Enantioselective hydrolysis of long chain α-amino acid esters by chiral sulfur-containing macrocyclic metallomicelles

Article abstract of DOI:10.1016/S0957-4166(98)00098-6

A novel chiral lipophilic sulfur-containing macrocyclic ligand 5 with bis-pendant alcohols in the proximity of the coordination center has been synthesized. Its metal ion complexes have been investigated as catalysts for the enantioselective hydrolysis of long chain α-amino acid esters in aqueous comicellar solution with Brij35. Large rate accelerations (up to 220 times) and moderate enantioselectivities (up to 4.85 (k(S)/k(R))) employing the macrocyclic 5-Cu²⁺ have been observed, whereas the acyclic 3-Cu²⁺ exhibits less reactivity and stereoselectivity. Taking the analogous ligand 4, lacking the hydroxy groups leads to a dramatic rate decrease, and an inversion of enantioselectivity is observed. The pKa value of the hydroxyl bound to Cu²⁺ is determined to be pKa=7.2 under our micellar reaction conditions.

Full text of DOI:10.1016/S0957-4166(98)00098-6

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 9 (1998) 1197–1203
Enantioselective hydrolysis of long chain α-amino acid esters by
chiral sulfur-containing macrocyclic metallomicelles
Jingsong You, Xiaoqi Yu, Xingshu Li, Qianshun Yan and Rugang Xie ∗
Department of Chemistry, Sichuan University, Chengdu, 610064, PR China
Received 28 January 1998; accepted 20 February 1998
Abstract
A novel chiral lipophilic sulfur-containing macrocyclic ligand 5 with bis-pendant alcohols in the proximity
of the coordination center has been synthesized. Its metal ion complexes have been investigated as catalysts for
the enantioselective hydrolysis of long chain α-amino acid esters in aqueous comicellar solution with Brij35.
Large rate accelerations (up to 220 times) and moderate enantioselectivities (up to 4.85 (kS/kR)) employing the
macrocyclic 5-Cu2+ have been observed, whereas the acyclic 3-Cu2+ exhibits less reactivity and stereoselectivity.
Taking the analogous ligand 4, lacking the hydroxy groups leads to a dramatic rate decrease, and an inversion of
enantioselectivity is observed. The pKa value of the hydroxyl bound to Cu2+ is determined to be pKa=7.2 under
our micellar reaction conditions. © 1998 Elsevier Science Ltd. All rights reserved.
1. Introduction
Metallomicelles are currently receiving interest because of their catalytic ability to mimic hydrolytic
metalloenzymes.1–7 They are made up of either transition metal complexes of ligand surfactants or
comicelles of lipophilic ligand complexes with nonfunctionalized surfactants. These artificial enzymes
are effective in promoting the cleavage of carboxylic and phosphoric acid esters or amides: in most
cases the ligands contain a hydroxyl group in close proximity to the metallocleft. The hydroxyl group
is activated by transition metal ions such as Cu2+ or Zn2+ and acts as an effective nucleophile in
a transacylation process. More recently, enantioselective cleavage of α-amino acid esters by chiral
metallomicelles has been reported, and remarkable enantioselectivities have been observed.8 Following
our interest in chiral sulfur-containing ligands,9 we have designed and synthesized a chiral lipophilic
sulfur-containing macrocyclic ligand 5 with two alcohol pendants. According to CPK models, a metal
ion bound to the metallocleft of ligand 5 is favorably positioned to activate the hydroxyl group for
nucleophilic attack. Moreover, the metal ion is able to coordinate and activate the substrate. We report
∗
Corresponding author. E-mail: Sichuan@rose.cnc.ac.cn
0957-4166/98/$19.00 © 1998 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(98)00098-6

1198
J. You et al. / Tetrahedron: Asymmetry 9 (1998) 1197–1203
here the first example of chiral macrocyclic metallomicelles and their effect on the enantioselective
cleavage of N-dodecanoylamino acid p-nitrophenyl esters. Macrocyclic ligand 4 lacks the hydroxymethyl
group with the aim of defining the role of the hydroxy function. In order to study the structural effect of
acyclic ligand and macrocyclic ligand on the catalytic activity and enantioselectivity, we have synthesized
ligand 3 that is a more flexible open stucture.
2. Results and discussion
2.1. Ligands and substrates
Macrocyclic ligand 5 was synthesized according to the procedures outlined in Scheme 1 using L-
cysteine and diethanolamine as the starting material. Reaction of diethanolamine with 1-bromododecane
yielded N,N-bis(β-hydroxyethyl)-n-dodecylamine, and then converted N,N-bis(β-chloroethyl)-n-
dodecylamine with SOCl2. In the presence of NaOH and NaI, L-cysteine reacted with the long
chain dichloroalkane to form a bridged bis-amino acid, and then converted to the diester derivative.
Subsequently the diester reacted with 2,6-bis(chlorocarbonyl)pyridine by the high-dilution technique to
give macrocyclic ligand 4. Reduction eventually yielded macrocyclic ligand 5. Model compound 3 was
obtained by reducing compound 2 with NaBH4.
Scheme 1.
Macrocyclic ligands 5 and 4 are barely soluble in neutral water. Their clear solutions, free or as
transition metal ion complexes, can be obtained only in the presence of micelles of inert surfactants
such as Brij35 as the matrix of comicellar aggregates. The kinetic experiments were carried out using
mixed micelles, composed of a metal ion complex and Brij35. Compound 3 is soluble in water and forms
a micellar aggregate: the critical micellar concentration (cmc) is 4.4×10−4 M in MES buffer (0.05 M,
pH 6.80).8c Scheme 2 shows the substrates investigated: they are all chiral, nonmetallophilic substrates,
i.e., the N-protected p-nitrophenyl esters of R(S)-phenylalanine, R(S)-leucine and R(S)-alanine.
Scheme 2.

J. You et al. / Tetrahedron: Asymmetry 9 (1998) 1197–1203
1199
Table 1
Pseudo-first-order rate constants (kR and kS,
s
−1) and enantioselectivities (kS/kR) for cleavage of the
N-protected α-amino acid esters by mixed micellar systems composed of chiral metal complexes and
Brij35
2.2. Kinetics
The rate of hydrolysis was followed by observing the release of p-nitrophenolate spectrophotometri-
cally (400 nm) under pseudo-first-order conditions. The pseudo-first-order constants (kR and kS) for the
cleavage of the substrates catalyzed by metal ion complexes comicellized with Brij35 are summarized
in Table 1. The enantioselectivity of each system is indicated by the ratios kS/kR. From the data in
Table 1, copper ion itself shows essentially no catalytic activity, whereas in the absence of metal ions,
the ligands exhibit only a slight rate enhancement and enantioselectivity. Large rate enhancement and
enantioselectivity (kS/kR) are observed in the presence of both ligand and metal ion. These indicate that
catalytic activity is the result of synergistic cooperation of metal ion and ligand. The catalytic activity
order of metal ion is Cu2+>Zn2+>Co2+. Most strikingly, the enantioselectivity of 5-M2+ is also very
sensitive to the nature of the transition metal ion and the highest value (4.85) is found for 5-Cu2+ (entries
4, 5 and 6). For all 5-M2+ the (S)-substrate is hydrolyzed faster than its (R)-enantiomer. Entries 4 and 9
in Table 1 show that acyclic metallocatalyst 3-Cu2+ are less reactive and enantoiselective compared with
macrocyclic metallocatalyst 5-Cu2+. Though these data are far from impressive, they clearly indicate that
the macrocyclic system is more efficient, as expected from the reduction in the degrees of freedom of the
system. Taking the analgous ligand 4 and 5 as an exemplary case (compare entries 4 and 10), lacking the
hydroxyl leads to a dramatic rate decrease, and an inversion of enantioselectivity is observed.3,8c These
results imply that the free hydroxy group of the ligand and the rigidity of the ligand are mandatory in
order to achieve both high rate accelerations and good enantioselectivities. Table 1 also indicates that
increasing the size and lipophilicity of the side chain in the substrate (entries 4, 7 and 8) causes an
increase in the rate and enantioselectivity.

1200
J. You et al. / Tetrahedron: Asymmetry 9 (1998) 1197–1203
2.3. Stoichiometry of the reactive complexes
In order to know the stoichiometry of the kinetically reactive complexes, the kinetic version of Job
Plots was examined by plotting kR and kS as a function of molar fraction of the ligand (γ), keeping the
total concentrations of ligand and metal ion constant. The results shown in Fig. 1 indicate that in the case
of Cu2+ and ligand 5 the rate maxima are observed at γ=0.5, which correspond to a stoichiometry of
ligand: Cu2+=1:1. Ligand 5 forms stable complexes with Cu2+ as indicated by the sharp maxima in the
Job plots.
Fig. 1. Kinetic Job plots for the cleavage of (R)-Phe-PNP (filled circles) and (S)-Phe-PNP (empty circles) by ligand 5 and Cu2+
in MES buffer, pH 6.80, 25°C. ([5]+[Cu2+])=4.0×10−4 M, [Brij35]=4.0×10−3 M, (R(S)-Phe-PNP)=2.0×10−5
M
2.4. pH–Rate profile
As said above, from the analysis of the data in Table 1, a free hydroxyl in the ligand structure is
a requisite for higher rate acceleration and enantioselectivity in micellar systems. A pH–rate constant
profile was determined for reactions of (S)-Phe-PNP with catalyst 5-Cu2+. The pH value was checked
before and after any kinetic run and proved to be constant within ꢀ0.05 pH unit. A plot of log k versus
pH (Fig. 2) gave a sharp break at pH 7.2, which we take as the systematic pKa of the hydroxyl bound
to Cu2+ under our micellar reaction conditions. A pKa=6.4 for a Cu2+ co-ordinated hydroxyl has been
recently reported by Tonellato.8c It would be very interesting to know the pKa of the Cu2+-bound water
molecule, which should be responsible for the cleavage of R(S)-Phe-PNP by the macrocyclic ligand 4.
Regrettably solutions of the complex 4-Cu2+ were unstable at pH>7.5. Up to that pH we did not observe
any break in the log kS versus pH profile.
3. Experimental
3.1. General methods and materials
Melting points were taken on a micro-melting apparatus and are uncorrected. 1H NMR spec-
tra were recorded at 299.95 MHz, and chemical shifts in ppm are reported relative to internal
Me4Si. Mass spectra data were recorded on a Finnigan MAT 4510 spectrometer. Elemental analy-

J. You et al. / Tetrahedron: Asymmetry 9 (1998) 1197–1203
1201
Fig. 2. Log k versus pH for the hydrolysis of (S)-Phe-PNP by Cu2+. See Table 1 for other conditions
ses were performed with a Carlo Erba 1106 instrument. Optical rotations were taken on a WZZ-
1 polarimeter. Kinetic runs were conducted on a Shimadzu UV-265FW spectrophotometer equipped
with a thermostated cell compartment. Zn(NO3)2·6H2O, Cu(NO3)2·6H2O, Co(NO3)2, and polyethylene
glycol dodecyl ether (Brij35) were purchased from commercial sources and used without purifica-
tion. The buffers were 2-morpholinoethanesulfonic acid (MES) (pH=5.0–6.8), 4-(2-hydroxyethyl)-1-
piperazineethanesulfonic acid (HEPES) (pH=6.8–7.8), 4-(2-hydroxyethyl)-1-piperazinepropanesulfonic
acid (EPPS) (pH=7.7–8.5). L-Cysteine hydrochloride was purchased from Sino-American Biotechnology
Company for use without further purification. The following compounds were prepared according to
literature procedures: 2,6-bis(chlorocarbonyl)pyridine,10 the p-nitrophenyl esters of the N-protected α-
amino acids.11 Methylene dichloride was purified according to the standard method. All other chemicals
and reagents were obtained commercially and used without further purification.
3.2. N,N-Bis[((R)-2-amino-2-methoxycarbonylethylthio)ethyl]-n-dodecylamine trihydrochloride 2
A mixture of diethanolamine 1 18.3 g (0.174 mol), 1-bromododecane 43.6 g (0.175 mol) and K2CO3
30.0 g (0.217 mol) was heated at 170°C for 7 h, C2H5OH was then added. The mixture was filtered and
washed with C2H5OH. The organic phase was evaporated to leave a dark brown oil, which was purified by
column chromatography (silica, CHCl3:CH3OH 10:1) to give N,N-bis(β-hydroxyethyl)-n-dodecylamine
as a pale yellow waxy solid (yield 70%). MS (m/z): 273 (M+, 40).
N,N-Bis(β-hydroxyethyl)-n-dodecylamine 4.0 g (14.7 mmol) in CHCl3 (50 mL) was cooled to 0°C.
Thionyl chloride (11 mL) was then added dropwise. After the addition the reaction mixture was refluxed
for 4 h, then brought down to room temperature and left to stand overnight. The solvent was evaporated.
The residue was treated with a solution of sodium carbonate (10%) until the pH was 8.0, then extracted
with CHCl3 (3×50 mL). The organic layer was then separated, dried over MgSO4, filtered and evaporated
to yield N,N-bis(β-chloroethyl)-n-dodecylamine as a pale yellow oil (yield 97%). MS (m/z): 310 (M+,
100).
L-Cysteine hydrochloride 4.1 g (26 mmol) and NaI 0.5 g were dissolved in a solution of 25 mL
water and 60 mL ethanol, and NaOH 2.08 g (52 mmol) was then addd slowly with stirring. N,N-Bis(β-
chloroethyl)-n-dodecylamine 4.0 g (12.9 mmol) was added dropwise over 30 min and continued for 4 h
at 60°C. The mixture was cooled in an ice bath. The precipitate was filtered off and sequentially washed
with water, ethanol, ether, and then dried to gain a bridged amino acid (yield 65%) as a white solid.

1202
J. You et al. / Tetrahedron: Asymmetry 9 (1998) 1197–1203
The bridged amino acid 4.0 g (8.35 mmol) was suspended in 50 mL of methanol and cooled to 0°C
in an ice bath. SOCl2 2.0 mL was added dropwise with stirring. The mixture was stirred at the same
temperature for 1 h and then at 60°C for another 4 h. The solvent was evaporated under vacuum to give
the ester of bridged amino acid hydrochloride 2 (yield 98%).
3.3. (4R,14R)-9-Dodecyl-4,14-dimethoxycarbonyl-6,12-dithia-3,9,15,18-tetraazabicyclo[15.3.1]hene-
icosa-1(18),17(19),20-triene-2,16-dione 4
Compound 2 1.5 g (2.4 mmol) and K2CO3 2.2 g were added to 500 mL dry CH2Cl2, and the mixture
was stirred for 4 h. 2,6-Bis(chlorocarbonyl)pyridine 0.5 g (2.4 mmol) in 50 mL CH2Cl2 was added
drop by drop. When the reaction was complete as monitored by TLC, water (10 mL) was added. The
organic layer was separated, and dried over MgSO4. The filtered solution was evaporated under reduced
pressure, and the raw product was chromatographed on silica gel (ethyl acetate:petroleum=2:1) to give a
white solid (yield 50%); m.p. 61–63°C; [α]D=+22.5 (c=1.0, CH3OH). 1H NMR(CDCl3) δ in ppm: 0.87
(t, J=6.6 Hz, 3H, CH2CH3), 1.24 (m, 18H, (CH2)9CH3), 1.40 (m, 2H, (CH2)9CH2CH2N), 2.36 (m, 4H,
2SCH2CH2N), 2.63 (m, 6H, 3NCH2), 3.04 (dd, J=4.8, 4.2 Hz, 2H, 2SCH2aCHNH), 3.10 (dd, J=4.8, 4.2
Hz, 2H, 2SCH2bCHNH), 3.78 (S, 6H, 2OCH3), 5.01 (m, 2H, 2CH), 8.05 (t, J=8.1 Hz, 1H, PyH4), 8.36
(d, J=7.8 Hz, 2H, PyH3 and PyH5), 8.78 (d, J=8.4 Hz, 2H, 2NH). Anal. calcd for C31H50N4O6S2: C,
58.28, H, 7.89, N, 8.77; found: C, 58.12, H, 7.99, N, 8.52. MS (m/z): 639 (M+, 20).
3.4. (4R,14R)-9-Dodecyl-4,14-dihydroxymethyl-6,12-dithia-3,9,15,18-tetraazabicyclo[15.3.1]heneicosa-
1(18),17(19),20-triene-2,16-dione 5
Compound 4 (1.3 mmol) was dissolved in 60 mL of EtOH containing NaBH4 0.3 g (7.9 mmol). After it
was stirred at room temperature overnight, the mixture was refluxed for 4 h. The solvent was evaporated
and the residue was treated with water and extracted with CHCl3 (5×30 mL). The organic layer was
dried over Na2SO4 and concentrated under reduced pressure. The residue was purified by column
chromatography (silica, CHCl3:CH3OH=100:5) to give a white solid (yield 36%); m.p. 172–174°C;
1
[α]D=+7.1 (c=1.0, CHCl3). H NMR (CDCl3) δ in ppm, 0.88 (t, J=6.6 Hz, 3H, CH3), 1.23 (m, 18H,
(CH2)9CH3), 1.39 (m, 2H, NCH2CH2(CH2)9), 2.40 (t, J=7.8 Hz, 4H, 2SCH2CH2N), 2.65 (m, 6H,
3NCH2), 2.78 (m, 4H, 2SCH2CHNH), 2.95 (m, 2H, 2NHCHCH2OH), 3.86 (m, 4H, 2CH2OH), 4.35
(m, 2H, 2OH), 7.98 (t, J=7.8 Hz, 1H, PyH4), 8.27 (d, J=7.8 Hz, 2H, PyH3 and PyH5), 8.65 (d, J=7.5 Hz,
2H, 2NH). Anal. calcd for C29H50O4N4S2: C, 59.76, H, 8.65, N, 9.61; found: C, 59.59, H, 8.80, N, 9.49.
MS (m/z): 584 (M++1, 100).
3.5. N,N-Bis[((R)-2-amino-3-hydroxylpropylthio)ethyl]-n-dodecylamine 3
This compound was synthesized by following the same procedure as followed in the synthesis
of ligand 5, using compound 2. The crude product was purified by column chromatography (silica,
CHCl3:CH3OH:NH3·H2O=7:1:0.15) to give a white solid (yield 40%); m.p. 49–51°C; [α]D=−20.1
1
(c=1.0, CHCl3). H NMR(CDCl3) δ in ppm, 0.88 (t, J=6.9 Hz, 3H, CH3), 1.26 (S, 18H, (CH2)9CH3),
1.42 (m, 2H, NCH2CH2(CH2)9), 2.45 (m, 4H, 2NH2), 2.64 (m, 6H, 3NCH2), 2.75 (m, 2H, 2CH), 3.01
(m, 2H, 2OH), 3.26 (s, 8H, 4SCH2), 3.46 (dd, J=6.3, 4.5 Hz, 2H, 2CH2aOH), 3.50 (dd, J=6.0, 4.2 Hz,
2H, 2CH2bOH). Anal. calcd for C22H49O2N3S2: C, 58.49, H, 10.93, N, 9.30; found: C, 58.24, H, 11.11,
N, 9.18. MS (m/z): 452 (M+, 30).

J. You et al. / Tetrahedron: Asymmetry 9 (1998) 1197–1203
1203
3.6. Kinetic studies
Solutions of the ligands, metal ions and Brij35 were prepared in the proper buffer (0.05 M). Reaction
temperature was mantained at 25ꢀ1°C. Kinetics were typically started by injecting an acetonitrile
solution (0.01 M) of substrate ester into a 1 cm cuvette containing 2.5 mL of buffered micellar solution
and the desired concentrations of metal ion and ligand. Pseudo-first-order rate constants (kR and kS)
for the hydrolysis of substrate ester were determined by monitoring the release of p-nitrophenolate at
320 nm (pH 5.0–6.3) or 400 nm (pH 6.3–8.5) for at least five half-lives, and obtained by linear plots
of ln(A −At) versus time. The rate constants for each reaction were determined three times from three
∞
separate runs with an uncertainty of less than 5%.
Acknowledgements
The project (No. 29502008 and 29632004) is supported by the National Natural Science Foundation
of China. The authors wish to thank Dr. J. Xie and Dr. Y. W. Ding for helpful discussions.
References
1. (a) Fornasier, R.; Milani, D.; Scrimin, P.; Tonellato, U. J. Chem. Soc., Perkin Trans. 2 1986, 233. (b) Scrimin, P.; Tecilla, P.;
Tonellato, U. J. Org. Chem. 1991, 56, 161. (c) Scrimin, P.; Tecilla, P.; Tonellato, U. J. Org. Chem. 1994, 59, 18. (d) Bunton,
C. A.; Scrimin, P.; Tecilla, P. J. Chem. Soc., Perkin Trans. 2 1996, 419.
2. (a) Tagaki, W.; Ogino, K.; Tanaka, O.; Machiya, K.; Kashihara, N.; Yoshida, T. Bull. Chem. Soc. Jpn 1991, 64, 74. (b)
Ogino, K.; Kashihara, N.; Ueda, T.; Isaka, T.; Yoshida, T.; Tagaki, W. Bull. Chem. Soc. Jpn 1992, 65, 373. (c) Ogino, K.;
Inoue, K.; Tagaki, W. Tetrahedron Lett. 1992, 33, 4191. (d) Ogino, K.; Yamamoto, H; Yoshida, T.; Tagaki, W. J. Chem.
Soc., Chem. Commun. 1995, 691.
3. Weijnen, J. G. J.; Koudijs, A.; Schellenes, G. A.; Engbersen, J. F. J. J. Chem. Soc., Perkin Trans. 2 1992, 829.
4. Kimura, E.; Hashimoto, H.; Koike, T. J. Am. Chem. Soc. 1996, 118, 10963.
5. Broxton, T. J.; Cox, R. A. Can. J. Chem. 1993,71, 670.
6. Faiver, V.; Brembilla, A.; Roizard, D.; Lochon, P. Tetrahedron Lett. 1991, 32, 193.
7. Menger, F. M.; Gan, L. H.; Johnson, E.; Durst, D. H. J. Am. Chem. Soc. 1987, 109, 2800.
8. (a) Cleij, M. C.; Mancin, F.; Scrimin, P.; Tecilla, P.; Tonellato, U. Tetrahedron 1997, 53, 357. (b) Cleij, M. C.; Scrimin, P.;
Tecilla, P.; Tonellato, U. Langmuir 1996, 12, 2956. (c) Scrimin, P.; Tecilla, P.; Tonellato, U. J. Org. Chem. 1994, 59, 4194.
9. (a) Li, X.; Xie, R. Tetrahedron: Asymmetry 1997, 8, 2283. (b) Li, X.; Xie, R. Tetrahedron: Asymmetry 1996, 7, 2779. (c)
Li, X.; Qin, J.; Zhang, X.; Xie, R. Synth. Commun. 1997, 27, 3079.
10. Newcomb, M; Timko, J. M.; Walba, D. M.; Cram, D. J. J. Am. Chem. Soc. 1977, 99, 6392.
11. (a) Ingles, D. W.; Knowles, J. R. Biochem. 1967, 104, 369. (b) Bodanszky, M.; Vigneaud, V. D. J. Am. Chem. Soc. 1959,
81, 5688.