The murine collagen antibody-induced arthritis (CAIA) model and primary mouse and human hepatocytes were used to investigate changes in cytochrome P450 expression incurred during inflammatory disease and by disease treatment with anti-cytokine monoclonal antibodies. Hepatic cytochrome P450-related effects of CAIA were compared to the effects of IL-6, IL-1β, and TNFa on mouse hepatocytes, to discern in vitro/in vivo correlation related to the contributions of individual cytokines and cytokines in combination. Moreover, the effects of cytokines and cytokine combinations on cytochrome P450 expression and acute phase response markers were compared between mouse and human hepatocytes to discern any overt species differences.
CAIA in female mice increased serum IL-1β, IL-6 and hepatic mRNA of the acute phase response marker, SAA. Hepatic cytochrome P450-related effects of CAIA and the effects of IL-6, IL-1β, and TNFa alone and in combination on mouse hepatocytes were compared. In vivo/in vitro suppression by cytokines was congruent for some cytochrome P450 isoforms (Cyp1a2, Cyp2c29, and Cyp3a11) but not for others (cytochrome P450 oxidoreductase (POR) and Cyp2e1). Anomalously in a pilot study, commercial murine IL-6 and IL-6 receptor antibodies did not lower liver mRNA levels of acute phase response markers and did not de-suppress cytochrome P450 levels in the CAIA mouse liver. In mouse hepatocyte culture, IL-1β and IL-6 were both equipotent at suppressing cytochrome P450 gene expression while TNFa caused only mild suppression at the highest concentrations used. This is in contrast to results obtained in human hepatocytes where IL-6 and TNFa were more potent than IL-1β at causing cytochrome P450 suppression. IL-1β and IL-6 dosed in combination in vitro caused a vast synergistic response in SAA expression in mouse hepatocytes, but mainly an additive response in human hepatocytes. In mouse hepatocytes, TNFa in combination with IL-1β, IL-6, or both had a protective effect against IL-1β and IL-6 mediated cytochrome P450 suppression, and for several cytochrome P450 isoforms low dose TNFa caused induction rather than suppression in the presence of IL-1β or IL-6. In contrast, when human hepatocytes were treated with IL-1β and IL-6 in combination at concentrations ranging from 1-100 pg/mL, IL-6 was the main determinant of mRNA increases in C-reactive protein and serum amyloid A and mRNA decreases in CYP3A4.
Although the CAIA mouse is an acceptable, quick turnaround model for the study of suppression of drug metabolizing enzymes by inflammation, factors underlying the lack of effect of the commercial anti-cytokine monoclonal antibodies, including dose and role of other cytokines, warrant more detailed experimentation. In vitro experiments with cytokine combinations revealed that the relative contributions of different cytokines to cytochrome P450 suppression are different in mouse and human. This may limit the human relevance of the mouse model in mechanistic investigations. In summary, in vitro and/or pre-clinical in vivo experimentation may afford semi-quantitative data indicating that ablating particular inflammatory cytokine effects might de-suppress cytochrome P450 activity. However, the predictive value of these pre-clinical approaches for discerning subtle disease treatment effects on cytochrome P450 expression in humans in vivo is likely to remain limited.