SGLT2 inhibitors: cardiorenal metabolic drugs for the ages

Having trained as an endocrinologist and diabetologist, as well as a nephrologist interested in phosphate and glucose transport in the kidney (1), it was logical to combine the two disciplines. It also made sense to utilize phlorizin, a dual SGLT2/SGLT1 inhibitor, as a therapeutic option for the treatment of type 2 diabetes (T2D) based on the premise that the induction of glucosuria would reduce the plasma glucose concentration (Figure 1). Because reduced glucose levels ameliorate glucotoxicity, lead to reversal of insulin resistance in muscle (2), and improve β cell function (3), we set out to test phlorizin using partially (90%) pancreatectomized diabetic rats.

Figure 1

SGLT2 inhibitors acutely induce glucosuria, which persists as long as the SGLT2i therapy is continued. Treatment with SGLT2 inhibitors, such as phlorizin, promotes glucosuria and leads to a decline in fasting plasma glucose concentration, i.e., from 160 to 100 mg/dL. When the fasting plasma glucose declines toward normoglycemic levels, activation of the renal sympathetic nerves stimulates the liver to undergo glucogenesis and glycogenolysis, and quantitatively this increases hepatic glucose production (HGP), matching the glucose amount excreted in the urine, thus preventing hypoglycemia. The decline in plasma glucose concentration creates a state of tissue energy deprivation and leads to a decrease in plasma insulin and an increase in plasma glucagon concentration. These resultant hormonal changes stimulate lipolysis, thus providing substrate for ketogenesis, and reset the liver into the ketogenic mode. Subsequent ketone (β-hydroxybutyrate [β-OH-B]) production provides a fuel-efficient substrate for the heart and the kidney. In muscle, glycogen breakdown is stimulated, providing the carbon skeleton for the synthesis of alanine (Ala), which is released into the circulation and transported to the liver to support gluconeogenesis (known as the alanine cycle). Stimulation of muscle proteolysis also results in the release of alanine as well as other amino acids, in particular the branched-chain amino acids (BCAAs): leucine (Leu); isoleucine (Isoleu); and valine (Val). The liver does not possess BCAA transferase, allowing the BCAAs to be taken up by the heart and kidney where they provide substrate to compensate for the reduction in plasma glucose concentration that occurs secondary to the glucosuric effect of the SGLT2i. PYR, pyruvate.

In partially pancreatectomized diabetic rats, insulin sensitivity, assessed with the gold standard euglycemic insulin clamp in awake, unstressed, chronically catheterized rats, was reduced by 30%–40%. In diabetic animals treated with phlorizin by subcutaneous injection for four weeks, whole-body insulin sensitivity, which primarily reflects that of the muscle, was completely normalized. The treatment also normalized the fasting and postmeal plasma glucose concentrations (2, 3). Discontinuation of phlorizin in diabetic rats reverted them into a state of insulin resistance with a diabetic plasma glucose profile. These studies unequivocally established the important role of glucotoxicity in the development of muscle insulin resistance. Thus, they showed that an initial defect in insulin secretion leading to hyperglycemia can result in the development of skeletal muscle insulin resistance. The experiments revealed that phlorizin, a drug that inhibits renal tubular glucose absorption but has no other known metabolic effects, had the capacity to completely reverse insulin resistance and restore normal glucose tolerance. Notably, the induction of glycosuria, with normalization of the plasma glucose profile, could reverse glucotoxicity, and this finding provided definitive in vivo proof for the glucotoxicity hypothesis (2, 3).

To examine the cellular mechanisms via which glucotoxicity induced insulin resistance, we examined in vivo glucose transport and whole-body insulin sensitivity in diabetic rats (4). As previously demonstrated (2), partial pancreatectomy resulted in severe insulin resistance and a parallel 50% decrease in insulin-stimulated glucose transport in adipocytes (4). Levels of GLUT4, measured by immunoblotting, were decreased in adipose cell subcellular membrane fractions, as were the corresponding mRNA levels, as assessed by Northern blotting. Restoration of normoglycemia with phlorizin restored insulin-stimulated glucose transport in adipocytes and insulin-mediated glucose disposal in vivo. However, GLUT4 protein levels within low-density microsomes in the basal and insulin-stimulated state remained reduced by 40%–50%. These studies demonstrated for the first time that (a) changes in the ambient glucose plasma concentration, independent of ambient insulin, can regulate the glucose transport response to insulin without changes in GLUT4 translocation to the plasma membrane and (b) changes in intrinsic GLUT4 functional activity play an important role in the enhanced insulin sensitivity observed with the reversal of glucotoxicity (2, 4).

The effects of chronic hyperglycemia on in vivo insulin secretion were examined after rats underwent 90% pancreatectomy, leaving intact the 10% of the pancreas between the common bile duct and the duodenal sweep (3). Notably, insulin secretion expressed within the remaining percentage of pancreas should be normal unless some acquired defect in β cell function develops. The hyperglycemic clamp technique in awake, unstressed, 90% pancreatectomized diabetic rats revealed that first- and second-phase insulin secretion were severely impaired when expressed per residual pancreas. Furthermore, both phases of insulin secretion completely returned to normal and plasma-glucose concentration normalized following treatment with phlorizin. These results provided conclusive evidence that chronic hyperglycemia impairs insulin secretion and showed that the reversal of glucotoxicity with phlorizin can restore normal β cell function (3).

Based upon these three landmark studies (2–4) in rodents, we approached Bristol-Myers Squibb/AstraZeneca and developed a clinical investigation program to examine the effect of inhibition of renal tubular glucose transport on glucose homeostasis, insulin sensitivity, and insulin secretion in patients with T2D (5). The first of these human studies was published in the JCI in 2014. In this double-blind placebo-controlled study (5), patients with insulin-resistant T2D were randomized to treatment with dapagliflozin (10 mg/d) or placebo for 14 days. Dapagliflozin induced glucosuria (75–91 g/d), markedly lowered the fasting plasma glucose concentration, and increased insulin-stimulated whole-body glucose disposal using the euglycemic insulin clamp technique. This study was the first to document an increase in insulin sensitivity following improvement in glycemic control with SGLT2 inhibition. This study also was the first to demonstrate that SGLT2 inhibitor (SGLT2i) administration stimulated endogenous glucose production (EGP) in patients with T2D. Of note, the increase in EGP was associated with an increase in plasma glucagon concentration and a decline in plasma insulin concentration, and the increase in EGP offset by approximately 50% the amount of glucose excreted in the urine (5–7), thereby blunting the decrease in fasting plasma glucose concentration. In another paper published in the JCI, Ferrannini and colleagues (8) used a dual-tracer technique to demonstrated that empagliflozin reduced the fasting and postmeal plasma glucose concentration, augmented insulin sensitivity, and stimulated basal EGP. In follow-up studies in patients with T2D, we examined β cell function more intensively using the oral glucose tolerance test (OGTT) and euglycemic insulin clamp (9, 10). Fasting and post-OGTT plasma glucose concentrations declined by 33 and 73 mg/dL, respectively; insulin secretion (Δ C-Pep_0–120/Δ G_0–120) increased 2-fold; and β cell function (disposition index) (Δ C-Pep_0–120/Δ G_0–120/insulin resistance) rose 2.4-fold, where C-Pep is C-Peptide and G is glucose. These findings document the deleterious impact of glucotoxicity on β cell function in patients with T2D and the improvement in β cell function following amelioration of glucotoxicity by the glucosuric effect produced by SGLT2 inhibition. Of note, a similar beneficial effect of SGLT2i on β cell function was observed by Ferrannini et al. following a meal tolerance test (8).

To examine the mechanism via which SGLT2i stimulated basal EGP, individuals with T2D received dapagliflozin under conditions in which the plasma insulin, glucagon, and glucose concentrations were clamped at the fasting level alone or in combination via the pancreatic clamp technique (11). In the absence of any change in plasma insulin, glucagon, or glucose concentrations, SGLT2i still caused a stimulation of EGP. To further pursue the mechanism responsible for the SGLT2i-induced rise in EGP, studies were performed in individuals with end-stage polycystic kidney disease who had undergone a renal transplant and in whom both native kidneys had been removed (12). When dapagliflozin was administered to these individuals, the stimulation of EGP was markedly blunted, indicating an important role for the renal nerves in mediating the SGLT2i-induced stimulation of EGP. This observation is consistent with the rapid stimulation of EGP, within 15–30 minutes after administration of an SGLT2 inhibitor (13), and uncovers a previously unrecognized renal-hepatic axis. Moreover, this stimulatory effect of SGLT2i on EGP persisted for at least 4 months (13). Furthermore, in normal glucose-tolerant individuals the increase in EGP quantitatively matches the increase in urinary glucose production, such that the fasting plasma glucose remains constant at the basal level (14, 15). This finding explains why SGLT2is do not promote hypoglycemia. Although, at the time these studies were performed, it was unknown whether the site responsible for the increase in EGP was the liver or kidney, follow-up studies combining arterial cannulation with renal vein catheterization and 3-³H-glucose and PAH infusion conclusively demonstrated that all of the increase in EGP is derived from the liver (16).

Based upon the established role of deranged tubuloglomerular feedback leading to increased intraglomerular pressure and diabetic nephropathy (17), we predicted that inhibition of proximal sodium along with glucose reabsorption with SGLT2i would enhance sodium delivery to the macular densa cells of the juxtaglomerular apparatus, thereby reducing intraglomerular pressure and slowing the rate of decline in glomerular filtration rate in patients with established diabetic nephropathy (6, 18). Not surprisingly, this prediction was confirmed in multiple prospective studies (6, 19). The effect of SGLT2i therapy to prevent heart failure and reduce cardiovascular mortality in patients with diabetes and established cardiovascular disease and/or risk factors for cardiovascular disease was totally unpredictable (20).

Who would have thought that studies initiated to control glycemia by reversing glucotoxicity in diabetic rodent models (2–4) and in people with T2D (5) would have eventuated in (a) the discovery of a previously unrecognized renal-hepatic axis involved in the regulation of glucose metabolism (Figure 1); (b) a therapeutic intervention providing both cardiovascular and renal protection; and (c) sustained improvement in glycemic control in patients with T2D? These findings lend credence to the adage that “an apple per day keeps the cardiologist and nephrologist away.”

The authors would like to thank Deena Murphy for her expert assistance in preparing the manuscript for submission and Marnie Palacios for her artistic talent in conceptualizing the figure. This work was supported by NIH grants R01 DK107680-06; R01 DK129676; PATHO-PH2-SUB_24_23/OT2HL16184; 2R01 DK024092-3; S-GRD-1516-MW33/U01DK098246; DK097554; and 5R01DK240923.

Address correspondence to: Ralph A. DeFronzo, 7703 Floyd Curl Drive, San Antonio, Texas, 78284-7886, USA. Phone: 210.567.6691; Email: defronzo@uthscsa.edu.

Conflict of interest: RAD serves on the advisory boards of AstraZeneca, Novo Nordisk, Boehringer Ingelheim, Intarcia, Aardvark, Renalytix, CORCEPT, and Alnylam. RAD has research grants from Boehringer Ingelheim, AstraZeneca, 89Bio, Amgen, Medality, and CORCEPT. RAD has served on the speakers bureau for AstraZeneca, CORCEPT, and Renalytix.

Copyright: © 2024, DeFronzo et al. This is an open access article published under the terms of the Creative Commons Attribution 4.0 International License.

Reference information:J Clin Invest. 2024;134(5):e177625. https://doi.org/10.1172/JCI177625.

1. DeFronzo RA, et al. The effects of glucose and insulin on renal electrolyte transport. J Clin Invest. 1976;58(1):83–90.
   View this article via: JCI CrossRef PubMed Google Scholar
2. Rossetti L, et al. Correction of hyperglycemia with phlorizin normalizes tissue sensitivity to insulin in diabetic rats. J Clin Invest. 1987;79(5):1510–1515.
   View this article via: JCI CrossRef PubMed Google Scholar
3. Rossetti L, et al. Effect of chronic hyperglycemia on in vivo insulin secretion in partially pancreatectomized rats. J Clin Invest. 1987;80(4):1037–1044.
   View this article via: JCI CrossRef PubMed Google Scholar
4. Kahn B, et al. Normalization of blood glucose in diabetic rats with phlorizin treatment reverses insulin resistant glucose transport in adipose cells without restoring glucose transporter gene expression. J Clin Invest. 1991;87(2):561–570.
   View this article via: JCI CrossRef PubMed Google Scholar
5. Merovci A, et al. Dapagliflozin improves muscle insulin sensitivity but enhances endogenous glucose production. J Clin Invest. 2014;124(2):509–514.
   View this article via: JCI CrossRef PubMed Google Scholar
6. Abdul-Ghani MA, et al. Renal sodium-glucose cotransporter inhibition in the management of type 2 diabetes mellitus. Am J Physiol Renal Physiol. 2015;309(11):F889–F900.
   View this article via: CrossRef PubMed Google Scholar
7. DeFronzo RA, et al. Renal, metabolic and cardiovascular considerations of SGLT2 inhibition. Nat Rev Nephrol. 2017;13(1):11–26.
   View this article via: CrossRef PubMed Google Scholar
8. Ferrannini E, et al. Metabolic response to sodium-glucose cotransporter 2 inhibition in type 2 diabetic patients. J Clin Invest. 2014;124(2):499–508.
   View this article via: JCI CrossRef PubMed Google Scholar
9. Merovci A, et al. Dapagliflozin lowers plasma glucose concentration and improves β cell function. J Clin Endocrinol Metab. 2015;100(5):1927–1932.
   View this article via: CrossRef PubMed Google Scholar
10. Merovci A, et al. Effect of dapagliflozin with and without acipimox on insulin sensitivity and insulin secretion in T2DM males. J Clin Endocrinol Metab. 2016;101(3):1249–1256.
    View this article via: CrossRef PubMed Google Scholar
11. Alatrach M, et al. Evidence against an important role of plasma insulin and glucagon concentrations in the increase in EGP caused by SLGT 2 inhibitors. Diabetes. 2020;69(4):681–688.
    View this article via: CrossRef PubMed Google Scholar
12. Daniele G, et al. Increase in endogenous glucose production with SGLT inhibition is attenuated in individuals who underwent kidney transplantation and bilateral native nephrectomy. Diabetologia. 2020;63(11):2423–2433.
    View this article via: CrossRef PubMed Google Scholar
13. Eugenio Cersosimo MD, et al. Emergence of new gluco-regulatory mechanism for glycemic control with dapagliflozin/exenatide therapy in type 2 diabetes. J Clin Endocrinol Metab. 2023;109(1):161–170.
    View this article via: CrossRef PubMed Google Scholar
14. Al-Jobori H, et al. Empagliflozin treatment is associated with improved beta cell function in T2DM. J Clin Endocrinol Metab. 2018;103(4):1402–1407.
    View this article via: CrossRef PubMed Google Scholar
15. Abdul-Ghani M, et al. Efficacy of exenatide plus pioglitazone versus basal/bolus insulin in T2DM patients with very high HbA1c. J Clin Endocrinol Metab. 2017;102(7):2162–2170.
    View this article via: CrossRef PubMed Google Scholar
16. Chen X, et al. Effect of dapagliflozin on renal and hepatic glucose kinetics in t2dm and NGT subjects. Diabetes. 2022;71(suppl_1):55-LB.
    View this article via: CrossRef Google Scholar
17. Hostetter TH, et al. The case for intrarenal hypertension in the initiation and progression of diabetic and other glomerulopathies. Am J Med. 1982;72(3):375–380.
    View this article via: CrossRef PubMed Google Scholar
18. DeFronzo RA, et al. Pathophysiology of diabetic kidney disease: impact of SGLT2 inhibitors. Nat Rev Nephrol. 2021;17(5):319–334.
    View this article via: CrossRef PubMed Google Scholar
19. Nuffield Department of Population Health Renal Studies Group; SGLT2 inhibitor Meta-Analysis Cardio-Renal Trialists’ ConsortiumImpact of diabetes on the effects of sodium glucose co-transporter-2 inhibitors on kidney outcomes: collaborative meta-analysis of large placebo-controlled trials. Lancet. 2022;400(10365):1788–1801.
    View this article via: CrossRef PubMed Google Scholar
20. McGuire DK, et al. Association of SGLT2 inhibitors with cardiovascular and kidney outcomes in patients with type 2 diabetes: a meta-analysis. JAMA Cardiol. 2021;6(2):148–158.
    View this article via: CrossRef PubMed Google Scholar