PABPC4 Broadly Inhibits Coronavirus Replication by Degrading Nucleocapsid Protein through Selective Autophagy

ABSTRACT Emerging coronaviruses (CoVs) can cause severe diseases in humans and animals, and, as of yet, none of the currently available broad-spectrum drugs or vaccines can effectively control these diseases. Host antiviral proteins play an important role in inhibiting viral proliferation. One of the isoforms of cytoplasmic poly(A)-binding protein (PABP), PABPC4, is an RNA-processing protein, which plays an important role in promoting gene expression by enhancing translation and mRNA stability. However, its function in viruses remains poorly understood. Here, we report that the host protein, PABPC4, could be regulated by transcription factor SP1 and broadly inhibits the replication of CoVs, covering four genera (Alphacoronavirus, Betacoronavirus, Gammacoronavirus, and Deltacoronavirus) of the Coronaviridae family by targeting the nucleocapsid (N) protein through the autophagosomes for degradation. PABPC4 recruited the E3 ubiquitin ligase MARCH8/MARCHF8 to the N protein for ubiquitination. Ubiquitinated N protein was recognized by the cargo receptor NDP52/CALCOCO2, which delivered it to the autolysosomes for degradation, resulting in impaired viral proliferation. In addition to regulating gene expression, these data demonstrate a novel antiviral function of PABPC4, which broadly suppresses CoVs by degrading the N protein via the selective autophagy pathway. This study will shed light on the development of broad anticoronaviral therapies. IMPORTANCE Emerging coronaviruses (CoVs) can cause severe diseases in humans and animals, but none of the currently available drugs or vaccines can effectively control these diseases. During viral infection, the host will activate the interferon (IFN) signaling pathways and host restriction factors in maintaining the innate antiviral responses and suppressing viral replication. This study demonstrated that the host protein, PABPC4, interacts with the nucleocapsid (N) proteins from eight CoVs covering four genera (Alphacoronavirus, Betacoronavirus, Gammacoronavirus, and Deltacoronavirus) of the Coronaviridae family. PABPC4 could be regulated by SP1 and broadly inhibits the replication of CoVs by targeting the nucleocapsid (N) protein through the autophagosomes for degradation. This study significantly increases our understanding of the novel host restriction factor PABPC4 against CoV replication and will help develop novel antiviral strategies.

clustering (2). They can infect various hosts, including humans and animals (dogs, cats, chickens, cows, and pigs), causing enteric, respiratory, neurological, and hepatic diseases. The human coronaviruses (HCoVs) were thought to cause mild, self-limiting respiratory infections before the severe acute respiratory syndrome coronavirus (SARS-CoV) outbreak (3). The SARS outbreak in China in 2002 and the Middle East Respiratory Syndrome-CoV emerging in the Middle East in 2012 as a worldwide epidemic virus caused a series of highly pathogenic respiratory syndromes with severe morbidity and high mortality rates (4)(5)(6)(7). The outbreak of emerging SARS-CoV-2 disease (COVID- 19) has recently been brought to global attention and declared a pandemic by the World Health Organization on 11 March 2020 (4,8,9). The avian coronavirus, infectious bronchitis virus (IBV), caused poor weight gain and reduced egg production in chickens (10). The swine coronavirus, porcine epidemic diarrhea virus (PEDV), transmissible gastroenteritis virus (TGEV), porcine deltacoronavirus (PDCoV), and swine acute diarrhea syndrome coronavirus (SADS-CoV) cause severe gastroenteritis in neonatal pigs leading to significant morbidity and mortality (11)(12)(13)(14)(15). Coronaviruses have the potential to spread rapidly and cause significant disease in humans and animals, but there is no widely available antiviral drug or vaccine that could be used for the treatment or prevention of the resulting infections.
The CoV genome typically encodes two polyproteins (pp1a and pp1ab) and four main structural proteins (spike protein, S; envelope protein, E; membrane protein, M; and nucleocapsid protein, N) (1). The S protein is a class I fusion protein responsible for virus attachment and membrane fusion (16). The N protein is abundantly expressed within infected cells; plays an important role in virus replication and immune evasions (17,18), such as binding to the viral RNA to form a helical nucleocapsid; and is involved in transcription, translation, and viral assembly (19)(20)(21)(22). Therefore, the N protein may serve as a useful drug target for treating viral infections (23,24).
Macroautophagy/autophagy is a major degradative process for maintaining cellular homeostasis by degrading the aggregated proteins or organelles and misfolded or long-lived cytoplasmic proteins during periods of starvation in eukaryotic cells (25). During autophagy, the misfolded proteins damaged organelles or invading pathogens, as substrate proteins are modified with ubiquitin and recognized by cargo receptors (25,26). The cargo receptors, including SQSTM1/p62, NDP52/CALCOCO2 (nuclear dot protein 52 kDa), OPTN (optineurin), and NBR1 (neighbor of BRCA1), recognize the ubiquitin-binding substrate proteins and directly interact with ATG8 (autophagy-related protein 8) for selective packaging of their targets into the autophagosome. Then, the autophagosomes fuse with lysosomes to degrade the cargo and release metabolic byproducts, such as amino acids (27). In addition to its physiological functions in cellular homeostasis, autophagy has been shown to play an important role in innate immunity and viral replication (28,29). For example, autophagy inhibits herpes simplex virus 1 (HSV-1) replication by protein kinase R (PKR)-dependent autophagic degradation (30). However, foot-and-mouth disease virus (FMDV)-induced autophagy facilitates viral replication through its capsid protein VP2, which activates the EIF2S1-ATF4 pathway to decrease the aggregation of HTT103Q (huntingtin polyglutamine expansion protein) (31,32). CoVs induce autophagy in viral replication (33)(34)(35), but the underlying mechanism of the cross talk between autophagy and CoV replication is yet to be understood.
Poly(A)-binding protein (PABP), a nucleocytoplasmic shuttling protein, has a fundamental role in promoting gene expression by enhancing mRNA translation and stability (36,37). Poly(A)-binding protein bound to the 39 end of the mRNA via the polyadenylated [poly(A)] tail and 59 cap structure of the mRNA through interactions with eukaryotic initiation factor 4G (eIF4G) to form a "closed loop" (38)(39)(40). This special PABP and mRNA complex could enhance ribosome subunit recruitment to promote translation initiation and protect mRNAs from exonucleolytic degradation (41,42). Poly (A)-binding protein, predominantly present in the cytoplasm, preferentially accumulates in the nucleus during heat shock and oxidative stress (43,44). PABP could be driven into the nucleus upon viral infections from viruses such as HSV-1 and rotavirus (45,46). Several studies have investigated that the coronavirus N proteins bind to PABP, and PABP binding by the N protein of bovine coronavirus (BCoV) negatively regulates translation of coronaviral RNA and host mRNA both in vitro and in cells (47)(48)(49). Poly(A)-binding protein 4 (PABPC4), one of the homologs of PABP, was initially identified as a human T-cell activation-induced protein (50), but its role in cellular processes and virus infection has not been further explored. In this study, we demonstrated that the expression of PABPC4 was regulated by transcription factor SP1 during PEDV infection. In addition, PABPC4 inhibited CoV replication by targeting and degrading the CoV N protein through selective autophagy. Furthermore, we revealed that PABPC4 utilized E3 ubiquitin ligase MARCH8 to ubiquitinate CoV N protein, which was delivered by cargo receptor NDP52 to autophagosomes for selective degradation. Our findings revealed a new antiviral mechanism for PABPC4, by which it suppresses CoV replication through selective autophagy.

RESULTS
PEDV infection downregulates the expression of PABPC4 via the transcription factor of SP1. To screen the potential proteins for their ability to inhibit PEDV replication, we used the proteomics approach of isobaric tagging for relative and absolute quantitation (iTRAQ) to analyze the proteins in PEDV-infected cells and uninfected cells as previously described (51). The results showed that PABPC4 was downregulated in Vero cells during PEDV infection. Porcine kidney cells (LLC-PK1) and Vero cells were infected with PEDV (strain JS-2013) at an multiplicity of infection (MOI) of 1, and the cell lysates were analyzed using quantitative real-time PCR (qRT-PCR) to confirm the expression of PABPC4. As shown in Fig. 1A and C, the expression of PABPC4 mRNA was downregulated in both LLC-PK1 and Vero cells in response to PEDV infection, relative to that in mock-infected cells. We also showed that the PABPC4 protein level was significantly decreased in PEDV-infected cells ( Fig. 1B and D). Thus, PEDV infection reduces the expression of PABPC4.
The accumulation of PABPC4 in the cytoplasm may be affected by heat shock, HSV-1, and murine gammaherpesvirus 68 (MHV68) (43, 46, and 52), but the mechanisms by which transcription factors regulate the expression of PABPC4 are not fully understood. To identify the transcription factors of PABPC4, we amplified 1,350 bp of the PABPC4 promoter and cloned it into a luciferase vector (pGL3-Basic). Truncated promoters (designated D1 to D5) were cloned, and their ability to induce luciferase expression in HEK 293T cells was tested. A promoter fragment containing the nucleotides 2279 to 21 (D5) induced luciferase activity comparable to that of the full-length promoter. In contrast, constructs that lacked the sequence 2279 to 21 induced low or undetectable luciferase expression. These results suggested that the PABPC4 core promoter region lies between positions 2279 and 21. To further confirm the boundaries of the minimal PABPC4 promoter, a series of truncated promoters (designated D5-1 to D5-4), based on the sequence from 2279 to 21, were cloned into the luciferase vector. The construct containing nucleotides from 2153 to 21 (D5-3) showed increased luciferase expression (Fig. 1E). Altogether, these results indicate that the minimal PABPC4 core promoter is located between 2153 and 21.
To analyze the transcriptional regulation of PABPC4, we examined the promoter for potential transcription factor-binding sites (TFBSs) using the JASPAR vertebrate database (http://jaspar.genereg.net/) (34). We found that the minimal PABPC4 core promoter region (approximately 2153 to 21 bp) contains several TFBSs, including Zfx-, HIC2-, EGR1-, EGR2-, and SP1-binding sites. Next, we detected the mRNA levels of all of the putative transcription factors with qRT-PCR and found that only SP1, HIC2, and Zfx were downregulated following PEDV infection (Fig. 1F). To further evaluate whether these transcription factors were involved in the regulation of PABPC4 expression, we synthesized small interfering RNAs (siRNAs) targeting their mRNA sequences. The qRT-PCR analysis revealed that the level of PABPC4 mRNA was significantly reduced in Vero cells transfected with SP1 siRNA (Fig. 1G) but remained stable in cells transfected with HIC2 or Zfx siRNA ( Fig. 1H and I). Subsequently, we overexpressed the putative transcription factor SP1 to investigate the inducible expression of endogenous PABPC4. qRT-PCR revealed that SP1 overexpression significantly enhanced PABPC4 transcript levels relative to that observed following control vector transfection (Fig. 1J), and it induced higher luciferase expression with the promoter containing nucleotides from 2153 to 21 (D5-3) (Fig. 1K). Overall, these results indicate that SP1 plays an important role in inducing PABPC4 expression by binding to the PABPC4 promoter.
PABPC4 inhibits PEDV replication. Vero cells were transfected with PABPC4-FLAG for 24 h and infected with PEDV at an MOI of 0.01 to investigate whether PABPC4 inhibits PEDV replication. The cells and supernatants of the infected cell cultures were collected at various time points and assayed for levels of viral mRNA and N protein as well as viral titers. As shown in Fig. 2A, the levels of PEDV N protein were significantly lower in cells overexpressing PABPC4-FLAG than in cells transfected with the control vector. The mRNA of PEDV N in the cells overexpressing PABPC4 was also significantly lower than that in the cells transfected with an empty vector (Fig. 2B). As expected, the viral titers in the supernatants of the cells overexpressing PABPC4 were lower than those from cells transfected with the control vector ( Fig. 2C and D). To further verify these conclusions, siRNAs targeting PABPC4 were synthesized and transfected into Vero cells for 24 h, followed by PEDV infection. In cells transfected with mPABPC4 siRNA, the viral yield was significantly higher than in non-target-control-siRNA-transfected cells ( Fig. 2E to H). Moreover, PEDV proliferation in LLC-PK1 cells was not only inhibited by the overexpression of PABPC4 but also significantly higher than that in non-target-control-siRNA-transfected cells ( Fig. 2I and J). These results indicate that PABPC4 plays a role in the inhibition of PEDV replication.
PABPC4 interacts with PEDV N protein and degrades N protein by autophagy. To determine the molecular mechanisms by which PABPC4 inhibits PEDV replication, we cotransfected the plasmid encoding FLAG-PABPC4 and PEDV structural proteins (S1, S2, E, M, and N) into HEK 293T cells to identify the PEDV structural protein that interacts directly with PABPC4 by coimmunoprecipitation (co-IP) analysis. The results showed that PABPC4 efficiently coimmunoprecipitated with PEDV N protein (Fig. 3A). Furthermore, PEDV N protein could efficiently coimmunoprecipitate with the endogenous PABPC4 protein (Fig. 3B). A glutathione S-transferase (GST) pulldown assay was also used to verify the binding between PABPC4 and PEDV N proteins. The results showed that GST-fused PEDV N (GST-N) could bind to PABPC4, whereas GST alone did not ( Fig. 3C), indicating that PEDV N is bound directly to PABPC4. We then investigated the subcellular localization of PABPC4 and the PEDV N proteins with confocal microscopy. Immunofluorescence in HeLa cells transfected with plasmids encoding enhanced green fluorescent protein (EGFP)-PABPC4 (green fluorescence) and mCherry-PEDV-N (red fluorescence) for 24 h revealed that PABPC4 and N proteins colocalized in the cytoplasm (Fig. 3D). These results demonstrated that PABPC4 directly interacts with the PEDV N protein.
To analyze whether PABPC4 is involved in regulating the stability of N proteins, the plasmids encoding hemagglutinin (HA)-N protein and FLAG-PABPC4 were cotransfected into HEK 293T cells, and the N protein abundances were examined by Western blotting. The results showed that the levels of N proteins decreased in a dose-dependent manner in response to PABPC4 (Fig. 3E). These data demonstrated that PABPC4 is involved in regulating PEDV N protein levels.
There are two major intracellular protein degradation pathways in eukaryotic cells: the ubiquitin-proteasomal system and the autolysosome pathway. Plasmids encoding the PEDV N and PABPC4 proteins were cotransfected into HEK 293T cells to assess the degradation pathways of N protein by PABPC4, and the cells were treated with the proteasome inhibitor MG132 and the autophagy inhibitors 3-methyladenine (3MA),  chloroquine (CQ), or bafilomycin A 1 (Baf A1). Western blots showed that PABPC4-mediated degradation of N protein was blocked by the autophagy inhibitors 3MA, CQ, and Baf A1 but not by the proteasome inhibitor MG132 (Fig. 3F). Thus, the degradation of N protein might be mediated by the autophagy-lysosome pathway. These results indicated that PABPC4 interacts with the PEDV N protein and degrades N protein by autophagy.
PABPC4 degrades PEDV N protein through the PABPC4-MARCH8-NDP52 autophagosome pathway. During selective autophagy, the substrate proteins are first ubiquitinated by an E3 ubiquitin ligase and then recognized by cargo receptors. The cargo receptors deliver the substrates to ATG8 family proteins and form autophagosomes to degrade the substrates. Recent studies have shown that interferon-inducible antiviral factor, BST2, could recruit the E3 ubiquitin ligase MARCH8 to catalyze the ubiquitination of the PEDV N protein, and the cargo receptor NDP52 recognizes and delivers the ubiquitinated N protein to autolysosomes for degradation (34). In this study, we found that PABPC4 degrades the N protein of PEDV via autophagy. Co-IP assays were performed to examine whether PABPC4 mediates the ubiquitination of N protein by recruiting MARCH8 and NDP52. This involved cotransfecting HEK 293T cells with plasmids encoding FLAG-PABPC4 and myc-MARCH8, or myc-NDP52, followed by co-IP using anti-FLAG monoclonal antibody (MAb) affinity gel. Western blotting showed that anti-FLAG MAb affinity gel successfully pulled down FLAG-PABPC4/myc-MARCH8 or FLAG-PABP4/myc-NDP52 (Fig. 4A), demonstrating interactions between FLAG-PABPC4 and myc-MARCH8 as well as FLAG-PABP4 and myc-NDP52. MARCH8 and NDP52 also could coimmunoprecipitate with the endogenous PABPC4 protein ( Fig. 4B and C). The GST pulldown further validated the direct binding of PABPC4 to NDP52 (Fig. 4D). We also confirmed the colocalization of PABPC4 with MARCH8 and NDP52 by immunostaining (Fig. 4E). It has been reported that the N protein of PEDV binds to MARCH8 and NDP52 (34). Our data suggested that PABPC4 targets the PEDV N protein and MARCH8 toward autophagic degradation by allowing their binding to the cargo receptor NDP52.
To determine whether MARCH8 and NDP52 are involved in PABPC4-induced autophagic degradation of the N protein, we individually transfected HEK 293T cells with plasmids encoding HA-PEDV-N and FLAG-PABPC4, together with small interfering RNAs (NDP52, MARCH8, ATG5, or negative-control siRNA). At 24 h posttransfection, Western blotting was performed. We found that interfering with the expression of NDP52, MARCH8, or ATG5 effectively prevented the degradation of PABPC4-induced N protein (Fig. 4F). The results also showed that interfering with NDP52, MARCH8, or ATG5 protein expression increased PEDV replication in Vero cells (Fig. 4G). These results suggest that the N protein of PEDV is degraded by selective autophagy through the PABPC4-MARCH8-NDP52 autophagosome pathway.
PABPC4 inhibits coronavirus replication through degrading N proteins. PEDV belongs to the Coronaviridae family, which is divided into the following four major genera: Alphacoronavirus, Betacoronavirus, Gammacoronavirus, and Deltacoronavirus. In this study, we proved that PABPC4 inhibited PEDV replication by degrading N protein.
To test whether PABPC4 was involved in the regulation of CoV replication via N protein degradation, we cloned seven N genes of CoVs, including Alphacoronavirus (HCoV 229E), Betacoronavirus (SARS-CoV-2, SARS-CoV, Middle East respiratory syndrome   [MERS]-CoV, and murine hepatitis virus [MHV]), Gammacoronavirus (IBV), and Deltacoronavirus (PDCoV) into the mammalian expression vector pCAGGS containing the hemagglutinin (HA) tag. These N-expressing plasmids were cotransfected with a plasmid encoding FLAG-PABPC4 into HEK 293T cells for 24 h; this was then followed by co-IP. The results showed that PABPC4 protein directly interacted with each of the seven CoV N proteins (Fig. 5A). Co-IP revealed that the N proteins of different CoVs interacted with the FLAG-tagged PABPC4.
To analyze whether PABPC4 was involved in regulating the stability of N proteins, the plasmids encoding HA-tagged N proteins derived from different CoVs were cotransfected with FLAG-PABPC4 or vector control, and the N protein abundances were examined by Western blotting. The results showed that PABPC4 also downregulated the expression of the other six N proteins (Fig. 5B), and the levels of SARS-CoV-2 N protein decreased in a dose-dependent manner in response to PABPC4. These data demonstrated that PABPC4 is involved in regulating the levels of CoV N proteins.
Cells were transfected with PABPC4-FLAG for 24 h and infected with Betacoronavirus (MHV), Gammacoronavirus (IBV), or Deltacoronavirus (PDCoV) to investigate whether PABPC4 regulates the levels of N protein during viral infection. N2a cells were infected with MHV, H1299 cells were infected with IBV, and ST cells were infected with PDCoV. The cells and supernatants of the infected cell cultures were collected at various time points and assayed for levels of viral mRNA and N protein, as well as viral titers. As shown in Fig. 5C, G, and K, the levels of CoV N protein were significantly lower in cells overexpressing PABPC4-FLAG than in cells transfected with the control vector. CoV mRNAs in the supernatants of the cells overexpressing PABPC4 were also significantly lower than those in the cells transfected with empty vectors (Fig. 5D, H, and L). To confirm these results, we selected siRNAs targeting the different regions of PABPC4 (Fig. 5E, I, and M). Cells were transfected with PABPC4 siRNA and infected with MHV, IBV, or PDCoV at an MOI of 0.01, and the viral yield was measured by quantitative PCR (qPCR). As expected, the viral titers were significantly higher in cells transfected with the PABPC4 siRNA (Fig. 5F, J, and N). Those results suggest that PABPC4 downregulates the protein abundance of the CoV N proteins and inhibits MHV (Betacoronavirus), IBV (Gammacoronavirus), and PDCoV (Deltacoronavirus) replication.
PABPC4 degrades coronavirus N proteins by selective autophagy. PABPC4 degrades PEDV N protein through selective autophagy and decreases the N proteins of other Coronaviridae genera. So, we speculated that the PABPC4 inhibited the coronavirus replication by degrading N protein with the PABPC4-MARCH8-NDP52 autophagosome pathway. Plasmids encoding the N protein of SARS-CoV-2 were cotransfected into HEK 293T cells with PABPC4 to identify the system that predominantly mediates the degradation of coronavirus N proteins by PABPC4, and the cells were treated with the proteasome inhibitor MG132 or the autophagy inhibitor 3MA, CQ, or Baf A1. Western blotting showed that PABPC4-mediated degradation of SARS-CoV-2 N protein was blocked by the autophagy inhibitors 3MA, CQ, and Baf A1 but not by the proteasome inhibitor MG132 (Fig. 6A), which is consistent as PABPC4 degrades the PEDV N protein. To determine whether MARCH8 and NDP52 were involved in PABPC4-induced autophagic degradation of the N protein, we individually transfected HEK 293T cells with plasmids encoding HA-tagged N proteins of CoVs (SARS-CoV-2, HCOV 229E, MERS-CoV, SARS-CoV, MHV, IBV, and PDCoV) and FLAG-PABPC4, together with small interfering RNAs (NDP52, MARCH8, ATG5, or negative-control siRNA). At 24 h posttransfection, Western blotting was performed. We found that interfering with the expression of NDP52, MARCH8, or ATG5 effectively prevented the degradation of N proteins (Fig. 6B to H). These results showed that the PABPC4-MARCH8-NDP52 autophagosome pathway was broadly involved in CoV N protein degradation.

DISCUSSION
Coronaviruses are broadly distributed among humans, pigs, birds, and other mammals and can cause respiratory and enteric diseases (1). In the last 2 decades, coronaviruses have caused three large-scale pandemics, SARS, MERS, and SARS-CoV-2. Due to its strong transmission ability and high mortality rate, these three coronaviruses have spread rapidly worldwide, causing thousands of deaths (53). Emerging coronaviruses can cause severe diseases in humans and animals, but none of the currently available drugs or vaccines can effectively control these diseases. Therefore, the identification of effective antiviral agents to combat novel viruses is a matter of urgency. Host antiviral proteins play an important role in inhibiting viral proliferation. In this study, we revealed a novel functional property of PABPC4 in response to coronavirus infections. We found that the transcription factor of SP1 downregulated the expression of PABPC4 during PEDV infection. Our results further demonstrated that PABPC4 suppresses PEDV (Alphacoronavirus), MHV (Betacoronavirus), IBV (Gammacoronavirus), and PDCoV (Deltacoronavirus) replication by degrading the viral N protein through the PABPC4-MARCH8-NDP52 autophagosome pathway (Fig. 7) and speculated that PABPC4 inhibits coronavirus replication by targeting the viral N protein for NDP52mediated selective autophagic degradation.
PABPC4 binds to the poly(A) tail of most eukaryotic mRNAs. It is expressed as a Tcell activation-induced protein and is also known as activated platelet protein 1 (50). PABPC4 plays an important role in the expression of erythroid mRNAs, erythroid differentiation, and regulation of telomerase activity and cell growth (54,55). However, the antiviral function of PABPC4 and the regulation mechanism of PABPC4 during virus infection remains unexplored. In our study, we found that PABPC4 was significantly downregulated by PEDV infection. Many factors regulated gene expression. To confirm the transcription factors of PABPC4, we amplified the PABPC4 promoter sequence and found that the minimal PABPC4 core promoter was located between 2153 and 21. Furthermore, we analyzed the regulatory elements of PABPC4, and the results indicated that SP1 plays a direct role in upregulating PABPC4 expression by binding to the PABPC4 promoter. SP1 is a zinc finger transcription factor, upregulated in response to transforming growth factor b (TGF-b) and zinc (56,57). Upregulation of SP1 protein could serve as a useful strategy for inhibiting CoVs by promoting PABPC4 expression.
PABPC4 is one of the isoforms of cytoplasmic poly(A)-binding protein (PABP). PABP is an RNA-processing protein, which plays an important role in promoting gene expression by enhancing translation and mRNA stability (36,37,41). PABPC1, the best-characterized cytoplasmic PABP, could bind to the polyadenylated [poly(A)] tail of mRNAs and interacts with eukaryotic initiation factor 4G (eIF4G), which associates with the 59 cap structure, which formed an mRNA "closed loop," protecting mRNAs from degradation and promoting translation initiation by enhancing ribosome subunit recruitment (39,41). PABPC4 and PABPC1 belong to the PABP family and share 75% identity at the protein level (39,58). PABP4 is highly similar to PABP1, so PABPC4 and PABPC1 might have similar functions in regulating mRNA translation and stability. It is suggested that PABPC4 may promote viral replication, but in our study, we found that PABPC4 could inhibit the replication of PEDV, and the downregulation of the expression of PABPC4 renders cells more susceptible to PEDV replication. Furthermore, PABPC4 has been shown to combat the replication of coronaviruses. So, there should be other antiviral functions of PABPC4 to inhibit replicating a wide variety of coronaviruses. It was noted that the PEDV N protein plays an important role in virus replication and immune evasion and could be degraded by the restriction factors BST2 through the selective autophagy pathway (17,34). Interestingly, we found that PABPC4 targeted the N proteins of coronaviruses and degraded the N proteins by selective autophagy pathway. These data indicate that PABPC4 suppresses coronavirus replication, possibly targeting and degrading viral N protein with selective autophagy.
Autophagy is a cellular catabolic process involved in maintaining intracellular homeostasis through the degradation of cellular misfolded or long-lived proteins and damaged organelles (59). Selective autophagy captures its cargo via selective autophagy receptors to degrade the viruses and virus-derived antigens (60). It has been reported that BST2 could induce selective autophagy to degrade the PEDV N protein and reduce PEDV replication (34). Here, we demonstrated that PABPC4 could recruit E3 ubiquitin ligase MARCH8 to catalyze CoV N proteins, recognized and degraded by NDP52-dependent selective autophagy. These results indicated that the induction of autophagy is beneficial for host cell antiviral factor-mediated viral restriction mechanisms. During the current coronavirus disease-19 (COVID-19; caused by SARS-CoV-2) pandemic, CQ has been administered as a first-line drug to treat this disease in several studies (61,62), but the treatment efforts are controversial. Some studies showed that CQ did not prevent illness compatible with COVID-19 (63,64). Chloroquine could inhibit the glycosylation of the ACE2 receptor, which acts as an entry receptor for SARS-CoV-2 (65). As an autolysosome inhibitor, it could also interfere with lysosome-mediated autophagy function. However, the inhibition of autophagy results in damaged mitochondria not being cleared via mitophagy and, thus, accumulating in cells, leading to oxidative stress and renal tubular dysfunction (66). Inhibition of autophagy pathways also inhibits the antiviral function of antiviral factors, suppressing viral replication by autophagy, such as BST2 and PABPC4. So, clinicians should carefully weigh the risks and benefits of CQ as a treatment for COVID-19.
Considering the rapid global spread of SARS-CoV-2, the development of novel strategies for the treatment or prevention of SARS-CoV-2 and other CoVs is urgent. The broad inhibition of CoV replication by the host protein, PABPC4, through the autophagy pathway provides novel targets for anticoronavirus therapeutics. Upregulating the expression of the transcription factor SP1 might increase PABPC4 expression and inhibit CoV replication. Furthermore, pharmacological modulation of the PABPC4-MARCH8-NDP52 autophagosome pathway might serve as a practical strategy for treating CoV infections, especially for COVID-19. Overexpression of PABPC4 using mammalian expression vectors or viral expression systems might be developed as plausible gene therapy for the treatment of CoV infection, especially for coronaviral diseases in animals. precipitated proteins were analyzed by co-IP and Western blotting as previously described (34). Briefly, lysates were centrifuged and incubated with anti-FLAG or anti-HA antibody-bound Dynabeads Protein G (Life Technologies). The precipitated proteins were washed with 0.02% PBST and eluted from beads with elution buffer (50 mM glycine, pH 2.8) for Western blotting.
Luciferase reporter assay. HEK 293T cells were seeded into 24-well plates and transfected with plasmids (including Luciferase reporter plasmid and pRL-TK plasmid) using Lipofectamine 3000 (Invitrogen). Twenty-four hours later, cells were collected, and luciferase activity was measured using a Dual-Luciferase reporter assay system (Promega). Luciferase values were calculated by normalization of the firefly luciferase activity to Renilla luciferase activity.
GST pulldown. PEDV N was cloned into pCold GST plasmid (Clontech), and PABPC4 was cloned into pCold TF plasmid (Clontech). Recombinant proteins were expressed in BL21 cells (Vazyme), following which cells were lysed using the lysis buffer provided in the GST protein interaction pulldown kit (Thermo) and incubated on ice for 30 min. After centrifugation, equal amounts of GST and GST-N supernatants were mixed with PABPC4. GST pulldown assays were performed using the GST protein interaction pulldown kit according to the manufacturer's protocol. The proteins were eluted with reduced glutathione and analyzed by Western blotting.
Immunofluorescence. Cells were cultured on coverslips and transfected with specific expression plasmids, then fixed and permeabilized with 4% paraformaldehyde (Sigma-Aldrich) for 15 min, and then washed three times with PBS followed by 0.1% Triton X-100 (Sigma-Aldrich) for 10 min. After washing, the cells were blocked using 5% bovine serum albumin for 1 h at 37°C and then incubated with primary antibody (anti-Myc) diluted in PBS containing 5% BSA for 1 h at 37°C. The cells were then washed three times and incubated with a fluorescently labeled secondary antibody (Alexa-Fluor 568-conjugated donkey anti-mouse IgG antibody; Thermo Fisher Scientific; catalog no.A10037) for 1 h at 37°C in the dark. After the cells were washed three times with PBS, they were stained with 49,6-diamidino-2-phenylindole (DAPI) (Beyotime Biotechnology; catalog no. C1002) at room temperature for 5 min. The cells were observed under a confocal immunofluorescence microscope (Carl Zeiss, Oberkochen, Germany).
Statistical analysis. All results are representative of three independent experiments. Statistical analysis was performed using GraphPad Prism 5 software. All data are expressed as mean 6 standard deviation (SD). P values were calculated using a two-tailed Student's t test or one-way analysis of variance (ANOVA).