Physiological role and mechanisms of action for a long noncoding haplotype region

Abstract

Direct targeting of noncoding genomic regions harboring common sequence variants associated with human traits through in vivo animal model studies and precise genome editing in human cells is essential for closing the critical gap between genetic discoveries and physiological understanding. However, such investigation has been impractical for many of these variants as they are in haplotypes containing multiple single-nucleotide polymorphisms (SNPs) spanning thousands of base pairs and have small effect sizes. We developed an integrated approach to address this challenge, combining an efficient two-step technique to precisely edit large haplotypes in human induced pluripotent stem cells and orthologous region deletion in phenotypically permissive animal models. As proof of principle, we applied this approach to examine a blood pressure-associated locus with a noncoding haplotype containing 11 SNPs spanning 17.4 kbp. We found a robust blood pressure effect of nearly 10 mmHg and identified the physiological and molecular mechanisms involved.

Keywords: CP: Genomics; animal model; blood pressure; chromatin interaction; genome editing; haplotype; hypertension; induced pluripotent stem cells; noncoding; salt; single-nucleotide polymorphism.

Figure 1.. An efficient, integrated, and targeted approach to examining the physiological roles and mechanisms of action for long noncoding haplotypes

SNP, single-nucleotide polymorphism; hiPSCs, human induced pluripotent stem cells.

Figure 2.. Deletion of the rs1173771 LD orthologous region attenuates salt-induced hypertension in male SS rats

(A) Identification of the rat genomic region orthologous to the human rs1173771 LD (linkage disequilibrium) region. See Figure S2 for more details. (B) Deletion of the rs1173771 LD orthologous region from the genome of SS rats. (C–H) Daily averages of mean arterial pressure (MAP), systolic blood pressure (SBP), diastolic blood pressure (DBP), pulse pressure, and heart rate and hourly averages of MAP on days 13–14 of 4% NaCl high-salt diet (HSD) of male SS-Δrs1173771LD^−/− rats (ΔLD) and SS littermates (WT). Data were expressed as mean ± SEM. n = 8; ##p < 0.01 for ΔLD vs. WT, two-way ANOVA repeated measures; *p < 0.05 for ΔLD vs. WT at the given time point, Holm-Sidak test.

Figure 3.. Deletion of the rs1173771 LD orthologous region leads to upregulation of arterial Npr3 expression and improves CNP-induced vasodilation

(A) Upregulation of Npr3 in mesentery arteries in male SS-Δrs1173771LD^−/− rats (ΔLD) compared with SS littermates (WT), based on RNA-seq. The schematic at the top shows the genomic positions of the rs1173771 LD orthologous region relative to Npr3, Sub1, and Tars genes. Data were expressed as mean ± SEM. n = 4; *p < 0.05, unpaired t test. (B) Western blot analysis of NPR-C in mesentery arteries from male ΔLD and WT rats. n = 9; *p < 0.05, unpaired t test. (C) Immunofluorescence analysis of NPR-C in mesentery arteries from male ΔLD and WT rats. n = 3; **p < 0.01, unpaired t test. (D) Vasodilation response of mesentery arteries from male ΔLD and WT rats in response to increasing concentrations of C-type natriuretic peptide (CNP). AP811, an NPR-C antagonist, was used at 100 nM. n = 6; **p < 0.01 for ΔLD vs. WT by Holm-Sidak test. (E) Npr3 expression in the kidney cortex in male ΔLD and WT rats based on RNA-seq. n = 4. (F) Western blot analysis of NPR-C in the kidney cortex from male ΔLD and WT rats. n = 9 WT and 10 ΔLD; *p < 0.05, unpaired t test. (G) Serum levels of atrial natriuretic peptides (ANPs), B-type natriuretic peptides (BNPs), and CNPs. n = 7.

Figure 4.. An efficient, two-step genome editing approach for generating isogenic iPSCs with homozygous alleles or haplotypes

(A) The rs1173771 haplotype contains 11 SNPs spanning 17.4 kbp. (B and C) Deletion and reconstitution approaches for generating isogenic human iPSC lines with homozygous alleles at the single SNP rs1173771 (B) and all 11 SNPs in the rs1173771 haplotype (C) are shown.

Figure 5.. Deletion of the rs1173771 haplotype region results in upregulation of NPR3 in iECs and iVSMCs

(A) Locations of various PCR amplicons relative to the rs1173771 haplotype region. (B–E) Confirmation of rs1173771 haplotype region deletion in Y4 and 39b cells by PCR (B and C) and Sanger sequencing (D and E). (F) mRNA expression (mean ± SEM) of neighboring genes NPR3, SUB1, and TARS in endothelial cells (CD144⁺) differentiated from iPSC Y4 (WT) and rs1173771 LD region-deleted Y4 (ΔLD). n = 8 (WT) and 12 (ΔLD) independently differentiated samples from 3 rounds of differentiation. *p ≤ 0.05, unpaired t test. (G) mRNA expression (mean ± SEM) of NPR3, SUB1, and TARS in endothelial cells (CD144⁺) differentiated from iPSC 39b (WT) and rs1173771 LD region-deleted 39b (ΔLD). n = 6 (WT) and 6 (ΔLD) independent differentiated samples from 3 rounds of differentiation. *p ≤ 0.05, unpaired t test.

Figure 6.. The BP-lowering rs1173771 haplotype increases NPR3 expression in iECs and iVSMCs

(A) Locations of various PCR amplicons relative to the rs1173771 haplotype region. (B and C) Confirmation of rs1173771 haplotype region re-insertion in Y4 (B) and 39b (C). The haplotype region insertion and the genotypes of all 11 SNPs in the haplotype were further confirmed by whole-genome sequencing. (D–G) Expression of neighboring genes NPR3, SUB1, and TARS in endothelial cells (iECs) and vascular smooth muscle cells (iVSMCs) differentiated from iPSCs containing BP-lowering and -elevating haplotypes in Y4 (D and E), and 39b (F and G). Data were expressed as mean ± SEM. n = 9–25 independently differentiated samples from 3–5 rounds of differentiation. *p ≤ 0.05, unpaired t test.

Figure 7.. Effect of the single SNP rs1173771 on the expression of neighboring genes

(A) Locations of various PCR amplicons relative to the rs1173771 single SNP region. (B) gRNA locations. (C–E) Confirmation of rs1173771 single SNP region deletion and reconstitution of homozygous BP-elevating and -lowering alleles in 39b cells. (F and G) Expression of neighboring genes NPR3, SUB1, and TARS in endothelial cells (iECs) (F) and vascular smooth muscle cells (iVSMCs) (G) differentiated from iPSCs containing homozygous BP-lowering and -elevating alleles at the single SNP rs1173771. Data were expressed as mean ± SEM. n = 19–20 (iECs) and 26 (iVSMCs) independently differentiated samples from 5 rounds of differentiation. *p ≤ 0.05, unpaired t test.

Figure 8.. The BP-elevating rs1173771 haplotype interacts more frequently with the NPR3 promoter region than the BP-lowering rs1173771 haplotype

(A–D) Chromatin contact maps at rs1173771 and the surrounding region, based on a Region Capture Micro-C analysis, are shown for iECs with BP-lowering rs1173771 haplotype (A), iECs with BP-elevating rs1173771 haplotype (B), iVSMCs with BP-lowering rs1173771 haplotype (C), and iECs with BP-elevating rs1173771 haplotype (D). The zoomed-in images for each image show chromatin contacts between the rs1173771 haplotype region with promoters of SUB1, NPR3, and TARS. Note the greater contacts for the NPR3 promoter shown in (B) vs. (A) and (D) vs. (C). (E–G) Significantly different, normalized chromatin contact frequencies are shown for regions overlapping the rs1173771 haplotype with the NPR3 promoter region in iECs (E; adjusted p [p.adj] = 0.009) and a region upstream of the NPR3 promoter in iECs (F; p.adj = 0.001) and iVSMCs (G; p.adj = 0.02). The sequence coordinates for these regions are available in Table S4. The statistical significance was based on HiCcompare analysis with p.adj < 0.05.