Sustained Klotho delivery reduces serum phosphate in a model of diabetic nephropathy

Abstract

Diabetic nephropathy (DN) is a primary cause of end-stage renal disease and is becoming more prevalent because of the global rise in type 2 diabetes. A model of DN, the db/db uninephrectomized ( db/db-uni) mouse, is characterized by obesity, as well as compromised renal function. This model also manifests defects in mineral metabolism common in DN, including hyperphosphatemia, which leads to severe endocrine disease. The FGF23 coreceptor, α-Klotho, circulates as a soluble, cleaved form (cKL) and may directly influence phosphate handling. Our study sought to test the effects of cKL on mineral metabolism in db/db-uni mice. Mice were placed into either mild or moderate disease groups on the basis of the albumin-to-creatinine ratio (ACR). Body weights of db/db-uni mice were significantly greater across the study compared with lean controls regardless of disease severity. Adeno-associated cKL administration was associated with increased serum Klotho, intact, bioactive FGF23 (iFGF23), and COOH-terminal fragments of FGF23 ( P < 0.05). Blood urea nitrogen was improved after cKL administration, and cKL corrected hyperphosphatemia in the high- and low-ACR db/db-uni groups. Interestingly, 2 wk after cKL delivery, blood glucose levels were significantly reduced in db/db-uni mice with high ACR ( P < 0.05). Interestingly, several genes associated with stabilizing active iFGF23 were also increased in the osteoblastic UMR-106 cell line with cKL treatment. In summary, delivery of cKL to a model of DN normalized blood phosphate levels regardless of disease severity, supporting the concept that targeting cKL-affected pathways could provide future therapeutic avenues in DN. NEW & NOTEWORTHY In this work, systemic and continuous delivery of the "soluble" or "cleaved" form of the FGF23 coreceptor α-Klotho (cKL) via adeno-associated virus to a rodent model of diabetic nephropathy (DN), the db/db uninephrectomized mouse, normalized blood phosphate levels regardless of disease severity. This work supports the concept that targeting cKL-affected pathways could provide future therapeutic avenues for the severe mineral metabolism defects associated with DN.

Keywords: FGF23; Klotho; cKL; diabetes; phosphate.

Fig. 1.

Various stages of diabetic nephropathy are exhibited by db/db uninephrectomized (db/db-uni) mice, and treatment with the circulating soluble form of α-Klotho (cKL) reduces hyperphosphatemia: representative differences in histopathological changes in kidneys from lean mice (A, C, and E) and db/db-uni LacZ mice (B, D, and F). A: hematoxylin-eosin (H&E)-stained section with normal renal pelvis (RP) and absence of inflammation. B: H&E-stained section showing a dilated renal pelvis with focally extensive inflammatory infiltrate (black arrows) that extends into the transitional epithelial lining. C: Masson’s trichrome (MTS)-stained section showing normal glomeruli with open capillary loops and normal mesangial matrix (stars). D: MTS-stained section depicting increased fibroplasia in glomeruli (increased blue matrix) with relative paucity of open capillary loops (yellow arrows). Inset shows a sclerotic glomerulus that has lost the normal glomerular architecture. Minimal interstitial fibrosis between the tubules is also highlighted in blue. E: periodic acid-Schiff (PAS)-stained section demonstrating normal mesangial matrix and open capillary loops (stars). F: significantly increased mesangial matrix deposition highlighted by PAS staining that frequently obliterates the normal capillary loops and cellularity in the glomeruli (yellow arrows). Glomeruli in db/db-uni LacZ mice are slightly larger than those in the lean mice because of functional hypertrophy subsequent to uninephrectomy. G: lean control (db/dm) mice were significantly lighter than all db/db-uni mice, regardless of albumin-to-creatinine ratio (ACR) or treatment group (*P < 0.05). H: all db/db-uni mice, regardless of ACR or treatment group, had a lower heart weight-to-body weight (BW) ratio compared with lean controls (*P < 0.05).

Fig. 2.

Administering adeno-associated virus 2/8 expressing the circulating soluble form of α-Klotho (AAV-cKL) to db/db uninephrectomized (db/db-uni) mice moderately improves kidney function and blood glucose. A: creatinine was elevated in low-albumin-to-creatinine ratio (ACR)/LacZ, high-ACR/LacZ, and high-ACR/cKL groups compared with lean controls (*P < 0.05). B: blood urea nitrogen (BUN) was significantly higher in high ACR/LacZ (**P < 0.01 vs. lean). AAV-cKL treatment reduced BUN after 6 wk (#P < 0.01 vs. high ACR/LacZ). C: blood glucose of lean control mice was significantly lower than that of all db/db-uni mice (*P < 0.01). Two weeks of AAV-cKL treatment significantly reduced blood glucose values in the high-ACR/cKL group (#P < 0.05).

Fig. 3.

Endocrine effects of adeno-associated virus 2/8 expressing the circulating soluble form of α-Klotho (AAV-cKL) in lean and db/db uninephrectomized (db/db-uni) mice. A: db/db-uni mice in both the low- and high-albumin-to-creatinine ratio (ACR) groups were hyperphosphatemic (**P < 0.01 vs. lean). Serum phosphate was significantly reduced in db/db-uni mice in both the low- and high-ACR groups 6 wk after AAV-cKL injection (#P < 0.01 vs. respective LacZ). AAV-cKL treatment in high-ACR db/db-uni mice significantly reduced serum phosphate compared with lean mice (**P < 0.01). B: serum Klotho levels in db/db-uni mice were significantly increased with AAV-cKL treatment 6 wk postinjection (**P < 0.01 vs. lean and respective LacZ). C: serum calcium was elevated in db/db-uni mice in both the low- and high-ACR groups treated with AAV-LacZ (*P < 0.05 and **P < 0.01 vs. lean). AAV-cKL treatment reduced serum calcium in both the low- and high-ACR groups (#P < 0.05 vs. respective ACR/LacZ). D: db/db-uni mice, regardless of ACR group, have elevated serum alkaline phosphatase (Alk Phos; *P < 0.05 and **P < 0.01 vs. lean). Treatment with AAV-cKL further elevated serum alkaline phosphatase in low- and high-ACR groups (#P < 0.05 vs. respective ACR/LacZ).

Fig. 4.

Changes in FGF23 processing with treatment with the circulating soluble form of α-Klotho (cKL) and downstream targets. A: intact FGF23 was significantly elevated in low- and high-albumin-to-creatinine ratio (ACR) groups of db/db uninephrectomized (db/db-uni) mice treated with adeno-associated virus 2/8 expressing cKL (AAV-cKL) at 4 and 6 wk postinjection (**P < 0.01). Six- and eight-week control values for intact FGF23 were as follows: db/dm: 133.8 ± 23.7 and 153.7 ± 8.2; low-ACR LacZ: 156.4 ± 25.2 and 233.5 ± 48.0; high-ACR LacZ: 209.0 ± 50.8 and 361.5 ± 102.5, respectively. B: COOH-terminal (C-term) fragments of FGF23 were significantly elevated in low- and high-ACR groups of db/db-uni mice treated with AAV-cKL at 4 and 6 wk postinjection (**P < 0.01). Control values for C-term FGF23 were as follows: db/dm: 1.1 ± 0.6; low-ACR LacZ: 29.9 ± 16.1; high-ACR LacZ: 415.5 ± 388.0.

Fig. 5.

Changes in downstream targets of FGF23 signaling and FGF receptor (FGFR) expression with treatment with the circulating soluble form of α-Klotho (cKL). A: cKL and FGF23 combination treatment (FGF23 + KL) increased polypeptide N-acetylgalactosaminyltransferase 3 (GALNT3; *P < 0.05), dentin matrix protein-1 (DMP1; **P < 0.01), and extracellular serine/threonine protein kinase FAM20C (FAM20C; *P < 0.05) mRNA expression. cKL treatment alone (KL) had no effect on mRNA expression. B: FGFR1, FGFR2, and FGFR3 mRNA expression was significantly reduced by combination treatment of cKL and FGF23 (*P < 0.05). FGFR4 mRNA expression was not statistically different from control following combination treatment.