A role for the Sts phosphatases in negatively regulating IFNγ‐mediated production of nitric oxide in monocytes

Abstract Introduction The atypical Sts phosphatases negatively regulate signaling pathways in diverse immune cell types, with two of their molecular targets being the related kinases Syk and Zap‐70. Mice lacking Sts expression (Sts −/−) are resistant to infection by the live vaccine strain (LVS) of Francisella tularensis. Although the mechanisms underlying the enhanced resistance of Sts −/− mice have not been definitively established, Sts −/− bone marrow‐derived monocytes (BMMs) demonstrate greater clearance of intracellular LVS following ex vivo infection, relative to wild type cells. To determine how the Sts proteins regulate monocyte bactericidal properties, we analyzed responses of infected cells. Methods Monocyte bacterial clearance was assayed using ex vivo coculture infections followed by colony‐forming unit analysis of intracellular bacteria. Levels of gene expression were quantified by quantitative reverse‐transcription polymerase chain reaction, levels of Nos2 protein levels were quantified by Western blot analysis, and levels of nitric oxide (NO) were quantified directly using the Griess reagent. We characterized monocyte cytokine production via enzyme‐linked immunosorbent assay. Results We demonstrate that Sts −/− monocyte cultures produce elevated levels of interferon‐γ (IFNγ) after infection, relative to wild type cultures. Sts −/− monocytes also demonstrate heightened responsiveness to IFNγ. Specifically, Sts −/− monocytes produce elevated levels of antimicrobial NO following IFNγ stimulation, and this NO plays an important role in LVS restriction. Additional IFNγ‐stimulated genes, including Ip10 and members of the Gbp gene family, also display heightened upregulation in Sts −/− cells. Both Sts‐1 and Sts‐2 contribute to the regulation of NO production, as evidenced by the responses of monocytes lacking each phosphatase individually. Finally, we demonstrate that the elevated production of IFNγ‐induced NO in Sts −/− monocytes is abrogated following chemical inhibition of Syk kinase. Conclusion Our results indicate a novel role for the Sts enzymes in regulating monocyte antibacterial responses downstream of IFNγ.

that the elevated production of IFNγ-induced NO in Sts −/− monocytes is abrogated following chemical inhibition of Syk kinase. Conclusion: Our results indicate a novel role for the Sts enzymes in regulating monocyte antibacterial responses downstream of IFNγ.

K E Y W O R D S
Francisella tularensis, IFNγ, monocytes, nitric oxide, signal transduction pathways, Syk

| INTRODUCTION
Francisella tularensis is a Gram-negative intracellular bacterial pathogen that is the causative agent of tularemia. [1][2][3] Following introduction into the host, F. tularensis proliferates rapidly and spreads to a variety of internal organs, reaching peak bacterial load within several days. 4,5 Bacterial spread is aided by the bacterial life cycle, which includes uptake by highly mobile cellular components of the innate immune response such as macrophages and other phagocytic cells, rapid escape from the phagosomal compartment before formation of the phago-lysosome, and extensive proliferation within the cytoplasm of host cells. 6,7 Simultaneously, it optimizes its intracellular niche by suppressing the activation of host apoptotic and inflammatory response pathways. 8 Unlike the pathogenic F. tularensis subspecies tularensis and holarctica, the F. tularensis live vaccine strain (LVS) is non-pathogenic in humans, but mimics in mice the clinical symptoms of tularemia. This makes it a useful model to study the regulation of host immune responses that are directed against a virulent intracellular bacterial pathogen. 9,10 IFNγ is a pleitropic cytokine that promotes a wide variety of host innate immune responses that are critical for overcoming an F. tularensis infection. 4,11,12 For example, within some innate cells such as macrophages it upregulates production of reactive oxygen and nitrogen species that are critical antimicrobial effector molecules. 13 It also upregulates expression of diverse members of the guanylate-binding protein (GBP) family of proteins that play a role in antibacterial host defense, 14,15 as well as chemokines such as Cxcl10. 16 Unsurpringly, Francisella species have evolved mechanisms to counteract IFNγ-mediated effector responses. For example, F. tularensis induces elevated expression of Socs-3, a known inhibitor of Stat1 phosphorylation. 17,18 As Stat1 mediates many of the downstream effects of IFNγ, upregulation of Socs-3 expression can allow F. tularensis to gain a survival advantage within infected cells.
Sts-1 and Sts-2, two homologous phosphatases that act as negative regulators of immune signaling pathways, are structurally and enzymatically very distinct from many phosphatases, including protein tyrosine phosphatases (PTPs). [19][20][21] Originally characterized as suppressors of TCR signaling through their ability to target the T cell kinase Zap-70, 22 the Sts phosphatases are now understood to negatively regulate diverse signaling pathways within multiple cell types, including mast cells, platelets, and BMDCs. [23][24][25] In addition to regulating TCR signaling, Sts-1 also regulates both GPVI-FcRγ signaling in platelets and FcεRI signaling in mast cells, by targeting the Zap-70 homologue Syk. 26,27 Previous studies have revealed that the absence of Sts expression can substantially alter the outcome of a pathogen infection. For example, Sts −/− mice are profoundly resistant to systemic infection by the human fungal pathogen Candida albicans, displaying significantly enhanced fungal clearance and reduced inflammation after a lethal bloodstream inoculum. 28 Sts −/− mice have also been shown to be significantly resistant to intradermal infection by F. tularensis LVS. 29 Although the cellular and molecular mechanisms underlying the increased resistance of Sts −/− mice are not well defined, we have demonstrated that bone marrow-derived monocytes (BMMs) isolated from Sts −/− animals exhibit greater bactericidal activity ex vivo relative to their wild type counterparts. 29 This suggests that the absence of Sts could potentiate antimicrobial effector activities within select monocyte populations, leading to increased in vivo resistance.
In this study, we extend our investigation of the role of Sts in regulating monocyte responses to LVS infection. We demonstrate that IFNγ plays an important role in the monocyte response to LVS, and that the Sts proteins regulate IFNγ-induced monocyte antimicrobial responses. These results suggest a novel function for Sts in regulating host responses to a virulent bacterial pathogen.
Resources (DLAR) guidelines, and all animal procedures were performed in accordance with the US National Research Council's Guide for the Care and Use of Laboratory Animals and approved by the Stony Brook University Institutional Animal Care and Use Committee (IACUC). In addition, ARRIVE guidelines established by the National Centre for the Replacement, Refinement, and Reduction for Animals in Research (NC3Rs) were strictly adhered to.

| Mouse strains and bacterial cells
Mice containing the Sts mutations, backcrossed 10 generations onto the C57/B6 background, have been described. 30 Mice were housed in the Stony Brook University Animal Facility under enhanced germ-free conditions. F. tularensis LVS (ATCC 29684) was grown on chocolate agar plates for 48 hours. Single colonies obtained were then used for an overnight culture in modified Mueller-Hinton broth containing 1% glucose, 0.025% ferric pyrophosphate, and 0.05% L-cysteine. Cultures were washed and resuspended in phosphatebuffered saline (PBS) to achieve desired bacterial concentrations.

| Ex vivo monocyte infections
The preparation of BMM was performed as described. 31 Briefly, cells were isolated from femurs of 6-to 8-weekold mice and cultured for 4 days in bone marrow medium; Dulbecco's modified Eagle's medium with GlutaMax (Invitrogen) supplemented with 30% L929 cell supernatant, 20% fetal bovine serum (FBS; Invitrogen), and 1 mM sodium pyruvate). Non-adherent cells were collected and suspended in BMM. BMM (2 × 10 6 ) were seeded in triplicate in 24-well plates and infected with F. tularensis LVS at the indicated multiplicity of infection (MOI): MOI 5 for non-cytokine treated cultures and MOI 20 for IFNγ-treated cultures containing activated cells. Plates were spun (700 rpm, 5 minutes, room temperature) to facilitate contact between cells and bacteria. Cells were infected for 2 hours, then washed with PBS and incubated for 1 hour with 50 μg/mL gentamicin. After antibiotic treatment, cells were washed with PBS and incubated with fresh media (lacking gentamicin) at 37°C until indicated time-points. To assess bacterial CFUs, cells were lysed with 0.2% deoxycholic acid, lysates were serially diluted, and dilutions were plated onto chocolate agar plates for colony enumeration after 2 to 3 days growth at 37°C. Where indicated, cells received a 2 hours treatment with chemical inhibitors followed by 50 ng/mL of mouse IFNγ (#5222; Cell Signalling Technology) or TNF-α (#5178; Cell Signalling Technology) for an additional 2 hours before infection. Cells were then infected in triplicate at the indicated MOI.

| Flow cytometry
Cells were suspended in PBS buffer containing 2% FBS, stained with Fc block for 20 minutes followed by staining with PE-conjugated antibody to IFNGR1 (CD119, 130-104-934; Miltenyi Biotec). Histogram plots represent the number of CD119 + cells in the FSCxSSC live-cell gate. Flow cytometric data were acquired with a FACScan cytometer (Cytek Biosciences, Inc.) and analyzed with software provided by FlowJo, LLC (www.flowjo.com).

| Increased IFNγ produced by F. tularensis LVS-infected Sts −/− monocytes
BMMs that are mobilized in response to bacterial infections within peripheral tissues are known to be potent inducers of pro-inflammatory interferons (eg, IFNγ), as well as additional cytokines and chemokines such as IL-1, IL-18, Ccl3, Cxcl9, and others. 32,33 Because monocytes lacking Sts expression display increased bactericidal activity toward internalized F. tularensis LVS ex vivo, 29 we examined whether cytokine production was deregulated. We infected wild type and Sts −/− BMMs ex vivo with LVS, and evaluated supernatant cytokine levels 24 hours postinfection. We detected significantly increased levels of IFNγ in Sts -/− culture supernatants relative to wild type cells, following infection at MOI 5 ( Figure 1A). For infections with a bacterial MOI 20, we detected equivalent levels of IFNγ in both cultures, at levels that were substantially reduced from the levels observed in the MOI 5 infection condition ( Figure 1A). In contrast to the for 24 hours, after which levels of IFNγ in culture supernatants was determined. Results represent mean ± SD of three independent experiments, each carried out in triplicate. *P < .05; **P < .01 (by Student's t test). ELISA, enzyme-linked immunosorbent assay; IFNγ, interferon-γ; IL-6, interleukin 6; TNF-α, tumor necrosis factor-α; WT, wild type increased IFNγ within Sts −/− cultures following infection at either MOI, no differences in the levels of proinflammatory cytokines tumor necrosis factor-α (TNF-α) or interleukin 6 (IL-6) were observed between wild type and Sts −/− cultures ( Figure 1B,C). The increased IFNγ produced within Sts −/− BMM cultures only occurred following infection with live pathogen, as treatment of cells with heat-killed bacteria yielded undetectable levels of IFNγ production ( Figure 1D).

| Role of IFNγ in potentiating responses of BMMs to intracellular F. tularensis LVS
Previously, we demonstrated that Sts −/− BMMs restrict intracellular F. tularensis LVS with greater efficiency than wild type cells. 29 To investigate a functional role for IFNγ during ex vivo monocyte responses, we cultured infected monocytes in the presence of exogenous IFNγ. Under these conditions, Sts −/− monocytes displayed significantly enhanced clearance of LVS relative to wild type cells (Figure 2A). This was not due to increased surface expression of IFNGR1 (CD119), because wild type and Sts −/− BMMs express equivalent levels of surface CD119 ( Figure 2B). To further investigate the functional role of IFNγ, we utilized a blocking antibody to inhibit its activity. Addition of an IFNγ-blocking antibody significantly impaired bacterial restriction, as evidenced by a greater than 100-fold increase in levels of recovered intracellular bacterial colony-forming units (CFUs) ( Figure 2C). Addition of the blocking antibody also eliminated the difference in recovered CFUs that was previously observed between wild type and Sts −/− cells ( Figure 2C). We also blocked IFNγ signaling by utilizing the purine analog fludarabine, which has been shown to inhibit the activation of Stat1. [34][35][36] Stat1 is an important transcription factor that mediates many responses downstream of IFNγ. 37 Similar to treatment of cells with IFNγ-blocking antibody, fludarabine abrogated the ability of infected (MOI 20) monocytes cultured with exogenous IFNγ to restrict intracellular LVS ( Figure 2D). Unlike IFNγ, addition of exogenous TNF-α did not potentiate the ability of Sts −/− monocytes to clear LVS (MOI 20) relative to wild type cells, nor did a blocking antibody abrogate bacterial clearance ( Figure 2E). Overall, these results support the hypothesis that the Sts proteins negatively regulate an antimicrobial response downstream of IFNγ.

| Regulation of IFNγ-induced nitric oxide production by the Sts proteins
IFNγ-dependent clearance of intracellular pathogens is mediated by diverse IFNγ-regulated antimicrobial effector mechanisms, including the generation of reactive nitrogen species. [38][39][40] To determine whether IFNγinduced nitric oxide (NO) contributed to the bacterial clearance, we determined the consequences of treating cells with the NO inhibitor 1400 W. 41 In the presence of 1400W, the ability of both wild type and Sts −/− cells to clear intracellular bacteria was attenuated (Figure 3), and the difference in recovered CFUs normally observed between wild type and Sts −/− cells was eliminated. These results suggest an important bactericidal role for IFNγmediated NO production in the clearance of LVS under ex vivo conditions, and also suggest a role for Sts in regulating the production of IFNγ-induced NO.
To further examine whether the Sts proteins have a role in regulating NO production downstream of IFNγ, we examined levels of NO in infected culture supernatants. IFNγ-primed infected Sts −/− monocytes displayed significantly higher NO production than WT cells by 24 hours postinfection ( Figure 4A). This effect was eliminated by fludarabine treatment (Figure 4B). The increased nitrite observed within Sts −/− culture supernatants correlated with higher levels of iNos gene induction in IFNγ-primed infected Sts −/− BMM relative to wild type cells ( Figure 4C).
To examine more closely the role of Sts in regulating the IFNγ-NO pathway, we stimulated cells with IFNγ in the absence of bacterial infection. Under these conditions, Sts −/− monocytes also released significantly greater nitrite into culture supernatants than comparably treated wild type cells ( Figure 4D). Consistently, addition of fludarabine to IFNγ-stimulated Sts −/− monocytes significantly reduced the levels of NO secreted into culture supernatants, and abrogated the difference in nitrite levels between wild type and Sts −/− cells (data not shown). The absence of Sts expression also led to heightened levels of iNos gene induction following stimulation with IFNγ ( Figure 4E), as well as increased expression of Nos2 protein ( Figure 4F). Finally, to assess the role of each individual Sts homologue, we stimulated Sts-1 −/− and Sts-2 −/− monocytes with IFNγ and assessed production of NO relative to wild type and Sts doubly-deficient monocytes. Deletion of each Sts homologue individually resulted in intermediate IFNγ-dependent levels of nitrite that were significantly greater than observed in wild type cultures, but less than that observed in cultures with monocytes lacking both Sts homologues ( Figure 5).

| A role for Sts in regulating expression of IFNγ-induced Gbp genes
To determine if IFNγ-induced genes other than Nos2 were regulated by Sts, we evaluated induction of additional IFNγ-induced immune effector genes. The guanylate-binding proteins (GBPs) are a family of 65 to 73 kDA GTPases (11 Gbp genes in mice), each with a conserved structure consisting of a C-terminal regulatory domain, an assembly domain, and a GTPase effector domain that binds GTP. They are known to be strongly induced by IFNγ treatment and recent studies have uncovered distinct antibacterial functions associated with different family members. 42,43 Cxcl10, also known as IP10, is a chemokine that is also strongly induced by IFNγ. We examined induction of Gbp2, Gbp4, Gbp5, Gbp6, and Ip10 in wild type and Sts −/− BMMs, following treatment with IFNγ. As illustrated in Figure 6A, all of the genes examined demonstrated two to four greater fold F I G U R E 3 iNos inactivation abrogates the bacterial clearance advantage displayed by Sts −/− monocytes. IFNγ-stimulated monocytes pretreated with iNos inhibitor 1400W as indicated were infected with Francisella tularensis LVS (MOI 20) for 24 hours. Cells were lysed at the indicated times and intracellular bacterial CFUs enumerated. Results represent mean ± SD of three independent experiments, each carried out in triplicate. **P < .01 (by Student's t test). CFU, colony-forming unit; IFNγ, interferon-γ; LVS, live vaccine strain; WT, wild type upregulation in IFNγ-stimulated Sts −/− BMMs compared with stimulated wild type cells. A similar result was obtained when cells were primed with IFNγ for 2 hours before a 4 hour infection with LVS ( Figure 6B). These results indicate that Sts-mediated regulation of IFNγ signaling is not confined to the Nos2 signaling axis, but extends to multiple biological activities of IFNγ.

| Syk inhibition abrogates IFNγ-induced NO production
Spleen tyrosine kinase (Syk), a non-receptor protein tyrosine kinase that is expressed in many myeloid cell types, has been implicated as an Sts substrate. 24,25 To investigate a potential role for Syk in the response of BMMs to IFNγ stimulation and F. tularensis LVS infection, we evaluated the effects of Syk inhibition in regulating the production of IFNγ-induced NO. Treatment of IFNγ-stimulated monocytes with two specific Syk inhibitors, piceatonnol 44 and R406, 45 reduced culture supernatant nitrite levels and also reduced the differential response observed between wild type and Sts −/− cells ( Figure S1). These observations support a role for Syk in mediating IFNγ-dependent signaling within BMMs, and identify a mechanism whereby the Sts proteins could regulate IFNγmediated production of NO.

| DISCUSSION
This study expands on our previous observation that mice lacking expression of the Sts phosphatases are significantly resistant to intradermal infection by the bacterial pathogen F. tularensis LVS. As part of our earlier study, we investigated potential underlying cellular factors contributing to the enhanced resistance. We observed that Sts −/− BMMs displayed a 10-fold greater restriction of intracellular bacteria relative to wild type monocytes, following infection in ex vivo culture. In the current study, we identify IFNγ as a critical cytokine that promotes the enhanced microbicidal activity of Sts −/− BMMs in the context of the ex vivo infection assay. We initially observed that excessive production of IFNγ, but not TNF-α or IL-6, within Sts −/− monocyte cultures correlated with the heightened ability of mutant monocytes to restrict intracellular LVS. This led us to examine how Sts regulates IFNγ biological activities within BMMs. Importantly, we provide evidence that the Sts proteins negatively regulate IFNγ-dependent antibacterial responses. For example, when exogenous IFNγ is added to an infection culture, Sts −/− cells are able to clear intracellular CFUs more effectively than wild type cells. The greater relative upregulation of Nos2 gene expression within infected Sts −/− cells as compared to wild type cells is also consistent with Sts-mediated regulation of IFNγ biological effects. This latter result suggests that the enhanced CFU clearance demonstrated by Sts −/− cells is in part due to higher levels of IFNγ-dependent NO produced by Sts −/− cells, a conclusion supported by the effects of Nos2 inhibition. Although Nos2 inhibitor treatment did not fully inhibit bacterial restriction, it did abrogate the Sts −/− clearance advantage. This suggests that while production of NO is not the sole effector response promoting bacterial clearance, it is prominently linked to the enhanced bacterial restriction displayed by Sts −/− cells. Regarding the regulation of the NO response, it is possible that the Sts proteins are among many redundant factors that converge on regulating levels of NO, a possibility that might also explain the modest and somewhat variable effect observed in Sts −/− cells.
Interestingly, the potentiation of IFNγ signaling in Sts −/− cells does not require the presence of intracellular Francisella, as we observed increased levels of NO in uninfected Sts −/− culture supernatants following IFNγ treatment. We also noted increased levels of iNos gene induction and higher levels of cellular Nos2 protein in uninfected Sts −/− culture supernatants following IFNγ treatment. Finally, heightened induction of Gbp2, Gbp4, Gbp5, Gbp6, and Ip10 in Sts −/− cells relative to wild type cells was observed following both IFNγ treatment alone and LVS infection of IFNγ-treated cells. Together these results support a role for the Sts proteins in negatively regulating pathways downstream of IFNγ. Future studies F I G U R E 5 Individual roles of Sts-1 and Sts-2 in regulating IFNγ-induced NO production. Bone marrow-derived monocytes were treated for 24 hours with recombinant IFNγ, following which nitrate levels in culture supernatants were evaluated. Results represent mean ± SD of three independent experiments. **P < .01, ****P < .0001 (by Student's t test). IFNγ, interferon-γ; NO, nitric oxide; WT, wild type (A) (B) F I G U R E 6 Enhanced upregulation of IFNγ-induced genes in Sts −/− monocytes. Monocytes were (A) treated with 50 ng/mL IFNγ for 4 hours or (B) treated with 50 ng/mL IFNγ followed by MOI 20 4 hours LVS infection. Fold change in gene expression relative to untreated controls was then analyzed via RT-qPCR. *P < .05, **P < .01, ***P < .001, ****P < .0001 (by Student's t test). GBP, guanylate-binding protein; IFNγ, interferon-γ; RT-qPCR, reverse-transcription polymerase chain reaction; WT, wild type are needed to investigate a potential role for individual GBP family members in restricting Francisella species within monocytes. The mechanisms by which Sts regulates signaling downstream of IFNGR1 are not clear. Canonical IFNγR1 signaling occurs via a Jak/Stat pathway, with the Jak1/ Jak2 kinases and the transcription factor Stat1 being the primary signaling molecules that mediate IFNγdependent responses. Interestingly, Sts-1 was originally identified in a screen for Jak2 interacting proteins, 46 suggesting that the Sts proteins could regulate IFNγdependent signaling by targeting Jak kinases. A recent report highlighted a potential role for Sts in regulating IFNα signaling via Jak1-Stat1. 47 However, to date the majority of Sts functional studies have focused on the role of the Sts proteins in targeting the related kinases Syk and Zap-70 for dephosphorylation and inactivation. For example, bone marrow-derived phagocytes lacking Sts expression display hyper-activation of Syk following stimulation of the anti-fungal CLR Dectin-1, 25,48 and within T cells the Sts proteins target the T cell kinase Zap-70 immediately downstream of TCR activation. 22 These previous observations led us to examine the potential involvement of Syk kinases in IFNγR signaling within BMMs. While the inhibition of IFNγ-dependent NO production by Syk inhibitors suggests Syk could be involved in regulating signaling downstream of IFNγR ( Figure S1), further studies will be necessary to strengthen this preliminary observation.
Our current results are consistent with a critical role for marrow monocytes in mediating the increased resistance of Sts −/− mice to LVS infection. In our model, Sts-/-monocytes that are recruited to peripheral tissues during the early stages of infection respond more acutely to IFNγ and clear the infection more effectively, in part by deploying increased levels of antibacterial NO. It is also possible that the lack of Sts expression impairs the ability of Francisella species to suppress host IFNγ responses and survive inside host cells. We are currently evaluating whether the key elements of this model accurately describe the underlying basis for the Sts −/− in vivo resistance phenotype. Additionally, mechanistic studies to determine how the Sts proteins regulate signaling responses downstream of IFNγ are also underway. Further insights into these pathways could lead to the development of therapeutic strategies to enhance immune-mediated resistance toward diverse Francisella species and other bacterial pathogens.