American journal of human genetics最新文献

Rare brain disorders often present with changes in brain volume, and variation in brain volume is known to be highly heritable. Recent work studying brain volume variation has largely focused on common variants and structural variants. Rare variants often have large effect sizes and clearer connections to biological mechanisms, but the role of rare variants has not been extensively studied. We performed rare-variant gene aggregation analysis for total brain volume and 43 regional brain volume phenotypes (n = 50,061) to identify genes associated with brain volume variation through loss-of-function and missense variants. We identified and replicated mutations in DISP1 and SCUBE2 that were associated with reduced cerebellar volume and suggest that this was mediated by modifying sonic hedgehog signaling. Additionally, we found an association between mutations in PTEN and macrocephaly that are likely mediated through the PI3K/mTOR pathway and hypothesize that mutations in FA2H influence cerebral white matter volume. Further, we identified 7 genes associated with volume variation in the population and rare brain diseases in ClinVar, supporting the role of mutations in these genes causing diseases and related subclinical phenotypes. Overall, we showed that rare-variant analysis can provide clarity on the biological processes connecting brain volume and disease.

Retinitis pigmentosa (RP) is an inherited retinal disease (IRD) characterized usually by progressive photoreceptor degeneration, leading to night blindness, peripheral visual field loss, and can progress to central vision impairment in some individuals. Despite advances in genomic diagnostics, many individuals with RP remain without a molecular diagnosis. We identified bi-allelic ultra-rare variants in fibronectin type II and Spry domain-containing protein 1-like (FSD1L) in six individuals with RP with or without neurological features from four unrelated families. FSD1L encodes a cytoplasmic protein, variants of which have not previously been associated with Mendelian disease. The gene is expressed in both human and mouse retinas that are enriched in cone and rod photoreceptors. Immunofluorescence and ultrastructure expansion microscopy show that FSD1L localizes along the photoreceptor microtubule axoneme, including the connecting cilium and outer segment, supporting a possible role in intracellular trafficking. A retina-enriched isoform of FSD1L includes an alternatively spliced exon (exon 10b), which we characterize as absent in minigene assays and affected individual-derived lymphocytes due to a deep intronic 26 nt deletion. Together, these findings support the association between bi-allelic disruption of FSD1L and IRD.

Newborn screening (NBS) for spinal muscular atrophy (SMA) enables rapid diagnosis and pre-symptomatic treatment of infants with bi-allelic SMN1 deletions. Standard PCR-based assays detect ∼95% of cases by identifying the absence of SMN1 exon 7; however, rare sequence variants can escape detection. We describe two newborns (in Germany and Australia) identified by NBS as lacking SMN1 but subsequently shown to carry a single SMN1 copy-with no SMN2 in P1 and one SMN2 copy in P2. Gene-specific long-range PCR and Sanger sequencing revealed two distinct 4-bp deletions in SMN1 exon 7 (c.855_858delAGAA [p.Arg288AlafsTer5] in P1 and c.861_864delAAGG [p.Arg288AlafsTer5] in P2). Both variants disrupt the reverse primer-binding site used in NBS assays and cause the same frameshift p.Arg288AlafsTer5, predicted to be deleterious. A plethora of assays demonstrated preserved exon 7 splicing, markedly reduced SMN protein abundance, and wild-type-like protein thermostability. In vivo, expression of the p.Arg288AlafsTer5 protein in zebrafish fully rescued the progressive motor and survival defects of smn1-deficient mutants. These findings raise the possibility that this novel SMN isoform has enhanced functional efficiency relative to the wild type. Population data (gnomAD) suggest that ∼800 individuals of European ancestry may carry these variants in trans with an SMN1 deletion, yet none have been reported with SMA. Based on our data, no therapy was initiated. Both children remain healthy at 24 months of age, avoiding >US$4 million in potential treatment costs. These findings challenge the assumption that complete loss of full-length SMN invariably causes SMA and suggest that very low levels of this novel SMN isoform can sustain normal motor development.

Myotonic dystrophy type 1 (DM1) is a multisystemic disorder caused by CTG repeat expansions in DM1 protein kinase (DMPK). Mutant transcripts containing expanded CUG repeats form ribonuclear foci that sequester muscleblind-like (MBNL) splicing regulator proteins, key regulators of RNA splicing and metabolism. This functional depletion leads to widespread mis-splicing and persistence of fetal transcript profiles, which underlie muscle weakness, myotonia, and muscle atrophy. In addition, miR-23b is upregulated in DM1 muscle and further represses MBNL1 translation, amplifying molecular defects. We developed chemically optimized microRNA (miRNA)-targeting antisense oligonucleotides (antimiRs) to inhibit miR-23b and restore functional MBNL1 levels. Using a multi-step screening process, we evaluated antimiRs with varying sequences, lengths, chemical modifications, and lipid conjugations. A key optimization was a 3'-oleic acid conjugation combined with specific chemical modifications, which enhanced muscle uptake and efficacy. Lead candidates showed strong activity in preclinical models (human skeletal actin [HSA]^LR and DMSXL mice and human myoblasts), increasing MBNL1 levels, correcting mis-splicing, improving muscle strength, and reducing myotonia. They also exhibited efficient biodistribution to skeletal muscle, a critical DM1-affected tissue. In vitro toxicology indicated a favorable safety profile with minimal immune or renal toxicity. The antimiR mechanism was conserved in rat and pig fibroblasts. Overall, two lead antimiRs emerged as promising therapeutic candidates for DM1, with improved pharmacokinetics, tissue targeting, and safety, supporting the potential of microRNA-based approaches to correct key molecular defects in this disorder.

Genomic medicine is increasingly being integrated into healthcare systems worldwide, requiring a skilled genomic workforce. Variant interpretation (VI) is crucial to genomic testing, yet there are no agreed professional competencies in VI, career pathways are ill-defined for some professions, and there is limited literature on educational needs and programs. We referenced education theory and curricula from nascent local VI training activities to co-develop learning outcomes in VI with content experts that informed scalable continuing education activities, evaluated using longitudinal cross-sectional surveys. We defined 16 learning outcomes in VI covering fundamental genetics and bioinformatics theory and the stages of variant identification, curation, and classification. The education program included options for self-directed online learning or blended learning (online pre-reading plus workshops). Program reach was 951 individuals (49.2% scientists and 27.8% clinicians). At completion, the majority reported increased understanding of learning outcomes (96.1%, 171/178) and increased confidence in the processes of VI (93.3%, 166/178). The majority (91.7%, 231/252) indicated that the learnings would impact their professional role. Our education program was effective at developing entry-level proficiency in VI across different professions. The learning outcomes can inform multiple aspects of VI education and training programs, including defining the desired level of mastery and program content and aligning evaluation measures. They also provide a basis for an educational framework to inform competencies in VI across multiple professions and for career benchmarking more broadly.