Vol. 9, 815-825,
September
Cell Growth & Differentiation
1998
Regulation of Apoptosis in Mouse Hepatocytes of Apoptosis by Nongenotoxic Carcinogens’
James Russell
G. Christensen,2 Andrea C. Cattley, and Thomas
J. Gonzales,3 L Goldsworth?
Chemical
Industry Institute of Toxicology, Research Triangle Park, 27709 [J. G. C., A. J. G., A. C. C., T. L G.]; Department of Toxicology, North Carolina State University, Aalelgh, North Carolina North
Carolina
27695 [J. G. C.]; and Curriculum in Toxicology, University Carolina, Chapel Hill, North Carolina 27599 [A. J. G.]
of North
Abstract Regulation of apoptosis is an important component of multistage hepatocarcinogenesis. The objectives of the present study were to characterize apoptosis regulation in primary mouse hepatocytes and to determine whether nongenotoxic carcinogens after apoptosis regulation. Bleomycin-induced apoptosis was accompanied by decreases in bcl-2 and bcl-xL and increases in p53, bak, and bax protein levels. Transforming growth factor fI’G9-fi-induced apoptosis was accompanied by decreased bcl-xL and increased bak. Bleomycin-induced apoptosis was partially dependent on p53, whereas TGF--induced apoptosis was independent of p53. Phenobarb’ital inhibited both TGF-13 and bleomycin-induced apoptosis and the normal regulation of p53, bcl-2, and bax. Nafenopin inhibited apoptosis through a mechanism dependent on PPAR-a and inhibited the normal regulation of bcl-2 and bak. 2,3,7,8-Tetrachborodibenzo-p-dioxin did not alter apoptosis or its regulation. Apoptosis was increased in hepatocytes from bcl-2-null mice, which indicated that the bcl-2 family contributes to hepatocyte apoptosis regulation. This study demonstrated that apoptosis regulation in mouse hepatocytes involves distinct pathways and that diverse nongenotoxic carcinogens differentially after molecular pathways that represent targets for hepatocarcinogenesis. Introduction Carcinogen risk assessment suIts from long-term rodent
Received
5/1 1/98;
The costs
revised
7/1/98;
often relies heavily on the recancer bioassays of selected
accepted
7/13/98.
of publication
of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to mdi-
cate this fact. 1
Supported
by National
Institute
of Environmental
Health
Sciences
Train-
ing Grant ES-07046-21 (to J. G. C.). 2 To whom requests for reprints should be addressed,
at CIIT, 6 Davis 12137, RTP, NC 27709-2137. Phone: (919) 558-1362; Fax: (919) 558-1300; E-mail:
[email protected]. 3 Present address: Parke-Davis, Department of Cancer Research, Ann Arbor, Ml 48105. Drive,
P. 0. Box
Present address: Park, NC 27709.
4
Integrated
Laboratory
Systems,
Aesearch
Triangle
and Alteration
chemicals. To better understand the risk and extrapolate rodent cancer bioassay data to predict effects from human exposure to chemicals, much emphasis has been placed recently on determining molecular mechanisms by which chemical exposure results in the development of cancer. The most frequently observed end point in rodent bioassays is liver neoplasia (1 , 2). A large number of chemicals that cause liver tumors in rodent bioassays do not directly mutate DNA and are classified as nongenotoxic (3, 4). Several of these agents, identified as possessing tumor-promoting activity, are believed to affect the cancer process through alterations in the regulation of cellular growth and survival. (5-8). One hypothetical mechanism of nongenotoxic carcinogenesis is the alteration of molecular regulation of apoptosis, which results in the decreased ability of a cell to respond to normal apoptotic stimuli and thus contributes to cellular transformation and tumor promotion and progression. Alteration of the regulation of apoptosis has been implicated in both human and rodent carcinogenesis (9, 10). Individual cells must integrate survival and death-inducing stimuli to determine their fate in the context of the organism. Important consequences of apoptosis include the elimination of cells that have accumulated genetic damage or cells that inappropriately express growth control proteins. Resistance to normal apoptotic stimuli may result in the accumulation of cells with genetic damage and genomic instability. In addition to dysregulation of apoptosis, the alteration of proliferation and cell cycle control may further result in the growth and clonab expansion of cells with genetic damage. Genetic damage, genomic instability, and altered cell growth signaling are all hallmark features of carcinogenesis. Certain nongenotoxic rodent liver carcinogens (e.g., phenobarbital and nafenopin) have been shown to promote cell survival and inhibit apoptosis both in vivo and in vitro (5-8). These carcinogens promote the growth of both normal and neoplatic cells. The inhibition of apoptosis in normal hepatocytes by nongenotoxic carcinogens could result in the accumulation of cells with genetic damage and altered cellular growth control and an initiated cell population. The continued inhibition of apoptosis by nongenotoxic carcinogens in neoplastic cells could result in further accumulation of genetic alterations and progression of neoplasia to malignancy. Preneopbastic foci, adenomas, and carcinomas exhibit increased rates of cell death relative to surrounding hepatocytes (9-1 1). Phenobarbital and nafenopin were found to reduce relative rates of cell death in both normal liver and hepatic neoplasia and thus accelerate growth and progression of cancer (5-8). Certain nongenotoxic carcinogens have also been shown to increase liver size and promote development of liver lesions in rodents. Withdrawal of certain nongenotoxic carcinogens, including phenobarbitab and nafenopin, has been shown to result in regression of hyperplastic liver, preneoplastic foci, adenomas, and in
815
816
Apoptosis
Regulation
and Nongenotoxic
Carcinogens
some reports, carcinomas, all with a concomitant increase in apoptosis (5-8, 12, 1 3). This regression is reversible upon readministration of these compounds (6, 8). In addition to in vivo effects, some nongenotoxic carcinogens also inhibited apoptosis, degeneration, and dedifferentiation in primary hepatocyte cultures (1 4-1 8). These studies indicate that apoptosis is important in rodent hepatocarcinogenesis and that certain nongenotoxic carcinogens alter apoptosis regulation in hepatocytes. Much attention has been given recently to the identification of molecular
targets
that
are involved
in the regulation
of
apoptosis and are altered in cancer. bcl-2 is a protein that was originally identified at the breakpoint of transbocations commonly occurring in follicular B-cell lymphomas that resuited in its overexpression (19-21). This protein was found to play a role in cell survival and inhibition of apoptosis rather than regulation of cell proliferation (22-24). Subsequent studies have
resulted
in the
identification
of a family
related homobogues that contribute to the grammed cell death in many species and members include apoptosis-inducing (e.g., inhibiting (e.g., bcb-xL) proteins that form heterodimers with themselves and other (25-28).
The
precise
mechanism
by which
of bcl-2-
regulation of procell types. Family bax and bak) and homodimers
family
and
members
these
proteins
regulate cell death is presently under investigation. However, the cellular equilibrium between inducing and inhibiting family members is known to be important in determining cell fate (26, 29). Both programmed cell death and expression of members of the bcl-2 family have been shown to be regubated through cellular signaling processes. For example, activation of p53 in response to DNA damage results in a p53-dependent change in bcl-2 family gene expression and apoptosis, whereas deletion of p53 from cells results in the inability
of cells
with
damaged
DNA
to undergo
apoptosis
(30, 31).
Two distinct types of stimuli, i.e., TGF5-f3 and DNA-damaging agents, have previously been shown to induce apoptosis in primary rat hepatocyte cultures (1 5, 32, 33). TGFsignaling is known to be important in hepatic apoptosis and growth regulation as well as hepatocarcinogenesis (34-36). Expression of TGF-f3 and its receptors has been shown to be altered in human and rodent hepatic neoplasia (37-39). Mutations and loss of heterozygosity in the TGF-13 processing M6P/lGFlb receptor have also been noted in human hepatocellular carcinoma (40, 41). The cellular response to DNA damage is also very important in carcinogenesis, as evidenced by the loss of p53 function in addition to other genes responsive to DNA damage in several cancers (31 , 42, 43). The objectives for the present investigation of the molecular regulation of apoptosis in primary mouse hepatocytes and its alteration by nongenotoxic carcinogens were to: (a) characterize apoptosis and its regulation by p53 and bcl-2 family members in primary mouse hepatocytes; (b) determine whether nongenotoxic chemicals alter apoptosis ragulation; and (c) determine whether p53 and bcl-2 actively
The abbreviations used are: TGF, transforming growth 2,3,7,8-tetrachlorodibenzo-p-dioxin; PVDF, polyvinylidene 5
factor; TCDD, difluoride.
contribute
to the apoptotic response using bcb-2-nubl and primary hepatocytes. Selected nongenotoxic carcinogens (nafenopin and phenobarbital, but not TCDD) inhibited apoptosis induced by TGF-j3 or bleomycin in mouse hepatocytes. Furthermore, distinct pathways of apoptosis (p53-dependent and -independent, PPAR-a-dependent) in mouse hepatocytes were identified, and bcb-2 family member regulation was found to be altered by apoptotic stimuli and nongenotoxic carcinogens. These data indicate that molecular regulation of apoptosis represents an important target in the mechanism of hepatocarcinogenesis by nongenotoxic chemicals. p53-null
Results and Quantitation of Apoptosis in Mouse Hepatocytes. To study the regulation of optosis, the response of primary mouse hepatocytes to optotic stimuli was evaluated by determining the number cells that stained positively using differential staining and
Priapapof had
morphological
(de-
Characterization mary
characteristics
associated
with
apoptosis
scribed in “Materials and Methods”) or adapted in situ endlabeling (Fig. 1 ; Refs. 44-47). The basal level of apoptosis in untreated culture remained at