Rosselló‑Tortella et al. Molecular Cancer (2022) 21:48 https://doi.org/10.1186/s12943-022-01532-w LETTER TO THE EDITOR Open Access DNA methylation‑associated dysregulation of transfer RNA expression in human cancer Margalida Rosselló‑Tortella1,2, Alberto Bueno‑Costa1, Laura Martínez‑Verbo1, Lorea Villanueva1 and Manel Esteller1,3,4,5,6* The human cytoplasmatic pool of tRNA for the 20 pro- transcription, and therefore may contribute to this teinogenic amino acids and selenocysteine is composed specificity. of 48 isoacceptor families –those tRNA with different The importance of tRNAs in cell physiology is empha- anticodons– divided into 253 different isodecoder spe- sized by the discovery that these molecules and their cies –those tRNAs that share the same anticodon but derived fragments are distorted in pathological pro- present sequence variations in other positions [1, 2]. All cesses, including cancer [7]. Cancer-associated tRNA these molecules cooperate to translate the genetic infor- deregulation has been overlooked for many years, but mation encoded in mRNA to enable protein synthesis. now these molecules have been directly connected to For many years, tRNAs have been considered as house- tumorigenesis and are postulated to undertake active keeping molecules without any additional regulatory roles in this process [7]. Tumor-related alterations in function, but compelling recent evidence of the intricacy tRNA occur throughout their bioprocessing, from tran- of tRNA biology have proven that this initial misconcep- scription to nucleoside modification and cleavage [7]. tion was far from reality. In fact, tRNAs actively engage in For instance, the altered expression of tRNA modifier protein synthesis regulation and in additional molecular enzymes reshapes the tRNA modification landscape and processes that are unrelated to translation, like apoptosis contributes to the disease [8]. The increased abundance prevention and the generation of small derivative non- of tRNAs in tumoral cells was first esteemed to be a natu- coding RNAs that perform further cellular functions [2]. ral outcome of the increased protein synthesis rate that The advances in high-throughput sequencing tech- is intrinsic of highly proliferative cells. This is in part due nologies have demonstrated that tRNAs are differentially to the number of oncogenic and tumor suppressive path- transcribed and expressed among tissues [3]. Such intri- ways that promote RNAPIII activity and lead to a global cacy in tRNA expression regulation cannot be exclusively boost in tRNA expression [7]. tRNA overexpression in explained by the global modulation of the RNA poly- tumoral cells is extensive to all tumor types, but not all merase III (RNAPIII) transcriptional activity. Thus, the tRNA species are equally altered [9]. The mechanisms definition of additional mechanisms is needed to clarify that orchestrate this specific control of tRNA expres- such precision in tRNA expression control. Nucleo- sion in cancer cells are not completely understood. These some positioning [4], histone modifications [5], and the changes in the tRNA pool composition facilitate the transcription of neighboring genes by the RNA poly- translation of transcripts with certain codons [10] and merase II (RNAPII) [6] are known to influence RNAPIII alter the abundance of small tRNA-derived fragments to foster tumorigenesis [11]. This is reflected in the differ- ential patient survival or drug response that arise from *Correspondence: mesteller@carrerasresearch.org alterations in tRNAs, their modifier enzymes and in their 6 Josep Carreras Leukaemia Research Institute (IJC), Carretera de Can Ruti, derived fragments [8, 12]. Camí de les Escoles s/n, 08916 Badalona, Barcelona, Catalonia, Spain Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​ mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Rosselló‑Tortella et al. Molecular Cancer (2022) 21:48 Page 2 of 8 Alterations in DNA methylation constitute a frequent Figure S1B). The separation of TCGA and cell line sam- mechanism by which tumor cells inactivate the expres- ples according to their origin showed that the different sion of many coding and non-coding genes to acquire normal and tumoral tissues also presented variations in their malignant features [13]. This epigenetic mechanism tDNA methylation levels according to the cancer type is reported to hamper the binding of the RNAPIII general (Fig. 1A). These findings indicate that tumorigenesis transcription factor 3 C (TFIIIC) [14] to DNA and pre- involves epigenetic lesions in tDNA genes that are char- vent tRNA transcription in vitro [15]. However, the real acteristic among different tissues. degree of DNA methylation involvement in tRNA expres- sion regulation remains unexplored. Herein, we show tDNA methylation is associated with reduced tRNA for the first time that variations in DNA methylation at expression in cancer cells tDNA influence tRNA expression and may account for The differences in tDNA methylation status between nor- the specific alterations in tRNA levels observed in cancer mal and malignant samples and among tissues prompted cells that can enhance the progression of the disease. us to investigate whether this phenomenon could drive cancer-associated and tissue-specific variations in the Results and Discussion expression of specific tRNAs. To address this, we evalu- Shifts in the DNA methylation patterns of tDNAs occur ated the in silico correlation between tDNA methyla- in human tumorigenesis tion and the expression of their cognate tRNA using a To identify DNA methylation defects in tDNA, we publicly available dataset of tRNA expression in TCGA retrieved the Infinium HumanMethylation450 (HM450) samples [9]. Three tRNAs presented a statistically signifi- methylation microarray probes’ identity that mapped cant Spearman’s correlation coefficient (ρ ≤ -0.2) when into the loci corresponding to the 416 high confidence considering all the samples: tRNA-Arg-TCT-4-1, tRNA- tDNA that are found in the human nuclear genome [1]. Ile-AAT-8-1, and tRNA-Val-CAC-2-1 (Fig. 1B). Since 138 tDNAs were represented in the HM450 microar- tDNA methylation presented differences depending on ray, and 95 of them were located further than 2 kb from the origin of the sample (Fig. 1 A), we also investigated RNAPII-transcribed genes (Supplementary Figure S1A). if the association between tDNA methylation and tRNA We established this exclusion criterion because the inter- expression was dependent on the tissue (Fig. 1B). From ferences mediated by elongating RNAPII on RNAPIII the total 1775 computed Spearman’s correlations, 1073 transcription of neighboring genes could mask the effects revealed a negative association between tDNA methyla- that DNA methylation would induce on tRNA expres- tion and tRNA expression (Fig. 1B). 114 of the computed sion. We identified 71 tDNA genes that were represented correlations displayed a significant anticorrelation (ρ ≤ by at least one CpG in the HM450 microarray whose -0.2; Supplementary Figure S2A). tRNA-Arg-TCT-4-1, methylation status could be efficiently interrogated (Sup- tRNA-Ile-AAT-8-1, and tRNA-Val-CAC-2-1 were those plementary Figure S1A). tRNAs with the highest frequency of tissues displaying The data mining of HM450 microarray-derived data a negative association between tDNA methylation and of thousands of samples from various cancer types and tRNA expression (Supplementary Figure S2B). This nega- normal tissues available at The Cancer Genome Atlas tive correlation between tDNA methylation and tRNA (TCGA) and of a panel encompassing approximately levels in TCGA suggests a plausible link between DNA 1000 cell lines revealed global differences in tDNA meth- methylation and tRNA expression. ylation in TCGA primary tumors and cancer cell lines Following the above described findings, we aimed compared to TCGA normal samples, mostly tending to analyze in more detail the relationship between towards hypermethylation in malignant samples (Supple- DNA methylation and tRNA expression by provid- mentary Figure S1B). Specifically, 60 of the 71 tDNA pre- ing additional experimental evidence of the epigenetic sented statistically significant differences between normal regulation of two tDNA whose methylation is sus- and tumoral TCGA sets of samples (Supplementary ceptible to impact tRNA expression according to our (See figure on next page.) Fig. 1 Tumor-associated changes in tDNA methylation promote differences in tRNA expression. A Heatmaps showing the average β-value for each tDNA in the different tissues from TCGA set of normal (left) and tumoral (middle) samples and in cancer cell lines (right). β-values span from 0 (green, unmethylated) to 1 (red, hypermethylated). Grey indicates missing values. B Heatmap showing the Spearman’s ρ of the correlation between tDNA gene methylation and the expression of their cognate tRNA in all TCGA tumor samples (left) and separated by tissue of origin (right). Yellow and blue denote positive and negative correlation, respectively. C ChIP-qPCR experiments indicate an increased binding of GTF3C1 (left) and POLR3A (right) to tRNA-Arg-TCT-4-1 (top) and tRNA-Ile-AAT-8-1 (below) genes upon treatment with 5’-azacytidine (AZA) in DND41 and SW48. Data represent the mean ± SD of three biological replicates analyzed using an unpaired two-tailed Student’s t-test. * p < 0.05; ** p < 0.01; *** p < 0.001. D qRT-PCR shows a recovery of tRNA-Arg-TCT-4-1 (left) and tRNA-Ile-AAT-8-1 (right) expression in DND41 and SW48 cell lines after the use of AZA. All qRT-PCR data represent the mean ± SD of biological triplicates analyzed using the unpaired two-tailed Student’s t-test. * p < 0.05; ** p < 0.01. Rosselló‑Tortella et al. Molecular Cancer (2022) 21:48 Page 3 of 8 Fig. 1 (See legend on previous page.) Rosselló‑Tortella et al. Molecular Cancer (2022) 21:48 Page 4 of 8 analysis: tRNA-Arg-TCT-4-1 and tRNA-Ile-AAT-8-1. expression in hypermethylated cell lines upon DNA tRNA-Arg-TCT-4-1 is methylated in most normal tis- demethylation are in accordance with the negative rela- sues, whereas tumoral samples show hypomethylation tionship between tDNA methylation and tRNA expres- events when compared to their matched healthy coun- sion that arose from the in silico analysis in TCGA terparts. (Figure 1A). Conversely, tRNA-Ile-AAT-8-1 (Fig. 1B). is unmethylated in normal tissues and hypermethyl- ated in cancer (Fig. 1A). We selected two cell lines that were hypermethylated for both genes according to tDNA methylation can predict the patients’ clinical the HM450 microarray data: DND41 and SW48 (Sup- outcome in TCGA cohorts plementary Figure S3A). Both tDNAs are represented Apart from modulating gene expression to promote in the HM450 microarray with a single CpG that is the malignant transformation of the cell and facilitate located within the internal promoter of the gene –the cancer progression, tumor-specific alterations in DNA conserved A and B boxes where TFIIIC binds to recruit methylation can be translated into the clinical prac- the RNAPIII machinery [2]. We confirmed by bisulfite tice as potential biomarkers to predict the evolution genomic sequencing of multiple clones the methylation of the disease [13]. The expression of some tRNA has status of the two selected tDNA (Supplementary Meth- also been associated to different clinical outcomes of ods) in these cell lines and obtained a comprehensive cancer patients [9, 12]. For this, because we observed view of the whole methylation landscape for each of that the changes in tDNA methylation that occur in them (Supplementary Figure S3B). DNA demethylation tumorigenesis modulate tRNA expression, we sought upon the use of the drug 5’-azacytidine on these cell to evaluate if these events had any impact on the out- lines increased the binding of the subunit 1 of TFIIIC come of cancer patients and therefore could serve as (GTF3C1) and the subunit A of RNAPIII (POLR3A) predictive biomarkers for this disease. To address this to the two tDNA genes as measured by chromatin hypothesis, we data mined the clinical information immunoprecipitation (ChIP)-qPCR (Supplementary from 31 TCGA projects to assess if differential tDNA Methods), indicating that DNA methylation prevents methylation could inform patients’ survival. We identi- the binding of the tDNA-associated transcriptional fied 86 cases in which the methylation status of a spe- machinery to the internal promoter of the tDNAs cific tDNA was significantly associated with differences (Fig. 1 C). Moreover, the treatment with the demethyl- in patients’ overall survival (Fig. 2A). 56 of these events ating agent 5’-azacytidine also restored the expression were confirmed to be prognostic predictive factors of both tRNAs in the two interrogated hypermethylated for the disease according to univariate Cox regression cell lines (Fig. 1D) as determined by qRT-PCR (Supple- analyses (Fig. 2 A, Supplementary Table S1). Since the mentary Methods), confirming that DNA methylation tRNA pool composition is reported to vary among tis- can repress RNAPIII-mediated gene expression. The sues to match the specific cellular needs and permit the use of the DNA methylation inhibitor, and its associ- expression of different sets of proteins [16], whether ated restoration of tRNA expression upon demethyla- each differential tDNA methylation event results in tion, reduced cell growth (Supplementary Figure S4A) an increased or a reduced risk of the disease should and induced apoptosis (Supplementary Figure S4B). be evaluated individually for every gene and tumor Thus, our in vitro results showing an increase in tRNA (Fig. 2 A, Supplementary Table S1). (See figure on next page.) Fig. 2 tDNA methylation is associated with different overall survival in TCGA cohorts. A Dot plots summarizing the logrank tests (left) and univariate Cox regression models (right) used to compare the overall survival of patients according to methylation status of the 71 tDNAs. Cases that are significantly associated with different prognosis are represented in large-sized bullets. Yellow and blue represent favorable and unfavorable prognosis according to univariate Cox regression analyses (right), respectively. HR, hazard ratio. B Kaplan-Meier curves show that tRNA-Arg-TCT-4-1 hypomethylation is associated with a shorter overall survival in KIRP (top) and UCEC (below) TCGA cohorts. HR, hazard ratio; CI, confidence interval. p-values correspond to logrank tests. C Bisulfite genomic sequencing confirms tRNA-Arg-TCT-4-1 hypomethylation in HEC1 cell line. The tDNA sequence is indicated with a blue bracket. The orange rectangles correspond to the A and B boxes of the tDNA. The TSS is marked with a black arrow. CpG dinucleotides are represented as short vertical lines, and their methylation status is denoted with black (methylated) or white (unmethylated) squares. The CpG represented in the HM450 microarray is marked with a red asterisk. D qRT-PCR exposes higher tRNA-Arg-TCT-4-1 levels in HEC1 compared to DND41 and SW48 cells. Data represent the mean ± SD of biological triplicates analyzed by an unpaired two-tailed Student’s t-test. *** p < 0.001. E Cell cycle analysis reveals an accumulation of tRNA-Arg-TCT-4-1 knockout HEC1 cells in G0/ G1 phase. Data shown represent the mean ± SD of biological triplicates analyzed by unpaired two-tailed Student’s t-test. * p < 0.05. F SRB assay shows a reduced growth of HEC1 knockout cells. Data at each time points are the mean ± SD of four biological replicates. Statistical differences were determined using an unpaired two-tailed Student’s t-test at the 144 h final time point. * p < 0.05. G Transwell assay shows a reduced migration of tRNA-Arg-TCT-4-1-silenced HEC1 cells. Data represent the mean ± SD of biological triplicates analyzed by an unpaired two-tailed Student’s t-test. *** p < 0.001. Representative images of the Transwell insert membranes are shown. Rosselló‑Tortella et al. Molecular Cancer (2022) 21:48 Page 5 of 8 Fig. 2 (See legend on previous page.) Rosselló‑Tortella et al. Molecular Cancer (2022) 21:48 Page 6 of 8 The silencing of tRNA‑Arg‑TCT‑4‑1 reduced endometrial effects of its hypermethylation-associated silencing (Sup- cancer cell growth plementary Figure S6A). This deletion was confirmed Our analyses identified tRNA-Arg-TCT-4-1 as a prom- by genomic PCR and Sanger sequencing in two HEC1 ising candidate to participate in tumorigenesis. Inter- knockout clones (Supplementary Figure S6B), which estingly, high levels of tRNA-Arg-TCT-4-1 are linked to showed minimal tRNA-Arg-TCT-4-1 expression com- a proliferative status of the cell [17], which fits with the pared to the wild-type cell line (Supplementary Figure cancer-associated demethylation of this tDNA in com- S6C). tRNA-Arg-TCT-4-1 CRISPR/Cas9-mediated dele- parison with the methylated status of the normal tissues tion delayed the progression of the cell cycle, as seen by (Fig. 1 and Supplementary Figure S1A) and its resulting an accumulation of cells in the G0/G1 phase in detriment overexpression in 12 cancer types represented in TCGA of the S and G2/M phases in these cells (Fig. 2E). This [9]. Importantly, tRNA-Arg-TCT-4-1 demethylation in resulted in a reduced growth of HEC1 knockout cells kidney renal papillary cell carcinoma (KIRP) and in uter- compared to the wild-type cell line, as assessed by the ine corpus endometrial carcinoma (UCEC) cohorts of SRB assay (Fig. 2 F). Additionally, tRNA-Arg-TCT-4-1 TCGA primary tumors is associated to a shorter overall silencing also reduced cell migration in the two HEC1 survival (Fig. 2B). This hypomethylation change in KIRP knockout clones (Fig. 2G). Overall, our results indicate and UCEC TCGA cohorts accounts for an 8% and 29% of that tRNA-Arg-TCT-4-1 demethylation and increased cases, respectively (Supplementary Figure S5A). In both expression exerts oncogenic-like roles in endometrial cases, the hypomethylation of the gene was associated cancer by promoting cell growth, in agreement with to an increased expression of the tRNA (Supplementary the observed reduced overall survival of those cancer Figure S5B). This is consistent with the overexpression of patients that undergo hypomethylation-associated acti- this tRNA in a subgroup of tumoral samples compared to vation of tRNA-Arg-TCT-4-1 (Fig. 2B and Supplemen- the normal matched tissue [9] that is methylated (Fig. 1 A tary Figure S5B). and Supplementary Figure S1A). The higher percentage of cases presenting tRNA-Arg-TCT-4-1 demethylation in the UCEC cohort of primary tumors (Supplementary Conclusions Figure S5A) prompted us to investigate the role of this Tumoral cells present defects in tRNA biology at multi- tRNA in endometrial cancer and confirm the proposed ple levels. The altered expression of tRNA in cancer cells oncogenic role of its demethylation. The prognosis for can support proliferation and drive tumorigenesis [7], advanced forms of this disease is poor, and therapeu- but the causes and consequences of this deregulation are tic options beyond surgery and first-line chemotherapy poorly understood. Together, our findings give insight on are limited [18]. Therefore, there is an urgent need to the contribution of DNA methylation to these changes investigate this malignancy and describe biomarkers in tRNA levels. Our study revealed that the malignant that predict its aggressiveness to spot high-risk patients transformation of a cell entails epigenetic lesions in that would need a more comprehensive vigilance or that tDNA genes and provided for the first time a plausible would benefit from receiving adjuvant therapy after the connection between DNA methylation and tRNA expres- surgical resection of the tumor [18]. sion. Most importantly from a clinical standpoint, these First, we verified by bisulfite genomic sequencing of differential DNA methylation events can serve as pre- multiple clones the unmethylated status of tRNA-Arg- dictive biomarkers for cancer patients’ prognosis, such TCT-4-1 gene in the endometrial cancer cell line HEC1 as the case for the tRNA-Arg-TCT-4-1 demethylation- (Fig. 2 C), as previously noted in the overall screening of associated overexpression in endometrial carcinoma. cancer cell lines (Fig. 1 and Supplementary Figure S1B). These results support that tRNAs are active participants We also confirmed that expression of this tRNA was in tumorigenesis and that shifts in the DNA methyla- higher in HEC1 than in the hypermethylated DND41 and tion patterns in their regulatory regions contribute to the SW48 cell lines (also derived from our in silico screen- altered profiles of tRNA expression observed in human ing, Fig. 1and Supplementary Figures S1B and S3A) to cancer. However, much remains to be discovered about support that tDNA hypermethylation represses tRNA how their disruption affect cancer cell biology and its expression (Fig. 2D). Changes in tRNA expression can molecular pathways. drive tumorigenesis as they can confer advantages to the malignant cell to promote cancer progression and Materials and method dissemination [7]. To further characterize the role of Detailed materials and methods are available in the Sup- tRNA-Arg-TCT-4-1 in endometrial carcinoma, we elimi- plementary Methods. The sequences of all primers used nated this gene from HEC1 cell line using the CRISPR/ are provided in Supplementary Table S2. Cas9 system (Supplementary Methods) to mimic the Rosselló‑Tortella et al. Molecular Cancer (2022) 21:48 Page 7 of 8 Abbreviations tRNA: Transfer RNA; tDNA: tRNA gene; RNAPII: RNA polymerase II; RNAPIII: RNA hypomethylated or hypermethylated when tRNA-Arg-TCT-4-1 β-value was polymerase III; HM450: HumanMethylation450; TFIIIC: RNAPIII general tran‑ lower or higher than 0.33, respectively. scription factor; ChIP: Chromatin immunoprecipitation; GTF3C1: TFIIIC subunit Additional file 6: Figure S6.Generation of a tRNA-Arg-TCT-4-1 knockout 1; POLR3A: RNAPIII subunit A; TCGA​: The Cancer Genome Atlas; KIRP: Kidney model in HEC1 cell line using the CRISPR/Cas9 system. (A) Overview of renal papillary cell carcinoma; UCEC: Uterine corpus endometrial carcinoma. the CRISPR/Cas9 system targeting tRNA-Arg-TCT-4-1 gene. Two different sgRNA (in green and orange) were simultaneously transfected to intro‑ duce a deletion of the gene of interest. (B) Genomic PCR using primers Supplementary Information flanking tRNA-Arg-TCT-4-1 (black arrows in the overview representation) The online version contains supplementary material available at https://​doi.​ shows a deletion in two HEC1 knockout cells. (C) qRT-PCR confirms the org/​10.​1186/​s12943-​022-​01532-w. silencing of tRNA-Arg-TCT-4-1 in the selected HEC1 knockout cells. Data correspond to the mean ± SD of three biological replicates analyzed using the unpaired two-tailed Student’s t-test. *** p < 0.001. Additional file 1: Figure S1. Description of tDNA methylation using the HM450 microarray. (A) 416 high confidence tDNAs annotated in the Additional file 7: Supplementary methods GRCh37/hg19 human genome (top). 138 of them are included in the Additional file 8: Table S1. Summary of the results obtained from the HM450 DNA methylation microarray. 95 of these 138 are located further Univariate Cox regression analyses performed to compare TCGA patients’ than 2 kb from any other RNAPII-transcribed gene. The withdrawal of prognosis according to their tDNA methylation levels. Blue-colored cells cross-reactive CpGs yielded the 71 different tDNAs that can be efficiently indicate statistical significance (FDR ≤ 0.05). interrogated with this approach (below). The distribution by amino acid and anticodon of the 416 total tDNA genes (top) and those of the 71 that Additional file 9: Table S2. Detailed list of primers used. All sequences are represented in the HM450 DNA methylation microarray (below) are are provided from 5’ to 3’. For tRNA qRT-PCR primers, the commercial refer‑ provided. (B) Average methylation of the 71 tDNA genes according to ence is provided. the HM450 DNA methylation microarray data in normal (left) and tumor (middle) TCGA samples and in cell lines (right). FDR-adjusted p-values Acknowledgements correspond to the Mann-Whitney U-test used to compare the methylation Not applicable. average between TCGA normal and tumor samples. ns, not significant; * FDR < 0.05; ** FDR < 0.01; *** FDR < 0.001. Authors’ contributions Additional file 2: Figure S2. The methylation status of tRNA-Arg-TCT-4-1, MRT and ME designed the study. MRT performed the research with the tRNA-Ile-AAT-8-1, and tRNA-Val-CAC-2-1 is associated with their expres‑ help of ABC, LMV and LV. MRT and ME wrote the manuscript the additional sion levels in TCGA tumors. (A) Volcano plot representing the Spearman’s contributions from all other authors. The author(s) read and approved the final correlation coefficient (ρ) and -log10 FDR of each computed Spearman’s manuscript. correlation from Figure 1B. Statistically significant associations are shown in blue. (B) Barplot summarizing the tissue frequency of all the statistically Funding significant negative associations between methylation and expression We thank the CERCA Programme/Generalitat de Catalunya and support by from panel A. the Health Department PERIS-project #SLT/002/16/00374, AGAUR-project Additional file 3: Figure S3. Bisulfite genomic sequencing confirms the #2017SGR1080; MCI/AEI/ERDF project #RTI2018-094049-B-I00; ERC EPIPHARM; methylation status of tRNA-Arg-TCT-4-1 andtRNA-Ile-AAT-8-1 according Cellex Foundation; and “la Caixa” Foundation (LCF/PR/GN18/51140001). to the HM450 microarray data. (A) DNA methylation status of tRNA-Arg- MRT is a fellow of the Instituto de Salud Carlos III, under i-PFIS contract no. TCT-4-1 (cg12798524) and tRNA-Ile-AAT-8-1 (cg21339923) in DND41 and IFI17/00006. ME is an ICREA Research Professor. SW48 according to the HM450 microarray data. b-values corresponding to single CpGs are shown for each tDNA. (B) Bisulfite genomic sequencing Availability of data and materials of tRNA-Arg-TCT-4-1 (top) and tRNA-Ile-AAT-8-1 (below) genes in DND41 Not applicable. and SW48 cell lines. The tDNA gene genomic sequence is indicated with a blue bracket. The orange rectangles correspond to the A and B boxes Declarations of the tDNA. The TSS is marked with a black arrow. CpG dinucleotides are represented as short vertical lines, and their methylation status is denoted Ethics approval and consent to participate with black (methylated) or white (unmethylated) squares. The CpG Not applicable. included in the HM450 microarray is marked with a red asterisk. Additional file 4: Figure S4. Impact on growth and apoptosis of the DNA Consent of publication methylation inhibitor 5’-azacytidine(AZA) in DND41 and SW48 cancer cell Not applicable. lines. (A) MTT assay growth assay show a reduced growth of treated cells in both DND41 and SW48 cells lines upon the use of the demethylating Competing interests agent. Data at each time points are the mean ± SD of four biological ME is a consultant of Ferrer International and Quimatryx. The remaining replicates. Statistical differences were determined using an unpaired authors declare that they have no conflict of interest. two-tailed Student’s t-test at the 96 hours final time point. *** p < 0.001 (B) Apoptosis quantification by Annexin V incorporation reveals a significant Author details 1 percentage of cells affected by the treatment in DND41 and SW48 cell Cancer Epigenetics Group, Josep Carreras Leukaemia Research Institute lines. Data shown represent the mean ± SD of four biological replicates (IJC), 08916 Badalona, Catalonia, Spain. 2 Germans Trias i Pujol Health Science analyzed by unpaired two-tailed Student’s t-test. * p < 0.05; *** p < 0.001. Research Institute (IGTP), 08916 Badalona, Catalonia, Spain. 3 Centro de Investi‑ gacion Biomedica en Red Cancer (CIBERONC), 28029 Madrid, Spain. 4 Institució Additional file 5: Figure S5. The demethylation of tRNA-Arg-TCT-4-1 Catalana de Recerca i Estudis Avançats (ICREA), 08010 Barcelona, Catalonia, gene is associated with an increased tRNA expression in KIRP and UCEC Spain. 5 Physiological Sciences Department, School of Medicine and Health TCGA tumors. (A) Frequency of tRNA-Arg-TCT-4-1 hypomethylation in Sciences, University of Barcelona (UB), 08907 Barcelona, Catalonia, Spain. tumors derived from TCGA according to the tissue of origin. (B) tRNA- 6 Josep Carreras Leukaemia Research Institute (IJC), Carretera de Can Ruti, Camí Arg-TCT-4-1 expression is higher in those KIRP (top) and UCEC (below) de les Escoles s/n, 08916 Badalona, Barcelona, Catalonia, Spain. TCGA tumors where this gene is hypomethylated. Statistical differences in tRNA expression between groups of samples were determined using a Received: 22 July 2021 Accepted: 5 February 2022 two-sided Mann-Whitney U-test; *** p < 0.001. Samples were considered Rosselló‑Tortella et al. Molecular Cancer (2022) 21:48 Page 8 of 8 References 1. Chan PP, Lowe TM. GtRNAdb 2.0: An expanded database of transfer RNA genes identified in complete and draft genomes. Nucleic Acids Res. 2016;44:D184–9. 2. Berg MD, Brandl CJ. Transfer RNAs: diversity in form and function. RNA Biol. 2021;18:316–39. 3. Pinkard O, McFarland S, Sweet T, Coller J. Quantitative tRNA-sequencing uncovers metazoan tissue-specific tRNA regulation. Nat Commun. 2020;11:4104. 4. Good PD, Kendall A, Ignatz-Hoover J, Miller EL, Pai DA, Rivera SR, et al. Silencing near tRNA genes is nucleosome-mediated and distinct from boundary element function. Gene. 2013;526:7–15. 5. Park JL, Lee YS, Kunkeaw N, Kim SY, Kim IH, Lee YS. Epigenetic regula‑ tion of noncoding RNA transcription by mammalian RNA polymerase III. Epigenomics. 2017;9:171–87. 6. Gerber A, Ito K, Chu CS, Roeder RG. Gene-Specific Control of tRNA Expres‑ sion by RNA Polymerase II. Mol Cell. 2020;78:765–78. 7. Santos M, Fidalgo A, Varanda AS, Oliveira C, Santos MAS. tRNA Deregula‑ tion and Its Consequences in Cancer. Trends Mol Med. 2019;25:853–65. 8. Rosselló-Tortella M, Llinàs-Arias P, Sakaguchi Y, Miyauchi K, Davalos V, Setien F, et al. Epigenetic loss of the transfer RNA-modifying enzyme TYW2 induces ribosome frameshifts in colon cancer. Proc Natl Acad Sci U S A National Academy of Sciences. 2020;117:20785–93. 9. Zhang Z, Ye Y, Gong J, Ruan H, Liu CJ, Xiang Y, et al. Global analysis of tRNA and translation factor expression reveals a dynamic landscape of translational regulation in human cancers. Commun Biol. 2018;1:234. 10. Goodarzi H, Nguyen HCB, Zhang S, Dill BD, Molina H, Tavazoie SF. Modu‑ lated expression of specific tRNAs drives gene expression and cancer progression. Cell. 2016;165:1416–27. 11. Torres AG, Reina O, Attolini CSO, De Pouplana LR. Differential expression of human tRNA genes drives the abundance of tRNA-derived fragments. Proc Natl Acad Sci U S A National Academy of Sciences. 2019;116:8451–6. 12. Kuang M, Zheng D, Tao X, Peng Y, Pan Y, Zheng S, et al. tRNA-based prog‑ nostic score in predicting survival outcomes of lung adenocarcinomas. Int J Cancer. 2019;145:1982–90. 13. Berdasco M, Esteller M. Clinical epigenetics: seizing opportunities for translation. Nat Rev Genet. 2019;20:109–27. 14. Bartke T, Vermeulen M, Xhemalce B, Robson SC, Mann M, Kouzarides T. Nucleosome-interacting proteins regulated by DNA and histone meth‑ ylation. Cell. 2010;143:470–84. 15. Besser D, Götz F, Schulze-Forster K, Wagner H, Kröger H, Simon D. DNA methylation inhibits transcription by RNA polymerase III of a tRNA gene, but not of a 5S rRNA gene. FEBS Lett. 1990;269:358–62. 16. Dittmar KA, Goodenbour JM, Pan T. Tissue-specific differences in human transfer RNA expression. PLoS Genet. 2006;2:2107–15. 17. Gingold H, Tehler D, Christoffersen NR, Nielsen MM, Asmar F, Kooistra SM, et al. A dual program for translation regulation in cellular proliferation and differentiation. Cell. 2014;158:1281–92. 18. Lee YC, Lheureux S, Oza AM. Treatment strategies for endometrial cancer: Current practice and perspective. Curr Opin Obstet Gynecol. 2017;29:47–58. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. Ready to submit your research ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations • maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions