Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

P450¹ monooxygenases comprise a large gene superfamily known to catalyze the oxidation of a wide variety of lipophilic drugs, endogenous substances, and environmental agents (Nelson et al., 1996). Although most of the products of metabolism are biologically less active, the involvement of P450s in the generation of electrophilic, cytotoxic, mutagenic, and carcinogenic metabolites from chemically stable parent compounds is well established (Gillette, 1995). The lung represents one of the major targets for exposure to environmental chemicals, not only because it is the primary site for the entrance of airborne agents but also because this tissue receives 100% of the cardiac output. Despite the fact that P450 enzyme levels are much lower in lung than in liver, a number of lung-toxic chemicals have been demonstrated to undergo metabolic activation via pulmonary P450 monooxygenases (Yost, 1997). CYP4B1, for example, converts 4-ipomeanol to reactive metabolites that appear to be associated with extensive lung damage (Boyd, 1977; Slaughter et al., 1983; Verschoyle et al., 1993).

Unlike that of the liver, the structural and cellular composition of the lung is highly heterogeneous. Chemical injury to the lung tends to be highly focal, which appears to be related, in part, to the cellular distribution of pulmonary P450s. Studies using immunohistochemistry, in situ hybridization, and isolated cells have revealed that, among the >40 cell types found in the lung, P450s are highly localized in only a few cell types, including Clara cells, alveolar type II cells, endothelial cells, and macrophages (Serabjit-Singh et al., 1988; Voigt et al., 1990; Overby et al., 1992; Forkert, 1995). Those studies support the view that differential expression of P450s is a key determinant in the selective toxicities of chemicals in target cell populations in the lung. However, although immunohistochemistry and in situ hybridization experiments yield excellent data on the distribution of antigenic protein and mRNA, these approaches do not assess the catalytic function of P450s. Although there have been several studies using isolated pulmonary cells to assay P450 activities (for review, see Devereux et al., 1993), those do not provide information regarding regional differences in P450 activities within the respiratory tract. In addition, isolated cells are recovered after enzymatic digestion, and the presence of proteolytic products in the solution raises the possibility that the products may affect catalytic activities. Previous work delineating regional differences in P450 activities has been based on the use of S9 supernatants (Gebremichael et al., 1995) or has been conducted in large animals such as dogs (Bond et al., 1988). Although studies with rats have demonstrated regional differences in P450 activities (Gebremichael et al., 1995), dilution by cytosolic proteins may confound the activity data.

The high degree of susceptibility of mice, compared with rats and hamsters, to both naphthalene- and 1,1-dichloroethylene-induced Clara cell toxicity has been ascribed to species differences in CYP2F (Buckpitt et al., 1992; Ritter et al., 1991) and CYP2E1 (Dowsley et al., 1996), respectively. The marked difference in species susceptibility to P450-dependent toxicity has highlighted the need to evaluate the metabolic fate of these toxicants in species similar to humans. Lungs of nonhuman primates, such as rhesus monkeys, are anatomically and cellularly similar to those of humans (Tyler and Plopper, 1985; Ten Have-Opbroek and Plopper, 1992) and thus could provide potential alternatives to human studies. However, our knowledge of xenobiotic metabolism in primates is limited. Earlier studies with microsomes from monkey (Buckpitt et al., 1992) and human (Shimada et al., 1992) whole-lung homogenates showed very small quantities of P450 monooxygenases. However, the distribution and expression of pulmonary P450 monooxygenase activities in primates remain unexplored.

Hepatic P450 monooxygenases are up-regulated by a number of environmental agents. In contrast, with the exception of CYP1A1, pulmonary monooxygenases appear to be refractory to induction. Previous studies on P450 inducibility in the lung were based on the use of either whole-lung homogenates (Serabjit-Singh et al., 1983), immunohistochemistry, or in situ hybridization (Forkert, 1995; Keith et al., 1987). Recent investigations in our laboratories have shown that exposure to sidestream tobacco smoke up-regulates P450 activities with a high degree of regioselectivity (Lee et al., 1996). Whether this results from differences related to the area of particle impact or inherent differences in the regulatory mechanisms within the lung is not certain. The goals of the present study were to measure P450 enzyme activities, using isoform-selective substrates, in different lung subcompartments of rats and monkeys and to determine whether the prototypical enzyme inducers -naphthoflavone, phenobarbital, and acetone altered activities in well-defined regions of this organ in rats.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Animals and Treatment. Adult male Sprague-Dawley rats were purchased from Zivic Miller Laboratories (Zelienople, PA). The animals were maintained on a 12/12-hr light/dark cycle, in wire-top metal cages, and were allowed free access to food and water throughout the study. Upon receipt, all animals were kept in our animal care facility for at least 1 week before use. -Naphthoflavone (Sigma Chemical Co., St. Louis, MO) was dissolved in corn oil and administered by ip injection (at 80 mg/kg body weight) 72 and 48 hr before euthanasia. A corresponding amount of corn oil alone was administered to the control group. Groups of rats were given 5% acetone (Fisher Scientific, Fair Lawn, New Jersey) or phenobarbital (0.1%) in the drinking water for 7 days or 14 days, respectively; the control group received regular drinking water. Rhesus monkeys (seven animals; age, 0.75-9.7 years) were obtained from the California Regional Primate Research Center, University of California, Davis.

Airway Microdissection. The procedure to obtain defined lung specimens was described previously (Plopper et al., 1991). Briefly, rats were euthanized with an overdose of sodium pentobarbital (administered ip), and the trachea was cannulated. The lung was separated from the body and inflated with 1% low-melting-point agarose (FMC Bioproducts, Rockland, ME). The inflated lung was quickly immersed in ice-cold Waymouth's medium (Gibco Laboratories, Greenland, NY). The monkeys were euthanized between 10:00 a.m. and 2:00 p.m. Lungs were rinsed, and one lobe was immersed in ice-cold Waymouth's medium. Under a dissecting microscope, the airways were separated from the blood vessels and the parenchyma. Because of the availability of sufficient amounts of tissue from rhesus monkey lungs, different segments of airway subcompartments, including the major daughter, minor daughter, and distal bronchioles, were obtained.

Microsomal Preparation. All steps were carried out at 0-4°C. Microsomes were obtained by differential centrifugation using standard techniques. Microsomes were resuspended in 0.1 M phosphate buffer (pH 7.4), and protein concentrations were determined with the micro Bio-Rad assay (Bradford, 1976). Standard curves were prepared using bovine serum albumin (Sigma).

CYP1A1, CYP2B, and CYP2E1 Activity Assays. The activities of CYP1A1 and CYP2B enzymes were measured by O-dealkylation of ethoxyresorufin and pentoxyresorufin (Sigma), respectively, using methods modified from the work of Rettie et al. (1986). Ethoxyresorufin, as obtained from the supplier, was contaminated with small amounts of resorufin. Therefore, the substrate was purified by preparative HPLC before use. The total incubation volume was 200 µl. Microsomes (25 µg protein) were mixed with each substrate (1 µM), and the reaction was initiated by addition of an NADPH-generating solution (0.14 mM NADP, 3.8 mM glucose-6-phosphate, 0.1 unit of glucose-6-phosphate dehydrogenase, and 10 mM MgCl₂). The incubations were conducted for 15 min at 37°C and terminated by addition of 2 volumes of ice-cold methanol. Samples were stored overnight at 20°C to precipitate protein. After centrifugation, the supernatants were used for determination of the amount of resorufin by HPLC, using a fluorescence detector set at an excitation wavelength of 535 nm and an emission wavelength of 585 nm, as described previously (Plopper et al., 1993). Standard curves were prepared by using a series of known concentrations of resorufin.

P450 Reductase Activity Assay. Microsomal P450 reductase was measured spectrophotometrically, as cytochrome c reductase activity, by standard methods (Guengerich, 1994). Changes in absorbance were measured at 550 nm at 30°C, after addition of NADPH.

Data Processing and Statistical Analysis. The results of all P450 monooxygenase activity assays are expressed as picomoles of product formed per milligram of microsomal protein per minute of incubation. P450 reductase activity is expressed as nanomoles of cytochrome c reduced per milligram of microsomal protein per minute. All data are presented as mean ± 1 SD. Comparisons of enzyme activities among lung compartments and/or subcompartments were performed using one-way analysis of variance. Post hoc tests, using Bonferroni/Dunn methods, were performed to define differences in P450 isoform activities in different compartments and/or subcompartments. Data from rat P450 induction studies were analyzed using two-tailed Student t tests. A p value of <0.05 was considered statistically significant.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

P450 Isoform Activities in Rat and Monkey Lung Compartments. Preliminary studies were conducted to assess the linearity of ethoxyresorufin and pentoxyresorufin O-dealkylation with time and microsomal protein concentration. All subsequent assays were conducted within the linear portion of the product/time and product/protein curves (data not shown). CYP1A1 (ethoxyresorufin O-dealkylase) activities were highest in rat lung parenchyma and 3-fold and 6-fold lower in the airways and trachea, respectively (fig. 1). Similarly, CYP2B1 (pentoxyresorufin O-dealkylase) activities were highest in the parenchyma; the airways and trachea exhibited 60% and 15%, respectively, of the parenchymal activities. In contrast, CYP2E1 activities were 6-fold higher in airway microsomes than in microsomes prepared from parenchyma. Activities in the trachea were below the level of detection (fig. 1).

View larger version (28K): [in this window] [in a new window]   Fig. 1.   Ethoxyresorufin, pentoxyresorufin, and p-nitrophenol metabolism and cytochrome c reductase activities in microsomal incubations from subcompartments of rat and monkey lung. Values are the mean ± SD from four animals, with the exception of the reductase measurements in monkey lung, where data are from three animals. TR, trachea; ARWY, airway; PAR, parenchyma; MAJ, major daughter axial pathway; MIN, minor daughter axial pathway; DB, distal bronchiole.

P450 Reductase Activity in Rat and Monkey Lung Compartments. P450 reductase activities were significantly lower in the trachea than in the airways and parenchyma of rats (fig. 1). In rhesus monkey lung, microsomal P450 reductase activities were relatively similar in all airway subcompartments measured.

Induction of Rat CYP1A1, CYP2B, and CYP2E Activities by -Naphthoflavone, Phenobarbital, or Acetone. Administration of -naphthoflavone at doses of 80 mg/kg (72 and 48 hr before euthanasia) resulted in 10-fold induction of hepatic microsomal CYP1A1 activities, as well as significant up-regulation of CYP1A1 in some subcompartments of the lung. Thirty-fold increases were observed in the airways and parenchyma of treated animals, compared with controls (table 1).

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The studies described here were designed to compare key xenobiotic-metabolizing enzyme activities in defined regions of the lungs of rats and monkeys and to determine whether up-regulation of the proteins in the lung is regioselective, using rats as a model. The striking species and regional differences in susceptibility to a variety of lung toxicants that undergo P450-dependent metabolic activation, combined with earlier work showing that human lung P450 monooxygenase activities are 20-50-fold lower than activities in rats, raise questions regarding the relevance of studies in rodent models to lung toxicity in humans (Buckpitt and Cruikshank, 1997) and underscore the need for comparative studies of defined regions of the lung. Although the metabolism of various agents has been investigated in human lung tissue, there is little information on the regional distribution of these xenobiotic-metabolizing enzyme activities in either human lungs or lungs of species that are anatomically similar to humans, such as rhesus macaques. The possibility that the P450 monooxygenases are highly localized and/or are highly up-regulated in a small subset of human lung cells could make these cells especially vulnerable to toxicants activated by the P450 monooxygenase system. High regional activities would necessarily be diluted by the use of whole-lung homogenates for microsomal preparations.

The finding that CYP1A1 and CYP2B activities are 3- and 1.6-fold higher, respectively, in the parenchyma than in other regions of the rat lung is somewhat surprising, in light of the relatively high sensitivity of airway Clara cells to a number of P450-activated cytotoxicants. Previous activity assays with Clara cells and isolated alveolar type II cells, as well as immunohistochemical studies using antibodies to CYP1A1 and CYP2B, demonstrated higher protein concentrations per cell (Jones et al., 1983; Devereux et al., 1993) in rat Clara cells and more antibody staining in Clara cells within airways than in the alveolar type II cells in the lung parenchyma (Domin et al., 1986). Similarly, CYP2B levels were 30% higher in isolated rabbit Clara cells than in alveolar type II cell preparations (Domin et al., 1986). However, our results are similar to the findings from studies of isolated dog airways, demonstrating higher ethoxycoumarin O-deethylase activities in peripheral lung than in either terminal or more proximal airways (Bond et al., 1988). The relatively higher activities of CYP1A1 and CYP2B in the parenchyma of rats were not observed in monkeys, where relatively similar specific activities were observed in all lung subcompartments examined. Although all P450 isoform activities were assayed using microsomes from the same group of monkey lung tissues, no clear association was observed for activity levels of the different isoforms or airway generations in any single animal.

In comparison with the distribution of CYP1A1 and CYP2B activities, CYP2E activities were concentrated in the airways of rats and in the distal bronchioles and parenchyma of monkeys. The distribution of this isoform in rats appears similar to the distribution of naphthalene monooxygenase in mice, an activity ostensibly of CYP2F2 (Ritter et al., 1991). In the current studies, CYP2E1 activities were not detected in microsomes prepared from rat tracheas; however, small amounts of activity are present (Watt et al., 1997), which were not observed under the conditions of assay used here.

The apparent differences in P450 protein levels assessed by immunohistochemistry and Western blotting or by activity assays in microsomes isolated from Clara and type II cells likely reflect differences in the abundance of metabolically active cells in particular lung subcompartments and emphasize the need to monitor P450 proteins using several different approaches. Isolation of cells suffers from the requirement for enzymatic digestion of the lung, a process that has been shown to result in some proteolysis of the P450 proteins (Devereux et al., 1993). As pointed out by Philpot (1993), measurements of microsomal activities assume similar purities of the microsomal preparations from each of the cell types or, as in the present experiments, from each of the lung subcompartments. The diversity of cell populations present in the airway means that the contribution of endoplasmic reticulum from metabolically active cells to the microsomal fraction may differ, thus leading to apparent differences in the specific metabolic activities observed. It is also likely that these differences contribute to the variations in activities noted in different preparations.

In addition, it is important to acknowledge the fact that the substrates used in a study such as this are isoform selective and that the selectivity has been adequately demonstrated in only a few species (Rettie et al., 1986). There are numerous examples in which mutation of a single amino acid in a P450 markedly alters the substrate specificity. Thus, the data obtained with rhesus macaques assume that, at the substrate concentrations used, CYP1A1, CYP2B, and CYP2E1 are the primary isoforms involved in the turnover of ethoxyresorufin, pentoxyresorufin, and p-nitrophenol, respectively. It is also probable that CYP1B1, an isoform found in the lungs of rats, mice, and humans (Savas et al., 1994; Willey et al., 1986), contributes to the deethylation of ethoxyresorufin. In recent studies using human recombinant proteins, Shimada et al. (1997) showed that CYP1B1 metabolizes this substrate at slightly less than 10% of the rate of metabolism catalyzed by CYP1A1. CYP1B1 is present in human lung and is likely to be in primate lung; therefore, this protein could also make a significant contribution to ethoxyresorufin metabolism.

Earlier work demonstrated that sidestream tobacco smoke exposure selectively up-regulates rat CYP1A1 in airways and parenchyma but not tracheal subcompartments of the lung (Lee et al., 1996). CYP1A1 activities in the airway appeared to be more sensitive to induction, because the activities were up-regulated in this region at lower doses than were required in the parenchyma. What was not clear from those studies was whether the regioselective effects of cigarette smoke were dependent on regional differences in the deposition of particles containing polycyclic aromatic hydrocarbon inducers or whether the effects involved an intrinsic property related to Ah receptor levels in different airways. The current work showing that pulmonary CYP1A1 responds in a regioselective fashion to a systemically administered inducer in rats suggests that particle deposition may not be the primary determinant of regioselective induction of CYP1A1 by environmental tobacco smoke. The recent finding that the Ah receptor is differentially distributed in the liver (Lindros et al., 1997) is consistent with the view that variations in the distribution of the receptor could control the regioselectivity of CYP1A1 induction in the lung. Those studies also provide convincing evidence that the lack of up-regulation of CYP2B and CYP2E proteins by prototypical inducers in whole-lung microsomal studies is not the result of dilution with microsomes from cells that do not contain these proteins. The fact that CYP2E1 is strikingly up-regulated in the nasal olfactory epithelium but apparently is not in the more distal portions of the respiratory tract suggests that the regulation of this protein is very different in these two areas. It is possible that the proteins in the lung are at maximal levels and therefore are not susceptible to up-regulation. Additional studies should apply the techniques described in this and other reports to examine the hypothesis that the P450 monooxygenases in human lung are highly localized in selected subcompartments of the lung and that the up-regulation of P450 proteins by cigarette smoke occurs in a highly regioselective manner.