Pseudomonas aeruginosa PcrV Enhances the Nitric Oxide-Mediated Tumoricidal Activity of Tumor-Associated Macrophages via a TLR4/PI3K/AKT/mTOR-Glycolysis-Nitric Oxide Circuit

Tumor-associated macrophages (TAMs), which display a tumor-supportive M2 phenotype, are closely related to tumor growth and metastasis. The reprogramming of TAMs toward a tumoricidal M1 profile has emerged as an attractive strategy for cancer immunotherapy. In this study, we found that the intratumoral injection of PcrV protein, a component of the Pseudomonas aeruginosa type 3 secretion system, suppressed tumor growth and increased apoptosis, inducible nitric oxide synthase (iNOS) expression, and the percentage of M1-polarized TAMs in tumor tissues. Furthermore, the intratumoral injection of PcrV-primed macrophages exerted a similar tumoricidal effect. In vitro analyses revealed that PcrV reeducated TAMs toward an antitumoral M1 phenotype and augmented their nitric oxide (NO)-mediated cytotoxicity against cancer cells. Mechanistically, we found that these effects were dependent on the activation of Toll-like receptor 4 (TLR4)/myeloid differentiation factor 88 (MyD88)-mediated regulation of a PI3K/AKT/mTOR-glycolysis-NO feedback loop via direct interaction with TLR4. Collectively, these results revealed a potential role for PcrV in cancer immunotherapy through the targeting of TAM plasticity.


INTRODUCTION
Tumor-associated macrophages (TAMs) form the major component of the immune cell infiltrate in the tumor microenvironment (TME) and are correlated with tumor development and progression. Macrophages infiltrated in the immunosuppressive TME are generally induced into the tumor-supportive M2 phenotype. These M2-like TAMs play roles in supporting tumor growth and metastasis and maintaining an immunosuppressive TME by generating a series of anti-inflammatory and protumoral cytokines and mediators (1). As macrophages are highly heterogeneous and plastic, TAMs exposed to M1-associated stimuli can undergo a reversal from the protumoral M2 to the tumoricidal M1 phenotype. M1-like TAMs exert tumoricidal activity through the production of proinflammatory cytokines (e.g., TNFA and IL12), cytotoxic nitric oxide (NO), and reactive oxygen species (ROS) and the activation of Th1 immune responses via increasing the expression of associated major histocompatibility complex class I and class II (MHCI/II) and costimulatory molecules (CD86 and CD80) (2). Accordingly, the reeducation of TAMs from an M2 to an M1 phenotype has emerged as an attractive strategy for cancer immunotherapy.
Recent studies have shown that drugs such as paclitaxel (3) and astragaloside IV (4) can suppress tumor growth by reprogramming TAMs into an M1 phenotype. In addition, several reports have highlighted the potential efficacy of bacteria or their products in antitumoral therapy. For instance, Mycobacterium bovis-derived bacillus Calmette-Gueŕin (BCG) has already been utilized as a preferred first-line treatment for non-muscle invasive bladder carcinoma (5,6). Additionally, modified lipopolysaccharide (LPS), a Toll-like receptor 4 (TLR4) agonist, has been tested for its antitumoral efficacy in a clinical trial (7). Given that a large number of bacterial components, such as Brucella abortus cell-surface protein 31 (BCSP31) protein (8) and Vibrio cholerae porin OmpU (9), can induce functional macrophage M1 polarization by activating the TLR-mediated signaling axis (8,9), these results imply that bacterial products have great potential for exploitation as therapeutic reagents for cancer treatment.
PcrV, a secretory needle tip protein component of the Pseudomonas aeruginosa type 3 secretion system (T3SS), helps the translocator proteins PopB and PopD form a functional pore on the target cell membrane (10). PcrV, a V-antigen, elicits protective immune responses against P. aeruginosa infection (11,12). However, the mechanism underlying the PcrV-mediated regulation of the host immune response remains unknown. We have previously reported that PcrV reverses host immune suppression elicited following bacterial biofilm infection via the activation of macrophagemediated immune responses (13). Considering that TAMs exhibit diverse phenotypes and given their function as tumor-resident macrophages, using PcrV to reeducate TAMs and thereby enhance M1 TAM-mediated antitumoral properties is an attractive possibility that requires further investigation.
In the current study, we found that PcrV inhibited tumor growth by reprogramming TAMs to a tumoricidal M1 phenotype. We further found that PcrV-mediated TAM reprogramming potentiated their NO-mediated cytotoxicity against cancer cells, effects that were exerted through the activation of a PI3K/AKT/mTOR-glycolysis-NO feedback loop via direct interaction with TLR4. Our findings demonstrated that PcrV exerts an immunomodulatory effect on TAMs, providing the basis for further investigation into the potential of PcrV as a therapeutic agent for cancer immunotherapy.

Animals and Ethics Statement
Male C57BL/6 mice were purchased from Biocytogen Co., Ltd (Beijing, China). The TLR4 −/− and myeloid differentiation factor 88 (MyD88) −/− C57BL/6 mice were provided by Professor Qingwu Yang. Animal experiments were conducted according to the experimental animal guidelines of the Army Medical University of China.

Expression and Purification of PcrV Protein
Recombinant PcrV protein was expressed from the expression strain Escherichia coli JM109/pQE31-PcrV as described previously (13). Protein purification was performed using His-Trap HP affinity columns (GE Healthcare, Sweden), and endotoxin was removed using Detoxi-Gel endotoxin removing gel (Thermo Fisher, USA), according to the manufacturer's instructions.

Induction of Bone Marrow-Derived Macrophages and Tumor-Associated Macrophages In Vitro
Mouse bone marrow cells were isolated from the tibia and femur of C57BL/6 mice. Bone marrow-derived macrophages (BMDMs) were induced by the administration of mouse macrophage colony-stimulating factor (M-CSF; 30 ng/ml) in DMEM supplemented with 10% FBS and 100 U/ml of penicillin for 7 days at 37°C with 5% CO 2 . For in vitro TAM induction, BMDMs were cultured in DMEM/FBS (10%) containing 20% (v/v) LLC cell culture supernatant for 24 h (14).

Mouse Tumor Models and Treatment
For PcrV administration, C57BL/6 mice were subcutaneously inoculated with 1 × 10 6 LLC cells. On day 9, the animals were randomly divided into two groups and were intratumorally injected with either phosphate-buffered saline (PBS; control group) or PcrV (0.5 mg/kg, every 4 days). For the BMDM inoculation experiment, mice were subcutaneously inoculated with 1 × 10 6 LLC cells (day 0). On days 0 and 7, the animals were peritoneally injected with 200 ml of clodronate liposomes (Liposoma BV, Netherlands) to deplete endogenous macrophages. PBS or PcrV (0.5 mg/kg) was intratumorally injected at days 12, 14, 16, and 18, while the wild type (WT), TLR4 −/− , or MyD88 −/− BMDMs (2 × 10 6 cells) primed or not with PcrV were intratumorally inoculated at days 16 and 18. At the end of the experiment, the mice were euthanized with pentobarbital, and tumor tissues were harvested for subsequent analyses.

TUNEL Assay
Tissue apoptosis was detected using the TdT-mediated dUTP nick end labeling (TUNEL) assay by staining the tissue sections using an In Situ Cell Death Detection Kit, POD (Roche, USA), at 37°C for 30 min. Nuclei were counterstained with Hoechst 33342, and images were obtained by laser scanning confocal microscopy (Leica TCS SP5, Germany).

Co-culture and Detection of Apoptosis
BMDMs were seeded at 5 × 10 5 cells/well in a six-well plate and treated or not with PcrV (10 mg/ml) for 24 h. After the supernatants were discarded, the macrophages (lower chamber of a Transwell plate [0.4-mm pore; Corning, USA]) were cocultured with LLC cells (upper chamber) in DMEM/FBS (10%) at a ratio of 1:1 for another 24 h. LLC cells were harvested and stained using an Annexin V/7-AAD apoptosis detection kit (BD Biosciences, USA) according to the manufacturer's instructions. Apoptosis was determined by FACS.

RNA Extraction and Real-Time Quantitative PCR
Total RNA extraction, reverse transcription, and RT-qPCR were performed according to the manufacturer's protocols. Primers were provided in the Supplementary Material (Table S1).

Cytokine Level
The levels of TNFA and IL12 p40/70 in the cell culture supernatants were quantified by ELISA kits (BD Biosciences, USA) according to the manufacturer's instructions.

Detection of Nitric Oxide Level
NO level in the culture supernatants was measured by Griess reagent (Beyotime, China) according to the manufacturer's instruction.

Measurement of Intracellular Reactive Oxygen Species
Cells were incubated with 2′7′-dichlorodihydrofluorescein diacetate (H2DCFDA) (Santa Cruz, USA) dye in DMEM at a final concentration of 5 mM at room temperature for 30 min. Intracellular ROS was measured by FACS.

Analysis of Lactic Acid Level
Lactic acid level in cell supernatant was measured according to the manufacturer's instruction (Dojindo Laboratories, Japan).

Extracellular Acidification Rate Determination
Cells were cultured in a Seahorse XFp cell culture microplate (Agilent, USA) at 5 × 10 4 cells/well. After pretreatment with the corresponding inhibitors or PcrV, the cells were washed with Seahorse XF Base Medium (103193-100, Agilent) supplemented with 2 mM of glutamine (103579, Agilent). The extracellular acidification rate (ECAR) was measured using a Seahorse XFp Glycolysis Stress Test Kit (103020-100, Agilent) with a Seahorse XFp Analyzer (Agilent) according to the manufacturer's instructions.

Migration Assay
LLC cells (3 × 10 4 cells in serum-free DMEM) that had been serum-starved overnight were seeded in the upper chamber of a Transwell insert (8-mm pore, 24-well, Corning). DMEM supplemented with 20% FBS and with or without PcrV (20 mg/ ml) was added to the lower chamber as a chemoattractant. After incubation for 24 or 48 h, the non-migrated LLC cells in the upper chambers were removed. Migrated cells in the bottom chamber were fixed and then stained with 0.5% crystal violet solution for 5 min. The number of migrated cells was counted under an optical microscope (Olympus, Tokyo, Japan).

Cytotoxicity Assay
Tumor cells and TAMs were treated with various concentrations of PcrV for 48 h. Cytotoxicity was assessed using a Cell Counting Kit-8 (CCK-8; Dojindo Laboratories, Japan) according to the manufacturer's protocol.

ATP Assay
Cells were lysed in lysis buffer, and the ATP concentration was tested using an enhanced ATP assay kit (Beyotime) according to the manufacturer's protocol. The luminescence data as determined using a microplate reader (Varioskan Flash, Thermo Fisher) were normalized against protein concentration.

Co-Immunoprecipitation
HEK293T cells were transfected with the pCDNA3.1-TLR4 vector using Lipofectamine 2000 (Invitrogen, USA) and cultured in an incubator at 37°C with 5% CO 2 for 48 h. The obtained cell lysates were incubated with purified PcrV (5 mg) at 4°C for 4 h before the addition of 20 ml of His Mag Sepharose ™ Ni (GE Healthcare) and incubation at 4°C for 1 h. PcrV and TLR4 were detected by Western blotting using anti-His and anti-HA-tag antibodies, respectively.

Surface Plasmon Resonance
For surface plasmon resonance (SPR) analysis, 30 mg/ml of human recombinant TLR4 protein containing a 6×His tag (Abcam, #ab233665) was fixed on the NH2 sensor chip (Nicoya, Canada) using Amine Coupling Kit (Nicoya). Then, PcrV at the indicated concentrations was sequentially injected into the chamber in PBS running buffer. TLR4-PcrV interaction was detected using OpenSPR (Nicoya). The parameters of the binding reactions were calculated and analyzed using Trace Drawer software (Nicoya).

Statistical Analysis
Data were analyzed using unpaired Student's t-tests or one-/two-way ANOVA in GraphPad Prism version 7.0. Tumor volume data were reported as means ± SEM. Other data were expressed as means ± SD.

PcrV Inhibits Tumor Growth by Reprogramming Tumor-Associated Macrophages to a Tumoricidal M1 Phenotype
As PcrV has been reported to activate the host immune response (11,12), we investigated whether PcrV exerts similar effects in the immunosuppressive TME. First, we assessed the efficacy of PcrV in vivo by subcutaneously inoculating LLC cells into the right flanks of C57BL/6 mice followed by intratumoral injection of PcrV ( Figure 1A). Compared with the control group, PcrV treatment decreased tumor growth ( Figure 1B) and weight ( Figure 1C). We previously demonstrated that PcrV significantly increased NO production in normal BMDMs (13). Considering that NO-mediated cytotoxicity is associated with tissue apoptosis and the inhibition of tumor growth, we then evaluated the levels of tumor cell apoptosis and the expression of the NO-generating enzyme-iNOS-in PcrV-treated tumor tissues. The results showed that PcrV treatment increased the levels of apoptosis ( Figure 1D) and iNOS expression ( Figure 1E) in the tumor tissues, indicating that PcrV-induced NO generation is associated with the suppression of tumor growth. In addition, we also treated LLC cells and TAMs with PcrV to assess whether PcrV administration might directly cause cytotoxicity or affect tumor cell growth and metastasis. We found that at the concentrations tested, PcrV did not cause significant cytotoxicity against either tumor cells or TAMs (Supplementary Figure 1A). Moreover, PcrV treatment did not affect the expression levels of genes (e.g., Cox2, Mmp9, Vegfa, and Hif1a; Supplementary Figure 1B) or proteins (e.g., COX2 and vimentin; Supplementary Figure 1C) related to tumor growth and metastasis, or the metastatic ability of LLC cells (Supplementary Figure 1D, E).
To further determine whether the PcrV-mediated M1 polarization of TAMs is responsible for tumor growth inhibition, LLC tumor-bearing mice were peritoneally injected with clodronate liposomes to deplete endogenous macrophages, after which the mice were intratumorally injected with PBS, PcrV, or BMDMs primed or not with PcrV ( Figure 3A). The results showed that clodronate liposome treatment reduced macrophage infiltration in tumor tissues ( Figure 3B). The treatment with PcrV-primed BMDMs decreased tumor growth ( Figure 3C) and weight ( Figure 3D) and increased the levels of apoptosis ( Figure 3E) and iNOS expression ( Figure 3F), and the percentage of iNOS + F4/ 80 + TAMs ( Figure 3F) in tumor tissues; however, PcrV treatment did not affect either tumor growth ( Figure 3C) or weight ( Figure 3D) in mice depleted of endogenous macrophages due to clodronate liposome administration. Collectively, these results demonstrated that the PcrV-mediated tumoricidal effect is associated with the reprogramming of TAMs to an M1 phenotype.

PcrV-Primed Tumor-Associated Macrophages Induce the Apoptosis of Cancer Cells by Enhancing Nitric Oxide-Associated Cytotoxicity In Vitro
Based on the above in vivo and in vitro results, we next investigated the direct cytotoxic effect of PcrV-primed BMDMs on tumor cells by co-culturing these cells with LLC cells. The coculture of BMDMs with tumor cells polarized normal BMDMs into TAMs. Hence, BMDMs are referred to as TAMs after their co-culture with tumor cells. We found that co-culture increased the rate of apoptosis among the tumor cells ( Figures 4A, B). Considering that ROS and NO are critical mediators of cancer cell cytotoxicity, we also examined the levels of these factors in the co-culture system. Unexpectedly, PcrV treatment did not and weight (C) were measured in LLC tumorbearing mice treated with PBS or PcrV. Apoptosis (D) and iNOS protein level (E) in tumor tissues were detected by TUNEL assay and immunofluorescence staining, respectively. Data were expressed as means ± SEM (B, n = 6) or means ± SD (C, n = 6) and were analyzed by two-way ANOVA (B) or unpaired Student's t-test (C). *p < 0.05; **p < 0.01. n.s, no significance; s.c, subcutaneous injection; i.t., intratumoral injection; iNOS, inducible nitric oxide synthase; LLC, Lewis lung cancer; PBS, phosphate-buffered saline. Figure 4). In contrast, NO production was found to be higher in the PcrV/TAM-LLC cell co-culture medium than in that of the TAM-LLC cell group ( Figure 4C). The levels of iNos ( Figure 4D) and iNOS ( Figures 4E, F) were also higher in both TAMs and LLC cells. The inhibition of NO production in PcrV/TAMs through S-methyl thiourea (SMT) treatment decreased apoptosis ( Figures 4A, B) in LLC cells. The addition of the NO donor DETA-NONOate (DETA-NO) into a co-culture system in which TAMs had been pretreated with SMT led to an increase in NO levels in the culture medium ( Figure 4C) while also promoting the apoptosis of LLC cells ( Figures 4G, H) in the PcrV/TAM-LLC cell co-culture group, thus providing further evidence that PcrV-primed TAMs display enhanced NO-associated cytotoxicity against tumor cells.

affect the levels of ROS either in TAMs or LLC cells (Supplementary
Even though the levels of NO were similar among the three groups [LLC cells (Supplementary Figure 5A), TAM-LLC cell co-culture group ( Figure 4C), and PcrV/TAM-LLC cell coculture group ( Figure 4C)] following the addition of DETA-NO to the culture medium, the increased concentrations of NO did not enhance the apoptosis rate of LLC cells cultured individually (Supplementary Figure 5B) or with BMDMs ( Figures 4G, H) compared with that of the PcrV/TAM-LLC cell co-culture system, suggesting that the PcrV-mediated augmentation of TAM-associated cytotoxicity against cancer cells relies on a synergistic effect of NO and other factors generated by PcrV-primed TAMs.
In addition, to observe whether PcrV-primed TAMs can affect tumor growth and metastasis, BMDMs treated or not with PcrV were co-cultured with LLC cells, and the levels of related genes were analyzed. No differences in the expression levels of genes related to tumor growth and metastasis, such as Cox2, Mmp9, Vegfa, and Hif1a, were found in LLC cells that were co-cultured with PcrV-primed TAMs ( Figure 4D). levels of NO ( Figure 5D), IL12 p40/70, and TNFA ( Figure 5E), indicating that PcrV reprograms TAMs toward an M1 profile through increasing glycolysis. In addition, to examine the role of glycolysis in PcrV-induced, TAM-mediated cytotoxicity against cancer cells, BMDMs pretreated with 2-DG and PcrV were cocultured with LLC cells. We found that 2-DG treatment decreased the PcrV-induced production of NO ( Figure 5F) and the rate of apoptosis in LLC cells ( Figures 5G, H), demonstrating that the enhancement of glycolysis by PcrV promotes the NO-associated tumoricidal activity of TAMs.

PcrV-Mediated Activation of a PI3K/AKT/ mTOR-Glycolysis-Nitric Oxide Feedback Loop Promotes Tumor-Associated Macrophage Repolarization and Cytotoxicity Against Cancer Cells
Studies have shown that the PI3K/AKT/mTOR pathway is closely related to macrophage activation (17) as well as glycolysis in pulmonary fibrosis (18). To observe whether the PcrV-mediated regulation of TAM repolarization might involve this signaling pathway, we examined AKT and mTOR phosphorylation levels in TAMs treated or not with PcrV. The results revealed that PcrV treatment increased the phosphorylation levels of both proteins ( Figure 6A) in TAMs. The treatment of PI3K, AKT, or mTOR with the corresponding inhibitor suppressed the levels of M1 markers in PcrV-treated TAMs, including that of iNos (Supplementary Figure 6A), iNOS ( Figure 6B), NO ( Figure 6C), IL12 p40/70, and TNFA (Supplementary Figure 6B) while increasing the levels of the M2 markers Fn1, c-Myc, and Egr2 (Supplementary Figure 6A), suggesting that PcrV repolarizes TAMs into an M1 phenotype through the activation of the PI3K/AKT/mTOR signaling pathway. To further investigate whether the PI3K/AKT/ mTOR signaling pathway is involved in the enhancement of the tumoricidal effect of TAMs elicited by PcrV, TAMs pretreated with the AKT or mTOR inhibitor plus PcrV were co-cultured with LLC cells. The suppression of AKT or mTOR activation reduced NO production in the co-culture medium ( Figure 6D) and also decreased the levels of PcrV/TAM-induced apoptosis in LLC cells ( Figures 6E, F).
We also investigated the cross-talk among PI3K/AKT/mTOR, glycolysis, and NO. The inhibition of PI3K, AKT, or mTOR led to impaired glycolysis-related ECAR and lactic acid production in PcrV-primed TAMs ( Figures 6G, H). Inversely, the suppression of glycolysis by 2-DG led to impaired AKT and mTOR phosphorylation and iNOS expression ( Figure 6I), suggesting that glycolysis positively regulates AKT/mTOR activation and NO generation in PcrV-primed TAMs. Given that NO promotes glycolysis in neurons (19) and also activates the PI3K/AKT axis in cancer cells (20,21), we next examined the role of NO in triggering AKT/mTOR activation and glycolysis in PcrV-primed TAMs. Inhibiting NO generation in PcrV-primed TAMs with SMT treatment reduced AKT and mTOR phosphorylation ( Figure 6J) as well as the ECAR ( Figure 6G) and lactic acid levels ( Figure 6H). Intriguingly, the exogenous administration of DETA-NO in combination with PcrV, but not DETA-NO alone, promoted AKT/mTOR activation in TAMs ( Figure 6K) and upregulated lactic acid production ( Figure 6L), indicating that NO-associated regulation of glycolysis and the AKT/mTOR signaling pathway in TAMs requires the involvement of other factors activated by PcrV. Collectively, these results indicated that the PcrV-mediated increase in the antitumoral effects of TAMs is associated with the activation of a PI3K/AKT/mTOR-glycolysis-NO feedback loop.  Figure 7B), suggesting that PcrV skews the TAM phenotype toward an M1 profile through the TLR4-MyD88 signaling pathway.
As it has been reported that the PI3K/AKT/mTOR pathway is under the control of the TLR4/MyD88 signaling axis (23), we next examined PI3K/AKT/mTOR pathway activation in TLR4 −/− or MyD88 −/− PcrV-primed TAMs. The results showed that the PcrV-induced activation of AKT and mTOR was impaired in TLR4 −/− or MyD88 −/− TAMs, respectively ( Figure 7C). Analyses of glycolysis-related factors demonstrated that the PcrV-induced increase in the ECAR ( Figure 7D) and lactic acid production ( Figure 7E TAMs, but not WT TAMs, reduced NO production in the culture medium ( Figure 7F) as well as the levels of apoptosis in LLC cells ( Figures 7G, H). In addition, the LLC tumor-bearing mice in which endogenous macrophages were depleted by clodronate liposomes were intratumorally injected with WT, TLR4 −/− , or MyD88 −/− BMDMs primed or not with PcrV (Supplementary Figure 8A). The result showed that treatment with PcrV-primed WT BMDMs, but not TLR4 −/− or MyD88 −/− BMDMs, decreased tumor growth (Supplementary Figure 8B) and weight (Supplementary Figure 8C). Collectively, these results indicated that the TLR4/ MyD88 signaling axis participates in the PcrV-mediated modulation of the PI3K/AKT/mTOR-glycolysis-NO circuit and the tumoricidal effect of TAMs.

Direct Interaction Between PcrV and TLR4 Is Required for the PcrV-Mediated Reeducation of Tumor-Associated Macrophages
To further explore whether PcrV alters TAM polarization through interaction with TLR4, we performed immunoprecipitation by incubating purified PcrV protein with the lysate of HEK293T cells overexpressing recombinant human TLR4 protein. The result showed a successful pull-down of TLR4 protein by PcrV ( Figure 7I). SPR analysis using purified protein further revealed a direct interaction between TLR4 and PcrV ( Figure 7J). Moreover, IF staining result showed that PcrV colocalized with TLR4 in Raw264.7 macrophages ( Figure 7K). In addition, to examine the effect of PcrV-TLR4 interaction on PcrV-mediated regulation of TAMs, we blocked TLR4 using an antibody, and we found that the PcrV-induced production of IL12 p40/70, TNFA (Supplementary Figure 7C), and NO (Supplementary Figure 7D) in TAMs was reduced, indicating that PcrVmediated regulation of TAM repolarization involves direct interaction between PcrV and TLR4.

DISCUSSION
TAMs can constitute up to 50% of a tumor mass, forming the major component of the tumor immune cell infiltrate. In the TME, M2-like TAMs promote tumor cell growth, invasion, and metastasis, angiogenesis, and infiltration of immune-suppressive cells, while suppressing antitumoral immune surveillance (24). In addition, TAMs are known to increase resistance to standard-ofcare therapeutics, including chemotherapy, irradiation, and angiogenic inhibitors (25). In contrast, tumoricidal M1-like macrophages, which express high levels of TNF, iNOS, and MHC molecules and low levels of ARG1, IL-10, CD163, and CD206 (26), play roles in reversing immune suppression in the TME and enhancing macrophage-or T cell-mediated killing of cancer cells. In tumor-initiating conditions, macrophages exhibit antitumoral activity; however, once tumors are established, macrophages are reeducated into a protumoral phenotype (27), likely because macrophages are highly plastic in terms of functional reprogramming in response to stimuli in the TME, such as hypoxia, cytokines, and chemokines, as well as in response to varied interactions with components of the extracellular matrix (24,28). Hence, approaches that can potentiate TAM reprogramming into a tumoricidal M1 phenotype show great promise in cancer immunotherapy. In this study, we found that intratumoral injection of PcrV reduces tumor growth and increases the rate of apoptosis in tumor tissues by reeducating TAMs into an M1 profile characterized by the elevated expression of M1 markers (e.g., iNOS, MHCII, and CD86) and reduced levels of M2 markers (e.g., ARG1 and CD206). Additionally, we found that PcrV-induced NO production increased M1 TAM-mediated cytotoxicity against cancer cells in vitro. These results highlight the feasibility of utilizing P. aeruginosa-derived PcrV in immunomodulation and cancer therapy. Similarly, other well-known bacterial molecules, such as MPT63 (M. bovis), arginine deiminase (Mycoplasma arginine), lapidated-azurin (Neisseria meningitidis), and azurin (P. aeruginosa), have been reported as potential anticancer drugs (29). However, factors such as bacterial endotoxin, manufacturing technique, protein stability, administration mode, and biosafety still need to be addressed when utilizing bacteria-derived proteins as antitumoral drugs (29,30).
TLR4, which is expressed on immune cells such as macrophages, dendritic cells (DCs), T cells, neutrophils, and epithelial cells, is one of the major sensors of PAMPs/damageassociated molecular patterns (DAMPs) that activate adaptive immune responses. The activation of TLR4 on multiple immune cells, such as T cells and DCs, represents a powerful means of suppressing tumor growth (31)(32)(33). In addition, several studies have reported that TLR4-dependent TAM reprogramming into an M1 profile reduces tumor growth (3,34). In this study, we found that PcrV directly interacts with TLR4 expressed on macrophages, which induces TAM M1 polarization and enhances TAM-mediated killing of tumor cells via the TLR4/ MyD88 signaling pathway. Surprisingly, the activation of TLR4 expressed on tumor cells has been proposed to also promote tumor development by increasing the production of oncogenic mediators (e.g., COX2, IL6, VEGF, and TGFb) via the activation of proinflammatory and protumoral signaling pathways, such as the NF-kB, MAPK, and COX2/PGE2 pathways (35,36). However, in this study, LLC cells treated with PcrV or PcrVprimed macrophages did not show significantly increased production of oncogenic mediators or metastatic ability, indicating that PcrV exerts its antitumoral effects mainly through the targeting of TAMs. The differences observed between macrophages and cancer cells might be partially attributable to differences in the expression levels of TLR4 or other accessory molecules expressed on these cells.
The PI3K/AKT/mTOR axis, which is regulated by TLR4, functions as a critical signaling pathway in modulating macrophage polarization. The inhibition of the PI3K/AKT/ mTOR or the PI3K/AKT signaling pathway in macrophages suppresses M1 macrophage polarization (17) and polarizes M1 macrophages into an M2-like phenotype (37). In addition, both the PI3K/AKT/mTOR and AKT/mTOR axes are reported to be involved in enhancing glycolysis in tumor cells (38) and TAMs (39), which, in turn, promotes tumor cell survival and proliferation. In line with these findings, our results showed that PcrV-induced activation of the PI3K/AKT/mTOR pathway promotes both M1 polarization and glycolysis in TAMs. However, in contrast to previously reported results, we found that PcrV-primed TAMs exert cytotoxic effects against cancer cells rather than protumoral activity. These discrepant results might be partially explained by the contrasting (cytoprotective/ cytotoxic) effects elicited by NO.
NO, a gas with diverse biological activities produced from arginine by NO synthases, has long been known as a cytotoxic agent that can directly induce the apoptosis of cancer cells (40), as well as enhance radiation-/chemotherapeutic agent-induced apoptosis of cancer cells (41,42). In our study, we found that the co-culture of LLC cells with PcrV-primed TAMs leads to a marked increase in NO levels in the co-culture medium as well as iNOS expression in LLC cells, which, in turn, induces LLC cell apoptosis in a NO-dependent manner. Intriguingly, the supplementation of exogenous NO to LLC cells or TAM-LLC cell co-culture medium did not enhance LLC cell apoptosis, whereas increased levels of LLC cell apoptosis were seen when exogenous NO was applied to co-cultures of LLC cells and PcrVprimed TAMs. These results indicate that NO-associated cytotoxicity relies on the involvement of other mediators supplied by macrophages following PcrV priming. In contrast to the NO-associated cytotoxic effects, NO has been suggested to enhance tumor cell growth and metastasis (43) and help tumor cells resist chemotherapeutic agent-induced apoptosis (44). These contradictory effects might be due to the concentration and duration of exposure to NO encountered by cancer cells (45). NO is cytoprotective at low/physiological levels or with short exposure time but is cytotoxic when produced at high concentrations or under long periods of exposure (46). We and others have reported that PI3K/AKT/mTOR activation and glycolysis rewire TAMs into an M1-polarized phenotype that, in turn, promotes NO production (47,48). Here, we further found that NO, in combination with PcrV, activates the PI3K/ AKT/mTOR-glycolysis signaling pathway in TAMs, resulting in the formation of a PI3K/AKT/mTOR-glycolysis-NO feedback loop that increases NO generation and, consequently, NOassociated cytotoxicity against cancer cells ( Figure 7L). Combined, our findings revealed a tumoricidal role for PcrV mediated by the reeducation of TAMs into a tumoricidal M1 phenotype through the modulation of a PI3K/AKT/mTORglycolysis-NO feedback loop via a direct interaction with TLR4. Our findings provide an alternative therapeutic approach for inhibiting tumor development.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
The animal study was reviewed and approved by Army Medical University of China.  Gene expression levels were analyzed by RT-qPCR. TAMs were pretreated with TLR4 antibody (5 mg/mL)at 37°C for 1 h to block TLR4 expressed on TAMs. Then, the cells were primed with PcrV (10 mg/mL) for 24 h. (C) Detection of IL12 p40/70 and TNFA levels in culture supernatants. (D) NO level in culture supernatant was measured by Griess reagent. Data were expressed as means ± SD and compared by unpaired Student's t-test. *P < 0.05, **P < 0.01 and ***P < 0.001. Ab indicates antibody.