Identifying Cancer Driver lncRNAs Bridged by Functional Effectors through Integrating Multi-omics Data in Human Cancers

Abstract

The accumulation of somatic driver mutations in the human genome enables cells to gradually acquire a growth advantage and contributes to tumor development. Great efforts on protein-coding cancer drivers have yielded fruitful discoveries and clinical applications. However, investigations on cancer drivers in non-coding regions, especially long non-coding RNAs (lncRNAs), are extremely scarce due to the limitation of functional understanding. Thus, to identify driver lncRNAs integrating multi-omics data in human cancers, we proposed a computational framework, DriverLncNet, which dissected the functional impact of somatic copy number alteration (CNA) of lncRNAs on regulatory networks and captured key functional effectors in dys-regulatory networks. Applying it to 5 cancer types from The Cancer Genome Atlas (TCGA), we portrayed the landscape of 117 driver lncRNAs and revealed their associated cancer hallmarks through their functional effectors. Moreover, lncRNA RP11-571M6.8 was detected to be highly associated with immunotherapeutic targets (PD-1, PD-L1, and CTLA-4) and regulatory T cell infiltration level and their markers (IL2RA and FCGR2B) in glioblastoma multiforme, highlighting its immunosuppressive function. Meanwhile, a high expression of RP11-1020A11.1 in bladder carcinoma was predictive of poor survival independent of clinical characteristics, and CTD-2256P15.2 in lung adenocarcinoma responded to the sensitivity of methyl ethyl ketone (MEK) inhibitors. In summary, this study provided a framework to decipher the mechanisms of tumorigenesis from driver lncRNA level, established a new landscape of driver lncRNAs in human cancers, and offered potential clinical implications for precision oncology.

Keywords: cancer drivers; cancer hallmarks; copy number alterations; immunosuppression; long non-coding RNAs.

Figure 1

The Overview of DriverLncNet (A) Construction of binary copy number profile of candidate driver lncRNAs. (B) Identification of dys-regulatory networks by differential network analysis integrating multi-omics data. (C) Identification of functional effectors of candidate driver lncRNAs based on non-coding regulatory mechanisms. (D) Determination of final driver lncRNAs using random walk and network distance. C-Lnc, PLSR, and PPI networks indicate the candidate driver lncRNAs, partial least-squares regression, and protein-protein interaction network, respectively.

Figure 2

Driver lncRNAs and Their Functional Effectors in Human Cancers (A) Copy number ratio in CNA samples (y axis), influence of copy number on expression (x axis; |log2(FC)|, where FC = mean expression [CNA samples]/mean expression [wild-type samples]), and alteration frequency (circle size) for each driver lncRNA. (B) The functional effectors in dys-regulatory network induced by ANRIL deletion in GBM. (C) Enrichment map for functional effectors of ANRIL.

Figure 3

Cancer Hallmarks and Driver lncRNAs with Functional Effectors (A) Enrichment map of functional effectors of PVT1 using an enrichment tool gProfiler (p ≤ 0.05). The size of the red circle is proportional to the size of a functional gene set. (B) EMT- and apoptosis-associated pathways and molecular markers correlated to PVT1 expression (blue for low expression or pathway activity, red for high; Pearson’s correlation, *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001). (C) Apoptosis pathway activity in PVT1 low- and high-expression groups (two-sided Wilcoxon test). (D) BiPartite graph in five tumor types. The curve between a cancer hallmark and a driver lncRNA indicates that the lncRNA is associated with the cancer hallmark. The percent under the icon of cancer hallmark indicates the fraction of GO terms of each hallmark. (E) Three cancer hallmarks of driver lncRNA PVT1.

Figure 4

Cancer Immunity of Driver lncRNAs (A) Fold change (FC) of mean gene expression in patients with high or low expression of each lncRNA across 5 cancer types. The size of the red (or blue) circle is proportional to the positive (or negative) log-fold change between patients with high or low expression. (B) Distribution of immunosuppressive scores in patients with high (red) or low (gray) expression of each lncRNA (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, two-sided Wilcoxon test). (C) Immunosuppressive score in patients with high (red) or low (blue) expression of RP11-571M6.8 in GBM. (D) PD-1, PD-L1, and CTLA4 expressions in patients with high (red) or low (blue) expression of RP11-571M6.8 in GBM. (E) Immune cell infiltration level and immune-related driver lncRNAs in 3 independent datasets across 5 cancer types (two-sided Wilcoxon test for immune cell infiltration level of patients with high and low expression, p ≤ 0.05). (F) Radar chart of 5 immune cells’ infiltrations in patients with high or low expression of RP11-571M6.8 in GBM in 3 three independent datasets (two-sided Wilcoxon test). (G) The correlation between expression of Treg cell markers and RP11-571M6.8 expression in GBM.

Figure 5

The Prognosis Value of Driver lncRNAs (A and E) KM survival curve for tumor patients with high and low expressions of (A) RP11-1020A11.1 or (E) FAM83A-AS1 for OS and DFS in BLCA. (B and F) Distribution of (B) RP11-1020A11.1 or (F) FAM83A-AS1 expression in diploid and amplified patients. (C and G) Enrichment map for functional effectors of (C) RP11-1020A11.1 or (G) FAM83A-AS1. (D and H) Results of multivariable cox regression model for OS and DFS in (D) BLCA or (H) LUAD. Red and green indicate risk factor and favorable factor, respectively.

Figure 6

The Association of Driver lncRNAs and Anti-cancer Drugs (A) Correlation between drug half-maximal inhibitory concentration (IC₅₀) and lncRNA copy number status in CGP. Dot size is proportional to the p value of correlation; dot color indicates the drug response effect (drug resistance, red; and sensitivity, green) in CNA samples compared to diploid for a specific lncRNA. (B) Selumetinib has a high IC₅₀ in samples with CTD-2256P15.2 amplification in cell models (left) and predicted spectrum (middle and right). (C) Copy number status of lncRNA PVT1 and drug zibotentan. (D) lncRNA CTD-2195M18.1 and drug GDC-0941. (E) lncRNA NKX2-1-AS1 and drug gemcitabine.