Multi-omics integration analysis identifies INPP4B as a T-cell-specific activation suppressor

Abstract

Background: Naïve T cells are maintained in a quiescent state prior to activation. As inappropriate T-cell activation can lead to impaired immune tolerance and autoimmune diseases, the transition from quiescence to activation must be under strict regulation. Despite its importance, the mechanisms underlying the maintenance of the quiescent state remain incompletely understood.

Methods and results: Through multi-omics integration analysis, we reveal that INPP4B, a phosphatase of the phosphoinositide 3-kinase pathway, is highly expressed specifically in T cells and is involved in suppressing T-cell activation and maintaining quiescence. Our findings uncover that INPP4B forms a T-cell-specific chromatin interaction domain and exhibits high expression levels in quiescent T cells. Upon T-cell activation, both the chromatin interaction and expression levels of INPP4B decrease. Functional studies further confirm that INPP4B suppresses T-cell activation and effector functions. Additionally, we observe increased expression level of INPP4B in exhausted T cells within the tumour microenvironment.

Conclusion: These results highlight the importance of maintaining optimal levels of INPP4B for T-cell function. Our findings suggest that INPP4B could be a potential target for enhancing the efficacy of T-cell-mediated immune responses against tumours.

Key points: A comprehensive multi-omics analysis characterizes the expression patterns of INPP4B across immune populations. INPP4B exhibits a T-cell-specific expression domain and functions as a T cell activation suppressor. INPP4B is significantly upregulated in exhausted T cells within the tumour microenvironment.

Keywords: INPP4B; T‐cell activation; multi‐omics.

FIGURE 1

INPP4B is downregulated in activated T cells. (A) Schematic of phosphatases in phosphoinositide 3‐kinase (PI3K) pathway. (B) Expression level of INPP4B in naïve and memory CD4⁺ T cells upon activation. Naïve T cells, CD4⁺CD45RO⁻CD27⁺; central memory T cells, CD4⁺CD45RO⁺CD27⁺; effector memory T cells, CD4⁺CD45RO⁺CD27⁻. Re‐analysis results of published datasets. (C) Transcription level and chromatin accessibility around INPP4B in naïve CD4⁺ T cells upon activation. Re‐analysis results of published datasets. (D) Chromatin interactions around INPP4B in naïve CD4⁺ T cells upon activation. Re‐analysis results of published datasets. (E) Expression level of Inpp4b in mouse T cells upon activation (n = 4). p‐Value was calculated with two‐sided Wilcoxon test. ^* p < .05. (F) Relative expression level of Inpp4b in mouse CD4⁺ T subsets. Th, helper T cells; Treg, regulatory T cells. Re‐analysis results of published datasets.

FIGURE 2

INPP4B suppresses T‐cell activation and effector functions. (A) Schematic of Inpp4b overexpression (OE) and knockdown (KD) in mouse T cells. (B) Expression level of Inpp4b in EV‐control (empty plasmid vector [EV] transfected cells), OE, shControl (non‐targeting shRNA control) and KD mouse CD8⁺ T cells (n = 4). (C) Percentage of CD25⁺, CD44⁺ and CD69⁺ subsets in anti‐CD3/CD28‐activated EV‐control, OE, shControl and KD mouse CD8⁺ T cells (n = 4). (D) Percentage of Ifng⁺, Tnfa⁺, Gzmb⁺ subsets in anti‐CD3/CD28‐activated EV‐control, OE, shControl and KD mouse CD8⁺ T cells (n = 4). (E) Expression level of Mki67, Ifng, Tnf and Gzmb in anti‐CD3/CD28‐activated EV‐control, OE, shControl and KD mouse CD8⁺ T cells (n = 4). (F) Accumulated proliferation of anti‐CD3/CD28‐activated EV‐control, OE, shControl and KD mouse CD8⁺ T cells. p‐Value was calculated with two‐sided Wilcoxon test. ^* p < .05.

FIGURE 3

INPP4B overexpression suppresses phosphoinositide 3‐kinase (PI3K)‒protein kinase B (AKT) pathway. (A) Enriched pathways of downregulated genes in overexpression (OE) group. (B) Expression heatmap of representative downregulated genes in PI3K‒AKT pathways.

FIGURE 4

INPP4B is actively expressed in T cells. RNA (A) and protein (B) levels of INPP4B in human immune cells (n = 4). Re‐analysis results of published datasets., (C) Chromatin accessibility around INPP4B in human immune cells. The top panel is the interaction heatmap around INPP4B gene‐body‐associated domain (GAD) structure in CD4⁺ T cells. (D) Chromatin interactions around INPP4B in human immune cells. (E) GAD score of all genes in human immune cells. Genes in phosphoinositide 3‐kinase (PI3K) pathway were marked out. (F) GAD score of members in PI3K pathway in human immune cells. (C‒F) Re‐analysis results of published datasets., , ,

FIGURE 5

BCL11B regulates the transcription of INPP4B. (A) Schematic of analysis pipeline for screening transcription factors (TFs) whose expression levels positively correlated with INPP4B. Line plot of the expression levels of Inpp4b and Bcl11b during T‐cell development (B) and upon T‐cell activation (C). Re‐analysis results of published datasets., Expression levels of Bcl11b (D) and Inpp4b (E) in the wildtype (WT) and Bcl11b‐knockdown (Bcl11b‐KD) naïve CD4⁺ T cells (n = 3). p‐Value was calculated with two‐sided Wilcoxon test. ^* p < .05. (F) Bcl11b binding signal and chromatin accessibility around Inpp4b in the WT and Bcl11b‐KD naïve CD4⁺ T cells. TSS, transcription start site. (G) Chromatin interaction heatmap around Inpp4b in the WT and Bcl11b‐KD naïve CD4⁺ T cells. (D‒G) Re‐analysis results of published datasets.

FIGURE 6

INPP4B is dysregulated in T cells under disease conditions. (A) Expression level of Inpp4b in naïve and exhausted OT‐I T cells isolated from mouse B78ChOVA melanomas. p‐Value was calculated with two‐sided Wilcoxon test. ^* p < .05. Re‐analysis results of published datasets. (B) UMAP projections of gene expression level for INPP4B and PDCD1. Re‐analysis results of published datasets. (C) Expression levels of INPP4B and PDCD1 in exhausted T cells with higher (exhaustion score > 0) and lower (exhaustion score < 0) exhaustion score. p‐Value was calculated with two‐sided Wilcoxon test. ^*** p < .001. Re‐analysis results of published datasets. (D) Expression levels of INPP4B in normal T cells, early T‐lineage progenitor acute lymphoblastic leukaemia (ETP‐ALL) and T‐cell acute lymphoblastic leukaemia (T‐ALL) cells. p‐Value was calculated with two‐sided Wilcoxon test. ^** p < .01. (E) Chromatin interactions around INPP4B in normal T cells, ETP‐ALL and T‐ALL cells. (F) INPP4B gene‐body‐associated domain (GAD) score in normal T cells, ETP‐ALL and T‐ALL cells. p‐Value was calculated with two‐sided Wilcoxon test. ^* p < .05. (D‒F) Re‐analysis results of published datasets. (G) Kaplan‒Meier overall survival curves for expression level of INPP4B in T‐ALL patients. (H) Schematic of expression levels of INPP4B in normal, exhausted and leukaemia T cells.