Discussion

Recently, several new SGLT2 inhibitors that block glucose reabsorption in the proximal tubule have been developed for the treatment of diabetes. One of the compounds in this class is luseogliflozin. It has an IC₅₀ value of 2.3 nmol/l for inhibition of SGLT2 and exhibits a 1765-fold selectivity for inhibition of SGLT2 over SGLT1 (Yamamoto et al., 2011). It has also been reported to effectively control hyperglycemia in both type 1 and 2 rat models of diabetes (Yamamoto et al., 2011). However, the potential long-term effects of luseogliflozin, or any other SGLT2 inhibitor, on renal hemodynamics and the progression of renal injury in animal models of diabetes have not been determined. Thus, the present study examined the effects of luseogliflozin on the development of renal injury in the T2DN rat model of type 2 diabetic nephropathy.

Plasma concentrations of luseogliflozin averaged approximately 50 ng/ml in the rats treated with luseogliflozin. This corresponds to an unbound plasma concentration of 5 nmol/l (95% protein bound), which exceeds the IC₅₀ by twofold. Chronic treatment of the rats with luseogliflozin reduced body weight by 10% although food intake increased. This likely reflects the large increase in glucose excretion and the associated loss of calories.

Luseogliflozin given alone, or in combination with lisinopril, produced an effective long-term control of blood glucose and HbA1c levels in T2DN rats. The antihyperglycemic actions of luseogliflozin were equivalent to that seen in a group of rats that received a slow-release insulin implant. However, the insulin implant increased plasma insulin levels by 10-fold, whereas luseogliflozin reduced plasma insulin levels. This likely reflects the effective control of hyperglycemia in the luseogliflozin-treated animals that lessen the stimulus for insulin secretion. Part of the very high insulin levels in the rats given the Linplant implant may be related to the fact that human insulin delivered by this implant crossreacts more strongly (167%) with the antibody used in the insulin assay. Nevertheless, the insulin levels were elevated at least fivefold in the rats receiving the Linplant implant.

The sustained increase in glucose excretion seen in the rats treated with luseogliflozin is unexpected. One would assume that blockade of the SGLT2 would increase glucose excretion in hyperglycemic T2DN rats but as the degree of hyperglycemia was normalized glucose excretion would decrease. Although the exact reason for the sustained increase in glucose excretion remains to be determined, one might speculate that perhaps the SGLT2 inhibitor facilitates the rapid excretion of the glucose after each meal and that this leads to the sustained increase in glucose excretion.

Consistent with the results of previous studies (Nobrega et al., 2004; Williams et al., 2011), the T2DN rats exhibited a progressive decline in GFR and increased proteinuria and developed severe glomerulosclerosis, renal interstitial fibrosis, and the formation of protein casts— an index of tubular necrosis. Control of hyperglycemia with luseogliflozin prevented the fall in GFR and significantly reduced the degree of glomerular injury, renal interstitial fibrosis, and tubular necrosis in comparison with the degree of injury observed in T2DN rats treated with vehicle. However, luseogliflozin produced a transient increase in proteinuria. We do not think that the transient increase in proteinuria in rats treated with luseogliflozin is because of an increase in the filtered load of protein because recent studies have indicated that SGLT2 inhibitors activate tubuloglomerular feedback response and reduce rather than increase GFR in diabetic animals (Thomson et al., 2012). One possibility that remains to be determined is that blockade of the sodium and glucose reabsorption in the proximal tubule may reduce reuptake of filtered protein secondary to an increase in flow rate reduced contact time. This could lead to an increase in protein excretion in the absence of glomerular injury and fit with the present observations that luseogliflozin significantly reduced the degree of glomerular injury and renal fibrosis.

Interestingly, the degree of renal protection produced by luseogliflozin was greater than that seen in rats treated with insulin. Although insulin effectively controlled hyperglycemia, it did not reduce proteinuria or prevent the fall in GFR in T2DN rats. More importantly, it had no effect on the degree of glomerular injury and renal interstitial fibrosis or the formation of protein casts. The reason for the failure of insulin to reduce renal injury in this model remains to be determined. It may be attributable the fact that we studied 1-year-old animals, and perhaps controlling blood glucose levels in these older T2DN rats with preexisting renal injury cannot slow the progression of the disease. The other possibility is that the plasma insulin levels had to be elevated approximately 10-fold to control the hyperglycemia in the T2DN rats that are insulin resistant. In this regard, elevated levels of insulin have been reported to increase the expression of TGF-β1 and the production of matrix in cultured mesangial cells (Anderson et al., 1996) and exacerbate glomerular and renal interstitial fibrosis in diabetic animals (Sarafidis and Ruilope, 2006). In contrast to the results obtained in the insulin-treated animals, controlling hyperglycemia with luseogliflozin was renoprotective. Again the reason for the difference remains to be determined. One possibility is that luseogliflozin lowers plasma insulin levels, and this may reduce the proliferative actions of insulin on the kidney. In addition, chronic inhibition SGLT2 with luseogliflozin reduces proximal reabsorption of sodium and shifts the burden for salt balance from the early proximal tubule to the distal nephron (Thomson et al., 2012). This may activate tubuloglomerular feedback to mitigate hyperfiltration, and the reduction in glomerular capillary pressure might reduce the development of glomerular injury. This idea fits with the present finding that chronic treatment of luseogliflozin reduced the degree of glomerular sclerosis and fibrosis and prevented the fall in GFR seen in T2DN rats over the course of the study. Moreover, elevated glucose concentrations have been reported to increase oxidative stress and stimulate the synthesis of collagen and fibronectin by proximal tubular cells (Ziyadeh et al., 1990; Phillips et al., 1997; Jones et al., 1999). Thus, it may be possible that chronic inhibition of glucose reuptake with luseogliflozin may suppress the progression of tubular injury and renal interstitial fibrosis in diabetic animals.

We also compared the degree of renal protection produced by luseogliflozin with that seen in rats treated with the ACE inhibitor lisinopril, which is the standard of care to reduce renal injury in diabetic patients. Long-term blockade of the renin angiotensin system with lisinopril reduced blood pressure in our T2DN rats. This fits with the idea that the renin angiotensin system is activated in diabetic animals secondary to volume depletion secondary to glucosuria. Lisinopril also reduced proteinuria and increased RBF and GFR relative to vehicle-treated animals. It also reduced the degree of glomerular injury and the formation of protein casts, an index of tubular necrosis, but it had no effect on the degree of renal interstitial fibrosis. The mechanisms of the renoprotective effects of blocking the renin angiotensin system in the T2DN rats remain to be determined. Part of the effect is likely attributable to the fall in systemic pressure and a reduction in efferent arteriolar tone, because we found that filtration fraction was reduced. Both of these effects lower glomerular capillary pressure and the filtration of protein. Blockade of the renin angiotensin system also reduces oxidative stress and the stimulatory effects of angiotensin II on the expression of a variety of growth factors and cytokines.

Blood glucose and HbA1c levels were effectively controlled and blood pressure was reduced in the rats that received combination therapy more than that seen in animals treated with either agent alone. The greater fall in blood pressure likely reflects combined effects of the volume loss associated with the diuretic effect of luseogliflozin combined with blockade of the renin angiotensin system with lisinopril. Moreover, the degree of glomerular injury, renal fibrosis, and the formation of protein casts was lower in the rats receiving combination therapy than that seen in the animals treated with luseogliflozin or lisinopril alone. This probably is attributable to the effective dual control of both hyperglycemia and blood pressure in the animals receiving combination therapy.

Urinary nephrin excretion has been reported to increase and the glomerular expression of nephrin decrease in diabetic animals (Bonnet et al., 2001; Saleh et al., 2011). Chronic treatment with luseogliflozin or lisinopril reduced urinary nephrin excretion in the present study. This indicates that luseogliflozin and lisinopril may reduce damage to the podocytes and the loss of glomerular filtration barrier function in T2DN rats.

These results indicate that chronic administration of a selective SGLT2 inhibitor luseogliflozin and the ACE inhibitor lisinopril slow the progression of renal injury in a rat model of type 2 diabetes. Combination therapy of luseogliflozin and lisinopril is more effective at lowering blood pressure and is more renoprotective than administration of either agent alone. These results suggest that SGLT2 inhibitors are an attractive therapeutic option for the treatment of diabetes and when used in combination with inhibitors of the renin angiotensin system may effectively reduce the renal complications associated with diabetes.