Fig 2: De-repressed IAPez LTR retroelement sequences drive the expression of a fusion TE-Capn11 transcript in Zfp92 KOs.A) 5’-RACE analysis of mRNA from Zfp92 KO islet sample revealed the production of a TE-Capn11 fusion transcript when the IAPez ERVK LTR element is derepressed. The panel shows a schematic representation of Capn11 gene (blue) and a relative location of B1 SINE (light green) and IAPez ERVK LTR (tan) TEs within the Capn11 intron 3, location of 5’ RACE primers (blue arrowheads), schematic, and the sequence of obtained 5’RACE product that represents fusion transcript. The IAPez-derived sequence is highlighted in tan, B1 SINE-derived sequence–in light green and a Capn11 downstream exon sequence—in blue. ZFP92 binding sequence is in red font. B) Structure of the full-length CAPN11 and N-terminally truncated CAPN11 protein that is encoded by the open reading frame within the TE-Capn11 fusion transcript.

Fig 3: RNA-seq analysis of islets from Zfp92 KOs shows an increase in Capn11 and a decrease in Mafb gene expression.A) Volcano plot showing differentially expressed genes (Log2Fold Change (FC) over padj-value) in Zfp92 KO vs WT islets from 15 week-old male mice (N = 4). The top 10 differentially expressed genes (based on padj-value) are indicated by names and total numbers of downregulated (green) and upregulated (red) genes are provided (padj<0.05). B) Heat map showing clustering of the top variant genes (based on p-value) dysregulated in Zfp92 KO vs WT islets. C. RT-qPCR analysis of RNA expression for select genes in islet RNA samples collected from 15 week-old male mice. Zfp92 wild type (WT), and knockout (KO) samples. N = 4, error bars: ± SEM. ***p=0.001; *p=0.05. p-value is determined by an unpaired t-test.

Fig 4: Genome-wide analysis reveals ZFP92 binds preferentially to B1/Alu SINE elements.A) Pie chart showing the percentage distribution of different TE classes within the ZFP92 binding peaks obtained by CUT&RUN analysis. 93% of peaks have TEs with SINE elements being the largest group. The inset pie chart shows the family distribution of ZFP92-bound SINEs, with Alu/B1 family being most represented. B) Pie chart showing the distribution of ZFP92 binding peaks within different genomic features. C) Pie chart showing the distribution of different ENCODE Candidate Cis-Regulatory Elements (cCREs) within the ZFP92 binding peaks. enhD, distal enhancer signature; enhP, proximal enhancer-like signature; prom, promoter-like signature; CTCF, CTCF-only; K4m3, DNase-H3K4me3. D) UCSC Genome Browser view of Capn11 locus showing ZFP92 binding peaks and RNA-seq reads in WT and Zfp92 KO samples (GRCm38/mm10 assembly). The main peak containing the B1 SINE element is highlighted in gray, with red lines denoting amplicons and primer pairs that were used for qPCR shown in (E). E) qPCR on CUT&RUN chromatin samples with the primers shown in (D) to different regions of Capn11 locus and TE elements found there. (N = 3) Error bars: ± SEM. **p=0.01; *p=0.05. p-value is determined by an unpaired t-test. F) ZFP92 binding motif predicted by MEME analysis of ZFP92 binding peaks. Under the ZFP92 motif are the two similar motifs (ZNF460 and ZNF135) that were identified by TomTom analysis from the MEME suite. G) Pie chart showing the distribution of ZFP92 binding motif within TE elements in ZFP92 peaks. The inset pie chart shows the family distribution of SINEs with FP92 motif, with Alu/B1 family being most represented.