One may wonder what an article on biotransformation enzymes is doing in a magazine dedicated to sport and supplementation science. Somehow, the topic is brought regularly to the surface in many unrelated discussions, and every time, similar questions are asked and need to be answered. This topic is actually quite relevant, since it directly addresses the way our body processes the chemical compounds we absorb daily, whether they be drugs, dietary supplements, environmental toxins or (pro)carcinogens. A better understanding of the systems involved would also allow for a better assessment of the risks and benefits when trying to manipulate these systems.
Biotransformation enzymes (also referred to as xenobiotics) or drug-metabolizing enzymes, play a major role in the metabolism of foreign (therapeutic drugs, chemical carcinogens, phytochemicals, etc.) and endogenous compounds (e.g. steroid hormones).
There are two main groups of biotransformation enzymes. The first group (phase I metabolic enzymes) converts hydrophobic compounds into more reactive metabolites by exposing or adding functional groups (mainly –OH, but also –SH, –NH2, –COOH). Among other enzymes, this group contains oxidative enzymes like the cytochromes P450 and the flavin monooxygenases (FMOs), as well as hydrolytic enzymes like esterases and amidases. The second group (phase II metabolic enzymes) is composed of enzymes that catalyze the conjugation of these functionalized compounds to a water-soluble moiety (e.g. glucuronic acid, sulfate, glutathione). Some of the enzymes involved in phase II metabolism are UDP-glucuronosyltransferases (UGTs), sulfotransferases (SULTs) and glutathione S-transferases (GSTs). The resulting conjugates can then be excreted out of the body.
The remainder of this article will focus mainly on the function and regulation of cytochrome P450 enzymes.
Cytochrome P450 enzymes (CYPs or P450s) constitute a superfamily of ubiquitous heme-dependent proteins that are expressed in mammals mainly in the liver, with lower levels of expression in the small intestine, lungs, kidneys, brain and placenta. In man, to date 57 different P450 isoforms have been identified, which were assigned to 18 families and 43 subfamilies based on their protein sequences . From these 18 families, only the first three are involved in xenobiotic metabolism and account for approximately 70% of the CYP content in human liver . Only 14 additional families are responsible for the biosynthesis and metabolism of endogenous hormones, and the function of the last family (CYP20) still remains to be elucidated. The following table gives more details on the role of the various P450 families characterized in humans.
|CYP1, CYP2, CYP3||Metabolism of drugs and xenobiotics|
|CYP4, CYP5, CYP8||Fatty acids hydroxylation, biosynthesis of prostaglandins, prostacyclins and thromboxanes|
|CYP7, CYP11, CYP17, CYP19 (=steroid aromatase), CYP21, CYP24, CYP27, CYP39, CYP46, CYP51||Biosynthesis and metabolism of cholesterol, steroid hormones and bile acids|
|CYP26||Retinoic acid hydroxylation|
P450 isoforms are typically designated by a number sometimes followed by a letter and a second number. The first number (P450 family) groups all isoforms, which show at least 40% protein sequence identity, whereas the letter (P450 subfamily) groups all isoforms showing at least 55% protein sequence identity within a given family. Finally, the last number designates the single gene products (also called isoforms or isozymes) within a given subfamily. Although most of the P450 family numbers do not have any special meaning, for historical reasons, some still reflect the specific reaction catalyzed by the corresponding enzyme family. For instance, the aromatase CYP19 owes its name to the fact that it oxidizes the C19 carbon atom of testosterone and androstenedione, whereas CYP7A1 is responsible for the hydroxylation of cholesterol at position 7alpha.
Function of P450s
As mentioned above, P450s are enzymes primarily involved in the chemical functionalization of xenobiotics and hormones. In particular, their role in the metabolic processing of therapeutic drugs is of uttermost importance. In fact, from the 315 most commonly prescribed drugs in the US, 175 (56%) are primarily cleared by P450-mediated metabolism . The liver expression levels of the various P450 isoforms as well as their relative contribution to drug metabolism is given in the table below. Some examples of known drugs primarily metabolized by P450s are also indicated. As can be seen, the CYP3A subfamily, which comprises CYP3A4, 3A5 and 3A7, is responsible for more than 50% of P450-mediated metabolism, followed by isoform CYP2D6 (24%) and by the CYP2C subfamily (2C8, 2C9 and 2C19; ~20%).
Abundance in liver
Contribution to drug metabolism
Also worth mentioning is the striking difference between the hepatic expression levels of CYP2D6 (1.5%) and the relative contribution of this isoform to overall drug metabolism (24%). Actually, CYP2D6 is a typical example of a P450 that is mainly expressed in peripheral tissues, in particular in the central nervous system. In fact, approximately 60% of drugs active on the CNS are metabolized by this isoform.
P450s are resource-consuming enzymes, where the synthesis of the iron-containing heme prosthetic group, multiple enzymes are required; NADPH is used as electron source and consequently, an NADPH-reductase system is needed to recycle NADP+ to NADPH. Furthermore, an overproduction of some hydroxylated compounds may be damaging for the cells. On the other side, the organism needs to be able to respond to exposure to potentially harmful chemicals. For these different reasons, most P450 enzymes have a variety of gene regulatory mechanisms, either at the transcriptional or at the post-transcriptional level.
One of the mechanisms used in living organisms to induce gene transcription (and P450s are no exception) is through the binding of a small signaling molecule to a transcription factor or to a nuclear hormone receptor, which then would bind to a short DNA sequence (so-called response element), triggering the expression of the genes under the control of that particular response element . This process works in the very same way as the activation of genes involved in satellite cell differentiation upon the binding of testosterone and other AAS to the androgen receptor. Over the years, a range of such factors and receptors involved in upregulation of P450 production have been identified.
The CYP1 isoforms (1A1, 1A2 and 1B1) are induced by polycyclic aromatic hydrocarbons (present, for instance, in cigarette smoke, and in charcoal-broiled or smoked meat). These compounds bind to a transcription factor called aryl-hydrocarbon receptor (AhR), which can then move into the cell nucleus and associate with the AhR nuclear translocator (Arnt). This dimer binds finally to specific DNA sequences (Xenobiotic Response Elements or XREs), turning on CYP1 gene transcription .
The inducible expression of the other two main subfamilies involved in xenobiotic metabolism (CYP2B/C and CYP3A) is regulated by two nuclear hormone receptors: CAR (Constitutive Active/Androstane Receptor) and PXR (Pregnane X Receptor). Both nuclear receptors are usually sequestered in the cytoplasm, although some contradictory results seem to privilege a localization of PXR in the nucleus [5,6]. Only after ligand binding, they translocate to the nucleus and heterodimerize with the Retinoic X Receptor (RXR). This dimer can then bind to the regulatory region of genes, activating their expression (mainly CYP2Bs and CYP2Cs, but also CYP3As for CAR, resp. mainly CYP3As, but also to a lesser extent CYP2Cs and CYP2Bs for PXR). A full list of genes regulated by CAR and PXR can be found in Table 2 of . Some agonists of CAR and PXR are given in the table below:
|PhenobarbitalAndrostanol/AndrostenolPolychlorinated biphenyls (PCB)||PregnenoloneProgesteroneRifampicin (macrolide antibiotic)Omeprazole
Glucocorticoids and anti-glucocorticoids
Hyperforin (St John’s wort)
Very early on, it was also recognized that some CYPs responsible for the synthesis and metabolism of endogenous compounds were also regulated via ligand-binding to a nuclear receptor. Soon after the discovery of the peroxisome proliferator-acitvated receptor alpha (PPARalpha), it became clear, that this nuclear receptor played a crucial role in the induction of the CYP4A subfamily . In response to binding of arachidonic acid, as well as drugs known as peroxisome proliferators (e.g. clofibrate), PPARalpha activates the expression of genes involved in the oxidation of fatty acids, among them the CYP4A cytochromes (omega-hydroxylation of fatty acids).
More recently, the role of two additional nuclear receptors, the Farnesoid X Receptor (FXR) and the Liver X Receptor (LXR), in the regulation of cholesterol and bile acid homeostasis has been established. LXR and FXR bind oxysterols and bile acids, respectively, and are key players in the regulation of CYP7A1, the first and tightly regulated step in the metabolism of cholesterol to bile acids . Under high cholesterol concentration, LXR directly induces the transcription of CYP7A1 in the liver as a mechanism to lower the cholesterol amount by producing bile acids, which can then be excreted. Well, at least this is how it works in rodents. In humans, due to a mutation in the response element of the promoter region of the Cyp7a1 gene, LXR is unable to bind and therefore, expression of CYP7A1 is limited to the basal level . The ability to modulate CYP7A1 expression in mice and rats makes these animals extremely resistant to a high cholesterol diet whereas other species, including man, rapidly develop hypercholesterolemia under comparable conditions.
On the other hand, FXR is a negative regulator of CYP7A1 and its role is to prevent the build-up of high, toxic and cell-damaging bile acid levels in the liver. In addition to the downregulation of CYP7A1 expression, FXR induces, among others, the production of transporter proteins to export bile acids from the liver into the bile duct for excretion into the feces .
Furthermore, it should be noted that, under pathological conditions such as cholestasis, bile acids tend to accumulate in the liver, causing cell damage. Interestingly, it was recently shown that under such conditions, the xenobiotic sensors CAR and PXR are able to bind bile acids, inducing several bile acid-hydroxylating CYPs (CYP3As, and to a lesser extent CYP2Cs and CYP2Bs), bile acids transporters, and sulfotransferases that promote the excretion of bile acids via blood and urine. Hence, offering an alternate mechanism in order to lower intrahepatic bile acid levels before they become hepatotoxic .
The take-on message of these last examples is that crosstalk among the various CYP induction pathways have been clearly identified. A given ligand can bind to many different nuclear receptors, activating the transcription of different CYP genes. This process is very likely concentration-dependent and may reflect the binding affinity difference of a compound for the various receptors. This redundancy can be seen as a compensatory mechanism in case of malfunction of the primary receptor, or when this receptor is saturated and alternative pathways need to be activated to avoid toxic build-up of certain substances. Additionally, an activated nuclear receptor can induce the expression of some P450 genes, and simultaneously repress the expression of other P450 genes.
P450 induction and procarcinogen activation
Procarcinogens are precursors of cancer-causing chemicals and are part of our everyday life: they can be found in the food we eat, in the air we breathe, in the water we drink. Many procarcinogens are metabolized by P450 enzymes, either to biologically inactive metabolites or to chemically reactive electrophilic metabolites that irreversibly react with DNA, causing mutations and eventually, tumors. This process is summarized in Figure 1.
Figure 1: Processing of procarcinogens by xenobiotic-metabolizing enzymes (from ).
Many procarcinogens and environmental pollutants have also been shown to induce the expression of phase I and phase II enzymes. In particular, polycyclic aromatic hydrocarbons (PAHs) like benzo[a]pyrene (BaP, present in cigarette smoke, charcoal-broiled and smoked meat, and frying oil used repeatedly ), dioxins like TCDD (from the smoke of waste-incineration plants and of “backyard” burning of domestic and garden waste ), and pesticides like DDT and PCBs are known to bind to the AhR transcription factor and activate the transcription of these genes, in particular CYP1A1, 1A2 and 1B1. Some phytochemicals (Indole-3-carbinol, flavone) are also known as agonists of AhR .
It has long been observed that induction of these CYP enzymes by subtoxic doses of an AhR agonist before exposure to known procarcinogens like PAHs inhibits chemical carcinogenesis. For instance, treatment of mice with nontoxic doses of TCDD enhanced CYP-dependent inactivation of PAHs in the epidermis and inhibited the production of the tumor-initiating metabolites , contradicting the results showing that the CYP1 family, in particular CYP1A1, is responsible for the metabolic activation of PAHs. Furthermore, an epidemiology study also showed that cigarette smokers who are heavily exposed in their daily diet to Aflatoxin B1 (a strong liver procarcinogen produced by a fungus in decaying spices and seeds, and which is a substrate for CYP1A2 and 3A4), had a decreased risk of developing liver cancer when compared with nonsmokers exposed to the same diet .
How can these conflicting results be explained? Part of the problem seems to be due to the tools that have been used over the years to determine the metabolic fate of procarcinogens. The use of transgenic cell lines (for instance, hepatoma cells) over-expressing a particular P450 isoform during the 1990’s yielded a lot of valuable results, but these in vitro systems need to be considered for what they are: artificial systems that rarely reflect the biological environment of a cell in an animal model or a human being. The recent use of transgenic animals has brought some clarification to these results. For instance, Uno et al.  have used knockout mice lacking CYP1A1 to evaluate the metabolic activation of BaP. The first step of the activation of BaP has long been thought to be due to CYP1A1.
In distinct contrast to this theory, the authors could show that CYP1A1 was essential for the detoxification of BaP, and mice lacking CYP1A1 were susceptible to BaP-induced carcinogenesis, whereas wild-type mice were protected. More recent results from the same group seem to plead for CYP1A1-mediated detoxification and CYP1B1-mediated metabolic activation . Additionally, the main site of CYP1A1 induction and BaP detoxification was the intestine, which acts as a protective barrier before BaP can reach organs with higher levels of CYP1B1 expression. Nevertheless, the last word on this issue has very likely not been told yet.
The Aflatoxin B1 example above is more understandable. Aflatoxin B1 is mainly activated into a reactive metabolite (Aflatoxin B1 dihydrodiol) by CYP3A4, whereas CYP1A2 seems predominantly involved in the formation of the non-toxic compound Aflatoxin M1 . Under normal conditions, the activation pathway dominates, leading to liver tumor formation, but after induction of the CYP1 family by agonists of AhR (for instance, the PAHs in cigarette smoke), the CYP1A2-mediated detoxification takes over the process, limiting the amount of reactive metabolite produced.
In summary, the effects of inducers of metabolic enzymes on the carcinogenicity of a chemical depend on the inducers’ effect on the different metabolic pathways illustrated in Figure 1 . Induction of phase I and phase II enzymes may be useful in individuals or populations exposed to high levels of carcinogens, but the application in the general population would require a careful assessment of risks and benefits . Additionally, it should be kept in mind that inducers could potentially alter the metabolism of therapeutic drugs, leading to unwanted adverse effects; either through an increased clearance of the drug or through the faster formation of an active metabolite. Similar effects on the synthesis and metabolism of endogenous hormones cannot be discounted either.
Alcohol induction of CYP2E1 and acetaminophen metabolism
CYP2E1 represents an interesting example of P450 induction. In the liver, ethanol is converted to acetaldehyde mainly by the alcohol dehydrogenase enzyme . But chronic ethanol consumption leads to a 4- to 10-fold induction of an accessory oxidative pathway involving CYP2E1. The molecular mechanism underlying CYP2E1 induction is still heavily debated, but seems to involve messenger RNA or protein stabilization and/or transcriptional activation [25-26].
Although both the alcohol dehydrogenase and CYP2E1 pathways convert ethanol to the same product (acetaldehyde), the substrate-specificity of CYP2E1 is much broader and includes, among others, organic solvents, anaesthetic agents (enflurane), illicit drugs (cocaine), and some over-the-counter anti-inflammatory drugs. In fact, CYP2E1 is even considered one of the major carcinogen-activating enzymes .
One of the most problematic compounds activated by CYP2E1 metabolism is the painkiller acetaminophen (also known as paracetamol), found in nonprescription drugs such as Tylenol. Under normal circumstances, due to the low expression levels of CYP2E1, only a small fraction of an acetaminophen dose is metabolized to the reactive compound N-acetyl-p-benzoquinone imine (NAPQI). This toxic metabolite is usually inactivated through conjugation to glutathione by glutathione S-transferase. But in cases where the hepatocellular glutathione stores are depleted and/or the NAPQI production is increased, it starts accumulating and reacts with liver cell proteins, leading to liver necrosis and finally to acute liver failure . Schiodt et al.  even showed that acetaminophen toxicity accounted for up to 40% of patients treated for acute liver failure in an urban county hospital in the US. The two conditions leading to the accumulation of NAPQI are often concomitant, the chronic alcohol consumption leading to the strong induction of CYP2E1, resulting in a rapid depletion of glutathione, which is further exacerbated by a poor nutrition (low protein intake), a very common occurrence among alcoholics.
This concludes the first part of this series of articles addressing the topic of drug-metabolizing enzymes and their modulation. In the next article, we will have a closer look at CYP inhibition and its consequences on drug-drug interactions.
2. Yan Z, Caldwell GW, Curr. Topics Med. Chem. 2001, 1, 403-425
3. Bertz RJ, Granneman GR, Clin. Pharmacokinet. 1997, 32, 210
4. MacDonald CJ, Ciolino HP, Yeh GC, Cancer Res. 2001, 61, 3919-3924
5. Moore JT, Moore LB, Maglich JM, Kliewer SA, Biochim. Biophys. Acta 2003, 1619, 235-238
6. Squirest EJ, Sueyoshi T, Negishi M, J. Biol. Chem. 2004, 279, 49307-49314
7. Waxman DJ, Arch. Biochem. Biophys. 1999, 369, 11-23
8. Handschin C, Meyer UA, Pharmacol. Rev. 2003, 55, 649-673
9. Johnson EF, Palmer CN, Griffin KJ, Hsu MH, FASEB J. 1996, 10, 1241-1248
10. Simpson AECM, General Pharmacology: The Vascular System 1997, 28, 351-359
11. Handschin C, Podvinec M, Amherd R, Looser R, Ourlin J-C, Meyer UA, J. Biol. Chem. 2002, 277, 29561-29567
12. Chiang JY, Kimmel R, Stroup D, Gene 2001, 262, 257-265
13. Francis GA, Fayard E, Picard F, Auwerx J, Annu. Rev. Physiol. 2003, 65, 261-311
14. Handschin C, Meyer UA, Arch. Biochem. Biophys. 2005, 433, 387-396
15. Conney AH, Cancer Res. 2003, 63, 7005-7031
16. Pandey MK, Yadav S, Parmar D, Das M. Toxicol. Appl. Pharmacol. 2005, Oct 28 (Epub ahead of print), PMID: 16263147
17. Hedman B, Naslund M, Nilsson C, Marklund S, Environ. Sci. Technol. 2005, 39, 8790-8796
18. Ebert B, Seidel A, Lampen A, Toxicol. Lett. 2005, 158, 140-151
19. DiGiovanni J, Kishore GS, Slaga TJ, Boutwell RK “2, 3, 7, 8-Tetrachlorodibenzo-p-dioxin (TCDD)-induced alterations in oxidative and nonoxidative biotransformation of PAH in mouse skin: role of anticarcinogenesis by TCDD”. In: Bjorseth O, Dennis AJ (eds.), Polynuclear Aromatic Hydrocarbons: Chemistry and Biological Effects 1980, 935-954. Columbus, OH, Battalle Press
20. Lin L, Yang F, Ye Z, Xu E, Yang C, Zhang C, Wu D, Nerbert DW, Pharmacogenetics 1991, 1, 79-85
21. Uno S, Dalton TP, Derkenne S, Curran CP, Miller ML, Shertzer HG, Nebert DW, Mol. Pharmacol. 2004, 65, 1225-1237
22. Uno, S, Dalton TP, Dragin N, Curran CP, Derkenne S, Miller ML, Shertzer HG, Gonzalez FJ, Nebert DW, Mol. Pharmacol. 2005, Dec 25 (Epub ahead of print), PMID: 16377763
23. Gallagher EP, Kunze KL, Stapleton PL, Eaton DL, Toxicol. Appl. Pharmacol. 1996, 141, 595-606
24. Moon YJ, Wang XW, Morris ME, Tox. In Vitro 2006, 20, 187-210
25. Lieber CS, Clin. Chim. Acta 1997, 257, 59-84
26. Park KS, Sohn DH, Veech RL, Song BJ, Eur. J. Pharmacol., Environ. Toxicol. Pharmacol. Sect. 1993, 248, 7-14
27. Yang CS, Chhabra SK, Hong JY, Smith TJ, J. Nutr. 2001, 131, 1041S-1045S
28. McClain CJ, Price S, Barve S, Devalarja R, Shedlofsky S, Curr. Gastroenterol. Rep. 1999, 1, 42-49
29. Schiodt FV, Rochling FA, Casey DL, Lee WM, N. Engl. J. Med. 1997, 337, 1112-1117