Integrated genomic approaches implicate osteoglycin (Ogn) in the regulation of left ventricular mass

Abstract

Left ventricular mass (LVM) and cardiac gene expression are complex traits regulated by factors both intrinsic and extrinsic to the heart. To dissect the major determinants of LVM, we combined expression quantitative trait locus1 and quantitative trait transcript (QTT) analyses of the cardiac transcriptome in the rat. Using these methods and in vitro functional assays, we identified osteoglycin (Ogn) as a major candidate regulator of rat LVM, with increased Ogn protein expression associated with elevated LVM. We also applied genome-wide QTT analysis to the human heart and observed that, out of 22,000 transcripts, OGN transcript abundance had the highest correlation with LVM. We further confirmed a role for Ogn in the in vivo regulation of LVM in Ogn knockout mice. Taken together, these data implicate Ogn as a key regulator of LVM in rats, mice and humans, and suggest that Ogn modifies the hypertrophic response to extrinsic factors such as hypertension and aortic stenosis.

Figure 1

Quantitative trait transcripts analysis of cis eQTLs with physiological traits in the rat. Expression profiles of the 1,444 cis eQTLs (P_GW = 0.05) were correlated with values of (a) indexed LVM, (b) DBP and (c) SBP measured in the BXH/HXB RI strain panel. For each cis eQTL across the rat genome, the absolute Pearson correlation coefficient with the physiological traits is plotted against the location of the probe set (Mb). Vertical lines, physical position of the end of each rat chromosome. The number of each chromosome is given on the upper y axis. The horizontal lines indicate empirical significance levels P = 0.05, P = 0.01 and P = 0.001 of the correlations (see Methods).

Figure 2

Candidate cis-regulated genes within the rat QTL for left ventricular mass. (a) Interval mapping of LVM to chromosome 17p14 in the BXH/HXB RI strain panel. Horizontal dashed line, genome-wide threshold for significance (P = 0.05) of linkage derived by 10,000 permutations; 2-lod support interval region, by vertical box (dotted lines). Positions of the genetic markers are also indicated along the x axis. (b) Volcano plot of significance of cis eQTLs against their relative change in expression. Circles represent 113 transcripts physically encoded within the 2-lod support interval of the QTL and showing cis linkage at their nearest markers (nominal P < 0.05, two-tailed Mann-Whitney U-test). For each transcript, the negative log₁₀ of the P value of linkage is plotted against the relative change in gene expression at the peak of linkage. Dashed horizontal line, significance threshold (P = 0.00033) after Bonferroni correction for multiple testing owing to the number of transcripts in the region that were tested for cis linkage. Transcripts representing Hbld2 (one probe set) and Ogn (two probe sets) are indicated by the arrows. (c) Sequence polymorphisms in Ogn genomic DNA from 2 kb upstream of the first exon to 3 kb downstream of the stop codon. For 5′ UTR, exonic and 3′ UTR variants, positions are reported relative to the start codon; for intronic variants, positions are given relative to the nearest exon. The sequences variation at position 348 in exon 3 of Ogn is synonymous. Ins, insertion. (d) Regulation of Ogn in a blood pressure–independent in vitro model of cardiac myocyte hypertrophy: neonatal rat ventricular myocytes stimulated with phenylephrine for the times shown and assayed by quantitative RT-PCR for changes in Ogn mRNA expression. Mean relative change (± s.e.m.) in gene expression as compared to control samples over the time course of the experiment; *P < 0.05, **P < 0.01.

Figure 3

Correlation between hemodynamic indices and indexed LVM in individuals with aortic stenosis. (a–d) We observed little correlation between LVM and peak velocity across the aortic valve (V_max) or indexed aortic valve area (AVAI), indices of aortic stenosis severity and hemodynamic pressure, in two independent cohorts of subjects with aortic stenosis characterized by echocardiography (a,b, n = 45) or cardiac MRI (c,d, n = 123), respectively. The percentage of variance of LVM accounted for by hemodynamic indices after including all significant covariates (R²) is indicated.

Figure 4

Molecular characterization of rat Ogn. (a) PCR products of Ogn 3′ UTR generated from BN and SHR cDNA, showing the presence of two splice variants. (b) Abundance of the 3′ UTR isoforms and total coding mRNA in the parental strains. Each data point represents the relative expression in one BN (circle) or one SHR (triangle) rat of total coding, short and long isoforms of Ogn. *P < 0.01, **P < 0.001. (c) Abundance of the 3′ UTR isoforms and total coding mRNA in the RI strains. Each data point represents relative Ogn expression in one rat from each of the RI strains carrying either the BN (circle) or SHR (triangle) allele; **P < 10⁻¹¹. (d) Effects of the BN and SHR 3′ UTR isoforms on protein synthesis as determined by luciferase assay. Isoforms: BN long, BN-L; BN short, BN-S; SHR long SHR-L; SHR short, SHR-S. **P < 0.0001. (e) Immunoblot of Ogn protein expression in 20 μg total protein from three BN and three SHR hearts. Immature pre-Ogn, ~50 kDa, mature Ogn protein, ~20 kDa. (f) Semiquantitative densitometry of data shown in e. **P < 0.0001. (g) Immunofluorescence confocal micrographs of isolated adult rat ventricular cardiac myocytes. Top left, sarcomeric proteins labeled with rhodamine-conjugated phalloidin; top right, DAPI counterstain of nuclei; bottom left, Ogn protein detected using antibodies to Ogn and a secondary antibody labeled with Alexa Fluor 488; bottom right, merged image. Mean ± s.e.m. (d,f); NS, not significant.

Figure 5

In vivo regulation of LVM in Ogn knockout mice. (a) Indexed LVM in mice at baseline and after hypertrophic stimulation by angiotensin II infusion over a 2-week period (n = 14–18 for Ogn^+/+, n = 6–10 for Ogn^+/− or Ogn^−/−). (b) SBP and DBP in mice at baseline and after angiotensin II infusion (n = 6 for Ogn^+/+, n = 3–5 for Ogn^+/− or Ogn^−/−). All results are given as mean ± s.e.m. P values were generated by one-way analysis of variance with the Dunnett's post hoc test for multiple comparisons using the wild-type mice (Ogn^+/+) as the reference group. *P < 0.05; **P = 0.002; NS, not significant.