Involvement of increased arginase-1 + group 2 innate lymphoid cells and myeloid-derived suppressor cells in endometrial carcinoma: A pilot study

Abstract

Background

Endometrial carcinoma is a prevalent gynecologic malignancy with a relatively low survival rate. Emerging studies have demonstrated that crosstalk between group 2 innate lymphoid cells (ILC2s) and myeloid-derived suppressor cells drives tumor progression. The immunopathological mechanisms underlying endometrial carcinoma are not yet fully understood.

Methods

This study assessed the frequency of innate lymphoid cells and myeloid-derived suppressor cells as well as related inflammatory mediators in peripheral blood, carcinoma tissue, and para-cancerous tissue of patients with endometrial carcinoma.

Results

Patients with endometrial carcinoma exhibited decreased levels of interleukin-22 and interferon-γ and increased levels of interleukin-25. Infiltration of ILC2s and monocytic myeloid-derived suppressor cells was elevated, while ILC3 levels were reduced. Functional analysis showed enhanced arginase-1 expression in ILC2s obtained from patients with endometrial carcinoma. Both arginase-1⁺ ILC2s and monocytic myeloid-derived suppressor cells were significantly associated with poorer progression-free survival. A direct correlation between arginase-1⁺ ILC2s and monocytic myeloid-derived suppressor cells suggests a synergistic role in endometrial carcinoma progression.

Conclusion

Our study indicates that the collaborative effects of ILC2s and myeloid-derived suppressor cells promote type II immunity and may contribute to the progression of endometrial carcinoma. Elevated levels of arginase-1⁺ ILC2s and monocytic myeloid-derived suppressor cells are associated with a poor prognosis in patients with endometrial carcinoma.

Introduction

Endometrial carcinoma (EC) arises from the endometrial epithelium, the innermost lining of the uterus. ¹ Globally, approximately 42,000 women die from EC each year. ² EC represents a significant challenge to women’s health; however, the immunological mechanisms underlying this disease remain unclear. ³ Recent studies have shown that the presence of a local and/or systemic immunosuppressive microenvironment can facilitate immune evasion and promote EC progression.^4–6 Dysregulation of various inflammatory factors and cytokines, including nitric oxide (NO), interleukin (IL)-25, and tumor necrosis factor-α (TNF-α), has been identified as a key contributor to the immunopathology of EC.^7,8

Recently, a family of newly identified immune cells known as innate lymphoid cells (ILCs) has gained attention. Unlike T and B cells, ILCs constitute a unique subset of lymphocytes that lack antigen-specific receptors. ⁹ Based on their developmental pathways, transcription factor dependence, and cytokine production profiles, ILCs are classified into three main subsets: group 1 ILCs (ILC1s), group 2 ILCs (ILC2s), and group 3 ILCs (ILC3s), in addition to natural killer (NK) cells and lymphoid tissue inducer (LTi) cells.^9–13 These ILC subsets play diverse and critical roles in shaping the tumor microenvironment (TME).^14–16 ILC1s, which depend on the transcription factor T-bet, are characterized by their ability to produce interferon-γ (IFN-γ). ¹⁷ ILC3s, which depend on the transcription factor RORγt, produce IL-22 and IL-17. Sivori et al. reported that ILC3s can mount adaptive immune responses and provide a protective barrier, thereby inhibiting tumor progression. ¹⁸ Additionally, Gronke et al. found that stem cells are more susceptible to DNA damage responses and can cause tumor development in TMEs lacking IL-22 produced by ILC3s. ¹⁴

In contrast, ILC2s are primarily associated with type II immune responses and exhibit potent immunosuppressive properties, largely regulated by the transcription factor GATA-3.^14,15 Studies have shown that ILC2 activation requires inflammatory factors, such as IL-25 and IL-33.^19–21 Upon activation, ILC2s promote type II immunity via effector molecules, including arginase-1 (Arg-1). Expression of Arg-1 by ILC2s leads to altered arginine metabolism, which inhibits T cell proliferation and amplifies type II inflammation. ²² Additionally, Arg-1 regulates the proliferative capacity of ILC2s.^23,24 In various tumor models, ILC2s have been associated with immunosuppressive type II immunity and impaired antitumor responses. ²⁰ Moreover, Jou et al. reported that dysregulated IL-25-producing ILC2s drive the differentiation of monocytes into myeloid-derived suppressor cells (MDSCs). ²⁵ Specifically, IL-25 activates ILC2s, enhances monocytic MDSC (Mo-MDSC)-mediated immunosuppression, and promotes colorectal cancer progression. ²⁵

MDSCs are immature myeloid progenitor cells with potent immunosuppressive effects. They represent a heterogeneous population commonly divided into two main subsets based on phenotypic and morphological criteria: Mo-MDSCs and polymorphonuclear MDSCs (PMN-MDSCs). ²⁶ A third subset, early-stage MDSCs (eMDSCs), is also occasionally described. The immunosuppressive effects of MDSCs within the TME have been reported to be largely mediated by Arg-1. ²⁷ As key regulators of tumor-induced immunosuppression, MDSCs can express Arg-1, which not only impairs T cell function but also inhibits B cell proliferation via the release of Arg-1.^27,28 Moreover, MDSCs can interact with ILC2s to form an immunosuppressive ILC2–MDSC axis, enhancing type II immunity and contributing to the progression of multiple tumor diseases.^25,29,30

Although Arg-1 is recognized as an important regulator in both MDSCs and ILC2s, the synergistic role of Arg-1⁺ILC2s and MDSCs in tumorigenesis remains unclear. In this study, we evaluated the correlation between ILCs and MDSCs in EC by analyzing ILC1s, ILC2s, ILC3s, Arg-1⁺ILC2s, MDSCs, and their associated inflammatory mediators in peripheral blood (PB) and carcinoma tissue (CT) from patients with EC. Furthermore, to explore their impact on tumor progression, we assessed the relationship between these cells and progression-free survival (PFS) in patients with EC.

This study was conducted in accordance with the Declaration of Helsinki for Medical Research and received appropriate ethical approval. Our findings have been published as a preprint on Research Square (DOI: 10.21203/rs.3.rs-3991479/v1) to rapidly disseminate our results to the academic community and solicit peer review and feedback, thereby promoting research transparency and quality.

Materials and methods

Preparation and recruitment of patients

The study included two distinct patient cohorts. The EC cohort (n = 41) comprised patients with an initial diagnosis of EC, while the control group (CG) (n = 40) consisted of patients with uterine leiomyoma, matched for age and body mass index to minimize potential confounding effects. The sample size was determined based on patient availability during the study period and feasibility for an initial pilot investigation. Table S1 summarizes the clinical characteristics of the participants. Patients were enrolled at the Department of Gynecology, The First Affiliated Hospital of Anhui Medical University, from October 2021 to December 2022. All participants provided informed consent, and the study was approved by the Clinical Medical Research Ethics Committee of The First Affiliated Hospital of Anhui Medical University (No. PJ2023-03-12). PFS was defined as the time from diagnosis to disease progression. Patients were followed up for 400 ±121.3 days. Patients with autoimmune diseases or immunodeficiencies or those receiving hormone or steroid therapy were excluded to maintain the study’s focus on EC. The handling and use of human participant data strictly adhered to confidentiality and privacy regulations to protect individual rights.

PB collection

PB samples were collected from all participants prior to surgery and immediately anticoagulated with heparin. Within the EC cohort, 10 patients with EC underwent post-surgical chemotherapy. One day after completion of chemotherapy, PB samples were collected from these 10 patients with EC. PB mononuclear cells (PBMCs) and serum were subsequently separated following established protocols. ³¹ Serum samples were stored at −80°C, and PBMCs were preserved at −80°C after the addition of a cytoprotective solution, as described previously. ³¹

Tissue collection

Tissue samples included CT and para-cancerous tissue (PCT), which is located 1 cm away from the tumor margin. CTs and PCTs were collected during surgery. Tumor single-cell suspensions were prepared using the gentleMACS Dissociator (Miltenyi Biotec, Bergisch Gladbach, Germany) and stored at −80°C in a cytoprotective solution, following the same protocol used for PB samples. ³¹

Flow cytometry

Samples were initially thawed at 37°C. Subsequently, 10 mL of phosphate-buffered saline was added, and the cells were centrifuged for 10 min at 412 ×g. This washing step was repeated twice. Then, 10⁶ cells were resuspended in 100 μL of phosphate buffer. A portion of the sample was stained with Trypan Blue (Sigma-Aldrich, USA) to assess cell viability, followed by staining with 5 μL of specific monoclonal antibodies (mAbs) for 30 min at 4°C. Fc receptor (FcR) blocking reagent (BioLegend, Germany) was applied prior to antibody staining according to the manufacturer’s instructions. Specific anti-human mAbs were used to stain ILCs and MDSCs, as listed in Table S2. Single-color stained samples served as compensation controls, and the compensation matrix was manually adjusted based on these controls and unstained cells to minimize spillover. Doublets were excluded using SSC-H versus SSC-A gating. Fluorescence-minus-one (FMO) controls were used to define positive populations. For tissue samples, CD45⁺ cells were selectively targeted. All subsequent experimental procedures, including cell collection and equipment operation, followed previously established protocols. ³¹ Cell sorting and flow cytometry analyses were performed using a FACSVerse flow cytometer (BD, USA) and analyzed with FlowJo software (Tree Star, Ashland, OR, USA).

ELISA

Cytokines, including IFN-γ, IL-4/22/25, and CCL3/4/5, were quantified in the serum samples of the two study groups using ELISA. Kits were obtained from Multisciences Biotech (Hangzhou, China). Subsequent procedures followed previously established protocols. ³¹

Statistical analysis

Statistical analyses were performed using GraphPad Prism 8.0 (GraphPad Software, San Diego, CA). An unpaired Student’s t-test was used to compare data between two independent groups, while a paired Student’s t-test was applied for comparisons before/after chemotherapy in patients with EC. One-way analysis of variance was employed to assess differences among multiple groups. Correlations between ILCs, MDSCs, and other factors were analyzed using Spearman correlation with Bonferroni correction to account for multiple testing. Survival analysis was performed using the Kaplan–Meier method, and log-rank tests were used to evaluate PFS. For survival analysis, patients were censored at the date of their last known follow-up without disease progression. A p-value <0.05 was considered statistically significant.

Results

Accumulation of ILCs in PB and CT from patients with EC

To investigate the correlation between ILCs and EC, we analyzed PBMCs and CTs obtained from patients with EC. Lin-CD127⁺ cells were identified as ILCs in PBMCs, whereas CD45⁺Lin-CD127⁺ cells were classified as ILCs in tissue samples. ILCs were further categorized into three subgroups based on their expression patterns. The gating strategy for this classification is shown in Figure 1(a).

Figure 1.

Analysis of different ILC subsets in patients with EC. (a) ILCs were identified as Lin⁻CD127⁺ cells and further classified into three subsets: ILC1s (c-Kit⁻CRTH2⁻), ILC2s (CRTH2⁺), and ILC3s (c-Kit⁺CRTH2⁻). (b–e) Frequencies of total ILCs and ILC2s were measured in PB from patients with EC and CG as well as in CT and PCT. (f–g) Ratios of ILC2s/ILCs and ILC3s/ILCs were determined in patients with EC and CG. (h) Percentage of ILC3s in the live PBMC gate was measured before and after chemotherapy. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. CG: control group; CT: carcinoma tissue; EC: endometrial carcinoma; ILCs: innate lymphoid cells; PB: peripheral blood; PCT: para-cancerous tissue.

Patients with EC exhibited elevated levels of total ILCs and ILC2s in PBMCs compared with the CG (p < 0.05, p < 0.001; Figure 1(b) and (c)). Similarly, the frequency of ILCs and ILC2s in CTs was significantly higher in patients with EC than in controls (p < 0.05 for both; Figure 1(d) and (e)). When assessing the proportions of ILC subgroups relative to total ILCs, we observed an increased ratio of ILC2s/ILCs in patients with EC compared with that in controls (p < 0.0001; Figure 1(f)), while the ratio of ILC3s/ILCs was reduced in patients with EC (p < 0.01; Figure 1(g)).

We also evaluated ILC levels in PBMCs obtained from patients with EC before and after chemotherapy. The frequency of ILC3s significantly increased after chemotherapy compared with prechemotherapy levels (p < 0.01; Figure 1(h)). However, comparative analysis between patients with early-stage and advanced-stage EC revealed no statistically significant differences in ILC frequencies.

Elevation of Arg-1⁺ILC2s in PB and CT of EC patients

To assess Arg-1 expression in ILC2s, we gated Arg-1⁺ILC2s (Figure 2(a)). Both PBMCs and CTs obtained from patients with EC displayed significantly elevated levels of Arg-1⁺ILC2s compared with those obtained from controls (p < 0.001, p < 0.05; Figure 2(b) and (c)). Similarly, the ratio of Arg-1⁺ILC2s/ILCs was higher in CTs than in PCTs (p < 0.01; Figure 2(d)). Although the ratio of Arg-1⁺ILC2s/ILCs in PBMCs was higher in patients with EC than in controls, this difference did not reach statistical significance (p = 0.056, Figure 2(e)).

Figure 2.

Elevated expression of Arg-1⁺ILC2s in patients with EC. (a) Arg-1⁺ILC2s were gated using FMO controls. (b–c) Frequency of Arg-1⁺ILC2s was measured in PB of patients with EC and CG as well as in CT and PCT. (d–e) Ratios of Arg-1⁺ILC2s/ILCs were calculated in PB and tissue samples. *p < 0.05, **p < 0.01, ***p < 0.001. Arg-1: arginase-1; CG, control group; CT: carcinoma tissue; EC: endometrial carcinoma; FMO: fluorescence-minus-one; ILCs: innate lymphoid cells; PCT: para-cancerous tissue.

Increased levels of MDSCs during EC progression

MDSCs were gated based on the expression of specific markers (CD33⁺HLA-DR-/^lowCD11b⁺/CD45⁺CD33⁺HLA-DR⁻/lowCD11b⁺) in PBMCs and CTs. Based on CD14 and CD15 expression, MDSCs were divided into two subsets: Mo-MDSCs (CD14⁺CD15⁻) and PMN-MDSCs (CD14⁻CD15⁺) via flow cytometry (Figure 3(a)).

Figure 3.

Gating strategy and analysis of MDSCs in PB and tissue. (a) MDSCs (CD33⁺HLA-DR⁻/lowCD11b⁺) were divided into Mo-MDSCs (CD14⁺CD15⁻ MDSCs) and PMN-MDSCs (CD14⁻CD15⁺ MDSCs). (b–e) Frequencies of total MDSCs and Mo-MDSCs were measured in PB of patients with EC and CG and in CT and PCT. (f–g) Percentage of MDSCs and Mo-MDSCs in live PBMCs was measured before and after chemotherapy. *p < 0.05, **p < 0.01, ***p< 0.001, ****p< 0.0001. CG: control group; CT: carcinoma tissue; EC: endometrial carcinoma; MDSCs: myeloid-derived suppressor cells; Mo-MDSCs: monocytic MDSCs; PB: peripheral blood; PCT: para-cancerous tissue; PMN-MDSCs: polymorphonuclear MDSCs.

Patients with EC showed a significant increase in total MDSCs and Mo-MDSCs in both PBMCs and CTs compared with controls (p < 0.001, p < 0.0001, p < 0.01, p < 0.05; Figure 3(b) to (e)). Furthermore, levels of MDSCs and Mo-MDSCs decreased in patients after chemotherapy compared with prechemotherapy levels (p < 0.01 for both; Figure 3(f) and (g)). The frequency of PMN-MDSCs was also evaluated, but no statistically significant differences were observed compared with the CG.

Levels of cytokines IFN-γ, IL-22, and IL-25 in serum

Compared with the CG, IL-25 levels were elevated in patients with EC, whereas IFN-γ and IL-22 were significantly decreased (p < 0.01, p < 0.05, p < 0.05; Figure 4(a) to (c)). No statistically significant differences were observed for the other tested cytokines.

Figure 4.

Serum levels of inflammatory factors. IFN-γ (a), IL-22 (b), and IL-25 (c) were measured in patients with EC and CG. *p < 0.05, **p < 0.01. IFN-γ: interferon-γ; IL-22/25: interleukin-22/25; PB: peripheral blood.

Correlation between ILCs, MDSCs, Arg-1⁺ILC2s, and Mo-MDSCs as predictors of PFS in EC

A significant positive correlation was observed between the frequencies of Arg-1⁺ILC2s and Mo-MDSCs (p < 0.05; Figure 5(a)). Similarly, ILC3s were positively correlated with ILC1s (p < 0.001; Figure 5(b)).

Figure 5.

Correlations between ILC subsets, Mo-MDSCs, and PFS in patients with EC. (a) Correlation between Arg-1⁺ ILC2s and Mo-MDSCs in PB. (b) Correlation between ILC1s and ILC3s in PB. (c, d) Kaplan–Meier curves showing associations of PB Arg-1⁺ILC2s and PB Mo-MDSCs with PFS. p < 0.05 is considered significant. Arg-1: arginase-1; ILCs: innate lymphoid cells; Mo-MDSCs: monocytic MDSCs; PB: peripheral blood; PFS: progression-free survival.

To evaluate the clinical implications of Arg-1⁺ILC2s and Mo-MDSCs in patients with EC, the mean value of each marker was used as a cutoff for subgroup categorization. The relationship between high and low subgroups of Arg-1⁺ILC2s and Mo-MDSCs with PFS was assessed using Kaplan–Meier curves and log-rank test. Elevated levels of both Arg-1⁺ILC2s (p < 0.05; Figure 5(c)) and Mo-MDSCs (p < 0.05; Figure 5(d)) were associated with PFS.

Discussion

The aim of this study was to investigate the relationship between key inflammatory cytokines, ILCs and MDSCs and their potential impact on PFS in patients with EC. Previous studies have highlighted the importance of tumor-associated immune responses in EC, including the activities of macrophages, T cells, NK cells, and related inflammatory factors.^1–3 Our principal findings indicate that alterations in the frequencies of ILCs and MDSCs may represent a critical immune mechanism contributing to EC progression.

ILCs are a class of immunomodulatory cells, and ILC2s are implicated in immunosuppression through IL-4, IL-13, and IL-25, with the potential to promote tumor development in various cancers. However, the role of ILC2s in EC has remained unclear. In our study, ILC2 levels were significantly increased in both PBMCs and CTs from patients with EC compared with controls. Similarly, Xu et al. reported proliferation and accumulation of ILC2s in a murine model of hepatocellular carcinoma. ³² These findings suggest that ILC2s may play a key role in establishing an immunosuppressive TME in EC.

Concurrently, our data show significant accumulation of MDSCs, particularly Mo-MDSCs, in the PB and tumor tissues of patients with EC. Higher frequencies of Mo-MDSCs were negatively correlated with PFS, mirroring the pattern observed with Arg-1⁺ILC2s. Together with previous studies, ³¹ our findings indicate that Mo-MDSCs may exert immunosuppressive effects in EC, contributing to disease progression.

Arg-1 is a key enzyme in the urea cycle that regulates T cell proliferation, thereby modulating the TME. ILC2s are among the immunosuppressive cells that express Arg-1. Therefore, we investigated Arg-1 expression in ILCs in PB and tumor tissues of patients with EC. Patients with EC exhibited a clear population of Arg-1⁺ILC2s. Elevated Arg-1⁺ILC2s were negatively associated with PFS in patients with EC followed for 400 ± 121.3 days. In a murine model of colon cancer, tumor-associated macrophages were shown to express Arg-1, conferring immunosuppressive properties that promote tumor progression. ³³ Interestingly, ILC2s have been shown to accelerate tumor growth by inducing MDSC infiltration into tumor tissue. Chevalier et al. reported that ILC2s promote bladder cancer recurrence by recruiting Mo-MDSCs and skewing immunity toward type II immunity.^29,34 Another study demonstrated that the immunosuppressive axis between ILC2s and MDSCs mutually reinforces their function, accelerating prostate cancer progression, and that blocking this axis improves survival in a murine model. ³⁰ In our study, IL-25, a key cytokine in the ILC2–MDSC axis, was significantly elevated in patients with EC. Furthermore, Arg-1⁺ILC2s were positively correlated with Mo-MDSCs. Consistent with previous studies, these findings suggest that the combined action of Arg-1⁺ILC2s and Mo-MDSCs can enhance type II immune responses via IL-25. This interaction underscores the important role of Arg-1-mediated immunosuppression in promoting EC progression.

We also examined other ILC subsets. The ratio of ILC3s to total ILCs was significantly decreased, accompanied by reduced serum IL-22 levels, suggesting potential impairment of the protective barrier function of IL-22-producing ILC3s. Notably, the frequencies of ILC1s and ILC3s were positively correlated in patients with EC, indicating a potential interplay between these subsets within the TME that warrants further investigation. Previous studies have shown that ILC1s, similar to NK cells, secrete IFN-γ, which inhibits tumor growth, while ILC3s can limit tumor development via IL-22. ¹⁷ Our observations are consistent with these reports, suggesting that the interaction between ILC1s and ILC3s may help attenuate immunosuppression and hinder EC progression.

In summary, our findings highlight the collaborative interaction between Arg-1⁺ILC2s and Mo-MDSCs in establishing an immunosuppressive TME in patients with EC. With a deeper understanding of the immune microenvironment in EC, future translational research can explore integrating these immunological insights into digital health platforms and patient education tools. Such integration could support personalized patient management, enhance doctor–patient communication, and ultimately improve long-term prognosis and quality of life for patients with EC.

Conclusion

In this study, we found that serum levels of IFN-γ and IL-22 were decreased in patients with EC. Notably, we observed a significant increase in the frequencies of ILC2s (including the Arg-1⁺ subset) and Mo-MDSCs in both PB and the TME. Our results indicate that elevated levels of Arg-1⁺ILC2s and Mo-MDSCs are associated with a poor prognosis, and their synergistic interaction contributes to the tumor-promoting processes in EC. We propose that the immunosuppressive effects arising from the interaction between Arg-1⁺ILC2s and Mo-MDSCs, potentially mediated through IL-25, play a central role in fostering an immunosuppressive TME in patients with EC.

Supplemental Material

sj-pdf-1-imr-10.1177_03000605261426163 - Supplemental material for Involvement of increased arginase-1 ⁺ group 2 innate lymphoid cells and myeloid-derived suppressor cells in endometrial carcinoma: A pilot study

Supplemental material, sj-pdf-1-imr-10.1177_03000605261426163 for Involvement of increased arginase-1 ⁺ group 2 innate lymphoid cells and myeloid-derived suppressor cells in endometrial carcinoma: A pilot study by Zhangyun Gong, Zhimin Lu, Anan Pan, Hao Wang, Haixing Yi, Peipei Guo, Kaihuan Bi, Xuqing Li, Ya Chen, Yuping Xu, Yunxia Cao, Caihua Li and Huanhuan Jiang in Journal of International Medical Research