Stabilin-1 plays a protective role against Listeria monocytogenes infection through the regulation of cytokine and chemokine production and immune cell recruitment

ABSTRACT Scavenger receptors are part of a complex surveillance system expressed by host cells to efficiently orchestrate innate immune response against bacterial infections. Stabilin-1 (STAB-1) is a scavenger receptor involved in cell trafficking, inflammation, and cancer; however, its role in infection remains to be elucidated. Listeria monocytogenes (Lm) is a major intracellular human food-borne pathogen causing severe infections in susceptible hosts. Using a mouse model of infection, we demonstrate here that STAB-1 controls Lm-induced cytokine and chemokine production and immune cell accumulation in Lm-infected organs. We show that STAB-1 also regulates the recruitment of myeloid cells in response to Lm infection and contributes to clear circulating bacteria. In addition, whereas STAB-1 appears to promote bacterial uptake by macrophages, infection by pathogenic Listeria induces the down regulation of STAB-1 expression and its delocalization from the host cell membrane. We propose STAB-1 as a new SR involved in the control of Lm infection through the regulation of host defense mechanisms, a process that would be targeted by bacterial virulence factors to promote infection.


Introduction
Listeria monocytogenes (Lm) is a major human foodborne pathogen that causes listeriosis, which is highly prevalent among high-risk groups including immunocompromised people, elderly, pregnant women, and neonates. Listeriosis is an overall public health concern associated with high hospitalization and mortality rates, being the most deadly foodborne infection in Europe [1]. Manifestations of the disease range from a selflimiting febrile gastroenteritis to septicemia, meningitis, and encephalitis [2]. The most severe aspects of the disease are related to the capacity of Lm to cross the intestinal, blood-brain, and maternal-fetal barriers, evading the immune response, multiplying within phagocytic and non-phagocytic cells and effectively disseminating throughout host tissues [3]. These properties are shaped by an arsenal of virulence factors [4].
The host innate immune response is critical to elicit an early defense toward Lm. The containment of infection requires both the participation of professional phagocytes that trap bacteria from target organs, and the activation of a number of pattern recognition receptors, including scavenger receptors (SRs) [5,6]. SRs comprise a diverse and conserved family of proteins, able to bind to a wide range of ligands stimulating the removal of non-self and modified-self targets [7]. They contribute to maintain homeostasis and control pathogen infections, playing key functions in the antimicrobial host immune response [7,8]. The role of SRs in Lm infection was first revealed for SR-A, SR-AI/II KO mice showing increased susceptibility to Lm infection and displaying increased hepatic granuloma formation [9]. Later, MARCO, CD36, and SR-BI were then shown to bind Lm and to modulate the immune response against Lm [10][11][12]. The first member of the Class H of SRs to be described was STABILIN-1 (STAB-1) [13]. It is a highly conserved type I transmembrane protein mainly expressed in sinusoidal endothelial cells of the spleen and liver, and on both afferent and efferent arms of the lymphatic vasculature, but also in subpopulations of monocytes/macrophages, and hematopoietic stem cells [13]. STAB-1 was implicated in lymphocyte adhesion and trafficking, angiogenesis and apoptotic cell clearance, therefore being crucial to maintain tissue homeostasis and resolving inflammation [14]. This SR has the ability to bind different ligands including modified low-density lipoproteins [15], phosphotidylserine expressed by apoptotic cells [16], and secreted protein acidic and rich in cysteine (SPARC). Importantly, STAB-1 was previously found to bind Gram-positive and Gram-negative bacteria in vitro [17]. Furthermore, it is known that this receptor controls inflammatory activity, modulates T cell activation, and also humoral immune response [18].
Here we address the role of STAB-1 in host defense against Lm infection and investigate the impact of STAB-1 deficiency on the host innate immune response against this bacterial pathogen. We reveal that STAB-1 KO mice display deregulated cytokine and chemokine expression, impaired recruitment of myeloid cells, and increased susceptibility to Lm infection. In addition, whereas STAB-1 appears to promote bacterial uptake by macrophages, Lm infection induces the down regulation of STAB-1 expression and its delocalization from the host cell membrane.

Macrophage infection
Macrophages were incubated for 30 min with 100 μg/ ml of fucoidan (Sigma-Aldrich), 50 μg/ml of Poly (I) or Poly (C) (Santa-Cruz-Biotechnology). Cells were infected for 30 min with exponential-phase bacteria at a multiplicity of infection (MOI) of 2 and treated with 20 μg/ml of gentamicin (Lonza) for 60 min as described [20]. Raw macrophages were incubated with 5 μg/ml or 25 μg/ml of mouse-IgG (SC-2025) or anti-STAB-1 antibody (sc-98,788) 1 h before bacterial infection at MOI of 50, during 30 min or 20 min plus 10 min with 50 μg/ ml of gentamicin. Cells were washed and lysed for CFU quantification.

Lentiviral vector production, transduction, and bacterial infection
Total RNA was extracted from HUVEC using TripleXtractor (GRISP) and reverse-transcribed, using RevertAid H Minus First-Strand cDNA Synthesis Kit (Thermo Scientific) and oligo(dT) 18 as recommended by the manufacturer. STAB-1 DNA fragment was amplified by polymerase chain reaction (PCR) (phusion high-fidelity DNA polymerase -Thermo Scientific) using HUVEC cDNA as a template. Digested and purified STAB-1 DNA fragment was colligated in pHR_citrine plasmid. STAB-1 signal peptide containing the HA tag (integrated DNA technologies-IDT) was fused with STAB-1 DNA sequence. Constructs were confirmed by PCR and DNA sequencing. Primers and signal peptide sequences are listed in Table S1. For virus assembly, STAB-1 cloned into pHR plasmid (STAB-1) or the empty plasmid (Mock), together with the envelope plasmid pMD-G and packaging plasmid pCMVR8.91 were co-transfected into HEK293T cells, using lipofectamine according the manufacturer's instructions. Viral supernatants were harvested after 72 h, filtered, and incubated with HeLa cells (supplemented with 1% of penicillin/streptomycin) for 72 h. Exponential-phase bacteria at MOI 50 (30 min or 20 min of infection plus 10 min with 20 μg/ml gentamicin) were added. Mock and STAB-1 cells were washed and lysed for CFU quantification.

Bone marrow-derived macrophages (BMDMs)
Mouse femurs were removed and flushed with Hank's Buffered Salt Solution (HBSS-Lonza) as described [21]. Bone marrow cells were collected by centrifugation and cultured overnight in DMEM supplemented with 10 mM HEPES (Gibco), 1 mM sodium pyruvate (Lonza), 10% FBS and 10% L929 cell-conditioned medium (LCCM). Non-adherent cells were collected and seeded. Upon 4 d of differentiation, 10% of LCCM was added, and on d 7 the medium was renewed. On d 10, latex beads of 1 μm (Invitrogen) (30 min of incubation) or exponential-phase bacteria at MOI 50 (20 min of infection plus 10 min with 50 μg/ml gentamicin) were added. Macrophages were washed and lysed for CFU quantification or used for immunofluorescence staining. Supernatants of WT and STAB-1 KO BMDMs were recovered and filtered to measure cytokine production by ELISA.

RNA techniques
RNAs were extracted from non-infected and infected cells (TripleXtractor, GRISP), as recommended by the manufacturer. Purified RNAs were reverse-transcribed (iScript, Bio-Rad-Laboratories) and analyzed by qPCR as described [22] or using specific primer probes (TaqMan). Gene expression data were analyzed by comparative Ct method [23], normalized to HPRT1 expression. For qualitative analysis, PCR was performed on cDNA (KAPA2G Mix, GRISP). Amplification products were resolved in 1% (w/v) agarose gel and analyzed with GelDoc XR+ System (Bio-Rad Laboratories). Primers and probes are listed in Table S1.

Immunofluorescence
Lm EGD GFP -infected BMDMs were fixed in 3% paraformaldehyde (15 min), quenched with 20 mM NH4Cl (1 h) and blocked with 1% BSA (sigma) in PBS (30 min). Cells were permeabilized with 0.1% Triton X-100 in PBS for 5 min and labeled with Alexa Fluor 647-conjugated phalloidin (Invitrogen), during 45 min on the dark. Cells were washed and slide preparations were mounted and dried at room temperature. Images were captured with an Olympus BX53 fluorescence microscope. The percentage of cells with intracellular bacteria or beads, and the number of intracellular bacteria or beads per cell were calculated. At least 300 cells were analyzed for each sample in three independent experiments. Non-infected and Lm-infected HUVEC were incubated for 1 h with primary antibody rabbit anti-STAB-1 (1:100, Millipore), diluted in 0.2% saponin (Merck) supplemented with 1% BSA. Cells were washed in 0.2% saponin and incubated 45 min with secondary anti-rabbit Alexa 488 antibody (Invitrogen). DNA was counterstained with DAPI (Sigma) and actin labeled with TRITC-conjugated phalloidin. Images were collected with an Olympus BX53 fluorescence microscope and processed using ImageJ.

Cell fractionation and immunoblotting
Cytoplasmic and membrane fractions from noninfected and Lm-infected cells were obtained using the Subcellular Protein Fractionation kit (Thermo Scientific). Cell samples and homogenized spleens were diluted in Laemmli buffer, resolved by SDS-PAGE on 8% gels. Samples were transferred onto nitrocellulose membrane (Bio-Rad Laboratories), blocked and blotted with rabbit anti-STAB-1 (1:500, Millipore), followed by HRP-conjugated goat antirabbit IgG (1:2000, P.A.R.I.S). Signals were detected using ECL (Thermo-Scientific) and digitally acquired in a ChemiDoc XRS+ system (Bio-Rad Laboratories). Signal intensity was quantified using Image J.

Flow cytometry and cell sorting
Ectopic expression of STAB-1 was confirmed by flow cytometry. Transduced HeLa cells were collected in icecold storage solution (PBS supplemented with 2% FBS and 0.1% azide), pelleted by centrifugation, washed, and labeled with mouse anti-HA (1:50, Abcam) during 30 min, followed by 20 min incubation with Alexa Fluor 647-conjugated anti-mouse IgG (1:500, Molecular Probes). The percentage of transduced cells (positive for both citrine and HA expression) was sorted, collected, and further cultured and expanded.

Animal infections
STABILIN-1 full knock-out (STAB-1 KO) mice and their wild-type (WT) littermates, both with a C57BL/ 6 N, 129SvJ mixed background have been described [24]. Infections were done as described [25]. Briefly, intravenous infections were performed through the tail vein with 5 × 10 5 colony-forming units (CFUs) in PBS. Mice were euthanized 72 h post-infection, spleens and livers were aseptically collected and CFUs counted. Blood was recovered from mice heart. Mouse survival was assessed upon intravenous infection of 10 5 CFUs. Animals were intraperitoneally injected with 10 5 CFUs (Lm) or 5 mg/kg of LTA from Staphylococcus aureus (L2515 Sigma) in PBS and euthanized 6 h or 24 h later. Animal procedures followed European Commission (directive 2010/63/EU) and Portuguese (Decreto-Lei 113/2013) guidelines and were approved by the IBMC Ethics Committee and Direção Geral de Veterinária (license 015301).

Statistics
Statistics were carried out with Prism (GraphPad), using unpaired two-tailed Student's t-test to compare means of two groups, and one-way ANOVA with Tukey's post-hoc test for pairwise comparison of means from more than two groups, or with Dunnett's post-hoc test for comparison of means relative to the mean of a control group.

Scavenger receptors are required for Lm uptake by macrophages
To evaluate the overall role of SRs in Lm uptake by eukaryotic cells, we chemically saturated SRs using different pleiotropic compounds (fucoidan, Poly (I)) known to inhibit SRs [26], before Lm infection of human (THP-1) macrophages. Pre-treatment of THP-1 cells with fucoidan severely impaired Lm uptake when compared to non-treated cells (Figure 1a). In addition, the number of intracellular bacteria was also reduced upon SR saturation with Poly (I), but not with its corresponding negative control Poly (C) (Figure 1a). In agreement, pre-treatment of murine macrophagelike cells (Raw and J774 cell lines) with fucoidan also compromised Lm uptake (Figure 1b). These data suggested a role for SRs in Lm uptake by macrophages.
To identify SRs potentially involved in Lm uptake by macrophages, we assessed SR expression profiles by analyzing total RNAs isolated from human and murine macrophage cell lines. Our analysis revealed that, although some of the selected SRs appeared broadly expressed, each cell line presented a specific SR expression profile (Figure 1c). In the tested conditions, SR-A, SRCL, SREC, LOX1 and STAB-1 appeared to be expressed in all cell lines. SR-A was previously proposed to play a crucial role in host defense against Lm infection [9]. Interestingly, STAB-1 was previously implicated in lymphocyte transmigration and apoptotic cell clearance [14], and shown to bind Gram-positive and Gram-negative bacteria in vitro [17]. Since the involvement of STAB-1 in infectious processes was never assessed so far, we further explore its potential role on Lm infection.

STAB-1 is required for Lm uptake by macrophages
During infection, systemic bacteria are sequestered by phagocytes both in the liver and spleen [27]. SRs are expressed by macrophages and may function as phagocytic receptors for bacteria [17]. Aiming at understanding the role of STAB-1 in Lm uptake, we pre-incubated Raw macrophages with anti-IgG (negative control) or anti-STAB-1 antibody before Lm infection. While the percentage of adherent bacteria was similar between IgG-and anti-STAB-1 treated cells, the percentage of intracellular Lm diminished upon macrophage treatment with anti-STAB-1 antibody (Figure 2a). These data suggest that saturating STAB-1 on the surface of macrophages reduces Lm uptake. To further address the role of STAB-1 in Lm uptake by macrophages, bone marrow-derived macrophages (BMDMs) from both WT and STAB-1 KO mice were infected with Lm. As compared to WT, STAB-1 KO macrophages displayed decreased numbers of intracellular Lm (Figure 2b). Immunofluorescence quantifications of the percentage of infected cells and the number of intracellular bacteria per cell confirmed the reduced capacity of STAB-1 KO macrophages to uptake Lm (Figure 2c-2e). STAB-1 KO macrophages also displayed a slight phagocytosis defect of nonpathogenic Listeria (Listeria innocua -Li) [28], as well as of latex beads (Figure 2b-2d). However this defect appeared more pronounced for Lm than for Li or beads. To investigate the potential role of STAB-1 in Lm internalization by non-phagocytic cells, we engineered an HeLa cell line ectopically and stably expressing STAB-1, using a lentiviral approach. Both control (Mock) and HeLa expressing STAB-1 (STAB-1) cells were infected with Lm and numbers of adherent and intracellular bacteria were quantified. Results indicated that STAB-1 plays no significant role in bacterial internalization into non-phagocytic cells ( Figure S1). STAB-1 appears thus to be involved in the uptake of foreign bodies by macrophages.

STAB-1 has a protective role against Lm infection
To investigate the contribution of STAB-1 during Lm systemic infection in vivo, WT and STAB-1 KO mice were intravenously infected with Lm. Three days later, mice were euthanized and bacterial loads in spleens and livers were quantified. Bacterial numbers appeared significantly higher in the organs of STAB-1 KO mice (Figure 3a), demonstrating a role for STAB-1 in the control of Lm infection. To test if this defect in the control of infection may lead to increased mortality, mice were intravenously infected with a lower dose of Lm and survival was monitored over time. Whereas WT mice survived throughout the infection, mortality in STAB-1 KO mice reached 80% by d 18 (Figure 3b).
Altogether, our data indicate that STAB-1 promotes protection against Lm infection.

STAB-1 is required for an efficient inflammatory response and immune cell accumulation in Lm-infected spleens
Mouse infection by Lm induces a robust innate inflammatory response that restricts bacterial growth prior to the development of protective T cell responses. Early protective immunity against Lm relies on the production and balance of pro-inflammatory cytokines, such as TNF-α and IL-6, and anti-inflammatory cytokines, such as IL-10 [29]. To analyze the potential role for STAB-1 in the production of microbicidal mediators in response to Lm infection, WT and STAB-1 KO mice were intravenously infected with Lm. The production of cytokines in the spleens, livers, and sera of Lminfected mice was evaluated by ELISA 3 d postinfection. As compared to WT animals, infected STAB-1 KO mice produced lower levels of TNF-α, IL-6 and IL-10 ( Figure 4a). Importantly, this reduction of the cytokine levels between WT and STAB-1 KO mice is not observed in absence of infection ( Figure S2). These results indicate that STAB-1 plays a role in the coordinated cytokine production elicited by Lm infection in targeted mouse organs.
Lm entering the bloodstream are rapidly taken up by various myeloid cells in tissues. In the spleen, bacteria are filtered by resident myeloid cells, including dendritic cells and professional phagocytes [30]. Inflammatory stimuli also induce the recruitment of inflammatory macrophages to infected tissues [31]. As STAB-1 appears to regulate the production of inflammatory cytokines in response to Lm infection, in particular in the spleen, we hypothesized that STAB-1 could impact innate immune cells recruitment to the infected spleen, a major site of bacteria replication. To test this hypothesis, WT and STAB-1 KO mice were intravenously infected with Lm, and 3 d post-infection, singlecell spleen suspensions were analyzed regarding myeloid cell populations by flow cytometry. As compared to WT-infected mice, Lm-infected STAB-1 KO mice showed a clear defect on myeloid CD11b int/hi cells, which resulted from the diminished number of neutrophils (CD11b hi Ly6G hi CD11 C − ) and macrophages (CD11b hi Ly6C int/-Ly6G − CD11c lo ) (Figure 4b and S3). Interestingly, within the macrophage population, the number of inflammatory monocytes (CD11b hi Ly6C hi Ly6G − CD11c lo ) was also reduced in infected STAB-1 KO animals (Figure 4b and S3). In the absence of infection, spleens of WT and STAB-1 KO mice showed comparable myeloid cell populations ( Figure S5). Taken together, these data show that STAB-1 is important in controlling the recruitment of neutrophils and macrophages to the spleen of Lminfected mice.
The migration and positioning of immune cells in tissues in response to infection is mainly controlled by chemokines [32]. We thus analyzed the expression of neutrophil-and monocyte-attracting chemokines in Lm-infected murine organs. In infected STAB-1 KO mouse livers, the expression of all chemokines tested was decreased as compared to WT infected mice (Figure 4c, left graph). In infected spleens, the expression of CCL7 and CXCL10 was also decreased in STAB-1 KO mouse spleens, whereas the expression of CCL2 was increased as compared to WT (Figure 4c, right graph). Differences observed between organs might be the result of niche/microenvironment disparities. Altogether, these results indicate a role for STAB-1 in the recruitment of immune cells to Lm-infection sites possibly through the expression control of attracting chemokines. These data also corroborate the role of CCL7 and CXCL10 in the recruitment of inflammatory monocytes to the spleen.

STAB-1 is important for early myeloid cells recruitment in response to Lm infection
The early recruitment of immune cells to infected tissues was shown to be crucial for an effective innate immune response against Lm [31]. To evaluate the role of STAB-1 in the early trafficking of myeloid cells to the site of Lm infection, WT and STAB-1 KO mice were intraperitoneally infected with Lm. Exudate cells from the peritoneal cavity were recovered 6 h or 24 h post-infection to evaluate myeloid cell populations by flow cytometry. When compared to non-infected mice, Lm infection appeared to trigger the recruitment of cells to the focus of infection, mainly neutrophils (CD45 + CD11b hi Ly6G hi CD11C − ) and inflammatory monocytes (CD45 + CD11b hi Ly6C hi Ly6G − CD11c lo ) (Figure 5a-C). While similar cell populations were detected in non-infected WT and STAB-1 KO mice (Figure 5a and S3), we observed a defect recruitment  Lipoteichoic acids are components of Gram-positive bacteria and potent inducers of inflammation. They stimulate immune cells and induce the migration of myeloid cells to the mouse abdominal cavity when injected intraperitoneally [33,34]. We used this experimental model to confirm the involvement of STAB-1 in the recruitment of innate immune cells to the infection site. Purified LTA was intraperitoneally injected into WT and STAB-1 KO mice and, 6 h post-stimulation, exudate cells from the peritoneal cavity were recovered to evaluate myeloid cell populations. In response to LTA, STAB-1 KO mice showed a significant reduction in the myeloid cell (CD45 + CD11b int/hi ) population when compared to WT mice, which correlates to a decreased recruitment of neutrophils (CD45 + CD11b hi Ly6G hi CD11C − ), macrophages (CD45 + CD11b hi Ly6C int/-Ly6G − CD11c lo ) and inflammatory monocytes (CD45 + CD11b hi Ly6C hi Ly6G − CD11c lo ) (Figure 5d). Altogether, these results indicate that, in vivo, STAB-1 potentiates the recruitment of immune cells to the infection site upon an inflammatory stimulus.

STAB-1 expression is decreased and re-localized in response to Lm infection
Since STAB-1 appeared to restrain Lm infection by regulating cytokine and chemokine production and controlling myeloid cell recruitment, we investigated the potential impact of Lm infection on STAB-1 expression. We analyzed STAB-1 expression in murine macrophage-like cells (J774) in response to Lm infection and showed a slight decrease of STAB-1 expression in infected as compared to non-infected macrophages ( Figure 6a). As observed in macrophage cell lines, Lm infection also induced the downregulation of STAB-1 expression in mouse BMDMs at the RNA and protein level (Figure 6(a,b)). Interestingly, this downregulation of STAB-1 expression in BMDMs was only observed with pathogenic Listeria (Lm) and not with the nonpathogenic species (Li) (Figure 6a and 6b). from two independent experiments per group. *p < 0.05; **p < 0.01; ***p < 0.001. (b) Spleen cells from Lm (5 × 10 5 CFU) infected WT and STAB-1 KO mice were isolated and analyzed by flow cytometry. Total numbers of myeloid cells (CD11b int/hi ), neutrophils (CD11b hi Ly6G hi CD11C − ), dendritic cells (CD11b int CD11c hi Ly6G − ), macrophages (CD11b hi Ly6C int/-Ly6G − CD11c lo ) and inflammatory monocytes (CD11b hi Ly6C hi Ly6G − CD11c lo ) are shown. Data are presented as scatter plots, with each animal represented by a dot and the mean indicated by a horizontal line. *p < 0.05; **p < 0.01. (c) WT and STAB-1 KO mice were intravenously infected with 5 × 10 5 CFU of Lm. Mice were sacrificed at d 3 post-infection to recover spleens and livers. The expression of chemokines CCL2, CXCL10 and CCL7 was quantified by qRT-PCR. Data are represented as an average of 10 mice from two independent experiments per group. *p < 0.05; **p < 0.01; ***p < 0.001.  As we showed that STAB-1 is required for an efficient immune response during Lm infection of mouse spleens, we thus assessed the impact of Lm infection on STAB-1 expression in vivo, in splenic tissue. The analysis indicated that Lm infection also induced a large decrease of STAB-1 expression in infected mouse tissues (Figure 6c). In the spleen, STAB-1 is not expressed by splenic macrophages but mainly by endothelial cells [13] that were shown to be active participants in the inflammatory response during Lm infection [35]. Therefore, we evaluated whether Lm infection could impact STAB-1 expression on endothelial cells (HUVECs) and showed again a significant decrease of STAB-1 levels in infected cells (Figure 6d). Immunofluorescence analysis corroborated the downregulation of STAB-1 expression upon Lm infection of endothelial cells (Figure 6e). In addition, by microscopy, we also observed a de-localization of STAB-1 from the host cell membrane upon infection (Figure 6e). This was further confirmed by cell fractionation that showed a significant reduction of membrane STAB-1 in Lm-infected HUVECs when compared to non-infected cells (figure 6f).
Together, these results indicate that infection by pathogenic Listeria induces a downregulation of STAB-1 expression in infected cells and tissues. This downregulation is accompanied by a de-localization of STAB-1 from the host cell membrane.

Discussion
Scavenger receptors (SRs) are transmembrane cell surface glycoproteins restricted to macrophages, dendritic cells, endothelial cells and a few other cell types [36]. Whereas SRs were initially defined by their ability to bind modified low-density lipoproteins, several SRs were since demonstrated to play an important role in innate immune defenses [37].
Here, we show for the first time the important role of the SR STAB-1 in the host protection against bacterial infection. We demonstrate that, during an infection by Listeria monocytogenes (Lm), STAB-1 is not only required for bacterial uptake by macrophages but also for an efficient inflammatory response, immune cell accumulation, and early myeloid cells recruitment to the infection site. Interestingly, we also show that infection by pathogenic Listeria induces the downregulation of STAB-1 expression and its de-localization from the cell membrane, suggesting a bacterial active virulence process targeting STAB-1 aiming to promote infection. STAB-1 appears thus as a new important player in the host protection against a major Gram-positive foodborne pathogen.
SRs were previously shown to represent an important part of the innate immune defense, in particular by acting as phagocytic receptors for microorganisms [38][39][40]. In addition, STAB-1 was previously found to bind Gram-positive and Gram-negative bioparticles in vitro [17], and was described to be a phagocytic receptor mediating efferocytosis by recognizing phosphatidylserine on apoptotic cells [16]. We show here the reduced ability of STAB-1 KO macrophages to uptake not only Lm but also nonpathogenic bacteria and beads. This could suggest a role for STAB-1 in the general phagocytic process. However, it was previously documented that antibody blockade or absence of STAB-1 is sufficient to skew macrophages from an anti-inflammatory to a more pro-inflammatory phenotype [15] ( Figure  S4), these later being inherently less phagocytic for Lm or latex beads than their anti-inflammatory counterparts [41]. This could be responsible, at least in part, for the decreased phagocytic capacity observed for STAB-1 KO macrophages.
We report here that STAB-1 contributes in the host response against Lm infection by controlling cytokine and chemokine production, thus controlling myeloid cell recruitment. SRs are strong players in the regulation of inflammation, such is the case of SR-A in Neisseria meningitidis and Porphyromonas gingivalis infections [42,43] or CD36 in response to Staphylococcus aureus [44]. STAB-1 was previously shown to control the activation of several proinflammatory cytokines in human monocytes [45]. Here, we show that infected STAB-1 KO mice produced reduced serum, liver and splenic levels of IL-6 and TNF-α as compared to WT mice, suggesting that infected with Lm for 30 min. STAB-1 expression levels in infected conditions were normalized to those in non-infected BMDMs, arbitrarily fixed to 1. (b-d) Independent immunoblots to detect STAB-1 protein in (B) BMDMs left uninfected (NI) or infected with Lm or Li for 30 min, (c) spleen of NI and Lm-infected mice for 3 d (5 × 10 5 CFU), (d) NI and Lm-infected HUVECs. Immunoblots quantification of STAB-1 signal intensity in NI and infected conditions, normalized to GAPDH. (e) Immunofluorescence images of NI and Lm infected HUVECs, stained with an anti-STAB-1 antibody. DNA counterstained with DAPI and actin labeled with TRITCconjugated phalloidin. Scale bar, 10 μm. (f) Immunoblots to detect STAB-1 protein in the cytoplasmic and membrane fractions of NI and Lm-infected HUVECs upon cell fractioning. cMet was used as a membrane loading protein control and GAPDH as a cytoplasmic loading protein control. Immunoblots quantification of STAB-1 signal intensity in NI and infected conditions, normalized to cMet or GAPDH.
STAB-1 participates in the regulation of the inflammatory cytokine response in Lm-targeted mouse organs upon infection. In agreement, IL-6-and TNF-α-deficient mice were shown to be more susceptible to Lm infection, with increased bacterial burden in the spleen and liver, and deficient neutrophil recruitment into the blood [46,47]. Surprisingly, we also observed an IL-10 decreased expression in STAB-1 KO mice upon Lm infection. IL-10 is a potent inhibitor of innate immunity and IL-10 deficiency was shown to improve resistance to Lm infection [48].
We show here that STAB-1 appears as an important regulator of the pro-/anti-inflammatory cytokine balance in response to Lm infection. Lm uses a balance of pro-and anti-inflammatory mechanisms to promote infection while inducing little inflammation in the host, both at the intestinal level and systemically [49]. In a rodent model of intestinal infection, differential induction of pro-or anti-inflammatory responses depending on the cell type used for entry were observed [50]. Thus, a picture emerges that the modulation of pro-and anti-inflammatory properties at the cellular level has important consequences for the course of infection when monitored in the complex host environment. Whereas STAB-1 was previously proposed as an immunosuppressive molecule, suggesting that STAB-1 may dampen pro-inflammatory reactions in vivo [45], our results rather tend to indicate a proinflammatory role for STAB-1 during Lm infection. In the context of sepsis, where the pro-inflammatory response predominates, STAB-1 was previously proposed to be both an immunosuppressive player to downregulate hyper-inflammation at early stages and to act as a vascular barrier keeper in later stages of sepsis. STAB-1 could thus appear as a regulator of inflammatory processes, acting both as a pro-and antiinflammatory molecule depending of the context and localization in the host.
Our findings also indicate that STAB-1 plays a role in the recruitment of myeloid cells in infected organs, a recruitment that appears to be dependent on chemokine expression, in particular CXCL10 and CCL7. CXCL10 was previously involved in immune cell migration, differentiation and activation [51], and is induced by TNF-α [52]. The reduced levels of CXCL10 in STAB-1 KO mice upon Lm infection might thus correlate with the concomitant decreased expression of TNF-α. Interestingly, CXCL10 has been shown to have direct antibacterial properties similar to α-defensins, in particular against Lm [53]. During Lm infection, the recruitment of inflammatory macrophages from bone marrow to sites of microbial infection was shown to be dependent on CCR2, a chemokine receptor that responds to CCL2 and CCL7 [54]. In our experimental model, CCL7 seems to play a more prominent role in the STAB-1-dependent recruitment of myeloid cells to Lm infected organs. However, the slight increase of CCL2 expression observed in the spleen of Lm-infected STAB-1 KO mice could also suggest a role of STAB-1 in myeloid cell chemotaxis.
We show that the deficiency on pro-inflammatory cytokine production and the defect on myeloid cell recruitment, which are crucial for the initial control of bacterial replication, lead to higher bacterial loads in the spleen and liver of STAB-1 KO mice. Neutrophils and macrophages, which are effective microbicidal cells, are among the first cells involved in the Lm-immune response. Mice deficient for these cells present increased bacterial burden and mortality [55]. The reduced capacity of STAB-1 KO mice to fight Lm infection appears thus to be more related to a deficiency in the recruitment of myeloid cells to target organs, than to a killing deficiency. In agreement, STAB-1, which was shown to be absent from all splenic macrophages, including red pulp, marginal zone and metallophilic macrophages, is solely expressed by the vascular endothelium [56], and is known to be involved in the transmigration of immune cells [57,58]. Nevertheless, impaired control of the infection by STAB-1 KO mice may not be only due to reduced myeloid cell recruitment to the sites of infection. STAB-1 could be required for the proper migration, position, and function of other immune cells, in particular CD8α + dendritic cells of the splenic marginal zone that were shown to be an obligate cellular entry point for a productive infection by Lm [59]. However, it was previously shown that the frequencies of splenic CD4+ and CD8 + T cells were comparable between wildtype and STAB-1 KO mice [18]. Our work focuses on the importance of STAB-1 in the early immune response of Lm, but further studies need to be performed to understand whether the deficiency of STAB-1 in the context of Lm infection interferes with T cell priming and the generation of T cell memory.
Whereas STAB-1 appears as an important player in the host protection against Lm, we also show that infection by this bacterial pathogen induces a decreased expression of STAB-1 in macrophages and endothelial cells but also in vivo in infected mice spleen, which is a major target organ for Lm replication. The expression of some other SRs, such as SR-A, MARCO, and LOX-1, was also reported to be modulated by microbial infection, either favoring host immune response or promoting pathogen survival [42,60,61]. In particular, a marked expression of MARCO was observed in response to Leishmania major and Lm infections [10,62]. Interestingly, we also found that Lm infection leads to a delocalization of STAB-1 from the membrane of endothelial cells. Importantly, the downregulation of STAB-1 expression was not observed with the nonpathogenic specie Listeria innocua, that essentially differs from Lm by the absence of major virulence factors [28]. This suggests an active mechanism driven by Lm virulence factors to control STAB-1 expression/localization, diminish host protective responses, and promote infection. However, the identification of the specific virulence factors potentially involved in this process requires further investigation.
Here, we highlight for the first time that STAB-1 plays a protective role during Lm infection. By regulating the inflammatory response and the recruitment of myeloid cells, STAB-1 appears as a new SR with an important role for the host response against Lm infection. Amplifying STAB-1-mediated host defenses may represent an innovative strategy against Gram-positive pathogens.