Human cytochrome P450 enzymes 5-51 as targets of drugs and natural and environmental compounds: mechanisms, induction, and inhibition - toxic effects and benefits

The function of P450 20A1 remains unknown; however, the sites of mRNA expression in human brain and the conservation among species may suggest possible neurophysiological function ( Stark et al. 2008 ). P450 20A1 appears to be the only human P450 Family 20 member. P450 20A1 EST sequences have recently been reported from kidney, brain, placenta, and urogenital tissues, according to the NCBI UniGene database ( http://www.ncbi.nlm.nih.gov/UniGene/ ). Orthologous nucleotide sequences have been found in a number of mammals, including rat (84% identity), mouse (85%), chimpanzee (99.6%), and chicken (72%) ( http://drnelson.utmem.edu/CytochromeP450.html ). None of these (putative) proteins has been studied to date. In contrast to our report in Stark et al. ( Stark et al. 2008 ), the Protein Atlas Database ( www.proteinatlas.org ) indicates widespread mRNA expression and possibly higher levels in endocrine tissues. Although expression in E. coli has been reported ( Stark et al. 2008 ), the levels were low and in our own hands this has not been easy to repeat. Although there is still no information about a substrate, deletion of the CYP20A1 gene in zebrafish has been reported to cause a hyperactivity disorder ( Lemaire et al. 2016 ). At this time P450 20A1 remains a true orphan P450 ( Guengerich and Cheng 2011 ) in the sense of not having any identified function. It has been noted that this protein lacks one amino acid of the conserved heme binding site. It also lacks the conserved I-helix motif AGX(D,E)T, and it has been suggested that its substrate may carry its own oxygen ( www.proteinatlas.org ). However, there is no actual evidence that this is the case.

Although the liver is generally considered to be the major organ involved in cholesterol synthesis, CYP51A1 mRNA was found to be high in testis, ovary, adrenal, prostate, kidney, and lung as well as liver ( Stromstedt et al. 1996 ). In mouse testis, mRNA levels were high and localized in both round and elongated spermatids ( Cotman et al. 2001 ). This high level has been proposed to lead to the production of steroids for signaling ( Rozman and Waterman 1998 ; Rozman et al. 2002 ). Cyp51a1 -knockout mice show a phenotype reminiscent of the Antley-Bixler Syndrome ( Keber et al. 2011 ). Several X-ray crystal structures of human P450 51A1 have now been published, both ligand-free and with ligands ( Strushkevich et al. 2010 ; Hargrove et al. 2016 ) (PDB 3JUS, 3JUV, 3LD6, 4UHI, 4UHL). In addition, several X-ray crystal structures of various parasite CYP51 enzymes have also been published ( Lepesheva et al. 2010 ; Hargrove et al. 2015 ; Hargrove et al. 2017 ). These structures have potential value in directing the synthesis of inhibitors towards the parasitic 14 α -demethylases and avoiding interactions with human P450 51A1. The specificity constant ( k cat / K m ) for human P450 51A1 has been reported to be only ~ 300 M −1 s −1 , which is relatively low among P450s and also enzymes in general ( Lamb et al. 1998 ). Nevertheless, the significance of this enzyme is clear, as further demonstrated by the transgenic mouse results ( Keber et al. 2011 ). An open question is the catalytic mechanism of the third step of the 14 α -demethylase reaction. Obviously this has similarity to the last step of the P450 19A1 reaction ( vide supra ). Although 18 O-labeling and other evidence has been presented in favor of a ferric peroxide (FeO 2 − ) nucleophilic attack on the C-32 formyl group ( Fischer et al. 1991 ; Shyadehi et al. 1996 ) instead of the general Compound I (FeO 3+ ) mechanism, the same limitations pointed out in the reassessment of the P450 19A1 mechanism ( Yoshimoto and Guengerich 2014 ) apply ( Guengerich and Yoshimoto 2018 ) and a reassessment with modern mass spectrometry methods is in order. The oxidative removal of the 14 α -methyl group from sterol precursors by sterol 14 α -demethylase (CYP51 Family genes) is a universal step in the biosynthesis of membrane sterols and steroid hormones. Thus, P450 51A is a primary target in treatment of fungal infections in organisms ranging from humans to plants, and development of more potent and selective inhibitors is an important biological objective in many therapeutic fields. Currently the CYP51 gene family is found in at least 82 organisms, spread over all biological kingdoms. Several plants and fungi contain multiple CYP51 genes (e.g. rice (10), black oats (2), tobacco (2), Arabidopsis thaliana (2), Fuzarium graminearum (3), and several of the Aspergillus species: e.g., Aspergillus fumigatus (2), A. nidulans (2), A. orizae (3)). As a result, the number of known CYP51 gene sequences is >100. All mammalian genomes examined contain only a single CYP51A1 gene but some have nonfunctional CYP51 pseudogenes (e.g., three in humans and one in rats, http://drnelson.uthsc.edu/cytochromeP450.html ). Azoles (imidazoles and triazoles) play a pivotal role in the treatment of systemic and dermal mycoses. The substrate analogs 7-oxo, 15-keto, 15-oxime, 15-hydroxy, 26-oxo derivatives of lanosterol are effective in blocking cholesterol biosynthesis in humans. 14 α -Amino derivatives of lanosterol were found effective to inhibit Candida albicans and Trypanasoma cruzi P450 51 orthologs and are effective to cure murine models of leishmania and Chagas disease ( Lepesheva and Waterman 2007 ). Sterol 14 α -demethylase is one of the hypothetical targets an alternative to classic statin therapy ( Kaluzhsiy et al. 2014 ). Most commercial and experimental inhibitors of microbial P450 51 orthologs (including all clinical antifungals) weakly inhibit human P450 51A1 activity. Only one relatively potent compound, ( R )- N -(1-(3,4´-difluorobiphenyl-4-yl)-2-(1 H -imidazol-1-yl)ethyl)-4-(5-phenyl-1,3,4-oxadiazol-2-yl)benzamide (VFV), was identified, and it was also found to decrease proliferation of different cancer cell types ( Hargrove et al. 2016 ).

The sole P450 involved in cholesterol formation, P450 51A1, is the lanosterol 14 α -demethylase ( Fig. 4A and Fig. 17 , vide infra ). There are three biologically important pathways from cholesterol catalyzed by P450 enzymes: (1) biosynthesis of steroid hormones, (2) formation of bile acids, and (3) vitamin D 3 metabolism. All mammalian steroids, bile acids, and active forms of vitamin D are formed from cholesterol ( Figs. 4A and 4B ). Detailed reactions leading to the formation of bile acids and catalyzed by P450 enzymes are presented in Figs. 14A and 15 , reactions related to biosynthesis and metabolism of vitamin D 3 in Figs. 19 and 20 , and thebiosynthesis of steroid hormomes in Figs. 7 to 13 and 14B . Clinical experience and studies with transgenic mice have shown the importance of all of these P450s ( Nebert and Russell 2002 ; Auchus and Miller 2015 ; Auchus 2017 ). The only other major enzymes involved in steroid hormones biosynthesis are dehydrogenases ( Fig. 4B ) (plus conjugating enzymes that form methyl ethers, sulfate esters, and glucuronides).

P450s are the major enzymes involved in the conversion of vitamin D 3 to more active products and also in the breakdown of vitamin D and its products ( Fig. 18 ). Because of the many roles of vitamin D, the balance is important and these enzymes are generally highly regulated. P450 24A1 is one of the seven human P450s normally localized in mitochondria, and accordingly it receives electrons from NADPH-adrenodoxin reductase and adrenodoxin. This P450 is localized in the kidney, plus a few other places ( Guengerich 2015 ). Its regulation is highly complex, as might be expected for an enzyme involved in a delicate regulation of the amount of an active vitamin ( Guengerich 2015 ). An X-ray crystal structure of rat P450 24A1 has been published ( Annalora et al. 2010 ) (PDB 3K9V, 3K9Y). No substrate is present, only a molecule of either steroidal detergent 3-[(3-cholamidopropyl)dimethylammonion]-1-propane sulfonate (CHAPS) or the totally synthetic detergent CYMAL-5. Some of the reactions catalyzed by P450 24A1 are shown in Fig. 19 , but the human enzyme will also oxidize a number of substrate analogs. The degradation of 1 α ,25-dihydroxy vitamin D 3 by P450 24A1 occurs through multiple, sequential oxidation reactions targeting specific carbon atoms in the side chain. These multicatalytic activities of P450 24A1 ( Makin et al. 1989 ; Reddy and Tserng 1989 ; Akiyoshi-Shibata et al. 1994 ; Beckman et al. 1996 ) can be divided into two regioselective oxidation pathways, involving an initial attack at C24 to yield a side-chain truncated product, calcitroic acid ( Makin et al. 1989 ; Reddy and Tserng 1989 ), or at C23 to give 1 α ,25-dihydroxy vitamin D 3 -26,23-lactone ( Sakaki et al. 2000 ) ( Fig. 19 ). The proportion of these catabolic products is species-dependent, with calcitroic acid production predominating in rodents ( Hamamoto et al. 2006 ), 26,23-lactone predominating in the opossum and guinea pig ( Pedersen et al. 1988 ; Horiuchi et al. 1995 ), and a mixture of the two products being formed in humans ( Prosser and Jones 2004 ). Substrate contact with a species-specific residue in the I-helix adjacent to the heme group of the cytochrome P450 is proposed to control the pathway selection, with Ala-326 directing substrate into the C24 pathway in humans and Gly-326 driving 23-hydroxylation in opossums ( Prosser et al. 2007 ). Other amino acids having a lesser influence on regioselectivity in rat include Thr-416 (Met-416 in humans) and Ile-500 (Thr-500 in humans) ( Hamamoto et al. 2006 ). Thus, the failure of the enzyme to oxidize certain substrates may result from an inability to correctly position the side chain rather than an inherent inability to bind the substrate ( Kaufmann et al. 2011 ). In melanocytic tumors the level of P450 24A1 was higher than in the normal epidermis, with the highest mean P450 24A1 level found in nevi and early stage melanomas. During melanoma progression P450 24A1 levels decreased and in the advanced stages was comparable to the normal epidermis and metastases. Lower P450 24A1 levels correlated with the presence of ulceration, necrosis, nodular type, and amelanotic phenotypes. A lack of detectable P450 24A1 expression was related to disease-free survival. Thus, an elevated level of CYP24A1 appears to have an important impact on the formation of melanocytic nevi and melanomagenesis, or progression, in the early stages of tumor development ( Brozyna et al. 2014 ). P450 27B1, a kidney mitochondrial enzyme, is the vitamin D 1 α -hydroxylase, acting on various derivatives of 25-hydroxyvitamin D. This is an activated form of vitamin D, and accordingly the enzyme has a very complex regulatory system ( Guengerich 2015 ). Further, P450 27B1 deficiency is associated with type I vitamin D-dependent rickets, and ≥ 20 variants are known ( Kitanaka et al. 1998 ; Wang et al. 1998 ; Smith et al. 1999 ; Portale and Miller 2000 ; Kim et al. 2007 ; Wu et al. 2007 ). No crystal structures have been reported, although roles of several amino acids have been identified through the work with the clinical variants.

Several P450 enzymes are involved in oxidations of retinoids ( Fig. 21 ), the regulation of which is very important, as in the case of vitamin D. The 4- and 18-oxidations are considered to be deactivating. The 3,4-desaturation by P450 27C1 in human skin is an efficient reaction but the biological significance has only been established in fish and amphibians. Retinoic acid is the endogenous active metabolite of vitamin A 1 (retinol) and acts as an essential regulator of cell growth, cell cycle, and differentiation ( Gudas 2012 ; Napoli 2012 ). all- trans -Retinoic acid is believed to be the main biologically active form, binding predominantly to nuclear retinoic acid receptors (RARs), and plays a critical role in many biological processes including reproduction, maintenance of skin and epithelia, regulation of the immune system and T- and B-cell function, and in fetal development ( Maden 2007 ; Ross 2012 ; Hogarth and Griswold 2013 ). The production of the enzyme P450 26A1 is regulated by its substrate, retinoic acid, via the RXR receptor. No crystal structures are available for the enzyme. P450 26A1 has relevance in developmental biology, in that Cyp26a1 −/− mice show distinct malformations and lethality ( Tay et al. 2010 ). Gudas has demonstrated a role of (mouse) P450 26a1 in stem cell differentiation, and inhibiting P450 26A1 has been considered as an approach in cell/differentiation therapy for neurodegenerative diseases ( Ricard and Gudas 2013 ). Imbalance in vitamin A and retinoic acid homeostasis have been implicated in development and progression of several human diseases, e.g., acne, psoriasis and ichthyosis, type II diabetes, neurodegenerative diseases, and some cancers ( Gudas 2012 ). The use and long term efficacy of all- trans retinoic acid has been limited by its high metabolic clearance, which limits oral administration of this agent, and the resistance that develops towards all- trans retinoic acid due to autoinduction of its metabolism ( Nelson et al. 2013 ). The clearance of all- trans retinoic acid is predominantly mediated by P450 Family 26 enzymes ( Thatcher and Isoherranen 2009 ; Ross and Zolfaghari 2011 ), particularly P450 26A1 in human liver, considered to be responsible for majority of the hepatic clearance ( Thatcher et al. 2010 ). It has been proposed that inhibition of the P450 Family 26 enzymes will increase all- trans retinoic acid concentrations in target tissues, and inhibition of P450 26 enzymes during could be used productively ( Njar et al. 2006 ). Many of the existing inhibitors of P450 26 enzymes have not been evaluated for specificity (among the P450 Family 26 enzymes) and therefore the benefits and shortcomings of selective or broad P450 26 inhibition have not been established. P450 Family 26 enzymes have cell type and tissue specific expression ( Topletz et al. 2012 ) and therefore specific P450 26A1 (or P450 26B1) inhibitors may be beneficial for increasing retinoic acid concentrations in selected target tissues without causing broad side effects. P450 26A1 is the predominant P450 Family 26 enzyme in the liver, and therefore selective inhibition of P450 26A1 could decrease systemic clearance of retinoic acid and be useful in therapy resistance with individuals being treated with all- trans retinoic acid or 13- cis -retinoic acid. A P450 26A1 selective inhibitor would avoid side effects of P450 26B1 inhibitors, which have potential reproductive and teratogenic pissues, and be more likely to be beneficial than when systemic retinoic acid concentrations are targeted ( Diaz et al. 2016 ). P450 26B1 is expressed in the brain and also in a number of other tissues but not liver ( Topletz et al. 2012 ). It is induced by all- trans -retinoic acid and also regulated by certain transcription factors ( Guengerich 2015 ). The reactions are similar to those catalyzed by P450 26A1, i.e. 4- and 18-hydroxylation of all- trans -retinoic acid and further oxidation of the 4-hydroxy product to 4-oxoretinoic acid. In contrast with the other Family 26 P450 enzymes, P450 26C1 is expressed mainly during development, although this conclusion is based largely on animal models ( Ross and Zolfaghari 2011 ). However, P450 26C1 mRNA has been detected not only in human fetal liver and braion but also in adult human brain, lung, liver, spleen, and testis, which are retinoid target tissues ( Xi and Yang 2008 ). The reactions catalyzed ( Fig. 23 ) are similar as for P450 26B1, although recent work by Zhong et al. (2018) has defined some important differences in both the substrate and product profiles. Zhong et al. (2018) reported that P450 26C1 has a specificty constant ( k cat / K m ) up to 10-fold higher than P450 26A1 or 26B1 in oxidizing all- trans -4-oxo retinoic acid. There are issues in defining the roles of the three different Family 26 P450s in retinoic acid metabolism, in light of the differneces in intrinssic catalytic acticity, cellular localizaiton, the concentration of each P450 protein in each tissue, and the potentialy different interactions of the P450s with cellular retinoic acid binding proteins ( Zhong et al. 2018 ). No information about the structure of this protein is available yet, although at least one homology model has been presented ( Zhong et al. 2018 ). Desaturation of retinol to 3,4-dehydroretinol (vitamin A 2 ) ( Figs. 21 , 24 ) has been reported in preparations of human breast skin ( Törmä and Vahlquist 1985 ) and keratinocytes in culture ( Rollman et al. 1993 ; Andersson et al. 1994 ). UV light exposure was reported to increase the formation of 3,4-dehydroretinol in cultured human keratinocytes ( Tafrova et al. 2012 ). Zebrafish P450 27C1 is an efficient retinol 3,4-desaturase and that this enzyme is expressed in the eyes of fish and amphibians where it acts to red-shift the visual chromophore ( Enright et al. 2015 ). No bird or mammal has been shown to express P450 27C1 in the eye, but the human CYP27C1 gene encodes a retinoid 3,4-desaturase, P450 27C1, with selectivity for all- trans retinol. This enzyme catalyzes the production of 3,4-dehydroretinoids in human skin ( Kramlinger et al. 2016 ; Johnson et al. 2017 ). The specificity constant for human P450 27C1 is relatively high, at least for mammalian P450 enzymes ( k cat/ K m 7.7 × 105 M-1 s-1 for all- trans -retinol), and clearly indicates that retinoids should be the natural substrates ( Kramlinger et al. 2016 ; Johnson et al. 2017 ). The rate-limiting steps in the catalytic cycle are the first-electron reduction and C-H bond-breaking, the latter revealed by kinetic deuterium isotope effects ( Johnson et al. 2017 ). In fish and some amphibians, the enzyme is localized in the retinal epithelium, and synthesis of P450 27C1 allows these animals to change their visual spectra ( Enright et al. 2015 ). However, in humans this (mitochondrial) enzyme is clearly localized in the skin, as shown by extensive proteomic analysis ( Johnson et al. 2017 ). Although there is little question that P450 27C1 is a skin retinoid desaturase ( Kramlinger et al. 2016 ; Johnson et al. 2017 ), what is not clear is why such a large fraction of the retinoids in human skin (1/4) is desaturated ( Vahlquist 1980 , 1982 ; Vahlquist et al. 1982 ). Rats and mice are devoid of the 27C1 gene and thus provide no insight into the function.

Some of the Family 4 P450s are involved in ω - and ω −1 hydroxylation of leukotrienes and prostaglandins ( Guengerich 2015 ), which are generally considered to be deactivating processes. The two P450s discussed here (P450s 5A1 and 8A1) catalyze rearrangements of prostaglandin H 2 and related molecules ( Fig. 1 ). P450s 8A1 and P450 5A1 are the only two human P450s that use an endoperoxide as a physiological substrate. These two enzymes isomerize prostaglandin endoperoxides without the use of molecular oxygen or any external electron donors ( Li et al. 2008 ). Both prostacyclin synthase (P450 8A1) and thromboxane synthase (P450 5A1) signaling affect a number of tumor cell survival pathways, e.g., cell proliferation, apoptosis, tumor cell invasion and metastasis, and angiogenesis. However, the effects of these respective synthases differ considerably with respect to the pathways described. Prostacyclin (prostaglandin I 2 (PGI 2 )) is a potent inhibitor of vasoconstriction, platelet activation, and aggregation. Widely known for its vasoprotective activity, prostacyclin is synthesized mainly in the endothelial and smooth muscle cells via the described isomerization of prostaglandin H 2 , a reaction catalyzed by prostacyclin synthase (CYP8A1; also termed PGI 2 synthase; PGIS). The balance of these oppositely-acting cyclooxygenase-derived prostanoids influences many processes throughout the body, e.g., blood pressure regulation, clotting, and inflammation. The prostacyclin/thromboxane ratio is important in vivo , with the corresponding synthases shown to be differentially regulated in a variety of disease states.

The human P450 family, which contains 57 members, can be divided into two major groups. Members of the first group are involved largely in the metabolism of xenobiotic compounds such as drugs and environmental chemicals (Families 1–4 with 35 enzymes). Members of the second group are enzymes that predominately catalyze biotransformation and metabolism of endobiotic compounds (Families 5–51 with 22 enzymes). However, there is overlap in these groups, i.e. the Family 5–51 enzymes can oxidize some non-physiological compounds. The enzymes from the first group have been extensively investigated ever since they were discovered. These enzymes have been the subject of a great number of research publications and review papers. Scientific interest in the enzymes belonging the gene Families CYP1 , CYP2 , and CYP3 ) was strengthened by the fact that these enzymes were found to be involved in most of the metabolism of clinically relevant drugs and toxicological responses to xenobiotic compounds, with some exceptions (e.g., Family 4 which in addition to some drugs catalyzes reactions in the metabolism of fatty acids and arachidonic acid). The enzymes belonging to the second group (P450s 5–51), reviewed in the present publication, preferentially catalyze biotransformations and metabolism of compounds that are of major importance in the physiological status of humans and consequently for human health. These enzymes catalyze the biosynthesis and metabolism of eicosanoids (P450s 5A1 and 8A1; Figs. 1 – 3 ), steroids (P450s 7A1, 11A1, 11B1, 17A1, 19A11, 21A1, 21A2, 27A1; Figs. 4A , 4B , 8 , 9 , 10 , 11 , 12 ), bile acids (P450s 7A1, 7B1, 8B1, 27A1, 39A1, 46A1; Figs 4A , 14A ), vitamin D 3 (P450 24A1, 27A1, 27B1; Figs. 4A , 14B ) and retinoids (26A1, 26B1, 26C1, 27C1; Fig. 21 ) and also the formation of cholesterol, the starting material for the biosynthesis of steroids, bile acids, and vitamin D) from lanosterol ( Figs. 4A and 17 ). Also in this group there are exceptions regarding drugs as substrates of the enzymes: P450 11B1, 11B2, 19A1, 21A2, and 46A1. P450 20A1 is considered as orphan enzyme in that its function still remains unknown. As humans are exposed to a great number of chemicals through food, environmental contamination, and medicines, there may be results with a significant positive or negative effect on catalytic activityies of the enzymes or expression of mRNAs or proteins. The data presented in the tables provide information to serve as a guide when considering possible drug-drug, drug-environmental chemicals, or chemical-physiological compounds interactions in clinic or in every day life. The positive effect of inhibition of P450 enzyme activity on human health is examplified by inhibition of P450 19A1, i.e. lowering estrogen levels by chemicals. Different chemicals can exert such an effect on the enzyme such as drugs (e.g., imidazoles, antiepileptic drugs, cyclooxygenase-selective inhibitors), natural compounds (e.g., flavonoids, polyphenols, and chalcones), physiological compounds (4-hydroxyandrostendione), and environmental chemicals, as well as number of developed drug candidates of different types explored for treatment of breast cancer ( Table 10 ). The following examples illustrate effects provoked by the action of drugs or other chemicals on the enzymes activity relating to human health: P450 8A1—elevated risk of hypertension and infarction when inhibited, chemo-preventive activity when induced ( Table 2 ); P450 5A1—anti-platelet aggregation and anti-tumor activity when inhibited, pro-carcinogenic role when induced ( Table 1 ); P450 11A1 and 17A1—prostate cancer treatment when inhibited ( Tables 6 and 9 , respectivaly). Analysis of the data from clinical samples showed that increases/or down-regulation of mRNA and/or protein expression of P450s 5A1, 7B1, 19A1, 26A1, 26B1, 26C1, 27A1, and 27B1 might be used as markers of the aggressive biological potential of tumors and association with poor patient survival. It has been suggested that up-regulation of these enzymes might be useful as tumor markers in the diagnosis and prognosis of different malignancies. In summary, the data presented show functional targets of P450 enzymes belonging to the Families 5–51. Of these enzymes, targets for drugs in clinical treatment are P450s 5A1, 11B1, 11B2, 17A1, and 19A1. The enzymes of this group synthesize important physiological molecules but over-production or down-regulation may be an issue in some diseases. The present review brings updated information and has focused on the properties of different chemical entieties (drugs, environmental, physiological, and natural compounds, as well as examples of drug candidates under investigation) as inhibitors and/or inducers of the enzymes activity and clinical implications.

The significance of the human cytochrome P450 (P450) enzymes in drug metabolism has been reviewed in detail in previous reviews ( Guengerich and Rendic 2010 ; Rendic and Guengerich 2010 , 2012 ; Guengerich 2015 ; Rendic and Guengerich 2015 ). In addition to a great number of compounds used as drugs or being found in the environment, and influencing the activity and/or expression of the cytochrome P450 enzymes ( Guengerich and Rendic 2010 ; Rendic and Guengerich 2012 ), the effects of diseases and environmental factors—including ionizing radiation, UV, γ -rays, X-rays—are also of interest and have been reviewed ( Semonin-Holleran 1991 ; Klammert et al. 2009 ; Guengerich and Rendic 2010 ; Rendic and Guengerich 2012 ). Such factors can have profound effects on enzyme activity and expression and therefore also on the final biological activity, efficacy, and safety of drugs and other chemicals. They can contribute to drug-drug, drug-chemical, or chemical-chemical interactions by modifying the disposition of xeno- and endobiotics and consequently their fate in the body. In some cases, analysis of the results on the effects of diseases and different environmental factors on human cytochrome P450 enzymes revealed inconsistency of the reported results, making it difficult to reach conclusions. The scope of this article is the human P450s in Families 5–51. The topic follows two relevant reviews that one of us wrote in the past four years ( Guengerich 2015 , 2017 ) on the P450s involved in metabolism of endogenous compounds. Why are we focusing on these P450s in a journal that deals with drug metabolism? There are several reasons. One is that some of the steroids and vitamins are used as drugs. Another is that most of these P450s have important roles in physiology and are subject to induction and/or inhibition by drugs. Finally, several of these P450s are functional targets for drugs, e.g. 5A1, 11B1, 11B2, 17A1, 19A1. These enzymes synthesize important molecules but over-production may be an issue in some diseases. Families 5–51 were covered in a chapter several years ago ( Guengerich 2015 ), but here we have focused on inhibitors and also updated the information. Most of the P450 enzymes in Families 5–51 are mainly extrahepatic, with several exceptions (7A1, 8B1, 26A1, 27A1, 39A1, 51A1). In contrast to the “drug-metabolizing” P450s in Families 1–4, the levels of expression of these enzymes are highly regulated and do not vary among individuals as much and, in general, are not very inducible by xenobiotics. Analyses of clinical tumor samples may link causality with changes in gene expression in some cases. For instance, increases/or downregulation in mRNA and/or protein expression of P450s 5A1, 7B1, 19A1, 26A1, 26B1, 26C1, 27A1, and 27B1 have been observed and suggested as markers of the aggressive biological potential of tumors and association with poor patient survival. It has been suggested that up-regulation of these enzymes might be useful as tumor markers in the diagnosis and prognosis of different malignancies. On the other hand, enhanced or lowered expression and/or activity of P450 enzymes in some diseases could result in clinically significant drug interaction potential, resulting in unfavorable clinical outcome or increased drug/chemical toxicity. Some examples include increased expression of P450 2E1 due to alcohol in healthy subjects and decreased enzyme expression in alcoholic liver disease, increased P450 2E1 expression in livers of transplant patients, high expression of P450 3A4 enzyme in lymphoid carcinoma (proposed as a useful predictor of poor response to the standard peripheral type lung cancer chemotherapy), and high expression of P450 3A enzymes in osteosarcomas (suggested as a predictor of metastasis and poor prognosis) ( Guengerich and Rendic 2010 ). For each of the 22 P450s we will review, the format will include a brief synopsis followed by a figure showing the main reaction and a table that includes a list of physiological substrates, function, and inhibitors and inducers. We have included references following the section on each P450 for the convenience of the reader. Collectively there is a total of 1,057 references for the entire review (not correcting for multiple entries in different sections). We have divided the review into several sections, based on the substrates for these P450s. The four sections include eicosanoids, steroids, vitamin D and related secosteroids, and retinoids. A fifth section is for P450 20A1, for which no substrates or functions have yet been characterized. This P450 remains an “orphan,” in our sense of the word ( Guengerich and Cheng 2011 ).