Effect of Sodium-Glucose Cotransport Inhibition on Polycystic Kidney Disease Progression in PCK Rats

The sodium-glucose-cotransporter-2 (SGLT2) inhibitor dapagliflozin (DAPA) induces glucosuria and osmotic diuresis via inhibition of renal glucose reabsorption. Since increased diuresis retards the progression of polycystic kidney disease (PKD), we investigated the effect of DAPA in the PCK rat model of PKD. DAPA (10 mg/kg/d) or vehicle was administered by gavage to 6 week old male PCK rats (n=9 per group). Renal function, albuminuria, kidney weight and cyst volume were assessed after 6 weeks of treatment. Treatment with DAPA markedly increased glucose excretion (23.6 ± 4.3 vs 0.3 ± 0.1 mmol/d) and urine output (57.3 ± 6.8 vs 19.3 ± 0.8 ml/d). DAPA-treated PCK rats had higher clearances for creatinine (3.1 ± 0.1 vs 2.6 ± 0.2 ml/min) and BUN (1.7 ± 0.1 vs 1.2 ± 0.1 ml/min) after 3 weeks, and developed a 4-fold increase in albuminuria. Ultrasound imaging and histological analysis revealed a higher cyst volume and a 23% higher total kidney weight after 6 weeks of DAPA treatment. At week 6 the renal cAMP content was similar between DAPA and vehicle, and staining for Ki67 did not reveal an increase in cell proliferation. In conclusion, the inhibition of glucose reabsorption with the SGLT2-specific inhibitor DAPA caused osmotic diuresis, hyperfiltration, albuminuria and an increase in cyst volume in PCK rats. The mechanisms which link glucosuria to hyperfiltration, albuminuria and enhanced cyst volume in PCK rats remain to be elucidated.


Introduction
Polycystic kidney diseases (PKD) are the most frequent entities among the genetically determined renal syndromes [1]. The autosomal dominant form of PKD (ADPKD) is twenty times more frequent than the autosomal recessive form (ARPKD) [2]. Approximately 5-8% of all patients with end-stage renal disease (ESRD) suffer from ADPKD [3]. Although progress has recently been made in the development of treatments which retard the cystic growth, no therapy was shown to be effective in delaying the occurrence of ESRD [4].
It has been shown that renal cAMP is a major driver of cyst growth in PKD [5]. The excessive cAMP production is a consequence of the genetic defect which underlies PKD [6]. Due to an early loss of the urine concentrating capability the production of vasopressin is upregulated in PKD, stimulating the production of cAMP directly through its V 2 receptor in the distal renal epithelium [7]. Therapeutic strategies which decrease the vasopressin-driven cAMP production have been successful in decreasing renal cyst growth and in slowing the decline of renal function in PKD [8][9][10][11]. Thus, treatment of mice, rats and humans with the vasopressin V 2 -receptor antagonist tolvaptan [12], crossing PKD rats (PCK strain) with vasopressin-deficient rats (Brattleboro strain) [13], or increasing fluid intake in rats by adding glucose to the drinking water [14] have all been effective to retard PKD disease progression. Patients with ADPKD tend to have a higher urine output because of a renal concentrating defect and a blunted release of vasopressin [15], but presumably also because drinking large amounts of water has been recommended to patients with ADPKD in an attempt to reduce cyst growth [16,17].
As mentioned, the aquaretic drug tolvaptan (vasopressin V 2 receptor antagonist) was shown to have beneficial effects on polycystic kidney disease progression. It is not known whether the induction of osmotic diuresis would also have such a beneficial effect. We have previously shown that the induction of osmotic diuresis by inhibiting renal proximal tubular sodium-glucose cotransport (SGLT) with phlorizin retards cyst growth and renal functional decline in the Han:SPRD rat model of PKD [18]. Phlorizin is a nonselective SGLT inhibitor which inhibits SGLT1 and SGLT2. In recent years, selective SGLT2 inhibitors have been developed and are now in clinical use for the treatment of hyperglycemia in patients with type 2 diabetes mellitus [19]. To evaluate whether the selective inhibition of SGLT2 is capable of retarding cyst volume progression and delaying renal functional loss, we tested the effect of oral dapagliflozin (DAPA) administration in PCK rats, an orthologous model of ARPKD.

Ethics statement
All animal work was conducted according to relevant national and international guidelines. The protocol was approved by the committee on the Ethics of Animal Experiments at the University of Zürich (Permit Number: 175-2012). All efforts were made to minimize any suffering to animals.

Animals
PCK rats (an orthologous model of autosomal recessive polycystic kidney disease) and normal Sprague-Dawley (SD) rats were used in this study. PCK rats (originally derived from SD rats) were obtained from Charles River Laboratories (Sulzfeld, Germany) while SD rats were obtained from the Rat Resource and Research Center (Columbia, MO, USA). All rats had free access to tap water and were fed a standard rat diet. Only male rats were used in this study as cysts develop more rapidly in male compared with female rats.

Experimental design
DAPA (Bristol-Myers Squibb, Princeton, New Jersey) was dissolved in a vehicle of polypropylene glycol, water and ethanol (45:45:10, v/v/v). At 5-6 weeks of age, male PCK or normal control SD rats (n = 8-9 per group) were given DAPA (10 mg/kg/d) or vehicle by gavage for 5-6 weeks. The doses of the drug and the vehicle were adjusted daily according to the body weight of the rats. Blood was drawn and 24 h urine was collected at baseline and after 2-3 and 5-6 weeks of treatment to assess different parameters of renal function. At the end of the treatment phase (at 12 week of age), the PCK rats were examined by ultrasound to determine the kidney and cyst volume in vivo. All rats were then sacrificed and the kidneys were excised, decapsulated and weighed to calculate the two kidneys to total body weight (2K/TBW) ratios. Kidney slices were fixed in 10% buffered formalin and were then embedded in paraffin for subsequent histological examinations.

Plasma and urine analyses
Plasma and urine aliquots were rapidly frozen and stored at -80°C until their measurement. Glucose, sodium, chloride, creatinine and blood urea nitrogen (BUN) concentrations were determined in plasma and urine using a Cobas 8000 Modular Analyzer from Roche Diagnostics AG (Rotkreuz, Switzerland). Plasma and urine osmolality were measured by using an Advanced Osmometer Model 2020 (Advanced Instruments Inc., Norwood, MA, USA). The urine protein content was analyzed by SDS-PAGE and Coomassie blue staining, adjusting the loading volumes of the samples to the 24 h urine volume. The GenWay rat albumin ELISA kit (San Diego, CA, USA) was used to measure urine albumin concentration, according to the instructions provided by the manufacturer.
Ultrasound imaging of rat kidneys PCK rats were anesthetized with isoflurane (1.5-2%) in oxygen. Physiological variables (heart rate, respiratory rate, rectal temperature) were continuously monitored by using a VisualSonics Advanced Physiological Monitoring Unit (Toronto, Ontario, Canada). The abdomen was clipped and all hair was removed using Veet hair removal cream. Acoustic coupling was ensured using ultrasound coupling gel. Ultrasound images of the kidneys were acquired using a high resolution ultrasound system (Vevo 2100, VisualSonics) equipped with a 18-38 MHz probe (MS400). Following image optimization, transverse 2D and power Doppler images of both kidneys were acquired with the aid of an automated 3D motor head. Care was taken to include the cranial and caudal poles of the kidney where possible (maximum scan distance 28 mm). The slice thickness between scanning planes was 0.072 mm with a maximum of 500 slices. The images were captured in digital raw format (RF). Offline processing of 3D reconstructions was performed using Vevo 2100 software (v1.6.0) on a dedicated workstation. The resulting 3D model was used to determine the total kidney and cyst volumes and the number of cysts.

Tissue sectioning, PAS staining, and cystic index determination
For histological examination, one of the kidneys from each rat was sliced perpendicularly to the long axis at approximately 2 mm intervals. Slices from the midportion of the kidneys were fixed in 10% buffered formalin overnight, and tissues were then embedded in paraffin. Three μm sections were stained with periodic acid-Schiff (PAS) following a routine protocol. The stained sections were subjected to cystic index analysis, using the HistoQuest image analysis software (TissueGnostics, Vienna, Austria). The total kidney area (TKA) and the medullary cystic area (MCA) were determined, and the cystic index was calculated in percent as MCA/ TKA Ã 100.

Immunohistochemical analysis
Immunohistochemistry for Ki67 was performed on 3 μm tissue sections which were deparaffinized and rehydrated. Antigen retrieval was performed in an autoclave oven. Mouse anti-Ki67 antibody (BD Pharmingen, San Jose, CA, USA) was applied for 1 hour. Sections were washed and incubated with the biotinylated secondary antibody (Vector Laboratories Inc. Burlingame, CA, USA) for 30 minutes. After washing, the ABC reagent (Vector) was applied to the sections using 3,3'-diaminobenzidine with metal enhancement as the detection reagent. The percentage of Ki-67 positive cells was determined from the total number of cells in cysts and non-cystic tubules from each kidney section. We counted the total number of cells in 10 random fields containing cysts and 10 random fields of non-cystic tissue in each kidney section of each PCK rat using HistoQuest image analysis software (TissueGnostics, Vienna, Austria).

Determination of cAMP content of whole kidneys
The frozen kidneys were ground to fine powder under liquid nitrogen in a stainless steel mortar. After the liquid nitrogen had evaporated, tissue was weighed and homogenized in 10 volumes of 0.1 M HCl. After centrifugation at 600 g for 10 min at room temperature, the supernatants were collected and assayed for cAMP without acetylation using an enzyme immunoassay kit (Sigma-Aldrich, Inc., St. Louis, MO, USA). The protein content was determined by using the BCA protein assay kit from Pierce (Rockford, Illinois, USA). The results were expressed in pmol/mg of protein.

Statistics
Data are expressed as means ± SE. The individual parameters of the DAPA group were compared with those of the CON group by a two-tailed Student's t-test for unpaired data by using the GraphPad Prism version 5.0 software (GraphPad, San Diego, CA, USA). P values of <0.05 were considered statistically significant.

Effect of DAPA treatment on body weight, diuresis and electrolytes
Treatment with DAPA was well tolerated in PCK rats and they appeared healthy throughout the 6 week treatment phase. Treatment of 6-week old male PCK rats with DAPA (10 mg/kg/d) induced immediate and sustained glucosuria that was accompanied by a 2-fold increase in water intake and a 3-fold increase in urine output when compared with vehicle-treated controls ( Table 1). Feces output was also increased by 23% at 3 week and by 45% at 6 week in DAPAtreated PCK rats compared with controls. Plasma glucose (P glucose ) levels did not change in response to DAPA, and plasma osmolality (P osm ), sodium (P Na + ) and chloride (P Cl -) concentrations remained stable at 3 and 6 weeks of treatment. DAPA-treated PCK rats displayed a slightly enhanced urine sodium and chloride excretion rate compared with vehicle-treated rats. Of note, DAPA-treated rats had a 6.4% lower body weight than controls at the end of the 6 week treatment phase (Table 1 and Fig 1A). Treatment with DAPA in normal SD rats also resulted in the induction of glucosuria and an increase in water intake and urine output and a slight increase in urine sodium and chloride, whereas P Na + and P Cl remained stable ( Table 2).
Effect of DAPA treatment on renal function and albuminuria Table 3 shows that the plasma creatinine concentrations were similar at 3 weeks but tended to increase at 6 weeks in DAPA-treated PCK rats. The plasma BUN concentrations were significantly higher at 3 and 6 weeks of DAPA treatment. The clearances of creatinine and BUN were significantly higher at 3 weeks, but the difference disappeared at 6 weeks of treatment with DAPA. Calculating the [clearance creatinine + clearance BUN / 2] confirmed that there was a higher clearance at 3 but not at 6 weeks of DAPA treatment in PCK rats (Fig 1B). Table 4 shows the data in normal SD rats. In DAPA-treated normal rats we did not see an increase in the creatinine and BUN clearances, and the [clearance creatinine + clearance BUN / 2] was also similar between DAPA-vs. vehicle-treated SD rats. These data suggest that there was transient hyperfiltration and subsequent deterioration of renal function upon treatment with DAPA in PCK but not in SD rats. Since albuminuria correlates with disease severity in PKD we analyzed the urine protein content by SDS-PAGE and determined the urine albumin excretion by ELISA (Fig 2). SDS-PAGE of urines (adjusted for 24 h urine volumes) revealed a marked increase of albuminuria in DAPA-treated PCK at 6 weeks (Fig 2A). The ELISA showed that albumin excretion  increased slightly at 3 weeks and a further increase was observed at 6 weeks in vehicle-treated PCK rats, as expected ( Fig 2B). In DAPA-treated rats the albuminuria increased markedly at 3 weeks and more importantly at 6 weeks. After 6 weeks the albumin excretion amounted to 1.7 ± 0.3 mg/d in vehicle-treated rats and to 6.8 ± 0.7 mg/day in DAPA-treated animals (P<0.0001). In normal SD rats, the excretion of urine albumin was lower than in PCK rats but also increased slightly upon treatment with DAPA for 5 weeks (0.4 ± 0.0 vs. 1.2 ± 0.2 mg/day, P<0.05).

In vivo high resolution ultrasound imaging of rat kidneys
After 6 weeks of treatment, the PCK rats were anesthetized for ultrasound analysis of both kidneys. Fig 3A shows that the kidneys with the typical medullary cysts could be easily visualized   in vehicle-treated male PCK rats. On the other hand, DAPA-treated rats displayed larger kidneys and had substantially bigger cysts (Fig 3B). Ultrasound-based measurements revealed a 35% higher total kidney volume (P<0.01), a 2-fold higher cyst volume (P<0.05) and a 47% higher cystic index (P<0.05) in DAPA-vs. vehicle-treated rats, whereas the number of cysts was similar in both groups after 6 weeks of treatment (P = 0.841) (Fig 3C-3F).

Effect of DAPA treatment on kidney weight and morphology
After 6 weeks of treatment with DAPA the total weight of both kidneys was 23% higher in DAPA-as compared with vehicle-treated male PCK rats (4.32 ± 0.24 vs. 5.33 ± 0.20 g, P<0.01). The two kidneys to body weight (2K/BW) ratio was also significantly increased by 34% in the DAPA-treated rats (P<0.001). PAS-staining of kidney sections revealed that the  cyst index was increased by 43% in DAPA-treated rats (P<0.05) while the total cyst number was not significantly changed (P = 0.074) (Figs 4 and 5). In normal SD rats the total kidney weight of both kidneys was also higher in DAPA-treated rats as compared to the total kidney weight in vehicle-treated rats (3.1± 0.3 vs. 2.3 ± 0.2 g, P = 0.0002). Effect of DAPA treatment on renal cAMP content and renal epithelial cell proliferation Since cAMP is a major driver of cyst growth we analyzed the content of cAMP in PCK kidney tissue by ELISA. Fig 6 shows that the amount of cAMP per mg of total protein was similar in DAPA-and vehicle-treated male PCK rats (P = 0.717). Immunohistochemistry staining for Ki67 was then used to examine whether DAPA treatment had an effect on tubular and cystic epithelial cell proliferation.  (Table 5).

Discussion
Here we show that the induction of osmotic diuresis with the SGLT2-specific inhibitor DAPA leads to an unexpected increase in the kidney and renal cyst volumes of PCK rats, an orthologous model of ARPKD. Furthermore, DAPA-treated PCK but not normal SD rats displayed a transient increase in the clearances of creatinine and BUN and a progressive increase in albuminuria which suggests that DAPA promotes hyperfiltration in PCK rats. The renal cAMP   content was similar, and there was no evidence for heightened epithelial cell proliferation which could explain the greater cyst volume upon treatment with DAPA. Thus, the induction of glucosuria with DAPA appeared to have a negative impact on cystic disease progression in PCK rats.
The increase in glomerular filtration in response to DAPA appeared to be transient, i.e. it was seen after 3 weeks but not after 6 weeks. However it was associated with an increase in albuminuria at 3 weeks and a further increase at 6 weeks. In general, an increase in albuminuria is seen with progressing stages of PKD [20]. The increased amount of urine albumin could be  the consequence of enhanced leakage of albumin from glomeruli, or alternatively it could be caused by decreased tubular reabsorption. Whether albuminuria is reflecting glomerular hyperfiltration damage needs to be examined further. Treatment with SGLT2 inhibitors has been associated with a decrease in the GFR in normal and diabetic rats and in humans with type 2 diabetes mellitus [21,22]. This contrasts with our findings in PCK rats which display evidence of transient hyperfiltration. This hyperfiltration could in part be explained by the enhanced food i.e. protein intake which is seen in response to DAPA [23], since protein loading is a known stimulus for increasing glomerular filtration. However this should have been counterbalanced by the higher sodium delivery at the macula densa which normally leads to afferent arteriolar vasoconstriction [24]. Since normal rats which are treated with DAPA do not develop hyperfiltration and albuminuria [25][26][27], the effects that we observed appear to be specific to the PCK model of PKD.
In a previous study, we found that treatment with the SGLT1/SGLT2 inhibitor phlorizin attenuated albuminuria in Han:SPRD rats, improved GFR, and decreased cyst growth [18]. Contrasting with these results, the DAPA-treated PCK rats displayed enhanced cyst growth. Han: SPRD rats are a non-orthologous model of PKD, with cyst formation exclusively in the proximal tubules [28,29]. In contrast, PCK rats display cyst formation in the distal nephron. Typically, the cysts disconnect from the original proximal tubule in Han:SPRD, whereas in PCK rats the cysts remain connected to the distal tubule. We speculate that the increased intratubular osmotic pressure which is caused by the glucosuria could promote the dilation of distal tubular segments and cysts. This could cause compression of adjacent healthy distal tubules and provide a stimulus to enhance GFR, as it may be seen in partial obstructive uropathy [30].
The enhanced cyst growth did not appear to be the consequence of vasopressin-mediated cAMP stimulation, since the renal cAMP content was similar between DAPA-and vehicletreated PCK rats. Despite a massive increase in the diuresis, DAPA-treated rats did not appear to be dehydrated, and the serum sodium concentrations did not increase. This suggests indirectly that vasopressin levels did not increase. Furthermore, Ki67 staining did not reveal an increase in proliferating cells in dilated tubules, and there was no change in the number of nuclei in the cysts upon DAPA treatment. This indicates again that the cysts increased in size due to osmotic pressure without affecting epithelial cell proliferation.
Diabetes may occur in patients with ADPKD and it is associated with greater cardiovascular morbidity [31]. With the possibility to treat ADPKD patients with SGLT2 inhibitors there could be a risk of albuminuria and increased cyst growth. Although such a complication has not yet been described caution should be exerted when prescribing these drugs to patients with ADPKD.
In summary, we have shown that the induction of osmotic diuresis by inhibiting SGLT2 promotes transient hyperfiltration, albuminuria and increased cyst volume in PCK rats. Further studies need to explore the mechanisms which link the enhanced glomerular filtration to the increase in cyst volume upon SGLT inhibition in this model of PKD.