Synthesis of glycero amino acid-based surfactants. Part 1. Enzymatic preparation of rac-1-O-(Nα-acetyl-L-aminoacyl)glycerol derivatives

Article abstract of DOI:10.1039/b103132p

An enzymatic procedure for the selective synthesis of a number of N^α-protected amino acid glyceryl esters is presented. Using N^α-Boc-arginine as a model substrate, changes in product yield are examined for the following reaction variables: aqueous buffer content, pH, biocatalyst configuration (free or deposited), enzyme concentration and synthetic approach (thermodynamically or kinetically controlled synthesis). The best yields (61-89%) are obtained at 50 °C in glycerol containing a range of aqueous buffer from 0 to 10% (v/v). Readily available industrial proteases and lipases are used as catalysts. The method allowed us to selectively acylate one of the primary hydroxy groups of the glycerol with the carboxylate groups of the amino acid. In the case of aspartic acid and glutamic acid, selective esterification of the α-carboxylate group is achieved. None of the enzymes tested could differentiate between the two enantiotopic hydroxymethyl groups of glycerol, giving diastereoisomeric mixtures in all cases.

Full text of DOI:10.1039/b103132p

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis of glycero amino acid-based surfactants. Part 1.
Enzymatic preparation of rac-1-O-(Nꢀ-acetyl-L-aminoacyl)glycerol
derivatives
1
Carmen Morán,a,b María Rosa Infanteb and Pere Clapés*a
a Department of Peptide and Protein Chemistry, Instituto de Investigaciones Químicas y
Ambientales de Barcelona-C.S.I.C., Jordi Girona 18-26, 08034-Barcelona, Spain
b Department of Surfactant Technology, Instituto de Investigaciones Químicas y Ambientales de
Barcelona-C.S.I.C., Jordi Girona 18-26, 08034-Barcelona, Spain
Received (in Cambridge, UK) 5th April 2001, Accepted 11th July 2001
First published as an Advance Article on the web 10th August 2001
An enzymatic procedure for the selective synthesis of a number of Nα-protected amino acid glyceryl esters is
presented. Using N α-Boc-arginine as a model substrate, changes in product yield are examined for the following
reaction variables: aqueous buffer content, pH, biocatalyst configuration (free or deposited), enzyme concentration
and synthetic approach (thermodynamically or kinetically controlled synthesis). The best yields (61–89%) are
obtained at 50 ЊC in glycerol containing a range of aqueous buffer from 0 to 10% (v/v). Readily available industrial
proteases and lipases are used as catalysts. The method allowed us to selectively acylate one of the primary hydroxy
groups of the glycerol with the carboxylate groups of the amino acid. In the case of aspartic acid and glutamic acid,
selective esterification of the α-carboxylate group is achieved. None of the enzymes tested could differentiate between
the two enantiotopic hydroxymethyl groups of glycerol, giving diastereoisomeric mixtures in all cases.
sources of starting materials we chose the simple glycerol. The
ester bond between the α-carboxy group of the amino acid and
Introduction
Amino acid and peptide lipid conjugates constitute an interest-
ing class of speciality surfactants with good surface properties,
antimicrobial activity, low potential toxicity and high bio-
degradability.1 Moreover, they can be synthesized using selec-
tive chemoenzymatic methods from natural renewable sources
of raw materials such as amino acids and fatty acids.2 All these
features make them very attractive as dispersants, drug carriers
and bactericides in food and pharmaceutical formulations.
Glycero amino acid-based surfactants constitute a novel
class of amino acid lipid conjugates, which can be considered
analogues of partial glycerides and phospholipids. They consist
of one or two aliphatic chains and one polar head, i.e. the
amino acid, linked together through a glycerol moiety. The
resulting structures resemble the acid esters of monoglycerides
such as lactic, citric, tartaric and succinic glyceride esters widely
used as food emulsifiers.3 The physicochemical and biological
properties of amino acid glyceride conjugates have not been
extensively explored yet.4 From preliminary observations
carried out in our lab, these novel compounds combine the
advantages of both partial glycerides and lipo-amino acids. For
instance, they possess antimicrobial activity, typically from
lipo-amino acids, and form lamellar phases and vesicles, char-
acteristic of partial glycerides and phospholipids. Moreover,
the possibility of introducing different ionic groups (i.e., selec-
ting the amino acid) increases the swelling properties by pro-
moting the electric repulsion between charged group bilayers.
For all these reasons, they constitute a promising family of
compounds with a great potential interest in pharmaceutical
and food formulations.
one of the primary hydroxy groups of glycerol can be per-
formed chemically or enzymatically. A specific catalyst such as
BF3 at 50 or 60 ЊC with an excess of glycerol is the chemical
method used.2a,6 In this case, cumbersome orthogonal protec-
tion steps for polyfunctional amino acids are often required.
Moreover, acid-sensitive protecting groups, such as tert-
butoxycarbonyl (Boc), cannot be used due to their instability in
the presence of BF3. Alternatively, enzymatic methodologies
using hydrolytic enzymes, notably papain, as catalysts for
the ester-bond formation have been employed successfully.7
However, in this strategy acidic and basic side functions were
protected, thereby negating the potential advantages of enzyme
selectivity over the chemical method.
The enzymatic preparation of amino acid-based surfactants,
of mono and diacylglyceride type, is a current topic of interest
in our laboratory. Our pursuit of this goal first led to the syn-
thesis of 1-O-(N α-acetylaminoacyl)glycerol derivatives or
α
amino acid COO-glyceryl esters, which are to constitute the
polar head of the surfactant. Thus, hydrophilic amino acids
with either ionic or non-ionic character, particularly those
found in the peptidic sequences of naturally occurring lipo-
peptides,8 were selected: arginine, aspartic acid, asparagine,
glutamic acid, glutamine and tyrosine. In this paper, we report
the enzymatic regioselective synthesis of the aforementioned
amino acid glyceryl esters with minimal protection. The amino
acids were solely Nα-protected with tert-butoxycarbonyl (Boc)
and, preferentially, with acetyl (Ac). Boc protecting group can
be easily removed under acidic conditions. Acetyl, being present
in numerous natural structures, allowed the permanent protec-
tion of amine functions, preventing their interaction with the
acidic side functions and preserving the non-ionic character of
the neutral amino acids. Therefore, the goal was to prepare the
following amino acid glyceryl esters: Ac-Arg-OGl 2, Ac-Asp-
OGl 3, Ac-Glu-OGl 4, Ac-Asn-OGl 5, Ac-Gln-OGl 6 and Ac-
Tyr-OGl 7 including Boc-Arg-OGlؒHCl 1. Special attention
One interesting strategy for the synthesis of glycero amino
acid-based surfactants consists of the acylation of the hydroxy
groups of amino acid and peptide glyceryl ester derivatives.2a
Amino acid glyceryl esters can be prepared from the corre-
sponding amino acid and either solketal, glycidol or glycerol.5
Having in mind the use of readily available natural renewable
DOI: 10.1039/b103132p
J. Chem. Soc., Perkin Trans. 1, 2001, 2063–2070
This journal is © The Royal Society of Chemistry 2001
2063

View Article Online
was paid to the purification procedures since a simple solvent
extraction cannot be applied in this case. The study was focused
on two aspects. First, the influence of water content, biocatalyst
type and configuration, enzyme concentration and the pH were
systematically investigated. Second, under the best reaction
conditions the preparation and purification of the derivatives
on a gram scale were performed. The methodology developed
provides a general strategy for the synthesis of amino acid and
oligopeptides glyceryl esters. These derivatives can also be
useful as acyl-donor ester substrates in irreversible lipase- and
trypsin-catalysed kinetically controlled peptide synthesis5–7 and
as a strategy to modulate the solubility of single amino acids in
pharmaceutical and dermatological formulations.9
Fig.
1
Kinetically controlled papain-catalysed synthesis of 1.
Influence of aqueous buffer concentration on the product 1 yield
(square symbols) and on acyl donor hydrolysis (Boc-Arg-OHؒHCl)
(circle symbols). Papain deposited onto polyamide (filled symbols) and
Celite (empty symbols) was used as catalyst. Reaction conditions are
described in the Experimental section.
Effect of buffer concentration and support for enzyme deposition
Papain deposited either onto Celite or polyamide was placed
in glycerol (1 ml), containing the Boc-Arg-OMeؒHCl, at 50 ЊC
and the reaction performance was examined as a function of
aqueous buffer concentration in the medium. The results
obtained, depicted in Fig. 1, show that, working with poly-
amide, maximum yields (65–67%) were achieved at buffer con-
centrations ranging between 13% and 15% v/v. Above and
below this range the transesterification yields decreased because
of an increase of the acyl donor hydrolysis rate (data not show)
and either low enzymatic activity or inactivation, respectively.
Using Celite the system followed a similar trend except that
the maximum yields were achieved at 10–13% v/v buffer. A
close inspection of the data revealed that the transesterification
product : hydrolysed acyl donor ratio for both supports was
rather constant (≈2) in the 7–15% v/v buffer range. This sug-
gests that, within this range, the product yield depended on the
enzymatic activity rather than on the relative rates of hydrolysis
and glycerolysis of the acyl enzyme (Scheme 1A). On this
basis, the different optimal buffer content observed with both
supports may be explained in terms of enzymatic activity and
support aquaphilicity10 (i.e., the partition of water between the
support material and the solvent). Polyamide adsorbed much
more water than did Celite so that the enzyme is insufficiently
hydrated, and thereby less active, requiring more water to be
present in the medium.
Results and discussion
The synthesis of Nα-protected amino acid glyceryl esters
involves the esterification of one of the primary hydroxy groups
of glycerol by the α carboxy group of the amino acid (Scheme
1). Initially, a systematic study was conducted to establish suit-
Scheme 1 Kinetically (A) and thermodynamically (B) controlled
enzymatic synthesis of 1-O-(Nα-protected aminoacyl) glycerol
derivatives (R1-AA-OGl): EH protease or lipase (serine or cysteine type
in A); R1-AA-OR2, acyl-donor ester; R1-AA-E, acyl-enzyme complex.
Influence of the biocatalyst
able reaction conditions for the selective enzymatic synthesis.
For experimental convenience, the synthesis of 1 was used as a
model. The ester-bond formation was performed first under
kinetically controlled conditions (Scheme 1A) using Boc-Arg-
OMeؒHCl as acyl donor. Recent results from our lab2c and
literature data7 suggested papain as putative catalyst in gly-
cerol, acting as both reactant and solvent, at 50 ЊC. Preliminary
experiments confirmed that the product could be obtained
under those conditions using papain on Celite as catalyst. A
selective monoacylation of one of the primary hydroxy groups
We have shown that papain from papaya latex can be success-
fully used in the transesterification of Boc-Arg-OMeؒHCl with
glycerol. Moreover, we were also interested in investigating the
ability of other industrial hydrolases to catalyse this reaction
and, in turn, use them for the synthesis of the target amino acid
glyceryl esters. To this end, the following lipases and proteases
were tested as catalysts: lipases from Candida antarctica
(Novozym 435), Aspergillus niger (lipase A), Pseudomonas
cepacia (lipase PS), subtilisins from Bacillus subtilis (Alca-
lase, Protease N), trypsin and bromelain. For comparison
purposes, highly purified subtilisin BPNЈ was also examined.
1
of the glycerol was ascertained by H NMR, 13C NMR and
mass spectrometric analysis.
2064
J. Chem. Soc., Perkin Trans. 1, 2001, 2063–2070

View Article Online
Proteases and lipases were deposited onto Celite, except for
Novozym 435 that was already immobilized. Reactions were
performed in glycerol containing 10% (v/v) buffer concen-
tration at 50 ЊC.
than those using the enzyme in solution. These results are con-
sistent with the fact that, in this reaction, deposition stabilizes
the enzyme. Nevertheless, it was observed that the ratio of
transesterification product : hydrolysed acyl donor, akin to the
acyl donor partition into water and one of the primary OH
groups of glycerol, was unfavourable using deposited enzymes.
This was particularly significant with papain; using the enzyme
deposited onto Celite the reaction was faster but the yield was
lower than that with soluble enzyme. The same behaviour was
observed with Alcalase (Table 2).
Table 1 shows that subtilisins were the best biocatalysts for
the transesterification reaction. An important advantage of
these enzymes is that scavengers, used with papain to prevent
the oxidation of its active site, can be avoided. Furthermore,
subtilisins are produced on an industrial scale for detergent
formulations, so that they are readily available and inexpensive.
Bromelain and trypsin, although specific for basic amino acids,
turned out to be unable to catalyse this reaction. The high
degree of acyl donor hydrolysis indicates that they have poor
affinity for glycerol at the S1Ј subsite.
Lipases were also inefficient biocatalysts for the synthesis
of Boc-Arg-OGlؒHCl 1. Valivety et al.2b reported that some
lipases, notably from Candida antarctica, readily accept N α-
(benzyloxycarbonyl)amino acids as substrates for esterification
reactions with long aliphatic alcohols and diols. In the present
case, as we will demonstrate below, the Nα protecting group of
the arginine had a marked influence on the lipase reactivity for
this particular amino acid.
Thermodynamically controlled synthesis
The synthetic pathway leading to Boc-Arg-OGlؒHCl 1 involved
the preparation of the acyl donor Boc-Arg-OMeؒHCl from
either Boc-Arg-OHؒHCl or H-Arg-OMe. For the sake of sim-
plicity, we investigated if Boc-Arg-OHؒHCl could be esteri-
fied directly under thermodynamically controlled conditions
(Scheme 1B).
Initially, reactions were carried out at 50 ЊC in glycerol con-
taining 10% v/v buffer (boric–borate, 0.1 M, pH 8.2 for papain
and HCl-Tris, 50 mM, pH 7.8 for Protease N and Alcalase)
using papain, Protease N and Alcalase deposited onto Celite
as catalysts. Papain–Celite preparation gave 60% esterifi-
cation yield in 48 h. Attempts to improve this yield by reducing
the buffer concentration failed: at 5% v/v buffer, a 10% yield
was achieved in 72 h. Mitin et al.7 suggested that denaturation
of papain may occur under these conditions. On the other
hand, additions of buffer up to 15% v/v did not affect the
yield substantially (58% after 48 h). Thus, the effect of buffer on
the reaction performance was similar to that observed under
kinetically controlled conditions. To our surprise, Protease N
and Alcalase gave less than 2% of Boc-Arg-OGlؒHCl 1.
Again, as we demonstrate below, the Boc protecting group
exerted a strong influence on the reactivity of this substrate, in
particular under thermodynamic control.
Biocatalyst concentration and configuration
To assess the best process for the enzymatic synthesis of
glyceryl esters, both possibilities, of free and deposited
enzymes, were considered. On the basis of the aforedescribed
results, subtilisins and papain were the biocatalysts of choice.
The solubility of Protease N, Alcalase and papain in glycerol
was ascertained, obtaining homogeneous solutions in all cases.
As Table 2 shows, reactions using deposited enzyme resulted in
higher Boc-Arg-OMeؒHCl conversions and/or reaction rates
Table 1 Effect of biocatalyst on the transesterification reaction of
Boc-Arg-OMeؒHCl with glycerola
The esterification yield under thermodynamically controlled
conditions depends crucially on the ionization equilibrium of
the carboxylic group involved. Therefore, in the present case,
the reaction pH should range between 3 and 4 according to the
Product
Acyl donor
hydrolysis yield (%)
Enzyme
yieldb (%)
Alcalase
69
65
68
20
10
10
12
3
18
10
17
70
15
2
α
pK of the COOH of the Nα-protected amino acid derivatives.
Protease N
Subtilisin BPNЈ
Trypsin
Bromelain
Lipase A
Lipase PS
Novozym 435
Thus, enzymatic reactions were also conducted in glycerol con-
taining citrate buffer, 0.1 M, pH 3.2 or 4.0 (10% v/v) using
papain deposited onto Celite at the corresponding pH. Under
these conditions, the yields dropped dramatically to 40% and
30%, respectively. The best result (60%) was achieved when the
enzyme was deposited at pH 8.2. Braun and Kuhl11 also found
that pH 8.6 was the optimum for papain-catalysed esterification
of Nα-protected amino acids in organic media. These results
differ from the optimal pH (pHopt 3.2) reported for papain-
catalysed ester-bond formation in predominantly aqueous
2
5
a Boc-Arg-OMeؒHCl (500 mM) was dissolved in glycerol (1 ml)
containing boric–borate 0.1 M, pH 8.2, buffer (10% v/v). Reactions
were carried out at 50 ЊC. Enzymes deposited onto Celite (200 mg
mlϪ1, 100 mg enzyme gϪ1 Celite) were used as catalysts. b After 48 h.
Table 2 Effect of biocatalyst configuration on the transesterification reaction of Boc-Arg-OMeؒHCl with glycerola
Biocatalyst
configuration
(mg)
Reaction
time
(t/h)
Acyl donor
hydrolysis
(%)
Substrate
conversion
(%)
Product : acyl
donor hydrolysis
ratio
Yield
(%)
Enzyme
Papain
Deposited (400)b
Free (40)
8
48
48
48
48
48
48
48
48
53
70
70
46
70
59
70
76
79
41
23
17
5
23
6
8
6
14
94
93
87
51
93
66
78
82
92
1.3
3.0
4.1
9.2
3.0
9.8
8.8
12.7
9.8
Protease N
Alcalase
Deposited (400)b
Free (40)
Deposited (400)b
Free (40)c
Free (66)c
Free (80)c
Free (132)c
a Boc-Arg-OMeؒHCl (500 mM) was dissolved in glycerol (1 ml) containing the corresponding aqueous buffer solution (10% v/v; boric–borate 0.1 M,
pH 8.2, for papain and Tris-HCl 50 mM, pH 7.8, for both Alcalase and Protease N). Reactions were carried out at 50 ЊC. b mg of enzyme-support
preparation (100 mg enzyme gϪ1 Celite, that is, 40 mg of enzyme in the reaction). c The enzyme was added as liquid preparation (66 mg protein mlϪ1
liquid preparation).
J. Chem. Soc., Perkin Trans. 1, 2001, 2063–2070
2065

View Article Online
Table 3 Hydrolase-catalysed synthesis of Nα-protected amino acid glyceryl ester derivativesa
Nα-Protected amino
acid glyceryl ester
Reaction
yield (%)
Isolated yield (g)
(% of theor.)
Enzyme
1
2
Papain
Papain
76
74
79
90
6
19
89
74
2
8 (55)
13 (67)
8 (59)
Alcalase
Novozym 435b
Papain
Alcalase
Novozym 435b
Papain
Alcalase
Novozym 435b
Papainc
Alcalase
Novozym 435b
Papain
Alcalase
Novozym 435b
Papain
Alcalase
Novozym 435b
Protease Nd
Papaine
15 (80)
3
4
5
6
7
12 (73)
9 (53)
6
61
10
2
72
8
10 (46)
13 (64)
3
12
10
3
87
62
10 (71)
a Reaction conditions as described in Experimental section. The aqueous buffer content was 10% v/v unless otherwise stated. b The water content in
the medium was 5% v/v. c The buffer content was 15% (v/v). d Reaction was carried out under kinetically controlled conditions (transesterification
reaction) and anhydrous conditions. e Reaction was carried out under kinetically controlled conditions.
media.7 There is evidence that in low-water systems the acid–
base conditions may change due to, as in the present case, acidic
substrates. Thus, the ionization state of the enzyme, akin to its
activity, cannot be controlled exclusively by pH but also by the
counter-ion.12 It seems likely that papain deposition at pH 8.2
can ensure its optimal ionization state in low-water conditions
and in the presence of acidic substrates.
Finally, it was observed that yields improved significantly
when free papain (40 mg mlϪ1) was used: 72% compared with
60% yield when using deposited enzyme. As in the kinetically
controlled approach, the support seems to have a negative effect
on the reaction yield.
of these enzymes seemed largely affected by both geometry and
hydrophobicity of the Nα protecting group of the amino acid.13
The acidic amino acids, Asp and Glu, showed a quite striking
behaviour. Novozym 435 gave the best yield in the synthesis
of 3, whereas it was found to be papain for 4. A number of
α
papers14,15 reported the regioselective COOH enzymatic ester
synthesis and hydrolysis of the aspartic acid and glutamic acid
derivatives. As expected, in the present case, both enzymes
α
esterified selectively the COOH groups of these amino acids.
The esterification position was established by comparing the
13C NMR spectra of the Nα-acetylamino acids with the corre-
sponding glyceryl ester derivatives. The α-ester substitution
induced 1.18 and 1.46 ppm upfield shifts on the α-carbon of
aspartic acid and glutamic acid, respectively (and no effect on
the chemical shifts of β and γ carbons).15
Synthesis of rac-1-O-(Nꢀ-acetyl-L-aminoacyl)glycerol
The scope of the synthetic methodology developed for the
preparation of 1 was extended to the other target amino acids.
Crude and/or industrial protease preparations, namely papain,
Alcalase, Protease N and Novozym 435, used as powder or
liquid (straight from the bottle), were screened as putative bio-
catalysts. The syntheses were conducted first at analytical level
and then on a preparative scale under the best reaction condi-
tions. The results obtained are presented in Table 3. Important
features are the following. Papain was a quite flexible catalyst
for the synthesis of 1-O-(Nα-protected aminoacyl)glycerol,
being reactive for most of amino acids. The direct esterification
under thermodynamically controlled conditions was used
except for the preparation of 7. In this case, Ac-Tyr-OH was
found to be inactive for the enzymes assayed and kinetically
controlled conditions had to be employed.
The negative influence of the support in papain-catalysed
reactions with Boc-Arg-OXؒHCl (X: H, Me) was confirmed
using Ac-Arg-OHؒHCl. Only a 22% yield of 2 was achieved
using papain deposited onto Celite, compared with 74% using
soluble enzyme. In contrast, the yields with glutamic acid,
asparagine or glutamine were similar regardless of enzyme
configuration.
In all the amino acid glyceryl ester synthesized, the pres-
ence of 2-O-regioisomer [2-O-(N α-acetyl--aminoacyl)-
glycerol] (≈10%) was detected by 1H NMR. The 1H NMR
signals were unequivocally assigned using the aspartic acid
derivative as a model. In this case, both regioisomers, namely
rac-1-O- (4) and 2-O-(Nα-Ac--aspart-1-yl)glycerol, were isolated
by HPLC (tr1-O = 7.5 min, tr2-O = 6.3 min; on a C-18 column,
eluted isocratically with 0.1% v/v TFA in water). Formation of
the ester bond in the secondary hydroxy group [2-O-(N α-Ac--
aspart-1-yl)glycerol] resulted in a downfield shift of the proton
at C2 of the glycerol moiety from δ 3.6 to δ 4.7 and a single
signal at δ 3.4 corresponding to the four symmetric protons at
C1 and C3. These NMR data helped us in identifying and
quantifying the 2-O-regioisomers of the other amino acid
glyceryl esters by NMR. The enzymatic esterification time
course of aspartic acid by the glycerol, as in the esterification
of the other amino acid derivatives, was followed by HPLC
analysis. As mentioned above, in the case of aspartic acid
HPLC baseline separation of both regioisomers was achieved.
The chromatograms revealed that a major product was formed
during the reaction time, which was identified as the rac-1-O-
regioisomer. A minor peak appeared simultaneously, which was
identified as the 2-O-regioisomer, constituting around 10% of
the product at the end of the reaction (24 h). This percentage
did not change during the reaction work-up and product purifi-
cation. Hence, in agreement with the literature data,16,17 this
result indicated that the enzymes exhibited a preference towards
In the previous section, we found that Alcalase and Novozym
435 failed as catalyst for the synthesis of 1 using Boc-Arg-
OXؒHCl (X: H, Me) as acyl donor. Interestingly, Ac-Arg-OH
turned out to be a good substrate for both Alcalase and
Novozym 435. As reported for chymotrypsin, the reactivity
2066
J. Chem. Soc., Perkin Trans. 1, 2001, 2063–2070

View Article Online
acylation of one of the primary hydroxy groups in the glycerol
molecule (i.e., alcohol groups in C1 or C3 positions). This sug-
gested that the 2-regioisomer, although it could be produced
enzymatically to some extent, it may be formed preferentially
by simple 3(or 1) → 2 acyl migration similar to that observed in
mono- and diglycerides18 and monoesters of Boc-Ala-OH with
propane-1,2-diol.16
particle size 26 µm; mean pore diameter 17 000 nm; specific
surface area BET method 2.19 m2 Ϫ1) were obtained from
g
Fluka. Polyamide-6 (EP 700, particle size < 800 µm; mean pore
diameter 50–300 nm; specific surface area BET method 8.4 m2
g
Ϫ1) was a generous gift from Akzo (Obernburg, Germany).
Solvents and other chemicals used in this work were of
analytical grade.
It was ascertained that the enzymes used could not dis-
tinguish between the two enantiotopic hydroxymethyls of
glycerol. The presence of both diastereoisomers was detected
Instrumentation
Mass spectrometric analysis were performed at the Servei
d’Espectrometria de Masses of the University of Barcelona.
Electrospray-mass spectrometric (ES-MS) analyses were
recorded on a VG-Quattro system from Fisons Instruments
(Altricham, UK). The carrier solution was water–CH3CN
(1 : 1) containing 1% (v/v) formic acid. Nuclear magnetic
resonance (NMR) analyses were carried out at the Instituto de
Investigaciones Químicas y Ambientales-CSIC. 1H and 13C
NMR spectra were recorded with a Unity-300 spectrometer
from Varian (Palo Alto, California, USA) for d6-DMSO and
CD3OD solutions.
1
and quantified (approximately 50% of each) by H NMR. The
1H NMR spectra of diastereoisomeric amino acid glyceryl
esters were distinguishable by the hydrogens of the C1 bearing
the amino acid residue (ROCOCH2CHOHCH2OH).19 At this
stage, however, it was not possible to separate them or their
derivatives (i.e. those obtained by acetylation or benzoylation
of the free hydroxy groups) by chromatographic techniques.
1
Hence, the specific assignation of the H NMR signals to S,R
or S,S diastereoisomers was not possible.
An important issue of the aforedescribed synthetic method-
ology was that of the purification procedures. All the products
were highly soluble in water, so that simple solvent extractions
were completely inefficient. Amino acid glyceryl derivatives
such as 1–4, containing charged amino acids, were purified by
anion- or cation-exchange chromatography. Purification of
uncharged compounds 5 and 6 was achieved using column
chromatography on activated charcoal–Celite 50 : 50,
whereas 7 was purified by means of hydrophobic interactions
on either styrene–divinylbenzene or C-18-type stationary
phases.
Methods
Enzyme adsorption. The general procedure for the deposition
of the enzymes onto solid support materials was the follow-
ing. An enzyme solution in the proper buffer (1 ml) was mixed
thoroughly with the support (1 g). The mixture was then
evaporated under vacuum overnight. For lipase A, lipase
PS, Alcalase, proteinase N, subtilisin BPNЈ and trypsin 100
mg of enzyme gϪ1 of Celite were prepared using 50 mM HCl-
Tris, pH 7.8, buffer. For bromelain and papain, the enzyme (100
mg) and DTT (50 mg) were dissolved in 0.1 M boric–borate pH
8.2 buffer and mixed with Polyamide or Celite depending on
the experiment (1 g).
Experimental
Materials
Papain (EC 3.4.22.2) from papaya latex crude powder [1.7
U mgϪ1 of solid, one unit (U) will hydrolyse 1.0 µmol of
N α-benzoyl--arginine ethyl ester hydrochloride per minute
(BAEE) at pH 6.2 at 25 ЊC], bromelain (EC 3.4.22.4) from
pineapple stem (2100 U mgϪ1, one unit will release 1.0 µmol
of p-nitrophenol from N α-Z--lysine-p-nitrophenyl ester per
min at pH 4.6 at 25 ЊC), subtilisin BPNЈ (EC 3.4.21.62)
crystallized and lyophilysed (7.8 U mgϪ1 of solid) and
trypsin (EC 3.4.21.4) from porcine pancreas (15,900 U mgϪ1
protein) crystallized, dialysed and lyophilysed were obtained
from Sigma (St. Louis, MO, USA). Bacillus subtilis protease
(Proteinase N) (EC 3.4.24.4) [7.3 U mgϪ1 of solid, one unit (U)
corresponds to the amount of enzyme which liberates 1 µmol of
folin-positive amino acid and peptides per minute at pH 7.0 and
37 ЊC using casein as substrate] was obtained from Fluka
(Buchs, Switzerland). Bacillus subtilis protease (Alcalase)
(EC 3.4.21.62) (0.6 U mgϪ1) and Candida antarctica lipase
(Novozym 435) (EC 3.1.1.3) (7000 PLU g-1, PLU: propyl
laurate units) were a generous gift from Novo Nordisk
A/S (Bagsvaerd, Denmark). Aspergillus niger lipase (lipase
A, 136,000 U gϪ1 Amano’s method) and Pseudomonas cepa-
cia lipase (lipase PS, 32100 U gϪ1 Amano’s method) were
generously donated by Amano Pharmaceutical Co., Ltd.
(Nagoya, Japan). N α-Acetyl--arginine (Ac-Arg-OH) [which
was converted to N α-acetyl--arginine hydrochloride (Ac--
Arg-OHؒHCl) by treatment with HCl], N α-acetyl--glutamine
(Ac-Gln-OH) and N α-tert-butoxycarbonyl--arginine hydro-
chloride (Boc-Arg-OHؒHCl) were obtained from Novabiochem
(Läufelfingen, Switzerland). N α-tert-Butoxycarbonyl--argin-
ine methyl ester hydrochloride (Boc-Arg-OMeؒHCl) was
synthesized in our laboratory as described below. N α-Acetyl-
-tyrosine ethyl ester was synthesized in our lab using the
standard thionyl dichloride procedure. Glycerol, N α-acetyl--
glutamic acid (Ac-Glu-OH), N α-acetyl--aspartic acid (Ac-
Asp-OH), 1,4-dithio--threitol (DTT) and Celite (type 545,
HPLC. The amount of reactants and products in the enzym-
atic reactions was measured by HPLC analysis. HPLC analyses
were performed on a Merck-Hitachi (Darmstadt, Germany)
Lichrograph system using a Lichrocart 250–4 HPLC cartridge,
250 × 4 mm filled with Lichrosphere® 100, RP-18, 5 µm
(Merck). Samples (8 µl) were withdrawn from the reaction
medium, mixed with acetic acid (8 µl) to stop any further
enzymatic reaction, evaporated under vacuum, and dissolved
in HPLC eluent before analysis. Quantitative analysis was
performed from peak areas by means of the external standard
method. Preparative HPLC runs were performed on a Waters
(Milford, MA, USA) Prep LC 4000 pumping system and a
Waters PrePack® 1000 module fitted with a PrePack® (Waters)
column (47 × 300 mm) filled with Bondapack C18, 300 Å,
15–20 µm stationary phase for the purification of Ac-Tyr-OGl.
Synthesis of Nꢀ-tert-butoxycarbonyl-l-arginine methyl ester
hydrochloride (Boc-Arg-OMeؒHCl). Arginine methyl ester
dihydrochloride (22.0 g, 84.23 mmol) was dispersed in
dimethylformamide (DMF) (250 ml). Then, triethylamine
(14 ml, 101.1 mmol) was added dropwise and the reaction mix-
ture was stirred overnight at 25 ЊC. The reaction mixture was
filtered and the solid collected was identified as triethylamine
hydrochloride. Then, the filtrate was evaporated to dryness
under reduced pressure. The residue was dissolved in CH3CN
(200 ml), and di-tert-butyl dicarbonate (23.9 g, 109.5 mmol)
was added dropwise over a period of 15 min. The reaction was
allowed to proceed at 25 ЊC for 24 h. Then, the solvent was
removed under vacuum. The solid obtained was dissolved in
water and washed successively with ethyl acetate (3 × 150 ml)
and diethyl ether (3 × 150 ml). The aqueous layer, containing
the product, was freeze dried, yielding a highly hygroscopic
white solid (20.5 g, 74%), δH (300 MHz; d6-DMSO; 45 ЊC)
7.87 (1H, s, CONHCH), 7.24 [5H, m, NHC(NH2)2], 3.93
(1H, m, NHCHCOOCH3), 3.61 (3H, s, COCH3), 3.04 (4H, m,
J. Chem. Soc., Perkin Trans. 1, 2001, 2063–2070
2067

View Article Online
CH2CH2CH2NH), 1.51 (2H, m, CHCH2CH2CH2NH), 1.39
(9H, s, (CH3)3CO); δC (75 MHz; d6-DMSO; 45 ЊC): 172.69
60–80 ml minϪ1, 3–4 cm minϪ1. Elution conditions for each
product using this technique are the following.
(COOCH ), 157.06 (C᎐N guanidine group), 155.37 (O-C-O-
Boc-Arg-OGlؒHCl and Ac-Arg-OGlؒHCl. The resin (Macro-
Prep High S Support) was equilibrated initially with 50 mM
phosphate buffer, pH 5.7 (1.5 l, 4 bed volumes), and washed
subsequently with plain, deionized water (2.4 l, 6.5 bed
volumes). Then, aliquots of the residue containing 6 or 7 g of
crude Boc-Arg-OGlؒHCl or Ac-Arg-OGlؒHCl, respectively
were loaded. First, the glycerol and Boc-Arg-OHؒHCl or
Ac-Arg-OHؒHCl were eliminated by washing The resin with
4.4 l (12 bed volumes) of plain, deionized water; at this pH
both arginine derivatives have the COOH group unprotonated
and consequently are electrically neutral. Then, the target
products were eluted isocratically by 0.74 l (2 bed volumes) of
250 mM NaCl in water–EtOH 3 : 2. Products were desalted by
precipitation with anhydrous ethanol.
Ac-Asp-OGl and Ac-Glu-OGl.. First, the resin (anion-
exchange resin AG 3-X4) was equilibrated with 1.0 M acetate
buffer, pH 7.2, and washed subsequently with plain, deionized
water. Then, aliquots of the residue containing 11 g of crude
Ac-Asp-OGl or 6 g of crude Ac-Glu-OGl were loaded. After
that, the glycerol was eliminated by washing the resin with 5.6 l
(15 bed volumes) of plain, deionized water. Elution of the
products were achieved isocratically by 1.5 l (4 bed volumes) of
aq. 0.5 M acetic acid.
᎐
3
NH), 78.14 [(CH ) C-O], 53.05 (NHCHC᎐O), 51.54 (OCH ),
᎐
3
3
3
45.27–25.07 (CH2), 28.02 [(CH3)3C-O]; m/z 289.1 ([M ϩ H])ϩ.
C12H25N4O4 requires m/z, 289.3).
Enzymatic reactions. Reactions on an analytical scale were
carried out in 5 ml closed flask vessels placed on a recipro-
cal shaker (200 rpm) at 50 ЊC. The amino acid derivative
(0.5 mmol) was dissolved in glycerol (13.7 mmol in 1 ml), con-
taining the corresponding buffer (0–15% v/v depending of
the experiment), at 50 ЊC. The reactions were started by the
addition of the enzyme, deposited onto a solid support or
straight from the bottle depending on the experiment. When the
reactions were catalysed by soluble papain and bromelain, the
reaction medium contained DTT (50% of the amount of
enzyme). Moreover, reactions with these two enzymes were
performed under argon atmosphere.
HPLC reaction monitoring. Synthesis of Boc-Arg-OGlؒHCl.
Solvent system: solvent [A]: aq. triethylamine–phosphoric acid
buffer, pH 6.0 (TEAP pH 6.0) in water; solvent [B]: TEAP, pH
6.0–CH3CN 1 : 4; gradient elution from 5 to 60% B in 30 min;
retention factor (kЈ) for each product: Boc-Arg-OH (4.9),
Boc-Arg-OMe (8.5), Boc-Arg-OGl (6.1).
Synthesis of Nα-acetylamino acid derivatives. Solvent system:
solvent [A]: 0.1% (v/v) trifluoroacetic acid (TFA) in water,
solvent [B]: 0.085% (v/v) TFA in water–CH3CN 1 : 4; isocratic
elution 0% B; kЈ-values for each product: Ac-Arg-OH (2.6),
Ac-Arg-OGl (5.2), Ac-Asp-OH (1.2), Ac-Asp-OGl (2.6),
Ac-Glu-OH (2.4), Ac-Glu-OGl (5.2), Ac-Asn-OH (0.7),
Ac-Asn-OGl (1.5), Ac-Gln-OH (0.9), Ac-Gln-OGl (2.2).
Synthesis of Nα-Ac-Tyr-OGl. Solvent system: solvent [A]:
0.05% v/v CH3COOH in water, solvent [B]: 0.05% v/v CH3-
COOH in water–CH3CN 1 : 4; gradient elution from 10 to 70%
B in 30 min; kЈ-values for each product: Ac-Tyr-OH (3.2),
Ac-Tyr-OEt (7.7), Ac-Tyr-OGl (4.0). In all cases, the flow rate
was 1 ml minϪ1, with UV detection at 215 nm.
Celite/activated charcoal chromatography. This chromato-
graphic support is used extensively for the separation of
polyhydroxylated compounds such as oligosaccharides from
monosaccharides.20 A mixture of Celite and activated char-
coal 1 : 1 (150–160 g of each) was packed in a Buchner fritted
funnel (10 × 13 cm, with a fine-porosity fritted-glass disk) to a
final bed volume of 940–1000 ml. The eluent was passed
through the stationary phase by vacuum. The chromatographic
system was treated initially with an aqueous solution of
acetic acid (10% v/v; 1.5 l) and washed subsequently with plain,
deionized water to pH 5. Then, aliquots of the residue contain-
ing 3.75 g of crude Ac-Asn-OGl or 7 g of crude Ac-Gln-OGl
were loaded. The glycerol was eliminated by washing the sup-
port with plain, deionized water (9 l, 10 bed volumes). Elution
of Ac-Asn-OGl or Ac-Gln-OGl was achieved isocratically by
1.25 l (1.2 bed volumes) and 2.7 l (2.7 bed volumes) of MeOH–
water 13 : 7, respectively.
Reversed-phase chromatography. In this case either C-18 or
bulk porous polymeric media based on styrenic polymers
(Amberchrom CG-161M, Montgomeryville, PA, USA) were
used as stationary phases. Both were effective for the purifi-
cation of the most hydrophobic, Ac-Tyr-OGl, derivative.
For the C-18 reversed phase the procedure was the following.
Aliquots of the residue containing crude Ac-Tyr-OGl (1.5 g)
were loaded onto the preparative PrePack® (Waters) column
(47 × 300 mm) filled with Bondapack C18, 300 Å; 15–20 µm
cartridge, and eluted using an aq. acetic acid solution (0.05%
v/v) and gradient elution from 0 to 16% CH3CN in 20 min. The
flow rate was 100 ml minϪ1 and the products were detected at
either 215 or 225 nm. Using the porous styrenic polymer the
conditions were the following. Aliquots of the residue contain-
ing crude Ac-Tyr-OGl (9 g) was loaded onto the Amberchrom
CG-161M resin (370 ml, bed dimensions 190 × 50 mm packed
on flash-chromatography-type column), and eluted using water
and gradient elution from 0 to 80% EtOH in 60 min.
Enzymatic reactions at preparative level. Scale-up of the
reactions was made on the basis of approximately 10 g of pure
product, considering the reaction and purification yields
obtained at analytical level. The Nα-protected amino acid
(35–70 mmol) was dissolved in glycerol (70–175 ml) at 50 ЊC,
containing the corresponding amount of buffer (0–15% v/v)
and the scavenger (DTT in the case of papain, 1.5–3.5 g). The
reactions were initiated by the addition of the enzyme (see
Table 3) and monitored by HPLC. At the end, the reaction
mixture was diluted three times with a mixture of MeOH–
acetic acid 4 : 1 to stop any further enzymatic reaction. Upon
the addition of MeOH–acetic acid, most of the enzyme became
denatured and a fluffy solid appeared in the solution. This was
especially visible in the reactions carried out using the enzyme
in solution. It was found that filtration through Celite was an
effective way to remove the precipitate and eliminate all possible
enzymatic activity. The filtrate obtained was evaporated under
vacuum to eliminate the methanol and acetic acid. Then, the
oily residue was dissolved with plain, deionized water and puri-
fied as described above. In the case of arginine derivatives, the
final water solution was adjusted to pH 4 with 0.2 M NaOH
before purification.
In all cases, the operation was repeated until the whole crude
was consumed. Analysis of the fractions was accomplished
under analytical RP-chromatography using the same elution
conditions, flow rate, and detection as previously described.
Pure fractions were pooled and lyophilised. Purity of the amino
acid glyceryl ester obtained was assessed by HPLC. NMR and
mass spectroscopy data for each compound are given below.
Purification procedures. Ion-exchange chromatography.
General: Bulk stationary phases MacroPrep High S Support,
50 µm strong cation-exchange resin (Bio-Rad, Hercules, CA,
USA) and Analytical Grade anion-exchange resin AG 3-X4,
100–200 mesh (Bio-Rad), were packed into glass columns of
the flash-chromatography type, to a final bed volume of 370 ml,
bed dimensions 190 × 50 mm. The eluent was passed through
the stationary phase by nitrogen pressure, at a flow rate of
rac-1-O-(Nꢀ-Boc-L-arginyl)glycerol hydrochloride (Boc-Arg-
OGlؒHCl) 1. Purity 92.3% by HPLC; δH (300 MHz; d6-DMSO;
2068
J. Chem. Soc., Perkin Trans. 1, 2001, 2063–2070

View Article Online
45 ЊC) 8.18 (1 H, d, CONHCH), 7.33 [5 H, m, NHC(NH2)2],
4.12–3.87 (3 H, m, NHCHCOOCH2), 3.64 (1 H, quintet, J 5.4,
COOCH2CHOHCH2OH), 3.34 (2 H, d, J 6, CHOHCH2OH),
3.06 (2 H, t, CH2NH), 1.54 [4 H, m, CH(CH2)2], 1.3 [9H, s,
(CH3)3CO]; δC (75 MHz; d6-DMSO; 45 ЊC) 172.15 (CHCOO),
COOCH2CHOH, diaster. I), 4.03 (0.5 H, dd, J 10.8 and 4.8,
COOCH2CHOH, diaster. II), 3.95 (0.5 H, dd, J 11.1 and
6.0, COOCH2CHOH, diaster. II), 3.92 (0.5 H, dd, J 11.1 and
6.0, COOCH2CHOH, diaster. I), 3.63 (1H, quintet, J 5.7,
COOCH2CHOHCH2OH), 3.35 (2 H, br s, J 5.4, CHOHCH2-
OH), 2.47 (2 H, m, H2OCCH2), 1.82 (3 H, s, CH3); δC (75 MHz;
d6-DMSO; 45 ЊC) 171.30 (CONH2), 170.88 (CHCOO), 169.11
(CH3CO), 69.15 (CHOH), 65.98 (OCH2CHOH), 62.48
(CH2OH), 48.89 (NHCHCO), 36.69 (CH2COOH), 22.18
(CH3); m/z (ES) 249.7 ([M ϩ H]ϩ. C9H17N2O6 requires m/z,
249.1).
156.96 (C᎐N guanidine function), 155.41 (OCONH), 78.15
᎐
[(CH3)3CO), 69.09 (CHOH), 65.87 (OCH2CHOH), 62.41
(CH2OH), 53.15 (NHCHCO), 40.33 (CH2NH), 28.04 [(CH3)3-
CO], 27.85 (CH2), 24.91 (CH2); m/z (ES) 348.8 ([M ϩ H]ϩ.
C14H29N4O6 requires m/z, 349.2).
rac-1-O-(Nꢀ-Ac-L-arginyl)glycerol hydrochloride (Ac-Arg-
OGlؒHCl) 2. Purity 97.9% by HPLC; δH (300 MHz; d6-DMSO;
45 ЊC) 8.26 (1 H, d, CONHCH), 7.85 [1 H, t, CH2NHC(CH2)2],
7.20 [4 H, s, CH2NHC(CH2)2], 4.23 (4 H, m, NHCH2CH2), 4.09
(0.5 H,dd, J 11.1 and 4.5, COOCH2CHOH, diaster. I), 4.03
(0.5 H, dd, J 11.1 and 4.8, COOCH2CHOH, diaster. II), 3.93
(0.5 H, dd, J 11.1 and 6.3, COOCH2CHOH, diaster. II), 3.91
(0.5 H, dd, J 11.1 and 6.3, COOCH2CHOH, diaster. I), 3.64
(1 H, quintet, J 5.7, COOCH2CHOHCH2OH), 3.35 (2 H, br s,
J 6.3, CHOHCH2OH), 3.11 (1 H, m, NHCHCO), 1.85 (3 H, s,
CH3CO); 1.73–1.48 (2 H, m, CH2CH); δC (75 MHz; d6-DMSO;
rac-1-O-(Nꢀ-Ac-L-glutaminyl)glycerol (Ac-Gln-OGl) 6. Purity
90.9% by HPLC; δH (300 MHz, d6-DMSO; 45 ЊC) 8.14 (1H, d,
CONHCH), 7.17 (1 H, s, CONH2), 6.67 (1 H, s, CONH2), 4.22
(1 H, m, NHCHCO), 4.09 (0.5 H, dd, J 10.8 and 4.2,
COOCH2CHOH, diaster. I), 4.03 (0.5 H, dd, J 11.1 and 4.5,
COOCH2CHOH, diaster. II), 3.97 (0.5 H, dd, J 11.1 and
6.3, COOCH2CHOH, diaster. II), 3.92 (0.5 H, dd, J 11.1
and 6.0, COOCH2CHOH, diaster. I), 3.64 (1 H, quintet,
J 5.7, COOCH2CHOHCH2OH), 3.36 (2 H, br s, J 5.4,
CHOHCH2OH), 2.12 (2 H, m, H2NCOCH2), 1.97 (2 H, m,
CH2CH), 1.84 (3 H, s, CH3); δC (75 MHz; d6-DMSO; 45 ЊC)
173.23 (γ-CONH2), 171.86 (CHCOO), 169.32 (CH3CO), 69.17
(CHOH), 65.86 (OCH2CHOH), 62.50 (CH2OH), 51.62
(NHCHCO), 31.10 (CH2CONH2), 26.74 (CH2), 22.13 (CH3);
m/z (ES) 263.0 ([M ϩ H]ϩ. C10H19N2O6 requires m/z, 263.1).
45 ЊC) 171.81 (CHCOO), 169.44 (CH CO), 157.06 (C᎐N
᎐
3
guanidine function), 69.15 (CHOH), 65.92 (OCH2CHOH),
62.48 (CH2OH), 51.61 (NHCHCO), 40.33 (CH2NH), 27.95
(CH2), 24.94 (CH2), 22.12 (CH3); m/z (ES) 292.1 ([M ϩ H]ϩ.
SC11H23N4O5 requires m/z, 291.2).
rac-1-(O-(Nꢀ-Ac-L-aspart-1-yl)glycerol (Ac-Asp-OGl) 3.
Purity 95.5% by HPLC; δH (300 MHz; d6-DMSO; 45 ЊC) 12.30
(1 H, s, CH2COOH), 8.11 (1 H, d, CONHCH), 4.56 (1 H, m,
NHCHCO), 4.06 (0.5 H, dd, J 11.1 and 4.2, COOCH2CHOH,
diaster. I), 4.05 (0.5 H, dd, J 11.1 and 4.5, COOCH2CHOH,
diaster. II), 3.92 (1 H, dd, J 11.4 and 4.2, COOCH2CHOH,
diaster. I ϩ II), 3.63 (1 H, quintet, J 6, CH2CHOHCH2), 3.46
(2 H, br s, J 7.8, CHOHCH2OH), 2.80–2.61 (2 H, m, CH2-
COOH), 1.82 (3 H, s, CH3CO); δH (300 MHz; CD3OD; 20 ЊC)
5.19 (1H, m, NHCHCO), 4.61 (0.5 H, dd, J 11.4 and 4.2,
COOCH2CHOH, diaster. I), 4.60 (0.5 H, dd, J 11.1 and 4.2,
COOCH2CHOH, diaster. II), 4.52 (0.5 H, dd, J 11.4 and 6.0,
COOCH2CHOH, diaster. II), 4.51 (0.5 H, dd, J 11.4 and 4.2,
COOCH2CHOH, diaster I), 4.25 (1H, quintet, J 6, COOCH2-
CHOHCH2OH), 3.9 (2 H, br s, J 7.8, CHOHCH2OH), 3.36–
3.28 (2H, m, CH2COOH), 2.40 (3 H, s, CH3CO); δC (75 MHz;
d6-DMSO; 45 ЊC) 172.38 (β-COOH), 170.28 (CHCOO), 169.96
(CH3CO), 69.24 (CHOH), 66.09 (OCH2CHOH), 62.61
(CH2OH), 48.54 (NHCHCO), 36.03 (CH2COOH), 22.42
(CH3); m/z (ES) 249.9 ([M ϩ H]ϩ. C9H16N1O7 requires m/z,
250.2).
rac-1-O-(Nꢀ-Ac-L-tyrosinyl)glycerol (Ac-Tyr-OGl) 7. Purity
99.2% by HPLC; δH (300 MHz, d6-DMSO; 45 ЊC) 9.09 [1 H, s,
C᎐C(OH)C], 8.10 (1 H, d, CONHCH), 6.98 [2 H, d,
᎐
HC᎐C(CH )C᎐CH], 6.66 [2 H, d, HC᎐C(OH)C᎐CH], 4.49
᎐
᎐
᎐
᎐
2
(1 H, m, NHCHCO), 4.04 (0.5 H, dd, J 11.1 and 4.5,
COOCH2CHOH, diaster. I), 4.03 (0.5 H, dd, J 10.8 and 4.2,
COOCH2CHOH, diaster. II), 3.93 (0.5 H, dd, J 10.8 and 6.0,
COOCH2CHOH, diaster. II), 3.93 (0.5 H, dd, J 10.8 and 6.0,
COOCH2CHOH, diaster. I), 3.62 (1 H, quintet, J 5.1,
COOCH2CHOHCH2OH), 3.33 (2 H, br s, J 5.1, CHOHCH2-
OH), 2.95–2.71 (2 H, m, CHCH2), 1.79 (3 H, s, CH3); δC (75
MHz, d6-DMSO; 45 ЊC) 171.57 (CHCOO), 169.19 (CH3CO),
155.79 (C᎐COH), 129.79 (CH C᎐C), 127.22 (CH C᎐C), 114.94
᎐
᎐
᎐
2
2
[C᎐C(OH)C], 69.13 (CHOH), 65.90 (OCH CHOH), 62.53
᎐
2
(CH2OH), 53.78 (NHCH CO), 35.98 (CH2), 22.13 (CH3); m/z,
(ES) 298.2 ([M ϩ H]ϩ. C14H20NO6 requires m/z 298.1).
Acknowledgements
Financial support from the Spanish C.I.C.Y.T., ref. PPQ2000-
1687-CO2-01, is gratefully acknowledged. We are thankful to
Dr Francisco Sánchez for the NMR analysis and discussion
and to Dr Irene Fernández for the mass spectrometry analysis.
rac-1-O-(Nꢀ-Ac-L-glutam-1-yl)glycerol (Ac-Glu-OGl) 4.
Purity 98.8% by HPLC; δH (300 MHz; d6-DMSO; 45 ЊC) 12.21
(1 H, s, COOH), 8.21 (1 H, d, CONHCH), 4.27 (1 H, m,
NHCHCO), 4.09 (0.5 H, dd, J 11.1 and 4.5, COOCH2CHOH,
diaster. I), 4.04 (0.5 H, dd, J 11.4 and 4.5, COOCH2CHOH,
diaster. II), 3.97 (0.5 H, dd, J 10.8 and 6.3, COOCH2CHOH,
diaster. II), 3.92 (0.5 H, dd, J 11.1 and 6.3, COOCH2CHOH,
diaster. I), 3.63 (1 H, quintet, J 5.7, COOCH2CHOHCH2OH),
3.35 (2 H, br s, J 5.1, CHOHCH2OH), 2.26 (2 H, m, CH2-
COOH), 1.94 (2 H, m, CH2COOH), 1.83 (3 H, s, CH3); δC (75
MHz; d6-DMSO; 45 ЊC) 173.44 (γ-COOH), 171.71 (CHCOO),
169.36 (CH3CO), 69.15 (CHOH), 65.90 (OCH2CHOH), 62.49
(CH2OH), 51.24 (NHCHCO), 29.86 (CH2 COOH), 26.25
(CH2), 22.10 (CH3); m/z (ES) 263.7 ([M ϩ H]ϩ. C10H18N1O7
requires m/z, 264.1).
References
1 J. Xia, Y. Xia and I. A. Nnanna, J. Agric. Food Chem., 1995, 43,
867; M. R. Infante, A. Pinazo and J. Seguer, Colloids Surf. A,
1997, 123–124, 49; L. Perez, J. L. Torres, A. Manresa, C. Solans and
M. R. Infante, Langmuir, 1996, 12, 5296; A. Pinazo, X. Wen,
L. Perez, M.-R. Infante and E. I. Franses, Langmuir, 1999, 15, 3134.
2 (a) R. Valivety, P. Jauregui and I. G. E. Vulfson, J. Am. Oil Chem.
Soc., 1997, 74, 879; (b) R. Valivety, I. S. Gill and E. N. Vulfson,
J. Surf. Deterg., 1998, 1, 177; (c) P. Clapés, C. Morán and M. R.
Infante, Biotechnol. Bioeng., 1999, 63, 333; (d ) E. Piera, M. R.
Infante and P. Clapés, Biotechnol. Bioeng., 2000, 70, 323.
3 N. J. Knog, in Food Emulsions, ed. S. E. Friberg and K. Larsson,
Marcel Dekker, New York, 3rd edn., 1997, p. 141.
4 S. Pegiadou, L. Perez and M. R. Infante, J. Surf. Deterg., 2000, 3,
517.
5 J. R. Matos, J. B. West and C. H. Wong, Biotechnol. Lett., 1987, 9,
233.
6 J. Wiese, H. G. Gattner and H. Zahn, Biomed. Biochim. Acta, 1991,
50, S90.
rac-1-O-(Nꢀ-Ac-L-asparaginyl)glycerol (Ac-Asn-OGl) 5. Pur-
ity 93.3% by HPLC; δH (300 MHz; d6-DMSO; 45 ЊC): 8.06 (1 H,
d, CONHCH), 7.29 (1 H, s, CONH2), 6.81 (1 H, s, CONH2),
4.57 (1 H, m, NHCHCO), 4.06 (0.5 H, dd, J 10.8 and 4.5,
J. Chem. Soc., Perkin Trans. 1, 2001, 2063–2070
2069

View Article Online
7 Y. V. Mitin, K. Braun and P. Kuhl, Biotechnol. Bioeng., 1997, 54,
287.
8 K. Jenny, O. Kaeppeli and A. Fiechter, Appl. Microbiol. Biotechnol.,
1991, 36, 5; A. Fiechter, Trends Food Sci. Technol., 1992, 3,
286.
9 P. Lozano, D. Combes, J. L. Iborra and A. Manjon, Biotechnol.
Lett., 1993, 15, 1223.
10 M. Reslow, P. Adlercreutz and B. Mattiasson, Eur. J. Biochem., 1988,
172, 573; P. Adlercreutz, Eur. J. Biochem., 1991, 199, 609.
11 K. Braun and P. Kuhl, Pharmazie, 1997, 52, 203.
12 N. Harper, B. D. Moore and P. J. Halling, Tetrahedron Lett., 2000,
41, 4223.
13 P. Clapes and P. Adlercreutz, Biochim. Biophys. Acta, 1991, 1118,
70; S. Calvet, P. Clapes, J. L. Torres, G. Valencia, J. Feixas and
P. Adlercreutz, Biochim. Biophys. Acta, 1993, 1164, 189.
14 C. F. Barbas and C.-H. Wong, J. Chem. Soc., Chem. Commun., 1987,
533; D. Cantacuzène, F. Pascal and C. Guerreiro, Tetrahedron, 1987,
43, 1823; D.-F. Tai, S.-L. Fu, S.-F. Chuang and H. Tsai, Biotechnol.
Lett., 1989, 11, 173; N. Xaus, P. Clapés, E. Bardají, J. L. Torres,
X. Jorba, J. Mata and G. Valencia, Tetrahedron, 1989, 45, 7421.
15 C. Chamorro, R. Gonzalez-Muniz and S. Conde, Tetrahedron:
Asymmetry, 1995, 6, 2343; S. Conde, P. Lopezserrano, M. Fierros,
M. I. Biezma, A. Martinez and M. I. Rodriguezfranco, Tetrahedron,
1997, 53, 11745.
16 D. Cantacuzene and C. Guerreiro, Tetrahedron, 1989, 45, 741.
17 L. Esteban-Cerdán, A. Robles-Medina, A. Gimenez-Gimenez,
M. J. Ibáñez-González and E. Molina-Grima, J. Am. Oil Chem.
Soc., 1998, 75, 1329; A. Robles-Medina, L. Esteban-Cerdán,
A. Gimenez-Gimenez, B. Camacho-Páez, M. J. Ibáñez-González
and E. Molina-Grima, J. Biotechnol., 1999, 70, 379; C. Virto and
P. Adlercreutz, Enzyme Microb. Technol., 2000, 26, 630.
18 A. M. Fureby, C. Virto, P. Adlercreutz and B. Mattiasson, Biocatal.
Biotransform., 1996, 14, 89.
19 P. E. Sonnet and R. L. Dudley, Chem. Phys. Lipids, 1994, 72, 185.
20 E. Montero, J. Alonso, F. J. Cañada, A. Fernandez-Mayoralas and
M. Martin-Lomas, Carbohydr. Res., 1998, 305, 383.
2070
J. Chem. Soc., Perkin Trans. 1, 2001, 2063–2070