The enzymatic hydroxylation of n-octane by Corynebacterium sp. strain 7E1C.

Abstract Cell-free extracts from sonically disrupted Corynebacterium sp. (7E1C) oxidized n-octane to 1-octanol and octanoic acid in the presence of NADH and O2. The hydroxylating activity, assayed by direct estimation of the reaction products, was found to be concentrated in the clarified S3 suprenatant fraction after centrifugation at 144,000 x g for 2 hours. By use of mass spectrometry it was shown that molecular oxygen is incorporated into the substrate during hydroxylation. The hydroxylating enzyme system was separated into two protein fractions, both of which were required for activity. One fraction, the S3(25–40)D, which precipitated between 25 and 40% ammonium sulfate saturation, appeared "particulate" and contained cytochrome P-450. The participation of this hemoprotein in n-octane hydroxylation was established by inhibition, induction, and spectral studies. The S3(60–100)D fraction, which precipitated between 60 and 100% ammonium sulfate saturation, was soluble, contained flavoprotein, and was functional in reducing cytochrome P-450 or cytochrome c in the presence of NADH. A tentative scheme for n-octane hydroxylation in the Corynebacterium 7E1C system is proposed.

sp. (7ElC) oxidized n-octane to 1-octanol and OCtanoic acid in the presence of NADH and 0 2. The hydroxylating activity, assayed by direct estimation of the reaction products, was found to be concentrated in the clarified Sa supernatant fraction after centrifugation at 144,000 X g for 2 hours.
By use of mass spectrometry it was shown that molecular oxygen is incorporated into the substrate during hydroxylation. The hydroxylating enzyme system was separated into two protein fractions, both of which were required for activity.
The participation of this hemoprotein in n-octane hydroxylation was established by inhibition, induction, and spectral studies.
The S3(60-100)D fraction, which precipitated between 60 and 100% ammonium sulfate saturation, was soluble, contained flavoprotein, and was functional in reducing cytochrome P-450 or cytochrome c in the presence of NADH.
A tentative scheme for n-octane hydroxylation in the Corynebaclerium 7ElC system is proposed.
Enzyme systems catalyzing the hydroxylation of n-alkanes have been isolated from several bacterial species (l-3), and have also been studied in intact microsomal fractions of vertebrate animals (4).  described a soluble enzyme system from Pseudomonas oleovorans which converts n-octane to 1 -octanol in the presence of NADH and oxygen.
The system consists of three protein fractions; a rubredoxin, a reductase, and an w-hydroxylase.
Only the first two fractions were purified and characterized.
This system also consists of three protein fractions; a nonheme iron protein, a flavoprotein, and an unidentified hydroxylase.
In contrast to the hydroxylase from the Pseudomonas the nalkane-hydroxylating system from liver microsomes contains a CO-binding hemoprotein, cytochrome P-450 (4). Cytochrome P-450 has been implicated in a number of biological hydroxylations (9-13).
Its function is not yet clear; however, it is believed to be involved in the terminal oxidation reaction of mixed function oxidases (14,15).
The lack of information on the terminal oxidase of bacterial n-alkane-hydroxylating systems prompted us to investigate the problem with the use of cell-free extracts of Corynebucterium sp. In a preliminary communication (3), we reported that the hydroxylation of n-octane was catalyzed in Corynebucterium 7ElC by a multienzyme system. The system was separated into two distinct protein fractions, one containing cytochrome P-450, and the other having the spectral characteristics of a flavoprotein. This paper describes further the studies on the characterization of this enzyme system.

EXPERIMENTAL PROCEDURE
Culture XethodsBatch cultures of Corynebucterium sp. strain 7ElC (ATCC 19067) were prepared with the use of Medium L (16) t.o which n-octane (99 moles 70 pure) was supplied in the vapor phase (essentially air saturated with n-octane) as the sole source of carbon.
Twenty-five l-liter batches were placed on reciprocal shakers and incubated for 34 days at 28" and a total of 35 g of cells (weight weight) were obtained.
After continuous flow centrifugation, and thorough washing in cold 0.067 M phosphate buffer, pH 7.4, the cells were resuspended in 150 ml of buffer and stored at -25".
For comparative studies on the induction of the hydroxylating enzyme system, 2% (w/v) sodium acetate in Medium L was used as the growth substrate.
In such studies the inoculum culture had been transferred at least 10 times on a hydrocarbon-free medium; otherwise the procedure was identical with that used for the octane-grown cells.
Preparation of Cell-free ExtractsUnless otherwise stated, all subsequent operations were carried out at 4". Cell-free extracts were obtained from the above resting cell suspensions by sonic oscillation, followed by differential centrifugation. During sonic treatment, the temperature of the cell suspension was maintained below 4" and the output of the sonifier was kept constant at 6.2 amp. The sonically disrupted cell suspension was centrifuged at 37,500 X g for 20 min to remove unbroken cells and cell debris. A second centrifugation at 144,000 x g for 2 hours yielded a clear orange supernatant, or Ss fraction, which contained the active hydroxylating enzyme system. The Ss fraction could be stored at -25" for 2 months without appre- Mixtures of authentic 1-heptanol and 1-octanol were diluted in diethyl ether to the concentration ranges comparable to those of the enzymatic samples. Samples of 5 ~1 were chromatographed on Column A as described.
The peak areas of 1-heptanol and loctanol were measured on the chromatogram by means of a compensating polar planimeter.
The area ratios (1-heptanol to loctanol) were then plotted as a function of molar ratios. The same procedure was used for the calibration curve for heptanoic acid and octanoic acid. The acids were separated on Column B as described.
ciable loss of hydroxylating activity. The hydroxylase in the SB fraction was resolved into two protein fractions by ammonium sulfate precipitation, and both fractions were required for activity (3). The &(25-40)D, a red fraction, precipitated between 25 and 40% ammonium sulfate saturation; the S3(60-lOO)D, a greenish yellow fraction, precipitated between 60 and 100% saturation.
Both ammonium sulfate fractions were very labile and had to be used within 2 days. Protein concentrations were determined by the biuret method of Gornall,Bardawill,and David (17).
Assay for n-Octane-hydroxylating Activity-The n-octanehydroxylating activity in cell-free extracts of Corynebacterium 7ElC was measured by the direct estimation of the reaction products, i.e. 1-octanol and octanoic acid. The enzyme assay was carried out in 50.ml Erlenmeyer flasks sealed with rubber stoppers.
The standard reaction mixture of 5.0 ml contained 0.335 mmole of phosphate buffer, pH 7.4, 0.03 ml of n-octane (99.85 moles % pure), 5 pmoles of NADH, and enzyme. The flasks were incubated on a rotary shaker (300 rpm) at 30". The reaction was stopped after 20 min by addition of 0.5 ml of concentrated HCl. At this time 1 pmole of I-heptanol (0.1 ml of 0.01 M heptanol in ethanol) or 1 pmole of 1-heptanoic acid (0.1 ml of a 0.01 M aqueous solution of sodium heptanoate), or both, was added to the reaction vessel as an internal standard.
The mixture was then extracted three times with a total volume of 12 ml of diethyl ether. The ether extracts were concentrated under Nz to approximately 20 ~1 and the reaction products were quantitatively determined by vapor phase chromatography.
Unless stated otherwise, enzymatic activities were expressed as millimicromoles of I-octanol and octanoic acid formed per min per mg of protein at 30". separation and Quantitative Estimation of i-Octanol-A Perkin-Elmer model 154 C vapor fractometer, fitted with a flame ionization detector, was used for all routine gas chromatographic analyses. Unless otherwise indicated, l-octanol and I-heptanol were separated on a 6-foot stainless steel column packed with 15% di-2-ethylhexyl sebacate on 100 to 120 mesh Gas-Chrom P (Column A). The operational conditions were the following: column temperature, 135"; carrier gas (helium) flow rate, 60 ml per min.
Between 2 and 5 ~1 of the concentrated ether samples were injected.
1-Octanol was quantitatively determined by comparing its normalized peak area to the peak area of the internal standard (I-heptanol).
A calibration curve had been previously obtained by chromatographing known mixtures of I-heptanol and I-octanol and by plotting peak area ratios against molar ratios.
As shown in Fig. 1, this plot is a straight line, indicating that (a) the detector response towards both l-heptanol and loctanol is a linear function of concentration, and (b) the relative detector response per mole of I-heptanol is 0.94 (detector response per mole of I-octanol = 1.00). The factor 0.94 was, therefore, used to normalize the peak area of 1-octanol to that of the standard.
The amount of product is given by equation where x = micromoles of octanol; 8-OH = area of the l-octanol peak; 7-OH = area of the l-heptanol peak; and 0.94 = normalization factor.
Assuming that 1-heptanol and I-octanol are extracted from the reaction mixture to the same extent, the internal standard technique gives quantitative results regardless of extraction losses that occur during the preparation of the sample (18). separation and Quantitative Estimation of Octanoic Acid-Octanoic acid and heptanoic acid were separated on a IO-foot stainless steel column, packed with 0.25g;h Carbowax 20 M and 0.4% isophthalic acid on 200 p glass beads (19) (Column B). The operational conditions were the following : column temperature, 140"; helium flow rate, 80 ml per min.
Between 2 and 5 ~1 of the concentrated ether extracts of the reaction mixtures were chromatographed on Column B. The quantitative determination of octanoic acid was obtained also by comparing the peak area of the product to that of the standard. The calibration curve in Fig. 1 shows that the relative detector response per mole of heptanoic acid is 0.78 (detector response per mole of octanoic acid = 1.00). Knowing the amount of standard added (1 pmole), the amount of product is calculated by the following equation where z = micromoles of octanoic acid; 8-COOH = area of the octanoic acid peak; 7-COOH = area of the heptanoic acid peak; and 0.78 = normalization factor.

Carbon Monoxide Inhibition
Studies-Mixtures of CO-air and Nz-air were prepared in 2-liter suction flasks. The reaction vessels were then flushed with 1 liter of the gas mixture and immediately sealed. These operations were carried out at 4". l*O Incorporation Study-The conditions for the la0 incorporation experiment were as follows: 70 mg of NADH were added to 100 ml of S3 fraction (10 mg of protein per ml) in a 500.ml suction flask. The mixture was allowed to freeze at -2O", and the chilled flask was twice evacuated and flushed with nitrogen. After a third evacuation, the oxygen vial (100 ml of oxygen gas, 95% enriched with 1802) was connected to the side arm and the gas was allowed to enter the flask. The pressure inside the flask increased from approximately -760 to -580 mm of mercury. Nitrogen was then added until a residual negative pressure of 50 mm of mercury was attained.
The flask was then sealed and allowed to equilibrate to 30" in a water bath. The reaction was initiated by injecting 0.15 ml of n-octane (99.85 moles $ZO pure) through a rubber septum with a hypodermic needle. The flask was incubated at 30" for 20 min on a rotary shaker (300 rpm). The reaction mixture was then extracted three times with a total volume of 150 ml of diethyl ether. The combined extracts were concentrated under reduced pressure to approximately 10 ml, filtered through anhydrous Na$Ol, and concentrated to a final volume of 0.2 ml. Pure 1-octanol was obtained from t.his ether extract by means of preparative vapor phase chromatography.
Pure samples of octanolJ60 were obtained by the same procedure from reaction mixtures that were incubated under normal atmosphere.
The ether extracts of reaction mixtures were chromatographed on a 12-foot copper column packed with 15% ethylene glycol succinate on Chromosorb W (60 to 80 mesh). Samples of 30 to 40 ~1 were injected.
A thermistor detector was used to monitor the emergence of the product.
The operational conditions were as follows: column temperature, 119"; helium flow rate, 50 ml per min.
Infrared Xpectrophotometry-Infrared absorption spectra of authentic 1-octanol and of the experimentally derived product were recorded on a model 137 Perkin-Elmer spectrophotometer, with the use of neat samples between NaCl prisms.
iVlass Spectrometry-Mass spectra of l0octanol and 180-octanol were recorded on a Consolidated 21-103 mass spectrometer, with an ionizing voltage of 70 e.v. and an ion source temperature of 225". Spectrophotometric Enzyme Assays-All spectrophotometric enzyme assays were carried out at 25", with a Beckman DU spectrophotometer and cuvettes of l-cm light path. The total volume of the reaction mixtures was 1.0 ml. The specific activities were calculated from an average of the readings that were taken at 15-see intervals during the first 2 min of reaction.
The reaction mixture for the NADH-cytochrome c reductase assay contained the following: 50 pmoles of Tris buffer, pH 8.0, 0.5 pmole of EDTA, 0.5 pmole of KCN, 0.2 pmole of NADH, 0.7 mg of oxidized cytochrome c, and enzyme.
Cytochrome c reduction was measured by following the increase in absorbance at 550 mp. The extinction coefficient (reduced minus oxidized) used for cytochrome c was 18.5 rnM+ cm+ (20).
The reaction mixture for the NADH-2,6-dichloroindophenol reductase assay contained 50 pmoles of phosphate buffer, pH 7.4, 0.5 pmole of KCN, 0.2 pmole of NADH, 0.035 pmole of DC&l and enzyme. The reduction of DC1 was measured by the decrease of absorption at 600 rnp. Activities were calculated by using the extinction coefficient of 20.5 mM-' cm-* (21).
The reaction mixture for the NADH oxidase assay contained 30 pmoles of phosphate buffer, pH 7.4, 0.2 pmole of NADH, and enzyme.
NADH oxidation was followed at 340 mp. The extinction coefficient of 6.22 rnM-r cm+ was used (22). NADH-ferricyanide reductase activity was measured under the same conditions as NADH-DC1 reductase activity. In this case 1.6 pmoles of K,Fe(CN)G were used, and the activity was estimated by changes in absorbance at 410 mp, with an extinction coefficient of 1 .O rn& cm-r (23).
Spectral AnalysesDifference and absolute spectra were recorded on a Cary model 14 recording spectrophotometer at room temperature, with cuvettes with a l-cm light path. When spectra of the CO compound of reduced pigments were to be measured, the sample was first reduced by addition of a few crystals of sodium dithionite, then saturated with CO by bubbling the gas for 30 sec.
Special Chemicals and ReagentsHydrocarbons were purchased from the Phillips Petroleum Company, Bartlesville, Oklahoma. Coenzymes were obtained from Sigma. The lsOZ sample was obtained from the Volk Radiochemical Company, Burbank, California.
The specifications were 170, 0.4227,; 1x0, 95.67,. Stationary phases and supports for vapor phase chromatography columns were purchased from Applied Science Laboratories. Column B was obtained from the same source.

RESULTS
Cofactor Requirements-The hydroxylation of n-octane by the cell-free extract of Corynebacterium 7ElC requires NADH as a cofactor (3). Maximal activity was recorded when 1 mM NADH was added to the reaction mixture.
No activity was observed with heated enzyme nor when the high speed supernatant, &, was assayed in the absence of cofactors.
Very little activity was observed when either NAD or NADPH was subst,ituted for NADH at the 1 mM level. The addition of ferrous ions (5 x 10w4~) appeared to inhibit the reaction.
Neither FMN nor FAD affected the n-octane-hydroxylating activity of the Ss fraction.
Product Formation as Function of Time and Protein Concentration-The enzymatic formation of l-octanol and octanoic acid as a function of time is shown in Fig. 2. The formabion of loctanol is a linear function of time during the first 10 to 15 min. At 20 min the concentration of 1-octanol reaches its peak. After this time, its concentration decreases. On the other hand, the rate of formation of octanoic acid is constant over the time period studied.
The curves of Fig. 2 indicate that (a) the over-all hydroxylation reaction is no longer significant after 20 min, and (b) I-octanol is converted to octanoic acid. The alcohol and aldehyde dehydrogenases, which presumably catalyze the latter reaction, are still active after a 60.min incubation interval. Fig.  2 shows that, at 20 min, the concentration of I-octanol is maximal and approximately 4 times the concentration of octanoic acid. This relationship is charact'eristic of all preparations that were studied.
Consequently a simplified assay was developed based on the measurement of I-octanol only after 20 min of incubation. This was used routinely to determine n-octane-hydroxylating activity of cell-free extracts. Fig. 3 shows the effect of enzyme concentration on the amount of I-octanol and octanoic acid formed after the ZO-min incubation interval.
This experiment was repeated several times. A linear relationship between enzyme concentration and amount of reaction product formed was never observed.
This finding is mixtures and separated by means of preparative vapor phase chromatography.
The identity of the enzymatic product with I-octanol was established by retention time and by infrared spectrophotometry.
The infrared spectra of the enzymatic product formed in an atmosphere of 1602 and of authentic l-octa-no1 were identical.
The mass spectrum of 1-octanol obtained from the enzymatic oxidation of n-octane was in good agreement with the data reported by Friedel, Schultz, and Sharkey (24). Since I-octanol gives a very weak parent peak, the oxonium ion CH2=+OH (at m/e 31) was selected to show '80 incorporation.
As shown in Fig. 4d, the oxygen containing fragment at m/e 31 has a relative abundance of 38% whereas the ion at m/e 33 is insignificant (relative abundance 27,).
In contrast, the mass spectrum of I-octanol obtained from the l0enriched mixture (Fig. 4B) shows a marked increase in the abundance of the ion at m/e 33 whereas the relative abundance of the m/e 31 ion is greatly decreased. These results are consistent with the presence of I80 in the hydroxyl group of octanol.
The unexpectedly high intensity of the m/e 31 peak (relative abundance ll%, Fig. 4B) can be attributed to some unlabeled octanol that was present in the mixture before the reaction was started.
In addition, it is likely that a H-C&O ion contributes to the m/e 29 peak (Fig. 4A). Thus, in the mass spectrum of octanol formed in i*02-enriched atmosphere, one observes a decrease of the m/e 29 peak, accompanied by an increase of the m/e 31 peak. These findings unequivocally indicate that molecular oxygen is incorporated into n-octane in the enzymatic conversion of I-octanol.
Requirement of Two Enzyme Fractions for Hydroxylation of n-Octane-Attempts at purifying the hydroxylating enzymes by use of ammonium sulfate precipitation resulted initially in the loss of the activity.
Subsequently, it was established that two enzyme fractions are required for the n-octane hydroxylation. As shown in Table I, neither the fraction that precipitates between 25 and 40 y0 ammonium sulfate saturation, nor that which precipitates between 60 and 100% saturation, is active alone. However, the activity is reconstituted when these two fractions are combined.
Optimal hydroxylation rates were observed when  is active by itself. However, its specific activity is approximately one-half of that observed for the combined fractions.

XtudiesTable
II shows the effect of varying concentrations of carbon monoxide on the n-octane hydroxylating activity of the SI fraction.
The hydroxylating activity is completely inhibited by CO-air, 50:50 (v/v) mixtures and is still strongly inhibited by 5:95 (v/v)

mixtures.
These results suggest the participation of a hemoprotein in the hydroxylation reaction.
The data of Table II also show that the hydroxylation of n-octane is inhibited under reduced oxygen pressure. Table III shows the effect of other inhibitors on the hydroxylation activity of the S3 fraction.
Heavy metals (CL?, Hg++, p-chloromercuribenzoate) inhibit the hydroxylation effectively at a concentration of 10W4 M. On the other hand, cyanide, azide, arsenite, and EDTA are ineffective inhibitors at a concentration Of lo+ M.

No inhibition
was observed with SKF-525 (lop5 M) which is known to inhibit cytochrome P-450 in the N-demethylation reaction of certain drugs by rat liver microsomes (25).
Cytochrome c and 2,6-dichloroindophenol produced approximately 50yo inhibition at a concentration level of lop3 M. These results suggest that cytochrome c and DC1 can react with the enzymes involved and drain electrons from the hydroxylating system. Ferricytochrome c inhibits n-octane hydroxylation even when NADH is present in excess concentration.
Spectral Studies on Hydroxylating Enzyme Fractions-As reported earlier (3) one of the two protein fractions that was required for hydroxylation contains cytochrome P-450. Fig. 5 shows the difference spectra of the S3(25-40)D fraction.
The dithionite-reduced minus oxidized preparation (solid line) is characterized by a broad absorption peak at 429 rnp with a shoulder at 440 to 445 rnp region and a weak absorption peak at 555 mb. Addition of CO to the reduced preparation brings about a marked change in the Soret region; a broad peak appears at 450 rnM and a small, sharp peak at 425 rnp. No changes are observed at higher wave lengths.
These observations, and the characteristic absence of (Y-and P-bands, are consistent with the presence of cytochrome P-450. Cytochrome P-450 in the S3-(25-40)D fraction was converted readily to the P-420 form by exposure to potassium deoxycholate (3). The absorption at 425 rnp is due to the presence of a small amount of cytochrome P-420. The concentration of cytochrome P-450, calculated from the extinction coefficient of 91 m&f-* cm-l (26), was 0.25 mlmole per mg of protein.
The f&(25-40)D fraction is essentially free from other hemoproteins as revealed by spectral analyses.
Although the cytochrome P-450 fraction is obtained from a high speed supernatant, it is unlikely that it is a soluble hemoprotein.
It was observed that a lipid-soluble, red carotenoid pigment was always present in this fraction, suggesting that lipid was present in the preparation.
Moreover, this S&(25-40)D fraction could be partially clarified by the addition of detergents (deoxycholate and Triton X-100).
The second fraction, S3(6&100)D, required for n-octane hydroxylation, appears to contain flavoprotein. This fraction is greenish yellow in color and the difference spectrum obtained for this fraction has been published previously (3). The characteristic bleaching attributed to flavins (450 rnp region) was noted after the addition of dithionite.
The component responsible for dithionite bleaching was extractable by acid, and the acid-extractable material revealed a typical flavin spectrum.
With the use of the extinction coefficient of 11.3 rnM-l cm-l (reduced minus oxidized at 450 mp), t,he acid-extractable flavin content was calculated to be 1.51 mpmoles per mg of protein (3) For these studies the protein concentration of the Sn(25-40)D fraction, which contained the cytochrome P-450 component, was 10 mg per ml, and the final concentration of NADH was 1.0 mM. protein component.
Although the presence of a relatively large concentration of flavin in a fraction obligately required for n-octane hydroxylat,ion, would not directly implicate its involvement in t,his reaction, it is difficult, however, to postulate a mechanism as to how NADH (a required oxidation-reduction component for n-alkane hydroadation) might function, if not via a flavoprotein intermediary.
A trace amount of a hemoprotein component was also noted in the S&(60-100)D fraction.
This and was subsequently identified as cytochrome a~. This cytochrome appeared to be a constitutive component in Coorynebacterium 7ElC since it was found in approximately equal concentrations in both octane-and acetate-grown cells.
Enzymatic Reduction of Cytochrome P-450--In view of the possible function of cytochrome P-450 in n-alkane hydroxylation and of the NADH requirement of this reaction, it was interesting to extablish whether cytochrome P-450 could be reduced by NADH.
CO does not react with the oxidized P-450 but combines readily with the reduced form of the hemoprotein, giving the characteristic peak at 450 mp. By using this property of cytochrome P-450, it was possible to follow its reduction upon addition of NADH or NADH and catalytic amounts of S3(60-1OO)D flavoprotein fraction.
As shown by the curve of Fig.  6B (.. . .) the addition of N&4DH partially reduced the cytochrome.
Complete reduction of cytochrome P-450 is brought about by the addition of NADH in the presence of catalyt'ic amounts of the S,(6@lOO)D fraction (Fig. 6B, --). Indeed, the CO peak of the enzymatically reduced cytochrome P-450 (Fig. 6B, ---) is comparable to the CO peak obtained by dithionite reduction (Fig. 611, --).
The comparable fraction obtained from acetate-grown cells contains only 0.04 mpmole per mg of protein of the CO-binding hemoprotein (in this case all the cytochrome was present in the P-420 form).
Thus a B-fold increase in concentration is induced by growth in the presence of n-octane.
A similar induction of cytochrome P-450 occurs in the liver microsomes of phenobarbitaltreated animals (13,27).
The n-octane-hydroxylating activities of enzyme fractions obtained from octane-and acetate-grown cells were tested. The results of these experiments are shown in Table IV. Experiment 1 shows that the high-speed supernatant from acetategrown cells (&ac) is neither active by itself nor when combined with the cytochrome P-450 fraction, S&(25-40)D, obtained from n-octane-induced cells. This indicates that the flavoprotein component of the hydroxylating system is not present in noninduced cells. However, the So&c fraction shows some activity when combined with the flavoprotein fraction from the n-octaneinduced cells, S3(60-100)D. Experiment 2 shows that this activity concentrated in the S&(25-40)ac fraction.
As mentioned above, the latter fraction contains low levels of cytochrome P-420 and is spectrally free of the P-450 form. The capability of obtaining reconstitutive activity with the P-420 form of cytochrome is in contrast with the observations of several investigators (27)(28)(29) who have shown that microsomal cytochrome P-420, obtained by treatment with detergents, sulfhydryl reagents, and lipolytic or proteolytic enzymes, is catalytically inactive. It is possible that the cytochrome P-420, obtained solely by sonic disruption of acetate-grown cells and precipitation by ammonium sulfate, allowed this form of the cytochrome to retain some activity.
NADH Oxidation by X3 and Derived Ammonium Sulfate Fraclions--The S&(60-1OO)D fraction is catalytically active in transferring electrons from NADH to the oxidized cytochrome P-450 (Fig. 6B). It was also observed ( Table III) that cytochrome c and DC1 inhibit the hydroxylation of the n-octane, suggesting that they may compete with cytochrome P-450 as electron acceptors. In view of these observations, it would be expected that the &(60-1OO)D fraction would catalyze the transfer of electrons from NADH to artificial electron acceptors, cytochrome c and DCI.
That this actually does occur can be seen from the following results which compare the activities for NADH oxidation (with various acceptors) by the Sp and the three derived ammonium sulfate fractions.
All fractions were completely devoid of NADH oxidase activity but possessed DCI, cytochrome c, and ferricyanide reductase activities. Of all the fractions assayed, the S,(6@100)D fraction exhibited the highest specific activities for NADH oxidation with these three electron acceptors.
On the basis of a 2-electron change, and in terms of millimicromoles per min per mg of protein (25"), the &(60-100)D fraction exhibited specific activities of 335, 22.5, and 485 with the use of DCI, cytochrome c, and ferricyanide, respectively, as electron acceptors for NADH oxidation.
Although the act.ivities were distributed throughout all three derivative ammonium sulfate fractions, highest specific activity for the NADH "diaphorasetype" activity was consistently found in the &(60-1OO)D fraction. Since this fraction is required for n-octane hydroxylation, it suggests that a functional NADH flavoprotein component is most probably associated with the n-alkane hydroxylation reaction.

DISCUSSION
The enzymatic hydroxylation of n-octane in Coynebacterium 7ElC requires NADH and oxygen.
The results of the IsO incorporation experiment establish conclusively that oxygen is incorporated into n-alkane during hydroxylation. The process can be described by the following reaction R-CHP + NADH + H+ + '*Ot + R-CH2'80H + NAD+ + H2180 where NADH is the electron donor and O2 the electron acceptor. Thus the conversion of n-octane to I-octanol is catalyzed by a mixed function oxidase (30). This term was previously proposed for the w-hydroxylating system of P. oleovorans; however, there was no direct evidence to indicate that oxygen is incorporated into the substrate.
The hydroxylating system of Coynebacterium 7ElC is similar in many aspects to the hydroxylating systems of P. oleovorans and P. o?enitri$cans.
All of these bacterial systems require NADH as electron donor; in contrast, the microsomal system required NADPH (12). The kinetics of end product formation in Coynebacterium 7ElC is comparable to that reported for P. oleovorans (6). Kusunose et al. (2,8) observed that the enzymatic hydroxylation of n-alkanes by P. denitrificans was stimulated by FAD or FMN.
There was no such flavin requirement for n-octane hydroxylation in the Coynebacterium system. The hydroxylating system of Coynebacterium consists of at least two distinct protein fractions.
The &(25-40)D, which contains cytochrome P-450, appears particulate or at least is associated with lipids.
A fundamental difference noted between the P. oleovorans and the Coynebacterium hydroxylating systems is in the sensitivity to CO. Whereas the former system is unaffected by 1 :l or 9:l carbon monoxide-oxygen mixtures (5), the latter is strongly inhibited by low concentrations of CO. Furthermore, the P. oleovorans system apparently does not contain cytochrome P-450 (5, B), while one of the two active protein fractions of Coynebacterium 7ElC did contain this CO-binding hemoprotein.
The studies with cell-free extracts of Coynebactetium 7ElC indicate that cytochrome P-450 is functional in the hydroxylation of n-alkanes.
The following lines of evidence support this conclusion: (a) Spectral studies show that cytochrome P-450 is the only major hemoprotein that is concentrated in one of the two active protein fractions.
This fraction is essentially free of other hemoproteins but may contain small amounts of cytochrome P-420 (the soluble form of cytochrome P-450).
With the exception of small amounts of cytochrome al, no other hemoprotein component can be detected spectrally in any of the purified active cell-free extracts.
(b) Studies with inhibitors showed that the over-all hydroxylation reaction is inhibited by carbon monoxide but is insensitive to cyanide and azide. Cytochrome P-450 reacts specifically with CO and if it were involved in the hydroxylation reaction one would expect the reaction to be CO-sensitive.
(c) Comparative studies with enzymes from n-octane-grown cells and acetate-grown cells indicate that cytochrome P-450 is an inducible hemoprotein. If cytochrome P-450 is involved in n-octane oxidation one would expect higher concentrations of this hemoprotein in cells grown on n-octane.