Obesity-Related CKD: When Kidneys Get the Munchies

A brief walk through a typical renal outpatient clinic or dialysis unit may give rise to a suspicion of a link between society’s dual public health problems of obesity and CKD. Indeed, epidemiologic studies have consistently found obesity to be an important risk factor for CKD, and it remains significant even after correcting for other obvious coassociations, such as hypertension and diabetes.^1,2 This effect of obesity in increasing the risk of CKD is seen in all age groups and implicated across the full spectrum of CKD spanning from early CKD to ESRD. Furthermore, obesity accelerates renal injury caused by other diseases, such as GN (Figure 1).

Figure 1.: eCB pathways involved in obesity-/HFD-associated CKD. Obesity/HFD upregulates the eCB tone via CB₁R and CB₂R found in the brain/central nervous system (CNS), peripheral tissues, and various cells of the kidneys. In the renal PTCs, obesity-induced lipotoxicity is modulated by the CB₁R-coupled α-subunit of heterotrimeric Gi/o protein (Gα_i/o)-protein kinase A (PKA) axis, which mediates the downstream activation of the liver kinase B1 (LKB1)/AMPK/ACC signaling pathway; this decreases fatty acid β-oxidation and increases inflammation and fibrosis, resulting in CKD. The CB₁R in the peripheral tissues increases hepatic de novo fatty acid synthesis and obesity-induced metabolic and hormonal abnormalities, whereas the CB₁R in the brain increases appetite, leptin, obesity, and insulin resistance, which contribute to the development of CKD. Although the CB₂R is found in all of these tissues, its role is still unclear. The pathway discussed by Udi et al. ⁸ in this issue of the Journal of the American Society of Nephrology has been highlighted in yellow. ACC, acetyl-CoA carboxylase; AMPK, AMP-activated protein kinase; eCB, endocannabinoid.

Despite the epidemiologic links between obesity and CKD, obesity is rarely considered as a distinct cause of CKD in humans. This is partly due to the lack of a generally identifiably pattern of injury linking obesity with CKD. Histologically, the pattern that is generally ascribed to obesity is that of obesity-related glomerulopathy (ORG), which consists predominantly of focal glomerulosclerosis and podocytopathy.³ However, ORG is mainly associated with morbid obesity, and although an important entity, the number of patients with diagnosed cases is low and clearly insufficient to explain the broader epidemiologic link between obesity and CKD. This holds true, even when accounting for the fact that ORG is likely underdiagnosed: obese patients with CKD are rarely biopsied, because they usually have bland urinary sediment and subnephrotic proteinuria and are more technically challenging to biopsy. This lack of biopsy data also makes it difficult to determine whether there are other pathologic findings due to the effects of obesity on the kidney, particularly on the tubulointerstitial structures. Interestingly, another morphologic link between fat and the kidney has been found in the computed tomography studies of the Framingham cohort, in which renal sinus fat was identified as a predictor of CKD progression.⁴

An important challenge for investigating the relationship between CKD and obesity is the development of appropriate experimental models of obesity-induced CKD. Several genetic models of obesity have provided us with fundamental information regarding mechanisms by which obesity predisposes to CKD. However, the human obesity epidemic was predominantly fueled by the industrialization of the world’s food systems, which led to a parallel increase in food consumption and a change in the composition of our diet (increased fat and refined carbohydrates).⁵ Consequently, experimental models should ideally mimic the dietary and environmental factors (decreased physical activity) that most commonly contribute to human obesity as well as genetic predispositions. To this end, investigators have used high fat and caloric intake models of obesity-induced CKD. For instance, Deji et al.⁶ found that 12 weeks of a high-fat diet (HFD) in mice resulted in a spectrum of renal changes, including albuminuria, mesangial expansion, renal lipid accumulation, macrophage infiltration, and sodium retention. However, it is important to note that the observed renal injury was relatively mild; the albuminuria was modest, fibrosis was mild, and GFR was preserved. Similar findings have subsequently been reproduced by other investigators.^7,8 Although the mild phenotype may be considered a limitation of the model, two considerations suggest the alternative perspective that it more closely resembles human disease and thus, may be clinically relevant. The first consideration is that, although it may appear that the direct effects of obesity on the kidney are mild, the timeframe of the intervention in these models is not sufficient to fully reflect the effects of human obesity, which can start to develop even in childhood, with the final harmful effects becoming clinically evident decades later. The second consideration is that it may be that the direct harmful effects of obesity on the kidney are mild in the absence of other factors but amplified by the presence of a second insult. Hence, this may be clinically very significant, because obesity is commonly associated with other risks factors, such as diabetes and hypertension, and lends credence to the epidemiologic associations observed between obesity and the increased risk of progression of CKD due to other renal diseases.

The potential global effect of obesity on CKD makes it of utmost importance to identify the mechanisms by which it contributes to CKD progression, so that specific therapeutic approaches can be explored. In this context, obesity leads to a spectrum of systemic metabolic changes that may be harmful to the kidney. Although the most obvious of these are hypertension and hyperglycemia, aberrant neuroendocrine signaling, such as sympathetic activation and altered adipokine levels, may also be important.⁹ A possible mechanism by which obesity may induce CKD was first proposed by Moorhead et al.,¹⁰ who put forward the lipotoxicity hypothesis. They surmised that, because obesity and other CKD models have been associated with increased renal lipid accumulation, particularly in the proximal tubular cells (PTCs), podocytes, and mesangial cells, the renal lipid accumulation may initiate a series of self-perpetuating cascades of secondary events, which can eventually result in progressive glomerular and tubulointerstitial disease.¹⁰ However, although evidence linking renal accumulation of lipids with CKD is strong, the concept that lipotoxicity triggers renal fibrosis has been disputed. Kang et al.¹¹ found that kidney-specific overexpression of the long-chain fatty acid transporter CD36 caused marked intrarenal lipid (triglyceride and long-chain fatty acid) accumulation, without inducing fibrotic signaling or fibrosis. This prompted an alternative hypothesis that proposes that renal lipid accumulation in CKD may be an epiphenomenon secondary to reduced energy metabolism, particularly fatty acid oxidation, and that this metabolic defect is instrumental in the development of renal fibrosis.¹¹ Consequently, several investigators have proposed that targeting of renal energy metabolism (for example, via molecules, such as the energy-sensor AMP-activated protein kinase [AMPK] or PPARα) could be used as a novel therapeutic target against CKD.^8,11 In this context, another endogenous lipid signaling system, the endocannabinoid (eCB) system, may also represent an appealing therapeutic target.

The eCB system is distributed throughout the central nervous system and peripheral organs, including the kidneys. It is best known for its effects on the regulation of mood, appetite, and pain. Its non-neural effects include control of whole-body energy regulation and modulation of insulin sensitivity by its actions on the gut, pancreas, liver, and muscle.¹² Indeed, systemic activation of the eCB is reported to be important in type 2 diabetes. In the kidneys, the eCB system is widely represented by cannabinoid receptor 1 (CB₁R) and CB₂R, in particular in podocytes, mesangial cells, and PTCs. Obesity activates the eCB/CB₁R system, and pharmacologic blockade of the CB₁R has been shown to improve renal function and reduce albuminuria and glomerular lesions in obese and diabetic mouse models, but it had more limited effect on the fibrotic response to obesity.¹³ A potential reason for the disparity may be due to systemic versus renal effects of the CB₁R, particularly because peripheral blockade could potentially either enhance renal protection (by increasing adiponectin) or blunt it by increasing renal uptake of leptin, causing a reduction in PTC metabolic activity.¹⁴ Whether specifically targeting the intrarenal eCB/CB₁R pathway is advantageous is unknown.

In this issue of the Journal of the American Society of Nephrology, Udi et al.⁸ generated proximal tubular–specific CB₁R−/− mice (PCT-CB₁R−/−) to describe their role in the development of obesity-related CKD. They found that these mice exhibited reduced obesity-induced lipid accumulation in the kidney along with less renal dysfunction, injury, inflammation, and fibrosis, despite having no effect on the degree of weight gain induced by the HFD. Thus, this study elegantly shows that specifically targeting pathways of lipid metabolism in the kidney can be beneficial, independent of any systemic actions. The PTC-CB₁R deletion was associated with enhanced fatty acid β-oxidation, which appeared to occur downstream of increased activation of liver kinase B1 and AMPK. This is consistent with previous studies that found that reduced AMPK activity is important in the initiation of HFD-induced kidney disease and that this could be ameliorated by pharmacologic activation of the AMPK pathway.⁷

Deleting the PTC-CB₁R in the HFD model by Udi et al.⁸ has several beneficial effects, including reduced mesangial expansion and albuminuria as well as reduced upregulation of proinflammatory and profibrotic markers (e.g., collagen 1, IL-18, TNF-α, and iNOS). However, it suffers from the same mild phenotype as previous studies in this field; the histologic changes with the HFD were not severe, GFR was not diminished, and there was no progression of the severity of injury when the model was extended for up to 43 weeks of high-fat feeding. Although these data from Udi et al.⁸ suggest a central role for CB₁R in PTCs in obesity-related CKD, this does not exclude the possibility that CB₁R expression by podocytes may also play a role, such as has been described in type 2 diabetes.¹⁵ However, for the reasons mentioned before, this concern should not detract from its potential clinical relevance.

It is sobering to reflect that the attributable risk of obesity on the rate of kidney disease has been previously estimated to be between 14% and 34%,¹ suggesting that a vast number of individuals are at risk. Given the clearly profound societal and economic implications of this situation, studies, such the one by Udi et al.,⁸ that improve our understanding of the mechanisms underlying obesity-related CKD are welcomed, because they leave us better placed to develop rational strategies that more specifically address the problem. These likely will need to be multifaceted, including a combination of public health, behavioral, and pharmacologic interventions.

Disclosures

None.