Deficiency of the Calcium-Sensing Receptor in the Kidney Causes Parathyroid Hormone–Independent Hypocalciuria

Abstract

Rare loss-of-function mutations in the calcium-sensing receptor (Casr) gene lead to decreased urinary calcium excretion in the context of parathyroid hormone (PTH)–dependent hypercalcemia, but the role of Casr in the kidney is unknown. Using animals expressing Cre recombinase driven by the Six2 promoter, we generated mice that appeared grossly normal but had undetectable levels of Casr mRNA and protein in the kidney. Baseline serum calcium, phosphorus, magnesium, and PTH levels were similar to control mice. When challenged with dietary calcium supplementation, however, these mice had significantly lower urinary calcium excretion than controls (urinary calcium to creatinine, 0.31±0.03 versus 0.63±0.14; P=0.001). Western blot analysis on whole-kidney lysates suggested an approximately four-fold increase in activated Na⁺-K⁺-2Cl⁻ cotransporter (NKCC2). In addition, experimental animals exhibited significant downregulation of Claudin14, a negative regulator of paracellular cation permeability in the thick ascending limb, and small but significant upregulation of Claudin16, a positive regulator of paracellular cation permeability. Taken together, these data suggest that renal Casr regulates calcium reabsorption in the thick ascending limb, independent of any change in PTH, by increasing the lumen-positive driving force for paracellular Ca²⁺ transport.

Calcium (Ca²⁺) is a vital cation involved in diverse biologic processes ranging from bone formation and neurotransmission to hormone secretion and muscle contraction. Large changes in the concentration of extracellular Ca²⁺ can disrupt the normal cellular activities and cause these systems to function abnormally.¹ The extracellular calcium-sensing receptor (Casr) plays a critical role in regulating extracellular Ca²⁺ concentrations and cellular responses to these concentrations.² Casr is a G-protein–coupled seven-transmembrane spanning receptor that resides on the cell surface as a dimer. It is highly expressed in the parathyroid glands, where it regulates the production and secretion of parathyroid hormone (PTH) in negative feedback fashion. It is also expressed in numerous other tissues, where it has diverse but less well defined functions.³ A key role of Casr is maintaining extracellular Ca²⁺ levels by means of ability to regulate PTH biosynthesis and release and also to alter the responsiveness of other target cells to Ca²⁺. This latter function of Casr is thought to be of substantial importance for renal reabsorption of Ca²⁺ and other essential ions.⁴

Although renal Ca²⁺ excretion is modulated by PTH, there are a number of other factors that play a role, including the calciotropic hormone 1,25(OH)₂ vitamin D. Casr is highly expressed at the basolateral membrane in the cortical and medullary thick ascending limb (TAL). Casr expression has also been reported in the apical membrane of the proximal tubule (PT). Casr has been hypothesized to take on many roles in this location, including regulating 1α-hydroxylation of 25(OH) vitamin D.⁵ A role for Casr has also been suggested for the regulation of water reabsorption and proton secretion in the collecting duct (CD).⁶

Both gain-of-function and loss-of-function mutations have been reported in human Casr. Gain-of-function mutations cause autosomal-dominant hypocalcemia with hypercalciuria as well as a variant form of Bartter’s syndrome.⁷ Heterozygosity for loss-of-function mutations causes familial hypocalciuric hypercalcemia (FHH), whereas loss-of-function mutations in both alleles cause neonatal severe hyperparathyroidism (NSHPT).⁵ Population studies have suggested a possible role for Casr variants in explaining human variation in serum Ca²⁺ and bone mineral density (BMD).^8,9 Various studies have suggested a role for Casr in the regulation of renal Ca²⁺ handling. Increasing extracellular Ca²⁺ concentration elicits a marked increase in urinary Ca²⁺ excretion, independently of any obvious changes in calcium-regulating hormones. There is abundant evidence that renal tubular Casr plays a role in the control of divalent cation reabsorption under both normal and pathologic conditions.¹⁰ Clinical studies in humans under PTH clamp conditions suggested that alterations in serum Ca²⁺ modulate excretion of Ca²⁺, magnesium (Mg²⁺), and sodium (Na⁺) through a Casr-dependent mechanism.¹¹ Recent PTH clamp studies in mice showed similar results.¹²

In this study, we explored the role of Casr in the kidney. We utilized the well described Cre/Lox system to target E3 of Casr encoding for a Ca²⁺-sensing peptide sequence in the N-terminal region of the extracellular domain.^13,14 The removal of this exon alters the reading frame and leads to absence of a functional Casr protein. We then generated and analyzed a novel mouse model in which Casr inactivation was specifically restricted to renal tubular epithelial cells by the Sine oculis homeobox homolog 2 (Six2) gene promoter.^15,16

Results

Germline Casr-Deficient Mice

The wild-type Casr gene (Figure 1A), targeting construct (Figure 1B), and excised allele (Figure 1C) were examined. Transgenic mice expressing Cre recombinase in the germline using the protamine 1 promoter were utilized to verify the validity of our floxed model.¹⁷ Excision of E3 in the Casr null (floxed) allele was confirmed (Figure 1, D and E), predicting Casr deficiency. At birth, Casr^−/− had significantly elevated serum Ca²⁺ levels (13.7±0.4 mg/dl) compared with heterozygous (11.7±0.3 mg/dl) and wild-type (10.1±0.2 mg/dl) animals (P<0.001). At postnatal day 8, Casr^−/− exhibited severe growth retardation, osteomalacia, and bone fractures without other histologic abnormalities (Supplemental Figure 1, A–C). This NSHPT-like phenotype in germline-deleted Casr mice confirmed the ability of the floxed Casr allele to behave as anticipated.

Figure 1.

Wild-type Casr gene, E3 targeting construct, and genotype analysis. (A) E2–E4 of the wild-type Casr gene. (B) A Sal1 restriction site in intron 2, the neomycin resistance gene (Neo; gray box) in intron 3 and LoxP sites (green arrowheads) were introduced into the targeting construct. Locations of genotyping primers are shown. (C) Excised allele. (D) Successful amplification with Postneo PCR (primers Casr-F/Postneo-R) indicates excision of E3 and Neo and presence of a Casr null (floxed) allele. Postneo PCR from Casr flox allele does not amplify due to large amplicon size. The absence of E3 amplification (primers Casr-E3-F/Casr-E3-R) indicates complete knockout of the Casr in targeted tissues. Neo PCR amplification (primers Neo-F/Neo-R) indicates presence of the Casr flox allele. (E) Amplification with primers Casr-F/Casr-E3-R followed by SalI digest indicates presence of the Casr flox allele.

Kidney-Specific Casr-Deficient Mice Driven by Six2 Promoter

After confirming the ability of our conditional model to efficiently target Casr via Cre/Lox recombination, we created a kidney-specific Casr-deficient model utilizing the transgenic animals expressing Cre recombinase under the Six2 promoter. Six2-Cre Casr floxed mice (experimental animals) were born in the predicted Mendelian ratio and displayed no size or weight difference compared with flox/flox littermates (control animals) or wild-type mice at birth and during observation for >1 year. There was no difference in size or life expectancy. Hematoxylin and eosin histology of kidneys and parathyroid glands revealed no apparent differences (data not shown).

PCR amplifications of whole-kidney cDNA across Casr exons 2–4 (E2–E4) in null animals generated a 200 bp amplicon, shorter than the wild-type 500 bp fragment (Figure 2A). DNA sequencing of this shorter amplicon confirmed absence of E3, with E2 joined to E4. The E3-deleted transcript encodes a frameshift at Casr codon 62 (E2: 5′-TGT ATC AG gtaaga-′3; E3: 5′-tccag G TAT AAT-′3; E4: 5′-ccag GTC AGT TAT…-′3), followed by 59 residues of novel amino acid sequence and terminator codon. The predicted protein encoded by the E3-deleted transcript lacks all of the known calcium binding sites and the entire transmembrane domain of Casr. Exon-spanning PCR in heterozygous mice amplified two fragments, one of wild-type length and one of length identical to the predicted E3-deleted fragment, both confirmed by sequencing (data not shown).

Figure 2.

Casr expression in germline and conditional Casr-deficient models. (A) PCR amplification of whole-kidney cDNA generates E3-deleted amplicon of 200 bp compared with the nonexcised 500 bp amplicon. Whole-kidney cDNA from Casr^flox/- mice show both the intact and the E3-deleted amplicons. (B) QPCR from whole-kidney cDNA with Casr primers located in E2 and E3 shows no Casr expression in null mice compared with age- and sex- matched control animals (P<0.001). (C) Immunoblot analysis of whole-kidney homogenate with the ADD antibody (targeting a peptide sequence encoded by E4). No Casr protein was detected in Casr null mice and in Six2-Cre Casr floxed mice (on both +/+ and +/− backgrounds). CTL, control.

We performed quantitative PCR (QPCR) from whole-kidney cDNA using exon-spanning Casr primers located in E2 and E3. Figure 2B shows that no Casr expression was detected in Casr null mice (n=12) compared with age- and sex-matched control animals (P<0.001).

To complement our finding of absence of Casr mRNA in the kidney, we tested Casr protein expression in whole-kidney homogenates by Western blot analysis. We used monoclonal and polyclonal anti-Casr antibodies toward a 21-peptide sequence encoded by E4 known as the ADD sequence (amino acid sequences 214–235).¹⁸ Figure 2C shows absence of Casr protein in whole-kidney lysates in E3-less germline Casr-deficient mice compared with lysates from the kidneys of Casr^+/+ and Casr^+/−. The absence of reactivity to this region of Casr, C-terminal to the deleted exon, indicates that no significant Casr is expressed by this altered allele.

We then performed immunohistochemistry on paraffin-embedded kidney sections of control (Figure 3A) and experimental mice (Figure 3B) with anti-Casr antibody. These experiments confirmed absence of Casr in renal tissue of Six2-Cre experimental mice. The immunohistochemistry experiments showed peritubular staining in both groups, which is likely a staining artifact due to the fixation process. Glomeruli also showed some staining in both groups, which has been reported previously by others.¹⁹ The glomerular mass is <1% of total kidney mass, which could explain absence of Casr expression in our Western blotting, if Casr is indeed expressed in glomeruli. Our Casr-deficient model driven by the Six2 promoter did not target glomeruli. To further study the expression localization of Casr relative to the TAL-specific NKCC2 and the CD-specific Aquaporin 2, we carried a series of immunohistochemistry staining on adjacent control mouse kidney sections with antibodies against Casr (Figure 3C), NKCC2 (Figure 3D), and Aquaporin 2 (Figure 3E). These immunohistochemistry data suggest Casr expression in the TAL; no staining was observed in the CD.

Figure 3.

Casr immunohistochemistry in control and Six2-Cre Casr floxed animals. (A) Immunohistochemistry with anti-Casr antibody shows Casr expression in renal tubular epithelia of control animals. (B) Six2-Cre Casr floxed animals show absence of Casr expression. (C) Magnified image of A suggests Casr expression in tubular structures consistent with TAL epithelia. (D and E) Immunohistochemistry of adjacent sections to C showing expression patterns of NKCC2 and Aquaporin 2, respectively, suggesting NKCC2 colocalization with Casr in TAL epithelia, but not the CD epithelia-specific Aquaporin 2. CTL, control.

Phenotype of Six2-Cre Casr Floxed Mice

After confirming absence of Casr expression at both RNA and protein levels, we tested the consequences of renal-specific Casr deficiency on several indices of systemic calcium physiology in mice fed a regular chow diet containing 0.8% calcium (Figure 4, A–C) or with 1.5% calcium chloride added to the drinking water (Figure 4, D–F). Serum Ca²⁺ levels did not differ between Casr-deficient mice and control mice either at baseline (10.35±0.19 versus 10.38±0.23 mg/dl; P=1.0) or after dietary calcium challenge (10.86±0.26 versus 10.89±0.34 mg/dl; P=0.83). Intact PTH serum levels did not differ between groups (62.23±45.86 versus 61.5±26.83 pg/ml; P=0.9), and both groups showed decreased PTH levels in response to increased dietary calcium intake (5.34±5.21 versus 6.72±5.82 pg/ml; P=0.65). Serum phosphate (PO₄³⁻) levels did not differ significantly between groups at baseline (4.65±0.15 versus 4.83±0.16 mg/dl; P=0.15) or when decreased in response to dietary calcium loading (1.81±0.66 versus 1.86±0.57 mg/dl; P=0.95). However, urinary calcium to creatinine ratios in Six2-Cre floxed Casr animals differed slightly but significantly from those of control animals (0.26±0.02 versus 0.31±0.02; P=0.003) (Figure 5A). This difference increased to approximately two-fold after challenge with higher dietary calcium load (0.31±0.03 versus 0.63±0.14; P=0.001) (Figure 5B). No significant differences between groups were evident in values of serum Mg²⁺ before (2.22±0.3 versus 2.11±0.1 mg/dl; P=0.82) (Figure 5D) and after dietary calcium challenge (2.46±0.4 versus 2.37±0.3 mg/dl; P=0.69) (Figure 5E). Mean urinary Mg²⁺ to creatinine ratios (0.035±0.008 versus 0.035±0.004; P=NS) at baseline did not differ (Figure 5F). Serum 1,25 vitamin D levels were not significantly different, but did show a tendency toward increased levels in Six2-Cre Casr floxed mice (49.14±24.13 versus 35±28.24; P=0.32) (Figure 5G). On the basis of this observation, we tested relative expression of CYP27B1 levels (1-α hydroxylase), which showed an approximately three-fold increase in Six2-Cre floxed animals (2.0±0.15 versus 0.7±0.15; P<0.001) (Figure 5H).

Figure 4.

Selected serum chemistries in age- and sex-matched Six2-Cre Casr floxed animals. On regular chow 0.8% calcium diet (A–C) and with dietary calcium challenge (D–F), no significant differences were observed between Casr-deficient and control mice in serum Ca²⁺ (A and D), PTH (B and E), and serum phosphate levels (C and F). CTL, control.

Figure 5.

Urine calcium to creatinine ratios. (A) Urinary calcium to creatinine ratios were significantly lower in the Six2-Cre Casr floxed mice than in control animals (P=0.003). (B) Addition to drinking water of 1.5% calcium chloride increased the urine calcium to creatinine ratio in control animals much higher than in Six2-Cre Casr floxed mice (P=0.001). (C) Control mice and Six2-Cre Casr floxed mice were injected with furosemide (25 mg/kg body weight). The discrepancy in the urinary calcium to creatinine ratio between the two groups observed at baseline (P=0.03) was no longer evident 2 hours after intraperitoneal injection (P=0.73). (D) No significant differences were observed in serum Mg²⁺ levels between experimental and control mice. (E) Serum Mg²⁺ levels increased equally between both groups with dietary calcium challenge; Mg²⁺ levels in experimental mice remained slightly higher. (F) Mean urinary Mg²⁺ to creatinine ratios (0.035±0.008 versus 0.035±0.004; P=0.74) at baseline did not differ. (G) Serum 1,25 vitamin D levels were increased in experimental mice; however, the difference was not significant. (H) CYP27B1 (1-α hydroxylase) expression levels were approximately three-fold increased in Six2-Cre Casr floxed mice (P<0.001). CTL, control; i.p., intraperitoneal.

Effects of Renal Casr Deficiency on the TAL

To study the role of the TAL in urinary Ca²⁺ reabsorption, we injected animals with furosemide (25 mg/kg body weight per intraperitoneal injection) and measured urine output and Ca²⁺ excretion over a 2-hour period. Figure 5C shows that the difference in urinary Ca²⁺ to creatinine ratio at baseline (0.23±0.01 versus 0.29±0.02 mg/dl; P=0.03) was not evident 2 hours post-treatment (0.55±0.09 versus 0.59±0.05 mg/dl; P=0.73). To further address this finding, we performed QPCR from whole-kidney cDNA and Western blot analysis from whole-kidney protein for NKCC2, which is the target of the loop diuretic furosemide. Relative quantification (RQ) of NKCC2 expression was compared with the housekeeping gene glyceraldehyde 3-phosphate dehydrogenase (GAPDH). The results show no baseline difference between experimental and control animals in levels of mRNA (P=0.77) (Figure 6A) and protein (Figure 6B). However, the anti-phospho-NKCC2 antibody revealed an approximately 4.3-fold increased expression of activated NKCC2 in Six2-Cre floxed animals (5.6±2.6 versus 1.3±0.2; P<0.001) (Figure 6, C and D). Although there is cross-reactivity for the antibody used with NKCC1,²⁰ we believe that the changes observed with this antibody reflect changes in NKCC2 phosphorylation because our Six2 Cre model is not targeting any sites of NKCC1 expression (outer and inner medullary CD cells, the glomerular and extraglomerular mesangium, and the glomerular afferent arteriole).^21,22 In addition, our immunohistochemistry data do not show Casr expression in the CD. It is extremely unlikely that the increased activation that we are seeing is not due to NKCC2 activation, which is more abundantly expressed in kidney and the TAL specifically, the same site in which renal Casr is heavily expressed and targeted by our experiments. On the basis of the validity of the R5 antibody to examine NKCC2 expression²³ and the experimental model that we used in our study, we are confident that our data indicated NKCC2 activation.

Figure 6.

NKKC2 expression. Quantitative PCR for NKCC2 from kidney tissue showed no difference in levels of mRNA (A) or protein (B). (C and D) Phospho-NKCC2 Western blot analysis revealed increased activation of NKCC2 in the same mouse kidney samples. Phospho-NKCC2 showed on average approximately four-fold increased expression in Six2-Cre Casr floxed mice (P<0.001). CTL, control.

QPCR studies on the TAL genes KCNJ1 (renal outer medullary potassium channel, ROMK) and CLCNKB (voltage-gated basolateral chloride channel) showed no difference in relative expression (Supplemental Figure 2, A and B).

To address the mechanism(s) of increased paracellular Ca²⁺ transport in the TAL, we measured relative expression of Claudins 14, 16, and 19 in experimental versus control mice. Claudins are tight junction proteins that have been implicated in paracellular cation transport in the TAL. Loss-of-function mutations in Claudin16 and Claudin19 can cause severe hypercalciuria and nephrocalcinosis in humans.^24,25 Claudin14-deficient mice exhibit hypocalciuria when challenged with dietary calcium.²⁶ Recent data have shown that Claudin14 is a negative regulator of paracellular cation (Ca²⁺) transport by interaction with the heteromeric Claudin16/19 complex in the TAL.

Our expression data for Claudin14 showed an approximately 80% downregulation in Six2-Cre floxed mice (RQ value, 1.1±0.1 versus 4.9±0.8; P<0.001) (Figure 7A). In contrast, Claudin16 expression showed a small but significant increase in experimental animals (RQ value, 2.3±0.4 versus 1.5±0.4; P=0.002) (Figure 7B). No difference was observed in Claudin 19 expression (RQ value, 1.1±0.2 versus 1.0±0.2; P=0.72) (Figure 7C). MicroRNAs have been implicated in the mechanism of Claudin14 downregulation²⁶; we therefore tested the expression levels of two microRNA molecules (miR-9 and miR-374) that are known to suppress Claudin14 transcript in response to dietary Ca²⁺ intake in reciprocal manner. Neither of these miRNAs showed any significant difference in expression level (P=0.67 and P=0.58, respectively) (Figure 7, D and E).

Figure 7.

Expression of Claudins. (A) Expression data from whole-kidney RNA for Claudin14 showed approximately 80% downregulation in Six2-Cre floxed mice (P<0.001). (B) In contrast, Claudin16 expression showed a small but significant increase in experimental animals (P=0.002). (C) No difference was observed in Claudin19 expression (P=0.72). Expression levels of two microRNAs, miR-9 (D) and miR-374 (E), implicated in Claudin14 expression in in vitro studies, showed no significant difference in experimental versus control mice. CTL, control.

Effects of Renal Casr Deficiency on the Distal Nephron

We did not observe any differences in urinary Mg²⁺ excretion despite a significant decrease in urinary Ca²⁺ excretion in our experimental model at baseline conditions. We measured relative expression levels of genes that could promote compensatory Mg²⁺ reabsorption in the distal nephron. Both SLC12A3 (Na⁺-Cl⁻ cotransporter) and TRPM6 (encoding for a luminal epithelial Mg²⁺ channel in the distal nephron) expression levels showed no significant difference in experimental versus control animals (Supplemental Figure 2, C and D). In addition, we measured relative expression levels of other genes that may be regulated or affected by renal tubular Casr deficiency. Both TRPV5 (encoding for ECaC, the epithelial Ca²⁺ channel in distal nephron) and ATPV06 (encoding for the H⁺-ATPase, the luminal proton pump in distal nephron) showed no significant difference in Six2-Cre Casr floxed mice versus controls (Supplemental Figure 2, E and F).

Discussion

The altered response to extracellular Ca²⁺ by the parathyroid gland in individuals with Mendelian disorders of abnormal Ca²⁺ homeostasis caused by Casr mutations demonstrates the critical role of Casr in parathyroid function.³ However, Casr is expressed in multiple tissues in addition to the parathyroid gland, including the kidney. Because of its multiple roles and this widespread expression, the phenotypic contribution of Casr mutations in the kidney and other nonparathyroid tissues to the overall phenotypes of Casr null mice and to the human disorders FHH and NSHPT has remained unclear. It has been difficult to determine which abnormalities are caused directly by inherited defects in Casr, which are direct effects of Casr alterations within different tissues or cell types, and which are secondary to hypercalcemia and to elevations in PTH.

In this study, we developed a new conditional Casr-deficient mouse model using the Cre/Lox approach. Our conditional (nonfloxed) Casr^flox/flox mouse model has no phenotypic differences compared with wild-type mice. However, when these animals are crossed with germline-Cre transgenic mice, the resulting Casr-deficient germline mouse model recapitulates the phenotypes seen in humans (NSHPT) and the Casr-deficient mouse.^17,27

This conditional Casr-deficient mouse model allowed dissection of the detrimental effects of hyperparathyroidism and severe hypercalcemia from the direct effects of Casr deficiency. We studied the effects of Casr deficiency in the kidney by utilizing Cre enzyme driven by the Six2 promoter, based on the essential role of Six2 in formation of the tubular epithelial cell layer during kidney development.¹⁶ All comparisons were performed between experimental mice (Six2-Cre Casr floxed mice) and sex- and aged-matched control littermates (Casr^flox/flox mice). The Six2-Cre Casr floxed mice were viable and fertile, and displayed no difference in serological markers, even when challenged with a high-calcium diet in the form of 1.5% calcium chloride added to the drinking water. Urine studies, however, indicated lower calcium to creatinine ratios than in control mice, a difference that was increased by a high-calcium diet. We did not observe any significant differences in serum and urinary Mg²⁺ excretion although FHH patients have been reported featuring hypermagnesemia, and Claudin14-deficient mice have displayed hypomagnesiuric hypermagnesemia when challenged with dietary calcium load.²⁶ Serum Mg²⁺ levels were slightly (but not significantly) higher in experimental animals in our study and increased equally in both the experimental and control animals on a high-calcium diet. Our model does not show significant effects on Mg²⁺ homeostasis, which is tightly linked to calcium homeostasis.^28,29 One explanation for these findings could be the compensation for increased Mg²⁺ reabsorption further down in the DCT (e.g., via decreased NCCT and TRPM6 activity).³⁰ We tested relative expression for both of these genes, but did not observe any significant differences between experimental and control mice. Further studies under both high and low calcium and magnesium diets will be necessary to understand the findings on renal Ca²⁺ and Mg²⁺ handling in our model.

Levels of 1,25 vitamin D did not significantly differ between the Six2-Cre Casr floxed and control mice, but showed a tendency to higher levels. Levels of 1-α hydroxylase expression were significantly increased in experimental mice consistent with the changes in 1,25 vitamin D levels. Casr deficiency in the PT, in which Casr expression was reported on the apical membrane, could lead to upregulation of 1-α hydroxylase as a compensatory mechanism.³¹ Given the reduced urinary Ca²⁺ excretion and the slightly increased 1,25 vitamin D levels in experimental mice while on a high-calcium diet, we would expect the Six2-Cre Casr floxed mice to be in positive calcium balance. Although we did not measure gastrointestinal calcium absorption or BMD in this study, it is of interest that patients with FHH exhibit generally normal levels of PTH, 1,25 vitamin D, and BMD despite their reduction in urinary Ca²⁺ excretion.³²

To study the pathway of relative hypocalciuria in our experimental mice, we evaluated the activation state of NKCC2 in the TAL. This experiment was primarily motivated by the results of our furosemide experiment, which equalized urine Ca²⁺ excretion between the mice lacking and expressing Casr in the kidney. We note that transepithelial Na⁺-Cl⁻ transport by the TAL is regulated by multiple, competing neurohumoral influences, including extracellular Ca²⁺. Our study confirms that effects of Ca²⁺ in this nephron segment are mediated through Casr, which is abundantly expressed at the basolateral membrane of TAL cells. Activation of Casr elevates intracellular Ca²⁺ and directly inhibits cAMP generation via pertussis toxin–sensitive G protein, Gi, by a Ca²⁺-inhibitable adenylate cyclase that is expressed in the TAL, and/or by increased phosphodiesterase-dependent cAMP degradation.³³ The relative importance of Casr in the regulation of salt transport by the TAL is illustrated by the phenotype of some patients with gain-of-function mutations in Casr. In addition to suppressed PTH, hypocalcemia, and hypercalciuria (the usual phenotype in autosomal-dominant hypocalcemia), these patients can exhibit features of Bartter’s syndrome such as renal salt wasting, mild alkalosis, mild polyuria, and sometimes increases in circulating renin and aldosterone levels.⁷

Our data suggest that NKCC2 activation is increased in Six2-Cre Casr floxed mice compared with control animals.^34,35 The dietary Ca²⁺ load-dependent hypocalciuria in these animals may reflect activation of TAL Cl⁻ reabsorption, leading to an increased lumen-positive transepithelial voltage gradient, thereby increasing paracellular calcium reabsorption.¹⁰ Casr in the TAL appears to upregulate calcium excretion in response to calcium loading, and other segments of the nephron are not able to compensate by upregulating calcium excretion to the level of the control mice under these conditions. Renal Casr thus provides a “protective” mechanism from developing hypercalcemia.

To examine other (functional) effects of Six2 promoter–driven tubular Casr deficiency, we tested relative expression of vacuolar-type H⁺-ATPase (ATP6V0), which is expressed in the CD and plays an important role in calcium-modulated urinary acidification.⁶ No differences in expression were observed in our model, which could be explained by absence of Casr expression in the CD (no CD Casr expression was observed in neither experimental nor control mice).

The tight junction protein Claudin14 is exclusively expressed in the TAL and plays an important role in Ca²⁺ reabsorption. Recent studies have shown that Claudin14 is crucial in regulating the paracellular cation channel permeability.²⁶ The tight junction proteins Claudin16 and Claudin19 assemble together in a heteromeric paracellular complex constituting either an intercellular pore or ion concentration sensor to regulate paracellular reabsorption of cations in the TAL.³⁶ Claudin14 appears to act as a negative gatekeeper of these heteromeric Claudin16/Claudin19 complexes and thereby modulates Ca²⁺ reabsorption in the TAL in response to dietary Ca²⁺ load.²⁶ The Claudin14-deficient mouse model displays hypocalciuria when loaded with dietary calcium, similar to our Six2-Cre Casr-deficient mouse model.²⁶ They also develop hypermagnesemia and hypomagnesuria under a high-calcium diet, which we have not seen in our model, although our Six2-Cre Casr-deficient mice did have slightly increased serum Mg²⁺ levels with calcium loading. Two microRNAs (miR-9 and miR-374) have been implicated in in vitro expression studies to regulate Claudin14. Cotransfection with antagomirs of both miRNAs in cultured TAL showed a significant increase of Claudin14 expression under various culture conditions. The fold increase of Claudin14 expression in these in vitro studies ranged from 1.87 to 3.86.²⁶ Renal Casr expressed on the basolateral membrane of the TAL has been suggested to act upstream of a Casr-microRNA-Claudin14-Claudin16/19 axis.

Our Six2-Cre Casr floxed mouse model shows a significant approximately 80% downregulation of Claudin14, providing in vivo evidence for a Casr-Claudin14 axis and suggesting that this pathway may be a main regulatory mechanism of paracellular Ca²⁺ reabsorption in the TAL. We also observed a small (approximately 50%) but statistically significant increase in Claudin16 expression, suggesting that Casr may also effect Claudin16 expression in the TAL tight junction structures. An increase in Claudin16 is suggestive of an increase in paracellular cation permeability given that a Claudin16 deficiency or knockdown decreases permeability and leads to renal Ca²⁺ and Mg²⁺ wasting. Claudin16 loss-of-function mutations in human cause familial hypomagnesemia with hypercalciuria and nephrocalcinosis (FHHNC) and Claudin16 knockdown in mice produces hypercalciuria, supporting the notion that an increase in Claudin16 expression may increase tight junction permeability and paracellular Ca²⁺ transport. Interestingly, in vitro studies in the epithelial cell model LLC-PK1 suggested that Claudin14 directly interacts with Claudin16, whereas no direct interaction or (permeability) repression was found with Claudin19.²⁶ We have not seen differences in Claudin19 expression in our model. Claudin14 downregulation with a concomitant increase in Claudin16 expression could further increase renal cation reabsorption in the TAL, which appears to be at least in part under the control of Casr. Patients with Casr gain-of-function mutations display autosomal-dominant hypocalcemia and renal calcium wasting. Although these patients have low PTH excretion, their clinical features, which are also classified as Bartter’s syndrome type 5, certainly support our findings of the role of Casr in regulating paracellular cation permeability. QPCR studies of the miRNAs miR-9 and miR-374, implicated in regulating Claudin14 expression in vitro, did not show any difference in experimental versus control mice. This does not rule out that these miRNAs are involved in Claudin14 regulation. Our negative data could be due to the use of whole-kidney RNA rather than TAL-specific RNA, which could have diluted existing differences in TAL expression of these miRNAs. It is also possible that these miRNAs do not play an important role in Claudin14 regulation in vivo.

Using a genetically engineered mouse model with a LoxP-flanked Casr allele targeting E3, we have demonstrated that renal Casr has a PTH-independent role for renal Ca²⁺ reabsorption that occurs in the TAL, accompanied by and at least in part caused by NKCC2 activation (phosphorylation) and Claudin14 downregulation, both increasing paracellular Ca²⁺ reabsorption without significantly affecting Mg²⁺ reabsorption. We do not yet know if the effects of Casr on NKCC2 activation and Claudin14 expression are independently regulated phenomena or if they are biochemically related. The Six2-Cre Casr floxed mouse model should prove helpful in further addressing the role of Casr in overall calcium homeostasis through regulating Ca²⁺ handling by the kidney.

Concise Methods

Targeted Disruption of the Casr

The targeting vector used to inactivate Casr is depicted in Figure 1. This construct contained the neomycin resistance gene (Neo) inserted in intron 3 of the Casr. Both E3 as well as the Neo gene were flanked by LoxP sites. The construct was electroporated into C1 129/Svj-derived embryonic stem (ES) cells, and targeted cells were selected in G418 and 1-(2-deoxy, 2-fluoro-β-D-arabinofuranosyl)-5-iodoracil (FIAU). Surviving ES cells were screened by Southern blot analysis using a genomic probe external to the targeting vector (data not shown). Targeted ES cells were injected into a C57BL/6 mouse blastocyst and re-implanted into a foster mother. The resultant chimeric offspring were bred with black Swiss mice; Casr^flox/+ offspring were identified by PCR. Approximately 50% of the offspring carried the Casr allele. These matings generated offspring with a ratio of wild-type/heterozygous/homozygous genotypes of 1:2:1.

Animals

Animals expressing Cre recombinase driven by the protamine 1 promoter (129S/Sv-Tg(Prm-cre)58Og/J mice) were obtained from the Jackson Laboratory (Bar Harbor, ME). These mice were crossed with our conditional Casr^flox/+ mice to generate germline Casr-deficient animals. Mice expressing Cre recombinase driven by the Six2-Cre promoter (gift from Andrew McMahon, Harvard Medical School, Boston, Massachusetts) were used to generate a kidney-specific Casr-deficient mouse model (specific for renal tubular epithelial cells across the entire nephron except CD).¹⁶ All animals were kept and treated following the regulations provided by Institutional Animal Care and Use Committee at the Harvard Medical School animal facility and the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Genotyping

DNA extraction was performed by digesting mice tails in 200 μl of 50 mM NaOH for 1 hour at 95°C followed by addition of 20 μl 1M Tris-HCl (pH 7.9). Genotyping was performed using the same PCR protocol (except for Sal1 PCR), as follows: 1× Master Mix (25 μl): 14.75 μl ddH2O, 2.5 μl 2 mM dNTP, 2.5 μl 10× buffer, 1.25 μl 40 pmol/μl forward primer, 1.25 μl 40 pmol/μl reverse primer, 0.25 μl Taq polymerase, and 2.5 μl of DNA sample. The following primers were used: Casr-F (5′-GACTTGCTATGTAGCCCAGAACTG-3′)/Post-Neo-R (5′-ACACCCCAAGTGCTCCTGATAACAG-3′) to test presence of Casr floxed allele, Neo-F (5′-AGGATCTCCTGTCATCTCACCTTGCT-3′)/Neo-R (5′-AAGAACTCGTCAAGAAGGCGATAGA-3′) to test for presence of Flox allele, Cre-F (5′-CCTGGAAAATGCTTCTGTCCG-3′)/Cre-R (5′-CAGGGTGTTATAAGCAATCCC-3′) to test for presence of Cre enzyme, and Casr-E3-F (5′-TGGATTCCGATGGTTACAAGCC-3′)/Casr-E3-R (5′-AAGGGATGTGCTCGGAGCA-3′) to test for presence of E3. The SalI PCR (primers Casr-F and Casr-E3-R) protocol was as follows: 1× Master Mix (25 μl): 20.3 μl ddH2O, 0.5 μl 2 mM dNTP, 2.5 μl 10× buffer, 0.25 μl 40 pmol/μl forward primer, 0.25 μl 40 pmol/μl reverse primer, 0.2 μl Taq polymerase, and 1 μl DNA sample. The SalI amplicon was digested using the New England BioLabs SalI restriction enzyme kit and using the following protocol: 1× Master Mix (20 μl): 7 μl ddH2O, 2 μl 10× buffer #3, 1 μl SalI, and 10 μl SalI PCR product. Restriction was performed at 37°C for 3 hours. The SalI PCR/digest was tested for the presence of one versus two alleles. All PCRs were performed using the following PCR program: 95°C for 10 minutes, followed by 30 cycles of 95° × 30 seconds, 55°C × 45 seconds, 72°C × 1 minute, and a final step of 72°C × 10 minutes. Only the annealing temperature for Neo PCR differed with 57°C.

RT-PCR and Expression Studies

Whole kidneys were extracted from control and experimental animals, immediately frozen in TRIzol, and stored at −80°C until used. Tissue was thawed and homogenized, and RNA extraction was performed using the RNeasy kit from Qiagen following the recommended protocol. Exon-spanning PCR was performed using primers located in Casr E2 and E4. The following primer sequences were used: 2F (5′-CTGTTTGGCCCTCCTGGCTC-′3) and 4R (5′-GCTTGTAACCATCGGAATCCA-3′). PCR master mixes and protocols shown above for Neo PCR were used for all amplification from cDNA. Products were run on 1.5% agarose gels, and bands were cut out and purified with a Qiagen kit and submitted for Sanger sequencing at our core facility following standard protocols. Casr, NKCC2, and GAPDH mRNA levels were assessed by quantitative RT-PCR as previously described utilizing the SYBR green method.³⁷ Primers and annealing temperatures were as follows: Casr (forward: 5′-GAGCACATCCCTTCAACCAT-′3; reverse: 5′-GCTAGAGGAGGCGTAGCTCA-′3; annealing temperature 57°C), NKCC2A³⁸ (forward: 5′-GGTAACCTCTATCACTGGGT-′3; reverse: 5′-GTCATTGGTTGGATCCACCA-′3; annealing temperature 59°C; GenBank ID NM_183354), and GAPDH (forward: 5′-CCATGGAGAAGGCTGGGG-′3; reverse: 5′-CAAAGTTGTCATGGATGACC-′3; annealing temperature 55°C; GenBank ID ×02231). RT-PCR was performed using the Transcriptor First Strand cDNA Synthesis Kit from Roche following the manufacturer’s instructions. Forty PCR cycles were used for all amplifications (using SYBR Green PCR Master Mix from Invitrogen), which were conducted in the linear range of amplification efficiency for each primer set. Data were normalized to GAPDH.³⁹

TaqMan assays were used according to the manufacturer’s recommendations for the following genes (Applied Biosystems by Life technologies). The assay IDs were Mm00462192_m1 (Claudin19), Mm01166037_m1 (TRPV5), Mm00475025_m1 (Claudin16), Mm00444727_s1 (KCNJ1), Mm00517109_s1 (Claudin14), Mm00490564_m1 (CLCNKB), Mm01165918_g1 (CYP27B1), Mm01275821_m1 (SLC12A1), Mm00463112_m1 (TRPM6), Mm00490213_m1 (SLC12A3), and Mm01222963_m1 (ATP6V02). Data were normalized to GAPDH (assay ID 99999915_g1). Relative expression levels of miRNAs were assessed by miRNA-specific RT-PCR (TaqMan MicroRNA Reverse Transcription Kit; Life Technologies) followed by QPCR based on the manufacturer’s recommendations (Applied Biosystems by Life Technologies). Assay IDs were 000583 (hsa-miR-9) and 001319 (mmu-miR-374–5p). The data were normalized to U6-snRNA (assay ID: 001973). The TaqMan PCR protocol was as follows: 95°C for 10 minutes, followed by 40 cycles of 95°C for 15 seconds and 60°C for 1 minute.

Western Blot Analyses

Kidneys were quartered, and individual quarters were homogenized at 4°C in radioimmunoprecipitation assay buffer containing 50 mM Tris (pH 7.4), 150 mM NaCl, 1% NP40, 1% Na⁺ deoxycholate, and 0.1% SDS, with added phosphatase and protease inhibitor cocktail tablets (Roche Diagnostics). The kidney homogenate was then sonicated at 4°C, centrifuged at 14,000 rpm (16,000×g), and the supernatant was transferred to a fresh tube. For Western blot analysis, one-tenth of the supernatant volume per sample was denatured at 95°C in the presence of sample loading buffer containing 2-mercaptoethanol and proteins were separated on 7.5% SDS-polyacrylamide gel for 1 hour at 120 V. Separated proteins were then transferred onto polyvinylidene fluoride membrane (semi-dry transfer; BioRad) at 25 V and 80 mA/blot for 1 hour). The membrane was incubated with blocking solution (5% nonfat milk and 0.1% Tween-20 prepared in 1× PBS) for 1 hour at room temperature. Antibodies to Casr (1:1,000, ADD/5C10, GTX 19347; Genetex), β-actin (1:5,000, A2066; Sigma-Aldrich), and NKCC2 (1:1000, GTX41968; Genetex) were diluted in skim milk blocking solution. Antibody recognizing phospho-NKCC2^20,23 (1:1,000, R5; a gift from Biff Forbush, Yale University) was prepared in BSA blocking solution (0.1% BSA and 0.1% Tween-20 prepared in 1× PBS). The membrane was washed three times in 1× PBS and 0.1% Tween-20 (PBST) followed by 30-minute incubation with horseradish peroxidase (HRP)–conjugated secondary antibody (Santa Cruz Biotechnology) at 1:2000 dilution in skim milk blocking solution. Excess secondary antibodies were washed three times by PBST, and ECL chemiluminescence reagent (Thermo Scientific) was utilized to visualize the protein signals that were recorded on Kodak films or imaged using FluoroChem Q (Cell Biosciences). All Western blot experiments were repeated at least three times with sex- and age-matched control and experimental animals. Western blot signal intensity of phospho-NKCC2 in experimental versus control animals was measured with ImageJ software (http://rsbweb.nih.gov/ij/). Statistical analysis of measured signal intensity was performed using nonparametric testing and was based on three experiments.

Immunohistochemistry

Kidneys were extracted from mice, preserved in 10% formalin, and fixed and embedded in paraffin for histologic sections. Kidney sections were treated for immunohistochemistry by deparaffinization (rehydration). Xylene wash was followed by a series of ethanol and distilled water washes. The tissue went through antigen retrieval using a low pH unmasking solution (Vector Laboratories, CA), followed by a wash in 1× PBS pH 7 (tissue was washed every other step). The tissue was then quenched with a 3% hydrogen peroxide solution to block any endogenous HRP. Blocking buffer was placed on the sections followed by Avidin and Biotin blocking solutions (Vector Laboratories). Sections were incubated with primary antibody mouse anti-Casr (NB120–19347; Novus Biologicals, CO) 1:10,000 in blocking buffer for 1 hour at room temperature followed by a wash in PBS and a application of secondary antibody 1:300 in blocking buffer (horse anti-mouse; Vector Laboratories). After washing off excess secondary antibody, ABComplex-HRP (Vector Laboratories) was added to sections to enhance signaling. Sections were washed and DAB substrate was added. Once optimal signal was achieved, sections were washed in distilled water, counterstained with hematoxylin, and mounted. The same protocol was used for immunohistochemistry with anti-NKCC2 (NBP1–57622, 1:50; Novus Biologicals), anti-Aquaporin (NB110–74682, 1:100, Novus Biologicals), and anti-GPRC6A (IMG-71143, 1:10,000; IMGENEX) antibodies.

Serum and Urine Measurements

Spot urine samples were obtained in the morning from experimental (Casr^−/−, Casr^-/+, Six2-Cre Casr^flox/flox, Six2-Cre Casr^flox/-) and control animals (Casr^flox/flox, Casr^+/+). Serum samples were obtained simultaneously using the cheek puncture technique for blood draw and utilizing serum separation tubes (Becton Dickinson microtainer). Samples were obtained at baseline and after exposure of the animals to autoclaved 1.5% calcium chloride in the drinking water. Water intake was assessed by measuring water volume daily, which was equal compared with regular water. All samples were analyzed for Ca²⁺, PO₄^3-, Mg²⁺, and creatinine levels. Ca²⁺ levels were measured using a Ca²⁺ (Arsenazo) reagent set (Pointe Scientific). Phosphorus levels were measured using a phosphorus (FAST 340) reagent set (Eagle Diagnostics). Mg²⁺ levels were measured using magnesium reagent set (Eagle Diagnostics). Urine creatinine levels were measured using the DCA 2000 Microalbumin/Creatinine Reagent Kit (Siemens). All experiments were performed according to the protocols provided by the manufacturers. Intact serum PTH levels and 1,25 vitamin D levels were measured using ELISA kits (Immunodiagnostic Systems) in all Six2-Cre Casr floxed animals and control animals, per the manufacturer's protocols. For 1,25 vitamin D testing, 100 μl of serum, rather than the recommended 500 μl, was used. Appropriate adjustments were made in the other reagents and the modified assay was validated using controls provided in the kit.

Furosemide Experiments

Mice were weighed, and baseline urine and serum samples (approximately 40 μl) were collected. Mice received an intraperitoneal injection of 25 mg/kg furosemide (Sigma) at a concentration of 2 mg/ml. Serum and urine samples were collected at baseline and 2 hours after furosemide administration. All collected samples were analyzed for Ca²⁺, PO₄³⁻, Mg²⁺, and creatinine following the above protocols.

Histology

Hematoxylin and eosin staining of Casr^−/− and Casr^+/+ whole animals at 8 days of age were performed at the rodent histopathology core at Harvard Medical School (Boston, MA) following standard procedures.

Statistical Analyses

All statistical analyses were performed with GraphPad Prism5 software and nonparametric analysis not assuming Gaussian distribution (Mann–Whitney test).

Disclosures

G.C. Curhan is editor-in-chief for the Clinical Journal of the American Society of Nephrology (CJASN) and section editor and author for UpToDate