Fig. 1: Detection/Identification of alternative NOS1AP/Nos1ap transcripts in mice and humans/ murine and human kidney tissue.

A Cartoon depicting the syntenic NOS1AP/Nos1ap genomic locus in humans and mice is shown (left). Exons are indicated that contribute to the canonical and a non-canonical (intergenic) transcript. The latter results from intergenic splicing that joins 153 nucleotides from exon 10 of NOS1AP/Nos1ap to coding exons 1 and 2 of C1orf226/Gm7694 (NM_001085375/NM_001198955). Some annotated transcripts include a non-coding exon, which is here labeled as “1*”. Arches symbolize splicing events in the Sashimi plot style. The intergenic splice junction is highlighted by a red line. mRNA transcription and processing yield two classes of transcripts (canonical and intergenic) seen in the middle, which are then translated to produce two classes of proteins (right). Both canonical and intergenic proteins contain the phosphotyrosine binding (PTB) domain, while only the canonical protein bears the NOS1 enzyme-binding PDZ domain (PDZ-BD). Exons/protein regions coded in blue arise from the canonical NOS1AP/Nos1ap locus, while those in green are generated from the C1orf226/Gm7694 locus. See Fig. S2 for further details of the repertoire of transcripts and protein products. B Cartoon depicting the murine Nos1ap intergenic splice product is shown. Exons are indicated that arise from the canonical Nos1ap (blue) or adjacent Gm7694 (green) loci. The non-canonical transcript results from intergenic splicing that joins the first 153 nucleotides from exon 10 of Nos1ap to coding exons 1 and 2 of Gm7694. An amplicon spanning from early Nos1ap into coding exon 2 of Gm7694 was identified by RT-PCR in adult C57BL/6 J mouse kidney total RNA. Aligned Sanger sequencing reads are shown below the transcript diagram. C Sanger sequencing chromatogram of the PCR amplicon in (B) was aligned to the intergenic splice transcript, demonstrating a contiguous transcript with the expected in-frame continuation from Nos1ap exon 10 to Gm7694 coding exon 1 ( = exon 11 of the intergenic transcript). The dashed gray box shows the nucleotides contributing to the UTR of the non-intergenic Gm7694 transcript NM_001198955. D Quantitative RT-PCR was performed from cerebrum and kidney total RNA of 5 individual adult C57BL/6J mice. Consistent with human and rat studies, relative levels of Nos1ap early exons reflecting both long canonical and intergenic transcripts (normalized to Actb) were expressed at higher levels in brain compared to kidney tissue (median 2.4-fold, p = 0.0079). Relative levels of canonical-specific exons (Nos1ap exon 9 to late canonical exon 10) were expressed at even higher levels in brain relative to kidney tissue (median 32.3-fold, p = 0.0079). In contrast, relative levels of intergenic-specific exons (Nos1ap exon 9 to Gm7694 coding exon 1) were increased in the kidney relative to brain levels (median 9.3-fold, p = 0.0079). Bar represents median values. Mann–Whitney test performed (**p < 0.01). E Bulk short-read RNA-sequencing data from four 8-week-old mice (NCBI GEO dataset GSE145053, https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE145053) were evaluated to quantify reads spanning the intergenic splice junction (GRCm38/mm10 chr1:170,318,737-170,318,738; representing canonical transcript) versus those split between exon 10 of Nos1ap and coding exon 1 of Gm7694 (GRCm38/mm10 chr1:170,302,842; representing intergenic transcript). The percentage of canonical and intergenic reads (total of 66 reads from four mice) is shown using a 5 bp sliding window approach and demonstrates that reads reflecting the intergenic transcript predominate (94.5%). F Cartoon is shown depicting the human NOS1AP intergenic splice product. Exons are indicated that arise from the canonical NOS1AP (blue) or adjacent C1orf226 (green) loci. The non-canonical transcript results from intergenic splicing that joins the first 153 nucleotides from exon 10 of NOS1AP to coding exons 1 and 2 of C1orf226 (NM_001085375). The lower image shows aligned Sanger sequencing reads spanning nearly the full-size amplicon (1943 bp), which was identified by RT-PCR from adult human kidney tissue. This product starts within the 5’ UTR of NOS1AP and ends close to the stop codon within coding exon 2 of C1orf226. G Sanger sequencing plot resulting from the 1943 bp amplicon in (F) is shown, demonstrating a contiguous transcript with the expected in-frame continuation from NOS1AP exon 10 to C1orf226 coding exon 1 ( = exon 11 of the intergenic transcript). The dashed gray box shows the nucleotides contributing to the UTR of the non-intergenic C1orf226 transcript NM_001085375. H Histogram displays percentage of NOS1AP transcripts reflecting canonical versus intergenic splice products by tissue in ENCODE long-read RNA-sequencing data. The total number of issue samples (t) and reads (r) is noted above each bar. I Violin plot displays percentage of NOS1AP transcripts reflecting canonical versus intergenic splice products from short-read RNA-sequencing data of glomerular samples from the NEPTUNE cohort. Red bar representing median values. Only samples with >5 reads at the intergenic splice junction in NOS1AP exon 10 were included. J Bulk short-read RNA-sequencing data from human fetal kidney (ENCODE) were evaluated to quantify reads spanning the intergenic splice junction versus those split between exon 10 of NOS1AP and coding exon 1 of C1orf226. The percentage of canonical and intergenic reads (r) was assessed, demonstrating that reads reflecting the intergenic transcript predominate in a fetal kidney from 20- and 24-weeks of gestation.