Diclofenac in Gyps vultures : a molecular mechanism of toxicity

Abstract

Over the last decade, three species of Gyps vultures on the Asian subcontinent have declined dramatically in population numbers, some as much as 97 to 99%. Although the initial cause was believed to be infectious, it was later shown to be due to an inadvertent exposure to diclofenac via the food chain. In order to protect the remaining wild vultures, diclofenac needed to be removed from the food chain. Unfortunately the Indian government was reluctant to ban diclofenac until an alternate veterinary non-steroidal anti-inflammatory drug (NSAID) that was both safe in vultures and effective in cattle could be identified. Although meloxicam was tentatively identified as this drug, toxicity testing still needed to be undertaken. Using a previously validated model, two studies were undertaken to determine the acute toxic effect of diclofenac in vulture as well as to ascertain if the drug had the potential to accumulate. In the first study, meloxicam in formulation was shown to be safe as a single oral dose up to 2mg/kg in African White Backed-Vultures (Gyps africanus). To further demonstrate the safety of food borne meloxicam, vultures were exposed to meat rich in meloxicam residues, with once again no signs of toxicity being evident. In the second study the drugs ability to accumulate was evaluated pharmacokinetically in Cape Griffon Vultures (Gyps corprotheres). From this study meloxicam was shown to have a very short half-life of elimination, making it unlikely that the drug could be a cumulative toxin. This was subsequently confirmed clinically by the absence of toxicity in birds receiving repeated doses of meloxicam. Although meloxicam was shown to be adequately safe, the safety of other veterinary NSAIDs still required elucidation. While further testing in vultures would have been possible, the small population size of the various vulture species made this unethical. Therefore a surrogate species needed to be identified. With the domestic chicken (Gallus domesticus) being commonly available, attempts were made to validate the chicken as a model. Although the dosed chickens did show similar toxicity patterns from clinical pathology to histopathology, a major problem was their higher tolerance making it impossible to use them as a surrogate. It was, however, concluded that the domestic chicken may be used in mechanistic studies in an attempt to establish an in vitro model. From the mechanistic studies both diclofenac and meloxicam were directly toxic to chicken and vulture renal tubular epithelial cells following 48h of incubation. It was later shown that this toxicity was associated with an increased production of reactive oxygen species (ROS), which could be temporarily ameliorated by pre-incubation with uric acid due to its anti-oxidant activity. When cultures were incubated with either drug for only two hours, meloxicam showed no toxicity in contrast to the cellular toxicity present for diclofenac. In both cases no increase in ROS production was evident. In addition diclofenac influenced the excretion of uric acid by interfering with p-amino-hippuric acid channels. The effect on uric acid excretion persisted after the removal of the diclofenac. It was therefore concluded that vulture susceptibility to diclofenac results from a combination of an increase in cellular ROS, a depletion of intracellular uric acid concentration and most importantly the drug’s long half-life in the vulture. Unfortunately the importance of the drug’s half-life in the toxicodynamics makes it unlikely that in vitro testing will be possible.