Engineering a mineralocorticoid- to a glucocorticoid-synthesizing cytochrome P450.

Site-directed mutagenesis of a domain (amino acids 299-338) aligning to the I-helix region of P450cam, P450BM3 and P450terp was used to investigate the different regioselectivities displayed in the hydroxylation reactions performed by human aldosterone synthase (P450aldo) and 11beta-hydroxylase (P45011beta). The two enzymes are 93% identical and are essential for the synthesis of mineralocorticoids and glucocorticoids in the human adrenal gland. Single replacement of P450aldo residues for P45011 beta-specific residues at positions 296, 301, 302, 320, and 335 only gave rise to slightly increased 11beta-hydroxylase activities. However, a L301P/A320V double substitution increased 11beta-hydroxylase activity to 60% as compared with that of P45011 beta. Additionally substituting Ala-320 for Val-320 of P45011 beta further enhanced this activity to 85%. The aldosterone synthase activities of the mutant P450aldo proteins were suppressed to a varying degree, with triple replacement mutant L301P/E302D/A320V retaining only 10% and double replacement mutant L301P/A320V retaining only 13% of the P450aldo wild type activity. These results demonstrate a switch in regio- and stereoselectivities of the engineered P450aldo enzyme due to manipulation of residues at three critical positions, and we attribute the determination of these features in P450aldo to the structure of a region analogous to the I-helix in P450cam.

Site-directed mutagenesis of a domain (amino acids 299 -338) aligning to the I-helix region of P450 cam , P450 BM3 and P450 terp was used to investigate the different regioselectivities displayed in the hydroxylation reactions performed by human aldosterone synthase (P450 aldo ) and 11␤-hydroxylase (P450 11␤ ). The two enzymes are 93% identical and are essential for the synthesis of mineralocorticoids and glucocorticoids in the human adrenal gland. Single replacement of P450 aldo residues for P450 11␤ -specific residues at positions 296, 301, 302, 320, and 335 only gave rise to slightly increased 11␤-hydroxylase activities. However, a L301P/A320V double substitution increased 11␤-hydroxylase activity to 60% as compared with that of P450 11␤ . Additionally substituting Ala-320 for Val-320 of P450 11␤ further enhanced this activity to 85%. The aldosterone synthase activities of the mutant P450 aldo proteins were suppressed to a varying degree, with triple replacement mutant L301P/E302D/A320V retaining only 10% and double replacement mutant L301P/A320V retaining only 13% of the P450 aldo wild type activity. These results demonstrate a switch in regio-and stereoselectivities of the engineered P450 aldo enzyme due to manipulation of residues at three critical positions, and we attribute the determination of these features in P450 aldo to the structure of a region analogous to the I-helix in P450 cam .
In the adrenal gland essential steroid hormones such as glucocorticoids, mineralocorticoids, and androgens are produced. Cortisol, the major glucocorticoid in humans, is synthesized in the zona fasciculata/reticularis under control of pituitary derived adrenocorticotropic hormone, whereas the most potent mineralocorticoid, aldosterone, is secreted from zona glomerulosa cells primarily in response to angiotensin II and potassium (1)(2)(3). This differential secretion is achieved by a diverging expression pattern of a series of monooxygenases, also called P450 1 enzymes, which catalyze a multistep process providing the organism with the effective hormones. In humans the final steps in cortisol and aldosterone production, precisely 11␤-hydroxylation in the zona fasciculata/reticularis and 11␤-hydroxylation, 18-hydroxylation, and 18-oxidation in the zona glomerulosa, are performed by two distinct enzymes, namely 11␤-hydroxylase (P450 11␤ ) and aldosterone synthase (P450 aldo ). The genes encoding these enzymes, CYP11B1 (P450 11␤ ) and CYP11B2 (P450 aldo ), have been isolated from a genomic library (4) and were shown by sequence comparison to be members of the superfamily of cytochrome P450 genes (5). Further structural characterization revealed nine exons spanning the genes, which are tandemly arranged on chromosome 8q22 (6,7). The molecular masses of the respective proteins have been determined to be 50 (P450 11␤ ) and 48.5 (P450 aldo ) kDa (8). Both enzymes, after being synthesized, are translocated into the mitochondrial matrix where they are bound to the inner membrane by as yet not precisely defined segments of the proteins. There, accompanied by an NADPH-dependent redox system consisting of a flavoprotein, adrenodoxin reductase, and an iron-sulfur protein, adrenodoxin, donating reducing equivalents, they participate in steroid hydroxylation. P450 11␤ enzymes of other species have been extensively studied, and it turned out that in bovine (9), porcine (10), and frog (11) adrenal cortex, synthesis of gluco-and mineralocorticoids is catalyzed by a single enzyme. Conversely, synthesis of human, rat (12), and mouse (13) gluco-and mineralocorticoids have been separated in evolution and are carried out by distinct enzymes, yet the reason for these interspecies differences is enigmatic. We intended to gain insight into the principles underlying the different regioselectivities involved in 11␤-hydroxylation and 18-hydroxylation/oxidation in the human enzymes. Since both proteins are 93% identical yet carry out separate reactions to yield different steroid hormones, it remained elusive on which structure-function relationships these diversities could be based. Recently, the cause of glucocorticoidremediable aldosteronism, an autosomal dominant disorder in humans, was reported to arise from unequal crossing-over events between the CYP11B1 and CYP11B2 genes (14). The resulting chimeric genes comprise a 5Ј CYP11B1 portion and a 3Ј CYP11B2 portion under control of the CYP11B1 regulatory region. Pascoe et al. (15), through the analyses of hybrid proteins, determined the C-terminal 247 amino acids of P450 aldo as crucial for aldosterone synthesis. Keeping this in mind, we carried out a computer-based sequence and structure alignment with three of the four by now crystallized P450 proteins, namely P450 cam from Pseudomonas putida (16), P450 BM3 from Bacillus megaterium (17), and P450 terp from another Pseudomonas species (18). We performed site-directed mutagenesis on a region supposedly analogous to the P450 cam I-helix and subsequent analyses of the mutants by transient transfection experiments using COS-1 cells. This led to the identification of mutant P450 aldo proteins having dramatically increased 11␤hydroxylase activity, which in the P450 aldo wild type protein is considerably lower than in the P450 11␤ wild type protein (19). Concomitantly, aldosterone synthase activity in these mutants was lost to a substantial degree, indicating that regioselectivities have successfully been switched from one position to the other.

Materials
COS-1 cells were obtained from the American Type Culture Collection. Culture media and antibiotics were from Seromed (Berlin), and fetal bovine serum was from Life Technologies, Inc.
Oligonucleotides were synthesized on an Applied Biosystems model 380A DNA synthesizer at BioTez (Berlin). Restriction enzymes, Klenow fragment of DNA polymerase I, T4 polynucleotide kinase, bovine alkaline phosphatase, T4 DNA ligase, and DH5␣ cells were purchased from New England Biolabs Inc. or Boehringer Mannheim. Taq polymerase was obtained from Perkin-Elmer. pALTER-1 plasmid, helper phage R408, JM109, and ES1301 mutS cells were obtained as part of the Altered Sites in vitro mutagenesis system (Promega Co., Madison, WI). pRc/CMV was from Invitrogen, and pBS SK(ϩ) was from Stratagene. The ⌬ Taq TM cycle sequencing kit was purchased from U.S. Biochemical Corp., and [ 35 S]dATP was from Amersham Corp. [ 3 H]Deoxycorticosterone and [ 3 H]deoxycortisol were obtained from DuPont NEN. Deoxycorticosterone, corticosterone, deoxycortisol, cortisol, chloroquine, cell culture-tested HEPES, and dimethyl sulfoxide were purchased from Sigma. DEAE-dextran was purchased from Pharmacia Biotech Inc. Radioimmunoassays were performed with Active TM coated radioimmunoassay kits from Diagnostic System Laboratories Inc. ECL Western blotting reagents were obtained from Amersham, polyclonal hemagglutinin (HA) 11 antibody was from the BAbCO Berkeley Antibody Company, and the polyclonal anti-bovine adrenodoxin (Adx) antibody was raised in rabbit by Eurogentec sa . Nitrocellulose membrane was used from Schleicher & Schuell.

Methods
Epitope Tagging of the CYP11B1 and CYP11B2 cDNAs and Construction of the Parent Plasmids Expressing P450 aldo and P450 11␤ -To be able to detect proteins in transfected COS-1 cells an HA epitope (32) was C-terminally fused to the P450 aldo and P450 11␤ encoding cDNAs via polymerase chain reaction. A common 5Ј-oligonucleotide, 5Ј-GGGGTCTAGAATGGCACTCAGGGCAAAGGC, harboring an XbaI site and the translation initiation site was used. BamHI-linked 3Јoligonucleotides were 5Ј-GGGCCGGATCCCTACAGGCTAGCGTAAT-CTGGAACATCGTATGGGTAGTTAATCGCTCTGAAAGTGAG for CY-P11B2 and 5Ј-GGGCCGGATCCTTACAGGCTAGCGTAATCTGGAA-CATCGTATGGGTAGTTGATGGCTCTGAAGGTGAG for CYP11B1. The stop codons are underlined, and the HA epitope encoding sequence is in italics. The 1548-base pair fragments were subcloned into pBS SK(ϩ) to yield pBS-B1-ha and pBS-B2-ha. pRc/CMV was digested with Bsp120I, blunted with Klenow fragment from DNA-polymerase I and further cut with XbaI to be subsequently ligated with a XbaI/EcoRV fragment from pBS-B1WT and pBS-B2WT, respectively. The sequences were verified by nucleotide sequencing using the chain termination method (20) and the ⌬ Taq cycle sequencing kit.
Insertion of Mutations into the CYP11B2 cDNA by Site-directed Mutagenesis-The XbaI/BamHI fragment from pBS-B2-ha was inserted into the pALTER-1 phagemid and transformed into JM109. Site-directed mutagenesis was performed according to Kunkel et al. (21) following instructions by the supplier (Promega). For the generation of multiple mutants several cycles of the mutagenesis procedure were performed by amp knock-out and tet repair or vice versa using the same vector molecule. The base substitutions were confirmed by sequencing first in pALTER-1 and a second time in pRc/CMV, into which they were inserted. All standard procedures were carried out as described by Sambrook et al. (22).
Cell Culture and Transfection-COS-1 cells were maintained at 37°C and 8% CO 2 in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 100 units/ml penicillin, and 0.1 mg/ml streptomycin. Transfections were done as described by Zuber et al. (23) except for some minor modifications. One day prior to transfection cells were plated at a density of 350,000/6-cm dish. For transfection cells were starved for 1 h in medium containing 20 mM HEPES and lacking fetal bovine serum, which then was aspired and substituted for the transfection mixture consisting of 10 g of P450 expression plasmid, 5 g of pBAdx4 (a kind gift from M. Waterman, Nashville, TN), and 0.5 mg of DEAE-dextran (M r 500,000) in starvation medium. After another incubation of 1 h, chloroquine was added in 2 ml of complete medium to a final concentration of 52 M, and cells were kept under these conditions for 4 h. Finally, a 2-min treatment with dimethyl sulfoxide in Dulbecco's modified Eagle's medium was carried out; cells were washed twice in Hank's balanced salt solution and grown for 24 h before substrate addition.
Hydroxylase Assays-Transfected cells were incubated for 48 h with For extraction of steroids the medium was combined with 4 volumes of an ethanol/acetone mixture (1:1) and incubated at room temperature (24). After pelleting the debris, the supernatant was transferred to a fresh tube and evaporated to the original volume. Steroids were extracted with 2 volumes of methylene chloride, and the organic phase was evaporated. The residue was dissolved in 0.5 ml of hexane and centrifuged at maximal speed, giving a small pellet. Then the supernatant was rotated under vacuum to dryness, and the extraction products were redissolved in 150 l of 10% (v/v) isopropyl alcohol in hexane. After the addition of a combination of internal steroid standards the samples were subjected to normal phase HPLC using a Lichrosorb ® Diol column (Merck, Darmstadt, FRG). A gradient solvent system was run starting with 15% (v/v) isopropyl alcohol in n-hexane at a flow rate of 1.3 ml/min. The standards were monitored by UV detection at 254 nm, while the radioactivity was assayed with a Betascan Sc radiation detector fitted to the HPLC. Alternatively, steroids were measured using an Active TM aldosterone or Active TM cortisol radioimmunoassay.
Immunodetection of Expressed Proteins-Transfected cells were washed twice with ice-cold phosphate-buffered saline and scraped in 0.5 ml of phosphate-buffered saline. Lysis was done in radioimmune precipitation buffer (10 mM Tris-Cl, pH 7.4, 1% sodium deoxycholate, 1% Triton X-100, 0.1% SDS, 150 mM NaCl, 1 mM EDTA, and 1 mM phenylmethylsulfonyl fluoride) supplemented with 0.04 units of aprotinin/ml (Sigma) and 0.1 g/ml leupeptin (Sigma) for 10 min on ice followed by 6 s of sonication, and lysates were boiled for 5 min in SDS-polyacrylamide gel electrophoresis sample buffer. Proteins were separated by SDS-polyacrylamide gel electrophoresis according to Laemmli (25), blotted to a nitrocellulose filter, and detected by hybridization with anti-HA or anti-bovine Adx antibodies using the ECL system from Amersham.
Alignment of P450 Sequences-An initial alignment of the P450 aldo and P450 11␤ sequences was performed using HOMOL, a data base created by D. Nelson in which he related 155 different P450 proteins to P450 cam . 2 The mode of alignment takes the identity and resemblance of single amino acid residues as well as secondary structure predictions into consideration. We added the structural alignments of P450 BM3 and P450 terp to our assignation in order to increase possible predictions for structural entities in P450 aldo and P450 11␤ . Updating of the sequences and further optimization was carried out by hand. In addition, we included the sequences of P450 scc , P450 arom , and P450 aldo/11␤ from other species by using CLUSTAL. The first amino acid of the different steroidogenic proteins to be considered was the first residue of the mature sequences.

RESULTS
Alignment of P450 aldo and P450 11␤ with P450 CAM , P450 BM3 , and P450 terp -Secondary structure is much better conserved in cytochrome P450 proteins than would be expected from an often rather low sequence homology, which in the case of P450 aldo and P450 11␤ compared with P450 cam ranks around 20%. This observation seems to hold especially true for the characteristic helices A to L. Comparison of the P450 aldo and P450 11␤ sequences with each other reveals 29 amino acid deviations in the mature and 32 amino acid deviations in the unprocessed proteins, respectively. Considering the distribu-2 D. Nelson, personal communication. Residues predicted to be contained in the putative I-helix are in white on black, and those positions (296, 301, 302, 320, 335, and 339) varying between the human P450 aldo and P450 11␤ sequences in this region are printed in boldface type. The P450 aldo and P450 11␤ were aligned to P450 cam as described under "Experimental Procedures," and the P450 BM3 and P450 terp are aligned with P450 cam as described by Ravichandran et al. (17) and Hasemann et al. (18). In addition we included the sequences of the human P450 arom and P450 scc in the alignment.
Site-directed Mutagenesis of P450 aldo tion of these mismatches, the alignment ( Fig. 1) shows that five of them cluster to a region that could be analogous to the P450 cam I-helix. P450 BM3 and P450 terp also possess the same helical structure as was found in crystallographic investigations and structural alignment to P450 cam (17,18). This putative domain in P450 aldo and P450 11␤ would be encoded by the second half of exon 5 and the first half of exon 6 of the genes, a region 3Ј of amino acid 256, which was reported to be a critical breaking point in the study of hybrid P450 aldo and P450 11␤ proteins. The P450 aldo portion C-terminal of this residue seems to be important for aldosterone synthase activity (15). Since the I-helix in P450 cam , apart from the heme binding site and a region between the B-and C-helices containing Tyr-96 (26), contributes to the active site, it also may be, in P450 aldo and P450 11␤ , a determinant of substrate binding and specificity. Furthermore, Thr-252 of P450 cam is highly conserved in P450 proteins (27) and in P450 cam was reported to play a critical role in oxygen activation (28 -30). Taking all these observations into account, positions 301, 302, 320, and 335 of P450 aldo and P450 11␤ could be candidate amino acids for determination of the different regioselectivities in hydroxylation reactions performed by these proteins. Positions 296 and 339 in our analyses did not map to the very 'putative' I-helix region but should be flanking it (Fig. 1) and thus may have an influence on the positioning of the helix.
Construction of the P450 aldo and P450 11␤ Parent Expression Plasmids and of P450 aldo Mutants-To obtain full-length fragments of P450 aldo and P450 11␤ encoding cDNAs we performed reverse transcription-polymerase chain reaction on total RNA from a human adrenal gland as described elsewhere (31). The P450 11␤ cDNA sequence corresponded to the one published by Mornet et al. (4), whereas in the P450 aldo sequence amino acid 173 was substituted from Arg to Lys, which was also found by Kawamoto et al. (3). After verifying the functionality of both cDNAs by expression studies we further engineered them by adding an HA epitope tag (32) to both C termini in order to enable protein detection in subsequent expression experiments.
To investigate the role of the putative I-helix in the regioselectivity of P450 aldo -and P450 11␤ -specific hydroxylation reactions, we performed site-directed mutagenesis on the P450 aldo cDNA. Derived from the alignment studies, the region corresponding to the I-helix in P450 cam extends from amino acid 299 to 338 (Fig. 2). This part of the total sequence includes four diverging amino acids at positions 301, 302, 320, and 335, but we also analyzed the role of Lys-296 in our investigations, since also a flanking amino acid could exert an influence on the positioning of the putative I-helix and thereby the regiospecificity of hydroxylation. Mutants were generated using the mutagenic oligonucleotides listed in Table I. Thus, P450 aldo -specific amino acid residues were substituted for P450 11␤ -specific ones, and single, double, and triple replacement mutants of P450 aldo were created.
Expression and Analyses of P450 aldo Mutants-To analyze the mutant versus the wild type proteins with respect to their hydroxylation specificities we cotransfected the resultant plasmids together with pBAdx4 into COS-1 cells. Expression of bovine adrenodoxin stimulates the activity of mitochondrial steroidogenic P450 enzymes, since the intrinsic amount of adrenodoxin in COS-1 cells seems inadequate for maximum electron transfer between the reductase and the P450 component of the system (23). Since the level in 11␤-hydroxylase activity toward 11-deoxycortisol is the most distinguishing feature of P450 aldo and P450 11␤ , transfected cells were first assayed for their ability to convert 11-deoxycortisol to cortisol, and subsequently their aldosterone-synthesizing potential was estimated. Incubation with low substrate concentrations (0.25 M 11-deoxycortisol) did not reveal significant differences among P450 aldo and P450 11␤ wild type proteins under the conditions tested. The prior enzyme, although 11␤-hydroxylating at a much lower rate than the latter one, was able to convert more than 90% of 11-deoxycortisol to cortisol within a 48-h interval. However, using higher substrate concentrations (2.5 to 5 M), clear differences were observed among P450 aldo and P450 11␤ so that intermediate activity mutants could be arranged between a maximal and a minimal limit. These flanking values are indicated in representative HPLC profiles in Fig. 3. In addition, at higher substrate concentrations the endogeneous ability of COS-1 cells to dehydrogenate cortisol to cortisone by the action of an 11␤-hydroxysteroid dehydrogenase became negligible, and the addition of glycyrrhetinic acid, an inhibitor of this activity, was unnecessary. Each mutant was assayed in cell culture by incubation with [ 3 H]deoxycortisol, and metabolites were either separated on HPLC or quantitated by radioimmune assay. The results are summarized in Fig. 4A. The small cortisol values detected in mock-transfected COS-1 cells were due to the antibody used in radioimmune assays, which exerted a slight cross-reactivity to 11-deoxycortisol. Relative to P450 aldo wild type, the activities of the single replacement P450 aldo mutants substituted at positions 296 and 335 were only slightly increased, whereas inserting a P450 11␤ residue at position 301, 302, or 320 did enhance the 11␤-hydroxylation potential 1.5-2-fold, suggesting that the latter amino acids could be of relative importance for 11␤-hydroxylase activity. Combination of single substitutions resulting in double and triple replacement mutants, however, did lead to much stronger effects. Triple mutant L301P/E302D/A320V, showing the most pronounced increase in 11␤-hydroxylation, accounted for about 85% of the P450 11␤ wild type activity. Also the combined substitutions L301P and A320V did result in an activity en- Amino acids differing between the two proteins are printed in boldface type. P450 aldo residues at positions 296, 301, 302, 320, and 335 were mutated to P450 11␤ residues, and the HA epitope was fused to the C termini of the proteins as described under "Experimental Procedures."

TABLE I
Mutagenic oligonucleotides used for the construction of P450 aldo mutants For construction of double and triple replacement mutants, oligos were either combined in a single mutagenesis reaction or in subsequent ones. The given sequences correspond to the noncoding cDNA strand, and the bases causing an amino acid substitution are printed in boldface type.
Oligonucleotides Substitution hanced to 60% as compared with P450 11␤ wild type. Because of the dual ability of the P450 aldo wild type enzyme to produce both cortisol and aldosterone in vitro we investigated aldosterone synthase activity in this set of P450 aldo mutants. The primary question was whether one of the two activities (cortisol synthesis) could be increased without affecting the other activity (aldosterone synthesis) or whether there is a reciprocal behavior to be found, meaning no increase in one activity without loss in the other. Examination of the aldosterone-synthesizing abilities of the mutants showed that triple mutant L301P/E302D/A320V only retained about 10% and double mutant L301P/A320V about 13% of the P450 aldo wild type activity (Fig. 4B). Mutants L301P/E302D and A320V were only slightly compromised in their activities, but the combination of both resulting in the triple mutant synergize drastically to a severe loss in 18-hydroxylation or 18-oxidation activity. Paradoxically, double mutant L301P/E302D was less severely affected in its aldosterone-synthesizing capacity (87% retained as compared with P450 aldo wild type) than the respective single mutants. Introducing P450 11␤ residues at positions 296 and 335 decreased activity to about 40% in the single mutants. The combined double mutant is reduced to about 15% in its aldosterone-synthesizing capacity as compared with P450 aldo wild type, and additionally substituting Asp at position 335 for Asn totally destroys aldosterone synthase activity. These data show that every alteration made in this region did negatively affect the aldosterone synthase activity of the recombinant enzyme although to a varying degree.
Immunodetection of Proteins-To examine protein synthesis in transfected COS-1 cells we performed Western blotting of whole cell extracts. P450 11␤ and P450 aldo wild type proteins and the mutants created were detectable to a very similar amount, indicating that the amino acid substitution did not notably impair protein translation and stability (Fig. 5). This, of course, is not a means to distinguish between an apo-and a holoprotein, but one would expect that an apoprotein tends to be malfolded and thus readily would become a target of the mitochondrial degradation machinery. Due to the 11 amino acids comprising the HA tag fused to the C termini of the proteins the molecular masses of the proteins were increased to about 56 kDa as compared with the wild type proteins having molecular masses of 50 and 48.5 kDa, respectively (8). Sequences added to the C termini of the two P450 enzymes did not decrease their hydroxylation abilities (data not shown). We also detected the cotransfected bovine adrenodoxin molecule (molecular weight 14 kDa) in the cells, since this also is a critical parameter determining the electron transport of the system and thus hydroxylation efficiency, and we found it to be present in equal amounts (Fig. 5). DISCUSSION In pursuing structure to function relationships in mammalian P450 proteins, alignment to the bacterial P450 cam and recently also to P450 BM3 and P450 terp has proved to be a useful approach. However, a relatively low amino acid sequence homology often hampers accurate alignment of specific residues The amount of cortisol arisen from the reaction was analyzed by radioimmunoassay. Plotted values with standard errors represent the means of four independent transfections for each of the proteins and are normalized against the activity of the P450 11␤ wild type. Mutants are designated at the bottom of the panel. B, the same experimental design as described for panel A was used, except that cells were incubated with 5 M 11-deoxycorticosterone and finally assayed for aldosterone arisen from the conversion. (33). We therefore benefited from the HOMOL data base developed by D. Nelson, which combines amino acid identities and secondary structure predictions. By including in our alignment also the P450 BM3 and P450 terp sequences, which (due to the revealed secondary structures) could be structurally aligned to P450 cam (17,18), we intended to improve the prediction of structural entities in P450 aldo and P450 11␤ . Since for P450 cam it is well established that the I-helix critically participates in substrate binding and because the P450 aldo and P450 11␤ proteins show some diversities in a region corresponding to the P450 cam I-helix, we hypothesized that these differences could be the basis for the different regioselectivities of the two steroid hydroxylases. The I-helix in P450 cam , like in P450 BM3 and P450 terp , runs like a tube through the interior of the molecule (Fig. 6) and in the case of P450 cam , together with the heme binding region, makes up part of the substrate binding pocket (16,34). One critical feature of this conserved helix in P450 proteins is Thr-252 in P450 cam , necessary for the activation of molecular oxygen (28 -30). In P450 cam the substrate camphor is found to be in tight association with the -VGGL-stretch, where the two Gly residues induce a bend in the helix and thus serve as a site where the substrate can fit into position correctly. There is only one Gly residue of the -VGGL-motif conserved in P450 aldo and P450 11␤ (Fig. 1) and it is questionable whether this is sufficient for bending the helix likewise.
A possible relevance of the putative I-helix region for the regioselectivity is also supported by the observations of Pascoe and colleagues (15). By studying artificially engineered hybrid proteins with variable N-terminal P450 11␤ and C-terminal P450 aldo portions reflecting an in vitro model of glucocorticoidremediable aldosteronism, a genetic disorder, His-256, was defined as a critical breakpoint, and the sequence C-terminal of it was identified as essential for aldosterone synthesis. In performing site-directed mutagenesis on the P450 aldo protein and assessing 11␤-hydroxylase activities in transfected COS-1 cells, we found that single amino acid replacements by P450 11␤specific residues at positions 296 and 335 only slightly increased 11␤-hydroxylase actvity and that there was up to a 2-fold increase detected when position 301, 302, or 320 harbored P450 11␤ residues. However, double substitution L301P/ E302D already conferred 60%, and triple replacement mutant L301P/E302D/A320V mounted up to an activity being about 85%, that of the P450 11␤ wild type protein, given that under these experimental conditions the P450 aldo wild type enzyme exhibited only 5% of the activity of the P450 11␤ wild type (Fig.  4A). These data indicate a synergistic rather than a mere additive effect contributed by these three residues. A nonconservative change from Leu to Pro at position 301 alone had no substantial impact, although it was expected to exert other than size effects, since Pro in many cases distorts if not destroys helix continuity. In contrast, the overall effect on cortisol production could be drastically enhanced by two additional conservative substitutions at positions 302 and 320, suggesting that the size of the side chains at these positions is critical for 11␤-hydroxylase activity.
In assaying the aldosterone-synthesizing activities of the mutants, we found mildly to strongly decreased activities. Interestingly, already single substitutions in some mutants markedly reduced aldosterone synthesis. Contrasting 11␤-hydroxylase activity, P450 aldo residues Lys-296 and Asp-335 obviously are important for maximum aldosterone synthase activity since single substitution at these positions to Asn lowered aldosterone synthesis to about 40%. Moreover, double replacement mutant L296N/D335N showed an aldosterone synthase activity decreased to 15%, which was completely abolished by adding the A320V mutation. Among the mutants still producing aldosterone, double replacement mutant L301P/ A320V and triple replacement mutant L301P/E302D/A320V gave rise to the most reduced aldosterone levels (13 and 10% of the P450 aldo wild type), which reciprocally parallels their increase in 11␤-hydroxylase activity. In conclusion, there is no strict correlation between 11␤-hydroxylase increase and decrease in aldosterone synthesis to be seen in this set of mutants. Whereas Pro-301, Asp-302, and Val-320 clearly are major contributors to 11␤-hydroxylation, aldosterone synthesis, apart from the residues at these positions in the P450 aldo sequence also seems to be influenced by Lys-296 and Asp-335. FIG. 5. Immunological detection of heterologously expressed P450 aldo mutants in COS-1 cells. Total proteins were separated on an SDS gel, followed by electroblotting to a nitrocellulose filter. P450 aldo wild type, P450 11␤ wild type, and P450 aldo mutant proteins were detected with a rabbit polyclonal anti-HA-directed antibody (upper panels), and bovine adrenodoxin was detected with a rabbit anti-BAdx antibody (lower panel). A peroxidase-conjugated secondary antibody and ECL chemiluminescence served for visualization of bands. A, detection of single replacement mutants in comparison with P450 aldo wild type and P450 11␤ wild type proteins. B, detection of multiple replacement mutants in comparison with P450 aldo wild type and P450 11␤ wild type proteins.
FIG. 6. Model of the P450 aldo and P450 11␤ structure. Three-dimensional structure of the entire P450 aldo and P450 11␤ structures with the incorporated heme is shown. The view is focused onto the putative I-helix region, and amino acid residues varying between P450 aldo and P450 11␤ are marked in black.
Aldosterone synthesis thus is dependent on a highly evolved structure in the P450 aldo protein and is susceptible to minor changes in this region.
Until now, no CYP11B2 defects have been linked to this region in patients suffering from hypoaldosteronism, but it is conceivable that their occurrence would deteriorate mineralocorticoid synthesis. Whether the positions targeted by sitedirected mutagenesis directly contact the substrate or decisively alter the position or structure of the I-helix remains an open topic.
The I-helix also was postulated in modelling studies of P450 arom by Graham-Lorence and co-workers (35) and P450 17␣ by Laughton and co-workers (36) to form part of the substrate binding pocket for the substrates androstenedion and pregnenolone or progesterone, respectively. Since these are also steroid-modifying enzymes, one is tempted to draw a general significance for this structure in steroid hydroxylation.
In contrast, for P450 proteins that belong to the CYP2 family and participate in liver microsomal steroid hydroxylation, it has been shown that also the N-terminal part of these proteins contributes to the substrate specificities and regioselectivities of hydroxylations (reviewed in Refs. 37 and 38). Lindberg and Negeshi (39) have shown that only one substitution, F209L, was sufficient to confer steroid 15␣-hydroxylase activity from P450 2A4 to P450 2A5, a coumarin hydroxylase. Recently, Halpert and He (40) were able to shift P450 2B1 androgen hydroxylation between the 15␣and 16␤-positions by manipulating two positions, namely Ile-114 and Gly-478, of the enzyme. Moreover, progesterone 21-hydroxylation of P450 2C4 and P450 2C5 seems to be dependent on a hypervariable Nterminal region (residues 95-123) of the proteins and could be conferred on P450 2C1 (41) by mutating the same region of the otherwise nonactive protein in 21-hydroxylation. The same activity could be generated by the creation of chimeric proteins composed of P450 2C2 and P450 2C1 portions, while the parent proteins do not carry out this reaction (42,43). Whether substrate access and binding in P450 aldo and P450 11␤ is also influenced by N-terminally residing amino acids currently remains elusive. However, with our current findings demonstrating a successful conversion of a mineralocorticoid-synthesizing enzyme to one that primarily has a glucocorticoid-synthesizing potential, the paradigm that small changes in the protein sequence at some critical positions can lead to dramatic changes in the activity and specificity can be extended to members of the P450 CYP11 family.