Non-coding cause of congenital heart defects: Abnormal RNA splicing with multiple isoforms as a mechanism for heterotaxy

Heterotaxy is a disorder characterized by severe congenital heart defects (CHDs) and abnormal left-right patterning in other thoracic or abdominal organs. Clinical and research-based genetic testing has previously focused on evaluation of coding variants to identify causes of CHDs, leaving non-coding causes of CHDs largely unknown. Variants in the transcription factor Zinc finger of the cerebellum 3 (ZIC3) cause X-linked heterotaxy. We identified an X-linked heterotaxy pedigree without a coding variant in ZIC3. Whole genome sequencing revealed a deep intronic variant (ZIC3 c.1224+3286A>G) predicted to alter RNA splicing. An in vitro minigene splicing assay confirmed the variant acts as a cryptic splice acceptor. CRISPR/Cas9 served to introduce the ZIC3 c.1224+3286A>G variant into human embryonic stem cells demonstrating pseudoexon inclusion caused by the variant. Surprisingly, Sanger sequencing of the resulting ZIC3 c.1224+3286A>G amplicons revealed several isoforms, many of which by-pass the normal coding sequence of the third exon of ZIC3, causing a disruption of a DNA binding domain and a nuclear localization signal. Short- and long-read mRNA sequencing confirmed these initial results and identified additional splicing patterns. Assessment of four isoforms determined abnormal functions in vitro and in vivo while treatment with a splice-blocking morpholino partially rescued ZIC3. These results demonstrate that pseudoexon inclusion in ZIC3 can cause heterotaxy and provide functional validation of non-coding disease causation. Our results suggest the importance of non-coding variants in heterotaxy and the need for improved methods to identify and classify non-coding variation that may contribute to CHDs. CRISPR/Cas9 was used in H1-OCT4-eGFP cells (a male human embryonic stem cell line) to produce two cell clones containing the ZIC3 intronic variant (NM_003413.4:c.1224+3286A>G) termed ZIC3 AtoG_C1 and ZIC3 AtoG_C2. Two ZIC3 knockout cell clones were also produced (ZIC3 KO_C1: NM_003413.4:c.190_200delinsG; ZIC3 KO_C2: NM_003413.4:c.200_201del). Unedited H1-OCT4-eGFP cells were termed ZIC3 WT. Short-read RNA sequencing was completed for ZIC3 WT ( n=1), ZIC3 AtoG_C1 (n=1), ZIC3 AtoG_C2 (n=1), ZIC3 KO_C1 (n=1), and ZIC3 KO_C2 (n=1) H1-OCT4-EGFP cells using the Illumina sequencing platform. Long-read RNA sequencing was completed in ZIC3 WT (n=3), ZIC3 AtoG_C1 (n=3), and ZIC3 KO_C1 (n=3) H1-OCT4-EGFP cells using the Oxford Nanopore sequencing platform.