A/(H1N1) pdm09 NS1 promotes viral replication by enhancing autophagy through hijacking the IAV negative regulatory factor LRPPRC

ABSTRACT

The quadrilateral reassortant IAV A/(H1N1) pdm09 is the pathogen responsible for the first influenza pandemic of the 21st century. The virus spread rapidly among hosts causing high mortality within human population. Efficient accumulation of virions is known to be important for the rapid transmission of virus. However, the mechanism by which A/(H1N1) pdm09 promotes its rapid replication has not been fully studied. Here, we found the NS1 of A/(H1N1) pdm09 mediated complete macroautophagy/autophagy, and then facilitated self-replication, which may be associated with the more rapid spread of this virus compared with H1N1_WSN and H3N8_JL89. We found that the promotion of self-replication could be mainly attributed to NS1_pdm09 strongly antagonizing the inhibitory effect of LRPPRC on autophagy. The interaction between NS1_pdm09 and LRPPRC competitively blocked the interaction of LRPPRC with BECN1/Beclin1, resulting in increased recruitment of BECN1 for PIK3C3 (phosphatidylinositol 3-kinase catalytic subunit type 3) and induction of the initiation of autophagy. In conclusion, we uncover the unique molecular mechanism by which A/(H1N1) pdm09 utilizes autophagy to promote self-replication, and we provide theoretical basics for the analysis of the etiological characteristics of the A/(H1N1) pdm09 pandemic and the development of anti-influenza drugs and vaccines.

Abbreviations: 293T: human embryonic kidney 293 cells; 293T_LRPPRC: stable LRPPRC expression 293T cells; 3-MA: 3-methyladenine; A549 cells: human non-small cell lung cancer cells; AA: amino acid; ACTB: actin beta; BECN1: beclin 1; BECN1 KO: BECN1 knockout 293T cells; Cal: calyculin A; Co-IP: co-immunoprecipitation; CQ: chloroquine; DC: dendritic cell; Eug: eugenol; GFP: green fluorescent protein; HA: hemagglutinin; HIV: human immunodeficiency virus; IAVs: Influenza A viruses; IFN: interferon; JL89: A/equine/Jilin/1/1989 (H3N8); LAMP2: lysosomal associated membrane protein 2; LRPPRC: leucine rich pentatriicopeptide repeat containing; LRPPRC KO: LRPPRC knockout 293T cells; M2: matrix 2; MAP1LC3B/LC3B: microtubule associated protein 1 light chain 3 beta; MDCK: Madin-Darby canine kidney cells; MOI: multiplicity of infection; MS: mass spectrometry; NP: nucleoprotein; NS1: non-structural protein 1; NS1_JL89: non-structural protein 1 of A/equine/Jilin/1/1989 (H3N8); NS1_pdm09: non-structural protein 1 of A/(H1N1) pdm09; NS1_SC09: non-structural protein 1 of A/Sichuan/2009 (H1N1); NS1_WSN: non-structural protein 1 of A/WSN/1933 (H1N1); PB1: polymerase basic protein 1; PB1-F2: alternate reading frame discovered in PB1 gene segment; PIK3C3: phosphatidylinositol 3-kinase catalytic subunit type 3; PR8: A/PR/8/34 (H1N1); Rapa: rapamycin; RFP: red fluorescent protein; SC09: A/Sichuan/2009 (H1N1); SQSTM1/p62: sequestosome 1; STK4/MST1: serine/threonine kinase 4; TEM: transmission electron microscopy; TOMM20: translocase of outer mitochondrial membrane 20; WHO: World Health Organization; WSN: A/WSN/1933 (H1N1); WSN-NS1_JL89: WSN recombinant strain in which NS1 was replaced with that of JL89; WSN-NS1_SC09: WSN recombinant strain in which NS1 was replaced with that of SC09.

Introduction

Influenza A viruses (IAVs) are enveloped, negative-strand RNA viruses causing epidemics and pandemics with relatively high human morbidity and mortality [1,2]. In April 2009, a new quadrilateral reassortant influenza A virus A/(H1N1) pdm09 emerged in Mexico and USA, causing the first influenza pandemic of the 21st century [3,4]. Globally, 135 countries were involved, and the World Health Organization (WHO) reported a total of 491,382 laboratory-confirmed cases and 18,449 deaths between April 2009 and August 2010, when the pandemic was declared over [5]. Some studies speculated the actual death toll may have been between 123,000 and 203,000 because of the deaths caused by secondary bacterial infection and exacerbation of preexisting chronic diseases [6]. A/(H1N1) pdm09 spread rapidly between hosts and resulted in high mortality. Most of the studies on A/(H1N1) pdm09 were based on molecular epidemiological investigations of clinical cases [7–9]. The comparison of A/(H1N1) pdm09 replication with that of other IAVs in host cells, as well as its effect on host cell homeostasis, is still lacking.

Previous study reported that A/(H1N1) pdm09 can enhance macroautophagy/autophagy in infected dendritic cells (DC) [10]. However, the effect of autophagy on the replication of A/(H1N1) pdm09 and the specific mechanism by which this occurs remain to be studied. Autophagy is known as a physiological process by which host organisms maintain homeostasis and capture or eliminate pathogens [11–14]. During viral infection, the host innate immune system regulates autophagy and performs selective degradation specifically of pathogen proteins or host factors necessary for the viral life cycle [11,15]. However, certain viruses have evolved different strategies to counter and even utilize autophagy to promote self-replication [11].

IAV is one of the viruses able to harness host autophagy and promote self-replication in a virus-specific manner [16–19]. It has been reported that the alternate reading frame discovered in the polymerase basic protein 1 (PB1) gene segment (PB1-F2), hemagglutinin (HA), matrix 2 (M2), non-structural protein 1 (NS1), and nucleoprotein (NP) can regulate host cell autophagy through distinctive pathways [19–23]. Among them, NS1 of IAVs is known as inhibitor of antiviral immunity and also an indirect regulator of autophagy [24,25]. The NS1 of A/PR/8/34 (H1N1) (PR8) indirectly regulates autophagy by enhancing the expression of HA and M2, and inhibiting the formation of MAPK8/JNK1-dependent autophagosomes mediated by RAB11A-recycled endosomes [25,26]. Besides, the NS1 of the H5N1 subtype IAV (A/mallard/Huadong/S/2005 H5N1) activates MAPK/JNK, which induces autophagy and aggravates the pathogenicity of this strain [26]. Therefore, there may be multiple virus-specific mechanisms by which NS1 regulates autophagy.

LRPPRC (leucine rich pentitricopeptide repeat containing) is mainly found in the mitochondria [27]. It is involved in embryo development, mitochondrial function in vivo, negative regulation of autophagy or mitophagy and the occurrence, development, and, as an oncogene, the outcome, of tumors [27–33]. LRPPRC can inhibit the initiation of autophagy and mitophagy through enhancing the stability of BCL2 by interacting with PRKN (parkin RBR E3 ubiquitin protein ligase) [34–36]. LRPPRC is known to affect the replication of certain viruses, such as ZIKA virus, hepatitis C virus (HCV), and human immunodeficiency virus (HIV) [27,37]. However, whether LRPPRC affects the replication of IAVs is still unknown.

This study aimed to explore the reasons for the rapid global spread of A/(H1N1) pdm09 through the investigation of viral replication ability and its mechanism. We demonstrated that A/(H1N1) pdm09 regulated macroautophagy/autophagy of host cells to facilitate self-replication in a unique manner. Unlike the NS1 of the other two studied IAVs, NS1 of A/(H1N1) pdm09 interacted with LRPPRC and competitively blocked the BECN1/Beclin1-BCL2 heterodimer that also interacted with LRPPRC. With the degradation of BCL2 from the BECN1-BCL2 heterodimer, the interaction between BECN1 and PIK3C3 was significantly enhanced, which in turn mediated BECN1-dependent autophagy. This process counteracted the suppression of IAV replication by LRPPRC.

Results

The NS1 of A/Sichuan/2009(H1N1) (SC09) promotes viral replication by enhancing autophagy

To assess the replication ability of A/(H1N1) pdm09 in vitro, A/Sichuan/2009 (H1N1), a representative strain of A/(H1N1) pdm09, a lab adapted H1N1 IAV A/WSN/1933 (H1N1) (WSN), and an equine IAV A/equine/Jilin/1/1989 (H3N8) (JL89) were selected to infect human non-small cell lung cancer cells (A549 cells) and human embryonic kidney 293 cells (293T cells) at different multiplicities of infection (MOIs), (multi-cycle: MOI=0.01, one-cycle: MOI=5.0). WSN is known as a pandemic IAV strain with an epidemic trend in humans similar to that of A/(H1N1) pdm09, while JL89 only causes localized epidemics in species of Equus [38]. The results showed that the SC09 strain displayed a higher growth curve than either the WSN or the JL89 strain in both A549 cells and 293T cells (Figure 1A and Fig. S1A-C). It has been reported that A/(H1N1) pdm09 can induce autophagy in infected cells, as well as autophagy promotes IAV replication in a virus-specific manner [10,16–18,39]. Therefore, through analysis of the differences in MAP1LC3B/LC3B (microtubule associated protein 1 light chain 3 beta) transformation, we found that the SC09 strain seemed to induce higher levels of autophagy than did either the WSN or the JL89 strain (Figure 1B and Fig. S1D–F). To further clarify this phenomenon, the formation of autophagosomes induced by these three strains was observed using transmission electron microscopy (TEM). The TEM results showed that A549 cells infected with the SC09 strain formed a large number of autophagosomes, while the number of autophagosomes induced by either the WSN or the JL89 strain was obviously less than that induced by the SC09 strain (Figure 1C). These results suggested a positive correlation between the higher levels of autophagy induced by the SC09 strain and its rapid replication.

Figure 1.

NS1_SC09 promotes virus replication by enhancing autophagy. (A and B) A549 cells were infected with SC09, WSN, or JL89 strains at an MOI of 5.0. The supernatants were sampled at 0, 12, 24, 36, and 48 h post-infection, and the viral titers were determined with endpoint titration in Madin-Darby canine kidney cells (MDCK cells). The cell lysates were prepared at 24 h post-infection and analyzed using western blotting. (C) A549 cells were infected with SC09, WSN, or JL89 strain at the MOI of 0.01. The formation of autophagosomes was observed with TEM. (D) 293T cells were transfected with empty vector, NS1_SC09-Flag, NS1_WSN-Flag, or NS1_JL89-Flag. Cell lysates were harvested 24 h post-transfection and analyzed using western blotting. (E) A549 cells were transfected with empty vector, NS1_SC09-Flag, NS1_WSN-Flag, or NS1_JL89-Flag. The formation of autophagosomes was observed using TEM. (F) HeLa cells were co-transfected with empty vector, NS1_SC09-Flag, NS1_WSN-Flag, or NS1_JL89-Flag and GFP-LC3B. The formation of GFP puncta and their colocalization with NS1 were analyzed. Treatment of rapamycin (20 μM) can induce autophagy as a control. Scale bar: 10 μm. It was representative of 20 cells. (G and H) A549 cells were infected with WSN-NS1_SC09, WSN-NS1_JL89, or WSN strain at an MOI of 5.0. The supernatants were sampled at 0, 12, 24, 36, and 48 h post-infection, and the viral titers were determined with endpoint titration in MDCK cells. The cell lysates were prepared at 24 h post-infection and analyzed with western blotting. (I and J) A549 cells were infected with WSN-NS1_SC09, WSN-NS1_JL89, or WSN strain at an MOI of 5.0, in the presence or absence of rapamycin (20 μM) or 3-MA (5 mM). The supernatants were sampled at 24 h post-infection, and the viral titers were determined with endpoint titration in MDCK cells. The cell lysates were prepared at 24 h post-infection and analyzed with western blotting. Error bars indicate the SD from three independent experiments. *, P < 0.05; **, P < 0.01.

Previous studies reported that HA, M2, NP, and PB1-F2 from IAVs each have specific mechanisms to induce autophagy or mitophagy to promote viral replication [19–22]. To investigate why SC09 induced higher levels of autophagy than the other two strains, 293T cells were transfected with the main structural or non-structural proteins from these three strains (Fig. S2A-I). We found that NP, M2, and HA of these three strains all independently induced an increase in LC3B-II, but there were no obvious differences among them (Fig. S2A-C). Interestingly, although a previous study reported that PB1-F2 of IAVs (A/duck/Hubei/Hangmei01/2006 (H5N1) [H5N1/HM] and PR8) results in mitophagy [22], we did not observed strong induction of LC3B-II by the PB1-F2s of selected stains in our study, indicating a strain specific effect of PB1-F2.

Unexpectedly, we found that the transformation of LC3B induced by NS1 of SC09 (NS1_SC09) increased more than that induced by either the NS1 of WSN (NS1_WSN) or the NS1 of JL89 (NS1_JL89) (Figure 1D). We then conducted TEM analysis of the differences in autophagosome formation of A549 cells transfected with NS1_WSN, NS1_SC09, or NS1_JL89. We found that the number of autophagosomes formed in A549 cells induced by NS1_SC09 was obviously greater than that induced by either NS1_WSN or NS1_JL89 (Figure 1E). Meanwhile, we also found that NS1_SC09 could induce more obvious aggregation of GFP-LC3B in HeLa cells than either NS1_WSN or NS1_JL89 (Figure 1F). These findings suggested that the NS1_SC09 may own special characters to promote autophagy. To characterize the NS1 of these three selected strains, we constructed an unrooted phylogenetic tree using their gene sequences and demonstrated that all these three strains were contained and distincted in their respective branches and evolutionary relationships (Fig. S2J).

In order to eliminate the influence of viral proteins other than NS1, we utilized the reverse genetic system to rescue recombinant WSN strains, in which NS1_WSN was replaced with NS1_SC09 (WSN-NS1_SC09) or NS1_JL89 (WSN-NS1_JL89). We found that the recombinant WSN-NS1_SC09 strain had a more rapid growth curve and induced more transformation of LC3B than either of the other two strains in both A549 and 293T cells at MOI of 0.01 and 5.0 (Figure 1G,H and Fig. S3A-F). Meanwhile treatment with autophagy inhibitor (3-MA) was able to reduce the replication of WSN-NS1_SC09 to a level comparable to WSN (Figure 1I,J and Fig. S3G-L), while treatment with autophagy activator (rapamycin) could also increase the replication of WSN to a level comparable to WSN-NS1_SC09 (Figure 1I,J and Fig. S3G-L). This further demonstrated that NS1_SC09 promoted the replication of the SC09 strain in an autophagy-dependent manner.

Considering that the most important biological activity of IAVs NS1 is to inhibit the innate immune response, we further analyzed the differences in the transcription of IFNB1 mRNA in A549 cells infected with WSN, WSN-NS1_SC09, or WSN-NS1_JL89. We found that the transcription of IFNB1 mRNA in A549 cells was significantly enhanced after infection with any of these three strains, but their increases were comparable (Fig. S4A). We next infected the type I IFN-defective Vero E6 cells [40] with WSN, WSN-NS1_SC09, or WSN-NS1_JL89 (MOI=0.01 and 5.0), and found that WSN-NS1_SC09 could induce higher levels of autophagy and replicate more rapidly than either the WSN or the WSN-NS1_JL89 at 36 h post-infection (Fig. S4B-E). These findings suggested that the regulation of viral replication by promotion of autophagy through NS1_SC09 was independent of its regulation of IFN expression.

Together, these results confirmed that NS1_SC09 enhanced autophagy in infected cells and promoted the replication of SC09.

NS1_SC09 induces complete autophagy

To investigate the characteristics of the autophagy induced by NS1_SC09, we examined the degradation of the autophagosome marker SQSTM1/p62 (sequestosome 1) induced by different NS1s. As a protein-polymer, SQSTM1 is bound to the autophagosome membranes and degraded by lysosomes [41,42]. We observed that NS1_SC09 induced higher levels of LC3B-II accumulation and degradation of SQSTM1 in 293T cells than did NS1_WSN (Figure 2 A and B), indicating that the autophagy induced by NS1_SC09 may be complete autophagy. To further confirm this, HeLa cells were transfected with mRFP-GFP-LC3B together with NS1_SC09, NS1_WSN, NS1_JL89, or M2 of PR8 (M2_PR8). M2_PR8, which induces incomplete autophagy, was selected as a control [20,21]. Immunofluorescence analysis revealed that NS1_SC09, but neither NS1_WSN nor NS1_JL89, was able to induce not only aggregation of RFP-GFP-LC3B, but also quenching of GFP fluorescence, providing further evidence that NS1_SC09 induced complete autophagy (Figure 2 C and D). Since M2_PR8 inhibited the fusion of autophagosomes with lysosome membranes [20], it induced the aggregation of mRFP-GFP-LC3B, but the GFP fluorescence was not quenched, resulting colocalization of GFP and RFP (Figure 2 C and D).

Figure 2.

NS1_SC09 mediates complete autophagy. (A and B) 293T cells were transfected with empty vector, NS1_SC09-HA, or NS1_WSN-HA. Cell lysates were prepared 24 h post-transfection and analyzed with western blotting. (C) HeLa cells were co-transfected with mRFP-GFP-LC3B and empty vector, NS1_SC09-Flag, NS1_JL89-Flag, NS1_WSN-Flag, or M2_PR8-Flag (as a control to induce incomplete autophagy). Cells were analyzed to detect the formation of autophagosomes. Fluorescence signals indicated the expression of RFP and GFP (yellow color: incomplete autophagy, red color: complete autophagy). Scale bar: 10 μm. It was representative of 20 cells. (D) The graph shows the quantification of mRFP+GFP+ autophagosomes and mRFP+GFP- autolysosomes by taking the average number of dots in 20 cells (n=average number of dots in 20 cells). (E and F) 293T cells were transfected with the indicated plasmids above for 8 h and then either treated with CQ (20 μM) or not. 12 h later, cell lysates were analyzed with western blotting. Error bars indicate the SD from three independent experiments. *, P < 0.05; **, P < 0.01.

We next investigated the effects of NS1_SC09 on the autophagosome. 293T cells were transfected with empty vector, NS1_SC09, or M2_PR8 in the presence or absence of chloroquine (CQ), which is an inhibitor that suppresses the lysosome hydrolysis and blocks the fusion of autophagosomes with lysosomes [43]. We observed that higher levels of LC3B-II accumulated in NS1_SC09 transfected cells with CQ treatment than in those without CQ treatment, indicating that transfection with NS1_SC09 led to the degradation of autolysosomes (Figure 2 E and F). Moreover, the aggregation of GFP-LC3B induced by NS1_SC09 in HeLa cells co-located with LAMP2 (lysosomal associated membrane protein 2) (Fig. S5). These findings indicated that NS1_SC09 induced complete autophagy.

NS1_SC09 promotes autophagy through the interaction with LRPPRC

In order to explore whether NS1_SC09 directly interact host proteins to induce autophagy, 293T cells were transfected with Flag-tagged NS1_SC09, NS1_JL89, or empty vector, and then Flag-pulldown tandem mass spectrometry was conducted to show any potential interactions between NS1_SC09 and host proteins. We discovered 246 host proteins that form potential interactions with NS1_SC09 but not with NS1_JL89 (Figure 3A and Table 1). Among them, LRPPRC which is a negative regulator of autophagy attracted our attention (Table 1) [27,35]. Co-immunoprecipitation (co-IP) assay further confirmed that NS1_SC09 has a strong interaction with LRPPRC (Figure 3 B and D), while the interaction between NS1_WSN and LRPPRC was weak and no interaction between NS1_JL89 and LRPPRC was observed (Figure 3 B and D). As LRPPRC is the mitochondrially-encoded subunit of complex IV [27,29,31], we used immunofluorescence to analyze the colocalization of NS1 and LRPPRC in the mitochondria. Consistent with previous research, LRPPRC was found to colocalize with TOMM20 (translocase of outer mitochondrial membrane 20) (Figure 3C). Furthermore, although NS1_SC09 was mainly located in the nucleus, it still obviously colocalized with LRPPRC in the mitochondria (Figure 3C). However, neither NS1_WSN nor NS1_JL89 colocalized significantly with LRPPRC (Figure 3C).

Figure 3.

NS1_SC09 interacts with LRPPRC. (A) 293T cells were transfected with NS1_SC09-Flag, NS1_JL89-Flag, or empty vector. Cell lysates were pulled down using the antibodies against Flag tag. Proteins were then separated using SDS-PAGE and stained with silver, after which they were identified using LC/MS. (B) 293T cells were transfected with the indicated plasmids above. Cell lysates were subjected to HA-IP. (C) HeLa cells were co-transfected with LRPPRC-Flag and NS1_SC09-HA, NS1_JL89-HA, or NS1_WSN-HA. Twenty-four h later, cells were assessed for the localization of NS1, LRPPRC, and TOMM20. Cyan color indicated the colocalization of LRPPRC and TOMM20 and white color indicated the colocalization of LRPPRC, NS1, and TOMM20. Scale bar: 5 μm. It was representative of 20 cells. (D) 293T cells were transfected with the indicated plasmids above. Cell lysates were subjected to Flag-IP.

Table 1.

Summary annotated proteins identified to interact with NS1_SC09-flag but not NS1JL89-flag and Vec-flag.

Accession	Description	ΣCoverage
P42704	LRPPRC	30.63
	(leucine rich pentatriicopeptide repeat containing)
Q5JVF3-3	PCID2	36.7
	(PCI domain containing 2)
Q9Y6Y0	IVNS1ABP(influenza virus NS1A binding protein)	25.08

To investigate the key domain responsible for the interaction between NS1_SC09 and LRPPRC, we compared the amino acid sequences of NS1_SC09 and NS1_JL89, and constructed a series of allelic replacement chimeric vectors of them (Figure 4A left). Co-IP experiments revealed that the 73–125 residues of NS1_SC09 were necessary for its interaction with LRPPRC (Figure 4A right). In our subsequent studies, we abstracted that the residues at 104–125 of NS1_SC09 formed the key domain necessary for the interaction of NS1_SC09 with LRPPRC (Figure 4B). Comparison of the amino acid sequences of residues 104–125 in NS1_SC09 and NS1_JI89 revealed a total of five amino acid differences. Then, site-directed mutagenesis was conducted for these five amino acid sites in NS1_SC09 and NS1_JL89 (Fig. S6A left). However, mutations at any of these five amino acids of NS1_SC09 did not result in loss of the ability to interact with LRPPRC (Fig. S6A right). Also, any single amino acid mutation of NS1_JL89 was unable to confer on the protein the ability to interact with LRPPRC (Fig. S6A right). Interestingly, we observed that the NS1_SC09^P114G slightly reduced binding of LRPPRC (Fig. S6A right). Our findings indicated that interaction between NS1_SC09 and LRPPRC may be “linearly dependent”. Subsequently, we found that these five amino acid sites were highly conserved in different A/(H1N1) pdm09 strains, indicating a unique evolution pattern of the strains. Meanwhile, based on previous reports [44], we divided LRPPRC into TI-T4 segments (Figure 4C left). Co-IP experiments demonstrated that the T3 domain of LRPPRC was the key domain for interaction with NS1_SC09 (Figure 4C right).

Figure 4.

The key domain responsible for the interaction between NS1_SC09 and LRPPRC. (A) 293T cells were transfected with the indicated plasmids above. Cell lysates were subjected to Flag-IP. (B) 293T cells were transfected with the indicated plasmids above. Cell lysates were subjected to HA-IP. (C) 293T cells were transfected with the indicated plasmids above. Cell lysates were subjected to Flag-IP.

We observed positive correlation between autophagy induction and the binding ability to LRPPRC of the NS1 mutations (Fig. S6B-D). Although the NS1_SC09 ^P114G mutant could also induce an increase in LC3B transformation, this induction was weaker than that by NS1_SC09 (Fig. S6D). Therefore, we concluded that NS1_SC09 enhanced autophagy through its interaction with LRPPRC.

LRPPRC is a key factor to block IAV replication via the autophagy pathway, and is antagonized by NS1_SC09

To explore the influence of LRPPRC on the replication of IAVs, we constructed LRPPRC stable expression cell lines (293T_LRPPRC) and LRPPRC knockout cell lines (LRPPRC KO) derived from 293T cells (Fig. S7A). Neither overexpression nor knockout of LRPPRC affected the viability of the cells (Fig. S7B). By analyzing the growth curve of the WSN strain in 293T_LRPPRC, LRPPRC KO, and 293T cells, we observed that, compared with 293T cells, enhancing the expression of LRPPRC inhibited the replication of WSN, while knockout of LRPPRC promoted replication of WSN at different MOI (Figure 5A,B and Fig. S7C,D). Eliminating the expression of LRPPRC significantly enhanced the transformation of LC3B induced by WSN compared with that in 293T cells, while overexpression of LRPPRC obviously inhibited the increase in the levels of LC3B-II post-infection (Figure 5B and S7D). These findings suggested that LRPPRC may negatively regulate the replication of IAVs through the inhibition of autophagy. Additionally, the replication of the WSN strain in LRPPRC KO cells and rapamycin-treated 293T cells was comparable, yet significantly higher than that in untreated 293T cells (Figure 5C,D and Fig. S7E,F). In contrast, both the overexpression of LRPPRC and treatment with 3-MA significantly inhibited the replication of the WSN strain in 293T cells to a similar level. (Figure 5C,D and Fig. S7E,F). To further investigate the autophagy, a mRFP-GFP-LC3B dual fluorescent reporter system was used in these three cell lines during infection with the WSN strain (Figure 5E). We observed significant aggregation of mGFP-RFP-LC3B and quenching of GFP fluorescence in LRPPRC KO cells compared with that in 293T cells (Figure 5 E and F). Conversely, the aggregation of mRFP-GFP-LC3B barely observed in 293T_LRPPRC cells compared with that in 293T cells (Figure 5 E and F). These results strongly suggested that LRPPRC inhibited the replication of WSN via an autophagy-dependent mechanism.

Figure 5.

LRPPRC is a key factor that regulates the replication of IAVs through the autophagy pathway and is antagonized by NS1_SC09. (A and B) 293T cells, 293T_LRPPRC cells, and LPRRPC KO cells were infected with WSN at the MOI of 0.01. The supernatants were sampled at 0, 12, 24, and 48 h post-infection, and the virus titers were determined using endpoint titration in MDCK cells. Cell lysates at 0, 12, 24, and 48 h post-infection were analyzed using western blotting. (C and D) 293T cells, 293T_LRPPRC cells, and LPRRPC KO cells were infected with WSN at the MOI of 5.0, in the presence or absence of rapamycin (20 μm) or 3-MA (5 mM). The supernatants were sampled at 24 h post-infection, and the virus titers were determined using endpoint titration in MDCK cells. The cell lysates were analyzed using western blotting. (E) 293T cells, 293T_LRPPRC cells, and LPRRPC KO cells were transfected with mRFP-GFP-LC3B and then infected with WSN at the MOI of 0.1. Cells were analyzed to assess the formation of autophagosomes. Fluorescence signals indicated the expression of RFP and GFP (yellow color: incomplete autophagy, red color: complete autophagy). Scale bar: 10 μm. It was representative of 20 cells. (F) The graph shows the quantification of mRFP+GFP+ autophagosomes and mRFP+GFP- autolysosomes by taking the average number of dots in 20 cells (n=average number of dots in 20 cells). (G and H) 293T cells and LPRRPC KO cells were infected with WSN-NS1_SC09 or WSN at the MOI of 0.1. The supernatants were sampled at 24 h post-infection, and the virus titers were determined using endpoint titration in MDCK cells. Cell lysates were analyzed using western blotting. (I) 293T cells and LPRRPC KO cells were transfected with NS1_SC09-HA or NS1_WSN-HA. 24 h post-transfection, cell lysates were analyzed using western blotting. Error bars indicate the SD from three independent experiments. *, P < 0.05; **, P < 0.01.

Next, we attempted to explore whether the autophagy induced by NS1_SC09 resulted from antagonism of the negative regulation of autophagy by LRPPRC. To this end, LRPPRC KO cells were infected with either WSN-NS1_SC09 or WSN. The replication of these two strains, as well as the transformation of LC3B induced by them, was found to be similar (Figure 5 G and H). Furthermore, similar levels of LC3B-II were observed when the LRPPRC KO cells were transfected with either NS1_SC09 or NS1_WSN (Figure 5I). These findings further confirmed that, without LRPPRC, NS1_SC09 was no longer able to promote self-replication in an autophagy-dependent manner.

The interaction between NS1_SC09 and LRPPRC results in initiation of BECN1-dependent autophagy

To reveal why NS1_SC09 is able to induce autophagy through interaction with LRPPRC, we first attempted to investigate whether infection with WSN-NS1_SC09 mediated the degradation of LRPPRC. Unfortunately, although the levels of autophagy- and lysosomal-associated proteins significantly decreased in WSN-NS1_SC09-infected A549 cells, there was no obvious change in the levels of LRPPRC compared with those in WSN-infected A549 cells (Fig. S8A). In view of LRPPRC is reported to interact with the BECN1-BCL2 heterodimer, forming a stable complex in the mitochondria, thereby inhibiting BECN1-dependent autophagy [35,44,45], we speculated that NS1_SC09 may interrupt the interactions among them, leading to a decrease in the stability of the complex. To prove this, co-IP was used to reveal the influence of NS1 on the stability of LRPPRC-BECN1-BCL2. Consistent with previous studies, we found that LRPPRC interacted with BECN1 and BCL2, and BECN1 also interacted with BCL2 (Figure 6A-C). When NS1_SC09 was present, the interactions of LRPPRC, BECN1, and BCL2 were significantly reduced. NS1_WSN, which interacted weakly with LRPPRC, also weakened the interactions of LRPPRC, BECN1, and BCL2, while the presence of NS1_JL89 did not affect the interactions among them at all (Figure 6A-C).

Figure 6.

The interaction between NS1_SC09 and LRPPRC results in initiation of BECN1-dependent autophagy. (A) 293T cells were transfected with the indicated plasmids above. Cell lysates were subjected to Flag-IP. (B) 293T cells were transfected with the indicated plasmids above. Cell lysates were subjected to Flag-IP. (C) 293T cells were transfected with the indicated plasmids above. Cell lysates were subjected to Flag-IP. (D) 293T cells were transfected with empty vector, NS1_SC09-HA, NS1_JL89-HA, or NS1_WSN-HA. Cell lysates were prepared and immunoprecipitated with the anti-BECN1 antibody or control IgG. (E) The indicated plasmids above were co-transfected into 293T cells. Cell lysates were subjected to Flag-IP. (F) 293T cells were transfected with empty vector, NS1_SC09-HA, NS1_JL89-HA, or NS1_WSN-HA. Cell lysates were prepared and immunoprecipitated with the anti-PIK3C3 antibody or control IgG. (G) The indicated plasmids above were co-transfected into 293T cells. Cell lysates were subjected to Flag-IP.

To eliminate the influence of exogenous transfection in this study, the interactions of endogenous BECN1 with LRPPRC and BCL2 were studied in the presence of NS1_SC09, NS1_WSN, or NS1_JL89. The interaction of endogenous BECN1 with either LRPPRC or BCL2 was also obviously attenuated after transfection with NS1_SC09 (Figure 6D). Consistent with the results from a previous study, we verified that the T3 domain of LRPPRC, which was the key domain responsible for the interaction with NS1_SC09, also determined the interaction with BECN1 (Figure 4C and S8B). Therefore, we speculated that NS1_SC09 and BECN1 interact competitively with LRPPRC, disturbing the downstream autophagy pathway. To investigate this, different dosages of NS1_SC09 (0.5 μg, 1.0 μg, and 2.0 μg) were transfected into 293T cells and the interactions of LRPPRC, BECN1, and BCL2 were evaluated. As predicted, the interactions of LRPPRC, BECN1, and BCL2 gradually decreased with increased concentrations of NS1_SC09 (Figure 6 E and G). Interestingly, NS1_SC09 affected not only the binding of LRPPRC with the BECN1-BCL2 heterodimer but also the interaction between BECN1 and BCL2 (Figure 6A-D and G). Meanwhile, in the presence of NS1_SC09, the expression of both exogenous and endogenous BCL2 in total cell lysates seemed to be reduced (Figure 6). These findings further suggested that NS1_SC09 was able to disrupt the stability of the LRPPRC-BECN1-BCL2 complex by blocking the binding between LRPPRC and the BECN1-BCL2 heterodimer.

The binding between BECN1 (dissociated from BCL2) and PIK3C3 induces the initiation of autophagy [35,45,46]. Therefore, we investigated whether the interaction between PIK3C3 and BECN1 could be enhanced by NS1_SC09. As expected, the interaction between endogenous PIK3C3 and BECN1 significantly improved in the presence of NS1_SC09 (Figure 6 D and F). Moreover, the interaction between BECN1 and PIK3C3 was enhanced in a dose-dependent manner by NS1_SC09 (0.5 μg, 1.0 μg, and 2.0 μg) (Figure 6G). Although the binding between Becin1 and PIK3C3 also slightly improved in the presence of NS1_WSN (Figure 6 D and F), we speculated that this slight increase in the binding of BECN1 and PI3KCII may not be sufficient to initiate autophagy. We observed that a low dosage expression of NS1_SC09 with similar ability to NS1_WSN in interrupting the LRPPRC-BECN1-BCL2 complex could not induce autophagy (Figure 6E), indicating that NS1_WSN has weak interaction with LRPPRC and resulting faint destruction of the stability among LRPPRC-BECN1-BCL2.

To further elucidate the importance of BECN1 in the NS1_SC09-induced autophagy pathway, WSN-NS1_SC09, WSN, and WSN-NS1_JL89 strains were used to infect 293T cells, BECN1 knockout 293T cells (BECN1 KO), and the BECN1 KO cells after transfection with Becin1-Flag. We found that the replication and the ability to induce autophagy of the WSN-NS1_SC09 strain in 293T cells and BECN1-expressed BECN1 KO cells were both significantly higher than those of either the WSN-NS1_JL89 or the WSN strain (Figure 7 B and C). However, in BECN1 KO cells, there were no obvious differences in replication among these three strains, so did the levels of induced transformation of LC3B (Figure 7 B and C). These data further confirmed the importance of BECN1 in the autophagy pathway induced by NS1_SC09.

Figure 7.

BECN1 plays a key role in the autophagy pathway induced by NS1_SC09. (A) 293T cells were stimulated with starvation or treated with Eug or Cal, and BECN1 KO cells were over-expressed with BECN^F123A-Flag as indicated. The cells mentioned above were transfected with indicated plasmids above. Cell lysates were subjected to Flag-IP. (B and C) 293T cells, BECN1 KO cells, and BECN1 KO cells overexpressing of BECN1-Flag or Becin1^F123A-Flag as indicated above were infected with WSN, WSN-NS1_SC09, or WSN-NS1_JL89 at the MOI of 5.0 and 0.01. The supernatants were sampled at 24 h post-infection, and the viral titers were determined with endpoint titration in MDCK cells. The cell lysates were prepared and analyzed using western bloting. (D-G) 293T cells and A549 cells were infected with WSN, WSN-NS1_SC09, or WSN-NS1_JL89 after stimulation with starvation or treatment with Eug or Cal as indicated above. The supernatants were sampled at 24 h post-infection, and the viral titers were determined using endpoint titration in MDCK cells. The cell lysates were prepared and analyzed using western blotting. Error bars indicate the SD from three independent experiments. *, P < 0.05; **, P < 0.01.

According to our study, we found that the dissociation of the BECN1-BCL2 heterodimer played a key role in the autophagy pathway induced by NS1_SC09. We further investigated viral replication under the disordered stabilities of BECN1-BCL2 heterodimer induced by different conditions or chemicals. Starvation and the expression of BECN1^F123A mutation are found to induce autophagy by promoting dissociation of the BECN1-BCL2 heterodimer [47,48]. Conversely, eugenol (Eug) has been reported to inhibit autophagy induced by IAVs through inhibition of the dissociation of BECN1 and BCL2 [49]. STK4/MST1 (serine/threonine kinase 4) is reported to phosphorylate the Thr108 residue in the BH3 domain of BECN1, which can inhibit autophagy through enhancing the interaction between BECN1 and BCL2. Calyculin A (Cal), as an agonist of STK4/MST1, may exert its inhibitory activity of autophagy through enhancing the activity of STK4/MST1 [50,51]. Therefore, we used co-IP experiments to evaluate the interactions of BECN1 with LRPPRC and BCL2 under the above conditions. From our results, which were consistent both with those from previous reports and our hypothesis, induction of starvation and the expression of BECN1^F123A mutation inhibited the bindings of BECN1 to BCL2 and LRPPRC similar to NS1_SC09 (Figure 7A). Eug and Cal significantly inhibited NS1_SC09-mediated BECN1-BCL2 dissociation (Figure 7A). In the subsequent virus infection experiments, A549 cells, 293T cells, and BECN1 KO cells were infected with WSN, WSN-NS1_SC09, or WSN-NS1_JL89, at the MOI of 0.01 and 5.0 under the conditions above. We found that the BECN1^F123A mutation (Figure 7 B and C) and starvation (Figure 7D-G) significantly promoted viral replication of the WSN and WSN-NS1_JL89, but not WSN-NS1_SC09, by enhancing autophagy of infected cells. Moreover, Cal and Eug obviously inhibited the replication of these three strains as well as autophagy induced by them, especially for the WSN-NS1_SC09 strain (Figure 7D-G). The above results further suggested that NS1_SC09-mediated dissociation of BECN1 and BCL2 played an important role in the regulation of autophagy and viral replication, and therefore this pathway could be a potential target for the treatment of A/(H1N1) pdm09 infection.

Discussion

Throughout human history, there are clear evidences for four major influenza pandemics [3,52,53]. Influenza pandemics are mainly caused by new viral subtypes resulting from antigen conversion of IAVs [3,54]. Due to general lack of corresponding immunity in human population, the new reassortant IAVs spread rapidly, causing a widespread epidemic at a global scale [3]. The pathogen of the 2009 influenza pandemic A/(H1N1) pdm09 is a four-source reassortant IAV, comprising genetic material from a three-source (classic H1N1 swine, avian, and human H3N2) reassortant swine IAV circulating in North America, as well as from an “avian-like” swine IAV circulating in Europe and Asia [4,55]. It is well known that when infected with a new type of virus, host cells often use the innate immune system and change their homeostasis to limit the replication of virus [3,5]. For a new type of virus to spread quickly, it must develop new pathogen-host interactions to break through the limitations imposed by hosts. Hence, research into the molecular characteristics of A/(H1N1) pdm09 is necessary to determine the cause of the 2009 IAV pandemic and to inform the effective prevention and control of the next. However, almost all previous studies have been based on the molecular epidemiological analysis of clinical data [7,8]. Knowledge of why A/(H1N1) pdm09 was able to spread so quickly in comparison with other characteristic subtypes of IAVs is still lacking.

In this study we found that the higher replication capacity of A/(H1N1) pdm09 might connect to their infectivity. Our findings revealed that when host cells were infected with SC09 strain, the NS1_SC09 was able to promote viral replication by enhancing autophagy which mainly attributed to the inhibitory effect of NS1_SC09 on the negative regulation of autophagy by LRPPRC. The interaction between NS1_SC09 and LRPPRC competitively blocked the interaction of LRPPRC with BECN1-BCL2 heterodimer, resulting in degradation of BCL2 and increased recruitment of BECN1 by PIK3C3 and then induction of the initiation of autophagy (Figure 8).

Figure 8.

A proposed model for the pathway of autophagy induced by NS1 _pdm09. In the normal physiological state, LRPPRC interacts with the BECN1-BCL2 heterodimer, forming a stable complex in the mitochondria, thereby inhibiting BECN1-dependent autophagy. During A/(H1N1) pdm09 infection, the NS1_pdm09 promotes viral replication by enhancing autophagy which is mainly attributed to the inhibitory effect of NS1_pdm09 on the negative regulation of autophagy by LRPPRC. The interaction between NS1_pdm09 and LRPPRC competitively blocks the interaction of LRPPRC with BECN1-BCL2 heterodimer, resulting in degradation of BCL2 and increased recruitment of BECN1 by PIK3C3, and then induction of the initiation of autophagy.

Numerous studies have shown that the activation of autophagy in host cells during IAV infection is beneficial to virus replication [18]. IAVs have evolved different strategies to activate autophagy in host cells and promote self-replication, and differences in the ability of different subtypes of IAVs to activate autophagy can also be related to viral virulence [16]. IAVs regulate autophagy mainly by activating initiation of autophagy and blocking autophagosome-lysosome fusion [19]. However, the exact mechanisms by which the IAVs activate and utilize autophagy remain unclear. Moreover, autophagy induced by IAVs may induce severe inflammatory response in infected hosts [56,57].

The IAV NS1 is a multi-functional virulence factor and an inhibitor of antiviral immunity [24]. There is a complex interaction network in host cells, which affects the replication of IAVs through a variety of pathways [58,59]. NS1 selectively enhances viral mRNA translation and regulates viral RNA synthesis in infected cells [24,60]. NS1 is additionally known to be an antagonist of the host innate immune response and inhibits the activity of host antiviral proteins such as RIGI (RNA sensor RIG-I) and EIF2AK2 (eukaryotic translation initiation factor 2 alpha kinase 2), as well as restrains proteins necessary for the activation of IFN system, such as NFKB (nuclear factor kappa B) and IRF3 (interferon regulatory factor 3) [61–65]. NS1 also acts on downstream molecules of the IFN signaling pathway, thereby antagonizing their activation [59,66,67]. Some studies have shown that NS1 cannot directly induce autophagy, but can promote autophagy by cooperating with other viral proteins [68]. In this study, we found for the first time that NS1_SC09 was able to directly induce autophagy in a virus-specific manner. Sequence alignment revealed that residues 104–125 in NS1_SC09 are highly conserved in different A/(H1N1) pdm09 strains and a few swine IAVs isolated around the world, but not in other seasonal IAVs. Thus, A/(H1N1) pdm09 may specifically induce autophagy by using NS1 to promote its replication. By comparing the residues of NS1 from A/(H1N1) pdm09 and other seasonal IAVs, we found that the main mutations occurred in T91S, F103L, K108R, A112I, M119L, I123V, and D125E. D125E mutation in the NS1 of A/(H1N1) pdm09 (NS1_pdm09) can affect apoptosis levels in host cells [69,70]. However, their effects on other physiological activities expect autophagy in host cells remain to be further determined.

In this study, we observed that the interaction between NS1_SC09 and LRPPRC led to the dissociation and degradation of BCL2 from the BECN1-BCL2 heterodimer, which improved the interaction between BECN1 and PIK3C3, and then initiated autophagy. As an apoptosis inhibiting protein, BCL2 mainly plays a role in regulating apoptosis in the mitochondria [71–73]. Therefore, whether NS1-dependent BCL2 degradation can regulate host cell apoptosis and its effect on viral replication or virulence, as well as the specific mechanism of degradation, are topics for future study. Meanwhile, phosphorylation modification at different sites on proteins in the autophagy pathway also plays a crucial role in the regulation of autophagy [74–76], although the regulatory mechanisms of phosphorylation and dephosphorylation, and how they are involved in NS1-dependent autophagy remain unclear.

There is considerable evidence that viruses have evolved different strategies to counter and utilize the various stages of autophagy to promote self-replication [11]. Certain viruses can directly reduce autophagy, such as Kaposi’s sarcoma-associated herpesvirus (KSHV) and murine gammaherpesvirus 68 (MHV68), which can inhibit autophagy by encoding viral homologs of cellular BCL2 pro-survival proteins (vBCL2), open reading frame 16 (ORF16) and M11 [77]. The vBCL2 has high homology with cellular BCL2, and vBCL2 can therefore directly interact with BECN1 to reduce autophagy [77,78]. Selective autophagy of host cells can be induced by certain viruses, such as HIV, which can utilize its own virion infectivity factor (Vif) to induce autophagous degradation of the antiviral protein APOBEC3G [79]. In addition, certain viruses are known to inhibit the maturation of autophagosomes or block the fusion of autophagosomes to lysosomes, thus inhibiting the autophagic degradation pathway [17,80]. The Nef of HIV-1 can interact with BECN1 to isolate TFEB (transcription factor EB) in the cytoplasm, thereby inhibiting autophagosome maturation [81,82].

Different kinds of viruses are known to regulate host cell autophagy through targeting the BECN1-BCL2 heterodimer [45]. If this common mechanism can be elucidated then the replication of certain kind of viruses could potentially be inhibited through inhibition of the BECN1-dependent initiation of autophagy, which might provide a candidate target for the development of antiviral drugs and vaccines.

Materials and methods

Antibodies and reagents

Antibodies used in this study include: mouse anti-HA tag monoclonal antibody (Sigma-Aldrich, H9658), mouse anti-Flag tag monoclonal antibody (Sigma-Aldrich, F1804), rabbit anti-HA tag polyclonal antibody (Sigma-Aldrich, F7425), mouse anti-ACTB monoclonal antibody (Sigma-Aldrich, A5441), rabbit anti-ACTB polyclonal antibody (ABclonal Technology, AC026), rabbit anti-LC3B monoclonal antibody (ABclonal Technology, A19665), rabbit anti-LAMP2 polyclonal antibody (ABclonal Technology, A0593) rabbit anti-BECN1 polyclonal antibody (Proteintech, 11,306-1-AP), rabbit anti-LRPPRC polyclonal antibody (Proteintech, 21,175-1-AP), rabbit anti-BCL2 monoclonal antibody (ABclonal Technology, A19693), rabbit anti-PIK3C3/VPS34 monoclonal antibody (ABclonal Technology, A12295), rabbit anti-NS polyclonal antibody (CUSABIO, H0228A), and NP monoclonal antibody (made by our lab). These antibodies were applied and then samples were incubated overnight at 4°C. Samples then treated with anti-rabbit IgG (H + L) antibody, DyLight™ 800-Labeled (KPL, 5230–0412), anti-rabbit IgG (H + L) antibody, DyLight™ 680-Labeled (KPL, 072–06-15-16), anti-rabbit IgG (H + L) antibody, DyLight™ 800-Labeled (KPL, 072–06-15-16), goat anti-rabbit IgG (H + L) Fluor 488 (Thermo Fisher Scientific, A-11034), goat anti-rabbit IgG (H + L) Alexa Fluor 647 (Thermo Fisher Scientific, A-2124), goat anti-mouse IgG (H + L) cross-adsorbed secondary antibody, Alexa Fluor 488 (Thermo Fisher Scientific, A-1101), or goat anti-mouse IgG (H + L) cross-adsorbed secondary antibody, Alexa Fluor 405, (Thermo Fisher Scientific, A-31553) were incubated 1 h at room temperature in the dark. Certein samples treated with DAPI (Sigma-Aldrich, D9542) were incubated for 10 min at room temperature in the dark. The reagents CQ (HY-B1370), 3-MA (HY-19312), rapamycin (HY-10219), eugenol (HY-N0337), and calyculin A (HY-18983) were ordered from MCE. Earle’s balanced salts solution (EBSS) was bought from Sigma-Aldrich (E2888).

Cells, viruses and plasmids

293T, (ATCC CRL3216), MDCK, (ATCC CCL-34), Vero E6 cells (ATCC CRL-1586), and HeLa cells (ATCC CRM-CCL-2) were maintained in Dulbecco modified Eagle medium (DMEM; Sigma-Aldrich, D0819) with 10% fetal bovine serum (FBS; Wisent, 086–150) and 1% penicillin-streptomycin (Thermo Fisher Scientific, 15,070,063), and kept at 37°C in 5% CO₂. A549 cells (ATCC CRM-CCL-185) were maintained in Ham’s F-12 K (Kaighn’s) medium (Thermo Fisher Scientific, 21,127,030) with 10% FBS and 1% penicillin-streptomycin, and kept at 37°C in 5% CO₂. Certain cells, viruses, and plasmids were kindly provided by the individuals listed below: The SC09 strain was provided by Hualan Chen (Harbin Veterinary Research Institute, China); the WSN strain was provided by Yoshihiro Kawoaka’s lab (Department of Pathobiological Sciences, School of Veterinary Medicine, University of Wisconsin-Madison, WI, USA). The reverse genetics system based on the pBD vector for the SC09 virus was established in Hualan Chen’s lab (Harbin Veterinary Research Institute, China). pCMV-GFP-LC3B and pCMV-mRFP-GFP-LC3B were obtained from Doctor Yuezhi Lin (Harbin Veterinary Research Institute, China). The BECN1 KO cells and pCAGGS-PIK3C3-Flag were obtained from Doctor Huiling Ren (Harbin Veterinary Research Institute, China). The JL89 strain was preserved in our own lab. The reverse genetics systems based on the pBD vector for the JL89 strain and that based on the pHH21 vector for the WSN strain were established in our own lab. The full-length cDNAs encoding NS1, PB1, PB2, PA, NP, M1, M2, HA, NA, and NS2 from three studied strains were amplified using RT-PCR from the total RNA of A549 cells infected with WSN, SC09, or JL89 strain using specific primers. The full-length cDNAs encoding LRPPRC, BECN1, and BCL2 were amplified using RT-PCR from the total RNA of A549 cells using specific primers. (available upon request). The full-length cDNA was then cloned into pCAGGS-Flag (LRPPRC, BCL2, BECN1, BECN1^F123A), pCAGGS-HA (LRPPRC, BCL2, BECN1), pCDNA3.1–3× HA (NS1_SC09, NS1_WSN, NS1_JL89), pCDNA3.1–3× Flag (NS1_SC09, NS1_WSN, NS1_JL89), and pLPCX-Flag (LRPPRC). The empty vectors pCAGGS-Flag, pCAGGS-HA, pCDNA3.1–3× HA, pCDNA3.1-Flag, and pCDNA3.1-V5 were stored in our laboratory. The NS1 mutants (NS1_{SC091-73-JL89-74-C terminal}, NS1_{SC091-125-JL89-126-C terminal}, NS1_{JL89 1–73-SC09-74-C terminal}, NS1_{JL89 1–125-SC09-126-C-terminal,} NS1_{SC09 72–82 JL89}, NS1_{SC09 83–92JL89}, NS1_{SC09 93–103JL89}, NS1_{SC09 104–125JL89}, NS1_{JL89 72–82SC09}, NS1_{JL89 83–92SC09}, NS1_{JL89 93–103SC09}, NS1_{JL89 104–125SC09}, NS1_SC09 ^I112T, NS1_SC09 ^P114G, NS1_SC09 ^C116M, NS1_SC09 ^R118K, NS1_SC09 ^L119M, NS1_JL89 ^T112I, NS1_JL89 ^G114P, NS1_JL89 ^M116C, NS1_JL89 ^K118R, and NS1_JL89 ^M119L) and the LRPPRC truncations (T1:1–337 AA; T2: 338–711 AA; T3: 712–1067 AA; T4: 1068–1394 AA) were constructed according to the online In-Fusion HD cloning kit user manual (http://www.clontech.com/CN/Products/Cloning_and_Competent_Cells/Cloning_Kits/xxclt_searchResults.jsp). Briefly, the fragments of the pcDNA3.1 or pCAGGS vector and each target gene were amplified with a 15-bp homologous arm and then fused using In-Fusion® HD Cloning Kit (Clontech, 639,650) and overlap cloning techniques. The lentivirus packaging system plasmids pCgp, VSV-G and pLPCX were held in our lab.

Establishment of LRPPRC knockout cell lines

293T cells were used to generate LRPPRC gene knockout cells using CRISPR-Cas9 technology. The gRNA was designed using the Broad Institute Zhang Lab Guide Design Resources. The sequence targeting LRPPRC was as follows: TTGCTAGTTGGACGTCGGA DNA fragments that contained the gRNA specific for LRPPRC, a guide RNA scaffold, an RNU6 promoter, and an RNU6 termination signal sequence were synthesized and subcloned into the pMD18-T backbone vector (Clontech, 6011). The Cas9-eGFP expression plasmid (pMJ920) was kindly provided by Dr. Jennifer Doudna at University of California, Berkeley. Briefly, 293T cells in 6-well plate were transfected with 1.0 μg of gRNA expression plasmid and 1.0 μg of pMJ920 plasmid with PolyJet DNA reagent (SignaGen Laboratories, SL100688). GFP-positive cells were sorted by fluorescence-activated cell sorting at 36 h post-transfection. The positive clones were validated with DNA sequencing and western blotting. The growth curves of positive and control cells were analyzed using the reagents from the CCK-8 kit (MCE, HY-K0301).

Generation of LRPPRC-stable expression 293T cells

For the generation of LRPPRC-stable expression 293T cells, the lentivirus packaging plasmids (pCgp, VSVG, and pLPCX-LRPPRC-Flag or empty pLPCX-Flag) were co-transfected into 293T cells using PolyJet DNA reagent (SignaGen Laboratories, SL100688). The medium was replaced 6 h later. Viral particles were collected 72 h post-transfection and filtered. When the 293T cells seeded on six-well plates were ∼40% confluent, they were infected with the lentivirus. After 48 dpi, the lentivirus-infected cells were selected using 2.0 μg/ml puromycin (Thermo Fisher Scientific, A1113803) for 5 days with the medium changed daily to select the positive clones. The expression of LRPPRC-Flag was analyzed using western blotting. The monoclonal cells were acquired using flow cytometry in 96-well plates. The LRPPRC-stable expression cell lines were obtained from these enlarged monoclonal cells and were confirmed using western blotting. The growth curves of positive and control cells were analyzed using the reagents from the CCK-8 kit.

Rescue of recombinant WSN-NS1_SC09 and WSN-NS1_JL89 viruses

293T cells were transfected with the reverse genetic system of WSN, in which the pHH21-NS1_WSN was replaced with either PBD-NS1_SC09 or PDZ-NS1_JL89. The cell supernatants were collected 24 h post-transfection and filtered. The cell supernatants were inoculated into SPF chicken embryos (Harbin Veterinary Research Institute, China), and the chicken embryo allantoic fluid was collected 48 h after inoculation. The NS1 gene in the samples were amplified using RT-PCR and sequenced for verification. The TCID50 of the viruses were determined for quantification [83].

Immunofluorescence analysis

HeLa cells, 293T cells, LRPPRC_293T cells, and LRPPRC KO cells plated in glass bottom cell culture dishs (NEST, 801,001) were transfected with the indicated plasmids or infected with the WSN strain at the MOI of 0.1. Twenty-four hours later, cells were washed with phosphate-buffered saline (PBS; Thermo Fisher Scientific, 10,010,023) and fixed with 4% paraformaldehyde (Beyotime, P0099-500 mL) for 20 min. Cells were then washed with PBS and incubated with 0.1% immunostaining permeable fluid (Beyotime P0096-500 mL) for 15 min. The cells were then blocked with 5% dry milk for 1 h at room temperature and incubated with the corresponding antibodies at 4°C overnight, followed by secondary antibodies as indicated above for 1 h at room temperature in dark. Finally, the cells were washed with PBS 5 times, and were either treated with DAPI Reagent or not. Samples were then assessed using confocal microscopy (Carl Zeiss LSM 800 Confocal Microscope and ZEN 2.3 LITE software).

Virus infection

The MDCK cells, A549 cells, 293T cells, 293T_LRPPRC cells, LRPPRC KO cells, BECN1 KO cells, and Vero E6 cells were infected with indicated viruses at the MOI of 0.01 and 5.0 for 1 h, washed twice with PBS, and cultured at 37°C in Opti-MEM (Thermo Fisher Scientific, 31,985,070) containing 0.5% FBS and tosylsulfonyl phenylalanyl chloromethyl ketone (TPCK)-trypsin (Sigma-Aldrich, 4,370,285) at 1 μg/ml (A549 cells), 0.2 μg/ml (293T related cells), or 2.0 μg/ml (Vero E6 cells and MDCK cells). At the indicated time points, the culture supernatant was harvested, and the virus titers in MDCK cells were determined as described above.

Transmission electron microscope analysis

A549 cells were infected with WSN, SC09, or JL89 strain at the MOI of 0.01 or transfected with pcDNA3.1-Flag, pcDNA3.1-NS1_WSN-Flag, pcDNA3.1-NS1_SC09-Flag, or pcDNA3.1-NS1_JL89-Flag. Cell samples were assessed for the formation of autophagosome using TEM 12 h post-infection or 24 h post-transfection. For negative staining: The sample was fixed with 2.5% glutaraldehyde (pH 7.2) overnight. A 20 Μl drop of sample was applied to a carbon coated grid that had been glow discharged for 20s in air, and the grids were immediately negatively stained using 2% phosphotungstic acid. Grids were examined in a H-7650 (Hitachi, Tokyo, Japan) operated at 80 Kv.

Co-IP assay and western blot

For immunoprecipitation and western blotting, infected or transfected cells were lysed using ice-cold lysis buffer (50 mM Tris-HCl [Biosharp, 0234]), pH 7.5, 50 mM NaCl [Sigma-Aldrich, S7653], 5 mM EDTA [Sigma-Aldrich, E6758]), 1% Triton X-100 [Sigma-Aldrich, T7878]) and centrifuged at 13,000 × g and 4°C for 10 min. After centrifugation, the crude lysates were incubated with anti-FLAG M2 magnetic beads (Sigma-Aldrich, M8823), anti-HA magnetic beads (MCE, HY-K0201), or protein A/G magnetic beads (MCE, HY-K0202) bound with anti-BECN1 antibody or anti-PIK3C3 antibody at 4°C for 2 h. After incubation, the beads were collected using a magnetic separator and washed three times with PBS. The bead-bound materials were eluted using a 3× Flag peptide (Sigma-Aldrich, F4799-4 MG) or by boiling in the SDS-PAGE loading buffer (prepared by our lab), and analyzed either using MS (Institute of Biophysics, Chinese Academy of Science, China) or western blotting. For western blotting, the samples were separated using SDS-PAGE and then transferred onto nitrocellulose membranes (Sigma-Aldrich, Z358657). Membranes were blocked with 5% milk in Tris-buffered saline with 0.05% Tween-80 (TBST; Solarbio Life Sciences, T1085) for 1 h. Incubation with the primary antibody (as indicated above) was performed for 2 h at room temperature. The secondary antibodies were then applied, and the samples were incubated at room temperature for 1 h in the dark. The membranes were then washed three times for 10 min with TBST. Signals were detected using an Odyssey imaging system (LI-COR, Lincoln, NE, USA). All bands from the western blots were detected within the linear range.

Viral titration of infectious particles by 50% tissue infective dose assay (TCID50)

Serial 10-fold dilutions of samples were prepared in Opti-MEM medium complemented with 2.0 mg/ml trypsin TPCK. The dilutions were inoculated into confluent MDCK cell monolayers prepared in 96-well microtiter plates, using 4 wells per dilution. Plates were incubated at 37°C under 5% CO₂, and the viral titer was determined 24 h after inoculation. The endpoint was taken to be the highest dilution of the virus that produced CPE in 50% of the inoculated cells. Viral titers (50% tissue infective doses TCID50/0.1 ml) were calculated following Reed and Muench [83].

Quantitative real-time PCR

Total RNA was extracted from Vero E6 cells and A549 cells using a RNeasy mini kit (Qiagen, 74,106) according to the manufacturer’s instructions, and then subjected to reverse transcription using PrimeScript RT reagent kit with a gDNA Eraser (Takara, RR047B). The expression levels of Human IFNB1 mRNA or Macaca mulatta IFNB1 mRNA were quantified using SYBR-Green (Takara, RR430A)-based real-time quantitative PCR analysis on Agilent Mx3000P, according to the manufacturer’s protocols. Real-time RT-PCR was performed using the Human IFNB1 mRNA primers: Hu IFNB1-forward (5ʹ AGTAGGCGACACTGTTCGTG3ʹ) and Hu IFNB1-reverse (5ʹ GCCTCCCATTCAATTGCCAC3ʹ) or Ma IFNB1-forward (5ʹ TCCTGTGGCAATTGAATGG3ʹ) and Ma IFNB1-reverse (5ʹ AATAGCGAAGATGTTCTGGAG3ʹ). ACTB was used as a housekeeping control to normalize the number of living cells. The ACTB primers were ACTB-forward (5ʹ ACGGCATCGTCACCAACTG3ʹ) and ACTB-reverse (5ʹCAAACATGATCTGGGTCATCTTCTC3ʹ). The difference in the transcription of IFNB1 mRNA was calculated from the RQ value.

Statistical analyses

Data are expressed as means ± standard deviations (SD). Statistical analysis was performed in the software GraphPad Prism 7. Independent samples were compared using paired Student’s two-tailed t-tests. P values equal to or lower than 0.05 were considered significant (*P < 0.05, ** P < 0.01). P values > 0.05 were considered statistically non-significant.

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Funding Statement

This work was supported by the grants from the National Natural Science Foundation of China (www.nsfc.gov.cn) to X.W. (as a part of grant to Dr. Hualan Chen’s grant 31521005) and Z.Z. (31702269) and the Natural Science Foundation of Heilongjiang Province (http://jj.hljkj.cn/zr/) to Z.Z. (No. JC2018010). The funders had no role in study design, data collection, and analysis, decision to publish or preparation of the manuscript.

Disclosure statement

No potential conflict of interest was reported by the authors

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/15548627.2022.2139922