Ouabain enhances renal cyst growth in a slowly progressive mouse model of autosomal dominant polycystic kidney disease

Abstract

Renal cyst progression in autosomal dominant polycystic kidney disease (ADPKD) is highly dependent on agents circulating in blood. We have previously shown, using different in vitro models, that one of these agents is the hormone ouabain. By binding to Na⁺-K⁺-ATPase (NKA), ouabain triggers a cascade of signal transduction events that enhance ADPKD cyst progression by stimulating cell proliferation, fluid secretion, and dedifferentiation of the renal tubular epithelial cells. Here, we determined the effects of ouabain in vivo. We show that daily administration of ouabain to Pkd1^RC/RC ADPKD mice for 1–5 mo, at physiological levels, augmented kidney cyst area and number compared with saline-injected controls. Also, ouabain favored renal fibrosis; however, renal function was not significantly altered as determined by blood urea nitrogen levels. Ouabain did not have a sex preferential effect, with male and female mice being affected equally. By contrast, ouabain had no significant effect on wild-type mice. In addition, the actions of ouabain on Pkd1^RC/RC mice were exacerbated when another mutation that increased the affinity of NKA for ouabain was introduced to the mice (Pkd1^RC/RCNKAα1^OS/OS mice). Altogether, this work highlights the role of ouabain as a procystogenic factor in the development of ADPKD in vivo, that the ouabain affinity site on NKA is critical for this effect, and that circulating ouabain is an epigenetic factor that worsens the ADPKD phenotype.

NEW & NOTEWORTHY This work shows that the hormone ouabain enhances the progression of autosomal dominant polycystic kidney disease (ADPKD) in vivo. Ouabain augments the size and number of renal cysts, the kidney weight to body weight ratio, and kidney fibrosis in an ADPKD mouse model. The Na⁺-K⁺-ATPase affinity for ouabain plays a critical role in these effects. In addition, these outcomes are independent of the sex of the mice.

INTRODUCTION

Autosomal dominant polycystic kidney disease (ADPKD) is the most common monogenetic disorder of the kidney, present in at least 1 in 1,000 births worldwide (1–3). It is caused by mutations in one allele, followed by the additional somatic mutation of the second allele of the PKD1 and PKD2 genes (4, 5). These genes encode the transmembrane proteins polycystin-1 (PC1) and polycystin-2 (PC2), respectively (6). Although the exact roles of the polycystin proteins are still under heavy investigation, it is known that they can form heteromeric PC1-PC2 complexes that function as an unspecific calcium channel, with PC2 belonging to the transient receptor potential (TRP) family of ion transporters (7) and PC1 functioning as a mechanosensory protein in cilia, with G protein-coupled receptor features (8). Through these functions, PC1 and PC2 control intracellular pathways that affect cell proliferation, adhesion, and migration, serving to maintain normal renal epithelial architecture and function (9–12).

ADPKD is characterized by the formation of fluid-filled cysts in the kidney that grow throughout a patient’s lifetime and eventually compress the normal renal tissue, affecting its function and leading to end-stage kidney disease (ESKD) (13). Although ADPKD cysts can begin to develop in utero, their progression and the functional damage to the kidney can be highly variable (14). This shows that factors beyond the disease-causing mutations can influence ADPKD cyst development, and, specifically, several agents circulating in blood have been identified as enhancers of the ADPKD phenotype (15, 16).

Using different in vitro models to study ADPKD, our laboratory has shown that ouabain is an agent that importantly contributes to the progression of renal cysts in ADPKD (17, 18). Ouabain is a member of a group of steroidal substances, the cardiotonic steroids, which normally circulate in the blood of humans and other mammals in nanomolar concentrations (19–21). Ouabain and ouabain-like substances come either from exogenous sources within the diet or are produced endogenously as hormones (21). We discovered that ouabain stimulates the proliferation of epithelial cells isolated from renal cysts of patients with ADPKD (ADPKD cells) (22). Ouabain also acts as a cofactor, enhancing the effects of forskolin and cAMP on fluid secretion and microcyst development in ADPKD cell cultures (23). In addition, ouabain promotes epithelial to mesenchymal transition (EMT) of ADPKD cells as manifested by the increase of mesenchymal and decrease of epithelial markers (24). These represent key events in the initiation and development of renal cystogenesis (13). In agreement with this, ouabain stimulated tubular dilations in embryonic renal explants of Pkd1^m1Bei mice, a model of ADPKD (25). These surprising effects of ouabain are unique to ADPKD cells, since normal human kidney (NHK) cells or metanephric organ cultures from wild-type (WT) mice do not significantly respond to the cystogenic effects of ouabain (17, 25).

The effects of ouabain require the binding of this substance to its specific receptor, Na⁺-K⁺-ATPase (NKA) (26). NKA is a primary ion transport system expressed at the plasma membrane of most animal cells that exchanges Na⁺ and K⁺ between the cell cytosol and the environment, using ATP as an energy source (27). Although relatively high concentrations of ouabain inhibit the ion transport function of NKA, low amounts of ouabain regulate proliferation, metabolism, motility, and adhesion in a cell- and tissue-specific manner. The actions of low ouabain concentrations depend on the role of NKA as a receptor and signal transducer for ouabain, which recruits and activates a cascade of intracellular messengers and stimulates downstream phosphorylation of different effector proteins (18, 28). In ADPKD, the proliferative and secretory effects of ouabain are mediated by the activation of the epidermal growth factor receptor (EGFR), the tyrosine kinase Src, and the extracellularly regulated kinase (ERK) (17, 29). This signaling pathway has been shown to be important for ADPKD cystogenesis due to an aberrant activation of the mediator kinase Ras, which maintains ERK in an uninhibited state (30). Therefore, ouabain contributes in an additive manner to further stimulate ERK to promote ADPKD cystogenesis.

Interestingly, the peculiar effect of ouabain on ADPKD appears to depend on the abnormal capacity of the ADPKD cells to bind ouabain. Unlike NHK cells, which present a single dose-response to ouabain, ∼25% of NKA in ADPKD cells have an abnormally increased affinity for ouabain (22). This suggests that ADPKD cells are more responsive to the circulating amounts of ouabain than NHK cells, which by activating NKA intracellular signaling cascades will exacerbate ADPKD cyst growth (17).

Although we have demonstrated that ouabain promotes ADPKD cystogenesis in vitro, it is still unknown whether ouabain is also able to mediate these effects in vivo. Moreover, it is unclear whether the increased affinity of NKA is important for the response of ADPKD cells to ouabain. Here, we show that ouabain enhances cystogenesis in Pkd1^RC/RC mice, a slowly progressive model of ADPKD (31). In addition, we describe that in Pkd1^RC/RC mice in which NKA is genetically modified to have a higher-than-normal ouabain affinity, the in vivo cystic effects of ouabain are enhanced.

METHODS

Generation of Mouse Models

All experimental protocols in this work involving mice were approved by the University of Kansas Medical Center (KUMC) Institutional Animal Care and Use Committee (IACUC). The Pkd1^RC/RC mice were originally obtained from the KUMC Rodent Models & Drug Testing Core. Genotyping for this mouse was performed according to Hopp et al. (31). The ouabain sensitive mouse model (NKAα1^OS/OS mice) was generated and generously provided to us by Jerry Lingrel (University of Cincinnati, Cincinnati, OH). This line contains mutations in two of the charged amino acids (R111Q and D122N) bordering the first extracytoplasmic loop of the NKAα1 isoform. These changes render NKAα1 with a higher-than-normal affinity for ouabain (32). We rederived this mouse and crossed it with Pkd1^RC/RC mice to obtain a new mouse in which all NKA have a higher capacity to bind ouabain under the ADPKD genotype (Pkd1^RC/RCNKAα1^OS/OS mice). All mouse lines were engineered on the C57 black 6 genetic background (C57BL/6 strain).

Ouabain Treatment

Male and female mice were randomly assigned to receive injections of ouabain dissolved in saline solution or saline alone. Ouabain solution was made fresh every 14 days and kept at 0.03 μg/μL. Pkd1^RC/RC mice received a daily dosage of 0.3 μg/g body wt, which is known to maintain constant blood ouabain at levels within those reported as being physiological (∼10–50 nM) (33–35). Pkd1^RC/RCNKA^OS/OS mice were administered 0.1 μg/g body wt. Mice were injected via intraperitoneal injection daily from day 9 after birth until months 1 to 5. At that point, mice were euthanized, blood was collected, and kidneys were dissected for the study.

Histological Analysis of Kidneys

Kidneys were analyzed from mice after 1, 2, 3, 4, or 5 mo of injections of saline or ouabain. Animals were euthanized by carbon dioxide per KUMC IACUC guidelines before organ harvest. Kidneys were isolated, weighed, and photographed before being fixed in Bouin’s solution for 16 h at 4°C on a rotator. After this time, kidneys were washed three times for 5 min in 70% ethanol and then stored in fresh 70% ethanol until processing. Tissue section and hematoxylin/eosin (H&E) staining were completed by the Biospecimen Repository Core Facility at KUMC. Slides were imaged using a ×1 objective lens on a Nikon 80i microscope. Where sections exceeded the visual field, images were photomerged using Adobe Photoshop software. The quantification of total kidney area, percent cyst area, and cyst number of each section were calculated using Image-Pro Premier 9.1 software as previously described (25). Fractional cyst area was calculated as the ratio of the total dilated cystic tubule area divided by the area of the whole kidney.

Determination of Kidney Fibrosis

Picrosirius red staining was used in order to detect fibrosis and collagen deposition on kidney sections, following previously described protocols (36). Quantification of the extent of fibrosis was determined using ImageJ 1.53k software.

Immunocytochemical Analysis

Unstained kidney sections were rehydrated in sequential washes of xylenes (2 × 5 min) and 100%, 95%, 70%, 50%, and 30% ethanol in saline (5 min each) and then rinsed in water for 5 min. Antigen retrieval was performed with citrate buffer (pH 6.0) and microwaved for 15 min. Then, sections were quenched for 15 min with 0.1 M ammonium chloride in saline and permeabilized for 30 min with 0.3% Triton X-100 in saline. Sections were then blocked with 2% BSA in saline for 60 min at room temperature (RT). A pan NKA-α antibody (Santa Cruz Biotechnology, Dallas, TX) was applied at 1:500 in 2% BSA for 16 h at 4°C in a hydration chamber. After three 5-min saline washes, a goat anti-mouse Alexa Fluor 488 secondary antibody (Abcam, Waltham, MA) at 1:500 was applied for 1 h at RT in the dark in a hydration chamber. Slides were mounted with ProLong Gold Antifade with DAPI (Thermo Fisher Scientific, Waltham, MA). Slides were imaged using a ×60 objective lens on a Nikon 80i fluorescence microscope.

Circulating Urea Nitrogen Measurements

Cardiac puncture was used to obtain blood from each mouse, which was left to coagulate at RT for 20 min. Blood plasma was separated by centrifugation (10,000 g for 5 min), and blood urea nitrogen (BUN) was measured using the BioAssay Systems QuantiChrom Urea Assay Kit (DIUR-100) following the manufacturer’s protocol (BioAssay Systems, Hayward, CA).

Kidney Membrane Preparation

To obtain a crude preparation of membranes, kidneys were minced with a razor blade and then homogenized in a glass homogenizer using 40 strokes in a buffer containing 250 mM sucrose, 25 mM imidazole pH 7.4, and 0.1 mM EGTA (suspension buffer). The solution was then centrifuged at 1,000 g for 5 min at 4°C. The supernatant was collected, and the pellet was resuspended in 1.0 mL of buffer and homogenized with 40 strokes in a glass homogenizer and centrifuged again as described earlier. The supernatants from both centrifugations were combined and spun at 30,000 g for 30 min in a Beckman TLA-55 rotor at 4°C (Beckman Coulter, Indianapolis, IN). The pellet was resuspended in 1.0 mL of buffer. Protein concentration of the sample was measured using the dye-binding assay based on the method of Bradford, from Bio-Rad (Hercules, CA). Samples were adjusted to 1.0 mg/mL protein and BSA and sodium deoxycholate were added at 0.75 μg/μg protein and 0.1% final concentration, respectively. After mixing for 30 min, the solution was ultracentrifuged at 100,000 g for 60 min at 4°C (Beckman TLA-55 rotor). The supernatant was discarded, and the pellet was resuspended in 200 µL of suspension buffer to be used in the assay.

NKA Assays

NKA activity was determined by the initial rate of release of P_i from ATP, using malachite green as described previously (37). The assay was adapted to a microtiter plate format with each well of a 96-well plate containing a final volume of 100 µL. In brief, the incubation solution contained 120 mM NaCl, 30 mM KCl, 3 mM MgCl₂, 0.2 mM EGTA, and 30 mM Tris·Cl, pH 7.4. The assay was performed in the absence or presence of different concentrations of ouabain. A total of 0.125 μg membrane protein was used for each sample. After a 5-min preincubation at 37°C, the reaction was started by the addition of ATP to a final concentration of 1 mM. Assay was performed for 30 min at 37°C and stopped by addition of the malachite green dye. The specific total NKA activity was defined as the Na⁺- and K⁺-dependent ATP hydrolysis sensitive to 10⁻³ M ouabain (38). Dose-response curves for the ouabain inhibition of NKA activity were fitted using a monophasic or a biphasic model, representing a single or two populations of NKA with different affinities for ouabain, as previously described (38).

Statistical Analysis

Statistical significance of the differences between means was determined by a two-tailed Student’s t test with equal variance using GraphPad Prism 9.0.0. Statistical significance was defined as P < 0.05.

RESULTS

Ouabain Increases Cyst Progression in Pkd1^RC/RC Mice

To determine whether the procystogenic effects of ouabain occur in vivo, we applied ouabain in normal saline or saline alone daily by intraperitoneal injection to WT and Pkd1^RC/RC mice from postnatal day 9 and for 1, 2, 3, 4, or 5 mo. To facilitate exploring the effects of ouabain, this compound was used at 0.3 μg/g body wt, allowing for maintenance of constant blood ouabain levels, but at a concentration within that reported as being physiological (∼10–50 nM), as previously described (33–35). At the preselected time points (1–5 mo), mice were euthanized and kidneys were dissected, fixed, sectioned, photographed, and analyzed for cyst area and number.

As shown in Fig. 1, in the absence of exogenous ouabain, the size and number of cysts were higher in Pkd1^RC/RC mice compared with WT mice, as expected. Importantly, although ouabain had no effect on WT kidneys, it significantly augmented cyst area (Fig. 1A) and cyst number (Fig. 1B) in Pkd1^RC/RC mice at all time points examined. Interestingly, ouabain caused a relatively greater increase in cyst number compared with cyst area, which was reflected by a significant decrease in the cyst area/cyst number (CA/CN) ratio from months 3 to 5 (data not shown).

Figure 1.

Ouabain increases cyst progression of Pkd1^RC/RC mice. A–C: quantifications of the average cyst area (A), cyst number (B), and percent kidney weight to body weight ratio (%KW/BW) (C) in WT and Pkd1^RC/RC mice after administration of saline or exogenous ouabain for 1, 2, 3, 4, or 5 mo. D: representative histological images from mixed male (blue ♂) and female (red ♀) mice stained with H&E. The bars represent the means ± SE of 6–10 determinations (4 or 5 males and 3–6 females per bar). Blue squares (□) represent measurements from male mice, whereas red triangles (△) represent measurements from female mice. Statistical values were obtained by Student’s t tests (#P < 0.05, ##P < 0.001, and ###P < 0.0001 vs. WT and *P < 0.05, **P < 0.001, and ***P < 0.0001 vs. ouabain untreated controls). Scale bar = 2,000 μm. H&E, hematoxylin and eosin.

The procystogenic effect of ouabain was reflected in the percent of kidney weight to body weight (%KW/BW), for which we detected increases at 2 and 3 mo (Fig. 1C). Representative histological images of WT and Pkd1^RC/RC mouse kidneys with and without exogenous ouabain from months 1 to 5 are shown in Fig. 1D. These results indicate that ouabain is an agent that has the capacity to promote ADPKD cystogenesis in vivo.

Ouabain Promotes Kidney Fibrosis in Pkd1^RC/RC Mice

A characteristic feature of ADPKD is the deposition of collagen and progressive fibrosis of the renal tissue, which slowly replaces the normal parenchyma and contributes to the decline in renal function (39). We investigated whether, in addition to its effects on ADPKD cystogenesis, ouabain could also influence renal fibrosis in the context of the whole mouse. The extent of fibrosis was measured from months 1 to 5 using picrosirius red staining, which reveals the presence of collagen in the tissue. As shown in Fig. 2A, in the first 2 mo of the experiment, the level of fibrosis stayed low and was similar between WT and Pkd1^RC/RC mice, whether in the absence or presence of injected ouabain. By 3 mo, Pkd1^RC/RC mice that received ouabain had significantly more fibrosis than the untreated controls, a trend that was maintained at 4 and 5 mo. Addition of ouabain did not modify fibrosis in WT mice, and by 5 mo, there was no difference in the extent of fibrosis between WT and Pkd1^RC/RC mice that were not administered exogenous ouabain. Figure 2B shows representative images of renal sections from control and ouabain-treated WT and Pkd1^RC/RC mice at 1 and 5 mo. These results show that ouabain exacerbates not only cyst progression but also renal fibrosis in ADPKD kidneys.

Figure 2.

Ouabain increases kidney fibrosis of Pkd1^RC/RC mice. A: quantification of the amount of collagen deposition relative to total kidney area for WT and Pkd1^RC/RC mice with and without ouabain treatment for 1–5 mo. B: representative histological sections from 1 and 5 mo stained with picrosirius red from male (blue ♂) and female (red ♀) mice. The bars represent the means ± SE of 4 or 5 determinations (2 or 3 males and 2 or 3 females per bar). Blue squares (□) represent measures from male mice, whereas red triangles (△) represent measurements from female mice. Statistical values were obtained by Student’s t tests (##P < 0.001 and ###P < 0.0001 vs. WT and **P < 0.001 vs. no ouabain). Scale bar = 2,000 μm.

Kidney Function Measurements

To assess whether the effects of ouabain had functional consequences to the kidneys, we measured BUN levels in the mice. As shown in Table 1, with the exception of the 4-mo time point, BUN levels were not statistically different between WT and Pkd1^RC/RC mice, under either untreated or ouabain-treated conditions. It should be noted that all groups had similar BUN values to those previously reported for same-age C57BL/6J mice (22–28 mg/dL) (40). This suggests that although ouabain induced higher cyst growth, this did not result in sufficient damage to the remaining normal renal parenchyma to affect kidney function during the duration of this study.

Table 1.

Kidney function is not affected for up to 5 mo after daily ouabain administration

	1 Mo		2 Mo		3 Mo		4 Mo		5 Mo
	BUN, mg/dL	n	BUN, mg/dL	n	BUN, mg/dL	n	BUN, mg/dL	n	BUN, mg/dL	n
Wild-type
Control	27.46 (7.05)	7	23.52 (6.63)	10	20.85 (3.26)	8	21.73 (2.40)	8	24.16 (5.50)	8
Ouabain	29.61 (11.34)	9	22.42 (7.77)	7	21.85 (4.30)	9	20.43 (5.01)	8	25.43 (5.19)	8
Pkd1 RC/RC
Control	25.38 (10.57)	10	21.55 (5.83)	8	25.09 (5.55)	9	25.55# (3.06)	10	27.73 (3.38)	8
Ouabain	29.37 (7.00)	8	22.00 (5.46)	8	22.31 (3.40)	8	26.59# (3.98)	9	30.30 (6.57)	8

Values for blood urea nitrogen (BUN) as a metric of kidney function are shown as means (SD) for the number of animals indicated under “n”. Statistical analysis was performed using Student’s t tests (#P < 0.05 vs. WT).

Sex Dependence of Ouabain Effects

ADPKD progression in humans has been associated with sex differences, with males being more susceptible than females to early end-stage renal disease and other ADPKD effects (41–44). To assess whether ouabain-induced cyst progression varied depending on the sex of mice, we separately analyzed cystic area, cyst number, and %KW/BW in male and female Pkd1^RC/RC mice. As shown in Table 2, no sustained significant differences were found between males and females. Although females that received ouabain for 3 mo had significantly fewer cysts than their male counterparts, this difference was not apparent at any other time point and the cyst area and %KW/BW were not statistically different. Therefore, ouabain does not seem to have a preferential effect due to the sex of the mice.

Table 2.

Ouabain effects in Pkd1^RC/RC mice are not sex dependent

	Cyst Area (%)		Cyst Number		%KW/BW		n
Ouabain	−	+	−	+	−	+	−	+
1 Mo
Male	8.47 (2.25)	17.23 (3.96)	495.6 (257.6)	1,873.0 (1,304.5)	1.69 (0.27)	1.72 (0.28)	5	4 or 5
Female	8.09 (3.42)	17.20 (5.86)	455.6 (65.4)	842.2 (481.0)	1.64 (0.21)	1.67 (0.19)	5	5
2 Mo
Male	7.91 (3.75)	23.51 (5.64)	627.8 (326.0)	2,400.5 (1,617.9)	1.50 (0.14)	1.89 (0.33)	4	4
Female	8.48 (3.36)	27.77 (13.56)	1,148.5 (1,188.3)	2,759.0 (1,805.5)	1.62 (0.26)	1.95 (0.40)	4	4
3 Mo
Male	11.21 (3.77)	29.81 (10.82)	573.8 (197.1)	4,695.5 (1,150.1)	1.66 (0.23)	1.88 (0.35)	4	4 or 5
Female	9.35 (5.12)	26.90 (7.82)	638.8 (210.7)	2,221.8* (1,020.6)	1.50 (0.05)	1.97 (0.26)	4	4
4 Mo
Male	7.88 (5.90)	23.21 (9.38)	510.5 (359.2)	3,350.5 (1,307.4)	1.73 (0.68)	1.89 (0.17)	4	4
Female	14.15 (5.48)	27.88 (10.07)	1,691.0 (1,673.5)	3,800.5 (1,800.1)	1.90 (0.14)	1.85 (0.60)	6	4–6
5 Mo
Male	11.30 (7.63)	29.12 (6.73)	777.5 (332.4)	3,107.5 (911.6)	1.92 (0.46)	2.10 (0.37)	4	4
Female	14.79 (2.74)	32.72 (9.17)	717.8 (106.3)	5,032.5 (2,590.1)	1.77 (0.38)	2.61 (1.17)	4	3 or 4

Values for cyst area, cyst number, and %KW/BW are presented as means (SD). Statistical analyses were performed using Student’s t test (*P < 0.05 females vs. males). %KW/BW, percent of kidney weight to body weight.

NKA Localization in Pkd1^RC/RC Mice

Previous studies have disagreed on the localization of NKA in ADPKD. Although one study reported it to be distributed apically (45), others found NKA to be at the serosal side of ADPKD cells (46–50). To assess the distribution of NKA in Pkd1^RC/RC mice, we performed immunocytochemistry studies. As shown in Fig. 3, both WT and Pkd1^RC/RC mice showed basolateral localization of NKA in normal and cystic tubular epithelial cells.

Figure 3.

NKA localized to the basolateral side of the ADPKD cystic epithelium. WT (A) and Pkd1^RC/RC (B) male mouse kidneys are shown. NKA was identified using a specific anti-α antibody and secondary goat anti-mouse Alexa Fluor 488 secondary antibody. Nuclei were stained with DAPI. Scale bar = 50 µm. ADPKD, autosomal dominant polycystic kidney disease; NKA, Na⁺-K⁺-ATPase.

NKA in Pkd1^RC/RC Mice Has Altered Ouabain Affinity

We have previously found that ADPKD cells have a pool of NKA with a higher affinity for ouabain than NHK cells (22). We have also observed that mouse cortical collecting duct cells engineered to stably express the C-terminal domain of polycystin-1, which functions as a dominant negative construct that induces the ADPKD phenotype, present an abnormally increased sensitivity to ouabain and a higher response to the proliferative and secretory effects of ouabain (51). Here, we tested the response of NKA from kidneys of WT and Pkd1^RC/RC mice to ouabain. As shown in Fig. 4, dose-response curves for the inhibition of NKA activity of kidney membranes exhibited a biphasic response, with relatively high and low ouabain affinity populations of NKA. The high ouabain affinity component was calculated to comprise 49.6 ± 3.0% of the total NKA activity and exhibited an IC₅₀ value of 1.26 ± 0.19 × 10⁻⁶ M. The fraction with low ouabain affinity corresponded to 50.4 ± 3.1% and had an IC₅₀ of 1.21 ± 0.38 × 10⁻⁴ M. In contrast, kidneys from WT mice show a single inhibition curve with an IC₅₀ of 1.63 ± 0.14 × 10⁻⁴ M (Fig. 3A). These data agree with our previous in vitro results and support the idea that increased affinity of NKA for ouabain is a characteristic of the ADPKD phenotype (18).

Figure 4.

NKA from Pkd1^RC/RC mouse kidneys exhibits two different affinities for ouabain. A: dose-response curve for ouabain inhibition of NKA activity of kidney membranes obtained from WT mice (n = 3, 3 males). The curve was best fitted to a monophasic response, corresponding to a single population of NKA. B: ouabain inhibition curve of NKA from kidneys of Pkd1^RC/RC mice (n = 3, 2 females and 1 male). Data were best fitted using a biphasic model, which showed 2 NKA populations with different affinity for ouabain. NKA, Na⁺-K⁺-ATPase.

NKA Ouabain Affinity Correlates With Cyst Progression in ADPKD Mice

The relevance of the peculiar high affinity for ouabain that NKA of ADPKD cells have on renal cystogenesis is unknown. We explored this here. To achieve this, we constructed a mouse in which NKA was modified to have a higher-than-normal affinity for ouabain on the background of the Pkd1^RC/RC mouse line (Pkd1^RC/RCNKAα1^OS/OS mouse). In dose-response curves for the inhibition of NKA by ouabain, all NKA of kidneys from these mice displayed a monophasic ouabain dose-response curve with an IC₅₀ of 9.51 ± 0.59 × 10⁻⁷ (Fig. 5). This is approximately three orders of magnitude lower than NKA from WT mice and is similar to the high-affinity pool of NKA from Pkd1^RC/RC mice shown in Fig. 4.

Figure 5.

All NKAs from Pkd1^RC/RCNKAα1^OS/OS mouse kidneys have greater ouabain affinity than WT mice. Shown is a dose-response curve for ouabain inhibition of NKA activity from kidneys of Pkd1^RC/RCNKAα1^OS/OS mice (n = 3, 3 males). The dose-response relationship was best fitted to a monophasic model with a relatively high affinity for ouabain. NKA, Na⁺-K⁺-ATPase.

With confirmation of their increased affinity for ouabain, we analyzed cyst progression in the Pkd1^RC/RCNKAα1^OS/OS mice, both without and with the addition of ouabain, following the protocols described for Fig. 1. However, in these experiments, the ouabain concentration was adjusted to lower amounts (0.1 µg/g) than those used for the Pkd1^RC/RC mice. This alteration was in response to the susceptibility of Pkd1^RC/RCNKAα1^OS/OS mice to the toxic effects of ouabain in all organs, which caused significant mortality when exposed to the higher concentrations used in Pkd1^RC/RC mice. In addition, because of significant renal symptoms observed, the study was only conducted up to 3 rather than 5 mo.

As shown in Fig. 6, in the absence of added ouabain, the kidneys from Pkd1^RC/RCNKAα1^OS/OS mice had significantly increased cyst area (Fig. 6A), cyst number (Fig. 6B), and %KW/BW (Fig. 6C) than Pkd1^RC/RC mice. With the administration of ouabain, the kidneys from Pkd1^RC/RCNKAα1^OS/OS mice also showed higher cyst area than those from Pkd1^RC/RC mice (Fig. 6A). The cystic number (Fig. 6B) and %KW/BW (Fig. 6C) increased at 1 mo compared with those of Pkd1^RC/RC mice, but this was not consistent across all time points. In this manner, our results here show a correlation between the affinity of NKA for ouabain and ADPKD cyst growth.

Figure 6.

Pkd1^RC/RCNKAα1^OS/OS mice with higher NKA ouabain affinity exhibit increased cyst progression. A–C: quantifications of the average cyst area (A), cyst number (B), and percent kidney weight to body weight ratio (%KW/BW) (C) in Pkd1^RC/RCNKAα1^OS/OS mice compared with Pkd1^RC/RC controls (data also represented in Fig. 1) after administration of saline or exogenous ouabain for 1, 2, or 3 mo. D: representative histological images from mixed male (blue ♂) and female (red ♀) mice. The bars represent the means ± SE of 7–15 determinations (4–9 males and 3–8 females per bar). Blue squares (□) represent measurements from male mice, whereas red triangles (△) represent measurements from female mice. Statistical values were obtained by Student’s t tests (#P < 0.05 and ##P < 0.001 vs. Pkd1^RC/RC). Scale bar = 2,000 μm.

In addition to an increase of cyst progression, the kidneys from Pkd1^RC/RCNKAα1^OS/OS mice showed higher levels of fibrosis with time, with a significant increase between 1 and 3 mo (Fig. 7). These values were also higher than those seen in Pkd1^RC/RC mice for the 1–3 mo considered (compare the values of fibrosis between Figs. 2A and 7A). Interestingly, the difference in the extent of renal fibrosis of Pkd1^RC/RCNKAα1^OS/OS mice between 1 and 3 mo was independent from the application of exogenous ouabain. This was clearly different from what we observed in Pkd1^RC/RC mice, in which the saline-only control group did not have an increase of fibrosis, even at the later time point of 5 mo (compare Figs. 2A and 7A). Representative histological images of picrosirius red-stained kidneys from Pkd1^RC/RCNKAα1^OS/OS mice with and without ouabain from 1 and 3 mo are shown in Fig. 7B. These results show that the higher affinity to ouabain of Pkd1^RC/RCNKAα1^OS/OS mice makes these animals more susceptible to its fibrotic effects, which respond in an exacerbated manner to both applied and endogenous ouabain.

Figure 7.

Pkd1^RC/RCNKAα1^OS/OS mice have increased kidney fibrosis. A: quantification of the amount of collagen deposition relative to total kidney area of renal sections from Pkd1^RC/RCNKAa1^OS/OS mice at 1, 2, and 3 mo after daily administration of saline or ouabain. B: representative histological sections stained with picrosirius red from male (blue ♂) and female (red ♀) mice. The bars represent the means ± SE of 4 or 5 determinations (2 or 3 males and 2 or 3 females per bar). Blue squares (□) represent measurements from male mice, whereas red triangles (△) represent measurements from female mice. Statistical analysis was performed using Student’s t tests (†P < 0.05 vs. 1 mo). Scale bar = 2,000 μm.

Despite the massively increased cyst burden, Pkd1^RC/RCNKAα1^OS/OS mice still had relatively preserved kidney function as shown in Table 3. Interestingly, although the BUN values were not statistically different at months 1 and 2 for both control and ouabain groups, there was a significant difference in BUN at the 3-mo time point between the untreated and ouabain-treated mice. However, the BUN values were close to those of Pkd1^RC/RC mice within the same time and treatment (see Table 1). These findings support the idea that the kidney function of Pkd1^RC/RCNKAα1^OS/OS mice is relatively preserved in the short term but may start to show a decline at month 3.

Table 3.

Kidney function of Pkd1^RC/RCNKAα1^OS/OS mice remains within normal range up to 3 mo

	1 Mo		2 Mo		3 Mo
	BUN, mg/dL	n	BUN, mg/dL	n	BUN, mg/dL	n
Pkd1 RC/RC NKAα1 OS/OS
Control	25.78 (4.42)	14	26.73 (4.35)	8	29.97 (4.53)	15
Ouabain	26.77 (6.32)	12	25.58 (6.13)	10	25.44* (3.08)	7

Values for blood urea nitrogen (BUN) as a metric of kidney function are presented as means (SD) for the number of mice indicated under “n”. Statistical analyses were performed using Student’s t tests (*P < 0.05 vs. no exogenous ouabain).

Finally, to further assess whether ouabain-induced cyst progression had a sex preference in the Pkd1^RC/RCNKAα1^OS/OS mouse model, we separately analyzed cystic area, cyst number, and %KW/BW in male and female mice. As shown in Table 4, no significant differences were found between males and females; while 1-mo-old females receiving exogenous ouabain had significantly more cysts and a greater %KW/BW than their male counterparts, this difference was not maintained at 2 and 3 mo. We can conclude that, like in Pkd1^RC/RC mice, ouabain-induced cyst progression is independent of the sex of the mice.

Table 4.

Ouabain effects in Pkd1^RC/RCNKAα1^OS/OS mice are not sex dependent

	Cyst Area (%)		Cyst Number		%KW/BW		n
Ouabain	−	+	−	+	−	+	−	+
1 Mo
Male	21.53 (11.27)	33.90 (13.98)	2,067.1 (2,471.3)	1,608.1 (1,169.7)	19.82 (3.59)	19.01 (3.38)	5–7	5–7
Female	36.57 (23.59)	35.89 (12.85)	1,889.0 (1,978.6)	3,741.5* (1,748.7)	24.77 (10.14)	35.05* (16.12)	5–8	5–7
2 Mo
Male	35.50 (22.91)	37.96 (12.37)	951.0 (516.6)	2,731.0 (2,036.0)	30.94 (15.32)	22.31 (6.43)	4 or 5	4–7
Female	37.10 (8.85)	42.45 (7.16)	702.3 (284.1)	1,355.8 (603.0)	20.85 (9.85)	26.58 (9.85)	3–5	3–5
3 Mo
Male	36.40 (25.93)	41.70 (19.58)	3,191.6 (1,748.9)	2,855.2 (2,006.4)	38.59 (19.12)	24.10 (8.49)	8 or 9	4 or 5
Female	40.10 (5.60)	46.50 (14.03)	2,773.0 (1,669.2)	3,487.3 (2,338.7)	22.42 (6.77)	31.84 (20.05)	5	3 or 4

Cyst area, cyst number, and %KW/BW in Pkd1^RC/RCNKAα1^OS/OS mice after treatment with and without ouabain for 1–3 mo analyzed by sex. Values are presented as means (SD). Statistical values were obtained by Student’s t tests (*P < 0.05 vs. males). %KW/BW, percent of kidney weight to body weight.

DISCUSSION

The progression of ADPKD is highly variable, even among patients carrying the same mutation (52, 53). A great amount of effort has been placed into investigating the factors that influence the course of this disease, and several epigenetic factors have been shown to accelerate ADPKD progression. For example, the hormone arginine vasopressin stimulates ADPKD cystogenesis (29); a lipophilic diterpene found in cystic fluid from ADPKD kidneys that resembles forskolin stimulates the proliferation and transepithelial chloride secretion of human cyst epithelial cells (54); caffeine worsens kidney function in PC1-deficient mice (55); and, in general, agents that act through the production of cAMP have procystic effects (56). In this work, we show that the hormone ouabain is one of those epigenetic factors enhancing ADPKD renal cyst development in vivo. Here, Pkd1^RC/RC mice that received ouabain had a drastic increase in kidney cyst area and number, a moderately elevated %KW/BW, and developed renal fibrosis during the 5 mo of treatment. These results expand our previous in vitro observations, which showed that ouabain stimulates several hallmark events of the ADPKD cystic phenotype, including 1) cell proliferation and fluid secretion of ADPKD monolayers and three-dimensional (3-D) collagen cultures (23); 2) dedifferentiation of ADPKD cell cultures by promoting a partial EMT phenotype (24); and 3) tubular dilation of mouse kidney explants in culture (25). Interestingly, ouabain treatment resulted in a decrease in CA/CN, which suggests a preferential effect on cell proliferation over fluid secretion. This has mechanistic implications and agrees with our previous results, which showed that ouabain by itself strongly stimulates cell proliferation but needs the simultaneous effect of cAMP-enhancing agents such as forskolin to promote fluid secretion (17, 23). In contrast with its effects on Pkd1^RC/RC mice, ouabain has no effects on cyst size and number in WT mouse kidneys and does not appear to induce cystic or fibrotic changes in WT animals, showing that the actions of ouabain are specific to enhancing the ADPKD phenotype.

The relevance of ouabain as a factor influencing ADPKD progression is highlighted by the fact that ouabain and ouabain-like compounds are substances that are normally circulating in the bloodstream of humans (57, 58). Ouabain and its derivatives are present in some plants, and, therefore, they represent an exogenous source of these compounds when ingested (21, 59). Moreover, ouabain and an ouabain isomer have been found to be endogenously synthesized by the adrenal glands and released into the circulation as a hormone upon different stimuli (60, 61). Notably, the levels of ouabain used in our study are within the range reported to be circulating in humans (35, 62), so the effects reported here are physiologically relevant and could be contributing to renal cystogenesis in patients with ADPKD.

In this study, we chose the slowly progressive Pkd1^RC/RC mouse as a model of ADPKD since this line mimics the slow growth of the kidney cysts in humans and allows for more reliable measurements of renal cyst changes on a longer time frame. In this respect, the acceleration of cyst growth caused by ouabain is especially striking because it occurs in a relatively short amount of time. Interestingly, although changes in cyst burden and kidney fibrosis were significant over the study period, we did not detect changes in circulating BUN levels. This suggests that the remaining normal parenchyma is still sufficient to support kidney function in these animals. These results agree with previous data that show that Pkd1^RC/RC mice only develop renal insufficiency later in life when extensive damage of the kidney tissue takes place (31).

The mechanisms leading to the effects of ouabain in ADPKD are not yet completely understood. However, we have previously reported that ouabain’s actions are mediated through NKA as a receptor and signal transducer for the activation of the EGFR-Src-MAPK/ERK pathway (29). Our present work suggests that circulating ouabain must be exerting its effects from the basolateral side of the cells since we found that its target, NKA, is localized to the basolateral region of the cyst lining epithelium. Initial studies have suggested that NKA was mislocalized apically in ADPKD and that this caused the secretory phenotype of ADPKD cysts (45). A series of other studies, using different ADPKD mouse models and intact cysts from human samples, showed that NKA conserves its canonical distribution at the basolateral side of the cells and that fluid secretion in ADPKD is driven by the vectorial apical transport of chloride and not sodium, which is mediated by the cystic fibrosis transmembrane regulator (CFTR) (46–50). Our experiments here provide further evidence for a basolateral distribution of NKA in the Pkd1^RC/RC mouse model, and histological examination of Pkd1^RC/RC and Pkd1^RC/RCNKAα1^OS/OS kidneys supports the basolateral distribution of NKA in normal and cystic epithelial cells. The effect of ouabain in ADPKD is specific to the disease and different from that of normal renal cells and kidneys. This is in line with the known phenomenon of the pleiotropic effects exerted by ouabain, which depend on the tissue and cell type considered (63). In the particular case of the normal kidney, it has been shown that when ouabain is present in concentrations similar to those found physiologically, the renal epithelium responds with an increase in NKA activity (34, 64, 65). This is not the case in ADPKD, where we discovered that cystic cells’ NKA have an exacerbated affinity for ouabain, which is approximately three orders of magnitude greater than that of NHK cells. This abnormal property could provide a sustained effect of ouabain through NKA signaling. This may explain the abnormal response to ouabain seen in ADPKD cells, where the already hyperactivated ERK pathway (15) is further stimulated by ouabain-NKA signaling, driving an increase in cystogenesis.

Our experiments here show that when the affinity of NKA for ouabain is increased on the background of the ADPKD phenotype, cyst progression is significantly augmented. Therefore, cyst area and number, %KW/BW, and fibrosis are increased in Pkd1^RC/RCNKAα1^OS/OS compared with Pkd1^RC/RC mice, in which only a fraction, but not all, NKA have a high affinity for ouabain. In addition, Pkd1^RC/RCNKAα1^OS/OS mice also started showing a decline in renal function, with BUN levels trending to higher levels at the 3-mo time point used in this study. This indicates more rapid damage to the renal parenchyma in these mice compared with Pkd1^RC/RC mice, where BUN measurements remained stable up to 5 mo. It is important to note that, although we did not find changes in BUN both in the Pkd1^RC/RC and Pkd1^RC/RCNKAα1^OS/OS mice, it is possible that a decrease in renal function is already occurring but could not be detected through BUN changes. The use of additional functional studies will be important to further assess renal function in these animals. Moreover, the increase in renal cyst development in Pkd1^RC/RCNKAα1^OS/OS mice occurred even in the absence of exogenously administered ouabain. This further supports the relevance of endogenous ouabain as a circulating compound that promotes ADPKD progression.

Interestingly, the exogenous administration of ouabain to the Pkd1^RC/RCNKAα1^OS/OS mice produced only a slight increase in renal cyst growth compared with the untreated controls. This suggests that the rate of cyst development through the ouabain-induced and NKA-mediated pathways may have reached a maximum in this ADPKD model, which is in accordance with the slowly progressive nature of the Pkd1^RC/RC mutation. Altogether, these results demonstrate that there is a link between NKA ouabain affinity and ADPKD cystogenesis and that this is a mechanism underlying the increased susceptibility of ADPKD cells to the cystic effects of ouabain.

At present, the reason for the higher ouabain affinity of some NKA in ADPKD is unclear. Different isoforms of the catalytic α-subunit of NKA (NKAα1, NKAα2, NKAα3, and NKAα4) are expressed in mammals, each of which possesses a different affinity for ouabain. Previous studies from our laboratory showed that the altered affinity of NKA in ADPKD cells for ouabain is not due to misexpression of NKA isoforms and that, like in normal renal epithelial cells, the naturally occurring NKAα1 is the only isoform expressed in ADPKD cells (22). Therefore, other factors may be modifying NKA ouabain affinity in ADPKD cells, a phenomenon that we will be exploring in subsequent studies.

ADPKD progression has been shown to be influenced by sex as a biological variable. Androgens can worsen the disease, whereas the castration of males reduces renal enlargement and cyst development (66, 67). In contrast, female sex appears to have a protective effect, lessening ADPKD progression, whereas ovariectomy accelerates renal cystogenesis (66, 68). In contrast to these results, Pkd1^RC/RCNKAα1^OS/OS females that received ouabain had a significant increase in cyst number and %KW/BW compared with their male counterparts at 1 mo, though this difference was not present at the other time points. Interestingly, this time point would be just as or before the mice reach sexual maturity; perhaps alterations of circulating sex hormones can influence NKA-ouabain binding, but the reason for this difference is currently unclear. Our results here show no significant differences on the overall cystic effects of ouabain based on the sex of the animals. This is also important if one considers that the structure of ouabain contains a steroidal backbone similar to estrogens and androgens and is therefore related to these sex hormones. Therefore, ouabain appears to influence ADPKD cyst progression over time by a different mechanism than circulating sex hormones.

Perspectives and Significance

In conclusion, this work highlights the role of circulating ouabain as an epigenetic factor that can worsen the ADPKD phenotype and demonstrates that the ouabain affinity of NKA is an important mechanism for cyst progression. This may represent a potential targetable mechanism to attenuate this currently uncurable disease. For example, this work supports the idea that immunoneutralizers of ouabain or other cardiotonic steroids, such as Digibind, which are Fab fragments from antidigoxin-specific antibodies (69), could be explored as cyst-blocking agents. More study is needed to decipher why some NKA in ADPKD have an increased ouabain affinity and whether diminishing the NKA-ouabain interaction may positively influence disease outcomes.

DATA AVAILABILITY

Data will be made available upon reasonable request.

GRANTS

This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK081431 (to G.B.).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

J.T., G.S., and G.B. conceived and designed research; J.T., G.S., and J.M. performed experiments; J.T. and G.S. analyzed data; J.T., G.S., and J.M. interpreted results of experiments; J.T. and J.M. prepared figures; J.T. drafted manuscript; J.T., G.S., J.M., and G.B. edited and revised manuscript; J.T., G.S., J.M., and G.B. approved final version of manuscript.