Paradigm Shift in Hyperglycemic Glomerular Hyperfiltration: Blunted Tubuloglomerular Feedback or Preglomerular Vasodilation?

Chronic kidney disease affects approximately 15% of US adults or 37 million people, resulting in $87 billion healthcare cost annually¹. Diabetes, obesity and hypertension are leading risk factors for chronic kidney disease, and glomerular hyperfiltration is a major pathophysiological mechanism contributing to proteinuria and glomerulosclerosis in these conditions. If left untreated, glomerular hyperfiltration is usually followed by rapid decline in glomerular filtration rate due to progressive nephron loss, ultimately leading to chronic kidney disease and end-stage renal disease². Pre-glomerular myogenic tone and tubuloglomerular feedback are the primary mechanisms of renal autoregulation that maintain constant renal blood flow and glomerular filtration rate in a wide range of renal perfusion pressure³. This autoregulatory capacity is compromised in diabetes and obesity such that even modest fluctuations in arterial pressure are transmitted to the glomeruli resulting in elevated glomerular capillary pressure, hyperfiltration, and glomerular damage. Blunted tubuloglomerular feedback was considered the chief cause of hyperglycemic glomerular hyperfiltration, but this paradigm may need to change. In this issue of Hypertension, Fei at al made an intriguing discovery that high glucose can increase glomerular filtration rate through direct vasodilatory effects in renal resistance vessels⁴.

Tubuloglomerular feedback is initiated when the macula densa senses an increased load of sodium chloride delivered to the distal nephron via the Na⁺-K⁺−2Cl⁻ co-transporter. This triggers macula densa cells to release adenosine triphosphate (ATP) which is hydrolyzed to adenosine by extracellular nucleases. ATP and adenosine activate afferent arteriolar P2X purinergic receptors and the A1 adenosine receptor respectively to stimulate preglomerular vasoconstriction³. These events lead to a decrease in glomerular capillary pressure and a subsequent reduction in the filtered sodium/chloride load. Dampened tubuloglomerular feedback was believed to be the primary cause of deleterious glomerular hyperfiltration in diabetes and obesity. Under hyperglycemia, augmented glucose reabsorption via sodium/glucose co-transporter (SGLT) 1 and 2 in the proximal tubule creates an electrical gradient that drives passive chloride re-uptake (Figure 1). The resultant decrease in chloride delivery to the macula densa leads to dampened tubuloglomerular feedback and weakened myogenic tone in the afferent arteriole. As a major breakthrough in the management of diabetic kidney disease, inhibition of SGLT2, which mediates 80–90% of glucose reabsorption in the proximal tubule, corrects glomerular hyperfiltration through improved tubuloglomerular feedback⁵.

Figure 1. Classical and new mechanisms of glomerular hyperfiltration during hyperglycemia.

The classical concept holds that in hyperglycemia increased sodium (Na⁺) reabsorption through SGLT1/2 and passive chloride (Cl⁻) re-uptake in the proximal tubule causes a decreased delivery of Cl⁻ to the macula densa. This blunts tubuloglomerular feedback (TGF) leading to weakened afferent arteriolar constriction and glomerular hyperfiltration. Red arrows denote pathological changes caused by hyperglycemia. The study by Fei et al supports a new paradigm where hyperglycemia and/or hyperosmolality directly stimulate vasodilation in renal microvessels through the Piezo1-CaMKII-eNOS pathway. In response to high glucose, there is greater vasodilation in the afferent arteriole (AA) than in the efferent arteriole (EA), resulting in elevated glomerular capillary pressure (P_GC) and augmented GFR. These hemodynamic derangements ultimately lead to proteinuria, glomerulosclerosis, and chronic kidney disease (CKD). EC, endothelial cell. CaMKII, Ca²⁺/Calmodulin-dependent protein kinase type II. SGLT, sodium-glucose co-transporter.

However, is impaired autoregulation in hyperglycemia solely caused by a blunted tubuloglomerular feedback response? This traditional concept is challenged by paradigm-shifting findings by Fei et al⁴. Using state-of-the-art microvascular perfusion and myography, the authors discovered that acute hyperglycemia induced profound afferent arteriolar vasodilation and glomerular hyperfiltration, as evidenced by markedly enhanced GFR, through a novel pathway involving Piezo1, Ca2⁺/Calmodulin-dependent protein kinase type II (CaMKII), and endothelial nitric oxide synthase (eNOS) (Figure 1).

Piezo1 is a mechanosensitive, Ca²⁺-permeable, non-selective cation channel that facilitates shear stress-induced eNOS activation and nitric oxide generation^{6, 7}. The Piezo1-CaMKII-eNOS pathway was supported by unequivocal pharmacological and molecular data obtained in perfused juxtaglomerular preparations, isolated renal microvessels, and cultured endothelial cells. Although Piezo1 is expressed in the vascular smooth muscle, it is dispensable for arterial myogenic tone. Thus, while smooth muscle actions cannot be ruled out, piezo1-dependent vasodilation is likely to be primarily mediated by eNOS as L-N^G-Nitro-arginine methyl ester completely abolished the dilatory effects. However, hyperglycemic afferent arteriolar dilation may not be entirely mechanosensitive as high glucose-induced nitric oxide generation and vasodilation is mediated, at least in part, by glucose transporter 1⁹, suggesting intracellular signaling triggered by glucose uptake may also contribute. Moreover, hyperosmotic mannitol had similar vasodilating effects as hyperglycemia that could also be prevented by Piezo1 inhibition, raising the question whether Piezo1 plays a role in osmosensing. Future genetic studies involving endothelium-specific deletion of Piezo1 and/or glucose transporter 1 are warranted to delineate the molecular pathways involved in hyperglycemic vasodilation in pre-glomerular resistance vessels.

The Fei et al study also revealed that hyperglycemia induced greater vasodilation in the afferent arteriole than efferent arteriole or vasa recta. This is consistent with early observations by Brenner and associates that afferent arteriolar resistance decreases more than the efferent arteriolar resistance in the ischemic renal mass ablation model, which may account for the single-nephron hyperfiltration and subsequent development of glomerulosclerosis and chronic kidney disease⁸. Calcium channel blockers (e.g. nifedipine) exacerbates glomerulosclerosis in the remnant kidney model despite significant blood-pressure lowering effects, further supporting a role of pathological vasodilation in the progression of chronic kidney disease¹⁰. These observations, along with convergent evidence from the laboratory of Liu and associates⁹, strongly support a pivotal role of pre-glomerular vasodilation in pathogenic glomerular hyperfiltration.

In comparison to diabetes, glomerular hyperfiltration in hypertension can occur without afferent arteriolar vasodilation. For example, we recently showed that female transgenic mice selectively expressing a human hypertension-causing mutation in Cullin3 ubiquitin ligase (Cul3Δ9, denoting lack of exon 9) in the endothelium developed profound salt-sensitive hypertension, glomerular hyperfiltration and microalbuminuria¹¹. The augmented glomerular filtration rate cannot be attributed to preglomerular vasodilation, and in fact, high salt-fed Cul3Δ9 transgenic mice exhibited a marked decline in endothelium-dependent vasodilation in renal microvessels due to impaired eNOS activation and diminished nitric oxide production. Indeed, impaired vasodilation in preglomerular resistance vessels is a hallmark of salt sensitive hypertension, and is often accompanied by elevated renovascular resistance and blunted renal blood flow^{12, 13}. These hemodynamic derangements are associated with increased glomerular capillary pressure and filtration fraction in salt-sensitive hypertension, causing a greater propensity to develop albuminuria and glomerulosclerosis¹⁴.

It is tempting to speculate whether the Piezo1 contributes to high salt-induced vasodilation, which is an adaptive response that offsets the pressor effect of volume expansion in salt resistant humans^{12, 13}. Whether the Piezo1 pathway is altered in salt-sensitive hypertension is opportunity for investigation.