Characterization of Alcohol Acyltransferase from Olive Fruit

Article abstract of DOI:10.1021/jf035433a

Alcohol acyltransferase catalyzes the esterification of volatile alcohols with acyl-CoA derivatives to produce volatile esters typically present in the aroma of some fruits. This enzyme was detected in extracts from the pericarp tissues of ripe olive fruits using hexanol and acetyl-CoA as the substrates. Alcohol acyltransferase showed a very low activity level in these fruits, with an optimum pH value at 7.5 and high K_m values for hexanol and acetyl-CoA. The substrate specificity of this enzyme for various alcohols was also studied. The involvement of the studied enzyme in the biogenesis of the volatile esters present in the aroma of virgin olive oil was discussed.

Full text of DOI:10.1021/jf035433a

J. Agric. Food Chem. 2004, 52, 3155−3158 3155
Characterization of Alcohol Acyltransferase from Olive Fruit
JOAQUIÄN J. SALAS*
Instituto de la Grasa, CSIC, Av. Padre Garc´ıa Tejero, 4, 41012, Sevilla, Spain
Alcohol acyltransferase catalyzes the esterification of volatile alcohols with acyl-CoA derivatives to
produce volatile esters typically present in the aroma of some fruits. This enzyme was detected in
extracts from the pericarp tissues of ripe olive fruits using hexanol and acetyl-CoA as the substrates.
Alcohol acyltransferase showed a very low activity level in these fruits, with an optimum pH value at
7.5 and high Km values for hexanol and acetyl-CoA. The substrate specificity of this enzyme for
various alcohols was also studied. The involvement of the studied enzyme in the biogenesis of the
volatile esters present in the aroma of virgin olive oil was discussed.
KEYWORDS: Olea europaea; Oleaceae; olive; fruit; oil aroma; alcohol acyltransferase
INTRODUCTION
butanol, hexanol, and 3(Z)-hexenol, which are compounds with
high perception thresholds (13) and confer positive attributes
to the final oil product (sweet, green, and fruity notes).
Volatile esters are major components of the aroma of many
fruits such as strawberries, bananas, and apples, sometimes being
the main compound responsible for the flavor that the customers
appreciate in them (1, 2).
EXPERIMENTAL PROCEDURES
Plant Material. Olive (Olea europaea) fruits were harvested from
30-year-old trees located in an orchard near Sevilla (Spain). These trees
were supplied with drop irrigation and fertirrigation during the whole
reproductive period (April-November).
It is well-known that these compounds are synthesized by
the enzyme alcohol acyltransferase (AAT, E.C. 2.3.1.84), which
catalyzes the esterification of a volatile alcohol with an acyl-
CoA moiety to produce a volatile ester and free CoA-SH. The
volatile ester biosynthesis was first described in bananas (3, 4).
These studies were completed with in vivo experiments carried
out by Yamashita et al. (5) in strawberries, in which aldehydes
were found to be the precursors of some volatile esters. Ueda
and Ogata (6) reported the CoA dependence of AAT, which
was purified and characterized from banana fruits almost a
decade later (7). Several AATs from different sources have been
characterized since then, resulting in a high degree of correlation
between the AAT activity level and the organoleptic quality of
the products studied (8, 9). More recently, microarray techniques
have been used to determine the expression pattern of the AAT
gene in strawberry fruits (10). Alcohol acyltransferase protein
displayed homology toward some hypersensitivity proteins and
other acyltansferases from tobacco or Arabidopsis. Consensus
sequences from these enzymes allowed the cloning of two newer
AAT genes from cantaloupe fruits (11). The expression of these
AAT clones in microbial hosts permitted a complete charac-
terization of their substrate specificity that was in good
agreement with the volatile ester composition of these fruits.
However, in the case of apples and strawberries, there were
important discordances between substrate specificity and ester
composition that have been explained on the basis of substrate
availability (9, 12).
Preparation of Enzyme Extracts. The olive fruits were rinsed with
tap water and distilled water before endocarp removal. Then, an acetone
powder was prepared from the fleshy pulp as described by Salas and
Sa´nchez (14). This whitish powder was stored at -20 °C in flasks
endowed with silica gel desiccator. In these conditions the enzyme
activities were stable for several months. The enzyme extracts were
prepared by homogenization of 1.5 g of olive pulp acetone powder in
50 mL of 50 mM HEPES pH 7.5, 7 mM 2-mercaptoethanol, 0.05%
Triton X-100 using an ice-cooled glass homogenator. The resulting
suspension was centrifuged at 20000g for 10 min, the pellet was
discarded, and the supernatant was fractionated by ammonium sulfate
precipitation. The cut from 30 to 60% of ammonium sulfate saturation,
containing most of the extracted protein, was resuspended in 6 mL of
25 mM HEPES pH 7.5, 0.05% ascorbate, and 50% glycerol and was
assayed for AAT.
Enzyme Assay. The enzyme AAT was assayed by a modification
of the method reported by Fellman et al. (9) based on the determination
of the free CoA concomitantly produced with volatile esters by AAT.
The assay mixture contained 50 mM HEPES pH 7.5, 20 mM MgCl2,
20 mM hexanol, 0.4 mM acetyl-CoA, and enzyme extract equivalent
to 50-100 µg of protein in a final volume of 0.5 mL. This mixture
was incubated at 30 °C for 10 min, and then a volume of 25 µL 20
mM of 5,5-dithio-bis-nitrobenzoic acid (DTNB) dissolved in 0.1 M
HCO3Na was added. Once the color was developed, a process that takes
about 10 min, the absorbance at 407 nm of the assay mixture was
measured. AAT activity was calculated by comparing the absorbance
of samples containing hexanol and blanks with the same composition
without the alcohol substrate. The molar extinction coefficient of the
chromophorous compound formed by DTNB and CoA-SH at 407 nm
was experimentally determined as 10 400.
Among the different groups of volatile compounds present
in virgin olive oil, esters account only for 1 to 2% of the total.
This fraction mainly consists of acetyl derivatives of ethanol,
Synthesis of volatile esters was assessed by gas-liquid chromatog-
raphy by injections of the headspace of the assay vials after 30 min of
* E-mail: jjsalas@cica.es.
10.1021/jf035433a CCC: $27.50 © 2004 American Chemical Society
Published on Web 04/15/2004

3156 J. Agric. Food Chem., Vol. 52, No. 10, 2004
Salas
Figure 1. Optimization of the alcohol acyltransferase assay. (a) Influence
of loaded protein in the assay. (b) Influence of reaction time in the release
of CoA-SH using hexanol as substrate.
equilibration at 70 °C. A Carbowax 20 M (2 × 2000 mm) packed
column (Supelco) was used for volatile separation. Volatile esters were
identified by comparison of their retention times with those of the
corresponding standards.
Figure 2. Effect of pH in alcohol acyltransferase from olive fruit. Sodium
acetate (-O-), HEPES (-b-), and TRIS (-4-) were used as the assay
buffers.
Protein Determination. Protein was determined by a modification
of the method described by Bradford (15) using bovine gamma globulin
as the standard.
of unspecific acetyl-CoA esterases in the enzyme extract.
Therefore, a concentration of protein of 0.2 mg/mL in the
enzyme assay was used provided it yielded a good activity level
as well as low absorbance blanks. Furthermore, the release of
free CoA catalyzed by AAT in function of time displayed a
hyperbolic curve, with a linear interval embracing the first 10
min of reaction (Figure 1B), and therefore it was the incubation
time used in the standard AAT assay.
RESULTS AND DISCUSSION
Volatile esters confer positive attributes such as green, fruity,
and sweet notes to olive oil (16), and thus it is interesting to
promote their synthesis within the olive oil elaboration process.
Previous attempts to characterize olive AAT failed to detect
any activity in olive fruit pulp crude extracts. However, olive
pulp homogenates were able to produce volatile esters from
acetic acid and volatile alcohols or aldehydes, implying the
sequential action of the acetyl-CoA synthetase, alcohol dehy-
drogenase, and AAT (17). In the present work, the AAT activity
was detected and characterized in crude cell-free preparations
from olive fruit pulp. Because of the low level of activity in
the crude extracts, it was necessary to concentrate the protein
as described in the Experimental Section. The AAT activity level
detected in the extracts from the pulp of mature olives ranged
from 30 to 90 pkat/mg prot. The preparations carried out with
green unripe olives yielded inactive extracts, indicating that this
enzyme is induced in the latter period of the fruit development,
as it occurs in other fruits such as strawberries or cantaloupes
(11, 18). Moreover, despite olive AAT being a soluble enzyme,
the addition of Triton X-100 in the extraction medium was
necessary to obtain active fractions. This confirms that in olive
fruit, as in banana, Triton X-100 may protect or alter the enzyme
from a coagulated insoluble form into a soluble form that allows
its detection in the extracts (7). Furthermore, the low activity
level of AAT in olive pulp, which was much lower than in other
sources such as bananas or apples (7, 9), makes difficult its
study and purification. Thus, several attempts to purify this
enzyme failed because the activity was lost after its elution from
ion exchange or exclusion columns.
Optimization of the Enzyme Assay. The gas chromatogra-
phy-based method was not used routinely because it provided
less sensibility and higher variability than the DTNB-based one,
which was not affected by esterases that could be present in
the crude extract (1). Since that method involved end point
determinations, the assay conditions were optimized by studying
the influence of the protein concentration and the incubation
time on this activity. AAT increased linearly with protein
concentration from 0.5 to 1.0 mg/mL (Figure 1A). However,
an increase in the blank absorbance was observed as well (data
not shown), which hampered the assay at high enzyme extract
loads. This increment of absorbance could be due to the presence
Effect of pH. The pH curves reported in previous works for
AATs from several sources exhibited maximum activity values
in the alkaline range (7, 8, 11). In contrast, when the synthesis
of volatile esters was studied in olive pulp homogenates a
slightly acidic optimum pH value of 6.8 was found (17). The
olive AAT was assayed within a pH range from 5.5 to 9.5, using
the buffers sodium acetate, HEPES, and TRIS. The resulting
pH curve displayed a maximum at pH 7.5, with a quick drop
of activity at acidic pH values (Figure 2), being much more
similar to that showed by AATs from other fruits. The difference
of optimum pH values between both works carried out in olive
fruit could be due to the fact that in the first case the participation
of other enzymes such as acetyl-CoA synthetase and alcohol
dehydrogenase was necessary to produce the volatile esters.
Kinetic Parameters. AATs from other fruits showed Km
values for their substrate alcohols in the millimolar order.
Therefore, the Km value of banana fruit AAT was 0.4 mM for
isoamyl alcohol (7) and 3 mM for butanol in the case of
strawberry (8). These numbers were lower than that reported
for yeasts, from which Km values of 30 mM have been reported
for isoamyl alcohol (19). Moreover, the Km values of the CoA
derivative substrates were usually 10 to 100-fold lower. Thus,
they were in the micromolar range in bananas and strawberries
(7, 8), and it was 0.2 mM for acetyl CoA in the case of brewer
yeast (19). Recombinant His-tagged AATs isolated from
strawberry and cantaloupe, expressed in Escherichia coli or yeast
and purified by affinity chromatography, displayed the same
magnitude order in their kinetic parameters for both substrates
(8.9 mM and 0.1 mM in AAT from strawberry (10) and 1.4
mM and 0.09 mM in AAT1 from cantaloupe (11) for hexanol
and acetyl-CoA, respectively). When the kinetic parameters of
olive AAT for hexanol and acetyl-CoA were studied, hyperbolic
curves that fit well with the model of Michaelis-Menten were
obtained. However, the hexanol curve did not reach saturation
even at the highest concentrations used of this alcohol (Figure
3) and showed a Km value for this substrate of 21 mM (Table
1). The Km value for Acetyl-CoA was about 10-fold lower than
that corresponding to hexanol, indicating as expected a higher

Olive Alcohol Acyltransferase
J. Agric. Food Chem., Vol. 52, No. 10, 2004 3157
Table 2. Substrate Specificity of the Alcohol Acyltransferase from
Olive Fruit Pulpa
alcohol acyltransferase
(pkat/mg prot)
relative activity
(%)
substrate
methanol
ndb
0
0
ethanol
ndb
butanol
3-methylbuthanol
hexanol
3(Z)-hexenol
2(E)-hexenol
12 ± 6
12 ± 2
94 ± 20
80 ± 10
42 ± 2
16
16
100
86
45
b
a Data are the means of three determinations ± standard deviation. nd )
not detected
commonly detected in olive oil (16), could be synthesized
through an alternative pathway. Furthermore, the substrate
specificity of this enzyme would explain why the ester 2(E)-
hexenyl acetate was so scarce in olive fruit although the
concentration of its precursor was addressed to be the highest
among the volatile alcohols (16). On the other hand, this
specificity profile kept certain similitude to that reported for
strawberry fruits, in which butyl- and isoamyl- (3-methylbutyl-)
alcohols were poorer substrates than hexanol (8, 10). Recom-
binant AAT1 from cantaloupe displayed, similarly to the form
from olive, high activities toward all C6 alcohols assayed
combined with lower activities toward the C4 ones and barely
detectable esterification rates for ethanol (11).
Figure 3. Effect of hexanol concentration on the activity alcohol
acyltransferase. The inset shows the double reciprocal plot of the data,
which yielded a K value of 21 mM and a Vmax of 128 pkat/mg prot.
m
Table 1. Kinetic Parameters of Olive Alcohol Acyltransferase for
a
Hexanol and Acetyl-CoA
substrate
K (mM)
m
Vmax (pkat/mg prot)
R2
hexanol
acetyl-CoA
21
2
128
213
0.9949
0.9671
a Kinetic parameters for Acetyl-CoA were calculated using a hexanol concentra-
tion of 20 mM.
This specificity profile was in agreement with the in vivo
experiment carried out by Ol´ıas et al. (17), in which the higher
rates of ester formation by disrupted olive pulp tissues took place
with 3(Z)-hexenol, explaining the fact that 3-hexenyl acetate
was one of the most abundant volatile esters in olive oil.
Summarizing, the enzyme responsible of the volatile ester
formation, AAT, was detected and investigated in olive pulp.
The characteristics of this enzyme broadly justify the composi-
tion of volatile esters present in olive oil, being responsible for
the most positive fruity and sweet notes in the aroma of this
product. Data retrieved from this work also indicates that AAT
limits the production of volatile esters in olives provided its
activity is several times lower than the previous enzymes in
the pathway synthesizing C6 volatile compounds: lipoxygenase,
hydroperoxide lyase, and alcohol dehydrogenase (14, 20, 21).
This limitation led to the accumulation of the volatile alcohol
precursors, which display lower perception thresholds and
sometimes are responsible for grassy and rancid negative notes
(16). Therefore, the increment of AAT activity during the olive
oil elaboration process, which implies steps of milling, malax-
ation, and phase separation (22), would produce a double effect
of removal of alcohols and production of esters that would be
translated in an increase in the organoleptic quality of the final
oil. The base of these future improvements could imply either
the selection of olive cultivars with a higher AAT activity or
modifications in the extraction process such as operating at lower
temperatures to prevent enzyme inactivation or the neutralization
of the acidic pH values of the olive paste during malaxation to
promote a higher esterification activity.
affinity of this enzyme for its cofactor (Table 1). These Km
values were much higher than those reported for other fruits,
being more similar, in the case of hexanol, to the Km value
displayed by the brewers yeast for isoamyl alcohol (19). On
the other hand, the Km value of olive AAT for acetyl-CoA
(Table 1) was also higher than the reported Km values of AATs
from other sources. The high apparent Km values showed by
olive AAT for its substrates could also contribute to the low
volatile ester content of olive oil. However, it has to be
considered that the kinetic parameters reported in previous works
corresponded to total or partially purified enzymes, which could
enlarge the differences on apparent kinetic parameters when they
are compared with activities measured in nonpurified enzyme
extracts.
Substrate Specificity. The most abundant volatile alcohols
present in olive oil are the six-carbon ones synthesized through
the lipoxygenase pathway (16), by the successive action of the
enzymes lipoxygenase (20), hydroperoxide lyase (21), and
alcohol dehydrogenase (14). Thus, olive AAT was assayed with
a variety of alcohol substrates that ranged from one to six carbon
atoms, including alcohols typically present in the olive oil
volatile fraction such as hexanol, 3(Z)-hexenol, and 2(E)-hexenol
(Table 2). This enzyme did not display any activity with the
short-chained alcohols methanol and ethanol and showed low
activities toward butanol and 3-methylbutanol. The maximum
activities were obtained when hexanol and 3(Z)-hexenol were
used as the substrates, although the six-carbon alcohol 2(E)-
hexenol, the most abundant in olive oil (16), was a poorer
substrate than its homologous compounds. Olive AAT specific-
ity profile is in agreement with the volatile ester composition
of olive oil, and thus hexanol and 3(Z)-hexenol, which are the
precursors of the preponderant volatile esters of olive oil, were
the substrates more efficiently used by this enzyme. Further-
more, the short-chain alcohol ethanol was not a substrate of
this enzyme, which indicates that ethyl acetate, an ester
LITERATURE CITED
(1) Seymour, G. B. Biochemistry of Fruit Ripening; Taylor, J. E.,
Tucker, G. A., Eds.; Chapman and Hall: London, U.K., 1993.
(2) Sanz, C.; Ol´ıas, J. M.; Pe´rez, A. G. Aroma biochemistry of fruits
and vegetables. In Phytochemistry of Fruits and Vegetables;
Toma´s-Barbera´n, F. A., Robins, R. J., Eds.; Clarendon Press:
Oxford, 1997; pp 125-155.

3158 J. Agric. Food Chem., Vol. 52, No. 10, 2004
Salas
(3) Myers, M. J.; Issenberg, P.; Wick, E. L. L-leucine as a precursor
of isoamyl alcohol and isoamyl acetate, volatile aroma constitu-
ents of banana fruit discs. Phytochemistry 1970, 9, 1693-1700.
(4) Tressl, R.; Drawert, F. Biogenesis of banana volatiles. J. Agric.
Food Chem. 1973, 21, 560-565.
(5) Yamashita, I.; Nemoto, Y.; Yoshikawa, S. Formation of volatile
alcohols and esters from aldehydes in strawberries. Phytochem-
istry 1976, 15, 1633-1637.
(6) Ueda, Y.; Ogata, K. Coenzyme A-dependent esterification of
alcohols and acids in separated cells of banana pulp and its
homogenate. Nippon Shokuhin Kogyo Gakkaishi 1977, 24, 624-
630.
(7) Harada, M.; Ueda Y.; Iwata, T. Purification and some properties
of alcohol acetyltransferase from banana fruit. Plant Cell Physiol.
1985, 26, 1067-1074.
(8) Pe´rez, A. G.; Sanz, C.; Ol´ıas, J. M. Partial purification and some
properties of alcohol acyltransferase from strawberry fruits. J.
Agric. Food Chem. 1993, 41, 1462-1466.
(9) Fellman, J. K.; Mattinson, D. S.; Bostick, B. C.; Mattheis, J. P.;
Patterson, M. E. Ester biosynthesis in “Rome” apples subjected
to low-oxygen atmospheres. PostharVest Biol. Technol. 1993,
3, 201-214.
(10) Aharoni, A.; Keizer, L. C. P.; Boummeeseter, H. J.; Sun, Z.;
Alvarez-Huerta, M.; Verhoeven, H. A.; Blaas, J.; Van Hou-
welingen, A. M. M. L.; De Vos, R. C. H.; Van der Voet, H.;
Jansen, R. C.; Guis, M.; Mol, J.; Davis, R. W.; Schena, M.; Van
Tunen, A. J.; O’Connell, A. Identification of the SAAT gene
involved in strawberry flavor biogenesis by use of DNA
microarrays. Plant Cell 2000, 12, 647-661.
(11) Yahyaoui, F. E. L.; Wongs-Aree, C.; Latche´, A.; Hackett, R.;
Grierson, D.; Pech, J. C. Molecular and biochemical character-
istics of a gene encoding an alcohol acyl-transferase involved
in the generation of aroma volatile esters during melon ripening.
Eur. J. Biochem. 2002, 269, 2359-2366.
(14) Salas, J. J.; Sa´nchez, J. Alcohol dehydrogenases from olive (Olea
europaea) fruit. Phytochemistry 1998, 48, 35-40.
(15) Bradford, M. M. A rapid and sensitive method for the quantifica-
tion of microgram quantities of protein utilizing the principle of
protein-dye binding. Anal. Biochem. 1976, 72, 248-254.
(16) Morales, M. T.; Alonso, M. V.; Rios, J. J.; Aparicio, R. Virgin
olive oil aroma: relationship between volatile compounds and
sensory attributes by chemometrics. J. Agric. Food Chem. 1995,
43, 2925-2931.
(17) Ol´ıas, J. M.; Pe´rez, A. G.; R´ıos, J. J.; Sanz, L. C. Aroma of
virgin olive oil: biogenesis of the “green” odour notes. J. Agric.
Food Chem. 1993, 41, 2368-2373.
(18) Pe´rez, A. G.; Sanz, C.; Ol´ıas, R.; R´ıos, J. J.; Ol´ıas, J. M.
Evolution of strawberry alcohol acyltransferase activity during
fruit development and storage. J. Agric. Food Chem. 1996, 44,
3286-3290.
(19) Minetoki, T.; Bogaki, T.; Iwamatsu, A.; Fujii, T.; Hamachi, M.
The purification, properties and internal peptide sequences of
alcohol acetyltransferase isolated from Sacharomyceps cereVisiae
Kyokai N° 7. Biosci. Biotechnol. Biochem. 1993, 57, 2094-
2098.
(20) Salas, J. J.; Williams, M.; Harwood, J. L.; Sa´nchez, J. Lipoxy-
genase Activity in Olive (Olea europaea) Fruit. J. Am. Oil Chem.
Soc. 1999, 7, 1-6.
(21) Salas, J. J.; Sa´nchez, J. Hydroperoxide Lyase from Olive (Olea
europaea) Fruits. Plant Sci. 1999, 143, 19-26.
(22) Alba, J. Elaboracio´n de aceite de oliva virgen. In El CultiVo del
OliVo; Barranco, D., Ferna´ndez-Escobar, D., Rayo, L., Eds.;
Ediciones Mundi-Prensa: Madrid, Spain, 1997; pp 509-539.
Received for review December 5, 2003. Revised manuscript received
March 5, 2004. Accepted March 14, 2004. This work has been carried
out with the support of the E.U. (AIR 3-93-1967), which is gratefully
acknowledged.
(12) Pe´rez, A. G.; R´ıos, J. J.; Sanz, C.; Ol´ıas, J. M. Aroma components
and free amino acids in strawberry variety Chandler during
ripening. J. Agric. Food Chem. 1992, 40, 2232-2235.
(13) Hatanaka, A.; Kajiwara, T.; Horino, H.; Inokuchi, K. Odor-
structure relationships in n-hexenols and n-hexenals. Z. Natur-
forsch. 1992, 47, 183-189.
JF035433A