Src-dependent Tyrosine Phosphorylation of Non-muscle Myosin Heavy Chain-IIA Restricts Listeria monocytogenes Cellular Infection*

Background: Non-muscle myosin IIA is involved in force generation, movement, and membrane reshaping. Its activity is regulated by phosphorylation of the light chain. Results: NMHC-IIA head domain is tyrosine-phosphorylated by Src and modulates Listeria intracellular levels. Conclusion: Tyrosine phosphorylation of NMHC-IIA affects the outcome of infection. Significance: This novel post-translational modification of NMHC-IIA possibly affects its functions. Bacterial pathogens often interfere with host tyrosine phosphorylation cascades to control host responses and cause infection. Given the role of tyrosine phosphorylation events in different human infections and our previous results showing the activation of the tyrosine kinase Src upon incubation of cells with Listeria monocytogenes, we searched for novel host proteins undergoing tyrosine phosphorylation upon L. monocytogenes infection. We identify the heavy chain of the non-muscle myosin IIA (NMHC-IIA) as being phosphorylated in a specific tyrosine residue in response to L. monocytogenes infection. We characterize this novel post-translational modification event and show that, upon L. monocytogenes infection, Src phosphorylates NMHC-IIA in a previously uncharacterized tyrosine residue (Tyr-158) located in its motor domain near the ATP-binding site. In addition, we found that other intracellular and extracellular bacterial pathogens trigger NMHC-IIA tyrosine phosphorylation. We demonstrate that NMHC-IIA limits intracellular levels of L. monocytogenes, and this is dependent on the phosphorylation of Tyr-158. Our data suggest a novel mechanism of regulation of NMHC-IIA activity relying on the phosphorylation of Tyr-158 by Src.


Bacterial pathogens often interfere with host tyrosine phosphorylation cascades to control host responses and cause infection. Given the role of tyrosine phosphorylation events in different human infections and our previous results showing the activation of the tyrosine kinase Src upon incubation of cells with
Listeria monocytogenes, we searched for novel host proteins undergoing tyrosine phosphorylation upon L. monocytogenes infection. We identify the heavy chain of the non-muscle myosin IIA (NMHC-IIA) as being phosphorylated in a specific tyrosine residue in response to L. monocytogenes infection. We characterize this novel post-translational modification event and show that, upon L. monocytogenes infection, Src phosphor-ylates NMHC-IIA in a previously uncharacterized tyrosine residue (Tyr-158) located in its motor domain near the ATP-binding site. In addition, we found that other intracellular and extracellular bacterial pathogens trigger NMHC-IIA tyrosine phosphorylation. We demonstrate that NMHC-IIA limits intracellular levels of L. monocytogenes, and this is dependent on the phosphorylation of Tyr-158. Our data suggest a novel mechanism of regulation of NMHC-IIA activity relying on the phosphorylation of Tyr-158 by Src.
Listeria monocytogenes is a human intracellular food-borne bacterial pathogen that causes serious disease in immunocompromised individuals. Within the host it finds suitable replication niches in the liver and spleen, disseminates, and then can reach the central nervous system. In pregnant women, L. monocytogenes targets the fetus, eliciting fetal infection and abortions (1). The ability of L. monocytogenes to cause disease relies on its capacity to invade nonphagocytic cells, replicate therein, and spread to the entire organism overcoming the intestinal, bloodbrain, and fetoplacental barriers (2). Through the expression of bacterial factors, L. monocytogenes establishes a cross-talk with host cells favoring the progression of the cellular infection (3). In epithelial cells, L. monocytogenes invasion is mainly driven by the bacterial surface proteins InlA and InlB that bind E-cadherin and c-Met, respectively, at the surface of host cells (4,5). This engagement of host cell receptors triggers tyrosine phosphorylation-mediated signaling, resulting in the local activation of the Arp2/3 complex that initiates actin polymerization at the site of L. monocytogenes attachment (6,7), causing membrane invagination that supports bacterial entry. InlB interaction with the receptor tyrosine kinase c-Met stimulates its autophosphorylation and induces the tyrosine phosphorylation and recruitment of adaptor proteins and the activation of phospho-inositide 3-kinase (PI3K) (5,8,9). Phosphatidylinositol 3,4,5triphosphate generated by PI3K accumulates at the cell membrane during L. monocytogenes infection (8) and plays a crucial role in the recruitment of molecules controlling actin polymerization, such as Rac1 and WAVE2 (6, 10 -12). In turn, InlA binding to E-cadherin induces the activation of Src tyrosine kinase that subsequently phosphorylates cortactin, E-cadherin, and the clathrin heavy chain (7,13,14). Although cortactin and clathrin tyrosine phosphorylations are critical events for actin polymerization and recruitment at the L. monocytogenes entry site (7,13), E-cadherin phosphorylation leads to its ubiquitination, internalization, and further degradation (14). The combined action of these events leads to the internalization the L. monocytogenes into epithelial cells.
In this study we aimed to identify new cellular proteins undergoing tyrosine phosphorylation in response to L. monocytogenes infection, and we address whether such post-translational modification would regulate cellular infection. The tyrosine-phosphorylated proteins were recovered from L. monocytogenes-infected epithelial cells and subjected to mass spectrometry identification. We identified the non-muscle myosin heavy chain IIA (NMHC-IIA) 6 as one of the enriched tyrosine-phosphorylated proteins recovered upon L. monocytogenes infection.
NMHC-IIA is an actin-binding protein with motor and contractile properties, involved in cellular processes requiring force generation, cell movement, and membrane reshaping (15). In infection, NMHC-IIA is critical for viral entry (16,17) and supports invasion (18) and dissemination (19) of various bacteria. Although the serine/threonine phosphorylation of the regulatory light chain is a well known mechanism to regulate non-muscle myosin IIA activity (15), our knowledge on the regulation of the heavy chain is limited, and NMHC-IIA tyrosine phosphorylation has never been characterized. Here, we show that NMHC-IIA undergoes tyrosine phosphorylation in response to several bacterial pathogens. Our data indicate that upon L. monocytogenes cellular infection NMHC-IIA was phosphorylated in tyrosine residue 158 by the host Src kinase. In the presence of blebbistatin, a chemical inhibitor of myosin II activity, the percentage of cells showing L. monocytogenes-associated actin foci was increased and correlated with higher levels of intracellular L. monocytogenes. In addition, increased numbers of intracellular L. monocytogenes were also found in cells depleted of NMHC-IIA as well as in conditions where NMHC-IIA tyrosine phosphorylation is prevented. These results show the involvement of NMHC-IIA in L. monocytogenes infection and point to the regulatory role of its phosphorylation in tyrosine 158, which could affect NMHC-IIA activity. Our findings describe a novel post-translational modification of NMHC-IIA with important implications in bacterial infection. Taking into account the central role of NMHC-IIA in key cell biology processes, our data also suggest the existence of a new mechanism of NMHC-IIA regulation that could be of critical importance in the canonical functions of non-muscle myosin IIA.

EXPERIMENTAL PROCEDURES
Bacterial Strains and Cell Lines-Listeria and Escherichia coli strains were grown aerobically at 37°C, with shaking, in brain-heart infusion and lysogeny broth (LB) media, respectively. Yersinia was grown aerobically at 26°C, with shaking, in LB media. When required, antibiotics were added to growth media. Details are provided in Table 1. Caco-2 cells (ATCC HTB-37) were cultivated in minimum Eagle's medium with L-glutamine, supplemented with nonessential amino acids, sodium pyruvate, and 20% fetal bovine serum (FBS). HeLa (ATCC CCL-2), HEK293 (ATCC CRL-1573), and COS-7 (ATCC CRL-1651) cells were cultivated in DMEM with glucose (4.5 g/liter) and L-glutamine, supplemented with 10% FBS. Cells were maintained at 37°C in a 5% CO 2 -enriched atmosphere. Cell culture media and supplements were from Lonza.
Determination of Intracellular Bacteria-The levels of intracellular bacteria were determined as described (21). When indicated, cells were incubated with serum-free medium containing blebbistatin, PP1, or DMSO. Cells were challenged with prewashed L. monocytogenes at a multiplicity of infection (m.o.i.) of 50 or with Yersinia pseudotuberculosis (m.o.i. 10) for 60 min, treated with 20 g/ml gentamicin for 90 min, washed in PBS, and lysed with 0.2% Triton X-100, and serial dilutions were plated for CFU counting. For immunofluorescence scoring, cells infected with L. monocytogenes (m.o.i. 50) were treated with 100 g/ml gentamicin for 10 min and washed with 20 g/ml gentamicin prior fixation.
Protein Identification by Mass Spectrometry (MS)-Protein identification was performed by MALDI TOF/TOF mass spectrometry as described (23). Protein bands were excised from SDS-polyacrylamide gels, reduced with dithiothreitol, alkylated with iodoacetamide, and in-gel digested with trypsin. Peptides were extracted, desalted, concentrated using Ziptips (Millipore), crystallized onto a MALDI sample plate, and analyzed using a 4700 Proteomics Analyzer MALDI-TOF/TOF (Applied Biosystems). Peptidic mass spectra were acquired in reflector positive mode at a 700 -4000 m/z mass window, and proteins were identified by peptide mass fingerprint using Mascot software (Matrix Science, UK) integrated in the GPS Explorer software (ABSCIEX) and searched against the SwissProt/UniProt Homo sapiens protein sequence database. The maximum error tolerance was 35 ppm, and up to two missed cleavages were allowed.
Immunoblotting-Proteins were resolved in SDS-polyacrylamide gels and transferred onto Nitrocellulose membranes (Hybond ECL, GE Healthcare). Membranes were blocked with 5% skimmed milk in buffer A (150 mM NaCl, 20 mM Tris-HCl, pH 7.4, and 0.1% Triton X-100) for 1 h at room temperature or overnight at 4°C. Primary and secondary antibodies were diluted in 2.5% skimmed milk in buffer A. Membranes used for anti-phosphotyrosine detection were blocked with Western Blocker solution (Sigma), also used to dilute primary and secondary antibodies.
Immunofluorescence Analysis-Cells were fixed in 3% paraformaldehyde (15 min), quenched with 20 mM NH 4 Cl (1 h), permeabilized with 0.1% Triton X-100 (5 min), and blocked with 1% BSA in PBS (30 min). Antibodies were diluted in PBS containing 1% BSA. Coverslips were incubated for 1 h with primary antibodies washed three times in PBS and incubated 45 min with secondary antibodies and phalloidin Alexa 555 or 647. DNA was counterstained with DAPI (Sigma). Coverslips were mounted onto microscope slides with Aqua-Poly/Mount (18606, Polysciences). Images were collected with a confocal laser-scanning microscope (Zeiss Axiovert LSM 510 or Leica SP2 AOBS S.E.) and processed using Adobe Photoshop software.
Transfection and Lentiviral Transduction-The lentiviral shRNA expression plasmids Mission pLKO.1-puro (control) and Mission shRNA-c-Src (Sigma) were used in combination with the envelope plasmid pMD.G and packaging plasmid pCMVR8.91. Packaging, envelope, and shRNA vector plasmids were co-transfected into HEK293 cells. Viral supernatants were harvested after 72 h, filtered, and incubated with target HeLa cells for 48 h at 37°C. Puromycin was used to select for individual clones. The knockdown was verified by immunoblot and/or real time RT-PCR.
Kinase Assay-Kinase assays were performed using the Src assay kit (17-131, Millipore), following the manufacturer's instructions. Anti-GFP-immunoprecipitated fractions from HEK293 cells expressing GFP-NMHC-IIA variants were incubated (10 min, 30°C) with 10 units of recombinant Src (14 -117, Millipore), in 30 l of kinase reaction buffer supplemented with 9 l of manganese/ATP mixture and 10 Ci of [␥-32 P]ATP (PerkinElmer Life Sciences). Reactions, including an Src-specific substrate or lacking Src, were used as controls. Reactions were precipitated with 40% TCA and spotted onto P81 phosphocellulose paper squares, washed three times with 0.75% phosphoric acid, once with acetone, and transferred to microtubes containing UniverSol liquid scintillation mixture (MP Biomedicals). Incorporation of 32 P was determined in a Wallac 1450 MicroBeta TriLux liquid scintillation counter (PerkinElmer Life Sciences), as counts/ min. Radioactivity measurements were performed in duplicate in two independent assays.
Statistical Analyses-Statistical analyses were performed with Prism 6 software (GraphPad Software, Inc.). One-way analysis of variance with post hoc testing analyses were used for pairwise comparison of means from at least three unmatched groups. Two-tailed Student's t test was used to compare means of two samples and one-sample t test to compare with samples arbitrarily fixed to 100. Differences were not considered statistically significant for p value Ն0.05.

NMHC-IIA Is Tyrosine-phosphorylated in Response to
Bacterial Infection-To identify new host proteins undergoing tyrosine phosphorylation (Tyr(P)) in response to L. monocytogenes and that could affect L. monocytogenes cellular infection, we compared the Tyr(P) protein profiles of L. monocytogenes-infected and noninfected (NI) HeLa cells. Cell extracts were collected at different time points post-inoculation and subjected to IP using anti-phosphotyrosine antibodies (anti-Tyr(P)). IP fractions were resolved by SDS-PAGE followed by silver staining. Bands showing variable intensities in L. monocytogenes-infected versus NI cells were excised and processed for mass spectrometry identification. A band corresponding to an Ϸ250-kDa protein and displaying increased intensity throughout the infection (Fig. 1A) was identified as the human NMHC-IIA (data not shown).
To validate this result, HeLa and Caco-2 cells were incubated with L. monocytogenes for different time periods, and the presence of NMHC-IIA in anti-Tyr(P) IP fractions was assessed by immunoblot using NMHC-IIA-specific antibodies. We detected a time-dependent increase of NMHC-IIA in IP fractions from L. monocytogenes-infected cells (Fig. 1B). Levels of NMHC-IIA in Tyr(P) fraction increased 3.5-fold after 60 min of L. monocytogenes incubation with HeLa cells and 15-fold in Caco-2 cells upon 20 min of L. monocytogenes infection (Fig.  1B). Levels of NMHC-IIA in whole cell lysates (WCL) were not affected by infection (Fig. 1B), showing that increased levels of NMHC-IIA in IP samples are not related to an augmentation of NMHC-IIA expression. Incubation of HeLa cells with the nonpathogenic species Listeria innocua for 60 min only induced a small enrichment of NMHC-IIA in the anti-Tyr(P) IP fractions as compared with L. monocytogenes (Fig. 1C). In addition, NMHC-IIA was barely detected in IP fractions from HeLa cells stimulated by E. coli DH5␣ or latex beads (Fig. 1C). Altogether, these results indicate that the enrichment of NMHC-IIA in the pool of Tyr(P) proteins is associated with the pathogenic features of L. monocytogenes and is not a broad cellular response to any extracellular stimuli.
To investigate whether the same response could be induced upon infection with other human bacterial pathogens, HeLa cells were incubated for 4 h with the extracellular pathogenic E. coli EPEC and EHEC or the invasive E. coli K12 expressing the Y. pseudotuberculosis invasin (K12-inv) (24), an infection model allowing the study of signaling pathways triggered downstream from the invasin-integrin interaction. As compared with NI conditions, NMHC-IIA appeared slightly increased in anti-Tyr(P) IP fractions from EPEC-and EHECinfected cells. Strikingly, K12-inv induced a robust enrichment of NMHC-IIA in IP samples that is abolished in cells incubated with bacteria harboring a disrupted invasin-encoding gene (K12-⌬inv, Fig. 1D). For comparison, cells were also incubated with L. monocytogenes for 1 and 4 h (Fig. 1D). These results indicate that the enrichment of NMHC-IIA in the pool of Tyr(P) proteins is an event triggered by several human bacterial pathogens.
Our data suggest that bacterial infection either induces the direct NMHC-IIA Tyr(P) or stimulates its interaction with a protein that itself undergoes Tyr(P). To address this issue, endogenous NMHC-IIA was immunoprecipitated from NI and L. monocytogenes-infected HeLa cells, and Tyr(P) proteins were detected by immunoblot. A band showing a consistent 1.5-fold increase in intensity in infected samples was detected at the molecular weight of NMHC-IIA (Fig. 1E). Immunoprecipitated levels of NMHC-IIA were similar in NI and L. monocytogenesinfected cells. These results support a direct Tyr(P) of NMHC-IIA triggered by L. monocytogenes infection.
NMHC-IIA-Tyr(P) Induced by L. monocytogenes Cellular Infection Requires the Activity of Src Tyrosine Kinase-Considering our previous findings revealing the key role of the tyrosine kinase Src during L. monocytogenes invasion (7), we addressed the role of this kinase in NMHC-IIA-Tyr(P) in the context of L. monocytogenes infection. Prior to L. monocytogenes incubation, HeLa cells were treated with PP1, an inhibitor of Src family kinases, or with Y-27632, an inhibitor of the serine/threonine kinase ROCK that regulates NMHC-IIA activity through the phosphorylation of the regulatory light chain of myosin II and limits L. monocytogenes internalization (25). Given that NMHC-IIA-Tyr(P) is hardly detected by using anti-Tyr(P) antibodies in immunoblot, cell lysates were subjected to anti-Tyr(P) IP assay and NMHC-IIA was detected in IP fractions.
The increase in NMHC-IIA-Tyr(P) induced by L. monocytogenes infection of nontreated cells (NT) was abolished in PP1treated cells while being not affected by Y-27632 treatment ( Fig.  2A), suggesting that NMHC-IIA-Tyr(P) requires Src kinase activity and occurs independently from ROCK activity. In addition, we interfered with Src activity by overexpressing an Src kinase-dead variant (Src-KD) (26). Levels of NMHC-IIA-Tyr(P) induced by L. monocytogenes infection were assessed in HeLa cells nontransfected (NT), transfected, with an empty plasmid (Ctr) or overexpressing Src-KD. In contrast to NT and Ctr cells showing increased levels of NMHC-IIA-Tyr(P) upon L. monocytogenes infection, in cells overexpressing Src-KD the NMHC-IIA-Tyr(P) was almost undetectable (Fig. 2B). To further confirm these data, we targeted the expression of endogenous Src by using specific shRNAs. We observed that L. monocytogenes-induced NMHC-IIA-Tyr(P) occurred in shRNA control (sh Ctr) and was clearly diminished in shRNA Src-expressing (sh Src) HeLa cells, in which Src expression is reduced by 60% (Fig. 2, C and D). Altogether, these data demonstrate that Src activity is required for NMHC-IIA-Tyr(P) triggered by bacterial infection.
Host Src Kinase Phosphorylates NMHC-IIA in Tyrosine Residue 158 -The NMHC-IIA amino acid sequence includes 34 tyrosine residues, most of which are located in the myosin motor domain (Fig. 3A). To identify the NMHC-IIA tyrosine residues phosphorylated by Src upon L. monocytogenes infection, we used combined in silico approaches (NetPhos 2.0 and NetPhosK). Nine tyrosine residues were predicted as potentially phosphorylated, among which only the tyrosine in position 158 (Tyr-158) was a putative substrate for Src kinase (Fig.   monocytogenes-infected cells (data not shown). Levels of endogenous NMHC-IIA were comparable in the different conditions, and GFP fusion proteins were expressed similarly in NI and infected cells (Fig. 3C). These results corroborate in silico predictions and suggest the central role of Tyr-158 in NMHC-IIA-Tyr(P) triggered upon infection. To validate our results, total lysates from NI and L. monocytogenes-infected cells were probed with an antibody raised against a peptide comprising the phosphorylated Tyr-158 residue of NMHC-IIA (Tyr(P)-158). In agreement with our data, levels of NMHC-IIA-Tyr(P)-158 were increased 1.5-fold in L. monocytogenes-infected cells (Fig. 3D). In addition, samples enriched in NMHC-IIA phosphopeptides from NI and L. monocytogenes-infected cells were analyzed by mass spectrometry.  (Fig. 3E, cluster II). In infected samples, the area of cluster I that is correlated with the abundance of the corresponding phosphopeptide was increased 4.8-fold. Cluster II appeared 2.1-fold more abundant in L. monocytogenes-infected samples as compared with NI. Cluster I was further characterized and validated by MS/MS sequencing. Altogether, our data show that phosphorylation occurs at Tyr-158.
We further evaluated whether NMHC-IIA-Tyr(P) occurs specifically on Tyr-158 through Src activity, performing an in vitro kinase assay. GFP-NMHC-IIA-WT or Y158F ectopically expressed in HEK293 cells was highly enriched through immunoprecipitation using an anti-GFP antibody and incubated with purified Src kinase and [␥-32 P]ATP. A synthetic peptide substrate for Src was used as positive control. In the absence of kinase, the control peptide (Ctr) and IP fractions of NMHC-IIA-WT and Y158F showed residual levels of [␥-32 P]ATP incorporation. In the presence of Src kinase, the NMHC-IIA-WT-enriched IP fraction and the control peptide became radiolabeled, whereas the radioactivity incorporation in the NMHC-IIA-Y158F enriched sample remained at a basal level (Fig. 3F).
Altogether these results strongly suggest that Tyr-158 of NMHC-IIA is a substrate for Src kinase, becoming phosphorylated in response to L. monocytogenes infection, and put forward the putative role of this event in cellular infection. In addition, Tyr-158 appears extremely conserved among species ranging from Saccharomyces cerevisiae to H. sapiens (Fig. 3G), pointing to the broad importance for Tyr-158 in the regulation of highly conserved canonical functions of NMHC-IIA.
Inhibition of NMHC-IIA Activity Affects Intracellular Levels of L. monocytogenes-To assess the role of NMHC-IIA activity in cellular infection, we measured intracellular levels of L. monocytogenes following chemical inhibition of NMHC-IIA. Blebbistatin, a specific inhibitor of myosin II activity (27), was added (10 or 100 M) to HeLa and Caco-2 cells, and L. monocytogenes infection efficiency was quantified by gentamicin protection assays. As control, we used an inactive form of blebbistatin. L. monocytogenes intracellular levels were increased by 2-8-fold, in a dose-dependent manner in both cell lines, following treatment with the active as compared with the inactive enantiomer of blebbistatin (Fig. 4A). Untreated and inactive blebbistatin-treated cells showed similar levels of intracellular L. monocytogenes (data not shown). Our data are in agreement with a previous report showing that blebbistatin treatment of L2 cells increases L. monocytogenes adhesion and invasion (25). Recruitment of NMHC-IIA and formation of actin foci at L. monocytogenes entry sites were both detected in control (DMSO) and active blebbistatin-treated HeLa cells (Fig. 4B). Although the percentage of L. monocytogenes-associated cells remained similar in both conditions, the percentage of cells showing L. monocytogenes-actin foci increased in the presence of active blebbistatin (Fig. 4C). Together, our results indicate that the ATPase activity of NMHC-IIA is not required for its localization to the sites of L. monocytogenes uptake and does not influence the interaction of L. monocytogenes with host cells. However, inhibition of NMHC-IIA ATPase activity fosters the formation of L. monocytogenes-actin foci, which correlates with increased rates of intracellular bacteria.
Reduced Expression of NMHC-IIA Increases the Level of Intracellular L. monocytogenes-To further address the role of NMHC-IIA in L. monocytogenes cellular infection, levels of adherent and intracellular L. monocytogenes were quantified by gentamicin protection assays in NMHC-IIA-depleted HeLa cells, using two siRNAs (si#1 and si#2). In accordance with the data described above, levels of intracellular L. monocytogenes increased 2-fold in NMHC-IIA-depleted (IIA-si#1 and IIA-si#2) as compared with control siRNA-transfected cells (Ctr)  (Fig. 5A). NMHC-IIA depletion assessed by immunoblot reached 85% in si#1-transfected cells and 65% when using si#2 (Fig. 5A). Levels of adhered L. monocytogenes were also aug-mented in NMHC-IIA-depleted cells (data not shown). Immunofluorescence analysis of L. monocytogenes-infected NMHC-IIA-depleted cells revealed a 2-fold increase in the percentage  (Fig. 5B). The number of bacteria and actin foci per cell were also increased in NMHC-IIA-depleted cells (Fig. 5C), correlating with increased levels of intracellular bacteria. Our data indicate that, although L. monocytogenes association with cells does not require NMHC-II activity, it is modulated by NMHC-IIA itself probably through the interaction with other proteins.
To discard the hypothesis that increased levels of intracellular L. monocytogenes detected in NMHC-IIA-depleted cells could result from the overexpression of the isoform B of nonmuscle myosin heavy chain (NMHC-IIB), we confirmed that expression levels of NMHC-IIB were similar in NMHC-IIAdepleted cells and control cells (Fig. 5D). In addition, we found that L. monocytogenes intracellular levels decreased 3-fold in NMHC-IIB-depleted HeLa cells (Fig. 5E), suggesting that NMHC-IIA and -IIB play opposite roles in L. monocytogenes infection and thus undermining the possibility of their mutual functional replacement. To definitively reinforce our findings and exclude potential uncontrolled off-target effects, we performed gentamicin protection assays following gene rescue experiments. We created an siRNA-resistant GFP-NMHC-IIA construct (NMHC-IIA-siRes) by introducing silent point mutations within the si#2 target sequence. We found that increased levels of intracellular L. monocytogenes detected upon NMHC-IIA depletion (IIA-si#2) dropped to control levels in NMHC-IIA-depleted cells expressing NMHC-IIA-siRes (Fig. 5F). In contrast, the expression of NMHC-IIA-WT in NMHC-IIA-depleted cells did not restore control levels of intracellular L. monocytogenes. Immunoblot analysis confirmed that the expression of endogenous NMHC-IIA was diminished in the presence of si#2 and that ectopically expressed NMHC-IIA was only detected in NMHC-IIA-siRes-transfected cells (Fig. 5F).
However, in the absence of si#2, both NMHC-IIA-WT and siRes variants are expressed at similar levels (Fig. 5G). Together, these results confirm that the increase in L. monocytogenes intracellular levels observed in NMHC-IIA-depleted cells is specifically due to NMHC-IIA depletion.
To analyze whether the role of NMHC-IIA on intracellular levels of bacteria was specific for L. monocytogenes or could be broadened to other bacterial infectious processes, we performed gentamicin protection assays using L. innocua expressing InlB (L. innocua-inlB), the major internalin driving L. monocytogenes entry in HeLa cells (28), K12-inv, and Y. pseudotuberculosis. Numbers of intracellular L. innocua-inlB were not significantly different in NMHC-IIA-depleted and Ctr cells (Fig. 5H). In contrast, levels of intracellular K12-inv and Y. pseudotuberculosis were significantly lower in NMHC-IIA-depleted cells (Fig. 5H). Our data indicate that NMHC-IIA is specifically triggered by pathogenic L. monocytogenes and is independent of an InlB-mediated uptake. In contrast, the invasin-mediated uptake requires NMHC-IIA. Interestingly, NMHC-IIA and -IIB were shown to be required for SopB-mediated invasion of Salmonella (18). Our findings, together with published reports, reveal that NMHC-IIA plays opposite roles in different infection models; although it is required for an utmost Y. pseudotuberculosis and Salmonella infection, it has a restrictive role in L. monocytogenes cellular infection.
Function of NMHC-IIA in L. monocytogenes Infection Relies on the Phosphorylation of Its Tyrosine 158 -We reported above two important observations. 1) NMHC-IIA is tyrosine-phosphorylated by Src kinase upon L. monocytogenes incubation with cells. 2) L. monocytogenes intracellular levels are increased in conditions of NMHC-IIA depletion or inhibition of its activity, demonstrating that NMHC-IIA activity limits L. monocytogenes infection. To investigate whether both findings could be interconnected, we evaluated levels of intracellular bacteria under conditions where NMHC-IIA-Tyr(P) does not occur. We used cells with compromised Src activity (PP1 treatment and Src-KD overexpression) and cells expressing an NMHC-IIA nonphosphorylatable variant (NMHC-IIA-Y158F). Levels of intracellular L. monocytogenes showed a 2.5-fold increase in PP1-treated HeLa cells as compared with control DMSOtreated cells (Fig. 6A). In agreement, we observed an increase in L. monocytogenes intracellular levels in cells expressing Src-KD (Fig. 6B). Inversely, intracellular levels of K12-inv decreased 2-fold in PP1-treated cells (Fig. 6C), as reported previously (29). Increased levels of intracellular L. monocytogenes detected in conditions of Src inactivation and thus in the absence of NMHC-IIA-Tyr(P) correlate with our data showing that reduced levels or inactivation of NMHC-IIA resulted in increased numbers of intracellular L. monocytogenes. Our data also suggest an association between the role of NMHC-IIA in Y. pseudotuberculosis invasin-mediated uptake and invasin-triggered NMHC-IIA-Tyr(P).
To further confirm the role of NMHC-IIA-Tyr(P) in the L. monocytogenes cellular infection, we evaluated intracellular levels of L. monocytogenes in HeLa and COS-7 cells transiently expressing either the GFP-NMHC-IIA-WT (WT) or the nonphosphorylatable variant GFP-NMHC-IIA-Y158F (Y158F). In contrast to HeLa cells, COS-7 cells naturally lack NMHC-IIA expression, thus appearing as a valuable experimental model to address the effect of exogenously expressed NMHC-IIA variants in absence of the endogenous protein. Equivalent expression levels of both constructs were verified by flow cytometry and immunoblot (data not shown). L. monocytogenes intracellular rates were determined by gentamicin protection assays in cell populations containing about 50% of transfected cells. As compared with NMHC-IIA-WT, the expression of NMHC-IIA-Y158F led to increased levels of intracellular L. monocytogenes in both cell lines (Fig. 6D). Thus, NMHC-IIA-Y158F expression recapitulates the increase of intracellular L. monocytogenes in NMHC-IIA-depleted or inactivated cells. Furthermore, both GFP-NMHC-IIA-WT and GFP-NMHC-IIA-Y158F showed the same localization and accumulate at the site of L. monocytogenes entry in HeLa cells (Fig. 6E). These results indicate that although NMHC-IIA subcellular localization and recruitment to the site of bacterial uptake are unrelated to Tyr-158, the phosphorylation of this specific NMHC-IIA tyrosine plays a key role in restraining L. monocytogenes infection.

DISCUSSION
Pathogens interfere with host phosphorylation cascades to foster adhesion, invasion, and intracellular survival. Here, we searched for new host proteins undergoing tyrosine phosphorylation upon L. monocytogenes infection. We showed that NMHC-IIA is tyrosine-phosphorylated in response to L. monocytogenes as well as to other human bacterial pathogens such as EPEC, EHEC, and K12-inv. In L. monocytogenes infection, this previously unknown tyrosine phosphorylation event is triggered by Src kinase on residue Tyr-158 of NMHC-IIA, and it limits intracellular bacterial levels.
Myosin II activity is regulated by phosphorylation events in serine and threonine residues of the regulatory light chain (15). NMHC-IIA also undergoes serine and threonine phosphoryla-tions, which regulate the assembly of myosin II filaments in vitro and are thought to control subcellular localization of NMHC-IIA and contractility that depends on the actin crosslinking activity of NMHC-IIA (15). Although NMHC-IIA was detected in studies aiming to unravel the global phosphotyrosine signaling in cancer tissues (30,31), its tyrosine phosphorylation has never been characterized. Our data constitute the first report showing and characterizing NMHC-IIA-Tyr(P). Our preliminary in silico analysis suggests an important and broad role for NMHC-IIA Tyr(P) in position 158 as follows. 1) Tyr-158 is highly conserved among species ranging from S. cerevisiae to H. sapiens. 2) An in silico study suggested that Tyr-163 of muscle myosin heavy chain (matching Tyr-158 in NMHC-IIA) could be phosphorylated (32). 3) Tyr-158 is located in the motor domain of NMHC-IIA near the ATP-binding pocket. 4) Analysis of the crystal structure of the myosin motor domain (33) showed that Tyr-158 is exposed at the surface of the protein and is thus accessible for phosphorylation. Thus, we hypothesize that the phosphorylation of NMHC-IIA Tyr-158 could modulate NMHC-IIA activity most probably by affecting its ability to bind and/or hydrolyze ATP. However at this point any other mechanism could be envisaged. In addition, it is likely that NMHC-IIA-Tyr(P) in Tyr-158 occurs in specific physiological conditions engaging NMHC-IIA activity and thus plays a role in the regulation of the highly conserved canonical functions of NMHC-IIA. The functional and structural outcomes of such modification are now critical to elucidate.
Our data suggest that, upon infection, only a small pool of NMHC-IIA becomes phosphorylated in Tyr-158, probably concentrated in a restricted subcellular localization and/or interacting with specific partners, which would impact infection. Yet, we observed that both NMHC-IIA-WT and Y158F concentrated around bacteria at the entry site. We also found that phosphorylation of Tyr-158 does not affect the phosphorylation of the myosin regulatory light chain, 7 which is achieved by MLCK and is required for activation of myosin II motor activity (15). Interestingly, Src was previously shown recruited to membrane blebs where it associates with MLCK and myosin II (34,35). In response to cell swelling, Src and MLCK form a complex in which Src activates MLCK, and both regulate a compensatory membrane retrieval that requires myosin II (35). It is thus conceivable that Src and MLCK could work together to fine-tune the activity of myosin II in the context of infection.
Myosin II isoforms were recently involved in viral and bacterial infections either promoting or limiting pathogen progression. However, their role in such processes is still mainly descriptive. NMHC-IIA is required for Kaposi sarcoma-associated herpesvirus and HSV1 entry into cells (16,17,36), facilitates Salmonella invasion, and regulates its intracellular growth (18,37) and promotes Chlamydia dissemination (19). Conversely, myosin II limits bacterial cell-to-cell spread by restraining L. monocytogenes protrusion formation (38) and participating in the formation of Shigella-associated septin cages (39). NMHC-IIB is involved in the formation of actin-rich structures that accumulate near the Salmonella-containing vacuole and restrain bacterial intracellular multiplication (40). Altogether, these data suggest that the different outcomes associated with myosin II function during infection are probably related to the cellular machinery engaged in the various infectious processes.
Our results indicate that NMHC-IIA activity limits L. monocytogenes infection most probably hindering cellular invasion by interfering with the formation of L. monocytogenes-induced actin foci. NMHC-IIA-depleted or inactivated cells were reported to lose cytoplasm cohesion and show increased membrane activity and plasticity (41,42). These phenotypes could thus suggest that the increased numbers of intracellular L. monocytogenes observed in such cells would be greatly due to the disruption membrane rigidity. However, if this was the case, cells displaying low NMHC-IIA activity should be more permissive to any extracellular pathogen, which was not observed in KSHV (17), HSV1 (16), and Salmonella (18) infections. In addition, we show here that NMHC-IIA sustains invasin-mediated Y. pseudotuberculosis infection, and the invasion rate of L. innocua expressing InlB was not significantly increased by NMHC-IIA depletion, thus excluding a nonspecific cell invasion mechanism.
NMHC-IIA participates in cellular processes associated with phosphotyrosine signaling, which are largely usurped by bacteria, namely L. monocytogenes and Y. pseudotuberculosis (43), during infection. NMHC-IIA regulates protrusion formation and cell migration through the generation of actin retrograde flow (44,45); it is required for integrin-mediated adhesion maturation (46); it controls cell-cell adhesion promoting E-cadherin clustering and stabilizing cellular junctions (47); and it governs the polarization of epithelial cells generating forces to maintain the epithelia (48). Whether NMHC-IIA is Tyr(P) in these processes is unknown.
In intercellular junctions, NMHC-IIA is critical for the E-cadherin localization (47), and Src activation is required for actin polymerization at cell-cell contacts (49) as it is during E-cadherin-mediated L. monocytogenes invasion (7). Interestingly, Src activation and recruitment of c-Cbl are key events to control c-Met signaling (50). Our data show that Src activity restricts intracellular levels of L. monocytogenes in HeLa cells in which L. monocytogenes uptake is mainly mediated by c-Met and present the hypothesis that Src is acting through the tyrosine phosphorylation of NMHC-IIA to inhibit entry. Remarkably, in KSHV infection, which depends on integrin and Src activation (51), NMHC-IIA interacts with the ubiquitin ligase c-Cbl (17). The complex c-Cbl⅐NMHC-IIA associates with the receptor tyrosine kinase EphA2 that amplifies Src signaling to promote viral macropinocytosis (36). It is thus possible that c-Cbl, which is required for L. monocytogenes infection (52), associates with NMHC-IIA and c-Met to modulate L. monocytogenes infection through tyrosine phosphorylation events. To invade cells, Y. pseudotuberculosis binds ␤1-integrin (53), which interacts with NMHC-IIA via its cytoplasmic tail to regulate cell migration (54). As in adhesion and cell migration processes (55), during Y. pseudotuberculosis infection the engagement of ␤1-integrin leads to the activation of Src kinase (56), which could also act on NMHC-IIA triggering its tyrosine phosphorylation at the site of bacterial attachment thereby promoting Y. pseudotuberculosis infection.
Our data open new perspectives in the regulatory mechanisms governing NMHC-IIA functions in infection and physiological cellular processes. Further work should reveal whether NMHC-IIA-Tyr(P) affects its motor activity, binding partners, and/or the formation of actomyosin filaments.