Induction of CYP1A1 and CYP1B1 in liver and lung by benzo(a)pyrene and 7,12-d imethylbenz(a)anthracene do not affect distribution of polycyclic hydrocarbons to target tissue: role of AhR and CYP1B1 in bone marrow cytotoxicity

Abstract

Polycyclic aromatic hydrocarbons (PAHs) (50 mg/kg, i.p.) selectively deplete mouse bone marrow (BM) hematopoietic cells through a process that is dependent on CYP1B1. 7,12-dimethylbenz(a)anthracene (DMBA), which forms greater amounts of dihydrodiol-epoxide-DNA adducts in BM, is much more effective in depleting BM cells than benzo(a)pyrene (BP). BM toxicity by BP is restored in congenic mice expressing a weakly responsive aryl hydrocarbon receptor (AhR^d replaces AhR^b). BP strongly induces CYP1A1 around the hepatic vein whereas DMBA produces a weaker diffuse response, paralleling differences in CYP1A1 protein. These responses are absent in AhR^d mice. BP and DMBA broadly and equally induce CYP1A1 in the lung, while CYP1B1 is induced in bronchial blood vessels. In sternum, CYP1B1 is induced in BM and white fat, whereas CYP1A1 is induced only in brown fat. BP and DMBA levels were similar within blood, lung, and BM and did not rise in AhR^d mice. In liver, selective decrease of BP was consistent with induced metabolism via CYP1A1, which nevertheless does not determine the blood levels and distribution to BM. Effective delivery of BP to BM is indicated by formation of BP-quinone DNA adducts and the effective induction of CYP1B1. The low formation of BP-dihydrodiol-epoxide-DNA adducts suggests effective AhR induction of BM detoxifying reactions that prevents their formation from dihydrodiols. These findings contrast with the substantial hepatic CYP1A1 contribution for PAHs previously seen for intragastric administration where first pass elimination limits the amount of PAHs reaching the BM.

Introduction

The role of cytochrome P450s (CYP) in the toxicity of xenobiotics, including drugs and environmental contaminants depends substantially on the ratio of bioactivation to detoxification. The CYP1 family members (1A1, 1A2 and 1B1) are of particular interest for their ability to metabolize polycyclic aromatic hydrocarbons (PAHs) to reactive metabolites. These CYP1 enzymes are expressed in a tissue-specific manner and are typically induced following aryl hydrocarbon receptor (AhR) activation by environmental toxicants such as polychlorinated biphenyls (PCBs), dioxin, and PAHs (Hankinson, 1995, Rowlands and Gustafsson, 1997). Benzo(a)pyrene (BP) and 7,12-dimethylbenz(a)anthracene (DMBA) are two prototypic PAHs known to be carcinogenic and immunosuppressive (Davila et al., 1999). Each PAH can be bioactivated by two CYP-dependent steps, which lead to the production of the ultimate carcinogen, PAH-dihydrodiol-epoxide (PAHDE) (Christou et al., 1994, Gelboin, 1980, Miyata et al., 1999, Sticha et al., 2000). In addition, alternate pathways can form toxic quinones that can contribute to toxicity if not removed (Jaiswal, 2000, Moorthy et al., 2003). Metabolism to phenols and dihydrodiols can lead to effective excretion via Phase II sulfotransferase or glucuronidation reactions (Ritter, 2000, Zheng et al., 2002).

The extrahepatic distribution of these CYPs forms is particularly important with respect to bioactivation. Although CYP1A1 is the most active member of the CYP1 family in the metabolism of many PAHs (Guengerich, 1988, Savas et al., 1997), this form is typically expressed at negligible levels unless induced by AhR activation (Bhattacharyya et al., 1995, Whitlock, 1999). Unlike CYP1A1, CYP1B1 is constitutively expressed in three types of extrahepatic tissues: (1) steroidogenic; (2) epithelia of dermal origin; and (3) mesodermal cells. In the latter, AhR activation can also lead to a modest increase in CYP1B1 (Bhattacharyya et al., 1995, Heidel et al., 1998, Murray et al., 2001).

PAHs binding affinity for AhR determines the induction of CYPs (Denison and Nagy, 2003, Wilson et al., 1984). This induction may limit the distribution and duration of PAH effects, as compared with the potent ligand 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), which is not metabolized. Hepatic metabolism may have a large impact on toxicity at peripheral sites such as the bone marrow (Legraverend et al., 1983, Uno et al., 2004). We have previously demonstrated that wild-type (AhR^b) mice treated with DMBA are susceptible to bone marrow cell depletion but are resistant to BP treatment. In contrast, congenic mice homozygous for the AhR^d allele exhibit an enhanced bone marrow toxicity for BP that is similar to levels seen for DMBA (Galvan et al., 2003). This difference between BP and DMBA may be due to the higher binding affinity of BP for the AhR resulting in more effective induction of hepatic CYP1A1. DMBA, which is a less potent AhR inducer of CYP1A1, bypasses hepatic metabolism and reaches peripheral sites, such as the bone marrow.

The importance of peripheral metabolism has been demonstrated by the key role of CYP1B1 in extrahepatic cancers (Buters et al., 2003). CYP1B1 is only minimally expressed in the liver (Bhattacharyya et al., 1995, Shimada et al., 2003). Previous studies with CYP1B1-null mice have shown that they are protected against the DMBA-induced bone marrow toxicity and lymphoblastoma that were observed in wild-type mice (Buters et al., 1999). CYP1B1-null mice retain the ability to induce hepatic CYP1A1, indicating that this is not a source of DMBA-induced bone marrow toxicity (Heidel et al., 2000). We have shown that CYP1B1 is expressed in the bone marrow and its induction is AhR-dependent (Galvan et al., 2003). This is consistent with a role for locally expressed CYP1B1 in bioactivation of DMBA to reactive metabolites within the bone marrow. In support of this model, BP and DMBA exhibit similar toxicity for preB cells cultured with adherent bone marrow stromal (BMS) cells in vitro. This likely involves intercellular transfer of proteins as well as bioactivation by the stromal cells (Allan et al., 2003, Near et al., 1999).

The relationship between CYP1B1 and bone marrow toxicity correlates with DMBA-dihydrodiol-epoxide-DNA adducts in the bone marrow (Heidel et al., 2000). Recent work has shown that CYP1B1 is essential to generate PAH-DNA adducts in many peripheral tissues (ovary, testis, uterus and skin) (Buters et al., 2003, Evans et al., 2004). However, lung tumor formation paradoxically increased for CYP1B1-null mice while other tumors decreased (Buters et al., 2003). Other enzymes, most notably CYP1A1, may mediate the carcinogenic response in susceptible tissues of CYP1B1-null mice. CYP1B1 in adjacent tissues of wild-type mice may even be protective against DMBA-mediated carcinogenesis if expressed in less susceptible cells.

In this study, we address the role of the induction of CYP1A1 and CYP1B1 by BP and DMBA in relation to the delivery of these PAHs to bone marrow and their subsequent toxicity. We have used intraperitoneal (i.p.) administration of PAHs to bypass the gastrointestinal tract thereby decreasing the first pass elimination in the liver. This allowed us to place more emphasis on the relative effects of these PAHs in the bone marrow. In situ hybridization is used to assess the mutually exclusive induction of CYP1A1 and CYP1B1 in the liver and bone marrow, respectively. We also examine lung as a tissue with immediate access to PAHs that expresses both CYP1A1 and CYP1B1. In liver and lung, the differences in CYP1A1 and CYP1B1 mRNA shown by in situ analyses are compared to protein levels measured by immunoblots. These CYP expression measurements assess the tissue and cell selectivity for AhR induction by activators of different affinity [BP >> DMBA]. The role of the AhR in the distribution and toxicity of these PAHs has been assessed by treatment of congenic mice that express an AhR variant with reduced affinity to PAHs (AhR^b >> AhR^d).

DNA-adduct analyses provide a measure of reactive metabolites generated from the PAHs in the bone marrow. Here, we show that differences in PAH-dihydrodiol-epoxide-DNA adducts are directly associated with bone marrow toxicity. Importantly, these differences are not determined by bone marrow CYP1B1 induction or by differences in delivery of PAHs to the bone marrow. This work suggests that diminished BP bone marrow toxicity is inversely related to AhR activity and that local alternative metabolism of PAH-dihydrodiols is critical to toxicity.