Fig 1.
Multivariate visualisation of correlations between serum vitamin D metabolites.
Serum 25OHD3 (ng/ml), 3-Epi-25OHD3 (ng/ml), 24,25(OH)2D3 (ng/ml) and 1 α,25(OH)2D3 (pg/ml) represented as a scatterplot matrix (black circles female, red triangles male), with robust linear trendlines for bivariate regressions. The curve plots on the main diagonal are univariate histograms; the off-diagonal panels are bivariate scatterplots between variables labelled at the plot edges (e.g. the top row, second-from-right panel is 24,25(OH)2D3 versus 25OHD3).
Table 1.
Subject characteristics for overall group.
Data are expressed as median (IQR). SBP = systolic blood pressure, DBP = diastolic blood pressure, BMI = body mass index.
Fig 2.
Serum vitamin D metabolites and body composition parameters in women.
Serum (a) 25OHD3, (b) 24,25(OH)2D3 correlated negatively with body fat (rho = -0.30, p = 0.02 and rho = -0.33, p = 0.01, respectively), but not (c) 1α,25(OH)2D3 (rho = -0.06, p = 0.79). Conversely, lean mass correlated positively with (f) 1α,25(OH)2D3 (rho = 0.47, p = 0.02), but not (d) 25OHD3 or (e) 24,25(OH)2D3 (rho = 0.03, p = 0.81, and rho = 0.03, p = 0.80). Data were analysed by Spearman correlations (rho) with p values and line of best fit shown.
Fig 3.
Serum vitamin D metabolites and urinary steroid metabolism in women.
Serum 25OHD3 and 24,25(OH)2D3 correlated negatively with urinary (tetrahydrocortisol+5α tetrahydrocortisol)/ tetrahydrocortisone ((THF+5αTHF)/THE ratios, and positively with urinary cortisol/cortisone (F/E) ratios. Data were analysed by Spearman correlations (rho) with p values and line of best fit shown.
Fig 4.
Serum active vitamin D and muscle strength.
1α,25(OH)2D3 correlated positively with jump plate measures of lower limb strength (a) Pmax (maximal power), (b) Vmax (maximum velocity), (c) jump height, all on standing 2-legged jump (S2LJ). Data were analysed by Spearman correlations (rho) with line of best fit shown.
Table 2.
Bivariate correlations between serum vitamin D metabolites and subject characteristics.
Fig 5.
Expression of VDR mRNA in human muscle biopsies.
Relationship between muscle expression of VDR (RT-PCR ΔCt value) and: age; serum 25OHD3 concentrations (ng/ml); fat mass (Total mass, kg); Vmax (m/s). p values for linear regression analyses are shown and significant correlations are shown as solid lines. Data were analysed by Spearman correlations (rho) with line of best fit shown.
Table 3.
Bivariate correlations between expression of Vitamin D Receptor (VDR) and other skeletal muscle target genes.
Data are Spearman correlation coefficients (rho)), with 95% confidence intervals (CI) and p values in brackets).
Table 4.
Bivariate correlations between serum 25OHD3, 1α,25(OH)2D3 and gene expression in skeletal muscle.
Data are Spearman correlation coefficients (rho)), with 95% confidence intervals (CI) and p values in brackets).