TO THE EDITOR
We read with interest a recent paper published in World Journal of Gastroenterology by Wang et al, who showed that 2 h pretreatment with an extract of the plant Gentiana manshurica Kitagawa (GM) could protect mice against acetaminophen (APAP) hepatotoxicity (300 mg/kg). The authors concluded that GM is hepatoprotective against acetaminophen-induced liver injury due to its antioxidant properties and anti-apoptotic capacity. A comparison of GM to the well-established clinically used antidote N-acetyl-L-cysteine (NAC) resulted in similar beneficial effects. According to the authors, their data suggest that GM is a potent hepatoprotective agent with a comparable mechanism to NAC. However, some of these mechanistic conclusions are clearly not justified by the data and whether or not GM is actually hepatoprotective under clinically relevant conditions has not been established.
The first concern is the authors’ conclusion that GM has an anti-apoptotic effect. Apoptotic cell death is characterized by cell shrinkage and nuclear condensation[2,3]. However, none of these morphological features was evident from the liver sections shown by Wang et al. In contrast, extensive centrilobular necrosis was shown, which correlates with the extensive release of liver enzymes into the plasma. The cells that are stained positive for DNA strand breaks with the transferase-mediated dUTP nick end-labeling (TUNEL) assay are not apoptotic. These cells show the characteristic nuclear and cytosolic staining of oncotic cells, which has been extensively described for APAP hepatotoxicity[2-4]. The nuclear fragmentation is caused by the mitochondrial release of endonuclease G and apoptosis-inducing factor[4,5], rather than by caspase-activated DNase, which is responsible for DNA fragmentation during apoptosis. Although the authors claimed that caspase-3 was activated based on the appearance of a 17 kD cleavage product, no other bands (pro-caspase-3) have been shown and no positive controls as reference points were included. Most importantly, a dramatic increase in procaspase-3 cleavage should go along with a dramatic increase in caspase-3 enzyme activity, which has never been shown in APAP-induced liver injury[2,7-9]. Consistent with these observations, caspase inhibitors do not protect against APAP hepatotoxicity arguing against a relevant role of caspase activation in the pathophysiology[8,9]. Furthermore, in both primary cultured mouse and human hepatocytes, APAP induces necrotic cell death. Taken together, oncotic necrosis is the predominant mode of cell death (> 95%) during APAP hepatotoxicity in animal models in vivo and in humans. The only exception is APAP-induced cell death in metabolically incompetent hepatoma cell lines. However, these mechanisms have no relevance to the in vivo hepatotoxicity of this drug.
A second major concern is related to the extensive mechanistic conclusion including the hypothesis that GM acts as an antioxidant. There is direct evidence that reactive oxygen species and peroxynitrite are formed in mitochondria during APAP hepatotoxicity and play a critical role in cell death[4,12,13]. However, the fact that at 12 h after APAP administration, there was less lipid peroxidation, together with other evidence for reduced tissue injury, does not prove that GM acts as an antioxidant. The same results would be obtained if GM protected and improved cell viability through other mechanisms with the consequence of less oxidant stress. In fact, one of the most likely mechanisms of protection, i.e. that one or several compounds in this plant extract may have inhibited cytochrome P450 activities or may have competed with APAP for metabolism, was not investigated. Toxicity of APAP is entirely dependent on its metabolic activation, which means that any interference with its reactive metabolite formation will substantially reduce or even eliminate toxicity. In the absence of clear evidence that this extract does not affect reactive metabolite formation, any conclusion regarding more distal mechanisms is not justified.
The third concern is the conclusion that GM acts as a hepatoprotectant similar to NAC. However, in clinically relevant situations of drug overdose, the antidote has to be effective when administered after the insult not as a pretreatment. The effectiveness of GM against APAP hepatotoxicity when treated after drug overdose has not been investigated. Furthermore, the comparison to NAC is not justified. NAC given as pretreatment to fasted animals will support glutathione (GSH) synthesis in the liver resulting in much higher GSH levels than the respective controls 2 h later. These elevated GSH levels will more effectively scavenge the reactive metabolite of APAP and therefore prevent initiation of liver injury. This protection mechanism of NAC is independent of the antioxidant effect of GSH mainly because no oxidant stress is generated at this point. However, if NAC is administered several hours after APAP, i.e. at a time when hepatic GSH is depleted and mitochondria have already generated an oxidant stress, GSH synthesized at this time is used to scavenge reactive oxygen species and peroxynitrite[13,17]. In addition, some of the excess NAC will also be used to support the impaired mitochondrial energy metabolism. Both mechanisms contribute to the late protection against APAP hepatotoxicity[13,17]. Thus, NAC can have 3 different mechanisms of action depending on the time of administration relative to APAP. The pretreatment with NAC as used by Wang et al will mainly scavenge the reactive metabolite of APAP, an effect that is unlikely to be relevant to GM.
Taken together, the protective effect of GM against APAP hepatotoxicity is an interesting observation. However, GM as a hepatoprotectant against drug toxicity under clinically relevant conditions has not been demonstrated. In addition, the actual protection mechanism of GM remains unclear. More mechanistic studies considering clinically relevant conditions are needed to evaluate the potential of this plant extract as an antidote against drug hepatotoxicity.