Conversion of oleic acid to 10-ketostearic acid by Sphingobacterium sp. strain O22

Article abstract of DOI:10.1007/s11746-999-0163-7

The conversion of oleic acid by a bacterium, tentatively identified as Sphingobacterium thalpophilum strain O22, was investigated. The microorganism was isolated as a stable culture from compost that was enriched with soybean oil outdoors and subsequently with oleic acid in the laboratory. Strain O22 converted oleic acid to products identified as 10-ketostearic acid (95% of the total conversion product) and 10-hydroxystearic acid (5%). This is in contrast to S. thalpophilum strain B-14797, which produces solely 10-hydroxystearic acid. Maximal conversion was reached in about 36 h after the addition of oleic acid to the fermentation broth. The yield of 10-ketostearic acid was approximately 75% from 0.26 g of oleic acid in 30 mL fermentation broth at 28 °C and 200 rpm for 48 h. This is the first report on the major production of 10-ketostearic acid by a microorganism in the genus Sphingobacterium.

Full text of DOI:10.1007/s11746-999-0163-7

Conversion of Oleic Acid to 10-Ketostearic Acid
by Sphingobacterium sp. Strain O22
T.M. Kuo*, A.C. Lanser, T. Kaneshiro, and C.T. Hou
Oil Chemical Research, NCAUR, ARS, USDA, Peoria, Illinois 61604
ABSTRACT: The conversion of oleic acid by a bacterium, ten- (10-KSA) is a major product of oleic acid bioconversion by a
tatively identified as Sphingobacterium thalpophilum strain
O22, was investigated. The microorganism was isolated as a
stable culture from compost that was enriched with soybean oil
outdoors and subsequently with oleic acid in the laboratory.
Strain O22 converted oleic acid to products identified as 10-ke-
tostearic acid (95% of the total conversion product) and 10-hy-
droxystearic acid (5%). This is in contrast to S. thalpophilum
strain B-14797, which produces solely 10-hydroxystearic acid.
Maximal conversion was reached in about 36 h after the addi-
tion of oleic acid to the fermentation broth. The yield of 10-ke-
tostearic acid was approximately 75% from 0.26 g of oleic acid
number of biological systems including a Corynebacterium
sp. (10), one Mycobacterium and two Nocardia (11), a Staphy-
lococcus sp. (12), and Flavobacterium sp. strain DS5 (13).
In our laboratory, we applied an enrichment culture tech-
nique to select stable microbial isolates able to effectively
convert soybean oil and fatty acids to value-added products.
The technique involved the use of composted plant and ani-
mal wastes that were mixed and enriched with soybean oil
and subsequently with oleic acid to enhance the propagation
of certain microbial species. In the study, we discovered that
in 30 mL fermentation broth at 28°C and 200 rpm for 48 h. This several isolates of Sphingobacterium sp. were able to convert
is the first report on the major production of 10-ketostearic acid
by a microorganism in the genus Sphingobacterium.
Paper no. J9033 in JAOCS 76, 709–712 (June 1999).
oleic acid predominantly to 10-KSA. In this paper, we de-
scribe the isolation and identification of a specific strain, O22,
and partial characterization of the bioconversion reaction in-
cluding the structure determination of products. This is the
first report on the predominant production of 10-KSA by a
microorganism in the genus Sphingobacterium.
KEY WORDS: Bioconversion, enrichment culture, 10-ke-
tostearic acid, oleic acid, Sphingobacterium thalpophilum.
Hydroxy- and keto-fatty acids are useful industrial chemicals EXPERIMENTAL PROCEDURES
used in plasticizer, surfactant, lubricant, and detergent formu-
Isolation of microorganism. Microorganisms were isolated
lation. They are often manufactured from fats and oils by
chemical reactions. Some of these products may be produced
by enzymes or whole microbial cells for greater specificity
and reduced environmental concerns.
from an aerobic compost after subculturing in an enrichment
medium (EM) and selection plates. The compost consisted
of a loose top soil, horse manure, and decaying garden
leaves–lawn grass mixture. Soybean oil (300 mL) was mixed
onto the top layer of this outdoor compost heap at regular in-
tervals, and samples for enrichment culture were taken from
the top layer after 3, 10, and 12 mon. The compost sample
(1.5–2.0 g) and oleic acid (0.5 mL) were added to 100 mL of
the enrichment medium (pH 7.3), which contained (per liter)
5 g glucose, 0.3 g yeast extract, 4.0 g K2HPO4, 0.5 g
MgSO4·7H2O, 15 mg FeSO4·7H2O, and 1 mL trace minerals
(5). The inoculated broths were incubated aerobically
overnight on a Gump rotary shaker (orbital radius, 5.08 cm;
New Brunswick Scientific, Edison, NJ) at 200 rpm, 28°C, and
readjusted to pH ≥ 7.0 with dilute NaOH. After the fourth day,
this regimen was repeated once or twice in fresh oleic acid-
containing EM (2% inoculum) before dilution-plating onto a
selective agar medium. The plates contained EM nutrient
with (per liter) 17 g agar, 25 mg bromo-cresol green indica-
tor, and 2.0 mL oleic acid (5). Discrete colonies were ran-
domly selected to test their bioreactivity. All cultures were
maintained on tryptone-glucose-yeast extract (TGY) (14)
Oleic acid is one of the major fatty acid components of corn
and soybean oils. Microbial conversion of oleic acid to form
value-added industrial products has been widely exploited.
Among the bioconversions of oleic acid to monohydroxy- or
keto-fatty acids, Wallen et al. (1) first reported that a
Pseudomonas sp. converted oleic acid to 10-hydroxystearic
acid (10-HSA) with a 14% yield. Similar bioconversions were
subsequently found with Rhodococcus rhodochrous (2), No-
cardia cholesterolicum (3), Micrococcus luteus and Sarcina
lutea (4), and by Sphingobacterium thalpophilum NRRL B-
14797 and Staphylococcus sp. NRRL B-14813 (5). Oleic acid
is converted to 15-,16-, and 17-hydroxy-9-octadecenoic acid
by Bacillus megaterium (6) and B. pumilus (7). It is also con-
verted to hydroxy- (8,9) and hydroperoxy-octadecenoic acid
(9) by Pseudomonas species. However, 10-ketostearic acid
*To whom correspondence should be addressed at NCAUR, ARS, USDA,
1815 N. University St., Peoria, IL 61604.
E-mail: kuotm@mail.ncaur.usda.gov
Copyright © 1999 by AOCS Press
709
JAOCS, Vol. 76, no. 6 (1999)

710
T.M. KUO ET AL.
agar slants and stored at 4°C. Identity of the bioreactive cul- resonance (NMR) and Fourier transform infrared (FTIR)
tures was established by Biolog System analysis using either spectroscopy. Electron impact mass spectra were obtained
a Gram-negative or Gram-positive data base (Biolog Inc., with a Hewlett-Packard (Palo Alto, CA) 5890 GC coupled to
Heyward, CA).
a Hewlett-Packard 5970 Series Mass Selective Dectector. The
Chemicals. Oleic acid (99+% purity) was purchased from column outlet was connected directly to the ion source. Sepa-
Nu-Chek-Prep, Inc. (Elysian, MN). 10-KSA used as a GC rations were effected in a methylsilicone capillary column (15
standard was obtained from a previous study (12). All other m × 0.25 mm i.d.) with a temperature gradient of 20°C/min
chemicals were reagent-grade and used without further purifi- to 200°C and then 10°C/min to 270°C after initially holding
cation. Thin-layer precoated Silica Gel 60 plates were ob- at 120°C for 2 min. Proton and 13C NMR spectra were deter-
tained from EM Separations Technology (Gibbstown, NJ).
mined in deuterated chloroform with a Bruker WM-300 spec-
Bioconversion assay. Cultures maintained on TGY slants trometer (Rheims Co., Tepten, Germany) operated at frequen-
were transferred to fresh TGY broth (pH 7.0) and incubated cies of 300 and 75.5 MHz, respectively. FTIR analyses of the
aerobically for 1-2 d before transfer (0.3 mL inoculum) into methyl ester of the product were done on a Perkin-Elmer
30-mL assay broths in 125 mL Erlenmeyer flasks. The assay (Oakbrook, IL) FTIR spectrum RX II system.
broths (pH 7.3) contained (per liter) 5.0 g yeast extract, 4.0 g
glucose, 4.0 g K2HPO4, 500 mg MgSO4·7H2O, and 15 mg
FeSO4·7H2O (1,15). Inoculated broths were shaken overnight
RESULTS AND DISCUSSION
for 20 h at 200 rpm and 28°C and checked for pH ≥ 7.0 before Enrichment culture selection. Commercial composted ma-
adding 0.3 mL oleic acid (0.26 g). The conversion reaction of nure of unknown origin yielded S. thalpophilum (NRRL B-
oleic acid by strain O22 was allowed for an additional 2–3 d 14797) and Acinetobacter (B-14920, B-14921, B-14923) cul-
under the same conditions. Lipids were recovered from the tures that converted oleic acid to 10-HSA and oleyl wax es-
acidified broth by extracting twice with an equal volume of ters, respectively (5,17). In our present study, a compost
methanol/ethyl acetate (1:9; v/v) (11,15). The solvent was then mixture amended with soybean oil yielded similar microor-
removed from the combined extracts with a rotary evaporator. ganisms after enrichment with oleic acid in the laboratory.
The concentrated lipid extracts were transferred to one-dram Nine of 74 randomly selected colonies from the agar plates
vials with methanol/chloroform (1:3; vol/vol) and ether and gave positive bioconversions to a mixture of 10-KSA and 10-
dried in an evacuated desiccator before weighing to obtain HSA identified by TLC and GC. Strain O22 was categorized
total lipid extracts. A portion of the extract was also dried im- by Biolog System analysis to be S. thalpophilum. The strain
mediately under a nitrogen stream for further analysis.
was chosen for further characterization because it was stable
Analytical procedures. Bioconversion was monitored by gas during maintenance, a fast grower, and a consistent 10-KSA
chromatography (GC) and thin-layer chromatography (TLC). producer.
Lipid extracts were esterified with diazomethane. The methyl
Structure determination. 10-KSA isolated by HPLC was
esters were injected into an HP (Hewlett Packard; Palo Alto, homogeneous on GC (data not shown). Its chemical structure
CA) model 5890 Series II gas chromatograph, equipped with a was confirmed by FTIR, GC–MS, and NMR analyses. FTIR
Supelco (Bellefonte, PA) SPB-1 capillary column (15 m × 0.32 spectroscopy of the methyl ester showed carbonyl bond ab-
mm, 0.25 µm film thickness), a flame-ionization detector and sorption at 1704 and at 1735 cm−1, indicating the presence of
an HP 7673 autosampler, and HP ChemStation software was saturated, acyclic ketone and ester groups, respectively (18).
used for data acquisition and integration. The GC conditions GC–MS analyses were performed both in electron impact
were set up as described previously (15). For quantitative ionization (EI) and chemical ionization (CI) (methane and
analysis, palmitic acid was added as an internal standard prior isobutane) modes. Methane CI provided evidence for molec-
to solvent extraction. A linear relationship was established for ular weight of 312 at m/z 281 (MH+ − 32, base peak), 313
the peak area ratios of products vs. methyl palmitate.
(MH+), 341 (M + 29) and 353 (M + 41). Isobutane CI con-
TLC analyses were carried out on Silica Gel 60 (0.25 mm firmed the proposed molecular weight with a base peak at m/z
thickness) plates (EM Science), developed in chloro- 313 (MH+) and a small ion at m/z 281. EI showed fragmenta-
form/methanol/acetic acid (9:1:0.1, vol/vol/vol). The chro- tion ions consistent with cleavage on each side of the car-
matograms were visualized first with sulfuric acid spray, bonyl (C9,10, m/z 141 and 171; C10, 11, m/z 113 and 199) (Fig.
followed by charring with a heat gun, and then with 1). The two even ions (m/z 156 and 214) were from β-cleav-
vanillin/sulfuric acid spray, followed by brief heating (16).
age with respect to the oxo-group in the chain, a shifting of
Isolation and identification of products. The bioconversion the double bond, and abstraction of a hydrogen atom (Fig. 1)
product was isolated by high-performance liquid chromatog- (19). These fragments determined the position of the oxo-
raphy (HPLC). HPLC was performed with a Spectra-Physics group to be at C-10. Proton and 13C NMR further confirmed
(San Jose, CA) SP8800 solvent delivery system and a Dyna- the identity of the main bioconversion product as being 10-
max 60-A silica column (25 cm × 21.4 mm i.d.; Rainin In- KSA. Resonance signals (ppm) and the corresponding mole-
strument, Woburn, MA) as described previously (12). The cular assignments are given in Table 1.
chemical structure of the purified product was confirmed by
One minor product of the bioconversion reaction (approxi-
GC–mass spectrometry (GC–MS) and by nuclear magnetic mately 5% of 10-KSA) having a GC retention time of 7.67 min,
JAOCS, Vol. 76, no. 6 (1999)

OLEIC ACID CONVERSION BY SPHINGOBACTERIUM
711
FIG. 1. Methyl ester and major mass spectral fragments of the conver-
sion product (10-ketostearic acid) from oleic acid by Sphingobacterium
sp. strain O22.
FIG. 2. Gas chromatograms of methyl esters recovered after the conver-
sion of oleic acid by Sphingobacterium sp. strain O22. Peak with reten-
tion time (RT) 2.76 min is internal standard; RT of 4.41 min is oleic acid;
RT of 7.47 min is 10-ketostearic acid; RT of 7.67 min is 10-hydroxy-
stearic acid.
as opposed to the 7.47 min of 10-KSA, 4.41 min of substrate
oleic acid, and 2.76 min of internal standard palmitic acid, was
also present (Fig. 2). GC–MS analysis of the methyl ester of
this minor product showed its molecular weight to be 314 with
a base peak at m/z 315 (MH+) and with prominent fragment
ions at m/z 201 (M − 113; C10,11 cleavage), 169 (201 − 32) and about 200 mg per 30 mL of media in about 36 h (Fig. 3). The
143 (M − 171; C9,10 cleavage). The data agree with those for maximal yield of 10-KSA production generally was obtained
10-HSA, determined previously (13,20). Therefore, S. in the range of 70–80%. The same product level was main-
thalpophilum strain O22, much like a Staphylococcus sp. (12) tained as the reaction time was extended to 4 d. This is indica-
and Flavobacterium sp. strain DS5 (13) discovered previously, tive that 10-KSA produced in the culture was not further me-
produces mainly 10-KSA from oleic acid. Similar bioconver- tabolized. The characteristic was also observed previously for
sion also occurs with Corynebacterium (10), Mycobacterium, the bioconversion as catalyzed by a Staphylococcus sp. (12)
and Nocardia strains (11). However, 10-HSA is the main prod- and Flavobacterium sp. strain DS5 (13).
uct in the bioconversion of oleic acid by Pseudomonas (1),
Rhodococcus rhodochrous (2), and Nocardia species (3).
The ability of microorganisms to convert oleic acid to ei-
ther 10-KSA or 10-HSA as a main product may strongly be
Bioconversion by strain O22. The maximal growth of related to the cellular regulation of both hydratase and sec-
strain O22 was reached at about 24 h after inoculation in 30 ondary alcohol dehydrogenase activities (13). In Nocardia
mL of culture medium. Therefore, the conversion reaction NRRL 5767, the product ratio of 10-KSA/10-HSA is influ-
was generally initiated by adding oleic acid substrate to the enced by different reaction conditions (3). Therefore, we
fermentation broth after 20–24 h of strain O22 cell growth. A compared the conversion reaction of strain O22 with the pre-
time course study on the production of 10-KSA from oleic viously characterized S. thalpophilum, B-14797, and Nocar-
acid by strain O22 indicated that the product increased lin- dia NRRL 5767 under the same conditions. The results, as
early up to 24 h of reaction and reached its maximal level of shown in Table 2, agree well with the previous report for
TABLE 1
Proton and 13C NMR Signals and Molecular Assignments for the
Methylated Bioconversion Producta
NMR type
Proton
Signal (ppm)
Hydrogen
Assignment
3.65
3
6
6
18
3
CH3, methyl ester
CH2, α to carbonyls
CH2, β to carbonyls
CH2, C4–7, C13–17
CH3, terminal
2.27–2.38
1.54–1.58
1.22–1.27
0.85–0.88
13C
211.56
C10
174.21
C1
51.36
C, methyl ester
42.70–42.78
34.02
C9,11
C2
31.76
C16
29.00–29.31
24.85
C4–7, C13–15
C3
23.77–23.84
22.57
14.00
C8,12
C17
C18
FIG. 3. Time course for the production of 10-ketostearic acid (10-KSA)
from oleic acid by Sphingobacterium sp. strain O22. Reaction condi-
tions: substrate oleic acid (0.3 mL) was added to a 24-h-old culture in
30-mL fermentation broths at 28°C and 200 rpm.
aNMR, nuclear magnetic resonance.
JAOCS, Vol. 76, no. 6 (1999)

712
T.M. KUO ET AL.
TABLE 2
and K. Soda, Transformation of Oleic Acid and Its Esters by
Comparison of Oleic Acid Conversion by Sphingobacterium
thalpophilum Strain O22 and Strain B-14797, and Nocardia NRRL
5767a
Sarcina lutea, Agric. Biol. Chem. 55:2651–2652 (1991).
5. Kaneshiro, T., L.K. Nakamura, and M.O. Bagby, Oleic Acid
Transformations by Selected Strains of Sphingobacterium
thalpophilum and Bacillus cereus from Composted Manure,
Curr. Microbiol. 31:62–67 (1995).
6. Miura, Y., and A.J. Fulco, (ω-2) Hydroxylation of Fatty Acids
by a Soluble System from Bacillus megaterium, J. Biol. Chem.
249:1880–1888 (1974).
7. Lanser, A.C., R.D. Plattner, and M.O. Bagby, Production of 15-,
16-, and 17-Hydroxy-9-octadecenoic Acid by Bioconversion of
Oleic Acid with Bacillus pumilus, J. Am. Oil Chem. Soc.
69:363–366 (1992).
8. Hou, C.T., and M.O. Bagby, 10-Hydroxy-8(Z)-Octadecenoic
Acid from Oleic Acid, An Intermediate in the Bioconversion of
Oleic Acid to 7,10-Hydroxy-8(E)-Octadecenoic Acid, J. Indust.
Microbiol. 9:103–107 (1992).
9. Guerrero, A., I. Casals, M. Busquets, Y. Leon, and A. Manresa,
Oxidation of Oleic Acid to (E)-10-Hydroperoxy-8-octadecenoic
and (E)-10-Hydroxy-8-octadecenoic Acids by Pseudomonas sp.
42A2, Biochem. Biophys. Acta 1347:75–81 (1997).
10. Seo, C.W., Y. Yamada, N. Takada, and H. Okada, Hydration of
Squalene and Oleic Acid by Corynebacterium sp. S-401, Agric.
Biol. Chem. 45:2025–2030 (1981).
11. El-Sharkawy, S.H., W. Yang, L. Dostal, and J.P.N. Rosazza,
Microbial Oxidation of Oleic Acid, Appl. Environ. Microbiol.
58:2116–2122 (1992).
12. Lanser, A.C., Conversion of Oleic Acid to 10-Ketostearic Acid
by a Staphylococcus Species, J. Am. Oil Chem. Soc. 70:543–545
(1993).
13. Hou, C.T., Production of 10-Ketostearic Acid from Oleic Acid
by Flavobacterium sp. Strain DS5 (NRRL B-14859), Appl. En-
viron. Microbiol. 60:3760–3763 (1994).
10-KSAb
(mg)
10-HSA Total product Total lipidc
Microbe
(mg)
(mg)
(mg)
Sphingobacterium
Strain O22
Strain B-14797
Nocardia
185
0
10
180
195
180
238
236
NRRL 5767
13
123
136
165
aOleic acid (0.26 g) was added to 24-h-old microbial culture in 30-mL fer-
mentation broth, and the bioconversion was allowed to proceed for 48 h at
28°C and 200 rpm prior to lipid extraction. (See Experimental Procedures
section). Each figure is the average of duplicate analyses.
bAbbreviations: 10-KSA, 10-ketostearic acid; 10-HSA, 10- hydroxystearic
acid.
cTotal lipid included 10-KSA, 10-HSA, oleic acid (substrate), and palmitic
acid (internal standard), as identified by gas chromatography.
NRRL 5767 (3) and confirm the observation that S.
thalpophilum strain B-14797 is a sole 10-HSA producer (20).
The result also demonstrates for the first time that a S.
thalpophilum strain O22 converts oleic acid predominantly to
10-KSA. Hydroxy- and keto-fatty acids are useful industrial
chemicals in the plasticizer, lubricant, and detergent formula-
tion. Investigations are in progress to optimize the culture and
reaction conditions for an increased production of 10-KSA
by strain O22, as well as 10-HSA by strain B-14797.
14. Difco Manual, 10th edn., Difco Laboratories, Detroit, MI, 1984,
pp. 679–680.
15. Kuo, T.M., L.K. Manthey, and C.T. Hou, Fatty Acid Bioconver-
sions by Pseudomonas aeruginosa PR3, J. Am. Oil Chem. Soc.
75:875–879 (1998).
16. Mangold, H.K., Aliphatic Lipids, in Thin-layer Chromatogra-
phy: A Laboratory Notebook, edited by E. Stahl, Academic
Press, New York, 1965, pp. 137–186.
17. Kaneshiro, T., L.K. Nakamura, J.J. Nicholson, and M.O. Bagby,
Oleyl Oleate and Homologous Wax Esters Synthesized Coordi-
nately from Oleic Acid by Acinetobacter and Coryneform
Strains, Curr. Microbiol. 32:336–342 (1996).
18. Dyer, J.R., Infrared Spectroscopy, in Applications of Absorption
Spectroscopy of Organic Compounds, Prentice-Hall, Inc., En-
glewood Cliffs, NJ, 1965, pp. 22–57.
19. Ryhage, R., and E. Stenhagen, Mass Spectrometric Studies. VI.
Methyl Esters of Normal Chain Oxo-, Hydroxy-, Methoxy-, and
Epoxy-acids, Arkiv. Kemi 15:545–574 (1960).
20. Kaneshiro, T., J.K. Huang, D. Weisleder, and M.O. Bagby,
10(R)-Hydroxystearic Acid Production by a Novel Microbe,
NRRL B-14797, Isolated from Compost, J. Indust. Microbiol.
13:351–355 (1994).
ACKNOWLEDGMENTS
We thank Tamy Leung, Sandra Su, Renee Tucker, and Richard Hsu
for technical assistance, Wanda Brown for cell growth study, and
Helen Gasdorf and Larry Nakamura for culture identification. We
also wish to thank Linda Manthey for TLC and preliminary GC–MS
analysis, and David Weisleder for NMR analysis.
REFERENCES
1. Wallen, L.L., R.G. Benedict, and R.W. Jackson, The Microbio-
logical Production of 10-Hydroxystearic Acid from Oleic Acid,
Arch. Biochem. Biophys. 99:249–253 (1962).
2. Litchfield, J.H., and G.E. Pierce, Microbiological Synthesis of
Hydroxy-fatty Acids and Keto-fatty Acids, U.S. Patent 4, 582,
804, 1986.
3. Koritala, S., L. Hosie, C.T. Hou, C.W. Hesseltine, and M.O.
Bagby, Microbial Conversion of Oleic Acid to 10-Hydroxy-
stearic Acid, Appl. Microbiol. Biotechnol. 32:299–304 (1989).
4. Blank, W., H. Takayanagi, T. Kido, F. Meussdoerffer, N. Esaki,
JAOCS, Vol. 76, no. 6 (1999)