Systemic disease diabetes mellitus (DM) is characterized by an inability of the body to either produce or effectively respond to the glucose-regulating hormone, insulin. The International Diabetes Federation 2017 estimates that there are 425 million people (both diagnosed and undiagnosed) with diabetes in the world, which will reach 629 million people by 2045. The kidney is a highly vulnerable tissue in the diabetes milieu as the prevalence of end-stage renal disease (ESRD) is up to 10 times higher in people living with diabetes (1). Diabetic nephropathy or diabetic kidney disease (DKD) is a key microvascular complication of diabetes, classically identified by the presence of proteinuria (microalbuminuria in early stages and macroalbuminuria as DKD advances) in people with diabetes. However, increasing evidence has shown that a significant number of patients with type 2 DM may have decreased glomerular filtration rate (GFR) without significant albuminuria, known as non-albuminuric DKD (2). Progression of DKD is more likely to occur in patients who have long-standing diabetes, poor glycemic control, or associated morbidities such as hypertension or obesity. But the progression rate to kidney failure in people who are non proteinuric but living with diabetes is much lower than those who are proteinuric (3). The lifetime risk of DKD is roughly equivalent to type 1 (insulin-dependent, juvenile-onset) and type 2 (adult-onset) diabetes (4, 5). There are no identified therapies that can specifically target reversal or slow down the progression of DKD injury. Current DKD management involves general measures including, lifestyle modification, blood pressure control, glycemic control, plus the use of lipid-lowering drugs, albuminuria-reducing drugs, and treatment with sodium-glucose co-transporter 2 (SGLT2) inhibitors (6–11). Among people with diabetes, the development of DKD carries a higher mortality risk. A substantial proportion of people with DKD will have progressive loss of kidney function and will develop ESRD (12). Hence, there is a clear unmet need to understand the specific biological basis of kidney injury in diabetes, to understand pathways that contribute to progressive DKD injury, and develop specific DKD-targeted therapies that can reverse or slow down the unrelenting progression of DKD and, improve both the quality and quantity of life for this patient group.

The development and progression of DKD is thought to involve a combination of hemodynamic, metabolic, ischemic, and inflammatory factors (13–16). However, the exact mechanisms of kidney injury in a hyperglycemic milieu, with underlying genetic and racial risk factors that may increase DKD disease risk, remain to be elucidated. The predominant structural changes include mesangial expansion, glomerular basement membrane thickening, podocyte injury, and, ultimately, glomerular sclerosis (17). Albuminuria and progressive chronic kidney disease (CKD) are major clinical manifestations of DKD (18). Although, classic lesions of diabetic nephropathy is similar in both type 1 and type 2 diabetes, however, renal lesions are more heterogeneous in patients with type 2 diabetes with some patients developing more advanced vascular or chronic interstitial lesions than diabetic glomerulopathy (19, 20). The principal biomarkers presently used to predict DKD progressions are albuminuria and estimated glomerular filtration rate (eGFR) (21). However, not all cases of classical DKD are accompanied by increases in albuminuria, which reduces the value of this biomarker, particularly in early DKD (22–24). Optimizing the management of diabetes can reduce the rate of kidney function decline, but with a lack of clarity on DKD pathogenesis, specific DKD therapies have not been developed to date (25).

We hypothesized that innate immunity and inflammation may play an important mechanistic role in DKD progression. Our study is designed to evaluate such perturbations in the kidney tissue and circulating blood of people with diabetes, and then interrogates the expression of specific markers in the biosamples with the increased severity of diabetic kidney damage and dysfunction. The hypothesis stems from the fact that dendritic cells, mast cells, and macrophages have all been reported to infiltrate diabetic kidneys (15, 26). A population study of diabetes revealed a positive correlation between plasma IgE and diabetes or prediabetes (27). Increased levels of eosinophilia positively correlated with worsening stages of CKD and DKD (28), thus providing evidence of dysregulated innate immunity. However, this is not an invariant finding in all people living with diabetes. Histological findings are consistent with apparent observations of mast cells and eosinophils in kidney tissues from DKD.

In this paper, we carefully mapped three independent, non-overlapping cohorts of individuals with or without diabetes, to identify genes that might play a crucial role in the development and progression of DKD. We performed an unbiased discovery analysis of gene expression changes in peripheral blood mononuclear cells (PBMC) collected from individuals with varying stages of DKD in Cohort 1. Analysis of genes in Cohort 1 was focused on evaluating pathways that may play a role in the proinflammatory component of DKD injury, and are also impacted by the advanced stages of kidney injury ( Table 1 ). With the assumption that infiltration of leukocytes in the diabetic kidney may impact the severity of glomerular disease progression in diabetes, in Cohort 2 we performed unbiased discovery of transcriptional changes in the glomerulus of DKD tissue. We aimed to find enrichment of DKD specific overlapping genes between the two cohorts, which can then be used as a biomarker for DKD. Finally, in Cohort 3, we independently validated the protein expression and localization of the most enriched gene in renal tissue samples from varying stages of DKD, assessing if genes identified from Cohort 1 and 2 could be localized in DKD tissue. Figure 1 summarizes the overall study design.

Current therapy for DKD is mostly focused on the use of angiotensin-converting enzyme inhibitors (ACEi) or angiotensin receptor blockers (ARBs) to reduce intra-glomerular pressure (44) along with multifactorial interventions such as improved control of blood pressure, blood glucose, and lipid, and smoking cessation as a means of slowing the atrophy of renal function in diabetes (45). Sodium-glucose-linked transporter 2 (SGLT2) inhibitors have been shown to further reduce renal events in patients with diabetes as an additive benefit over ACEi or ARB therapy. SGLT2 inhibitors suppress renal glucose reabsorption to reduce blood glucose and body weight, alter renal hemodynamics, reduce intraglomerular pressure, attenuate diabetes-associated hyperfiltration, tubular hypertrophy, and tubular toxicity of glucose to directly protect the kidney. Furthermore, SGLT2 inhibitors reduce the workload of the proximal tubules to improve tubulointerstitial hypoxia, and then allow fibroblasts to resume normal erythropoietin production, and thereby protecting the kidney (46).

While DKD is not considered a primarily “immune-mediated” form of kidney disease, extensive evidence supports the involvement of many immune system components in DKD progression and even initiation. Emerging knowledge about the dysregulation of the immune response and inflammation, suggests a key pathogenic link between aberrant innate and adaptive immunity, metabolism, and progressive kidney damage, and thus open the door to explore new immune-modulatory therapies in DKD. Inflammation is being increasingly understood to be a prominent pathological feature of diabetic nephropathy. Increased inflammation and inflammatory markers have been shown to predict worse renal outcomes (15). Recently, baricitinib and a CCR2 antagonist CCX140 have been shown to reduce proteinuria in DKD, likely because of a global effect on renal tissue inflammation (47, 48). CCR2 expressed by monocytes and macrophages are the main receptor for the pro-inflammatory chemokine CCL2. Urinary excretion of CCL2 correlates with the severity of DKD, again supporting the role of increased inflammation and renal injury in diabetes (49–52). Additionally, the importance of pharmacologic intervention to block inflammation in the diabetic kidney, can be seen in pre-clinical models, where either blockade of CCR2, or lack of CCR2 in a knockout model, can reduce renal macrophage and monocyte infiltration and may translate to a downstream reduction of interstitial fibrosis in the kidney (45, 53).

Recognizing that there is a critical immune-mediated component of renal injury in diabetes, we undertook an innovative study design ( Figure 1 ) of unbiased transcriptional discovery and independent gene and protein validation of key dysregulated pro-inflammatory pathways in independent patients with diabetes with varying stages of DKD injury, examining different tissue sources, PBMC ( Figure 2 ) and kidney tissue ( Figure 3 ). This analysis reveals a skewed gene expression profile of an overlapping hub of genes, in both specific immune cells in the blood and similar infiltrating immune cells in the DKD kidneys, with significant enrichment in Fc epsilon receptor (FcER1), and T-cell receptor-mediated signaling pathways ( Figures 2E , 3B ). From the list of candidate genes, tumor necrosis factor (TNF), toll-like receptor (TLR), and chemokine receptor 1 (CCR1) have been previously described in the literature as playing a role in DKD; supporting the biological relevance of the identified gene-set in this study (54–58). Microarray analysis of the PBMC dataset (Cohort 1), identified FcER1, TRPC3, SNX20, FAM20A, and SLC12A7 as genes showing increased expression with the severity of diabetes ( Supplementary Table 1 ). FcER1 and TRPC3, both are expressed and associated with inflammatory signaling in mast cells. SNX20 plays a role in cellular vesicle trafficking, FAM20A is expressed in hematopoietic cells, and SLC12A7 is required for basolateral Cl (–) extrusion in the kidney and contribute to renal acidification. In Cohort 2, our eGWAS meta-analysis data evaluated up-regulated genes in microdissected glomeruli from DKD patients. Comparing both the PBMC and kidney tissue in independent DKD cohorts surprisingly identified a common set of 28 genes with significantly increased expression in both DKD PBMCs and glomerulous ( Table 2 ), with high significance for FcER1 in DKD disease ( Figure 4 ). Further expression and co-localization of FcER1 in yet another independent set of human DKD kidney tissue samples, with varying stages of renal injury, confirmed increased FcER1 gene and protein levels in human DKD injury, where FcER1 expression co-localized in infiltrating mast cells ( Figures 5 , 6 ). It is important for us to understand that glomerular genes in cohort 2 includes the entire glomerulus consists of glomeruli and glomerular-interstitial spaces. Tubulointerstitium is a continuation of the glomerular interstitium. So it is understandable that the inflammatory response is spilling into these continual mesenchymal spaces as the DKD pathology recruits more cells. Also, even in the case of early DKD though the major involvement is of the glomerulus, there is inflammatory cells recruitment from the tubular interstitium. All these could explain the expression of FcER1 protein in tubules as well ( Figures 5 and 6 ). In future, we will address the tubular involvement in DKD and its association with FcER1.

FcER1 is a high-affinity IgE receptor expressed on mast cells, basophils, eosinophils, and antigen-presenting cells (59). Mast cells are an innate immune cell type that has been previously shown to infiltrate the renal parenchyma in pre-clinical models of DKD and human diabetic kidney disease. We hypothesize that mast cells increasingly infiltrate the human diabetic kidney and support an inflammatory milieu, through increased expression of FcER1, which then may play a critical role in DKD progression. Increased intra-renal FcER1activation in infiltrating mast cells in a patient with diabetes may promote the release of many inflammatory mediators, including TGF-β, TNF-α, IL6, Tryptase, IL1, which can then subsequently drive the development of renal fibrosis. In diabetic nephropathy, mast cells participate in renal fibrosis by contributing to excessive accumulation of extracellular and mesangial matrix and also producing non-fibrillar short-chain type VIII collagen (60). Pre-clinical studies support this hypothesis, as mast cell inhibitors such as cromolyn or ketotifen (Zaditor) have been found to protect mice with diabetes from the development of renal injury (27). Some of this reno-protective effect may likely be mediated by reducing the expression of FcER1. In this study, we found coexpression of the mast cell marker tryptase and FcER1 in DKD with a trend towards higher tissue expression of both with DKD progression ( Figure 6 ). This fact argues for the biological relevance of this axis in DKD damage rather than this only being a non-specific effect of the kidney being just exposed to a diabetic milieu. Also, we noted that the PBMC transcriptome of healthy and ESRD cluster closer together in Figure 2 , which is consistent with the IF results in Figure 6 , where the kidney FcER1 expression appears to be dampened in patients with severe diabetic nephropathy compared with moderate disease. We believe this is because both, healthy and ESRD, have lower levels of immune response genes compared to active DKD kidney injury.

In conclusion, we demonstrate how transcriptomic datasets may be combined and integrated to highlight the most robust markers. This study highlights the importance of both immune and non-immune mechanisms driving diabetic nephropathy. The innate immune response occurring in the diabetic kidney is an anticipated consequence of the chronic stresses and injury in the diabetic kidney. Ultimately, the ongoing diabetic renal inflammation results in substantial kidney damage with progressive fibrosis that eventually leads to end-stage renal disease. Therefore, therapeutic strategies targeting the innate immune system will be important for the treatment of diabetic nephropathy. This is the first study implicating a direct role of the IgE receptor FcER1 in DKD progression, thus uncovering a new druggable target for improving DKD or slowing DKD progression. Further validation studies are planned to evaluate the role of FcER1 blockade in preclinical models of diabetes to directly address the efficacy of targeting this molecule as a novel therapeutic for DKD.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://www.ncbi.nlm.nih.gov/, GSE142153.

Written informed consent was obtained from all the participants, and the study was approved by the institutional review boards of the University of Wisconsin and University of California, San Francisco

MN conducted experiments. SS, PB, and MN performed data analysis, data interpretation, manuscript writing. HS participated in the sample collection and study design. TS contributed in experiments, data analysis and manuscript writing. MS conceived the study, analysis and data interpretation, writing and revising the manuscript critically for intellectual content. All authors contributed to the article and approved the submitted version.

This work was funded by two grants, NCATS/NIH (R21TR001761) and NIDDK/NIH (R01DK109720).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.