Parathyroid hormone activates TRPV5 via PKA-dependent phosphorylation

Abstract

Low extracellular calcium (Ca(2+)) promotes release of parathyroid hormone (PTH), which acts on multiple organs to maintain overall Ca(2+) balance. In the distal part of the nephron, PTH stimulates active Ca(2+) reabsorption via the adenylyl cyclase-cAMP-protein kinase A (PKA) pathway, but the molecular target of this pathway is unknown. The transient receptor potential vanilloid 5 (TRPV5) channel constitutes the luminal gate for Ca(2+) entry in the distal convoluted tubule and has several putative PKA phosphorylation sites. Here, we investigated the effect of PTH-induced cAMP signaling on TRPV5 activity. Using fluorescence resonance energy transfer, we studied cAMP and Ca(2+) dynamics during PTH stimulation of HEK293 cells that coexpressed the PTH receptor and TRPV5. PTH increased cAMP levels, followed by a rise in TRPV5-mediated Ca(2+) influx. PTH (1 to 31) and forskolin, which activate the cAMP pathway, mimicked the stimulation of TRPV5 activity. Remarkably, TRPV5 activation was limited to conditions of strong intracellular Ca(2+) buffering. Cell surface biotinylation studies demonstrated that forskolin did not affect TRPV5 expression on the cell surface, suggesting that it alters the single-channel activity of a fixed number of TRPV5 channels. Application of the PKA catalytic subunit, which phosphorylated TRPV5, directly increased TRPV5 channel open probability. Alanine substitution of threonine-709 abolished both in vitro phosphorylation and PTH-mediated stimulation of TRPV5. In summary, PTH activates the cAMP-PKA signaling cascade, which rapidly phosphorylates threonine-709 of TRPV5, increasing the channel's open probability and promoting Ca(2+) reabsorption in the distal nephron.

Figure 1.

PTH rapidly induces a sustained stimulation of TRPV5-mediated Ca²⁺ influx. (A) ⁴⁵Ca²⁺ influx in HEK293 cells transiently coexpressing PTH1R and TRPV5, pretreated for 30 min with 25 μM BAPTA-AM. All procedures were performed at 37°C, and PTH or DMSO solvent (CTR) was administered simultaneously with ⁴⁵Ca²⁺. *P < 0.05 versus CTR. TRPV5-specific ⁴⁵Ca²⁺ influx was determined by addition of 10 μM ruthenium red (RR). (B) Representative trace of intracellular Ca²⁺ upon PTH stimulation using yellow cameleon 2.1 as a FRET sensor. (C) Basal calibrated Ca²⁺ levels in HEK293 cells transfected with mock (n = 8) and TRPV5 (n = 9). *P < 0.05 versus mock. (D) Average trace of intracellular Ca²⁺ levels of TRPV5 upon treatment with PTH (n = 9). (Inset) Relative FRET Ca²⁺ increase from basal to PTH-stimulated conditions, as shown by time periods “A” and “B,” respectively. *Significant difference (P < 0.05) from basal conditions.

Figure 2.

Dynamic effects of PTH and its fragments on PIP₂, cAMP, and Ca²⁺ levels. (A) ⁴⁵Ca²⁺ uptake of HEK293 cells transiently coexpressing PTH1R and TRPV5-PKC6 was performed as previously performed with TRPV5-WT. *P < 0.05 versus CTR. (B) PIP₂ response upon PTH stimulation in cells coexpressing PIP₂ FRET sensors, PTH1R, and TRPV5-WT. (C) Consequence of PTH treatment on intracellular cAMP response. (D and E) Average cAMP trace upon stimulation of full-length PTH (n = 7; D) and PTH (1 to 31) (n = 10) and PTH (3 to 34) (n = 4; E). (F) Average FRET data representing intracellular Ca²⁺ of cells expressing TRPV5-WT upon stimulation with PTH (1 to 31) (n = 9) and PTH (3 to 34) (n = 5). Ca²⁺ levels during unstimulated (as depicted by “A”) and forskolin-treated conditions (time period “B”) for both PTH fragments, as shown by inset figure. *P < 0.05 versus basal conditions.

Figure 3.

Forskolin (FSK) specifically stimulates TRPV5 activity under Ca²⁺-chelating conditions. (A) ⁴⁵Ca²⁺ uptake of HEK293 cells transiently expressing TRPV5 and pretreated for 30 min with 25 μM BAPTA-AM. FSK was administered simultaneously with ⁴⁵Ca²⁺. (B) Potential role of PKA was investigated by 30 min of pretreatment of 10 μM H-89. (C) FSK effect on ⁴⁵Ca²⁺ uptake of HEK293 cells transiently expressing TRPV6, pretreated with 25 μM BAPTA-AM and DMSO as a control. (D) Both pretreatment of 25 μM EDTA-AM and coexpression of calbindin-D_28K (CaBP_28K) enabled an FSK-induced stimulation of TRPV5 function. (E) Amphotericin B perforated whole-cell patch-clamp was used to study Na⁺ currents of cells expressing TRPV5-WT. At t = 2 min, cells were treated with 10 μM FSK. For investigation of the effect of intracellular Ca²⁺ on TRPV5 activity by FSK, cells were treated with 100 μM BAPTA-AM. (Inset) Current increase of CTR (n = 8), FSK (n = 10), and FSK + BAPTA-AM (n = 11) after 14 min compared with the current measured at t = 2 min. *P < 0.05 versus CTR.

Figure 4.

Threonine 709 is necessary for forskolin-induced TRPV5 stimulation. (A) Localization of putative PKA phosphorylation sites within several TRPV5 species. (B) Putative phosphorylation residues S142, S680, and T709 of TRPV5 were individually mutated into alanine. ⁴⁵Ca²⁺ uptake of HEK293 cells expressing TRPV5-WT, TRPV5 point mutants, or TRPV5-PKC6 was conducted in the presence of 25 μM BAPTA-AM. *P < 0.05 versus CTR. (C) Protein expression of TRPV5-WT and mutants. (D) Average calibrated FRET Ca²⁺ levels of TRPV5-WT (black trace, n = 9) and TRPV5-T709A (gray trace, n = 5), both stimulated with forskolin. (Inset) Comparison of basal Ca²⁺ levels (time period “A”) and Ca²⁺ levels after forskolin stimulation (“B”) of TRPV5-WT and TRPV5-T709A. *P < 0.05 versus basal conditions. (E) In vitro phosphorylation of TRPV5 N- and C-tail fused to GST. (Top) Protein input via Coomassie staining. (Bottom) ³²P radiation.

Figure 5.

Forskolin does not affect TRPV5 cell surface abundance. Biotinylation studies were performed with HEK293 cells expressing TRPV5 pretreated with BAPTA-AM to chelate intracellular Ca²⁺. (A and B) Forskolin had no effect on protein input (A) or plasma membrane expression (B) of TRPV5-WT or TRPV5-T709A.

Figure 6.

Forskolin stimulates TRPV5 channel activity. (A) Cell-attached single-channel recordings were made from HEK293 cells expressing TRPV5-WT. Channel activity was elicited by step potentials varying from −100 to 80 mV (left). The averaged calculated slope conductance was 90 pS (top right). Amplitude histograms were constructed from regions of the single-channel recordings obtained at −80 mV (bottom right). The histograms were fit by three Gaussian functions corresponding to a closed, one open, and two open levels. (B) Representative recording of TRPV5-WT in time using cell-attached patch-clamp (holding potential −80 mV). NPo values were the average of 5-s intervals. (C and D) Representative channel openings of TRPV5-WT (C) and TRPV5-T709A (D) stimulated with forskolin. (E and F) Statistical analysis demonstrates a significant elevation for TRPV5-WT (n = 8; E), whereas TRPV5-T709A channel activity (n = 8; F) was not affected. *P < 0.05 versus CTR.

Figure 7.

Isolated PKA catalytic subunit directly activates TRPV5-WT and not TRPV5-T709A. (A) The plasma membrane was excised when channel activity was detected in cell-attached configuration. Channel activity was elicited by step potentials varying from −100 to 80 mV (left). Slope conductances of TRPV5 channels were obtained from single-channel current-voltage (I-V) relationship (top right). Amplitude histogram of single-channel recordings at −80 mV is shown (bottom right). The histograms were fit by two Gaussian functions corresponding to a closed and one open level. (B and C) Representative current traces from inside-out patch of TRPV5-WT (B) and TRPV5-T709A (C) in response to −80 mV holding potential. Inside-out patch was performed in the presence of 1 mM ATP and isolated PKA catalytic subunit. (D and E) Relative average NPo of TRPV5-WT (n = 7; D) was increased after PKA catalytic subunit stimulation, whereas relative average NPo of TRPV5-T709A (n = 5; E) was not significantly altered. *P < 0.05 versus CTR.