Question: 

The association between inhibition of renal prostaglandin synthesis by non-steroidal anti-inflammatory drugs (NSAID) and renal dysfunction is well known, however, the underlying pathophysiology is incompletely understood. We have recently observed in MDCK cells that prostaglandin E₂ (PGE₂) promotes cell survival during hypertonic stress by phosphorylation and inactivation of the pro-apoptotic protein Bcl-2 antagonist of death (BAD). We addressed the role of COX-2-derived PGE₂ on phosphorylation of BAD and medullary apoptosis in COX-2 deficient mice.

Methods: 

COX-2^-/- and COX-2^+/+ animals had free access to water or were deprived of water. After 24 h, urine and kidney tissue samples were taken for analysis. Urinary concentration of PGE2 was measured by ELISA assay. Expression of COX-2 and BAD and phosphorylation of BAD were assessed by Western blot analysis. Caspase-3 activity was used as indicator of apoptosis in renal medullary homogenates. For analysis of localization of phosphorylated and unphosphorylated BAD and identification of apoptotic cells in the renal medulla, immunohistochemistry was used. The role of BAD and COX-2 during tonicity-induced apoptosis was further investigated by siRNA-mediated knockdown of BAD and pharmacological inhibition of COX-2 in MDCK cells.

Results: 

Both COX-2 deficient mice and wild type animals constitutively expressed BAD along the inner and outer medullary collecting duct. Dehydration caused a robust increase in papillary COX-2 expression, PGE₂ excretion and Bad phosphorylation in COX-2^+/+, but not in COX-2^-/- animals. Accordingly, caspase-3 activity, as an indicator of apoptosis, was significantly higher in renal medullary epithelial cells of COX-2^-/- mice. Accordingly, inhibition of COX-2 in MDCK cells decreased BAD phosphorylation and increased apoptosis during hypertonic stress. The important role of BAD for induction of apoptosis was further illustrated by the fact that siRNA-mediated knockdown of BAD reduced caspase-3 activation during hypertonic stress. Furthermore, the addition of PGE₂ to cells with knockdown of BAD had no effect on induction of apoptosis, whereas in mock transfected cells PGE₂ caused phosphorylation of BAD, probably mediated by the EP₂ receptor, and substantially improved cell survival.

Conclusions: 

In summary, dehydration and hence osmotic stress increases COX-2 expression and PGE₂ production in the renal medulla, which entails phosphorylation and inactivation of the pro-apoptotic protein BAD. This mechanism counteracts tonicity-induced apoptosis in renal medullary cells and may provide novel insights into the development of NSAID-induced renal medullary injury.