Hemoglobin alters vitamin carrier uptake and vitamin D metabolism in proximal tubule cells: implications for sickle cell disease

Abstract

Kidney disease, including proximal tubule (PT) dysfunction, and vitamin D deficiency are among the most prevalent complications in sickle cell disease (SCD) patients. Although these two comorbidities have never been linked in SCD, the PT is the primary site for activation of vitamin D. Precursor 25-hydroxyvitamin D [25(OH)D] bound to vitamin D-binding protein (DBP) is taken up by PT cells via megalin/cubilin receptors, hydroxylated to the active 1,25-dihydroxyvitamin D [1,25(OH)₂D] form, and released into the bloodstream. We tested the hypothesis that cell-free hemoglobin (Hb) filtered into the PT lumen impairs vitamin D uptake and metabolism. Hb at concentrations expected to be chronically present in the ultrafiltrate of SCD patients competed directly with DBP for apical uptake by PT cells. By contrast, uptake of retinol binding protein was impaired only at considerably higher Hb concentrations. Prolonged exposure to Hb led to increased oxidative stress in PT cells and to a selective increase in mRNA levels of the CYP27B1 hydroxylase, although protein levels were unchanged. Hb exposure also impaired vitamin D metabolism in PT cells, resulting in reduced ratio of 1,25(OH)₂D:25(OH)D. Moreover, plasma levels of 1,25(OH)₂D were reduced in a mouse model of SCD. Together, our data suggest that Hb released by chronic hemolysis has multiple effects on PT function that contribute to vitamin D deficiency in SCD patients.

Keywords: endocytosis; megalin; proximal tubule; sickle cell disease; vitamin D.

Fig. 1.

Hemoglobin competes with vitamin D-binding protein (DBP) for uptake by proximal tubule cells. A: filter-grown opossum kidney (OK) cells were incubated with 25 nM apically added Alexa Fluor-647-DBP in the presence of the indicated concentrations of oxyhemoglobin (oxyHb; 0–20 µM) for 1 h at 37°C and then solubilized, and cell-associated DBP was quantified by spectrofluorimetry. Data from 5 experiments done in duplicate are plotted, with each experiment represented by a different symbol. Graphed values were normalized by dividing each point by the respective experimental mean for that condition. *P ≤ 0.01, ****P ≤ 0.0001, by one-way ANCOVA (Dunnett’s multiple comparisons test). B: filter-grown OK cells incubated for 1 h at 37°C with 25 nM apically added Alexa Fluor-647 DBP (red) in the presence of 0, 0.6, or 5 µM Alexa Fluor-488 oxyHb (green) were fixed and processed for confocal imaging. Scale bar, 10 µm.

Fig. 2.

Hemoglobin weakly inhibits retinol-binding protein (RBP) uptake by proximal tubule cells. Filter-grown opossum kidney (OK) cells were incubated with 0.1 µM apically added Alexa Fluor-647 RBP in the presence of the indicated concentrations of oxyhemoglobin (oxyHb; 0–20 µM), and cell-associated RBP was quantified as in Fig. 1. Data from four experiments performed in duplicate are plotted, with each experiment represented by a different symbol. Graphed uptake values were normalized by dividing each point by the respective experimental mean for that condition. *P ≤ 0.05, by one-way ANCOVA (Dunnett’s multiple comparisons test). Expt, experiment.

Fig. 3.

Hemoglobin induces oxidative stress and selectively increases proximal tubule (PT) CYP27B1 transcript levels. Filter-grown LLC-PK1 (A) or opossum kidney (OK) (B–F) cells were incubated with concentrations of oxyhemoglobin (oxyHb) or methemoglobin (metHb) for 72 h at the concentrations indicated. Cells were then washed and lysed, and total RNA and protein were collected. HMOX1 (A), CYP24A1 (B), and CYP27B1 (C) transcripts were quantified by quantitative PCR and normalized to ACTB transcript levels. Relative fold change (RFC) of treated conditions compared with an untreated control (normalized to 1; dotted line) is plotted. Protein expression of CYP24A1 and CYP27B1 was quantified by Western blotting. D: representative Western blots for CYP24A1 and CYP27B1 with migration of molecular mass markers (in kDa) indicated. RFC of CYP24A1 (E) and CYP27B1 (F) in treated conditions compared with an untreated control (normalized to 1; dotted line) is plotted. Data from 5 experiments and the mean ± SD of the aggregate data are plotted. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, by one-sample t test. Similar statistical results were obtained when the raw data were analyzed by one-way ANCOVA (Dunnett’s multiple comparisons test).

Fig. 4.

Vitamin D metabolism in proximal tubule cells is modulated by hemoglobin exposure. Filter-grown LLCPK1 cells were incubated for 72 h with 1 µM or 20 µM concentrations of oxyhemoglobin (oxyHb) and then treated with 100 nM 25-hydroxyvitamin D [25(OH)D] for 4 h. A and B: cells were then washed and solubilized, and 1,25(OH)₂D (A) and 25(OH)D (B) concentrations were quantified as described in materials and methods. C: the ratio of 1,25(OH)₂D to 25(OH)D is plotted. Data from three independent experiments (means ± SD) are plotted. *P < 0.05, by one-way ANOVA (Dunnett’s multiple comparisons test).

Fig. 5.

Vitamin D status is altered in Townes sickle cell HbSS mice. A and B: 1,25-dihydroxyvitamin D [1,25(OH)₂D] (A) and 25(OH)D (B) levels were quantified in plasma from 8- to 20-wk-old male and female HbSS-Townes (sickle cell disease, SCD) and HbAA-Townes (control) mice. C: the ratio of 1,25(OH)₂D to 25(OH)D is plotted. *P < 0.05, **P ≤ 0.01, by unpaired t test.