View as PDF
Robert R. Maronpot, Katsuhiko Yoshizawa, Abraham Nyska, Takanori Harada, Gordon Flake, Gundi Mueller, Bhanu Singh, and Jerrold M. Ward
View as PDF

Hepatic enzyme induction is generally an adaptive response associated with increases in liver weight, induction of gene expression, and morphological changes in hepatocytes. The additive growth and functional demands that initiated the response to hepatic enzyme induction cover a wide range of stimuli including pregnancy and lactation, hormonal fluctuations, dietary constituents, infections associated with acute-phase proteins, as well as responses to exposure to xenobiotics. Common xenobiotic enzyme inducers trigger pathways involving the constitutive androstane receptor (CAR), the peroxisome proliferator-activated receptor (PPAR), the aryl hydrocarbon receptor (AhR), and the pregnane-X-receptor (PXR). Liver enlargement in response to hepatic enzyme induction is typically associated with hepatocellular hypertrophy and often, transient hepatocyte hyperplasia. The hypertrophy may show a lobular distribution, with the pattern of lobular zonation and severity reflecting species, strain, and sex differences in addition to effects from specific xenobiotics. Toxicity and hepatocarcinogenicity may occur when liver responses exceed adaptive changes or induced enzymes generate toxic metabolites. These undesirable consequences are influenced by the type and dose of xenobiotic and show considerable species differences in susceptibility and severity that need to be understood for assessing the potential effects on human health from similar exposures to specific xenobiotics.


hepatocellular hypertrophy, liver toxicity, hepatocellular hyperplasia, liver carcinogenesis, xenobiotic enzyme inducers, P450.


Hepatic enzyme induction may be associated with changes in liver weight, histological evidence of abnormal hepatocytes, alteration of serum clinical chemistry analytes (Ennulat et al. 2010), and pleotrophic gene expression. General background information relative to enzyme induction as covered in the introductory overview is typically compartmentalized into Phase I and Phase II induction (Botts et al. 2010). Phase I oxidative metabolism is catalyzed by various isoforms of the P450 superfamily and occurs in the microvesicles of the hepatocyte smooth endoplasmic reticulum. The classification and functions of the large family of P450 (CYP) enzymes have been reviewed with identification of species-specific isoforms (Martignoni et al. 2006).

The liver is responsible for maintenance of normal homeostasis and physiological functions. It functions as a conditional system capable of relatively rapid responses to a variety of stimuli. Liver size is governed both by genetic factors and by the rate of biochemical activity to maintain optimal functional mass. Following stimuli such as toxic insult, infection, or partial hepatectomy, the liver rapidly restores its optimal mass to maintain normal function. The liver also readily responds to some stimuli by undergoing additive growth and function. Growth and functional demands that promote a liver response span a range of stimuli, including pregnancy and lactation; hormonal fluctuations; dietary constituents such as fat, carbohydrates and proteins; viral and bacterial infections that induce acute-phase proteins; and enzyme inductive responses to a variety of xenobiotics. The hepatic response to these various stimuli may involve an increase in liver size and functional capacity attributable to an increase in size and/or number of hepatocytes.

This review addresses factors influencing common hepatic enzyme inducers and the morphological features of hepatic enzyme induction including adaptive and adverse responses.

Common Xenobiotic Induction Pathways and Effect On Liver Morphology

Induction Involving Constitutive Androstane Receptor

Barbiturates and chemicals that induce CYP2B initially interact with the constitutive androstane receptor (CAR) that translocates to the nucleus and dimerizes with the retinoid-X-receptor (RXR). The dimer then binds to specific response elements, resulting in transcriptional activation of genes regulating P450 expression (Liang et al. 1995; Sueyoshi and Negishi 2001). Phenobarbital is a prototypical xenobiotic that acts through the CAR receptor. Phenobarbital-induced liver enlargement is associated with initial transient hyperplasia in rodents as an early event after treatment and a substantial proliferation of smooth endoplasmic reticulum, causing hepatocellular hypertrophy. The hypertrophy typically affects all of the centrilobular hepatocytes and 45% of periportal hepatocytes in rats (Massey and Butler 1979). Differences in the response may be influenced by the species, sex, age of the animals, and the barbiturate dose (Crampton et al. 1977a; Massey and Butler 1979; Schlede and Borowski 1974). Increasing the dose of phenobarbitone in the rat will correlate with liver enlargement and enzyme induction only up to a point, after which there is no further increased liver enlargement and hepatotoxicity observed (Crampton et al. 1977a). Prolonged phenobarbitone exposure results in induction of hepatocellular neoplasms in mice and rats (Greaves 2007; Holsapple et al. 2006; Kunz et al. 1983; Wastl et al. 1998; Whysner et al. 1996).

Following administration of phenobarbital via drinking water to F344 rats for three weeks, increased liver weight was found to be a result of hepatocellular hypertrophy, hyperplasia, and enlargement of the hepatic blood space (Massey and Butler 1979). The authors report the presence of increased rough endoplasmic reticulum (RER) in addition to smooth endoplasmic reticulum (SER). In the centrilobular hepatocytes, there was substantial proliferation of rough endoplasmic reticulum, as shown by electron microscopy with slightly less than half of the periportal hepatocytes showing rough endoplasmic reticulum proliferation. Most authors associate microsomal hepatic enzyme induction with increases in smooth endoplasmic reticulum, however, this paper indicated both rough and smooth forms were affected in their model.

In addition to inducing P450 enzymes, phenobarbital can also induce glutathione S-transferase A1/A2 (rGSTA1/A2). Twelve hours after a single dose of phenobarbital, induction of rGSTA1/A2 as assessed by in situ hybridization was threefold higher in centrilobular versus periportal hepatocytes in rats (Selim et al. 2000). The half-life of rGSTA1/A2 was increased three-fold by Phenobarbital, and the higher centrilobular induction was attributed to localized stabilization of mRNA transcripts.

Other xenobiotics eliciting hepatic changes similar to phenobarbital in rodents include butylated hydroxytoluene (Crampton et al. 1977a), octamethylcyclotetrasiloxane (McKim et al. 2001), phenytoin, carbamazepine, dieldrin (Kolaja et al. 1996), and chlordane (Malarkey et al. 1995). Reversal of pleotrophic hepatic responses associated with enzyme induction has been documented following discontinuation of treatment. In the case of butylated hydroxytoluene, liver weight and enzyme activity returned to normal levels within thirty days of cessation of treatment (Crampton et al. 1977a). Increased liver weight, increased replicative DNA synthesis, and inhibition of gap-junction intercellular communication induced by a two-week treatment with diethlyhexyl phthalate or phenobarbital returned to control levels within one to four weeks following cessation of treatment (Isenberg et al. 2001). Reversal of hepatic proliferative focal lesion growth and even hepatocellular neoplasms has been reported following cessation of treatment with phenobarbital, dieldrin, and chlordane in these models (Kolaja et al. 1996; Malarkey et al. 1995).

Enzyme Induction Involving the Peroxisome Proliferator-activated Receptor

Peroxisome proliferator-activated receptors (PPAR) are members of the steroid hormone receptor superfamily (Peters et al. 1997), and PPAR-α is associated with pleotrophic responses in rodents, including liver enlargement. Peroxisome proliferator-activated receptor-α is highly expressed in rat and mouse liver, and there is less expression in guinea pigs, nonhuman primates, and humans (Cariello et al. 2005; Klaunig et al. 2003). Several xenobiotics induce PPAR activity, resulting in liver enlargement by stimulating the proliferation of hepatocyte peroxisomes and inducing the fatty acid oxidation enzyme CYP4A in rats and mice (Bentley et al. 1993; Stott 1988). Following activated PPAR binding to the RXR to form a heterodimer, there is binding to multiple specific response elements, leading to transcriptional activation of genes responsible for the pleotrophic effects of peroxisome proliferators (Green 1995; Yu et al. 2003).

As is the case with phenobarbitone-like enzyme induction, peroxisome proliferating compounds are associated with liver tumors in rats and mice following prolonged exposure (Greaves 2007; Peters et al. 1997). As non-genotoxic agents, peroxisome proliferators have demonstrated apparent thresholds for carcinogenesis under the experimental conditions studied (Bentley et al. 1993; Stott 1988). The mechanisms associated with hepatocarcinogenesis include production of reactive oxygen species; early, intermittent, or persistently enhanced cell proliferation and reduced apoptosis; promotion of spontaneously initiated hepatocytes; and combinations of one or more of these mechanisms (Grasso et al. 1991).

Peroxisome proliferator-activated receptor-α knockout mice do not develop hepatocellular neoplasia following long-term exposure to peroxisome proliferators (Lake 1995; Peters et al. 1997). Knockout mice expressing a human PPAR-α receptor developed fewer (5% vs. 71%) liver tumors than did wild-type mice exposed to a peroxisomal proliferator (Wy-14,643), suggesting carcinogenic response is a possible consequence of structural differences in human and mouse PPAR-α(Morimura et al. 2006).

There are important species and strain differences in sensitivity to peroxisome proliferators (O’Brien et al. 2005). In a comparative study of the herbicide oxadiazon, a xenobiotic with peroxisome proliferative activity but structurally distinct from other peroxisome proliferators, in Sprague-Dawley rats, CD1 mice, and Beagle dogs, treatment for fourteen (rats) or twenty-eight (mice and dogs) days produced liver enlargement in all three species (Richert et al. 2008). Although the magnitude of the change in absolute and/or relative liver weights was similar across species, morphological evidence of peroxisome proliferation based on transmission electron microscopy occurred only in rats and mice. Peroxisome proliferation was correlated with increased palmotyl CoA oxidase and acetyl carnitine transferase activities. Cultured rat hepatocytes also had elevated activities of these enzymes, whereas cultured human hepatocytes did not.

Enzyme Induction Involving the Ah Receptor

Halogenated aromatic hydrocarbons including polybrominated biphenyls, polychlorinated biphenyls, various dioxins including the potent 2,3,7,8-tetrachlorodibenzodioxin (TCDD), and dibenzylfurans, as well as the protypical inducer 3-methylcholanthrene, act by binding to the intracellular Ah receptor (AhR) (Carlson and Perdew 2002; Shimada et al. 2003; Whitlock 1993). This action results in multiple effects on the liver, including induction of CYP1A1, CYP1A2 and CYP1B1; hepatocyte hypertrophy; increased cellular replication along with decreased apoptosis; and liver enlargement. Many of the dioxins and other halogenated aromatic hydrocarbons are hepatocarcinogenic in rats.

Enzyme Induction Involving the Pregnane-X-receptor

Cyproterone acetate is a gestagen and antiandrogen that interacts with the nuclear pregnane-X-receptor (PXR) to bring about activation of several genes, including CYP3A. Liver enlargement is apparent after three days in rats (SchulteHermann and Parzefall 1980); DNA synthesis in periportal hepatocytes increased up to twenty-fold in twelve hours after commencement of treatment. The induction of liver growth is primarily by hepatocellular hyperplasia. Chronic treatment of rats results in liver tumors (Schulte-Hermann and Parzefall 1980; Schulte-Hermann, Hoffman, Parzefall et al. 1980; Schulte-Hermann, Hoffman, and Landgraf 1980). Spironolactone, pregnenolone 16a-carbonitrile, and other xenobiotics also act by binding PXR, with results similar to cyproterone acetate (Kliewer 2003; Schuetz et al. 1998; Schulte-Hermann and Parzefall 1980). In humans, CYP3A is readily induced by xenobiotics, is activated by CYP3A inducers, and leads to activation of PXR (Kliewer 2003; Schuetz et al. 1998).

In summary, liver enlargement can be the consequence of hepatocellular hypertrophy, hepatocellular hyperplasia, or both in response to enzyme induction. For example, α-hexachlorocyclohexane, phenobarbital, and 3-methylcholanthrene all increase liver size in rodents primarily owing to hypertrophy, in contrast to enzyme induction involving PXR mechanisms (Batt et al. 1992; Green 1995; Glaister 1986; Porter and Coon 1991; Schulte-Hermann 1974). However, xenobiotics that cause liver enlargement primarily by hepatocellular hypertrophy may also produce transient (phenobarbital-like xenobiotics) or sustained (WY-14,643) hepatocyte proliferation. It has also been shown that enzyme induction of phase II enzymes can lead to hepatomegaly without hepatocellular hypertrophy (Makino et al. 1998; Makino et al. 2008). Both tert-butylated hydroxyanisole (BHA) and 1,2-bis(2-pyridyl)ethylene (2PY-e) caused increased liver weight without any histopathological evidence of hepatocellular hypertrophy in rats (Makino et al. 1998).

Lobe and Lobular Zonation of Induced Enzymes

Several factors and conditions are associated with and influence hepatocyte enzyme induction. These factors include age, hormonal effects, diet, infection, and other environmental influences and may result in differential lobe and lobular effects on hepatocytes.

Published data on lobe differences in endogenous and induced enzymes is scarce, although many researchers suspect there may be important lobe differences based on differential blood flow (Wilson and Hiley 1983) and the observation of lobe differences in toxic responses. Using liver homogenates as the enzyme source, activity of cytochrome P450 in the rat was higher in the median and right lobes of the liver, and mitochondrial enzymes were higher in the left lobe (Matsubara et al. 1982). Even after phenobarbital treatment or after fasting, cytochrome P450 activity remained higher in the median and right lobes. In contrast, treatment with β-naphthoflavone resulted in a uniform increase in P450 throughout the liver. Interestingly, fasting for twenty-four hours resulted in marked decrease in glycogen content throughout the liver, with the most dramatic decrease occurring in the median and right lobes.

Zonation differences are associated with phase I and phase II metabolism enzymes, with a predominant centrilobular localization of most constitutively expressed P450s (Figure 1A and 1B) and other phase I enzymes and a predominant periportal expression of most phase II enzymes. Typical inducers of centrilobular CYP enzyme induction include phenobarbital, TCDD, 3-methylcholanthrene, ethanol, dexamethasone, and peroxisome proliferators (Lindros 1997). Following enzyme induction, the extent of lobular involvement increases to include mid-lobular and sometimes periportal hepatocytes (Figure 1C and 1D). However, phenobarbital and TCDD are also phase II enzyme inducers with induction of some glutathione transferases and UDP-glucuronyltransferases in centrilobular hepatocytes and predominant periportal expression of glutathione peroxidase (Lindros 1997). There is some evidence to suggest that zonation associated with P450 induction is modulated by circulating levels of growth hormone and thyroid hormones, whereas the zonation of constitutive P450 expression is governed by other factors (Lindros 1997; Tani et al. 2001).

epatic Enzyme Induction Histopathology-fig1

FIGURE 1.—Immunohistochemical staining for CYP3A1 in Fischer 344 rats. (A) Low magnification of constitutive expression involving centrilobular hepatocytes in a control rat. (B) Higher magnification of 1A. (C) Low magnification of enzyme induction following gavage exposure to kava kava extract for fourteen weeks. (D) Higher magnification of 1C. (Reproduced with permission of Elsevier from publication by Clayton et al. (2007). Exp Toxicol Pathol 58, 223-236.). C, central vein; G, Glison’s sheath.

Other factors that can also influence the zonal distribution of induced enzymes include age and sex, as well as different xenobiotics and the dosing regimen used.

The influence of age and sex on the distribution of induced enzymes was observed in a study of the lobular distribution of cytochrome P450a in Sprague-Dawley and Long-Evans rats (Moody et al. 1983). The lobular distribution of P450a staining by immunohistochemistry was identical between the two stocks. Baseline P450a fluorescence could be separated into two groups: (1) mature males, with fluorescence in centrilobular, midzonal and periportal regions; and (2) mature females and immature males and females, with fluorescence localized primarily to centrilobular hepatocytes. Following three days of daily injection of phenobarbital or 3-methylcholanthrene, the P450a fluorescence pattern did not change in mature males. However, in mature females and immature males and females, the intensity of fluorescence was increased and the distribution was similar to that of mature males. Additionally, a separate study of lobular distribution of hepatic P450 in SpragueDawley rats concluded that there were pronounced changes in hepatic lobular distribution of P450 enzymes from day 19 of gestation to forty-five days of age (Watanabe et al. 1993). From gestation day 19 to five days of age, P450 increased 1046% in periportal hepatocytes and 819% in centrilobular hepatocytes. The centrilobular content of P450 increased progressively between five and forty-five days of age, with the characteristic sublobular distribution favoring centrilobular hepatocytes first becoming apparent around seven days of age.

Endogenous substances (e.g., glucocorticoids) and exogenous substances (e.g., xenobiotics) can also cause enzyme induction. Following an intraperitoneal injection of 50 mg of hydrocortisone sodium succinate/kg body weight, hepatic induction of tyrosine transaminase was demonstrated in male Sprague-Dawley rats (Welsh 1972). Prior to treatment, tyrosine transaminase was distributed uniformly in the hepatic lobule. Four hours after treatment, there was a six-fold increase in periportal cells and a four-fold increase in centrilobular cells based on chemical quantitation. In the same study, glutamic pyruvate transaminase was five-fold higher in periportal hepatocytes versus centrilobular hepatocytes but increased in both areas of the lobule during a three-day fast. In a separate study, phenobarbital induced NADPHcytochrome c reductase, glutathione reductase, cytochrome P450, and UDP-glucuronosyltransferase equally in the centrilobular and periportal hepatocytes. In contrast, in untreated rats, higher activities of microsomal NADPH-cytochrome c reductase, 7-ethoxycoumarin o-deethylase, 7-ethoxyresorufin o-deethylase, and cytosolic glutathione transferase were found in the centrilobular hepatocytes (Bengtsson et al. 1987). The authors point out that, contrary to other published studies, the enzyme induction by phenobarbital in their study was panlobular. Further, a study was conducted to assess the effects of 3-methylcholanthrene, phenobarbital, and trans-stilbene oxide on epoxide hydrolase immunohistochemical staining. 3-Methylcholanthrene did not cause significant alterations in epoxide hydrolase immunohistochemical staining, but phenobarbital and trans-stilbene oxide produced significant alterations in the intensity and lobular pattern of distribution (Kawabata et al. 1983). Phenobarbital treatment produced increased staining in the mid-lobular regions seven days after treatment and increased staining in periportal as well as centrilobular hepatocytes. Trans-stilbene oxide increased staining up to 80% in both mid-lobular and periportal hepatocytes. These findings indicate that phenobarbital and trans-stilbene oxide both induce epoxide hydrolase non-uniformly within the hepatic lobule.

Lastly, the zonal distribution and intensity of immunopositivity for induced enzymes is also dependent on the specifics of the methodology used in the immunostaining procedures. For example, auto-induction of CYP2E1 can be demonstrated by immunostaining in rats and humans (Tsutsumi et al. 1989). It appears exclusively in the centrilobular hepatocytes at low antibody concentrations, but at higher antibody concentrations, panlobular immunostaining is observed. Immunohistochemical staining for e