The role of phosphodiesterase 3 in endotoxin-induced acute kidney injury

Abstract

Background: Acute kidney injury frequently accompanies sepsis. Endotoxin is known to reduce tissue levels of cAMP and low levels of cAMP have been associated with renal injury. We, therefore, hypothesized that endotoxin induced renal injury by activating phosphodiesterase 3 (PDE3) which metabolizes cAMP and that amrinone an inhibitor of PDE3 would prevent the renal injury.

Methods: Animals were divided into three groups (n = 7/group): 1) Control (0.9% NaCl infusion without LPS); 2) LPS (0.9% NaCl infusion with LPS); 3) Amrinone+LPS (Amrinone infusion with LPS). Either lipopolysaccharide (LPS) or vehicle was injected via the jugular vein and the rats followed for 3 hours. We explored the expression of PDE3 isoenzymes and the concentrations of cAMP in the tissue.

Results: The PDE3B gene but not PDE3A was upregulated in the kidney of LPS group. Immunohistochemistry also showed that PDE3B was expressed in the distal tubule in the controls and LPS caused PDE3B expression in the proximal as well. However, PDE3A was not expressed in the kidney either in the control or LPS treated groups. Tissue level of cAMP was decreased after LPS and was associated with an increase in blood urea nitrogen, creatinine, ultrastructural proximal tubular changes, and expression of inducible nitric oxide synthase (iNOS) in the endotoxemic kidney. In septic animals the phosphodiesterase 3 inhibitor, amrinone, preserved the tissue cAMP level, renal structural changes, and attenuated the increased blood urea nitrogen, creatinine, and iNOS expression in the kidney.

Conclusion: These findings suggest a significant role for PDE3B as an important mediator of LPS-induced acute kidney injury.

Figure 1

Effects of Lipopolysaccharide (LPS) and type 3 phosphodiesterase (PDE3) inhibitor on mean arterial pressure (MAP) in rats (n = 7 per group). PDE3 inhibitor, amrinone, mildly reduced MAP. *P < 0.05 versus Control or LPS; ^†P < 0.05 versus Control.

Figure 2

A representative RT-PCR on total RNA from rat kidneys (A). A representative immunoblotting for PDE3A in the kidney and heart tissue extracts prepared from control and LPS treated group (B). LPS treatment induced 9-fold greater expression of PDE3B mRNA transcript than control. GAPDH was used as a loading control (A). PDE3A was not detected in the kidney tissue either with LPS or without LPS. However, PDE3A was detected in the heart tissue, and LPS caused an increase in its expression in the heart (B).

Figure 3

PDE3B expression in a control kidney (A) and one treated with LPS (B). In the control group (A), immunohistochemical stain for PDE3B showed positive expression in the epithelial cells of the distal convoluted tubules (arrows). The glomerulus and proximal convoluted tubules showed no prominent expression for PDE3B. In the LPS group (B), both proximal and distal convoluted tubules showed strong expression for PDE3B (B). PCT: proximal convoluted tubule, DCT: distal convoluted tubule, G: glomerulus. Original magnification, ×200.

Figure 4

cAMP concentration (as determined by ELISA) in renal cortex at the end of the experiment (n = 7/group). *P < 0.05 versus control or Amrinone+LPS.

Figure 5

Effects of LPS administration and amrinone infusion on serum BUN (A) and creatinine (B) (n = 7/group). *P < 0.05 versus control or Amrinone+LPS; ^†P < 0.01 versus control or Amrinone+LPS.

Figure 6

Electron micrographs of the proximal tubule. Control animal showed a normal tubule and normal appearance of mitochodria (A). Loss of polarity of mitochondria and increased perinuclear clear space occurred in the endotoxemic kidney (B). Amrinone infusion restored mitochondrial integrity and intracellular edema (C). Scale Bar = 5 μm. Nu, nucleus; M, mitochondria; Ly, lysosome.

Figure 7

Endotoxin caused increased inducible nitric oxide synthase (iNOS) expression in the kidney tissue. Amrinone infusion attenuated iNOS expression. (A) A representative immunoblotting for iNOS in the kidney tissue extracts prepared from control, LPS, and Amrinone+LPS groups (n = 6 per group). (B) Densitometric results are expressed as a percentage compared with the mean value in LPS group (100%). *P < 0.01 versus Control or Amrinone+LPS.