Lack of an effect of collecting duct-specific deletion of adenylyl cyclase 3 on renal Na⁺ and water excretion or arterial pressure

Abstract

cAMP is a key mediator of connecting tubule and collecting duct (CD) Na⁺ and water reabsorption. Studies performed in vitro have suggested that CD adenylyl cyclase (AC)3 partly mediates the actions of vasopressin; however, the physiological role of CD AC3 has not been determined. To assess this, mice were developed with CD-specific disruption of AC3 [CD AC3 knockout (KO)]. Inner medullary CDs from these mice exhibited 100% target gene recombination and had reduced ANG II- but not vasopressin-induced cAMP accumulation. However, there were no differences in urine volume, urinary urea excretion, or urine osmolality between KO and control mice during normal water intake or varying degrees of water restriction in the presence or absence of chronic vasopressin administration. There were no differences between CD AC3 KO and control mice in arterial pressure or urinary Na⁺ or K⁺ excretion during a normal or high-salt diet, whereas plasma renin and vasopressin concentrations were similar between the two genotypes. Patch-clamp analysis of split-open cortical CDs revealed no difference in epithelial Na⁺ channel activity in the presence or absence of vasopressin. Compensatory changes in AC6 were not responsible for the lack of a renal phenotype in CD AC3 KO mice since combined CD AC3/AC6 KO mice had similar arterial pressure and renal Na⁺ and water handling compared with CD AC6 KO mice. In summary, these data do not support a significant role for CD AC3 in the regulation of renal Na⁺ and water excretion in general or vasopressin regulation of CD function in particular.

within the renal collecting duct (CD), adenylyl cyclase (AC)-derived cAMP plays a major role in the regulation of water and Na⁺ transport by promoting the phosphorylation and/or increased plasma membrane abundance of aquaporin (AQP)2 (7, 15, 16, 36), urea transporters (17, 39), and the epithelial Na⁺ channel (ENaC) (25, 33). To date, nine membrane-bound AC isoforms and one soluble AC isoform have been identified (3). Only three membrane-bound AC isoforms (AC3, AC4, and AC6) are expressed within mouse CD principal cells (35). Recent in vitro studies have found that small interfering RNA against AC3 and AC6, but not AC4, reduced arginine vasopressin (AVP)-stimulated cAMP accumulation in inner medullary CD (IMCD) cells (35). Global knockout (KO) of AC6 caused increased fluid intake, elevated urine volume, and decreased urine osmolality (29). Similarly, we (31) have recently demonstrated that mice lacking AC6 specifically in the CD (CD AC6 KO) have reduced AVP-stimulated cAMP accumulation in the CD and decreased urine concentrating ability. Moreover, these CD AC6 KO mice completely lack AVP-stimulated ENaC activity and have an exaggerated hypotensive response to angiotensin receptor blockade (30). In contrast to AC6, it is uncertain whether AC3 plays a physiological role in the regulation of CD water and Na⁺ transport. In addition to the small interfering RNA findings mentioned above, there is indirect evidence supporting a role of AC3 in AVP regulation of CD function. Several studies have found that Ca²⁺ and calmodulin are required for AVP-stimulated cAMP in cultured (2) or acutely isolated CDs (4, 13), suggesting that Ca²⁺/calmodulin-stimulated AC isoforms are involved (AC1, AC3, and/or AC8) (5). Since only AC3 is expressed in the CD (35), these data raise the possibility that AC3 may modulate AVP-stimulated water and/or Na⁺ transport in the CD. Notably, mice with global AC3 deficiency tend to have increased urine volume (2.0 ± 0.5 ml/day in AC3 KO mice vs. 1.2 ± 0.3 ml/day in control mice) and urinary Na⁺ excretion (0.39 ± 0.14 meq·min⁻¹·100 g body wt⁻¹ in AC3 KO mice vs. 0.23 ± 0.05 meq·min⁻¹·100 g body wt⁻¹ in control mice) despite a 50% reduction in the glomerular filtration rate (GFR) in AC3 KO mice (27), although AC3 KO mice were significantly larger than control mice. Thus, to examine a role of AC3 in mediating AVP-regulated CD Na⁺ and water transport, and to avoid the confounding factor of altered GFR, mice with CD-specific KO of AC3 were generated and examined with respect to modulation of blood pressure (BP) and renal function.

MATERIALS AND METHODS

Animal study approval.

All animal use and welfare adhered to the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Animal breeding, housing, and protocols were approved by the Institutional Laboratory Animal Care and Use Committees of the University of Utah Health Sciences Center and the University of Texas Health Science Center at San Antonio.

Generation of mouse lines.

Floxed (loxP-flanked) AC3 mice were generated with loxP sites flanking exons 4–6 of the Adcy3 gene, the key exons encoding the first catalytic domain. In addition, excision of exons 4–6 causes a frame shift if exons 3 and 7 recombine; if exon 7 or additional exons are skipped, the entire first catalytic domain and some of the transmembrane-spanning domain is deleted in addition to still causing a frame shift. In brief, a targeting construct was made containing loxP sites in introns 3 and 6 of the Adcy3 gene with ∼2-kb flanking homology arms. A FRT-flanked neomycin resistance cassette was inserted immediately 3′ to the loxP site in intron 3. Mice were generated using homologous recombination in embryonic stem cells, blastocyst injection, and identification of founders conferring germline transmission of the floxed allele. The neomycin resistance cassette was eliminated by breeding with mice expressing Flp recombinase under control of the ROSA26 promoter (23). Flp recombinase was then bred out of floxed AC3 mice by crossing with wild-type mice. To generate CD-specific deletion of the Adcy3 gene, floxed AC3 mice were bred with AQP2-Cre mice; the latter mice contain a transgene with 11 kb of the mouse AQP2 gene 5′-flanking region driving expression of Cre recombinase (26). Female AQP2-Cre mice were mated with male floxed AC3 mice; female offspring heterozygous for floxed AC3 and hemizygous for AQP2-Cre were bred with males homozygous for floxed AC3. Animals homozygous for floxed AC3 and hemizygous for AQP2-Cre (CD AC3 KO) were used in all experiments. Sex-matched littermates that were homozygous for the floxed Adcy3 gene but without Cre were used as control mice in all experiments.

Mice with CD-specific disruption of both AC3 and AC6 receptor genes were generated in a manner similar to that described above and as previously described for CD AC6 KO mice (31). Mice were bred to obtain homozygotes for both floxed AC3 and AC6. Through a series of successive breedings, mice were obtained that were homozygous for both floxed AC3 and AC6 (control group) or also were hemizygous for AQP2-Cre (CD AC3/6 KO).

Genotyping.

Tail DNA from floxed AC mice was PCR amplified with the following primers: AC3, forward 5′-CTGCTTTGTCATTACAATTTCC-3′ and reverse 5′-TGAGGACTGCCTTTCTAGAG-3′, which yields a 275-bp product from the floxed Adcy3 gene and a 241-bp product from the wild-type allele; and AC6, forward 5′-GGAAAGTAGATCCTCGCTTCGG-3′ and reverse 5′-CCTACTTACAGAACCGCAGGAG-3′, which yields a 350-bp product from the floxed Adcy6 gene and a 316-bp product from the wild-type allele. The AQP2-Cre transgene was detected using the following primers: murine AQP2, forward 5′-GAGACGTCAATCCTTATCTGGAG-3′; creTag, reverse 5′-GCGAACATCTTCAGGTTCTGCGG-3′; and R₂D₂, reverse 5′-GGCTACTCACAGCATTGACAGC-3′, which yield 600- and 650-bp products for AQP2-Cre and wild-type DNA, respectively.

Analysis of Adcy3 gene recombination.

The brain, heart, lung, spleen, liver, intestine, testis, and kidneys were excised. Kidneys were cut longitudinally into sections containing the entire corticomedullary axis. Kidney sections were then incubated with 1 ml HBSS containing 2 mg/ml collagenase and 2 mg/ml hyaluronidase for 20 min at 37°C. The incubated tissue was rinsed with HBSS and stored on ice until dissection of the tubules. Dissection of proximal tubules, cortical CDs (CCDs), and IMCDs was performed at 4°C. DNA from selected organs and microdissected tubules was isolated and PCR amplified to evaluate Adcy3 gene recombination using the following primers spanning exons 4–6 of the Adcy3 gene: forward 5′-CAGGTAGAATTCTTGCTGGTTC-3′ and reverse 5′-TGAGGACTGCCTTTCTAGAG-3^′. Recombination of the Adcy3 gene yields a 400-bp product; the size of the unrecombined Adcy3 gene is ∼2,000 bp.

Metabolic cage experiments.

CD AC3 KO mice, CD AC3/6 KO mice, and their floxed controls were placed in metabolic cages and given 9 ml of a gelled diet made from 62 g of PMI rodent powdered diet (LD101, LabDiet, Richmond, IN), 7 g gelatin, and 110 ml water (1) with free access to drinking water for 3 days (baseline). For moderate water restriction, mice were switched to 9 ml of the gelled diet made from 124 g of the PMI rodent powdered diet, 7 g gelatin, and 110 ml water with free access to drinking water for 3 days. For marked water restriction, mice were given 9 ml of the gelled diet made by 248 g of the PMI rodent powdered diet, 7 g gelatin, and 110 ml water for 2 days with no access to drinking water. For experiments with 1-deamino-8-d-AVP (dDAVP), floxed control and CD AC3 KO mice were given dDAVP (Sigma, St. Louis, MO) subcutaneously at 1 ng/h via osmotic minipump (model 1002, Alzet, Cupertino, CA). After 4 days of recovery, mice were placed in metabolic cages, and baseline measurements and marked water restriction were performed as described above. In all experiments, urine was analyzed for volume and osmolality.

For Na⁺ balance experiments, mice were fed a normal (0.3%) Na⁺ diet for 3 days followed by a high (3.15%)-Na diet for 7 days. The diets consisted of the normal water gelled diet described above with NaCl added to achieve the high-Na diet. At the end of each diet, blood was taken from the tail vein for the determination of plasma renin concentration (PRC). Twenty-four-hour urine collection was done each day for 3-day normal and 7-day high-salt diet feeding, and urine for days 2 and 3 of normal or high-salt diet was analyzed for volume, Na⁺, and K⁺.

BP monitoring.

BP was monitored in CD AC3 KO mice, CD AC3/6 KO mice, and their floxed controls by radiotelemetry (TA11-PAC10, Data Sciences, St. Paul, MN) with catheters inserted into the right carotid artery. Mice were allowed to recover for 1 wk after surgery. BP and heart rate were monitored during normal and high-salt intake.

Isolated, split-open CD preparation and electrophysiological analyses.

As previously described, CDs were microdissected from kidney slices (<1 mm), and electrophysiological analysis was performed (30). CDs were split open to gain access to the apical membrane of principal cells. ENaC activity was quantified in cell-attached patches of the apical membrane made under voltage-clamp conditions (negative voltage potential: −60 mV) using standard procedures. Patch-clamp analysis was done within 2 h after euthanization. For the current experiments, the typical bath solution was (in mM) 150 NaCl, 5 mM KCl, 1 CaCl₂, 2 MgCl₂, 5 glucose, and 10 HEPES (pH 7.4) and the pipette solution was (in mM) 140 LiCl, 2 MgCl₂, and 10 HEPES (pH 7.4). For each experimental condition, CDs from at least three to seven different mice were assayed.

cAMP experiments.

Inner medullas were isolated, minced in HBSS, and centrifuged to bring down intact tubules and cells. The pellet was resuspended in HBSS + 10 mM HEPES (pH 7.4) and incubated with 1 mM IBMX for 15 min at 37°C followed by AVP (10 pM–1μM, Sigma) or 100 nM ANG II (Sigma) for 10 min. Cells were ethanol extracted, and cAMP levels measured by enzyme immunoassay (Enzo Life Sciences, Farmingdale, NY). Protein concentration was determined by the Bradford assay (Bio-Rad, Hercules, CA).

Analysis of mRNA.

RNA from the inner medulla was extracted using guanidinium isothiocyanate and acid phenol and reverse transcribed, and cDNA levels were determined for AC3, AC6, and GAPDH using real-time PCR (StepOne Plus, Applied Biosystems, Foster City, CA). PCR was performed according to instructions provided by the manufacturer using the Taqman Gene Expression Assay (Applied Biosystems, Carlsbad, CA) with Adcy3 (catalog no. Mm00460371_m1), Adcy6 (catalog no. Mm00475773_g1), and GAPDH (catalog no. Mm03302249_m1) primers.

Hormone and electrolyte analysis.

Plasma and urine Na⁺ and K⁺ were analyzed using an Easylyte Analyzer (Medica, Bedford, MA). Urine osmolality was determined by freezing point depression (Osmett II, Precision System, Natick, MA). Urine creatinine was measured using a quantitative colorimetric creatinine assay (Quantichrom, BioAssay Systems, Hayward, CA). Urine urea was measured using a quantitative colorimetric urea assay (Quantichrom). For plasma AVP, mice were decapitated, blood was collected, plasma was extracted using acetone and petroleum ether, and AVP was determined by enzyme immunoassay (Enzo Life Sciences, Farmingdale, NY). PRC was measured by enzyme immunoassay as the amount of ANG I generated after an incubation with excess angiotensinogen (Peninsula Laboratories, San Carlos, CA) and expressed as the amount of ANG I generated per hour per milliliter of plasma. Urine was hydrolyzed with HCl, ethyl acetate was extracted, and aldosterone was determined by enzyme immunoassay (Enzo Life Sciences).

Immunostaining.

Kidneys from wild-type mice were cut longitudinally into two pieces and fixed in 4% paraformaldehyde in PBS containing 75 mM sucrose overnight at 4°C. Kidney slices were then infiltrated with 30% sucrose in PBS for at least a day, placed on a support in optimal cutting temperature compound (Sakura FineTek, Torrance, CA) for 10 min, and frozen. Cryostat sections (4 μm) were rehydrated in PBS for 5 min and blocked with 1% BSA in PBS for 20 min. Incubations with polyclonal anti-AC3 (1:50, Sigma) and anti-AQP2 (1:100, Santa Cruz Biotechnology, Dallas, TX) primary antibodies were done overnight at 4°C. Incubations with Alexa fluor 488 donkey anti-rabbit secondary antibodies (1:50) and Alexa fluor 555 donkey anti-goat secondary antibodies (1:400) in PBS were carried out at the indicated dilutions for 1 h. Three wash-rinse steps of 10 min with PBS with 1% BSA were included after antibody incubation. Immunostained sections were mounted using a 1:1 mixture of Vectashield and 0.3M Tris·HCl (pH 8.9) before examination with a Nikon FXA epifluorescence microscope.

Western blot analysis.

Inner medullas from CD AC3 KO and floxed control mice were homogenized in ice-cold buffer [10 mM triethanolamine, 250 mM sucrose (pH 7.6), 1 μg/ml leupeptin, and 2 mg/ml PMSF], SDS was added to a final concentration of 1%, samples centrifuged at 13,000 rpm for 20 min at 4°C, and the protein concentration was determined. Samples were diluted in buffer [containing 125 mM Tris (pH 6.8), 4% SDS, 10% glycerol, 10% β-mercaptoethanol, and 0.02% bromophenol blue] and heated at 100°C for 2 min. Proteins (10 μg/lane) were subjected to SDS-PAGE and then electroblotted onto polyvinylidene difluoride membranes (Millipore, Bedford, MA). Membranes were probed with primary antibodies for AQP2 (Santa Cruz Biotechnology) and phosphorylated AQP2 at Ser²⁶⁹ (Phosphosolutions, Aurora, CO) overnight at 4°C. The secondary antibody used for detection was Alexa fluor 680-linked anti-rabbit or anti-goat IgG (Invitrogen, Chicago, IL). Proteins were visualized using infrared detection with the LI-COR Odyssey protein analysis system (LI-COR, Lincoln, NE). Laser densitometry was used to quantify the intensity of the resulting bands. Densitometry values were normalized to β-actin loading controls.

Statistical analysis and data presentation.

Data are reported as means ± SE. Experiments involving varying Na⁺ and water intakes and hemodynamics were analyzed by ANOVA. Patch-clamp, mRNA, protein content, and cAMP experiments were analyzed by a two-sided unpaired Student's t-test except where indicated otherwise. The criterion for significance was P ≤ 0.05. In patch-clamp experiments, for presentation, slow baseline drifts were corrected and current data from some patches were software filtered at 50 Hz.

RESULTS

Confirmation of CD AC3 KO mice.

CD AC3 KO mice were born at the expected frequency and had normal growth rates (as determined by body weight) and no apparent gross abnormalities. Kidneys had normal gross morphology and histology by light microscopy. Analysis of DNA from an organ panel (heart, lung, spleen, liver, and bowel) showed no recombination of the Adcy3 gene, whereas there was evidence of recombination in the brain and testis (Fig. 1A). Analysis of DNA from microdissected tubules from CD AC3 KO mice revealed no Adcy3 gene recombination in proximal tubules (Fig. 1D), some recombination in CCDs (Fig. 1C), and complete recombination (no unrecombined allele detected) in IMCDs (Fig. 1B). The finding that IMCDs had complete Adcy3 gene recombination, whereas CCDs had incomplete recombination, is consistent with the known renal expression of the AQP2-Cre transgene solely in CD principal cells.

Fig. 1.

Recombination of adenylyl cyclase (AC)3 gene (Adcy3) DNA in control mice and mice with collecting duct (CD)-specific disruption of AC3 [CD AC3 knockout (KO) mice] from organ panels (A), inner medullary CDs (B), cortical CDs (CCD; C), and proximal tubules (PT; D). The upper 2-kb band is the unrecombined allele (large arrowhead in B–D), and the 0.4 kb band (small arrowhead in B–D) is the recombined allele. Representative gels from five different mice are shown.

Attempts were also made to confirm CD-specific AC3 KO by immunostaining. As shown in Fig. 2, AC3 was mainly localized to the cortex with relatively faint staining in the medulla. Moreover, AC3 did not show colocalization with AQP2, indicating that its expression in the CD is too low to detect by conventional immunostaining. This experiment was not designed to determine which cortical structure was stained so prominently for AC3; however, the pattern of staining suggested it was distal tubule.

Fig. 2.

Immunostaining with anti-AC3 antibody (green) and anti-AQP2 antibody (red) in kidney sections from control mice. A: the renal cortex and medulla. AC3 was mainly expressed in the cortex region and did not colocalize with aquaporin (AQP)2. B: higher-magnification image of the cortex. C: higher-magnification image of the medulla. Nuclei are labeled blue in all images. Images are representative of five separate kidneys.

Additional attempts were made to confirm CD-specific AC3 KO using mRNA analysis. Real-time PCR of microdissected CCDs showed that CD AC3 KO mice had a 70% reduction of AC3 mRNA compared with control mice (Fig. 3A). Papilla from CD AC3 KO mice had 60% less AC3 message compared with control mice (Fig. 3B). Notably, papillary AC3 mRNA was present in 32-fold less abundance compared with AC6 mRNA in control mice (change in threshold cycle: 6.1 ± 0.4 for AC6 vs. 10.3 ± 0.3 for AC3, normalized to GAPDH mRNA, n = 5).

Fig. 3.

Relative mRNA expression of AC3/GAPDH in microdissected CCDs (A) and renal papilla (B) from CD AC3 KO and floxed control mice (n = 5–6 mice/group). *P < 0.05 vs. floxed control mice.

While the DNA recombination and mRNA expression strongly suggested that AC3 was indeed effectively targeted, some functional evidence of Adcy3 gene disruption was sought. To assess this in a relatively straightforward manner, the effect of agonist-stimulated cAMP in the isolated inner medulla was assessed with varying concentrations of AVP (10 pM–1 μM). There were no differences in cAMP accumulation at any concentration of AVP between CD AC3 KO and floxed control mice (Fig. 4A). In contrast, ANG II (100 nM) elicited significantly less cAMP in the inner medulla from CD AC3 KO compared with control mice (Fig. 4B). Thus, based on the ANG II cAMP and DNA recombination data, it does indeed appear that the Adcy3 gene is effectively and selectively (within the kidney) targeted in CD principal cells.

Fig. 4.

cAMP production in the minced inner medulla from CD AC3 KO and floxed control mice (n = 8–17 mice/group) after arginine vasopressin (AVP; A) or ANG II (B) stimulation. Samples were stimulated with 1 nM AVP or 100 nM ANG II for 10 min in the presence of 1 mM IBMX. *P < 0.05 vs. floxed control mice.

Effect of CD AC3 KO on renal water and urea excretion.

The first series of experiments examined the effect of CD AC3 KO on renal water and urea excretion. Water intake was similar between CD AC3 KO and control mice during baseline conditions (6.5–7.0 ml/day), moderate water restriction (4.0–4.5 ml/day), or marked water restriction (1.2–0.15 ml/day; Table 1). There were no differences in urine volume, urine osmolality, or urinary urea excretion between both genotypes during normal water intake. Urinary creatinine excretion was similar between control and CD AC3 KO mice during the normal-salt diet (0.48 ± 0.04 vs. 0.53 ± 0.04 mg/day). Similarly, despite the expected changes in urine volume and osmolality associated with reduced water intake, there were no differences in urine volume, urine osmolality, or urinary urea excretion between the two genotypes during moderate or marked water restriction (Table 1). Urinary osmolyte excretion was also not different between control and CD AC3 KO mice in any of the varied fluid intakes (Table 1). The lack of any effect of CD AC3 KO on renal water handling was supported by the finding that inner medullary AQP2 protein expression was similar between CD AC3 KO and control mice (Fig. 5, A and B). Moreover, the abundance of AQP2 phosphorylated at Ser²⁶⁹, which is required for AQP2 localization to the apical plasma membrane (24), was not different between the two groups (Fig. 5, C and D).

Table 1.

Effect of CD AC3 KO on urine volume, osmolality, and urea excretion during normal, moderate, and marked water restriction

	Control	CD AC3 KO
Normal water intake
Fluid intake, ml	6.9 ± 0.3	6.7 ± 0.5
Urine volume, ml	3.3 ± 0.2	3.4 ± 0.3
Urine osmolality, mosmol/kgH2O	1,460 ± 43	1,392 ± 23
Osmolyte excretion, mosmol/day	4.9 ± 0.3	4.7 ± 0.4
Urea excretion, mg/day	138 ± 13	135 ± 16
Moderate water restriction
Fluid intake, ml	4.4 ± 0.4	4.1 ± 0.4
Urine volume, ml	1.5 ± 0.2	1.6 ± 0.2
Urine osmolality, mosmol/kgH2O	2,765 ± 278	3,179 ± 329
Osmolyte excretion, mosmol/day	3.8 ± 0.4	4.6 ± 0.4
Urea excretion, mg/day	141 ± 16	134 ± 13
Marked water restriction
Fluid intake, ml	1.3 ± 0.1	1.4 ± 0.1
Urine volume, ml	0.8 ± 0.1	0.8 ± 0.1
Urine osmolality, mosmol/kgH2O	3,870 ± 188	3,781 ± 261
Osmolyte excretion, mosmol/day	3.0 ± 0.4	3.1 ± 0.5
Urea excretion, mg/day	136 ± 16	123 ± 15

Values are means ± SE; n = 7–15 mice/group. CD AC3 KO, collecting duct (CD)-specific adenylyl cyclase (AC)3 knockout (KO).

Fig. 5.

Effect of CD AC3 KO on protein expression of AQP2 and AQP2 phosphorylated at Ser²⁶⁹ (pS269-AQP2) in the inner medulla. A and C: blots of AQP2 (A) and pS269-AQP2 (C) from the inner medulla of control and CD AC3 KO mice. B and D: densitometry results for AQP2 (B) and pS269-AQP2 (D; n = 5 mice/group).

One possibility for the apparent lack of an effect of CD AC3 KO on renal water and urea excretion could be compensatory changes in plasma AVP levels. As shown in Table 2, the plasma AVP concentration was similar between controls and CD AC3 KO mice. However, there was a high variability in plasma AVP values, so to additionally evaluate for differential responsiveness to AVP, exogenous dDAVP (1 ng/h in osmotic minipump set to deliver agent for 14 days) was administered to control and CD AC3 KO mice. Mice were placed on normal water intake or marked water restriction (Table 3). Under both water intake conditions, there were no differences in fluid intake, urine volume, urine osmolality, urine osmolyte excretion, or urine urea excretion between control and CD AC3 KO mice under any conditions during dDAVP infusion. Taken together, the data on renal function during varying water intake with or without dDAVP, AQP2 expression, and AVP-stimulated cAMP in the inner medulla (discussed above) indicate that CD AC3 KO does not appreciably affect CD water or urea handling or AVP responsiveness with regard to these parameters.

Table 2.

Effect of CD AC3 KO on plasma AVP, plasma renin concentration, and urine aldosterone during normal salt intake

	Control	CD AC3 KO
Plasma AVP, pg/ml	32 ± 9	40 ± 10
Plasma renin concentration, ng·min−1·h−1	143 ± 22	101 ± 34
Urine aldosterone, ng/day	4.6 ± 0.9	3.6 ± 0.8

Values are means ± SE; n = 7–14 mice/group. AVP, arginine vasopressin.

Table 3.

Effect of CD AC3 KO on urine volume, osmolality, and urea excretion during normal and water restriction in the presence of exogenous AVP

	Control	CD AC3 KO
dDAVP + normal water intake
Fluid intake, ml	6.2 ± 0.2	6.3 ± 0.2
Urine volume ml	3.1 ± 0.2	3.4 ± 0.3
Urine osmolality, mosmol/kgH2O	1,578 ± 60	1,501 ± 47
Osmolyte excretion, mosmol/day	4.9 ± 0.2	5.0 ± 0.3
Urea excretion, mg/day	126 ± 9	141 ± 11
dDAVP + water restriction
Fluid intake, ml	1.6 ± 0.1	1.7 ± 0.1
Urine volume, ml	1.1 ± 0.1	1.2 ± 0.1
Urine osmolality, mosmol/kgH2O	3,976 ± 149	4,005 ± 151
Osmolyte excretion, mosmol/day	4.3 ± 0.3	4.5 ± 0.2
Urea excretion, mg/day	140 ± 12	142 ± 8

Values are means ± SE; n = 12 mice/group. Mice were infused with 1-deamino-8-d-AVP (dDAVP; 1 ng/h sc) and then exposed to normal water intake followed by water restriction.

Effect of CD AC3 KO on BP and renal Na⁺ handling.

Telemetric BP and heart rate were determined in CD AC3 KO mice during 5 days of normal salt intake and 7 days of high-salt intake. There were no differences in systolic BP (Fig. 6A), diastolic BP (Fig. 6B), or heart rate (Fig. 6C) between CD AC3 KO and control mice on either diet on any of the days.

Fig. 6.

Effect of CD AC3 KO on systolic blood pressure (SBP; A), diastolic blood pressure (B), and heart rate [HR; in beats/min (bpm); C] during normal (0.3% Na⁺) and high-salt (3.15% Na⁺) intake (n = 11–14 mice/group).

Metabolic balance experiments were conducted during 3 days of normal Na⁺ intake and during 7 days of high-Na⁺ intake. To limit the magnitude of the data, we report here baseline data from days 2 and 3 of normal Na⁺ intake and data from days 2 and 3 of high-Na⁺ intake, the time periods during which renal adjustments to a new high-Na⁺ diet are typically most evident (Table 4). Food and fluid intake were similar between control and CD AC3 KO mice on normal or high-salt diets. Regardless of the amount of salt in the diet, there were no differences in urine volume, urinary Na⁺ or K⁺ concentrations, or urine Na⁺ or K⁺ excretion between the two genotypes.

Table 4.

Effect of CD AC3 KO on metabolic balances on days 2 and 3 after the initiation of normal or high-salt intake

	Day 2		Day 3
	Control	CD AC3 KO	Control	CD AC3 KO
Normal salt intake
Food intake, g	9.6 ± 0.4	9.3 ± 0.2	9.6 ± 0.4	9.8 ± 0.7
Fluid intake, ml	6.7 ± 0.3	6.2 ± 0.5	6.9 ± 0.3	6.7 ± 0.5
Urine volume, ml	3.4 ± 0.3	3.2 ± 0.3	3.3 ± 0.2	3.4 ± 0.3
Urine Na+ concentration, mmol/l	85.7 ± 6.2	81.4 ± 1.5	85.5 ± 2.7	83.9 ± 2.6
Urine K+ concentration, mmol/l	98.2 ± 4.4	84.8 ± 1.5*	84.8 ± 1.5	88.2 ± 3.5
UNaV, μmol/day	280 ± 19	263 ± 19	284 ± 15	281 ± 23
UKV, μmol/day	327 ± 23	276 ± 23	291 ± 14	292 ± 20
High-salt intake
Food intake, g	9.8 ± 1.0	9.4 ± 1.2	10.0 ± 1.1	10.0 ± 0.4
Fluid intake, ml	12.3 ± 1.8	12.2 ± 2.2	14.1 ± 2.1	13.9 ± 1.7
Urine volume ml	4.8 ± 1.8	5.3 ± 2.2	5.6 ± 2.0	7.6 ± 2.7
Urine Na+ concentration, mmol/l	480 ± 43	435 ± 48	431 ± 38	351 ± 29
Urine K+ concentration, mmol/l	100 ± 11	89 ± 13	82 ± 8	85 ± 20
UNaV, μmol/day	1,933 ± 607	2,099 ± 821	2,166 ± 614	2,417 ± 838
UKV, μmol/day	381 ± 104	381 ± 141	393 ± 104	453 ± 144

Values are means ± SE; n = 7–9 mice/group. UNaV, urinary Na⁺ excretion; UKV, urinary K⁺ excretion.

^*P < 0.05 vs. floxed control mice on the same day and similar diet.

Given that changes in urinary Na⁺ excretion can be difficult to detect in mice, and given that the goal of this study was to assess the effect of CD AC3 KO on AVP-regulated CD function, the effect of CD AC3 KO on baseline and AVP-stimulated ENaC activity in acutely isolated split-open CCDs was determined using patch-clamp electrophysiology. Incubation with AVP increased ENaCs [number of channels (N); Fig. 7E], channel open probability (P_o; Fig. 7F), and ENaC activity (NP_o; Fig. 7G) in control mice. Similarly, AVP increased ENaC P_o and NP_o in CD AC3 KO mice (Fig. 7, F and G). AVP tended to elevate ENaC N in CD AC3 KO mice (Fig. 7E), although this did not achieve statistical significance. Finally, there were no significant differences in the degree of AVP-stimulated ENaC N, P_o, and NP_o between control and CD AC3 KO mice (Fig. 7 and Table 5).

Fig. 7.

Effect of CD AC3 KO on epithelial Na⁺ channel (ENaC) activity during baseline conditions and after AVP stimulation. Representative gap-free current traces from cell-attached patches made on the apical membrane of principal cells in split-open CCDs from control mice (A and C) and CD AC3 KO mice (B and D) on a normal salt diet before (A and B) and after treatment with 1.0 μM AVP for 30 min (C and D) are shown. E: number of ENaCs (N). F: channel open probability (P_o). G: ENaC activity (NP_o). *P < 0.05 vs. vehicle within the same group; **no difference from control littermates with identical treatment. See Table 5 for values and number of patches.

Table 5.

Effect of AVP on ENaCs in cortical CDs from CD AC3 KO mice and control littermates fed a normal Na⁺ intake

	NPo	N	Po	f
Control mice
Vehicle	0.52 ± 0.14	2.7 ± 0.41	0.17 ± 0.03	0.59 (13/22)
+AVP	1.92 ± 0.22*	4.5 ± 0.34*	0.41 ± 0.02*	0.78 (38/49)
CD AC3 KO
Vehicle	0.65 ± 0.19†	2.6 ± 0.43**	0.19 ± 0.04†	0.62 (13/21)
+AVP	1.53 ± 0.25*†	3.8 ± 0.46†	0.35 ± 0.03*†	0.83 (24/29)

Values are means ± SE. Isolated cortical CD were treated with 1.0 μM AVP for 30 min. NP_o, epithelial Na⁺ channel (ENaC) activity; N, number of channels; P_o, channel open probability; f, frequency of ENaCs detected in patches.

^*P < 0.05 vs. vehicle within the same group (by t-test);

^**not different from control littermates with identical treatment (by ANOVA + Student-Newman-Keuls posttest).

Since small changes in ENaC activity could be compensated for by alterations in the renin-angiotensin-aldosterone system, these parameters were assessed in CD AC3 KO and control mice. As shown in Table 2, there were no significant differences in PRC or urinary aldosterone excretion between the two genotypes.

Evaluation of the role of AC6 in compensating for CD AC3 KO.

One possibility for the apparent lack of an effect of CD AC3 KO on renal Na⁺ and water handling is compensation by AC6. No differences in AC6 mRNA levels were detected in microdissected IMCDs from CD AC3 KO and control mice (change in threshold cycle: 6.12 ± 0.36 in control mice and 6.15 ± 0.27 in CD AC3 KO mice, n = 13 each group); however, mRNA levels do not necessarily reflect protein abundance, and AC6 protein levels are too low to be reliably measured. Furthermore, as mentioned above, AC6 is in much greater abundance than AC3 in the CD, and previous studies have indicated that AC6 regulates CD Na⁺ and water transport (30, 31). To address the possibility of AC6 compensation, mice were developed with combined AC3 and AC6 KO in the CD (CD AC3/6 KO) and compared with mice with either CD AC3 KO or CD AC6 KO [the latter using historical data (30, 31)].

As expected, CD AC3/6 KO mice were born at the expected frequency and developed normally as described above for CD AC3 KO mice. Recombination of AC3 and AC6 genes occurs specifically in the CD and to the same degree as for single AC3 and AC6 KO mice (data not shown). Such findings are predictable given that the same AQP2-Cre transgenic and floxed AC3 and AC6 mice are used for the double KOs as for the single KOs.

Mice with CD AC3/6 KO were first examined for renal water handling and inner medullary cAMP production. CD AC3/6 KO mice had reduced AVP-stimulated cAMP accumulation compared with control mice (Fig. 8). During normal water intake, CD AC3/6 KO mice had similar fluid intake and tended to have increased urine volume (Table 6). Urine osmolality was modestly but significantly reduced in CD AC3/6 KO mice compared with control mice. Urine volume was significantly increased in CD AC3/6 KO mice during moderate and severe water restriction, whereas urine osmolality was reduced during moderate water restriction and tended to be decreased during severe water restriction (Table 6). There were no differences in water intake, urinary osmolyte excretion, or urinary urea excretion between the two genotypes. Plasma AVP was not significantly different between the two groups, although there was substantial variability (122 ± 45 pg/ml in control mice vs. 92 ± 32 pg/ml in CD AC3/6 KO mice, n = 13–14).

Fig. 8.

AVP-stimulated cAMP generation in the minced inner medulla from CD AC3/6 KO and floxed control mice (n = 8–10 mice/group). All samples were stimulated with 1 nM AVP for 10 min in the presence of 1 mM IBMX. *P < 0.05 vs. floxed control mice.

Table 6.

Effect of CD AC3/6 KO on urine volume, urine osmolality, and urea excretion during normal, moderate, and marked water restriction

	Control	CD AC3/6 KO
Normal water intake
Fluid intake, ml	4.6 ± 0.3	4.7 ± 0.3
Urine volume, ml	2.4 ± 0.2	2.8 ± 0.2
Urine osmolality, mosmol/kgH2O	2,784 ± 103	2,247 ± 183*
Osmolyte excretion, mosmol/day	6.4 ± 0.5	5.8 ± 0.2
Urea excretion, mg/day	185 ± 16	152 ± 15
Moderate water restriction
Fluid intake, ml	3.7 ± 0.5	3.9 ± 0.4
Urine volume, ml	1.4 ± 0.1	2.0 ± 0.3*
Urine osmolality, mosmol/kgH2O	3,225 ± 133	2,705 ± 219*
Osmolyte excretion, mosmol/day	4.3 ± 0.3	4.6 ± 0.2
Urea excretion, mg/day	127 ± 9	136 ± 6
Marked water restriction
Fluid intake, ml	1.4 ± 0.1	1.4 ± 0.1
Urine volume, ml	0.9 ± 0.1	1.2 ± 0.1*
Urine osmolality, mosmol/kgH2O	5,411 ± 405	4,658 ± 379
Osmolyte excretion, mosmol/day	4.3 ± 0.4	5.1 ± 0.3
Urea excretion, mg/day	132 ± 14	136 ± 15

Values are means ± SE; n = 13–14 mice/group. AC3/6, combined AC3 and AC6.

^*P < 0.05 vs. floxed control mice on the same water intake.

Analysis of BP and urinary Na⁺ and K⁺ excretion was conducted for AC3/6 KO and control mice as described for CD AC3 KO mice (Table 7; note that the data are the same in Table 6 for normal water intake and in Table 7 for day 3 normal salt intake since the same mice were used for these two experiments). Food and water intakes were similar between control and CD AC3/6 KO mice during normal salt intake. As described above, urine volume tended to be higher (day 3) or was higher (day 2) in CD AC3/6 KO mice, and KO mice also had reduced urinary Na⁺ concentration. With a high-salt diet, urine volume was significantly increased in CD AC3/6 KO mice, which correlated with an increase in water intake. CD AC3/6 KO mice had reduced urine Na⁺ and K⁺ concentrations during high-salt intake. No differences in urinary Na⁺ or K⁺ excretion between the two genotypes were detected during normal or high-Na⁺ intake. No differences in systolic BP (Fig. 9A), diastolic BP (Fig. 9B), and heart rate (Fig. 9C) between CD AC3/6 KO and control mice were detected during either normal or high-salt intake.

Table 7.

Effect of CD AC3/6 KO on metabolic balances on day 2 and day 3 after normal and high-salt intake

	Day 2		Day 3
	Control	CD AC3/6 KO	Control	CD AC3/6 KO
Normal salt intake
Food intake, g	7.7 ± 0.4	7.2 ± 0.3	7.3 ± 0.4	7.0 ± 0.3
Fluid intake, ml	4.6 ± 0.2	4.7 ± 0.2	4.6 ± 0.3	4.7 ± 0.3
Urine volume, ml	2.1 ± 0.2	2.7 ± 0.3*	2.4 ± 0.2	2.8 ± 0.2
Urine Na+ concentration, mmol/l	159 ± 4	140 ± 10*	145 ± 3	125 ± 7*
Urine K+ concentration, mmol/l	179 ± 7	157 ± 12	151 ± 6	136 ± 8
UNaV, μmol/day	329 ± 26	350 ± 21	339 ± 35	330 ± 12
UKV, μmol/day	364 ± 26	388 ± 20	343 ± 28	360 ± 12
High salt intake
Food intake, g	7.8 ± 0.2	7.8 ± 0.4	7.1 ± 0.4	7.0 ± 0.3
Fluid intake, ml	5.7 ± 0.4	6.7 ± 0.6	4.6 ± 0.2	5.6 ± 0.5*
Urine volume, ml	3.3 ± 0.2	4.7 ± 0.5*	2.7 ± 0.2	3.6 ± 0.3*
Urine Na+ concentration, mmol/l	221 ± 8	193 ± 4*	220 ± 9	196 ± 4*
Urine K+ concentration, mmol/l	130 ± 7	116 ± 11*	151 ± 7	125 ± 11*
UNaV, μmol/day	721 ± 54	909 ± 88	583 ± 35	699 ± 62
UKV, μmol/day	413 ± 23	521 ± 40*	396 ± 20	414 ± 20

Values are means ± SE; n = 13–14 mice/group.

^*P < 0.05 vs. floxed control mice on the same day and diet.

Fig. 9.

Effect of combined CD-specific knockout of AC3 and AC6 (CD AC3/6 KO) on SBP (A), DBP (B), and HR (C) during normal (0.3% Na⁺) and high-salt (3.15% Na⁺) intake (n = 7–9 mice/group).

The magnitude of the decreased AVP-stimulated inner medullary cAMP and impaired urinary concentrating ability in CD AC3/6 KO mice was similar to that seen in CD AC6 KO mice using comparisons with previously published data (Table 8). These findings support the notion that the lack of an apparent effect of CD AC3 KO on renal Na⁺ and water handling is not due to a compensatory response by AC6.

Table 8.

Comparison of effects of CD-specific KO of AC3, AC6, and AC3/6 on AVP-stimulated cAMP production, urine volume, and urine osmolality during normal water intake and water restriction

	CD AC3 KO	CD AC6 KO	CD AC3/6 KO
AVP-stimulated cAMP production	109 ± 17	58 ± 16*	57 ± 10*
Normal water intake
Urine volume	103 ± 8	143 ± 23	117 ± 9
Urine osmolality	95 ± 2	74 ± 6*	81 ± 5*
Marked water restriction
Urine volume	100 ± 12	110 ± 13	133 ± 13*
Urine osmolality	98 ± 6	82 ± 5*	86 ± 7*

Values are means ± SE. Data are shown as percentages of floxed control littermates.

^*P < 0.05 vs. floxed control mice.

DISCUSSION

The present study demonstrates that CD-specific KO of AC3 in mice has no effect on 1) urine volume and urine osmolality during different levels of water intake and/or dDAVP infusion, 2) Na⁺ excretion and BP during normal or high-salt diets, 3) cAMP production after AVP incubation in the inner medulla compared with control mice, 4) hormone levels (AVP, PRC, and aldosterone), 5) basal or AVP-stimulated ENaC activity in isolated CCDs, and 6) medullary AQP2 levels. The present study also shows that CD-specific KO of both AC3 and AC6 causes a modest decrease in AVP responsiveness associated with impaired urinary concentrating ability. Compared with previous studies examining CD AC6 KO mice (30, 31), CD AC3/6 KO mice have a similar degree of impaired urine concentrating capacity as CD AC6 KO mice. Taken together, these findings suggest that AC6, but not AC3, is involved in CD Na⁺ and water handling in general and in AVP regulation of CD solute and water transport in particular.

The failure to observe an effect of Adcy3 gene targeting in our model is unlikely to be due to ineffective KO of AC3. The floxed region of the AC3 gene contains most of the first catalytic domain, which is critical for cyclase activity; we observed essentially 100% Adcy3 gene recombination in the IMCD. It is impossible to assess the completeness of Adcy3 gene targeting in other regions (the outer medulla and cortex) of the CD since intercalated cells are not targeted. However, mRNA expression from microdissected CCDs showed a 70% reduction of AC3 in CD AC3 KO mice. Since the CCD contains ∼60–70% principal cells (6, 37), the CCD AC3 mRNA data strongly suggest that the Adcy3 gene in the CCD was completely targeted. Finally, as discussed below, CD AC3 KO mice did have reduced ANG II-induced cAMP accumulation in the inner medulla, indicating that these mice do manifest a new phenotype.

The finding that CD AC3 KO had no detectable effect on BP or urinary Na⁺, K⁺, urea, or water excretion was unexpected. As stated above, mice with global AC3 deficiency tended to have increased Na⁺ and water excretion despite having a markedly reduced GFR (27). Previous studies by our group (35) and Hoffert et al. (13) demonstrated that AC3 mRNA and/or protein are detected in the mouse CD, whereas both of these studies provided evidence in vitro implicating AC3 in AVP-stimulated cAMP in the mouse CD. The quantitative mRNA analysis and immunostaining performed in the present study, when taken together, indicate that CD AC3 is present in relatively small amounts in the mouse CD, at least compared with AC6. However, compensatory changes in AC6 did not account for the lack of an apparent effect of CD AC3 KO on BP or renal function, as evidenced by no change in CD AC6 mRNA levels in CD AC3 KO mice and the lack of a difference in the renal phenotype between CD AC3/6 KO and CD AC6 KO mice. Thus, while the present study does not rule out a role for CD AC3 in the regulation of any aspect of CD function, it is likely that the in vitro data are reflective of in vivo CD AC3 function, at least with respect to action of AVP. These considerations beg the question as to which other AC isoforms in the CD mediate AVP effects; if CD AC6 KO or CD AC3/6 KO causes only a partial reduction in AVP-stimulated cAMP and a modest impairment in urine concentration, and CD AC3 KO has no effect, then perhaps another AC isoform is involved. AC4 and soluble AC are the only other AC isoforms that have been identified in the mouse CD (12, 35); further exploration of the role of these AC isoforms is warranted. Finally, it is possible that downregulation of cAMP phosphodiesterase activity could mitigate the severity of the AC3 KO phenotype; however, this could not completely compensate for the absence of AC3 if AC3 was involved in mediating CD salt and water reabsorption.

Our immunostaining results indicate that AC3 is in greatest abundance in cortical nephron segments; in this regard, it is notable that Pluznick et al. (27) reported prominent AC3 and associated olfactory G protein immunostaining in the mouse distal tubule. Ultimately, studies on the role of AC3 in the regulation of distal tubule function will likely be merited.

The present study determined a decrease in ANG II-stimulated cAMP accumulation in the acutely isolated inner medulla from CD AC3 KO mice compared with control mice. This effect of ANG II was examined to provide support for inactivation of AC3 in the CD; however, these findings are of potential interest with regard to actions of ANG II in the CD. Previous studies have suggest that ANG II may regulate CD water handling through ACs. ANG II, via ANG II type 1 (AT₁) receptors, stimulates cAMP production and increases AQP2 expression in cultured CD cells (18, 19). Mice with either global deficiency or CD-specific KO of AT_1a receptors have impaired urinary concentrating ability (21, 22, 34), and inner medullary expression of both AC3 and AC6 is reduced in whole animal AT_1a KO mice (21), suggesting that ANG II may regulate both of these AC isoforms in the CD. While the present study focused on AVP-regulated CD function, future studies are needed to determine whether actions of ANG II (via the AT₁ receptor) in the CD are mediated, at least in part, through AC3 and/or AC6.

In summary, we have demonstrated, using mice lacking AC3 specifically in principal cells of the CD, that principal cell AC3 is not significantly involved in the regulation of renal Na⁺ or water handling under varying physiological conditions. In particular, principal cell AC3 does not appear to be involved in mediating the actions of AVP in the CD. Future studies examining the role of intercalated cell AC3 in the regulation of renal salt excretion would be of interest; such studies could be performed by targeting both intercalated and principal cells together [using HoxB7-Cre mice (34)]. Finally, whether and how principal cell AC3 is involved in the regulation of other aspects of CD function [such as nontransport effects or actions of ANG II, norepinephrine (38), or aldosterone (32)], whether CD AC3 plays a role in CD pathology under disease conditions, and which other CD AC isoforms regulate CD Na⁺ and water transport remain to be determined.

GRANTS

This work was supported in part by a Merit Review from the Department of Veterans Affairs and by National Institute of Diabetes and Digestive and Kidney Diseases Grant R01-DK-097007 (both to D. E. Kohan).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: W.K., D.S., A.N.V.H., J.D.S., M.A.B., and D.E.K. conception and design of research; W.K., D.S., A.N.V.H., V.B., E.M., and M.A.B. performed experiments; W.K., D.S., A.N.V.H., J.D.S., V.B., E.M., M.A.B., and D.E.K. analyzed data; W.K., A.N.V.H., J.D.S., M.A.B., and D.E.K. interpreted results of experiments; W.K., A.N.V.H., J.D.S., M.A.B., and D.E.K. prepared figures; W.K. and D.E.K. drafted manuscript; W.K., A.N.V.H., J.D.S., and D.E.K. edited and revised manuscript; W.K., D.S., A.N.V.H., J.D.S., V.B., E.M., M.A.B., and D.E.K. approved final version of manuscript.