Discovery and characterization of a novel telomerase alternative splicing isoform that protects lung cancer cells from chemotherapy induced cell death

Abstract

All cancer cells must adopt a telomere maintenance mechanism to achieve replicative immortality. Most human cancer cells utilize the enzyme telomerase to maintain telomeres. Alternative splicing of TERT regulates the amount and function of telomerase, however many alternative splicing isoforms of TERT have unknown functions. Single molecule long read RNA/cDNA sequencing of TERT revealed 45 TERT mRNA variants including 13 known and 32 novel variants. Among the variants, TERT Delta 2-4, which lacks exons 2-4 but retains the original open reading frame, was selected for further study. Induced pluripotent stem cells and cancer cells express higher levels of TERT Delta 2-4 compared to primary human bronchial epithelial cells. Overexpression of TERT Delta 2-4 enhanced clonogenicity and resistance to cisplatin-induced cell death. Knockdown of endogenous TERT Delta 2-4 in Calu-6 cells reduced clonogenicity and resistance to cisplatin. Our results suggest that TERT Delta 2-4 enhances cancer cells' resistance to cell death. RNA sequencing following knockdown of Delta 2-4 TERT indicates that translation is downregulated and that mitochondrial related proteins are upregulated compared to controls. Overall, our data indicate that TERT produces many isoforms that influence the function of TERT and the abundance and activity of telomerase.

Keywords: Alternative RNA splicing; Lung cancer; TERT; Telomerase.

Fig. 1

Long-read sequencing pipeline to discover novel TERT Delta 2–4 isoform and validation of Delta 2–4 expression. (A) Cartoon of full-length TERT exons. Four functional domains (TEN telomerase N-terminal, TRBD telomerase RNA-binding domain, RT reverse transcriptase, CTE C-terminal extension) and three localization signals (MTS mitochondrial-targeting signal in exon 1, NLS nuclear localization signal in exon2, NES nuclear export signal in exon 12) are shown. (B) Long-read sequencing pipeline to discover novel TERT isoforms. Four cell lines: induced pluripotent stem cells (iPSC-CHiPSC22), Calu-6, A549, and H1299 were pelleted from three biological replicates and RNA was extracted from each pellet. First strand cDNA was synthesized with oligo dT using superscript 4, followed by PCR with gene specific primers (GSP; TERT exons targeting primers) tailed with SSP or VNP (SSP-Exon 1 or VNP-Exon 16). After size exclusion of the TERT specific library, ONT (Oxford nanopore technology) barcodes and rapid 1D sequencing adapters were attached. Samples were loaded into MinION Mk1C and raw sequence files were processed by a bioinformatics pipeline consisting of Guppy, minimap2, TranscriptClean, TALON, SWAN, and R. (C) Heatmap showing expression levels of relatively abundant TERT isoforms. Numeric values are Log2-transformed TPM values and colors indicate expression level. Cancer average is the average of A549, Calu-6, and H1299. (D) Transcript models of the isoform (Left to right: 5ʹ to 3ʹ). Boxes indicate exons and solid lines between exons indicate introns. (E) Sanger Sequencing result confirmed TERT exon 2–4 splicing event from full length cDNA clone in pTOPO vector. (F) TERT Delta 2–4 expression levels in cell panels (determined by ddPCR; n = 3 biological replicates per condition). Total delta 2–4 indicates both exons 2–4 skipping variants including Delta 2–4 and delta2-4/delta 7–8, and Delta 2–4 indicates exons 2–4 skipping with exon 7/8 inclusion (note the capital D) (determined by ddPCR; n = 3 biological replicates per condition). (G) Percentage of localization between cytoplasm and nucleus was determined for nuclear non-coding RNA MALAT1, GAPDH, and TERT isoforms. Total delta 2–4: Delta 2–4 and delta 2–4/delta 7–8; Delta 2–4: exons 2–4 skipping with exons 7/8; Potential FL: exons 7/8 including TERT; Minus beta: exons 6/9 junction containing TERT; i11: intron 11 detained TERT; i14: intron 14 detained TERT (TERT isoforms: determined by ddPCR; MALAT1 and GAPDH: determined by gel-based PCR; n = 3 biological replicates). Unpaired t-test was performed to compare TERT mRNA variants’ percentage of localization compared to MALAT1 or GAPDH (^$$$$p < 0.0001 compared to MALAT1; ****p < 0.0001 and *p < 0.05 compared to GAPDH). Data are presented as means ± standard deviations where applicable.

Fig. 2

Impact of TERT Delta 2–4 isoform on growth rate, telomere length, and telomerase activity. (A) Total TERT delta 2–4 (Delta 2–4 and delta 2–4/delta 7–8) was measured following 48 h of siRNA treatment (determined by ddPCR; n = 6 biological replicates). (B) TERT Delta 2–4 expression level was measured following 48 h of siRNA treatment (determined by ddPCR; n = 6 biological replicates). (C) Telomerase activity did not change with TERT Delta 2–4 knockdown (48 h) in Calu-6 cells. Telomerase activity was determined by ddTRAP (n = 6 biological replicates per condition). (D) Flag-V5-Delta 2–4 expression was confirmed by TERT Y182 antibody, V5 antibody, and Flag antibody in Calu-6 cells (left) and U-2 OS cells (right). Beta-actin was used as a loading control. (E) Telomerase activity did not change with TERT Delta 2–4 overexpression in Calu-6 cells. Telomerase activity was determined by ddTRAP (n = 11 biological replicates per condition). (F,G) Telomere length did not change with TERT Delta 2–4 overexpression in Calu-6 (F) and U-2 OS (G) (determined by terminal restriction fragment (TRF) assay). Student’s t-test set at *p ≤ 0.05 for significance compared to siRNA control conditions. Data are presented as means ± standard deviations where applicable.

Fig. 3

Impact of TERT Delta 2–4 isoform on clonogenicity and resistance to cisplatin. (A,B) Overexpression Delta 2–4 in telomerase deficient cell U-2 OS enhanced clonogenicity. Representative image (A) and quantification (B) are shown. (C,D) Knockdown of TERT Delta 2–4 reduced clonogenicity in telomerase positive cells Calu-6, whereas overexpression of TERT Delta 2–4 enhanced clonogenicity in Calu-6. Knockdown effect was rescued by siRNA resistant Delta 2–4 overexpression. Representative image (C) and quantification (D) are shown. (E) Calu-6 cells became more sensitive to cisplatin treatment (1.59 ug/mL, 48 h) by TERT Delta 2–4 knockdown and more resistant by TERT Delta 2–4 overexpression. Knockdown effect was rescued by siRNA resistant Delta 2–4 overexpression. For multiple group comparisons, p-value was calculated by Tukey’s multiple comparisons test following one-way ANOVA (ns > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001). Data are presented as means ± standard deviations where applicable.

Fig. 4

Transcriptomics analyses following TERT Delta 2–4 knock down in Calu-6 cells. (A) A heatmap visualizing clusters of differentially expressed genes (q < 0.1; 24 upregulated genes and 55 downregulated genes) in TERT Delta 2–4 knocked down Calu-6 cells. (B) Visualization of differentially expressed genes by volcano plot showing -log(p valule) and LFC (log twofold change). Dashed lines are at p value = 0.01 and LFC = 1.5 or − 1.5. Significantly differentially expressed seven genes are indicated (p < 0.01 and LFC <  − 1.5 or > 1.5). (C) Four significantly (q < 0.1) enriched biological processes by TERT Delta 2–4 were revealed by gene ontology analysis using differentially expressed genes. Two biological processes in red colors (cytoplasmic translation and translation) are related to downregulated genes, and the other two biological processes in blue colors (respiratory chain and electron transport) are related to upregulated genes with Delta 2–4 knockdown. (D) Upregulated DKK1 expression by TERT Delta 2–4 knockdown was rescued by siRNA resistant Delta 2–4 overexpression (determined by ddPCR; n = 3 biological replicates) in Calu-6 cells. For multiple group comparisons, p-value was calculated by Šídák's multiple comparisons test following one-way ANOVA (ns > 0.05, ***p < 0.001). Data are presented as means ± standard deviations where applicable.