L. M. Frydrych1, G. Bian1, K. He1, G. K. Wakam1, M. S. Anderson1, F. Fattahi2, P. A. Ward2, M. J. Delano1 1University Of Michigan,Department Of Surgery, Division Of Acute Care Surgery,Ann Arbor, MI, USA 2University Of Michigan,Department Of Pathology,Ann Arbor, MI, USA
Introduction: Sepsis is the leading cause of death in the critically ill population. In diabetic patients with sepsis, acute kidney injury (AKI) frequently occurs and leads to increased mortality. Although the specific mechanisms responsible for diabetic AKI during sepsis are unknown, a growing body of evidence suggest that ongoing inflammation and renal complement deposition provoke chronic kidney injury in other disease states. However, the role that acute inflammation and complement deposition plays in the development of diabetic AKI following sepsis has yet to be investigated. We hypothesize that sepsis induced inflammation enables renal complement production, drives AKI evolution, and potentiates diabetic kidney injury following sepsis.
Methods: 30 week old C57BL/6, C3-/-, TLR2-/-, TNF-/-, and Diet Induced Obese (diabetic) mice with lean controls underwent sublethal cecal ligation and puncture (CLP) or sham procedure with and without TNF inhibition (Etanercept 100μg/mouse/day intraperitoneal). At serial time points between days 1 and 21 after CLP or sham, n=5 mice/group were euthanized. Kidneys and plasma were harvested for histology using in-situ hybridization or Periodic acid–Schiff staining, as well as analyzed for TLRs 1-9, C3, C5a, TNF, BUN, creatinine, and NGAL expression using RT-PCR or specific ELISA.
Results: In lean mice, AKI peaked 24 hours after CLP with a tenfold elevation in plasma and renal tissue NGAL levels compared to sham controls (p=0.001). In diabetic mice, AKI did not peak until 2 weeks after sepsis initiation. Renal C3, but not C5a, mRNA levels were eight fold higher 24 hours after CLP compared to sham in both lean and diabetic mice (p=0.001). However, in lean animals C3 levels returned to baseline at 7 days while C3 levels remained elevated, peaked at 2 weeks, and finally returned to baseline at 3 weeks in diabetic mice. Renal in-situ hybridization showed significantly more glomerular and proximal convoluted tubal C3 transcripts in septic diabetic animals compared to lean controls. RT-PCR analysis of kidney homogenates after sepsis revealed greater renal TNF mRNA expression over a 3 week period in diabetic compared with lean mice (p=0.01). TNF-/- and C3-/- lean mice each produced significantly less renal C3 and manifested less AKI, as determined by NGAL levels. TLR2 inhibition with TLR2-/- mice or Etanercept prevented renal C3 production (p=0.01), NGAL elevation (p=0.01), and sepsis induced AKI (p=0.01).
Conclusion: In lean mice, sepsis induced AKI occurs early, resolves quickly, and is dependent on local renal production of C3 through TLR2 and TNF signaling pathways. In diabetic mice, AKI also presents early; however, it persists for 2 weeks and does not fully resolve until week 3. TNF and TLR2 inhibition after sepsis represent novel therapeutic strategies to prevent kidney injury and improve diabetic survival following sepsis.