Whole genome sequences discriminate hereditary hemorrhagic telangiectasia phenotypes by non-HHT deleterious DNA variation

Mutations causing premature termination codons discriminate and generate cellular and clinical variability in HHT

ABSTRACT

For monogenic diseases caused by pathogenic loss-of-function DNA variants, attention focuses on dysregulated gene-specific pathways, usually considering molecular subtypes together within causal genes. To better understand phenotypic variability in hereditary hemorrhagic telangiectasia (HHT), we subcategorized pathogenic DNA variants in ENG/endoglin, ACVRL1/ALK1, and SMAD4 if they generated premature termination codons (PTCs) subject to nonsense-mediated decay. In 3 patient cohorts, a PTC-based classification system explained some previously puzzling hemorrhage variability. In blood outgrowth endothelial cells (BOECs) derived from patients with ACVRL1+/PTC, ENG+/PTC, and SMAD4+/PTC genotypes, PTC-containing RNA transcripts persisted at low levels (8%-23% expected, varying between replicate cultures); genes differentially expressed to Bonferroni P < .05 in HHT+/PTC BOECs clustered significantly only to generic protein terms (isopeptide-bond/ubiquitin-like conjugation) and pulse-chase experiments detected subtle protein maturation differences but no evidence for PTC-truncated protein. BOECs displaying highest PTC persistence were discriminated in unsupervised hierarchical clustering of near-invariant housekeeper genes, with patterns compatible with higher cellular stress in BOECs with >11% PTC persistence. To test directionality, we used a HeLa reporter system to detect induction of activating transcription factor 4 (ATF4), which controls expression of stress-adaptive genes, and showed that ENG Q436X but not ENG R93X directly induced ATF4. AlphaFold accurately modeled relevant ENG domains, with AlphaMissense suggesting that readthrough substitutions would be benign for ENG R93X and other less rare ENG nonsense variants but more damaging for Q436X. We conclude that PTCs should be distinguished from other loss-of-function variants, PTC transcript levels increase in stressed cells, and readthrough proteins and mechanisms provide promising research avenues.

New genetic drivers in hemorrhagic hereditary telangiectasia

BACKGROUND

Hereditary hemorrhagic telangiectasia (HHT) is a rare vascular disease inherited in an autosomal dominant manner. Disease-causing variants in endoglin (ENG) and activin A receptor type II-like 1 (ACVRL1) genes are detected in around 90% of the patients; also 2% of patients harbor pathogenic variants at SMAD4 and GDF2. Importantly, the genetic cause of 8% of patients with clinical HHT remains unknown. Here, we present new putative genetic drivers of HHT.

METHODS

To identify new HHT genetic drivers, we performed exome sequencing of 19 HHT patients and relatives with unknown HHT genetic etiology. We applied a multistep filtration strategy to catalog deleterious variants and prioritize gene candidates based on their known relevance in endothelial cell biology. Additionally, we performed in vitro validation of one of the identified variants.

RESULTS

We identified variants in the INHA, HIF1A, JAK2, DNM2, POSTN, ANGPTL4, FOXO1 and SMAD6 genes as putative drivers in HHT. We have identified the SMAD6 p.(Glu407Lys) variant in one of the families; this is a loss-of-function variant leading to the activation of the BMP/TGFβ signaling in endothelial cells.

CONCLUSIONS

Variants in these genes should be considered for genetic testing in patients with HHT phenotype and negative for ACVRL1/ENG mutations.

Pathogenic Variant Frequencies in Hereditary Haemorrhagic Telangiectasia Support Clinical Evidence of Protection from Myocardial Infarction

Hereditary haemorrhagic telangiectasia (HHT) is a vascular dysplasia inherited as an autosomal dominant trait, due to a single heterozygous loss-of-function variant, usually in ACVRL1 (encoding activin receptor-like kinase 1 [ALK1]), ENG (encoding endoglin [CD105]), or SMAD4. In a consecutive single-centre series of 37 positive clinical genetic tests performed in 2021-2023, a skewed distribution pattern was noted, with 30 of 32 variants reported only once, but ACVRL1 c.1231C>T (p.Arg411Trp) identified as the disease-causal gene in five different HHT families. In the same centre's non-overlapping 1992-2020 series where 110/134 (82.1%) HHT-causal variants were reported only once, ACVRL1 c.1231C>T (p.Arg411Trp) was identified in nine further families. In a 14-country, four-continent HHT Mutation Database where 181/250 (72.4%) HHT-causal variants were reported only once, ACVRL1 c.1231C>T (p.Arg411Trp) was reported by 12 different laboratories, the adjacent ACVRL1 c.1232G>A (p.Arg411Gln) by 14, and ACVRL1 c.1120C>T (p.Arg374Trp) by 18. Unlike the majority of HHT-causal ACVRL1 variants, these encode ALK1 protein that reaches the endothelial cell surface but fails to signal. Six variants of this type were present in the three series and were reported 6.8-25.5 (mean 8.9) times more frequently than the other ACVRL1 missense variants (all p-values < 0.0039). Noting lower rates of myocardial infarction reported in HHT, we explore potential mechanisms, including a selective paradigm relevant to ALK1's role in the initiating event of atherosclerosis, where a plausible dominant negative effect of these specific variants can be proposed. In conclusion, there is an ~9-fold excess of kinase-inactive, cell surface-expressed ACVRL1/ALK1 pathogenic missense variants in HHT. The findings support further examination of differential clinical and cellular phenotypes by HHT causal gene molecular subtypes.

Pharmacogenomic Considerations for Anticoagulant Prescription in Patients with Hereditary Haemorrhagic Telangiectasia

Hereditary haemorrhagic telangiectasia (HHT) is a vascular dysplasia that commonly results in bleeding but with frequent indications for therapeutic anticoagulation. Our aims were to advance the understanding of drug-specific intolerance and evaluate if there was an indication for pharmacogenomic testing. Genes encoding proteins involved in the absorption, distribution, metabolism, and excretion of warfarin, heparin, and direct oral anticoagulants (DOACs) apixaban, rivaroxaban, edoxaban, and dabigatran were identified and examined. Linkage disequilibrium with HHT genes was excluded, before variants within these genes were examined following whole genome sequencing of general and HHT populations. The 44 genes identified included 5/17 actionable pharmacogenes with guidelines. The 76,156 participants in the Genome Aggregation Database v3.1.2 had 28,446 variants, including 9668 missense substitutions and 1076 predicted loss-of-function (frameshift, nonsense, and consensus splice site) variants, i.e., approximately 1 in 7.9 individuals had a missense substitution, and 1 in 71 had a loss-of-function variant. Focusing on the 17 genes relevant to usually preferred DOACs, similar variant profiles were identified in HHT patients. With HHT patients at particular risk of haemorrhage when undergoing anticoagulant treatment, we explore how pre-emptive pharmacogenomic testing, alongside HHT gene testing, may prove beneficial in reducing the risk of bleeding and conclude that HHT patients are well placed to be at the vanguard of personalised prescribing.

Functional filter for whole-genome sequencing data identifies HHT and stress-associated non-coding SMAD4 polyadenylation site variants >5 kb from coding DNA

Despite whole-genome sequencing (WGS), many cases of single-gene disorders remain unsolved, impeding diagnosis and preventative care for people whose disease-causing variants escape detection. Since early WGS data analytic steps prioritize protein-coding sequences, to simultaneously prioritize variants in non-coding regions rich in transcribed and critical regulatory sequences, we developed GROFFFY, an analytic tool that integrates coordinates for regions with experimental evidence of functionality. Applied to WGS data from solved and unsolved hereditary hemorrhagic telangiectasia (HHT) recruits to the 100,000 Genomes Project, GROFFFY-based filtration reduced the mean number of variants/DNA from 4,867,167 to 21,486, without deleting disease-causal variants. In three unsolved cases (two related), GROFFFY identified ultra-rare deletions within the 3' untranslated region (UTR) of the tumor suppressor SMAD4, where germline loss-of-function alleles cause combined HHT and colonic polyposis (MIM: 175050). Sited >5.4 kb distal to coding DNA, the deletions did not modify or generate microRNA binding sites, but instead disrupted the sequence context of the final cleavage and polyadenylation site necessary for protein production: By iFoldRNA, an AAUAAA-adjacent 16-nucleotide deletion brought the cleavage site into inaccessible neighboring secondary structures, while a 4-nucleotide deletion unfolded the downstream RNA polymerase II roadblock. SMAD4 RNA expression differed to control-derived RNA from resting and cycloheximide-stressed peripheral blood mononuclear cells. Patterns predicted the mutational site for an unrelated HHT/polyposis-affected individual, where a complex insertion was subsequently identified. In conclusion, we describe a functional rare variant type that impacts regulatory systems based on RNA polyadenylation. Extension of coding sequence-focused gene panels is required to capture these variants.

Executive summary of the 14th HHT international scientific conference

Hereditary Hemorrhagic Telangiectasia (HHT) is an autosomal dominant vascular disorder characterized by small, dilated clustered vessels (telangiectasias) and by larger visceral arteriovenous malformations (AVMs), which directly connect the feeding arteries with the draining veins. These lesions are fragile, prone to rupture, and lead to recurrent epistaxis and/or internal hemorrhage among other complications. Germline heterozygous loss-of-function (LOF) mutations in Bone Morphogenic Protein 9 (BMP9) and BMP10 signaling pathway genes (endoglin-ENG, activin like kinase 1 ACVRL1 aka ALK1, and SMAD4) cause different subtypes of HHT (HHT1, HHT2 and HHT-juvenile polyposis (JP)) and have a worldwide combined incidence of about 1:5000. Expert clinicians and international scientists gathered in Cascais, Portugal from September 29th to October 2nd, 2022 to present the latest scientific research in the HHT field and novel treatment strategies for people living with HHT. During the largest HHT scientific conference yet, participants included 293 in person and 46 virtually. An impressive 209 abstracts were accepted to the meeting and 59 were selected for oral presentations. The remaining 150 abstracts were presented during judged poster sessions. This review article summarizes the basic and clinical abstracts selected as oral presentations with their new observations and discoveries as well as surrounding discussion and debate. Two discussion-based workshops were also held during the conference, each focusing on mechanisms and clinical perspectives in either AVM formation and progression or current and future therapies for HHT. Our hope is that this paper will represent the current progress and the remaining unanswered questions surrounding HHT, in order to serve as an update for those within the field and an invitation to those scientists and clinicians as yet outside of the field of HHT.

Unsupervised machine learning algorithms identify expected haemorrhage relationships but define unexplained coagulation profiles mapping to thrombotic phenotypes in hereditary haemorrhagic telangiectasia

Hereditary haemorrhagic telangiectasia (HHT) can result in challenging anaemia and thrombosis phenotypes. Clinical presentations of HHT vary for relatives with identical casual mutations, suggesting other factors may modify severity. To examine objectively, we developed unsupervised machine learning algorithms to test whether haematological data at presentation could be categorised into sub-groupings and fitted to known biological factors. With ethical approval, we examined 10 complete blood count (CBC) variables, four iron index variables, four coagulation variables and eight iron/coagulation indices combined from 336 genotyped HHT patients (40% male, 60% female, 86.5% not using iron supplementation) at a single centre. T-SNE unsupervised, dimension reduction, machine learning algorithms assigned each high-dimensional datapoint to a location in a two-dimensional plane. k-Means clustering algorithms grouped into profiles, enabling visualisation and inter-profile comparisons of patients' clinical and genetic features. The unsupervised machine learning algorithms using t-SNE and k-Means identified two distinct CBC profiles, two iron profiles, four clotting profiles and three combined profiles. Validating the methodology, profiles for CBC or iron indices fitted expected patterns for haemorrhage. Distinct coagulation profiles displayed no association with age, sex, C-reactive protein, pulmonary arteriovenous malformations (AVMs), ENG/ACVRL1 genotype or epistaxis severity. The most distinct profiles were from t-SNE/k-Means analyses of combined iron-coagulation indices and mapped to three risk states - for venous thromboembolism in HHT; for ischaemic stroke attributed to paradoxical emboli through pulmonary AVMs in HHT; and for cerebral abscess attributed to odontogenic bacteremias in immunocompetent HHT patients with right-to-left shunting through pulmonary AVMs. In conclusion, unsupervised machine learning algorithms categorise HHT haematological indices into distinct, clinically relevant profiles which are independent of age, sex or HHT genotype. Further evaluation may inform prophylaxis and management for HHT patients' haemorrhagic and thrombotic phenotypes.