Multi-omics analysis of bidirectional liquid-liquid phase separation signatures reveals immunotherapy response and prognosis in renal cell carcinoma

Abstract

Renal cell carcinoma (RCC) is a common malignant tumor of the urinary system, accounting for approximately 80% of urinary tract malignancies, with clear cell renal cell carcinoma (ccRCC) being the most prevalent subtype. Despite its high incidence, the pathogenesis of ccRCC remains incompletely elucidated, and limitations persist in early diagnosis, targeted therapy, and immunotherapy. Based on emerging evidence highlighting the critical role of liquid-liquid phase separation (LLPS) in tumor proliferation and invasion, this study leveraged multiple ccRCC cohorts to employ Mendelian randomization (MR) and ultimately developed a novel Bidirectional Liquid-Liquid Phase Separation-associated Index (B-LLPSI) using LASSO regression, aiming to predict patient prognosis and therapeutic response. Notably, B-LLPSI demonstrates substantial potential as a universal prognostic indicator across multiple cancer types, underpinning its prospective application in individualized oncology. Paradoxically, while high B-LLPSI levels correlate with poorer patient prognosis, they simultaneously predict enhanced responsiveness to immunotherapy. Furthermore, the B-LLPSI signature was found to be upregulated in tumor tissues and capable of promoting tumor proliferation and invasion. In summary, B-LLPSI holds promise as a robust and independent prognostic tool for diverse cancers, providing critical insights for formulating individualized treatment strategies.

Keywords: Immune microenvironment; Immunotherapy efficient; Liquid-liquid phase separation; Molecular docking; Pancancer analysis; Prognosis; Renal clear cell carcinoma.

Fig. 1

Screening for characteristic liquid-liquid phase separation genes in ccRCC. A Screening for ccRCC signatures using SMR. B Intersection of ccRCC signatures with LLPS differentially expressed genes. C Heatmap of Intersecting genes in (B).(D) Unicox analyses based on TCGA and E-MATB-1980 cohorts

Fig. 2

Construction and validation of Bidirectional Liquid-Liquid Phase Separation-associated Index (B-LLPSI). A Error plots for cross-validation of Lasso regressions. B Correlation coefficients of B-LLPSI genes. C B-LLPSI gene expression landscape in pan-cancer. D and E Prognostic prediction efficacy of B-LLPSI and B-LLPSI genes expression landscape in TCGA (training set) and E-MTAB-1980 (validation set). F Nomogram based on B-LLPSI, ccRCC clinical staging and pathological grading. G Unicox analysis (blue) and multicox analysis (red) based on B-LLPSI and multiple clinical traits in ccRCC patients. H Correlation of B-LLPSI with clinical traits and immunophenotyping of ccRCC patients

Fig. 3

Comparison of the immune microenvironment and immunotherapeutic effects between high and low B-LLPSI groups. Differences of TIDE score (A) and IPS scores (B) between high and low B-LLPSI groups. Differences in immune cell (C) and immune function scores (D) between high and low B-LLPSI groups. (E) Differences in immune checkpoint expression between high and low B-LLPSI groups. (F) B-LLPSI difference between 6 immune subtypes. (G) Heatmap of B-LLPSI and B-LLPSI gene expression correlating with immune cell infiltration. (H) Comparison of tumor microenvironment scores between high and low B-LLPSI groups. (I) Differences in survival and immunotherapy response between high and low B-LLPSI groups using the IMvigor 210 immunotherapy cohort as a validation set

Fig. 4

Correlation of B-LLPSI with gene mutations. Waterfall plots of gene mutations in low (A) and high (B) B-LLPSI populations. C Survival analysis of different tumor mutation burden(TMB) and B-LLPSI populations. D Correlation of B-LLPSI with tumor mutation burden. E Mutation rates of genes constructing B-LLPSI in renal clear cell carcinoma and non-clear cell carcinoma. Summary of mutant landscapes in the high (F) and low (G) B-LLPSI groups

Fig. 5

Screening of renal cancer-sensitive small molecule drugs based on B-LLPSIs. A Correlation between GDSC drug sensitivity and gene mRNA expression. B Correlation between CTRP drug sensitivity and gene mRNA expression. C Correlation of drug sensitivity with B-LLPSI signatures. D Differences in sensitivity to commonly used chemotherapeutic agents between high and low B-LLPSI groups. E Screening of small molecule drugs targeting TTK using molecular docking

Fig. 6

Predictive value of B-LLPSI in pan-cancer: A Overall Survival. B Disease Specific Survival. C Progression Free Interval. D Disease Free Interval

Fig. 7

TTK, DTL, and GPRIN1 were highly expressed in renal cancer cells, and knockdown of TTK inhibited the proliferation and migration of RCC cells. A–C The expression of TTK, DTL and GPRIN1 in HK2, 786-O, 769-P, CAKI-1, ACHN and OSRC-2 cells was detected by Realtime PCR. D The expression of TTK, DTL, GPRIN1 in HK2, 786-O, 769-P, CAKI-1, ACHN and OSRC-2 cells was detected using Western blot. E, F Expression of TTK, DTL, GPRIN1 in 786-O and 769-P after transfection with siRNA. EdU cell proliferation capacity assay G, J Transwell cell migration capacity (H, K,n = 3), and wound healing capacity assay (I, L) in 786-O and 769-P cells transfected with two si-TTK, respectively. Note* p < 0.05, **p < 0.01, ***p < 0.001,****p < 0.0001. Data are mean ± SD of three independent experiments.All experiments were performed in triplicate

Fig. 8

Knockdown of DTL, GPRIN1 inhibits RCC cell proliferation and migration. EdU cell proliferation capacity assay (A, D), Transwell cell migration capacity (B, E), and wound healing capacity assay (C, F) in 786-O and 769-P cells transfected with two si-DTL, respectively. EdU cell proliferation capacity assay (G, J), Transwell cell migration capacity (H, K), and wound healing capacity assay (I, L) in 786-O and 769-P cells transfected with two si-GPRIN1, respectively. Note* p < 0.05, **p < 0.01, ***p < 0.001,****p < 0.0001. Data are mean ± SD of three independent experiments. All experiments were performed in triplicate