Heterogeneous Expression of TREM2 And STAB1 Among Tumor-Associated Macrophages in Clear-Cell Renal Cell Carcinoma

Abstract

Tumor-associated macrophages (TAMs) play important roles in shaping the tumor microenvironment and regulating antitumor immune responses. Triggering receptor expressed on myeloid cells 2 (TREM2) and stabilin-1 (STAB1/CLEVER1) have recently attracted attention as molecules associated with immunosuppressive macrophage phenotypes; however, their expression patterns in clear-cell renal cell carcinoma (ccRCC) remain unclear. This study aimed to examine TREM2 and STAB1 expression in ccRCC using public transcriptomic datasets and public single-cell RNA sequencing data, as well as immunohistochemistry using ccRCC specimens. In the TCGA cohort, high TREM2 and STAB1 mRNA expression was associated with poor overall survival and advanced tumor stage. In an independent immunohistochemistry cohort, TREM2 was expressed mainly in TAMs, whereas STAB1 was expressed in both TAMs and endothelial cells. TREM2-positive TAM density was significantly associated with high nuclear grade, whereas STAB1-positive TAM density was not. Double immunohistochemistry and single-cell RNA sequencing analyses suggested different, but partially overlapping, expression patterns of TREM2 and STAB1 among TAMs. TREM2-high TAMs showed features of monocyte-derived macrophages, whereas STAB1-high TAMs showed features associated with tissue-resident macrophages. These findings suggest that TREM2-positive TAMs and STAB1-positive TAMs contribute differently to the immune microenvironment of ccRCC. Further studies are needed to clarify their functional significance and therapeutic potential.

I. Introduction

Renal cell carcinoma (RCC) is a malignant tumor arising from renal tubular epithelial cells that accounts for more than 90% of all kidney cancers [1]. Approximately 40% of patients with RCC reportedly die from disease progression, making RCC one of the most lethal malignancies of the genitourinary system [2]. RCC comprises multiple histological subtypes, among which, clear-cell RCC (ccRCC) is the most common, accounting for approximately 75% of cases [1, 2].

Macrophages are multifunctional antigen-presenting cells that play central roles in inflammatory responses. Macrophages that extensively infiltrate tumor tissues are collectively referred to as tumor-associated macrophages (TAMs), and their functions are reprogrammed through interactions with tumor cells. TAMs, particularly those with a protumor phenotype expressing CD163 or CD204, promote tumor invasion, metastasis, angiogenesis, cancer stemness, and immunosuppression, thereby contributing to tumor progression [3, 4].

TAMs have been investigated as therapeutic targets in cancer immunotherapy from several perspectives, including inhibition of monocyte recruitment, depletion of TAMs, reprogramming of TAMs, and targeting of inhibitory molecules [5, 6]. Current therapeutic strategies focus on the key signaling pathways involved in monocyte recruitment and TAM differentiation, such as the CCL2–CCR2 and CSF1–CSF1R axes [7]. Indeed, clinical trials targeting CCL2, such as carlumab, and CSF1R, such as pexidartinib, have been conducted; however, their therapeutic efficacy has been limited [8]. This limited efficacy may be attributable to the plasticity of TAMs and the presence of redundant immunosuppressive pathways within the tumor microenvironment. Thus, inhibition of a single pathway may be insufficient to reverse immunosuppression fully. These findings highlight the importance of identifying molecules that directly mediate the immunosuppressive functions of TAMs, beyond merely regulating their recruitment or survival.

In recent years, triggering receptor expressed on myeloid cells 2 (TREM2) and stabilin-1 (STAB1; also known as FEEL-1 or CLEVER-1) have attracted attention as molecules associated with the immunosuppressive phenotype of TAMs [9, 10]. TREM2 is a receptor expressed on myeloid cells, including macrophages, and is involved in the maintenance of tissue homeostasis by promoting phagocytosis and suppressing inflammation [9, 11]. By contrast, STAB1 is a scavenger receptor expressed on immunosuppressive monocytes and macrophages that plays a role in maintaining immune tolerance [12]. These molecules have been reported to contribute to the formation of an immunosuppressive tumor microenvironment in various cancer types, and their expression is associated with poor prognosis [9, 10]. Although therapeutic strategies targeting TREM2 and STAB1 may represent novel approaches in cancer immunotherapy, the expression patterns of TREM2 and STAB1 in the same cohort of patients with ccRCC have not been fully characterized. Therefore, in the present study, we aimed to examine the expression and distribution of TREM2 and STAB1 in ccRCC tissues using immunohistochemistry (IHC) and publicly available single-cell RNA sequencing datasets.

II. Materials and Methods

The Cancer Genome Atlas (TCGA) data analysis

RNA sequencing data and corresponding clinical information for RCC were obtained from TCGA database. The expression levels of TREM2 and STAB1 were analyzed using normalized gene expression data. For survival analysis, patients were divided into high- and low-expression groups based on the median expression level of each gene. Overall survival was evaluated using the Kaplan–Meier method, and differences between groups were assessed using the log-rank test.

Patients

Patients included in this study were those with ccRCC who underwent radical or partial nephrectomy at Kumamoto University Hospital between 2019 and 2020. Tissue samples were fixed in 10% neutral-buffered formalin and embedded in paraffin. A total of 45 ccRCC cases were included in the pathological analysis. TREM2 immunoreactivity was evaluable in all 45 cases. In contrast, 10 cases showed absent or markedly weak STAB1 staining, including in endothelial cells serving as internal positive controls, and were therefore considered technically non-evaluable. Consequently, 35 cases were included in the quantitative analysis of STAB1-positive TAMs. The clinicopathological characteristics of the patients are summarized in Supplementary Table S1.

Immunohistochemistry (IHC)

Sections with a thickness of 3 μm were immersed in EDTA solution, pH 8.0, and heated in a pressure cooker for antigen retrieval. Anti-TREM2 antibody (rabbit monoclonal, clone: EPR26209-22, x500 dilution, Abcam, Cambridge, UK), anti-STAB1 antibody (mouse monoclonal, clone: 840449, x300 dilution, R&D Systems, Minneapolis, MN, USA), anti-Iba-1 antibody (rabbit polyclonal; x2000 dilution, Fujifilm, Tokyo, Japan), anti-FOLR2 antibody (mouse monoclonal, clone: OT14G6, x500 dilution, Origene, Rockville, MD, USA), anti-CD163 antibody (mouse monoclonal, clone: 10D6, x500 dilution, Leica Biosystems, Nussloch, Germany), and anti-CD204 antibody (mouse monoclonal, clone: SRA-E5p6, x500 dilution, CosmoBio, Tokyo, Japan) were used as primary antibodies. After incubation with the primary antibodies, sections were incubated with horseradish peroxidase-labeled secondary anti-rabbit or anti-mouse antibodies (Nichirei Bioscience, Tokyo, Japan). Immunoreactions were visualized using the diaminobenzidine system (Nichirei Bioscience). Stained slides were digitally scanned using a NanoZoomer S20 scanner (Hamamatsu Photonics, Shizuoka, Japan). Because STAB1 was also expressed in vascular structures, only STAB1-positive cells with round or macrophage-like morphology within the tumor area were counted as STAB1-positive TAMs. TREM2- and STAB1-positive cells in the tumor area were counted using HALO (version 3.6.4134; Indica Labs, Albuquerque, NM, USA), and the average density, expressed as cells per square millimeter, was calculated in solid tumor area without necrosis. For double IHC, the second immunoreaction was visualized using the HistoGreen chromogen substrate system. Single and double IHC procedures were performed as previously described [13, 14].

Single-cell RNA sequencing

Single-cell RNA sequencing data from ccRCC cohorts (GSE159115, 7 patients; GSE207493, 19 patients; GSE152938, 2 patients; and GSE171306, 2 patients) were obtained from the Gene Expression Omnibus. Data processing was performed using R (version 4.5.2; R Foundation for Statistical Computing, Vienna, Austria) with Seurat (version 5.4.0) and Harmony for batch correction. Low-quality cells, defined as cells with ≤500 or ≥5000 detected genes or ≥20% mitochondrial reads, were removed.

Each sample was converted into a Seurat object and merged using the merge function, followed by layer integration using JoinLayers, normalization using NormalizeData, and identification of variable features using FindVariableFeatures. Gene expression was scaled using ScaleData, and batch effects were regressed using RunHarmony with orig.ident as the covariate. Principal component analysis using 13 principal components was performed for dimensionality reduction and clustering, followed by FindNeighbors and FindClusters at a resolution of 0.5, which yielded 17 clusters.

Statistical analysis

All statistical analyses were performed using GraphPad Prism (version 9.4.0; GraphPad Software, San Diego, CA, USA). Comparisons between two groups were performed using the Mann–Whitney U test, as appropriate. Values of P < 0.05 were considered statistically significant.

Research involving Human Participants and/or Animals

The study design was approved by the Institutional Review Board of Kumamoto University (approval no. #2226) in accordance with the guidelines for Good Clinical Practice and the Declaration of Helsinki.

Informed consent

The present study was a retrospective analysis, which does not require individual consent; however, all participants had the opportunity to opt out.

Data availability statement

The datasets generated during and/or analyzed during the current study are not publicly available due to the privacy of research participants or ethical restriction but are available from the corresponding author on reasonable request.

III. Results

TCGA data analysis showed that increased expression of TREM2 and STAB1 was associated with poor clinical outcomes

First, the association between TREM2 or STAB1 expression and prognosis in patients with ccRCC was examined using TCGA data available from The Human Protein Atlas. Kaplan–Meier analysis demonstrated that the TREM2 high-expression group showed significantly poorer overall survival than the TREM2 low-expression group (Fig. 1A). Similarly, the STAB1 high-expression group exhibited significantly reduced overall survival compared with the STAB1 low-expression group. In addition, both TREM2 and STAB1 expression levels were significantly higher in stage III/IV cases than in stage I/II cases (Fig. 1B).

Fig. 1.

In silico analysis of TREM2 and STAB1 expression in ccRCC. (A) Kaplan–Meier survival analysis of TREM2 and STAB1 expression in TCGA ccRCC cohort. Patients were stratified into high- and low-expression groups according to the median expression level of TREM2 or STAB1. Overall survival was compared between groups using the log-rank test. (B) Comparison of TREM2 and STAB1 expression according to tumor stage in TCGA ccRCC cohort. Each dot represents an individual patient, and horizontal lines indicate the mean ± standard error of the mean. Statistical analysis was performed using the Mann–Whitney U test.

Increased numbers of TREM2-positive TAMs were associated with high nuclear grade

Next, IHC was performed to evaluate the TREM2 and STAB1 expression in ccRCC tissues (Fig. 2A). TREM2 was exclusively detected in macrophage-like cells, whereas STAB1 was expressed in not only macrophages, but also vascular structures. Double IHC using anti-Iba-1 antibody, one of widely used markers for monocytes/macrophages, demonstrated that TREM2 was detected in Iba-1-positive cells and thus considered to be expressed in TAMs (Fig. 2B). STAB1-positive cells with round morphology were also detected among Iba-1-positive cells and considered to represent TAMs.

Fig. 2.

Immunohistochemistry (IHC) analysis of TREM2 and STAB1 in ccRCC tissues. (A) Representative figures of TREM2 and STAB1 IHC. Arrowheads indicate STAB1-positive endothelial cells with flattened and elongated morphology lining vascular lumina. Bars = 20 μm. (B) Representative figures of double IHC of TREM2(brown)/Iba1(green) and STAB1(brown)/Iba1(green). Iba-1 was used as a marker for pan-monocyte/macrophages. Nuclei were counterstained with hematoxylin. Bars = 20 μm.

We then examined the densities of TREM2- and STAB1-positive cells with round morphology in ccRCC tissues and analyzed their associations with clinicopathological factors. As shown in Fig. 3A, the number of TREM2-positive cells was significantly increased in grade 3/4 tumors compared with grade 1/2 tumors. By contrast, although the number of STAB1-positive cells tended to be higher in grade 3/4 tumors, the difference did not reach statistical significance.

Fig. 3.

Correlation of TREM2 and STAB1 to nuclear grade. (A) Comparison of TREM2- and STAB1-positive cell densities according to WHO/ISUP nuclear grade in ccRCC tissues. Each dot represents an individual case, and horizontal lines indicate the mean ± standard error of the mean. Statistical analysis was performed using the Mann–Whitney U test. (B) Representative figures of double IHC of STAB1 (brown) and TREM2 (green) in ccRCC tissues. Nuclei were counterstained with hematoxylin. Bars = 20 μm.

Furthermore, double IHC for TREM2 and STAB1 demonstrated that TREM2-positive cells had larger cytoplasms than STAB1-positive cells, and many TREM2-positive cells were negative for STAB1 (Fig. 3B). A small number of cells were positive for both TREM2 and STAB1 (Supplementary Fig. S1), whereas many cells were positive for either TREM2 or STAB1.

TREM2 and STAB1 showed different expression patterns among TAMs

Because double IHC suggested different, but partially overlapping, expression patterns of TREM2 and STAB1 among TAMs in ccRCC, publicly available single-cell RNA sequencing datasets were reanalyzed. TREM2 expression was restricted to AIF1 (Iba-1-expressing TAMs), whereas STAB1 was expressed in both endothelial cells and TAMs (Fig. 4).

Fig. 4.

Single-cell RNA sequencing analysis. UMAP visualization of clustered single-cell RNA sequencing data from 30 patients with ccRCC. Cell clusters were annotated according to cell identity. Feature plots showing the expression of AIF1, TREM2, and STAB1.

Next, gene expression profiles of AIF1-positive/TREM2-high/STAB1-low cells and AIF1-positive/TREM2-low/STAB1-high cells were examined; the heatmap profile is shown in Supplementary Fig. S2A. TREM2-high cells showed higher expression of genes related to monocyte-derived macrophages, such as APOE and SPP1, whereas STAB1-high cells showed higher expression of genes related to tissue-resident macrophages, such as SIGLEC1, VSIG4, C1Q, and FOLR2. Violin plots of macrophage-related genes are shown in Supplementary Fig. S2B.

CD163 is known to be specifically expressed in tissue-resident macrophages in mice, but is expressed in both monocyte-derived and tissue-resident macrophages in humans [15]. In the present RNA sequencing dataset, CD163 expression was significantly higher in STAB1-high cells. CD204 is a macrophage-associated marker, and RNA sequencing data indicated that CD204 was expressed in both TREM2- and STAB1-high cells.

Double IHC showed that FOLR2 was expressed in STAB1-positive cells, and that FOLR2- and TREM2-positive cells were distinct populations (Supplementary Fig. S3). STAB1-positive TAMs expressed CD163; however, little or no CD163 expression was observed in TREM2-positive TAMs. CD204 was positive in both STAB1- and TREM2-positive TAMs.

IV. Discussion

In the present study, we examined TREM2 and STAB1 expression in ccRCC. In the TCGA cohort, increased mRNA levels of these molecules were significantly associated with poor clinical outcomes. IHC analysis using pathological specimens from our ccRCC cohort showed that TREM2 was expressed in TAMs, whereas STAB1 was expressed in both TAMs and endothelial cells. Although survival data were not available in our ccRCC cohort, an increased density of TREM2-positive TAMs was observed in cases with a high nuclear grade, which is reportedly associated with poor clinical outcomes. By contrast, STAB1-positive TAMs were not significantly associated with nuclear grade. This discrepancy may be explained, at least in part, by STAB1 expression in endothelial cells. Whereas TCGA bulk RNA-sequencing data reflect STAB1 expression in all cell types within the tumor tissue, our IHC analysis specifically quantified STAB1-positive TAMs after excluding vascular structures. Therefore, total STAB1 mRNA expression may show a different association with clinicopathological factors from that observed for the density of STAB1-positive TAMs.

A previous study demonstrated the significance of TREM2 in single-cell RNA sequencing analysis and showed that TREM2-expressing TAMs were associated with postoperative recurrence [16]. TREM2 has been identified in TAMs in various organs, and animal studies have shown that targeting TREM2 enhances anticancer immunotherapy [11, 16]. Increased STAB1 expression has been reported to be associated with poor clinical outcomes in several solid cancers, including gastric and colorectal cancer [12]; however, to our knowledge, no previous study has examined STAB1 expression in ccRCC. Our observations suggest that TREM2 and STAB1 show different, but partially overlapping, expression patterns among TAMs in ccRCC. In addition, the single-cell RNA sequencing data indicated that TREM2-high and STAB1-high TAMs were enriched for different gene expression profiles. TREM2-high TAMs were enriched for features associated with monocyte-derived macrophages, whereas STAB1-high TAMs showed features associated with tissue-resident macrophages. However, since the extent of TREM2 and STAB1 co-expression was not quantitatively assessed, these findings should not be interpreted as demonstrating completely separate macrophage subpopulations. These findings raise the hypothesis that TREM2- and STAB1-expressing TAMs may represent different therapeutic targets; however, this possibility requires validation in functional studies.

Morphological observations using IHC showed that TREM2-positive TAMs had larger cytoplasms than STAB1-positive TAMs. Recent studies have indicated that TREM2-positive macrophages are enriched for lipid metabolism-related genes and exhibit a lipid-rich phenotype [17]. Lipid-rich macrophages are also referred to as lipid-associated macrophages and known to have protumor and immunosuppressive functions in solid cancers, such as prostate and breast cancer [18, 19]. However, no studies have investigated the significance of lipid-rich or lipid-associated macrophages in ccRCC. Our findings suggest that TREM2 may be a useful marker of lipid-rich or lipid-associated macrophages in ccRCC.

In the single-cell RNA sequencing analysis, TREM2-expressing macrophages expressed APOE and SPP1, which are known markers of monocyte-derived macrophages, whereas STAB1-expressing macrophages expressed MRC1, VSIG4, and FOLR2, which are associated with tissue-resident macrophages [20, 21]. TREM2-expressing TAMs have been reported to have protumor functions in several cancers. By contrast, FOLR2-expressing TAMs are thought to be involved in anticancer immune responses through the development of tertiary lymphoid structures, and increased numbers of FOLR2-positive TAMs have been correlated with better clinical outcomes in breast cancer [22]. However, in the present study, IHC analysis showed that FOLR2-positive macrophages expressed higher levels of CD163 than TREM2-positive macrophages. CD163 is a transmembrane scavenger receptor involved in the protumor activation of TAMs [23]. Therefore, the functional significance of FOLR2-/STAB1-positive macrophages in ccRCC requires further investigation, particularly in relation to their potential protumor or antitumor roles.

This study has several limitations. First, the number of cases in our IHC cohort was limited, and survival analysis could not be performed. Second, the functional roles of TREM2- and STAB1-positive TAMs were not directly examined. Third, STAB1 was also expressed in vascular structures; therefore, morphological criteria were required to distinguish STAB1-positive TAMs from endothelial cells. Further studies using larger cohorts and functional assays are needed to clarify the biological and clinical significance of these TAM subpopulations in ccRCC.

In conclusion, TREM2 and STAB1 showed different but partially overlapping expression patterns among TAMs in ccRCC. TREM2-high and STAB1-high TAMs exhibited different gene expression profiles, suggesting phenotypic heterogeneity among TAMs. These findings suggest that TREM2- and STAB1-expressing TAMs may have different functional roles; however, further functional studies are required to determine whether either or both populations represent appropriate therapeutic targets in ccRCC.

V. Conflicts of Interest

The Authors have no conflicts of interest to declare in relation to this study.

VI. Acknowledgments

This work was supported by JSPS KAKENHI (#25K02477 to Y.K.).

Supplementary Material

Supplementary Fig. S1

Supplementary Fig. S2

Supplementary Fig. S3

Supplementary Table S1