Microbial fatty acid specificity

Folia Microbiol. 42 (2), 117-120 (1997)

Microbial Fatty Acid Specificity G. AGGELISa*, G. PAPADIOTISb and M. KOMArrIs b aDepartment of Agricultural Biotechnology bDepartment of Food Science and Technology Agricultural Universityof Athens, Athens 118.55, Greece Received July 30, 1996 Revised version October 21, 1996 ABSTRAL-'r. Strains of Rhodotorula sp., Candida spp. and Langermania sp. cultivated on polyunsaturated oil preferentially incorporated more unsaturated fatty acids. These fatty acids were used mainly for growth needs whereas the saturated ones accumulated in the microbial cell. The cellular oil and the remaining oil in the culture had a lower degree of unsaturation as compared to the initial oil, and a modified fatty acid composition. Candida lipo~ytica, in a chemostat continuous culture, incorporated C18 fatty acids in the order of C18:3 > C18:2 > C18:1 > C18:0, and accumulated mostly the saturated ones. The specific productivity of the cellular oil and of the oil remaining in the culture medium was 0.036 and 0.487 g g - 1 h - l , respectively, at dilution rate D = 0.2/h.

The fats and oils industry is able to modify crude materials by chemical and physico-chemical methods, e.g. hydrogenation, interesterification and crystallization. These processes are limited as to the type and specificity of changes that can be brought about. For example, it is not possible to remove et-linolenic acid from soybean oil economically and specifically by conventional technology (Glatz et al. 1984). However, the use of specific lipases yields valuable products unobtainable by chemical methods (Macrae 1984). Moreover, conversion of common fats to more desirable forms can be attained by growing oil-accumulating microorganisms on fatty substrates (Glatz et al. 1984; Koritala et al. 1987). During microbial growth and/or accumulation of fat reserves, the dominating phenomena defining the composition of the cellular and residual culture medium fats, are the specific hydrolysis of the substrate (Aggefis et al. 1993, 1995a), the specific incorporation of substrate fatty acids into the cell (Button 1993; Aggelis et al. 1995c) and the intracellular transformations of fatty acids (Aggelis et al. 1991a,c). Fatty acids are either degraded for growth needs or act as the substrate of intracellular biotransformations, thus leading to compositional changes (Montet et al. 1985; Samelis et al. 1993; Aggelis et al. 1995b,c). It may thus be possible to obtain oils with altered but nonrandom glyceride structure and composition (Glatz et aL 1984; Ratledge and Boulton 1985; Aggelis et aL 1991b). The aim of this work was to study the specificity of some microorganisms from our collection against the most common fatty acids of the C18 group. The specificity of Candida lipolytica was studied in a carbon-limited chemostat continuous culture.

MATERIALS AND METHODS

Microorganisms and culture conditions. The microorganisms used were Rhodotorula sp., Candida tropicalis, C. lipolytica, C. cremoris and Langerrnania gigantea from the culture collection of the Laboratory of Food Microbiology, Agricultural University of Athens. They were kept on potato-dextrose

agar at 8 ~ Evening Primrose Oil (EPO) was used as the model carbon source because it contains all the common fatty acids of the C18 group: C18:0 (stearic acid), C18:1 (oleic acid), C18:2 (linoleic acid) and Clg:3 (y-linolenic acid). Thus, it was possible to examine the specificity which various microorganisms show for fatty acids bearing the same aliphatic chain length but different number of double bonds. Shaker cultures were performed in 250-mL conical flasks containing 50 mL of an organic medium (g/L): EPO (degree of unsaturation a, A/mol = 1.83) 10, peptone 5, yeast extract 1; pH was 5.5. The medium was sterilized in an autoclave at 121 ~ The flasks were inoculated with approximately 107 cells and incubated in a rotary shaker (frequency 3.3 Hz) at 28 +- 1 ~ for 4 d. Fermentor culture was performed in a Gallenkamp modular fermentor with a working volume of 1 L on a carbon-limited mineral medium (ratio C/N = 3). It contained (g/L): EPO (A/mol -- 1.71) 5, NH4CI 5, Na2HPO4"2H20 2, KH2PO4"7H20 7, MgSO4"7H20 1.5, CaCI2"H20 0.15, yeast extract 1,

*Corresponding author. a(monoene x 1 + diene x 2 + triene x 3)/100.

11 $

G. AGGELIS et al.

Voi. 42

polyoxyethylene sorbitan monostearate 0.5. The medium pH was adjusted to 5 and automatically controlled by adding 1 mol/L NaOH. The medium was sterilized in an autoclave at 121 *C. The cultivation temperature was 27 _+ 0.5 ~ Aeration rate was 0.5 VVM and stirrer frequency 10 Hz. A 0.5 % inoculure was added and grown as a batch culture for 2 d. Steady-state conditions were maintained for at least five complete changes of medium in the vessel. Harvest of biomass, lipid determination and analysis. The biomass was harvested by centrifugation at 16 000 g for 20 rain and washed three times with hexane and distilled water in order to remove external lipids or soaps. To verify the complete removal of lipids from the cell surface the last hexane eluate was checked by TLC (60G silica gel, Merck). The development solvent was composed of petroleum ether-diethyl ether-acetic acid 70:30:I(V/V/V), and visualization was effected by exposing the plate to iodine vapor. The collected cells were lyophilized and the cellular lipids were quantitatively extracted according to Folch. The lipids remaining in the broth were extracted three times with hexane after acidification of the broth with an equal volume of 4 mol/L HCI. The two fractions were esteritied and the methyl esters produced were analyzed in duplicate by GLC (Aggelis et al. 1995c).

RESULTS A N D DISCUSSION

Shaker cultures. The growth of C. lipolytica and L. gigantea on the organic medium was satisfactory (production of dry mass: 7.8 and 10.6 g/L, respectively) and significant amounts of lipids were accumulated in the microbial cells (21.4 and 40.4 %, fipids per dry mass for C. lipolytica and L. gigantea, respectively). The biomass yield was significantly lower with the other microorganisms studied. Table L Fatty acid composition of the Evening Primrose Oil, cellular lipids and residual medium lipidsa

Source of oilb

16:0

16:1 Shaker

Evening Primrose

17:0

18:0

18:1

18:2

18:3

A/mol

cultures

6.5

0

0

1.3

10.0

71.0

10.4

1.83

Rhodotorula sp.

C R

30.1 7.3

0 0

2.0 tr

4.5 1.6

35.8 9.5

25.4 72.4

tr 8.3

0.87 1.71

Candida tropicalis

C R

32.6 8.2

0 0

8.8 0

10.1 1.4

24.1 10.3

16.3 69-5

tr 10.4

0.57 1.81

C. lipotytica

C R

8.0 7.1

2.1 0

0_5 0

0.7 1.6

12.5 10.5

75.7 70.9

tr 9.7

1.66 1.81

C. cremoris

C R

38.9 7.3

8.8 0

11.4 0

11.0 1.5

18.4 10.5

7.7 72.7

tr 7.7

0.43 1.79

Langermania gigantea

C R

6.7 9.0

0 0

tr 0

1.3 2.2

8.8 10.1

73.9 71.5

6.9 6.9

1.77 1.74

Chemostat Evening Primrose

Candida lipo~tica

culture 6.7

C R

22.2 7.6

(carbon-limited

medium)

tr

tr

1.8

10.8

69.2

7.4

1.71

0.7 0

2.4 tr

8.6 2.3

26.3 13.2

6.7 66.1

tr 6.5

0.40 1.65

atr -- traces. b c -- cellular, R -- residual.

The produced cellular and residual culture oils exhibited a degree of unsaturation (A/mol) lower than that of the EPO (Table I). The absence of y-linolenic acid in cellular lipids was noted in all

1997

MICROBIAL FATYY ACID SPECIFICITY

119

strains but L. gigantea. C. tropicalis and C. cremotis produced significant quantities of margaric acid (C17:0). On the other hand, only C. lipolytica and C. cremoris produced palmitoleic acid (C16:1). The concentration coefficient CFA was used for characterizing the microbial fatty acid specificity: CFA = ([FA] - [FAIEPO)/[FAIEPo where [FA] represents the concentration (%) of the particular fatty acid in the culture residual or cellular lipid, and [FA]EPO represents its concentration in the EPO. When a fatty acid is preferentially incorporated in the microbial cell, the CFA in the residual lipid is negative. When the fatty acid is preferentially accumulated, the CFA in the cellular lipid is positive. The strains preferentially incorporated unsaturated fatty acids with the exception of Rhodotorula spp. which incorporated oleic acid more effectivelly than linoleic acid, and C. tropicalis which incorporated linoleic more than y-linolenic acid (Table II). On the other hand, the strains preferentially incorporated palmitic more than stearic acid. The only exception was the yeast C. tropicalis. C. tropicalis and C. cremoris preferentially accumulated saturated fatty acids, and more specifically stearic acid. Rhodotorula spp. accumulated fatty acids in the order C16:0 > C18:1 > C18:0 > C18:2 > C18:3, while C. lipolytica in the order C18:1 > C16:0 > C18:2 > C18:0 > C18:3. Finally, L. gigantea preferentially accumulated linoleic than oleic acid. These results show that it is possible to modifij an oil by using selected microbial cultures. Table II. Concentration coefficient (CFA) values of each fatty acid in the cellular (C) and residual medium (R) lipids

Source of oil

CC16:0

CC18:0

CC18:1

CC18:2

CC18:3

Rhodotorula sp.

C R

3.63 0.12

2.46 0.23

2.58 -0.05

-0.64 0.02

-1 -0.2

Candida tropicalis

C R

4.02 0.26

6.77 0.08

1.41 0.03

-0.77 -0.02

-1 0

C. lipo~ytica

C R

0.23 0.09

-0.46 0.23

0.25 0.05

0.07 0

-1 -0.07

C. cremoris

C R

5 0.12

7.46 0.15

0.84 0.05

-0.89 0.02

-1 -0.26

Langermania gigantea

C R

0.03 0.38

0 0.69

-0.12 0.01

0.04 0.01

-0.34 -0.34

Chemostat culture. Modification of the initial oil composition was more pronounced in a carbon-limited culture medium. C. lipolytica, cultivated in a chemostat on EPO at D = 0.2/h, produced 0.028 g L -1 h -1 of cellular oil and 0.380 gL -1 h -1 of residual culture oil, with a specific productivity 0.036 and 0.487 gg-1 h-l, respectively. The productivity and specific productivity were excellent with a regard to the oil remaining in the culture medium, which should be considered as an exocellular product. The cellular and the residual oils produced had a lower degree of unsaturation than the EPO (Table I). The microbial cells accumulated significant quantities of palmitic and oleic acids and contained only 6.7 % linoleic acid in their cellular lipid. Fig. 1 describes the effect of the number of double bonds on the CFA coefficient. A linear relationship between the CFA values and the number of double bonds of the C18 fatty acids was observed. The microbial population preferentially incorporated polyunsaturated fatty acids. The less unsaturated fatty acids were preferentially accumulated in the microbial cell, whereas the oligounsaturated ones were used for growth needs. Another strain of C. lipolytica (Glatz et al. 1984), cultivated on vegetable oils rich in 0t-linolenic acid, tends to lower the concentration of this acid in the cellular oil. On the other hand, C. lipolytica strain YB 423-12 was found to have a significant A9-desaturase activity for accumulated palmitic and stearic acids, while An-eicosenoic and erucic acids were shortened to oleic (Montet et al. 1985).

120

G. AGGELIS et al.

This work demonstrated that it is possible to produce less unsaturated fats or to change the fatty acid composition in an oil by using selected microbial cultures. Such microbial specificity depends on culture conditions (Ratledge and Boulton 1985). This might be particularly interesting for production of new fats, e.g. with low 0t-linolenic acid content, or for production of some animal or vegetable fat substitutes. The authors wish to thank Prof. Y. Clonis for critically reviewing the manuscript.

Vol. 42

0.3

c~, 0.2

0

0.1 0 -0.1 -0.2 t. 3

,

I

I

I

I

0

!

2

2 1 0 -1 -2

II

Fig. 1. Effect of the number of double bonds of the C18 fatty acids on the CFA values in the residual broth (top) and cellular (bouom) lipids of Candida lipotytica after culture on a carbonlimited mineral medium; n -- number of double bonds.

REFERENCES AGGELISG., KoMAms M., DIMITROULIASG., PINA M., GRAILLEJ.: Possibilite de production d'acide gamma linolenique par culture de Mucor circinelloides CBS 172-27 sur quelques huiles vegetales. Oleagineux 46, 208-212 (1991a). AGGEIaS G., BALATSOURASG., KOMArrlS M., ANAGNOSTOPOULOUG., DIMITROULIASG., PINA M., GRAILLE J.: Production d'acide gamma linolenique par bioconversion de i'acide linoleique de quelques huiles vegetales. Rev.Franc.Corps Gras 38, 95-101 (1991b). AGGELISG., KOMAITISi . , PINAi . , GRAILLEJ.: Specificity of Mucor meihei lipase on methyl ester substrates. Grasas y Aceites 44, 331-334 (1993). AGGELISG., MAMALAKISG., KOMAITISM.: Fatty acid specificity (typospecificity) of some microbial lipases. Riv.ltal.Sostanze Grasse 72, 211-213 (1995a). AC,GELlSG., KOMAmS M., PAPANICOLAOUS., PAPADOPOULOSG.: A mathematical model for the study of lipid accumulation in oleaginous microorganisms. I. Lipid accumulation during growth of Mucor circinelloides CBS 172-27 on a vegetable oil. GrasasyAceites 46, 169-173 (1995b). AC~ELtS G., KOMAmS M., PAPA~ICOLAOUS., PAPADOPOULOSG.: A mathematical model for the study of lipid accumulation in oleaginous microorganisms. II. Study of cellular lipids of Mucor circinelloides during growth on a vegetable oil. Grasas y Aceites 46, 245-250 (1995c). BUTTOND.IC: Nutrient-limited microbial growth kinetics: Overview and recent advances. Antonie van Leeuwenhoek 63, 225-235

(1993). GLATZB.A., HAMMONDE.G., Hsu K.H., BAEHMANL., BATIN., BEDNARSKIW., BROWND., FLOETENMEYERM.: Production and modification of fats and oils by yeast fermentation, pp. 163-175 in Biotechnology for the Oils and Fats Industry (C. Ratledge, P. Dawson, J. Rattray, Eds). American Oil Chemists' Society, New York 1984. KORITALAS., HESSELTINEC.W., PRYDE E.H., MOUNTS,T.L.: Biochemical modification of fats by microorganisms: A primary survey. JAm.Oil Chern.Soc. 64, 509-513 (1987). MACRAEA.IL: Microbial lipases as catalysts for the interesterification of oils and fats, pp. 189-197 in Biotechnologyfor the Oils and Fats Industry (C. Ratledge, P. Dawson, J. Rattray, Eds). American Oil Chemists' Society, New York 1984. MOtCrETD., RATOMAHEN[NAR., GALZYP., PIN^ M., GRAILLEJ.: A study of the influence of the growth media on the fatty acid composition in Candida lipo~ytica DIDDENSand LODDER.Biotechnol.Let~. 10, 733-736 (1985). RATLEDGE C., BOULTOr~ C.A.: Fats and oils, pp. 984-1003 in Comprehensive Biotechnology (M. Moo-Young, Ed.), Vol. 3 (C.L. Cooney, A.E. Humphrey, vol. Eds). Pergamon Press, Oxford 1985. SAMELISJ., AGGELISG., MLzrAxotaaULOSJ.: Lipolytic and microbial changes during the natural fermentation and ripening of greek dry sausages. Meat Science 35, 371-385 (1993).