Filtered glucose can alter the cellular metabolism of endothelial, glomerular, and proximal tubule (PT) cells and progressively contribute to impair kidney function [24,25].

At the kidney level, excessive glucose load, even if transitory, induces massive and long-lasting cellular modifications, like adaptive hypertrophy, glomerular hyperfiltration, rearrangement of metabolic pathways, and epigenomic changes in kidney cells [26,27,28]. For instance, proximal tubular (PT) cells in the presence of glycosuria require an abnormal amount of energy and oxygen consumption for ATP synthesis to sustain continuous glucose reabsorption into the bloodstream through sodium–glucose cotransporters (SGLTs) [29]. This altered energy-metabolism flux generates an oxidative burst that leads to mitochondrial dysfunction by triggering inflammatory and fibrotic pathways, which ultimately determines the decline of the glomerular filtration rate (GFR) and renal scarring [30,31,32].

At the molecular level, PT cells are forced by the excessive glucose availability to overwhelm its catabolism via the glycolytic and mitochondrial oxidative pathways, inducing the generation of damaging levels of reactive oxygen species (ROS) and depleting intracellular content of oxygen. Under normal conditions, PT cells utilize fatty acid oxidation (FAO) as the preferential source for producing ATP, and the impairment of oxidative phosphorylation (OXPHOS) downstream of FAO is thought to be one of the key drivers of tubular injury and fibrinogenesis [33,34]. Notably, positive regulation of FAO transcriptional regulators (e.g., PPARα and PGC-1α) and mediators (e.g., CPT1α) plays a protective role against severe cellular damage in PT cells [35,36,37]. Moreover, the uptake control of free fatty acids (FFAs) via the inhibition of the free fatty acid transporters CD36 or FATP2 in PT cells permits a reduction in the lipotoxicity produced following an excess of lipid accumulation in cells that are unable to efficiently perform FAO [35,38,39,40].

Hyperglycemia-induced ROS overproduction is responsible for a plethora of detrimental effects due to the alteration of the cellular oxidant–antioxidant balance. Indeed, a reduced ROS scavenging ability brings an accumulation of irreversibly modified cellular sugars, lipids, nucleic acids, and proteins that can stimulate secondary metabolisms like the polyol, hexosamine, AGE/RAGE, and protein kinase C (PKC) pathways, which are able to disrupt cellular homeostasis [41,42,43,44,45,46]. These activated pathways induce several downstream signaling events as the activation of an inflammatory response mediated by the nuclear factor-κB (NF-κB) and/or JAK/STAT3 pathways, thus mediating the release of inflammatory cytokines (e.g., TNFα), chemokines, adhesion molecules, and fibrinogenic factors (e.g., TGF-β1) [47,48,49,50]. Additionally, the activity of phosphoinositide 3-kinase (PI3K)/protein kinase B (Akt) exerts a tight control over cell proliferation, differentiation, and apoptotic processes by in turn regulating the glycogen synthase 3 β (GSK3β), mTOR (mammalian target of rapamycin), and Forkhead Box Protein O1 (FoxO1) signaling effectors [51,52,53,54]. Nonetheless, the oxidative stress-induced nuclear translocation of the nuclear factor erythroid-2-related factor 2 (Nrf2) transcription factor can promote the expression of antioxidant enzymes while counteracting NF-κB activation, which simultaneously ameliorates cellular defense against oxidative damage and dampens the resulting inflammatory response [55,56,57]. The combination of an augmented inflammatory response, which is responsible for the activation of immune cells and fibroblast-mediated collagen deposition, and a hypertrophic expansion of renal cells altogether contributes to tubular atrophy and glomerular hyperfiltration accompanied by persistent proteinuria and, in the end, loss of kidney functionality.

On the other hand, PT cells of patients with diabetic kidney disease (DKD) exhibit aberrant oxygen consumption that creates a cellular environment characterized by a low oxygen tension and ATP content. Under these conditions, oxygen and energy substrate deprivation can promote the activation of hypoxia-inducible factor 1 α (HIF1α) and of AMP-activated protein kinase (AMPK), which are responsive for oxygen and AMP levels, respectively. HIF-1α is transcription factor that can regulate the expression of the glycolytic enzymes hexokinase-1 (HK1) and phosphofructokinase (PFKL) to promote ATP production with anerobic glycolysis instead of aerobic mitochondrial OXPHOS [58]. AMPK is a modulator of mTOR, a serine/threonine kinase that mediates cell growth control based on the cellular nutrient status [59]. Both HIF-1α and AMPK pathways contribute to fuel the uncontrolled tissue growth and injury accumulation in the PT cells of diabetic individuals with CKD [59].

Therefore, the preservation of mitochondrial function is crucial for the prevention of kidney failure, and it could be achieved by limiting the glucose reuptake mediated by SGLT transporters. The use of SGLT inhibitors represents a potential therapeutic strategy to prevent the establishment of metabolic anomalies, oxidative damage, inflammatory bursts, and hypoxic environments, which are at the basis of renal fibrotic tissue formation and drive DKD progression/severity, at early stages.

Glucagon-like peptide-1 (GLP-1) is a peptide hormone that plays a critical role in glucose homeostasis and insulin secretion [60]. It is produced from the pro-glucagon gene (GCG), which is subjected to post-translation processing by prohormone convertase enzymes (PCSKs). PCSK1, expressed mainly in intestinal cells and in the brain, cleaves the pro-hormone in GLP-1, while PCSK2 produces a longer peptide that includes both GLP-1 and GLP-2 [60]. Therefore, the production of pro-glucagon peptides is regulated by the tissue-specific expression of proteases, and the intestine and brain are the main organs producing GLP-1 [60]. The effects of GLP-1 are mediated by the interaction the GLP-1 receptor (GLP-1R) that is expressed at high levels in the pancreas [61]. Activation of GLP-1R by GLP-1 induces pleiotropic effects on glucose metabolism, promoting insulin secretion and reducing blood glucose levels. In parallel, GLP-1R activation inhibits the release of glucagon from pancreatic alpha cells, helping the suppression of hepatic glucose production and contributing to the overall reduction in blood glucose levels [61]. The metabolic effects of GLP-1 are not restricted to glucose metabolism but may influence other physiological districts, like the cardiovascular system and the kidneys. Indeed, GLP-1 can stimulate vasodilation via the induction of nitric oxide (NO) production [62], decreasing platelet activation and inflammation and contributing to overall cardiovascular health [62]. In the kidneys, GLP-1 seems to have important functions related to renal protection. GLP-1R is expressed in the kidneys across different cell types [63], and its expression is lower in DKD patients [64]. The nephroprotective effect of GLP-1 does not seem to correlate with glucose lowering. In an animal model of diabetic nephropathy, GLP-1R agonists were able to reduce proteinuria and ameliorate glomerular filtration without modifying blood pressure or body weight [65]. In humans, GLP-1 infusion can increase sodium excretion and improve the glomerular filtration rate [66,67].

Aldosterone acts on Na⁺/K⁺-ATPase to promote sodium reabsorption, potassium excretion, and water retention, thus leading to hypertension, but it also mediates some untoward pro-inflammatory and pro-fibrotic effects in the kidneys [68]. It is known that in diabetes, hyperglycemia can increase Ang II production, as well as upregulate MR synthesis, thus further worsening tissue damage [69]. Notably, local aldosterone synthesis promotes renal fibrosis by stimulating fibroblast proliferation through growth factor activation, as well as fibronectin production, thus confirming how the hyperactivation of the RAAS system plays a key role on the pathogenesis of DKD.

The action of aldosterone depends on the activation of a nuclear receptor called the mineralocorticoid receptor (MR), and the expression of 11β-hydroxysteroid dehydrogenase type 2 in the cells of the distal nephron is crucial to confer specificity for aldosterone over glucocorticoids [68]. In the resting state, MR localizes in the cell cytosol, bound to heat-shock proteins. Upon aldosterone binding, it translocates to the nucleus, where it binds the hormone-responsive elements to initiate the transcription of target genes. Preclinical data indicate that the molecular effects of MR activation in kidney disease are related to (i) oxidative stress, due to increased ROS production through the induction of NADPH oxidases and/or decreased ROS detoxification through repression of glucose-6-phosphate dehydrogenase; (ii) inflammation, due to increased leucocyte infiltration, pro-inflammatory cytokine production, and adhesion molecule production; and (iii) promotion of fibrosis through the increased production of connective tissue growth factors, leading to the proliferation of fibroblasts and the deposition of the extracellular matrix [70].

All these finding have prompted the targeting of MR using MR antagonists (MRAs). The first MRA approved was spironolactone, but its reduced specificity led to the development of the more selective eplerenone, which attenuated renal fibrosis and reduced oxidative stress in preclinical studies [71]. Recently, to improve the efficacy and reduce the side effects of MRAs, new-generation MRAs with a non-steroidal structure have been developed. The most effective is finerenone, which behaves as an inverse agonist and inhibits cofactor recruitment to MRs in the ligand-free state [72].

Hypoxia-induced factor (HIF) is the master regulator of adaptative responses to reduced oxygen availability [73]. HIF is a heterodimeric protein complex generally composed of two proteins; the first one, HIF-1 alpha (HIF1α), is codified by the HIF1A gene, while the second subunit, HIF-1 beta (HIF1β), or aryl hydrocarbon receptor nuclear translocator protein, is codified by the ARNT gene [74] and is expressed constitutively. Upon heterodimerization, the complex has transcriptional activity and binds to hypoxia response elements (HREs) within the promoters of many different downstream targets, activating their transcription [73]. Major targets of the HIF complex are genes involved in angiogenesis, red blood cell development, glycolysis, oxidative stress, and other genes involved in oxygen sensing and metabolic conditioning to hypoxia [75].

The HIF complex is regulated in a variety of ways. A major pathway is the PHD–VHL–HIF axis, where, besides HIF, prolyl hydroxylase domain-containing proteins (PHDs) and Von Hippel–Lindau tumor suppressor (VHL or pVHL) are the key proteins regulating HIF activation/deactivation. In normoxia, HIF1α is hydroxylated by PHDs, and particularly by PHD2, an α-ketoglutarate/2-oxoglutarate-dependent dioxygenase that contains Fe²⁺ in the active site. Hydroxylated HIF can interact with VHL, acting as a substrate-recognition subunit of the Cullin-RING E3 ubiquitin ligase complex and therefore targeting HIF for proteasomal degradation [76]. In hypoxia, hydroxylation is inhibited, thereby decreasing HIF degradation and promoting the formation of the heterodimeric complex HIF1α/HIF1β that can activate transcription of target genes in the nucleus [73].

HIF is also regulated by another dioxygenase called factor-inhibiting HIF-1 (FIH1). FIH1 can hydroxylate an asparagine residue on HIF1α, preventing interaction with the p300 and CBP proteins which act as coactivators of the transcriptional heterodimeric complex, eventually inhibiting the transcription of hypoxia-related genes [77]. FIH1 has more affinity for O₂ when compared to members of the PHD family; therefore, it is still active when oxygen levels are very low [78]. Combined regulation by PHD proteins and FIH1 can provide a dynamic response to hypoxia within a wide range of oxygen tensions.

In the kidney, hypoxia represents one of the main consequences of fibrosis, since expansion of fibrotic tissue is associated with the loss of peritubular capillaries, reduced blood flow, and renal anemia [79,80]. In addition, kidneys show low oxygen tension when compared to other organs [81] and display an increase in oxygen demand according to several models mimicking different pathologies with renal involvement, such as diabetes and hypertension [82,83]. These studies underscore the dynamic regulation of oxygen demand in the kidney to supply enough energy for cellular activities such as transport, electrolyte regulation, and several other processes through the aerobic metabolism [84].

Three isoforms of HIFα are expressed in the kidney: HIF-1α, HIF-2α, and HIF-3α. Studies have mainly investigated the roles of HIF-1α and HIF-2α in the kidney. The two isoforms seem to have a different pattern of expression in kidney cells, where HIF-1α is enriched in tubular cells (TECs), while HIF-2α is mainly expressed in endothelial cells and interstitial fibroblasts [85]. However, recent single-cell atlas data have also shown relatively high expression of HIF-2α in TEC cells at the mRNA level [86]. Both isoforms have shown clear associations with renal fibrosis. In a model of cisplatin-induced renal fibrosis, HIF-1α was increased and promoted at the onset of the pathology, activating NOTCH-1 signaling [87]. Other studies have shown that HIF-1α is able to activate a series of pro-fibrotic pathways, including secretion of inflammatory mediators such as IL-1β and TNF-α, induction of epithelial–mesenchymal transition (EMT), and vascular remodeling [88]. HIF-2α is also prominently involved in renal fibrosis. In a unilateral ureteral obstruction (UUO) model of mice, knock-out of HIF-2α attenuated renal fibrosis through a mechanism involving sirtuin-1 (SIRT1) [89]. On the other hand, it has been shown that long-term activation of HIF-2α may alter the course of renal fibrosis, improving renal function and decreasing several EMT markers [90].

Xue Li and coworkers [91] showed that in a mouse model of tubular proximal cells, damage induced by folic acid (FA) and relative mitochondrial failure, treatment with the HIF stabilizer Roxadustat prior to FA administration reduced cell apoptosis, stabilized intercellular junctions, and promoted the expression of aquaporin 1 (AQP1) and aquaporin 2 (AQP2). Moreover, interstitial inflammatory cell infiltration after FA injection was reduced as well as collagen deposition. The concentration of HIF-1α in mice treated with Roxadustat was elevated until day 3 from the FA injection and then decreased, while in the control group, it remained elevated until day 7, the time at which the histologic signs of tubular toxicity were evaluated. In the latter group, persistently elevated levels of HIF-1α were related to mitochondria fragmentation and reduction in crest expression, a sign of metabolic suffering. As a consequence, the intracellular levels of ATP decreased in the control group, while the group treated with Roxadustat showed stable ATP levels and decreased ROS levels. A meaningful finding was that in the damaged tubular cells of the non-treated group, the mitochondrial pattern shifted from a fusion pattern to a fission pattern, as demonstrated by the hyperexpression of fission proteins such as FIS 1 e Drp 1 and the lower expression of fusion proteins such as Opa1 and Mfn1. In the Roxadustat group, the pattern shifted to a fusion pattern, ameliorating the aerobic metabolism [91].

Taken together, these studies show how regulation of HIF activation and/or expression may directly influence the development of pro-fibrotic pathways, even though contrasting results are often reported and may be associated to different outcomes for renal health [92].