Disruption of hypoxia-inducible factor-2α in neutrophils decreases colitis-associated colon cancer

Abstract

Neutrophils are abundant immune cells in the colon tumor microenvironment. Studies have shown that neutrophils are recruited into hypoxic foci in colon cancer. However, the impact of hypoxia signaling on neutrophil function and its involvement in colon tumorigenesis remain unclear. To address this, we generated mice with a deletion of hypoxia-inducible factor (HIF)-1α or HIF-2α in neutrophils driven by the MRP8Cre (HIF-1α^ΔNeu) or (HIF-2α^ΔNeu) and littermate controls. In an azoxymethane (AOM)/dextran sulfate sodium (DSS) model of colon cancer, the disruption of neutrophils-HIF-1α did not result in any significant changes in body weight, colon length, tumor size, proliferation, or burden. However, the disruption of HIF-2α in neutrophils led to a slight increase in body weight, a significant decrease in the number of tumors, and a reduction in tumor size and volume compared with their littermate controls. Histological analysis of colon tissue from mice with HIF-2α-deficient neutrophils revealed notable reductions in proliferation as compared with control mice. In addition, we observed reduced levels of proinflammatory cytokines, such as TNF-α and IL-1β, in neutrophil-specific HIF-2α-deficient mice in both the tumor tissue as well as the neutrophils. Importantly, it is worth noting that the reduced tumorigenesis associated with HIF-2α deficiency in neutrophils was not evident in already established syngeneic tumors or a DSS-induced inflammation model, indicating a potential role of HIF-2α specifically in colon tumorigenesis. In conclusion, we found that the loss of neutrophil-specific HIF-2α slows colon tumor growth and progression by reducing the levels of inflammatory mediators.

NEW & NOTEWORTHY Despite the importance of hypoxia and neutrophils in colorectal cancer (CRC), the contribution of neutrophil-specific HIFs to colon tumorigenesis is not known. We describe that neutrophil HIF-1α has no impact on colon cancer, whereas neutrophil HIF-2α loss reduces CRC growth by decreasing proinflammatory and immunosuppressive cytokines. Furthermore, neutrophil HIF-2α does not reduce preestablished tumor growth or inflammation-induced colitis. The present study offers novel potential of neutrophil HIF-2α as a therapeutic target in CRC.

INTRODUCTION

Colorectal cancer (CRC) is the second-leading cause of cancer mortality worldwide (1, 2). Chronic inflammation as observed in patients with ulcerative colitis and Crohn’s disease, collectively known as inflammatory bowel disease (IBD) is a risk factor for CRC (3–5). Moreover, the extent of inflammation and duration of the disease correlate to the increased cumulative risk for CRC (6–8). There are several overlapping mutations that drive sporadic CRC and IBD-associated CRC (9, 10). However, adenomatous polyposis coli (APC) mutations are slightly less common, and TP53 mutations are detected earlier and more frequently in IBD-associated CRC (3, 10, 11). Hence, the mechanisms that contribute to chronic inflammation play a significant role in risk stratification, as well as the survival and proliferation of cancer cells, ultimately leading to the development of colon tumorigenesis.

Hypoxia is a common characteristic shared by colon tumors and sites of inflammation. Tissue hypoxia can be attributed to the increased metabolic demands of rapidly proliferating cancer cells, the infiltration of immune cells, and perturbations in vascular dynamics, culminating in diminished oxygen levels within tumor or inflamed colon tissue. Consequently, hypoxia-inducible factors (HIFs) are activated. HIFs are composed of an oxygen-sensitive α subunit (HIF-1α, HIF-2α, and HIF-3α) and a constitutively expressed β-subunit (ARNT) (12, 13). Under normal oxygen concentrations (normoxia), oxygen-sensing enzymes known as prolyl hydroxylases (PHDs) utilize oxygen as a substrate for hydroxylation, leading to von Hippel-Lindau gene (VHL)-induced rapid proteasomal degradation of HIF-1α and HIF-2α (14, 15). As tissue O₂ levels decrease, HIFs stabilize and heterodimerize with ARNT. This heterodimer complex translocates to the nucleus to induce transcription of target genes (16). HIF-1α and HIF-2α have diverse functions in intestinal inflammation (17). HIF-2α increases inflammation through the activation of several proinflammatory mediators triggering inflammatory response in intestinal epithelial cells (18–20). The overexpression of HIF-2α in the intestinal epithelial cells in mice induces spontaneous colitis and the mice are susceptible to intestinal injury (19). On the other hand, HIF-1α acts as a protective factor to maintain intestinal barrier function and mucosal immune response. Disruption of HIF-1α in intestinal epithelial cells leads to increased susceptibility to intestinal injury (21–25). In CRC, overexpression of intestinal epithelial HIF-2α but not HIF-1α increases proliferation, tumor progression, and resistance to standard chemotherapeutics (19, 26).

The colon tumor microenvironment involves a complex setting of both innate and adaptive immune cells of which granulocytic cells such as neutrophils are the most prevalent (27). The polymorphonuclear neutrophils can exhibit antitumor and protumor phenotypes depending upon the environmental cues (28, 29). Antitumor neutrophils promote tumorigenesis through activation of oncogenic signaling via secretion of elastase, angiogenic factors, and suppression of antitumor immunity (30–32). We have previously shown that colon tumors are highly infiltrated with neutrophils, which are recruited to hypoxic niches and promote colon tumorigenesis (33). It is well known that HIF signaling in immune cells plays an important role in regulating inflammatory responses that mediate progression to cancer (34). HIF-1α and HIF-2α are expressed in neutrophils and regulate various functions. HIF-1α in neutrophils has been shown to regulate glycolysis and ATP generation and its deficiency leads to impairment in neutrophil mobility, aggregation, phagocytosis, and bacterial killing. (35, 36). Neutrophil HIF-2α has a role in hypoxia-induced neutrophil survival. The deficiency of HIF-2α promotes neutrophil apoptosis both in vivo and ex vivo, which suppresses neutrophil inflammatory responses during acute lung injury (37). The precise role of neutrophil HIF-1α and HIF-2α in the tumor microenvironment is not well explored.

We established a mouse model where HIF-1α or HIF-2α was selectively and constitutively disrupted in neutrophils. Abrogating the hypoxic response of neutrophils through HIF-1α loss yielded no discernible effects on colon tumorigenesis. In contrast, the deletion of neutrophil-specific HIF-2α decreased tumor growth, proliferation, survival, and the progression of CRC in mouse models. Neutrophil HIF-2α plays a pivotal role as a modulator of proinflammatory cytokine levels. Disruption of HIF-2α led to a decrease in the expression of neutrophil and tissue TNF-α and IL-1β correlating with the attenuation of inflammation-driven cancer progression. This study not only provides novel insights into the precise role of hypoxia-driven inflammatory responses within neutrophils during the progression of colon tumors but also posits a robust foundation for considering HIF-2α within tumor-associated neutrophils as a promising target for therapeutic intervention.

METHODS

Mice and Treatments

All animal studies were approved by the Institutional Animal Care and Use Committee at the University of Michigan. For all experiments, 6- to 8-wk-old male and female mice on C57BL/6 background were used. HIF-1α^F/F and HIF-2α^F/F mice were crossed with Mrp8-Cre-expressing mice (Mrp8^Cre) as previously used by other research groups (38) to produce neutrophil-specific constitutive HIF-1α^F/F (HIF-1α^ΔNeu) or HIF-2α^F/F knockout (HIF-2α^ΔNeu) mice. For the syngeneic model, 2 million MC38 cells were subcutaneously injected in both flanks of HIF-2α^F/F and HIF-2α^ΔNeu mice. After 3 wk, mice were euthanized, and tumors were collected, weighed, and processed. For the colitis-associated colon tumor model, HIF-1α^ΔNeu or HIF-2α^ΔNeu mice and their respective littermates were injected intraperitoneally with azoxymethane (AOM) (Sigma, A5486; 10 mg/kg). After 3 days, mice were cycled on and off with initially 1.5% dextran sulfate sodium (DSS) (MP Biomedicals, 160110) for one cycle and 2% DSS for the following two cycles in their drinking water for 7 days. Body weights were recorded every 5 days. Mice were euthanized 4 wk after the last cycle of dextran sulfate sodium. Tumors were counted, measured, imaged, and Swiss-rolled for histological analysis. The normal as well as tumor colon tissue was flash-frozen in liquid N₂ and stored for gene expression assays. For dextran sulfate sodium (DSS) studies, HIF-2α^ΔNeu mice and their respective littermates mice were placed on 3.0% DSS in drinking water for 7 days and then the DSS water was replaced with normal drinking water for 4 days after which the mice were euthanized, and colonic mucosal cells were scraped and frozen in liquid N₂ and stored for further assays. A portion of the colon was Swiss-rolled and fixed in formalin for hematoxylin and eosin (H&E) staining.

Quantitative Real-Time Reverse Transcriptase PCR

RNA was isolated from frozen normal or colon tumor tissue, syngeneic tumor tissue, or colon tissue using TRIzol lysis reagent (Invitrogen, Thermo Fisher) as previously described (39–41). After quantification with NanoDrop 2000 (NanoDrop products), 1 μg RNA with a purity of ∼2.0 (260/280 ratio) was reverse transcribed using M-MLV Reverse Transcriptase (Thermo Fisher Scientific). cDNA was then quantified by SYBR Green master mix (Applied Biosystems, Thermo Fisher) and run on a QuantStudio 5 Real-Time PCR System (Applied Biosystems; primer sequence listed in Table 1). Cycle threshold (Ct) values were normalized to β-actin and expressed as fold differences from controls.

Table 1.

Primers for quantitative PCR

Sr. No.	Gene Name	Forward Sequence	Reverse Sequence
1.	HIF-1α	ACCTTCATCGGAAACTCCAAAG	ACTGTTAGGCTCAGGTGAACT
2.	HIF-2α	CTGAGGAAGGAGAAATCCCGT	TGTGTCCGAAGGAAGCTGATG
3.	TNF-α	CCCTCACACTCAGATCATCTTCT	GCTACGACGTGGGCTACAG
4.	IL-1β	AAGAGCTTCAGGCAGGCAGTATCA	TGCAGCTGTCTAATGGGAACGTCA
5.	IFN-γ	CAGCAACAGCAAGGCGAAAAAGG	TTTCCGCTTCCTGAGGCTGGAT
6.	TGF-β	TCAGACATTCGGGAAGCAGT	TCGAAAGCCCTGTATTCCGT
7.	Arg1	ACCTGGCCTTTGTTGATGTCCCTA	AGAGATGCTTCCAACTGCCAG ACT
8.	IL-10	TGCACTACCAAAGCCACAAAGCAG	AGTAAGAGCAGGCAGCATAGCAGT
9.	MCP-1	TCACCTGCTGCTACTCATTCACCA	TACAGCTTCTTTGGGACACCTGCT
10.	CXCL1	TGCACCCAAACCGAAGAAGTC	CAAGGGAGCTTCAGGGTGAAG
11.	CXCL3	TGAGACCATCCAGAGCTTGACG	CCTTGGGGGTTGAGGCAAACTT
12.	β-Actin	GGCTGTATTCCCCTCCACG	CCAGTTGGTAACAATGCCATGT

Histology and Immunofluorescence Staining

Normal and tumor colon tissue were Swiss-rolled and fixed in PBS-buffered formalin for a day. H&E analysis was performed in paraffin-embedded tissue sections (5 μm) as previously described (13, 19, 42). Histological analysis was done on H&E-stained paraffin sections and microscopically analyzed by a gastrointestinal pathologist for inflammation score. For Ki67 immunofluorescence staining to assess proliferation, paraffin-embedded tissue sections were subjected to antigen retrieval, followed by blocking with 5% goat serum and probed with primary antibody against Ki67 (1:200 dilution; D3B5; Cell Signaling Technology) overnight. Slide sections were then washed three times with PBST and were incubated with anti-rabbit IgG Alexa Fluor-488 (1:500 dilution) for 1 h (Thermo Fisher Scientific). Sections were then washed three times with PBST, and slides were mounted using the Prolong Gold Antifade Reagent with DAPI (Thermo Fisher Scientific).

Confocal Microscopy

Tissues were harvested and sectioned as mentioned in Histology and Immunofluorescence Staining. After deparaffinization and hydration, slides were incubated in citrate-based antigen unmasking solution (Vector Labs H-3300-250) at 95°C for 20 min. Slides were cooled to room temperature, washed three times with PBS, and blocked with 10% bovine serum albumin (BSA)/PBS for 30 min at room temperature. Slides were incubated with 1:200 Ly-6g antibody (BioLegend 127601, Clone 1A8) overnight at 4°C. After being washed three times with PBS, slides were incubated with 1:200 goat anti-rat Alexa Fluor 647 (1:200, Invitrogen A-21247) for 1 h at room temperature. Slides were washed three times with PBS and counterstained with 1:2,000 Hoechst 33342 (Thermo Scientific 62249) for 10 min at room temperature. After being washed, slides were mounted with ProLong Gold antifade mountant (Invitrogen P36930) and imaged the next day on a Leica SP8 confocal microscope.

Hematological Analysis

Approximately 300–500 μL of blood were collected from HIF-1α^ΔNeu or HIF-2α^ΔNeu mice and their respective littermates in the EDTA vacutainers. After 15 min, the serum was collected by centrifugation at 8,000 rpm for 10 min and submitted for hematological parameters quantification. Complete blood count analysis was performed by the Unit for Laboratory Animal Medicine Pathology Core at The University of Michigan.

Neutrophils and Macrophage Isolation

Neutrophils were purified and macrophages were differentiated from bone marrow. Bone marrow cells were isolated from femurs and tibia of HIF-1α^ΔNeu or HIF-2α^ΔNeu mice and their respective littermates and then neutrophils were purified using a two-layer density gradient method with 62.5% Ficoll (GE Healthcare) as a separation reagent made in 1X HBSS-Prep (HBSS + 0.5% FBS + 20 mM Na-HEPES (pH 7.4) + Pen/Strep). Cells were layered over the 62.5% Percoll and centrifuged for 30 min at 1,000 g without brakes. At the end of the gradient centrifugation, a sharp interface on top of the 62.5% Percoll (these are immature cells and nongranulocytic lineages) is separated from a cloudier pellet (the neutrophils). The neutrophil pellet was washed in HBSS prep, and cells were used for RNA extraction. Macrophages on the other hand were differentiated for 6 days from bone marrow cells using RPMI with 10% heat-inactivated FBS + 20 mM HEPES + and 1% penicillin-streptomycin solution supplemented with 20 ng/mL granulocyte-macrophage colony-stimulating factor (GM-CSF) (BioLegend). The media was changed after 3 days and supplemented with GM-CSF again.

ELISA

The protein lysates were prepared from ∼5 mg of neutrophil cell pellet or normal colon and tumor tissues using ∼300 μL of complete extraction buffer (100 mM Tris, pH 7.4, 150 mM NaCl, 1 mM EGTA, 1 mM EDTA, 1% Triton X-100, 0.5% sodium deoxycholate, supplemented with protease/phosphatase inhibitor cocktail and PMSF). Mouse TNF-α, IL-1β, IFN-γ, and IL-10 assays were performed using paired antibody ELISA kits (R&D Systems, Minneapolis, MN) as per the suggested protocol from the manufacturer.

Statistical Analysis

Sample description and classification were not provided to the core personnel during data extraction or analyses. No sample or data were excluded from the study for statistical purposes. Results are expressed as the means ± SE or means ± SD. Significance between two groups was tested using a two-tailed, unpaired t test. GraphPad Prism 9.0 was used to conduct statistical analyses. Statistical significance is described in the figure legends as P < 0.05, P < 0.01, P < 0.001, and P < 0.0001.

RESULTS

Generation and Basal Characterization of Mice with Deletion of HIF-1α or HIF-2α in Neutrophils (HIF-1α^ΔNeu) or (HIF-2α^ΔNeu)

To investigate the role of neutrophil-specific HIF-1α or HIF-2α on colon tumorigenesis, we generated neutrophil-specific HIF-1α- or HIF-2α-deficient mice model by crossing mice carrying the HIF-1α or HIF-2α-floxed allele (referred to as HIF-1α^F/F or HIF-2α^F/F) to MRP8-Cre (38, 43) transgene-expressing mice to generate MRP8-Cre+ HIF-1α^F/F or HIF-2α^F/F (referred to as HIF-1α^ΔNeu or HIF-2α^ΔNeu) mice with lineage-specific deletion of HIF-1α or HIF-2α specifically in the neutrophil compartment (Fig. 1A). Gene expression analysis confirmed the deficiency of HIF-1α or HIF-2α only in neutrophils (Fig. 1B) and not in other myeloid cells such as macrophages (Fig. 1C). The genetic disruption of HIF-1α or HIF-2α in neutrophils did not lead to a change in the number of neutrophils or the lymphocytes in HIF-1α^ΔNeu (Fig. 1D) or HIF-2α^ΔNeu mice (Fig. 1E) compared with their littermate controls. In addition, no differences were observed with respect to the red blood cells (RBC), hematocrits (HCT), hemoglobin (HGB), and mean corpuscular volume (MCV) in HIF-1α^ΔNeu (Fig. 1D) or HIF-2α^ΔNeu (Fig. 1E) mice. Moreover, deletion of neutrophil-specific HIF-1α or HIF-2α does not lead to basal histological changes. We found no major differences in the small intestine, colon, kidney, and liver in HIF-1α^ΔNeu mice (Fig. 1F) and HIF-2α^ΔNeu mice (Fig. 1G) in comparison with their respective littermates.

Figure 1.

Deletion of neutrophil hypoxia-inducible factor (HIF)-1α or HIF-2α. A: schematic of the mouse cross (HIF-1α^ΔNeu or HIF-2α^ΔNeu mice). Gene expression analysis from neutrophils and macrophages for HIF-1α in HIF-1α^ΔNeu mice (n = 3) (B) or HIF-2α in HIF-2α^ΔNeu mice (n = 4) (C) compared with littermate controls (n = 4). Quantifications of neutrophil, lymphocyte, red blood cell (RBC), hemoglobin (HGB), hematocrit (HCT), and mean corpuscular volume (MCV) from HIF-1α^ΔNeu (n = 4) (D) or HIF-2α^ΔNeu mice (n = 6) (E). Hematoxylin and eosin (H&E) staining in the duodenum, colon, kidney, and liver tissue of HIF-1α^ΔNeu (F) or HIF-2α^ΔNeu (G) mice. **P < 0.01, ***P < 0.001, ns, nonsignificant, using unpaired t test compared with littermate HIF-1α^F/F or HIF-2α^F/F mice. Error bars indicate standard deviations. n, independent animals.

Neutrophil-Specific HIF-2α but Not HIF-1α Regulates Tumorigenesis in Colitis-Associated Colon Cancer Model

Our prior research has demonstrated that the activation of epithelial HIF-2α can induce the growth of colon tumors through both cell-autonomous and cell-nonautonomous pathways (26). In addition, our work has underscored the role of epithelial HIF-2α in regulating the recruitment of neutrophils to colon tumors, an essential factor contributing to the establishment of a protumorigenic inflammatory microenvironment in colon cancer (33). These studies establish an important link between hypoxia, inflammation, and colon cancer. However, the precise function of neutrophil-specific HIFs in the progression of inflammation to cancer has not been assessed. To dissect the role of neutrophil-specific HIF-1α or HIF-2α in the inflammatory progression of colon cancer, we utilized the azoxymethane/dextran sulfate sodium (AOM)/DSS) model of colitis-associated cancer (CAC) in mice with neutrophil-specific deletion of HIF-1α (HIF-1α^ΔNeu) or HIF-2α (HIF-2α^ΔNeu) or littermate control mice (HIF-1α^F/F or (HIF-2α^F/F) (Fig. 2A). No difference in body weights were observed between HIF-1α^F/F and HIF-1α^ΔNeu mice subjected to cycles of DSS (Fig. 2B). Although we did not observe a significant difference in body weights between HIF-2α^F/F and HIF-2α^ΔNeu mice, a noticeable trend indicated that HIF-2α^ΔNeu mice exhibited a degree of protection against DSS, as evidenced by their improved body weight recovery compared with their littermates (Fig. 2C). No differences were observed in colon length between HIF-1α^F/F and HIF-1α^ΔNeu mice (Fig. 2D) or HIF-2α^F/F and HIF-2α^ΔNeu mice (Fig. 2E). Tumor number, volume, and size remain unchanged in HIF-1α^ΔNeu mice compared with their littermates (Fig. 2, F and H). In contrast, HIF-2α^ΔNeu mice had significantly reduced colon tumor numbers, tumor volume, and size compared with littermate controls, suggesting that HIF-2α expression in neutrophils is essential for inflammation-induced colon tumorigenesis (Fig. 2, G and I). This data demonstrate that neutrophil-HIF-2α but not neutrophil-HIF-1α is critical for the increase in colon tumor formation and progression.

Figure 2.

Neutrophil HIF-2α modulates inflammation-induced colon tumorigenesis. A: HIF-1α^ΔNeu (n = 6) or HIF-2α^ΔNeu mice (n = 6) and littermate controls (n = 6) were treated with AOM (10 mg/kg) for 5 days after which they were subjected to 3 cycles of DSS treatment. B and C: percent weight change colon length (D and E), tumor number, tumor volume, and tumor size (F and G), and gross images (H and I) of the colons were assessed. Statistical analysis was performed with Student’s unpaired t test. *P < 0.05; ns, not significant for the comparisons between HIF-1α^ΔNeu and HIF-1α^F/F or HIF-2α^ΔNeu and HIF-2α^F/F. Error bars indicate standard error of mean. n, independent animals. AOM, azoxymethane; DSS, dextran sulfate sodium; HIF, hypoxia-inducible factor.

Neutrophil HIF-2α Regulates Tumor Proliferation and Architecture in a CAC Model

Proliferation was assessed by Ki67 staining, revealing no significant differences in tumors between HIF-1α^F/F and HIF-1α^ΔNeu mice (Fig. 3, A and B). In contrast, the number of Ki67-positive proliferating cells in colon tumors of HIF-2α^ΔNeu was significantly lower than that in tumors of HIF-2α^F/F mice (Fig. 3, C and D). This suggests the role of neutrophil-HIF2α in increasing tumor cell proliferation, which may be important for the increase in colon tumor formation and progression. Subsequently, we examined tumor formation and macroscopic changes using H&E staining. All tumors assessed in the colon from HIF-1α^F/F and HIF-1α^ΔNeu and HIF-2α^F/F and HIF-2α^ΔNeu mice. We did not observe major histological changes at the time point we assessed. All tumors were mainly high-grade adenomas irrespective of genotype (Fig. 3, E and F). All together these data suggest that HIF-2α activation in neutrophils is indeed crucial for the growth of colon tumors. Since HIF-2α plays a critical role in neutrophil migration, particularly in tight spaces such as tumors (44), we wanted to see if tumor resolution in HIF-2α^ΔNeu mice is due to less migration of HIF-2α-deficient neutrophils. We address this by analyzing the distribution of neutrophils within tumors from the HIF-2α^ΔNeu mice compared with littermate controls. Surprisingly, the Ly6G (neutrophil-specific marker) staining as well as the neutrophils number did not show a significant difference in tumors from HIF-2α^ΔNeu mice in comparison with tumors from HIF-2α^F/F mice (Fig. 3, G and H), suggesting that neutrophil HIF-2α does not contribute to neutrophil infiltration into the tumors.

Figure 3.

Neutrophil HIF-2α increases tumor proliferation. Representative images of Ki67 staining (A and C) and quantification of the data (B and D) in tumors following AOM/DSS treatment in HIF-1α^ΔNeu (n = 6) or HIF-2α^ΔNeu mice (n = 6), respectively. E and F: representative hematoxylin and eosin (H&E) images. *P < 0.05; **P < 0.01; ns, not significant for comparisons between HIF-1α^ΔNeu and HIF-1α^F/F or HIF-2α^ΔNeu and HIF-2α^F/F. Green fluorescence: Ki67; Blue: DAPI. G and H: representative confocal microscopy images and quantification of Ly6G (yellow) and DAPI (blue) staining on tumor sections from HIF-2α^F/F(n = 6) and HIF-2α^ΔNeu mice (n = 6). Statistical analysis was performed with Student’s unpaired t test. Error bars indicate standard deviations. n, independent animals. AOM, azoxymethane; DSS, dextran sulfate sodium; HIF, hypoxia-inducible factor.

Neutrophil-Specific HIF-2α Is Not Essential for Established Tumor Growth or Acute Inflammatory Stress

Since our results indicate an important role of neutrophil-specific HIF-2α in inflammatory progression of colon cancer, we next wanted to see if neutrophil HIF-2α is also important for the growth of established tumors. We assessed this in a syngeneic tumor model using mouse CRC cell line MC38. HIF-2α^F/F and HIF-2α^ΔNeu mice were injected subcutaneously with MC38 colon carcinoma cells (Fig. 4A). Interestingly, we observed no differences in the growth or and size of the tumors between HIF-2α^F/F and HIF-2α^ΔNeu mice (Fig. 4B). In addition, no significant differences were observed in terms of tumor weight between HIF-2α^F/F and HIF-2α^ΔNeu mice (Fig. 4C). This indicates that neutrophil-derived HIF-2α does not play a significant role in influencing the growth of pre-established tumors. Furthermore, we treated HIF-2α^F/F and HIF-2α^ΔNeu mice with 3% DSS in their drinking water for 7 days followed by a switch to normal water for 4 days (Fig. 4D). Notably, there were no substantial differences in the body weight of HIF-2α^F/F and HIF-2α^ΔNeu mice (Fig. 4E). Moreover, the length of the colon exhibited no significant difference between HIF-2α^F/F and HIF-2α^ΔNeu mice (Fig. 4F). Histological analysis revealed tissue damage with loss of crypt architecture, and inflammatory cell infiltrations in the HIF-2α^F/F mice while interestingly HIF-2α^ΔNeu mice also showed severe crypt damage, erosion, and robust inflammation (Fig. 4, G and H). These findings collectively suggest that neutrophil-derived HIF-2α does not contribute to the development of colitis. Instead, its significance lies in the progression of inflammation-induced colon cancer.

Figure 4.

Neutrophil HIF-2α does not impact growth of established syngeneic tumors or DSS-induced colitis. A: schematic of the syngeneic MC38 tumor study. Gross tumor images (B) and endpoint weight of tumors (C) of MC38 implanted in HIF-2α^ΔNeu (n = 7) and littermate control mice (n = 7). D: schematic of DSS induced-acute colitis study. Body weight (E), colon length (F), representative H&E images (G), and histological scoring (H) after DSS treatment in HIF-2α^ΔNeu (n = 5) and littermate control mice (n = 5). ns, not significant for the comparisons between HIF-2α^ΔNeu and HIF-2α^F/F mice. Statistical analysis was performed with Student’s unpaired t test. Error bars indicate standard error of mean. n, independent animals. DSS, dextran sulfate sodium; HIF, hypoxia-inducible factor.

Neutrophil-Specific HIF-2α Promotes Colon Tumorigenesis by Regulating Inflammatory Cytokines Level

Our previous work has demonstrated the necessity of HIF-2α to activate several proinflammatory genes (19, 33, 45). However, it remains unclear whether neutrophil-specific HIF-2α drives the expression of inflammatory genes within neutrophils that contribute to colon tumorigenesis, tumor growth in established tumors, or inflammation following injury. To gain deeper insights, we assessed changes in the expression of inflammatory and immunosuppressive cytokines and chemokines in HIF-2α^ΔNeu and age-matched wild-type littermates (HIF-2α^F/F) in neutrophils, normal and tumor tissue following AOM/DSS treatment, syngeneic MC38 tumor, and acute DSS injury model. We observed decreased expression of proinflammatory cytokines such as TNF-α and IL-1β and increased expression of anti-inflammatory cytokine TGF-β in neutrophils from HIF-2α^ΔNeu mice compared with their control littermates (Fig 5A). In addition, immunosuppressive cytokines such as Arg1 and IL-10 showed significantly lower mRNA levels in neutrophils from HIF-2α^ΔNeu mice compared with their control littermates indicating the role of neutrophil HIF-2α in suppressing the immune response (Fig. 5A). Moreover, we found increased mRNA levels of migratory chemokine MCP-1 suggesting enhanced migration of HIF-2α-deficient protumorigenic neutrophils toward colon tumors in HIF-2α^ΔNeu mice compared with their control littermates (Fig. 5A). Interestingly, both normal colon tissue (Fig. 5B) and tumor tissue (Fig. 5C) from HIF-2α^ΔNeu mice exhibited reduced levels of TNF-α and IL-1β, well-known proinflammatory cytokines implicated in tumor progression. In syngeneic MC38 tumor tissue, no significant differences in inflammatory gene expression were observed between HIF-2α^F/F and HIF-2α^ΔNeu mice (Fig. 5D). This suggests that neutrophil-specific HIF-2α does not regulate tumor growth through alterations in inflammatory markers in established tumors. Furthermore, no significant differences in inflammatory gene expression were observed in colon tissue from acute DSS treatment from HIF-2α^F/F and HIF-2α^ΔNeu mice (Fig. 5E), indicating that neutrophil-specific HIF-2α is not essential for driving colitis. Collectively, these findings underscore the significance of neutrophil-specific HIF-2α in elevating proinflammatory cytokine levels, thereby promoting inflammation-induced colon cancer.

Figure 5.

Neutrophil hypoxia-inducible factor (HIF)-2α is essential in proinflammatory and immunosuppressive cytokines expression. Gene expression analysis using quantitative PCR in neutrophils (n = 4) (A), normal colon (n = 6) (B), colon tumors (n = 5) (C), syngeneic tumors (n = 5) (D), or colitic tissue (n = 5) (E) from HIF-2α^ΔNeu and littermate control mice. *P < 0.05; **P < 0.01; ***P < 0.001; ns, not significant for the comparisons between HIF-2α^ΔNeu and HIF-2α^F/F. Statistical analysis was performed with Student’s unpaired t test. Error bars indicate standard error of mean. n, independent animals.

We next confirmed gene expression of inflammatory cytokines using ELISAs. Neutrophils, normal or tumor colon tissue from the AOM-DSS model, syngeneic tumors, and colon tissue from acute DSS model were assessed. Interestingly, the proinflammatory cytokines TNF-α and IFN-γ and immunosuppressive cytokine IL-10 showed decreased levels in the neutrophils (Fig. 6A) from HIF-2α-deficient mice compared with controls. Both the normal colon tissue (Fig. 6B) and tumor tissue (Fig. 6C) from the HIF-2α^ΔNeu mice displayed reduced protein levels of TNF-α and IL-1β compared with their control littermates. Consistent with gene expression data, we did not observe any changes in cytokines from syngeneic MC38 tumor tissue or colon tissue from acute DSS treatment between HIF-2α^F/F and HIF-2α^ΔNeu mice. These results strongly suggest the role of neutrophil-specific HIF-2α in regulating inflammation-induced colon tumorigenesis.

Figure 6.

Loss of neutrophil hypoxia-inducible factor (HIF)-2α reduces protein levels of proinflammatory cytokines in inflammation-induced tumorigenesis mouse model. Protein levels of inflammatory cytokines measured through ELISA in neutrophils (n = 6) (A), normal colon (n = 6) (B), colon tumors (n = 5) (C), syngeneic tumors (n = 5) (D), or colitic tissue (n = 5) (E) from HIF-2α^ΔNeu and littermate control mice. *P < 0.05; **P < 0.01; ***P < 0.001; ns, not significant for the comparisons between HIF-2α^ΔNeu and HIF-2α^F/F. Statistical analysis was performed with Student’s unpaired t test. Error bars indicate standard error of mean. n, independent animals.

DISCUSSION

Neutrophils are a critical player in the tumor microenvironment. Evidence shows the dual role of neutrophils during tumor progression, illustrating that these cells can both stimulate tumor growth by suppressing immune response or reduce tumor development by contact-dependent T cell cytotoxicity (46, 47). The tumor promotion as well as antitumor roles of neutrophils are largely dependent on the tumor microenvironmental cues (48). Neutrophils promote tumor growth via different mechanisms, including the suppression of T cell cytotoxic response, the promotion of genetic instability, tumor cell proliferation, angiogenesis, and metastasis (49–51). Neutrophils can inhibit tumor growth by direct killing of tumor cells mediated through release of reactive oxygen species or reactive nitrogen species, recruitment of proinflammatory macrophages and activation of cytotoxic T cells (52–54). Neutrophils are a crucial element of tumor-promoting inflammation in many cancers, and neutrophil extracellular traps (NETs) that are associated with tissue damage have been observed in different cancers including gastric, intestinal, liver, and breast. Hypoxia has been linked with the production of NETs (55–59). Despite existing research, there is still a lack of understanding regarding the mechanisms through which neutrophils contribute to colon tumorigenesis. In addition, it remains uncertain whether neutrophil-specific HIF-1α or HIF-2α directly affects the growth of colon tumors. Our findings, derived from mouse models featuring inflammation-induced colon cancer and neutrophil-specific deletion of either HIF-1α or HIF-2α, suggest that the disruption of neutrophil HIF-2α, rather than HIF-1α, can effectively halt the growth and progression of colon tumor cells. HIF-2α has been shown to regulate neutrophil motility in tumors (44), our data suggest a nonsignificant role of neutrophil-specific HIF-2α in regulating their distribution within tumors. Neutrophils play a critical role in tumor microenvironment (60), and neutrophil HIF-2α might impact tumor metabolism, which could be favorable for tumor growth. Studies are required to dissect the impact of neutrophil-specific HIF-2α in regulating metabolic interactions in the tumor microenvironment.

Neutrophils isolated from HIF-2α^ΔNeu mice expressed decreased levels of proinflammatory and immunosuppressive mediators. Tumor and normal tissues isolated from these mice also expressed decreased IL-1β and TNFα. IL-1β and TNFα have been shown to be critical in the progression of sporadic and CAC. HIF-2α is directly linked to proinflammatory cytokines and chemokines. HIF-2α in epithelial cells has been shown to directly regulate TNF-α. In macrophages, HIF-2α was shown to suppress NLRP3 inhibiting IL-1β (19). This suggests that there is cell type-dependent regulation of proinflammatory mediators by HIF-2α. In our study, we found a direct correlation of HIF-2α loss in neutrophils with decreased expression of proinflammatory cytokines TNF-α and IL-1β in tumor tissue. The decrease in levels of inflammatory cytokines in tumor tissue could be responsible for reduced tumor growth in HIF-2α^ΔNeu mice. More work is needed to understand if the proinflammatory or immunosuppressive milieu driven by neutrophil HIF-2α is the major mechanism that leads to enhanced tumorigenesis.

Interestingly, unlike the inflammation-induced CRC model, our work shows no significant change in tumor growth and progression with loss of neutrophil HIF-2α in established MC38 colon carcinoma model. Moreover, constitutive HIF-2α activation in intestinal epithelial cells promotes spontaneous colitis and previous research shows the pathological role of HIF-2α in increasing sensitivity toward intestinal injury in acute models of DSS-colitis by enhancing levels of proinflammatory genes and decreasing barrier function (19, 24, 61–64). However, disruption of neutrophil HIF-2α does not alter susceptibility to acute intestinal injury models. Together, these data suggest that neutrophil HIF-2α does not modify CAC tumorigenesis by reducing intestinal injury. Furthermore, data from the syngeneic models may indicate that the colon microenvironment, characterized by high hypoxia and adjacent microbiota, plays a more significant role in the relevance of neutrophils, which could affect the importance of HIF-2α when compared with subcutaneously implanted tumors. In addition, the CAC model is relatively more chronic than the syngeneic or acute colitis model. Our previous data has suggested that temporal influences of the local microenvironment could alter neutrophil function (65). Future studies will be directed toward a better understanding of the detailed molecular mechanisms by which neutrophil-specific HIF-2α controls tumorigenesis in inflammation-induced colon cancer.

The overlapping as well as opposing roles of myeloid HIF-1α and HIF-2α in the tumor setting have been reported. The loss of macrophage HIF-1α has been shown to decrease tumor mass and to limit tumor progression in a mouse model of breast cancer (66). However, our data showed no significant effect on tumor proliferation and development with deletion of HIF-1α in neutrophils. The deficiency of HIF-1α has been associated with abolishing prolonged survival of neutrophils. HIF-1α is also associated with maintenance of intracellular ATP levels regulating neutrophil granule protease production, and bacterial killing (35–37, 67). Our results indicate no significant role of HIF-2α in regulating neutrophil inflammatory functions. It has been demonstrated that deficiency of HIF-2α in macrophages leads to reduced tumor volume and progression in a mouse model of CAC. The loss of HIF-2α in macrophages has been associated with defects in inflammatory macrophage migration and invasion into the tumors (68). HIF-2α in neutrophils has been shown to exhibit an immunomodulatory phenotype by regulating its survival, which can have direct effects on tissue inflammation and repair. It has also been established that loss of neutrophil HIF-2α increases the susceptibility toward apoptosis, which reduces the inflammation (37). Our study depicts a significant role of neutrophil-HIF-2α in the tumor microenvironment as constitutive loss of HIF-2α in neutrophils decreases proinflammatory and immunosuppressive cytokines, which controls cell growth and progression of colon tumors in a murine model of CAC. Currently, there is an on-target HIF-2α inhibitor in clinical trials for clear cell renal carcinoma (69, 70). Given its efficacy and the recognized significance of HIF-2α in tumor epithelium, as well as in several critical cell types (such as macrophages and neutrophils) within the tumor microenvironment of CRC, the pharmacological targeting of HIF-2α in CRC has the potential to be highly effective.

DATA AVAILABILITY

Data will be made available upon reasonable request.

GRANTS

This work was supported by the National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) grants (R01DK095201), National Cancer Institute (NCI) grants (R01CA245546 and R01CA148828), and the Center for Gastrointestinal Research grant (DK034933) (to Y.M.S.); University of Michigan, Postdoctoral Fellowship (032650) from American Physiological Society and Research Fellow Award (1003279) from Crohn’s and Colitis Foundation (to R.S.); Crohn’s and Colitis Foundation Research Fellow Award (623914) and the American Heart Association Postdoctoral Fellowship (19POST34380588) (to S.S.); NCI Predoctoral Fellowship (5F30CA257292-02) (to H.N.B.), NIDDK F30 predoctoral grant (F30DK131851) (to W.H.); and NIDDK diversity supplement grant (3R01DK095201-10S1) (to C.C.).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

R.S., N.K.K., and Y.M.S. conceived and designed research; R.S., N.K.K., S.S., W.H., H.N.B., M.O.E., and C.C. performed experiments; R.S., N.K.K., S.S., and W.H. analyzed data; R.S., N.K.K., S.S., W.H., H.N.B., and Y.M.S. interpreted results of experiments; R.S. and N.K.K. prepared figures; R.S. and Y.M.S. drafted manuscript; R.S., N.K.K., S.S., W.H., and Y.M.S. edited and revised manuscript; R.S. and Y.M.S. approved final version of manuscript.