Delayed myelosuppression with acute exposure to hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) and environmental degradation product hexahydro-1-nitroso-3,5-dinitro-1,3,5-triazine (MNX) in rats.

Hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX), a widely used munitions compound, and hexahydro-1-nitroso-3,5-dinitro-1,3,5-triazine (MNX), its N-nitroso product of anaerobic microbial nitroreduction, are contaminants of military sites. Previous studies have shown MNX to be the most acutely toxic among the nitroreduced degradation products of RDX and to cause mild anemia at high dose. The present study compares hematotoxicity with acute oral exposure to MNX with parent RDX. Both RDX and MNX caused a modest decrease in blood hemoglobin and ~50% loss of granulocytes (NOAELs=47 mg/kg) in female Sprague-Dawley rats observed 14 days post-exposure. We explored the possibility that blood cell loss observed after 14 days was delayed in onset because of toxicity to bone marrow (BM) progenitors. RDX and MNX decreased granulocyte/macrophage-colony forming cells (GM-CFCs) at 14, but not 7, days (NOAELs=24 mg/kg). The earliest observed time at which MNX decreased GM-CFCs was 10 days post-exposure. RDX and MNX likewise decreased BM burst-forming units-erythroid (BFU-Es) at 14, but not 7, days. Granulocyte-erythrocyte-monocyte-megakaryocyte (GEMM)-CFCs were unaffected by RDX and MNX at 7 days suggesting precursor depletion did not account for GM-CFC and BFU-E loss. MNX added to the culture media was without effect on GM-CFC formation indicating no direct inhibition. Flow cytometry showed no differential loss of BM multilineage progenitors (Thy1.1(+)) or erythroid (CD71(+)) precursors with MNX suggesting myeloid and erythroid lineages were comparably affected. Collectively, these data indicate that acute exposure to both RDX and MNX caused delayed suppression of myelo- and erythropoiesis with subsequent decrease of peripheral granulocytes and erythrocytes.

Nimesulide-induced electrophile stress activates Nrf2 in human hepatocytes and mice but is not sufficient to induce hepatotoxicity in Nrf2-deficient mice.

Nimesulide is a widely prescribed nitroaromatic sulfoanilide-type nonsteroidal anti-inflammatory drug that, despite its favorable safety profile, has been associated with rare cases of idiosyncratic drug-induced liver injury (DILI). Because reactive metabolites have been implicated in DILI, we aimed at investigating whether hepatic bioactivation of nimesulide produces a protein-reactive intermediate in hepatocytes. Also, we explored whether nimesulide can activate the transcription factor Nrf2 that would protect from drug-induced hepatocyte injury. We found that [(14)C]-nimesulide covalently bound to human liver microsomes (<50 pmol/mg under standard conditions) or immortalized human hepatocytes in a sulfaphenazole-sensitive, rifampicin-inducible manner; yet the overall extent of binding was modest. Although exposure of hepatocytes to nimesulide was not associated with increased net levels of superoxide anion, nimesulide (100 microM, 24 h) caused nuclear translocation of Nrf2 in a sulfaphenazole-sensitive manner, indicating a role of electrophilic metabolites. However, knockdown of Nrf2 with siRNA did not make the cells more sensitive to nimesulide-induced cell injury. Similarly, exposure of wild-type C57BL/6x129 Sv mice to nimesulide (100 mg/kg/day, po, for 5 days) was associated with nuclear translocation of immunoreactive Nrf2 in a small number of hepatocytes and induced >2-fold the expression levels of the Nrf2-target gene Nqo1 in wild-type but not Nrf2-null mice. Nimesulide administered to Nrf2(-/-) knockout mice did not cause increases in serum ALT activity or any apparent histopathological signs of liver injury. In conclusion, these data indicate that nimesulide is bioactivated by CYP2C to a protein-reactive electrophilic intermediate that activates the Nrf2 pathway even at nontoxic exposure levels.

Underlying mitochondrial dysfunction triggers flutamide-induced oxidative liver injury in a mouse model of idiosyncratic drug toxicity.

Flutamide, a widely used nonsteroidal anti-androgen, but not its bioisostere bicalutamide, has been associated with idiosyncratic drug-induced liver injury. Although the susceptibility factors are unknown, mitochondrial injury has emerged as a putative hazard of flutamide. To explore the role of mitochondrial sensitization in flutamide hepatotoxicity, we determined the effects of superimposed drug stress in a murine model of underlying mitochondrial abnormalities. Male wild-type or heterozygous Sod2(+/-) mice were injected intraperitoneously with flutamide (0, 30 or 100 mg/kg/day) for 28 days. A kinetic pilot study revealed that flutamide (100 mg/kg/day) caused approximately 10-fold greater exposure than the reported therapeutic mean plasma levels. Mutant (5/10), but not wild-type, mice in the high-dose group exhibited small foci of hepatocellular necrosis and an increased number of apoptotic hepatocytes. Hepatic GSSG/GSH, protein carbonyl levels, and serum lactate levels were significantly increased, suggesting oxidant stress and mitochondrial dysfunction. Measurement of mitochondrial superoxide in cultured hepatocytes demonstrated that mitochondria were a significant source of flutamide-enhanced oxidant stress. Indeed, mitochondria isolated from flutamide-treated Sod2(+/-) mice exhibited decreased aconitase activity as compared to vehicle controls. A transcriptomics analysis using MitoChips revealed that flutamide-treated Sod2(+/-) mice exhibited a selective decrease in the expression of all complexes I and III subunits encoded by mitochondrial DNA. In contrast, Sod2(+/-) mice receiving bicalutamide (50 mg/kg/day) did not reveal any hepatic changes. These results are compatible with our concept that flutamide targets hepatic mitochondria and exerts oxidant stress that can lead to overt hepatic injury in the presence of an underlying mitochondrial abnormality.

Comparative cytotoxicity of alachlor, acetochlor, and metolachlor herbicides in isolated rat and cryopreserved human hepatocytes.

Noncancerous adverse effects observed at the lowest dose for chloroacetanilide herbicides alachlor [2-chloro-2',6'-diethyl-N-(methoxymethyl)-acetanilide] and acetochlor [2-chloro-2'-methyl-6'-ethyl-N-(ethoxymethyl)acetanilide], but not metolachlor [2-chloro-2'-ethyl-6'-methyl-N-(1-methyl-2-methoxymethyl)acetanilide], are hepatotoxicity in rats and dogs. Liver microsomal N-dealkylation, a step in the putative activating pathway, of acetochlor exceeds that of alachlor and is negligible for metolachlor. In the present investigation, cytotoxicity of the three chloroacetanilides was ranked using isolated rat and cryopreserved human hepatocytes to correlate this endpoint with CYP3A-dependent metabolism. Chloroacetanilide cytotoxicity in rat hepatocyte suspensions was time dependent (e.g., LC(50 - alachlor/2 h) vs. LC(50 - alachlor/4 h) = 765 vs. 325 muM). Alachlor and acetochlor were more potent than metolachlor after 2 and 4 h, times when N-dealkylated alachlor product 2-chloro-N-(2,6-diethylphenyl)acetamide (CDEPA) formation was readily detectable. Alachlor and acetochlor potencies with cryopreserved human hepatocytes at 2 h were comparable to freshly isolated rat hepatocytes, and alachlor metabolism to CDEPA was likewise detectable. Unlike rat hepatocytes, metolachlor potency was equivalent to acetochlor and alachlor in human hepatocytes. Furthermore, chloroacetanilide cytotoxicity from two sources of human hepatocytes varied inversely with CYP3A4 activity. Collectively, while cytotoxicity in rat hepatocytes was consistent with chloroacetanilide activation by CYP3A, an activating role for CYP3A4 was not supported with human hepatocytes.