PRRSV-2 impedes lysosomes from eliminating secondary infected bacteria

TL;DR

PRRSV-2 weakens the ability of pig immune cells to clear bacteria by disrupting a key structural component of the cell, which could lead to new treatments for secondary infections.

Contribution

This study reveals a novel mechanism by which PRRSV-2 reprograms the cytoskeleton to facilitate secondary bacterial survival.

Findings

PRRSV-2 disrupts F-actin in macrophages, impairing lysosome transport to bacteria.

nsp5 protein downregulates RhoA via ubiquitination, leading to F-actin suppression and bacterial persistence.

Targeting cytoskeletal F-actin could be a new strategy to control secondary bacterial infections.

Abstract

As an immunosuppressive virus, the occurrence of secondary bacterial infection following porcine reproductive and respiratory syndrome virus type 2 (PRRSV-2) infection is widely recognized. The immune escape capability of PRRSV-2 enables the virus to maintain efficient proliferation even within macrophages. In this study, we report that PRRSV-2 infection disrupts the intracellular F-actin, thereby causing the inability of macrophage lysosomes to transport to secondary infected bacteria promptly for bacterial clearance. RhoA is a crucial molecule in the polymerization of G-actin to F-actin within the cell. Silencing RhoA suppresses the production of F-actin in the cell, delays the targeted clearance of bacteria by lysosomes, and leads to an increase in the number of viable bacteria within the cell. Overexpression of RhoA promotes the production of F-actin, accelerates the targeted clearance of lysosomes to bacteria, and effectively reduces the number of viable bacteria. After PRRSV-2 infection, the expression of RhoA protein is down-regulated by nsp5 to inhibit the production of F-actin. Mechanistically, nsp5 interacts with the E3 ubiquitin ligase Smurf1 to mediate K63-linked ubiquitination of RhoA at lysine 187 (K187), which subsequently leads to its degradation via the autophagy-lysosome pathway under the guidance of the selective autophagy receptor TOLLIP. Therefore, our study presents a novel mechanism through which PRRSV-2 reprograms the cytoskeleton to facilitate the survival of bacteria in secondary infections, providing a theoretical foundation and target for the prevention and control of PRRSV-2 secondary bacterial infection.

Since its first discovery in the 1980s, porcine reproductive and respiratory syndrome virus type 2 (PRRSV-2) has caused incalculable economic damage to the global pig industry. Due to the immunosuppressive characteristics of PRRSV-2, co-infection with PRRSV-2 and secondary bacteria is frequent and intensifies the production of pro-inflammatory cytokines, resulting in more severe clinicopathological damage. A better understanding of the mechanisms of PRRSV-2 secondary bacterial infection is essential for the development of safe and effective treatments. Here, we report that PRRSV-2 impedes the phagocytic function of lysosomes in macrophages by reprogramming the cytoskeleton F-actin, thereby facilitating the survival of secondary infected bacteria. Our work provides a novel mechanism of secondary bacterial infection with PRRSV-2 and suggests that targeting cytoskeletal F-actin may be a new strategy to control secondary bacterial infection.

Figures

PRRSV-2 facilitates the secondary infection of A. viridans.(A) Immortalized PAMs were infected or not infected with PRRSV-2 for 24 h and then infected with A. viridans for 4 h, the cells were harvested to detect the number of bacteria by CFUs. (B) Immortalized PAMs were infected or not infected with PRRSV-2 for 24 h and then infected with A. viridans for 4 h, the cells were harvested to detect the number of bacteria by immunofluorescence assay. (C-F) Immortalized PAMs were infected with PRRSV-2 for 24 h and then infected with A. viridans for 4 h and 8 h, the cells were harvested to detect (C) IL-1β mRNA expression, (D) IL-6 mRNA expression, (E) CCL4 mRNA expression, (F) P65 mRNA expression. The cells infected with PRRSV-2 and A. viridans respectively were taken as the control. (G) Immortalized PAMs were infected with PRRSV-2 for 24 h and then infected with A. viridans for 4 h, the cells were harvested to detect the bacterial morphology by transmission electron microscopy. Scale bar: 10 μm. P values were calculated using Student’s t-test. ns, not significant; *, P < 0.05; , P < 0.01; *, P < 0.001.

PRRSV-2 facilitates A. viridans secondary infection in piglets and triggers pronounced inflammatory responses.(A) Flow chart of the animal experiments. Fifteen piglets were randomly divided into three groups, with 5 piglets in each group. PRRSV-2 was infected on day 1, followed by A. viridans infection on day 7, as illustrated in the figure. On day 9, the surviving piglets were humanely euthanized in accordance with established ethical guidelines. (B) The rectal temperature of the piglets was recorded daily. (C) Following the death of piglets, PAMs were isolated and the expression of the PRRSV-2 N protein was examined by Western blot, tubulin served as an internal control. (D) The survival rate of the piglets in each group. (E) The lung tissues of piglets were homogenized, and bacterial load was quantitatively assessed via CFUs. (F-G) The expression levels of (F) IL-1β, (G) IL-6 in PAMs of piglets were quantified using RT-qPCR. (H-I) The expression levels of (H) IL-1β, (I) IL-6 in serum of piglets were quantified using ELISA. (J) Lung tissue sections from piglets in the three groups were stained with hematoxylin and eosin (HE). P values were calculated using Student’s t-test. *, P < 0.05; *, P < 0.001.

PRRSV-2 induces the depolymerization of F-actin and thereby impedes the lysosome-mediated clearance of bacteria.(A) Immortalized PAMs were infected or not infected with PRRSV-2 for 24 h and then infected with A. viridans for 4 h, the cells were harvested to detect the subcellular localization for lysosome and A. viridans. Colocalization between lysosome and A. viridans was quantified using ImageJ, with results expressed as the Manders’ correlation coefficients (M1 and M2). Specifically, M1 represents the proportion of lysosomal signal that overlaps with the bacterial signal relative to the total lysosomal signal, while M2 represents the proportion of bacterial signal overlapping with the lysosomal signal relative to the total bacterial signal. n = 20 cells from three independent experiments. (B) Immortalized PAMs were infected or not infected with PRRSV-2 for 24 h, the cells were harvested to observe the changes in F-actin. The fluorescence intensity of F-actin was quantified using ImageJ, n = 20 cells from three independent experiments. (C) Immortalized PAMs were incubated with DMSO or LAT-A for 12 h, the cells were harvested to observe the changes in F-actin. The fluorescence intensity of F-actin was quantified using ImageJ, n = 20 cells from three independent experiments. (D) Immortalized PAMs were incubated with DMSO or LAT-A for 12 h and then infected with A. viridans for 4 h, the cells were harvested to detect the number of bacteria by CFUs. (E) Immortalized PAMs were incubated with DMSO or LAT-A for 12 h and then infected with A. viridans for 4 h, the cells were harvested to detect the subcellular localization for lysosome and A. viridans. Colocalization between lysosome and A. viridans was quantified using ImageJ, with results expressed as the Manders’ correlation coefficients (M1 and M2). Specifically, M1 represents the proportion of lysosomal signal that overlaps with the bacterial signal relative to the total lysosomal signal, while M2 represents the proportion of bacterial signal overlapping with the lysosomal signal relative to the total bacterial signal. n = 20 cells from three independent experiments. Scale bar: 10 μm. P values were calculated using Student’s t-test. , P < 0.01, *, P < 0.001.

RhoA promotes F-actin polymerization and accelerates lysosomal targeting of bacteria.(A) Immortalized PAMs were transfected with siNC and three distinct RhoA-specific siRNA for 36 h, the cells were harvested to detect the RhoA protein expression by Western blot, tubulin served as an internal control. (B) Immortalized PAMs were transfected with siNC or siRhoA for 36 h, the cells were harvested to observe the changes in F-actin. The fluorescence intensity of F-actin was quantified using ImageJ, n = 20 cells from three independent experiments. (C) Immortalized PAMs were transfected with siNC or siRhoA for 36 h and then infected with A. viridans for 4 h, the cells were harvested to detect the number of bacteria by CFUs. (D) Immortalized PAMs were transfected with siNC or siRhoA for 36 h and then infected with A. viridans for 4 h, the cells were harvested to detect the subcellular localization for lysosome and A. viridans. Colocalization between lysosome and A. viridans was quantified using ImageJ, with results expressed as the Manders’ correlation coefficients (M1 and M2). Specifically, M1 represents the proportion of lysosomal signal that overlaps with the bacterial signal relative to the total lysosomal signal, while M2 represents the proportion of bacterial signal overlapping with the lysosomal signal relative to the total bacterial signal. n = 20 cells from three independent experiments. (E) Immortalized PAMs were transfected with p3 × Flag (vector) or p3 × Flag-RhoA for 36 h, the cells were harvested to observe the changes in F-actin. The fluorescence intensity of F-actin was quantified using ImageJ, n = 20 cells from three independent experiments. (F) Immortalized PAMs were transfected with p3 × Flag (vector) or p3 × Flag-RhoA for 36 h and then infected with A. viridans for 4 h, the cells were harvested to detect the number of bacteria by CFUs. (G) Immortalized PAMs were transfected with p3 × Flag (vector) or p3 × Flag-RhoA for 36 h and then infected with A. viridans for the indicated periods, the cells were harvested to detect the subcellular localization for lysosome and A. viridans. Colocalization between lysosome and A. viridans was quantified using ImageJ, with results expressed as the Manders’ correlation coefficients (M1 and M2). Specifically, M1 represents the proportion of lysosomal signal that overlaps with the bacterial signal relative to the total lysosomal signal, while M2 represents the proportion of bacterial signal overlapping with the lysosomal signal relative to the total bacterial signal. n = 20 cells from three independent experiments. Scale bar: 10 μm. P values were calculated using Student’s t-test. ns, not significant; , P < 0.01; *, P < 0.001.

PRRSV-2 down-regulates RhoA expression through nsp5.(A-B) PAMs were infected with PRRSV-2 (MOI = 1) for the indicated periods, and the cells were harvested to detect (A) RhoA mRNA expression by RT-qPCR, (B) RhoA protein expression by Western blot, tubulin served as an internal control, and the PRRSV-2 N protein was used as an infection indicator. (C) HEK-293T cells were co-transfected with the plasmids of PRRSV-2 proteins and RhoA for 36 h, and the cells were harvested to detect the PRRSV-2 proteins and RhoA protein expression by Western blot, tubulin served as an internal control. (D) HEK-293T cells were co-transfected with the plasmids of RhoA and nsp5 with an increasing concentration gradient for 36 h, the cells were harvested to detect the RhoA and PRRSV-2 nsp5 protein expression by Western blot, tubulin served as an internal control. (E-F) HEK-293T cells were co-transfected with the plasmids of RhoA and nsp5 for 36 h, the cells were harvested to detect the interaction between RhoA and nsp5 by Co-IP. (G) HEK-293T cells were co-transfected with the plasmids of RhoA and nsp2 for 36 h, the cells were harvested to detect the interaction between RhoA and nsp2 by Co-IP. (H) HEK-293T cells were co-transfected with the designated plasmids for 36 h, the cells were harvested to detect the subcellular localization of nsp5 and nsp2 in relation to RhoA, respectively. The co-localization of nsp2 and nsp5 with RhoA was analyzed using ImageJ, and the results were quantified using the Pearson’s correlation coefficient. n = 10 cells from three independent experiments. (I) Online prediction of the transmembrane domain distribution of nsp5 polypeptide sequences. (J) The truncated plasmids were constructed based on the transmembrane domain distribution of nsp5. (K) HEK-293T cells were co-transfected with the plasmids of RhoA and the truncated plasmid of nsp5 for 36 h, the cells were harvested to detect the RhoA and the truncated nsp5 protein expression by Western blot, tubulin served as an internal control. (L) The plasmid of RhoA was co-transfected with nsp5 and nsp5-D5 into HEK-293T cells for 36 h respectively, the cells were harvested to detect the RhoA, nsp5 and nsp5-D5 protein expression by Western blot, tubulin served as an internal control. Scale bar: 10 μm. P values were calculated using Student’s t-test. *, P < 0.001.

Nsp5 degrades RhoA through the autophagy pathway by interaction with TOLLIP.(A) HEK-293T cells were co-transfected with the plasmids of RhoA and nsp5 for 24 h and then incubated with MG132, Z-VAD and 3-MA for 6 h, the cells were harvested to detect the RhoA and nsp5 protein expression by Western blot, tubulin served as an internal control. (B) HEK-293T cells were co-transfected with the designated plasmids for 36 h, the cells were harvested to detect the subcellular localization of RhoA and LC3. The co-localization of RhoA and LC3 was analyzed using ImageJ, and the results were quantified using the Pearson’s correlation coefficient. n = 10 cells from three independent experiments. (C) ATG7+/+ and ATG7-/- HEK-293T cells were co-transfected with the plasmids of RhoA and nsp5 for 36 h, the cells were harvested to detect the RhoA, nsp5 and ATG7 protein expression by Western blot, tubulin served as an internal control. (D) ATG7-/- HEK-293T cells were co-transfected with the plasmids of RhoA, nsp5 and ATG7 for 36 h, the cells were harvested to detect the RhoA, nsp5 and ATG7 protein expression by Western blot, tubulin served as an internal control. (E) The plasmid of nsp5 was co-transfected with TAX1 BP1, NBR1, TOLLIP, PSMD4 and OPTN into HEK-293T cells for 36 h respectively, the cells were harvested to detect autophagy receptors that interact with nsp5 by Co-IP. (F) HEK-293T cells were co-transfected with the plasmids of nsp5 and TOLLIP for 36 h, the cells were harvested to detect the interaction between nsp5 and TOLLIP by Co-IP. (G) The plasmid of RhoA was co-transfected with TAX1 BP1, NBR1, TOLLIP, PSMD4 and OPTN into HEK-293T cells for 36 h respectively, the cells were harvested to detect autophagy receptors that interact with RhoA by Co-IP. (H) HEK-293T cells were co-transfected with the plasmids of RhoA and TOLLIP for 36 h, the cells were harvested to detect the interaction between RhoA and TOLLIP by Co-IP. (I) PAMs were infected with recombinant lentivirus expressing nsp5 or control lentivirus for 24 h, the cells were harvested to detect the interaction between nsp5 and endogenous RhoA, as well as endogenous TOLLIP by Co-IP. (J) HEK-293T cells were co-transfected with Flag-TOLLIP, Myc-RhoA, and HA-nsp5 plasmids for 36 h, the cells were harvested to detect the subcellular localization of TOLLIP, RhoA and nsp5. The co-localization of TOLLIP, RhoA and nsp5 was analyzed using ImageJ, and the results were quantified using the Pearson’s correlation coefficient. n = 10 cells from three independent experiments. (K) HEK-293T cells were co-transfected with HA-nsp5, Flag-TOLLIP and Myc-RhoA plasmids for 36 h, the cells were harvested to detect the interaction between nsp5, RhoA and TOLLIP by ternary Co-IP. (L) HEK-293T cells were co-transfected with the plasmids of RhoA, nsp5 and TOLLIP for 36 h, the cells were harvested to detect the RhoA, nsp5 and TOLLIP protein expression by Western blot, tubulin served as an internal control. (M) TOLLIP+/+ and TOLLIP-/- HEK-293T cells were co-transfected with the plasmids of RhoA and nsp5 for 36 h, the cells were harvested to detect the RhoA, nsp5 and TOLLIP protein expression by Western blot, tubulin served as an internal control. (N) TOLLIP-/- HEK-293T cells were co-transfected with the plasmids of RhoA, nsp5 and TOLLIP for 36 h, the cells were harvested to detect the RhoA, nsp5 and TOLLIP protein expression by Western blot, tubulin served as an internal control. (O) PAMs were transfected with siNC and siTOLLIP for 24 h, and then infected with recombinant lentivirus expressing nsp5 or control lentivirus for 24 h, the cells were harvested to detect the RhoA, nsp5 and TOLLIP protein expression by Western blot, tubulin served as an internal control. Scale bar: 10 μm. P values were calculated using Student’s t-test. *, P < 0.001.

Nsp5 induces polyubiquitination of RhoA.(A) HEK-293T cells were co-transfected with the plasmids of RhoA, nsp5 and Ub for 36 h, the cells were harvested to detect the ubiquitination of RhoA induced by nsp5. (B) PAMs were infected with recombinant lentivirus expressing nsp5 or control lentivirus for 24 h, the cells were harvested to detect the ubiquitination of endogenous RhoA induced by nsp5. (C) HEK-293T cells were co-transfected with the designated plasmids for 36 h, the cells were harvested to detect the subcellular localization of RhoA and LAMP1. The co-localization of RhoA and LAMP1 was analyzed using ImageJ, and the results were quantified using the Pearson’s correlation coefficient. n = 10 cells from three independent experiments. (D) Seven types of ubiquitin plasmids (K6-Ub, K11-Ub, K27-Ub, K29-Ub, K33-Ub, K48-Ub, and K63-Ub) were co-transfected in HEK-293T cells with nsp5 and RhoA respectively for 36 h, the cells were harvested to detect the ubiquitination type of RhoA induced by nsp5. (E-H) The indicated lysine residue mutant plasmids of RhoA were co-transfected in HEK-293T cells with nsp5 for 36 h, the cells were harvested to detect the RhoA and nsp5 protein expression by Western blot, tubulin served as an internal control. (I) HEK-293T cells were co-transfected with the plasmids of RhoAK187R, nsp5 and Ub for 36 h, the cells were harvested to detect the ubiquitination of RhoAK187R induced by nsp5. (J) Immortalized PAMs were transfected with siNC and siRhoA for 24 h, and then transfected with the plasmids of RhoAK187R for 24 h, the cells were harvested to detect the RhoA protein expression by Western blot, tubulin served as an internal control. (K) Immortalized PAMs were transfected with siNC and siRhoA for 24 h, and then transfected with the plasmids of RhoAK187R for 24 h. Subsequently, cells were infected with A. viridans for 4 h, the cells were harvested to detect the number of bacteria by CFUs. (L) Immortalized PAMs were transfected with siNC and siRhoA for 24 h, and then transfected with the plasmids of RhoAK187R for 24 h. Subsequently, cells were infected with A. viridans for 4 h, the cells were harvested to detect the subcellular localization for lysosome and A. viridans. Colocalization between lysosome and A. viridans was quantified using ImageJ, with results expressed as the Manders’ correlation coefficients (M1 and M2). Specifically, M1 represents the proportion of lysosomal signal that overlaps with the bacterial signal relative to the total lysosomal signal, while M2 represents the proportion of bacterial signal overlapping with the lysosomal signal relative to the total bacterial signal. n = 20 cells from three independent experiments. Scale bar: 10 μm. P values were calculated using Student’s t-test. , P < 0.01, *, P < 0.001.

Nsp5 potentiates Smurf1-mediated ubiquitination and degradation of RhoA.(A) The plasmid of nsp5 was co-transfected with Smurf1, Smurf2, Cul3 and RhoA into HEK-293T cells for 36 h respectively, the cells were harvested to detect E3 ubiquitin ligase that interact with nsp5 by Co-IP. (B) The plasmid of Smurf1 was co-transfected with nsp5, RhoA and nsp2 into HEK-293T cells for 36 h respectively, the cells were harvested to detect nsp5 and RhoA that interact with Smurf1 by Co-IP. (C) PAMs were infected with recombinant lentivirus expressing nsp5 or control lentivirus for 24 h, the cells were harvested to detect the interaction between nsp5 and endogenous RhoA, as well as endogenous Smurf1 by Co-IP. (D) HEK-293T cells were co-transfected with Flag-Smurf1, Myc-RhoA, and HA-nsp5 plasmids for 36 h, the cells were harvested to detect the subcellular localization of Smurf1, RhoA and nsp5. The co-localization of Smurf1, RhoA and nsp5 was analyzed using ImageJ, and the results were quantified using the Pearson’s correlation coefficient. n = 10 cells from three independent experiments. (E) HEK-293T cells were co-transfected with HA-nsp5, Flag- Smurf1 and Myc-RhoA plasmids for 36 h, the cells were harvested to detect the interaction between nsp5, RhoA and Smurf1 by ternary Co-IP. (F) HEK-293T cells were co-transfected with the plasmids of RhoA, Smurf1 and Ub for 36 h, the cells were harvested to detect the ubiquitination of RhoA induced by Smurf1. (G) HEK-293T cells were co-transfected with the plasmids of RhoA, Smurf1, nsp5 and Ub for 36 h, the cells were harvested to detect the effect of nsp5 on Smurf1-induced polyubiquitination of RhoA. (H) HEK-293T cells were co-transfected with the plasmids of RhoA, Smurf1, nsp5 and K63-Ub for 36 h, the cells were harvested to detect the effect of nsp5 on Smurf1-induced polyubiquitination of RhoA. (I) HEK-293T cells were co-transfected with the plasmids of RhoA, nsp5 and Smurf1 for 36 h, the cells were harvested to detect the RhoA, nsp5 and Smurf1 protein expression by Western blot, tubulin served as an internal control. (J) Smurf1+/+ and Smurf1-/- HEK-293T cells were co-transfected with the plasmids of RhoA and nsp5 for 36 h, the cells were harvested to detect the RhoA, nsp5 and Smurf1 protein expression by Western blot, tubulin served as an internal control. (K) Smurf1-/- HEK-293T cells were co-transfected with the plasmids of RhoA, nsp5 and Smurf1 for 36 h, the cells were harvested to detect the RhoA, nsp5 and Smurf1 protein expression by Western blot, tubulin served as an internal control. (L) PAMs were transfected with siNC and siSmurf1 for 24 h, and then infected with recombinant lentivirus expressing nsp5 or control lentivirus for 24 h, the cells were harvested to detect the RhoA, nsp5 and Smurf1 protein expression by Western blot, tubulin served as an internal control. Scale bar: 10 μm.

RhoA has no effect on PRRSV-2 replication.(A-D) Immortalized PAMs were transfected with siNC or siRhoA for 24 h and then infected with PRRSV-2 for 36 h, cells and supernatants were harvested to determine (A) RhoA mRNA expression, (B) PRRSV-2 ORF7 mRNA expression, (C) RhoA and PRRSV-2 N protein expression, tubulin served as an internal control, (D) supernatant virus titer. (E-F) Immortalized PAMs were transfected with pEGFP-C1 (vector) or pEGFP-C1-RhoA for 24 h, followed by infection with PRRSV-2 for the indicated periods, cells and supernatants were harvested to determine (E) RhoA and PRRSV-2 N protein expression, tubulin served as an internal control, (F) supernatant virus titer. P values were calculated using Student’s t-test. ns, not significant.

Schematic model for the mechanisms that PRRSV-2 facilitates the survival of secondary infected bacteria.During PRRSV-2 infection, PRRSV-2 nsp5 interacts with RhoA, inducing the autophagy-lysosome degradation of RhoA and resulting in a decrease in F-actin. The reduction of F-actin impedes the efficiency of lysosomal targeted clearance of bacteria, thereby facilitating the survival of secondary infected bacteria.

Full text

Introduction

Porcine reproductive and respiratory syndrome (PRRS) causes sow reproductive failure and respiratory disease in swine of all ages [1]. The causative agent of the disease is PRRS virus (PRRSV). PRRSV is a positive-sense, single stranded RNA virus belonging to the Arteriviridae family in the order Nidovirales that is divided into two species, PRRSV-1 (formerly European genotype 1) and PRRSV-2 (formerly North American genotype 2) [2,3]. Current treatment strategies cannot effectively control PRRSV, which affects the global swine industry and leads to significant economic losses each year. As a typical immunosuppressive virus, PRRSV infection typically damages the thymus, lymph nodes, spleen and other immune organs, resulting in a low immune capacity of the pigs and thereby creating favorable conditions for the secondary infection of other viruses or bacteria, such as type 2 porcine circovirus (PCV2), Streptococcus suis (S. suis), Hemophilus parasuis (Hps), Salmonella cholera (S. choleraesuis) and Mycoplasma hyopneumoniae (Mhp), which play important roles in the pathogenesis of PRRSV infection [4–6]. In recent years, due to the abuse of antibiotics, an increasing number of pathogenic bacteria have been isolated from swine infected with PRRSV. Among them, Aerococcus viridans (A. viridans) has gradually attracted attention due to its high isolation rate in the swine industry [7–9].

A. viridans is a gram-positive microaerophilic coccus that belongs to the family Aerococcaceae. For a long time, A. viridans was considered to be the only species of Aerococcus [10]. In recent years, the development of microbial detection and identification technology has greatly aided microbial classification and identification. Thus far, multiple species of Aerococcus have been described, of which A. viridans is representative [11]. As an opportunistic pathogen, A. viridans is widely distributed in the environment, that is, in the air, water, and soil, and it has been clinically associated with endocarditis, arthritis and urinary infections in humans and swine and with septicemia and fatal infection in crustaceans and fish. A. viridans is easily misidentified as Staphylococcus and Streptococcus species, which may be why the pathogenicity of A. viridans has been underestimated [12–14]. Previously, we isolated a strain of A. viridans from pigs infected with PRRSV-2. Pigs co-infected with A. viridans and PRRSV-2 caused more severe inflammatory damage compared to those infected with PRRSV-2 alone. However, it remains unknown whether PRRSV-2 promotes secondary infection with A. viridans.

Macrophages form the host’s first line of innate immune defense against bacterial invasion. Lysosomes, which are rich in various hydrolytic enzymes, play a crucial role in the process of phagocytosis and bacterial killing [15–17]. Invading bacteria are engulfed through the endocytic or phagocytic pathway and delivered to the lysosomes to form the phagolysosome, an organelle that functions to ultimately eliminate the bacteria [18,19]. Some bacteria develop elaborate regulatory strategies for lysosomes and phagosomes to avoid lysosomal degradation. For instance, Mycobacterium tuberculosis (Mtb) can prevent the fusion of phagosomes with lysosomes by expressing lipoarabinomannan on its cell envelope [20]. Leishmania donovani (L. donovani), and Salmonella enterica (S. enterica) also form membrane-bound vacuoles that acquire lysosomal membrane markers to escape lysosomal phagocytosis [21,22]. On the other hand, host cells also develop countermeasures to regulate lysosomal homeostasis in response to pathogen invasion. For example, upon sensing Mtb infection, host cells induce an increase in lysosomal content and activity in macrophages in response to Mtb infection via mechanistic target of rapamycin complex 1 (mTORC1) and the transcription factor EB (TFEB), and macrophages also facilitate the transportation of phagosomes to lysosomes in order to promote the degradation of Mtb [23].

The actin cytoskeleton, which is the assemblage of actin filaments with their accessory and regulatory proteins, serves as the principal mechanism for generating force within the cell [24,25]. Through the dynamic recombination of globular actin monomer (G-actin) into actin filament (F-actin), cells can move, deform, and drive the intracellular movement of membrane organelles [26,27]. The process of lysosome targeting phagocytic bacteria is highly reliant on the movement of organelles driven by F-actin. For instance, Caspase-11 modulates actin polymerization via cofilin, facilitates the fusion of Legionella pneumophila with lysosomes, thereby inhibiting Legionella pneumophila infection in macrophages [28]. NLR family, CARD domain containing 4 (NLRC4) induces F-actin formation via the Arp2/3 complex, thereby enhancing lysosomal-targeted phagocytosis of Vibrio splendidus in sea cucumbers [29]. It is widely known that the primary target cells for PRRSV-2 infection are porcine alveolar macrophages (PAMs), and bacterial secondary infection frequently occurs following PRRSV-2 infection [30,31]. However, it remains unknown whether PRRSV-2 has a role in the lysosomal phagocytosis of the secondary infected bacteria. Here, we report that the degradation of the host protein RhoA due to PRRSV-2 infection results in intracellular depolymerization of F-actin, thereby impeding the lysosomal targeted clearance of secondary infected bacteria. Our study is the first to reveal the molecular mechanism through which PRRSV-2 reprograms the cytoskeleton to facilitate the survival of bacteria with secondary infections. The details are described below.

Results

PRRSV-2 facilitates A. viridans secondary infection and induces a violent inflammatory response in vitro

PRRSV-2 is known to cause secondary bacterial infections and more intense inflammatory responses. Here, we take A. viridans as a model to elaborate on the molecular mechanism of PRRSV-2 secondary bacterial infection in detail. To explore whether PRRSV-2 enhances the invasion of A. viridans, immortalized PAMs (3D4/21-CD163) were infected or not infected with PRRSV-2, and then infected with A. viridans. The colony-forming units (CFUs) corresponded to the numbers of intracellular live bacteria. We detected the number of invading bacteria in the bacteria-only group and the co-infection group and found that there were significantly more invading bacteria in the co-infection group than in the bacteria-only group (Fig 1A). Upon observing the impact of PRRSV-2 on A. viridans by immunofluorescence, we noticed that compared with the control, PRRSV-2 significantly promoted invasion by A. viridans (Fig 1B). Subsequently, we investigated the level of cellular inflammatory response in the context of PRRSV-2 and A. viridans co-infection. The results showed that the expression level of inflammatory factors in the cells with co-infection was generally higher than that in the cells with PRRSV-2 or bacterial infection alone (Fig 1C-1F). To visualize A. viridans in PRRSV-2-infected and uninfected immortalized PAMs, we analyzed macrophages inoculated with A. viridans using transmission electron microscopy. The bacteria observed in macrophages in the absence of PRRSV-2 infection exhibited morphological features indicative of degradation, such as irregular or disrupted margins (arrows in Fig 1G). In contrast, intracellular bacteria within PRRSV-2-infected macrophages maintained an intact structure and displayed no evidence of degradation. These results imply that PRRSV-2 facilitates the secondary infection of A. viridans and triggers a more intense inflammatory response.

*PRRSV-2 facilitates the secondary infection of A. viridans.(A) Immortalized PAMs were infected or not infected with PRRSV-2 for 24 h and then infected with A. viridans for 4 h, the cells were harvested to detect the number of bacteria by CFUs. (B) Immortalized PAMs were infected or not infected with PRRSV-2 for 24 h and then infected with A. viridans for 4 h, the cells were harvested to detect the number of bacteria by immunofluorescence assay. (C-F) Immortalized PAMs were infected with PRRSV-2 for 24 h and then infected with A. viridans for 4 h and 8 h, the cells were harvested to detect (C) IL-1β mRNA expression, (D) IL-6 mRNA expression, (E) CCL4 mRNA expression, (F) P65 mRNA expression. The cells infected with PRRSV-2 and A. viridans respectively were taken as the control. (G) Immortalized PAMs were infected with PRRSV-2 for 24 h and then infected with A. viridans for 4 h, the cells were harvested to detect the bacterial morphology by transmission electron microscopy. Scale bar: 10 μm. P values were calculated using Student’s t-test. ns, not significant; *, P < 0.05; **, P < 0.01; **, P < 0.001.

PRRSV-2 facilitates A. viridans secondary infection in piglets and triggers pronounced inflammatory responses

To further elucidate the impact of PRRSV-2 on A. viridans infection in vivo, 15 piglets were randomly assigned to three experimental groups (n = 5 per group) and inoculated with PRRSV-2 and A. viridans as specified in Fig 2A. The two groups of piglets infected with PRRSV-2 developed fever by day 4, with peak body temperatures observed on day 6 (Fig 2B), averaging 40.8°C and 41.0°C, respectively, indicating successful establishment of the PRRSV-2 infection model. Post-mortem examination further confirmed this finding through the detection of PRRSV-2 nucleocapsid (N) protein in PAMs of the infected piglets (Fig 2C). During the experiment, 1 piglet died in the A. viridans group, whereas 4 piglets died in the PRRSV-2 + A. viridans group (Fig 2D). Quantitative analysis of A. viridans in the lungs of piglets by CFUs revealed that the bacterial load in the PRRSV-2 pre-infection group was significantly higher than that in the group infected with A. viridans alone (Fig 2E), indicating that PRRSV-2 infection facilitated the secondary infection caused by A. viridans in piglets. We further examined the levels of inflammatory factors in PAMs and serum samples from piglets, and found that the expression levels of IL-1β and IL-6 in the PRRSV-2 + A. viridans group were significantly higher than those in both the PRRSV-2-only and A. viridans-only groups (Fig 2F-2I). Moreover, more obvious microscopic lesions were observed in the PRRSV-2 + A. viridans group, including thickening of the alveolar septum and shrinkage of the alveolar space (Fig 2J). These findings suggest that PRRSV-2 facilitates secondary infection with A. viridans and exacerbates the host inflammatory response.

*PRRSV-2 facilitates A. viridans secondary infection in piglets and triggers pronounced inflammatory responses.(A) Flow chart of the animal experiments. Fifteen piglets were randomly divided into three groups, with 5 piglets in each group. PRRSV-2 was infected on day 1, followed by A. viridans infection on day 7, as illustrated in the figure. On day 9, the surviving piglets were humanely euthanized in accordance with established ethical guidelines. (B) The rectal temperature of the piglets was recorded daily. (C) Following the death of piglets, PAMs were isolated and the expression of the PRRSV-2 N protein was examined by Western blot, tubulin served as an internal control. (D) The survival rate of the piglets in each group. (E) The lung tissues of piglets were homogenized, and bacterial load was quantitatively assessed via CFUs. (F-G) The expression levels of (F) IL-1β, (G) IL-6 in PAMs of piglets were quantified using RT-qPCR. (H-I) The expression levels of (H) IL-1β, (I) IL-6 in serum of piglets were quantified using ELISA. (J) Lung tissue sections from piglets in the three groups were stained with hematoxylin and eosin (HE). P values were calculated using Student’s t-test. *, P < 0.05; **, P < 0.001.

PRRSV-2 infection induces F-actin depolymerization thereby preventing lysosomes from targeting bacteria

Macrophages are typically able to actively phagocytose pathogens and eliminate them via lysosomal digestion within the cell. We were curious about how A. viridans, secondary to PRRSV-2 infection, evades targeted clearance by macrophage lysosomes. Here, we explored whether there is a defect in the fusion of A. viridans with lysosomes during PRRSV-2 infection, enabling the bacteria to evade degradation within macrophages. Immortalized PAMs were infected or not infected with PRRSV-2 and then infected with A. viridans for 4 h. The intracellular lysosomes and bacterial subcellular localization were analyzed through immunofluorescence assay, and the degree of co-localization between bacteria and lysosomes was quantitatively assessed using the Manders’ colocalization coefficients. As shown in Fig 3A, the proportion of bacteria co-localizing with lysosomes was significantly higher in uninfected cells compared to those infected with PRRSV-2, indicating that PRRSV-2 infection hindered the targeted clearance function of the lysosomes on the bacteria. This finding serves as crucial evidence that the number of viable bacteria in PRRSV-2-infected cells is higher than that in non-PRRSV-2-infected cells (Fig 1A). As intracellular organelles, the directed transport of lysosomes is highly reliant on the cytoskeletal structure of actin filaments. We hypothesize that PRRSV-2 infection disrupts the cytoskeleton, resulting in the impairment of lysosomal directed transport efficiency and thus hindering the targeted clearance of bacteria by lysosomes. Therefore, we examined the influence of PRRSV-2 infection on intracellular F-actin, and the results indicated that F-actin was notably decreased and disordered in PRRSV-2-infected cells as compared with uninfected cells (Fig 3B), which might be the direct cause for PRRSV-2 preventing lysosomal targeting of bacteria. Subsequently, we performed a conservative validation using an additional prevalent PRRSV-2 subtype—the NADC30-like strain. Our findings demonstrate that infection with the NADC30-like PRRSV-2 strain similarly disrupts the intracellular F-actin (S1A Fig), and effectively prevents the co-localization of lysosomes and bacteria (S1B Fig), thereby facilitating the survival of the bacteria within infected cells (S1C Fig). Furthermore, S. suis 2 and K. pneumoniae were also used for confirmatory validation. The results showed that PRRSV-2 infection significantly impaired the co-localization signal of bacteria and lysosomes, thereby effectively facilitated the survival of S. suis 2 and K. pneumoniae (S1D-S1G Fig in S1 File). Finally, cells were pre-treated with F-actin inhibitor Latrunculin A (LAT-A) in the absence of PRRSV-2 infection, followed by A. viridans infection. The immunofluorescence results indicated that LAT-A markedly inhibited F-actin synthesis (Fig 3C), and the number of bacteria within the cells increased significantly (Fig 3D). By observing the intracellular subcellular localization of lysosomes and bacteria, we discovered that the application of LAT-A also impeded the transport efficiency of lysosomes targeting bacteria (Fig 3E). Here, we demonstrate that PRRSV-2 infection induces F-actin depolymerization thereby preventing lysosomes from targeting bacteria.

*PRRSV-2 induces the depolymerization of F-actin and thereby impedes the lysosome-mediated clearance of bacteria.(A) Immortalized PAMs were infected or not infected with PRRSV-2 for 24 h and then infected with A. viridans for 4 h, the cells were harvested to detect the subcellular localization for lysosome and A. viridans. Colocalization between lysosome and A. viridans was quantified using ImageJ, with results expressed as the Manders’ correlation coefficients (M1 and M2). Specifically, M1 represents the proportion of lysosomal signal that overlaps with the bacterial signal relative to the total lysosomal signal, while M2 represents the proportion of bacterial signal overlapping with the lysosomal signal relative to the total bacterial signal. n = 20 cells from three independent experiments. (B) Immortalized PAMs were infected or not infected with PRRSV-2 for 24 h, the cells were harvested to observe the changes in F-actin. The fluorescence intensity of F-actin was quantified using ImageJ, n = 20 cells from three independent experiments. (C) Immortalized PAMs were incubated with DMSO or LAT-A for 12 h, the cells were harvested to observe the changes in F-actin. The fluorescence intensity of F-actin was quantified using ImageJ, n = 20 cells from three independent experiments. (D) Immortalized PAMs were incubated with DMSO or LAT-A for 12 h and then infected with A. viridans for 4 h, the cells were harvested to detect the number of bacteria by CFUs. (E) Immortalized PAMs were incubated with DMSO or LAT-A for 12 h and then infected with A. viridans for 4 h, the cells were harvested to detect the subcellular localization for lysosome and A. viridans. Colocalization between lysosome and A. viridans was quantified using ImageJ, with results expressed as the Manders’ correlation coefficients (M1 and M2). Specifically, M1 represents the proportion of lysosomal signal that overlaps with the bacterial signal relative to the total lysosomal signal, while M2 represents the proportion of bacterial signal overlapping with the lysosomal signal relative to the total bacterial signal. n = 20 cells from three independent experiments. Scale bar: 10 μm. P values were calculated using Student’s t-test. **, P < 0.01, **, P < 0.001.

RhoA promotes F-actin polymerization and accelerates lysosomal targeting of bacteria

The Ras-related GTPase family plays a significant role in the assembly of actin stress fibers during cytoskeletal regulation. Among the members of the GTPase family, RhoA is the crucial protein that promotes the polymerization of G-actin to F-actin [32]. We wanted to determine the role of RhoA-mediated F-actin polymerization in the process of A. viridans infection from the perspective of the host. The specific siRNA of RhoA was transfected into immortalized PAMs, and the 3# siRNA, which demonstrated the most effective knockdown, was selected for subsequent experiments (Fig 4A). As expected, RhoA knockdown inhibited intracellular F-actin polymerization (Fig 4B), and the number of viable bacteria within the cells was higher than that in the control cells (Fig 4C). By observing the intracellular lysosomes and the subcellular localization of bacteria, it was found that RhoA knockdown effectively hindered the lysosomal targeting of bacteria (Fig 4D). Correspondingly, we discovered that overexpression of RhoA facilitated intracellular F-actin polymerization (Fig 4E), and the number of viable bacteria within the cells was lower than that in the control cells (Fig 4F). Finally, we evaluated the impact of RhoA overexpression on the transport efficiency of lysosomal targeted bacteria. The results showed that during the first and second hours following bacterial infection, RhoA overexpression significantly enhanced the co-localization of bacteria with lysosomes (Fig 4G). These data imply that RhoA promotes the lysosomal targeting of bacteria by facilitating F-actin polymerization.

*RhoA promotes F-actin polymerization and accelerates lysosomal targeting of bacteria.(A) Immortalized PAMs were transfected with siNC and three distinct RhoA-specific siRNA for 36 h, the cells were harvested to detect the RhoA protein expression by Western blot, tubulin served as an internal control. (B) Immortalized PAMs were transfected with siNC or siRhoA for 36 h, the cells were harvested to observe the changes in F-actin. The fluorescence intensity of F-actin was quantified using ImageJ, n = 20 cells from three independent experiments. (C) Immortalized PAMs were transfected with siNC or siRhoA for 36 h and then infected with A. viridans for 4 h, the cells were harvested to detect the number of bacteria by CFUs. (D) Immortalized PAMs were transfected with siNC or siRhoA for 36 h and then infected with A. viridans for 4 h, the cells were harvested to detect the subcellular localization for lysosome and A. viridans. Colocalization between lysosome and A. viridans was quantified using ImageJ, with results expressed as the Manders’ correlation coefficients (M1 and M2). Specifically, M1 represents the proportion of lysosomal signal that overlaps with the bacterial signal relative to the total lysosomal signal, while M2 represents the proportion of bacterial signal overlapping with the lysosomal signal relative to the total bacterial signal. n = 20 cells from three independent experiments. (E) Immortalized PAMs were transfected with p3 × Flag (vector) or p3 × Flag-RhoA for 36 h, the cells were harvested to observe the changes in F-actin. The fluorescence intensity of F-actin was quantified using ImageJ, n = 20 cells from three independent experiments. (F) Immortalized PAMs were transfected with p3 × Flag (vector) or p3 × Flag-RhoA for 36 h and then infected with A. viridans for 4 h, the cells were harvested to detect the number of bacteria by CFUs. (G) Immortalized PAMs were transfected with p3 × Flag (vector) or p3 × Flag-RhoA for 36 h and then infected with A. viridans for the indicated periods, the cells were harvested to detect the subcellular localization for lysosome and A. viridans. Colocalization between lysosome and A. viridans was quantified using ImageJ, with results expressed as the Manders’ correlation coefficients (M1 and M2). Specifically, M1 represents the proportion of lysosomal signal that overlaps with the bacterial signal relative to the total lysosomal signal, while M2 represents the proportion of bacterial signal overlapping with the lysosomal signal relative to the total bacterial signal. n = 20 cells from three independent experiments. Scale bar: 10 μm. P values were calculated using Student’s t-test. ns, not significant; **, P < 0.01; **, P < 0.001.

PRRSV-2 down-regulates RhoA expression through nsp5

Since RhoA plays a crucial role in the process of F-actin polymerization, and PRRSV-2 infection leads to F-actin depolymerization, we wondered whether PRRSV-2 achieves this function by regulating RhoA. Thus, we examined the expression trend of RhoA in PRRSV-2-infected cells. As shown in Fig 5A and 5B, PRRSV-2 infection had no impact on the transcription of RhoA mRNA, but inhibited the expression of RhoA protein. Based on this, we further sifted through the specific viral proteins that cause the down-regulation of RhoA protein expression. The results showed that the expression of RhoA protein was significantly decreased by nsp5 in a concentration-gradient-dependent manner (Fig 5C and 5D), and there was an interaction between nsp5 and RhoA (Fig 5E and 5F), another viral protein nsp2 did not interact with RhoA as a negative control (Fig 5G). The subcellular localization of nsp5 and nsp2, along with the calculation of the Pearson’s correlation coefficient with RhoA, further substantiated the specific interaction between nsp5 and RhoA (Fig 5H). To identify the crucial functional regions of nsp5-induced down-regulation of RhoA protein expression, we analyzed the polypeptide sequence of nsp5 and constructed four truncated plasmids named D1 to D4 based on the four transmembrane structures of nsp5 (Fig 5I and 5J). The results showed that the full-length nsp5, D2, D3, and D4 could markedly induce the down-regulation of RhoA protein expression, except for D1 (Fig 5K). Finally, we targeted the functional region of nsp5 to amino acids 86–122 and constructed a truncated plasmid named D5. As expected, both D5 and nsp5 could significantly induce the down-regulation of RhoA protein expression (Fig 5L). Here, we show that PRRSV-2 leads to the down-regulation of RhoA protein expression through nsp5 and locks the key functional region of nsp5 to amino acids 86–122.

*PRRSV-2 down-regulates RhoA expression through nsp5.(A-B) PAMs were infected with PRRSV-2 (MOI = 1) for the indicated periods, and the cells were harvested to detect (A) RhoA mRNA expression by RT-qPCR, (B) RhoA protein expression by Western blot, tubulin served as an internal control, and the PRRSV-2 N protein was used as an infection indicator. (C) HEK-293T cells were co-transfected with the plasmids of PRRSV-2 proteins and RhoA for 36 h, and the cells were harvested to detect the PRRSV-2 proteins and RhoA protein expression by Western blot, tubulin served as an internal control. (D) HEK-293T cells were co-transfected with the plasmids of RhoA and nsp5 with an increasing concentration gradient for 36 h, the cells were harvested to detect the RhoA and PRRSV-2 nsp5 protein expression by Western blot, tubulin served as an internal control. (E-F) HEK-293T cells were co-transfected with the plasmids of RhoA and nsp5 for 36 h, the cells were harvested to detect the interaction between RhoA and nsp5 by Co-IP. (G) HEK-293T cells were co-transfected with the plasmids of RhoA and nsp2 for 36 h, the cells were harvested to detect the interaction between RhoA and nsp2 by Co-IP. (H) HEK-293T cells were co-transfected with the designated plasmids for 36 h, the cells were harvested to detect the subcellular localization of nsp5 and nsp2 in relation to RhoA, respectively. The co-localization of nsp2 and nsp5 with RhoA was analyzed using ImageJ, and the results were quantified using the Pearson’s correlation coefficient. n = 10 cells from three independent experiments. (I) Online prediction of the transmembrane domain distribution of nsp5 polypeptide sequences. (J) The truncated plasmids were constructed based on the transmembrane domain distribution of nsp5. (K) HEK-293T cells were co-transfected with the plasmids of RhoA and the truncated plasmid of nsp5 for 36 h, the cells were harvested to detect the RhoA and the truncated nsp5 protein expression by Western blot, tubulin served as an internal control. (L) The plasmid of RhoA was co-transfected with nsp5 and nsp5-D5 into HEK-293T cells for 36 h respectively, the cells were harvested to detect the RhoA, nsp5 and nsp5-D5 protein expression by Western blot, tubulin served as an internal control. Scale bar: 10 μm. P values were calculated using Student’s t-test. **, P < 0.001.

Nsp5 degrades RhoA through the autophagy-lysosome pathway by interaction with TOLLIP

The ubiquitin-proteasome pathway and the autophagy-lysosome pathway are the main intracellular protein degradation pathways. To determine the degradation pathway of RhoA, we initially co-transfected the eukaryotic expression plasmids of nsp5 and RhoA, and subsequently treated the cells with the proteasome inhibitor MG132, the autophagy inhibitor 3-MA, and the apoptosis inhibitor Z-VAD. As shown in Fig 6A, the down-regulation of RhoA protein expression induced by nsp5 can only be restored under 3-MA treatment. Further confocal microscopy analyses confirmed that the presence of nsp5 enhances the co-localization of RhoA with the autophagy marker LC3 (Fig 6B). Autophagy-related gene 7 (ATG7) serves as a critical mediator in the process of autophagic degradation. Knockout of ATG7 was found to completely block nsp5-mediated RhoA degradation (Fig 6C), and exogenous expression of ATG7 restored the nsp5-mediated degradation of RhoA (Fig 6D), indicating that nsp5 induces RhoA protein degradation via an autophagy-dependent mechanism. To identify the cargo receptor participating in nsp5-mediated autophagy, we used Co-IP experiments and discovered that both nsp5 and RhoA interacted with TOLLIP, but not with TAX1 BP1, NBR1, PSMD4, OPTN (Fig 6E-6H). The interactions between nsp5 and RhoA as well as TOLLIP were further confirmed in primary PAMs (Fig 6I). Colocalization of nsp5, RhoA, and TOLLIP was also confirmed by confocal microscopy analysis (Fig 6J). Moreover, a ternary Co-IP assay provided direct evidence for the formation of a nsp5-RhoA-TOLLIP complex (Fig 6K). Overexpressed TOLLIP enhanced the nsp5-induced degradation of RhoA (Fig 6L), whereas TOLLIP knockout attenuated this effect (Fig 6M). Furthermore, in TOLLIP-knockout cells, exogenous expression of TOLLIP restored nsp5-mediated RhoA protein degradation (Fig 6N), indicating that TOLLIP acts as a receptor for nsp5-mediated autophagy pathway responsible for RhoA degradation. The critical role of TOLLIP in mediating nsp5-induced degradation of endogenous RhoA protein was further validated in primary PAMs (Fig 6O).

*Nsp5 degrades RhoA through the autophagy pathway by interaction with TOLLIP.(A) HEK-293T cells were co-transfected with the plasmids of RhoA and nsp5 for 24 h and then incubated with MG132, Z-VAD and 3-MA for 6 h, the cells were harvested to detect the RhoA and nsp5 protein expression by Western blot, tubulin served as an internal control. (B) HEK-293T cells were co-transfected with the designated plasmids for 36 h, the cells were harvested to detect the subcellular localization of RhoA and LC3. The co-localization of RhoA and LC3 was analyzed using ImageJ, and the results were quantified using the Pearson’s correlation coefficient. n = 10 cells from three independent experiments. (C) ATG7+/+ and ATG7-/- HEK-293T cells were co-transfected with the plasmids of RhoA and nsp5 for 36 h, the cells were harvested to detect the RhoA, nsp5 and ATG7 protein expression by Western blot, tubulin served as an internal control. (D) ATG7-/- HEK-293T cells were co-transfected with the plasmids of RhoA, nsp5 and ATG7 for 36 h, the cells were harvested to detect the RhoA, nsp5 and ATG7 protein expression by Western blot, tubulin served as an internal control. (E) The plasmid of nsp5 was co-transfected with TAX1 BP1, NBR1, TOLLIP, PSMD4 and OPTN into HEK-293T cells for 36 h respectively, the cells were harvested to detect autophagy receptors that interact with nsp5 by Co-IP. (F) HEK-293T cells were co-transfected with the plasmids of nsp5 and TOLLIP for 36 h, the cells were harvested to detect the interaction between nsp5 and TOLLIP by Co-IP. (G) The plasmid of RhoA was co-transfected with TAX1 BP1, NBR1, TOLLIP, PSMD4 and OPTN into HEK-293T cells for 36 h respectively, the cells were harvested to detect autophagy receptors that interact with RhoA by Co-IP. (H) HEK-293T cells were co-transfected with the plasmids of RhoA and TOLLIP for 36 h, the cells were harvested to detect the interaction between RhoA and TOLLIP by Co-IP. (I) PAMs were infected with recombinant lentivirus expressing nsp5 or control lentivirus for 24 h, the cells were harvested to detect the interaction between nsp5 and endogenous RhoA, as well as endogenous TOLLIP by Co-IP. (J) HEK-293T cells were co-transfected with Flag-TOLLIP, Myc-RhoA, and HA-nsp5 plasmids for 36 h, the cells were harvested to detect the subcellular localization of TOLLIP, RhoA and nsp5. The co-localization of TOLLIP, RhoA and nsp5 was analyzed using ImageJ, and the results were quantified using the Pearson’s correlation coefficient. n = 10 cells from three independent experiments. (K) HEK-293T cells were co-transfected with HA-nsp5, Flag-TOLLIP and Myc-RhoA plasmids for 36 h, the cells were harvested to detect the interaction between nsp5, RhoA and TOLLIP by ternary Co-IP. (L) HEK-293T cells were co-transfected with the plasmids of RhoA, nsp5 and TOLLIP for 36 h, the cells were harvested to detect the RhoA, nsp5 and TOLLIP protein expression by Western blot, tubulin served as an internal control. (M) TOLLIP+/+ and TOLLIP-/- HEK-293T cells were co-transfected with the plasmids of RhoA and nsp5 for 36 h, the cells were harvested to detect the RhoA, nsp5 and TOLLIP protein expression by Western blot, tubulin served as an internal control. (N) TOLLIP-/- HEK-293T cells were co-transfected with the plasmids of RhoA, nsp5 and TOLLIP for 36 h, the cells were harvested to detect the RhoA, nsp5 and TOLLIP protein expression by Western blot, tubulin served as an internal control. (O) PAMs were transfected with siNC and siTOLLIP for 24 h, and then infected with recombinant lentivirus expressing nsp5 or control lentivirus for 24 h, the cells were harvested to detect the RhoA, nsp5 and TOLLIP protein expression by Western blot, tubulin served as an internal control. Scale bar: 10 μm. P values were calculated using Student’s t-test. **, P < 0.001.

In the autophagy-lysosome pathway, the substrate protein is ubiquitinated and subsequently transported by the cargo receptor to the lysosome for degradation. Substrate proteins typically undergo ubiquitination at lysine residues. Therefore, we investigated the effect of nsp5 on RhoA ubiquitination. As shown in Fig 7A and 7B, the presence of nsp5 significantly enhanced the ubiquitination of both exogenous and endogenous RhoA proteins. Consequently, we observed that nsp5 enhanced the co-localization of RhoA with the lysosomal marker protein LAMP1 (Fig 7C). To investigate which polyubiquitination modification of RhoA was mediated by nsp5, seven types of ubiquitin plasmids (K6-Ub, K11-Ub, K27-Ub, K29-Ub, K33-Ub, K48-Ub, and K63-Ub) were co-transfected with nsp5 and RhoA, respectively. The results showed that K63-linked ubiquitination of RhoA occurred most frequently upon nsp5 overexpression (Fig 7D). Through peptide sequence analysis, a total of 17 lysine residues in RhoA were identified. These 17 lysine residues in RhoA were converted to alanine residues for ubiquitination site confirmation. We found that co-transfection of RhoA^K187R^ with nsp5 prevented the degradation of RhoA by nsp5, while the other 16 mutants did not (Fig 7E-7H). Furthermore, we co-transfected the RhoA^K187R^ and nsp5 plasmids with HA-Ub. As shown in Fig 7I, the polyubiquitination of RhoA induced by nsp5 was inhibited after the K187 residues mutation of RhoA. Together, these results indicate that nsp5 targets K187 residues of RhoA and promotes K63-linked polyubiquitination, which is degraded through the autophagy-lysosomal pathway by the autophagy receptor TOLLIP. Finally, we assessed the functional impact of the RhoA^K187R^ on bacterial infection using a rescue experiment. The results showed that overexpression of RhoA^K187R^ in the condition of knocked-down RhoA significantly reduced the number of viable bacteria within the cells (Fig 7J and 7K). Furthermore, confocal microscopy analysis revealed that RhoA^K187R^ significantly enhanced lysosomal targeting of internalized bacteria (Fig 7L).

*Nsp5 induces polyubiquitination of RhoA.(A) HEK-293T cells were co-transfected with the plasmids of RhoA, nsp5 and Ub for 36 h, the cells were harvested to detect the ubiquitination of RhoA induced by nsp5. (B) PAMs were infected with recombinant lentivirus expressing nsp5 or control lentivirus for 24 h, the cells were harvested to detect the ubiquitination of endogenous RhoA induced by nsp5. (C) HEK-293T cells were co-transfected with the designated plasmids for 36 h, the cells were harvested to detect the subcellular localization of RhoA and LAMP1. The co-localization of RhoA and LAMP1 was analyzed using ImageJ, and the results were quantified using the Pearson’s correlation coefficient. n = 10 cells from three independent experiments. (D) Seven types of ubiquitin plasmids (K6-Ub, K11-Ub, K27-Ub, K29-Ub, K33-Ub, K48-Ub, and K63-Ub) were co-transfected in HEK-293T cells with nsp5 and RhoA respectively for 36 h, the cells were harvested to detect the ubiquitination type of RhoA induced by nsp5. (E-H) The indicated lysine residue mutant plasmids of RhoA were co-transfected in HEK-293T cells with nsp5 for 36 h, the cells were harvested to detect the RhoA and nsp5 protein expression by Western blot, tubulin served as an internal control. (I) HEK-293T cells were co-transfected with the plasmids of RhoAK187R, nsp5 and Ub for 36 h, the cells were harvested to detect the ubiquitination of RhoAK187R induced by nsp5. (J) Immortalized PAMs were transfected with siNC and siRhoA for 24 h, and then transfected with the plasmids of RhoAK187R for 24 h, the cells were harvested to detect the RhoA protein expression by Western blot, tubulin served as an internal control. (K) Immortalized PAMs were transfected with siNC and siRhoA for 24 h, and then transfected with the plasmids of RhoAK187R for 24 h. Subsequently, cells were infected with A. viridans for 4 h, the cells were harvested to detect the number of bacteria by CFUs. (L) Immortalized PAMs were transfected with siNC and siRhoA for 24 h, and then transfected with the plasmids of RhoAK187R for 24 h. Subsequently, cells were infected with A. viridans for 4 h, the cells were harvested to detect the subcellular localization for lysosome and A. viridans. Colocalization between lysosome and A. viridans was quantified using ImageJ, with results expressed as the Manders’ correlation coefficients (M1 and M2). Specifically, M1 represents the proportion of lysosomal signal that overlaps with the bacterial signal relative to the total lysosomal signal, while M2 represents the proportion of bacterial signal overlapping with the lysosomal signal relative to the total bacterial signal. n = 20 cells from three independent experiments. Scale bar: 10 μm. P values were calculated using Student’s t-test. **, P < 0.01, **, P < 0.001.

Nsp5 potentiates Smurf1-mediated ubiquitination and degradation of RhoA

Given that nsp5 lacks E3 ubiquitin ligase activity, we hypothesize that nsp5 may serve as a molecular scaffold to facilitate the interaction between RhoA and its cognate E3 ubiquitin ligase, thereby promoting ubiquitination and subsequent degradation of RhoA. Previous studies have identified several E3 ubiquitin ligases involved in RhoA degradation, including Smurf1, Smurf2, and Cul3 [33–35]. Co-IP experiments demonstrate that nsp5 specifically interacts with Smurf1, but not with Smurf2 or Cul3 (Fig 8A and 8B). The interactions between nsp5 and RhoA as well as Smurf1 were further confirmed in primary PAMs (Fig 8C). Confocal microscopy analysis and ternary Co-IP assay provided direct evidence for the formation of a nsp5-RhoA-Smurf1 complex (Fig 8D and 8E). Subsequently, we investigated whether Smurf1 mediates the polyubiquitination and degradation of RhoA. It was observed that Smurf1 mediates the polyubiquitination of RhoA (Fig 8F), while nsp5 enhances this modification by promoting the interaction between Smurf1 and RhoA, thereby increasing the level of RhoA polyubiquitination (Fig 8G), and K63-linked polyubiquitination (Fig 8H). We also found that Smurf1 promotes the degradation of the RhoA protein, with a more pronounced reduction in RhoA protein levels in the presence of nsp5 (Fig 8I). However, the knockout of Smurf1 reversed the nsp5-induced degradation of the RhoA protein (Fig 8J). Furthermore, in Smurf1-knockout cells, exogenous expression of Smurf1 restored nsp5-mediated RhoA protein degradation (Fig 8K), indicating that nsp5 promotes the polyubiquitination of RhoA in a Smurf1-dependent manner, thereby leading to RhoA degradation. The critical role of Smurf1 in mediating nsp5-induced degradation of endogenous RhoA protein was further validated in primary PAMs (Fig 8L).

Nsp5 potentiates Smurf1-mediated ubiquitination and degradation of RhoA.(A) The plasmid of nsp5 was co-transfected with Smurf1, Smurf2, Cul3 and RhoA into HEK-293T cells for 36 h respectively, the cells were harvested to detect E3 ubiquitin ligase that interact with nsp5 by Co-IP. (B) The plasmid of Smurf1 was co-transfected with nsp5, RhoA and nsp2 into HEK-293T cells for 36 h respectively, the cells were harvested to detect nsp5 and RhoA that interact with Smurf1 by Co-IP. (C) PAMs were infected with recombinant lentivirus expressing nsp5 or control lentivirus for 24 h, the cells were harvested to detect the interaction between nsp5 and endogenous RhoA, as well as endogenous Smurf1 by Co-IP. (D) HEK-293T cells were co-transfected with Flag-Smurf1, Myc-RhoA, and HA-nsp5 plasmids for 36 h, the cells were harvested to detect the subcellular localization of Smurf1, RhoA and nsp5. The co-localization of Smurf1, RhoA and nsp5 was analyzed using ImageJ, and the results were quantified using the Pearson’s correlation coefficient. n = 10 cells from three independent experiments. (E) HEK-293T cells were co-transfected with HA-nsp5, Flag- Smurf1 and Myc-RhoA plasmids for 36 h, the cells were harvested to detect the interaction between nsp5, RhoA and Smurf1 by ternary Co-IP. (F) HEK-293T cells were co-transfected with the plasmids of RhoA, Smurf1 and Ub for 36 h, the cells were harvested to detect the ubiquitination of RhoA induced by Smurf1. (G) HEK-293T cells were co-transfected with the plasmids of RhoA, Smurf1, nsp5 and Ub for 36 h, the cells were harvested to detect the effect of nsp5 on Smurf1-induced polyubiquitination of RhoA. (H) HEK-293T cells were co-transfected with the plasmids of RhoA, Smurf1, nsp5 and K63-Ub for 36 h, the cells were harvested to detect the effect of nsp5 on Smurf1-induced polyubiquitination of RhoA. (I) HEK-293T cells were co-transfected with the plasmids of RhoA, nsp5 and Smurf1 for 36 h, the cells were harvested to detect the RhoA, nsp5 and Smurf1 protein expression by Western blot, tubulin served as an internal control. (J) Smurf1+/+ and Smurf1-/- HEK-293T cells were co-transfected with the plasmids of RhoA and nsp5 for 36 h, the cells were harvested to detect the RhoA, nsp5 and Smurf1 protein expression by Western blot, tubulin served as an internal control. (K) Smurf1-/- HEK-293T cells were co-transfected with the plasmids of RhoA, nsp5 and Smurf1 for 36 h, the cells were harvested to detect the RhoA, nsp5 and Smurf1 protein expression by Western blot, tubulin served as an internal control. (L) PAMs were transfected with siNC and siSmurf1 for 24 h, and then infected with recombinant lentivirus expressing nsp5 or control lentivirus for 24 h, the cells were harvested to detect the RhoA, nsp5 and Smurf1 protein expression by Western blot, tubulin served as an internal control. Scale bar: 10 μm.

RhoA has no effect on PRRSV-2 replication

These studies reveal the mechanism by which PRRSV-2 infection leads to the degradation of RhoA, thereby impeding the lysosome from targeting phagocytic bacteria. We were also curious to know whether the degradation of RhoA induced by PRRSV-2 had any impact on viral replication. To assess the impact of ablating RhoA expression on viral expression, we transfected the specific siRNA into immortalized PAMs and infected with PRRSV-2. The results showed that knockdown of RhoA had no effect on the expression level of PRRSV-2 ORF7 mRNA (Fig 9A and 9B). At the same time, PRRSV-2 N protein level and virus titer in the RhoA knockdown group were consistent with those in the NC group (Fig 9C and 9D). Subsequently, the eukaryotic expression plasmid of RhoA was transfected into immortalized PAMs and infected with PRRSV-2. As shown in Fig 9E and 9F, overexpression of RhoA exerted no influence on the expression level of PRRSV-2 N protein and viral titer. These data imply that the degradation of RhoA induced by PRRSV-2 infection has no impact on viral replication.

RhoA has no effect on PRRSV-2 replication.(A-D) Immortalized PAMs were transfected with siNC or siRhoA for 24 h and then infected with PRRSV-2 for 36 h, cells and supernatants were harvested to determine (A) RhoA mRNA expression, (B) PRRSV-2 ORF7 mRNA expression, (C) RhoA and PRRSV-2 N protein expression, tubulin served as an internal control, (D) supernatant virus titer. (E-F) Immortalized PAMs were transfected with pEGFP-C1 (vector) or pEGFP-C1-RhoA for 24 h, followed by infection with PRRSV-2 for the indicated periods, cells and supernatants were harvested to determine (E) RhoA and PRRSV-2 N protein expression, tubulin served as an internal control, (F) supernatant virus titer. P values were calculated using Student’s t-test. ns, not significant.

Discussion

In the natural environment, pathogen infection often predisposes the host to secondary infection by other pathogens. The first infecting pathogen changes the host’s internal environment, including metabolism, immunity, and tissue morphology, which makes the host more susceptible to other infectious pathogens [36,37]. Secondary bacterial infection after viral infection is more common and serious than other scenarios [38]. Bacterial superinfection poses a severe threat to human health. For example, influenza virus infection increases susceptibility to several respiratory bacterial pathogens, and retrospective research has shown that more than 50 million people died of pneumonia exacerbated by bacterial superinfection during the 1918 influenza outbreak [39]. Therefore, research on bacterial superinfection in the context of ongoing viral infection is of great significance, but this situation is very complex. In this study, we report a novel mechanism through which the cytoskeleton is reprogrammed following PRRSV-2 infection, thereby inhibiting the phagocytosis of secondary infected A. viridans by lysosomes and consequently enhancing the survival of these bacteria.

The harm to animals caused by secondary bacterial infections cannot be ignored. With the existence of immunosuppressive diseases and the abuse of antibiotics, an increasing number of bacterial superinfections in the context of ongoing viral infection have begun to plague the breeding industry. Due to the lack of effective prevention and treatment measures, the industry experiences huge economic losses every year. Secondary bacterial infection after PRRSV infection is particularly common, Hps, S. suis, Actinobacillus pleuropneumoniae (A. pleuropneumoniae) and S. enterica are frequently observed as secondary infections following PRRSV infection [5,40,41]. In recent years, A. viridans has been increasingly identified in pig farms experiencing outbreaks of PRRSV. We also isolated a strain of A. viridans from pigs infected with PRRSV-2, and for the first time, we have demonstrated that PRRSV-2 facilitates secondary infection by A. viridans, resulting in a more pronounced inflammatory response (Fig 1). Given the high morphological similarity between A. viridans and Streptococcus, this resemblance may contribute significantly to the underestimation of A. viridans’ pathogenic potential. It is imperative to employ more precise detection technologies to differentiate these two bacterial species in swine farms.

Viruses can promote bacterial infection of hosts in many ways. For example, influenza virus decreases the production of CCL2, thereby inhibiting the recruitment of macrophages and reducing host clearance of bacteria, or by inducing glucocorticoids production to compromise innate host defense against secondary bacterial infection. PRRSV is acknowledged as an immunosuppressive virus that facilitates secondary bacterial infections by destroying the host’s immune function. However, the mechanisms by which these secondary infecting bacteria avoid macrophage phagocytosis remain unclear. We observed that in WT macrophages, lysosomes and bacteria typically co-localize, which directly reflects the bactericidal function of macrophages. However, in macrophages infected with two distinct PRRSV-2 strains, lysosomes failed to co-localize with bacteria, suggesting that PRRSV-2 infection effectively inhibited the phagocytic activity of lysosomes towards secondarily infected bacteria (Figs 3A and S1B), thereby allowing more time for bacterial proliferation. Given the physical separation between lysosomes and secondary-infected bacteria in PRRSV-2-infected macrophages, we hypothesize that the cytoskeletal F-actin, which serves as the transport channel for organelles, may have been compromised. Subsequent experiments validated our hypothesis, demonstrating that the organized cytoskeletal F-actin in macrophages became depolymerized and disordered following two distinct PRRSV-2 strains infection (Figs 3B and S1A in S1 File). Our study is not the first to describe the interaction between PRRSV-2 and cytoskeleton [42], however, the contribution of PRRSV-2 reprograms cytoskeleton to facilitate the survival of secondary bacterial infections within macrophages has not been previously reported. The clinical phenotypes caused by the two PRRSV genotypes, PRRSV-1 and PRRSV-2, are highly similar. Both severely compromise the immune defense in pigs, thereby promoting secondary bacterial infections. Given the significant effect of the F-actin inhibitor LAT-A (Fig 3C-3E), if PRRSV-1 is also capable of damaging F-actin within macrophages, the core conclusion of this study likely extends to PRRSV-1 as well.

Previous research has shown that changing the cytoskeleton affects the host’s ability to eliminate bacteria. The Ras-related GTPase family plays an important role in the assembly of actin stress fibers during cytoskeletal regulation. Among the GTPase family members, RhoA is a critical participant in many cell functions, such as cytokinesis, adhesion, migration and actin cytoskeleton dynamics [43]. Research has shown that RhoA plays different roles in the process of various bacterial infections. For instance, RhoA contributes to type III group B streptococcal invasion of human brain microvascular endothelial cells. On the contrary, RhoA inhibits Legionella infection of mouse macrophages. In invertebrates, RhoA functions as an antibacterial molecule, preventing Vibrio anguillarum infection in shrimp [44–46]. In this study, we found that RhoA also acts as an antibacterial molecule by promoting F-actin polymerization, which accelerates the lysosome-targeted phagocytosis of A. viridans in porcine alveolar macrophages (Fig 4). However, PRRSV-2 induces the autophagy-lysosome degradation of RhoA via the viral protein nsp5, thereby effectively inhibiting RhoA-mediated antibacterial functions and consequently providing a more favorable environment for the proliferation of secondary bacterial infections (Figs 6 and 7). Another interesting finding of this study is that despite the downregulation of RhoA expression following PRRSV-2 infection, RhoA does not participate in PRRSV-2 replication (Fig 9). Previous studies have shown that RhoA has opposite functions in the face of different viral infections. For example, RhoA is not only required for respiratory syncytial virus-induced syncytium formation and filamentous virion morphology but also plays an active role in the internalization process of classical swine fever virus [47,48]. Conversely, polypeptide sequences derived from RhoA can effectively inhibit the replication of human immunodeficiency virus type 1 and human parainfluenza virus-3 [49]. Given the crucial role of the cytoskeleton, many viruses have evolved sophisticated strategies to modulate cytoskeletal dynamics to promote their replication cycle. For instance, during late phases of SARS-CoV-2 infection, the viral structural proteins M, E, and N induce a strong rearrangement of F-actin nanostructures, resulting in the intracellular vesicles containing viral components are surrounded by F-actin ring-shaped structures. This rearrangement of F-actin nanostructures suggests a functional role in facilitating the trafficking of viral particles toward the plasma membrane for subsequent release [50]. Similarly, HIV-1 progeny virions require the host enzyme MICAL to induce localized disassembly of F-actin at the plasma membrane in order to complete budding [51]. Additionally, vaccinia virus F11 protein promotes viral spread by modulating the cortical actin cytoskeleton by inhibiting RhoA signaling [52]. However, our study specifically investigated the downstream effects of PRRSV-2 induced F-actin reprogramming on bacterial secondary infection. Despite the fact that RhoA knockdown or overexpression does not influence PRRSV-2 replication, its regulatory impact on secondary bacterial infections following PRRSV-2 infection is evident. The activation of RhoA-mediated cytoskeletal F-actin polymerization might represent a novel strategy to deal with the secondary bacterial infections.

In summary, we find for the first time that PRRSV-2 reprograms the cytoskeleton to facilitate the survival of bacteria in secondary infections. During PRRSV-2 infection, PRRSV-2 nsp5 interacts with RhoA, inducing the autophagy-lysosome degradation of RhoA and resulting in a decrease in F-actin. The reduction of F-actin impedes the efficiency of lysosomal targeted clearance of bacteria, thereby facilitating the survival of secondary infected bacteria (Fig 10). Our study provides novel insights into secondary bacterial infections following PRRSV-2. It fills knowledge gaps related to PRRSV-2-bacterial coinfection and offers a theoretical basis for the prevention and control of secondary infections in swine farms.

Schematic model for the mechanisms that PRRSV-2 facilitates the survival of secondary infected bacteria.During PRRSV-2 infection, PRRSV-2 nsp5 interacts with RhoA, inducing the autophagy-lysosome degradation of RhoA and resulting in a decrease in F-actin. The reduction of F-actin impedes the efficiency of lysosomal targeted clearance of bacteria, thereby facilitating the survival of secondary infected bacteria.

Materials and methods

Ethics statement

All animals were handled in strict accordance with good animal practice according to the Animal Ethics Procedures and Guidelines of the People’s Republic of China, and the study was approved by The Animal Ethics Committee of Lanzhou Veterinary Research Institute, Chinese Academy of Agricultural Sciences (Permit No. LVRIAEC-2025–128).

Cells, viruses and bacteria

The immortalized PAMs (3D4/21-CD163), MARC-145 and HEK-293T cells were cultured in DMEM (BasalMedia) supplemented with 10% fetal bovine serum (FBS) (TransGen Biotech). PAMs were isolated from 4- to 6-week-old PRRSV-negative pigs and maintained in RPMI 1640 medium (BasalMedia) supplemented with 10% FBS. All cells were grown at 37°C with humidity and 5% CO_2_. A highly pathogenic PRRSV-2 strain GD-HD (GenBank ID: KP793736.1), a NADC30-like PRRSV-2 strain SX-YL1806 (GenBank ID: OR208175.1), were used in this study. The A. viridans, K. pneumoniae and S. suis 2 strain was preserved in our laboratory (Lanzhou Veterinary Research Institute, Chinese Academy of Agricultural Sciences, Lanzhou, China).

Antibodies and chemicals

Anti-RhoA monoclonal antibody (#2117) was purchased from Cell Signaling Technology, Anti-Smurf1 (sc-100616) was purchased from Santa Cruz Biotechnology, anti-RhoA antibody (10749–1-AP, 66733–1-Ig), anti-ATG7 polyclonal antibody (10088–2-AP), anti-TOLLIP polyclonal antibody (11315–1-AP), anti-LAMP1 monoclonal antibody (CL647–65051), anti-ubiquitin Polyclonal antibody (80992–1-RR) were purchased from Proteintech, anti-tubulin monoclonal antibody (HC-101) was purchased from TransGen Biotech, anti-Flag antibody (20543–1-AP, CL594–66008), anti-Myc antibody (16286–1-AP, CL488–60003), anti-GFP polyclonal antibody (50430–2-AP), anti-HA antibody (51064–2-AP, CL647–81290) were purchased from Proteintech, anti-PRRSV-2 N monoclonal antibody was prepared by our laboratory; Lipofectamine RNAiMAX Transfection Reagent (13778150), Lipofectamine 3000 Transfection Reagent (L3000015) and LysoTracker Red DND-99 (L7528) were purchased from Thermo Fisher Scientific; CFDA (S1076) was purchased from Solarbio; YF 488-Phalloidin (YP0114L) was purchased from UElandy; Latrunculin A (HY-16929), 3 × Flag peptide (HY-P0319) were purchased from MedChemExpress; ELISA assay kits for porcine IL-1β (SEKP-0001) and IL-6 (SEKP-0004) were procured from Solarbio.

RNA interference assay

Small interfering RNAs (siRNAs) against RhoA and negative control (NC) siRNA were designed and synthesized by Tsingke Biotechnology, and the sequences of the siRNAs used are listed in Table 1. Immortalized PAMs were seeded in 12-well plates at 1 × 10^5^ cells/well, and then transfected with the indicated siRNAs at a final concentration of 50 nM using Lipofectamine RNAiMAX according to the manufacturer’s instructions. The effects of RhoA knockdown were detected by RT-qPCR and Western blotting.

Table 1: List of siRNAs, sgRNAs and RT-qPCR primers.

CRISPR-Cas9-mediated knockout in HEK-293T cells

The CRISPR-Cas9 design and analysis method was described in a previous study [53]. sgRNAs were ligated into the Lenti-CRISPRv2 plasmid and co-transfected with psPAX2 and pMD2. G into HEK-293T cells to obtain recombinant lentivirus. The recombinant lentiviruses were used to infect HEK-293T cells. Puromycin (1.5 μg/ml) was added to the cell cultures to select the knockout cells 48 h later. The selected cells were subcloned and inserted into 96-well plates for single-clone growth and analyzed by PCR and Western blotting. The sgRNA sequences of TOLLIP and Smurf1 used are listed in Table 1.

Quantitative reverse transcriptase PCR (RT-qPCR)

RT-qPCR was performed as described previously with minor modifications [54]. According to the manufacturer’s instructions, total RNA was isolated using TRIzol reagent and reverse transcribed using a PrimeScript RT reagent kit (TaKaRa). RT-qPCR analysis was performed using ChamQ SYBR qPCR Master Mix (Vazyme), and the relative expression levels were calculated by the 2^-ΔΔCT^ method. The primers used are listed in Table 1. HPRT1 was used as an internal reference in immortalized PAMs.

Ternary Co-IP

HA-tagged nsp5, Flag-tagged TOLLIP (or Smurf1), and Myc-tagged RhoA were co-transfected into HEK-293T cells for 36 h. Cell lysates were first incubated with Anti-Flag Magnetic Beads to capture the Flag-TOLLIP (or Smurf1) complex. After washing, the bound complexes were competitively eluted using 3 × Flag peptide. The eluate was then subjected to a second immunoprecipitation with Anti-Myc Magnetic Beads to enrich for Myc-RhoA and its interacting proteins. Finally, the presence of HA-nsp5 in the complex was detected by immunoblotting.

Western blot

Western blotting was performed as described previously with minor modifications [55], and the cells were harvested and lysed in RIPA buffer (Solarbio) containing a protease inhibitor mixture. The cell lysates were separated by 12% SDS‒PAGE and transferred onto PVDF membranes. The membranes were blocked with 5% nonfat milk in PBST for 2 h at room temperature and then incubated with the primary antibody at 4°C overnight. The membranes were washed with PBST and then incubated with a secondary antibody for 1 h at room temperature. After washing, the target proteins were detected with an enhanced chemiluminescence (ECL) kit (Beyotime).

Immunofluorescence assay

IFA was performed as described previously with the following modifications [56]. Cells grown on confocal dish were fixed with 4% paraformaldehyde (PFA) for 15 min and permeabilized in 0.1% Triton X-100 for 15 min. Then, the cells were incubated in 1% BSA in PBS for 2 h, followed by incubation with primary antibody for 1 h at 37°C. After washing with PBS three times, the cells were incubated with the fluorescent secondary antibody for 1 h. Finally, DAPI was used to stain the nucleus for 5 min. The images were collected using a Zeiss LSM980 confocal microscope.

Bacterial carboxy fluorescein diacetate (CFDA) staining

Bacterial CFDA staining was carried out according to the manufacturer’s instructions. Briefly, CFDA was diluted in the bacterial culture medium at a concentration of 50 μM. The mixtures were incubated at 37°C for 30 min and the bacteria were then washed twice with PBS. The labeled bacteria were used for subsequent experiments.

Lysosome staining

Lysosome staining was carried out according to the manufacturer’s instructions. LysoTracker was diluted in the preheated DMEM at a concentration of 100 nM. Cells were incubated with the diluted working solution for 2 h in the cell incubator, and then washed twice with PBS. The labeled lysosomes were used for subsequent experiments.

F-actin Staining

F-actin staining was carried out according to the manufacturer’s instructions. Briefly, 100 μL of ready-to-use fluorescent Phalloidin solution was added into the confocal dish to fully cover the cells and was incubated at room temperature for 90 min in the dark, and then washed twice with PBS. The labeled F-actin were used for subsequent experiments.

Viral titration

A 50% tissue culture infective dose (TCID_50_) assay was performed to assess viral titration as described previously with minor modifications [57]. MARC-145 cells were seeded into 96-well plates 24 h in advance, and virus supernatants were prepared by 10-fold continuous dilution, and 100 μl volumes of the dilutions were added per well in replicates of eight. The cells were cultured for another 5–6 d, and the TCID_50_ value was calculated by the Reed-Muench method.

Animal experiment

Fifteen 4-week-old piglets (free of PRRSV, PRV, PCV2, and CSFV), were randomly divided into three groups: the PRRSV-2 group, the A. viridans group, and the PRRSV-2 +A. viridans co-infection group, with 5 piglets in each group. On day 1, 200 μl of each piglet was inoculated intramuscularly, and 200 μl of each piglet was inoculated intranasally with PRRSV-2 GD-HD (10^5.5^ TCID_50_/ml). On day 7, 200 μl of each piglet was inoculated intranasally with A. viridans (2 × 10^8^ CFUs). All piglets were carefully monitored throughout the study, and rectal temperatures were recorded daily until euthanasia. Serum samples were collected from piglets in each group at the time of death, and PAMs were isolated from each group accordingly.

Statistical analysis

The data were obtained from at least three independent experiments and statistically analyzed using GraphPad Prism. The results are expressed as the means ± standard deviations (SDs). Statistical significance was determined by Student’s t-test, and asterisks indicate statistical significance: NS, not significant; *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Supporting information

S1 FigPRRSV-2 impairs the lysosome-mediated clearance of bacteria, exhibiting a conserved pattern across different viral and bacterial strains.(A) Immortalized PAMs were infected or not infected with NADC30-like PRRSV-2 for 24 h, the cells were harvested to observe the changes in F-actin. The fluorescence intensity of F-actin was quantified using ImageJ, n = 20 cells from three independent experiments. (B) Immortalized PAMs were infected or not infected with NADC30-like PRRSV-2 for 24 h and then infected with A. viridans for 4 h, the cells were harvested to detect the subcellular localization for lysosome and A. viridans. Colocalization between lysosome and A. viridans was quantified using ImageJ, with results expressed as the Manders’ correlation coefficients (M1 and M2). Specifically, M1 represents the proportion of lysosomal signal that overlaps with the bacterial signal relative to the total lysosomal signal, while M2 represents the proportion of bacterial signal overlapping with the lysosomal signal relative to the total bacterial signal. n = 20 cells from three independent experiments. (C) Immortalized PAMs were infected or not infected with NADC30-like PRRSV-2 for 24 h and then infected with A. viridans for 4 h, the cells were harvested to detect the number of bacteria by CFUs. (D) Immortalized PAMs were infected or not infected with PRRSV-2 for 24 h and then infected with S. suis 2 for 4 h, the cells were harvested to detect the subcellular localization for lysosome and S. suis 2. Colocalization between lysosome and S. suis 2 was quantified using ImageJ, with results expressed as the Manders’ correlation coefficients (M1 and M2). Specifically, M1 represents the proportion of lysosomal signal that overlaps with the bacterial signal relative to the total lysosomal signal, while M2 represents the proportion of bacterial signal overlapping with the lysosomal signal relative to the total bacterial signal. n = 20 cells from three independent experiments. (E) Immortalized PAMs were infected or not infected with PRRSV-2 for 24 h and then infected with S. suis 2 for 4 h, the cells were harvested to detect the number of bacteria by CFUs. (F) Immortalized PAMs were infected or not infected with PRRSV-2 for 24 h and then infected with K. pneumoniae for 4 h, the cells were harvested to detect the subcellular localization for lysosome and K. pneumoniae. Colocalization between lysosome and K. pneumoniae was quantified using ImageJ, with results expressed as the Manders’ correlation coefficients (M1 and M2). Specifically, M1 represents the proportion of lysosomal signal that overlaps with the bacterial signal relative to the total lysosomal signal, while M2 represents the proportion of bacterial signal overlapping with the lysosomal signal relative to the total bacterial signal. n = 20 cells from three independent experiments. (G) Immortalized PAMs were infected or not infected with PRRSV-2 for 24 h and then infected with K. pneumoniae for 4 h, the cells were harvested to detect the number of bacteria by CFUs. Scale bar: 10 μm. P values were calculated using Student’s t-test. **, P < 0.01, ***, P < 0.001.(DOCX)