Use of minigene assays as a useful tool to confirm the pathogenic role of intronic variations of the <scp><i>ANK1</i></scp> gene: Report of two cases of hereditary spherocytosis

ANK1 is one of the most frequently mutated genes in hereditary spherocytosis (HS) in European countries and encodes the protein Ankyrin-1 which is a major membrane erythrocyte protein linking horizontal protein network to the vertical one. Inheritance can be either autosomal dominant or recessive depending on the type of variations.1, 2 According to the American College of Medical Genetics and Genomics (ACMG) classification,3 loss of function variations (non-sense, frameshift, canonical splicing regions) of the ANK1 gene is dominant and can be undoubtedly classified at least as probably pathogenic (class 4) in a context of HS. On the contrary, variations affecting deeper regions of introns cannot be classified higher than class 3 (variation of uncertain significance [VUS]) without functional characterization. We recently encountered that situation in two typical cases of HS (a 61-year-old man, patient 1 and a 15-year-old boy, patient 2). We performed NGS analysis targeting known genes implicated in red blood cell membrane diseases for these two patients (Supplementary data: NGS panel genes list). Molecular analysis revealed the presence of an ANK1 intronic variation at heterozygous state: NM_020476 (ANK1): c.1405-9G>A (intron 13) in patient 1 and c.5097-33G>A (intron 38) in patient 2. Both variations were absent in the gnomAD v3.1.2 database and all used splicing prediction software (Splice Site Finder-like, MaxEntScan, NNSPLICE, GeneSplicer) concluded to altered splicing (Figure 1). The intron 13 variation led to the decrease in the canonical acceptor site strength and to the creation of a new strong cryptic acceptor site, whereas the intron 38 variation led to the creation of a cryptic acceptor site with higher scores than the canonical one. Moreover, these two intronic ANK1 variations had already been referenced in a 2020 publication by Tole et al.4 who established a list of potentially implicated variations in a cohort of paediatric HS cases. Interestingly, thanks to the corresponding author, we learned that all the described HS patients harbouring the −33 variation were family related. Their common ancestor was a farmer who lived in Toronto region at the end of the 19th century, which strongly suggested a founder effect. The −9 variation was associated with a new HS description 1 year later in a Chinese family.5 Taken together, these arguments were strongly in favour of these two variations of causality in the diagnosis of HS. Unfortunately, no functional studies were performed in these two publications. We performed a minigenes analysis on both variations. Those results allowed us to reclassify it as pathogenic variations according to ACMG classification. It is to note that these two minigenes were performed in two distinct laboratories. The precise and complete description of each protocol is available in supplementary data but the principle remains the same. Briefly, total RNA was purified from cultured cells transfected with empty plasmid, minigene containing partial wild-type (WT) or mutated ANK1 intron, prior natural n + 1 exon and partial n + 1 intron. RT-PCR products using primers located in the plasmid exons before and after the ANK1 exon of interest were obtained (Figure 2A,D), analysed on agarose gel (Figure 2B,E) and sequenced using Sanger analysis (Figure 2C–F). For the ANK1:c.1405-9G>A mutation (intron 13), both the normal and mutated minigene gives rise to a major RT-PCR product of about 450 bp which may correspond to the normal splicing of exon 14 (expected size = 445 bp) (Figure 2B). Interestingly, a very minor product of about 250 bp was also observed which may correspond to exon 14 skipping (expected size = 247 bp). Sanger sequencing confirmed these hypotheses but, for the 450-bp product of the mutated minigene, exon 14 was preceded by the last seven bases of intron 13 (Figure 2C). The ANK1:c.1405-9G>A mutation has thus a complete modifying effect on splicing with the exonisation of the 7 last bases of intron 13 that gives rise to a frameshift leading to a premature stop-codon a few bases later. For the ANK1:c.5097-33G>A mutation (intron 38), the wild-type minigene gives a major RT-PCR product of 537 bp while the corresponding one for the mutated minigene was 31 bp longer (568 bp) (Figure 2E). A minor PCR product of about 411 bp for WT and 442 bp for mutated construct was also observed which may correspond to the creation of an artefactual donor site in the exon 39 sequence. It was thus confirmed that the ANK1:c.5097-33G>A mutation creates a new splicing site triggering frameshift due to 31 bp insertion leading to a premature non-sense codon UGA in position c.5217 (p.1739) in exon 39 (Figure 2F). No normal transcript was produced after the transfection of the mutated minigene. In conclusion, our study proves again the high usefulness of minigenes studies in functional testing of intronic VUS in the field of hereditary hemolytic anaemia.6 It also emphasizes the importance of searching for deep intronic variations out of usual flanking intronic regions in next generation-sequencing analysis pipelines. Ariane Lunati-Rozie wrote the paper, Alexandre Janin and Emmanuelle Faubert performed minigene study, Severine Nony made analytical technic, Céline Renoux made genetic diagnosis, Manuel D. Carcao gave clinical information about patients, Pascale Fanen designed minigen assay, Benoît Funalot performed genetic analysis, Lamisse Mansour-Hendili and Philippe Joly wrote and directed the study. None. None. The authors have no competing interests. All patients provided written informed consent. Not applicable. Data S1. 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.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.