Purpose of review

The purpose of this review is to correlate predictions based on pre-clinical data with outcomes from clinical trials that examine the effects of incretin-based diabetes treatments on the kidney. The incretin-based treatments include agonists of the glucagon-like peptide 1 receptor (GLP-1R) and inhibitors of the enzyme, dipeptidyl peptidase-4 (DPP-4). In addition, what is known about the incretin-based therapies will be compared to what is known about the renal effects of SGLT2 inhibitors.

Recent findings

Large-scale clinical trials have shown that SGLT2 inhibitors reduce albuminuria and preserve estimated glomerular filtration rate (eGFR) in patients with diabetic nephropathy. A concise and plausible hemodynamic mechanism is supported by pre-clinical research on the physiology and pharmacology of SGLT2. Large-scale clinical trials have shown that incretin-based therapies mitigate albuminuria but have not shown beneficial effects on eGFR. Research on the incretin-based therapies has yielded a diverse array of direct effects throughout the body, which fuels speculation as to how these drugs might benefit the diabetic kidney and affect its function(s). But in vivo experiments have yet to confirm that the proposed mechanisms underlying emergent phenomena, such as proximal tubular fluid reabsorption, are the ones predicted by cell and molecular experiments. There may be salutary effects of incretin-based treatments on the diabetic kidney, but the system is complex and not amenable to simple explanation or prior prediction. This contrasts with the renal effects of SGLT2 inhibitors, which can be explained concisely.

Healthy kidneys filter ∼160 g/day of glucose (∼30% of daily energy intake) under euglycaemic conditions. To prevent valuable energy from being lost in the urine, the proximal tubule avidly reabsorbs filtered glucose up to a limit of ∼450 g/day. When blood glucose levels increase to the point that the filtered load exceeds this limit, the surplus is excreted in the urine. Thus, the kidney provides a safety valve that can prevent extreme hyperglycaemia as long as glomerular filtration is maintained. Most of the capacity for renal glucose reabsorption is provided by sodium glucose cotransporter (SGLT) 2 in the early proximal tubule. In the absence or with inhibition of SGLT2, the renal reabsorptive capacity for glucose declines to ∼80 g/day (the residual capacity of SGLT1), i.e. the safety valve opens at a lower threshold, which makes it relevant to glucose homeostasis from day-to-day. Several SGLT2 inhibitors are now approved glucose lowering agents for individuals with type 2 diabetes and preserved kidney function. By inducing glucosuria, these drugs improve glycaemic control in all stages of type 2 diabetes, while their risk of causing hypoglycaemia is low because they naturally stop working when the filtered glucose load falls below ∼80 g/day and they do not otherwise interfere with metabolic counterregulation. Through glucosuria, SGLT2 inhibitors reduce body weight and body fat, and shift substrate utilisation from carbohydrates to lipids and, possibly, ketone bodies. Because SGLT2 reabsorbs sodium along with glucose, SGLT2 blockers are natriuretic and antihypertensive. Also, because they work in the proximal tubule, SGLT2 inhibitors increase delivery of fluid and electrolytes to the macula densa, thereby activating tubuloglomerular feedback and increasing tubular back pressure. This mitigates glomerular hyperfiltration, reduces the kidney's demand for oxygen and lessens albuminuria. For reasons that are less well understood, SGLT2 inhibitors are also uricosuric. These pleiotropic effects of SGLT2 inhibitors are likely to have contributed to the results of the EMPA-REG OUTCOME trial in which the SGLT2 inhibitor, empagliflozin, slowed the progression of chronic kidney disease and reduced major adverse cardiovascular events in high-risk individuals with type 2 diabetes. This review discusses the role of SGLT2 in the physiology and pathophysiology of renal glucose reabsorption and outlines the unexpected logic of inhibiting SGLT2 in the diabetic kidney.

Sodium-glucose co-transporter 2 (SGLT2) inhibitors immediately reduce the glomerular filtration rate (GFR) in patients with type 2 diabetes mellitus. When given chronically, they confer benefit by markedly slowing the rate at which chronic kidney disease progresses and are the first agents to do so since the advent of angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs). Salutary effects on the kidney were first demonstrated in cardiovascular outcomes trials and have now emerged from trials enriched in subjects with type 2 diabetes mellitus and chronic kidney disease. A simple model that unifies the immediate and long-term effects of SGLT2 inhibitors on kidney function is based on the assumption that diabetic hyperfiltration puts the kidney at long-term risk and evidence that hyperfiltration is an immediate response to a reduced signal for tubuloglomerular feedback, which occurs to the extent that SGLT2 activity mediates a primary increase in sodium and fluid reabsorption by the proximal tubule. This model will likely continue to serve as a useful description accounting for the beneficial effect of SGLT2 inhibitors on the diabetic kidney, similar to the hemodynamic explanation for the benefit of ACEIs and ARBs. A more complex model will be required to incorporate positive interactions between SGLT2 and sodium-hydrogen exchanger 3 in the proximal tubule and between sodium-glucose co-transporter 1 (SGLT1) and nitric oxide synthase in the macula densa. The implication of these latter nuances for day-to-day clinical medicine remains to be determined.

Kidney size and glomerular filtration rate (GFR) often increase with the onset of diabetes, and elevated GFR is a risk factor for the development of diabetic kidney disease. Hyperfiltration mainly occurs in response to signals passed from the tubule to the glomerulus: high levels of glucose in the glomerular filtrate drive increased reabsorption of glucose and sodium by the sodium-glucose cotransporters SGLT2 and SGLT1 in the proximal tubule. Passive reabsorption of chloride and water also increases. The overall capacity for proximal reabsorption is augmented by growth of the proximal tubule, which (alongside sodium-glucose cotransport) further limits urinary glucose loss. Hyperreabsorption of sodium and chloride induces tubuloglomerular feedback from the macula densa to increase GFR. In addition, sodium-glucose cotransport by SGLT1 on macula densa cells triggers the production of nitric oxide, which also contributes to glomerular hyperfiltration. Although hyperfiltration restores sodium and chloride excretion it imposes added physical stress on the filtration barrier and increases the oxygen demand to drive reabsorption. Tubular growth is associated with the development of a senescence-like molecular signature that sets the stage for inflammation and fibrosis. SGLT2 inhibitors attenuate the proximal reabsorption of sodium and glucose, normalize tubuloglomerular feedback signals and mitigate hyperfiltration. This tubule-centred model of diabetic kidney physiology predicts the salutary effect of SGLT2 inhibitors on hard renal outcomes, as shown in large-scale clinical trials.

A number of oxylipins have been described as endogenous PPAR ligands. The very short biological half-lives of oxylipins suggest roles as autocrine or paracrine signaling molecules. While coronary arterial atherosclerosis is the root of myocardial infarction, aortic atherosclerotic plaque formation is a common readout of in vivo atherosclerosis studies in mice. Improved understanding of the compartmentalized sources of oxylipin PPAR ligands will increase our knowledge of the roles of PPAR signaling in diverse vascular tissues. Here, we performed a targeted lipidomic analysis of ex vivo-generated oxylipins from porcine aorta, coronary artery, pulmonary artery and perivascular adipose. Cyclooxygenase (COX)-derived prostanoids were the most abundant detectable oxylipin from all tissues. By contrast, the coronary artery produced significantly higher levels of oxylipins from CYP450 pathways than other tissues. The TLR4 ligand LPS induced prostanoid formation in all vascular tissue tested. The 11-HETE, 15-HETE, and 9-HODE were also induced by LPS from the aorta and pulmonary artery but not coronary artery. Epoxy fatty acid (EpFA) formation was largely unaffected by LPS. The pig CYP2J homologue CYP2J34 was expressed in porcine vascular tissue and primary coronary artery smooth muscle cells (pCASMCs) in culture. Treatment of pCASMCs with LPS induced a robust profile of pro-inflammatory target genes: TNFα, ICAM-1, VCAM-1, MCP-1 and CD40L. The soluble epoxide hydrolase inhibitor TPPU, which prevents the breakdown of endogenous CYP-derived EpFAs, significantly suppressed LPS-induced inflammatory target genes. In conclusion, PPAR-activating oxylipins are produced and regulated in a vascular site-specific manner. The CYP450 pathway is highly active in the coronary artery and capable of providing anti-inflammatory oxylipins that prevent processes of inflammatory vascular disease progression.

Vascular pathologies are associated with changes in the presence and expression of morphologically distinct vascular smooth muscle cells. In particular, in complex human vascular lesions and models of disease in pigs and rodents, an intimal smooth muscle cell (iSMC) which exhibits a stable epithelioid or rhomboid phenotype in culture is often found to be present in high numbers, and may represent the reemergence of a distinct developmental vascular smooth muscle cell phenotype. The CYP450-oxylipin - soluble epoxide hydrolase (sEH) pathway is currently of great interest in targeting for cardiovascular disease. sEH inhibitors limit the development of hypertension, diabetes, atherosclerosis and aneurysm formation in animal models. We have investigated the expression of CYP450-oxylipin-sEH pathway enzymes and their metabolites in paired intimal (iSMC) and medial (mSMC) cells isolated from rat aorta. iSMC basally released significantly larger amounts of epoxy-oxylipin CYP450 products from eicosapentaenoic acid > docosahexaenoic acid > arachidonic acid > linoleic acid, and expressed higher levels of CYP2C12, CYP2B1, but not CYP2J mRNA compared to mSMC. When stimulated with the pro-inflammatory TLR4 ligand LPS, epoxy-oxylipin production did not change greatly in iSMC. In contrast, LPS induced epoxy-oxylipin products in mSMC and induced CYP2J4. iSMC and mSMC express sEH which metabolizes primary epoxy-oxylipins to their dihydroxy-counterparts. The sEH inhibitors TPPU or AUDA inhibited LPS-induced NFκB activation and iNOS induction in mSMC, but had no effect on NFκB nuclear localization or inducible nitric oxide synthase in iSMC; effects which were recapitulated in part by addition of authentic epoxy-oxylipins. iSMCs are a rich source but not a sensor of anti-inflammatory epoxy-oxylipins. Complex lesions that contain high levels of iSMCs may be more resistant to the protective effects of sEH inhibitors.

Relative abundance of fibroblast growth factor-23 (FGF23) measured by the C-terminal (cFGF23, which measures both intact FGF23 and C-terminal fragments) versus intact (iFGF23, measures only intact hormone) assays varies by kidney function in humans. Differential kidney clearance may explain this finding. We measured cFGF23 and iFGF23 in the aorta and bilateral renal veins of 162 patients with essential hypertension undergoing renal angiography. Using multivariable linear regression, we examined factors associated with aorta to renal vein reduction of FGF23 using both assays. Similar parameters and with addition of urine concentrations of cFGF23 and iFGF23 were measured in six Wistar rats. Mean ± standard deviation (SD) age was 54 ± 12 years, 54% were women, and mean creatinine clearance was 72 ± 48 mL/min/100 g. The human kidney reduced the concentrations of both cFGF23 (16% ± 12%) and iFGF23 (21% ± 16%), but reduction was higher for iFGF23. Greater kidney creatinine and PTH reductions were each independently associated with greater reductions of both cFGF23 and iFGF23. The greater kidney reduction of iFGF23 compared to cFGF23 appeared stable and consistent across the range of creatinine clearance evaluated. Kidney clearance was similar, and urine concentrations of both assays were low in the rat models, suggesting kidney metabolism of both cFGF23 and iFGF23. Renal reduction of iFGF23 is higher than that of creatinine and cFGF23. Our data suggest that FGF23 is metabolized by the kidney. However, the major cell types involved in metabolization of FGF23 requires future study. Kidney clearance of FGF23 does not explain differences in C-terminal and intact moieties across the range of kidney function. © 2022 American Society for Bone and Mineral Research (ASBMR).