A universal approach to the expression of human and rabbit cytochrome P450s of the 2C subfamily in Escherichia coli

A conserved proline-rich sequence between the N-terminal signal-anchor and catalytic domains is required for assembly of functional cytochrome P450 2C2

In cytochrome P450 2C2, the region which links the N-terminal signal anchor with the catalytic domain contains a highly conserved proline-rich region with the sequence, 30-PPGPTPFP-37. Mutation of proline-30 or proline-33 diminished activities of the mutants expressed in COS-1 cells (Chen, C., and Kemper, B. (1996) J. Biol. Chem. 271, 28697-28611). Substitution of alanine, proline, or arginine for glycine-32 abolished laurate hydroxylase activity of the proteins expressed in COS-1 cells, which suggests that this residue is also functionally important. To determine the basis for the decreased activity in COS-1 cells, the activities and spectral properties of mutant proteins expressed in insect cells and bacteria were determined. Substitution of alanine for either proline-30 or -33 resulted in reduced expression in insect cells of functional cytochrome P450 hemoprotein and an increase in the expression of inactive cytochrome P420. In contrast, substitution of alanine for proline-31, -35, or -37 resulted in hemoproteins with spectra similar to cytochrome P450 2C2 so that the amount of cytochrome P450 expressed in insect cells correlated with the activities of the mutants in COS-1 cells. The laurate hydroxylase activities per nanomole of cytochrome P450 in insect microsomes were similar for wild type and all mutants, indicating that, once folded, the catalytic activity of membrane-bound cytochrome P450 was not affected by the mutations. Expression in bacteria resulted in diminished expression of cytochrome P450 for all mutants, with the greatest decrease for the proline-30 and -33 mutants, and increased cytochrome P420. In contrast to the insect cell studies, the proline-30 and -33 mutants were inactive, while the other mutants had specific activities 30-70% of cytochrome P450 2C2. These data are consistent with a role for the proline-rich region in efficient assembly of cytochrome P450 2C2 in eukaryotic cells. Mutations of this region also may affect the conformational integrity of the proteins, which was revealed by assays of solubilized bacterially expressed proteins.

Progesterone and testosterone hydroxylation by cytochromes P450 2C19, 2C9, and 3A4 in human liver microsomes

Roles of human cytochrome P450 (P450 or CYP) 2C9, 2C19, and 3A4 in the oxidation of progesterone and testosterone were studied in recombinant P450 enzymes and in human liver microsomes. In vitro inhibition experiments showed that progesterone and its 17alpha- and 21-hydroxylated metabolites and 11-deoxycortisol suppressed the CYP2C19-dependent R-warfarin 7-hydroxylation activities, with progesterone being the most active. These steroid chemicals also inhibited CYP2C9-dependent S-warfarin 7-hydroxylation activities though lesser extents seen with those in CYP2C19 enzyme. Progesterone was found to be a competitive inhibitor of CYP2C19 and CYP2C9 in human liver microsomes. Recombinant CYP2C19 catalyzed progesterone to form 21-hydroxyprogesterone as a major product and 16alpha- and 17alpha-hydroxyprogesterone as minor products. CYP2C9 also had progesterone 21-hydroxylation activities, although the activities were lower than those catalyzed by CYP2C19. Vmax/Km ratios for the progesterone 21-hydroxylation activity of CYP2C19 were determined to be 13- and 32-fold higher than those of CYP2C9 and 3A4, respectively. CYP3A4 oxidized progesterone to form 16alpha-, 6beta-, and 2beta-hydroxyprogesterone as major products and 21-hydroxyprogesterone as a minor product, but did not produce detectable levels of 17alpha-hydroxyprogesterone. Immunoinhibition experiments suggested that anti-CYP2C9 (which inhibits both CYP2C9 and CYP2C19 catalytic activities) suppressed the progesterone 21-hydroxylation activities catalyzed by liver microsomes of humans and monkeys and that anti-CYP2C11 inhibited the progesterone 21-hydroxylation activities catalyzed by liver microsomes of male rats. CYP2C19 was also found to oxidize testosterone at 17-position to form androstenedione. Androstenedione formation catalyzed by liver microsomes of humans and monkeys and of male rats was suppressed by anti-CYP2C9 and anti-CYP2C11, respectively. These results suggest that CYP2C19 plays important roles in the oxidation of progesterone and testosterone in human liver microsomes, although the physiological significance of these metabolic pathways remains unclear. CYP2C9 may have some, but lesser extent than those by CYP2C19, of the catalytic roles for the metabolism of progesterone and testosterone by human liver microsomes.

A general strategy for the expression of recombinant human cytochrome P450s in Escherichia coli using bacterial signal peptides: expression of CYP3A4, CYP2A6, and CYP2E1

Heterologous expression of unmodified recombinant human cytochrome P450 enzymes (P450s) in Escherichia coli has proved to be extremely difficult. To date, high-level expression has only been achieved after altering the 5'-end of the native cDNA, resulting in amino acid changes within the P450 protein chain. We have devised a strategy whereby unmodified P450s can be expressed to high levels in E. coli, by making NH2-terminal translational fusions to bacterial leader sequences. Using this approach, we initially tested two leader sequences, pelB and ompA, fused to CYP3A4. These were compared with an expression construct producing a conventional NH2-terminally modified CYP3A4 (17alpha-3A4). Both leader constructs produced spectrally active, functional protein. Furthermore, the ompA-3A4 fusion gave higher levels of expression, and a marked improvement in the recovery of active P450 in bacterial membrane fractions, when compared with 17alpha-3A4. We then tested the ompA leader with CYP2A6 and CYP2E1, again comparing with the conventional (17alpha-) approach. As before, the leader construct produced active enzyme, and, for CYP2E1 at least, gave a higher level of expression than the 17alpha-construct. The ompA fusion strategy thus appears to represent a significant advance for the expression of P450s in E. coli, circumventing the previous need for individual optimization of P450 sequences for expression.

Efficient assembly of functional cytochrome P450 2C2 requires a spacer sequence between the N-terminal signal anchor and catalytic domains

Cytochromes P450 (P450) are anchored to the endoplasmic reticulum membrane by an N-terminal transmembrane sequence with the catalytic domain facing the cytoplasmic side. Within the peptide sequence linking these two domains in P450 2C2 is a glycine-rich region from residues 22 to 28. To examine the role of this region, deletion and substitution mutations were constructed, and the activities and spectral properties were determined for the mutant proteins expressed in COS-1 cells, insect cells, and bacteria. Deletion of residues 22 to 28 or substitution of 7 valines for this region inactivated the proteins in COS-1 cells, and no P450 species was detected for these mutations in bacteria or insect cells. Substitution of the three glycine residues with alanine or proline or the entire sequence from 22 to 28 with 7 alanines did not reduce lauric acid hydroxylase activity of the proteins expressed in COS-1 cells. Reducing the number of alanines substituted to 4, 3, and 2 progressively decreased activity in COS-1 cells to undetectable levels when 2 alanines were substituted. The loss of activity in COS-1 cells correlated with decreased expression of hemoprotein with a reduced difference spectrum of 450 nm (P450 species) and a corresponding increase in the inactive P420 species in insect cells and bacteria. The activities expressed per nanomole of P450 in insect microsomes were similar for P450 2C2 and the alanine substitution mutants, including the mutant with 2 alanines which was inactive in COS-1 cells. The rates of conversion of P450 to P420 resulting from incubation at 48 degrees C in vitro were not changed sufficiently to explain the increase in expressed P420 observed for the mutants with 3 or 7 alanines substituted. These data are consistent with a role for the residue 22-28 region as a linker that facilitates the folding of P450; however, once the protein is properly folded into the functional P450 species, this region has little influence on the stability and activity of the enzyme.

Expression, purification, and enzymatic properties of recombinant human cytochrome P450c27 (CYP27)

A large number of microsomal P450s have been expressed in Escherichia coli in quantities sufficient for structure/function analysis. However, only one mitochondrial P450 has been successfully overexpressed, that being cholesterol side chain cleavage cytochrome P450 (P450scc). We report here overexpression, purification, and characterization of a second mitochondrial P450, human sterol C-27 hydroxylase (P450c27). The conditions used for expression are very similar to those applied for P450scc, although a quite different purification protocol was necessary to achieve highly purified P450c27. The catalytic properties of purified recombinant human P450c27 resemble those of purified, endogenous rat and rabbit P450c27, regarding both specificity and turnover numbers. Like endogenous P450c27 from rat and rabbit liver, human recombinant P450c27 is only functional in the presence of adrenodoxin and adrenodoxin reductase and shows no activity in the presence of the microsomal P450 reductase. We conclude that P450c27 is most likely not the 1alpha-hydroxylase of 25-hydroxyvitamin D3, contrary to a previous suggestion (Axen, E., Postlind, H., Sjöberg, H., and Wikvall, K. (1994) Proc. Natl. Acad. Sci. USA 91, 10014-10018) because this activity of P450c27 (28 pmol/min/nmol P450) seems far too low to be physiologically relevant. This activity is 10(3) times lower than the 27-hydroxylase activity toward 5beta-cholestane-3alpha,7alpha,12alpha-triol and 40 times lower than the 27-hydroxylation of cholesterol by this enzyme. The development of this expression system and purification procedure creates the potential for structure/function analysis of P450c27, including possible crystallization of this important enzyme.

Reconstitution of recombinant cytochrome P450 2C10(2C9) and comparison with cytochrome P450 3A4 and other forms: effects of cytochrome P450-P450 and cytochrome P450-b5 interactions

Tolbutamide methyl hydroxylation and S-warfarin 7-hydroxylation activities were reconstituted in systems containing recombinant human cytochrome P450 (P450 or CYP) 2C10(2C9) and the optimal conditions for the systems were compared with those of bufuralol 1'-hydroxylation by CYP1A1, theophylline 8-hydroxylation by CYP1A2, bufuralol 1'-hydroxylation by CYP2D6, chlorzoxazone 6-hydroxylation by CYP2E1, and testosterone 6 beta-hydroxylation by CYP3A4. CYP2C10 required cytochrome b5 (b5) for optimal rates of tolbutamide and S-warfarin oxidations and b5 could be replaced by apo-b5; apo-b5 and b5 effects on the reconstituted systems have already been reported in systems containing CYP3A4 for the oxidation of testosterone and nifedipine and for the rapid reduction of CYP3A4 by NADPH-P450 reductase (H. Yamazaki et al., 1996, J. Biol. Chem. 271, 27438-27444). Stopped-flow studies, however, suggested that apo-b5 as well as b5 did not cause stimulation of the reduction of CYP2C10 by NADPH-P450 reductase, while the reduction rates were dependent on the substrates in reconstituted systems. Chlorzoxazone 6-hydroxylation by CYP2E1 was stimulated by b5, but not by apo-b5, in reconstituted systems. Neither apo- nor holo-b5 increased bufuralol 1'-hydroxylation activity by CYP1A1 or 2D6 or theophylline 8-hydroxylation by CYP1A2. Interestingly, we found that testosterone 6 beta-hydroxylation by CYP3A4 was stimulated by CYP1A2 (and also by a modified form in which the first 36 residues of the native human protein were removed) and CYP1A1 as well as by b5, and such stimulations were not seen when other P450 proteins (e.g., CYP2C10, 2D6, or 2E1) were added to the reconstituted systems. In contrast, substrate oxidations by CYP2C10 and CYP2E1 were not stimulated by other P450 proteins. The present results suggest that there are differences in optimal conditions for reconstitution of substrate oxidations by various forms of human P450 enzymes, and in some P450-catalyzed reactions protein-protein interactions between P450 and b5 and other P450 proteins are very important in some oxidations catalyzed by CYP2C10, 2E1, and 3A4.

Microsomal P450 2C3 is expressed as a soluble dimer in Escherichia coli following modification of its N-terminus

A hydrophobic segment present in the N-terminus of microsomal P450s is thought to serve as a membrane anchor. A variant of P450 2C3 was constructed, P450 2C3d, that lacked the putative membrane-spanning segment of the N-terminus, residues 3-20. This construct also incorporated substitutions of an alanine for 2Asp to facilitate expression in Escherichia coli and of serines for 24His and 25Gly to introduce a restriction site. P450 2C3d is expressed at relatively high levels in E. coli, 800-1200 nmol/liter of culture medium. In contrast to P450 2C3mod, which retains a membrane-spanning N-terminal sequence modified for expression in E. coli, the subcellular distribution of P450 2C3d in E. coli is dependent on the ionic strength of the buffer used for cell disruption. In low ionic strength buffers, 2C3d was mainly localized in the membrane fraction, whereas in buffers containing 1 M NaCl or 0.5 M KPi, P450 2C3d was predominantly found in the soluble fraction, indicating that deletion of the hydrophobic segment converted the intrinsic membrane protein to an extrinsic one. P450 2C3d was further modified by the incorporation of four histidine residues at the C-terminus (P450 2C3dH), and this enzyme could be purified in the absence of detergent using immobilized metal affinity chromatography following extraction from isolated membranes in high salt buffers. The catalytic properties of the purified, modified enzymes are similar to those of the native enzyme. Size-exclusion chromatography indicated that 2C3dH and 2C3d are predominantly dimers, whereas 2C3 is a larger oligomer (> 8-mer). Moreover, the detergents sodium cholate and Chaps each dissociate the dimers of 2C3dH to monomers at concentrations that do not alter the aggregation state of 2C3. These modifications are likely to facilitate attempts to crystallize the catalytic domains of microsomal P450s.

Targeted antipeptide antibodies to cytochrome P450 2C18 based on epitope mapping of an inhibitory monoclonal antibody to P450 2C51

The epitope recognized by the inhibitory monoclonal antibody designated 2F5, which was raised against P450 2C5, was mapped to amino acids 237-260 by immunoblotting using a combination of recombinant antigens and chimeric and partial fusion proteins constructed from rabbit P450s 2C2, 2C4, 2C5, and 2C16, which are recognized by 2F5, and from 2C1 and 2C3, which are not. When the sequence of the epitope for 2F5 (amino acids 237-260) was compared with those of other rabbit 2C P450s, a single lysine residue at position 253 appeared to be a likely determinant of 2F5 immunoreactivity. Substitution of lysine for glutamic acid 253 in P450 2C3 (2C3E253K) conferred immunoreactivity and the ability of 2F5 to inhibit progesterone metabolism catalyzed by P450 2C3E253K. Sequence alignment revealed that this epitope lies in close proximity to the epitope identified for LKM-1 autoantibodies to P450 2D6. Based on these results, an antipeptide antibody was raised to the corresponding region (amino acids 252-263) of human P450 2C18. The resulting antipeptide antiserum recognizes P450 2C18 but not P450 2C8, 2C9, or 2C19. However, the antipeptide 2C18 antiserum did not inhibit 2C18-catalyzed diazepam N-demethylation. Human 2C P450s were also quantitated by immunoblot analysis in a panel of six human liver microsomes using Escherichia coli expressed P450s as standards. Analysis of immunoblots indicated that, if present, P450 2C18 was expressed at very low levels (<2.5 pmol/mg), whereas P450s 2C8, 2C9, and 2C19 were easily detected.

Cloning and functional expression in yeast of a cDNA coding for an obtusifoliol 14alpha-demethylase (CYP51) in wheat

Screening of a wheat cDNA library with an heterologous CYP81B1 probe from Helianthus tuberosus led to the isolation of a partial cDNA coding a protein with all the characteristics of a typical P450 with high homology (32-39% identity) to the fungal and mammalian CYP51s. Extensive screening of several wheat cDNA libraries isolated a longer cDNA (W516) coding a peptide of 453 amino acids. Alignment of W516 with other P450 sequences revealed that it was missing a segment corresponding to the N-terminal membrane anchor of the protein. The corresponding segment from the yeast lanosterol 14alpha-demethylase was linked to the partial wheat cDNA and the chimera expressed in Saccharomyces cerevisiae. Compared to microsomes from control yeasts, membranes of yeast expressing the chimera catalysed 14alpha-demethylation of obtusifoliol with an increased efficiency relative to lanosterol demethylase activity. W516 is thus a plant member of the most ancient and conserved P450 family, CYP51.

Optimization of heterologous protein production in Escherichia coli

Escherichia coli has long been the primary prokaryotic host for the synthesis of heterologous proteins. Recent advances have been made in the expression of complex proteins as soluble, functional molecules, complete with prosthetic groups, disulfide bonds, and quaternary structure. The development of alternative promoter and induction strategies has improved the options available for manipulating the expression conditions, which are frequently critical to soluble yield.

New applications of bacterial systems to problems in toxicology

Bacterial systems have long been of use in toxicology. In addition to providing general models of enzymes and paradigms for biochemistry and molecular biology, they have been adapted to practical genotoxicity assays. More recently, bacteria also have been used in the production of mammalian enzymes of relevance to toxicology. Escherichia coli has been used to express cytochrome P450, NADPH-cytochrome P450 reductase, flavin-containing monooxygenase, glutathione S-transferase, quinone reductase, sulfotransferase, N-acetyltransferase, UDP-glucuronosyl transferase, and epoxide hydrolase enzymes from humans and experimental animals. The expressed enzymes have been utilized in a variety of settings, including coupling with bacterial genotoxicity assays. Another approach has involved expression of mammalian enzymes directly in bacteria for use in genotoxicity systems. Particularly with Salmonella typhimurium. Applications include both the reversion mutagenesis assay and a system using a chimera with an SOS-response indicator and a reporter.