Lipase-catalyzed alcoholysis of triglycerides for short-chain monoglyceride production

Article abstract of DOI:10.1007/s11746-006-0936-1

Lipase from Pseudomonas fluorescens efficiently catalyzed the alcoholysis of various TG in dry alcohols. For TG with short-chain FA, more MG were accumulated. The yields of MG were affected by the alcohols used. The maximum yields of MG were as follows: 8

Full text of DOI:10.1007/s11746-006-0936-1

Lipase-Catalyzed Alcoholysis of Triglycerides
for Short-Chain Monoglyceride Production
Guan-Chiun Leea, Dong-Lin Wanga, Yi-Fang Hob, and Jei-Fu Shawa,*
aInstitute of Botany, Academia Sinica, Nankang, Taipei, Taiwan 11529, and bInstitute of Bioscience
and Biotechnology, National Taiwan Ocean University, Keelung, 20224, Taiwan
ABSTRACT: Lipase from Pseudomonas fluorescens efficiently these publications, several additional works on the alcoholysis
catalyzed the alcoholysis of various TG in dry alcohols. For TG
with short-chain FA, more MG were accumulated. The yields of
MG were affected by the alcohols used. The maximum yields
of MG were as follows: 85% for monoacetin in n-butanol, 96%
for monobutyrin in ethanol or n-butanol, 50% for monocaprylin
in n-butanol, 48% for monolaurin in isopropanol, and 45% for
monopalmitin in isopropanol. The MG produced were judged
to be 2-MG by TLC analysis. The presence of organic cosolvent
affected the reaction rate of the lipase-catalyzed alcoholysis of
TG. For the alcoholysis of various TG in ethanol and cosolvent
(1:1, vol/vol), the rates had the following orders: (i) for tribu-
of oils and long-chain TG were reported (8,13,14). Lipase pro-
duced by the bacterium P. fluorescens is used extensively for a
large number of different synthetic reactions in organic chem-
istry, with special emphasis on the chiral production of
racemic compounds (15). The P. fluorescens lipases have been
shown to have broad substrate specificity towards TG and
preferential regioselective hydrolysis of the sn-1 position of a
TG (16,17). In the present work, we further report that the
Pseudomonas lipase can catalyze the alcoholysis of short-
chain TG in lower alcohols such as ethanol and isopropanol,
tyrin, hexane > toluene > acetone > ethyl acetate > chloroform which could be useful for the production of short-chain MG.
> acetonitrile > pyridine; (ii) for tricaprylin, hexane > acetone >
toluene > acetonitrile > ethyl acetate > pyridine > chloroform;
and (iii) for trilaurin, hexane > acetonitrile = acetone > ethyl ac-
etate > pyridine = chloroform > toluene.
The yields of MG are dependent on reaction conditions such
as the chain length of TG and alcohols, the organic solvents,
and the reaction times.
Paper no. J10744 in JAOCS 81, 533–536 (June 2004).
MATERIALS AND METHODS
KEY WORDS: Alcoholysis, lipase, monoglycerides, triglyc-
Reagents. Lipase from P. fluorescens was a generous gift
from Amano Pharmaceutical Co. Ltd. (Nagoya, Japan) and
had a specific activity of 3.2 Amano units (AU) per milligram
of solid. TG (triacetin, tributyrin, tricaproin, tricaprylin, tri-
laurin, and tripalmitin) were obtained from Sigma Chemical
Co. (St. Louis, MO), and their purities were all 99% by GC.
Celite 545 was purchased from Hayashi Pure Chemical In-
dustries (Osaka, Japan). All other biochemicals and chemi-
cals were of reagent grade.
Lipase-catalyzed alcoholysis. Lipase-catalyzed alcoholy-
sis was carried out as described previously (9–11). Typically,
the reactions were initiated by the addition of 0.3 g Celite-im-
mobilized lipase to 2 mL of a 25-mM TG solution in alcohols
predried with 3A molecular sieves (Merck Chemical Co.,
Darmstadt, Germany). The mixture was placed in a stoppered
glass vial and shaken in an orbital shaker incubator at 250
rpm and 28°C. The lipase was immobilized as follows: 40 g
of Celite 545 washed with l0 mM potassium phosphate buffer
(pH 7.0) was mixed with 20 g of Pseudomonas lipase (Amano
P), which was previously dissolved in 120 mL of 10-mM
phosphate buffer (pH 7.0) and stirred at 4°C. A fivefold vol-
ume excess of cold (−20°C) acetone was added to precipitate
the enzyme on the Celite surface. The immobilized lipase was
filtered and dried under vacuum at room temperature.
erides.
MG are useful as food flavors, pharmaceuticals, fragrances,
and emulsifiers (1,2). MG of different chain lengths have differ-
ent properties, e.g., HLB values, and therefore possess different
biological functions and have various industrial applications (3).
Conventional processes employing the direct esterification of
FA with glycerol catalyzed by inorganic catalyst require high
temperatures (200–250°C), which not only waste energy but
also have undesirable side reactions. Lipases can be used as
biocatalysts for the esterification process, and they have many
advantages over the chemical process, namely, mild reaction
conditions, high catalytic efficiency, and stereo- and positional
specificities (2,4–8). The lipase-catalyzed production of MG
has been shown to be quite successful with long-chain FA
(9,10). However, it may not be applicable to the synthesis of
short-chain MG owing to the relatively high acidity of short-
chain FA. In a previous work (11), we discovered that As-
pergillus niger lipase efficiently catalyzed the alcoholysis of
glucose pentaacetate to prepare glucose tetraacetate in lower
alcohols. Later, we discovered that Pseudomonas fluorescens
lipase could also be used for the alcoholysis of olive oils and
other long-chain TG for fatty ester production (12). Following
TLC. Positional specificity of the enzyme was examined by
TLC of the products of the enzymatic reaction with pure TG
(Sigma) as substrates (18). Aliquots of the reaction mixtures
*To whom correspondence should be addressed.
E-mail: boplshaw@gate.sinica.edu.tw
Copyright © 2004 by AOCS Press
533
JAOCS, Vol. 81, no. 6 (2004)

534
G.-C. LEE ET AL.
TABLE 2
were applied on a Silica gel 60 plate (Merck) and developed
with a solvent mixture of chloroform, acetone, and acetic acid
(96:4:1, by vol). Pure 1(3)-monocaprylin, 1(3)-monolaurin, 1,2-
dicaprylin, 1,3-dilaurin, tricaprylin, and trilaurin (Sigma) were
used as reference glycerides. Spots were visualized by heating
at 180°C for 25 min after spraying with 20% sulfuric acid.
GC. The reaction products were analyzed by GC as de-
scribed previously (11). One milliliter of sample was dried by a
rotary evaporator and derivatized with 1,1,1,3,3,3-hexamethyl-
disilazane according to the procedures of Brobst and Lott (19).
A 2-mL glass column packed with 3% OV-17 (Supelco, Belle-
fonte, PA) on 80/l00 Chromosorb WHP (Supelco) was used to
analyze the reaction products of triacetin, tributyrin, tricaprylin,
and trilaurin. The packing material for the analysis of tripalmitin
reaction products was 1% Dexsil 300 (Dexsil, Hamden, CT) on
100/120 Supelcoport (both from Supelco).
Effect of Solvent Polarity on the Lipase-Catalyzed Alcoholysis
of Tricaprylina
Product composition (%)
Solvent
Tricaprylin Dicaprylin Monocaprylin Glycerol
Acetonitrile
Acetone
Ethyl acetate
Pyridine
Chloroform
Toluene
Hexane
49
41
47
63
88
43
4
35
37
36
26
12
33
30
6
9
5
3
0
10
13
12
8
0
17
37
7
29
aThe initial tricaprylin concentration was 94 mM. The amount of Celite-im-
mobilized lipase (Amano P; Amano Pharmaceutical Co. Ltd.) was 0.1 g and
the reaction time was 30 min. Other conditions are the same as described in
Table 1.
TABLE 3
Effect of Solvent Polarity on the Lipase-Catalyzed Alcoholysis
of Trilaurina
RESULTS AND DISCUSSION
Product composition (%)
Effect of solvent polarity. It has been proposed that biocatalytic
activity in organic media correlates with the hydrophobicity
index, logP (20). However, the rate of TG alcoholysis cat-
alyzed by the Pseudomonas lipase did not correlate with logP.
Among the organic cosolvents tested, the velocity of the alco-
holysis reaction was increased only when hexane was added to
the reaction mixture (data not shown). The effect of the organic
cosolvent appeared to be different with different TG as sub-
strates. For the alcoholysis of various TG in ethanol and cosol-
vent (1:1, vol/vol), the rates of lipase-catalyzed alcoholysis de-
creased in the following order: (i) for tributyrin, hexane >
toluene > acetone > ethyl acetate > chloroform > acetonitrile >
pyridine (Table 1); (ii) for tricaprylin, hexane > acetone >
toluene > acetonitrile > ethyl acetate > pyridine > chloroform
(Table 2); (iii) for trilaurin, hexane > acetonitrile = acetone >
ethyl acetate > pyridine = chloroform > toluene (Table 3).
Effect of various substrates. In agreement with Sugihara et
al. (18), the results of the present TLC analysis (Fig. 1) showed
Solvent
Trilaurin
Dilaurin
Monolaurin
Glycerol
Acetonitrile
Acetone
Ethyl acetate
Pyridine
Chloroform
Toluene
Hexane
10
18
38
60
71
65
12
36
33
30
19
12
17
31
20
20
8
7
7
34
29
24
14
10
12
31
6
26
aThe initial trilaurin concentration was 25 mM. The amount of Celite-immo-
bilized lipase (Amano P; Amano Pharmaceutical Co. Ltd.) used was 0.1 g
and the reaction time was 30 min. Other conditions are the same as de-
scribed in Table 1.
TABLE 1
Effect of Solvent Polarity on the Lipase-Catalyzed Alcoholysis
of Tributyrina
Product compositionc (%)
Solvent
LogPb
Tributyrin
Dibutyrin Monobutyrin
Acetonitrile
Acetone
Ethyl acetate
Pyridine
Chloroform
Toluene
Hexane
−0.33
−0.23
0.68
0.71
2.00
2.50
3.50
55
46
49
69
52
30
5
38
42
43
28
41
55
48
7
11
8
2
7
15
47
aCelite-immobilized Pseudomonas lipase (0.3 g Amano P; Amano Pharma-
ceutical Co. Ltd., Nagoya, Japan) was incubated with 0.16 mM tributyrin in
2 mL of organic solvent/ethanol mixture (l:l) in a stoppered glass vial and
shaken on an orbital shaker at 250 rpm and 28°C for 1 h.
FIG. 1. TLC plate showing the products of Pseudomonas fluorescens li-
pase-catalyzed alcoholysis in ethanol. The TG substrates were tributyrin
(lane 1), tricaproin (lane 2), tricaprylin (lane 3), and trilaurin (lane 4),
respectively. Abbreviations: 1-Mono-CA, 1(3)-monocaprylin; 1-Mono-
LA, 1(3)-monolaurin; 1,2-Di-CA, 1,2-dicaprylin; 1,3-Di-LA, 1,3-dilau-
rin; Tri-CA, tricaprylin; Tri-LA, trilaurin.
bLogP is defined as the logarithm of the partition coefficient in a standard
octanol/water two-phase system.
cThe product composition was analyzed by a Hitachi 263-30 gas chromato-
graph (Tokyo, Japan) equipped with a 2-m glass column packed with 3%
OV-17 (Supelco, Bellefonte, PA) on 80/l00 Chromosorb WHP (Supelco).
JAOCS, Vol. 81, no. 6 (2004)

LIPASE-CATALYZED ALCOHOLYSIS OF TG
535
that 1,3-DG and 1(3)-MG migrated faster than 1,2(2,3)-DG and
2-MG, respectively. The 1,2- and 2,3-DG were not distinguish-
able by TLC. The MG produced by the P. fluorescens lipase
were likely 2-MG, and the DG were probably 1,2(2,3)-DG. As
shown in Table 4, the maximal yield of 2-MG depended on the
TG and alcohols used in the lipase-catalyzed alcoholysis of TG.
For TG with short-chain FA, more MG were accumulated. The
maximal yields of 2-MG were as follows: 85% for monoacetin
in n-butanol, 96% for monobutyrin in ethanol, 50% for mono-
caprylin in n-butanol, 48% for monolaurin in isopropanol, and
45% for monopalmitin in isopropanol.
From the product concentration vs. reaction time profiles
(Figs. 2–4), it is clear that the alcoholysis reaction followed a
sequential kinetic process:
FIG. 3. The concentration vs. reaction time profile for P. fluorescens li-
pase-catalyzed alcoholysis of trilaurin in isopropanol. The reaction con-
ditions are the same as described in Table l. (◆) Isopropyl laurate; (●)
trilaurin; (●) dilaurin; (◆) monolaurin; (■) glycerol.
k1
k2
k3
TG → DG → MG → glycerol
[1]
where k1, k2, and k3 represent the rate constants of the sequen-
tial alcoholysis step. The acyl group was transferred to the re-
spective alcohols used, and led to the production of FA esters isopropyl palmitate (Fig. 4). Figure 2 shows that tributyrin was
such as ethyl butyrate (Fig. 2), isopropyl laurate (Fig. 3), and completely converted into dibutyrin and monobutyrin in 2 h
in the presence of 0.3 g Celite-immobilized lipase at 28°C.
After 12 h almost all of the dibutyrin was converted into
monobutyrin, but no significant amount of glycerol appeared.
TABLE 4
Effect of Various Alcohols on the Yield of MG in the Lipase-Catalyzed
Alcoholysis of Various TGa
We concluded that the alcoholysis rate of monobutyrin (k3)
was very slow compared with k1 and k2. Therefore, a high con-
centration of monobutyrin can be accumulated. However, for
TG with medium- and long-chain FA such as trilaurin and tri-
palmitin, glycerol appeared before the TG was completely al-
coholyzed (Figs. 3 and 4). The alcoholysis rate of MG (k3) ap-
peared to increase as the chain length of FA in MG increased.
Therefore, the alcoholysis of TG catalyzed by P. fluorescens
lipase was the most suitable for the production of MG with
short-chain FA such as monoacetin and monobutyrin. Since
by-products such as ethyl acetate and ethyl butyrate are low-
b.p. compounds, they are easily separated from the MG.
The catalytic efficiency and substrate specificity of reactions
of this type are also presumably affected by the source of the li-
pase. Although lipases share a similar catalytic triad Ser-
Asp/Glu-His (similar to serine proteases), their overall sequence
homologies are usually quite low; therefore, their substrate bind-
ing sites are quite different. No single lipase can be the best for
the synthesis of MG of all different chain lengths. For the syn-
thesis of a particular MG, lipases from various sources should
be tested to find the best one for the desired application.
Maximal yield of 2-MG (%)b
Substrates
Ethanol
Isopropanol n-Butanol Amyl alcohol
Triacetin
30 (3.3)c
96 (8.0)
23 (8.0)
10 (0.3)
38 (2.0)
50 (2.6)
90 (4.0)
48 (7.0)
45 (0.3)
48 (0.3)
45 (2.0)
85 (3.3)
95 (8.0)
65 (3.0)
50 (0.4)
35 (0.3)
75 (2.0)
92 (4.0)
10 (8.0)
42 (0.3)
35 (0.3)
Tributyrin
Tricaproin
Tricaprylin
Trilaurin
Tripalmitin
aCelite-immobilized lipase (0.3 g Amano P; Amano Pharmaceutical Co. Ltd.)
was incubated with 25 mM of various TG in various alcohols (2 mL) in stop-
pered glass vials and shaken on an orbital shaker at 250 rpm and 28°C for
various time intervals; the product composition was then assayed by GC.
bThe yield of MG = (MG)/[(TG) + (DG) + (MG) + (glycerol)].
cThe numbers in parentheses denote the reaction time (hours) to reach the
maximal production of MG under the specified conditions.
In the present work, we discovered that the P. fluorescens li-
pase is excellent for the production of monoacetin and
monobutyrin, with 85 and 96% yields, respectively. However,
yields of other MG were not as high; therefore, other lipases
should be tested for the best production yields. In a number of
cases, organisms produced lipase isoforms that exhibited quite
different substrate specificities (21). Crude enzyme prepara-
tions are used in most applications, and lipases obtained from
different suppliers have been shown to exhibit variations in cat-
alytic efficiency and stereospecificity (22). We have discovered
multiple enzyme forms with different substrate specificities and
FIG. 2. The concentration vs. reaction time profile for the P. fluorescens
lipase-catalyzed alcoholysis of tributyrin in ethanol. The reaction con-
ditions are the same as described in Table l. (◆) Ethyl butyrate; (●) tri-
butyrin; (●) dibutyrin; (◆) monobutyrin; (■) glycerol.
JAOCS, Vol. 81, no. 6 (2004)

536
G.-C. LEE ET AL.
8. Millqvist, A., P. Adlercreutz, and B. Mattiasson, Lipase-Catalyzed
Alcoholysis of Triglycerides for the Preparation of 2-Monoglyc-
erides, Enzyme Microb. Technol. 16:1042–1047 (1994).
9. Irimescu, R., K. Furihata, K. Hata, Y. Iwasaki, and T. Yamane,
Utilization of Reaction Medium-Dependent Regiospecificity of
Candida antarctica Lipase (Novozym 435) for the Synthesis of
1,3-Dicapryloyl-2-docosahexaenoyl (or eicosapentaenoyl) Glyc-
erol, J. Am. Oil Chem. Soc. 78:285–289 (2001).
10. Irimescu, R., K. Furihata, K. Hata, Y. Iwasaki, and T. Yamane,
Two-Step Enzymatic Synthesis of DHA-rich Symmetrically
Structured Triacylglycerols via 2-Monoacylglycerols, Ibid.
78:473–478 (2001).
11. Shaw, J.F., and E.T. Liaw, Preparation of Acyl Derivatives of l-
Hydroxy Aldose by Lipase-Catalyzed Hydrolysis or Alcoholy-
sis of Fully Acylated Aldose in Organic Solvent, in Biocatalysis
in Organic Media, edited by C. Laane, J. Tramper, and M.D.
Lilly, Elsevier, Amsterdam, 1987, pp. 233–239.
12. Shaw, J.F., D.L. Wang, and Y.J. Wang, Lipase-Catalysed
Ethanolysis and Isopropanolysis of Triglycerides with Long-
Chain Fatty Acids, Enzyme Microb. Technol. 13:544–546 (1991).
13. Dossat, V., D. Combes, and A. Marty, Efficient Lipase Catalysed
Production of a Lubricant and Surfactant Formulation Using a
Continuous Solvent-Free Process, J. Biotechnol. 97:117–124
(2002).
14. Vacek, M., M. Zarevucka, Z. Wimmer, K. Stransky, M. Mack-
ova, and K. Demnerova, Enzymatic Alcoholysis of Blackcur-
rant Oil, Biotechnol. Lett. 23:27–32 (2001).
15. Coffen, D.L., Enzyme-Catalyzed Reactions, in Chiral Separa-
tions: Applications and Technology, edited by S. Ahuja, Ameri-
can Chemical Society, Washington, DC, 1997, pp. 59–91.
16. Gupta, R., N. Gupta, and P. Rathi, Bacterial Lipases: An
Overview of Production, Purification and Biochemical Proper-
ties, Appl. Microbiol. Biotechnol. DOI: 10.1007/s00253-004-
1568-8 (2004).
FIG. 4. The concentration vs. reaction time profile for the P. fluorescens
lipase-catalyzed alcoholysis of tripalmitin in isopropanol. The reaction
conditions are the same as described in Table l. (◆) Isopropyl palmi-
tate; (●) tripalmitin; (●) dipalmitin; (◆) monopalmitin; (■) glycerol.
thermostabilities in a commercial Candida rugosa lipase prepa-
ration (23). Recently, we demonstrated that five C. rugosa li-
pase genes are differentially expressed in the presence of dif-
ferent inducers (24). Traditionally, the culture conditions are
optimized for the maximal production of enzyme activity units.
Our results indicate that different culture conditions might re-
sult in heterogeneous compositions of isozymes, which display
different catalytic activities and specificities. By engineering
the culture conditions, selected enzymes can be enriched for
particular biotechnological applications.
Many factors affect the lipase-catalyzed synthesis of MG, in-
cluding enzyme structure, immobilization methods (25), reac-
tion media (organic solvents and supercritical fluid), tempera-
ture, pH (or pH memory in organic solvents), water activity (and
logP in organic media), acyl donors, and receptors. To obtain
the best yield of FA esters, all these factors should be optimized.
17. Rogalska, E., C. Cudrey, F. Ferrato, and R. Verger, Stereoselec-
tive Hydrolysis of Triglycerides by Animal and Microbial Li-
pases, Chirality 5:24–30 (1993).
18. Sugihara, Y. Shimada, and Y. Tominaga, A Novel Geotrichum
candidum Lipase with Some Preference for the 2-Position on a
Triglyceride Molecule, Appl. Microbiol. Biotechnol. 35:738–740
(1991).
19. Brobst, K.M., and C.E. Lott, Determination of Some Carbohy-
drates in Corn Syrups by Gas Chromatography of Trimethylsi-
lyl Derivatives, Cereal Chem. 43:35–42 (1966).
20. Laane, C., S. Boeren, R. Hilhorst, and C. Veeger, Optimization
of Biocatalysis in Organic Media, in Biocatalysis in Organic
Media, edited by C. Laane, J. Tramper, and M.D. Lilly, Else-
vier, Amsterdam, 1987, pp. 65–84.
REFERENCES
1. Arcos, J.A., and C. Otero, Enzyme, Medium, and Reaction En-
gineering to Design a Low-Cost, Selective Production Method
for Mono- and Dioleoylglycerols, J. Am. Oil Chem. Soc.
73:673–682 (1996).
2. Bornscheuer, U.T., Lipase-Catalyzed Syntheses of Monoacyl-
glycerols, Enzyme Microb. Technol. 17:578–586 (1995).
3. Stutz, R.L., A.J. del Vecchio, R.J. Tenney, and C.J. Patterson,
The Role of Emulsifiers and Dough Conditioners in Foods, Food
Prod. Dev. 7:52–60 (1973).
4. Elfman Borjesson, I., and M. Harrod, Synthesis of Monoglyc-
erides by Glycerolysis of Rapeseed Oil Using Immobilized Li-
pase, J. Am. Oil Chem. Soc. 76:701–707 (1999).
5. Fan, H.L., Y. Chu, G.X. Yang, W. Zhang, J.L. Liu, Z.S. Wu,
S.G. Cao, and D.L. You, Lipase-Catalyzed Syntheses of Mono-
glycerides by Hydrolysis of Soybean Oil in AOT/Isooctane Re-
versed Micelles, Ann. N.Y. Acad. Sci. 864:267–272 (1998).
6. Hoq, M.M., T. Yamane, S. Shimizu, T. Funada, and S. Ishida,
Continuous Synthesis of Glycerides by Lipase in a Microporous
Membrane Bioreactor, J. Am. Oil Chem. Soc. 61:776–781 (1984).
7. Langone, M.A., M.E. De Abreu, M.J. Rezende, and G.L. Sant-
Anna, Jr., Enzymatic Synthesis of Medium Chain Monoglyc-
erides in a Solvent-Free System, Appl. Biochem. Biotechnol.
98–100:987–996 (2002).
21. Chang, R.C., S.J. Chou, and J.F. Shaw, Multiple Forms and Func-
tions of Candida rugosa Lipase, Biotechnol. Appl. Biochem.
19:93–97 (1994).
22. Barton, M.J., J.P. Hamman, K.C. Fichter, and G.J. Calton, En-
zymatic Resolution of (R,S)-2-(4-hydroxyphenoxy) Propionic
Acid, Enzyme Microb. Technol. 12:577–583 (1990).
23. Shaw, J.F., C.H. Chang, and Y.J. Wang, Characterization of Three
Distinct Forms of Lipolytic Enzyme in a Commercial Candida Li-
pase Preparation, Biotechnol. Lett. 11:779–784 (1989).
24. Lee, G.C., S.J. Tang, K.H. Sun, and J.F. Shaw, Analysis of the
Gene Family Encoding Lipases in Candida rugosa by Competi-
tive Reverse Transcription-PCR, Appl. Environ. Microbiol.
65:3888–3895 (1999).
25. Shaw, J.F., R.C. Chang, F.F. Wang, and Y.J. Wang, Lipolytic
Activities of a Lipase Immobilized on Six Selected Supporting
Materials, Biotechnol. Bioeng. 35:132–137 (1990).
[Received October 24, 2003; accepted April 12, 2004]
JAOCS, Vol. 81, no. 6 (2004)