Metabolic stress through Na+-glucose cotransport (SGLT) activity promotes collectin binding, complement activation, and injury in renal proximal tubules and podocytes in diabetes

Diabetic nephropathy (DN) is a complication of diabetes mellitus and is characterized by persistent albuminuria and a progressive decline in kidney function. DN is a leading cause of end-stage kidney disease. Sodium-glucose cotransporter 2 (SGLT-2) inhibitors (gliflozins) have improved cardiovascular and kidney outcomes, but the molecular mechanisms behind the nephroprotection are not fully understood. The complement system has been linked to DN and is activated through three interrelated pathways: the classical, the alternative, and the lectin pathway. Mannan binding lectin (MBL), a recognition molecule of the lectin pathway, is associated with the progression and development of DN. MBL belongs to the collectin family and several more collectins are known: collectin liver 1 (CL-L1) and collectin kidney 1 (CL-K1). Collectins bind to molecular patterns on pathogens, but also altered self-antigens, and allow serine proteases (MASPs) to initiate complement activation. MBL and CL-L1 are mainly expressed in the liver, while CL-K1 is expressed predominantly in the kidney. The purpose of the project was to explore new molecular mechanisms in the development of diabetes-mediated kidney injuries and to gain insight into the nephroprotection of SGLT-2 inhibitors. The focus of the study was to explore the association between CL-K1 and DN, and perhaps relate it to SGLT-2 inhibitors. It was hypothesized that hyperglycemia in early diabetes leads to stress and hypoxia in SGLT-2-associated kidney cells due to increased workload and oxygen consumption. Stress and hypoxia in combination with glycationinduced dysfunction of the complement system will lead to the binding of CL-K1 and result in complement activation. SGLT-2 inhibitors lower the workload and prevent the binding of CL-K1 by reducing stress and alleviating diabetes-mediated kidney injuries. Furthermore, in a discovery-based approach, we focused on identifying protective signaling pathways in glomerular podocytes.

The first study (Manuscript #1) aimed to investigate if dysregulated diabetes with hyperglycemia and stressed proximal tubular cells is associated with increased synthesis and release of collectins (CL-K1 and MBL) and complement components of the lectin pathway; to reveal if SGLT-2 inhibitors attenuate these parameters and if deletion of CL-K1 confers protection against diabetes-mediated kidney injury. We induced type-1 diabetes mellitus (T1DM) by streptozotocin (STZ) in FVB/NHan®Hsd and CL-K1 knockout (KO) mice, and used a combination of enzyme-linked immunosorbent assay (ELISA), western blotting, in-situ hybridization, and real-time PCR for quantification of collectins and complement products of the lectin pathway (MBL-associated serine protease 1 (MASP-1), complement factor 3a (C3a), complement factor 5a (C5a)) in organs and plasma. Experimental T1DM (14 and 35 Days) resulted in kidney injury (increased u-albumin and KIM-1) and increased protein abundance and mRNA levels of CL-K1 and MBL-C in kidney tissue, alongside an increase in plasma together with MASP-1. CL-K1 mRNA was located in the juxtaglomerular apparatus and intrarenal arteries in diabetic kidneys compared to vehicle. SGLT-2 inhibition by dapagliflozin lowered plasma glucose and attenuated the protein abundance of CL-K1 and MBL-C, but did not affect the mRNA levels, the level of kidney injury markers (KIM-1 and u-albumin), or anaphylatoxins in kidney tissue. Ongoing studies of CL-K1 KO mice indicate a protection against diabetes-mediated injuries. It is concluded that experimental diabetes increases CL-K1 mRNA and protein abundance, and SGLT-2 inhibitors could exert protection by lowering collectin abundance and binding.

The second study (Manuscript #2) explored similar research questions as in study 1 with plasma samples from a clinical trial. We hypothesized that SGLT-2 inhibitors lower the circulating level of collectins and attenuate systemic complement activation in patients with type 2 diabetes mellitus (T2DM) and albuminuria. We analyzed paired plasma samples from a published, placebo-controlled, cross-over, intervention study (The DapKid study) of patients with T2DM and albuminuria treated with dapagliflozin and placebo for 12 weeks (n=36). ELISA was used to determine the plasma concentration of collectins, MASP-2, C3a, and the stable C3 split product C3dg and a neoepitope only present in the assembled membrane attack complex (sC5b-9). We found that SGLT-2 inhibition by dapagliflozin resulted in a slight but significantly increased concentration of C3a (23.8 %) and C3dg (9.5%) in plasma. This was not associated with changes in the circulating concentration of collectins or sC5b-9. The hypothesis that SGLT-2 inhibition leads to lower levels of collectins and/or complement activation split products in plasma was refuted. Organ-specific local protective effects of SGLT-2 inhibitors against complement
activation cannot be excluded but are not reflected in systemic circulation.

The third study (Manuscript #3) focused on the effect of SGLT-2 inhibitors on collectins and complement activation products in patients with T2DM with (n=17) and without (n=16) chronic kidney disease (CKD) compared to patients with non-diabetic CKD (n=16) and included both plasma and spot urine samples. We hypothesized that 1) CKD with and without diabetes is associated with aberrant glomerular filtration of collectins and complement precursors, 2) SGLT-2 inhibition by empagliflozin lowers circulating collectins and complement activation in plasma, 3) SGLT-2 inhibition attenuates complement activation in kidney reflected in urine samples from patients with albuminuria and CKD. We analyzed paired plasma samples from a placebo-controlled cross-over intervention study of patients treated with empagliflozin and placebo for 4 weeks (The SiRENA study). We used ELISA as in manuscript #2 to determine the levels of collectins and complement components. Urine microvesicles were isolated to examine the presence of membrane-bound C5b-9. There was no change in collectins (CL-K1, CL-L1, and MBL) and complement activation split products C3a, C5a, and sC5b-9 in plasma in response to SGLT-2 inhibition except for an 18.1 % increase in C3dg concentration in patients with diabetes and CKD. Collectins were not detectable in crude or ex vivo concentrated urine. Following treatment with empagliflozin there was a significant decrease in urinary C3a and sC5b-9-creatinine ratios in patients with diabetes and CKD. This was recapitulated in urine microvesicles by immunoblotting for C5b-9. SGLT-2 inhibitors do not lower collectins and complement activation products in plasma they significantly reduce indices of complement activation and membrane deposition in the kidney. Long-term kidney protection by SGLT-2 inhibitors could involve reduced complement-mediated injury.


Taken together, these studies support the hypothesis that CL-K1 in the kidney might be a pathogenic contributor in the early stages of diabetic kidney affection and SGLT-2 inhibitors could potentially exert local protective effects in the kidney by inhibition of the complement system. However, the exact molecular mechanism by which CL-K1 recognizes ligands and where it binds needs further investigation.

A fourth study (Manuscript #4) was performed with a discovery-based open approach to map endogenous pathways that maintain the glomerular filtration barrier and protect podocytes. Podocytes are a part of the glomerular filtration barrier, and their dysfunction contributes to albuminuria and DN. We have previously shown that NPRA, the receptor for natriuretic peptides (ANP, BNP) is expressed in podocytes in the human kidney and is necessary to sustain filtration barrier proteins in mice. Transgenic overexpression of ANP and BNP by infusion in mice ameliorates kidney injuries and DN progression by reduction of albuminuria. We hypothesized that natriuretic peptides exert a direct protective effect on the glomerular barrier integrity through activation of NPRA and modulation of gene expression patterns in podocytes. We isolated green fluorescence-positive podocytes from mice with a conditional knockout of Npr1 encoding NPRA by fluorescence-assisted cell sorting (FACS). Differentially expressed genes (DEGs) in podocytes were assessed by RNA-sequencing of podocytes from wildtype and NPR-A knockout mice. Identified transcripts were analyzed by real-time PCR and ELISA of cultured human and mouse glomeruli. A total of 158 DEGs were identified, and among the downregulated genes were protein S which is known to have a protective effect in podocytes. The findings show that endogenous levels of natriuretic peptides through NPRA in mice support protective differentiation pathways in glomerular podocytes.