Responses of renal hemodynamics and tubular functions to acute sodium–glucose cotransporter 2 inhibitor administration in non-diabetic anesthetized rats

The aim of this study is to examine the effects of acute administration of luseogliflozin, the sodium–glucose cotransporter 2 (SGLT2) inhibitor, on renal hemodynamics and tubular functions in anesthetized non-diabetic Sprague Dawley (SD) rats and 5/6 nephrectomized (Nx) SD rats. Renal blood flow (RBF), mean arterial pressure (MAP), and heart rate (HR) were continuously measured and urine was collected directly from the left ureter. Intraperitoneal injection of luseogliflozin (0.9 mg kg−1) did not change MAP, HR, RBF, or creatinine clearance (CrCl) in SD rats (n = 7). Luseogliflozin significantly increased urine volume, which was associated with significantly increased urinary glucose excretion rates (P < 0.001). Similarly, luseogliflozin significantly increased urinary sodium excretion (from 0.07 ± 0.01 µmol min−1 at baseline to 0.76 ± 0.08 µmol min−1 at 120 min; P < 0.001). Furthermore, luseogliflozin resulted in significantly increased urinary pH (P < 0.001) and decreased urinary osmolality and urea concentration (P < 0.001) in SD rats. Similarly, in Nx SD rats (n = 5–6), luseogliflozin significantly increased urine volume and urinary glucose excretion (P < 0.001) without altering MAP, HR, RBF, or CrCl. Luseogliflozin did not elicit any significant effects on the other urinary parameters in Nx SD rats. These data indicate that SGLT2 inhibitor elicits direct tubular effects in non-diabetic rats with normal renal functions.

non-diabetic subjects. It is of particular importance that the pharmacological effects of SGLT2 inhibitors in non-diabetic subjects are studied to minimize any indirect influence induced by changes in blood glucose levels. In this regard, a study with pooled urine reported that treatment with an SGLT2 inhibitor increased the 24-h urine volume, urinary glucose, and sodium excretion in non-diabetic mice 12 . Similarly, an SGLT2 inhibitor tended to increase 24-h urinary excretion of sodium, potassium, and chloride in dogs 13 . However, it is difficult to exclude the possibility that urinary sodium excretion was influenced by sodium intake in these urine-storing studies. Thus, it remains unclear whether SGLT2 inhibitor-induced urinary changes are actually mediated by its direct tubular action or by other indirect mechanisms.
In the current study, we investigated the direct effects of SGLT2 inhibition on renal hemodynamics and tubular functions in vivo by examining the acute effects of luseogliflozin, the selective SGLT2 inhibitor 1, 3 , in anesthetized non-diabetic rats with normal kidney function. Clinically, side effects of SGLT2 inhibitors, such as polyuria and polydipsia, have been reported during the early stages of treatment 1,2 , which leads to the restriction of use of an SGLT2 inhibitor, particularly in patients with chronic kidney disease (CKD). Therefore, we also examined the effects of a SGLT2 inhibitor on renal hemodynamics and functions in 5/6 nephrectomized (Nx) rat, a model of CKD.

Results
A major limitation of these studies with intravenous infusion of luseogliflozin was that intravenously administered vehicle (2-hydroxylpropyl-β-cyclodextrin (HP-β-CD)) had considerable effects on urinary parameters. Specifically, vehicle significantly increased urine flow from 3.27 ± 0.57 to 8.47 ± 1.17 µl min −1 and urinary sodium excretion from 0.20 ± 0.02 to 0.44 ± 0.03 µmol min −1 after 60 minutes in Sprague Dawley (SD) rats (Figs 1 and 2). To minimize the effects of vehicle, preliminary experiments with intraperitoneal injection of luseogliflozin were performed. We found that intraperitoneal administration of vehicle did not change urine flow or urinary sodium excretion. Based on these findings, further experiments using intraperitoneal injection of luseogliflozin were also performed.

Effects of intravenous injection of the SGLT2 inhibitor. Renal hemodynamics. Intravenous injection
of luseogliflozin did not change mean arterial pressure (MAP), renal blood flow (RBF), heart rate (HR), or creatinine clearance (CrCl) in either SD or Nx SD rats (Table 1).
Urine flow and urinary glucose excretion. In SD rats, luseogliflozin significantly increased urine flow from 5.40 ± 1.27 µl min −1 at baseline to 10.33 ± 0.73 µl min −1 at 120 min (P < 0.001; Fig. 1a), which was associated with a significant increase in the urinary glucose excretion rate from 0.003 ± 0.001 mg min −1 at baseline to 0.81 ± 0.18 mg min −1 at 120 min (P < 0.001; Fig. 1b). These changes induced by luseogliflozin were significantly greater than those induced by vehicle ( Fig. 1a and b). On the other hand, luseogliflozin-induced changes in glucose excretion were significantly less in Nx SD rats compared with SD rats (Fig. 1b).
Urinary excretions of sodium. In SD rats, intraperitoneal administration of luseogliflozin induced an increase in urinary sodium excretion from 0.07 ± 0.01 µmol min −1 at baseline to 0.76 ± 0.08 µmol min −1 at 120 min, by approximately 10-fold (P < 0.001; Fig. 4). On the other hand, intraperitoneal administration of luseogliflozin did not have a significant effect on urinary sodium excretion in Nx SD rats (from 0.26 ± 0.15 µmol min −1 at baseline to 0.66 ± 0.17 µmol min −1 at 120 min; Fig. 4). Urinary potassium excretion was not significantly affected by acute intraperitoneal injection of luseogliflozin in either SD rats or Nx SD (data not shown).
Urinary osmolality and free water clearance. In SD rats, luseogliflozin generate a significant reductions in urinary osmolality from 2325 ± 44 mOsm kg −1 at baseline to 1236 ± 121 mOsm kg −1 at 120 min (P < 0.001; Fig. 6a). Accordingly, augmentation of calculated negative values of free water clearance were concomitant with these  urinary changes induced by luseogliflozin in these rats (P < 0.001; Fig. 6b). Luseogliflozin did not have any significant effects on urinary pH, urea concentration, or osmolality in Nx SD rats.

Discussion
Chronic treatment of a SGLT2 inhibitor elicits several systemic effects on blood glucose, body weight, blood pressure, etc. To avoid any systemic changes, the present study utilized acute renal clearance method to examine the effects of luseogliflozin on renal hemodynamics and tubular functions. We found that acute administration of an SGLT2 inhibitor induced natriuresis without changing renal hemodynamics in non-diabetic SD rats. In Protocol 1, rats were anesthetized with sodium pentobarbital and maintained with isoflurane, and luseogliflozin was injected intravenously. Luseogliflozin significantly increased urine flow and urinary glucose excretion in SD rats without affecting MAP, RBF, or CrCl. To avoid any possible effects from the vehicle, luseogliflozin was also injected intraperitoneally (Protocol 2). We confirmed that intraperitoneal injection of vehicle did not change any parameters. However, intraperitoneal injection of luseogliflozin resulted in substantial glycosuria and natriuresis without changing renal hemodynamics in SD rats. The findings that these acute tubular effects of the SGLT2 inhibitor were observed in non-diabetic rats strongly support the hypothesis that the SGLT2 inhibitor elicits direct tubular effects independently of blood-glucose changes. Our data showed that acute administration of luseogliflozin did not have any effects on MAP, mean RBF, or CrCl in non-diabetic SD rats. Previous studies have reported that chronic SGLT2 inhibitor treatment reduced GFR in type 1 diabetic patients with glomerular hyperfiltration 14 and in type 1 diabetic Akita mice 10 . Similarly, chronic SGLT2 blockade significantly decreased GFR in type 2 diabetic rats 1, 15 and patients 16,17 . These studies have indicated that SGLT2 inhibitor-induced reduction in GFR is associated with an increase in the activity of tubuloglomerular feedback (TGF) mechanism. However, SGLT2 inhibitor did not have a significant effect on GFR in the early diabetic condition 18 . In agreement with the results in non-diabetic mice 9 and rats 19 , our data also showed that acute infusion of luseogliflozin did not have any immediate significant effects on renal hemodynamics in non-diabetic SD rats. Thus, it seems that SGLT2 inhibitor-induced renal hemodynamic changes are significant in diabetes and not obvious in non-diabetes. Less amount of glucose reabsorption with SGLT2 inhibition may be related to less effect of SGLT inhibition on the activation of TGF mechanism 11 . Acute administration of luseogliflozin significantly increased urine flow and urinary sodium excretion without affecting renal hemodynamics in non-diabetic SD rats. These acute responses are consistent with a previous study in which it was reported that luseogliflozin increased urinary sodium excretion 3 days after its administration in metabolic syndrome rats 6 . A clinical study has also shown that canagliflozin, a SGLT2 inhibitor, significantly increased urine volume and urinary sodium excretion in patients with type 2 diabetes 8 . As previously noted, a micropuncture study reported that an SGLT2 inhibitor increased urinary sodium excretion in diabetic rats 11 . Collectively, these data suggest that SGLT2 inhibitor elicits direct tubular effects in both diabetic and non-diabetic subjects.
In the current study, we expected urinary osmolality to increase following treatment with the SGLT2 inhibitor owing to a significant urinary glucose excretion. However, acute administration of the SGLT2 inhibitor Figure 6. Urine osmolality and free water clearance in Protocol 2. Effect of intraperitoneal injection of luseogliflozin on urine osmolality (a) and free water clearance (b). SD, Sprague Dawley; Nx, 5/6 nephrectomized; vehicle, SD rats treated with vehicle; luseogliflozin, SD rats treated with luseogliflozin; Nxvehicle, 5/6 Nx SD rats treated with vehicle; Nx-luseogliflozin, 5/6 Nx SD rats treated with luseogliflozin. Values are mean ± SEM. ***P < 0.001 vs. baseline, # P < 0.05 and ### P < 0.001 vs. vehicle.
consistently reduced urinary osmolality in SD rats (P < 0.001). We also observed that the urea concentration significantly decreased with the SGLT2 inhibitor (P < 0.001) in SD rats. As suggested previously 20 , high urinary glucose retains water in the tubular lumen, which dilutes urinary urea. Further studies are needed to examine the effects of SGLT2 inhibitors on the activities of water and urea transporters.
In this study, we also examined the effects of luseogliflozin in Nx SD rats, a model of CKD. In agreement with previous findings 21 , MAP elevation was observed in anesthetized Nx SD rats. Studies have indicated that in the remnant kidney model, chronic reduction in renal mass leads to an increase in blood pressure 22,23 . In these animals, glomerular hyperfiltration of the remaining nephrons was observed, probably to compensate for nephron loss 24 and following structural damage, and the kidney may have allowed an increase in the GFR for adaptation 25 . However, acute administration of the SGLT2 inhibitor did not alter RBF or CrCl in Nx SD rats. These data are inconsistent with a previous clinical study in which SGLT2 inhibitor significantly decreased the GFR in type 1 diabetic patients with glomerular hyperfiltration 14 . Therefore, it is possible that SGLT2 inhibitor-induced changes in renal hemodynamics are not obvious in non-diabetics with glomerular hyperfiltration. Alternatively, the possibility cannot be excluded that responses in the GFR to SGLT2 inhibitors are species specific.
When compared with SD rats, rats after 2 weeks of 5/6 nephrectomy displayed a marked increase in urine volume at baseline. Previous studies have indicated that increases in urine volume were associated with downregulation of the water channels in the collecting duct and proximal tubule 26 . Intraperitoneal administration of luseogliflozin induced a significant increase in urine output in Nx SD rats (P < 0.001). In agreement with previous studies 24 , after administering luseogliflozin, Nx SD rats exhibited lower glucose excretion compared with the SD rats, probably due to renal functional impairments 27 . In agreement with previous studies [28][29][30] , Nx SD rats showed greater urinary sodium and potassium excretion compared with SD rats. However, acute blockade of SGLT2 by luseogliflozin did not alter urinary sodium excretion in Nx SD rats. We have no good explanation as to why the SGLT2 inhibitor does not increase urinary sodium excretion in 5/6 Nx SD rats. One possibility is that reduced renal mass diminishes the functional activities of the sodium transporters, as shown by previous studies 28,31,32 . Consistent with a previous study 33 , we observed a significant reduction in the ability of urine to be concentrated and the urinary urea concentration in Nx SD rats. Luseogliflozin treatment similarly did not result in further reductions in urine osmolality or urea concentration in Nx SD rats. Interestingly, we found a significant increase in urinary pH in SD rats after luseogliflozin administration (P < 0.001), which was not observed in Nx SD rats. These results are consistent with a previous study in which systemic administration of phlorizin (a non-selective SGLT inhibitor) resulted in bicarbonaturia 34 . However, precise mechanism responsible for tubular effect of SGLT2 inhibitor is not clear, due to the limitations of in vivo clearance studies. Further studies are needed to explore the molecular mechanisms for tubular effects of SGLT2 inhibitor.
The major challenge in this study was the solubility of luseogliflozin. In this study, we used dimethyl sulfoxide (DMSO) and HP-β-CD as vehicle. Luseogliflozin could be dissolved in only 100% DMSO, but acute administration of high concentration DMSO caused hematuria. Intravenous administration of HP-β-CD caused neither hematuria nor renal hemodynamic changes; however, urinary parameters were significantly affected by administration of HP-β-CD. On the other hand, we confirmed that intraperitoneal administration of HP-β-CD showed minimal effect on systemic and renal parameter. Preliminary experiments showed that 0.9 mg kg −1 is the maximum dose of luseogliflozin and the minimum concentration of HP-β-CD is 4.5%. We have also documented that intraperitoneal administration of luseogliflozin at 0.9 mg kg −1 caused significant glucosuria, whereas 4.5% HP-β-CD did not change any renal parameters. Another crucial point is the application of an adequate anesthetic. In Protocol 3, we administered luseogliflozin in SD rats under inactin anesthesia. However, the predicted effect on urinary sodium excretion by luseogliflozin was absent, possibly because of the effects of inactin. Previous studies have clearly shown that inactin has a strong ability to inactivate urinary sodium excretion 35 . For these reasons, we used sodium pentobarbital and isoflurane as anesthetics in the current study.
In conclusion, acute administration of the SGLT2 inhibitor induced significant increases in urinary glucose and sodium excretion without altering renal hemodynamics in non-diabetic SD rats. These renal effects of the SGLT2 inhibitor were markedly attenuated in 5/6 Nx SD rats. These findings support the hypothesis that SGLT2 inhibitors elicit direct tubular effects without changes in plasma glucose levels.

Methods
Animals. All experimental procedures were performed according to the guidelines for the care and use of animals established by Kagawa University (Kagawa, Japan). Male SD rats (Japan SLC, Inc., Shizuoka, Japan) were housed in specific pathogen-free animal facilities at a controlled temperature (24 ± 2 °C) and humidity (55 ± 5%) on a 12-h light-dark cycle. Animals had free access to standard chow and water. All surgical and experimental procedures were approved by the Animal Care and Use Committee, Kagawa University.
By following the excision remnant kidney model 22 , an experimental Nx rat model was induced by ablation of two thirds of the left kidney at 7 weeks of age and, after a further week, a total right nephrectomy was performed. After a 2-week acclimation period, rats were assigned to one of two groups: Nx-luseogliflozin and Nx-vehicle. Experimental procedures. SD and Nx SD rats were anesthetized with sodium pentobarbital (50 mg kg −1 , i.p.) and isoflurane (0.5-1.5% in air) or inactin (100 mg kg −1 , i.p.) at 9-11 weeks of age. Then, animals were placed on a heated pad to maintain body temperature at 37 °C. A polyethylene catheter (PE-60; Becton Dickinson and Company, Sparks, MD, USA) was inserted into the abdominal aorta via the right femoral artery for blood pressure measurement and collection of arterial blood. We collected urinary samples for every 30 minutes. The left kidney was exposed through a retroperitoneal flank incision. The renal artery was carefully isolated from the tissue connecting the renal hilum cephalic. A Doppler flow probe (HDP 10.20 R; Crystal Biotech, Northborough, MA, USA) was placed around the renal artery and RBF was continuously monitored. A polyethylene catheter (tapered PE-50; Becton Dickinson and Company) was inserted into the left ureter for urine collection. After surgery, each rat was kept isolated for 60 min to allow for stabilization of MAP, RBF, and urine flow. All rats were given saline as a maintenance fluid at a dose of 0.7 ml h −1 . Blood was collected in EDTA-containing tubes. At the end of experiments, animals were euthanized by overdose of pentobarbital (250 mg kg −1 ). Supplementary Figures 4 and 5.

Experimental protocols. Full details of experimental protocols are given in
Protocol 1. SD and Nx SD rats were anesthetized with pentobarbital and isoflurane, and randomly assigned to one of four groups (n = 5-7 for each): Group 1, SD rats administered luseogliflozin; Group 2, SD rats administered vehicle (HP-β-CD at 4.5%); Group 3, Nx SD rats administered luseogliflozin; and Group 4, Nx SD rats administered vehicle. Luseogliflozin (0.3 mg kg −1 ) or vehicle at 0.1 ml was administered intravenously. Preliminary experiments showed that intravenous injection of luseogliflozin (0.3 mg kg −1 ) significantly increased urinary glucose excretion.
Protocol 2. SD and Nx SD rats were anesthetized with pentobarbital and isoflurane, and randomly assigned to one of four groups (n = 5-7 for each): Group 1, SD rats administered luseogliflozin; Group 2, SD rats administered vehicle; Group 3, Nx SD rats administered luseogliflozin; and Group 4, Nx SD rats administered vehicle. Luseogliflozin (0.9 mg kg −1 ) or vehicle at 0.1 ml was administered intraperitoneally. Preliminary studies showed that the minimum dose of HP-β-CD was 4.5% in 0.1 ml to dissolve 0.9 mg kg −1 luseogliflozin. Protocol 3. SD rats were anesthetized with inactin and assigned to one of two groups (n = 7 for each): Group 1, SD rats administered luseogliflozin; and Group 2, SD rats administered vehicle. Luseogliflozin (0.9 mg kg −1 ) or vehicle was administered at 0.1 ml intraperitoneally.

Statistical analysis.
Data are mean ± standard error of the mean (SEM). The statistical significance of difference was determined using two-way analysis of variance followed by Bonferroni post-tests using GraphPad Prism software ver. 5.0 (GraphPad, La Jolla, CA, USA). A P-value < 0.05 was considered statistically significant.