The extracellular calcium-sensing receptor regulates human fetal lung development via CFTR

Abstract

Optimal fetal lung growth requires anion-driven fluid secretion into the lumen of the developing organ. The fetus is hypercalcemic compared to the mother and here we show that in the developing human lung this hypercalcaemia acts on the extracellular calcium-sensing receptor, CaSR, to promote fluid-driven lung expansion through activation of the cystic fibrosis transmembrane conductance regulator, CFTR. Several chloride channels including TMEM16, bestrophin, CFTR, CLCN2 and CLCA1, are also expressed in the developing human fetal lung at gestational stages when CaSR expression is maximal. Measurements of Cl(-)-driven fluid secretion in organ explant cultures show that pharmacological CaSR activation by calcimimetics stimulates lung fluid secretion through CFTR, an effect which in humans, but not mice, was also mimicked by fetal hypercalcemic conditions, demonstrating that the physiological relevance of such a mechanism appears to be species-specific. Calcimimetics promote CFTR opening by activating adenylate cyclase and we show that Ca(2+)-stimulated type I adenylate cyclase is expressed in the developing human lung. Together, these observations suggest that physiological fetal hypercalcemia, acting on the CaSR, promotes human fetal lung development via cAMP-dependent opening of CFTR. Disturbances in this process would be expected to permanently impact lung structure and might predispose to certain postnatal respiratory diseases.

Figure 1. Expression of chloride channels in the developing mouse lungs and human fetal lungs.

Paraffin-embedded, 5 μm-thick sections from E12.5 mouse (left panel) and week 9–11 human fetal lungs (right panel) were dewaxed and used for immunohistochemistry. Expression of the Ca²⁺-activated chloride channels TMEM16A, bestrophin-1 and CFTR were visualised using DAB (brown straining) in the lung epithelium. Sections were counterstained with Harris’ hematoxylin (blue staining). Negative controls were carried out in serial sections form the same lungs by substituting the primary antibody with an isotype control (inset). Block arrows show apical expression in the epithelium, arrowheads show basolateral expression and open arrows show expression in the mesenchyme. Scale bar = 100 μm.

Figure 2. CaSR activation drives fluid secretion in fetal mouse and human lungs.

(a) E12.5 lungs were cultured for 48 hours in the presence of medium containing either 1.05 mM, 1.05 mM Ca²⁺_o + the calcimimetic NPS-R568 or 1.05 mM Ca²⁺_o + NPS-R568 + the CFTR blocker, Inh-172, before measurement of transluminal potential differences were carried out. Pharmacological CaSR activation almost doubled the PD from 100 ± 13% to 184 ± 19%, effect which was completely abolished by Inh-172. Data were pooled from 3–4 separate isolations, n = 9, for all conditions and are presented as mean (as a percentage of 1.05 mM Ca²⁺_o control) ± SEM. **p < 0.01, one-way ANOVA with Tukey post-test. (b) Human lung rudiments were separated into two halves and then kept in culture for 72 h in medium containing either 1.05 mM Ca²⁺_o in the presence or absence of 100 nM NPS-R58 (n = 4), or 1.05 mM Ca²⁺_o + 100 nM NPS-R568 (n = 4) in the presence or absence of 10 μM Inh-172. Culturing human fetal lungs rudiments in the presence of 1.05 mM Ca²⁺_o + 100 nM NPS-R568 induced an increase in transluminal potential difference compared to its paired lung half cultured in 1.05 mM Ca²⁺_o alone. Furthermore, culturing fetal lung halves in the presence of medium containing 1.05 mM Ca²⁺_o + 100 nM NPS-R568 + 10 μM Inh-172 decreased transluminal potential difference in comparison to its paired 1.05 mM Ca²⁺_o + 100 nM NPS-R568 lung half. Data are presented as mean difference from 1.05 mM Ca²⁺_o + 100 nM R568 ± SEM. **p < 0.01, paired t-test. (c) CaSR and CFTR are co-localised in the human fetal lung epithelium. Week 10 gestation human fetal lungs were obtained from maternal donors. Sections from ethically consented week 10 human fetal lungs were incubated with anti-CFTR (1:200) or anti-CaSR antibodies (1:200) and immunoreactivities were detected using Alexa Fluor 594 goat anti-rabbit secondary antibodies (1:200). Staining indicates that both CFTR and CaSR are present in the columnar and cuboidal epithelium cells of the primitive airways of human fetal lung.

Figure 3. Fetal hypercalcaemia drives fluid secretion via activation of CFTR in the fetal human, but not mouse lung.

(a) Inh-172 is a specific inhibitor for the CFTR channel. E12.5 lungs were cultured for 48 hours in the presence of medium containing 1.70 mM Ca²⁺_o with or without Inh-172, before measurements of transepithelial potential differences were carried out. Culturing E12.5 mouse lungs in the presence of Inh-172 did not significantly alter transepithelial potential differences in lungs cultured in medium containing 1.70 mM Ca²⁺_o. (b) Human lung rudiments were separated into two halves and then kept in culture for 72 h in medium containing either 1.05 mM or 1.70 mM Ca²⁺_o in the presence or absence of the CFTR inhibitor, Inh-172. Culturing human lung rudiments in medium containing 1.70 mM Ca²⁺_o induced an increase in transluminal potential difference that was inhibited by Inh-172. Data are pooled from 4–6 different lungs for all conditions. Data are presented as mean difference from 1.70 mM Ca²⁺_o ± SEM. **p < 0.01, ***p < 0.001, paired t-test.

Figure 4. CaSR activation leads to opening of CFTR: involvement of a calcium-activated adenylate cyclase.

(a) Fischer Rat Thyroid FRT cells, which endogenously express the CaSR, were engineered to express the CFTR channel and a halide-sensitive YFP as an indicator of channel opening. FRT cells were fixed with 4% paraformaldehyde before undergoing CaSR immunostaining. Primary antibody binding was detected using DAB (brown straining, right panel). Sections were counterstained with Harris’ hematoxylin (blue staining). Negative controls were carried out through the substitution of the primary antibody with an isotype control (left panel). Scale bar = 1000 μm. (b) FRT cells were pre-incubated for at least 2 min with NPS-R568 (1 μM) in the presence or absence of either the CFTR specific inhibitor, Inh-172 (10 μM) or the adenylate cyclase inhibitor, MDL-12330A (25 μM). 20 μM forskolin was used as a positive control, and pre-incubation with Dulbecco’s PBS was used as the time control. After 2 min the cell are perfused with iodide-rich Dulbecco’s PBS and the fluorescence of the halide-sensitive YFP is quenched at a rate dependent upon the halide permeability of the cell and therefore the activity of anion channels or transporters. The calcimimetic R568 quenched YFP fluorescence, demonstrating that CaSR activation in FRT cells leads to increased halide permeability of the cell. Inhibition of adenylate cyclase by MDL-12330A or CFTR by Inh-172 brought YFP quenching to levels similar to that of the time control – demonstrating that the increased halide permeability was due to activation of both adenylate cyclase and CFTR. (b) shows average traces of YFP quenching (±SD) over time from N = 6–13 separate experiments (n = 48–232 cells).

Figure 5. Proposed model for CaSR-mediated increase in fetal lung fluid secretion.

In both the mouse and human activation of the CaSR via the calcimimetic NPS R-568 leads to an increase in cytosolic Ca²⁺ through activation of the PI-PLC pathway and release of Ca²⁺ from intracellular stores. This rise in cytosolic calcium leads to an increase in intracellular cAMP level via a Ca²⁺-stimulated adenylate cyclase (AC1), and in turn activation of the cAMP-dependent enzyme protein kinase A (PKA). Phosphorylation of the CFTR’s regulatory ‘R’-domain by PKA allows opening of the channel and conductance of Cl⁻ ions through the channel. A similar pathway also appears to be in place in response to fetal hypercalcemia in the human fetal lung, however this is not the case in the mouse where as yet unknown pathway, not involving an apical CFTR channel, appears to induce this increase in fluid secretion.