Fatty Acid Oxidation by roqueforti'

R. L. GIROLAMI AND S. G. KNIGHT Department of Bacteriology, University of Wisconsin, Madison, Wisconsin Received for publication April 7, 1955 Although the oxidation of fatty acids by molds has Production of 2-heptanone was determined by the received only superficial study, it has been found that addition of the following components to 500-ml Erlen- forms methyl ketones as products meyer flasks: 2 g damp mycelium blended in 50 ml of the oxidation. The distinctive taste and odor of water, 45 ml 0.065 M phosphate buffer at pH 7.3, bleu cheese is suggestive of 2-heptanone (Hammer and 5 ml 0.02 M magnesium sulfate, and 10 ml 0.1 M sodium Bryant, 1937), and this compound, together with caprylate. After incubation on a rotary shaker (330 other methyl ketones, has been isolated from a ripe rpm) at 30 C for 3 to 4 hours, the flask contents was cheese and identified (Patton, 1950). distilled directly into a solution of 0.1 per cent 2,4- Experimental work on fatty-acid oxidation by dinitrophenylhydrazine. The precipitate was recrystal- ketone-forming molds has been limited to growing lized twice from hot 95 per cent ethyl alcohol before cultures from which methyl ketones were isolated after identification was attempted. Acetone was determined a period of incubation (Starkle, 1924; Stokoe, 1928; by the method of Greenberg and Lester (1944). Hammer and Bryant, 1937). These experimenters Paper chromatograms were made at room tempera- found that the molds produced methyl ketones of one ture in a closed glass cylinder with 90 per cent methanol less carbon atom than the fatty acids supplied; however, as the developing solvent. The compounds were spotted the results were complicated by the many reactions of on a sheet of Whatman no. 1 filter paper and the Rf growing cultures. The present study was limited, of the unknown derivative was compared to that of therefore, to relatively short experiments with resting known methyl ketone hydrazones. Melting points of cells and a single substrate in order to eliminate syn- the unknown compounds were compared with the thetic reactions. values listed in Shriner and Fuson (1948) and Huntress and Mulliken (1941). EXPERIMENTAL METHODS A strain of Penicillium roqueforti used in the manu- RESULTS facture of bleu cheese was used throughout these studies. The oxidation of fatty acids by whole cells of P. The mold was grown in a medium of 2 per cent lactose roqueforti was established manometrically in the and 4 per cent corn steep liquor. Then, 1 ml of a spore presence of low substrate concentrations, magnesium suspension of the organism was added to 100 ml of ions, and phosphate buffer at pH 7.3. Figure 1 illustrates medium in 500-ml Erlenmeyer flasks; incubation was the effect of substrate concentration on the amount of at 25 C for 48 hours on a reciprocating shaker which oxygen consumed during the oxidation of caprylate. completed one hundred 4-in strokes per minute. The The toxic effect of this acid was evident at concentra- cells were harvested by filtration on a Buchner funnel, tions greater than 10 ,M per flask. This general effect resuspended several times in distilled water, and was noted with all the fatty acids studied and bore a refiltered. A 2-g portion of the damp mycelium was relationship to the carbon chain length of the acid; suspended in 100 ml of distilled water, or 0.065 M the cells had a greater concentration tolerance for the phosphate buffer at pH 7.3, and blended in a chilled shorter fatty acids. Butyrate, for example, was oxidized Waring blendor cup for 10 seconds. Either 1 or 2 ml rapidly at a concentration of 50 ,uM, whereas 2 ,uM of of this suspension were added to Warburg vessels, and laurate were inhibitory in several experiments. Palmitate the oxidation of various substrates was measured at and stearate were not oxidized at a substrate level of 30 C by standard manometric techniques (Umbreit 1 ,UM. The sodium salts of the acids were used, but no et al., 1949). differences in rates of oxidation or tolerance were noted The salts of the fatty acids were made by neutralizing when the potassium salts were substituted. the acids with alkali, usually sodium hydroxide, to pH Early work on fatty-acid oxidation by animal 7.3 and making to volume. Acetoacetic acid was pre- preparations established a requirement for both pared from the ethyl ester by the method of Davis orthophosphate and magnesium. Table 1 shows the (1943). Other substrates were commercial preparations. stimulatory effect of orthophosphate on the oxidation I Published with the permission of the Director of the of caprylate by P. roqueforti; the degree of stimulation Wisconsin Agricultural Experiment Station. varied greatly between experiments but could be 264 1955] FATTY ACID OXIDATION BY PENICILLIUM ROQUEFORTI 265 demonstrated routinely to some extent. If 20 to 40 gM vations, because of the variation among different of fluoride were added instead of phosphate, somewhat batches of cells. Often the point of inflection had to be less stimulation of fatty-acid oxidation was obtained. estimated since the break in the oxygen uptake curve When both phosphate and fluoride were added to- was not so pronounced as in figure 2, especially with gether, the amount of stimulation was no greater than the short-chain fatty acids. with phosphate alone. Stimulation of oxidation by Figure 3 shows the relationship between the amount magnesium was inconsistent, but the metal was added of oxygen consumed during fatty-acid oxidation and routinely because of its general requirement by oxida- the carbon chain length of the acid. Except for the tive systems (Lardy, 1951). Figure 2 shows the results of a typical experiment in which two levels of substrate were used in order to establish whether oxidation was stopping before com- pletion or if it was proceeding as far as substrate con- centration permitted. Experiments similar to this were done with a series of straight-chain, saturated fatty acids as substrates. The results of these experiments are summarized in table 2 and are compared to the theoretical values calculated for the complete oxidation of the fatty acid to carbon dioxide and water. The values reported were established from a series of obser-

eW - 400 - ii

0

3001 -

MINUTES FIG. 2. The oxidation of sodium caprate by Penicillium 02200 roqueforti. Each vessel contained about 3 mg blended myce- lium, 1.5 ml 0.065 M phosphate buffer at pH 7.3, 0.1 ml 0.02 M ENDOGENOUS MgSO4, the indicated amount of substrate, and distilled water to 3 ml. co I I 0 10 20 30 40 M CAPRYLATE TABLE 2. Oxygen consumption during the oxidation offatty acids FIG. 1. The effect of substrate concentration on the oxida- by Penicillium roqueforti* tion of caprylate by Penicillium roqueforti. Each vessel con- 02 Uptake During Oxidation of 1 pM tained about 4 mg blended mycelium, 1.5 ml 0.065 M phosphate Fatty Acid buffer at pH 7.3, 0.1 ml 0.02 M MgSO4, and distilled water to Theoretical Observedt 3 ml.

Acetic ...... 2.0 1.1 TABLE 1. The effect of inorganic phosphate on the oxidation of Propionic ...... 3.5 1.1 caprylate by Penicillium roqueforti* Butyric ...... 5.0 1.6 Phosphate Added 02 Uptake in 120 Min. Valeric ...... 6.5 2.2 Caproic ...... 8.0 2.3 pM L Heptylic ...... 9.5 3.2 0.0 132 Caprylic ...... 11.0 3.3 6.7 171 Pelargonic ...... 12.5 3.4-4.2 13.4 244 Capric ...... 14.0 4.2-4.8 26.8 281 53.6 277 * Each vessel contained about 4 mg blended mycelium, 1.5 107.2 266 ml 0.065 M phosphate buffer at pH 7.3, 0.1 ml 0.02 M MgSO4, the indicated substrate in low concentration (1 to 2 pM), and * Each vessel contained about 4 mg blended mycelium, distilled water to 3 ml. 1.0 ml 0.1 M tris(hydroxymethyl)-aminomethane buffer at pH t Determined by subtracting the endogenous oxygen uptake 7.3, 0.1 ml 0.02 M MgSO4, 5,Mm sodium caprylate, and distilled from the uptake in the presence of substrate at the point at water to 3 ml. which exogenous parallels endogenous. 266 R. L. GIROLAMI AND S. G. KNIGHT [VOI,. 3

short-chain fatty acids, as the carbon chain lengthened TABLE 4. The production of methyl ketones by Penicillium roqueforti up to 9 and 10 carbon atoms the amount of oxygen consumed increased in paired steps. Fatty acids of Substrate Ketone Produced longer than 10 carbon atoms showed the toxic effect described earlier. Butyrate Acetone Valerate 2-Butanone A comparison of the amount of oxygen consumed Caproate 2-Pentanone during the oxidation of butyrate and its oxidation Heptylate 2-Hexanone intermediates is summarized in table 3. None of the Caprylate 2-Heptanone substrates was oxidized so rapidly as the original acid, Pelargonate 2-Octanone* and the ethyl esters were oxidized more rapidly than * No 2-octanone was available. Huntress and Mulliken the sodium salts. Acetoacetate, which accumulates (1941) list 58 C as the melting point of the 2,4-dinitrophenyl- during fatty-acid oxidation by animal tissues, was hydrazone; 52 to 54 C was observed. oxidized rapidly by these cells. After the oxidation of caprylate for several hours, the phenylhydrazone of 2-heptanone and that the sample contents of Warburg flasks had a distinctive ketone was not contaminated with other methyl ketones. odor. This odor, not evident when other fatty acids Although no odor of ketones could be detected during were oxidized, was suggestive of a methyl ketone and the oxidation of other fatty acids in Warburg vessels, was identified as 2-heptanone by melting-point (74 to methyl-ketone formation from these acids was studied 78 C) and mixed melting-point (76 to 77 C) determina- by following the methods described for the production tions of the hydrazone derivative. Paper chromatograms and detection of 2-heptanone from caprylate. Very of this compound indicated that it was the 2,4-dinitro- little or no ketone was formed in these experiments. However, when cells grown in corn steep-lactose medium of half normal concentration were used, the 400 - amount of ketone produced during the oxidation of other fatty acids was increased greatly. Table 4 lists the fatty acids and the corresponding ketones which 300 were isolated. The amount of ketone produced was variable, but approximately 2 per cent of the caprylate 0 supplied was oxidized to 2-heptanone. Variations in -~200 gas phase and pH indicated that conditions favoring most rapid substrate oxidation, that is, near neutral pH

100 and an 02 concentration of air or greater, also allowed the greatest formation of ketone. Only a few crystals of derivative were formed after oxidation either of

0 valerate or pelargonate. Although a ketone odor could 2 3 4 5 6 7 8 9 10 12 14 ENDO. be detected after the oxidation of caprate, not enough CARBON CHAIN LENGTH of the compound (2-nonanone?) was formed to allow FIG. 3. The relationship between carbon chain length and a melting-point determination. amount of oxygen consumed during fatty-acid oxidation by Penicillium roqueforti. Each vessel contained about 3 mg Inasmuch as methyl ketones are a product of fatty- blended mycelium, 1.5 ml 0.065 M phosphate buffer at pH 7.3, acid oxidation by P. roqueforti, the effect of these com- 0.1 ml 0.02 M MgSO4, 2 ,uM of the indicated substrate, and dis- ponents on the growth and activity of this organism tilled water to 3 ml. The vessels were incubated for 210 min. was studied. It was found that unadapted cells were un- able to oxidize the TABLE 3. The oxidation of intermediatss in the scheme of beta ketones further and that the presence oxidation of butyrate by Penicillium roqueforti* of the ketone did not affect the oxidation of the fatty acid from which it was produced. The presence of 0.01 Substrate 02 Uptake in 150 Min. per cent or more 2-heptanone in corn steep-lactose IAL medium inhibited all mold growth; if the amount was Sodium butyrate 290 less than 0.01 per cent, the mold grew slowly and Sodium crotonate 130 gained the ability to oxidize the ketone slightly, about Sodium beta-hydroxybutyrate 120 10 per cent as rapidly as caprylate oxidation. Ethyl beta-hydroxybutyrate 220 Sodium acetoacetate ...... 200 DISCUSSION Ethyl acetoacetate ...... 210 Endogenous ...... 110 Nieman (1954) suggested that fatty acids are toxic

* Each vessel contained about 3 mg blended mycelium, 1.5 because of their adsorption on the cell wall. If this is ml 0.065 M phosphate buffer at pH 7.3, 0.1 ml 0.02 M MgSO4, true with P. roqueforti, then increased inhibition of 50,Mm substrate, and distilled water to 3 ml. oxidation with increasing chain length as well as with 1955] FATTY ACID OXIDATION BY PENICILLIUM ROQUEFORTI 267 increasiing conceiitration could be strictly physical. SUMMARY The toxic effect of fatty acids with increasing chain length was also noted by Rolinson (1954), who found The oxidation of fatty acids by Penticillium roqueforti that was able to oxidize was investigated, and it was found that resting cell palmitate and stearate at concentrations as high as 1.0 suspensions of the organism were capable of oxidizing per cent (w/v); here, a physical explanation alone would fatty acids in the presence of phosphate, magnesium, not suffice. and low substrate concentration. The acids became Determination of oxygen consumption during the increasingly toxic as the carbon chain lengthened. oxidation of fatty acids resulted in figures which could Acids of longer than 10 carbon atoms were oxidized account for approximately one-third of the acid's only slowly, if at all, even at low substrate concentra- being oxidized to carbon dioxide and water. As expected, tions. the amount of oxygen consumed during the oxidation It was found that a methyl ketone containing one increased as the carbon chain lengthened. All the acids less carbon atom than the fatty-acid substrate was an through C10 were oxidized without lag and to approxi- end-product of the oxidation. These compounds were mately the same extent. Intermediates in the scheme of not metabolized further by resting cell suspensions but butyrate oxidation were oxidized, although not so were inhibitory to a growing culture of the organism. rapidly as butyrate. Partial impermeability of the cells The amount of ketone produced was related to the rate to these substrates or failure to supply the "active" intermediate could account for the decreased rate of of fatty-acid oxidation.

oxidation. - REFERENCES Part of the discrepancy between theoretical and observed could be for DAVIS, R. 1943 Studies on the acetone-butanol fermenta- oxygen uptake accounted by the tion; Acetoacetic acid decarboxylase of Clostridium production of methyl ketones. A possible mechanism acetobutylicum (BY). Biochem. J. (London), 37, 230-238. for the formation of these ketones is the oxidation of GREENBERG, L. A., AND LESTER, D. 1944 A micromethod the fatty acid to the beta-keto acid and then the for the determination of acetone and ketone bodies. J. decarboxylation of this acid (Stiirkle, 1924; Thaler Biol. Chem., 154, 177-190. and Geist, 1939). Stokoe (1928) suggested that in HAMMER, B. W., AND BRYANT, H. W. 1937 A flavor constit- uent of ( type). Iowa State Coll. ketone-forming fungi the acid was sufficiently poisonous J. Sci., 11, 281-285 to alter the normal scheme of oxidation. These theories HUNTRESS, E. H., AND MULLIKEN, S. P. 1941 Identification do not appear to account for the formation of only one of Pure Organic Compounds. Order 1. John Wiley and ketone from a fatty acid, since several beta-keto acids Sons, New York. would be formed during the oxidation of the longer- LARDY, H. A. 1951 The influence of inorganic ions on phos- phorylation reactions. In McElroy, W. D., and Glass, B. chain acids. Attempts to clarify the mechanism have (editors): Phosphorus Metabolism, Vol. 1. The Johns been impeded by the impermeability of P. roqueforti Hopkins Press, Baltimore, Md. to these substrates. Efforts to obtain preparations free NIEMAN, C. 1954 Influence of trace amounts of fatty acids from the limitations imposed by permeability have on the growth of microorganisms. Bacteriol. Revs., 18, been unsuccessful. 147-163. PATTON, S. 1950 The methyl ketones of blue cheese and their Because it was found that methyl ketones had an relation to its flavor. J. Dairy Sci., 33, 680-684. inhibitory effect on growth when present in sufficient ROLINSON, G. N. 1954 The effect of saturated fatty acids concentration, it is interesting to speculate that the on oxygen uptake by Penicillium chrysogenum. J. Applied production of these ketones in a ripening cheese may Bacteriol., 17, 190-195. serve as a self-regulating mechanism. As the organism SHRINER, R. L., AND FuSON, R. C. 1948 Identification of Organic Compounds. John Wiley and Sons, New York. grows, the production of these compounds may inhibit STARKLE, M. 1924 Die Methylketone im oxydativen Abbau overgrowth of the mold during the ripening period. einiger Triglyceride (bzw. Fettsauren) durch Schimmel- One might expect that the reduced PO2 within a cheese pilze unter Berucksichtigung der besonderen Ranziditat might favor the formulation of such incomplete oxida- des Kokosfettes. Biochem. Z., 151, 371-415. tion products as methyl ketones. Results with standing STOKOE, W. N. 1928 The rancidity of coconut oil produced cultures did not corroborate this possibility, since it by mould action. Biochem. J. (London), 22, 80-93. was observed that lowered oxygen tension actually THALER, H., AND GEIST, G. 1939 Zur Chemie der Ketonran- reduced the amount of ketone and that a zigkeit; tlber die Bildung von Methylketonen aus ,-Oxy- formed PO2 fettsiiuren durch Penicillium glaucum. Biochem. Z., 302, greater than that of air did not increase ketone yields. 369-383. The amount of ketone produced was found to be re- UMBREIT, W. W., BURRIS, R. H., AND SrAUFFER, J. F. 1949 lated to the rate of fatty-acid oxidation and not the Manometric Techniques and Tissue Metabolism. (Rev. ed.) amount of available oxygen. Burgess Publishing Co., Minneapolis, Minn.