Macrophage CCL7 promotes resistance to immunotherapy for colorectal cancer by regulating the infiltration of macrophages and CD8+ T cells

Abstract

Background

Immune checkpoint inhibitors (ICIs) have been proven to be one of the most promising and effective immunotherapies; however, their efficacy in colorectal cancer (CRC) remains significantly limited. Therefore, understanding the mechanism of resistance to ICIs therapy in CRC patients is of great significance for the development of new anti-tumor immunotherapy targets.

Methods

Ccl7 myeloid cell-specific knockout mice and MC38 tumor-bearing mouse models were established to investigate the role of Ccl7 during CRC progression. Proteomic analysis, RNA-seq, and flow cytometry analysis were used to determine the role of Ccl7 in the tumor immune microenvironment.

Results

Herein, we found that elevated CCL7⁺ tumor-associated macrophages (TAMs) in tumors correlated with tolerance to ICIs blockage therapy in patients with CRC. Deletion of CCL7 in myeloid cells resulted in reduced accumulation of immunosuppressive TAMs and increased infiltration of activated CD8⁺ T cells within the tumor. Mechanistically, CCL7 modulates peroxisome biogenesis and fatty acid oxidation, thereby promoting the immunosuppressive functions of TAMs via the PI3K-AKT-PEX3 signaling pathway. Furthermore, CCL7 inhibits the expression of chemokine CXCL10 by suppressing the AKT2-STAT1 signaling pathway, which reduces the infiltration of activated CD8⁺ T cells in the tumor. Blocking CCL7 delays CRC progression and enhances the therapeutic efficacy of PD-L1.

Conclusion

Our study highlights the novel role and regulatory mechanisms of CCL7⁺ TAMs in ICIs immunotherapy resistance, suggesting that CCL7 may serve as a potential combined therapeutic target for ICIs immunotherapy.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• CCL7 is a chemotactic factor that promotes the proliferation, migration, and invasion of CRC cells by activating the Janus kinase/signal transducer and activator of transcription-signal transducer and activator of transcription (JAK-STAT) and extracellular regulated protein kinases-Jun N-terminal kinase (ERK-JNK) signaling pathways. Nevertheless, the role of CCL7 in CRC progression remains largely unexplored.

WHAT THIS STUDY ADDS

• Elevated levels of CCL7⁺ tumor-associated macrophages (TAMs) in colorectal cancer (CRC) tissues correlate with tolerance to ICIs blockage therapy in CRC patients.

• CCL7 promotes peroxisome biogenesis and fatty acid oxidation in TAMs, thereby affecting their immunosuppressive functions through the PI3K-AKT-PEX3 signaling pathway.

• CCL7 inhibits CD8⁺ T cell infiltration by obstructing the AKT2-STAT1-CXCL10 signaling pathway.

• Blockade of CCL7 significantly enhanced the antitumor efficacy of anti-PD-L1 antibodies.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE, OR POLICY

• CCL7 is highly expressed by a distinct subpopulation of TAMs in CRC tissues and is associated with poor survival outcomes in CRC patients. Blocking CCL7 delayed CRC progression and enhanced the therapeutic efficacy of the immune checkpoint inhibitor PD-L1, suggesting that targeting CCL7 may represent a promising immunotherapy strategy for patients with CRC.

Introduction

Colorectal cancer (CRC) is the fourth most prevalent cancer globally and the third leading cause of cancer-related mortality.¹ Immune checkpoint inhibitors (ICIs), particularly PD-1/PD-L1 inhibitors, have demonstrated greater efficacy in patients with CRC characterized by high microsatellite instability or defective mismatch repair (MSI-H/dMMR). However, its efficacy in CRC remains notably limited, up to 50% of CRC patients with metastatic MSI-H/dMMR are resistant to immunotherapy, followed by disease progression and recurrence.²

A growing body of research indicates that the effectiveness of immunotherapy is significantly influenced by the tumor immune microenvironment (TIME) in CRC.³ The immunosuppressive characteristics of this microenvironment promote immune evasion in tumor cells, thereby facilitating the occurrence, progression, recurrence, and metastasis of CRC, which ultimately leads to treatment resistance.⁴ Therefore, to address this challenge, it is essential to elucidate the characteristics of the immunosuppressive microenvironment in CRC, as well as its potential mechanisms and emerging targeted therapies aimed at reversing immunosuppression.⁵ The TIME is a complex milieu consisting of tumor cells, immune cells, secretory factors, extracellular matrix components, and metabolic molecules.³ It is characterized by the infiltration of various immunosuppressive cells, especially tumor-associated macrophages (TAMs) and bone marrow-derived suppressor cells (MDSCs).⁶ These tumor-infiltrating myeloid cells have been reported to accelerate tumor growth, metastasis, and the development of resistance to treatment.⁷ For example, TAMs express immune checkpoint regulators and promote the growth and progression of multiple tumor types, including CRC, by secreting a variety of inflammatory cytokines, growth factors, and chemokines.⁸

C-C motif ligand 7 (CCL7), also known as monocyte chemoattractant protein-3, is a well-characterized chemokine that was initially identified in the supernatant of human osteosarcoma cells.⁹ The elevated expression of CCL7 recruits monocytes to the site of injury and mediates the local inflammatory response, which exacerbates the clinical symptoms associated with various inflammatory diseases.¹⁰ Furthermore, CCL7 has been reported to be highly expressed in multiple cancer types, including non-small cell lung cancer and CRC.¹¹ It is associated with more aggressive malignant phenotypes, such as cellular proliferation, epithelial-mesenchymal transition, invasion, and metastasis.¹² Some studies suggest that the overexpression of CCL7 in CRC cells can enhance tumor invasion and metastasis.^{13 14} CCL7 has been shown to bind to several receptors, including CCR1, CCR2, and CCR3, thereby activating downstream signaling pathways such as JAK-STAT and ERK-JNK, which promote the proliferation, migration, and invasion of CRC cells.^{15 16} To date, most research on CCL7 has focused on its role in tumor proliferation and migration. However, a recent study indicated that CCL7 secreted by MDSCs plays a significant role in the growth and metastasis of CRC cells.¹⁷ Nevertheless, the function of CCL7 within the TIME in CRC remains largely unexplored.

This study aimed to explore the role of tumor-infiltrating bone marrow cells in the resistance of CRC patients to immunotherapy. Our findings indicate that a higher proportion of CCL7⁺ macrophages in CRC tissues correlates with a lack of response to anti-PD-1/PD-L1 therapy. CCL7⁺ TAMs play a crucial role in regulating the accumulation and activity of both tumor-infiltrating macrophages and CD8⁺ T cells, thereby influencing the efficacy of ICI therapy. Additionally, we investigated whether the inhibition of CCL7 could suppress CRC tumor growth and evaluated its potential as a target for combination therapy with ICIs.

Methods

Human samples and databases

Human CRC and adjacent tissues were collected from Chongqing University Cancer Hospital in Chongqing, China, and the clinical characteristics of all patients are described in online supplemental table S1. Peripheral blood samples were obtained from healthy adult volunteers and patients with CRC at the same institution. All experiments were conducted in compliance with local, national, and international regulations and received approval from the ethics committees of Chongqing University Cancer Hospital, along with informed consent from the patients or their guardians (Ethics Number is CZLS2023114-A). Prior to enrollment, all patients provided written informed consent in accordance with the Declaration of Helsinki. RNA sequencing (RNA-seq) data for patients with CRC (PanCancer Atlas datasets; n=594) were retrieved from cBioPortal, and single-cell RNA-seq (scRNA-seq) data were obtained from the GEO database. After excluding samples that lacked information on AJCC stage or overall survival, a total of 440 samples collected prior to chemotherapy or targeted therapy were retained for subsequent analyses.

Cell lines and treatment

THP1, LLC, CT26, and MC38 cell lines were acquired from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China) in 2020. HT29 cells were acquired from Wuhan Pricella Biotechnology (Wuhan, China; CL-0118). The morphology and function of these cell lines were assessed routinely. All the cell lines tested negative for mycoplasma contamination. MC38 and CT26 cells expressing luciferase reporter gene (MC38-luc and CT26-luc) were generated in our laboratory. Bone marrow-derived macrophages (BMDMs) were obtained as previously described.⁸ The cells were cultured in dulbecco's modified eagle medium (DMEM) supplemented with high glucose, 10% Fetal Bovine Serum (FBS), and 100 U/mL penicillin/streptomycin. Mouse Ccl7 recombinant protein was procured from PeproTech (Cat No. 250–08); BEZ235 was obtained from Selleck (Cat No. S1009).

Animals and tumor models

Ccl7-Flox (Ccl7^f/f) and Lyz2^cre mice were purchased from the Shanghai Biomodel Organism Center, and all mice were on a C57BL/6 genetic background. We crossed Ccl7^f/f with Lyz2^cre mice to generate Ccl7 myeloid cell-conditional knockout (Ccl7 MKO) mice. Genotyping of these mice was performed in accordance with the protocols provided by the Shanghai Biomodel Organism Center. C57BL/6 and BALB/c mice aged 6–8 weeks were obtained from the Animal Institute of the Academy of Medical Sciences in Beijing, China. Both male and female mice were used, ensuring that all experiments employed age- and sex-matched littermate controls. All mice were housed in a controlled environment, free from specific pathogens, and subjected to a 12-hour light/dark cycle. They were fed a standard chow diet at Chongqing University Cancer Hospital. For animal experiments, 6–8-week-old male or female mice were randomly assigned to different treatment groups, with 5×10⁵ MC38-luc or CT26-luc cells being intraperitoneally implanted into Ccl7^f/f, Ccl7 MKO, or C57BL/6 mice.¹⁸ Tumor growth progression was monitored using the IVIS Lumina imaging system every 3–4 days, with the relative photon flux normalized to the intensity of bioluminescence measured at the initial establishment of the tumor model. In TAMs depletion experiments, 200 µL of clodronate liposomes (Apexbio, Cat No. K2721) was administered intraperitoneally 1 day prior to MC38 tumor inoculation and subsequently every 3 days. For the Bindarit treatment (5 mg/kg, Cat No. HY-B0498, MCE), Bindarit was administered by gavage every 3 days, commencing once the tumors became palpable. For in vivo treatment, Recombinant Mouse CCL7 (50 µg/kg, Cat No. 250–08, PeproTech), anti-Ccl7 antibody (100 µg, Cat No. AF-456-NA, R&D Systems), and anti-PD-L1 antibody (200 µg, Cat No. BE0101, BioXcell) were administered intraperitoneally 3 days after tumor inoculation. All mouse experiments adhered to national and international guidelines for the care and use of laboratory animals. The study was approved by the Animal Care and Use Committee (IACUC) of Chongqing University Cancer Hospital, in compliance with the Declaration of Helsinki and the Ethics Number is SYXK2021-001.

FCM

Single-cell suspension samples were isolated from tumor tissues and pre-incubated in a staining buffer Phosphate Buffered Saline (PBS) containing 2%FBS for a minimum of 20 min on ice. Subsequently, the cells were stained with the specified antibodies (1:100) for 30 min on ice. Dead cells were excluded using Fixable Viability Dye Efluor 780 (Cat No. 65-0865-14; eBioscience). The panel of antibodies utilized in these experiments included CD45 (Cat No. 147712), CD8α (Cat No. 100706), interferon (IFN)-γ (Cat No. 505810), granzyme B (GzmB; Cat No. 515403), CD3 (Cat No. 100206), CD4 (Cat No. 100412), CD11b (Cat No. 101208), Gr-1 (Cat No. 108426), CD11c (Cat No. 117308), MHC II (Cat No. 107608), F4/80 (Cat No. 123116), B220 (Cat No. 103206), PD-L1 (Cat No. 124308), NK1.1 (Cat No. 108718), CCR1 (Cat No. 152508), CCR2 (Cat No. 150608), and CCR3 (Cat No. 144512), all sourced from Biolegend (San Diego, CA). For intracellular staining of ARG1 (Cat No. 42284, GeneTex) and iNOS (Cat No. MA5-17139, Thermo), cells were first stained with surface markers, then fixed and permeabilized using the Foxp3/Transcription Factor Staining Kit (Cat No. 00-5523-00, eBioscience), followed by intracellular antibody staining. The proliferation and functional assays of CD8⁺ T cells were conducted as previously described.⁸ Cells were subsequently collected by trypsinization, washed twice with PBS, and re-suspended in 500 µL of PBS. The CFSE probe was obtained from Dojindo (Cat No. C309). Flow cytometry (FCM) was performed on BD FACS Canto II platforms, and the results were analyzed using FlowJo software version 10 (TreeStar).

Lipidomic analyses

2×10⁶ BMDMs from Ccl7^f/f, Ccl7 MKO, and C57BL/6 mice were treated with or without Ccl7 for 24 hours. The adherent cells were collected and lipidomic analysis was conducted as previously described.⁸

Seahorse analyses

2×10⁴ BMDMs were seeded in Seahorse XFe96 Spheroid Microplates. Cellular oxygen consumption rate (OCR) was measured using Agilent Seahorse XFe96 Analyzer according to manufacturer’s instructions.

Migration assay

Ccl7^f/f and Ccl7 MKO TAMs were cultured in RPMI 1640 supplemented with 2.5% FBS for 48 hours to collect the conditioned medium. Subsequently, 5×10⁴ CD8⁺ T cells, isolated from the spleens of C57BL/6 mice, were seeded into a 24-well Transwell Multiple Well Plate (Corning, Cat No. 07-200-149) that had been precoated with Matrigel (0.4 mg/mL; BD Biosciences). The conditioned medium from the mouse TAMs served as a chemoattractant in the bottom chamber of the 24-well plate. After a 24-hour incubation, the migrated T cells were fixed and counted under a microscope. For the in vivo CD8⁺ T cell migration assay, CD8⁺ T cells were labeled with CFSE (Dojindo, Cat No. C309). A total of 1×10⁶ CFSE-labeled CD8⁺ T cells were injected intravenously into Ccl7^f/f and Ccl7 MKO mice 15 days after MC38 inoculation. The mice were euthanized approximately 18 hours after CD8⁺ T cell transplantation, and a single-cell suspension of tumor tissues was prepared and stained with surface markers for further FCM analysis.

Transfection assays

Control siRNAs (siNC) and siRNAs targeting mouse Pex3, Ccr1, Ccr2, and Ccr3 were acquired from RiboBio (Guangzhou, China). Pex3 overexpression lentiviruses were obtained from GeneChem (Shanghai, China). Transfection assays for the siRNAs were conducted using Lipofectamine RNAiMAX (Thermofisher, Cat No. 13778150), following the manufacturer’s protocols. The efficiency of the siRNAs was assessed through qPCR or Western blot analysis.

ELISA

The expressions of Ccl7, Ccl6, Ccl17, Ccl12, and Cxcl10 in serum or cultured medium were quantified using mouse or human ELISA kits (Solarbio Life Science), following the manufacturer’s instructions.

Quantitative real-time PCR

Total RNA from the sample was extracted using RNAiso Plus (Cat No. 9108Q, Takara). The RNA concentration was quantified, and 1 µg of total RNA was subsequently converted into complementary DNA (cDNA) utilizing the PrimeScript RT-PCR Kit (Cat No. RR014A, Takara). Quantitative PCR (qPCR) was conducted with the TB Green Fast qPCR Mix Kit (Cat No. RR430A, Takara). Each qPCR experiment was performed in triplicate. All primer sequences were sourced from PrimerBank (https://pga.mgh.harvard.edu/primerbank/).

Immunofluorescence

Human or mouse CRC tumor samples were fixed in 4% formaldehyde for 15 min, after which the tissue sections were incubated in 10% normal goat serum for 1 hour. Subsequently, the sections were incubated with primary antibodies against CD68 (1:100; Cat No. ab53444, Abcam) and CCL7 (1:50; Cat No. MA5-29089, Invitrogen) and PMP70 (1:100, Cat No. sab4200181; Sigma-Aldrich) overnight at 4°C. Following this, secondary antibodies, Anti-rat IgG (H+L) Alexa Fluor 488 conjugate (Cat No. 4416, CST) and Anti-rabbit IgG (H+L) Alexa Fluor 647 (Cat No. 4414, CST), were applied at a dilution of 1:200 for 1 hour. DAPI was used to stain nuclei at a concentration of 100 ng/mL. Finally, the sections were imaged using a TCS SP5 confocal microscope (Leica Microsystems).

RNA-seq) library construction

Total RNA was extracted from TAMs isolated from MC38 tumor-bearing Ccl7^f/f and Ccl7 MKO mice. The RNA-seq library for these samples was constructed following the strand-specific RNA-seq library preparation protocol. The mRNA transcripts were enriched through two rounds of poly-(A⁺) selection using Dynabeads oligo-(dT) 25 (Invitrogen) prior to library construction. The prepared libraries were sequenced using an Illumina NovaSeq 6000 platform.

Proteomic assay analysis

TAMs (5×10⁶) isolated from MC38 tumor-bearing Ccl7^f/f and Ccl7 MKO mice were collected and transported to Novogene Co., Ltd. (Beijing, China) for proteomic analysis.

Western blot

Cells were lysed using radioimmunoprecipitation assay lysis buffer, followed by a 30-min incubation on ice. The lysates were then centrifuged at 13,000×g for 15 min at 4°C, and the supernatant was collected. Western blotting was performed as previously described.⁸ The primary antibodies used were CCL7 (1:1,000; Cat No. PA5-86885, Invitrogen), Pex3 (1:1,000; Cat No. 30 424–1-AP, Proteintech), Acox1 (1:1,000; Cat No. 10 957–1-AP, Proteintech), p-PI3K (1:1,000; Cat No. 4228, CST), PI3K (1:1,000; Cat No. 4249T, CST), p-AKT1 (1:1,000; Cat No. 2938, CST), AKT1 (1:1,000; Cat No. 9018, CST), AKT2 (1:1,000; Cat No. 3063, CST), Stat1 (1:1,000; Cat No. 9172, CST), and β-actin (1:1,000; Cat No. A1978, Sigma-Aldrich).

Statistical analysis

Statistical methods and sample sizes (n values) are detailed in the figure legends. All results were validated through a minimum of three independent experiments and are presented as the mean±SD. To compare two groups, the Mann-Whitney test or one- or two-way analysis of variance with Sidak’s multiple comparisons test was employed to assess statistical significance, utilizing GraphPad Prism software (V.8.0). For survival analysis, the Kaplan-Meier method was applied, and differences in survival curves were evaluated using the log-rank test. A p-value less than 0.05 is considered statistically significant.

Results

Elevated CCL7⁺ macrophages in CRC tissues are associated with poor treatment response to ICIs immunotherapy and patient survival

We hypothesized that tumor-infiltrating myeloid cells play a crucial role in the resistance of CRC patients to ICIs therapy. To verify this hypothesis, we analyzed a published scRNA-seq dataset of immune cells isolated from non-responsive and responsive patients with CRC after neoadjuvant ICIs treatment.¹⁹ Analysis of tumor-infiltrating myeloid populations revealed changes in subset composition between responsive and non-responsive CRC patients after neoadjuvant ICI treatment (figure 1A and B). To better understand the roles of these different myeloid populations in resistance to ICIs therapy, we examined differentially expressed genes in these myeloid cells between responsive and non-responsive patients. Interestingly, tumor-infiltrating myeloid cells in the ICIs non-response group expressed higher levels of chemokines, such as CXCL1/3/5, CCL7, and CCL20, compared with the ICIs response group (figure 1C and online supplemental table S2). We concentrated on the CCL7⁺ myeloid cell population due to the significant increase in CCL7 transcripts observed in tumors resistant to ICI therapy. Furthermore, CCL7 transcripts were found to be highly expressed in tumor-infiltrating myeloid cells. CCL7 is upregulated in CRC tissues (online supplemental figure S1A and B) and expressed in various cell types, including macrophages, dendritic cells (DCs), and fibroblasts (online supplemental figure S1C).²⁰ CCL7 expression was predominantly observed in a subset of macrophages in the CRC tissues (online supplemental figure S1D). By analyzing scRNA-seq data from an additional cohort of CRC patients, we found that CCL7⁺ macrophages were significantly upregulated in samples resistant to ICI therapy (online supplemental figure S1E). Additionally, the number of CCL7⁺ macrophages was significantly higher in human CRC tissues than that in the adjacent non-cancerous tissues (figure 1D). Moreover, the accumulation of CCL7⁺ macrophages marked increased during CRC progression, and a higher infiltration of CCL7⁺ macrophages was associated with a lower survival rate of CRC patients (figure 1E and F and online supplemental figure S1F). Furthermore, compared with healthy donors, patients with CRC exhibited elevated serum concentrations of CCL7 (figure 1G). We also noted an increase in the percentage of CCL7⁺ macrophages in the peripheral blood of CRC patients (figure 1H). Immunofluorescence staining further demonstrated an increased presence of CCL7⁺ macrophages within the CD68⁺ macrophages of tumor tissues from CRC patients (figure 1I). Consistently, the proportion of CCL7⁺ macrophages in the MC38 tumor tissue was found to be higher than that in the spleen (figure 1J). Importantly, serum CCL7 expression levels were positively correlated with MC38 tumor size (figure 1K). Overall, these data suggest that CCL7⁺ TAMs are increased in CRC tissues and their abundance correlates with resistance to ICIs therapy in CRC patients.

Figure 1. CCL7⁺ TAMs abundance in CRC tissues correlates with resistance to ICI therapy in CRC patients. (A) Unsupervised clustering of broad cell types in non-responsive and responsive CRC patients using the GSE205506 dataset. (B) Subset of myeloid cells in (A). (C) Differentially expressed genes in myeloid cells between the non-responsive and responsive CRC samples. (D) The proportion of CCL7⁺ TAMs in both tumor and adjacent human CRC tissues was quantified. (E) The frequency of CCL7⁺ TAMs in different CRC stages was examined. (F) Patient survival probabilities were stratified based on the expression levels of the CCL7⁺ TAM gene signature, normalized by CD68. (G) Serum CCL7 levels in healthy donors (HD) and CRC patients (n=5) were evaluated using ELISA. (H) FCM plots depict the presence of CCL7⁺ TAMs in the peripheral blood of CRC patients compared with HD (n=6). (I) Immunofluorescence images showing CCL7 expression in macrophages from human CRC tissues (T) and adjacent tissues (N), with red indicating CCL7, green indicating CD68, and blue indicating DAPI. Scale bars represent 5 µm. (J) The proportion of CCL7⁺ macrophages in the spleen (Spl) and MC38 tumor tissues was measured by FCM (n=5). (K) The correlation between serum Ccl7 protein levels and tumor volume in MC38 tumor-bearing mice (n=25) was assessed using Pearson’s correlation coefficient. Data are presented as mean±SD. Statistical significance was determined with *p<0.05, **p<0.01, and ***p<0.001 using the Mann-Whitney test for all comparisons. CRC, colorectal cancer; FCM, flow cytometry; ICI, immune checkpoint inhibitor; TAMs, tumor-associated macrophages.

CCL7 regulates the accumulation and activity of tumor-infiltrating macrophages during CRC progression

Since CCL7⁺ macrophages are associated with resistance to ICIs therapy and accumulate significantly during the progression of CRC, we investigated their role in tumor growth. To this end, we generated myeloid cell-specific Ccl7 knockout mice (Ccl7 MKO) (online supplemental figure S2A and B). These Ccl7 MKO mice exhibited smaller tumors than Ccl7^f/f mice in both the intraperitoneal and subcutaneous MC38 tumor models (figure 2A and online supplemental figure S2C). In addition, LLC tumors in Ccl7 MKO mice grew slower (online supplemental figure S2D), indicating that Ccl7 expressed by macrophages promoted tumor growth. Consistently, the serum Ccl7 level in Ccl7 MKO tumor-bearing mice was significantly lower than that in Ccl7^f/f tumor-bearing mice (figure 2B). To characterize the effects of Ccl7 on the TIME of tumor-bearing mice, we conducted FCM analysis of immune cells within the TIME. The infiltration of CD8⁺ T cells in the tumor tissues of Ccl7 MKO mice was significantly increased (figure 2C). Furthermore, the percentage of IFN-γ⁺ CD8⁺ T cells was elevated in tumors from Ccl7 MKO mice (figure 2D). In contrast, the frequency of tumor-infiltrating DCs, MDSCs, and macrophages decreased in Ccl7 MKO tumors (figure 2E–I). Notably, the percentage of iNOS⁺ macrophages (M1-like) increased, whereas the percentage of Arg1⁺ macrophages (M2-like) decreased in the tumors of Ccl7 MKO mice (figure 2J,K and online supplemental figure S2E). FCM analysis of typical TAM markers also indicated that CCL7⁺ TAMs from MC38 tumors expressed higher levels of the M2-like markers CD206 and Arg1, but lower levels of M1-like markers, such as reactive oxygen species (ROS), MHC II, iNOS, and CD86 (online supplemental figure S2F and G). Consistent with our observations in the mouse model, CRC patients exhibiting elevated CCL7 expression in tumor tissue tended to show an increased presence of M2-like macrophages and a decreased percentage of CD8⁺ T cells (figure 2L). These results indicated that CCL7 derived from TAMs plays a crucial role in regulating the accumulation and activity of tumor-infiltrating CD8⁺ T cells and macrophages, thereby influencing tumor growth.

Figure 2. CCL7 regulates the accumulation and activity of tumor-infiltrating CD8⁺ T cells and macrophages. (A, B) 5×10⁵ MC38-luc cells were intraperitoneally injected into Ccl7^f/f and Ccl7 MKO mice, and tumor growth was monitored every 4 days (A). Serum Ccl7 levels in both Ccl7^f/f and Ccl7 MKO mice were quantified using ELISA (B, n=5). (C-I) Ccl7^f/f and Ccl7 MKO mice were intraperitoneally injected with MC38 tumor cells and euthanized on day 14. The frequencies of various immune cell types, including T cells (C), IFNγ⁺ CD8⁺ T cells (D), B cells (E), MDSCs (F), DCs (G), NK cells (H), and macrophages (I) within tumor tissues were assessed by FCM, with n=5 for each group. (J) Expression levels of iNOS and Arg1 in macrophages within MC38 tumors, as determined by FCM (n=5). (K) The statistical analysis of immune cells in MC38 tumors from both Ccl7^f/f and Ccl7 MKO mice. (L) Immune cell infiltration in tumor tissues with high and low CCL7 expression in patients with colon adenocarcinomas (COAD). Data are expressed as mean±SD, with statistical significance indicated as *p<0.05, **p<0.01, ***p<0.001. The Mann-Whitney test was employed for all comparisons, and 'ns’ indicates no significant difference. FCM, flow cytometry; IFN, interferon; MDSC, bone marrow-derived suppressor cell.

CCL7 potentiates the immunosuppressive activity of TAMs in mouse CRC model

CCL7 is upregulated in TAMs of CRC tissues. Based on this observation, we hypothesized that the expression of CCL7 is influenced by tumor-derived factors. Notably, enhanced expression of CCL7 was observed in both M1 and M2a macrophages, with M2a macrophages exhibiting slightly higher levels of CCL7 than M1 macrophages (figure 3A and online supplemental figure S3A and B). Furthermore, CCL7 mRNA expression levels were elevated in M2 macrophages derived from CRC tumor tissues (figure 3B). Co-culture experiments involving BMDMs and MC38 cells showed significantly upregulated Ccl7 expression in BMDMs (online supplemental figure S3C). Additionally, stimulation of naïve CD14⁺ monocytes with the conditioned medium from HT29 cells led to a significant increase in the proportion of CCL7⁺ macrophages (figure 3C). Consistently, treatment with HT29-conditioned medium enhanced CCL7 concentration in the supernatant of macrophages (figure 3D), indicating that tumor cell-derived factors promote the accumulation of CCL7-producing macrophages. Moreover, recombinant Ccl7 treatment accelerated MC38 and CT26 tumor growth in vivo (figure 3E). Strikingly, we found that the tumor-promoting effects of Ccl7 were completely eliminated by macrophage depletion in MC38 tumor-bearing mice (figure 3F and online supplemental figure S3D).

Figure 3. CCL7 enhances the immunosuppressive activity of TAMs. (A) BMDMs were stimulated with 20 ng/mL LPS to induce M1 polarization or 20 ng/mL IL-4 to induce M2a polarization for 24 hours, after which CCL7 expression was analyzed using WB. (B) The gene expression levels of CCL7 in monocytes, M0, M1, and M2 macrophages from tumor tissues of COAD patients were evaluated using the GEPIA2021 database. (C, D) Naive CD14⁺ monocytes obtained from HD were treated with 10% conditioned medium from HT-29 cells for 24 hours; the frequency of CCL7⁺ macrophages was analyzed by FCM (C), and the concentration of CCL7 in the supernatants of macrophages was assessed (D), with n=5. (E) The 6-week-old mice were subcutaneously inoculated with MC38 or CT26 cells and treated with PBS (Veh) or recombinant Ccl7 every 3 days, starting on day 3 post-tumor injection and the tumor volume was measured, n=5. (F) Mice were subcutaneously injected with MC38 tumor cells and treated with recombinant Ccl7 starting on day three post-tumor injection, along with clodronate liposomes (CL) or control liposomes. The tumor volumes were measured (n=5). (G-I) BMDMs were stimulated with 1 µg/mL recombinant CCL7 for 48 hours, and the expression of iNOS, Arg1 (G), MHC II (H), and DCF-DA (I) was determined using FCM (n=5). (J-L) CD8⁺ T lymphocytes were co-cultured with TAMs derived from CCL7^f/f and Ccl7 MKO mice in a 2:1 ratio; the frequencies of IFN-γ⁺ (J) and GzmB⁺ T cells (K) were determined using FCM. The proliferation of CD8⁺ T cells induced by anti-CD3 and anti-CD28 was measured by FCM (L), with n=5. Data are presented as mean±SD. Statistical significance was assessed using the Mann-Whitney test, with *p<0.05, **p<0.01, ***p<0.001, and ns indicating no significance. FCM, flow cytometry; HD, healthy donors; IFN, interferon; TAMs, tumor-associated macrophages.

We subsequently investigated whether CCL7 regulates the immunosuppressive activity of TAMs. As anticipated, treatment with recombinant Ccl7 resulted in an increase in Arg1 expression while simultaneously impairing the expression of iNOS and MHC II in BMDMs (figure 3G,H). Specifically, Ccl7 treatment enhanced the expression of Arg1 and Mrc1 in macrophages, whereas it inhibited the mRNA expression of Nos2 and Tnfa (online supplemental figure S3E). Additionally, CCL7-treated BMDMs exhibited reduced ROS levels (figure 3I). To further characterize the immunosuppressive activity of CCL7 on TAMs, we co-cultured CCL7-deficient TAMs with CD8⁺ T cells. Compared with the control group, CCL7-deficient TAMs significantly increased the proportion of CD8⁺ T cells producing IFN-γ and GzmB (figure 3J,K). Furthermore, CCL7-deficient TAMs showed decreased suppressive function of co-cultured CD8⁺ T cells (figure 3L), indicating that deletion of CCL7 compromises the immunosuppressive activity of TAMs. Collectively, these findings demonstrate that tumor cell-derived factors promote the accumulation of CCL7-producing TAMs within the TIME, and that CCL7 plays a crucial role in enhancing the immunosuppressive activity of these macrophages.

CCL7-mediated lipid metabolism reprogramming enhances immunosuppressive function of TAMs

Reprogramming of lipid metabolism is essential for maintaining the immunosuppressive functions of tumor-infiltrating TAMs.²¹ Consequently, we hypothesized that lipid metabolism may play a role in the immunosuppressive function of CCL7⁺ TAMs. Gene set enrichment analysis (GSEA) indicated that CCL7⁺ TAMs were enriched in fatty acid metabolic processes and PPAR signaling pathways (figure 4A and online supplemental figure S4A and B). Moreover, scRNA-seq analysis of differentially expressed genes associated with fatty acid metabolic processes revealed a significant upregulation of ACADVL, HADH, and PPARG in CCL7⁺ TAMs (figure 4B). FCM and immunofluorescence analyses demonstrated a marked reduction in fatty acid accumulation in CCL7^- TAMs (figure 4C,D). Similarly, the number of intracellular lipid droplets in macrophages was significantly increased following Ccl7 treatment (figure 4E). Lipidomic analysis further confirmed that CCL7-deficient TAMs exhibited decreased free fatty acid (FFA) content, with elevated FFA levels after Ccl7 treatment in BMDMs (figure 4F and online supplemental figure S4C).

Figure 4. The role of CCL7 in regulating fatty acid metabolism of TAMs. (A) GSEA was performed on genes ranked by the fold change between CCL7⁺ and CCL7^- TAM clusters, with the normalized enrichment score (NES) reported. (B) A dot plot displaying representative genes associated with fatty acid metabolism in both CCL7⁺ and CCL7^- TAMs. (C) Bodipy staining of CCL7⁺ and CCL7^- TAMs in MC38 tumor tissues. (D) Bodipy staining of BMDMs from Ccl7^f/f and Ccl7 MKO mice was evaluated using FCM with a scale bar of 50 µm. (E) Bodipy staining of BMDMs following 48 hours of treatment with 1 µg/mL recombinant Ccl7 is shown, with n=5. (F) BMDMs were treated with Ccl7 for 48 hours, and fatty acid intermediates were quantified using Liquid Chromatograph Mass Spectrometer (LC-MS), with n=4. (G) The number of mitochondria in Ccl7^f/f and Ccl7 MKO TAMs derived from MC38 tumors was assessed via mitogreen staining, with n=5. (H) Mitogreen staining of BMDMs after 48 hours of Ccl7 treatment is presented, with n=5. (I) BMDMs were stimulated with recombinant Ccl7 for 24 hours, and the OCR and SRC of control and Ccl7-treated BMDMs were measured using a Seahorse XFe 96 analyzer; n=4. (J) The OCR and SRC of BMDMs from Ccl7^f/f and Ccl7 MKO mice were measured, with n=4. (K) BMDMs were treated with or without IL-4 for 24 hours, and ATP production in Ccl7^f/f and Ccl7 MKO macrophages was determined, with n=4. (L) BMDMs were treated with Veh (0.1% DMSO) or ETO (200 µM) for 24 hours, and the expression of iNOS and Arg1 was analyzed via FCM; n=5. Data are presented as mean±SD. Statistical significance was assessed using the Mann-Whitney test, with *p<0.05, **p<0.01, and ***p<0.001. BMDM, bone marrow-derived macrophages; ETO, etomoxir; GSEA, gene set enrichment analysis; OCR, oxygen consumption rate; SRC, spare respiratory capacity; TAMs, tumor-associated macrophages; DMSO, dimethylsulfoxide.

Mitochondria play a crucial role in lipid metabolism,²¹ prompting us to investigate the impact of CCL7 on the number of mitochondria in macrophages. FCM analyses revealed that CCL7-deficient TAMs exhibited a reduced number of mitochondria compared with KCs, whereas CCL7-treated macrophages demonstrated an increased mitochondrial count relative to the control group (figure 4G,H). Consistent with these findings, CCL7 treatment enhanced the mitochondrial OCR and spare respiratory capacity (SRC) in macrophages. Notably, CCL7-deficient macrophages displayed lower OCR and SRC than Ccl7^f/f macrophages, indicating a reduction in fatty acid oxidation (FAO) in the absence of CCL7 (figure 4I,J). Furthermore, a decrease in ATP production was observed in CCL7-deficient macrophages (figure 4K). Moreover, treatment with the CPT1 inhibitor etomoxir (ETO) resulted in a suppression of Arg1 expression and an enhancement of iNOS expression in BMDMs (figure 4L), suggesting that lipid metabolism plays a significant role in the immunosuppressive function of TAMs. Collectively, these results suggested that CCL7 regulates the immunosuppressive function of TAMs by upregulating lipid metabolism.

CCL7 promotes peroxisome biogenesis and FAO in TAMs through PEX3

To elucidate the molecular mechanisms by which CCL7 regulates fatty acid metabolism, we conducted proteomic analysis. Our investigation identified 523 upregulated and 409 downregulated proteins in the Ccl7^f/f TAMs group (figure 5A and online supplemental figure S5A). Notably, our findings revealed a significant upregulation of both peroxisomes and lysosomes in the Ccl7^f/f group (figure 5B and online supplemental figure S5B). Emerging evidence underscores the critical role of peroxisomes in lipid metabolism and ROS production.²² Consequently, we hypothesized that CCL7 promotes fatty acid metabolism and ROS elimination in TAMs, which is dependent on peroxisomes. The proteomic data and qPCR results indicated the upregulation or downregulation of peroxisome-related genes, with the most pronounced change observed in Pex3 (figure 5C,D). Furthermore, Western blot analysis demonstrated a significant reduction in Pex3 expression in the Ccl7 MKO group, whereas Pex3 levels increased following Ccl7 treatment (figure 5E and online supplemental figure S5C). Pex3 has been identified as a critical player in peroxisome biogenesis.²² We noted that the peroxisome marker PMP70 showed a decreasing trend in Ccl7 MKO TAMs, while PMP70 expression was significantly increased in macrophages after Ccl7 treatment (figure 5F,G). The overexpression of Pex3 led to an increase in PMP70 expression, FAO, and ROS elimination in Ccl7 MKO TAMs (online supplemental figure S5D-F). Additionally, Pex3 overexpression resulted in elevated Arg1 levels and reduced iNOS levels in Ccl7 MKO TAMs (online supplemental figure S5G), indicating that Pex3 is involved in both FAO and ROS elimination in TAMs. Importantly, the silencing of Pex3 using siRNA decreased FAO in macrophages (figure 5H). Moreover, Pex3 inhibition led to increased ROS accumulation in macrophages (figure 5I and online supplemental figure S5H). Knockdown of Pex3 significantly elevated iNOS expression while concurrently reducing Arg1 expression in macrophages (figure 5J and online supplemental figure S5I). Additionally, a significant increase in the proportion of anti-CD3- and anti-CD28-induced CD8⁺ T cell proliferation and IFNγ-producing CD8⁺ T cells was observed in siPex3 macrophages (figure 5K,L and online supplemental figure S5J). These findings provide compelling evidence that CCL7 enhances peroxisome biogenesis and FAO in TAMs, thereby supporting its immunosuppressive function through PEX3.

Figure 5. CCL7 activates PEX3 to regulate fatty acid metabolism. (A-C) Volcano plot (A) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis (B) of differentially expressed proteins in Ccl7^f/f and Ccl7 MKO TAMs derived from MC38 tumors, along with a heatmap of peroxisome-related differentially expressed proteins (C), with n=3. (D) The expression levels of peroxisome-related genes in TAMs from Ccl7^f/f and Ccl7 MKO tumor tissues (n=3). (E) The expression levels of Pex3 in BMDMs from Ccl7^f/f and Ccl7 MKO mice were assessed using western blot analysis. (F) Representative images of PMP70 staining (red) in TAMs from Ccl7^f/f and Ccl7 MKO tumor tissues, with a scale bar of 5 µm. (G) PMP70 expression in BMDMs from both Ccl7^f/f and Ccl7 MKO mice, as well as BMDMs stimulated with recombinant Ccl7 for 24 hours. (H) BMDMs were transfected with siRNA targeting Pex3, and OCR and SRC were measured (n=5). (I) BMDMs transfected with siPex3 and subsequently stimulated with Ccl7 for 24 hours exhibited ROS production, which was determined by FCM (n=4). (J) iNOS and Arg1 expression in BMDMs transfected with siPex3 was assessed by FCM (n=4). (K, L) CD8⁺ T lymphocytes were co-cultured with either siNC or siPex3 TAMs at a 2:1 ratio; the proliferation of CD8⁺ T cells induced by anti-CD3 and anti-CD28 was measured by FCM (K), and the production of IFN-γ in CD8⁺ T cells was measured by FCM (L; n=4). (M) Intracellular staining of p-Akt1 in TAMs from Ccl7^f/f or Ccl7 MKO tumor-bearing mice (n=5). (N) BMDMs were pretreated with or without BEZ235 (200 nM). and subsequently stimulated with Ccl7 for 24 hours, Arg1 expression was determined by FCM, n=4. (O, P) BMDMs were treated with or without BEZ235 (200 nM), and the expression of Pex3 and Acox1 was quantified using western blot analysis (O). (P) PMP70 expression in BMDMs was assessed (P). (Q, R) BMDMs were transfected with siRNA targeting Ccr1, Ccr2, and Ccr3, and the expression of Arg1 (Q) and iNOS (R) was analyzed using FCM. Data are presented as mean±SD. Statistical significance was determined using the Mann-Whitney test, with *p<0.05, **p<0.01, ***p<0.001, and ns indicating no significant difference. BMDM, bone marrow-derived macrophages; FCM, flow cytometry; IFN, interferon; TAMs, tumor-associated macrophages; KEGG, Kyoto Encyclopedia of Genes and Genomes; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.

To investigate the potential mechanism by which CCL7 enhances PEX3 expression in TAMs, we conducted RNA-seq analysis of tumor-infiltrating TAMs. Ccl7^f/f TAMs displayed differential expression of 498 downregulated and 903 upregulated genes compared with Ccl7 MKO TAMs (online supplemental figure S6A). These differentially expressed genes were enriched in cytokine-cytokine receptor interaction, the PI3K-Akt signaling pathway, and cell cycle (online supplemental figure S6B). Given that the PI3K-Akt signaling pathway was the highest in our gene enrichment analysis, we selected it for further investigation. We identified 33 differentially expressed genes within the PI3K-Akt signaling pathway, including Itga3, Irs1, and Ccnd1, in CCL7 knockout TAMs (online supplemental figure S6C). FCM analysis confirmed that the expression levels of p-Akt were significantly reduced in CCL7 knockout TAMs (figure 5M and online supplemental figure S6D). To further validate the regulatory role of CCL7 in the PI3K-Akt signaling pathway, we randomly selected 10 genes from this pathway for qPCR validation. Consistent with the RNA-seq findings, the mRNA levels of these 10 genes were either upregulated or downregulated (online supplemental figure S6E). Additionally, western blot analysis revealed that the protein levels of p-PI3K and p-Akt were diminished following CCL7 knockout in TAMs (online supplemental figure S6F). Importantly, treatment with recombinant CCL7 significantly elevated p-PI3K and p-AKT levels in BMDMs (online supplemental figure S6G and H). To investigate whether CCL7 promotes PEX3 expression through the activation of the PI3K-Akt pathway, we treated macrophages with the PI3K inhibitor BEZ235. BEZ235 treatment significantly inhibited CCL7’s capacity to upregulate the expression of immunosuppressive markers Arg1 and Mrc1 in macrophages (figure 5N and online supplemental figure S6I and J). In addition, Pex3 and Acox1 expression levels decreased after BEZ235 treatment (figure 5O). Consistently, BEZ235 suppressed the expression of PMP70, suggesting a reduction in the number of peroxisomes in macrophages (figure 5P). These findings indicated that CCL7 mediates the increase in PEX3 and peroxisomes by activating the PI3K-AKT signaling pathway in TAMs.

To elucidate the molecular mechanism by which CCL7 influences the PI3K-AKT signaling pathway, we analyzed the expression of CCL7 receptors, specifically CCR1, CCR2, and CCR3, across various cell types within the CRC TIME. Our findings indicated that CCR1 was broadly expressed in multiple cell types, whereas CCR3 was predominantly expressed in CD4⁺ T cells (online supplemental figure S7A). In contrast, CCR2 expression was confined to myeloid cells within the CRC TME, exhibiting high levels in macrophages and monocytes (online supplemental figure S7A and B). Notably, all three chemokine receptors were upregulated in TAMs compared with splenic macrophages (online supplemental figure S7C). To determine which receptor mediates CCL7-induced immunosuppressive activity and PEX3 expression in TAMs, we transfected BMDMs with siRNAs targeting Ccr1, Ccr2, and Ccr3 (online supplemental figure S7D). Our results revealed that knockdown of Ccr2 and Ccr3 significantly reduced the expression of M2-associated markers Arg1 in CCL7-induced BMDMs (figure 5Q and online supplemental figure S7E). Furthermore, silencing of Ccr2 and Ccr3 resulted in an increase in M1 markers, including iNOS, TNFα, and IL-1β, in macrophages (figure 5R and online supplemental figure S7F and G). But our data suggest that reduced MDSCs play a more predominant role than M1-like macrophages. Since CCR3 is mainly expressed by CD4⁺ T cells in CRC TIME, and our data indicate that CCR2 plays a more significant role in the immunosuppressive activity of macrophages induced by Ccl7, we chose CCR2 for further study. We evaluated the effect of Ccr2 knockdown on the PI3K-AKT signaling pathway and Pex3 expression levels. Western blot analysis revealed that Ccr2 knockdown resulted in decreased expression levels of p-PI3K, p-Akt1, and Pex3 in BMDMs (online supplemental figure S7H). Consistent with these findings, Ccr2 knockdown inhibited the immunosuppressive activity of TAMs (online supplemental file 1). Collectively, these results suggest that CCL7 promotes PEX3 expression and the immunosuppressive activity of TAMs through the PI3K-AKT-CCR2 signaling pathway.

Inhibition of the AKT2-STAT1-CXCL10 pathway by CCL7-CCR2 interaction for impeding T cell recruitment in the TIME

CCL7 is known to promote the migration of tumor cells;¹³ we further investigated the effect of Ccl7 on immune cell chemotaxis within the TIME. Compared with the Ccl7^f/f group, TAMs derived from Ccl7 MKO mice exhibited decreased expression levels of chemokines Ccl7, Ccl6, Ccl17, and Ccl12 (figure 6A,B and online supplemental figure S8A). We measured the concentrations of these chemokines in the supernatants of TAMs from both Ccl7^f/f and Ccl7 MKO mice using ELISA. As expected, the loss of Ccl7 resulted in reduced expression of these chemokines in the TAMs’ supernatant. Notably, we also observed upregulation of Cxcl10 in TAMs from Ccl7 MKO mice (figure 6A). CXCL10 is a chemokine that plays a critical role in the recruitment of cytotoxic CD8⁺ T cells into tumors.²³ We hypothesized that the increased priming of CD8⁺ T cells in Ccl7 MKO mice could be attributed to the enhanced secretion of Cxcl10 by TAMs. To test this hypothesis, we assessed serum Cxcl10 concentrations in MC38 tumor-bearing mice. As anticipated, serum Cxcl10 levels were significantly elevated in Ccl7 MKO mice (figure 6C). In contrast, the levels of other chemokines important for T cell migration, such as Ccl5 and Cxcl9, showed no significant differences between TAMs from Ccl7^f/f and Ccl7 MKO tumor-bearing mice (online supplemental figure S8B). Furthermore, the invasion of CD8⁺ T-cells into the conditioned media from Ccl7 MKO TAMs was significantly enhanced compared with that in the control group (figure 6D and online supplemental figure S8C). To further elucidate the effect of Ccl7 MKO TAMs on CD8⁺ T cell recruitment, we conducted an in vivo invasion assay by transferring CD8⁺ T cells into MC38 tumor-bearing recipients. Ccl7 knockout resulted in increased migration of CD8⁺ T cells to the tumor (figure 6E), indicating that the absence of Ccl7 in TAMs promotes CD8⁺ T cell migration through the upregulation of Cxcl10.

Figure 6. CCL7 inhibits the infiltration of CD8⁺ T cells by inactivating the AKT2-STAT1-CXCL10 signaling pathway. (A) Differentially expressed genes involved in the cytokine-cytokine receptor interaction pathway in TAMs derived from either Ccl7^f/f or Ccl7 MKO tumors. (B) The concentrations of Ccl7, Ccl6, Ccl17, and Ccl12 in the supernatant of TAMs from Ccl7^f/f and Ccl7 MKO mice are presented. (C) Cxcl10 expression levels in the serum of Ccl7^f/f and Ccl7 MKO MC38 tumor-bearing mice are depicted. (D) The invasion of CD8⁺ T cells toward the conditioned medium from Ccl7^f/f and Ccl7 MKO TAMs is quantified, with n=5. (E) CFSE⁺ labeled CD8⁺ T cells were transferred into MC38 tumor-bearing mice, which were subsequently euthanized approximately 18 hours post-transplantation; and the number of CFSE⁺ CD8⁺ T cells in the tumor was measured, with n=5. (F) Differentially expressed genes of the JAK-STAT pathway in TAMs from either Ccl7^f/f or Ccl7 MKO tumors. (G) The expression levels of Stat1 in TAMs from Ccl7^f/f or Ccl7 MKO tumors. (H) BMDMs were treated with recombinant Ccl7, and Stat1 expression was quantified using FCM analysis. (I) Akt2 expression levels in Ccl7^f/f and Ccl7 MKO TAMs are shown. (J) BMDMs were transfected with Ccr2 siRNA and the subsequent expression of Stat1 in BMDMs was measured. Data are presented as mean±SD. Statistical significance was determined using the Mann-Whitney test for all comparisons, with *p<0.05, **p<0.01, and ***p<0.001. BMDM, bone marrow-derived macrophages; FCM, flow cytometry; TAMs, tumor-associated macrophages.

Previous studies have demonstrated that IFN-γ induces the expression of CXCL10 through activation of the JAK/STAT1 signaling pathway.²⁴ In line with these findings, Ccl7 MKO TAMs exhibited significant upregulation of Stat1 compared with the Ccl7^f/f group (figure 6F,G). Notably, treatment with recombinant Ccl7 suppressed Stat1 expression in BMDMs, whereas Stat1 expression was significantly upregulated following Ccr2 knockdown (figure 6H,I). It has been previously reported that AKT2 is upregulated in M1 macrophages and plays a critical role in the regulation of STAT1 and NF-κB signaling.^{25 26} Our results indicate that Akt2 expression was significantly enhanced in Ccl7 MKO TAMs (figure 6J), and it was also notably upregulated after Ccr2 knockdown (online supplemental figure S8E). These findings suggest that CCL7 interacts with CCR2, inhibiting the AKT2-STAT1-CXCL10 pathway and impeding CD8⁺ T-cell recruitment into the tumor.

CCL7 blockage delays CRC progression and enhances anti-tumor immune responses of anti-PD-L1 antibodies

We further investigated whether CCL7 blockade inhibited CRC tumor growth. Treatment with Ccl7 neutralizing antibodies or the Ccl7 inhibitor bindarit effectively suppressed MC38 and CT26 tumor growth and enhanced the survival of tumor-bearing mice (figure 7A,B and online supplemental figure S9A). Additionally, both anti-Ccl7 antibody and Ccl7 inhibitor treatments led to a reduction in tumor-infiltrating TAMs (figure 7C and online supplemental figure S9B). FCM analysis revealed that treatment with the Ccl7 inhibitor or anti-Ccl7 antibody increased the proportion of M1-like TAMs in MC38 tumors, whereas the percentage of M2-like macrophages decreased (figure 7D,E and online supplemental figure S9C and D). Notably, administration of bindarit resulted in a reduction of Ccl7⁺ TAMs within the tumor tissues (online supplemental figure S9E). Furthermore, serum levels of Ccl7 were significantly decreased following bindarit treatment in mice bearing MC38 tumors (online supplemental figure S9F). TAMs from MC38 tumors treated with anti-Ccl7 antibodies exhibited elevated Stat1 expression compared with those from control tumors (figure 7F). Mice bearing tumors that received bindarit or anti-Ccl7 antibodies demonstrated a significantly higher proportion of IFN-γ⁺ and TNFα⁺ CD8⁺ T cells (figure 7G and online supplemental figure S9G and H). Collectively, these findings suggest that Ccl7 blockade in vivo not only delays tumor progression but also remodels the TIME.

Figure 7. Targeting CCL7 enhances the efficacy of immunotherapy with anti-PD-L1 antibody. (A-G) MC38-luc and CT26-luc cells were intraperitoneally injected into mice, which subsequently received 100 µg of anti-Ccl7 antibody (α-Ccl7) or Veh (PBS) every 3 days, starting on day 7 post-tumor injection. Tumor growth was monitored (A), and survival of tumor-bearing mice was assessed (B, n=9–12). The percentage of TAMs was determined (C), along with the expression of iNOS and Arg1 in macrophages (D), the presence of MHC II⁺ TAMs (E), Stat1 expression in TAMs (F), and the frequency of IFNγ⁺ and TNF-α⁺ CD8⁺ T cells in tumors (G), with n=5 for these analyses. (H-L) C57 and BALB/c mice were intraperitoneally injected with MC38-luc and CT26-luc cells, respectively. Tumor-bearing mice were injected with Veh, α-Ccl7, anti-PD-L1 (αPD-L1), or a combination of α-Ccl7 and anti-PD-L1 (α-C+α-P). Tumor growth was monitored (H), survival of tumor-bearing mice was analyzed (I, n=10–14), and the percentages of TAMs (J), CD8⁺ T cells (K), and IFNγ⁺ CD8⁺ T cells (L) in the tumor tissues were evaluated using FCM (n=5). (M) Correlation between CCL7 and CD163, MRC1, and PEX3 in human CRC tissues. Data are presented as mean±SD. Statistical significance was determined using the Mann-Whitney test (A, C-G), log-rank (Mantel-Cox) test (B and I), and one-way or two-way analysis of variance with Sidak multiple comparisons test (H, J-L), with significance levels indicated as *p<0.05, **p<0.01, ***p<0.001. CRC, colorectal cancer; IFN, interferon; TAMs, tumor-associated macrophages.

Given that CCL7 blockade significantly increases the presence of IFN-γ⁺ CD8⁺ T cells in tumor tissues and that IFN-γ signaling enhances PD-L1 expression in TAMs,²⁷ we investigated whether CCL7 neutralizing antibodies could enhance the immunotherapy effect of PD-L1 neutralizing antibodies. The combination of anti-CCL7 and anti-PD-L1 antibodies resulted in a more pronounced reduction in tumor volume in tumor-bearing mice, along with a substantial increase in mouse survival (figure 7H,I). Additionally, the proportion of TAMs within the tumor was further diminished, and CD8⁺ T cell activation was notably enhanced (figure 7J–L). These findings suggest that the combination of checkpoint blockade and CCL7 inhibition may yield a more effective anti-tumor response in vivo. Bioinformatics analysis revealed a positive correlation between CCL7 and CD163, MRC1, and PEX3 in CRC patients (figure 7M). These results further substantiate our hypothesis that CCL7 is integral to the regulation of tumor-infiltrating CD8⁺ T cells and macrophages, thereby affecting tumor progression and ICIs treatment resistance.

Discussion

Tumor immunotherapy utilizing ICIs has emerged as a promising strategy for tumor eradication. However, certain CRC patients exhibiting MSI-H/dMMR do not respond effectively to anti-PD-1/PD-L1 therapies. Therefore, there is an urgent need to understand how TIME affects non-response of patients with refractory CRC to PD-1/PD-L1 treatment. Since the intensity of the immune response is crucial in determining the effectiveness of immunotherapy, the limited therapeutic effect may be due to the immunosuppressive state of the TIME.²⁸ In our study, we observed that CCL7⁺ TAMs promoted the tolerance of CRC patients to ICIs blockade therapy, suggesting that CCL7 may be a target for patients resistant to PD-1/PD-L1 immunotherapy.

CCL7 is a well-established chemokine primarily known for its chemotactic properties, which facilitate the recruitment of monocytes, DCs, and activated T lymphocytes to sites of injury, thereby mediating the inflammatory response.²⁰ Increasing evidence suggests that CCL7 expression may be advantageous for the development and progression of various cancers, including breast cancer, renal cell carcinoma, and CRC.²⁰ In our study, we found that knockout of CCL7 inhibited both the chemotactic and immunosuppressive activities of TAMs. This finding indicates that CCL7 should not be regarded as a proinflammatory mediator in the CRC TIME. Although previous studies have indicated that CCL7 can promote tumor invasion and metastasis, other studies have suggested that it may also exert tumor-suppressive effects.²⁹ Moreover, studies have demonstrated that CCL7 secreted by monocyte-derived MDSCs interacts with CCR2 on tumor cells, activating the JNK/STAT3 signaling pathway and promoting the proliferation of dormant CRC cells.¹⁷ Additionally, CCL7 was found to interplay with CCR3, resulting in enhanced cellular proliferation, invasion, and migration via ERK and JNK signaling pathways.¹⁵ Herein, our data indicate that CCR2 and CCR3 play a role in the immunosuppressive activity of macrophages induced by Ccl7. And CCR2 may play a more significant role in the immunosuppressive activity of macrophages induced by Ccl7. Our findings further demonstrate that CCL7 promotes peroxisome biogenesis and FAO in TAMs, thereby enhancing their immunosuppressive function and contributing to the immunosuppressive state of TIME in CRC.

Peroxisomes are highly specialized membrane-bound organelles that perform critical metabolic functions, including FAO and ROS metabolism.²² In mammalian cells, peroxisome biogenesis is regulated by a group of heterogeneous proteins known as peroxins, specifically PEX3, PEX16, and PEX19, which are essential for peroxisome regeneration.²² We found that several peroxins, particularly PEX3, were highly expressed in CCL7⁺ TAMs, suggesting that PEX3 plays a significant role in regulating FAO and ROS metabolism in these cells. Previous studies have shown that PEX3 regulates phospholipid metabolism and activates the AKT/GSK3β signaling pathway via plasma membrane localization of ITGB3.³⁰ Our findings indicate that CCL7 enhances PEX3 expression in TAMs, thereby augmenting its immunosuppressive function via the CCR2-PI3K-AKT signaling pathway. These results imply that the PEX3 and PI3K-AKT signaling pathways may form a feedback loop that promotes FAO and the immunosuppressive capabilities of TAMs.³¹ The PI3K/Akt signaling pathway is activated by signals from various receptors and plays a crucial role in mediating both proinflammatory and anti-inflammatory responses in macrophages.³² AKT comprises three serine-threonine kinases: AKT1, AKT2, and AKT3. Notably, AKT1 and AKT2 exhibit opposing effects on the polarization of M1 and M2 macrophages.³³ Specifically, AKT1 promotes M2 polarization and enhances the immunosuppressive functions of macrophages, while AKT2 supports M1 polarization and facilitates the antitumor responses of these cells. In our study, we found that CCL7 binds to CCR2, inhibits AKT2, and activates the AKT1 pathway, suggesting its critical role in the regulation of AKT subtypes. However, the underlying mechanisms require further investigation.

Our study demonstrated that anti-CCL7 treatment not only inhibits the infiltration of TAMs but also enhances the infiltration of CD8⁺ T cells. CCL7 deficiency in TAMs enhances CD8⁺ T cell migration via the AKT2-STAT1-CXCL10 axis. IFN-γ secreted by these infiltrating T cells may ultimately contribute to tumor immune escape by upregulating negative immune checkpoints, such as PD-L1.³⁴ Furthermore, the combination of anti-CCL7 and anti-PD-L1 therapies exhibits a synergistic effect in inhibiting CRC growth in murine models. These findings indicate that CCL7 facilitates tolerance to ICIs blockade therapy by regulating the accumulation and activity of tumor-infiltrating macrophages and CD8⁺ T cells. Therefore, targeting CCL7 in conjunction with PD-L1 checkpoint blockade represents a promising therapeutic strategy for CRC immunotherapy. Previous studies have shown that Bindarit treatment of human melanoma cells can reduce tumor growth and macrophage infiltration, causing tumor cell necrosis.³⁵ Treatment of mouse breast and prostate cancer tumors with Bindarit inhibited tumor growth and significantly reduced TAM infiltration.³⁶ In addition, Bindarit has been reported to protect mice with lupus erythematosus from kidney damage and prolong their life span.³⁷ In animal models of arterial injury, Bindarit effectively reduces neointima formation by directly acting on vascular smooth muscle cells and by reducing the content of macrophages in the neointima.³⁸ These results suggest that Bindarit or anti-CCL7 antibodies may be widely used in the treatment of tumors and autoimmune-related diseases. To further understand the role of CCL7 in tumor immunity and autoimmunity, it is necessary to further study the therapeutic potential of CCL7 inhibitors in diseases such as tumors and autoimmunity.

Our research primarily emphasizes that the elevated levels of CCL7⁺ TAMs in CRC tissues are associated with patient tolerance to ICI therapy. However, the clinical data presented in this study remain preliminary. Whether CCL7 can be used as a biomarker for ICI resistance needs to be verified in a larger cohort of CRC patients, especially those who have received immunotherapy and whose immunotherapy outcomes are known. Furthermore, orthotopic tumor models can be established more rapidly and effectively simulate the TME associated with the growth of in situ tumors. Therefore, before clinical application, it is essential to develop an orthotopic tumor model to further elucidate the role of CCL7-CCR2 signaling within the TIME.

In summary, CCL7⁺ TAMs were upregulated in CRC patients that tolerated ICIs blockade therapy. CCL7 derived from TAMs plays a crucial role in regulating the accumulation and activity of tumor-infiltrating CD8⁺ T cells and macrophages, thereby influencing tumor growth. Blockade of CCL7 significantly improved the anti-tumor efficacy of anti-PD-L1 antibodies. Our findings highlight its function in promoting the immunosuppressive capabilities of TAMs while concurrently inhibiting the chemotaxis of CD8⁺ T cells. These insights suggest that CCL7 may serve as a promising therapeutic target for ICIs combined immunotherapy of CRC.

Data availability statement

Data are available upon reasonable request.