Kinetic analysis of hydroxylation of saturated fatty acids by recombinant P450foxy produced by an Escherichia coli expression system 7 Tatsuya 7 Kitazume 1 , Akinori Tanaka 1 , Naoki Takaya 1 , Akira Nakamura 1 , Shigeru Matsuyama 1 , Takahisa Suzuki 1 and Hirofumi Shoun 2 1 Institute of Applied Biochemistry, University of Tsukuba, Tsukuba, Ibaraki, Japan; 2 Department of Biotechnology, Graduate School of Agricultural and Life Sciences, The University of Tokyo, Japan Cytochrome P450foxy (P450foxy, CYP505) is a fused pro- tein of cytochrome P450 (P450) and its reductase isolated from the fungus Fusarium oxysporum, which catalyzes the subterminal (x-1x-3) hydroxylation of fatty acids. Here, we produced, purified and characterized a fused recombin- ant protein (rP450foxy) using the Escherichia coli expression system. Purified rP450foxy was catalytically and spectrally indistinguishable from the native protein, but most of the rP450foxy was recovered in the soluble fraction of E. coli cells unlike the membrane-bound native protein. The results are consistent with our notion that the native protein is targeted to the membrane by a post-translational modifica- tion mechanism. We also discovered that P450foxy could use shorter saturated fatty acid chains 1 (C9 and C10) as a sub- strate. The regiospecificity (x-1x-3) of hydroxylation due to the enzymatic reaction for the short substrates (decanoate, C10; undecanoate, C11) was the same as that for longer substrates. Steady state kinetic studies showed that the k cat values for all substrates tested (C9-C16) were of the same magnitude (1200–1800 min )1 ), whereas the catalytic effi- ciency (k cat /K m ) was higher for longer fatty acids. Substrate inhibition was observed with fatty acid substrates longer than C13, and the degree of inhibition increased with increasing chain length. This substrate inhibition was not apparent with P450BM3, a bacterial counterpart of P450foxy, which was the first obvious difference in their catalytic properties to be identified. Kinetic data were consistent with the inhibition due to binding of the second substrate. We discuss the inhibition mechanism based on differences between P450foxy and P450BM3 in key amino acid residues for substrate binding. Keywords: fatty acid hydroxylase; cytochrome P450; P450foxy; dodecanoic acid; Fusarium oxysporum. Cytochrome P450 (P450) is a group of heme proteins that are widespread in nature [1–3]. It is generally accepted that all P450s originated from the same, ancient gene (P450 superfamily), which has acquired unparalleled molecular and functional diversity during evolution [1,4]. Most of the P450 enzymes function as monooxygenases that act on various lipophilic compounds, whereas others catalyze a variety of reactions [2]. P450s can be classified into several classes according to their redox partners [5,6]. Bacterial and mitochondrial P450 systems are of class I; they receive electrons from NAD(P)H via ferredoxin reductase and ferredoxin coupling. Eukaryotic microsomal P450 systems are of class II; they receive electrons from P450 reductase, which contains both FAD and FMN. These two classes comprise typical, multicomponent P450 monooxygenase systems, whereas the functions of P450 are most diversified in other classes. Class III P450s are not monooxygenases but catalyze isomerization [7] or dehydration [8], and require neither external redox equivalents nor any redox partners. P450nor is the only class IV P450, and catalyzes the reduction of nitric oxide (NO) to nitrous oxide (N 2 O) using NAD(P)H as the direct electron donor [9]. The class III and IV P450s are self-sufficient, meaning that they can complete their functions without other proteinaceous components. This laboratory has isolated two unique P450s from the fungus Fusarium oxysporum. One is P450nor (CYP55) as described above [9–12], and the other is P450foxy (CYP505) [13,14], both of which are self-sufficient. P450foxy is of the class II type, but its self-sufficiency depends on fusion of the P450 and the reductase domains on one gene and is thus produced as a single polypeptide [14]. It catalyzes the subterminal (x-1x-3) hydroxylation of fatty acids [13,15]. P450foxy closely resembles in some aspects, P450BM3 (CYP102) from Bacillus megaterium. The identity of the predicted amino-acid sequence of P450foxy is closest to that of P450BM3 in both the P450 and reductase domains. We therefore concluded that P450foxy is the eukaryotic coun- terpart of P450BM3. This conclusion raises the evolutio- nary question of why phylogenetically distant organisms, such as eukaryotes (F. oxysporum) and prokaryotes (B. megaterium), share such closely related P450s. The only obvious difference between P450foxy and P450BM3 iden- tified so far is that of their intracellular localization. P450foxy is exclusively recovered in the membrane fraction Correspondence to H. Shoun, Department of Biotechnology, Graduate School of Agricultural and Life Sciences, The University of Tokyo, Bunkyo-ku, Tokyo 113-8657, Japan. Fax: + 81 35841 5148, Tel.: + 81 35841 5148, E-mail: ahshoun@mail.ecc.u-tokyo.ac.jp Abbreviations: GC, gas chromatography; GC-EIMS, gas chromato- graphy-electron impact mass chromatography; P450, cytochrome P450, rP450foxy, recombinant P450foxy. (Received 19 November 2001, revised 22 February 2002, accepted 25 February 2002) Eur. J. Biochem. 269, 2075–2082 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.02855.x like other eukaryotic P450s [14], whereas P450BM3 is a soluble protein like other bacterial P450s. Several P450 functions are of interest with respect to potential industrial applications [16–18] and for basic studies. The most inconvenient properties of the P450 system when considering industrial applications would be the complexity of its electron transport. In vitro P450 function can only be exhibited in a reconstituted mixture of many components, yet the activity is usually very low under such conditions. One way to overcome this obstacle would be to use a fused protein consisting of P450 and its reductase. Okawa et al. originally constructed an artificial fused protein using the yeast expression system and applied the system to the bioconversion of fine chemicals [19]. Thereafter, P450BM3 [20] and P450foxy [13,14] were identified as naturally fused proteins. The catalytic turnover of both P450s is exception- ally high, possibly because they are catalytically self- sufficient due to the fusion of two domains. The naturally fused protein would be more useful for such applications and attempts have made using P450BM3 [21,22]. We have produced recombinant P450foxy (rP450foxy) in the host-vector system of Saccharomyces cerevisiae [14]. However, a more efficient production system is required for advancing both basic and application studies of P450foxy. Here, we describe the expression of P450foxy cDNA in Escherichia coli, which resulted in the large-scale production of rP450foxy. We also characterized the substrate specificity and other catalytic properties. MATERIALS AND METHODS Strain, culture and media Plasmids were constructed and rP450foxy was produced using E. coli strains JM109 and DH5a, respectively. E. coli strains were cultivated in Luria–Bertani broth (1% tryp- tone, 0.5% yeast extract, 0.5% NaCl) and Terrific Broth (1.2% tryptone, 2.4% yeast extract, 0.4% glycerol, 0.23% KH 2 PO 4 , 0.125% K 2 HPO 4 ) containing 50 lgÆmL )1 amp- icillin (LBA and TBA, respectively). Construction of the expression plasmid Plasmid pCWfoxy was constructed to produce recombinant P450foxy in E. coli as follows. The cDNA of P450foxy [14] was prepared by PCR using the respective 5¢ and 3¢ PCR oligonucleotide primers: 5¢-CATATGGCTGAATCTGT TCCGATTCCGGAAC C GCCGGGTTATCCGCTT-3¢ and 5¢-TGTTTGCTTG ATCTCCAAAGCGTAGTT-3¢ (mutated residues are underlined), which have homology to the nucleotide sequences of the 5¢ and the 3¢ ends of the P450foxy cDNA, respectively [14]. PCR products were purified, digested by NdeIandBamHI, then ligated to the plasmid vector pCWori+ [23] that had been digested with the same restriction enzymes. The resulting plasmid was designated pCWfoxy. Standard DNA techniques proceeded according to Sambrook et al. [24]. Preparation of recombinant P450foxy E. coli DH5a was transformed with pCWfoxy, cultured in LBA overnight, transferred to 2 L of TBA in a 5-L volume flask, and rotated at 120 r.p.m. 2 at 30 °C. When D 600 ¼ 0.5, 1 m M isopropyl thio-b- D -galactoside, 0.5 m M 5-aminolevulinic acid and 1 lgÆmL )1 chloram- phenicol (final concentration) were added to the medium and the flask was further incubated for 48 h under the same conditions. The cells were then harvested by centrifugation, suspended in 50 mL of buffer A (50 m M Mops/KOH, 10% glycerol, 1.0 m M dithiothreitol, 0.1 m M EDTA, pH 7.4) and disrupted using a French Pressure Cell Press (Sim-Aminco, New York, USA) 3 at 20 000 psi. The homogenate was centrifuged at 1800 g for 15 min to remove cellular debris and unbroken cells. The resulting cell free extract was centrifuged at 100 000 g for 60 min to separate the supernatant (soluble) and pellet (membrane) fractions. The soluble fraction was applied to a DEAE- cellulose column (Whatman DE52) equilibrated with buffer A. The column was washed, then proteins were eluted with a 0–0.3 M KCl gradient in buffer A. The fraction containing heme was collected, dialyzed against buffer A, and applied to a 2¢-,5¢-ADP Sepharose column (Amersham Pharmacia Biotech) equilibrated with the same buffer. A dark brown fraction that was eluted with 3m M NADPH in the same buffer was directly passed through a Superdex 200 HR 10/30 column (Amersham Pharmacia Biotech) equilibrated with buffer A. The sam- ple after this elution step was used as purified rP450foxy. Spectroscopy Optical and fluorescence spectra were measured using a Beckman DU 7500 spectrophotometer and a Hitachi F-3010 fluorescence spectrophotometer, respectively. Heme was identified and determined by the pyridine ferrohemo- chromogen method using the molar absorption coefficient (e) of the chromogen of the protoheme as 34.4 m M )1 Æcm )1 at 557 nm [25]. FAD and FMN were identified and quantified as described by Faeder & Siegel [26] using e at 450 nm ¼ 11.5 m M )1 Æcm )1 . The P450 content was deter- mined using an extinction coefficient of 91 m M )1 Æcm )1 for the difference in the carbon monoxide (CO) difference spectrum between 450 nm and 490 nm [13]. The ratio in the high/low spin states in the bound heme of P450 was calculated as reported previously [27]. Enzyme assays Fatty acid hydroxylase was assayed as described previously [13]. The reaction mixture contained 50 m M Mes (pH 6.5), 1 l M FAD, 1 l M FMN, 10% glycerol, 125 l M NADPH, and 125 l M fatty acid (final concentration). The reaction was initiated by adding rP450foxy, then the A 340 was followed at 30 °C using a Beckman DU-7500 spectropho- tometer. NADPH-cytochrome c reductase was assayed as described previously [13] in the same buffer containing 125 l M NADPH and 50 l M horse heart cytochrome c (Sigma). Protein conxentration was determined using the Bio-Rad protein assay reagent (Bio-Rad Laboratories Inc., CA, USA). Data analysis Steady state kinetic analyses for P450foxy were examined with varying concentrations of fatty acid and a saturating 2076 T. Kitazume et al. (Eur. J. Biochem. 269) Ó FEBS 2002 concentration of the second substrate (NADPH, 125 l M ). Substrate inhibition was observed for longer fatty acid substrates. Assuming that the inhibition depended on the second binding of fatty acid substrate at its higher concen- trations, the data were fitted to Eqn (1) using ORIGIN Software, in which v, e, and [S] represent the experimentally determined initial velocity, and the enzyme and the substrate (fatty acid) concentrations, respectively. K s represents the dissociation constant for the second binding of S. v=e ¼ k cat ½S=fK m þ½Sð1 þ½S=K S Þg ð1Þ Determination of the reaction products The structures of the reaction products (hydroxyl fatty acids) of rP450foxy were determined fundamentally as described previously [15] by gas chromatography (GC) and GC-electron impact-mass spectrometry (GC-EIMS). Prior to analysis, the carboxyl group of the products was blocked by methylation with diazomethane and the hydroxyl group was trimethyl-silylated (TMSlated) with TMSI-H (GL Science Ltd, Tokyo, Japan) 4 . RESULTS Production of recombinant P450foxy A cDNA of P450foxy was expressed in E. coli DH5a cells under the control of the lac promoter resident in the pCWori+ vector. To improve production, the nucleotide sequence on the 5¢ end was mutated without changing the encoding amino-acid sequence so as to become more A/T rich, with the exception of proline codons that were mutated to CCG, the most frequently used proline codon in E. coli genes. The expression and production of the recombinant protein was confirmed by the presence of the characteristic chromophore in the cell free extract of the transformants, which gave a specific content of 0.021 nmol P450 Æ(mg protein) )1 and a yield of 10 mg P450Æ(g wet cells) )1 .This yield was much higher than that attained using the yeast system [0.2 mg P450Æ(g wet cells) )1 ] [14]. Most P450 (>88%) was recovered in the soluble fraction like the recombinant protein produced by the yeast system [14] but in contrast to native P450foxy, which is cofractionated with themembranefractionofF. oxysporum [13]. The produc- tion of rP450foxy in the E. coli system was further confirmed by Western blotting that gave a specific signal with the predicted M r of 118 000 (data not shown) and dodecanoic acid-dependent NADPH oxidase activity. None of these properties characteristic of P450foxy were detected in the extract of E. coli cells that harbored only the vector. The fraction containing P450 was purified to homogeneity from the soluble fraction at a yield of 26%. The M r estimated by SDS/PAGE and gel filtration 5 was 118 000 and 132 000, respectively, indicating that rP450foxy exists as a monomer-like native protein and rP450foxy produced by yeast [14]. Spectral properties The absorption spectra of purified rP450foxy (Fig. 1) in its resting oxidized, dithionite-reduced ferrous, and car- bon monoxide (CO)-ligated forms are identical to the corresponding spectra of native P450foxy purified from F. oxysporum [13]. The CO-difference spectrum with a peak at 448 nm and a trough at 407 nm (Fig. 1, inset) was also identical. The calculated heme content was 0.4 mol of protoheme per mol of protein. In contrast to these spectral characteristics due to the bound heme, the absorbance around 450 nm (characteristic of flavin) was not prominent, with resting rP450foxy similar to the native protein, possibly because of a low flavin content. However, the presence of flavin in rP450foxy was confirmed by the characteristic fluorescence that accom- panied the emission (excited at 450 nm) peak at 528 nm. The calculated specific contents of FAD and FMN were 0.15 and 0.65 molÆ(mol protein) )1 , respectively. Native P450foxy also has a low content of these cofactors (heme and flavins) [13]. These results indicate that rP450foxy is correctly folded in heterologous bacterial cells. Catalytic activities Because of the low flavin content, the activity of purified rP450foxy was low. A prior incubation with free FAD and FMN remarkably accelerated the specific activity of rP450foxy as observed with the native enzyme [28]. Dodec- anoic acid hydroxylase and NADPH-cytochrome c reduc- tase were assayed for rP450foxy and compared with the findings obtained using the native [13] and recombinant protein produced in the yeast host [14] (Table 1). We found that the NADPH-cytochrome c reductase activities of P450foxy are enhanced in the presence of the substrate (fatty acid) to be hydroxylated [13]. We also found the same phenomenon with rP450foxy (dodecanoic acid) (Table 1). The specific activities of rP450foxy with respect to fatty acid hydroxylase and cytochrome c reductase were similar to those for the recombinant protein produced in the yeast or the native fungal protein [13,14]. These results demonstrated that rP450foxy produced by E. coli is kinetically and spectrally (above results) indistinguishable from the native protein. Fig. 1. Absorption spectra of rP450foxy. Solid line, resting (oxidized); dotted line, dithionite-reduced; dashed line, dithionite-reduced + CO. Inset, CO-difference spectrum (CO-bound minus dithionite-reduced); 7.2 l M purified rP450foxy in 100 m M sodium phosphate buffer (pH 7.3), 10% glycerol, at room temperature. Ó FEBS 2002 Fungal P450foxy catalyzing fatty acid hydroxylation (Eur. J. Biochem. 269) 2077 We determined the substrate specificity of rP450foxy against saturated fatty acids. Figure 2 shows that rP450foxy was active against fatty acids with a chain length of C9 (nine carbon atoms, nonanoic acid) to C18 (18 carbon atoms, octadecanoic acid), with the activity on tridecanoic acid (C13) being the highest. These results are similar to those obtained using the native protein [13], but rP450foxy can also efficiently use the shorter substrate, nonanoic acid (C9). Stoichiometry between the consumption of NADPH and O 2 was 1.3 : 1 with all of the enzymatic reactions on substrates with chains of C10 to C15 in length (data not shown), consistent with the theoretical value for 2 electron reduction coupling to monooxygenation of the substrates. Interaction of the resting rP450foxy with the substrate fatty acids Spectral changes were observed upon mixing the resting rP450foxy with saturated fatty acids with chain length from C8 to C18. They were typical of type I spectral change that is generally observed upon binding to P450 of the substrate to be hydroxylated [29]. That is, the ratio in the high-spin state heme with a Soret peak at around 388 nm increased and the low-spin state heme with a peak at around 418 nm decreased (Fig. 3). The extent of the fatty-acid-chain length that could afford the spectral change (C8-C18) almost completely agreed with the substrate specificity determined with respect to the overall activity above (C9-C18). The K d value of the enzyme–substrate complex (Table 2) and the maximal absorbance change can be obtained by the spectrophotometric titration using this spectral change. As observed in Fig. 3, addition of the substrate did not cause a complete exchange from low to high spin states. The ratio of high-spin state heme was obtained for each rP450foxy– fatty-acid complex from the maximal absorbance change (Table 2) [27]. The ratio of high spin was higher for the complexwithalongerfattyaciduptoC15.Wecouldnot determine this ratio with the longer fatty acids (C16-C18) because of the extremely low water-solubility of these fatty acids. The spectral change with octanoic acid linearly increaseduptoasubstrateconcentrationof3m M ,whichis the determination limit defined by the solubility of the fatty acid, indicating that the K d for octanoate would be over 3m M . These results showed that the heme in rP450foxy is in equilibrium between high and low spin states when a fatty acid substrate binds and that the chain length of the substrate affects the equilibrium. Steady-state kinetics Apparent K m and k cat values for fatty acid substrates were determined (Table 2). The K m value was in a similar range (8–36 l M ) between the substrates from C12 to C16, and each value approximately agreed with the respective K d value for the same substrate. In contrast, the K m value increased with decreasing chain length of the substrate Table 1. Catalytic activities of native and recombinant P450foxy. Data are mean values of three experiments. Enzyme Activity (nmol NADPHÆmin )1 Ænmol P450 )1 ) Dodecanoic acid hydroxylase Cytochrome c reductase Cytochrome c reductase (with 150 l M dodecanoic acid) rP450foxy (E. coli) 1460 ± 100 1400 ± 190 3300 ± 710 rP450foxy (Yeast) a 1210 ± 110 890 ± 50 1590 ± 140 P450foxy (Fusarium) b 1200 900 2,000 a Kitazume et al. [14]. b Nakayama et al. [13]. Fig. 2. Apparent substrate specificity of rP450foxy for saturated fatty acids. Enzymatic activity was assayed at a fixed concentration (125 l M ) of fatty acid (C9-C18). C means control reaction without fatty acid. Data are mean values of three experiments. Fig. 3. Spectral perturbation in rP450foxy caused by pentadecanoic acid. (A) Absorption spectra of rP450foxy [6 l M in 50 m M Mes (pH 6.5), 1 l M FAD, 1 l M FMN, 10% glycerol, at 30 °C] in the presence of 0, 2.5, 5, 7.5, 10, 15, 20, 30, 40, 50 l M pentadecanoic acid, respectively (lines 1–10). (B) Difference spectra. Each difference spec- trum was obtained by subtracting line 1 from each of lines 2–10 in A. 2078 T. Kitazume et al. (Eur. J. Biochem. 269) Ó FEBS 2002 below C12 (dodecanoic acid), and became larger than the K d value for the same substrate. The kinetic constants for fatty acids over C16 could not be obtained because of the substrate inhibition. The k cat value was of the same magnitude for all substrates tested (1200–1800 min )1 ). Substrate inhibition occurred for fatty acids with a chain length of C13 or longer (Fig. 4). The inhibition was apparent at higher substrate concentration and was eluci- dated by the mechanism described in Eqn (1), in which a second substrate binds to form an abortive enzyme- substrate complex. Such inhibition was not evident when the chain length of the fatty acids was shorter than C13 (C9 to C12). The data fit closely with the respective curves calculated on the basis of Eqn (1) with the K s for tetradecanoic, pentadecanoic and hexadecanoic acids being 1100, 770, and 70 l M , respectively (Table 2). These results indicate that the substrate inhibition is stronger for longer fatty acid substrates. The substrate inhibition can elucidate well the apparent discrepancy that the apparent activity decreased for substrates longer than C13 (Fig. 2) although the catalytic efficiency (k cat /K m ) paradoxically increased (Table 2). Such substrate inhibition with P450BM3 has not been reported. We also conducted assays to determine the kinetic constants for the electron donors, NADPH and NADH, using a saturated concentration (150 l M ) of dodecanoic acid as the substrate. The K m for NADPH was too low to be determined, and may have been in the order of 10 )6 M or lower. The K m for NADH was 74 l M . These results are same as those obtained with the native enzyme [13]. Determination of the reaction products We identified the metabolites (reaction products) of fatty acids due to the enzymatic reaction of rP450foxy by GC- EIMS. The results using dodecanoic acid (C12) are shown in Fig. 5. The derivatives of products were separated into three peaks on GC (Fig. 5A), each of which gave a fragment pattern typical of TMSlated alcohol on electron impact mass chromatography (EIMS) 6 (Fig. 5B–D). The mass number of each fragment identified these metabolites as x-1, x-2 and x-3 hydroxy derivatives of dodecanoic acid, respectively. The ratio of these metabolites was estimated from the signal intensity on GC (Fig. 6). These results using a C12 fatty acid are similar to those we obtained using cell-free extracts of F. oxysporum [15]. However, the present study is the first to identify the reaction products of purified P450foxy. We also confirmed that the time-dependent decrease of the substrate during the enzymatic reaction approximately agreed with the accompanying increase in the sum of the products (data not shown). The same sets of experiments were replicated for, decanoic and undecanoic acids as substrates, and the ratio of the three products is shown in Fig. 6. DISCUSSION We produced rP450foxy using an E. coli expression system in a yield that was 50-fold higher than that obtained using the yeast system [14]. The recombinant protein was produced as a soluble protein unlike native fungal P450foxy that is membrane-bound. However, the known catalytic and spectral properties of rP450foxy are indistinguishable from those of the native protein, supporting our previous conclusion that native P450foxy is targeted to the mem- brane by post (or co) translational modification in fungal cells. We also showed that the enzymatic reaction of P450foxy exhibits a similar regiospecificity (x-1x-3)ofthe reaction products for fatty acids with shorter chains (C10, C11) to that for longer fatty acids [15]. We generated far more purified P450foxy using this system than the original fungal cells [13] or the yeast host-vector system [14]. Thus, this unique P450 can be more extensively studied. Table 2. Kinetic constants of rP450foxy for various saturated fatty acids. Data are mean values for more than five experiments. Standard errors are below 20%. ND, not determined. Substrate a K d (l M ) K m (l M ) K s (l M ) k cat (min )1 ) k cat /K m (min )1 Æl M )1 ) Spin state high spin(%) Nonanoic acid 170.0 3200 > 2000 1500 0.5 25 Decanoic acid 8.7 260 > 2000 1200 4.6 23 Undecanoic acid 14.0 160 > 2000 1900 11.5 30 Dodecanoic acid 9.4 30 > 2000 1500 49.1 33 Tridecanoic acid 9.0 36 > 2000 1800 62.7 46 Tetradecanoic acid 2.8 19 1100 1300 68.4 43 Pentadecanoic acid 8.4 8 770 1300 163 74 Hexadecanoic acid ND 10 70 1800 180 ND a Kinetic constants could not be determined using octanoic, heptadecanoic and octadecanoic acids as substrates. Fig. 4. Steady state kinetics of long-chain fatty acids. Substrates are tridecanoic acid (j), tetradecanoic acid (d), pentadecanoic acid (s), and hexadecanoic acid (h). Ó FEBS 2002 Fungal P450foxy catalyzing fatty acid hydroxylation (Eur. J. Biochem. 269) 2079 Analyses by steady state kinetics revealed a novel feature of P450foxy with respect to substrate specificity. We examined substrates shorter than C10 (C8, C9), and found that the C9 fatty acid can be a significant substrate although the K m value was very large (Table 2). The present and previous results [13] showed that P450foxy has activity for C9-C18 fatty acids, although fatty acids longer than C18 have not been examined. Regardless, the substrate specif- icity of P450foxy is distributed among saturated fatty acids with a rather shorter chain length than those for P450BM3 (C12-C20). The apparent activity at a fixed substrate concentration was highest against a C13 fatty acid (Fig. 2) whereas the catalytic efficiency (k cat /K m ) was higher against longer substrates (Table 2). This discrepancy can be explained by the inhibition that increased with longer substrates. Both K m and K d values were similar for longer fatty acids whereas K m became much larger than K d for fatty acids shorter than C12. As K m ¼ (k cat + k off )/k on and K d ¼ k off /k on ,wherek on and k off are, respectively, the rate constants for association and dissociation of the substrate, and k cat is almost constant for all substrates tested (Table 2), the results mean that both k on and k off values became much smaller when the chain length of the fatty acid decreased. In other words, rapid equilibrium cannot be assumed for the enzymatic reaction with short chain substrates. Here, we discovered a prominent difference in the catalytic properties of P450foxy and P450BM3, namely, substrate inhibition with P450foxy. This could be explained by the binding of the second substrate (Eqn 1). At present, the structural basis of the substrate-binding site of P450foxy is not known, but alignment of the amino-acid sequences between P450foxy and P450BM3 implies similarities with the interaction of P450foxy with fatty acid substrates. Crystallographic and mutational studies have shown that several amino-acid residues are critical for the binding of substrate to P450BM3. Arg47 and Tyr51 are located at the entrance of the substrate-accessing channel, and their guanidinium and hydroxyl groups, respectively, play crucial roles in the binding of the substrate by interacting with the carboxyl group of the substrate fatty acids [30–32]. Phe42 is located close to the entrance and its aromatic residue may be a lid that excludes solvent water to strengthen the electro- static interaction of Arg47 and the substrates [30–32]. The corresponding amino-acid residues to those in P450foxy are Leu43, Lys48, and Phe52, respectively [14]. Inside the entrance, the hydrophobic residues, Leu75, Phe87, Leu181, Ile263, and Leu437, form a hydrophobic stretch in P450BM3 that allow access to the aliphatic chains of fatty acids. All of these residues are conserved in P450foxy. These alignments indicate that all of the key amino-acid residues at the entrance of the active site pocket of P450BM3 (Phe42, Arg47, and Tyr51) are replaced by others (Leu43, Lys48, and Phe52) in P450foxy although all of other key residues inside the pocket are conserved. The positive charge essential for fixing the carboxylate of fatty acids is maintained by replacing Arg with Lys. However, hydrophobicity at the entrance to P450foxy should be significantly increased as the result of the substitutions from Phe to Leu and from Tyr to Phe. Why substrate inhibition is observed only with P450foxy may be explained by these substitutions. The first interaction of P450BM3 with fatty acids may occur between the carboxylate anion of the Fig. 5. Gas chromatographic separation (A) and EIMS-spectra of TMSlated and methyla- ted derivatives of the reaction products (B–D) from dodecanoic acid. Mass spectra of deriva- tives 1–3 (A) are shown in B–D, respectively. Relative abundance of ions in panel D was expanded by threefold for fragments with high m/z-values (indicated by horizontal arrow). Fig. 6. Regio-specificity of reaction products determined with the substrates, decanoic (C10), undecanoic (C11), and dodecanoic (C12) acids. Relative amount of each product was determined by the extent to which corresponding peaks separated on GC. Closed, open, and striped bars represent x-1, x-2, and x-3 hydroxy fatty acids, respectively. 2080 T. Kitazume et al. (Eur. J. Biochem. 269) Ó FEBS 2002 substrates and Arg47 and Tyr51 residues of the protein [30– 32]. The aliphatic head of fatty acids would then turn and penetrate the access channel. The more hydrophobic environment around the entrance of P450foxy would permit another fatty acid molecule to partially penetrate the channel from its aliphatic head even after the first molecule has already occupied the channel. The stronger substrate inhibition by longer fatty acids suggests a larger contribu- tion of the aliphatic chain of fatty acids to their binding, as compared with P450BM3. In other words, the interaction of P450foxy with the carboxylate anion of fatty acids at the entrance is relatively weaker than that of P450BM3. This notion is supported by the fact that Phe52 in P450foxy replaces Tyr51 in P450BM3, which makes it impossible to have the hydrogen bonding that supports the electrostatic interaction of the positive charge (Lys48 in case of P450foxy) with the carboxylate. The effectiveness of shorter fatty acids as substrates for P450foxy also agrees with this replacement, as the interaction at the entrance must be more flexible for the aliphatic head with a shorter chain to reach the active site near heme. The fact that such important amino-acid residues are not conserved between P450foxy and P450BM3 is notable. These results demonstrate that P450foxy and P450BM3 would be an interesting basis for protein engineering studies from both basic and application aspects. ACKNOWLEDGEMENTS This study was supported by PROBRAIN (Program for Promotion of Basic Research Activities for Innovative Biosciences), SBPB (Structural Biology Sakabe Project) of FAIS (Foundation for Advancement of International Science), and Grant-in-Aid for Scientific Research from Ministry of Education, Science, Culture and Sports of Japan. REFERENCES 1. Omura, T. (1999) Forty years of cytochrome P450. Biochem. Biophys. Res. Commun. 266, 690–698. 2. Sono, M., Roach, M.P., Coulter, E.D. & Dawson, J.H. (1996) Heme-containing oxygenases. Chem. Rev. 96, 2841–2887. 3. Nelson, D.R., Koymans, L., Kamataki, T., Stegeman, J.J., Feyereisen, R., Waxman, D.J., Waterman, M.R., Gotoh, O., Coon, M.J., Estabrook, R.W. et al. (1996) P450 superfamily: update on new sequences, gene mapping, accession numbers and nomenclature. Pharmacogenetics 6, 1–42. 4. Nelson, D.R. (1999) Cytochrome P450 and the individuality of species. Arch. Biochem. Biophys. 369, 1–10. 5. Boddupalli, S.S., Hasemann, A.C., Ravichandran, G.K., Lu, Y.J., Goldsmith, J.E., Deisenhofer, J. & Peterson, A.J. (1992) Crystal- lization and preliminary X-ray diffraction analysis of P450terp and the hemoprotein domain of P450BM-3, enzymes belonging to two distinct classes of the cytochrome P450 superfamily. Proc. Natl Acad.Sci.USA89, 5567–5571. 6. Shimizu, H., Park, Y.S., Lee, S.D., Shoun, H. & Shiro, Y. (2000) Crystal structures of cytochrome P450nor and its mutants (Ser286 fi Val, Thr) in the ferric resting state at cryogenic tem- perature: a comparative analysis with monooxygenase cyto- chrome P450s. J. Inorgan. Biochem. 81, 191–205. 7. Haurand, M. & Ullrich, V. (1985) Isolation and characterization of thromboxane synthase from human platelets as a cytochrome P-450 enzyme. J. Biol. Chem. 260, 15059–15067. 8. Song, W.C. & Brash, A.R. (1991) Purification of an allene oxide synthase and identification of the enzyme as a cytochrome P-450. Science 253, 781–784. 9. Nakahara, K., Tanimoto, T., Hatano, K., Usuda, K. & Shoun, H. (1993) Cytochrome P-450 55A1 (P-450dNIR) acts as nitric oxide reductase employing NADH as the direct electron donor. J. Biol. Chem. 268, 8350–8355. 10. Kizawa, H., Tomura, D., Oda, M., Fukamizu, A., Hoshino, T., Gotoh, O., Yasui, T. & Shoun, H. (1991) Nucleotide sequence of the unique nitrate/nitrite-inducible cytochrome P-450 cDNA from Fusarium oxysporum. J. Biol. Chem. 266, 10632–10637. 11. Park, S.Y., Shimizu, H., Adachi, S., Nakagawa, A., Tanaka, I., Nakahara, K. & Shoun, H. (1997) Crystal structure of nitric oxide reductase from denitrifying fungus Fusarium oxysporum. Nat. Struct. Biol. 10, 827–832. 12. Kudo, T., Tomura, D., Liu, D., Dai, X. & Shoun, H. (1996) Two isozymes of P450nor of Cylindrocarpon tonkinense:molecular cloning of the cDNAs and genes, expressions in the yeast, and the putative NAD(P) H-binding site. Biochimie 78, 792–799. 13. Nakayama, N., Takemae, A. & Shoun, H. (1996) Cytochrome P450foxy, a catalytically self-sufficient fatty acid hydroxylase of the fungus Fusarium oxysporum. J. Biochem. 119, 435–440. 14. Kitazume, T., Takaya, N., Nakayama, N. & Shoun, H. (2000) Fusarium oxysporum fatty-acid subterminal hydroxylase (CYP505) is a membrane-bound eukaryotic counterpart of Bacillus megaterium cytochrome P450BM3. J. Biol. Chem. 275, 39734–39740. 15. Shoun, H., Sudo, Y. & Beppu, T. (1985) Subterminal hydro- xylation of fatty acids by a cytochrome P-450-dependent enzyme system from a fungus, Fusarium oxysporum. J. Biochem. 97, 755– 763. 16. Tsujita,Y.,Ikeda,T.,Serizawa,N.,Hishimura,A.&Komai,T. (1997) Pravastatin sodium: new clinical aspects and recent pro- gress. Ann. Report Sankyo Res. Lab. 49, 1–61. 17. Shiota, N., Nagasawa, A., Sakaki, T., Yabusaki, Y. & Ohkawa, H. (1994) Herbicide-resistant tobacco plants expressing the fused enzyme between rat cytochrome P4501A1 (CYP1A1) and yeast NADPH-cytochrome P450 oxidoreductase. Plant Physiol. 106, 17–23. 18. Carmichael, A.B. & Wong, L.L. (2001) Protein engineering of Bacillus megaterium CYP102. The oxidation of polycyclic aro- matic hydrocarbons. Eur. J. Biochem. 268, 3117–3125. 19. Murakami, H., Yabusaki, Y., Sakaki, T., Shibata, M. & Ohkawa, H. (1987) A genetically engineered P450 monooxygenase: con- struction of the functional fused enzyme between rat cytochrome P450c and NADPH-cytochrome P450 reductase. DNA 6, 189– 197. 20. Narhi, L.O. & Fulco, A.J. (1986) Characterization of a catalyti- cally self-sufficient 119,000-dalton cytochrome P-450 mono- oxygenase induced by barbiturates in Bacillus megaterium. J. Biol. Chem. 261, 7160–7169. 21. Schneider, S., Wubbolts, M.G., Sanglard, D. & Witholt, B. (1998) Biocatalyst engineering by assembly of fatty acid transport and oxidation activities for In vivo application of cytochrome P-450BM-3 monooxygenase. Appl. Environ. Microbiol. 64, 3784– 3790. 22. Schneider, S., Wubbolts, M.G., Oesterhelt, G., Sanglard, D. & Witholt, B. (1999) Controlled regioselectivity of fatty acid oxida- tion by whole cells producing cytochrome P450BM-3 mono- oxygenase under varied dissolved oxygen concentrations. Biotechnol. Bioeng. 64, 333–341. 23. Muchmore, D.C., McIntosh, L.P., Russell, C.B., Anderson, D.E. & Dahlquist, F.W. (1989) Expression and nitrogen-15 labeling of proteins for proton and nitrogen-15 nuclear magnetic resonance. Methods Enzymol. 177, 44–73. 24. Sambrook, J., Fritsch, E.F. & Maniatis, T. (1989) Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory, Cold Spring Harbor Press, New York. 25. Falk, J.E. (1964) Porphyrins and Metalloporphyrins.Elsevier, Amsterdam. Ó FEBS 2002 Fungal P450foxy catalyzing fatty acid hydroxylation (Eur. J. Biochem. 269) 2081 26. Faeder, E.J. & Siegel, L.M. (1973) A rapid micromethod for determination of FMN and FAD in mixtures. Anal. Biochem. 53, 332–336. 27. Okamoto, N., Imai, Y., Shoun, H. & Shiro, Y. (1998) Site-directed mutagenesis of the conserved Threonine (Thr243) of the distal helix of fungal cytochrome P450nor. Biochemistry 37, 8839–8847. 28. Nakayama, N. & Shoun, H. (1994) Fatty acid hydroxylase of the fungus Fusarium oxysporum is possibly a fused protein of cyto- chrome P-450 and its reductase. Biochem. Biophys. Res. Commun. 202, 586–590. 29. Truan, G. & Peterson, A.J. (1998) Thr268 in substrate binding and catalysis in P450BM-3. Arch. Biochem. Biophys. 349, 53–64. 30. Li, H Y. & Poulos, T.L. (1997) The structure of the cytochrome p450BM-3 haem domain complexed with the fatty acid substrate, palmitoleic acid. Nat. Struct. Biol. 4, 140–146. 31. Oliver, C.F., Modi, S., Primrose, W.U., Lian, L Y. & Roberts, G.C.K. (1997) Engineering the substrate specificity of Bacillus megaterium cytochrome P-450 BM3: hydroxylation of alkyl trimethylammonium compounds. Biochem. J. 327, 537–544. 32. Noble, M.A., Miles, C.S., Chapman, S.K., Lysek, D.A., MacKay, A.C.,Reid,G.A.,Hanzlik,R.P.&Munro,A.W.(1999)Rolesof key active-site residues in flavocytochrome P450 BM3. Biochem. J. 339, 371–379. 2082 T. Kitazume et al. (Eur. J. Biochem. 269) Ó FEBS 2002 . Kinetic analysis of hydroxylation of saturated fatty acids by recombinant P450foxy produced by an Escherichia coli expression system 7 Tatsuya 7 Kitazume 1 ,. Fungal P450foxy catalyzing fatty acid hydroxylation (Eur. J. Biochem. 269) 2079 Analyses by steady state kinetics revealed a novel feature of P450foxy with respect