The inflammatory microenvironment of solid tumors creates a protumorigenic milieu that resembles chronic inflammation akin to a subverted wound healing response. Here, we investigated the effect of converting the tumor microenvironment from a chronically inflamed state to one of acute microbial inflammation by injecting microbial bioparticles directly into tumors. Intratumoral microbial bioparticle injection led to rapid and dramatic changes in the tumor immune composition, the most striking of which was a substantial increase in the presence of activated neutrophils. In situ photoconversion and intravital microscopy indicated that tumor neutrophils transiently switched from sessile producers of VEGF to highly motile neutrophils that clustered to make neutrophil-rich domains in the tumor. The neutrophil clusters remodeled tumor tissue and repressed tumor growth. Single-cell transcriptional analysis of microbe-stimulated neutrophils showed a profound shift in gene expression towards heightened activation and antimicrobial effector function. Microbe-activated neutrophils also upregulated chemokines known to regulate neutrophil and CD8+ T-cell recruitment. Microbial therapy also boosted CD8+ T-cell function and enhanced the therapeutic benefit of checkpoint inhibitor therapy in tumor-bearing mice and provided protection in a model of tumor recurrence. These data indicate that one of the major effector mechanisms of microbial therapy is the conversion of tumor neutrophils from a wound healing to an acutely activated cytotoxic phenotype, highlighting a rationale for broader deployment of microbial therapy in the treatment of solid cancers.

Significance:

Intratumoral injection of microbial bioparticles stimulates neutrophil antitumor functions, suggesting pathways for optimizing efficacy of microbial therapies and paving the way for their broader utilization in the clinic.

Most solid tumors generate a sub-acute inflammatory response that favors a tumor microenvironment (TME) that stimulates angiogenesis, tumor growth, and suppresses antitumor immunity (1). Certain microbial preparations can shift the TME towards an acute inflammatory response that inhibits cancer growth (2). Indeed, bladder cancer micrometastases have been successfully eradicated by treatment with Mycobacterium bovis Bacillus Calmette Guerin (BCG; ref. 3). Yet why some cancers are sensitive to acute inflammation despite being promoted by chronic inflammation is not understood (4).

Several microbial and Toll-like receptor (TLR) adjuvants are currently in clinical trials for various solid cancers; however, their success as cancer monotherapies has been limited (1). This suggests that a better understanding of the effects of acute inflammation in cancer is required to advance microbial adjuvants as cancer therapeutics. To date, most of the investigations of microbial therapy have been focused on dendritic cells (DC) and macrophages (1), yet there are many immune cell types known to respond to microbial stimulation. The role of neutrophils has not been explored in detail despite several studies demonstrating neutrophil recruitment in response to BCG (2, 3).

Neutrophils are an essential first line of immune response to microbes but also infiltrate most solid cancers and are frequently associated with a poor prognosis (4). Neutrophil pro-tumor functions include stimulating tumor angiogenesis, aiding metastasis and producing molecules that suppress antitumor immune responses and establish a pro-tumor immunosuppressive microenvironment (5).

Neutrophils can also have potent antitumor functions including direct cytotoxicity against tumor cells and attenuation of metastasis (6, 7). Increased neutrophil infiltration in a model of colon adenocarcinoma had potent antitumor effects mediated in part through cross-talk with CD8 T cells (8, 9). The opposing roles for neutrophils in cancer led to the development of the N1/N2 paradigm, where N1 neutrophils have antitumor properties, whereas N2 neutrophils are tumor promoting. N2 neutrophils can be converted into N1 antitumor effectors by modulating the TME (10), indicating that these cells rely on external cues in the TME to determine their activation and phenotypic state. However, the mechanisms that govern the switch between pro- and antitumor neutrophils and how the antitumor state can be maintained in the TME are all unknown but could reveal how to manipulate neutrophils to achieve an anticancer benefit.

Here, we investigate the impact of switching tumor inflammation from chronic to acute using microbial bioparticles with the outcome of converting intratumoral neutrophils from a wound healing to a tumor-killing phenotype. Our findings identify neutrophil- and T-cell–based mechanisms of microbial therapy and highlight approaches to successfully target inflammation in cancer therapy.

Mice

All mice used in this study were maintained on C57BL/6 (RRID:MGI:5656552) background and housed in specific pathogen-free conditions. All animal experiments and procedures were approved by the Garvan Institute of Medical Research/St Vincent's Hospital Animal Ethics Committee. Male and female mice were randomly assigned to treatment groups once tumors were established. C57BL/6 mice (RRID:MGI:5656552) were obtained from Australian BioResources (Moss Vale, NSW). Kaede mice (RRID:IMSR_RBRC05737; ref. 11) were a gift from Professor Michio Tomura and were maintained on C57BL/6 background. Ly6GCre-tdTomato (C57BL/6-Ly6 g (tm2621(Cre-tdTomato)Arte) neutrophil-specific reporter mice (12) were a gift from Professor Matthias Gunzer and crossed with B6.LSL td-Tomato (B6.Cg-Gt(ROSA)26Sortm14(CAG-tdTomato)Hze/J (IMSR, catalog no. JAX:007914 RRID:IMSR_JAX:007914) to generate BigRed/CatchupIVM-red mice and crossed to Albino.B6 or C57BL/6 mice with spontaneous mutations in the tyrosinase gene (B6(Cg)-Tyrc-2J/J) for imaging. Lysozyme M fluorescent reporter mice were generated by crossing lysozyme M Cre mice (Jackson Laboratory, catalog no. 004781 RRID:IMSR_JAX:004781) to B6-ROSA/kikGR(floxed) mice (Riken, # RBRC09254, RRID: IMSR_RBRC09254) or B6.Cg-Gt(ROSA)26Sortm14(CAG-tdTomato)Hze/J mice (Jackson Laboratory, catalog no. 007914 RRID: IMSR_JAX:007914).

Tumor cell lines

Mouse Lewis lung carcinoma (LLC; ATCC, catalog no. CRL-1642, RRID:CVCL_4358) cell line was purchased from ATCC (RRID:SCR_001672). LLC-eGFP cell line was a gift from Professor Robert Brink. B16F10–3C melanoma cell line (13) was a gift from Professor Wolfgang Weninger. Murine AT-3 mammary carcinoma (RRID:CVCL VR89) cell line was a gift from Dr. Scott Abrams. KPC primary pancreatic ductal adenocarcinoma (PDAC) cell line (RRID:CVCL_XD11; ref. 14) was a gift from Professor Paul Timpson.

Neutrophil recruitment into tumors

LLC cells were inoculated into ear pinnae of Kaede mice. When tumors reached 4 to 8 mm3, they were photoconverted for 20 minutes with a violet light from a cold-light source fitted with a filter (Zeiss) to minimize thermal and phototoxicity (15, 16) and immediately injected with 20×106Staphylococcus aureus (S. aureus) bioparticles. Twenty-four, 48, and 72 hours later cell suspensions were analyzed by flow cytometry.

Microbial control of tumor growth

A total of 1×105 to 2×105 LLC, B16F10–3C, AT-3 or KPC tumor cells (in 5 μL volume) were inoculated into ear pinnae of C57BL/6 mice. Once tumors were detected, 4×106 to 20×106S. aureus bioparticles or 5×106 CFU BCG or relevant vehicle control was administered intratumorally every 2 days unless otherwise specified. Tumor dimensions were measured using calipers and volume calculated with the modified ellipsoidal formula V = ½ (Length × Width2).

Neutrophil depletion

LLC tumors were grown in the ear pinnae of C57BL/6 mice. Once tumors became visible, neutrophils were depleted by intraperitoneal injection of 500 μg anti-Ly6G clone 1A8 (Bio X Cell, catalog no. BE0075–1, RRID:AB_1107721) or rat IgG2a isotype control clone 2A3 (Bio X Cell, catalog no. BE0089, RRID:AB_1107769). Twenty-four hours later, 4×106 to 10×106S. aureus bioparticles were injected into LLC tumors. The mice then had alternating days of maintenance dose of 250 μg anti-Ly6G/isotype intraperitoneally and S. aureus bioparticles injected into tumors every 2 days. A total of 4 doses of 250 μg anti-Ly6G or isotype was administered whilst S. aureus bioparticles were given until mice reached ethical endpoints.

Neutrophil depletion in mice treated with anti-Ly6G was confirmed by flow cytometry. Red cells in blood samples were lysed with 10 mmol/L KHCO3, 0.1 mmol/L EDTA, and 166 mmol/L NH4Cl solution and then blocked with 5% normal mouse serum and stained with unlabeled rat anti–mouse-Ly6G clone 1A8 (Bio X Cell, catalog no. BE0075–1, RRID:AB_1107721) primary antibody, washed and then stained with a secondary goat anti-rat IgG DyLight 649 (BioLegend, catalog no. 405411, RRID:AB_1575141) antibody and washed with FACS buffer and blocked with 5% normal rat serum. Samples were washed again and stained with labeled cell surface antibodies including CD11b-APCef780 clone M1/70 (Thermo Fisher Scientific, catalog no. 47–0112–80, RRID:AB_1603195).

CD8 T-cell depletion

A total of 1×105 LLC tumor cells were inoculated into ear pinnae of C57BL/6 mice. Once tumors were visible, CD8 T cells were depleted by intraperitoneal injection of 250 μg anti-mouse CD8 clone 53–6.7 (Bio X Cell, catalog no. BE0004–1 RRID:AB_1107671) or isotype clone IgG2a (Bio X Cell, catalog no. BE0089 RRID:AB_1107769). Twenty-four hours later, 20 × 106S. aureus bioparticles or vehicle control, were injected into tumors. Subsequent anti-CD8 antibodies were administered every 3 days and S. aureus bioparticles (or vehicle control) every 2 days. A total of five anti-CD8 or isotype injections were administered and S. aureus bioparticles (or vehicle) were given until mice reached ethical endpoints.

Effect of C5a receptor inhibition on microbial therapy

AT-3 tumor cells (1×105 per mouse) were inoculated into ear pinnae of C57BL/6 mice. Once tumors were visible, 200 μg/mouse of C5a receptor (C5aR) antagonist, PMX205 (17) (a gift from Professor Trent M. Woodford) or vehicle control was injected intraperitoneally daily. On alternative days 20×106 of S. aureus bioparticles or vehicle control were injected into tumors and tumor growth was measured.

Effect of ROS inhibition on microbial therapy

A total of 1×105 AT-3 tumor cells were inoculated into ear pinnae of C57BL/6 mice. Tumor bearing mice received 500 μg/mouse of N-acetylcysteine (NAC; Sigma-Aldrich, catalog no. A9165 CAS 616–91–1) or vehicle control daily intraperitoneally from the time of tumor inoculation. Tumors were treated with 20×106S. aureus bioparticles or vehicle control every second day once tumors were visible and tumor growth was measured.

Microbial and checkpoint inhibitor combination therapy

A total of 1×105 AT-3 tumor cells were inoculated into ear pinnae of C57BL/6 mice. Once tumors were detected, mice were treated with 2.5 mg anti-mouse programmed cell death protein 1 (PD-1; clone: RMP1–14; Bio X Cell, catalog no. BE0146, RRID: AB10949053) and 1 mg anti-mouse CTLA4 (Bio X Cell, catalog no. BE0131, RRID: AB10950184) or isotype control antibodies (clones: 2A3; Bio X Cell, catalog no. BE0089, RRID:AB_1107769, polyclonal Syrian hamster IgG; Bio X Cell, catalog no. BE0087, RRID: AB1107782) via intraperitoneal injection every 3 days for a total of 5 injections. The day after the first antibody injection, some mice were treated with 20×106S. aureus bioparticles administered intratumorally every 3 days (a total of 3 injections) or vehicle control.

Tumor recurrence model

AT-3 inoculated C57BL/6 mice treated with microbial therapy were followed for >60 days after tumor resolution to confirm complete recovery. These mice and naïve C57BL/6 mice were injected with 1×105 AT-3 tumor cells into ear pinnae. Tumor growth was measured as previously.

Adoptive transfer of bone marrow cells

For the single adoptive transfer, bone marrow was harvested by flushing the femurs of BigRed/CatchupIVM-red mice using cold PBS. Erythrocytes were removed by lysing with 10 mmol/L KHCO3, 0.1 mmol/L EDTA, and 166 mmol/L NH4Cl and washing twice with PBS. The resulting cell suspension (6 × 106 cells) was transferred intravenously into recipient C57BL/6 mice bearing LLC tumors.

For the double adoptive transfer, bone marrow from the first donor mouse strain (lysozyme M tdTomato) was harvested as described above and 5 × 106 cells transferred intravenously into recipient C57BL/6 mice bearing LLC tumors. One day later, bone marrow from the second donor mouse strain (lysozyme M Kikume) was harvested as described above and transferred intravenously into the same recipient. Tumors were imaged 2 hours later using intravital two-photon microscopy, and again at 24 hours after the second adoptive transfer.

Statistical analysis

The statistical distribution of experimental data was determined using a D'Agostino-Pearson omnibus normality test. The statistical significance of experimental data for comparisons of two groups was determined using either an unpaired t test when distribution was normal and unpaired Mann–Whitney test if distribution was not normal. Statistical significance, for comparison of three or more groups was calculated using ordinary one-way ANOVA test when distribution was normal and unmatched Kruskal–Wallis test if distribution was not normal. All analysis was done on GraphPad Prism (GraphPad Prism, RRID:SCR_002798).

Data availability

The single-cell RNA sequencing data generated in this study are available in ArrayExpress: accession number E-MTAB-12642. Other data generated in this study are available within the article and its Supplementary Data files or upon request from the corresponding author. Additional Materials and Methods are available as Supplementary Data.

Acute microbial inflammation reshapes the tumor immune landscape

To investigate how converting the TME from chronic to acute antimicrobial inflammation would alter the tumor immune landscape, we used the LLC model established in the ear pinnae of C57Bl/6 mice, which we developed to track tumor immune cells by intravital microscopy and in situ photoconversion (16). Unmanipulated tumors 10 to 14 days after inoculation showed evident leukocyte infiltration (Fig. 1A; Supplementary Fig. S1A–S1E), whereas injection of killed S. aureus bioparticles directly into tumors markedly increased the size of the infiltrate (Fig. 1A; Supplementary Fig. S1A and S1B). We observed increased recruitment of monocytes and DCs, but the proportions of migratory DCs, cDC1, and cDC2 subsets, the number of macrophages and the proportions of F4/80+ and MHC class IIhigh macrophage subsets were unchanged after treatment (Supplementary Figs. S2A and S2B, S3A–S3F). Among lymphocytes, the number of B cells was unchanged, while the number of CD3+ T cells increased in treated tumors (Supplementary Fig. S3G and S3H). This increase was due to an increase in CD4CD8 rather than conventional CD8+ or CD4+ T cells (Supplementary Fig. S3I). Unlike CD4CD8 T cells in autoimmune conditions (18), very few tumor CD4CD8 T cells were B220+ before and after microbial treatment (Supplementary Fig. S3J).

Figure 1.

Microbe-driven recruitment and activation of intratumoral neutrophils. A–C,S. aureus (S.a) bioparticle injection into tumors leads to an influx of leukocytes (A) and especially neutrophils (B), which made up the majority of tumor leukocytes after challenge (C). D–F, Neutrophil activation was assessed by expression of CD11b (D), CD62L (E), and CXCR2 (F) detected by flow cytometry 24 hours post S.a injection. G, Phagocytosis of labeled S.a bioparticles by intratumoral neutrophils. Representative flow cytometry plots (left) and pooled data (right) are shown. H–K, Proportion of VEGF (H), MMP9 (I), MPO (J), and iNOS (K) expressing neutrophils was quantitated by flow cytometry in control and S.a-treated tumors. Mean + SEM from 3 to 5 (AI and K) and 2 (J) independent experiments. Each circle represents a tumor. Data analyzed using nonparametric Mann–Whitney test (A, B, E, F, J, and K) and t test (C, D, H, and I). ***, P ≤ 0.001; ****, P ≤ 0.0001.

Figure 1.

Microbe-driven recruitment and activation of intratumoral neutrophils. A–C,S. aureus (S.a) bioparticle injection into tumors leads to an influx of leukocytes (A) and especially neutrophils (B), which made up the majority of tumor leukocytes after challenge (C). D–F, Neutrophil activation was assessed by expression of CD11b (D), CD62L (E), and CXCR2 (F) detected by flow cytometry 24 hours post S.a injection. G, Phagocytosis of labeled S.a bioparticles by intratumoral neutrophils. Representative flow cytometry plots (left) and pooled data (right) are shown. H–K, Proportion of VEGF (H), MMP9 (I), MPO (J), and iNOS (K) expressing neutrophils was quantitated by flow cytometry in control and S.a-treated tumors. Mean + SEM from 3 to 5 (AI and K) and 2 (J) independent experiments. Each circle represents a tumor. Data analyzed using nonparametric Mann–Whitney test (A, B, E, F, J, and K) and t test (C, D, H, and I). ***, P ≤ 0.001; ****, P ≤ 0.0001.

Close modal

In contrast, after microbial treatment tumor neutrophils (identified as Ly6G+CD11b+ cells) increased by ∼28-fold to comprise the majority (>85%) of all immune cells in the treated tumors (Fig. 1B and C; Supplementary Figs. S1C, S3K). Although neutrophil marker Ly6G can be expressed on eosinophils (19) and eosinophil blood counts in patients have been linked to disease recurrence during BCG treatment (20), eosinophil number in tumors or peripheral blood did not change after microbial treatment (Supplementary Fig. S3L and S3M) and only a small subset of tumor neutrophils expressed the eosinophil markers SiglecF and F4/80 following microbial stimulation (Supplementary Fig. S3N). Thus, microbial bioparticle treatment dramatically reshapes the tumor immune landscape promoting recruitment of immune cells, and especially neutrophils, into tumors.

Microbial therapy induces a switch in neutrophil phenotype and function

To investigate how S. aureus bioparticle treatment affects tumor neutrophil state, we examined their phenotype. The expression of GR-1 on total tumor Ly6G cells was significantly higher following S. aureus treatment, indicating that recruited neutrophils have a mature rather than myeloid-derived suppressor cell phenotype (Supplementary Fig. S3O). Furthermore, the majority of tumor neutrophils prior to injection of microbial therapy, at 8 hours (when many neutrophils have recently entered the tumor) and 24 hours after S. aureus treatment had a mature Ly6GhighCXCR4lo (21) phenotype with only a small number of neutrophils of intermediate maturity present in unmanipulated tumors (Supplementary Fig. S3P).

The expression of markers CD11b, CD62L, and CXCR2 is regulated upon neutrophil activation (22–24). Consistent with the neutrophil phenotype in response to a bacterial stimulus (15), microbial treatment led to an increase in the activation marker CD11b and concomitant downregulation of CD62L and CXCR2 (Fig. 1DF).

Analysis of neutrophil function following microbial treatment showed that neutrophils were the main immune subset to phagocytose microbial particles (assessed by uptake of labeled S. aureus bioparticles; Fig. 1G). Neutrophils in unmanipulated tumors expressed higher levels of VEGF, which is involved in angiogenesis and wound healing and is considered a pro-tumor molecule due to its ability to promote tumor growth and metastasis (Fig. 1H; ref. 25). While matrix metalloproteinase (MMP)9 and myeloperoxidase (MPO; Fig. 1I and J) were not significantly altered following treatment, we observed an increase in the levels of inducible nitric oxide synthase (iNOS; Fig. 1K), an antimicrobial agent produced in response to phagocytosis that can also stimulate apoptosis and kill nearby cells by releasing reactive oxygen species (ROS; refs. 26, 27). These data show that microbial treatment polarizes the phenotype of tumor neutrophils to one characteristic of an antimicrobial response.

Microbial bioparticle treatment alters intratumoral neutrophil dynamics

Next, we leveraged intravital two-photon microscopy to visualize neutrophil dynamics within intact tumors established in neutrophil-specific reporter (BigRed/CatchupIVM-red) mice where tdTomato fluorescent protein is expressed in Ly6G+ neutrophils (12). Prior to treatment, neutrophils, which were scattered throughout the tumor mass (Fig. 2A), displayed limited motility (median average speed 0.04 μm/second) and displacement (median displacement 3 μm; Figs. 2B and C; Supplementary Video 1), unlike neutrophils responding to acute microbial inflammation, which migrate rapidly to coalesce in dynamic swarms (28). However, neutrophil dynamics and distribution rapidly changed following microbial treatment. As early as 4 hours after treatment, we detected large clusters of microbe-activated (MA) neutrophils throughout the tumor mass (Fig. 2A) and a 2.5-fold increase in neutrophil speed and a 4-fold increase in displacement (Figs. 2B and C; Supplementary Video 1).

Figure 2.

Tumor neutrophil dynamics in vivo. A, Neutrophils (red) were visualized in steady state or in S. aureus (S.a) bioparticle–treated LLC tumors using intravital two-photon microscopy. Yellow tracks indicate neutrophils with confined motility (track displacement length < 13 μm); green tracks indicate migrating neutrophils (track displacement length > 47 μm). Blue, second harmonic generation (SHG)/collagen. Bar, 20 μm. B, Mean track speed of intratumoral neutrophils. C, Track displacement length of intratumoral neutrophils. Data from at least four independent imaging experiments per time point were analyzed using a one-way ANOVA with Dunn correction for multiple comparisons (B and C). Median and quartiles are shown. **, P ≤ 0.01; ****, P ≤ 0.0001; n.s., not significant.

Figure 2.

Tumor neutrophil dynamics in vivo. A, Neutrophils (red) were visualized in steady state or in S. aureus (S.a) bioparticle–treated LLC tumors using intravital two-photon microscopy. Yellow tracks indicate neutrophils with confined motility (track displacement length < 13 μm); green tracks indicate migrating neutrophils (track displacement length > 47 μm). Blue, second harmonic generation (SHG)/collagen. Bar, 20 μm. B, Mean track speed of intratumoral neutrophils. C, Track displacement length of intratumoral neutrophils. Data from at least four independent imaging experiments per time point were analyzed using a one-way ANOVA with Dunn correction for multiple comparisons (B and C). Median and quartiles are shown. **, P ≤ 0.01; ****, P ≤ 0.0001; n.s., not significant.

Close modal

Adoptive transfer of fluorescent bone marrow cells into tumor bearing mice led to the recruitment of labeled neutrophils into tumors and this recruitment substantially increased after microbial injection (Supplementary Fig. S4A), indicating that circulating neutrophils exported from the bone marrow are recruited into tumors. Neutrophils transferred into tumor-bearing mice 24 hours apart had similar motility in unmanipulated tumors (Supplementary Fig. S4B), demonstrating that the changes in neutrophil motility are driven by the microbial stimulus rather than the time spent in the TME.

Microbe-induced increase in neutrophil motility was transient and by 24 hours after microbial treatment neutrophil speed started to decline. Detailed analysis of MA neutrophil dynamics 24 hours posttreatment revealed two distinct modes of behavior, where a proportion of neutrophils remained motile, while most neutrophils formed large stable clusters of low motility (Fig. 2A; Supplementary Video 1). Our results show that neutrophils undergo a striking change in their behavior in response to microbial therapy by increasing their motility and forming large clusters within the tumor mass, indicating a change in how neutrophils interact with tumor cells.

MA neutrophils remodel tumor matrix

To test whether MA neutrophils carry out tumor remodeling in vivo similarly to tissue remodeling mediated by neutrophil swarms in response to infectious and sterile inflammation (28, 29), we investigated neutrophil interactions with tumor cells in intact LLC-eGFP tumors established in the ear pinnae of neutrophil reporter mice. Twenty-four hours following microbial treatment large neutrophil clusters could be observed within the tumor mass (Fig. 3A). These clusters corresponded to areas cleared of tumor cells and collagen, indicating that MA neutrophils in treated tumors remodel tumors.

Figure 3.

Visualizing neutrophil interactions with tumor cells. A, LLC-GFP tumor cells (green), neutrophils (red), and second harmonic generation (SHG)/collagen (blue) were visualized in frozen sections from bioparticle-treated and control tumors using two-photon microscopy. White arrows indicate areas of tumor cell clearance by neutrophils. Scale bar, 500 μm. B, Neutrophils (red) and LLC-GFP cells (green) were visualized in tumors prior or 24 hours following S. aureus (S. a), treatment using intravital microscopy. Region I, an area of LLC loss; region II, partial LLC destruction; region III, largely intact but rounded LLCs. Quantitation of LLC-GFP cell loss (right panel). C, Killing of LLC-GFP cells was assessed using SYTOX labeling to detect dead cells 24 hours following incubation with S.a-stimulated or control neutrophils in vitro. D, Purple tracks show neutrophils (red) interacting and crawling over LLC cells (green). Bar, 20 μm. E, Quantitation of neutrophil motility in areas of intact tumor cells and in areas of extensive tumor remodeling and cell debris. F, Two examples of neutrophil-LLC interactions and tumor cell blebbing (white arrows). Bar, 20 μm. G, LLC-GFP tumors (green), neutrophils (red), and SYTOX (blue) were visualized in frozen sections from bioparticle-treated and control tumors using two-photon microscopy. White arrows, SYTOX staining of LLC cells. Scale bar, 50 μm. Data analyzed using Mann–Whitney test (B and E) and one-way ANOVA (C). **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001.

Figure 3.

Visualizing neutrophil interactions with tumor cells. A, LLC-GFP tumor cells (green), neutrophils (red), and second harmonic generation (SHG)/collagen (blue) were visualized in frozen sections from bioparticle-treated and control tumors using two-photon microscopy. White arrows indicate areas of tumor cell clearance by neutrophils. Scale bar, 500 μm. B, Neutrophils (red) and LLC-GFP cells (green) were visualized in tumors prior or 24 hours following S. aureus (S. a), treatment using intravital microscopy. Region I, an area of LLC loss; region II, partial LLC destruction; region III, largely intact but rounded LLCs. Quantitation of LLC-GFP cell loss (right panel). C, Killing of LLC-GFP cells was assessed using SYTOX labeling to detect dead cells 24 hours following incubation with S.a-stimulated or control neutrophils in vitro. D, Purple tracks show neutrophils (red) interacting and crawling over LLC cells (green). Bar, 20 μm. E, Quantitation of neutrophil motility in areas of intact tumor cells and in areas of extensive tumor remodeling and cell debris. F, Two examples of neutrophil-LLC interactions and tumor cell blebbing (white arrows). Bar, 20 μm. G, LLC-GFP tumors (green), neutrophils (red), and SYTOX (blue) were visualized in frozen sections from bioparticle-treated and control tumors using two-photon microscopy. White arrows, SYTOX staining of LLC cells. Scale bar, 50 μm. Data analyzed using Mann–Whitney test (B and E) and one-way ANOVA (C). **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001.

Close modal

We applied intravital imaging to examine the interactions between neutrophils and GFP-tagged tumor cells in real time. In contrast to elongated intact LLC cells in unmanipulated tumors, we observed an increase in rounded tumor cells and tumor cell debris in areas of neutrophil infiltration in microbe-treated tumors, (Fig. 3B; Supplementary Video 2). Quantitation of eGFP signal showed loss of LLC tumor cells in treated tumors (Fig. 3B). We tested whether microbial stimulation enhances neutrophil capacity to kill tumor cells by incubating LLC-GFP cells with purified neutrophils and S. aureus bioparticles or unstimulated neutrophils in vitro and found that that LLC cells incubated with MA neutrophils for 24 hours showed significantly more cell death (Fig. 3C).

We also observed that neutrophils engaged in multiple interactions with tumor cells and neutrophil motility was significantly lower in areas of tumor remodeling where a large number of rounded tumor cells was observed (Fig. 3D and E; Supplementary Video 3). Our analysis revealed tumor cells within neutrophil clusters undergoing blebbing, a characteristic feature of cell death (Fig. 3F; Supplementary Video 4). We used labeling with SYTOX to confirm tumor cell death within neutrophil clusters (Fig. 3G). Taken together our results indicate that MA neutrophils remodel tumor matrix and may contribute to tumor cell death.

Neutrophil extracellular traps are induced in response to microbial therapy

Neutrophil extracellular traps (NET) are one of the mechanisms of neutrophil pathogen defense but have also been observed in sterile inflammation (22). In cancer, NETs play a role in tumor metastasis and tissue remodeling (5). We stained tumor sections for citrullinated histone H3 and neutrophil elastase to detect NETs and observed a significant increase in both markers within tumors after microbial treatment (Supplementary Fig. S5A). To test whether inhibiting NETosis in the TME affects neutrophil recruitment and function, we administered GSK484, a potent inhibitor of protein arginine deiminase 4 (PAD4), which stops the formation of NETs in both mouse and human neutrophils (30). GSK484 administration did not alter neutrophil recruitment to tumors in response to S. aureus (Supplementary Fig. S5B), nor their functions including iNOS, MMP9, and MPO (Supplementary Fig. S5C–S5E). Finally, we employed two-photon microscopy (as in Fig. 3A and B) to assess the extent of tissue remodeling and loss of LLC-GFP cells when S. aureus was coadministered with GSK484. We found no significant changes in tumor remodeling after NETosis was inhibited in microbe-treated tumors (Supplementary Fig. S5F), indicating that factors other than NETosis contribute to neutrophil tumor remodeling in response to microbial therapy.

Microbe- and tumor-derived signals shape neutrophil phenotype and turnover

To investigate whether the stimulatory effect of microbial bioparticles persists over a prolonged period or is suppressed by the TME, we examined neutrophil number and activation over time. Following rapid recruitment in response to microbial treatment, intratumoral neutrophil number remained stable over the course of 72 hours (Fig. 4A). In contrast, neutrophil activation rapidly decreased as evidenced by a decline in the expression of CD11b (Fig. 4B), suggesting that repeated stimulation is required to maintain the acutely activated cytotoxic phenotype.

Figure 4.

Modulation of tumor neutrophil phenotype and persistence by signals in the TME. A and B, Neutrophil number (A) and CD11b expression (B) over time after an injection of S. aureus (S. a) bioparticles. C–F, Tumors were photoconverted and treated with S.a bioparticles and CD11b (C), CD62L (D), CXCR2 (E), and CXCR4 (F) expression was analyzed 24 hours later on circulating (blue), recently recruited (non-photoconverted, green) and tumor-experienced (photoconverted, magenta) neutrophils using flow cytometry. G, Tumors were photoconverted and treated with S.a bioparticles immediately after photoconversion. Photoconverted neutrophils in tumors were quantified 24, 48, and 72 hours later. H, Flow cytometric analysis of photoconverted neutrophils in tumor draining lymph nodes 24 hours after photoconversion. Each circle represents a tumor, blood, or lymph node sample. Mean +SEM are shown (C-E and H) or the mean + SEM of at least 7 tumor samples (G). Mean + SEM from data pooled from at least two independent experiments and analyzed using Kruskal–Wallis test (A, B, and DF), one-way ANOVA (C), multiple t test (G), or Mann–Whitney test (H). *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001; n.s., not significant. In G, blue * indicates 24 versus 72 hours comparison; black * indicates S.a versus control comparison.

Figure 4.

Modulation of tumor neutrophil phenotype and persistence by signals in the TME. A and B, Neutrophil number (A) and CD11b expression (B) over time after an injection of S. aureus (S. a) bioparticles. C–F, Tumors were photoconverted and treated with S.a bioparticles and CD11b (C), CD62L (D), CXCR2 (E), and CXCR4 (F) expression was analyzed 24 hours later on circulating (blue), recently recruited (non-photoconverted, green) and tumor-experienced (photoconverted, magenta) neutrophils using flow cytometry. G, Tumors were photoconverted and treated with S.a bioparticles immediately after photoconversion. Photoconverted neutrophils in tumors were quantified 24, 48, and 72 hours later. H, Flow cytometric analysis of photoconverted neutrophils in tumor draining lymph nodes 24 hours after photoconversion. Each circle represents a tumor, blood, or lymph node sample. Mean +SEM are shown (C-E and H) or the mean + SEM of at least 7 tumor samples (G). Mean + SEM from data pooled from at least two independent experiments and analyzed using Kruskal–Wallis test (A, B, and DF), one-way ANOVA (C), multiple t test (G), or Mann–Whitney test (H). *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001; n.s., not significant. In G, blue * indicates 24 versus 72 hours comparison; black * indicates S.a versus control comparison.

Close modal

Next, we took advantage of a photoconversion-based approach that we have developed to label tumor-infiltrating immune cells (16) to compare the phenotypes of tumor-experienced neutrophils to neutrophils in circulation and neutrophils recently recruited to tumors. As neutrophils left circulation and infiltrated tumors, they became activated (as indicated by increased expression of CD11b and downregulation in CD62L and CXCR2; Fig. 4CE). This activation was more pronounced in photoconverted tumor-experienced neutrophils compared with non-photoconverted recently recruited neutrophils, indicating that circulating neutrophils undergo progressive activation upon entering tumors. We also observed that neutrophil activation was further enhanced by S. aureus bioparticles (Fig. 4CE). However, tumor-experienced neutrophils appeared less susceptible to stimulation because S. aureus treatment did not alter CD11b and CD62L levels on photoconverted neutrophils, suggesting that aged neutrophils lose some of their plasticity and ability to respond to signals in their environment.

As chemokine receptor CXCR4 expression is associated with neutrophil migration and aging (31, 32), we found that CXCR4 expression was higher in photoconverted neutrophils that are likely to be more aged than non-photoconverted neutrophils that have recently entered the tumor from circulation (Fig. 4F). CXCR4 expression was unchanged in neutrophils from microbial particle–treated tumors, suggesting that tumor neutrophils rapidly acquire an aging phenotype regardless of microbial signals.

Neutrophils are short-lived cells with a half-life of just several hours in circulation (33). Their lifespan can be extended to several days when neutrophils enter tissues (22), but how long they survive in the TME is not yet known. We used photoconversion to label tumor neutrophils and analyze the number of photoconverted neutrophils remaining in tumors at various time points. The number of photoconverted neutrophils in tumors rapidly declined over time—neutrophil number decreased by 63% between 24 and 48 hours and by 92% between 48 and 72 hours (Fig. 4G), indicating that most tumor-infiltrating neutrophils do not survive for extended periods of time. However, the decline in photoconverted neutrophils was less rapid following S. aureus injection—neutrophil number decreased by 37% between 24 and 48 hours and by 74% between 48 and 72 hours (Fig. 4G), suggesting that microbial activation prolongs tumor neutrophil survival.

We applied photoconversion as previously (16) to determine whether neutrophils can emigrate from tumors to draining lymph nodes as this could contribute to tumor neutrophil turnover. By photoconverting tumor-infiltrating cells prior to microbial bioparticle treatment and then analyzing draining lymph nodes for the presence of photoconverted tumor-egressing neutrophils, we found that neutrophils emigrated poorly from unmanipulated tumors (Fig. 4H). However, microbial treatment substantially increased egress of photoconverted neutrophils to draining lymph nodes (Fig. 4H). Our data indicate that microbial treatment modulates tumor neutrophil turnover and egress from tumors but the stimulatory effect of microbial therapy is transient and neutrophil phenotype rapidly evolves in response to the predominant signals in the TME.

Microbial treatment inhibits tumor growth

Microbial treatment led to neutrophil influx into tumors and a switch in neutrophil function (Fig. 1), suggesting that MA neutrophils may have a potent antitumor effect. However, analysis of neutrophil activation over time showed a rapid peak and loss of activation (Fig. 4), indicating that repeated stimulation is required to maintain neutrophil antitumor phenotype. Consistent with this, we observed that a single injection of microbial bioparticles was not sufficient to repress tumor growth (6 of 7 mice treated once reached ethical endpoints for tumor size within 3 weeks). Therefore, we assessed whether repeated microbial treatment could repress tumor growth. In the unmanipulated (or vehicle treated) state, LLC tumors showed an exponential growth pattern up to 2 to 3 weeks post inoculation, at which point, animals reach ethical endpoints and are euthanized (Fig. 5A). However, we observed a striking suppression of tumor growth after treatment with S. aureus microbial bioparticles on alternate days (Fig. 5A). This shows that treatment with microbial bioparticles could override the normally pro-tumorigenic TME to promote a potent antitumor response.

Figure 5.

Microbial treatment leads to neutrophil-dependent inhibition of tumor growth. A, Tumor volume was measured in S. aureus (S.a), saline (vehicle)-treated or untreated LLC tumors grown in C57BL/6 mice, and treated every second day. B, Tumor volume in BCG-treated or vehicle control LLC tumors in C57BL/6 mice. C, LLC tumor volume in C57BL/6 mice following anti-Ly6G neutrophil depletion on days −1, 1, 3, 5, 7. S.a was injected on days 0, 2, 4, 6, 8. D, LLC tumor volume following CD8+ T-cell depletion. Anti-CD8 was administered intraperitoneally every 3 days and S.a every 2 days. E, AT-3 tumor volume in C57BL/6 mice treated with NAC to inhibit ROS. NAC was administered daily from the time of tumor inoculation and S.a was injected every second day once tumors were detectable. F, AT-3 tumor volume in C57BL/6 mice treated with C5aR antagonist, PMX205, intraperitoneally daily and S.a every second day. Data shown as a mean + SEM from 5 (A), 2 (B and C), 1 (DF) experiments. Data analyzed using Mann–Whitney (AC, and F) or t test (D and E). No significant difference between S.a + Vehicle and S.a.+ NAC (E) or S.a + Vehicle and S.a.+ PMX205 (F); differences between vehicle and NAC were significant as indicated on the graph (E). *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001.

Figure 5.

Microbial treatment leads to neutrophil-dependent inhibition of tumor growth. A, Tumor volume was measured in S. aureus (S.a), saline (vehicle)-treated or untreated LLC tumors grown in C57BL/6 mice, and treated every second day. B, Tumor volume in BCG-treated or vehicle control LLC tumors in C57BL/6 mice. C, LLC tumor volume in C57BL/6 mice following anti-Ly6G neutrophil depletion on days −1, 1, 3, 5, 7. S.a was injected on days 0, 2, 4, 6, 8. D, LLC tumor volume following CD8+ T-cell depletion. Anti-CD8 was administered intraperitoneally every 3 days and S.a every 2 days. E, AT-3 tumor volume in C57BL/6 mice treated with NAC to inhibit ROS. NAC was administered daily from the time of tumor inoculation and S.a was injected every second day once tumors were detectable. F, AT-3 tumor volume in C57BL/6 mice treated with C5aR antagonist, PMX205, intraperitoneally daily and S.a every second day. Data shown as a mean + SEM from 5 (A), 2 (B and C), 1 (DF) experiments. Data analyzed using Mann–Whitney (AC, and F) or t test (D and E). No significant difference between S.a + Vehicle and S.a.+ NAC (E) or S.a + Vehicle and S.a.+ PMX205 (F); differences between vehicle and NAC were significant as indicated on the graph (E). *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001.

Close modal

To demonstrate that this is a broad phenomenon not restricted to a specific cancer model, we tested our microbial treatment in the B16F10 melanoma model, but also in the AT-3 model of triple-negative breast cancer and KPC model of PDAC because these cancers are considered to be ‘cold’ cancers refractory to immune attack and resistant to checkpoint immunotherapy (34, 35). Remarkably, in these models, microbial particle treatment again substantially inhibited tumor growth (Supplementary Fig. S6A–S6C), indicating that our microbial bioparticle treatment approach is effective in a range of solid tumors and has potential utility for cold cancers like pancreatic cancer, for which few effective therapies exist (36).

Analysis of tumor neutrophil kinetics showed that neutrophil numbers are maintained for at least 3 days following microbe inoculation while neutrophil activation declines over the same period (Fig. 4A and B). This suggests that restimulation of intratumoral neutrophils during their peak recruitment period to boost their antitumor function and maintain the acute phenotype, as well as recruitment of neutrophils from circulation, would be beneficial. Consistent with this hypothesis, analysis of tumor growth over time showed that extending treatment of tumors to every 4 days could inhibit tumor growth (Supplementary Fig. S6D). This suggests that an effective microbial treatment regime could be developed on the basis of the kinetics of neutrophil phenotype switching.

Tumor growth inhibition achieved by treatment with S. aureus bioparticles suggested that microbial activation may represent a common mechanism by which antitumor neutrophils can be recruited. Administration of BCG is in clinical practice to eradicate micrometastases in bladder cancer, although its mechanism of action is unclear (2). As was observed with S. aureus bioparticles, intratumoral treatment with BCG led to both a rapid influx of neutrophils (Supplementary Fig. S7A and S7B) and substantial suppression of tumor growth (Fig. 5B). These data demonstrate that the effect of microbial treatment on tumor growth is not restricted to particular microorganisms like S. aureus, but represents a general principle of microbe-mediated tumor growth control.

Neutrophils are required for tumor growth inhibition

To determine whether MA neutrophils provide a causal contribution to the control of tumor growth in response to microbial treatment, we treated LLC tumors with S. aureus bioparticles to suppress tumor growth but also depleted tumor-infiltrating neutrophils using neutrophil-specific anti-Ly6G antibody (Supplementary Fig. S8A and S8B). Although neutrophils are hard to deplete long-term due to their rapid replenishment by the bone marrow (37), short-term neutrophil depletion reversed the tumor suppression effect of S. aureus (Fig. 5C). These data demonstrate that MA neutrophils are essential for microbe-mediated inhibition of tumor growth.

CD8 T cells are co-effectors of microbial therapy

To test whether adaptive immunity, specifically CD8 T cells, also contribute to the antitumor effect of microbial therapy, we administered S. aureus bioparticles as previously but also depleted CD8 T cells. Our analysis showed that the tumor growth inhibitory effect of microbial therapy was lost in the absence of CD8 T cells (Fig. 5D), indicating that CD8 T cells are important mediators of the antitumor effects of microbial therapy.

Mechanisms of microbe-mediated tumor growth inhibition

The increase in cytotoxicity coupled with iNOS upregulation suggests that the release of ROS is a potential mechanism for tumor growth inhibition by MA neutrophils. We used NAC to inhibit ROS in vivo and observed that NAC treatment significantly increased the rate of tumor growth (Fig. 5E). Notably, while anticancer potential of antioxidants has been recognized for some time, several recent studies have shown that blocking ROS and specifically, administering NAC in vivo can also have pro-tumor effects (38–40). This effect of NAC on tumor growth makes it hard to interpret the effect of coadministration of microbial therapy and NAC. However, our results show that suppressing ROS using NAC is insufficient on its own to inhibit the antitumor effect of microbial therapy (Fig. 5E).

Previous studies reported that neutrophils can kill antibody-coated tumor cells via tropotosis (41). Although we did not observe transfer of cancer cell plasma membrane to neutrophils, indicating that, unlike ADCC-mediated cancer cell death, tumor cell death did not occur by neutrophil trogocytosis, we wanted to test the role of complement in mediating the antitumor effects of microbial therapy. Analysis of tumor sections following microbial treatment showed a significant increase in complement C1q expression (Supplementary Fig. S9), suggesting that complement activation and tumor cell opsonization may contribute to tumor cell removal following microbial therapy. As complement activation generates the potent neutrophil mobilizer and chemotaxin, C5a (42), we tested the role of this complement factor using the C5aR1 antagonist PMX205 (17). Daily administration of the inhibitor did not significantly alter tumor growth, and when administered in combination with microbial therapy, did not affect tumor growth suppression mediated by microbial therapy (Fig. 5F), suggesting that multiple neutrophil chemotaxins can mediate neutrophil recruitment and activation.

Microbial therapy shapes tumor neutrophil transcriptional state

To investigate the neutrophil transcriptional response to microbial stimulation, we conducted single-cell gene expression analysis of neutrophils from tumors treated with S. aureus bioparticles or vehicle control (control) or from unmanipulated tumors, as well as non-neutrophil leukocytes in each tumor (Supplementary Fig. S10A and S10B). After quality control we obtained 7,729 (stimulated), 5,226 (control), and 3,895 (unmanipulated) high-quality cells with an average of 2,042 genes per cell and a total of 20,950 genes detected across all conditions. The combined dataset contained a prominent population of neutrophils among other leukocyte populations, including basophils and eosinophils, macrophages, DCs, B, T, and natural killer (NK) cells within our samples (Fig. 6A).

Figure 6.

Neutrophils in the TME acquire an activated transcriptional state and upregulate cytokines following microbial stimulation. A, Uniform Manifold Approximation and Projection (UMAP) embedding of TME immune cell populations profiled by single-cell sequencing. B, UMAP embedding of neutrophil populations Neu1–5. C, Overlay of experimental conditions (Unmanipulated, Vehicle control, S.a treated), otherwise as in B. D, Stacked bar plot of proportion of cells from three experimental conditions across Neu1–5. E, Heatmap of scaled smoothed expression for the top 15 marker genes for Neu1–5 based on fold enrichment, with columns and rows representing cells and genes, respectively. F, Bubble plot of top biological processes enriched among marker genes for Neu1–5. Bubbles represent significant terms (FDR < 0.05, one-sided Fisher exact test), with size reflecting –log10(P value). G, Heatmap of scaled mean expression aggregated by cell state Neu1–5 for marker genes involved in chemotaxis (Gene Ontology term Neutrophil Chemotaxis GO:0030593). H, UMAP embeddings colored by smoothed expression of chemokines identified as marker genes.

Figure 6.

Neutrophils in the TME acquire an activated transcriptional state and upregulate cytokines following microbial stimulation. A, Uniform Manifold Approximation and Projection (UMAP) embedding of TME immune cell populations profiled by single-cell sequencing. B, UMAP embedding of neutrophil populations Neu1–5. C, Overlay of experimental conditions (Unmanipulated, Vehicle control, S.a treated), otherwise as in B. D, Stacked bar plot of proportion of cells from three experimental conditions across Neu1–5. E, Heatmap of scaled smoothed expression for the top 15 marker genes for Neu1–5 based on fold enrichment, with columns and rows representing cells and genes, respectively. F, Bubble plot of top biological processes enriched among marker genes for Neu1–5. Bubbles represent significant terms (FDR < 0.05, one-sided Fisher exact test), with size reflecting –log10(P value). G, Heatmap of scaled mean expression aggregated by cell state Neu1–5 for marker genes involved in chemotaxis (Gene Ontology term Neutrophil Chemotaxis GO:0030593). H, UMAP embeddings colored by smoothed expression of chemokines identified as marker genes.

Close modal

We identified five distinct transcriptional states in the TME neutrophils, Neu1-Neu5, that segregated closely with treatment condition and were robust to in silico batch correction (Fig. 6B and C; Supplementary Fig. S10C). Indeed, two states, Neu1 and Neu2, contained neutrophils from tumors stimulated with S. aureus bioparticles almost exclusively (respectively 99.5% and 97.5% of cells; Fig. 6D). In contrast, Neu5 mostly contained neutrophils from unmanipulated tumors (96%). The majority of neutrophils in Neu4 were from vehicle control tumors (91%), while Neu3 was a mix of neutrophils from vehicle and unmanipulated tumors (24% and 71% respectively). This indicates that neutrophils in the TME undergo a pronounced change in their transcriptional state in response to microbial stimulation.

Transcriptome diversity differed between neutrophil expression states. While total transcript counts were comparable between Neu1-Neu5, fewer genes were detected in Neu1 and Neu2 compared with Neu5 (Supplementary Fig. S10D). Also, compared with Neu5, Neu1 and Neu2 had a higher percent of transcripts in the top 50 most highly expressed genes, suggesting that S. aureus stimulation strongly upregulates a specific subset of highly expressed genes. We hypothesized that the greater transcriptome diversity in the unstimulated Neu5 neutrophils may reflect a less mature unspecialized effector state induced by the TME. Consistent with this, Neu5 had the lowest maturation score based on a previously identified gene signature (43) compared with all other clusters (Supplementary Fig. S10E). Conversely, the neutrophils responding to S. aureus in Neu1/Neu2 had a transcriptional profile consistent with mature neutrophils.

We identified marker genes for Neu1-Neu5 and determined biological processes enriched among them (Fig. 6E and F; Supplementary Table S1). Neu1 and Neu2, consisting of S. aureus–stimulated neutrophils, upregulated genes associated with an acute response to bacterial infection (Fig. 6F). Among the top markers for Neu1 were mediators of neutrophil effector function such as the iNOS gene Nos2 involved in cell killing by ROS generation, as well as granule proteins lipocalin 2 (Lcn2) and Chitinase-like 1 (Chil1; Fig. 6E). Neu2 was enriched for genes involved in cytokine-mediated signaling, and top markers included members of the CXC chemokine family (Cxcl2, Cxcl3, Cxcl10; Fig. 6E). Neu4, consisting mostly of neutrophils after treatment with vehicle control, was enriched for genes associated with leukocyte adhesion to vascular endothelium, with intercellular adhesion molecule 1 (Icam1) among top markers. Neu3, enriched for neutrophils from unmanipulated and control conditions, expressed genes associated with cell migration including Resistin-like gamma (Retnlg) and Ccl6. Notably, Neu5 was enriched for genes associated with eicosanoid biosynthesis (prostaglandins and leukotrienes; Supplementary Table S1) including prostaglandin-endoperoxide synthase 1 (Ptgs1) as one of the top markers, suggesting that these neutrophils may modulate the chronic inflammatory response in the TME.

Cytokine secretion, chemotaxis and migration were among few processes regulated by most neutrophil states (Fig. 6F), suggesting that changes in transcription underpin the changes in neutrophil behavior observed using intravital microscopy following microbial therapy (Fig. 2). Both Neu2 and Neu3 strongly regulated genes associated with migration (Fig. 6F and G), but while Neu3 expressed several receptors that mediate neutrophil recruitment (C5ar2 and Cxcr2), Neu2 upregulated a number of chemokines including Cxcl2, Cxcl3, Cxcl10, Ccl3, and Ccl4 (Fig. 6H). The induction of neutrophil recruiting chemokines Cxcl2 and Cxcl3 (44) following S. aureus stimulation suggests that neutrophils themselves contribute to maintaining neutrophil numbers in tumors following microbial therapy. On the other hand, Cxcl10 as well as Ccl3 and Ccl4 can stimulate NK and CD8 T-cell recruitment to tumors (45), suggesting that MA neutrophils may promote adaptive immunity. Conversely, Ccl6, which binds to CCR1 and is thought to promote tumor metastasis (46), was most highly expressed by Neu3 neutrophils. This indicates that potent immune modulatory cytokines differentiate neutrophil states in tumors. Together, this shows that neutrophils in the TME acquire a mature effector transcriptional state upon S. aureus stimulation that may create a distinct tumor immune microenvironment through differential regulation of chemokines.

Activation of adaptive immunity in response to microbial therapy

We showed that MA neutrophils expressed chemokines that promote T-cell recruitment to tumors (Fig. 6G) and depletion of CD8 T cells demonstrated that they are important effectors of microbial therapy (Fig. 5D). Concomitantly, three rounds of microbial therapy increased recruitment of activated CD8 T cells to tumors (Fig. 7A and B) and the number of CD8 T cells expressing of IFNγ, perforin, granzyme (increasing trend) and CD107a (Fig. 7CF). The number of tumor infiltrating CD8 T cells expressing the exhaustion markers PD-1 and Tim-3 was also augmented (Fig. 7G and H). Similarly, CD8 T cells in the tumor draining lymph nodes also showed enhanced recruitment, activation (Fig. 7I and J), and effector function ((Fig. 7KN) but no change in the expression of exhaustion markers (Fig. 7O and P). These data indicate that CD8 T-cell effector function is enhanced both in tumors and draining lymph nodes in response to microbial therapy.

Figure 7.

Microbial therapy enhances CD8 T-cell effector function, response to checkpoint inhibitor therapy, and protects from tumor rechallenge. CD8 T-cell number, activation, and effector function in LLC tumors and draining lymph nodes following three rounds of S. aureus (S.a) bioparticle treatment every second day were assessed 72 hours after the last treatment. Cell number (A and I), expression of CD69 (B and J), IFNγ (C and K), perforin (D and L), granzyme (E and M), CD107 (F and N), PD-1 (G and O), and TIM3 (H and P) in tumor/lymph node CD8 T cells. Q, Tumor volume in C57BL/6 mice bearing AT-3 tumors that were treated with checkpoint inhibitor antibodies (anti–PD-1 and anti-CTLA4, 5 injections every third day) and either S.a (three injections every second day) or vehicle control or isotype treated. R, Tumor volume in C57BL/6 mice inoculated with AT-3 tumor cells that have previously received AT-3 cells and were treated with microbial therapy to suppress tumor growth and rested for at least 60 days or tumor naïve mice. Data shown as mean + SEM, with each circle representing one tumor or lymph node (AP) or at least four tumors per time point (Q and R). Data analyzed using Mann–Whitney (AE and GR) or t test (F). *, P ≤ 0.05; **, P ≤ 0.01; ****, P ≤ 0.0001.

Figure 7.

Microbial therapy enhances CD8 T-cell effector function, response to checkpoint inhibitor therapy, and protects from tumor rechallenge. CD8 T-cell number, activation, and effector function in LLC tumors and draining lymph nodes following three rounds of S. aureus (S.a) bioparticle treatment every second day were assessed 72 hours after the last treatment. Cell number (A and I), expression of CD69 (B and J), IFNγ (C and K), perforin (D and L), granzyme (E and M), CD107 (F and N), PD-1 (G and O), and TIM3 (H and P) in tumor/lymph node CD8 T cells. Q, Tumor volume in C57BL/6 mice bearing AT-3 tumors that were treated with checkpoint inhibitor antibodies (anti–PD-1 and anti-CTLA4, 5 injections every third day) and either S.a (three injections every second day) or vehicle control or isotype treated. R, Tumor volume in C57BL/6 mice inoculated with AT-3 tumor cells that have previously received AT-3 cells and were treated with microbial therapy to suppress tumor growth and rested for at least 60 days or tumor naïve mice. Data shown as mean + SEM, with each circle representing one tumor or lymph node (AP) or at least four tumors per time point (Q and R). Data analyzed using Mann–Whitney (AE and GR) or t test (F). *, P ≤ 0.05; **, P ≤ 0.01; ****, P ≤ 0.0001.

Close modal

Microbial treatment synergizes with checkpoint inhibitor therapy

Increased CD8 T-cell effector function suggests that microbial therapy may improve effectiveness of T-cell–based immune therapies, such as checkpoint inhibition. To test this, we used the AT-3 tumor model and administered three rounds of S. aureus bioparticles intratumorally in combination with checkpoint inhibitors anti–PD-1 and anti–CTLA4. Compared with mice that received checkpoint inhibitor antibodies only, we found that microbial bioparticles significantly enhanced suppression of tumor growth by checkpoint inhibitors (Fig. 7Q).

Because tumor recurrence represents a major clinical challenge, we asked whether microbial therapy could improve therapeutic outcomes in a model of tumor recurrence. Mice were inoculated with AT-3 tumor cells and treated with microbial therapy to repress tumor growth, then rested for at least 60 days and subsequently rechallenged with AT-3 tumor cells. We found that in the absence of any additional microbial treatment, mice that were previously treated with microbial therapy successfully suppressed tumor growth compared with naïve mice (Fig. 7R). Together these results indicate that microbial therapy can induce an effective CD8 T-cell response and enhance the therapeutic impact of checkpoint immunotherapy.

Our study demonstrated that changing the TME from a chronic, aberrant wound healing response to acute microbe-triggered inflammation induced extensive changes in tumor neutrophil transcription, migration, and function and ultimately repressed tumor growth in a neutrophil-dependent manner. Tumor neutrophils were either slow moving or sessile but increased their motility following microbial therapy and formed large clusters corresponding to areas cleared of collagen and tumor cells, indicating that microbial signals in the TME induced neutrophils to take on some of the features of the antimicrobial response in tissues and to mediate tumor remodeling by removing tumor cells. Indeed, in the context of microbial infection with Toxoplasma gondii, neutrophils migrate rapidly between foci of inflammation to form dynamic swarms, and remodel underlying tissue (28, 47). MA neutrophils also shifted their function from VEGF production to release of ROS, indicating that they mediate their antitumor function by coopting antimicrobial effector mechanisms, which is consistent with the transition from a wound repair to a tumor killing phenotype.

The switch in neutrophil functional state was dependent upon a pronounced reprogramming of the neutrophil transcriptional landscape. In response to microbial treatment tumor neutrophils upregulated genes associated with immune activation and microbial defense. MA neutrophils also upregulated expression of chemokines that recruit neutrophils as well as NK and CD8 T cells, suggesting these neutrophils contribute to broader antitumor immunity by reshaping the TME and enhancing the recruitment of antitumor lymphocytes.

Analysis of neutrophil plasticity in the TME revealed that the bioparticle-driven switch in neutrophil phenotype was transient and restimulation was required to maintain neutrophils in the antitumor state and repress tumor growth. These results have important implications for clinical application of microbial therapy, which could provide benefit as neoadjuvant therapies to shrink solid tumors prior to surgery, as adjuvant therapy to remove residual cancer cells and as a way of reducing inoperable tumors. Although tumor accessibility can limit the scope of intra-tumoral injections, image-guided procedures may provide an opportunity to target less accessible tumors. There are currently several clinical trials using intratumor injections in non-superficial cancers such as pancreatic (source: https://clinicaltrials.gov). Advances in structural chemistry may also permit depot formulations that maintain efficacy with fewer injections with the eventual goal of broadening the use of microbial therapy. Delivery platforms such as the bacterially derived nanocells (48) can be used to deliver microbial therapy to poorly accessible tumors or to tumors that spread to multiple sites where an intravenous delivery system would be advantageous.

Immune suppression mediated by cancer and immune cells is a key obstacle for an effective T-cell antitumor response. Here, we show that successive administration of microbial bioparticles led to an increase in effector CD8 T cells in tumors and draining lymph nodes and significantly enhanced the efficacy of checkpoint inhibitor therapy, suggesting that combination strategies targeting both innate and adaptive immunity may synergize to overcome checkpoint blockade resistance and promote tumor killing. Furthermore, microbial treatment of primary tumors conferred protection in a rechallenge model, indicating that microbial therapy-mediated tumor repression may establish a protective memory response. Therefore, microbial therapy may provide a pathway to increase adaptive immune responses in poorly infiltrated cancers (e.g., breast or pancreatic cancers), where insufficient immune infiltration and immunosuppression have been major obstacles for checkpoint inhibitor therapy (34, 35). This substantially expands potential applications of microbial immunotherapy in treatment of solid tumors.

The remarkable tumor growth inhibition achieved by treatment with both S. aureus bioparticles and BCG supports a general principle whereby microbial therapy switches neutrophils from a wound healing program to a potent cytotoxic response. Despite its weakened state, live attenuated BCG has the potential to cause disseminated mycobacterium infection in patients (49), whereas using killed bacterial bioparticles may represent a safer alternative especially for immunocompromised patients. TLR agonist formulations such as TLR3 and TLR9 agonists can stimulate DC maturation and enhance CD8 T-cell responses (50) but show only limited clinical efficacy as monotherapies (1). This may reflect the existence of multiple nonredundant pathways that have to be targeted simultaneously to improve clinical efficacy. Further, TLR3 and TLR9 agonists as standalone therapies (or as vaccine adjuvants) may be insufficient to overcome immunosuppression mediated by other immune subsets. Microbial therapy successfully inhibited tumor growth in several preclinical tumor models, indicating that neutrophil plasticity in cancer could be exploited across a range of solid tumors. Our work supports the use of killed microbes as a promising strategy to target multiple TLRs and ensure robust activation of a broad range of immune subsets.

A.O. Yam reports grants from Australian Government Research Training Program, Royal Australasian College of Physicians Fellows Research Entry Scholarships and from Phil Salter Immuno-Oncology Fellowship during the conduct of the study. A. Jakovija reports grants from UNSW and other support from scholarship and government during the conduct of the study. L.D. Goldstein reports other support from Kinghorn Foundation during the conduct of the study and from Genentech Inc. outside the submitted work; in addition, L.D. Goldstein has a patent for "Methods for diagnosing and treating cancer by means of the expression status and mutational status of NRF2 and downstream target genes of said gene" issued. S.T. Grey reports grants from NHMRC during the conduct of the study. T. Chtanova reports grants from the National Breast Cancer Foundation (IIRS-22-053, IIRS-19-027), PanKind Australian Pancreatic Cancer Foundation (the Michael Luscombe Grant), the UNSW Cellular Genomics Futures Institute, UNSW Sydney, and interdisciplinary funding scheme during the conduct of the study, and grants from Noxopharm Limited outside the submitted work. No disclosures were reported by the other authors.

A.O. Yam: Conceptualization, formal analysis, validation, investigation, visualization, methodology, writing–original draft, writing–review and editing. J. Bailey: Formal analysis, validation, investigation, visualization. F. Lin: Data curation, formal analysis, validation, investigation, methodology. A. Jakovija: Data curation, formal analysis, investigation, methodology. S.E. Youlten: Software, formal analysis, validation, investigation, visualization, methodology, writing–review and editing. C. Counoupas: Resources, investigation, methodology. M. Gunzer: Resources, investigation, methodology. T. Bald: Resources, investigation, writing–original draft. T.M. Woodruff: Resources, investigation, methodology, writing–review and editing. J.A. Triccas: Resources, supervision, investigation, methodology, writing–original draft. L.D. Goldstein: Software, formal analysis, supervision, validation, investigation, visualization, methodology, writing–review and editing. D. Gallego-Ortega: Resources, validation, investigation, methodology. S.T. Grey: Conceptualization, formal analysis, investigation, writing–original draft, writing–review and editing. T. Chtanova: Conceptualization, resources, formal analysis, supervision, funding acquisition, validation, investigation, visualization, methodology, writing–original draft, project administration, writing–review and editing.

This research was supported by funding to T. Chtanova from the National Breast Cancer Foundation (IIRS-22–053, IIRS-19–027), UNSW Cellular Genomics Futures Institute, UNSW Sydney, interdisciplinary funding scheme grants, and Avner Grant PanKind Australian Pancreatic Cancer Foundation (the Michael Luscombe Grant). A.O. Yam is supported by the Australian Government Research Training Program, Royal Australasian College of Physicians Fellows Research Entry Scholarships, and a Phil Salter Immuno-Oncology Fellowship. Intravital microscopy supported by Peter and Val Duncan and ACRF INCITe Centre. S.E. Youlten and L.D. Goldstein acknowledge funding from the Kinghorn Foundation. D. Gallego-Ortega is a National Breast Cancer Foundation Elaine Henry Fellow (IIRS-21–096). S.T. Grey is a NHMRC Senior Research Fellow (GNT1140691).

The publication costs of this article were defrayed in part by the payment of publication fees. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.

Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

1.
Urban-Wojciuk
Z
,
Khan
MM
,
Oyler
BL
,
Fahraeus
R
,
Marek-Trzonkowska
N
,
Nita-Lazar
A
, et al
.
The role of TLRs in anticancer immunity and tumor rejection
.
Front Immunol
2019
;
10
:
2388
.
2.
Pettenati
C
,
Ingersoll
MA
.
Mechanisms of BCG immunotherapy and its outlook for bladder cancer
.
Nat Rev Urol
2018
;
15
:
615
25
.
3.
Suttmann
H
,
Riemensberger
J
,
Bentien
G
,
Schmaltz
D
,
Stockle
M
,
Jocham
D
, et al
.
Neutrophil granulocytes are required for effective Bacillus Calmette-Guérin immunotherapy of bladder cancer and orchestrate local immune responses
.
Cancer Res
2006
;
66
:
8250
7
.
4.
Coffelt
SB
,
Wellenstein
MD
,
de Visser
KE
.
Neutrophils in cancer: neutral no more
.
Nat Rev Cancer
2016
;
16
:
431
46
.
5.
Quail
DF
,
Amulic
B
,
Aziz
M
,
Barnes
BJ
,
Eruslanov
E
,
Fridlender
ZG
, et al
.
Neutrophil phenotypes and functions in cancer: a consensus statement
.
J Exp Med
2022
;
219
:
e20220011
.
6.
Blaisdell
A
,
Crequer
A
,
Columbus
D
,
Daikoku
T
,
Mittal
K
,
Dey
SK
, et al
.
Neutrophils oppose uterine epithelial carcinogenesis via debridement of hypoxic tumor cells
.
Cancer Cell
2015
;
28
:
785
99
.
7.
Granot
Z
,
Henke
E
,
Comen
EA
,
King
TA
,
Norton
L
,
Benezra
R
.
Tumor-entrained neutrophils inhibit seeding in the premetastatic lung
.
Cancer Cell
2011
;
20
:
300
14
.
8.
Stoppacciaro
A
,
Melani
C
,
Parenza
M
,
Mastracchio
A
,
Bassi
C
,
Baroni
C
, et al
.
Regression of an established tumor genetically modified to release granulocyte colony-stimulating factor requires granulocyte-T cell cooperation and T cell–produced interferon gamma
.
J Exp Med
1993
;
178
:
151
61
.
9.
Colombo
MP
,
Ferrari
G
,
Stoppacciaro
A
,
Parenza
M
,
Rodolfo
M
,
Mavilio
F
, et al
.
Granulocyte colony-stimulating factor gene transfer suppresses tumorigenicity of a murine adenocarcinoma in vivo
.
J Exp Med
1991
;
173
:
889
97
.
10.
Fridlender
ZG
,
Sun
J
,
Kim
S
,
Kapoor
V
,
Cheng
G
,
Ling
L
, et al
.
Polarization of tumor-associated neutrophil phenotype by TGF-beta: "N1" versus "N2" TAN
.
Cancer Cell
2009
;
16
:
183
94
.
11.
Tomura
M
,
Yoshida
N
,
Tanaka
J
,
Karasawa
S
,
Miwa
Y
,
Miyawaki
A
, et al
.
Monitoring cellular movement in vivo with photoconvertible fluorescence protein "Kaede" transgenic mice
.
Proc Natl Acad Sci USA
2008
;
105
:
10871
6
.
12.
Hasenberg
A
,
Hasenberg
M
,
Mann
L
,
Neumann
F
,
Borkenstein
L
,
Stecher
M
, et al
.
Catchup: a mouse model for imaging-based tracking and modulation of neutrophil granulocytes
.
Nat Methods
2015
;
12
:
445
52
.
13.
Tikoo
S
,
Jain
R
,
Tomasetig
F
,
On
K
,
Martinez
B
,
Heu
C
, et al
.
Amelanotic B16-F10 melanoma compatible with advanced three-dimensional imaging modalities
.
J Invest Dermatol
2021
;
141
:
2090
4
.
14.
Morton
JP
,
Karim
SA
,
Graham
K
,
Timpson
P
,
Jamieson
N
,
Athineos
D
, et al
.
Dasatinib inhibits the development of metastases in a mouse model of pancreatic ductal adenocarcinoma
.
Gastroenterology
2010
;
139
:
292
303
.
15.
Hampton
HR
,
Bailey
J
,
Tomura
M
,
Brink
R
,
Chtanova
T
.
Microbe-dependent lymphatic migration of neutrophils modulates lymphocyte proliferation in lymph nodes
.
Nat Commun
2015
;
6
:
7139
.
16.
Torcellan
T
,
Hampton
HR
,
Bailey
J
,
Tomura
M
,
Brink
R
,
Chtanova
T
.
In vivo photolabeling of tumor-infiltrating cells reveals highly regulated egress of T-cell subsets from tumors
.
Proc Natl Acad Sci USA
2017
;
114
:
5677
82
.
17.
Kumar
V
,
Lee
JD
,
Clark
RJ
,
Noakes
PG
,
Taylor
SM
,
Woodruff
TM
.
Preclinical pharmacokinetics of complement C5a receptor antagonists PMX53 and PMX205 in mice
.
ACS Omega
2020
;
5
:
2345
54
.
18.
Mixter
PF
,
Russell
JQ
,
Morrissette
GJ
,
Charland
C
,
Aleman-Hoey
D
,
Budd
RC
.
A model for the origin of TCR-alphabeta+ CD4-CD8- B220+ cells based on high affinity TCR signals
.
J Immunol
1999
;
162
:
5747
56
.
19.
Limkar
AR
,
Mai
E
,
Sek
AC
,
Percopo
CM
,
Rosenberg
HF
.
Frontline science: cytokine-mediated developmental phenotype of mouse eosinophils: IL5-associated expression of the Ly6G/Gr1 surface Ag
.
J Leukoc Biol
2020
;
107
:
367
77
.
20.
Temiz
MZ
,
Colakerol
A
,
Ulus
I
,
Kilic
E
,
Paslanmaz
F
,
Sahin
S
, et al
.
Prediction of non–muscle-invasive bladder cancer recurrence during intravesical BCG immunotherapy by use of peripheral blood eosinophil count and percentage: a preliminary report
.
Cancer Immunol Immunother
2021
;
70
:
245
52
.
21.
Evrard
M
,
Kwok
IWH
,
Chong
SZ
,
Teng
KWW
,
Becht
E
,
Chen
J
, et al
.
Developmental analysis of bone marrow neutrophils reveals populations specialized in expansion, trafficking, and effector functions
.
Immunity
2018
;
48
:
364
79
.
22.
Amulic
B
,
Cazalet
C
,
Hayes
GL
,
Metzler
KD
,
Zychlinsky
A
.
Neutrophil function: from mechanisms to disease
.
Annu Rev Immunol
2012
;
30
:
459
89
.
23.
Grieshaber-Bouyer
R
,
Nigrovic
PA
.
Neutrophil Heterogeneity as therapeutic opportunity in immune-mediated disease
.
Front Immunol
2019
;
10
:
346
.
24.
Silvestre-Roig
C
,
Hidalgo
A
,
Soehnlein
O
.
Neutrophil heterogeneity: implications for homeostasis and pathogenesis
.
Blood
2016
;
127
:
2173
81
.
25.
Nozawa
H
,
Chiu
C
,
Hanahan
D
.
Infiltrating neutrophils mediate the initial angiogenic switch in a mouse model of multistage carcinogenesis
.
Proc Natl Acad Sci USA
2006
;
103
:
12493
8
.
26.
Wheeler
MA
,
Smith
SD
,
Garcia-Cardena
G
,
Nathan
CF
,
Weiss
RM
,
Sessa
WC
.
Bacterial infection induces nitric oxide synthase in human neutrophils
.
J Clin Invest
1997
;
99
:
110
6
.
27.
Finisguerra
V
,
Di Conza
G
,
Di Matteo
M
,
Serneels
J
,
Costa
S
,
Thompson
AA
, et al
.
MET is required for the recruitment of antitumoral neutrophils
.
Nature
2015
;
522
:
349
53
.
28.
Chtanova
T
,
Schaeffer
M
,
Han
SJ
,
van Dooren
GG
,
Nollmann
M
,
Herzmark
P
, et al
.
Dynamics of neutrophil migration in lymph nodes during infection
.
Immunity
2008
;
29
:
487
96
.
29.
Lammermann
T
,
Afonso
PV
,
Angermann
BR
,
Wang
JM
,
Kastenmuller
W
,
Parent
CA
, et al
.
Neutrophil swarms require LTB4 and integrins at sites of cell death in vivo
.
Nature
2013
;
498
:
371
5
.
30.
Lewis
HD
,
Liddle
J
,
Coote
JE
,
Atkinson
SJ
,
Barker
MD
,
Bax
BD
, et al
.
Inhibition of PAD4 activity is sufficient to disrupt mouse and human NET formation
.
Nat Chem Biol
2015
;
11
:
189
91
.
31.
Martin
C
,
Burdon
PC
,
Bridger
G
,
Gutierrez-Ramos
JC
,
Williams
TJ
,
Rankin
SM
.
Chemokines acting via CXCR2 and CXCR4 control the release of neutrophils from the bone marrow and their return following senescence
.
Immunity
2003
;
19
:
583
93
.
32.
Devi
S
,
Wang
Y
,
Chew
WK
,
Lima
R
,
AG
N
,
Mattar
CN
, et al
.
Neutrophil mobilization via plerixafor-mediated CXCR4 inhibition arises from lung demargination and blockade of neutrophil homing to the bone marrow
.
J Exp Med
2013
;
210
:
2321
36
.
33.
Lord
BI
,
Molineux
G
,
Pojda
Z
,
Souza
LM
,
Mermod
JJ
,
Dexter
TM
.
Myeloid cell kinetics in mice treated with recombinant interleukin-3, granulocyte colony-stimulating factor (CSF), or granulocyte-macrophage CSF in vivo
.
Blood
1991
;
77
:
2154
9
.
34.
Gatti-Mays
ME
,
Balko
JM
,
Gameiro
SR
,
Bear
HD
,
Prabhakaran
S
,
Fukui
J
, et al
.
If we build it they will come: targeting the immune response to breast cancer
.
NPJ Breast Cancer
2019
;
5
:
37
.
35.
Bonaventura
P
,
Shekarian
T
,
Alcazer
V
,
Valladeau-Guilemond
J
,
Valsesia-Wittmann
S
,
Amigorena
S
, et al
.
Cold tumors: a therapeutic challenge for immunotherapy
.
Front Immunol
2019
;
10
:
168
.
36.
Manji
GA
,
Olive
KP
,
Saenger
YM
,
Oberstein
P
.
Current and emerging therapies in metastatic pancreatic cancer
.
Clin Cancer Res
2017
;
23
:
1670
8
.
37.
Boivin
G
,
Faget
J
,
Ancey
PB
,
Gkasti
A
,
Mussard
J
,
Engblom
C
, et al
.
Durable and controlled depletion of neutrophils in mice
.
Nat Commun
2020
;
11
:
2762
.
38.
Le Gal
K
,
Ibrahim
MX
,
Wiel
C
,
Sayin
VI
,
Akula
MK
,
Karlsson
C
, et al
.
Antioxidants can increase melanoma metastasis in mice
.
Sci Transl Med
2015
;
7
:
308re8
.
39.
Sayin
VI
,
Ibrahim
MX
,
Larsson
E
,
Nilsson
JA
,
Lindahl
P
,
Bergo
MO
.
Antioxidants accelerate lung cancer progression in mice
.
Sci Transl Med
2014
;
6
:
221ra15
.
40.
Zhang
VX
,
Sze
KM
,
Chan
LK
,
Ho
DW
,
Tsui
YM
,
Chiu
YT
, et al
.
Antioxidant supplements promote tumor formation and growth and confer drug resistance in hepatocellular carcinoma by reducing intracellular ROS and induction of TMBIM1
.
Cell Biosci
2021
;
11
:
217
.
41.
Matlung
HL
,
Babes
L
,
Zhao
XW
,
van Houdt
M
,
Treffers
LW
,
van Rees
DJ
, et al
.
Neutrophils kill antibody-opsonized cancer cells by trogoptosis
.
Cell Rep
2018
;
23
:
3946
59
.
42.
Pellas
TC
,
Boyar
W
,
van Oostrum
J
,
Wasvary
J
,
Fryer
LR
,
Pastor
G
, et al
.
Novel C5a receptor antagonists regulate neutrophil functions in vitro and in vivo
.
J Immunol
1998
;
160
:
5616
21
.
43.
Xie
X
,
Shi
Q
,
Wu
P
,
Zhang
X
,
Kambara
H
,
Su
J
, et al
.
Single-cell transcriptome profiling reveals neutrophil heterogeneity in homeostasis and infection
.
Nat Immunol
2020
;
21
:
1119
33
.
44.
Griffith
JW
,
Sokol
CL
,
Luster
AD
.
Chemokines and chemokine receptors: positioning cells for host defense and immunity
.
Annu Rev Immunol
2014
;
32
:
659
702
.
45.
Franciszkiewicz
K
,
Boissonnas
A
,
Boutet
M
,
Combadiere
C
,
Mami-Chouaib
F
.
Role of chemokines and chemokine receptors in shaping the effector phase of the antitumor immune response
.
Cancer Res
2012
;
72
:
6325
32
.
46.
Li
F
,
Du
X
,
Lan
F
,
Li
N
,
Zhang
C
,
Zhu
C
, et al
.
Eosinophilic inflammation promotes CCL6-dependent metastatic tumor growth
.
Sci Adv
2021
;
7
:
eabb5943
.
47.
Yam
AO
,
Chtanova
T
.
Imaging the neutrophil: intravital microscopy provides a dynamic view of neutrophil functions in host immunity
.
Cell Immunol
2019
;
350
:
103898
.
48.
Sagnella
SM
,
Yang
L
,
Stubbs
GE
,
Boslem
E
,
Martino-Echarri
E
,
Smolarczyk
K
, et al
.
Cyto-immuno-therapy for cancer: a pathway elicited by tumor-targeted, cytotoxic drug-packaged bacterially derived nanocells
.
Cancer Cell
2020
;
37
:
354
70
.
49.
Koya
MP
,
Simon
MA
,
Soloway
MS
.
Complications of intravesical therapy for urothelial cancer of the bladder
.
J Urol
2006
;
175
:
2004
10
.
50.
Smith
M
,
Garcia-Martinez
E
,
Pitter
MR
,
Fucikova
J
,
Spisek
R
,
Zitvogel
L
, et al
.
Trial watch: Toll-like receptor agonists in cancer immunotherapy
.
Oncoimmunology
2018
;
7
:
e1526250
.
This open access article is distributed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) license.