the SQSTM1-mediated hMPV M2-2 reduction. These results indicate that SQSTM1 also mediates the degradation of hMPV M2-2 via the process of selective autophagy.

Fig 7.

SQSTM1 induces hMPV M2-2 degradation. (A) The evolutionary relationship between the M2-2 gene of aMPV subgroups (A, B, C, and D) and hMPV was analyzed. A phylogenetic tree was constructed using MEGA 6.06 software, and the reference strains were obtained from GenBank. The scale bar indicates estimated phylogenetic divergence. (B) HEK-293T cells transfected with plasmids encoding GFP-hM2-2 and FLAG-SQSTM1 or FLAG for 24 h were processed and immunoprecipitated with anti-FLAG, and analyzed for FLAG, GFP, and ACTB. (C) HEK-293T cells cotransfected with pFLAG-SQSTM1 and pEGFP-hM2-2 or pEGFP-C1 plasmids for 24 h were processed and immunoprecipitated with anti-GFP, and analyzed for FLAG, GFP, and ACTB. (D) HEK-293T cells transfected with plasmids encoding GFP-hM2-2 or GFP for 24 h were processed and immunoprecipitated with anti-GFP, and analyzed for SQSTM1, GFP, and ACTB. (E) Vero cells were cotransfected with plasmids FLAG-SQSTM1 or FLAG and pEGFP-hM2-2 for 24 h, the cells were fixed and stained with DAPI (blue), and the colocalization between M2-2 (green) and FLAG-SQSTM1 (red) was examined by confocal fluorescence microscopy. (F) HEK-293T cells were transfected with different concentrations of plasmids encoding FLAG-SQSTM1 and GFP-hM2-2 for 24 h, and the extracted proteins were detected with anti-FLAG, GFP, and ACTB antibodies. (G) HEK-293T cells treated with siATG7 or siCon were cotransfected with the plasmids pEGFP-M2-2 and pFLAG-SQSTM1 for 24 h, and the extracted proteins were detected with anti-FLAG, GFP, ATG7, and ACTB antibodies.

DISCUSSION

Autophagy is a degradative process associated with the delivery of cytoplasmic contents to autophagosome for lysosome-mediated degradation (14, 34, 35). This basal autophagic process serves as a non-selective intracellular energy supply and quality control mechanism, which is associated with the indiscriminate degradation of cytosolic components (36). Compared to this non-selective type of autophagy, the process of selective autophagy mainly involves the recognition and degradation of specific and exclusive ubiquitinated cargoes by designated receptors, such as SQSTM1, NDP52, or NBR1 (37, 38). These selective autophagic receptors play important roles in connecting cargoes with lapidated LC3 via their ubiquitin-binding domains and LIR, respectively, thereby promoting the subsequent autophagic degradation of the selected cargoes (39–42). SQSTM1, a ubiquitous and multifunctional protein, serves as a selective autophagy receptor that induces viral protein degradation, thereby inhibiting viral replication (27, 43). In our previous study, we found that aMPV/C infection reduced SQSTM1 expression (28), thereby indicating that the expression of SQSTM1 may be involved in aMPV/C replication as a selective autophagic receptor. To gain further insights in this regard, we analyzed the regulatory relationships between SQSTM1 and aMPV/C replication and the specific regulatory mechanisms. Our results revealed that SQSTM1 negatively regulates aMPV/C replication via selective autophagy, and that the 67th amino acid of aMPV/C M2-2 is essential for the SQSTM1-mediated aMPV/C M2-2 degradation.

In recent years, an increasing number of research data have shown that selective autophagy participates in the regulation of viral replication, and among these, it has been reported that SQSTM1 participates in the autophagic degradation of viral components or host proteins. For example, it has been demonstrated that the capsid proteins of dengue virus (DENV), Seneca Valley virus (SVV), Sindbis virus, and infectious bursal disease virus can be degraded via SQSTM1-mediated autophagy (26, 27, 44). Similarly, SQSTM1 has been found to mediate the autophagic degradation of pattern recognition receptors (45, 46). In this study, we demonstrated that SQSTM1 negatively regulates aMPV/C replication, which is related to the autophagy (Fig. 1). Subsequently, the results further showed that the degradation of aMPV/C M2-2 is dependent on an autophagic pathway, which involves the specific recognition of M2-2 by the interaction between SQSTM1 and aMPV/C M2-2 in the presence of Rap (Fig. 2 and 3). Collectively, these observations provided evidence to indicate that SQSTM1 may participate in the autophagic degradation of M2-2.

SQSTM1 can interact with LC3 in autophagosomes and then mediate the degradation of substrate proteins via the autophagic pathway (39). Our results revealed that SQSTM1 promotes the autophagic degradation of M2-2 (Fig. 4A through E) and further confirmed that SQSTM1 clearly enhances the colocalization and interaction between M2-2 and LC3 (Fig. 4F through I), thereby indicating that SQSTM1 serves as an intermediate in mediating the autolysosomal degradation of M2-2. In short, our previous study confirmed that aMPV infection-induced autophagy contributes to virus replication (28), while SQSTM1, as a selective autophagy receptor, can inhibit virus replication by degrading virus protein in this study. These results are similar to some other viruses, such as SVV (27, 47) and DENV (48). During virus infection, SVV- and DENV-induced autophagy facilitates viral replication while also undergoing selective autophagy mediated by SQSTM1 for degradation of viral proteins.

The main functional domains of SQSTM1 associated with the autophagic process have been established to be the PB1, ZZ, and UBA domains and LIR motif (20), among which the UBA domain and LIR motif that combine with ubiquitin substrates and LC3, respectively, are responsible for the degradation of cargoes (49). Surprisingly, we found that SQSTM1 lacking the PB1 domain lost its capacity to interact with M2-2 and was characterized by a dispersed cytoplasmic distribution rather than by a punctate accumulation. Moreover, this domain was associated with a significant attenuation of the degradation of M2-2 and the inhibition of aMPV/C replication (Fig. 5), which contrasts to previously reported findings (29). It has previously been established that a phase separation of SQSTM1 induced by polyubiquitin chains drives the concentration and segregation of autophagic contents and subsequent degradation (49). Given the important role of SQSTM1 in recognizing ubiquitinated proteins in the process of SQSTM1-mediated selective autophagy, we subsequently focused on whether ubiquitination occurs in M2-2 and whether this ubiquitination plays a key role in the degradation of M2-2 by SQSTM1. Ubiquitin is a conserved molecule that binds to targeted proteins via a cascade of enzymic reactions (49), and then the ubiquitinated proteins are degraded either via the ubiquitin–proteasome system characterized by K48- or K11-linked ubiquitination or by autophagic degradation characterized by K63-linked ubiquitination (23, 50, 51). In this study, we identified the 67th amino acid of M2-2 as the key site for its ubiquitinated degradation mediated by SQSTM1. By analyzing the interaction between these two proteins, we discovered that this site is also involved in the specific recognition of M2-2 by SQSTM1 (Fig. 6). Consequently, these results indicate that SQSTM1 induces the autophagic degradation of M2-2 via recognition of the 67th amino acid of this protein.

The M2-2 protein of hMPV plays important roles in viral RNA transcription and translation (52), and thereby maintains the pathogenicity and immunogenicity of this virus within the host. Notably, it has been found that the replication efficiency of recombinant aMPV/C lacking the M2-2 gene is less than that of the parental virus (9). On the basis of a comparison of the M gene, it has previously been demonstrated that aMPV/C and hMPV are relatively close with respect to phylogeny and molecular evolution, which prompted us to analyze the evolutionary relationship between these two viruses with respect to the M2-2 gene. We indeed found that the evolutionary relationship between the M2-2 gene of aMPV/C and hMPV was similar to that described for the M gene (6). Similar to its interaction with aMPV/C M2-2, SQSTM1 was found to show a significant colocalization and interaction with hMPV M2-2 and mediated its degradation via selective autophagy (Fig. 7), thereby indicating that M2-2 has a similar biological function in these two viruses.

In conclusion, the findings of this study reveal, for the first time, that SQSTM1 recognizes the 67th amino acid of aMPV/C M2-2, followed by recruiting LC3, and degrades M2-2 via SQSTM1-mediated selective autophagy and finally inhibits aMPV/C replication (Fig. 8). The mechanisms underlying the inhibitory effects of SQSTM1 on aMPV/C replication will provide important insights for the development of antiviral drugs.

Fig 8.

Mechanisms underlying the SQSTM1 regulation of aMPV/C replication. SQSTM1, a selective receptor, recognizes the 67th ubiquitinated amino acid of aMPV/C M2-2 and recruits the LC3 as a platform, and then degrades aMPV/C M2-2 via the SQSTM1-mediated selective autophagy and finally inhibits aMPV/C replication.

MATERIALS AND METHODS

Cells and viruses

African Green Monkey Kidney cells (Vero) (CCL-81, ATCC) and HEK-293T cells (CRL-3216, ATCC) were originally purchased from the American Type Culture Collection. The cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) (12800082, Gibco, Life Technologies) supplemented with 10% fetal bovine serum (FBS) (10099141, Gibco, Life Technologies) and 1% streptomycin and penicillin (P1400, Solarbio) at 37°C in a 5% CO₂ incubator.

The JC strain of aMPV subgroup C (aMPV/C) was isolated from local broilers in China and preserved in the laboratory. For propagation, aMPV/C was inoculated into Vero cells, and the virus titer of the TCID₅₀ was 10^4.5 per 0.1 mL.

Plasmid construction, antibodies, and reagents

Plasmids (GFP tag) for cloning different viral genes were preserved in the laboratory. The mutant of the M2-2 gene (K8R, K25R, or K67R) was cloned into a pEGFP-C1, and the M2-2 gene was constructed into pCMV-HA vector. The full-length SQSTM1 gene and SQSTM1 mutated genes (ΔPB1, ΔZZ, ΔLIR, ΔUBA) derived from Vero cells were inserted into a pFLAG, pEGFP-C1, or pCherry-C1 vector. All the primers used in plasmid construction are listed in Table 1. Plasmids encoding GFP-hM2-2 were constructed commercially by Sangon Biotech (Shanghai, China).

TABLE 1.

Primers used in the construction of plasmids

Primers	Sequence (5’−3’)
GFP-M2-2-F	AAGAATTCTATGACTCTACAGCTGCCATGCAAGA
GFP-M2-2-R	AAGGTACCTTAACTTAGAAATGTTTTAACATAA
GFP-M2-2-8K-R-F	AGCTGCCATGCAGGATAGTGCAAAC
GFP-M2-2-8K-R-R	GTTTGCACTATCCTGCATGGCAGCT
GFP-M2-2-25K-R-F	TGATCTTCCTGAGGATGAGATTGGA
GFP-M2-2-25K-R-R	TCCAATCTCATCCTCAGGAAGATCA
GFP-M2-2-67K-R-F	CATTTATGTTAAGACATTTCTAAGT
GFP-M2-2-67K-R-R	ACTTAGAAATGTCTTAACATAAATG
Monkey-FLAG-SQSTM1-F	ATGAATTCAGATGGCGTCGCTCACCGTGAAGGCCT
Monkey-FLAG-SQSTM1-R	AAGGTACCTCACAATGGCGGGGAGTGCTTTGAA
Monkey-GFP/mCherry-SQSTM1-F	ATGAATTCAATGGCGTCGCTCACCGTGAAGGCCTA
Monkey-GFP/mCherry-SQSTM1-R	ATGGTACCTCACAATGGCGGGGAGTGCTTTGAATACT
FLAG-SQSTM1△PB1-F	ATGAATTCATTGCCGGCGGGACCACCGCCCACCGT
FLAG-SQSTM1△PB1-R	ATGGTACCTCACAATGGCGGGGAGTGCTTTGAA
FLAG-SQSTM1△ZZ-F	TGCTCAGGAGGCGCCCCGCAACATGTTCCCCAGCCCCTTCG GGCACCTGT
FLAG-SQSTM1△ZZ-R	ACAGGTGCCCGAAGGGGCTGGGGAACATGTTGCGGGGCGC CTCCTGAGCA
FLAG-SQSTM1△UBA-F	ATGAATTCATATGGCGTCGCTCACCGTGAAGGCCT
FLAG-SQSTM1△UBA-R	ATGGTACCTGCCTTGTACCCGCATCTCCCACCA
FLAG-SQSTM1△LIR-F	TAACTGCTCAGGAGGAGACGATGACTCTTCAAAAGAAGTGG ACCCGTCC
FLAG-SQSTM1△LIR-R	TGGACGGGTCCACTTCTTTTGAAGAGTCATCGTCTCCTCCTG AGCAGTTA
Monkey-mCherry-LC3-F	ATGAATTCTATGCCGTCGGAGAAGACTTTCAAGC
Monkey-mCherry-LC3-R	ATGGATCCTTACACTGACAATTTCATCCCGAAC

The aMPV/C N monoclonal antibody used in this study was prepared in our laboratory. The commercial antibodies and reagents used in the study were as follows: rabbit anti-SQSTM1 antibody (P0067, Sigma-Aldrich), rabbit anti-LC3B antibody (L7543, Sigma-Aldrich), tetramethyl rhodamine isothiocyanate (TRITC)-conjugated goat anti-rabbit or anti-mouse secondary antibody (T6779 or AP181R, Sigma-Aldrich), horseradish peroxidase (HRP)-conjugated anti-mouse (A9044, Sigma-Aldrich) or anti-rabbit (A0545, Sigma-Aldrich) secondary antibody, Anti-Flag M2 Affinity Gel (A2220, Sigma-Aldrich), Rap (R0395, Sigma-Aldrich), CQ (C6628, Sigma-Aldrich), rabbit anti-ATG5 antibody (2630, Cell Signaling Technology), rabbit anti-ATG7 antibody (2631, Cell Signaling Technology), mouse anti-FLAG antibody (F1804, Sigma-Aldrich), mouse anti-ACTB antibody (D191047, Sangon Biotech), mouse anti-mCherry-Tag mAb (AE002, ABclonal), Alexa Fluor 647-labeled goat anti-rabbit IgG (A21245, Invitrogen), TRNzol (Y1321, TIANGEN), GFP-Nanoab-Agarose (GNA-50100, Lablead), phenylmethanesulfonyl fluoride (PMSF) (ST506, Beyotime), cell lysis buffer NP-40 (P0013F, Beyotime), RIPA (P0013B, Beyotime), Protein A Plus-Agarose (sc-2002, Santa Cruz), rabbit anti-GFP antibody (SR48-02, Huaan Biological Technology), and HiScript III RT SuperMix (R323-01, Vazyme).

Virus infection and TCID₅₀ assay

According to the experimental requirements, Vero cells were inoculated with aMPV/C at a multiplicity of infection (MOI) of 0.1. After 1.5 h of incubation, 2% FBS culture medium was added to the cells, followed by culturing at 37°C in a 5% CO₂ incubator. To determine the effect of SQSTM1 or SQSTM1 mutants on viral replication, Vero cells were transfected with plasmids encoding SQSTM1 or SQSTM1 mutants 24 h prior to viral infection.

TCID₅₀ assay: monolayer Vero cells were cultured in 96-well plates, and the virus-containing cell supernatants collected in a different experiment were serially diluted and inoculated on these cells. After adding 2% FBS culture medium, the cells were incubated in a 5% CO₂ incubator at 37°C. The cells were examined under a microscope 5 d after infection for an assessment of a cytopathic effect, and the TCID₅₀ was calculated using the Reed–Muench methods (53).

RNA interference (RNAi)

The siRNA sequences used for this study are as follows: siATG7 in the HEK-293T cells (sense: 5ʹ-GCAGUUUGCUGCUCCCUUUAAUTT-3ʹ; antisense: 5ʹ-AUUAAAGGGAGCAAACUGCTT-3ʹ), siATG5 in the HEK-293T cells (sense: 5ʹ-GCAGAUGGACAGCUGCAUATT-3ʹ; antisense: 5ʹ-UAUGCAGCUGUCCAUCUGCTT-3ʹ), siSQSTM1 in the Vero cells (sense: 5ʹ-GCUCUGGACACCAUCCAGUTT-3ʹ; antisense: 5ʹ-ACUGGAUGGUGUCCAGAGCTT-3ʹ), or siATG7 in the Vero cells (sense: 5ʹ-GCAGUUUGCUGCUCCCUUUAAUTT-3ʹ; antisense: 5ʹ-AUUAAAGGGAGCAAACUGCTT-3ʹ) were used to silence the expression of corresponding proteins using Lipofectamine RNAiMAX (13778-150, Invitrogen).

Confocal imaging

Vero cells grown in 24-well plates containing sterile cover slides were transfected with the indicated plasmids for 24 h. The cells were fixed with 4% paraformaldehyde (PFA) and subsequently blocked with 5% skim milk for 1 h, followed by treatment with the corresponding primary antibody and secondary antibodies. The nuclei were stained with 4ʹ,6-diamidino-2-phenylindole (DAPI) and examined under a laser confocal microscope (LSM 880NLO, Zeiss, Germany).

Immunoprecipitation (IP), co-IP, ubiquitination assay, and Western blotting

For co-IP, HEK-293T cells transfected with appropriate plasmids for 36 h were processed with NP-40 lysis buffer containing 1% PMSF, followed by centrifugation. The supernatants were immunoprecipitated with GFP-Nanoab-Agarose or Anti-Flag M2 Affinity Gel for 5 h. After washing with NP-40, the beads were collected and then analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).

For ubiquitination assay, HEK-293T cells were transfected with plasmids encoding GFP-M2-2 or GFP, GFP-SQSTM1, and HA-Ub; encoding GFP-M2-2 and HA-Ub, HA-Ub-K48, or HA-Ub-K63; and encoding HA-Ub and GFP, GFP-M2-2, or GFP-M2-2 mutants (K8R, K25R, or K67R). Cell lysates were obtained in IP buffer (Beyotime, P0013), followed by centrifugation. The supernatants were precipitated with GFP-Nanoab-Agarose and then gently rocked at 4°C. After washing with IP buffer, proteins eluted from the beads were analyzed by SDS-PAGE.

For Western blotting, the cell proteins were transferred to nitrocellulose membranes (PALL, 66485) for 2 h, followed by blocking with 5% skim milk at room temperature for 2 h. The blocked membranes were thereafter incubated overnight with the appropriate primary antibodies and subsequently washed three times with phosphate-buffered saline (PBS) (each for 10 min). The membranes were then incubated with HRP-labeled secondary antibodies, and the results were subsequently visualized using chemiluminescence and chemical imaging systems (GE, USA).

Genetic evolutionary analysis

Based on similarity of amino acid sequence between aMPV M2-2 in different subgroups and hMPV M2-2, a phylogenetic tree was constructed using the neighbor-joining clustering method through MEGA 6.06 software. Bootstrap values are indicated at each branching point. Scale bar indicates estimated phylogenetic divergence.

Statistical analyses

Data were expressed as means ± standard deviations (SD). Differences were analyzed by one-way analysis of variance or Student’s t-test using the GraphPad Prism 8.3.0 Software (NS, not significant; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001).

DATA AVAILABILITY

The data that support the findings of this study are available from the corresponding author upon reasonable request.