Single-cell landscape and spatial transcriptomic analysis reveals macrophage infiltration and glycolytic metabolism in kidney renal clear cell carcinoma

Abstract

The present study investigates the clinical relevance of glycolytic factors, specifically PGAM1, in the tumor microenvironment of kidney renal clear cell carcinoma (KIRC). Despite the established role of glycolytic metabolism in cancer pathophysiology, the prognostic implications and key targets in KIRC remain elusive. We analyzed GEO and TCGA datasets to identify DEGs in KIRC and studied their relationship with immune gene expression, survival, tumor stage, gene mutations, and infiltrating immune cells. We explored Pgam1 gene expression in different kidney regions using spatial transcriptomics after mouse kidney injury analysis. Single-cell RNA-sequencing was used to assess the association of PGAM1 with immune cells. Findings were validated with tumor specimens from 60 KIRC patients, correlating PGAM1 expression with clinicopathological features and prognosis using bioinformatics and immunohistochemistry. We demonstrated the expression of central gene regulators in renal cancer in relation to genetic variants, deletions, and tumor microenvironment. Mutations in these hub genes were positively associated with distinct immune cells in six different immune datasets and played a crucial role in immune cell infiltration in KIRC. Single-cell RNA-sequencing revealed that elevated PGAM1 was associated with immune cell infiltration, specifically macrophages. Furthermore, pharmacogenomic analysis of renal cancer cell lines indicated that inactivation of PGAM1 was associated with increased sensitivity to specific small-molecule drugs. Altered PGAM1 in KIRC is associated with disease progression and immune microenvironment. It has diagnostic and prognostic implications, indicating its potential in precision medicine and drug screening.

Keywords: PGAM1; glycolytic metabolism; immune infiltration; kidney renal clear cell carcinoma; single cell-RNA sequencing.

Figure 1

Gene landscape and characteristics of PGAM1 in KIRC. (A) PGAM1 gene expression levels in the male urinary system were examined. (B) The relationship between PGAM1 and nine highly mutated genes in KIRC was investigated, with mutation sites indicated by red lines. (C) The frequency of mutations was compared between PGAM1-high and PGAM1-low groups using Fisher’s exact test. Mutation types, driver mutation types, and groups are shown in the right panel. (D) A PGAM1 interaction network was generated using the Reactome database. (E, F) Univariate and multivariate Cox regression models were used to calculate hazard ratios for PGAM1 at different stages of KIRC.

Figure 2

Evaluation of the diagnostic potential of PGAM1 expression in KIRC biopsy specimens. (A) PGAM1 gene expression levels in renal cancer. (B) Comparison of PGAM1 expression between KIRC tumor and non-tumor tissues. Boxplots depicting PGAM1 expression levels across different stages (C), metastasis status (D), and ccRCC subtypes (E) of KIRC. (F) Comparative immunohistochemical analysis of PGAM1 expression in KIRC tissue samples from four different patients based on the Human Protein Atlas. (G) Prognostic significance of PGAM1 mRNA levels for overall survival, as determined using the Kaplan-Meier plotter dataset.

Figure 3

Resolving spatial relationships of cell type and gene expression using spatial transcriptomics in a mouse kidney injury model. (A) H&E-stained sections of 3 mouse models: sham operation, ischemia/reperfusion injury (IRI), and cecal ligation and puncture (CLP), respectively. Different regions of the cortex (Lrp2) and medulla (Aqp2) were labeled using tissue-specific biomarkers. (B) Analysis of Pgam1 and different biomarkers of renal injury (Lcn2, Kim1, Havcr1).

Figure 4

The co-expression genes of PGAM1 in KIRC were subjected to enrichment analysis. The target genes were analyzed using (A) Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway, (B) cellular component, focusing on (C) molecular function and (D) biological process.

Figure 5

The use of single-cell RNA sequencing analysis has allowed for the identification of immune cell populations. (A, B) The relative proportions of each cell type found within the two datasets, while showcasing the proportion of integrated immune cells present within the databases. UMAP is an abbreviation for the Unified Manifold Approximation and Projection technique used in this study. Visual representations of all KIRC cells from both GEO datasets are depicted in (C, D) through the utilization of Unified Manifold Approximation and Projection (UMAP) and assigned specific colors according to clusters.

Figure 6

The single-cell transcriptomes of patient-derived cultures treated with PGAM1 were presented. The expression clusters of PGAM1 were visualized using UMAP plots in (A, B), while the UMAP plots of each distinct cluster were analyzed through Gene Set Enrichment Analysis (GSEA).

Figure 7

The impact of PGAM1 on the Tumor Immune MicroEnvironment (TIME) was investigated. Immunological analyses of immune infiltrates and immunosuppressants were carried out using GSE159115 and GSE121636 databases, respectively, as shown in (A). Furthermore, a heatmap was presented to depict the correlation between PGAM1 expression and lymphocytic infiltration in human cancers (B).

Figure 8

Investigating the link between PGAM1 and immunization. (A) TIMER analysis determines the correlation between PGAM1 expression and six different immune cells in KIRC. (B) The correlation between PGAM1 levels and various types of macrophages is assessed. (C) Additionally, the correlation between PGAM1 and genes related to macrophages is investigated.

Figure 9

Analyzing the co-occurrence of PGAM1 and CD163 in tumor biopsies during KIRC development. (A) Merge indicates PGAM1/CD163/DAPI; inset indicates local magnification; 2.5D reconstructed image shows the local fluorescence changes, morphology and fluorescence intensity. The merge image displays PGAM1/CD163/DAPI, while the inset exhibits a local magnification. The 2.5D reconstructed image showcases local fluorescence changes, morphology, and fluorescence intensity. (B) The Pearson's correlation coefficient was employed to visualize the degree of overlap between PGAM1 and CD163 fluorescence signals.

Figure 10

Analysis of drug sensitivity and cytotoxicity in renal cancer cells. (A) The PGAM1 gene was queried in the pharmacogenetics database to identify gene signatures and potential drugs. Predictivity refers to the fold change in efficacy of short hairpin PGAM1 (shPGAM1), which indicates the efficiency of PGAM1 knockdown using shRNA, between cells with high and low response to the target drug. The drug sensitivity of the shPGAM1 gene to various chemical drugs was evaluated in KIRC cell lines. The boxplots show the log of the half maximal inhibitory concentration (IC₅₀) values of GNE-317 (B), MS-275 (C), AC45971100 (D), and NSC-35468 (E).

Figure 11

The proposed model depicts the potential significance of PGAM1 in various aspects of KIRC, such as diagnosis, prognosis, tumor immune microenvironment, and potential precision treatments.