Evidence for a role of prolactin in calcium homeostasis: regulation of intestinal transient receptor potential vanilloid type 6, intestinal calcium absorption, and the 25-hydroxyvitamin D(3) 1alpha hydroxylase gene by prolactin

Abstract

Increased calcium transport has been observed in vitamin D-deficient pregnant and lactating rats, indicating that another factor besides 1,25-Dihydroxyvitamin D(3) (1,25(OH)(2)D(3)) is involved in intestinal calcium transport. To investigate prolactin as a hormone involved in calcium homeostasis, vitamin D-deficient male mice were injected with 1,25(OH)(2)D(3), prolactin, or prolactin + 1,25(OH)(2)D(3). Prolactin alone (1 microg/g body weight 48, 24, and 4 h before termination) significantly induced duodenal transient receptor potential vanilloid type 6 (TRPV6) mRNA (4-fold) but caused no change in calbindin-D(9k). Combined treatment with 1,25(OH)(2)D(3) and prolactin resulted in an enhancement of the 1,25(OH)(2)D(3) induction of duodenal TRPV6 mRNA, calbindin-D(9k) mRNA, and an induction of duodenal calcium transport [P < 0.05 compared with 1,25(OH)(2)D(3) alone]. Because lactation is associated with an increase in circulating 1,25(OH)(2)D(3), experiments were done to determine whether prolactin also has a direct effect on induction of 25-hydroxyvitamin D(3) 1alpha hydroxylase [1alpha(OH)ase]. Using AOK B-50 cells cotransfected with the prolactin receptor and the mouse 1alpha(OH)ase promoter -1651/+22 cooperative effects between prolactin and signal transducer and activator of transcription 5 were observed in the regulation of 1alpha(OH)ase. In addition, in prolactin receptor transfected AOK B-50 cells, prolactin treatment (400 ng/ml) and signal transducer and activator of transcription 5 significantly induced 1alpha(OH)ase protein as determined by Western blot analysis. Thus, prolactin, by multiple mechanisms, including regulation of vitamin D metabolism, induction of TRPV6 mRNA, and cooperation with 1,25(OH)(2)D(3) in induction of intestinal calcium transport genes and intestinal calcium transport, can act as an important modulator of vitamin D-regulated calcium homeostasis.

Figure 1

The effect of repeated administration of prolactin or 1,25(OH)₂D₃ on duodenal TRPV6 and calbindin-D_9k mRNA levels. Sixty-day-old mice were made 1,25(OH)₂D₃ deplete by feeding 0.8% strontium diet for 7 d. Mice were injected with prolactin [1 μg/g bw, vehicle (control), or 1,25(OH)₂D₃ (1 ng/g bw as a positive control)] three times over 48 h (48, 24, and 4 h before termination). Serum calcium levels: control −D mice, 6.3 ± 0.04; prolactin-treated mice, 7.1 ± 0.1; and 1,25(OH)₂D₃-treated mice, 8.2 ± 1.5 mg/dl [P < 0.05 prolactin and 1,25(OH)₂D₃-treated mice vs. control −D]. Left panel, RT-PCR analysis of TRPV6 mRNA in duodenum. Right panel, Summary of densitometric scans of Northern blot analyses of calbindin-D_9k mRNA in duodenum. Data were corrected for GAPDH or β-actin mRNA expression and are presented relative to control (vitamin D-depleted mice). Bottom right, Western blot analysis of calbindin-D_9k protein C, Control; P, prolactin; 1,25D₃, 1,25(OH)₂D₃. Bars represent the mean ± sem, n = 6–8/group. *, P < 0.05 compared with control.

Figure 2

Prolactin (P) has cooperative effects with 1,25(OH)₂D₃ (D) in regulating duodenal calcium transport genes. RT-PCR analysis of TRPV6 mRNA expression (left panel) or Northern blot analysis of calbindin-D_9k mRNA (right panel) in the duodenum of vitamin D-deficient male mice injected with either vehicle (C, control), 1,25(OH)₂D₃ (2 ng/g bw D12 h), or prolactin (1 μg/g bw P4 h) at 12 or 4 h before termination, respectively, or 1,25(OH)₂D₃ + prolactin (D12, P4 h) at 12 and 4 h before termination, respectively. Bottom right, Western blot analysis of calbindin-D_9k protein. Bars represent the mean ± sem, n = 4–6/group. *, P < 0.05 compared with control; +, P < 0.05 compared with D12 h. P4 h vs. C. P > 0.5; a single injection of prolactin and termination at 2, 6, or 12 h after prolactin injection also did not result in a significant effect on TRPV6 mRNA or calbindin-D_9k mRNA (data not shown).

Figure 3

Active intestinal calcium transport in the duodenum of vitamin D-deficient mice. Calcium transport was measured using everted intestinal sacs from the duodenum of vitamin D-deficient male mice injected with vehicle (-D) or 1,25(OH)₂D₃ (D; 1 ng/g bw) three times over 48 h (48, 24, and 6 h before termination) as a positive control. Mice were also injected with 1,25(OH)₂D₃ (2 ng/g bw D12 h) or prolactin (P; 1 μg/g bw P6 h) at 12 or 6 h before termination, respectively, or with a combination of both 1,25(OH)₂D₃ and prolactin (D12P6 h) at 12 and 6 h before termination, respectively. Data are expressed relative to duodenal calcium transport in the deficient mice injected with vehicle (-D). Values represent the mean ± sem, n = 12–16/ group. *, P < 0.05 compared with -D, vehicle-treated mice.

Figure 4

Lack of transcriptional activation of the −7 kb/+160 TRPV6 promoter construct by prolactin. Caco-2 cells were transfected with 50 ng prolactin receptor and 0.3 μg hTRPV6 promoter (−7 kb/+160) in the presence or absence of STAT5. Empty vectors were used to keep the total DNA concentration the same. After 24 h, cells were treated with vehicle, prolactin, 1,25(OH)₂D₃ (10⁻⁹ or 10⁻⁸ m), or prolactin + 1,25(OH)₂D₃ for 24 h. TRPV6 promoter activity was normalized to values for Renilla luciferase activity as an internal control and is expressed as fold induction by comparison with basal levels; mean ± sem, n = 6–8 separate experiments. Prolactin alone (400–800 ng/ml) or in the presence of STAT5a (400 ng/ml prolactin + 50 ng STAT5a) did not affect TRPV6 transcriptional activation above basal levels (P > 0.5). In addition, prolactin did not enhance 1,25(OH)₂D₃-induced TRPV6 transcriptional activation [1,25(OH)₂D₃ (10⁻⁹ or 10⁻⁸ m) + prolactin (400, 600, or 800 ng/ml) vs. 1,25(OH)₂D₃ alone (10⁻⁹ or 10⁻⁸ m); P > 0.5]. Time-course studies (4–24 h) also did not indicate an effect of prolactin at earlier times in the presence or absence of 1,25(OH)₂D₃ on TRPV6 transcription (data not shown). When human prolactin, which is also an agonist for the long form of prolactin receptor, was substituted for ovine prolactin (used in these studies), similar results were observed.

Figure 5

Induction of 1α(OH)ase transcription by prolactin/STAT5. AOK B-50 cells were cotransfected with the prolactin receptor and the mouse 1α(OH)ase promoter construct (−1651/+22) in the presence or absence of STAT5a. After 24 h, cells were treated with vehicle or prolactin (400 ng/ml) for 24 h. For all transcription experiments, empty vectors were used to keep the total DNA concentration the same, Renilla luciferase was used as an internal control, and 1α(OH)ase promoter activity is represented as fold induction by comparison with basal levels. A, Prolactin alone (PRL) or STAT5 alone (50 ng) had a minimal effect on 1α(OH)ase transcription (1.5- to 2-fold increase in luciferase activity). However, a combination of prolactin and STAT5a resulted in a maximum induction of transcription of 13.2 ± 1.0 fold (P < 0.05 compared with prolactin alone or STAT5 alone at all concentrations of STAT5). Cotransfection with DN STAT5 (25 ng) resulted in a significant decrease in prolactin/STAT5 induction of 1α(OH)ase transcription (*, P < 0.05 PRL + STAT5a + DN STAT5 vs. PRL + STAT5a). There was no effect of DN STAT5 at the concentration used on basal levels of 1α(OH)ase transcription (data not shown). Prolactin treatment (400 ng/ml) in combination with STAT1 (PRL + STAT1) did not affect 1α(OH)ase transcription above levels observed with STAT1 alone [PRL + STAT1 (50 ng) vs. STAT1 (50 ng); P > 0.5)]. Results represent the mean ± sem, n = 4–10 observations/group. Similar results were observed using mouse proximal tubule (MPCT) cells (data not shown). B, The JAK2 inhibitor AG490 inhibits the prolactin/STAT5 induction of 1α(OH)ase transcription. AOK B-50 cells were cotransfected with the prolactin receptor and the mouse 1α(OH)ase promoter as described in A. Cells were also transfected with STAT5a expression vector (50 ng). AOK B-50 cells were pretreated for 1 h with JAK2 inhibitor AG490 (200 μm) followed by treatment with prolactin (PRL; 400 ng/ml) for 4 h. Results are expressed as mean ± sem of at least three observations/group. The JAK2 inhibitor AG490 significantly suppresses prolactin/STAT5 induction of 1α(OH)ase transcription (P < 0.05 PRL + STAT5a + AG490 vs. PRL + STAT5a).

Figure 6

Regulatory region for prolactin/STAT5 stimulation of 1α(OH)ase transcription and induction of 1α(OH)ase protein by prolactin. A, AOK B-50 cells were transfected with 0.3 μg mouse 1α(OH)ase promoter −1651/+22 or deletion constructs −477/+22, −233/+22, −144/+22, and −85/+22. In some wells, STAT5a expression vector (50 ng) was also transfected. After 24 h, cells were treated with vehicle or prolactin (400 ng/ml) for 24 h. Results are the mean ± sem of six to eight observations/group. B, Western blot analysis was performed using total extracts from AOK B-50 cells. Cells were transfected with STAT5a and prolactin receptor and treated with vehicle (0) or prolactin (400 ng/ml) for 3–48 h. Prolactin treatment resulted in a significant increase in 1α(OH)ase protein levels at 3, 6, 12, and 24 h after prolactin treatment (P < 0.05 compared with vehicle).