Integrating single-cell RNA sequencing with spatial transcriptomics reveals immune landscape for interstitial cystitis

Abstract

Interstitial cystitis (IC) is a severely debilitating and chronic disorder with unclear etiology and pathophysiology, which makes the diagnosis difficult and treatment challenging. To investigate the role of immunity in IC bladders, we sequenced 135,091 CD45⁺ immune cells from 15 female patients with IC and 9 controls with stress urinary incontinence using single-cell RNA sequencing (scRNA-seq). 22 immune subpopulations were identified in the constructed landscape. Among them, M2-like macrophages, inflammatory CD14⁺ macrophages, and conventional dendritic cells had the most communications with other immune cells. Then, a significant increase of central memory CD4⁺ T cells, regulatory T cells, GZMK⁺CD8⁺ T cells, activated B cells, un-switched memory B cells, and neutrophils, and a significant decrease of CD8⁺ effector T cells, Th17 cells, follicular helper T cells, switched memory B cells, transitional B cells, and macrophages were noted in IC bladders. The enrichment analysis identified a virus-related response during the dynamic change of cell proportion, furthermore, the human polyomavirus-2 was detected with a positive rate of 95% in urine of patients with IC. By integrating the results of scRNA-seq with spatial transcriptomics, we found nearly all immune subpopulations were enriched in the urothelial region or located close to fibroblasts in IC bladders, but they were discovered around urothelium and smooth muscle cells in control bladders. These findings depict the immune landscape for IC and might provide valuable insights into the pathophysiology of IC.

Fig. 1

Overview of workflow for single-cell RNA-sequence (scRNA-seq) and spatial transcriptomics (ST). a CD45⁺ single cells from IC (n = 15) and control (n = 9) bladders were collected to construct an immune landscape. ST views were established based on 3 IC bladders and 3 control bladders. The constructed atlases were combined via multimodal intersection analysis. b Pathological features of bladders of IC patients and controls were presented by H&E images (bar = 200 μm). c Violin plots showing the quality control of single-cell data. d A high correlation coefficient of 0.8995 between cell counts and genes was observed, but not in cell counts and mitochondrial genes. e Canonical marker genes for immune cell subsets. f The expression of selected marker genes in the immune landscape. g UMAP plot presenting the immune landscape of IC bladders. h The proportion of each cell type. IC interstitial cystitis, UMAP uniform manifold approximation and projection, CD8⁺ Tef CD8⁺ effector T cell, CD4⁺ Tcm central memory CD4⁺ T cell, Treg regulatory T cell, CD4⁺ Tem effector memory CD4⁺ T cell, Tfh follicular helper T cell, Act CD8⁺ T activated CD8⁺ T cell, NK CD56^bright CD16^low natural killer cell, Act B activated B cell, SMB switched memory B cell, USMB un-switched memory B, PLASMA plasma cell, TBC transitional B cell, Neu neutrophil, CD14⁺ M inflammatory CD14⁺ macrophage, 2-M M2-like macrophage, cDC conventional dendritic cell, pDC plasmacytoid dendritic cell, MAST mast cell, Fib fibroblast, EPC epithelial cell, ENC endothelial cell

Fig. 2

Focused analysis of T and B cells. a The expression of CD3E, CD4, CD8A, and NCAM1 across the clusters. b Heatmap showing the expression of selected marker genes across the T cell clusters. c The changes of proportion in T cell subsets between the two groups (***P < 0.001). d Bar chart showing the significant GO and KEGG terms of Tregs in the IC group compared with the control group. e Significant KEGG terms of Th17 in the IC group compared with the control group. f The relative expression level of all exhausted genes in CD8⁺ T cells between the IC and control groups. g The expression level of representative exhausted markers in the two groups. h Violin plot showing the expression of B cell marker gene (CD79A) across the clusters. i Heatmap showing the expression of selected marker genes across the B cell clusters. j The changes of proportion in B cell subsets between the two groups (***P < 0.001). k The relative expression level of aging-related genes in B cells between the IC and control groups. l The expression level of immunoglobulins including IgA, IgM, IgG, and IgE in plasma or urine of IC patients and controls (mean ± standard deviation, Mann–Whitney U-test, two-tailed; IC, n = 20, control, n = 14). IC interstitial cystitis, CD8⁺ Tef CD8⁺ effector T cell, CD4⁺ Tcm central memory CD4⁺ T cell, Treg regulatory T cell, CD4⁺ Tem effector memory CD4⁺ T cell, Tfh follicular helper T cell, Act CD8⁺ T activated CD8⁺ T cell, NK CD56^bright CD16^low natural killer cell, Act B activated B cell, SMB switched memory B cell, USMB un-switched memory B, PLASMA plasma cell, TBC transitional B cell

Fig. 3

Focused analysis of myeloid cells. a Violin plots showing the expression level of selected marker genes across the clusters. b The changes of proportion in myeloid-cell subsets between the two groups (***P < 0.001). c Significant GO and KEGG terms of neutrophils in the IC group compared with the control group. d, e Pseudotime analysis depicted the developmental trajectory from CD14⁺ macrophages to M2-like macrophages. f Scatter plots showing the expression changes of inflammatory genes (IL-1β, S100A8, and NLRP3) and phagocytosis genes (CD163, FOLR2, STAB1, C1QA, and MSR1) from CD14⁺ macrophages to M2-like macrophages over time. g Functional analysis of DEGs of inflammatory CD14⁺ macrophages. h Functional analysis of DEGs of M2-like macrophages. i Disease enrichment analysis of DEGs of M2-like macrophages. j Immunofluorescence staining of CD68, S100A8, and CD163 on frozen sections (Det detrusor, Int interstitial layer; IC, n = 15; control, n = 9). IC interstitial cystitis, DEGs differentially expressed genes

Fig. 4

Mass cytometry (CyTOF) confirmed the phenotypical heterogeneity of immune cells in IC bladders. a Overview of workflow for CyTOF (IC, n = 5; control, n = 5). b The heatmap showed the relative protein expression level of selected markers in bladder tissues, and generated clusters are visualized by T-SNE plots (c). d Feature plots showing the expression of the canonical marker gene (CD45) across immune cells. e The selected markers of T cell (CD3), CD4⁺ T cell (CD4), CD8⁺T cells (CD8), B cell (CD19), macrophage (CD68), dendric cell (CD11C), mast cell (C-KIT), and activated gene (HLA-DR) were detected in the IC and control groups. IC interstitial cystitis

Fig. 5

Cellphone DB analysis based on scRNA-seq showed the crosstalk in the immune network. a Heatmap showing the distribution of interaction pairs across the cell types. b Cell–cell communication network in IC based on the top 50 interaction pairs. c Violin plot showing the expression of interaction pairs including CCL15-CCR1 on epithelial cells and FGF10-FCGR1 on fibroblasts. d The expression of inflammatory genes including TNF, IL-1β, and IFNG across the clusters. e Violin plot showed the upregulated expression of CCL5 across immune cells, which was validated by Enzyme-linked-immunosorbent serologic assay tests in urine (P < 0.0001), but not in blood (P = 0.6595) (Mann–Whitney U-test, two-tailed; IC, n = 20; control, n = 14). f The downregulated expression level of IL-10 and IL-35 across immune cells. g The differentially expressed genes (DEGs) in Tregs in the IC group compared with the control group were enriched and upregulated in TGF-β signaling pathway. h Enrichment analysis of DEGs of epithelial cells. i Enrichment analysis of DEGs of fibroblasts. j Top 5 transcription factors of epithelial cells. k Bar plot showing the virus-related enriched results in selected cell types. l Human polyomavirus-2 was detected in IC urine with 95% positive rates, but 0% in control urine (IC, n = 20; control, n = 14). IC interstitial cystitis, CD8⁺ Tef CD8⁺ effector T cell, CD4⁺ Tcm central memory CD4⁺ T cell, Treg regulatory T cell, CD4⁺ Tem effector memory CD4⁺ T cell, Tfh follicular helper T cell, Act CD8⁺ T activated CD8⁺ T cell, NK CD56^bright CD16^low natural killer cell, Act B activated B cell, SMB switched memory B cell, USMB un-switched memory B, PLASMA plasma cell, TBC transitional B cell, Neu neutrophil, CD14⁺ M inflammatory CD14⁺ macrophage, 2-M M2-like macrophage, cDC conventional dendritic cell, pDC plasmacytoid dendritic cell, MAST mast cell, Fib fibroblast, EPC epithelial cell, ENC endothelial cell

Fig. 6

Mapping distinct immune populations across IC bladder sections using multimodal intersection analysis (MIA). a–f for the first IC bladder. a The spatial transcriptomics (ST) map of IC bladder (bar = 500 μm). b The correlation between the Genes and UMIs was calculated. c Spatial plots showing the marker genes for smooth muscle cells (SMCs), urothelial cells, myofibroblasts, interstitial cells (ICs), and mast cells. d T-SNE plot showing the clusters in the ST map of IC bladder. e Violin plots showing the expression of fibrotic genes (FN1, TGF-β1, COL1A1, and COL3A1) across the clusters. f Heatmap showing the distribution of immune cells in the ST map (the most mapped cell populations were labeled with purple box). g–l for control bladder, using the same presentation form with IC bladder. IC interstitial cystitis