*3.1. Transcriptional Regulation*

The promoter region of CYP1A1 and CYP1A2 genes contains several Aryl hydrocarbon receptor (AhR) response elements [42,43], which, after binding of respective compounds, initiate coordinated transcription of both genes. Consequently, both genes are highly inducible by AhR ligands such as polycyclic aromatic hydrocarbons (PAHs), dioxins and numerous xenobiotics [44]. Examples for well-established experimental inducers are 2,3,7,8-tetrachlorodibenzo-p-dioxon (TCDD), 3-methlycholanthrene and β-naphthoflavone [45,46]. In addition, omeprazole was shown to be a potent model inducer for CYP1A1 and CYP1A2 [47–49]. There is evidence that the induction of CYP1A1 by AhR is stronger than that of CYP1A2 [44,49,50]. This is also in agreemen<sup>t</sup> with the fact that tissues possessing high expression of CYP1A1 (i.e., lung, placenta, intestine, urinary bladder) also show high expression levels of AhR [51]. Table 2 provides an overview of in vitro, in vivo, and ex vivo data on CYP1A1/1A2 induction by clinically relevant drugs. In this regard, albendazole, carbamazepine, omeprazole, lansoprazole, primaquine, and rosiglitazone were shown to be strong inducers of both, expression and metabolic activity of CYP1A1 and CYP1A2 [48–50,52]. As discussed later, these findings do not necessarily translate for all compounds to clinically relevant drug–drug interactions (e.g., for omeprazole). Relevant exogenous sources of AhR activators are charcoal grilled food, tobacco smoking as well as other natural sources including broccoli or fish oil supplementation, which strikingly induce endogenous eicosanoids [11,53,54]. Consequently, some of these environmental factors may contribute to the substantial variability in the expression and function of both CYP1A isoenzymes. Associated with this—tobacco smoking, especially, was shown to have significant effects on the pharmacokinetics and actions of many CYP1A substrates (see paragraph "Drug–Drug Interactions").


**Table 2.** Impact of clinically relevant drugs, smoking, and diet on the induction of CYP1A1/1A2 mRNA, protein, and activity.

**Table 2.** *Cont.*

↑, increase; ±, unchanged; CBT, caffeine breath test; EROD, 7-ethoxyresorufin O-deethylase; POD, phenacetin O-deethylation.

As already described for genes regulated by other nuclear receptors (e.g., *ABCB1* by pregnane-X-receptor (PXR)), a partial transactivation of human CYP1A by nuclear receptors other than AhR is possible. In this regard, CYP1A1 and 1A2 were also shown to be induced upon activation of the human constitutive androstane receptor (CAR) [63]. This explains considerable induction of CYP1A enzymes by typical CAR ligands, such as carbamazepine, phenobarbital and phenytoin. On the other side, the relevance of PXR in the regulation of CYP1A seems to be negligible as shown in vitro [47,59,60] and in vivo [64].

#### *3.2. Impact of Gender, Age, and Diseases*

In addition to the described transcriptional regulation, also several nongenetic factors seem to influence CYP1A2 expression and function. For example, protein abundance and metabolic CYP1A2 activity for different substrates was shown to be considerably lower in woman than in men [31,65–67]. However, as smoking and the intake of oral contraceptives (inhibitors of CYP1A function) represent substantial confounders of CYP1A2 expression and function, those data have to be interpreted with caution and need further verification. Moreover, in analyzing potential gender differences in the pharmacokinetics of CYP1A substrates, dose-adjustment was shown to be essential as demonstrated for tizanidine [68]. There is also evidence that CYP1A2 activity is significantly higher in younger (<20 years) than in older people (>20–60 years and >60 years) [65].

Disease-related changes have been also reported for CYP1A. Here, CYP1A2 expression in liver dysfunction and cholestasis was found to be decreased [40,69]. Other studies failed to confirm those differences in vivo [53,54] and in human liver tissue at both, mRNA and protein levels [40,70]. More recent mRNA expression data demonstrated that the expression of CYP1A2 was decreased by about 90% in hepatocellular carcinoma livers, 80% in alcoholic cirrhosis, and 65% in severe cirrhosis [71]. In parallel, analysis of liver biopsy samples of patients with chronic hepatitis C revealed significantly lower gene expression levels of CYP1A1 and CYP1A2 [72,73]. These data have been recently confirmed by a targeted proteomic analysis [74]. Likewise, nonalcoholic fatty liver disease (NAFLD) was associated with decreased mRNA, protein amount, and functional activity of microsomal CYP1A2 compared to healthy liver tissue [75].

#### *3.3. Genetics and Epigenetics*

The large inter-individual variability in the elimination of drugs undergoing CYP1A2 metabolism has been attributed to genetic and environmental factors [11,76,77]. In this regard, Rasmussen and colleagues demonstrated in a large study in 378 mono- and dizygotic twins for the ca ffeine metabolic ratio (a surrogate for CYP1A2 activity) a strong overall heritability of 0.72 [78].

Common polymorphisms in the CYP1 gene have been found to be only of limited relevance for human drug metabolism. However, considering the involvement of CYP1A enzymes in bioactivation of procarcinogens, many studies investigated certain single nucleotide polymorphisms in association to various types of cancer [10,11]. The Pharmacogene Variation Consortium website (www.pharmvar. org) lists 15 alleles for CYP1A1 [79]. Of the most frequent variants m1 to m4, only the common non-synonymous variant CYP1A1\*2C (rs1048943, 2454A>G, Ile462Val), which has a global minor allele frequency of about 12%, was shown to be associated with substantially modified enzymatic activity, i.e., 6- to 12-fold higher for its substrates 17β-estradiol and estrone [80]. This variant was associated with an increased risk for lung cancer in Chinese and breast, and prostate cancer in Caucasians [10,11].

Several alleles, namely 24, have been also reported for CYP1A2 [79], of which only the most established will be briefly mentioned here. The CYP1A2\*6 variant was shown to result in a nonfunctional protein [81]. However, considering the rare occurrence of this and other variants [82], they are expected to be of limited clinical relevance. The CYP1A2\*1C was associated with reduced CYP1A2 induction by cigarette smoking in Japanese [83]. On the contrary, the CYP1A2\*1F variant (–163C>A) was linked with enhanced enzyme inducibility in Caucasian smokers [53,84] and heavy co ffee drinkers [85]. Interestingly, carriers of the combined genotype CYP1A2\*1C/\*1F were not inducible by the AhR ligand omeprazole [86]. Both variants were described to increase the susceptibility to certain cancers. Despite the described multiplicity of CYP1A2 polymorphisms, clear gene dose relationships by comparing common SNPs to the respective protein abundance or metabolic phenotype could not be demonstrated yet. Thus, so far no single SNP or haplotype in the CYP1A2 gene seems to be predictive [41]. In this regard, a multivariate linear modeling by Klein et al. revealed that genetic polymorphisms contribute about 35% of hepatic CYP1A2 activity variation, whereas some 40% of the variation were explained by nongenetic factors together [40].

However, the clinical impact of genetic variation in terms of susceptibility factors for cancer or pharmacokinetics, e fficacy and safety of certain CYP1A substrates is not systematically covered here but was excellently summarized by others [10,11,87–90].

Finally, there is also evidence for an epigenetic regulation of CYP1A2 expression as concluded from the observation that the extent of DNA methylation of a CpG island close to the translation start site was inversely correlated to the hepatic CYP1A2 mRNA expression [53,54]. Recent studies point also to an involvement of certain microRNAs in the expression and induction of CYP1A2 [91,92].

#### **4. Metabolic Function, Substrates, and Inhibitors**

### *4.1. Metabolic Features*

Considering that CYP1A2 shares about 80% amino acid sequence identity with CYP1A1, it is not surprising that the substrate specificities of these enzymes often overlap, owing to a CYP1 family–specific distortion of the F helix in the area of the substrate binding cavity, which produces bending of the helix and results in the formation of an enclosed and planar substrate binding site observed in both CYP1A1 and CYP1A2 [93]. It has been demonstrated that commonly used probe drugs for CYP1A2 such as ca ffeine, theophylline, phenacetin, propranolol, and 7-ethoxyresorufin are metabolized by both CYP1A isoenzymes [94,95]. Despite this considerable similarity, CYP1A1 shows a preference for planar aromatic hydrocarbons (e.g., naphthalene, PAHs), while CYP1A2 prefers aromatic amines and heterocyclic compounds (e.g., 2-naphthylamine, xanthines) (Table 3). The metabolic feature of CYP1A1 in combination with its expression pattern in tissues potentially exposed to high amounts of PAHs (e.g., the lung via tobacco smoke, the intestine via charbroiled food) makes it plausible that increased expression and function of CYP1A1 may result in higher formation rates of potentially carcinogenic metabolites. In this regard, benzo[a]pyrene and other procarcinogens (e.g., arylarenes, nitroarenes, arylamines) are bioactivated by CYP1A1 to reactive and carcinogenic intermediates such as epoxides which may cause DNA damage and in long term malignancies. In the same manner, CYP1A2 is involved in the bioactivation of heterocyclic aromatic amines (HAAs) originating from cook muscle meats such as beef, pork, or fish to carcinogenic hydroxylamines. Thus, it can be assumed that induction of CYP1A1/1A2 in smokers by inhaling frequently high amounts of PAHs may contribute to strikingly increased risk for lung cancer [96]. However, the toxicological impact of both isoenzymes on the bioactivation of carinogenes from environmental compounds is beyond the scope of this article but summarized elsewhere [11,97].


**Table 3.** Overview for clinically relevant drugs undergoing significant CYP1A2-mediatedmetabolism (≥25%).

5-HT, 5-hydroxy tryptamine; ALS, Amyotrophic lateral sclerosis; CNS, central nervous system; COPD, Chronic obstructive pulmonary disease; NSAID, non-steroidal anti-inflammatory drug.
