Renal NMDA Receptors Are Increased in the Akita Mouse Model of Type 1 Diabetes and NMDA Antagonists Reduce Diabetic Nephropathy

Abstract

NMDA receptors are expressed in glomeruli, proximal tubules, and medullary collecting ducts, as well as in cell lines derived from these structures [1]. There is evidence that certain endogenous agonists of NMDA receptors, such as L‐homocysteate, quinolinate, and guanadinosuccinate, are increased in diabetes [1]. Moreover, sustained activation of NMDA receptors induces marked changes in cytoskeletal organization, and, in due time, induces apoptosis in cultured podocytes [2, 3]. Here we have studied NMDA receptors in the MPC5 immortalized mouse podocyte cell line and in the Akita mouse (on the DBA/2J background), a classic model for type‐1 diabetes. Experiments in mice were approved by the University of Houston IACUC. We observed that exposing cultured podocytes to high glucose (25 mM) for 24 hr caused marked increases in NR1, NR2A, NR2B, NR2C, and NR3B protein subunits and transcripts compared to controls cultured in 9 mM glucose + 16 mM mannitol. Using immunoblot and immunohistochemistry, we observed marked increases in NR1, NR2A and NR2C subunits in Akita mice compared to DBA/2J controls by 7 weeks of age, when albuminuria is still modest and light microscopic signs of nephropathy are not yet present. Increases in NMDA receptors were seen in glomeruli but were more pronounced in several tubular structures. To test the functional significance of NMDA receptors in diabetes, Akita mice and DBA/2J mice were treated with the NMDA antagonist MK‐801 for four weeks, starting at 8 weeks of age, using subcutaneous osmotic minipumps. Controls received saline. MK‐801‐treated Akita mice had significantly reduced 24‐hr albumin excretion compared to saline‐treated controls (P < 0.05, N = 9–10 per group, two‐way ANOVA). MK‐801 had no effect on albumin excretion in DBA/2J mice. MK‐801 treatment also eliminated the increase in glomerular basement membrane thickness that occurs in Akita mice (P < 0.05, N = 4) and improved foot process morphology and organization, as assessed by transmission and scanning EM. MK‐801 also caused a significant reduction in mesangial matrix expansion in Akita mice (P < 0.05). Reduced severity of diabetic nephropathy was also seen in Akita mice treated for 4 weeks with with memantine, an NMDA antagonist that is typically weaker than MK‐801, but which is already approved for clinical use. Memantine for 4 weeks produced significant improvements in 24‐hr albumin excretion (P < 0.05, N = 10), glomerular basement thickness (P < 0.05, N = 4) and podocyte foot process effacement. There was also a trend toward reduced mesangial matrix expansion that approached statistical significance (P = 0.0743, N = 4 mice per group, two‐way ANOVA). Akita mice with diabetes are smaller than DBA/2J controls, and NMDA antagonists did not change this trend; indeed, MK‐801 tended to make it worse. In addition, MK‐801 did not affect blood glucose in mice with type‐1 diabetes. These data suggest that peripheral NMDA receptors may be valid targets for therapies designed to slow the progression of diabetic nephropathy, possibly using agents such as memantine or dextromethorphan that are already used clinically.Support or Funding InformationJuvenile Diabetes Research Foundation