Exploring the role of splicing in TP53 variant pathogenicity through predictions and minigene assays

BackgroundThe incorrect interpretation of missense and synonymous variants can lead to improper molecular diagnosis and subsequent faulty genetic counselling. The aim of this study was to evaluate the pathogenicity of presumed COL4A3/COL4A4 missense and synonymous variants detected by next-generation sequencing to provide evidence for diagnosis and genetic counselling.MethodsPatients' clinical findings and genetic data were analysed retrospectively. An in vitro minigene assay was conducted to assess the effect of presumed COL4A3/COL4A4 missense and synonymous variants on RNA splicing.ResultsFive unclassified COL4A3/COL4A4 variants, which were detected in five of 343 patients with hereditary kidney diseases, were analysed. All of them were predicted to affect splicing by Human Splicing Finder. The presumed COL4A3 missense variant c.4793T > G [p. (Leu1598Arg)] resulted in a loss of alternative full-length transcript during the splicing process. The COL4A3 transcript carried synonymous variant c.765G > A [p. (Thr255Thr)], led to an in-frame deletion of exon 13. Nevertheless, variants c.3566G > A [p. (Gly1189Glu)] in COL4A3 and c.3990G > A [p. (Pro1330Pro)], c.4766C > T [p. (Pro1589Leu)] in COL4A4 exhibited no deleterious effect on splicing. Among the five patients harbouring the abovementioned COL4A3/COL4A4 variants, three patients were genetically diagnosed with autosomal recessive Alport syndrome, one patient was highly suspected of having thin basement membrane nephropathy, and the other patient was clinically diagnosed with Alport syndrome.ConclusionsCOL4A3 presumed missense variant p. (Leu1598Arg) and synonymous variant p. (Thr255Thr) affect RNA splicing, which highlights the prime importance of transcript analysis of unclassified exonic sequence variants for better molecular diagnosis and genetic counselling. Meanwhile, the reliability of splicing predictions by predictive tools for exonic substitutions needs to be improved.

T2. IDENTIFICATION OF PATHOGENIC VARIANTS IN PROTEIN CODING GENES

Triple negative breast cancer (TNBC) is a subtype of breast cancer of heterogeneous nature that is negative for estrogen receptor (ER), progesterone receptor (PR) and growth factor human epidermal 2 (HER2) following immunohistochemical analysis. TNBC is frequently characterized by relapse and reduced survival. To date, there is no targeted therapy for this type of cancer. Chemotherapy, radiotherapy, and surgery remain as the standard treatments options. The lack of a target therapy and the heterogeneity of TNBC highlight the need to seek new therapeutic options. In this study, fresh tissue samples of TNBC were analyzed with a panel of 48 driver genes (212 amplicons) that are likely to be therapeutic targets. We found intron variants, missense, stop gained and splicing variants in TP53, PIK3CA and FLT3 genes. Interestingly, all the analyzed samples had at least two variants in the TP53 gene, one being a drug response variant, rs1042522, found in 94% of our samples. We also found seven additional variants not previously reported in the TP53 gene, to the best of our knowledge, with probable deleterious characteristics of the tumor suppressor gene. We found four genetic variants in the PIK3CA gene, including two missense variants. The rs2491231 variant in the FLT3 gene was identified in 84% (16/19) of the samples, which not yet reported for TNBC, to the best of our knowledge. In conclusion, genetic variants in TP53 were found in all TNBC tumors, with rs1042522 being the most frequent (94% of TNBC biopsies), which had not been previously reported in TNBC. Also, we found two missense variants in the PIK3CA gene. These results justify the validation of these genetic variants in a large cohort, as well as the extensive study of their impact on the prognosis and therapy management of TBNC.

Variants in the MYO7A gene are increasingly identified among patients suffering from Usher syndrome type 1B (USH1B). However, such mutations are less commonly detected among patients suffering from nonsyndromic hearing loss (NSHL), including autosomal recessive deafness (DFNB2) and autosomal dominant deafness (DFNA11). This research attempts to clarify the genetic base of DFNB2 in a Chinese family and determine the pathogenicity of the identified mutations. Targeted next-generation sequencing (TGS) of 127 known deafness genes was performed for the 14-year-old proband. Then, Sanger sequencing was performed on the available family members. A minigene splicing assay was performed to verify the impact of the novel MYO7A synonymous variant. After performing targeted next-generation sequencing (TGS) of 127 existing hearing loss-related genes in a 14-year-old proband, Sanger sequencing was carried out on the available family members. Then, to confirm the influence of the novel MYO7A synonymous variants, a minigene splicing assay was performed. Two heteroallelic mutants of MYO7A (NM_000260.3) were identified: a maternally inherited synonymous variant c.2904G > A (p.Glu968=) in exon 23 and a paternally inherited missense variant c.5994G > T (p.Trp1998Cys) in exon 44. The in vitro minigene expression indicated that c.2904G > A may result in skipping of exon 23 resulting in a truncated protein. We reported a novel missense (c.5994G > T) and identified, for the first time, a novel pathogenic synonymous (c.2904G > A) variant within MYO7A in a patient with DFNB2. These findings enrich our understanding of the MYO7A variant spectrum of DFNB2 and can contribute to accurate genetic counseling and diagnosis of NSHL patients.

Oculocutaneous albinism type 2 (OCA2) is the second most frequent form of albinism and represents about 30% of OCA worldwide. As with all types of OCA, patients present with hypopigmentation of hair and skin, as well as severe visual abnormalities. We focused on a subgroup of 29 patients for whom genetic diagnosis was pending because at least one of their identified variants in or around exon 10 of OCA2 is of uncertain significance (VUS). By minigene assay, we investigated the effect of these VUS on exon 10 skipping and showed that not only intronic but also some synonymous variants can result in enhanced exon skipping. We further found that excessive skipping of exon 10 could be detected directly on blood samples of patients and of their one parent with the causal variant, avoiding invasive skin biopsies. Moreover, we show that variants, which result in lack of detectable OCA2 mRNA can be identified from blood samples as well, as shown for the most common OCA2 pathogenic missense variant c.1327G>A/p.(Val443Ile). In conclusion, blood cell RNA analysis allows testing the potential effect of any OCA2 VUS on transcription products. This should help to elucidate yet unsolved OCA2 patients and improve genetic counseling.

Breast cancer is a multifaceted disease with several clinical, pathological and molecular attributes contributing to disease prognosis or treatment outcome. Treatment measures in breast cancer are based on hormone/growth factor receptor -estrogen/progesterone receptor (ER/PR) or human epidermal growth factor receptor 2 (Her2) status. TP53 pathway inactivation in breast cancer is well-established. Although TP53's therapeutic relevance is well-recognized, it remains under-utilized in patient-management, since all TP53 mutants are treated equally in the diagnostic context. In reality, enormous heterogeneity exists in nature, type and functional impact of TP53 variants. Therefore, understanding the diversity of TP53 variants in breast cancer subtypes may enhance its diagnostic utility in this cancer. We utilized clinical NGS data, obtained using commercially available targeted panels, TruSeq Amplicon Cancer Panel (Illumina) and Ion AmpliSeq Cancer Hotspot Panel v2 (Thermofisher) to analyze tumor DNAs from cancer patients at the Advanced Molecular Diagnostic Laboratory (Princess Margaret Cancer Centre, Toronto, Canada). We focused on data from 105 advanced breast cancer patients. We consolidated several schemes proposed in the literature to classify TP53 variants, and evaluated patient molecular profiling and pathology data based on: (1) presence of TP53 variants; (b) coding effect; and (c) transcriptional activity. We further investigated whether TP53 variants were associated with reportable variant load, co-occurrence with other molecular changes and hormone/growth-factor receptor status. In our study group, 70.4% cases carried one or more variants. TP53 alterations were prevalent (40.9%) in our cohort, followed by PIK3CA variants (36.2%). 15/105 cases (14.3%) carried variants in both genes. Unlike in other cancer types, where missense TP53 variants predominate (e.g., colorectal, 72.6%), missense (49%) and nonsense/frameshift (42%) variants were similarly distributed in breast cancers. Gain-of-Function (GOF) and Loss-of-Function (LOF) TP53 variants were also equally distributed (32% vs. 33%). However, TP53mut PIK3CAmut breast cancer cases were more likely to carry missense and/or LOF variants (10/15 cases). TP53 variants were also associated with hormone/growth-factor receptor status. A greater proportion of ER- vs ER+, PR- vs PR+, and ER-PR-Her2- vs ER+PR+Her2- breast cancer cases carried missense GOF TP53 variants respectively when compared to missense LOF and variants of unknown significance taken together(80-85% vs 50-55%; p<0.0001). Finally, TP53mut cases were more likely to carry multiple variants in contrast to TP53wt cases (37.2% vs. 12.9%). Taken together, we define a stratification strategy for TP53 that takes into account the diversity of TP53 variants, and demonstrate its application to molecular profiling and clinico-pathological data in breast cancer. Citation Format: Swati Garg, Mahadeo A. Sukhai, Maksym Misyura, Mariam Thomas, Tong Zhang, Lillian L. Siu, Philippe L. Bedard, Tracy L. Stockley, Suzanne Kamel-Reid. Impact of TP53 status and functional classification on molecular profiles in breast cancer subtypes. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 4765.

Global developmental delay with speech and behavioral abnormalities (OMIM: 619243) is an autosomal dominant disease caused by variants in TNRC6B gene. We reviewed and summarized clinical manifestations and genotypes in patients previously reported with TNRC6B gene variants. We used several prediction tools to predict pathogenicity and performed minigene assays to verify the function of the synonymous variant affecting RNA splicing. The patient presented with convulsive seizures and developmental delay. WES combined with functional studies diagnosed a child with a synonymous variant in TNRC6B gene. Through minigene assay and Sanger sequencing, we demonstrated that c.3141G > A variant induced exon 7 skipping and the synonymous variant was pathogenic. Synonymous variants do not change the amino acids encoded by the codon, so we usually consider synonymous variants to be benign and ignore their pathogenicity. Minigene assay is a valuable tool to identify the effect of variation on RNA splicing and identify synonymous variants' benign or pathogenic. We showed that the synonymous variant was pathogenic by minigene assay. WES combined with minigene assay establishes a robust basis for genetic counseling and diagnosing diseases.

Paroxysmal kinesigenic dyskinesia (PKD) is the most common hereditary paroxysmal movement disorder. The PRRT2 gene is the first identified causative gene and accounts for the majority of PKD. In this study, we investigated the pathogenicity of PRRT2 variants in the splice regions. Patients with clinically suspected PKD and no detectable pathogenic variants in the PRRT2 gene were included. Targeted next-generation sequencing technology was used to screen the full-length sequence of PRRT2. In silico analyses were performed on splice region variants identified in our cohort and compiled from the Human Gene Mutation Database (HGMD). Subsequently, a minigene system carrying these variants was constructed and introduced into HEK293T cells for functional assays to assess the pathogenicity. Fourteen PRRT2 variants were analyzed, including four identified in patients with clinically suspected PKD from our center and 10 retrieved from HGMD. These variants comprised 10 intronic variants, two synonymous variants, one deletion, and one missense variant. In silico predictions suggested that all variants, except for one deep intronic variant, had the potential to affect normal splicing. Functional assays showed that 11 PRRT2 variants, including missense and intronic variants, caused aberrant splicing events, such as exon skipping and intron retention. The two synonymous variants and one deep intronic variant exhibited no splicing abnormalities. Based on these results, five patients with PRRT2 variants previously classified as variants of uncertain significance can now be genetically diagnosed with PKD or other PRRT2-related disorders. Combining in silico analyses with functional assays is essential for determining the pathogenicity of splice variants. It can help confirm the diagnosis of patients with clinically suspected PKD and other PRRT2-related disorders.

Variants in GCK, HNF1A, and HNF4A genes are the three main causes of monogenic diabetes. Determining the molecular etiology is essential for patients with monogenic diabetes to benefit from the most appropriate treatment. The increasing number of variants of unknown significance (VUS) is a major issue in genetic diagnosis, and assessing the impact of variants on RNA splicing is challenging, particularly for genes expressed in tissues not easily accessible as in monogenic diabetes. The in vitro functional splicing assay based on a minigene construct is an appropriate approach. Here, we performed in silico analysis using SpliceAI and SPiP and prioritized 36 spliceogenic variants in GCK, HNF1A, and HNF4A. Predictions were secondarily compared with Pangolin and AbSplice-DNA bioinformatics tools which include tissue-specific annotations. We assessed the effect of selected variants on RNA splicing using minigene assays. These assays validated splicing defects for 33 out of 36 spliceogenic variants consisting of exon skipping (15%), exonic deletions (18%), intronic retentions (24%), and complex splicing patterns (42%). This provided additional evidence to reclassify 23 out of 31 (74%) VUS including missense, synonymous, and intronic noncanonical splice site variants as likely pathogenic variants. Comparison of in silico analysis with minigene results showed the robustness of bioinformatics tools to prioritize spliceogenic variants, but revealed inconsistencies in the location of cryptic splice sites underlying the importance of confirming predicted splicing alterations with functional splicing assays. Our study underlines the feasibility and the benefits of implementing minigene-splicing assays in the genetic testing of monogenic diabetes after a prior in-depth in silico analysis.

Introduction: TP53 is a multi-functional tumor suppressor gene, with several important roles in tumorigenesis. Females who harbor germline TP53 pathogenic/likely pathogenic variants (GPV) have a very high lifetime risk of developing breast cancer (BC), especially hormone receptor positive and HER2 positive tumors. Specific TP53 variants have different functional consequences for molecular pathways and somatic aberrations in cancers, leading to the hypothesis that the type of TP53 GPV could modulate breast tumor phenotype. Objective: To examine differences in TP53 GPV and their predicted functional impact on age at first BC diagnosis and BC subtype (ER and HER2 status). MATERIAL AND METHODS: This multicenter international cohort was comprised of female TP53 GPV carriers diagnosed with any invasive BC (non-metastatic or de novo metastatic). We included TP53 variants classified as GPV in ClinVar, or classified as GPV by at least one major laboratory, or truncating variants. Carriers of a second GPV in other BC genes and carriers of TP53 GPV downgraded to a variant of uncertain significance by the TP53 ClinGen-VCEP were excluded. Clinical data were abstracted from medical records. TP53 GPV were classified according to mutation type: nonsense, frameshift, missense affecting the DNA binding domain (Missense_DBD) or the tetramerization domain (Missense_TD), variants affecting splicing, and copy number variations. For this report, functional classification was performed using Fortuno’s classification based on two germline TP53 databases from commercial labs. Functional classification categories included: missense variants with dominant negative effect (Missense_DNE) and without DNE (Missense_notDNE), truncating and hotspots variants. The bivariate association between tumor phenotype and each of the variables (mutation type and Fortuno’s functional classification) was assessed using two-sample Wilcoxon test for continuous and Fisher’s exact test for nominal categorical variables. Results: Among 301 females who met study criteria,mean age of BC diagnosis was 37.0 years (SD 10.46), 187 (62.1%) met Li-Fraumeni syndrome clinical criteria (revised Chompret or Classic criteria), and 41 (13.6%) had bilateral synchronous BC. The distribution of BC subtypes was: 108 (35.9%) ER+/HER2-, 79 (26.2%) ER+/HER2+, 62 (20.6%) ER-/HER2+, 20 (6.6%) ER-/HER2-, and 32 (10.6%) ER unknown and/or HER2 unknown. Most TP53 GPV were missense variants (n=217; 141 Missense_DBD, 76 Missense_TD). In comparison to Missense_TD, Missense_DBD carriers had a younger age at BC diagnosis (35.7 vs 42.3, p<0.01), lower rates of ER+ disease (63.1% vs 80.3%, p<0.01), and higher rates of HER2+ disease (48.9% vs 27.6%, p<0.01). When considering ER and HER2 subtypes, Missense_TD carriers had more ER+/HER2- BC (51.3% vs 34.8%, p=0.02) and Missense_DBD carriers had more ER-/HER2+ tumors (22.0% vs 5.3%, p=<0.01). According to Fortuno’s classification, 89 (29.6%) TP53 GPV were Missense_notDNE, 59 (19.6%) Missense_DNE, 80 (26.6%) truncating, 49 (16.3%) hotspots variants, and 24 (8.0%) variants unclassified by the method. Among all these categories, Missense_notDNE variants were associated with older ages of BC diagnosis (51.7% above 40 years; p<0.001), higher rates of ER+ disease (79.8%; p<0.001), lower rates of HER2+ tumors (29.2%; p<0.001), and lower rates of bilateral synchronous BC (p=0.003). CONCLUSIONS: These findings suggest that TP53 GPV functional status can influence age at breast cancer presentation and tumor ER/HER2 status which can potentially improve targeted treatment strategies and inform risk prediction and risk reduction strategies. Citation Format: Renata L. Sandoval, Michele Bottosso, Miki Horiguchi, Natalia Polidorio, Anh Le, Brittany L. Bychkovsky, Benjamin Verret, 
Alessandra Gennari, Sophie Cahill, Alison Schwartz-Levine, Olivier Caron, Marion ImbertBouteille,
Catherine Noguès, Pauline Rochefort, Kara N. Mawell, Maria Isabel Achatz, Fabrice Andre, Judy E. Garber. Differences in breast cancer phenotype by germline TP53 variant functional classification [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2024; 2024 Dec 10-13; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(12 Suppl):Abstract nr PS16-04.

Somatic variation is a major type of genetic variation contributing to human diseases including cancer. Of the vast quantities of somatic variants identified, the functional impact of many somatic variants, in particular the missense variants, remains unclear. Lack of the functional information prevents the translation of rich variation data into clinical applications. We previously developed a method named Ramachandran Plot-Molecular Dynamics Simulations (RP-MDS), aiming to predict the function of germline missense variants based on their effects on protein structure stability, and successfully applied to predict the deleteriousness of unclassified germline missense variants in multiple cancer genes. We hypothesized that regardless of their different genetic origins, somatic missense variants and germline missense variants could have similar effects on the stability of their affected protein structure. As such, the RP-MDS method designed for germline missense variants should also be applicable to predict the function of somatic missense variants. In the current study, we tested our hypothesis by using the somatic missense variants in TP53 as a model. Of the 397 somatic missense variants analyzed, RP-MDS predicted that 195 (49.1%) variants were deleterious as they significantly disturbed p53 structure. The results were largely validated by using a p53-p21 promoter-green fluorescent protein (GFP) reporter gene assay. Our study demonstrated that deleterious somatic missense variants can be identified by referring to their effects on protein structural stability.

Missense variants in ABCA4 constitute approximately 50% of causal variants in Stargardt disease (STGD1). Their pathogenicity is attributed to their direct effect on protein function, whilst their potential impact on pre-mRNA splicing disruption remains poorly understood. Interestingly, synonymous ABCA4 variants have previously been classified as 'severe' variants based on in silico analyses. Here, we systemically investigated the role of synonymous and missense variants in ABCA4 splicing by combining computational predictions and experimental assays. To identify variants of interest, we used SpliceAI to ascribe defective splice predictions on a dataset of 5579 biallelic STGD1 probands. We selected those variants with predicted delta scores for acceptor/donor gain > 0.20, and no previous reports on their effect on splicing. Fifteen ABCA4 variants were selected, four of which were predicted to create a new splice acceptor site and eleven to create a new splice donor site. In addition, three variants of interest with delta scores < 0.20 were included. The variants were introduced in wild-type midigenes that contained 4 to 12kb of ABCA4 genomic sequence, which were subsequently expressed in HEK293T cells. By using RT-PCR and Sanger sequencing, we identified splice aberrations for 16 of 18 analyzed variants. SpliceAI correctly predicted the outcomes for 16 out of 18 variants, illustrating its reliability in predicting the impact of coding ABCA4 variants on splicing. Our findings highlight a causal role for coding ABCA4 variants in splicing aberrations, improving the severity assessment of missense and synonymous ABCA4 variants, and guiding to new treatment strategies for STGD1.

Germline TP53 splicing variants are uncommon, and their clinical relevance is unknown. However, splice-altering variants at exon 4-intron 4 junctions are relatively enriched in pediatric adrenocortical tumors (ACT). Nevertheless, family histories of cancer compatible with classic Li-Fraumeni syndrome are rarely seen in these patients. We used conventional and in silico assays to determine protein stability, splicing, and transcriptional activity of 10 TP53 variants at exon 4-intron 4 junctions and analyzed their clinical correlates. We reviewed public databases that report the impact of TP53 variants in human cancer and examined individual reports, focusing on family history of cancer. TP53 exon 4-intron 4 junction germline variants were identified in 9 of 75 pediatric ACTs enrolled in the International Pediatric Adrenocortical Tumor Registry and Children's Oncology Group ARAR0332 study. An additional eight independent TP53 variants involving exon 4 splicing were identified in the Pediatric Cancer Genome Project (n = 5,213). These variants resulted in improper expression due to ineffective splicing, protein instability, altered subcellular localization, and loss of function. Clinical case review of carriers of TP53 exon 4-intron 4 junction variants revealed a high incidence of pediatric ACTs and atypical tumor types not consistent with classic Li-Fraumeni syndrome. Germline variants involving TP53 exon 4-intron 4 junctions are frequent in ACT and rare in other pediatric tumors. The collective impact of these germline TP53 variants on the fidelity of splicing, protein structure, and function must be considered in evaluating cancer susceptibility. IMPLICATIONS: Taken together, the data indicate that splice variants at TP53 codon 125 and surrounding bases differentially impacted p53 gene expression and function.

Genetic diseases are mostly implicated with genetic variants, including missense, synonymous, non-sense, and copy number variants. These different kinds of variants are indicated to affect phenotypes in various ways from previous studies. It remains essential but challenging to understand the functional consequences of these genetic variants, especially the noncoding ones, due to the lack of corresponding annotations. While many computational methods have been proposed to identify the risk variants. Most of them have only curated DNA-level and protein-level annotations to predict the pathogenicity of the variants, and others have been restricted to missense variants exclusively. In this study, we have curated DNA-, RNA-, and protein-level features to discriminate disease-causing variants in both coding and noncoding regions, where the features of protein sequences and protein structures have been shown essential for analyzing missense variants in coding regions while the features related to RNA-splicing and RBP binding are significant for variants in noncoding regions and synonymous variants in coding regions. Through the integration of these features, we have formulated the Multi-level feature Genomic Variants Predictor (ML-GVP) using the gradient boosting tree. The method has been trained on more than 400,000 variants in the Sherloc-training set from the 6th critical assessment of genome interpretation with superior performance. The method is one of the two best-performing predictors on the blind test in the Sherloc assessment, and is further confirmed by another independent test dataset of de novo variants.

Accurate prediction of damaging missense variants is critically important for interpreting a genome sequence. Although many methods have been developed, their performance has been limited. Recent advances in machine learning and the availability of large-scale population genomic sequencing data provide new opportunities to considerably improve computational predictions. Here we describe the graphical missense variant pathogenicity predictor (gMVP), a new method based on graph attention neural networks. Its main component is a graph with nodes that capture predictive features of amino acids and edges weighted by co-evolution strength, enabling effective pooling of information from the local protein context and functionally correlated distal positions. Evaluation of deep mutational scan data shows that gMVP outperforms other published methods in identifying damaging variants in TP53, PTEN, BRCA1 and MSH2. Furthermore, it achieves the best separation of de novo missense variants in neuro developmental disorder cases from those in controls. Finally, the model supports transfer learning to optimize gain- and loss-of-function predictions in sodium and calcium channels. In summary, we demonstrate that gMVP can improve interpretation of missense variants in clinical testing and genetic studies.