Portal de Periódicos da CAPES

Predictive medicine in a testis trio‐family through a combined multi‐omics approach

2024; Springer Science+Business Media; Volume: 14; Issue: 4 Linguagem: Inglês

10.1002/ctm2.1643

2001-1326

Federica Di Maggio, Gianluca Damaggio, Marcella Nunziato, Silvia Buonaiuto, Felice Crocetto, Alessandra Calabrese, Achille Aveta, Gioacchino Vino, Giacinto Donvito, Savio Domenico Pandolfo, Ciro Imbimbo, Vincenza Colonna, Francesco Salvatore,

Genetics and Neurodevelopmental Disorders

Dear Editor, This study takes a comprehensive approach that combines targeted gene panel sequencing, whole genome sequencing (WGS), and DNA methylation analysis of a family trio with a history of seminoma and other tumours, in an attempt to identify pathogenic variants and epigenetic changes related to early onset of testicular cancer (TC) in the proband (p1, 18 years old). We found potentially pathogenic variants in the CHEK2 and ELAC2 genes, differentially methylated regions (DMRs) between the proband/father and the mother in the PPP1R13L gene involved in TP53-mediated apoptosis and near the ERCC2 DNA repair gene, with suggestive gene expression changes, thus indicating possible roles in the proband's tumour development. Testicular cancer is one of the most common malignancies in young men.1 Risk factors for TC are undescended testis, a personal or family history of TC, HIV, age and ethnicity.1 However, the pathogenetic mechanisms underlying TCs are mostly unknown. The goals of this study were to (i) find putatively causative variants in two sets of genes associated with several cancer types including testis cancer, (ii) compare what was found with next-generation sequencing (Illumina) with that of the third-generation sequencing approach (Oxford Nanopore Technology, ONT) and (iii) look for differences in the DNA methylation profiles within the study group. The study of genetic variants was based on a candidate-gene method using two separate sets of genes. The first set is a customized panel of 56 genes that we designated as "cancer-set". We sequenced the latter using a short read sequencing strategy (Illumina).2, 3 The second set consists of 133 genes associated with TCs that we designated "testis-cancer-set", which we sequenced with DNA long-reads (ONT),4 (Figure 1 and Tables S2 and S3). From each subject (p1, p2 and p3) we extracted and isolated nucleic acids (both DNA and RNA) from whole peripheral blood, as well as from testicular tumour tissue taken from the proband (p1) (see Supporting Information). In the "cancer-set", we prioritized in the proband two "high-quality" mapped variants using Alissa software (Agilent Technologies), and we applied the pipeline described in Figure S1. The latter data were also confirmed with ONT (see Table 1A). The first variant is shared by the proband (p1) and his mother (p3). It is an intronic variant in the CHEK2 gene, c.721+3 A > T, rs587782849, which is reported in the ClinVar database as a conflicting interpretation of pathogenicity but, predicted to be a Likely Pathogenic (LP) by the Franklin tool according to ACMG guidelines, considering the evidence PS4, PS3, PM2 and PP3.3, 5 The second variant found is shared with his father (p2), it is a nonsense mutation in the ELAC2 gene, c.2392G > T, p. Glu798Ter, rs771225119, that is not reported in the ClinVar database, but with a CADD score of 34. However, based on the ACMG guidelines, it is predicted to be LP considering the evidence of PVS1 and PM2.5, 6 The genotype concordance being 98.9% among DNA sequences, obtained with the two approaches used, was assessed with the Picard tool, which suggests that the long-read sequencing data is a reliable method with which to detect single nucleotide variants (SNVs) (see Table 1A and Tables S4–S6). Moreover, with the "testis-cancer-set" four more missense variants, all in a heterozygous state were also identified.7, 8 The first 2 were in the PLEC and PCNT genes and shared by p1 and p2; the other two were in the TFCP2L1 and FANCM genes and shared by p1 and p3. Based on the ACMG, the four variants were classified as a variant of uncertain significance (VUS) (Table 1A), and could be considered candidates for further investigation. Data from ONT were used to analyze DNA methylation patterns, focusing on where the methylation profiles of p1 and p2 were similar but different from those of the control (p3). This comparison revealed 256 DMRs, (p-value < .001), of which 115 differentially methylated in the son-father couple (p1-p2) significantly different from that of the mother (methylated: n = 31; unmethylated: n = 84, Figure 2A, B). We found that 51% of the DMRs-p1p2 overlap annotated CpG island, suggesting that our method is robust to identify novel methylation sites. Furthermore, 23 DMRs-p1p2 fall within regulatory gene elements (promoters/enhancers) and 67 within the panels used (Tables S7 and S8). We focused on two DMRs-p1p2 specifically: the first, localized on chromosome 19, contains the PPP1R13L gene and upstream ERCC2 gene; in the second one, on chromosome 17, falls SEPTIN4 (not included in our panels) and its antisense and downstream TEX14 gene (Figure 2D,E and Table 1B). PPP1R13L belongs to the regulative pathway of TP53 activity (Reactome pathway ID R-HSA-6804759), adjusted p-value of .05 (Figure 2C), it is related to DNA repair and cell survival and inhibits the function of TP53 that suppresses the activation of apoptosis.9 Moreover, the ERCC2 gene is involved in the nucleotide excision repair (NER) pathway, which is responsible for repairing DNA damage. The TEX14 gene is related to spermatogenesis and fertility, and changes in its expression are related to spermatogenic deficiency and male infertility. Therefore, to evaluate the effect of methylation on the expression of candidate genes, we evaluated it using real-time polymerase chain reaction of the above genes: ERCC2, PPP1R13L, TEX14, TP53 and GAPDH used as a normalizer (see Supporting Information). This evaluation was performed on RNA extracted from the peripheral whole blood of the 3 subjects (p1, p2 and p3) and the proband's tumour tissue (p1_k). Indeed, the gene PPP1R13L is poorly expressed in the blood of all members of the trio, however, we note an increased expression in the tumour tissue of the proband (p1_k, Figure 3). On the other hand, TP53 expression levels are lower in the proband (p1 and p1_k) than in the parents' blood (p2 and p3). PPP1R13L and TP53 expression patterns are consistent with the effect of reduced methylation of PPP1R13L. The expression of ERCC2, particularly in the blood of p1 and tumour tissue, is lower than in the blood of the parents, but it is higher in the tissue (p1_k) than in the blood of p1 (Figure 3). This pattern of expression is compatible with dysregulation mechanisms of NER and DNA repair pathway existing in tumour tissue, which is much less evident in blood because of the mechanism that regulates secretion from tumour tissue; this is also reported in the literature.10 The TEX14 gene is poorly expressed in the blood of all trio members. However, its expression in the tumor tissue is 40-fold greater than that in the blood of p1. Specifically, this high increase in tumor tissue can be due to the higher specificity in TCs,11 being also this gene connected with spermatogenesis and subsequent fertility. In conclusion, this study takes a comprehensive and combined multi-omics approach to analyze a family trio with a history of TC and other tumours. Using this approach, we identified highly putative pathogenic alterations at the genomic level, as well as DMRs between the proband/father and different from the mother considered also as control, particularly along the pathway of apoptosis and DNA-repair related genes. Gene expression analysis provided supporting evidence of functional effects of these genetic and epigenetic alterations, that all together may contribute to the early onset of TCs. A limitation of the study is that it is a trio-case study, and therefore, starting from the data described in this paper any direct functional assays, aimed at investigating the links between methylation and gene expression, will be a matter of further studies. Francesco Salvatore, Vincenza Colonna and Ciro Imbimbo designed the project and supervised the whole process; Federica Di Maggio, Marcella Nunziato, Gianluca Damaggio and Silvia Buonaiuto conducted the experiments, performed the data analyses and contributed to elaboration and first draft writing; Achille Aveta, Alessandra Calabrese, Felice Crocetto, Savio Domenico Pandolfo and Ciro Imbimbo enrolled the patients and performed the clinical study; Federica Di Maggio, Marcella Nunziato, Gianluca Damaggio and Silvia Buonaiuto helped the first draft of the manuscript and edit the images; Gioacchino Vino and Giacinto Donvito helped in bioinformatics for database search. All authors contributed to some parts of the writing including tables and figures, and agreed to the complete, final version of the manuscript under the supervision of Francesco Salvatore and Vincenza Colonna. We thank Jean Ann Gilder for the English language revision. The authors declare no conflict of interest. This research was supported by Ministero della Salute [RF-2010-23183729] to Professor Francesco Salvatore; Grant from Regione Campania [CIRO project: infrastructures and scientific instrumentation to CEINGE (Coordinator Prof. Francesco Salvatore) D.D. 366/2018; SATIN "Neoplasia studies" POR Campania FESR 2014/2020; Grants from Regione Campania in the context of studies on the fight against neoplastic diseases, BURC: Legge 38/2020 art.16, D.D. Regione Campania 48 of 04/03/2021, D.D. Regione Campania 359 of 20/12/2022 and D.D. Regione Campania 9 of 12/01/2023 to Prof. Francesco Salvatore. The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Federico II Ethics Committee Number 318/20. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.