Chemo-enzymatic synthesis of amino acid-based surfactants

Article abstract of DOI:10.1007/s11746-997-0232-8

The application of lipases to the synthesis of amino acid-based surfactants was investigated. Low yields (2-9%) were obtained in the acylation of free amino acids, such as L-serine and L-lysine, as well as their ethyl esters and amides with fatty acids, o

Full text of DOI:10.1007/s11746-997-0232-8

Chemo-Enzymatic Synthesis of Amino
Acid-Based Surfactants
Rao Valivety, Paula Jauregi, Iqbal Gill, and Evgeny Vulfson*
FMS Department, BBSRC Institute of Food Research, Earley Gate, Reading RG6 6BZ, United Kingdom
ABSTRACT: The application of lipases to the synthesis of starting materials. This point has been clearly demonstrated
amino acid-based surfactants was investigated. Low yields
(2–9%) were obtained in the acylation of free amino acids, such
by a number of recent reports dealing with the application of
enzymes to synthesis and/or modification of sugar fatty acid
esters, phospholipids, and alkyl glycosides (5). This paper is
aimed at exploring the usefulness of enzymes in the prepara-
tion of yet another interesting group of surface-active agents,
amino acid-based surfactants.
Recently, amino acid esters and amides have been a sub-
ject of intensive investigations owing to their excellent emul-
sifying properties, biocompatibility, and strong antimicrobial
activity (6–14). These features have made them attractive for
as
with fatty acids, owing in part to low miscibility of the reactants.
When the N-carbobenzyloxy (Cbz)- -amino acids were used in
L-serine and L-lysine, as well as their ethyl esters and amides
L
an effort to improve miscibility of the amino acid derivatives
with the acyl donor, a dramatic improvement was observed for
N-Cbz-
sine (7 and 2% yield, respectively). As an alternative, an effi-
cient synthesis of Nζ-acyl- -lysines was developed, based on
the regiospecific chemical acylation of copper(II) lysinate. In
L
-serine (92% yield) but not for Nα-Cbz- or Nζ-Cbz-L-ly-
L
pursuit of a general route to amino acid–fatty acid surfactants, applications in cosmetics and personal-care products, food,
the utility of a polyol linker was investigated. Thus, the glycerol
ester of Nα,Nζ-di-Cbz- -lysine was prepared and evaluated as a
substrate for acylation. As expected, this and other glycer-1-yl
esters of N-protected amino acids were excellent substrates for
lipase-catalyzed acylation. Their reaction with myristic acid in
the presence of Novozyme resulted in the regioselective acyla-
tion of the primary hydroxyl group of the glycerol moiety to af-
and pharmaceutical formulations. Additionally, there are
good reasons for considering amino acid-based surfactants as
potential bulk detergents for industrial and household clean-
ing usages. Indeed, their production would rely on the use of
renewable and inexpensive raw materials (e.g., amino and
fatty acids) with expected rapid and complete biodegradabil-
ity in the environment. Furthermore, the structural diversity
of amino acids should enable manufacturers to tailor the func-
tional properties of these surfactants to suit particular appli-
cations.
L
ford the corresponding 1-O-(N-Cbz-L-aminoacyl)-3-O-myris-
toylglycerols with conversions of 50–90%. These were readily
deprotected to give a range of 1-O-(aminoacyl)-3-O-myristoyl-
glycerols with overall yields of 27–71%.
JAOCS 74, 879–886 (1997).
The wide range of functionalities displayed by amino acids
should be a great asset where any “fine-tuning” of the surfac-
tant’s performance is concerned. However, this characteristic
is also a potential drawback from a synthetic viewpoint owing
to the requirement for temporary protection of any reactive
functional groups. It is somewhat surprising therefore to find
that relatively few attempts to circumvent some of the syn-
thetic difficulties, by utilizing the selectivity of enzymes,
have been reported (15–17). In this communication, we re-
port the results of our initial investigations, aimed at the de-
velopment of a facile chemo-enzymatic approach to the
preparation of amino acid-based surfactants. In particular, the
usefulness of lipases as catalysts for acylation of glycerol es-
ters of amino acid was investigated in some detail.
KEY WORDS: Amidation, amino acids, copper(II) lysinate,
enzymatic synthesis, fatty acids, glycerol ester, lipase, regiose-
lectivity, solvent-free acylation, surfactants.
Surfactants constitute an important class of bulk and special-
ity chemicals with applications in many sectors of industry.
Until recently, the synthesis of surfactants has been consid-
ered solely within the capabilities of organic chemistry. How-
ever, the rapid advance of biotechnology has led to a surge of
interest in “biosurfactants,” i.e., surface-active agents pro-
duced via mediation of biological catalysts in the form of mi-
croorganisms (1–4) or isolated enzymes (5). The main advan-
tages associated with the use of biocatalysts are mild reaction
conditions and high enzymatic specificity, which often elimi-
nates the need for regioselective protection of multifunctional
EXPERIMENTAL PROCEDURES
Materials. N-Carbobenzyloxy (Cbz)-glycine, N-Cbz-
sine, N-Cbz- -phenylalanine, N-Cbz- -aspartic acid, N-Cbz-
serine, and Nα,Nζ-di-Cbz- -lysine were obtained from Sigma
L-tyro-
L
L
L-
*To whom correspondence should be addressed at FMS Department,
BBSRC Institute of Food Research, Earley Gate, Whiteknights Rd., Reading
RG6 6BZ, United Kingdom. E-mail: jenya.vulfson@bbsrc.ac.uk.
L
Chemical Company (Poole, Dorset, England). Novozyme (im-
Copyright © 1997 by AOCS Press
879
JAOCS, Vol. 74, no. 7 (1997)

880
R. VALIVETY ET AL.
mobilized Candida antarctica lipase) and Lipozyme (immobi- vals, quenched with methanol containing 5% vol/vol acetic
lized Mucor miehei lipase) were purchased from Novo Nordisk acid, and analyzed by HPLC.
A/S (Bagsvaerd, Denmark).
caproyl chloride, lauroyl chloride, myristoyl chloride, palmi- lysinato)copper(II) complex was prepared by adding aqueous
toyl chloride, stearoyl chloride, myristic acid, glycerol, formic potassium -lysinate (30 mL of 3.3 M solution, 100 mmol) to
L
-Lysine, capryloyl chloride,
Preparation of Nζ-acyl-
L-lysines. Aqueous bis(L-
L
acid, palladium on alumina catalyst (10% w/w), boron trifluo- ice-cold aqueous copper(II) sulfate (40 mL of 1.3 M solution,
ride etherate, 2,2-dimethyl-1,3-dioxolane-4-methanol (solke- 52 mmol) with vigorous stirring and cooling in an ice bath.
tal), dicyclohexylcarbodiimide (DCC), trifluoroacetic acid, d6- Solid potassium bicarbonate (15 g, 150 mmol) was then added
dimethylsulfoxide (d6-DMSO), EDTA disodium salt, and basic with stirring, followed by dropwise addition of the acid chloride
alumina (Brockmann grade 1) were acquired from Aldrich (125 mmol in 10 mL of acetone) over a period of 1.5–2 h. The
Chemical Company (Poole, Dorset, England). All solvents em- mixture was left stirring in the ice bath for a further 2–3 h, then
ployed were of the highest available purity and were dried over allowed to warm to room temperature; stirring was continued
3A molecular sieves before use.
for another 3–5 h, to give a viscous solution or suspension of
-lysinato)copper(II) complex. The mixture was
Analytical methods. Reactions were followed by reverse- the bis(Nζ-acyl-
L
phase high-performance liquid chromatography (HPLC), with then heated to 50°C, an aqueous suspension of EDTA disodium
a Gilson 305/306 system (Anachem Luton, United Kingdom), salt (500 mL of 0.25 M, 125 mmol) was added with vigorous
connected with a Gilson 231 Autosampler, an Applied Chro- stirring, and the pH of the mixture was adjusted to between 4.0
matography Systems (ACS; Cheshire, United Kingdom) and 6.0 with 2.0 M aqueous hydrochloric acid. The precipitate
750/12 ultraviolet detector or an ACS 750/14 light-scattering formed was filtered off and washed successively with 2 × 50 mL
detector, and a Hewlett-Packard Chemstation 35900 (Win- water, 2 × 50 mL of 1:1 water/ethanol, 1 × 50 mL ethanol, 1 ×
nersh, United Kingdom) for data acquisition and processing. 50 mL of 1:1 ethanol/ether, and finally 4 × 50 mL ether, to give
Analyses were carried out on a 0.46 × 15 cm Hichrom RPB/5- the required Nζ-acyl-
µm column (Reading, United Kingdom), maintained at 45°C The analytical data are presented below for a representative
and operated at a flow rate of 1 mL · min−1. Methanol and 4:1 member of this class of compounds, Nζ-myristoyl-
-lysine (1):
water/methanol, both containing 0.05% vol/vol phosphoric Nζ-Myristoyl- -lysine (1). M.p. > 230°C (decomp.); [α]D,25
L-lysine as a white solid (51–64% yield).
L
L
acid, were employed as mobile phases. Fast atom bombard- + 10.07° (c 1.38, 1:1 methanol/trifluoroacetic acid); FAB-MS:
ment mass spectroscopy (FAB-MS) was carried out on a Kratos (M + H) calcd. for C20H41N2O3 357.3117, (M + H) obsvd.
MS9/50TC spectrometer (Manchester, United Kingdom) with 357.3070.
xenon at 5–7 kV and a source potential of 5.33 kV. Glycerol
was used as the matrix, and spectra were recorded at 0.1 mm in Cbz-glycyl)-, 1-O-(N-Cbz-
positive or negative ionization mode, with polyglycerol ions as rosyl)-, 1-O-(N-Cbz- -seryl)-, and 1-O-(Nα,Nζ-di-Cbz-
reference. Optical rotations were measured on a Thorn NPL- glycerol were synthesized from the corresponding N-Cbz-
type 243 polarimeter (Nottingham, United Kingdom) at 589.3 amino acids by esterification with glycerol and boron trifluorode
nm, 25°C, in a 0.5-cm cell, with methanol or 1:1 methanol/tri- etherate as catalyst (18). N-Cbz- -amino acid (50 mmol) was
Preparation of 1-O-(N-Cbz-
-phenylalanyl)-, 1-O-(N-Cbz-
-lysyl)-
L-aminoacyl)glycerols. 1-O-(N-
L
L-ty-
L
L
L
-
L
fluoroacetic acid as the solvent. Melting points were deter- dissolved in a mixture of dimethylformamide (50 mL) and glyc-
mined with a Stuart Scientific SMP-1 apparatus (Red Hill, erol (500 mL). The solution was heated to 60°C, boron trifluo-
United Kingdom) using sealed capillaries and are uncorrected. ride etherate (25 mL) was added over a period of 1 h to the
1
All compounds were also fully characterized by H, 13C, stirred mixture, and the reaction was allowed to proceed for 20
DEPT, 1H-1H correlation spectroscopy (COSY), and 1H-13C h. Brine (300 mL) and 0.6 M sodium bicarbonate (400 mL) was
COSY nuclear magnetic resonance (NMR) experiments. These then added, and the reaction mixture was extracted with 3 ×
spectra were recorded on a Jeol JNM EX 270 Fourier transform 500-mL portions of ethyl acetate. The organic layers were
spectrometer (Tokyo, Japan) at 35°C, with d6-DMSO as sol- pooled and washed successively with 4 × 50 mL of aqueous 0.1
vent. Proton spectra were recorded at 270.0 MHz and carbon- M sodium bicarbonate containing 0.2 M sodium chloride, 4 ×
13 spectra at 67.8 MHz with broad-band proton decoupling, 50 mL of aqueous 0.1 M citric acid containing 0.2 M sodium
both by using a DMSO field lock. 1H-1H COSY spectra were chloride, and finally with 4 × 50 mL of 0.4 M sodium chloride.
recorded in symmetrized mode.
General procedure for acylation of free amino acids and evaporation, gave the pure product in 80–90% yield.
their derivatives. The acylation of -serine, -lysine, and their 1-O-(N-Cbz-glycyl)glycerol (2a). M.p. 68–70°C; [α]D,25
Drying over anhydrous magnesium sulfate, followed by rotary
L
L
amide and methyl/ethyl ester derivatives was conducted as 0° (c 1.10, methanol); FAB-MS: (M + H) calcd. for
follows. The amino acid substrate (1 mmol) was dispersed in C13H17NO6 284.1137, (M + H) obsvd. 284.1104.
fatty acid or fatty acid methyl ester (2 or 4 mmol) at 50, 60,
1-O-(N-Cbz-L-phenylalanyl)glycerol (3a). M.p. 43–44°C;
or 70°C, with or without the addition of solvent (5 or 10 mL [α]D,25−9.43° (c 1.96, methanol); FAB-MS: (M + H) calcd.
where employed). Immobilized lipase (100–500 mg) was for C20H24NO6 374.1606, (M + H) obsvd. 374.1588.
then added, and the mixture was stirred vigorously at 50, 60
1-O-(N-Cbz-L-tyrosyl)glycerol (4a). M.p. 70–72°C;
or 70°C by means of a heating mantle/stirrer. Samples for [α]D,25−5.00° (c 1.57, methanol); FAB-MS: (M + H) calcd.
HPLC analysis were withdrawn at the appropriate time inter- for C20H24NO7 390.1555, (M + H) obsvd. 390.1519.
JAOCS, Vol. 74, no. 7 (1997)

CHEMO-ENZYMATIC SYNTHESIS OF AMINO ACID-BASED SURFACTANTS
881
1-O-(N-Cbz-
(c 1.72, methanol); FAB-MS: (M + H) calcd. for C14H20NO7 crude product as a semisolid. This product was then dissolved
314.1242, (M + H) obsvd. 314.1260. in ether, and applied to a basic alumina column (3 × 50 cm with
1-O-(Nα,Nζ-Di-Cbz- -lysyl)glycerol (6a). M.p. 47–48°C; 400 g of alumina, Brockmann grade II), and purified by eluting
[α]D,25 −8.35° (c 2.10, methanol); FAB-MS: (M + H) calcd. with ether [for 1-O-(N-Cbz- -tyrosyl)-, 1-O-(N-Cbz- -pheny-
for C25H33N2O8 489.2242, (M + H) obsvd. 489.2238. lalanyl)-, and 1-O-(N-Cbz- -seryl)-3-O-myristoylglycerol].
Preparation of α,γ-diglycer-1-yl-N-Cbz- -aspartate (7a). For 1-O-(Nα,Nζ-di-Cbz- -lysyl)-3-O-myristoylglycerol, 1-O-
The direct esterification of N-Cbz- -aspartic acid with glyc- (N-Cbz-glycyl)-3-O-myristoylglycerol, and α,γ-di(3-O-myris-
erol was unsuitable for the preparation of α,γ-di-glycer-1-yl- toylglycer-1-yl)-N-Cbz- -aspartate, the product was eluted
N-Cbz- -aspartate, because of its decomposition to a mixture with pure ethanol.
of monoesters during the extraction procedures. Therefore, 1-O-(N-Cbz-glycyl)-3-O-myristoylglycerol (2b). M.p.
the following route, based on the disolketal ester, was em- 81–82°C; [α]D,25 0° (c 1.01, methanol); FAB-MS: (M + H)
ployed. N-Cbz- -aspartic acid (20 g, 75 mmol) and 4-(N,N- calcd. for C27H42NO7 492.2963, (M + H) obsvd. 492.2990.
dimethylamino)pyridine (DMAP, 0.4 g, 3 mmol) were dis- 1-O-(N-Cbz- -phenylalanyl)-3-O-myristoylglycerol (3b).
L
-seryl)glycerol (5a). M.p. Oil; [α]D,25−6.56° the extract was rotary-evaporated to dryness at 30oC to give the
L
L
L
L
L
L
L
L
L
L
L
solved in a mixture of solketal (38 mL, 300 mmol) and 400 M.p. Oil; [α]D,25 −6.73° (c 1.17, methanol); FAB-MS: (M + H)
mL of acetonitrile. DCC (34 g, 165 mmol), dissolved in ace- calcd. for C34H50NO7 584.3590, (M + H) obsvd. 584.3617.
tonitrile (50 mL), was then added over a period of 0.5 h to the
1-O-(N-Cbz-L-tyrosyl)-3-O-myristoylglycerol (4b). M.p.
stirred solution, and the reaction was allowed to proceed over oil; [α]D,25 −3.80° (c 1.19, methanol); FAB-MS: (M + H)
a period of 24 h. Glacial acetic acid (10 mL) was then added, calcd. for C34H50NO8 600.3539, (M + H) obsvd. 600.3527.
and the mixture was stirred for a further 0.5 h, after which
1-O-(N-Cbz-L-seryl)-3-O-myristoylglycerol (5b). M.p. oil;
time it was filtered and rotary-evaporated to a syrup. This was [α]D,25 −4.17° (c 1.60, methanol); FAB-MS: (M + H) calcd.
dissolved in 500 mL ethyl acetate and washed successively for C28H46NO8 524.3235, (M + H) obsvd. 524.3246.
with 4 × 50 mL of aqueous 0.05 M potassium bicarbonate
1-O-(Nα,Nζ-Di-Cbz- -lysyl)-3-O-myristoylglycerol (6b).
L
containing 0.3 M sodium chloride, 4 × 50 mL of aqueous 0.05 M.p. 28–29°C; [α]D,25 −6.20° (c 2.19, methanol); FAB-MS:
M citric acid containing 0.3 M sodium chloride, and finally 4 (M + H) calcd. for C39H59N2O9 699.4226, (M + H) obsvd.
× 50 mL of 0.4 M sodium chloride (0.05 M). The organic 699.4245.
layer was dried over anhydrous magnesium sulfate, then ro-
Di-(3-O-myristoylglycer-1-yl)-N-Cbz-L-aspartate (7b).
tary-evaporated to give the pure product as a syrup. Further M.p. oil; [α]D,25 +1.84° (c 2.23, methanol); FAB-MS: (M +
purification by flash column chromatography on silica gel (3 H) calcd. for C46H77NO12 836.5524, (M + H) obsvd.
× 50 cm column containing 400 g of silica gel 40/60) with 7:1 836.5546.
petroleum ether/ethanol as eluent gave the pure α,γ-di(2,3-
isopropylideneglycer-1-yl)-N-Cbz- -aspartate as a colorless glycerols and α,γ-di-(3-O-myristoylglycer-1-yl)-N-Cbz-
syrup (25.9 g, 70% yield). The disolketal ester was depro- partate. The N-Cbz-derivatives were deprotected by transfer-
tected by dissolving it in 200 mL of 4:1 acetonitrile/1 M aque- hydrogenation with formic acid as donor. 1-O-(N-Cbz-
ous hydrochloric acid, followed by stirring at room tempera- aminoacyl)-3-O-acyl-glycerol or α,γ-di-(3-O-myristoyl-
ture for 6 h. The solution was then concentrated by rotary glycer-1-yl)-N-Cbz- -aspartate (300 mg) was dissolved in
evaporation at 30°C, followed by freeze-drying, to give α,γ- 4% vol/vol formic acid in methanol (10 mL). Palladium on
diglycer-1-yl-N-Cbz-L-aspartate as a colorless viscous liquid alumina catalyst [300 mg, except for 1-O-(Nα,Nζ-di-Cbz-
Deprotection of 1-O-(N-Cbz-
L
-aminoacyl)-3-O-myristoyl-
L
L
-as-
L
-
L
L
-
(21.4 g, 98% deprotection yield, 69% overall yield). M.p. oil; lysyl)-3-O-myristoylglycerol, where 600 mg was employed)
[α]D,25 −6.80° (c 1.04, methanol); FAB-MS: (M + H) calcd. was then added, and the mixture was stirred at room tempera-
for C18H26NO10 416.1561, (M + H) obsvd. 416.1556.
ture until deprotection was complete (typically less than 0.5
h). The reaction mixture was then filtered through a bed of
Solvent-free enzymatic esterification of 1-O-(N-Cbz-L-
aminoacyl)glycerols and α,γ-diglycer-1-yl-N-Cbz-
tate. The 1-O-(N-Cbz- -aminoacyl)glycerols and α,γ- at 30°C. The syrup obtained was then diluted with water (10
diglycer-1-yl-N-Cbz- -aspartate were enzymatically esteri- mL) and freeze-dried, and the solid obtained was purified by
fied as follows. The glycer-1-yl ester (20 mmol) was crystallization from petroleum ether, to give pure 1-O-
thoroughly mixed with pentan-3-one (20% w/w), myristic aminoacyl-3-O-myristoylglycerol or α,γ-di(3-O-myristoyl-
acid (40 mmol, twofold excess except for α,γ-diglycer-1-yl- glycer-1-yl)- -aspartate in almost quantitative yield as a
N-Cbz- -aspartate, where a threefold excess of the fatty acid white solid.
was employed), and Novozyme (2% w/w of substrate mix- 1-O-Glycyl-3-O-myristoylglycerol (formate salt, 2c). M.p.
L-aspar- Celite, and the filtrate was concentrated by rotary evaporation
L
L
L
-
L
L
ture) in a widemouthed reaction flask, which was then trans- 60–61°C; [α]D,25 0° (c 0.85, methanol); FAB-MS: (M + H)
ferred to a heating mantle–stirrer maintained at 70°C. After calcd. for C19H37NO5 360.2747, (M + H) obsvd. 360.2734.
25–60 h, over 95% of the glycer-1-yl ester substrate had re-
1-O-L-Phenylalanyl-3-O-myristoylglycerol (formate salt,
acted, and the conversion to the 3-O-myristoyl ester had 3c). M.p. 63–64°C; [α]D,25 +5.50° (c 1.03, methanol); FAB-
reached 50–90%. The reaction mixture was then extracted MS: (M + H) calcd. for C26H43NO5 450.3219, (M + H)
with methanol, the immobilized enzyme was filtered off, and obsvd. 450.3225.
JAOCS, Vol. 74, no. 7 (1997)

882
R. VALIVETY ET AL.
1-O-
L
-Tyrosyl-3-O-myristoylglycerol (formate salt, 4c).
1-O-(N-Cbz-L
-tyrosyl)glycerol (4a). 1H NMR, δ: 2.90 (m,
M.p. 69–71°C; [α]D,25 +4.67° (c 0.96, methanol); FAB-MS: (M 1H, Tyr-Hβ1), 3.15 (m, 1H, Tyr-Hβ2), 3.50 (d, 2H, J = 5.3 Hz,
+ H) calcd. for C26H43NO6 466.3163, (M + H) obsvd. 466.3140. Glc-H3), 3.78 (t, 1H, J = 5.1 Hz, Glc-H2), 4.15 (m, 2H, Glc-
1-O- -Seryl-3-O-myristoylglycerol (formate salt, 5c). M.p. H1), 4.40 (m, 1H, Tyr-Hα), 5.10 (s, 2H, Cbz-CH2), 6.80 (d,
L
40–41°C; [α]D,25 +3.50° (c 0.43, methanol); FAB-MS: (M + 2H, Jo,m = 8.6 Hz, Tyr-Arom-Ho), 7.14 (d, 2H, Jo,m = 8.6 Hz,
H) calcd. for C20H40NO6 390.2852, (M + H) obsvd. 390.2860. Tyr-Arom-Hm), 7.33–7.44 (m, 5H, Cbz-Arom-H), 7.82 (d,
1-O-
-Lysyl-3-O-myristoylglycerol (diformate salt, 6c). 1H, Jα,NH = 8.3 Hz, Tyr-NH); 13C NMR, δ: 36.6 (Tyr-Cβ),
L
M.p. 58–59°C; [α]D,25 +3.96° (c 0.37, 1:1 methanol/trifluo- 55.9 (Tyr-Cα), 62.6 (Glc-C3), 65.4 (Cbz-CH2), 66.2 (Glc-C1),
roacetic acid); FAB-MS: (M + H) calcd. for C23H46N2O5 69.3 (Glc-C2), 115.1/127.2/127.7/128.1/130.0/136.9 (Tyr-
431.3485, (M + H) obsvd. 431.3479.
Di-(3-O-myristoylglycer-1-yl)- -aspartate (formate salt,
Arom-C + Cbz-Arom-C), 156.0/171.6 (carbonyl).
1-O-(N-Cbz-
-seryl)glycerol (5a). 1H NMR, δ: 3.50 (m, 2H,
L
L
7c). M.p. 39–41°C; [α]D,25 −0.68° (c 1.49, methanol); FAB- Glc-H3), 3.60 (m, 1H, Glc-H2), 3.80 (m, 2H, Ser-Hβ), 4.20 (t,
MS: (M + H) calcd. for C38H71NO10 702.5156, (M + H) 1H, Jα,NH = 4.9 Hz, Ser-Hα), 4.70 (m, 2H, Glc-H1), 5.16 (s, 2H,
obsvd. 702.5171.
Cbz-CH2), 7.33–7.42 (m, 5H, Cbz-Arom-H); 13C NMR, δ:
Preparation of 3-O-myristoyl-N-Cbz-
L
-serine (8a). N- 57.0 (Ser-Cα), 61.6 (Glc-C3), 62.8 (Ser-Cβ), 65.9 (Cbz-CH2),
Cbz- -Serine (4.78 g, 20 mmol) was mixed with myristic acid 69.5 (Glc-C2), 72.4 (Glc-C1), 127.0/127.8/128.5/137.0 (Cbz-
L
(9.14 g, 40 mmol), pentan-3-one (5 mL) and Novozyme (0.5 Arom-C), 156.0/170.0 (carbonyl).
g), and the mixture was incubated with stirring in an open 1-O-(Nα,Nζ-Di-Cbz-
L
-lysyl)glycerol (6a). 1H-NMR, δ: 1.24
flask at 70°C. After 60 h, when the esterification level had (m, 2H, Lys-Hβ), 1.78 (m, 2H, Lys-Hδ), 2.50 (m, 2H, Lys-Hγ),
reached 93%, the reaction mixture was dissolved in ether (100 3.10 (t, 2H, Jδ,ζ = 6.0 Hz, Lys-Hζ), 3.50 (m, 4H, Glc-H1 + Glc-
mL) and filtered. Rotary evaporation, followed by recrystal- H3), 3.80 (m, 1H, Glc-H2), 4.20 (m, 1H, Lys-Hα), 5.13 (s, 2H,
lization from petroleum ether, afforded the title product as a Cbzζ-CH2), 5.16 (Cbzα-CH2), 7.30 (m, 1H, Lys-NHζ),
white solid (7.2 g, 80% yield). M.p. 49–50°C; [α]D,25 +3.47° 7.35–7.45 (m, 10H, Cbzα-Arom-H + Cbzζ-Arom-H), 7.70 (d,
(c 1.08, methanol); FAB-MS (M + H) calcd. for C25H40NO6 1H, Jα,NH = 8.0 Hz, Lys-NHα); 13C-NMR, δ: 22.8 (Lys-Cβ),
450.2858, (M + H) obsvd. 450.2846.
3-O-Myristoyl- -serine (formate salt, 8b). 3-O-Myristoyl- 63.3 (Glc-C3), 65.2 (Cbzζ-CH2), 65.5 (Cbzα-CH2), 69.4 (Glc-
N-Cbz- -serine was deprotected by transfer hydrogenation C1), 72.4 (Glc-C2), 127.0/127.3/128.0/136.9/137.3/138.7
with formic acid as described above, to give 3-O-myristoyl- (Cbzα-Arom-C + Cbzζ-Arom-C), 156.3/172.5 (carbonyl).
-serine formate salt in quantitative yield. M.p. 43–44°C; α,γ-Diglycer-1-yl-N-Cbz-
-aspartate (7a). 1H NMR, δ: 2.90
[α]D,25 −0.90° (c 0.55, methanol); FAB-MS: (M + H) calcd. (d, 2H, Jα,β = 7.7 Hz, Asp-Hβ), 3.40 (m, 4H, Glcα-H3 + Glcγ-H3),
for C17H33NO4 316.2488, (M + H) obsvd. 316.2476. 3.75 (m, 2H, Glcα-H2 + Glcγ-H2), 4.10 (m, 4H, Glcα-H1 + Glcγ-
Nζ-Myristoyl-
29.0 (Lys-Cδ), 30.6 (Lys-Cγ), 40.4 (Lys-Cζ), 55.1 (Lys-Cα),
L
L
L
L
L
-lysine (1). 1H NMR, δ: 0.82 (t, 3H, J = 6.6 H1), 4.50 (m, 1H, Asp-Hα), 5.10 (s, 2H, Cbz-CH2), 7.32–7.42 (m,
Hz, Myr-H14), 1.22 (m, 22H, Myr-H4–13), 1.50 (m, 6H, Myr- 5H, Cbz-Arom-H); 13C NMR, δ: 36.4 (Asp-Cβ), 51.0 (Asp-Cα),
H3 + Lys-Hγ + Lys-Hδ), 1.82 (m, 2H, Lys-Hβ), 2.20 (t, 2H, J 63.0 (Glcγ-C1), 63.5 (Glcα-C1), 66.2 (Cbz-CH2), 66.9 (Glcα-C3 +
= 7.9 Hz, Myr-H2), 3.20 (m, 2H, Lys-Hζ), 3.80 (m, 1H, Lys- Glcγ-C3), 69.7 (Glcα-C2 + Glcγ-C2), 128.1/128.4/128.9/137.2
Hα), 8.20 (m-broad, 3H, Lysα-NH + Lysζ-NH); 13C NMR, δ: (Cbz-Arom-C), 156.4/170.5/171.4 (carbonyl).
13.2 (Myr-C14), 21.6/21.9 (Myr-C13 + Lys-Cβ), 25.5 (Myr-
1-O-(N-Cbz-glycyl)-3-O-myristoylglycerol (2b). 1H NMR,
C3), 28.0 (Lys-Cγ), 28.5–29.0 (Myr-C4–11), 29.5 (Lys-Cδ), δ: 0.85 (t, 3H, J = 6.4 Hz, Myr-H14), 1.24 (m, 20H, Myr-
31.3 (Myr-C12), 35.0 (Myr-C2), 38.6 (Lys-Cζ), 52.3 (Lys-Cα), H4–13), 1.52 (m, 2H, Myr-H3), 2.30 (t, 2H, J = 7.3 Hz, Myr-
171.0/175.0 (carbonyl).
H2), 3.85 (d, 2H, Jα,NH = 6.0 Hz, Gly-Hα), 4.00 (m, 1H, Glc-
1-O-(N-Cbz-glycyl)glycerol (2a). 1H NMR, δ: 3.45 (m, 2H, H2), 4.10 (m, 4H, Glc-H1 + Glc-H3), 5.10 (s, 2H, Cbz-CH2),
Glc-H3), 3.75 (m, 1H, Glc-H2), 3.85 (d 2H, Jα,NH = 6.3 Hz, Gly- 7.31–7.42 (m, 5H, Cbz-Arom-H), 7.60 (t, 1H, Jα,NH = 6.0 Hz,
Hα), 4.10 (m, 2H, Glc-H1), 5.15 (s, 2H, Cbz-CH2), 7.33–7.45 Gly-NH); 13C NMR, δ: 14.1 (Myr-C14), 22.2 (Myr-C13), 24.6
(m, 5H, Cbz-Arom-H), 7.60 (t, 1H, Jα,NH = 6.1 Hz, Gly-NH); (Myr-C3), 28.4–29.2 (Myr-C4–11), 31.5 (Myr-C12), 33.6 (Myr-
13C NMR, δ: 42.1 (Gly-Cα), 62.5 (Glc-C3), 65.5 (Cbz-CH2), C2), 42.9 (Gly-Cα), 64.9 (Glc-C3), 65.7 (Glc-C1 + Cbz-CH2),
66.1 (Glc-C1), 69.2 (Glc-C2), 127.0/128.0/128.5/137.0 (Cbz- 69.2 (Glc-C2), 127.8/128.2/128.5/137.9 (Cbz-Arom-C),
Arom-C), 156.0/170.0 (carbonyl).
156.7/170.2/172.9 (carbonyl).
1-O-(N-Cbz-
L
-phenylalanyl)glycerol (3a).1H NMR, δ:
1-O-(N-Cbz- -phenylalanyl)-3-O-myristoylglycerol (3b).
L
2.95 (m, 1H, Phe-Hβ1), 3.25 (m, 1H, Phe-Hβ2), 3.50 (d, 2H, J 1H NMR, δ: 0.94 (t, 3H, J = 6.6 Hz, Myr-H14), 1.30 (m, 20H,
= 5.0 Hz, Glc-H3), 3.80 (m, 1H, Glc-H2), 4.15 (m, 2H, Glc- Myr-H4–13), 1.60 (m, 2H, Myr-H3), 2.33 (t, 2H, J = 7.3 Hz,
H1), 4.45 (m, 1H, Phe-Hα), 5.10 (s, 2H, Cbz-CH2), 7.30–7.44 Myr-H2), 3.05 (m, 1H, Phe-Hβ1), 3.20 (m, 1H, Phe-Hβ2), 4.00
(m, 10H, Phe-Arom-H + Cbz-Arom-H), 7.82 (d, 1H, Jα,NH
=
(m, 1H, Glc-H2), 4.10–4.20 (m, 4H, Glc-H1 + Glc-H3), 4.40
8.3 Hz, Phe-NH); 13C NMR, δ: 36.6 (Phe-Cβ), 55.6 (Phe-Cα), (m, 1H, Phe-Hα), 5.06 (s, 2H, Cbz-CH2), 7.30–7.40 (m, 10H,
62.6 (Glc-C3), 65.4 (Cbz-CH2), 66.3 (Glc-C1), 69.3 (Glc-C2), Phe-Arom-H + Cbz-Arom-H), 7.82 (d, 1H, Jα,NH = 8.0 Hz,
126.5/129.0/136.9/137.5 (Phe-Arom-C + Cbz-Arom-C), Phe-NH); 13C NMR, δ: 13.7 (Myr-C14), 22.0 (Myr-C13), 24.4
156.0/171.7 (carbonyl).
(Myr-C3), 28.5–29.0 (Myr-C4–11), 31.3 (Myr-C12), 33.4 (Myr-
JAOCS, Vol. 74, no. 7 (1997)

CHEMO-ENZYMATIC SYNTHESIS OF AMINO ACID-BASED SURFACTANTS
883
C2), 36.6 (Phe-Cβ), 55.4 (Phe-Cα), 64.6 (Glc-C3), 65.1 (Glc- NMR, δ: 0.94 (t, 3H, J = 6.3 Hz, Myr-H14), 1.33 (m, 20H,
C1), 65.5 (Cbz-CH2), 66.1 (Glc-C2), 126.3/128.9/136.7/137.3 Myr-H4–13), 1.61 (m, 2H, Myr-H3), 2.37 (t, 2H, J = 7.3 Hz,
(Phe-Arom-C + Cbz-Arom-C), 155.8/171.8/172.1 (carbonyl). Myr-H2), 3.60 (m, 2H, Gly-Hα), 3.86 (m, 1H, Glc-H2), 4.05
1-O-(N-Cbz-
-tyrosyl)-3-O-myristoylglycerol (4b). 1H (m, 4H, Glc-H1 + Glc-H3); 13C NMR, δ: 13.7 (Myr-C14), 21.9
L
NMR, δ: 0.94 (t, 3H, J = 6.3 Hz, Myr-H14), 1.32 (m, 20H, Myr- (Myr-C13), 24.3 (Myr-C3), 28.6–28.9 (Myr-C4–11), 31.4 (Myr-
H4–13), 1.60 (m, 2H, Myr-H3), 2.30 (m, 2H, Myr-H2), 2.90 (m, C12), 33.4 (Myr-C2), 44.2 (Gly-Cα), 62.6 (Glc-C3) 65.3 (Glc-
1H, Tyr-Hβ1), 3.05 (m, 1H, Tyr-Hβ2), 4.00 (m, 1H, Glc-H2), C1), 69.2 (Glc-C2), 166.0/173.0 (carbonyl).
4.15 (m, 4H, Glc-H1 + Glc-H3), 4.35 (m, 1H, Tyr-Hα), 5.10 (s,
1-O-L-Phenylalanyl-3-O-myristoylglycerol (formate salt,
2H, Cbz-CH2), 6.75 (d, 2H, Jo,m = 8.3 Hz, Tyr-Arom-Ho), 7.10 3c). 1H NMR, δ: 0.94 (t, 3H, J = 6.6 Hz, Myr-H14), 1.32 (m,
(d, 2H, Jo,m = 8.3 Hz, Tyr-Arom-Hm), 7.30–7.40 (m, 5H, Cbz- 20H, Myr-H4–13), 1.60 (m, 2H, Myr-H3), 2.35 (t, 2H, J = 7.4
Arom-H), 7.82 (d, 1H, Jα,NH = 8.6 Hz, Tyr-NH); 13C NMR, δ: Hz, Myr-H2), 2.85 (dd, 1H, Jβ1,β2 = 14.0 Hz, Jα,β1 = 7.3 Hz,
13.8 (Myr-C14), 22.0 (Myr-C13), 24.3 (Myr-C3), 28.4–29.3 Phe-Hβ1), 3.00 (dd, 1H, Jβ1,β2 = 14.0 Hz, Jα,β2 = 6.2 Hz, Phe-
(Myr-C4–11), 31.2 (Myr-C12), 33.3 (Myr-C2), 35.8 (Tyr-Cβ), Hβ2), 3.70 (m, 1H, Phe-Hα), 3.90 (m, 1H, Glc-H2), 3.96 (m, 4H,
55.8 (Tyr-Cα), 64.6 (Glc-C3), 65.0 (Glc-C1), 65.3 (Cbz-CH2), Glc-H1 + Glc-H3), 7.30–7.40 (m, 5H, Phe-Arom-H); 13C NMR,
66.1 (Glc-C2), 115.0/127.3/127.4/127.6/128.2/129.8/136.8 δ: 13.8 (Myr-C14), 21.9 (Myr-C13), 24.3 (Myr-C3), 28.3–28.6
(Tyr-Arom-C + Cbz-Arom-C), 156.0/171.6/172.6 (carbonyl).
(Myr-C4–11), 31.2 (Myr-C12), 33.3 (Myr-C2), 40.5 (Phe-Cβ),
1-O-(N-Cbz-
L
-seryl)-3-O-myristoylglycerol (5b). 1H 55.6 (Phe-Cα), 64.5 (Glc-C3), 64.9 (Glc-C1), 66.0 (Glc-C2),
NMR, δ: 0.95 (t, 3H, J = 7.0 Hz, Myr-H14), 1.33 (m, 20H, 126.1/127.9/129.1/137.8 (Phe-Arom-C), 172.6/174.6 (car-
Myr-H4–13), 1.62 (m, 2H, Myr-H3), 2.36 (t, 2H, J = 5.0 Hz, bonyl).
Myr-H2), 3.82 (m, 2H, Ser-Hβ), 4.00 (m, 1H, Glc-H2), 4.10
1-O-L-Tyrosyl-3-O-myristoylglycerol (formate salt, 4c).
(m, 4H, Glc-H1 + Glc-H3), 4.15(m, 1H, Ser-Hα), 5.16 (s, 2H, 1H NMR, δ: 0.94 (t, 3H, J = 7.1 Hz, Myr-H14), 1.32 (m, 20H,
Cbz-CH2), 7.32–7.41 (m, 5H, Cbz-Arom-H); 13C NMR, δ: Myr-H4–13), 1.60 (m, 2H, Myr-H3), 2.38 (m, 2H, Myr-H2),
13.7 (Myr-C14), 22.2 (Myr-C13), 24.5 (Myr-C3), 28.6–29.1 2.80 (m, 2H, Tyr-Hβ), 3.65 (m, 1H, Tyr-Hα), 3.90 (m, 1H, Glc-
(Myr-C4–11), 31.4 (Myr-C12), 33.5 (Myr-C2), 56.8 (Ser-Cα), H2), 4.10 (m, 4H, Glc-H1 + Glc-H3), 6.73 (d, 2H, Jo,m = 8.6
61.4 (Ser-Cβ), 64.9 (Glc-C3), 65.5 (Glc-C1), 65.7 (Cbz-CH2), Hz, Tyr-Arom-Ho), 7.05 (d, 2H, Jo,m = 8.6 Hz, Tyr-Arom-
66.2 (Glc-C2), 127.3/127.5/128.2/136.9 (Cbz-Arom-C), Hm); 13C NMR, δ: 13.8 (Myr-C14), 21.9 (Myr-C13), 24.3
156.6/170.0/172.6 (carbonyl).
1-O-(Nα,Nζ-Di-Cbz-
(Myr-C3), 28.4–29.2 (Myr-C4–11), 31.2 (Myr-C12), 33.3 (Myr-
L
-lysyl)-3-O-myristoylglycerol (6b). C2), 40.5 (Tyr-Cβ), 55.6 (Tyr-Cα), 64.5 (Glc-C3), 64.9 (Glc-
1H NMR, δ: 0.94 (t, 3H, J = 6.5 Hz, Myr-H14), 1.33 (m, 24H, C1), 66.0 (Glc-C2), 126.1/127.9/129.1/137.8 (Tyr-Arom-C),
Myr-H4–13 + Lys-Hβ + Lys-Hδ), 1.60 (m, 4H, Myr-H3 + Lys- 172.6/174.6 (carbonyl).
Hγ), 2.40 (m, 2H, Myr-H2), 3.10 (m, 2H, Lys-Hζ), 4.00 (m,
1-O-L
-Seryl-3-O-myristoylglycerol (formate salt, 5c). 1H
1H, Glc-H2), 4.10 (m, 5H, Lys-Hα + Glc-H1 + Glc-H3), 5.10 NMR, δ: 1.00 (t, 3H, J = 6.6 Hz, Myr-H14), 1.40 (m, 20H,
(s, 2H, Cbzζ-CH2), 5.15 (Cbzα-CH2), 7.25 (m, 1H, Lys-NHζ), Myr-H4–13), 1.65 (m, 2H, Myr-H3), 2.33 (m, 2H, Myr-H2),
7.30–7.40 (m, 10H, Cbzα-Arom-H + Cbzζ-Arom-H), 7.70 (d, 3.80 (m, 2H, Ser-Hβ), 3.95 (m, 1H, Glc-H2), 4.10 (m, 5H, Glc-
1H, Jα,NH = 7.3 Hz, Lys-NHα); 13C NMR, δ: 13.7 (Myr-C14), H1 + Glc-H3 + Ser-Hα); 13C NMR, δ: 13.8 (Myr-C14), 22.0
22.2 (Myr-C13), 22.7 (Lys-Cβ), 24.4 (Myr-C3), 28.5–29.0 (Myr-C13), 24.3 (Myr-C3), 28.8–29.0 (Myr-C4–11), 31.2 (Myr-
(Lys-Cδ + Myr-C4–11), 30.4 (Lys-Cγ), 31.4 (Myr-C12), 33.4 C12), 33.3 (Myr-C2), 56.1 (Ser-Cα), 62.6 (Ser-Cβ), 65.0 (Glc-
(Myr-C2), 40.1 (Lys-Cζ), 54.0 (Lys-Cα), 65.1 (Glc-C3), 65.5 C3), 65.4 (Glc-C1), 69.2 (Glc-C2), 172.8/174.0 (carbonyl).
(Glc-C1), 66.2 (Glc-C2), 64.6 (Cbzζ-CH2), 65.3 (Cbzα-CH2),
127.2/127.3/127.5/127.6/128.0/136.0/137.0 (Cbzζ-Arom-C + NMR, δ: 0.95 (t, 3H, J = 6.6 Hz, Myr-H14), 1.33 (m, 24H,
Cbzζ-Arom-C), 156.0/156.1/172.0/172.4/174.0 (carbonyl). Myr-H4–13 + Lys-Hβ + Lys-Hδ), 1.60 (m, 4H, Myr-H3 + Lys-
Di-(3-O-myristoylglycer-1-yl)-N-Cbz-
-aspartate (7b). 1H Hγ), 2.40 (m, 2H, Myr-H2), 2.80 (t, 2H, Jδ,ζ = 7.3 Hz, Lys-
1-O-L
-Lysyl-3-O-myristoylglycerol (diformate salt, 6c). 1H
L
NMR, δ: 0.94 (t, 6H, J = 6.6 Hz, Myrα-H14 + Myrγ-H14), 1.32 (m, Hζ), 3.60 (m, 1H, Lys-Hα), 4.00 (m, 1H, Glc-H2), 4.10 (m,
40H, Myrα-H4–13 + Myrγ-H4–13), 1.60 (m, 4H, Myrα-H3 + Myrγ- 4H, Glc-H1 + Glc-H3); 13C NMR, δ: 13.8 (Myr-C14), 21.9
H3), 2.33 (t, 4H, J = 7.2 Hz, Myrα-H2 + Myrγ-H2), 2.90 (m, 2H, (Myr-C13), 22.0 (Lys-Cβ), 24.3 (Myr-C3), 27.0 (Lys-Cδ),
Asp-Hβ), 3.50 (2d, 4H, J2,3 = 5.2 Hz, Glcα-H3 + Glcγ-H3), 3.60 (m, 28.8–29.0 (Myr-C4–11 +Lys-Cγ), 31.2 (Myr-C12), 33.5 (Myr-
2H, Glcα-H2 + Glcγ-H2), 4.00 (m, 4H, Glcα-H1 + Glcγ-H1), 4.50 C2), 38.5 (Lys-Cζ), 53.6 (Lys-Cα), 64.6 (Glc-C3), 65.0 (Glc-
(m, 1H, Asp-Ha), 5.10 (s, 2H, Cbz-CH2), 7.32–7.41 (m, 5H, Cbz- C1), 66.0 (Glc-C2), 172.6/174.0 (carbonyl).
Arom-H), 7.82 (d, 1H, J = 7.5 Hz, Asp-NH); 13C NMR, δ: 13.8
Di-(3-O-myristoylglycer-1-yl)-L-aspartate (formate salt,
(Myrα-C14 + Myrγ-C14), 22.1 (Myrα-C13 + Myrγ-C13), 24.4 (Myrα- 7c). 1H NMR, δ: 0.93 (t, 6H, J = 6.6 Hz, Myrα-H14 + Myrγ-
C3 + Myrγ-C3), 28.5–29.0 (Myrα-C4–11 + Myrγ-C4–11), 31.4 (Myrα- H14), 1.32 (m, 40H, Myrα-H4–13 + Myrγ-H4–13), 1.60 (m, 4H,
C12 + Myrγ-C12), 33.5 (Myrα-C2 + Myrγ-C2), 36.0 (Asp-Cβ), 50.4 Myrα-H3 + Myrγ-H3), 2.25 (m, 4H, Myrα-H2 + Myrγ-H2),
(Asp-Cα), 62.7 (Glcα-C1 + Glcγ-C1), 65.4 (Cbz-CH2), 65.6 (Glcα- 2.90 (m, 2H, Asp-Hβ), 3.45 (m, 4H, Glcα-H3 + Glcγ-H3), 3.75
C3 + Glcγ-C3), 69.8 (Glcα-C2 + Glcγ-C2), 127.6/127.7/128.3 (Cbz- (m, 2H, Glcα-H2 + Glcγ-H2), 3.80 (m, 1H, Asp-Hα), 4.00 (m,
Arom-C), 156.0/172.0/173.0/174.0 (carbonyl).
4H, Glcα-H1 + Glcγ-H1); 13C NMR, δ: 13.8 (Myrα-C14
+
1-O-Glycyl-3-O-myristoylglycerol (formate salt, 2c). 1H Myrγ-C14), 22.0 (Myrα-C13 + Myrγ-C13), 24.4 (Myrα-C3 +
JAOCS, Vol. 74, no. 7 (1997)

884
R. VALIVETY ET AL.
Myrγ-C3), 28.5–28.9 (Myrα-C4–11 + Myrγ-C4–11), 31.2 (Myrα-
C12 + Myrγ-C12), 33.4 (Myrα-C2 + Myrγ-C2), 39.0 (Asp-Cβ),
51.0 (Asp-Cα), 62.6 (Glcα-C1 + Glcγ-C1), 65.4 (Glcα-C3 +
Glcγ-C3), 69.2 (Glcα-C2 + Glcγ-C2), 172.8/174.0 (carbonyl).
3-O-Myristoyl-N-Cbz-
-serine (8a). 1H NMR, δ: 0.94 (t,
L
3H, J = 6.6 Hz, Myr-H14), 1.33 (m, 20H, Myr-H4–13), 1.60 (m,
2H, Myr-H3), 2.33 (t, 2H, J = 7.0 Hz, Myr-H2), 4.25 (m, 1H,
Ser-Hβ1), 4.45 (m, 2H, Ser-Hα + Ser-Hβ2), 5.15 (s, 2H, Cbz-
CH2), 7.40–7.50 (m, 5H, Cbz-Arom-H), 7.63 (d, 1H, J = 8.2
Hz, Ser-NH); 13C NMR, δ: 13.6 (Myr-C14), 22.0 (Myr-C13),
24.3 (Myr-C3), 28.2–29.0 (Myr-C4–11), 31.3 (Myr-C12), 33.3
(Myr-C2), 54.0 (Ser-Cα), 62.9 (Ser-Cβ), 65.6 (Cbz-CH2),
127.2/127.6/128.2/136.9 (Cbz-Arom-C), 156.0/172.5/174.2
(carbonyl).
1
3-O-Myristoyl-
L
-serine (formate salt, 8b). H NMR, δ:
0.94 (t, 3H, J = 6.6 Hz, Myr-H14), 1.32 (m, 20H, Myr-H4–13),
1.60 (m, 2H, Myr-H3), 2.24 (t, 2H, J = 7.3 Hz, Myr-H2), 4.10
(m, 1H, Ser-Hβ1), 4.50 (m, 2H, Ser-Hα + Ser-Hβ2); 13C NMR,
δ: 13.8 (Myr-C14), 22.0 (Myr-C13), 24.5 (Myr-C3), 28.5–29.0
(Myr-C4–11), 31.2 (Myr-C12), 33.7 (Myr-C2), 55.0 (Ser-Cα),
62.0 (Ser-Cβ), 174.0 (carbonyl).
FIG. 1. Lipase-catalyzed acylation of
with myristic acid. Substrates: ■, N-Cbz-
Cbz- -lysyl)glycerol; ■, Nα-Cbz- -lysine methyl ester; ▲, Nα-Cbz-
sine; ●, Nζ-Cbz- -lysine methyl ester; ◆, Nα-Cbz- -lysine.
L
-serine and
-serine; ●, 1-O-(Nα,Nζ-di-
-ly-
L-lysine derivatives
L
L
L
L
L
L
RESULTS AND DISCUSSION
sible for the disappointing performance of lipases in these re-
Initially, we investigated the feasibility of lipase-catalyzed actions. Indeed, it may well be that lipases are inherently poor
acylation of -lysine and -serine with a range of long-chain amidation catalysts where N-acylation of amino acids is con-
L
L
fatty acids and their methyl esters as acyl donors. Special at- cerned, perhaps reflecting the lack of suitable structural pro-
tention was focused on the synthesis of myristoyl derivatives visions for the binding of these substrates as acyl acceptors
because N-myristoyl-
hibit optimal surface-active and antimicrobial properties (14).
L
-amino acids have been shown to ex- (19–22).
In pursuit of an alternative route to Nζ-acyl-
L
-lysines, we
The reactions were carried out with a range of immobilized examined the chemical acylation of copper(II) lysinate, in
lipases, including Lipozyme and Novozyme, with up to 3 mol which the α-amino (but not the ζ-amino group) is coordinated
equivalents of the acylating agent in the absence of solvent at to Cu(II). Chemical amidation of this chelate complex with
60–80°C. However, the reaction rates were poor, with only fatty acid chlorides proceeded readily and regiospecifically in
traces of product (2–4%) being detected after several days of aqueous media, thereby providing access to Nζ-acyl-
L-lysines
incubation. The yields remained below 10% even when the in good yields. For example, the capryloyl, caproyl, lauroyl,
corresponding amino acid amides or ethyl or methyl esters and myristoyl derivatives were synthesized with preparative
were used as substrates. No significant improvement was ob- yields of 69, 64, 51, and 63%, respectively, by using this ap-
served upon the addition of organic solvents nor on replace- proach (see the Experimental Procedures section). Although
ment of the fatty acid with triglycerides, such as triolein. The this approach proved to be more facile for regiospecific Nζ-
low conversions obtained in these experiments were initially amidation of lysine than use of enzymes, it showed little
attributed to low solubility of the amino acid derivatives in promise where the side-chain modification of other multi-
the oil phase.
functional amino acids, such as serine, threonine, and argi-
In an attempt to improve the miscibility of the reactants, nine, was concerned.
the N-Cbz-
L
-amino acids were examined as alternative sub-
As alcohols are known to be good substrates for lipases,
strates. We found that N-Cbz-
L
-serine was an excellent sub- we reasoned that use of a polyol, such as glycerol, as a linker
strate for lipases and provided the required O-acyl-N-Cbz-
L
-
between amino and fatty acids (Fig. 2) may prove to be a su-
serine in over 90% yield (Fig. 1). In contrast, the correspond- perior strategy for the enzymatic synthesis of a range of
ing lysine derivatives were poor substrates, regardless of their amino acid-based surfactants. As well as providing a facile
solubility in the oil phase. Thus, Nα-Cbz-
Cbz- -lysine) was acylated only to a minor extent (Fig. 1), glycerol-linked compounds should possess interesting
despite its structural similarity to the serine compound. Even physico-chemical properties that are intermediate to those
use of Nα- and Nζ-Cbz- -lysine methyl esters failed to signif- displayed by long-chain amino acid esters and monoglyc-
L
-lysine (and Nζ- method for attaching lipid functionalities to amino acids, such
L
L
icantly improve the reactions (Fig. 1), in spite of their much erides, and should also retain the good biocompatibilities of
higher solubility in the fatty acid phase. These results indi- the constituents. To investigate the feasibility of this ap-
cated that substrate miscibility was not the sole factor respon- proach, we prepared the glycer-1-yl ester of Nα,Nζ-di-Cbz-
L-
JAOCS, Vol. 74, no. 7 (1997)

CHEMO-ENZYMATIC SYNTHESIS OF AMINO ACID-BASED SURFACTANTS
885
BF3/Et2O
−
(CH2)4NH3+HCO2
HCO2−N
+
H3
(CH2CO2GlcMyr
FIG. 2. Chemo-enzymatic approach to the synthesis of 1-O-(L-aminoacyl)-3-O-myristoylglycerols.
lysine (Fig. 2, 6a) and studied its enzymatic esterification. As and under these conditions, 0.31–0.61 g of product was
expected, 6a was a good substrate for Novozyme and formed per gram of reaction mixture. After purification, the
Lipozyme catalysts, and a highly regioselective acylation of 1-O-(N-Cbz-
the primary hydroxyl of the glycerol moiety was observed tained with preparative yields of 35–80% from the corre-
(Fig. 1). Thus, a conversion of 80% was attained in a matter sponding N-Cbz- -derivatives of glycine, phenylalanine, ty-
L-aminoacyl)-3-O-myristoylglycerols were ob-
L
of hours when the reaction was performed in a 1:2 (mole rosine, serine, and aspartic acid (Table 1). The effects of vary-
ratio) mixture of 6a and myristic acid in the presence of a ing the substrate ratio and the amount of immobilized enzyme
small amount of pentan-3-one (20% w/w, added at the begin- added on the synthesis of a representative 1,3-di-O-acylglyc-
ning of the reaction, then allowed to evaporate as acylation erol, 1-O-(N-Cbz-
proceeded), which was included to improve miscibility of the (3b), are shown in Figures 3A and B, respectively. Subse-
substrates. quent N-deprotection by transfer hydrogenation gave the re-
Encouraged by this result, we proceeded to investigate quired 1-O-( -aminoacyl)-3-O-myristoylglycerols in overall
whether this methodology was equally applicable to the syn- yields of 27–71%.
L-phenylalanyl)-3-O-myristoylglycerol
L
thesis of other glycerol-linked amino acid–fatty acid esters
The developed methodology provides easy access to a
(Fig. 2). This was found to be the case, and conversions of range of cationic and zwitterionic surfactants, which can be
50–90% were attained for Novozyme-catalyzed acylations used as obtained, or alternatively, can serve as efficient build-
with myristic acid. The substrates were used at 0.85–1.09 M, ing blocks for further transformations, such as the synthesis
TABLE 1
Novozyme-Catalyzed Synthesis of 1-O-(N-Cbz-
L
-aminoacyl)-3-O-Myristoylglycerols
Product
concentration
(gg−1 b
Time
(h)
Conversion
(%)a
Yield
(%)c
FAB-MS
(M + H, calcd.)
FAB-MS
(M + H, obsvd.)
Product
)
2b
3b
4b
5b
6b
7b
8a
25
25
40
25
25
35
60
50
91
90
78
80
60
93
0.31
0.54
0.61
0.50
0.58
0.51
0.55
35
80
68
55
56
40
80
492.2963
584.3590
600.3539
524.3235
699.4226
836.5524
450.2858
492.2990
584.3617
600.3527
524.3246
699.4245
836.5546
450.2846
aPercentage of product formed, as determined by high-performance liquid chromatography analysis upon completion of
the reaction.
bCalculated on the basis of the mass of the reaction mixture excluding biocatalyst, gram per gram.
cYield after purification. In all cases 20% w/w of pentan-3-one was added at the beginning of the reaction, then allowed
to evaporate as acylation progressed. A 1:2 molar ratio of glycer-1-yl ester/myristic acid was employed in all cases, ex-
cept in the preparation of (7b) where a 1:3 ratio was used. Abbreviation: FAB-MS, fast atom bombardment mass spec-
troscopy; 2b: 1-O-(N-Cbz-glycyl)-3-O-myristoylglycerol; 3b: 1-O-(N-Cbz-
O-(N-Cbz- -tyrosyl)-3-O-myristoylglycerol; 5b: 1-O-(N-Cbz- -seryl)-3-O-myristoylglycerol; 6b: 1-O-(Nα,Nζ-di-Cbz-
lysyl)-3-O-myristoyl glycerol; 7b: di-(3-O-myristoylglycer-l-yl)-N-Cbz- -aspartate; 8a: 3-O-myristoyl-N-Cbz- -serine.
L-phenylalanyl)-3-O-myristoylglycerol; 4b: 1-
L
L
L-
L
L
JAOCS, Vol. 74, no. 7 (1997)

886
R. VALIVETY ET AL.
2. Fiechter, A., Biosurfactants: Moving Towards Industrial Appli-
A
cations, Trends Biotechnol. 10:208–217 (1992).
3. Georgiou, G., S. Lin, and M.M. Sharma, Surface-Active Com-
pounds from Microorganisms, Bio/Technology 10:60–65
(1992).
4. Desai, A.J., and R.M. Patel, Advances in the Production of Bio-
surfactants and Their Commercial Applications, J. Sci. Ind. Res.
53:619–629 (1994).
5. Sarney, D.B., and E.N. Vulfson, Application of Enzymes to the
Synthesis of Surfactants, Trends Biotechnol. 13:164–172
(1995).
6. Infante, M.R., J. Molinero, P. Bosch, M.R. Julia, and P. Erra,
Lipopeptide Surfactants: I. Neutral N-Lauroyl-L-Arginine
Dipeptides from Pure Amino Acids, J. Am. Oil Chem. Soc. 66:
1835–1839 (1989).
7. Mhaskar, S.Y., R.B.N. Prasad, and G. Lakshminarayana, Syn-
thesis of N-Acyl Amino Acids and Correlation of Structure with
Surfactant Properties of Their Sodium Salts, Ibid. 67:1015–1019
(1990).
8. Infante, M.R., J. Molinero, and P. Erra, Lipopeptide Surfactants:
B
II. Acidic and Basic Nα-Lauroyl-L-Arginine Dipeptides from
Pure Amino Acids, Ibid. 69:647–652 (1992).
9. Mhaskar, S.Y., and G. Lakshminarayana, Synthesis of Di-
ethanolamides of N-Lauroyl Dipeptides and Correlation of Their
Structure with Surfactant and Antibacterial Properties, Ibid. 69:
643–646 (1992).
10. Desai, A., P.H. Kothwala, and P. Bahadur, Surface Activity and
Micellar Behavior of Sodium Salts of N-Acyl-ω-Amino Acids
in Aqueous Solution, Tenside Surf. Det. 29:58–61 (1992).
11. Wu, Z.G., C.N. Ji, and M.L. Shi, Synthesis of Aromatic Amino
Acids for Antibacterial Surfactants, J. Am. Oil Chem. Soc. 70:
109–110 (1993).
12. Nasreddine, M., S. Szönyi, and A. Cambon, New Amphoteric
Surfactants Containing the 2-Perfluoroalkyl 2-Hydroxy Ethyl
Group and an Amino Acid Residue, Ibid. 70:105–107 (1993).
13. Mhaskar, S.Y., G. Lakshminarayana, and L. Saisree, N-Acyl-L-
leucines of Biologically Active Uncommon Fatty Acids: Syn-
thesis and Antibacterial Activity, Ibid. 70:23–27 (1993).
14. Xia, J., Y. Xia, and I.A. Nnanna, Structure–Function Relation-
ship of Acyl Amino Acid Surfactants: Surface Activity and An-
timicrobial Properties, J. Agric. Food Chem. 43:867–871
(1995).
FIG. 3. Synthesis of 1-O-(N-Cbz-
A: Effect of substrate/fatty acid ratio on Novozyme-mediated synthesis.
1-O-(N-Cbz- -phenylalanyl)glycerol/myristic acid molar ratio: ●, 1:2;
■, 1:1.5; ▲, 1:1; ■, 1:0.75; ◆, 1:0.5. B: Effect of Novozyme concentra-
tion on synthesis. Amount of Novozyme added (% w/w): ●, 20%; ▲,
10%; ■, 5%; ◆, 2%.
L-phenylalanyl)-3-O-myristoylglycerol.
15. Nagao, A., and M. Kito, Synthesis of O-Acyl-L-homoserine by
Lipase, J. Am. Oil Chem. Soc. 66:710–713 (1989).
16. Montet, D., F. Servat, M. Pina, J. Graille, P. Galzby, A. Arnaud,
H. Ledon, and L. Marcou, Enzymatic Synthesis of N-ε-
Acyllysines, Ibid. 67:771–774 (1990).
17. Godtfredsen, E., and F. Bjorkling, An Enzyme-Catalyzed
Process for Preparing N-Acyl Amino Acids and N-Acyl Amino
Acid Amides, Novo Nordisk A/s, World Patent:WO 90/14429
(1990).
L
18. Wiese, J., H.G. Gattner, and Z. Zahn, Use of Z-Amino
Acid–Glyceryl Esters in Protease Catalyzed Peptide Synthesis,
Biomed. Biochim. Acta 50:S90–S93 (1991).
19. Astorga, C., F. Rebolledo, and Gotor, V, Enzymatic Aminolysis
of Nonactivated Diesters with Diamines, J. Chem. Soc. Perkin
Trans. I:829–832 (1994).
of peptide–lipid conjugates. The properties of these products
and their usefulness for the preparation of lipidated dipeptides
are currently under investigation in our laboratory.
ACKNOWLEDGMENTS
20. Sanchez, V., F. Rebolledo, and V. Gotor, Highly Efficient En-
zymatic Aminolysis of Alpha, Beta-Unsaturated esters, Synlett:
529–530 (1994).
21. Bistline, R.G., A. Bilyk, and S.H. Feairheller, Lipase-Catalyzed
Formation of Fatty Amides, J. Am. Oil Chem. Soc. 68:95–98
(1991).
We would like to express our thanks to John Eagles and Dr. Fred
Mellon, IFR, Norwich, for providing the FAB-MS data reported in
this work.
22. Tawaki, S., and A.M. Klibanov, Chemoselectivity of Enzymes
in Anhydrous Media Is Strongly Solvent Dependent, Biocataly-
sis 8:3–19 (1993).
REFERENCES
1. Van Dyke, M.I., H. Lee, and J.T. Trevors, Applications of Mi-
crobial Surfactants, Biotechnol. Adv. 9:241–252 (1991).
[Received July 12, 1996; accepted January 21, 1997]
JAOCS, Vol. 74, no. 7 (1997)