A survey of the nature of glucose acylation reactions in plant extracts

Article abstract of DOI:10.1016/S0031-9422(99)00319-2

A variety of plant species have been shown to catalyse anomeric acyl exchange from a number of 1-O-fatty-acyl-β-glucoses as donor substrates to [¹⁴C]-glucose as an acceptor. The activity in wild tomato Lycopersicon pennellii has been analysed in detail by using analogs of glucose as acceptors and a number of 1-O-acyl-β-glucoses and 1-O-acyl-2-deoxyglucoses as acyl donors. Compared to 1-O-isobutyryl-β-glucose, the analogous 1-O- isobutyryl-β-2-deoxyglucose is an effective donor both to glucose (ca. 75%) and to 2-deoxyglucose (ca. 95%). On the contrary, compared to 1-O-isobutyryl- β-glucose, 1-O-isobutyryl-α-2-deoxyglucose is a poor donor both to glucose (ca. 4%) and to deoxyglucose (ca. 6%). The glucose analogs free at the anomeric center such as 3-O-methylglucose and 2-deoxyglucose are competent acyl acceptors from various 1-O-acyl-β-glucoses whereas 1-O-methylglucose is not. The primary initial product of acyl transfer from 1-O-isobutyryl-β- glucose to [¹⁴C]-glucose is β-glucosidase-cleavable 1-O-acyl-[¹⁴C]-β- glucose, whereas the isobutyryl-[¹⁴C]-3-O-methylglucose generated from 1-O- isobutyryl-β-glucose and [¹⁴C]-3-O-methylglucose is β-glucosidase- resistant. The transfer of the acyl group to 3-O-methylglucose occurs at the anomeric center: therefore the resistance to β-glucosidase reflects the strict specificity of β-glucosidase for glucose.

Full text of DOI:10.1016/S0031-9422(99)00319-2

Phytochemistry 52 (1999) 785±792
A survey of the nature of glucose acylation reactions in plant
extracts
Gurdev S. Ghangas*
Department of Plant Breeding and Biometry, Cornell University, Ithaca, NY 14853, USA
Received 29 September 1998; received in revised form 10 May 1999
Abstract
A variety of plant species have been shown to catalyse anomeric acyl exchange from a number of 1-O-fatty-acyl-b-glucoses as
donor substrates to [14C]-glucose as an acceptor. The activity in wild tomato Lycopersicon pennellii has been analysed in detail
by using analogs of glucose as acceptors and a number of 1-O-acyl-b-glucoses and 1-O-acyl-2-deoxyglucoses as acyl donors.
Compared to 1-O-isobutyryl-b-glucose, the analogous 1-O-isobutyryl-b-2-deoxyglucose is an e€ective donor both to glucose (ca.
75%) and to 2-deoxyglucose (ca. 95%). On the contrary, compared to 1-O-isobutyryl-b-glucose, 1-O-isobutyryl-a-2-deoxyglucose
is a poor donor both to glucose (ca. 4%) and to deoxyglucose (ca. 6%). The glucose analogs free at the anomeric center such as
3-O-methylglucose and 2-deoxyglucose are competent acyl acceptors from various 1-O-acyl-b-glucoses whereas 1-O-
methylglucose is not. The primary initial product of acyl transfer from 1-O-isobutyryl-b-glucose to [14C]-glucose is b-glucosidase-
cleavable 1-O-acyl-[14C]-b-glucose, whereas the isobutyryl-[14C]-3-O-methylglucose generated from 1-O-isobutyryl-b-glucose and
[14C]-3-O-methylglucose is b-glucosidase-resistant. The transfer of the acyl group to 3-O-methylglucose occurs at the anomeric
center; therefore the resistance to b-glucosidase re¯ects the strict speci®city of b-glucosidase for glucose. # 1999 Elsevier Science
Ltd. All rights reserved.
Keywords: Lycopersicon pennellii; Solanaceae; Acyl exchange; Triacylglucose biosynthesis; Acylglucose; Acylsugar; Isobutyroyl-glucose
1. Introduction
tomato are the subject of continuous studies primarily
because of their insect-deterring properties (Go€reda,
Ste€ens & Mutschler, 1990; Hawthorne, Shapiro,
Tingey & Mutschler, 1992; Juvik, Shapiro, Young &
Mutschler, 1994; Liedl et al., 1995; Rodriguez, Tingey
& Mutschler, 1993). Some of these recent studies have
focused on the biosynthesis of these acylsugars
(Ghangas & Ste€ens, 1993, 1995; Kuai, Ghangas &
Ste€ens, 1997; Li, 1998). Such studies are likely to lead
to economical and/or environment-friendly methods
for the production of acylsugars.
Because of their surfactant nature, fatty acid esters
of glucose and sucrose are widely used in food and
cosmetic industries as emulsi®ers and emollients.
Recently, Olestra, a sucrose polyester, was approved
by the US Food and Drug Administration as a cook-
ing-oil alternative. So far the production of these esters
has depended on partial or purely chemical processes.
Interestingly, similar esters of glucose and sucrose
(acylsugars) are produced by a variety of Solanaceous
plant species including wild tomato, Lycopersicon pen-
nellii (Burke, Goldsby & Mudd, 1987; King, Calhoun,
Singh & Boucher, 1993). The acylsugars of the wild
The wild tomato, L. pennellii LA716, contains a
unique fatty acid activation and transacylation mech-
anism implicated in the biosynthesis of its 2,3,4-triacy-
lacylglucoses (Ghangas & Ste€ens, 1993, 1995). The
activation reaction generates high-energy monoacylglu-
coses (1-O-acyl-b-glucoses) from UDPGlc and free
fatty acids. The energetics for the formation of similar
1-O-acyl-b-glucoses and the standard free energy of
* Corresponding author. Present address: The Pennsylvania State
University, Department of Horticulture, 103 Tyson Building,
University Park, PA 16802, USA.
E-mail address: gsg123@psu.edu (G.S. Ghangas)
0031-9422/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved.
PII: S0031-9422(99)00319-2

786
G.S. Ghangas / Phytochemistry 52 (1999) 785±792
Table 1
A survey for the presence of anomeric acyl exchange activities in plants: transfer of isobutyrate and laurate from 1-O-isobutyryl-b-D-glucose and
1-O-lauroyl-b-D-glucose, respectively, to [14C]-glucose by L. pennellii and other speciesa
pmol acyl groups incorporated/mg protein leading to
Plant
monoisobutyryl-[14C]-Glc
monolauroyl-[14C]-Glc
Lycopersicon pennellii
Lycopersicon esculentum
Solanum berthaultii
Nicotiana tabacum
Arabidopsis thaliana
Triticum aesticum
Phaseolus vulgaris
Cucumis melo
976
150
747
467
330
42
1088
26
374
282
115
16
611
115
282
23
a Reactions were carried out for 1 h as described in the text using 40 mM [14C]-glucose (300 mCi/mmol), 1.25 mg/mL leaf protein, 2 mM donor
in 50 mM NaPi pH 6.5 at 378C. The products were separated and monitored by TLC (Ghangas & Ste€ens, 1993).
their hydrolysis have recently been estimated (Mock &
Strack, 1993). Two soluble proteins from L. pennellii,
responsible for the activation reactions have been par-
tially puri®ed (Kuai et al., 1997). An acyltransferase
enzyme has also been puri®ed whose sequence is hom-
ologous to serine-type carboxypeptidases and contains
a catalytic triad (Asp±His±Ser) in the active site, simi-
lar to those found in a variety of serine proteases (Li,
1998). This acyltransferase catalyzes acyl exchange i.e.
the transfer of anomeric acyl moiety of the 1-O-acyl-b-
glucoses to the anomeric position of b-glucose
(Ghangas & Ste€ens, 1995). The enzyme is inhibited
by diisopropyl¯uorophosphate, an inhibitor of serine
proteases. Presumably, the acyltransferase forms an
Enz±Ser±acyl intermediate from which the acyl group
is transferred to a glucose molecule near the active site
of the enzyme. The same acyltransferase catalyzes the
transfer of the acyl group from a molecule of 1-O-
acyl-b-glucose to another 1-O-acyl-b-glucose molecule
resulting in the formation of a diacylglucose and
release of a glucose. This reaction is termed dispropor-
tionation (Ghangas & Ste€ens, 1995). While the acti-
vation and disproportionation reactions appear crucial
to the synthesis of polyacylated glucoses, the role of
acyl exchange in the synthesis of polyacylated glucoses
is not clear. In vitro, nevertheless, acyl exchange can
be used to prepare radiolabeled 1-O-acyl-b-glucoses
(Ghangas & Ste€ens, 1995; Denzel, Weisemann &
Gross, 1988).
In this communication the results of a study on acyl
exchange in various plants are reported, together with
the ®rst chemical synthesis of a- and b-1-O-isobutyryl-
Fig. 1. Stereocontrolled syntheses of a- and b-1-O-isobutyryl-2-deox-
yglucoses. (1) 3,4,6-Tri-O-benzyl-D-glucal. (2) 1-O-isobutyryl-2-
2-deoxyglucoses. Using these and other analogs, the
donor and acceptor substrates for acyl exchange and
disproportionation in L. pennellii (LA716) were ana-
lysed. During this work features of b-glucosidase speci-
®city were uncovered which might be useful in
preserving the group transfer potential of 1-O-acyl lin-
kages generated from UDPGlc and free fatty acids.
deoxy-3,4,6-tri-O-benzyl-a-glucopyranose.
(3)
1-O-isobutyryl-2-
deoxy-a-glucopyranose. (4) 2-deoxy-2-selenophenyl-3,4,6-tri-O-ben-
zyl-a-glucopyranose. (5) 1-O-isobutyryl-2-deoxy-2-selenophenyl-3,4,6-
tri-O-benzyl-b-glucopyranose. (6) 1-O-isobutyryl-2-deoxy-3,4,6-tri-O-
benzyl-b-glucopyranose. (7) 1-O-isobutyryl-2-deoxy-b-glucopyranose.
(a) Ph3PÁHBr, isobutyric acid. (b) Pd/C/H2, EtOAc. (c) PhSeCl. (d)
Na2CO3, H2O. (e) Ph3P, isobutyric acid, DEAD. (f) ACHN, Bu3Sn
reduction. (g) Pd/C/H2, EtOAc.

G.S. Ghangas / Phytochemistry 52 (1999) 785±792
787
Table 2
Synthesis of monoacyl glucoses and monoacyl-2-deoxyglucoses from acyl donor analogs and 14C-labeled sugar-acceptorsa
% of [14C]- monoacyl product generated from
14C]-Glc
Donor
[
[4C]-2-d-Glc
1-O-Isobutyryl-b-Glc
1-O-Isobutyryl-2-deoxy-b-Glc
1-O-Isobutyryl-2-deoxy-a-Glc
8.19
6.20
0.31
3.0
2.83
0.18
a Reactions were carried out as described in the text using 50 mM [14C]-glucose (or [14C]-2-deoxy-glucose), 2.5 mg/mL L. pennellii protein, 2.5
mM donors in 50 mM NaPi pH 6.5, 378C. The products were separated and monitored by TLC (Ghangas & Ste€ens, 1993).
2. Results and discussion
gave the pure a- precursor 2. Catalytic hydrogenolysis
resulted in pure 1-O-isobutyryl-2-deoxy-a-glucopyra-
nose 3 with an overall yield of 40%.
2.1. A survey of acyl exchange in various plants
Starting again with 3,4,6-tri-O-benzyl-D-glucal 1, the
synthesis of the b- precursor 6 and the ®nal product 7
are also shown in Fig. 1. In a one-pot procedure the
protected D-glucal was ®rst reacted with phenylselene-
nyl chloride followed by treatment with aqueous
Na2CO3. The reaction with phenylselenenyl chloride
generates a mixture of unstable anomeric chlorides.
Hydrolysis of the anomeric chlorides with aqueous
Na2CO3 generates an equilibrium mixture from which
the predominant a- isomer 4 was puri®ed by ¯ash
chromatography. Esteri®cation of a-OH of 4 with iso-
butyric acid using the Mitsunobu set of reagents
(Mitsunobu, 1981) resulted in invertion of con®gur-
ation at the anomeric carbon thus giving pure b-isomer
5. Tributyltin hydride reduction of 5 cleanly removed
the selenophenyl group and gave pure b-precursor 6.
Catalytic hydrogenolysis of 6 followed by ¯ash puri®-
cation gave 7 in an overall yield of 4% starting from
1.
Our previous work shows that acyl exchange is pre-
sent in wild tomato L. pennellii (LA716) that produces
acylglucoses. Measurable levels of acyl exchange were
also found in L. esculentum, a plant that does not pro-
duce such acylsugars (Ghangas & Ste€ens, 1995). The
new data show that acyl exchange from 1-O-isobu-
tyryl-b-glucose (also termed 1-O-isobutyroyl-b-glucose)
to glucose can be demonstrated in a wide variety of
plants (Table 1). Catalytic acyl transfer occurs also
from 1-O-lauroyl-b-glucose, although the ratios of iso-
butyryl and lauroyl transfer activities are not pro-
portional across the species investigated. The
di€erences in acyl transfer from 1-O-isobutyryl-b-glu-
cose vs. 1-O-lauroyl-b-glucose in various plants could
result from a single enzyme that shows chain-length
di€erences or from isozymes that prefer substrates
with di€erent chain length. However, other protein
and nonprotein factors could also modulate these ac-
tivities. It was also observed that activation activity re-
sponsible for the generation of 1-O-acyl-b-glucoses
from UDPGlc and free fatty acids is found in a wide
variety of plants (unpublished). These ®ndings suggest
multiple biological roles in plants for these activities.
As compared to the a-isomer 2 the b-isomer 6
underwent slow hydrogenolysis. When hydrogenolysis
was carried out in EtOH or MeOH instead of in
EtOAc, loss of the isobutyryl group was observed.
2.3. Acyl exchange from 1-O-isobutyryl-2-deoxy-a/b-
glucopyranoses
2.2. Chemical syntheses of 1-O-isobutyryl-2-deoxy-a/b-
glucopyranoses
The results in Table 2 show that catalytic acyl trans-
fer to free glucose and 2-deoxy-glucose occurs e-
ciently from 1-O-isobutyryl-b-glucose and 1-O-
isobutyryl-b-2-deoxy-glucose. These results are consist-
ent with the enzyme forming an active Enz±Ser±isobu-
tyryl intermediate both from 1-O-isobutyryl-b-glucose
and 1-O-isobutyryl-b-2-deoxy-glucose. It is interesting
to note that acyl transfer from 1-O-isobutyryl-b-2-
deoxy-glucose is more ecient to glucose than it is to
2-deoxy-glucose. Furthermore, the transfer from 1-O-
isobutyryl-b-2-deoxy-glucose is about 20 fold faster
than it is from 1-O-isobutyryl-a-2-deoxy-glucose.
In order to further probe the speci®city of acyl
exchange, the 1-O-isobutyryl-2-deoxy-a/b-glucoses
were synthesized from 3,4,6-tri-O-benzyl-D-glucal (Fig.
1). The synthetic strategy was designed to obtain the
fully benzylated a- and b-precursors 2 and 6 of the
®nal products 3 and 7 respectively (Fig. 1). Such pre-
cursors (2 and 6) are stable and yield the ®nal products
upon hydrogenolysis.
The synthesis of the a-precursor was carried out by
reacting 3,4,6-tri-O-benzyl-D-glucal 1 with isobutyric
acid in the presence of catalytic amounts of triphenyl-
phosphine
Mioskowski, Lee & Falck, 1990). Silica gel puri®cation
hydrobromide
(Ph3PÁHBr)
(Bolitt,
The acyl exchange from 1-O-isobutyryl-b-2-deoxy-
glucose to glucose has a novel synthetic utility as fol-

788
G.S. Ghangas / Phytochemistry 52 (1999) 785±792
Table 3
Synthesis of mono-, di- and triacyl glucoses from various acyl donors and 14C-labeled sugar-acceptors by L. pennellii leaf extractsa
% of [14C]-acceptor incorporated into
Donor
[
14C]-acceptor
monoacyl
diacyl
triacyl
1-O-Isobutyryl-b-Glc
1-O-Lauroyl-b-Glc
1-O-Palmitoyl-b-Glc
Glc
Glc
Glc
13.58
0.59
0.26
0.19
0.35
0.19
0.0
0.08
0.0
1-O-Isobutyryl-b-Glc
1-O-Lauroyl-b-Glc
1-O-Palmitoyl-b-Glc
2-d-Glc
2-d-Glc
2-d-Glc
6.52
1.42
0.61
0.0
0.0
0.0
0.0
0.35
0.34
1-O-Isobutyryl-b-Glc
1-O-Lauroyl-b-Glc
1-O-Palmitoyl-b-Glc
3-O-Me-Glc
3-O-Me-Glc
3-O-Me-Glc
11.17
0.91
0.19
0.15
0.19
0.13
0.0
0.0
0.0
a Reactions were carried out for 6 h as described in the text using 50 mM [14C]-glucose, 2.5 mg/mL protein, 2.5 mM donors in 50 mM NaPi
pH 6.5, 378C. The products were separated and monitored by TLC (Ghangas & Ste€ens, 1993).
lows. Since 1-O-isobutyryl-b-2-deoxy-glucose moves
faster on silica gel than 1-O-isobutyryl-b-glucose, the
availability of 1-O-isobutyryl-b-2-deoxy-glucose pro-
vides a new method for synthesizing 1-O-isobutyryl-
[14C]-b-glucose from [14C]-glucose and 1-O-isobutyryl-
b-2-deoxy-glucose. The chemical synthesis of 1-O-iso-
butyryl-[14C]-b-glucose of known and high speci®c ac-
tivity is expensive and time-consuming.
in such reactions. These results are consistent with the
NMR spectral data that show 1,2-diisobutyryl glucose
as a major product of 1-O-isobutyryl-b-glucose dispro-
portionation by the enzyme found in L. Pennellii (not
shown). Furthermore, on silica gel TLC plates the syn-
thetic 1,2-diisobutyryl-b-glucose comigrates with an
enzyme-generated diisobutyryl glucose from 1-O-isobu-
tyryl-b-glucose. Low concentrations (up to 1 mM) of
other sugars do not compete with anomeric transfer to
glucose from 1-O-isobutyryl-b-D-glucose. Such compe-
tition becomes apparent at 200 fold molar excess of
some other sugars, provided the anomeric center is
available (Table 4). Note that 1-O-methyl-glucose does
not compete even at this high level. This is consistent
with the observation that 1-O-methyl-[U-14C]glucose is
not an acceptor substrate in acyl exchange assays (data
not shown).
2.4. Acyl transfer from various 1-O-acyl-b-glucoses to
sugar acceptors
Isobutyryl-, lauroyl- and palmitoyl-groups are trans-
ferred from their respective 1-O-acyl-b-glucoses not
only to glucose but also to 2-deoxy-glucose and 3-O-
methyl-glucose (Table 3). This monoacylation is fol-
lowed by the formation of diacyl-glucose(s) and diacyl-
3-O-methyl-glucose(s), whereas the formation of diiso-
butyryl-2-deoxy-glucose(s) is characteristically absent
The formation of [14C]-diacylglucoses from 1-O-iso-
butyryl-b-glucose and [14C]-glucose (or [14C]-3-O-
methylglucose) is the result of disproportionations
(Table 3). The absence of such diacylglucoses from 1-
Table 4
Inhibition of monoisobutyryl-[14C]-glucose biosynthesis generated
O-isobutyryl-b-glucose
and
[14C]-2-deoxy-glucose
through catalytic acyl exchange from 1-O-isobutyryl-b-D-glucose and
suggests that the 2-OH group on glucose plays a major
role in disproportionation and thus in the enzyme-cat-
alyzed synthesis of higher-order acylated-glucoses.
Disproportionation enzymes involved in the biosyn-
thesis of other diacyl molecules have been identi®ed in
other plant systems (Dahlbender & Strack, 1984, 1986;
Kojima & Kondo, 1985; Schmidt, Denzel, Schilling &
[
14C]-glucose by the L. pennellii protein extracta
Competitor
% cpm in monoacylglucosea
None
+10 mM glucose
100
14
37
43
62
98
43
+10 mM 2-deoxy-glucose
+10 mM 3-O-methyl-glucose
+10 mM 1-thio-glucose
+10 mM 1-O-methyl-b-D-glucose
+10 mM galactose
Gross, 1987; Villegas, Shimokawa, Okuyama
&
Kojima, 1986). The activity from oak leaves is
involved in the disproportionation of b-glucogallin (1-
O-galloyl-b-glucose) for the biosynthesis of gallotan-
nins (Schmidt et al., 1987). However, in that case the
primary initial product is 1,6-di-galloyl-glucose and
not 1,2-di-galloyl-glucose.
a Reactions were carried out for 1 h as described in the text using
50 mM [14C]-glucose, 2.5 mg/mL protein, 2.5 mM donor in 50 mM
NaPi pH 6.5, 378C. The products were separated and monitored by
TLC (Ghangas & Ste€ens, 1993).

G.S. Ghangas / Phytochemistry 52 (1999) 785±792
789
2.5. Susceptibility of acyl exchange products to b-
glucosidase
the concentration of in vivo donor generated by feed-
ing 2 mM isobutyrate (in a total volume of 100 mL/g
leaf) is also very low (10±20 mM).
The initial primary product of acyl transfer from 1-
O-isobutyryl-b-glucose to [14C]-glucose (Table 3) is sus-
ceptible to almond b-glucosidase. The acyl product
generated from 3-O-methyl-glucose under the same
conditions is completely resistant to b-glucosidase (not
shown). Mass spectral analysis of this b-glucosidase-re-
sistant product showed it to be 1-O-isobutyryl-3-O-
methyl-glucose. The peaks at m/z 319 and m/z 71 con-
®rmed the identity of the isobutyryl moiety. The origin
of other peaks at m/z 303, m/z 242, m/z 200, m/z 169,
m/z 140, m/z 115, m/z 98 were all characteristic of 3-
We previously hypothesized a biological role for
acyl exchange in regenerating 1-O-acyl-b-glucoses
during the terminal steps in triacylglucose biosynthesis
(Ghangas & Ste€ens, 1995). The presence of acyl
exchange activity in a variety of plants that do not se-
crete such acylsugars now leads one to believe that the
reaction may have other functions as well. The acti-
vated 1-O-acyl-b-glucoses are perhaps mobilized as
diacylglucoses to escape b-glucosidase action. Thus dis-
proportionation and acyl-exchange could provide a
mechanism to preserve/regenerate these linkages. Yet
another explanation for the entire UDPGlc-dependent
pathway is to provide a detoxi®cation mechanism for
the plant cell against free fatty acids which are ®rst
converted to 1-O-acyl-b-glucoses and then to diacylglu-
coses.
O-methylglucopyranose structure (De Jongh
&
Bieman, 1963). The monoacyl-2-deoxy-glucose gener-
ated from [14C]-2-deoxy-glucose and 1-O-isobutyryl-b-
glucose is also susceptible to b-glucosidase although
the rate is somewhat slower as compared with the pro-
duct generated from [14C]-glucose and 1-O-isobutyryl-
b-glucose (data not shown).
The complete resistance of the 1-O-isobutyryl-3-O-
methyl-glucose to almond b-glucosidase is consistent
with previous observations that glucosidases usually
require an unsubstituted glucosidic moiety for activity
(Selmar, 1993; Esen, 1993). This would indicate that
1,n-diacyl-b-glucoses, such as generated through dis-
proportionation of 1-O-isobutyryl-b-glucose, should be
resistant to b-glucosidase action as well. Indeed, syn-
thetic 1,n-diisobutyryl-b-glucoses (1-O-[1-14C-isobutyr-
yl],n-isobutyryl-b-glucoses) are completely resistant to
b-glucosidase (data not shown).
3. Experimental
The reagents for the synthesis of 1-O-acyl-b-glucoses
have been previously described (Ghangas & Ste€ens,
1995). 3,4,6-tri-O-benzyl-D-glucal was purchased from
Aldrich. Dry solvents (methylene chloride, carbon tet-
rachloride, benzene and toluene) were used as pur-
chased from Aldrich. Ethyl acetate was shaken with
anhydrous potassium carbonate (2 g/100 mL), ®ltered
and redistilled from calcium hydride. Preparative silica
gel chromatography (Still, Kahn & Mitra, 1978) was
performed with silica gel 60 (230±400 mesh or 70±230
mesh, EM Science). Radiochemicals ([U-14C]glucose
and [U-14C]2-deoxy-glucose) were purchased from
American Radiolabeled Chemicals (St. Louis, MO).
Before use [U-14C]-2-deoxy-glucose was puri®ed by
TLC over silica gel using CHCl3/MeOH/H2O
(75:22:3). 1H-NMR spectra were recorded with a
Varian spectrometer operating at 200 MHz. GC-MS
(electron impact) was carried out with a DB-5 column
in a Hewlett-Packard 5890 gas chromatograph/5970
mass-selective detector operating at 70 eV (Ghangas &
Ste€ens, 1993). Melting points were determined on an
electrothermal melting point apparatus.
1-O-lauroyl-b-glucose was prepared from 2,3,4,6-
tetra-O-benzylglucose and lauroyl chloride in benzene
at 628C as previously described for the synthesis of 1-
O-isobutyryl-b-glucose (Ghangas & Ste€ens, 1995;
Pfe€er, Rothman & Moore, 1976). 1-O-[1-14C-isobu-
tyryl]-b-glucose was converted to 1,n-diisobutyryl-b-
glucose (1-O-[1-14C-isobutyryl],n-isobutyryl-b-glucose)
using isobutyryl-chloride and pyridine as previously
described for the synthesis of donor 1,n-dibutyryl-b-
glucose (Ghangas & Ste€ens, 1995). 1-O-methyl-a/b-
[14C]-glucose was synthesized from [14C]-glucose
2.6. Acyl exchange in vivo
Exogenous 1-O-isobutyryl-b-glucose is hydrolysed
during uptake by L. pennellii leaves and thus cannot
be used for donor studies in vivo. Therefore the nas-
cent donor 1-O-isobutyryl-b-glucose generated from
isobutyrate co-fed with the nonmetabolizable acceptor
sugars 3-O-methyl-glucose or 2-deoxy-glucose was
relied upon. In contrast to the results in vitro, no sig-
ni®cant formation of monoisobutyryl 2-deoxy-glucose
occurred in the wild tomato leaves fed with 2-deoxy-
glucose and isobutyric acid; what appeared to be a
weak formation of monoisobutyryl-3-O-methylglucose,
however, could be observed in such in vivo exper-
iments (not shown). Administration of [14C]-glucose to
leaves co-fed with isobutyrate was uninformative since
the label was distributed in a random manner. The
lack of appreciable in vivo transacylation from 1-O-
isobutyryl-b-glucose to 2-deoxy-glucose and 3-O-
methyl-glucose may be due to compartmental separ-
ation of the donor from the acceptors. Furthermore,
only a small portion of the isobutyrate fed to the
leaves is converted to 1-O-isobutyryl-b-glucose. Thus,

790
G.S. Ghangas / Phytochemistry 52 (1999) 785±792
according to (Fischer, 1895). 40 mg [14C]-glucose (300
mCi/mmol) was mixed with anhydrous methanol that
was made 2% with conc. HCl (1 mL) and kept at 60±
648C for 8 h. The solvent was then removed with a
stream of N2 and the residue puri®eded by silica gel
chromatography. The synthesis of a- and b- derivatives
of 1-O-isobutyryl-2-deoxy-glucose is detailed below
(Fig. 1). The bold-faced numbers correspond to the
structures in Fig. 1. A preliminary report of this work
at 208C was slowly added dropwise phenylselenenyl
chloride (PhSeCl) (1 g, 5.24 mmol dissolved in 4 mL
dry CCl4). The resulting brown±yellow solution was
stirred for 60 min at 208C and then CCl4 was
removed in vacuo. The resulting orange±brown oil was
dissolved in THF-H2O (20 mL, 1:1) and after adding
Na2CO3 (0.77 g) the mixture was stirred at room tem-
perature for 15 min and then brought to 508C for 4 h.
The mixture was cooled to room temperature, diluted
with H2O and extracted with ether (Et2O). The ether
extract was washed with saturated NaCl and dried
over anhydrous MgSO4. The solvent was removed in
vacuo and the residue puri®ed by a ¯ash column using
two petroleum ether±ethyl acetate solvents (4:1, 1:1).
This was followed by a second ¯ash column using hex-
ane±ethyl acetate (7:2). The product was further puri-
®ed by crystallizing from EtOH. Melting point
(uncorrected) 102±1038C. Yield 0.75 g (41%). Rf 0.33
[(petroleum ether±ethylacetate (4:1)]. 1H-NMR (200
MHz,CDCl3), d 7.56±7.64 (m, 2H; aromatic protons
ortho to Se), 7.1±7.4 (m, 18H; aromatic protons), 5.45
(d or dd, J = 2.9±3.17 Hz, 1H; a-anomeric), 5.0 (d,
J = 11.4 Hz, 1H), 4.8±4.9 (m, 2H), 4.45±4.65 (m,
3H),4.05±4.2 (m, 2H or 3H), 3.55±3.75 (m, 3H), 3.35
(dd, J = 10.86, 3.21 Hz, 1H)
was presented at a recent meeting (Ghangas
Ste€ens, 1996).
&
3.1. 1-O-isobutyryl-2-deoxy-3,4,6-tri-O-benzyl-a-
glucopyranose (2)
Isobutyric acid (0.7 mL, 7.5 mmol) was added to a
magnetically stirred solution of 3,4,6-tri-O-benzyl-D-
glucal (1) (1.04 g, 2.5 mmol) in anhydrous CH2Cl2 fol-
lowed by triphenylphosphine hydrobromide (Bolitt et
al., 1990) (Ph3PÁHBr)(43 mg, 0.125 mmol). After stir-
ring at ambient temperature for 2 h, the reaction mix-
ture was washed with saturated NaHCO3 and NaCl,
dried over anhydrous Na2SO4 and concentrated. The
residue (1.2 g, 99% yield) was puri®ed by two succes-
sive silica gel ¯ash columns using benzene±ethyl acet-
ate (19:1) followed by hexane±EtOAc (8:1). Yield was
1
0.63 g (52%). Rf 0.18 (hexane±EtOAc: 8:1). H-NMR
(200 MHz,CDCl3), d 7.15±7.4 (m, 15H; aromatic ring
protons), 6.25 (dd, J = 1.8,1.3 Hz, 1H; a-anomeric),
4.9 (d, J = 10.5 Hz, 1H), 4.45±4.7 (m, 5H), 3.6±4.0
(m, 5H), 2.5 (septet, J = 7 Hz, 1H; (CH3)2CH ), 2.25
(m, 1H), 1.85 (m, 1H), 1.19 (overlapping d, J = 7 Hz,
6H; (CH3)2CH)
3.4. 1-O-isobutyryl-2-deoxy-2-selenophenyl-3,4,6-tri-O-
benzyl-b-glucopyranose (5)
A
solution of 2-deoxy-2-selenophenyl-3,4,6-tri-O-
3.2. 1-O-isobutyryl-2-deoxy-a-glucopyranose (3)
benzyl-a-glucopyranose (4) (118 mg, 0.2 mmol), triphe-
nylphosphine (Ph3P) (78.7 mg, 0.3 mmol), isobutyric
acid (25 mL, 0.27 mmol) and 20 mg molecular sieves (4
A) in dry benzene (1 mL) was prepared. Diethyl azodi-
carboxylate (Mitsunobu, 1981; Smith, Hale & Rivero,
1986) (DEAD) (52 mL, 0.34 mmol) was slowly added
to this solution. The mixture was stirred at 08C for 30
min and then at room temperature for 3 h. The solvent
was removed in vacuo and the crude product was puri-
®ed by ¯ash chromatography over silica gel using pet-
roleum ether±ethyl acetate (4:1). A second round of
¯ash chromatography using hexanes±ethyl acetate
(10:1) was used to further purify product 5. Yield 61
mg (46%). Rf 0.86 [petroleum ether±EtOAc (4:1)], Rf
A
solution of 1-O-isobutyryl-2-deoxy-3,4,6-tri-O-
benzyl-a-glucopyranose (2) (60.6 mg, 0.12 mmol) in
anhydrous EtOAc was sparged with dry N2 and trea-
ted with palladium catalyst (Pd on activated carbon,
Pd 10%). The suspension was hydrogenated using a
Paar apparatus. The hydrogenated product was puri-
®ed by silica gel ¯ash chromatography using CHCl3±
MeOH (7:1). Yield was 23 mg (82%). Rf 0.36 [CHCl3±
1
MeOH (7:1)]. H-NMR (200 MHz, acetone-d6), d 6.12
(dd, J = 2.2, 1.3 Hz, 1H; a-anomeric), 3.55±3.95 (b,
4H), 3.4 (t, J = 9.1, 1H), 2.55 (sep, J = 7 Hz, 1H;
(CH3)2CH ), 2.05 (m, 1H), 1.70 (m, 1H), 1.13 (overlap-
ping d, J = 7 Hz, 6H; (CH3)2CH).
0.40
[(hexane±EtOAc(9:1)].
1H-NMR
(200
MHz,CDCl3), d 7.56±7.64 (m, 2H; aromatic protons
ortho to Se), 7.1±7.4 (m, 18H; aromatic protons), 5.72
(d, J = 9.4 Hz, 1H; b-anomer), 4.75±5.0 (m, 3H),
4.45±4.65 (m, 3H), 3.45±3.87 (m, 5H), 3.35 (dd,
J = 9.5, 9.4 Hz, 1H), 2.45 (septet, J = 7 Hz, 1H;
(CH3)2CH ), 1.20 (overlapping d, J = 7 Hz, 6H;
(CH3)2CH).
3.3. 2-deoxy-2-selenophenyl-3,4,6-tri-O-benzyl-a-
glucopyranose (Kaye, Neidle & Reese, 1988; Roush &
Lin, 1991) (4)
To a magnetically stirred solution of 3,4,6-tri-O-ben-
zyl-D-glucal (1) (1.28 g, 3.0 mmol) in dry CCl4 (10 mL)

G.S. Ghangas / Phytochemistry 52 (1999) 785±792
791
3.5. 1-O-isobutyryl-2-deoxy-3,4,6-tri-O-benzyl-b-
glucopyranose (6)
transfer of isobutyrate from the synthetic monoisobu-
tyryl donors to [U-14C]glucose, [U-14C]2-deoxy-glucose
or 3-O-methyl-[U-14C]glucose. Reaction mixes con-
tained 2.5 mM donor, 50 mM acceptor sugar (8 Â 104±
4.5 Â 105 cpm), 50 mg protein in 20 mL (total volume)
50 mM NaPi, pH 6.5. Reaction progress (378C) was
monitored by TLC (Ghangas & Ste€ens, 1993, 1995).
b-glucosidase treatments with Sigma sweet almond
enzyme were carried out as previously described
(Ghangas & Ste€ens, 1995).
A magnetically stirred solution of 1-O-isobutyryl-2-
deoxy-2-selenophenyl-3,4,6-tri-O-benzyl-b-glucopyra-
nose (5) (33 mg, 0.05 mmol) in anhydrous toluene (1
mL) and containing catalytic amount of 1,1'-azobis(cy-
clohexanecarbonitrile (Overberger, Biletch, Finestone,
Lilker & Herbert, 1953) (ACHN; Aldrich trade name
VAZO1) (1.22 mg, 5 mmol) under dry N2 gas was pre-
pared. Tributylltin hydride (Bu3SnH) (67.25 mL, 0.25
mmol) was added to this solution and the reaction stir-
red at 908C for 5 h. The solvent was then removed in
vacuo and the residue puri®ed by ¯ash chromatog-
raphy using hexane±ethyl acetate (9:1). Yield was 14
mg (68%). Rf 0.265 [(hexane±petroleum ether (10:1)],
Rf 0.187 (hexane±EtOAc (9:1). 1H-NMR (200 MHz,
CDCl3), d 7.15±7.4 (m, 15H; aromatic ring protons),
5.68 (dd, J = 2.2,10 Hz, 1H; b-anomeric), 4.88 (d,
J = 10.8 Hz, 1H), 4.48±4.74 (m, 5H), 3.45±3.8 (m,
5H), 2.6 (septet, J = 7 Hz, 1H; (CH3)2CH ), 2.35 (m,
1H), 1.75 (m, 1H), 1.21 (overlapping d, J = 7 Hz, 6H;
(CH3)2CH).
3.8. Acyl exchange in vivo
Leaves from L. pennellii (LA716) plants grown in
the greenhouse were washed with EtOH/H2O system
as previously described (Ghangas & Ste€ens, 1993).
After 1 day recovery, the leaves were administered
with various combinations of substrates and analysed
as previously described (Ghangas & Ste€ens, 1993).
Acknowledgements
Supported in part by the USDA-National Research
Initiative Competitive Grants program and the Cornell
Center for Advanced Technology in Biotechnology.
The author wishes to acknowledge the contribution
made to this work by John C. Ste€ens. I also thank
Manish Tandon, Tadhg P. Begley, David Collum and
Susan E. Blauth for helpful suggestions.
3.6. 1-O-isobutyryl-2-deoxy-b-glucopyranose (7)
A
solution of 1-O-isobutyryl-2-deoxy-3,4,6-tri-O-
benzyl-b-glucopyranose (6) (20.2 mg, 0.04 mmol) in
anhydrous EtOAc (10 mL) was ¯ushed with dry N2
and treated with palladium catalyst (Pd on activated
carbon, Pd 10%). The suspension was hydrogenated
using a Paar apparatus (25 psi). The hydrogenated
product was puri®ed by silica gel ¯ash chromatog-
raphy using CHCl3±MeOH (7:1). Rf 0.39 [CHCl3±
MeOH (7:1)]; Rf 0.60 [CHCl3±MeOH±H2O (75:22:3)].
References
1
Yield 2.9 mg (31%). H-NMR (200 MHz, acetone-d6),
Bolitt, V., Mioskowski, C., Lee, S.-G., & Falck, J. R. (1990). J. Org.
Chem., 55, 5812.
d 5.7 (dd, J = 10.1, 2.1 Hz, 1H; b-anomeric), 3.45±
3.85 (b, 4H), (t, 1H), 2.55 (sep, 1H; (CH3)2CH ), 2.1
(m, 1H), 1.55 (m, 1H), 1.14 (overlapping d, J = 7 Hz,
6H; (CH3)2CH).
Burke, B. A., Goldsby, G., & Mudd, J. B. (1987). Phytochemistry ,
26, 2567.
Dahlbender, B., & Strack, D. (1984). J. Plant Physiol., 116, 375.
Dahlbender, B., & Strack, D. (1986). Phytochemistry, 25, 1043.
De Jongh, D. C., & Bieman, K. (1963). J. Am. Chem. Soc., 85,
2289.
3.7. Enzyme reactions
Denzel, K., Weisemann, S., & Gross, G. G. (1988). J. Plant Physiol.,
133, 113.
Foliage samples of greenhouse-grown wheat
(Triticum aestivum ), bean (Phaseolus vulgaris) and
melon (Cucumis melo) were provided by M. E.
Sorrells, D. H. Wallace and M. Kyle (Cornell
University, Ithaca, NY) and were immediately frozen
after harvest. Lycopersicon pennellii (LA716),
Lycopersicon esculentum and Solanum berthaultii plants
were grown in the greenhouse. Arabidopsis thaliana
and Nicotiana tabacum plants were grown in the lab-
oratory. Leaf extracts were prepared, the proteins par-
tially puri®ed and quanti®ed as described (Ghangas &
Ste€ens, 1995). Anomeric acyl exchange (Ghangas &
Ste€ens, 1993, 1995) was analysed by measuring the
Esen, A. (1993). In A. Esen, b-glucosidases: biochemistry and molecu-
lar biology (p. 1). Washington, DC: American Chemical Society.
Fischer, E. (1895). Ber., 28, 1145.
Ghangas, G. S., & Ste€ens, J. C. (1993). Proc. Natl. Acad. Sci.,
USA, 90, 9911.
Ghangas, G. S., & Ste€ens, J. C. (1995). Arch. Biochem. Biophys.,
316, 370.
Ghangas, G. S., & Ste€ens, J. C. (1996). Glycobiology, 6 Abstract
No. 38.
Go€reda, J. C., Ste€ens, J. C., & Mutschler, M. A. (1990). J. Am.
Soc. Hort. Sci., 115, 161.
Hawthorne, D. J., Shapiro, J. A., Tingey, W. M., & Mutschler, M.
A. (1992). Entomol. Exp. Appl., 65, 65.
Juvik, J. A., Shapiro, J. A., Young, T. E., & Mutschler, M. A.
(1994). J. Econ. Entomol., 87, 482.

792
G.S. Ghangas / Phytochemistry 52 (1999) 785±792
Kaye, A., Neidle, S., & Reese, C. B. (1988). Tetrahedron Lett., 29,
2711.
Pfe€er, P. E., Rothman, E. S., & Moore, G. G. (1976). J. Org.
Chem., 41, 2925.
King, R. R., Calhoun, L. A., Singh, R. P., & Boucher, A. (1993). J.
Agric. Food Chem., 41, 469.
Rodriguez, A. E., Tingey, W. M., & Mutschler, M. A. (1993). J.
Econ. Entomol., 86, 34.
Kojima, M., & Kondo, T. (1985). Agric. Biol. Chem., 49, 2467.
Kuai, J.-P., Ghangas, G. S., & Ste€ens, J. C. (1997). Plant Physiol.,
115, 1581.
Roush, W. R., & Lin, X.-F. (1991). J. Org. Chem., 56, 5740.
Schmidt, S. W., Denzel, K., Schilling, G., & Gross, G. G. (1987). Z.
Naturforsch., 42C, 87.
Li, X. (1998). Ph.D. Dissertation, Cornell University, Ithaca, NY.
Liedl, B. E., Lawson, D. M., White, K. K., Shapiro, J. A., Cohen,
D. E., Carson, W. G., Trumble, J. T., & Mutschler, M. A.
(1995). J. Econ. Entomol., 88, 742.
Selmar, D. (1993). In A. Esen, b-glucosidases: biochemistry and mol-
ecular biology (p. 191). Washington, DC: American Chemical
Society.
Smith III, A. B., Hale, K. J., & Rivero, R. A. (1986). Tetrahedron
Lett., 27, 5813.
Mitsunobu, O. (1981). Synthesis, 1±28.
Mock, H.-P., & Strack, D. (1993). Phytochemistry, 32, 575.
Overberger, C. G., Biletch, H., Finestone, A. B., Lilker, J., &
Herbert, J. (1953). J. Am. Chem. Soc., 75, 2078.
Still, W. C., Kahn, M., & Mitra, A. (1978). J. Org. Chem., 43, 2923.
Villegas, R. J. A., Shimokawa, T., Okuyama, H., & Kojima, M.
(1986). Phytochemistry, 26, 1577.