Nonmuscle Myosin Heavy Chain IIA Recognizes Sialic Acids on Sialylated RNA Viruses To Suppress Proinflammatory Responses via the DAP12-Syk Pathway

NMHC-IIA, a subunit of nonmuscle myosin IIA (NM-IIA), takes part in diverse physiological processes, including cell movement, cell shape maintenance, and signal transduction. Recently, NMHC-IIA has been demonstrated to be a receptor or factor contributing to viral infections. Here, we identified that NMHC-IIA recognizes sialic acids on sialylated RNA viruses, vesicular stomatitis virus (VSV) and porcine reproductive and respiratory syndrome virus (PRRSV). Upon recognition, NMHC-IIA associates with the transmembrane region of DAP12 to recruit Syk. Activation of the DAP12-Syk pathway impairs the host antiviral proinflammatory cytokine production and signaling cascades. More importantly, sialic acid mimics and sialylated RNA viruses enable the antagonism of LPS-triggered proinflammatory responses through engaging the NMHC-IIA–DAP12-Syk pathway. These results actually support that NMHC-IIA is involved in negative modulation of the host innate immune system, which provides a molecular basis for prevention and control of the sialylated RNA viruses and treatment of inflammatory diseases.

IL-8, IL-1␤, and other effectors (2). All these responses contribute to restraining viral infections, promoting infected cell clearance, and activating the host adaptive immune system. However, excessive host immune responses usually lead to individuals' dysfunctions and disorders, thereby requiring a fine-tuned modulation by various negative regulators.
In particular, an appropriate production of proinflammatory cytokines induces acute inflammation to suppress viral replication and prevent other opportunistic pathogens (3). Once viral triggers disappear, proinflammatory responses are immediately terminated by different mechanisms (4). However, if viruses are not eliminated during acute inflammation, the chronic inflammatory state will be established with continuous production of inflammatory cytokines, which may be detrimental to hosts (5). For example, AIDS, with the symptom of sustained inflammation, is caused by HIV persistent infection (6). In some cases, overproduction of proinflammatory cytokines, namely, cytokine storm, can be life-threatening, e.g., lethal viral septic shock (7). Consequently, antiviral proinflammatory responses are normally under precise control to achieve an efficient clearance of invading viruses and avoid immune damage.
Proinflammatory signaling cascades converge on the activation of nuclear factor kappa-light-chain-enhancer of activated B cells (NF-B) as well as the mitogen-activated protein kinase (MAPK) family members, p38 MAPK, c-Jun N-terminal kinase (JNK), and extracellular signal-regulated kinase 1/2 (ERK1/2). NF-B activation is characterized by degradation of the phosphorylated inhibitory protein IB␣ and translocation of phosphorylated p65/p50 dimers to the nuclei, whereas MAPKs are activated with phosphorylation (8,9). Their activation eventually increases the transcription of various proinflammatory cytokines. These proinflammatory signaling pathways are therefore often hijacked by viruses to establish infections or targeted by anti-inflammation modulators to prevent uncontrolled proinflammatory responses (10,11).
Nonmuscle myosin heavy chain IIA (NMHC-IIA) participates in a variety of cellular physiological processes, such as cell contractility, shape maintenance, and signal transduction (12). In addition, cell surface NMHC-IIA has been reported to facilitate viral infections (13,14). Here, we revealed a novel mechanism of negative regulation of host proinflammatory responses, where NMHC-IIA recognizes sialic acids on the sialylated RNA viruses or sialic acid mimics to suppress proinflammatory responses through the DAP12-Syk pathway.
We subsequently sought to determine the inhibitory mechanism of DAP12. Viral infections have been shown to induce tyrosine (Y) phosphorylation within the DAP12 immunoreceptor tyrosine-based activation motif (ITAM) and then recruit phosphorylated Syk (17). In this study, obvious phosphorylation of DAP12 and Syk was observed during PRRSV early infection (Fig. 1E). Immunoprecipitation (IP) assays indicated that DAP12 was constitutively associated with Syk, and the association was enhanced upon PRRSV infection in PAMs (Fig. 1F) and CRL-2843-CD163 cells (Fig. S3A). The interaction of these two proteins was also verified in the overexpression system of human embryonic kidney 293T (HEK-293T) cells ( Fig. S3B and C). Y86 and Y97 within the DAP12 ITAM were further shown to be indispensable for its interaction with Syk, while aspartic acid (D) 50 within the DAP12 transmembrane domain (TMD) was dispensable (Fig. S3D).
Syk knockdown ( Fig. S4A and B) or noncytotoxic Syk inhibitor R406 (Fig. S4D) in PAMs significantly promoted PRRSV-induced proinflammatory cytokine transcription and restrained PRRSV infection as DAP12 knockdown did ( Fig. 1G and Fig. S4C, E, and F). Additionally, DAP12 overexpression failed to inhibit the transcription of proinflammatory cytokines in Syk knockdown CRL-2843-CD163 cells, resulting in the decreased PRRSV replication (Fig. S4G to I). These results illustrated that the DAP12-Syk pathway was involved in antagonism of PRRSV-triggered proinflammatory cytokine production.
To determine which proinflammatory cascades were targeted by the DAP12-Syk pathway, we used noncytotoxic inhibitors of NF-B and MAPKs to treat PAMs and then inoculated PAMs with PRRSV (Fig. S5A). We observed that PRRSV-induced proinflammatory cytokine transcription was mediated by activation of NF-B and MAPK (p38 and ERK1/2) pathways ( Fig. S5B and C). DAP12 or Syk knockdown enforced these activations in response to PRRSV (Fig. 1H). In CRL-2843-CD163 cells, DAP12 overexpression repressed PRRSV-triggered NF-B activation and proinflammatory cytokine transcription, while Syk knockdown restored the proinflammatory responses ( Fig. S5D to F). Phosphorylated p38 and ERK1/2 were hardly detected in CRL-2843-CD163 cells (data not shown). The results demonstrated that the DAP12-Syk pathway suppressed MAPK (ERK1/2, p38)-and NF-B-mediated proinflammatory responses triggered by PRRSV.
NMHC-IIA is identified to interact with DAP12. In general, association of a receptor with DAP12 is responsible for activating the DAP12-Syk pathway (18). Here, we carried out IP assays followed by mass spectrometry (MS) analysis to screen potential DAP12-associated receptors. Among the MS-identified proteins, NMHC-IIA was one of the most prominent proteins binding to DAP12 ( Fig. 2A). The binding of NMHC-IIA to DAP12 was also confirmed by immunoblotting (IB) analysis (Fig. 2B). We further found that the association of endogenous NMHC-IIA and DAP12 was augmented upon PRRSV infection (Fig. 2C). Moreover, localization of NMHC-IIA and DAP12 was visualized in CRL-2843-CD163 cells cotransfected with enhanced green fluorescent protein (EGFP)-NMHC-IIA and DAP12-monomeric red fluorescent protein (mRFP). Though NMHC-IIA and DAP12 sparsely colocalized in the cell membranes, their colocalization was in-FIG 1 DAP12-Syk axis is activated to inhibit PRRSV-triggered proinflammatory responses. (A to C) DAP12 knockdown enhances PRRSV-triggered production of proinflammatory cytokines and restricts PRRSV infection. PAMs were transfected with siDAP12-433# or siRNA-NC for 36 h and infected with PPRSV (MOI ϭ 0.1) for indicated time periods (0, 3, 6, 9, and 12 h). "0 h" indicated that PRRSV was added at this time point and washed off immediately. qRT-PCR was used to measure proinflammatory cytokine (TNF-␣, IL-6, IL-8, and IL-1␤) mRNAs (A) and PRRSV ORF7 (C). TNF-␣ production was measured by ELISA (B). (D) DAP12 knockdown promotes PRRSV-induced transcription of proinflammatory cytokines at different MOIs. PAMs with DAP12 knockdown were infected by PRRSV (MOI ϭ 0.1 or 1) for 6 h. qRT-PCR was performed to detect mRNA abundance of TNF-␣, IL-6, IL-8, and IL-1␤. (E) PRRSV early infection induces phosphorylation of DAP12 and Syk. PAMs were infected with PRRSV (MOI ϭ 1) for indicated time periods (0, 0.5, 1, 2, and 3 h). DAP12 was immunoprecipitated by anti-DAP12 antibody. Phosphorylated DAP12 and Syk were detected by IB. (F) PRRSV infection enhances the association of DAP12 and Syk. PAMs were infected with PRRSV for 1 h. IB was conducted for DAP12-immunoprecipitated protein detection. (G) Syk knockdown promotes PRRSV-induced proinflammatory cytokine transcription. PAMs were transfected with siSyk-443# or siRNA-NC for 36 h and then infected with PRRSV (MOI ϭ 0.1) for indicated time periods (0, 6, 9, and 12 h). TNFA, IL-6, and IL1B transcription was detected by qRT-PCR. (H) DAP12 or Syk knockdown promotes PRRSV-induced phosphorylation of NF-B, p38, and ERK1/2. DAP12 or Syk knockdown PAMs were infected with PRRSV (MOI ϭ 1) for 1 h. IB analysis was performed with indicated antibodies. Experiments in all panels were repeated at least three times, and similar results were obtained. Quantitation data were shown as mean Ϯ SD from three replicates. Statistical analysis used in qRT-PCR was determined by Student's t test: *, P Ͻ 0.05; **, P Ͻ 0.01; ***, P Ͻ 0.001; ns, not significant.

Liu et al.
® creased after PRRSV infection (Fig. S6A). We also immunoprecipitated NMHC-IIA from membrane proteins of PRRSV-infected PAMs. The results of immunoblotting (IB) analysis showed that PRRSV infection intensified the association of NMHC-IIA and DAP12 on the cell surface (Fig. 2D). We further observed that NMHC-IIA specifically interacted with DAP12 but not with Syk in CRL-2843-CD163 cells transfected with 3ϫFlag-DAP12 and Syk-myc-His (Fig. 2E). Additional evidence from immunofluorescence assay (IFA) indicated that punctately distributed NMHC-IIA was located at the cell membranes where DAP12 colocalized with Syk in mock-infected cells, while PRRSV infection increased their colocalization (Fig. S6B). Collectively, NMHC-IIA was identified to be a novel binding protein for DAP12.
Next, we explored how NMHC-IIA interacted with DAP12. On one hand, NMHC-IIA was divided into three fragments from N to C terminus, denominated IIA-A (residues 1 to 742; the numbering is according to UniProt entry F1SKJ1), containing myosin N-terminal SRC homology 3 domain (SH3)-like domain and actin-binding domain; IIA-B (residues 743 to 1560), containing the coiled-coil rod domain; and IIA-C (residues 1561 to 1957), with the nonhelical tail. Glutathione S-transferase (GST) pulldown assays showed that IIA-B was responsible for the interaction with DAP12 (Fig. 2F). IP assay further indicated that the interaction was independent of D50 and the five Y's in DAP12 (Fig. S7A). On the other hand, DAP12 was separated into two parts, ΔICD with deletion of the intracellular domain (ICD) and ΔECD with deletion of the extracellular domain (ECD). Pulldown assays determined that both ΔICD and ΔECD bound to IIA-B, suggesting that the DAP12 TMD might be the binding region ( Fig. 2G and Fig. S7B). Furthermore, a short stretch (residues 51 to 57; the numbering is according to UniProt entry Q9TU45) within the DAP12 TMD was proven to be indispensable for the interaction ( Fig. S7C and D).
NMHC-IIA recognizes the sialic acids on PRRSV to inhibit proinflammatory responses via the DAP12-Syk pathway. To identify the ligands required for NMHC-IIA-DAP12-Syk pathway activation, we inoculated PAMs with the same amounts of naive, UV-inactivated, and heat-inactivated PRRSV virions, respectively. Almost identical to naive ones, both UV-and heat-inactivated virions induced phosphorylation of DAP12 and Syk, as well as the interaction of NMHC-IIA or Syk with DAP12 (Fig. 4A). Various amounts of heat-inactivated virions all enhanced DAP12 phosphorylation and Syk binding to DAP12 (Fig. S9A). MYH9 knockdown significantly inhibited phosphorylation of DAP12 and Syk induced by heat-inactivated PRRSV (Fig. S9B). The results suggested that neither PRRSV structural proteins nor viral RNA genome affected activation of the DAP12-Syk pathway.

NMHC-IIA Suppresses Antiviral Responses
® acids, respectively (20). The virions where ␣2-6-linked sialic acids remained induced the activation of the DAP12-Syk pathway as the naive ones did, while the virions with removal of ␣2-3,6-linked sialic acids failed to activate the DAP12-Syk pathway (Fig. 4B). These results demonstrated that sialic acids on PRRSV were required for activation of the DAP12-Syk pathway.
To further elucidate whether sialic acids were the ligands of NMHC-IIA, we conducted Fc pulldown assays with Fc-fused NMHC-IIA and purified PRRSV virions treated with the indicated enzymes (21). The results indicated that PRRSV interacted with NMHC-IIA partially dependent on the ␣2-3,6-linked sialic acids, because the removal of N-glycans or ␣2-3,6-linked sialic acids decreased by about 50% or 60% the amount of PRRSV bound to NMHC-IIA, respectively (Fig. 4C). In addition, competitive Fc pulldown showed that the mixtures of sialic acid mimics (3=-sialyllactose or 6=-sialyllactose mimicking ␣2-3or ␣2-6-linked sialic acids) competed against PRRSV to interact with NMHC-IIA (Fig. S9C). Moreover, NMHC-IIA nonhelical tail IIA-C was demonstrated to strongly interact with PRRSV in GST pulldown assays (Fig. 4D). PRRSV glycoprotein 5 (GP5) modified with sialic acids was shown to be important for PRRSV infection in vitro and in vivo (22,23). We conducted pulldown assays with naive or PNGase F-or ␣2-3,6,8 neuraminidase-treated GP5 and IIA-C (21) and observed that GP5 interacted with IIA-C partially dependent on the sialic acids ( Fig. 4E and F). The above results identified that the sialic acids on PRRSV were the ligands of NMHC-IIA.
We also evaluated the effects of heat-inactivated PRRSV virions on LPS-stimulated proinflammatory responses. The virions attenuated NF-B activation ( Fig. 4G and H) as well as TNFA, IL-6, and IL1B transcription in response to LPS (Fig. S9D). In contrast, virions with removal of N-glycans or ␣2-3,6-linked sialic acids lost the inhibitory function ( Fig. 4G and H and Fig. S9D). Considering that sialic acids on the cell surface might cis-interact with NMHC-IIA, we removed sialic acids on PAMs with ␣2-3,6,8 neuraminidase treatment before LPS stimulation. Sialic acid mimics still inhibit LPSinduced transcription of proinflammatory cytokines in trans (Fig. S9E).
Taken together, these results showed that NMHC-IIA recognized sialic acids on PRRSV to repress the proinflammatory responses via the DAP12-Syk pathway.
NMHC-IIA recognizes the sialic acids on VSV to inhibit proinflammatory responses by activating the DAP12-Syk pathway. To determine whether other sialylated RNA viruses were recognized by the NMHC-IIA-DAP12-Syk pathway for suppressing proinflammatory responses, we used vesicular stomatitis virus (VSV) with abundant ␣2-3-linked sialic acids (24) to inoculate murine peritoneal macrophage-like RAW 264.7 cells. Noncytotoxic knockdown of NMHC-IIA, DAP12, or Syk enhanced the TNFA and IL1B transcription in response to VSV (Fig. 5A). However, knockdown of NMHC-IIA decreased the amounts of VSV virions after 1 h of incubation at 37°C (Fig. 5A), suggesting that NMHC-IIA might be involved in VSV invasion. Therefore, we used heat-inactivated VSV virions instead of naive ones in the following experiments. IP analysis showed that heat-inactivated VSV (MOI ϭ 20) enhanced phosphorylation of DAP12 and binding of NMHC-IIA or Syk to DAP12 at indicated time points (Fig. 5B). The virions also augmented the association of NMHC-IIA with DAP12 on cell membranes (Fig. 5C). Next, we investigated their effects on LPS-triggered proinflammatory responses. The virions inhibited the LPS-triggered proinflammatory responses in control RAW 264.7 cells but not in NMHC-IIA inhibitor-treated or NMHC-IIA, DAP12, or Syk knockdown cells (Fig. 5D to F). Differently, MAPK activation was not influenced (Fig. 5F), and IL-8 transcription was not detected (data not shown). Intriguingly, we found that sialic acid mimics exerted a similar function as the virions in response to LPS, and 3=-sialyllactose had a stronger inhibition of NF-B activation and IL-6 transcription than 6=-sialyllactose ( Fig. 5G and H). Furthermore, PNGase F-or ␣2-3,6,8 neuraminidasetreated virions lost the inhibitory function ( Fig. 5G and H). All these findings demonstrated that NMHC-IIA recognized sialic acids on VSV to suppress the proinflammatory responses via the DAP12-Syk pathway.
Sialic acid mimics inhibit LPS-induced proinflammatory responses through the NMHC-IIA-DAP12-Syk pathway. Based on the above results, we wondered whether sialic acids repressed LPS-induced proinflammatory responses through the NMHC-IIA-DAP12-Syk pathway independent of sialylated RNA viruses. Sialic acid mimics were shown to induce the activation of the NMHC-IIA-DAP12-Syk pathway and antagonize NF-B activation ( Fig. 6A and B). Interestingly, after analyzing the relative level of phosphorylated DAP12 (pDAP12) induced by sialic acid mimics, we found that pDAP12 levels in 3=-sialyllactose-treated RAW 264.7 cells were about twice those of 6=sialyllactose-treated ones. In contrast, pDAP12 levels elevated by 6=-sialyllactose were about 2.4-fold higher than those elevated by 3=-sialyllactose in PAMs (Fig. 6C). We also observed that sialic acid mimics inhibited LPS-induced transcription of proinflammatory cytokines, and the inhibitory function of 3=-sialyllactose was stronger than 6=sialyllactose in RAW 264.7 cells, while 6=-sialyllactose was the stronger one in PAMs ( Fig. 6D and E). Additionally, sialic acid mimics suppressed LPS-triggered proinflammatory responses in RAW 264.7 cells (Fig. 7A to D) or PAMs (Fig. 8A to D). As expected, MYH9, DAP12, and Syk knockdown all terminated the inhibitory function of sialic acid mimics ( Fig. 7 and 8) at different time points. Collectively, these results demonstrated that NMHC-IIA recognized sialic acids to inhibit LPS-induced proinflammatory responses by activating the NMHC-IIA-DAP12-Syk pathway.

DISCUSSION
To eradicate invading viruses, various host signaling cascades are activated to induce effective proinflammatory responses. However, excessive host antiviral proinflammatory responses may develop into acute or chronic inflammatory disorders, and therefore, multiple negative regulatory mechanisms are needed to maintain homeostasis (25). In particular, Siglecs are a family of type I membrane proteins which specifically recognize sialic acid-modified glycans. It is known that several pathogens have evolved the capacity to gain sialic acids from their hosts and produce their own sialylated glycoconjugates (26). This capacity seems to be crucial for their survival in mammalian hosts, possibly by mimicking the host cell surface molecules to negatively regulate the innate immune responses and avoid the immune attack. For example, Siglec 1, Siglec H, and Siglec G were reported to be exploited by viruses to antagonize antiviral immune responses (27)(28)(29). Siglec 1 also played a role in establishing an immunosuppressive state of inflammation (30).
NMHC-IIA plays important roles in cell adhesion, cell migration, and cell division (31). Increasing evidence indicates that NMHC-IIA is required for entry of viruses such as herpes simplex virus 1 (13), Epstein-Barr virus (14,32), severe fever with thrombocytopenia syndrome virus (33), and PRRSV (34). Here, we identified that NMHC-IIA functioned as a Siglec to negatively regulate NF-Band MAPK-mediated proinflammatory responses. GST pulldown assays demonstrated that NMHC-IIA directly interacted with sialylated RNA viruses partially dependent on sialic acids ( Fig. 4; see also Fig. S9 in the supplemental material). Furthermore, knockdown of NMHC-IIA led to augmented transcription of proinflammatory cytokines and activation of NF-B, p38, and ERK1/2 in response to sialylated RNA viruses ( Fig. 3; Fig. S5 and S8). To distinguish the immunoregulation function from its role in viral invasion, we used heat-inactivated virions or monovalent sialic acid mimics instead of naive viruses to inoculate LPS-stimulated cells, which still resulted in decreased proinflammatory cytokine production and attenuated NF-B activation ( Fig. 4 and 8 and Fig. S9). In contrast, MYH9 knockdown recovered the proinflammatory responses ( Fig. 5 and 8). Interestingly, we found that the suppression of LPS-triggered inflammation by sialic acid mimics in RAW 264.7 cells was different from that in PAMs. 3=-Sialyllactose showed higher inhibitory potency than 6=sialyllactose in RAW 264.7 cells, while 6=-sialyllactose was the more potent one in PAMs. . Experiments in all panels were repeated at least three times, and similar results were obtained. Quantitation shown was from three independent experiments (mean Ϯ SD from three replicates). Statistical analysis used in qRT-PCR was performed using Student's t test: *, P Ͻ 0.05; **, P Ͻ 0.01; ***, P Ͻ 0.001; ns, not significant.  . IB was performed to detect p65 phosphorylation. Experiments in all panels were repeated at least three times, and similar results were obtained. Quantitation data are mean Ϯ SD (n ϭ 3). Statistical analysis used in qRT-PCR was performed using Student's t test: *, P Ͻ 0.05; **, P Ͻ 0.01; ***, P Ͻ 0.001; ns, not significant.

NMHC-IIA Suppresses Antiviral Responses
Liu et al.

®
The difference might be due to cell species. We did not obtain the appropriate multivalent sialic acid mimics to investigate their effects on LPS-triggered proinflammatory responses, which needed a further exploration. These findings revealed a novel role of NMHC-IIA in negative modulation of innate immune responses and may be applied to design of anti-inflammatory drugs. DAP12, also called TYROBP (tyrosine kinase binding protein) and KARAP (killer cell activating receptor-associated protein), functions as an adaptor in various immune cells, including macrophages, microglia, monocytes, DCs, and natural killer (NK) cells (35). DAP12 contains a small ECD, a TMD, and an ICD. The ECD possesses an essential cysteine required for the homodimer formation, the TMD possesses an aspartic acid indispensable for interaction with DAP12-associated receptors, and the ICD possesses an ITAM responsible for interaction with Syk and delivering signals (18). The receptor-DAP12-Syk axis is a classic pathway involved in immune responses such as synergistic activation of proinflammatory cytokine production (36) and IFN augmentation (37). Interestingly, the axis is also involved in negative regulation of proinflammatory responses. DAP12-and Syk-deficient macrophages from the corresponding knockout mice displayed an increased secretion of proinflammatory cytokines in response to LPS, CpG DNA, and synthetic lipopeptide (38). In our study, we observed activation of the DAP12-Syk pathway during viral early infection ( Fig. 1 and 5). Knockdown of DAP12 or Syk enhanced both virus-triggered production of proinflammatory cytokines and NF-B/MAPK activation ( Fig. 1 and 5; Fig. S1 and S4). In contrast, DAP12 overexpression antagonized these responses (Fig. S2). These results suggested that the DAP12-Syk pathway was hijacked to suppress the antiviral proinflammatory responses.
So far, only the triggering receptor expressed on myeloid cells 2 (TREM-2) was identified to be a DAP12-associated receptor mediating inhibition of Toll-like receptor (TLR) and FcR signaling (39). DAP12-associated receptors in negative regulation of proinflammatory responses were not fully illustrated. Here, we identified a new DAP12associated receptor, NMHC-IIA, through IP-MS, IFA, and co-IP assays ( Fig. 2 and Fig. S6). The results showed a direct interaction between NMHC-IIA and DAP12, with a requirement of NMHC-IIA domain B (mainly containing ␣-helices) and a short stretch (residues 51 to 57) within the DAP12 TMD ( Fig. 2 and Fig. S7). In addition, we identified the sialic acids on RNA viruses (VSV and PRRSV) as the ligands of NMHC-IIA to activate the DAP12-Syk pathway (Fig. 4 and 5 and Fig. S9).
PRRSV, a sialylated RNA virus, has been shown to inhibit NF-B-mediated inflammatory responses during early infection while triggering the responses in the late infection (40). VSV, another sialylated RNA virus (41), has also been reported to induce a delayed NF-B activation (42). Consequently, the two sialylated RNA viruses are utilized to explore the interaction between sialic acids and host innate immunity during viral infections. Knockdown of NMHC-IIA, DAP12, and Syk all promoted virus-triggered proinflammatory responses ( Fig. 1 and 3 and Fig. S8). We also found that DAP12 knockdown promotes proinflammatory cytokine transcription at time zero (Fig. 1), which might be explained by DAP12 being involved in negative regulation of immune responses for maintaining homeostasis (16,43). However, IL-10 transcription was not induced during PRRSV infection, a finding which was different from previous reports where PRRSV upregulated IL-10 expression (44,45). This divergence might be due to the PRRSV strains and PAMs used and needs to be further elucidated. Next, we identified that sialic acids on viruses were the stimuli and exerted the anti-inflammatory effects (Fig. 4 and 5 and Fig. S9). This may be a common mechanism in negative regulation of antiviral proinflammatory responses (Fig. 9), which needs further demonstration in other sialylated RNA viruses, and even DNA viruses or bacteria.
Indeed, there are some issues left unsolved in the current work. First, we performed knockdown instead of knockout of NMHC-IIA in vitro since MYH9 knockdown with more than 70% efficiency was detrimental to the cells (Fig. S8). We failed to get the MYH9 knockout mice and perform in vivo experiments because the loss of NMHC-IIA was lethal for mice, which is consistent with a previous study (46). Second, it might be better to generate a stable cell line with DAP12 or Syk knockdown to determine their inhibitory functions. However, it is hard to obtain such stable cell lines for primary PAMs. Third, factors downstream of Syk delivering the inhibitory signals have not been discovered, and the related work is being carried out in our laboratory. Fourth, DAP10 is similar to DAP12 in its TMDs. There are certain receptors associated with DAP12 which are also identified to pair with DAP10 (43). Whether DAP10 is associated with NMHC-IIA needs to be further studied. Our initial work showed that DAP12 knockdown increased type I IFN production. Whether NMHC-IIA-DAP12-Syk has effects on IFN responses is our next issue to be explored.
In conclusion, we identify that NMHC-IIA recognizes sialic acids on sialylated RNA viruses and activates the DAP12-Syk pathway to suppress virus-triggered proinflammatory responses. More importantly, the NMHC-IIA-DAP12-Syk pathway is shown to inhibit LPS-induced proinflammatory responses on recognition of heat-inactivated sialylated RNA viruses and sialic acid mimics. Taken together, we have revealed a novel negative regulation mechanism of proinflammatory responses, which expands our knowledge of the host innate immune system and provides clues for struggling with the sialylated RNA viruses and inflammatory diseases.  HEK-293T, and RAW 264.7 cell lines were purchased from Cellbio (Shanghai, China). CRL-2843-CD163, a cell line stably expressing CD163 in CRL-2843, was constructed in our laboratory. PAMs and CRL-2843-CD163 cells were routinely maintained in Roswell Park Memorial Institute 1640 medium (RPMI 1640) supplemented with 10% heat-inactivated fetal bovine serum (FBS; Gibco), penicillin (100 U/ml), and streptomycin (100 mg/ml) in a humidified 37°C, 5% CO 2 incubator. MARC-145, HEK-293T, and RAW 264.7 cells were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% FBS and antibiotics.

MATERIALS AND METHODS
Highly pathogenic (HP) PRRSV strain HN07-1 (isolated during an HP-PRRSV outbreak in the Henan province of China in 2007 by our laboratory) and VSV (Indiana serotype kept in our laboratory) were prepared according to our laboratory's previous study (47). UV-inactivated viruses were generated by irradiation with short-wave UV light (254 nm) for 1 h, and heat-inactivated ones were gained by a water bath at 65°C for 15 min. The viral infectivity was confirmed to be completely lost. The viruses were purified through tangential flow filtration and Sepharose 4 Fast Flow gel chromatography, and their infectivity was comparable to that of naive ones.  Table 1.
qRT-PCR. Total RNAs were extracted with TRIzol reagent (Invitrogen), and the reverse transcription cDNAs were prepared from total RNAs using the PrimeScript RT reagent kit with gDNA Eraser (TaKaRa). The cDNAs from different samples were amplified by quantitative real-time PCR (qRT-PCR) to measure mRNA abundance on a 7500 Fast RT-PCR system (Applied Biosystems, Foster City, CA, USA). Relative mRNA level was evaluated by the 2 ϪΔΔCT method with glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA as an endogenous control (48). The primers for qRT-PCR analysis are listed in Table S1 in the supplemental material.
IB and IP. Cells were harvested and lysed in radioimmunoprecipitation assay (RIPA) lysis buffer (Beyotime Biotechnology), supplemented with protease and phosphatase inhibitors. Whole-cell lysates (WCLs) were normalized to equal amounts of ␤-actin, separated by 8 to 15% gradient sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and electrotransferred onto Immobilon-P membranes (Merck Millipore). The membranes were blocked in 5% skimmed milk for 1 h and probed with the indicated primary antibodies. After incubation with horseradish peroxidase (HRP)-labeled goat antimouse or rabbit IgG antibody as secondary antibodies, the indicated proteins were visualized with enhanced chemiluminescence (ECL) reagent (Solarbio). For IP, the indicated primary antibodies were first bound to protein A/G beads at 4°C for 4 h. Samples were subsequently incubated with the beads at 4°C overnight, and potential associated proteins were tested by IB as stated above. The relative levels of target proteins were analyzed using Image J software, and the ratio was displayed as fold change below the images.
ELISA. PAMs with DAP12 knockdown were inoculated with PRRSV at an MOI of 0.1 for indicated time periods (3, 6, 9, 12, 24, and 36 h). The cell supernatants were collected for measurement of TNF-␣ using ELISA kits according to the manufacturer's instructions.
Virus titration assay. The transfected cells were inoculated with PRRSV at an MOI of 0.1, 1, or 5. At indicated time points (24,36, and 48 h) postinfection, the progeny virus titers were measured by the 50% tissue culture infective dose (TCID 50 ) assay in MARC-145 cells.
Eukaryotic expression was conducted by transfection of each expression vector into HEK-293T or CRL-2843-CD163 cells for 36 h using Lipofectamine 2000 or Lipofectamine LTX with Plus reagent according to the manufacturer's instructions (Thermo Fisher Scientific). The transfected cells were lysed in RIPA lysis buffer supplemented with protease and phosphatase inhibitors and clarified by centrifugation at 12,000 rpm at 4°C for 15 min to collect supernatants. Protein A/G beads were incubated with the indicated antibodies and WCLs or cell supernatants at 4°C and eluted by 0.05 M glycine-HCl buffer, pH 2.2 (0.2 M glycine, 0.2 M HCl).
Inhibitor treatments. PAMs were seeded onto 24-well plates and treated with specific inhibitors of MAPKs (ERK1/2, p38, and JNK) and NF-B at different concentrations (5, 10, 15, 20, 25, and 30 M), Syk inhibitor at 5 M for 12 h, or NMHC-IIA inhibitor at 20 M for 1 h. Cell viability measurement was performed as stated above. Then, PAMs were inoculated with PRRSV at an MOI of 0.1 for 1 h, and transcription of proinflammatory cytokines was tested by qRT-PCR. Phosphorylation of indicated proteins was detected by IB.

TABLE 1 siRNAs in this study
Mass spectrometry analysis. PAMs were inoculated with PRRSV for 1 h and lysed for IP with DAP12 primary antibody or isotype control IgG. DAP12-associated proteins were eluted and subjected to SDS-PAGE and silver staining. Discrepant bands with intensive signal compared with isotype control IgG were cut and digested, followed by analysis using matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) by Shanghai Sangon Biotech Co. Ltd.
IFA. CRL-2843-CD163 cells were grown on coverslips in 6-well plates at 30 to 50% confluence. For visualization of the distribution of NMHC-IIA, DAP12, or Syk, cells were transfected with EGFP-NMHC-IIA or/and DAP12-mRFP for 36 h and then mock infected or infected with PRRSV (MOI ϭ 20) for 1 h. The cells were fixed with 4% paraformaldehyde for 15 min at room temperature and then blocked with 5% BSA-PBST for 30 min. Next, cells were incubated with anti-Syk MAb at 4°C for 2 h, followed by incubation with DyLight 350 (blue)-conjugated anti-mouse IgG for an additional 45 min. The localization of NMHC-IIA, DAP12, or Syk was observed under an inverted fluorescence and phase-contrast microscope (Carl Zeiss AG, Oberkochen, Germany). Images were taken at a ϫ400 magnification.
In vitro pulldown assays. GST resins or protein A/G beads were incubated with purified GST-or Fc-tagged proteins at 4°C for 2 h and then with the indicated purified Flag-tagged proteins or differently treated virions at 4°C for another 30 min or overnight. After extensive washing with TBS 4 times, proteins were eluted and subjected to IB with the indicated antibodies.
Deglycosylation treatments. Heat-inactivated virions were treated with PNGase F in PBS at 37°C for 90 min to remove all N-glycans (high-mannose and complex, sialic acid-containing glycans). To remove sialic acids, the virions were incubated with specific neuraminidases in PBS at 37°C for 90 min or only the neuraminidase buffer (0 unit of neuraminidase) as a negative control. PAMs or RAW 264.7 cells were stimulated with LPS (10 g/ml) for indicated time periods (0, 15, and 30 min or 0, 30, and 120 min) in the presence of deglycosylation virions for 1 h. After extensive washing, cells were collected to conduct protein or mRNA extraction. The removal of sialic acids on the purified PRRSV GP5 was performed in parallel. The desialylated protein was further applied for pulldown assays. To remove sialic acids on the cells before infection, PAMs or RAW 264.7 cells were incubated with ␣2-3,6,8 neuraminidase at 37°C for 1 h and washed extensively to remove the enzyme (20,49).
Competition experiments with monovalent sialic acid mimics. The protein A/G beads with Fc-tagged NMHC-IIA were incubated with the purified PRRSV virions in the presence or absence of mixtures of 20 M 3=-sialyllactose and 6=-sialyllactose sodium salt at 4°C for 3 h. The virions were eluted and measured by IB.
Protein extraction. The nuclear and cytoplasmic proteins were separately extracted using a nuclear and cytoplasmic protein extraction kit and applied for IB detection. The membrane and cytoplasmic proteins were harvested using the ProteoExtract native membrane protein extraction kit.
Treatment with virions and sialic acid mimics. PAMs were incubated with the same amount (MOI ϭ 10 or 20) of heat-inactivated virions, desialylatd virions, or 10 M sialic acid mimics (soluble 3=-sialyllactose and 6=-sialyllactose sodium salt) for 1 h and later stimulated with LPS (10 g/ml) for indicated time periods (0, 15, and 30 min). RAW 264.7 cells were stimulated with LPS (10 g/ml) at indicated time points (30 and 120 min) in the presence of the specifically treated virions or the sialic acid mimics for 1 h. The indicated proteins and proinflammatory cytokines were analyzed by IB and qRT-PCR, respectively.
Statistical analysis. Three replicates were included in all experiments, and each experiment was independently repeated at least three times. The experimental data were presented as group mean and standard deviation (SD) and analyzed by the unpaired two-tailed Student t test with GraphPad (Graph-Pad Software, San Diego, CA, USA). Asterisks indicate statistical significance as follows: ns, not significant; *, P Ͻ 0.05; **, P Ͻ 0.01; ***, P Ͻ 0.001.

SUPPLEMENTAL MATERIAL
Supplemental material for this article may be found at https://doi.org/10.1128/mBio .00574-19. funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.
The authors declare that there is no conflict of interest.