Figure 4. Serum RBP and Vitamin A Analysis.

(A) ECL western blot (reducing conditions) showing dog RBP monomers (21 kDa) with human ^HARBP (22 kDa) as a positive control (HeLa-CM). The higher MW signal is dog IgG (150 kDa), which cross-reacts with the secondary reagent (donkey anti-rabbit IgG) and is relatively resistant to DTT (Singh and Whitesides, 1994).

(B) Parallel blot (non-reducing conditions) showing RBP monomers (WT) and homodimers (K12del mutant). Both species are present in heterozygotes, in roughly a 4:1 molar ratio.

(C) Histograms comparing immunoreactive serum RBP and vitamin A (mean ± SD) versus genotype. There is a linear dosage relationship, with heterozygotes having an intermediate level of RBP (0.66 ± 0.20) that is close to the arithmetic mean of del/del and +/+ samples (0.62 ± 0.22).

(D) Scatterplot comparing serum vitamin A (μg/mL, ordinate) and relative RBP (abscissa) for all samples, including WT (black), HET (gray), and DEL (white) dogs with normal eyes (circles) and microphthalmic DEL dogs (white squares). The values are well correlated (p < 0.0001, r² = 0.88); however, the regression line is shifted rightward from the origin, indicating that the K12del protein binds less retinol in vivo than WT. The homozygous DEL samples have similar RBP and vitamin A levels, regardless of phenotype (one-way ANOVA, p = 0.18 for RBP and p = 0.45 for vitamin A, comparing 5 normal and 3 microphthalmic dogs).

See also Tables S1 and S3.