Transcriptomic analysis uncovers hidden intronic and synonymous pathogenic variants in red cell membrane disorders Free

Recessive hereditary spherocytosis (HS) is frequently caused by biallelic pathogenic variants in SPTA1, while monoallelic variants in SPTB is a cause of dominant HS. SPTA1 and SPTB encode α- and β-spectrin respectively (two essential proteins of the red cell membrane skeleton). Severe forms of HS may present with in utero hydrops fetalis, neonatal jaundice, transfusion dependence, and iron overload. While next-generation sequencing has improved diagnostic yield, its typical focus on exonic and near-splice site regions may leave many patients undiagnosed or harboring variants of uncertain significance (VUS), especially noncoding or synonymous variants that do not have functional interpretation.

To address this limitation, we implemented an integrated genomic-transcriptomic approach combining whole genome sequencing (WGS), RNA sequencing (RNAseq) of patient-derived reticulocytes, and red cell phenotyping to a cohort of patients with suspected red cell membrane disease. This strategy enabled us to functionally characterize deep intronic or synonymous VUS in red cell structural genes. This approach led to the discovery and mechanistic validation of two novel deep intronic pathogenic SPTA1 variants and one synonymous splicing-disruptive SPTB variant. In one case, an infant with hydrops fetalis, requiring in-utero transfusions and persistent transfusion dependency with reticulocytopenia after delivery, was found to be homozygous for an intronic SPTA1 VUS, c.6531-16G>A. Severe reduction of SPTA1 expression was confirmed by qRT-PCR in bone marrow CD71⁺ erythroid precursors; RNAseq prepared from these cells revealed a 14 bp intronic retention (r.6530_6531ins[6531-14_6531-1]) that causes a frameshift (p.Ala2178Profs*37) and is predicted to result in nonsense-mediated mRNA decay (NMD), establishing a diagnosis of complete α-spectrin deficiency and guiding management toward chronic transfusions and aggressive chelation followed by hematopoietic stem cell transplant.

In a second family, a child with clinical HS was compound heterozygous for the well-characterized α-LEPRA (c.4339-99C>T) and three maternally inherited intronic VUS. RNAseq identified c.5189+40C>T as the sole variant causing splicing disruption, specifically, a 38 bp cryptic exon inclusion (r.5189_5190ins[5189+1_5189+38]) leading to a frameshift (p.Glu1731Tyrfs*16) and predicted NMD. No splicing changes were associated with the other two variants (c.5432+181G>T and c.5911-279G>A), allowing their reclassification as benign. Quantification of red cell α-spectrin showed a ~70% reduction, consistent with recessive HS and predictive of favorable response to splenectomy.

In a third family, the proband presented in early infancy with transfusion requirement and was found to carry a synonymous SPTB variant, c.4224C>T p.(Gly1408=), originally classified as a VUS. RNAseq demonstrated that this variant activated a cryptic donor site 2 bp upstream, resulting in a 44-bp deletion at the 3' end of exon 20 in the mature transcript (r.4223_4226del), producing a frameshift (p.Gln1408Alafs*68) and predicted NMD. The variant segregated with HS across three generations on the paternal side. β-spectrin protein levels were reduced in the red cells of the proband and affected father, confirming its pathogenic impact and justifying reclassification of this synonymous variant as pathogenic.

These findings expand the mutational spectrum of spectrin-associated membrane disorders and underscore the diagnostic blind spots left by exon-centric sequencing. While α-LEPRA is well established as a pathogenic intronic allele, our results demonstrate that additional deep intronic or synonymous changes in SPTA1 and SPTB can disrupt splicing, reducing spectrin protein expression, and cause clinically significant HS. Functional transcriptomic integration allowed mechanistic dissection of VUS and direct links to patient phenotype, offering diagnostic resolution where conventional sequencing failed.

Conclusion: Deep intronic and synonymous variants are underrecognized contributors to hereditary red cell membrane disorders. Transcriptomic analysis, particularly RNAseq of reticulocytes, enables visualization of splicing defects, supports functional reclassification of VUS, and transforms previously inconclusive genetic findings into actionable diagnoses. Our findings establish transcriptome-based diagnostics as a powerful complement to WGS and a necessary tool in the precision evaluation of rare anemias.