Stepwise Approach toward First Generation Nonenzymatic Hydrolases

Article abstract of DOI:10.1021/jo971935+

The synthesis and reactivity study of a first generation serine protease mimic is described. Central in the design stands the possibility of stabilization of the transition state by an amino triol such as 8t. En route to 8t, a series of amino alcohols (4-8) was obtained, the reactivity of which was studied toward esterification by acetylimidazole (AcIm) and by p-nitro-2,2,2-trifluoroacetanilide (PNTFA). Interesting reactivity differences were observed between the cis- and the trans-series, especially between 7c and 7t (AcIm), and between 8c and 8t (PNTFA). In both cases the results are explained by invoking extra stabilization of the tetrahedral oxyanion.

Full text of DOI:10.1021/jo971935+

2548
J . Org. Chem. 1998, 63, 2548-2559
Step w ise Ap p r oa ch tow a r d F ir st Gen er a tion Non en zym a tic
Hyd r ola ses
Annemieke Madder and Pierre J . De Clercq*
University of Gent, Department of Organic Chemistry, Krijgslaan 281, (S4), B-9000 Gent, Belgium
J ean-Paul Declercq
Laboratoire de Chimie Physique et de Cristallographie, Universite´ Catholique de Louvain,
1 Place Louis Pasteur, B-1348 Louvain-la Neuve, Belgium
Received October 20, 1997
The synthesis and reactivity study of a first generation serine protease mimic is described. Central
in the design stands the possibility of stabilization of the transition state by an amino triol such as
8t. En route to 8t, a series of amino alcohols (4-8) was obtained, the reactivity of which was
studied toward esterification by acetylimidazole (AcIm) and by p-nitro-2,2,2-trifluoroacetanilide
(PNTFA). Interesting reactivity differences were observed between the cis- and the trans-series,
especially between 7c and 7t (AcIm), and between 8c and 8t (PNTFA). In both cases the results
are explained by invoking extra stabilization of the tetrahedral oxyanion.
In tr od u ction
same catalytic activity and specificity as the pancreatic
serine protease, trypsin.8 However, several groups have
subsequently questioned these results.9 Probably the
most successful mimic until now is the one described by
Stewart and co-workers.10 Computer modeling was used
to design a bundle of four short parallel amphipatic
helical peptides bearing the serine protease catalytic
residues. The peptide has affinity for chymotrypsin ester
substrates. However, no activity toward amides, the
natural substrates of chymotrypsin, was reported. So,
the real challenge is still intact.
Several years ago we have embarked on a program
aiming at the development of synthetic hydrolases with
the focus on gaining a better understanding of the
enzymatic process. In this respect we wish to set forth
the following criteria for evaluating a successful under-
taking. The process should follow the basic steps of the
enzymatic mechanism, should allow for turnover, and
eventually operate on unactivated amides. In pursuance
of this goal we decided to work first at the optimization
of the chemical machinery in terms of the nature and
the relative orientation of the required functional groups,
and to include an appropriate binding site only in a
subsequent phase of the program. In an ideal case this
would allow for an eventual combination of general
efficiency and specificity on demand. In the present
paper we report on the synthesis and the reactivity of a
first generation of model hydrolases.
To match enzymatic efficiency remains one of the major
quests in organic chemistry.1 Even if impressive suc-
cesses in synthetic efficiency have been recorded, they
more often than not relate to unnatural processes, e.g.,
the Sharpless epoxidation, where a high level of selectiv-
ity, usually expressed in terms of ee’s, is considered as
evidence of performance. However, the efficiency of the
enzymatic machinery as expressed in turnover and rate,
remains largely unequalled. A typical example in the
field of classic organic chemistry is the hydrolysis of the
unactivated amide bond. Compared to the acidic or basic
hydrolysis the efficiency of chymotrypsin, a protease
responsible for the cleavage of peptide bonds with a
specificity for large hydrophobic residues, is enormous.2
In contrast, the development of synthetic mimics for
proteases has only been marginally successful.3 The
design of hydrolase candidates has mainly rested on the
incorporation of functional groups on a potential binding
site such as a cyclodextrin,4 a crown ether,5 or a struc-
tural moiety that can bind a possible candidate via
hydrogen bonding.6 Almost as a rule the substrates of
which the cleavage was studied were activated esters,7
with a preference for phenyl esters, and in most cases
the observed reaction involved a transesterification rather
than a hydrolysis. Recently, a 29 residue cyclic peptide
was synthesized that was reported to possess nearly the
Even if there still exists some controversy with regard
to the relative importance of the individual steps, the
mechanism of the hydrolysis by chymotrypsin is rather
well understood and involves in the first place an almost
cooperative action of a triad consisting of serine, histi-
(1) Murakami, Y.; Kikuchi, J .; Hisaeda, Y.; Hayashida, O. Chem.
Rev. 1996, 96, 721.
(2) Kirby, A. J . Angew. Chem., Int. Ed. Engl. 1996, 35, 707.
(3) “... but amide cleavage is one of the most daunting challenges
for manmade catalysts, and so far we are not even close.”: see ref 2,
p 709.
(4) (a) Bender, M. L.; D’Souza, V. T. Acc. Chem. Res. 1987, 20, 146.
(b) Breslow, R. Acc. Chem. Res. 1995, 28, 146.
(5) Lehn, J . M.; Sirlin, C. J . Chem. Soc., Chem. Commun. 1978, 949.
(6) (a) Cram, D. J .; Lam, P. Y.; Ho, P. S. J . Am. Chem. Soc. 1986,
108, 839. (b) Tecilla, P.; J ubian, V.; Hamilton, A. D. Tetrahedron 1995,
51, 435.
(7) For a recent and noteworthy exception involving the esterifica-
tion of 4-(hydroxymethyl)pyridine by acetylimidazole both being bound
inside the cavity of a cyclic porphyrin trimer, see: Mackay, L. G.; Wylie,
R. S.; Sanders, J . K. M. J . Am. Chem. Soc. 1994, 116, 3141.
(8) Atassi, M. Z.; Manshouri, T. Proc. Natl. Acad. Sci. U.S.A. 1994,
90, 8282.
(9) (a) Matthews, B. W.; Craik, C. S.; Neurath, H. Proc. Natl. Acad.
Sci. U.S.A. 1994, 91, 4103. (b) Corey, D. R.; Phillips, M. A. Proc. Natl.
Acad. Sci. U.S.A. 1994, 91, 4106. (c) Wells, J . A.; Fairbrother, W. J .;
Otlewski, J .; Laskowski J r., M.; Burnier, J . Proc. Natl. Acad. Sci.
U.S.A. 1994, 91, 4110.
(10) Hahn, K. W.; Klis, W. A.; Stewart, J . M. Science 1990, 248, 1544.
(11) Blow, D. M. Acc. Chem. Res. 1976, 9, 145.
S0022-3263(97)01935-X CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/19/1998

First Generation Nonenzymatic Hydrolases
J . Org. Chem., Vol. 63, No. 8, 1998 2549
Sch em e 1
potential of Scheme 1 in the context of the hydrolysis of
an unactivated amide bond as final goal. The molecule
represented in the center corresponds to the protonated
tetrahedral intermediate, a species that we consider as
being closest to the two transition states (cf. GBC and
GAC) that are involved in the acylation process. The
dashed lines indicate the various hydrogen bonds that
are implicated in the transition state stabilization. Next
to the hydrogen bonding to the oxyanion, we consider the
presence of four oxygen atoms that can stabilize the
ammonium part of the intermediate as an important
asset in the mechanism. Indeed, Gandour and co-
workers recently reported on the catalytic behavior of
open chain polyethers in the butylaminolysis of aryl
acetates.15 They found that catalysis proceeds via a
complex in which four oxygens in a polyether subunit
stabilize the ammonium part of the transition state. The
core of the proposed molecule consists of an array of
functional groups, i.e., one tertiary amine, two primary
hydroxy groups, one secondary hydroxy group, and two
ether functions, that are incorporated in a 15-atom linear
framework designed in such a way as to enable all proton
transfers or hydrogen bonds to occur via six-membered
ring geometries. In a stepwise fashion the proposed
mechanism involves GBC/GAC steps in which the ter-
tiary amine functions as the general base and the
secondary hydroxy group as a proton shuttle in the acid-
catalyzed breakdown of the tetrahedral intermediate. The
second primary hydroxy group should assist in stabilizing
the oxyanion.
F igu r e 1. Charge stabilization mechanism for the peptide
hydrolysis by R-chymotrypsin.
dine, and aspartate (Figure 1).11 Attack by serine, which
is made more nucleophilic by the proximate histidine
functioning as a general base (GBC), leads to a tetrahe-
dral intermediate, the oxyanion in which is stabilized by
hydrogen bonding with the backbone peptide bonds. The
role of aspartate probably consists here in stabilizing the
protonated imidazole moiety.12 The latter is essential in
a general acid-catalyzed (GAC) breakdown of the tetra-
hedral intermediate leading to an acylated enzyme
derivative. Further cleavage of the acylated derivative
by water following the reverse mechanism completes the
process with regeneration of the enzyme. The crucial
part in the process are the GBC and GAC steps which
have been suggested to occur in an almost simultaneous
(cooperative) manner.13 The knowledge that this very
triad is operative as well in the mammalian serine
proteases (cf. chymotrypsin) as in bacterial serine pro-
teases (cf. subtilisin) is a striking example of independent
convergent evolution.14
As mentioned above, the molecule has been further
conceived in a way that all heavy atoms occupy corner
positions on a diamond grid (Scheme 2). As a corollary
the transition state stabilization is expected to occur via
a low energy fully staggered geometry. To ease the
entropic burden it is further desirable to incorporate
Resu lts a n d Discu ssion
Design of t h e F ir st Gen er a t ion . Inspired by the
mechanism of chymotrypsin we set out to evaluate the
(12) Zimmerman, S. C.; Korthals, J . S.; Cramer, K. D. Tetrahedron
1991, 47, 2649.
(13) Warshel, A.; Russell, S. J . Am. Chem. Soc. 1986, 108, 6569.
(14) Neurath, H. Science 1984, 224, 350.
(15) Hogan, J . C.; Gandour, R. D. J . Org. Chem. 1991, 56, 2821.

2550 J . Org. Chem., Vol. 63, No. 8, 1998
Madder et al.
Ta ble 1. Ha lf-Live Va lu es (pseu d o fir st-or d er con d ition s)
for Rea ction of Am in o Alcoh ols 1-8 w ith Acetylim id a zole
(AcIm ) a n d p-Nitr o-2,2,2-tr iflu or oa ceta n ilid e (P NTF A) in
Aceton itr ilea
Sch em e 2
t1/2
compound
AcImb
PNTFAc
1,6-hexanediol + Et3Nd
2240c
721
198
151
925
6900
255
420
306
450
26
412
112
107
206
-
1
2
3
4c
4t
5c
5t
6c
6t
7c
7t
8c
8t
-
-
-
-
-
353
271
873
181
480
169
446
a
Determined via 1H NMR integration, 500 MHz, 23 °C,
concentration (amino alcohol): 0.05 mol L-1 (see Supporting
Information). Half-live values in minutes. c Half-live values in
b
hours. Concentrations: 0.05 mol L-1
.
d
the latter the GAC portion of the mechanism. Activated
esters, such as phenyl esters, are considered inappropri-
ate for our purpose since evidence has been presented
that in most cases the cleavage involves nucleophilic
catalysis.16,17
A Con ver gen t Syn th esis of 8. During the develop-
ment of our target as defined in Scheme 2, a series of
amino alcohols was obtained of which the reactivity
toward the cleavage of AcIm was studied (Table 1). The
13 derivatives that were selected for study of reactivity
are shown in Chart 1. The synthesis of the most evolved
amino triol 8t involves a convergent strategy implying
first a reductive amination and second an alkylation as
key-steps for the attachment of the two side-chains on
the central cyclohexane ring (Scheme 3). The order that
was chosen for performing these key transformations
allowed us to synthesize a series of amino alcohols of
increasing complexity, the reactivity of which toward a
GBC esterification by AcIm was studied in the first place.
The synthesis of the central aldehyde that will be used
in the reductive amination is shown in Scheme 4.
Starting from the known 4,4-dimethylcyclohexanone 9,18
reaction with dimethyl carbonate led to 10,19 and further
selective alkylation of the corresponding dianion (LDA,
THF) with (benzyloxy)methyl chloride gave cyclohex-
structural features that will ensure the proximity of the
different functional groups that need to cooperate. Among
many possibilities, we like to focus here on derivatives
in which a tert-butyl substituent at C-2 and a chair
cyclohexane ring incorporated at C-6, C-7, and C-8 assist
in the GBC and the GAC steps, respectively. Finally,
four oxygen atoms are involved in the stabilization of the
ammonium moiety in accord with Gandours findings. The
derivative depicted in Scheme 2 is one in which the
substitution pattern of the cyclohexane ring involves the
all-trans stereochemistry. The R-substituent at C-7
shows the possibility of incorporating in a next generation
a binding site designed so as to fit the acyl group R1 of
the amide.
1
anone rac-11,20 which appears as the enol form (cf. H
The absence of a substrate binding site at this stage
of the program is obviously a major shortcoming. Indeed
missing the entropic advantage of the formation of a
Michaelis complex, which renders the chemical reaction
an essentially unimolecular transformation, will result
in an intrinsic low reactivity. Hence, if any reaction is
to be observed at all, one will be forced to have recourse
to activated substrates, the choice of which is crucial. In
the context of this work we will focus on the cleavage of
two electronically activated amides, acetyl imidazole
(AcIm) and p-nitro-2,2,2-trifluoroacetanilide (PNTFA).
The former will be used in order to evaluate the GBC,
NMR). Reduction of the carbonyl with sodium borohy-
dride and cerium(III) chloride led to a ca. 1:1 mixture of
alcohols rac-12c and rac-12t (77% combined yield), that
was separated by HPLC. The structural assignment
1
follows from analysis of the H NMR spectral data: (1)
the relative cis-orientation of the two equatorial substit-
uents at C-2 and C-6 follows from the coupling pattern
of H2 and H6 in both derivatives, i.e., J (H2-H3ax) and
J (H6-H5ax) ≈ 13.3 Hz in 12c and 12t, (2) distinction
between the all-cis and all-trans alcohols follows from
the coupling pattern of H1 (sum of coupling constants )
3 Hz for 12c and 20 Hz for 12t). The further synthesis
(16) (a) Anoardi, L.; Tonellato, U. J . Chem. Soc., Chem. Commun.
1977, 401. (b) Somayaji, V.; Skorey, K. I.; Brown, R. S. J . Org. Chem.
1986, 51, 4866. (c) Khan, M. N. J . Org. Chem. 1985, 50, 4851.
(17) However, examples of GBC-hydrolysis of PNPA were reported,
see: (a) Hine, J .; Khan, M. N. J . Am. Chem. Soc. 1977, 99, 3847. (b)
Werber, M. M.; Shalitin, Y. Bioorg. Chem. 1973, 2, 202.
(18) (a) Flaugh, M. E.; Crowell, T. A.; Farlow, D. S. J . Org. Chem.
1980, 45, 5399. (b) Meyer, W. L.; Brannon, M. J .; da Burgos, C. G.;
Goodwin, T. E.; Howard, R. W. J . Org. Chem. 1985, 50, 438.
(19) Evans, D. A.; Nelson, J . V. J . Am. Chem. Soc. 1980, 102, 774.
(20) Huckin, S. N.; Weiler, L. J . Am. Chem. Soc. 1974, 96, 1082.

First Generation Nonenzymatic Hydrolases
J . Org. Chem., Vol. 63, No. 8, 1998 2551
Ch a r t 1
Sch em e 3
Sch em e 4a
of the aldehyde was performed in each separate series
and involved a three-step sequence: (1) protection of the
secondary hydroxyl group as silyl ether (triethylsilyl
ether (TES) in the all-cis series; tert-butyldimethylsilyl
ether (TBDMS) in the all-trans series);21 (2) reduction of
the ester rac-13c and rac-16t with DIBALH; (3) oxidation
of the obtained primary alcohols rac-14c and rac-17t with
sulfur trioxide-pyridine complex in DMSO.
The synthesis of the three amino diols rac-4c, 5c, and
6c in the cis-series (see Chart 1) is shown in Scheme 5.
Comparison of the reactivity of 4 and 5 should reveal the
effect of the anchoring tert-butyl substituent on the GBC
part of the mechanism. Central in the synthesis stands
a reductive amination sequence. The therefore required
secondary amine derivatives 19 and 20 were obtained
using classical procedures that are included in the
Supporting Information section. After considerable ex-
perimentation the successful transformation was realized
using titanium(IV) isopropoxide in the condensation step,
followed by sodium cyanoborohydride reduction.22 Cou-
pling of the aldehyde rac-15c with amine 19 led to rac-
21c (54% yield) and after desilylation with tetrabutyl-
a
TES ) triethylsilyl. Reagents: (a) NaH, MeOCOOMe, 95%;
(b) i. LDA, THF, ii. PhCH2OCH2Cl; 65%; (c) CeCl3‚7H2O, NaBH4,
MeOH, 77%; (d) TESCl, imidazole, 99%; (e) DIBALH, toluene, -78
°C, 99%; (f) SO3‚pyridine, Et3N, DMSO, CH2Cl2, 93%; (g) TBDM-
SCl, imidazole, 99%; (h) DIBALH, toluene, -78 °C, 99%; (i)
SO3‚pyridine, Et3N, DMSO, CH2Cl2, 94%.
ammonium fluoride (TBAF) to amino diol rac-4c. The
analogous coupling with the tert-butyl-substituted amine
was performed with the enantiomerically pure (R)-
derivative 20.23 The reductive amination sequence led
to a ca. 1:1 mixture of the two diastereomers 22c and
23c that could not be separated at this stage. Desilyla-
(21) Originally the TBDMS protective group was also chosen for the
all-cis series. Problems encountered during later deprotection steps
led us eventually to use the TES ether as protective group.
(22) Mattson, R. J .; Pham, K. M.; Leuck, D. J .; Cowen, K. A. J . Org.
Chem. 1990, 55, 2552.

2552 J . Org. Chem., Vol. 63, No. 8, 1998
Madder et al.
Sch em e 5a
F igu r e 2. Ball-and-stick model of the X-ray structure of the
(S)-camphor sulfonate salt of 7c (top) and the calculated global
minimum energy conformation of 7c (bottom).
Ta ble 2. 1H NMR (500 MHz) Vicin a l Cou p lin g Con sta n ts
for Am in o Tr iols 7c a n d 7t
a
TES ) triethylsilyl. Reagents: (a) i. 2 equiv Ti(iPrO)4; ii.
Na(CN)BH3, EtOH, 54%; (b) Bu4NF, THF, 58%; (c) i. 2 equiv 20,
2 equiv Ti(iPrO)4; ii. Na(CN)BH3, EtOH, 94%; (d) Bu4NF, THF,
73%.
J (Hz)
tion led to the corresponding amino diols 5c and 6c that
were isolated in pure form after HPLC separation. The
analogous sequence was performed on the all-trans
aldehyde rac-18t. In this manner amino diols rac-4t, 5t,
and 6t were obtained (see Supporting Information).
The assignment of the structures of the diastereomeric
amino diols 5c and 6c became possible via X-ray diffrac-
tion analysis of a later intermediate. After debenzylation
of 5c, the obtained amino triol 7c was converted to the
corresponding ammonium sulfonate salt by treatment
with (S)-camphorsulfonic acid in dichloromethane. Fig-
ure 2a shows the ammonium salt as deduced from the
X-ray data. At this point it is interesting to compare this
result with the preferred calculated conformation of the
free amino triol 7c which is represented in Figure 2b.24
position
7ca
7tb
1a-2
1b-2
2-3a
2-3b
4a-5
4b-5
5-6a
5-6b
5-11
8a-9
8b-9
9-10a
9-10b
9-11
9.9
3.4
11.7
2.6
c
6.4
3.8
10.9
2.9
9.5
4.9
13.4
3.5
9.7
c
c
13.7
3.1
c
13.1
3.6
6.7
5.0
c
3.3
7.7
4.1
9.7
Recorded in CD3CN. Recorded in CDCl3. c The individual
a
b
coupling constants could not be determined from the observed 1H
NMR signal.
1
Table 2 shows H NMR coupling constants for the amino
triols 7c and 7t which provide a nice indication of the
anchoring capacity of the tert-butyl substituent in the
GBC part of the molecule.
The structural assignment of the diastereomeric trans-
derivatives 5t and 6t, that were obtained in a similar
way, rests on the inversion sequence that is shown in
Scheme 6. After selective protection of the primary
hydroxy group in 5c (leading to 24c) and Dess-Martin
oxidation of the secondary alcohol, the obtained cyclo-
hexanone derivative was reduced under dissolved metal
conditions which led to the expected all-trans diol 25t
exclusively. Desilylation gave amino triol 7t which was
(23) The synthesis of the corresponding enantiomer has been
described before: Steels, I.; De Clercq, P. J .; Declercq, J . P. Tetrahe-
dron: Asymmetry 1992, 3, 599.
(24) Figure 2b shows the global energy minimum among several
minimal energy conformations that were found. The geometry was
calculated using MacroModel V3.0: Still, W. C.; Mohamadi, F.;
Richards, N. G. J .; Guida, W. C.; Lipton, M.; Liskamp, R.; Chang, G.;
Hendrickson, T.; DeGunst, F.; Hasel, W.; Department of Chemistry,
Columbia University, New York, 10027.

First Generation Nonenzymatic Hydrolases
J . Org. Chem., Vol. 63, No. 8, 1998 2553
Sch em e 6a
Sch em e 7a
a
Reagents: (a) Li, liquid NH3, 69%; (b) TBDMSCl, Et3N,
DMAP, DMF, 78%; (c) Dess Martin, CH2Cl2, 55%; (d) Li, liquid
NH3, MeOH, 61%; (e) Bu4NF, THF, 82%; (f) Li, liquid NH3, 83%.
a
Reagents: (a) Li, liquid NH3, 100%; (b) i. 1.2 equiv n-BuLi,
THF, -78 °C to -30 °C; ii. 26, -78 °C, 68%; (c) Bu4NF, THF, 80%;
(d) Li, liquid NH3, 93%.
found identical with the derivative obtained via deben-
zylation of diastereomer 5t.
this is indicative of a mechanism in which the tertiary
amino group is actively involved. A somewhat reduced
set of amino alcohols was also examined in their behavior
toward p-nitro-2,2,2-trifluoroacetanilide (PNTFA). Here
also half-live values were determined following the
disappearance of PNTFA (Table 1). In this case the
corresponding trifluoro-acetylated derivatives were not
isolated. Presumably if formed, the acylated derivatives
are further cleaved in a fast process involving the tertiary
amine (vide infra).
With regard to AcIm, it has been shown that the amino
alcohol catalyzed hydrolysis is subject to GBC.16 More-
over, in a recent study of the influence of anchoring
substitution in 1,3-amino alcohols on the rate of esteri-
fication by AcIm and by p-nitrophenylacetate (PNPA) in
acetonitrile,27 anchoring substitution was found to lead
to (1) a rate decrease in the esterification with PNPA,
and (2) a rate increase in the reaction with AcIm. These
effects were ascribed to a nucleophilic mechanism for the
reaction with PNPA and to a GBC mechanism for the
reaction with AcIm.28
The attachment of the final side chain involves an
alkylation sequence that is shown in Scheme 7 for the
cis-series. After selective protection of the primary
hydroxy group in 5c as TBDMS-ether (24c), debenzyla-
tion leads to diol 25c. Upon deprotonation (1 equiv of
n-butyllithium) followed by alkylation with chloride 26,
a selective reaction at the less hindered primary hydroxy
group is observed.25 Desilylation leads further to 27c and
final debenzylation to 8c. The same sequence starting
from 5t, but involving tert-butyldiphenylsilyl as protec-
tive group, led to 8t (see Supporting Information).
Rea ctivity Stu d ies. As a measure of reactivity we
first studied the esterification of the amino alcohols in
Chart 1 by AcIm in acetonitrile using a large excess of
amino alcohol to ensure the pseudo first-order conditions.
Half-live values at 23 °C were determined by following
the disappearance of AcIm (see Supporting Information)
and are shown in Table 1.26 In all cases the primary
hydroxyl group at C-1 was found to be preferentially
acylated. Especially in the cases of amino triols 7 and 8
All studied derivatives in Chart 1 can be considered
as derivatives of N-methyl-N-alkyl-3-amino-1-propanols.
The simple derivatives 1-3 are included here for com-
parison. Their synthesis and reactivity behavior have
been reported in detail before.29 Comparison of the
reactivity of 1 and 2 shows that the presence of an
anchoring tert-butyl group is responsible for a reasonable
4-fold rate enhancement. This observation is in accord
with a GBC mechanism. Indeed, the presence of a tert-
butyl group induces the combined anti-orientation of the
dimethylamino and hydroxy groups relative to the tert-
(25) Proof for the selective alkylation at the primary alcohol site
was obtained indirectly via preparation of the mesylate; the 1H NMR
resonance at δ ) 4.68 ppm was found to integrate for the expected
1H.
(26) Reactions were followed by 1H NMR over a period of at least
one half-life. Since the reactions were monitored by following the
relative integrations of AcIm, a compound present in a very low
concentration (cf. pseudo first-order conditions), monitoring over more
extended time periods was inappropriate. However, in cases when half-
lives were determined over at least five lifetimes using more sensitive
UV methodology, values of the same order of magnitude are obtained
(see ref 29b): 649, 144, and 108 min for 1, 2, and 3, respectively. These
values are consistently lower than those mentioned in Table 1, which
is in line with the somewhat higher temperature at which the UV
experiments were run (25 °C).
(27) Steels, I.; De Clercq, P. J .; Maskill, H. J . Chem. Soc., Chem.
Commun. 1993, 294.
(28) (a) Oakenfull, D. G.; J encks, W. P. J . Am. Chem. Soc. 1971,
93, 178. (b) Oakenfull, D. G.; Salvesen, K.; J encks, W. P. J . Am. Chem.
Soc. 1971, 93, 188. (c) Fife, T. H. Acc. Chem. Res. 1993, 26, 325.
(29) (a) Madder, A.; De Clercq, P. J .; Maskill, H. J . Chem. Soc.,
Perkin Trans. 2 1997, 851. (b) Madder, A.; Sebastian, S.; Van Haver,
D.; De Clercq, P. J .; Maskill, H. J . Chem. Soc., Perkin Trans. 2 1997,
2787.

2554 J . Org. Chem., Vol. 63, No. 8, 1998
Madder et al.
Sch em e 8
Sch em e 9
Sch em e 10
In the cis-derivative 4c we note that the intramolecular
hydrogen bonded form VIII (Scheme 9) is destabilized
by nonbonded interactions originating from the N-methyl
substituent so that the normal hydrogen bonded form VII
is the expected one. The difference in reactivity that is
now observed between 5c and 4c closely parallels the one
that is observed between 2 and 1.
A very interesting reactivity difference is observed
when comparing amino triols 7c and 7t. The reactivity
of 7t is similar to the observed reactivities of the
previously discussed trans-derivatives 5t and 6t. As for
the cis-derivative the rate enhancement clearly indicates
that an extra effect has to be involved. We like to suggest
that the primary hydroxyl group that is not involved in
the GBC portion of the mechanism can effectively help
in stabilizing the tetrahedral oxyanion via intramolecular
hydrogen bonding as illustrated in Scheme 10.
The introduction of the side chain (cf. 8c and 8t) that
is expected to stabilize the ammonium intermediate is
of course irrelevant here since the imidazole moiety does
not allow for a GAC breakdown of the tetrahedral
intermediate, both because of electronic and geometric
reasons. For the sake of completeness, they have been
included in the list and show, as can be expected, a
reactivity behavior that is almost identical to 5c and 5t,
respectively.
As a conclusion of this part we believe that as far as
the GBC portion of the mechanism is concerned the study
of this set of derivatives has been quite instructive.
Especially, the knowledge that a correctly positioned
hydroxy group for hydrogen bonding to the oxyanion
intermediate (cf. Scheme 10) is effective, will be useful
in the future. At the same time it is obvious that a
further evaluation of the hydrolytic capacity of our set
requires a substrate that can benefit from eventual
hydrogen bond stabilization as shown in Scheme 2. On
the other hand, for reasons that have been mentioned
earlier, the intrinsic low reactivity will require an
electronically activated substrate. We consider trifluo-
roacetanilides as a reasonable compromise. Indeed, the
trifluoroacyl group should exhibit an enhanced reactivity
toward nucleophilic attack, and the anilide moiety should
have a propensity toward GAC breakdown of the tetra-
hedral intermediate. Previous studies involving trifluo-
roacetanilides have shown that the imidazole-catalyzed
hydrolysis of a series of trifluoroacetanilides, among
butyl group (cf. conformation I and II, Scheme 8); the
induced proximity of the polar groups in the propane
fragment favors the intramolecular hydrogen bonding
between the tertiary amino group and the hydroxyl
function, and hence the GBC mechanism. The presence
of a second hydroxyl group in 3 further accelerates the
reaction presumably through hydrogen bond assistance
in stabilizing the tetrahedral oxyanion.29a
Inspection of Table 1 reveals that, with regard to the
transesterification with AcIm, the maximum reactivity
difference equals 28 (cf. t1/2 of 1 and 7c). It is further
interesting to note that the presence of a tert-butyl group
in the more elaborate derivatives 5 when compared to 4
leads to a sensible rate enhancement in the cis- and the
trans-series. A following and unexpected observation is
that the derivatives in the cis-series are consistently more
reactive than the trans-analogues, i.e., a factor of ap-
proximately 7, 1.5, 1.5, 20, and 2.5 for 4, 5, 6, 7, and 8,
respectively. The very low reactivity of 4t suggests that
the normal intramolecular GBC mechanism is not opera-
tive here. Presumably, preferred hydrogen bonding of
the tertiary amino group with the secondary hydroxyl
group is taking place so that the product behaves as an
ordinary primary alcohol. In this compound conforma-
tion IV is the preferred one (Scheme 8). Control experi-
ments have shown that when AcIm is treated with 1,6-
hexanediol in the presence of triethylamine the rate of
the transesterification is negligible (cf. half-live of ≈ 2200
h). In contrast when considering the tert-butyl substi-
tuted derivative 5t, the intramolecular hydrogen bonded
form IV suffers from severe nonbonded interactions (i.e.,
tert-butyl (R) and N-methyl). Hence the III conformation
should be the preferred one. This is in line with the
observed rate enhancement compared to 4t. In the case
of the diastereomeric derivative 6t the preferred confor-
mation is expected to be V rather than VI,30 with a very
similar reactivity as observed for 5t (cf. form III).
(30) Note that for reasons of clarity the enantiomer of 6t is
represented in the scheme.

First Generation Nonenzymatic Hydrolases
J . Org. Chem., Vol. 63, No. 8, 1998 2555
Sch em e 11
oxyanion by the terminal primary hydroxy group will be
enforced in the first place. The incorporation of anchor-
ing structural moieties in the flexible polyether part of
the molecule and/or the inclusion of unreactive terminal
functional groups with more efficient hydrogen bonding
capacity, such as a phosphonamidate or a sulfonamide,34
should help solve this problem. These aspects will be
further investigated in the future.
Finally, we like to stress the fact that the results in
terms of intrinsic hydrolytic performance are not spec-
tacular but that they were not expected to be so in the
first place. Rather, the relative differences that have
been observed within the series of Chart 1 indicate
reactivity trends that are not only of general interest but
very encouraging in the context of our final endeavor.
Exp er im en ta l Section
Gen er a l In for m a tion . All reactions involving air and/or
moisture sensitive materials were conducted under atmo-
spheres of N2 or Ar, which were dried and purified by passage
through a column containing copper on charcoal, anhydrous
CaSO4, and 4 Å molecular sieves. All solvents were purified
before use. Et2O and THF were distilled from sodium ben-
zophenone ketyl. Toluene was distilled from sodium. Et3N and
diisopropylamine (DIA) were distilled from CaH2. CH2Cl2 was
distilled from P2O5. DMSO was distilled from CaH2 under
vacuum and immediately stored over freshly activated 4 Å
molecular sieves. Analytical and preparative thin-layer chro-
matography were performed using glass plates precoated (0.25-
mm layer) with silica gel 60 F254. Flash chromatography was
performed using Merck silica gel 60 (70-235 or 230-400
which the p-nitro derivative, proceeds via imidazolium
GAC breakdown of the tetrahedral intermediate.31
The cleavage of PNTFA by the amino alcohols 1, 2, 3,
7, and 8 was followed in the same way as described above
for AcIm (monitoring by 1H NMR integration). The
observation that the reaction led to the hydrolysis of
PNTFA without isolation of the trifluoroacetyl ester of
the amino alcohol implies that water is participating in
the process.32 At this point it is not clear whether water
is involved in a direct hydrolysis of PNTFA via a process
involving GBC or rather in the fast hydrolysis of the
previously formed trifluoroacetyl ester of the amino
alcohol under study. Although the mechanism is unclear,
inspection of the reduced set that was investigated
remains instructive (Table 1). The expected drop in
overall reactivity is observed and can be roughly evalu-
ated at a factor of 50-100 (cf. from minutes to hours).
For the simple derivatives 1 and 2, no anchoring effect
is observed as in the case of AcIm. Furthermore, the
trans-derivatives are now found to be more reactive than
the corresponding cis-derivatives. Also, a general trend
is observed in that the more structurally complex the
molecule becomes (cf. 2 f 3 f 7 f 8) the less reactive it
behaves except for 8t.
In the context of our work it is tempting to ascribe the
higher reactivity of 8t, when compared to 8c, to the
occurrence of GAC catalysis (compare IX and X in
Scheme 11). However, this possibility has to be ruled
out since a competition experiment in which the reactiv-
ity of 8t was followed in the presence of a 1:1 mixture of
PNTFA and the corresponding p-MeO-trifluoroacetanil-
ide clearly showed the p-nitro derivative to be hydrolyzed
faster.33 Rather we believe that in the case of 8t, reaction
is occurring via GBC in which the oxyanion formation is
assisted by the proximate secondary hydroxy group (cf.
XI). It is likely that the expected mechanism represented
in X will only be effective when the stabilization of the
1
mesh). The H NMR spectra were obtained at 500, 360, and
200 MHz and the 13C NMR spectra at 50.3 MHz. Chemical
shifts are reported in parts per million (ppm) downfield from
TMS using residual CHCl3 (7.27 ppm) as internal standard.
Melting points are uncorrected. Infrared (IR) were recorded
on a FT-IR spectrometer using KBr plates with film or in
solution. Mass spectra (MS) were recorded at 70 eV. Com-
mercial AcIm was recrystalized from dry Et2O. PNTFA was
35
prepared according to a literature procedure, recrystalized
from water, and dried in vacuo (P2O5).
r a c-[Meth yl (1R,2S,3R)-3-[(Ben zyloxy)oxy]-2-h yd r oxy-
5,5-d im eth ylcycloh exa n e-1-ca r boxyla te] (12c) a n d r a c-
[Met h yl (1R,2R,3R)-3-[(Ben zyloxy)oxy]-2-h yd r oxy-5,5-
dim eth ylcycloh exan e-1-car boxylate] (12t). To a suspension
of oil-free sodium hydride (8.94 g of a 60% suspension in
mineral oil, 372 mmol) in 400 mL of THF was added dimethyl
carbonate (74.6 mL, 885 mmol). The mixture was heated to
reflux and a solution of 4,4-dimethylcyclohexanone (21.53 g,
170 mmol) in 200 mL of THF was added dropwise over 45 min.
The mixture was maintained at reflux for 2 h after the addition
was complete. The reaction mixture was added to 300 mL of
saturated aqueous NH4Cl and the aqueous phase extracted
twice with Et2O (2 × 300 mL). The combined organic extracts
were washed with saturated aqueous sodium bicarbonate and
brine, dried (Na2SO4), and concentrated in vacuo. The crude
product was bulb-to-bulb distilled at 89 °C (0.2 mmHg) to yield
29.83 g (95%) of 10 as a colorless oil: IR (neat): 3434, 2953,
2867, 1725, 1697, 1616 cm-1 1H NMR (CDCl3, 500 MHz) δ
;
0.95 (6H, s), 1.44 (t, 2H, J ) 6.7 Hz), 2.02 (s, 2H), 2.29 (t, 2H,
J ) 6.7 Hz), 3.74 (s, 3H), 12.15 (s, 1H); 13C NMR (CDCl3) δ
26.22, 27.46, 28.59, 33.99, 35.74, 50.96, 95.98, 170.90, 172.82;
MS m/z (relative intensity) 184 (M+, 70).
To a solution of diisopropylamine (4.47 g, 6.20 mL, 44.21
mmol) in 80 mL of THF at 0 °C n-butyllithium (18.47 mL of a
2.5 M solution in hexanes, 46.18 mmol) was slowly added.
After stirring for 20 min at 0 °C a solution of 10 (4 g, 21.71
mmol) in 20 mL of THF was added. After stirring for another
(31) Komiyama, M.; Bender, M. L. J . Am. Chem. Soc. 1978, 100,
5977.
(32) It should be stressed that under the pseudo first-order condi-
tions (10-fold excess of amino alcohol) minute amounts of water are
sufficient for complete hydrolysis.
(34) We acknowledge this particular suggestion by one of the
referees.
(35) Henderson, J . W.; Haake, P. J . Org. Chem. 1977, 42, 3989.
(33) Caplow, M. J . Am. Chem. Soc. 1969, 91, 3639.

2556 J . Org. Chem., Vol. 63, No. 8, 1998
Madder et al.
20 min at 0 °C (benzyloxy)methyl chloride (3 mL, 3.40 g, 21.71
mmol) was added. The reaction mixture was stirred overnight
at room temperature. The reaction was quenched with 50 mL
of a 5 N HCl-solution and 100 mL of Et2O. After separation
of the layers, the aqueous phase was extracted with Et2O. The
combined organic extracts were washed with H2O until
neutral. Purification by silica gel chromatography with pen-
tane/EtOAc (98:2) yielded 4.28 g of the desired 11 (65%): IR
(neat) 3500, 2953, 2864, 1748, 1715, 1651, 1615 cm-1; 1H NMR
(CDCl3, 500 MHz) δ 0.89 (s, 3H), 1.02 (s, 3H), 1.55 (dd, 1H, J
) 13.2 and 11.4 Hz), 1.60 (ddd, 1H, J ) 13.3, 6.5 and 2.2 Hz),
2.04 (dd, 1H, J ) 15.6 and 2.5 Hz), 2.10 (ddd, 1H, J ) 15.7,
2.0 and 1.4 Hz), 2.66 (m, 1H), 3.71 (dd, 1H, J ) 9.1 and 6.2
Hz), 3.74 (dd, 1H, J ) 9.1 and 3.8 Hz), 3.75 (s, 3H), 4.55 (s,
2H), 7.27-7.39 (m, 5H), 12.41 (s, 1H); 13C NMR (CDCl3) δ
24.43, 28.51, 31.08, 36.00, 37.34, 38.78, 51.09, 70.23, 72.88,
97.09, 127.12, 127.55, 127.94, 138.08, 170.75, 173.06.
Cerium(III) chloride heptahydrate (45 g, 123.19 mmol) was
added to a solution of rac-11 (25 g, 82 mmol) in 500 mL of
MeOH. The reaction mixture was stirred until a clear solution
was obtained and subsequently cooled to -55 °C. Sodium
borohydride (4.66 g, 123.19 mmol) was added portionwise.
After complete conversion (TLC-control) the mixture was
transferred to a 1 L flask. 200 mL of saturated aqueous NH4-
Cl solution and 50 mL of H2O were added, and most of the
methanol was removed under reduced pressure. The residue
was transferred into a separating funnel and, after addition
of another 200 mL of NH4Cl solution, extracted with Et2O. The
organic layer was dried over Na2SO4 and concentrated under
reduced pressure. The resulting crude product was purified
via flash silica gel chromatography followed by HPLC with
isooctane:EtOAc (80:20) leading to 19.27 g of the pure isomers
12c and 12t (77% combined yield).
Hz), 3.67 (s, 3H), 4.43 (d, 1H, J AB ) 11.6 Hz), 4.54 (d, 1H, J AB
) 11.5 Hz), 4.55 (m, 1H), 7.30-7.40 (m, 5H); 13C NMR (CDCl3)
δ 5.27, 6.96, 24.87, 30.01, 32.84, 34.26, 35.46, 39.59, 45.47,
51.32, 68.15, 72.13, 73.14, 127.53, 127.81, 128.30, 138.27,
174.20; MS m/z (relative intensity) 391 (M•+ - 29, 8), 91 (100),
45 (23), 41 (24). Anal. Calcd for C24H40O3Si: C, 68.53; H, 9.58.
Found: C, 68.58; H, 9.54.
r a c-[(1S,2R,3R)-[3-[(Ben zyloxy)oxy]-2-[(tr ieth ylsilyl)-
oxy]-5,5-d im eth ylcycloh exyl]m eth a n ol] (14c). Ester 13c
(1 g, 2.38 mmol) was dissolved in dry toluene (1.5 mL) and
cooled to -78 °C. After stirring for 1 h at -78 °C, 3.45 mL of
DIBALH (1.5 M in toluene, 5 mmol) was added, and the
reaction mixture was stirred for 2 h. Subsequently 4 mL of
Na-K tartrate was added, and the resulting mixture stirred
for 1 h at room temperature. After extraction with Et2O, the
organic layer was dried (MgSO4). After filtration and evapora-
tion of the solvent under reduced pressure, the residue was
purified by flash silica gel chromatography with pentane:
EtOAc (from 95:5 to 90:10). This yielded 937 mg (100%) of
alcohol 14c: IR (neat) 3410, 2955, 2921, 2875, 1494 cm-1; 1H
NMR (CDCl3, 500 MHz) δ 0.60 (q, 6H, J ) 7.9 Hz), 0.92 (s,
3H), 0.93 (s, 3H) 0.95 (t, 9H, J ) 8.0 Hz), 1.00-1.04 (m, 2H),
1.31 (dd, 2H, J ) 12.9 and 12.9 Hz), 1.70-1.77 (m, 1H), 1.82-
1.88 (m, 1H), 3.18 (dd, 1H, J ) 8.9 and 5.7 Hz), 3.39-3.43 (m,
1H), 3.43 (dd, 1H, J ) 8.7 and 8.7 Hz), 3.53-3.56 (m, 1H),
4.17 (m, 1H), 4.44 (d, 1H, J AB ) 11.7 Hz), 4.54 (d, 1H, J AB
)
11.7 Hz), 7.27-7.34 (m, 5H); 13C NMR (CDCl3) δ 5.41, 7.14,
25.50, 30.23, 33.04, 35.96, 36.21, 39.48, 41.84, 64.55, 67.33,
72.52, 73.09, 127.52, 127.83, 128.32, 138.30; MS m/z (relative
intensity) 391 (M•+ - H, 1), 363 (M•+ - CH2CH3, 2), 91 (100),
49 (19). Anal. Calcd for C23H40O3Si: C, 70.36; H, 10.27.
Found: C, 70.46; H, 10.14.
r a c-(1R,2S,3R)-3-[(Ben zyloxy)oxy]-2-(tr ieth ylsilyloxy)-
5,5-d im et h ylcycloh exa n eca r b a ld eh yd e (15c). Triethyl-
amine (3.19 mL, 22.92 mmol) was added to a solution of alcohol
14c (3 g, 7.64 mmol) in CH2Cl2 (17 mL) and DMSO (17 mL)
at 0 °C. Then sulfur trioxide-pyridine complex (3.04 g, 19.10
mmol) was added portionwise. After stirring for 1.5 h at room
temperature, the reaction mixture was diluted with 20 mL of
CH2Cl2 and poured into 20 mL of ice-water. Extraction with
CH2Cl2 was followed by drying over Na2SO4. After filtration
and removal of the solvent in vacuo the residue was purified
by flash silica gel chromatography hexane:EtOAc (from 98:2
to 95:5). This yielded 2.68 g (90%) of aldehyde 15c: IR (neat)
Alcoh ol 12c: Rf (pentane:EtOAc, 8:2) ) 0.23; IR (neat)
1
3512, 2951, 1711 cm-1; H NMR (CDCl3, 500 MHz) δ 0.93 (s,
3H), 0.99 (s, 3H), 1.08 (m, 1H), 1.42 (m, 1H), 1.48 (dd, 1H, J )
13.3 and 13.3 Hz), 1.77 (dd, 1H, J ) 13.3 and 13.3 Hz), 1.87
(m, 1H), 2.55 (ddd, 1H, J ) 13.5, 3.7 and 2.2 Hz), 3.46 (dd,
1H, J ) 9.1 and 4.9 Hz), 3.59 (dd, 1H, J ) 9.1 and 7.0 Hz),
3.71 (s, 3H), 4.35 (m, 1H), 4.52 (s, 2H), 7.22-7.38 (m, 5H); 13
C
NMR (CDCl3) δ 24.25, 29.95, 32.26, 34.68, 34.89, 37.42, 43.55,
54.41, 66.44, 72.35, 70.02, 127.22, 127.97, 128.00, 137.89,
175.63. Anal. Calcd for C18H26O4: C, 70.56; H, 8.55. Found:
C, 70.54; H, 8.45.
Alcoh ol 12t: Rf (pentane:EtOAc, 8:2) ) 0.15; IR (neat) 3498,
3031, 2952, 1738, cm-1; 1H NMR (CDCl3, 500 MHz) δ 0.94 (s,
3H), 0.99 (s, 3H), 1.01 (dd, 1H, J ) 13.2 and 13.2 Hz), 1.31
(ddd, 1H, J ) 13.5, 3.2 and 3.2 Hz), 1.37 (dd, 1H, J ) 13.3
and 13.3 Hz), 1.58 (ddd, 1H, J ) 13.3, 3.3 and 3.3 Hz), 1.96
(m, 1H), 2.60 (ddd, 1H, J ) 13.6, 10.2 and 3.6 Hz), 3.50 (dd,
1H, J ) 8.7 and 8.7 Hz), 3.56 (dd, 1H, J ) 4.3 and 9.1 Hz),
3.70 (dd, 1H, J ) 10.1 and 10.1 Hz), 3.71 (s, 3H), 3.95 (m, 1H),
4.51 (s, 2H), 7.22-7.38 (m, 5H); 13C NMR (CDCl3) δ 24.79,
30.02, 31.77, 38.74, 39.76, 40.45, 47.22, 51.83, 73.57, 75.62,
75.80, 127.61, 127.78, 128.40, 137.47, 175.55; MS m/z (relative
intensity) 306 (M•+, 1), 200 (23), 167 (43), 107 (26), 91 (100).
Anal. Calcd for C18H26O4: C, 70.56; H, 8.55. Found: C, 70.51;
H, 8.55.
2953, 2875, 1727, 1455 cm-1 1H NMR (CDCl3, 500 MHz) δ
;
0.55 (q, 6H, J ) 8.0 Hz), 0.90 (s, 3H), 0.91 (t, 9H, J ) 7.9 Hz),
0.99 (s, 3H) 1.01 (ddd, 1H, J ) 12.9, 3.1 and 2.6 Hz), 1.32 (dd,
1H, J ) 13.0 and 13.0 Hz), 1.40 (ddd, 1H, J ) 13.2, 3.1 and
2.3 Hz), 1.71 (dd, 1H, J ) 13.1 and 13.1 Hz), 1.92 (m, 1H),
2.33 (ddd, 1H, J ) 13.0, 3.4 and 1.9 Hz), 3.20 (dd, 1H, J ) 8.9
and 5.2 Hz), 3.43 (dd, 1H, J ) 9.0 and 9.0 Hz), 4.44 (d, 1H,
J AB ) 11.6 Hz), 4.55 (d, 1H, J AB ) 11.6 Hz), 4.64 (m, 1H), 7.25-
7.40 (m, 5H), 9.71 (s, 1H); 13C NMR (CDCl3) δ 5.27, 6.92, 24.86,
29.83, 32.52, 32.72, 35.71, 39.44, 52.56, 66.32, 71.69, 73.14,
127.57, 127.79, 128.30, 138.00, 204.64; MS m/z (relative
intensity) 391 (1), 239 (10), 91 (100), 89 (56). Anal. Calcd for
C
23H38O3Si: C, 70.72; H, 9.80. Found: C, 70.62; H, 9.98.
r a c-[(1S ,2S ,3R )-3-[(Be n zyloxy)oxy]-2-[(t er t -b u t yld i-
r a c-[Met h yl (1R,2S,3R)-3-[(Ben zyloxy)oxy]-2-(t r iet h -
ylsilyloxy)-5,5-dim eth ylcycloh exan e-1-car boxylate] (13c).
To a solution of alcohol 12c (0.1 g, 0.33 mmol) in DMF (1 mL)
was added imidazole (67 mg, 0.99 mmol), followed by triethyl-
silyl chloride (82 µL, 0.49 mmol). The reaction mixture was
stirred overnight at room temperature after which it was
poured into ice-water, extracted with Et2O, and washed with
a saturated NaHCO3 solution. The organic layer was dried
(Na2SO4) and concentrated in vacuo. The residue was purified
by flash silica gel chromatography with hexane:EtOAc (95:5).
This yielded 137 mg (99%) of the desired product 13c: IR
m eth ylsilyl)oxy]-5,5-d im eth ylcycloh exylm eth a n ol] (17t).
Imidazole (11.33 g, 166.53 mmol) and a few milligrams of
DMAP were added to a solution of TBDMSCl (12.55 g, 83.24
mmol) in 100 mL of dry DMF. The solution was cooled to 0
°C after which a solution of alcohol rac-12t (12.07 g, 39.38
mmol) in 30 mL of dry DMF was added. The reaction mixture
was stirred for 12 h at room temperature and was subse-
quently poured into ice-water (100 mL). After extraction with
pentane, the combined organic layers were washed with brine
and dried (MgSO4). The excess of solvent was removed under
reduced pressure and the residue purified by flash silica gel
chromatography with hexane:EtOAc (95:5) yielding 16.39 g
(99%) of 16t: IR (neat) 2928, 2856, 1740 cm-1; 1H NMR (CDCl3,
500 MHz) δ -0.06 (s, 3H), 0.015 (s, 3H), 0.82 (s, 9H), 0.94 (s,
3H), 0.98 (s, 3H), 1.29 (dd, 1H, J ) 13.3 and 13.3 Hz), 1.40
(dd, 1H, J ) 13.2 and 13.2 Hz), 1.50 (ddd, 1H, J ) 13.2, 3.5
and 3.5 Hz), 1.56 (ddd, 1H, J ) 13.6, 3.4 and 3.4 Hz), 1.78 (m,
1
(neat) 2951, 2874, 1742 cm-1; H NMR (CDCl3, 500 MHz) δ
0.53 (q, 6H, J ) 7.9 Hz), 0.89 (s, 3H), 0.92 (t, 9H, J ) 7.9 Hz),
0.97 (s, 3H), 1.00 (m, 1H), 1.30 (dd, 1H, J ) 13.0 and 13.0
Hz), 1.39 (m, 1H), 1.80 (dd, 1H, J ) 13.3 and 13.3 Hz), 1.85-
1.92 (m, 1H), 2.46 (ddd, 1H, J ) 13.3, 3.5 and 1.9 Hz), 3.18
(dd, 1H, J ) 9.0 and 5.6 Hz), 3.42 (dd, 1H, J ) 8.8 and 8.8

First Generation Nonenzymatic Hydrolases
J . Org. Chem., Vol. 63, No. 8, 1998 2557
1H), 2.63 (ddd, 1H, J ) 13.4, 9.9 and 3.8 Hz), 3.43 (dd, 1H, J
) 8.8 and 6.5 Hz), 3.50 (dd, 1H, J ) 8.8 and 2.9 Hz), 3.65 (s,
with Et2O. The organic layer was dried (MgSO4) and the
solvent removed in vacuo. The residue was purified by flash
silica gel chromatography with pentane:acetone:25% ammonia
solution (from 20:9.8:0.2 to 10:9.8:0.2). This yielded 75 mg
3H), 3.78 (dd, 1H, J ) 10.0 and 10.0 Hz), 4.48 (d, 1H, J AB
)
12.1 Hz), 4.54 (d, 1H, J AB ) 12.1 Hz), 7.30-7.35 (m, 5H); 13C
NMR (CDCl3) δ -4.41, -4.24, 18.24, 24.68, 26.03, 30.12, 32.16,
40.93, 41.42, 42.20, 49.16, 51.48, 71.60, 72.95, 127.41, 127.50,
128.28, 138.62, 175.96; MS m/z (relative intensity) 363 (M•+
- 57, 3), 271 (10), 91 (100).
(58%) of amino diol rac-4c: IR (neat) 3417, 2922, 1644 cm-1
;
1H NMR (CDCl3, 500 MHz) δ 0.91 (s, 3H), 0.93 (s, 3H), 1.01-
1.09 (m, 2H), 1.23 (dd, 1H, J ) 13.0 and 13.0 Hz), 1.27 (dd,
1H, J ) 13.0 and 13.0 Hz), 1.58-1.69 (m, 2H), 1.72-1.78 (m,
1H), 1.80-1.87 (m, 1H), 2.10 (dd, 1H, J ) 12.3 and 6.4 Hz),
2.18 (s, 3H), 2.42 (dd, 1H, J ) 12.3 and 7.9 Hz), 2.45-2.52 (m,
2H), 3.33 (dd, 1H, J ) 9.1 and 5.7 Hz), 3.50 (dd, 1H, J ) 9.1
and 7.3 Hz), 3.58-3.64 (m, 2H), 3.85 (m, 1H), 4.47 (s, 2H),
7.26-7.35 (m, 5H); 13C NMR (CDCl3) δ 25.08, 27.42, 30.52,
32.96, 35.59, 36.23, 37.77, 38.43, 42.74, 59.37, 60.27, 64.52,
67.54, 73.40, 73.62, 127.56, 127.66, 128.38, 138.28; MS m/z
(relative intensity) 349 (M•+, 2), 334 (M•+ - CH3, 5), 102 (100),
91 (25), 58 (57).
The ester 16t was then reduced to the corresponding alcohol
using DIBALH in toluene at -78 °C as described for 13c: IR
1
(neat) 3443, 2976, 2956 cm-1; H NMR (CDCl3, 500 MHz) δ
0.02 (s, 3H), 0.08 (s, 3H), 0.88 (s, 9H), 0.95 (s, 3H), 0.98 (s,
3H), 1.13 (dd, 1H, J ) 13.2 and 13.2 Hz), 1.18 (dd, 1H, J )
13.2 and 13.2 Hz), 1.40 (br, s, 1H), 1.45 (ddd, 1H, J ) 13.3,
3.3 and 3.3 Hz), 1.59 (ddd, 1H, J ) 13.6, 3.5 and 3.5 Hz), 1.72
(m, 1H), 1.83 (m, 1H), 3.30 (dd, 1H, J ) 10.0 and 10.0 Hz),
3.33 (dd, 1H, J ) 8.8 and 7.0 Hz), 3.52 (dd, 1H, J ) 8.8 and
3.1 Hz), 3.54 (m, 1H), 3.68 (dd, 1H, J ) 10.5 and 3.8 Hz), 4.41
(d, 1H, J AB ) 12.1 Hz), 4.53 (d, 1H, J AB ) 12.2 Hz), 7.30-7.35
(m, 5H); 13C NMR (CDCl3) δ 0.00, 17.92, 24.92, 25.72, 29.71,
32.31, 41.01, 41.29, 41.62, 43.08, 64.65, 71.59, 72.57, 127.01,
127.17, 127.91, 138.32; MS m/z (relative intensity) 335 (M•+
- 57, 4), 91 (100), 75 (25). Anal. Calcd for C23H40O3Si: C,
70.36; H, 10.27. Found: C, 70.24; H, 10.33.
(1R ,2R ,6S )-2-[(B e n z y lo x y )o x y ]-6-[[N -[(2R )-2-(h y -
dr oxym eth yl)-3,3-dim eth ylbu tyl]-N-m eth ylam in o]m eth yl]-
4,4-d im eth ylcycloh exa n -1-ol (5c) a n d (1S,2S,6R)-2-[(Ben -
zyloxy)oxy]-6-[[N-[(2R)-2-(h yd r oxym eth yl)-3,3-d im eth yl-
bu tyl]-N-m eth yla m in o]m eth yl]-4,4-d im eth ylcycloh exa n -
1-ol (6c). A mixture of aldehyde rac-15c (250 mg, 0.64 mmol),
amine 20 (332 mg, 1.28 mmol), and titanium(IV) isopropoxide
(0.377 mL, 1.28 mmol) was stirred at room temperature in a
round-bottomed flask with a CaCl2 drying tube. After 2 h the
viscous solution was diluted with 2.5 mL of absolute EtOH
(freshly distilled). Sodium cyanoborohydride (27 mg, 0.43
mmol) was added portionwise, and the resulting solution was
stirred for 20 h. Dilution with 5 mL of Et2O was followed by
addition of 2 mL of a 25% ammonia solution, and the resulting
inorganic precipitate was filtered and thoroughly washed with
EtOH. The filtrate was then concentrated in vacuo. The
residue was taken up in Et2O, and water was added. The
layers were separated, and the aqueous phase was extracted
with Et2O. The combined organic layers were then washed
with brine and dried (MgSO4). After filtration and concentra-
tion the residue was purified by flash silica gel chromatogra-
phy with pentane:EtOAc (96.5:3.5) leading to 325 mg (0.51
mmol, 80%) of a mixture of diastereomers which could not be
separated and was used as such in the following reaction.
r a c-[(1R,2R,3R)-3-[(Ben zyloxy)oxy]-2-[(ter t-b u t yld i-
m et h ylsilyl)oxy]-5,5-d im et h ylcycloh exa n e-1-ca r b a ld e-
h yd e] (18t). The alcohol 17t (2.31 g, 5.88 mmol) was oxidized
to aldehyde 18t using sulfur trioxide-pyridine complex as was
described for the synthesis of aldehyde 15c and led to 1.86 g
1
(81%) of aldehyde 18t: IR (neat) 2928, 1727 cm-1; H NMR
(CDCl3, 500 MHz,) δ -0.02 (s, 3H), 0.03 (s, 3H), 0.83 (s, 9H),
0.97 (s, 3H), 0.98 (s, 3H), 1.26 (dd, 1H, J ) 13.1 and 13.1 Hz),
1.31 (dd, 1H, J ) 13.1 and 13.1 Hz), 1.35 (ddd, 1H, J ) 13.4,
3.5 and 3.5 Hz), 1.54 (ddd, 1H, J ) 13.8, 3.3 and 3.3 Hz), 1.82
(m, 1H), 2.59 (m, 1H), 3.45 (dd, 1H, J ) 8.9 and 3.0 Hz), 3.49
(dd, 1H, J ) 8.9 and 6.1 Hz), 3.81 (dd, 1H, J ) 10.0 and 10.0
Hz), 4.41 (d, 1H, J AB ) 12.1 Hz), 4.55 (d, 1H, J AB ) 12.0 Hz),
7.30-7.35 (m, 5H), 9.64 (d, 1H, J ) 3.79 Hz); 13C NMR (CDCl3)
δ -4.14, -3.71, 18.09, 24.80, 25.90, 29.88, 31.20, 38.06, 40.93,
41.36, 55.11, 71.00, 71.64, 72.99, 127.41, 127.46, 128.22,
138.47, 204.73; MS m/z (relative intensity) 333 (M•+ - 57, 3),
213 (18), 121 (18), 105 (28), 91 (100), 77 (18). Anal. Calcd for
C
23H38O3Si: C, 70.72; H, 9.80. Found: C, 70.58; H, 9.79.
r a c-[(1R,2R,6S)-2-[(Ben zyloxy)oxy]-6-[[N-(3-h yd r oxy-
The mixture of diastereomeric amines 22c and 23c (100 mg,
0.16 mmol) was dissolved in 0.2 mL of THF and 0.79 mL of a
1 M TBAF solution (0.79 mmol) was added. The reaction
mixture was stirred overnight at room temperature. After
removal of the solvent under reduced pressure the crude
product was purified via HPLC with isooctane:acetone:25%
ammonia solution (45:9.8:0.2) leading to 28 mg of amino diol
6c and 19.3 mg of 5c (combined yield 73%).
pr opyl)-N-m eth ylam in o]m eth yl]-4,4-dim eth ylcycloh exan -
1-ol] (r a c-4c). A solution of aldehyde rac-15c (300 mg, 0.77
mmol), secondary amine 19 (312 mg, 1.54 mmol), and tita-
nium(IV) isopropoxide (0.452 mL, 1.54 mmol) was stirred at
room temperature in a round-bottomed flask with a calcium
chloride drying tube for 4 h. The reaction mixture was diluted
with absolute ethanol (0.8 mL) and subsequently sodium
cyanoborohydride (31 mg, 0.5 mmol) was added. After stirring
for 24 h, 1 mL of a 25% aqueous ammonia solution was added,
and the mixture was diluted with Et2O. After filtration of the
inorganic salts the solvent was removed in vacuo. The residue
was purified by flash silica gel chromatography with pentane:
EtOAc (9:1). This yielded 239 mg (54%) of rac-21c: IR (neat)
For 6c: Rf (isooctane:acetone:25% ammonia solution (40:
9.8:0.2)) ) 0.28; IR (neat) 3428, 2958, 1459 cm-1 1H NMR
;
(CDCl3, 500 MHz) δ 0.88 (s, 9H), 0.92 (s, 3H), 0.94 (s, 3H),
0.98 (m, 1H), 1.12 (m, 1H), 1.29 (dd, 1H, J ) 12.8 and 12.8
Hz), 1.40 (dd, 1H, J ) 13.1 and 13.1 Hz), 1.75-1.84 (m, 4H),
2.24 (s, 3H), 2.55 (dd, 1H, J ) 12.2 and 12.2 Hz), 2.58 (m, 1H),
2.77 (dd, 1H, J ) 10.7 and 10.7 Hz), 3.44 (dd, 1H, J ) 9.0 and
5.3 Hz), 3.56 (dd, 1H, J ) 9.0 and 6.6 Hz), 3.67 (dd, 1H, J )
10.4 and 10.4 Hz), 3.91 (m, 1H), 3.92 (m, 1H), 4.51 (s, 2H),
7.28-7.34 (m, 5H); 13C NMR (CDCl3): 25.12, 27.90, 30.06,
30.46, 31.16, 35.27, 36.35, 37.99, 38.34, 43.35, 44.27, 58.95,
62.20, 66.08, 66.53, 73.35, 127.50, 127.68, 128.35, 138.42; MS
m/z (relative intensity) 405 (M•+, 1), 91 (34), 58 (100). For 5c:
Rf (isooctane:acetone:25% ammonia solution (40:9.8:0.2)) )
2952, 2789, 1463 cm-1; H NMR (CDCl3, 500 MHz) δ 0.04 (s,
1
6H), 0.42 (q, 6H, J ) 7.8 Hz), 0.89 (s, 9H), 0.90 (s, 3H), 0.91
(s, 3H), 0.95 (t, 9H, J ) 8.0 Hz), 1.00 (m, 1H), 1.20 (m, 1H),
1.28 (dd, 1H, J ) 12.5 and 12.5 Hz), 1.29 (dd, 1H, J ) 13.1
and 13.1 Hz), 1.60-1.70 (m, 3H), 1.80-1.87 (m, 1H), 2.08 (dd,
1H, J ) 12.1 and 5.2 Hz), 2.14 (s, 3H), 2.20-2.30 (m, 2H),
2.38-2.47 (m, 1H), 3.15 (dd, 1H, J ) 8.9 and 5.6 Hz), 3.41
(dd, 1H, J ) 8.7 and 8.7 Hz), 3.60-3.68 (m, 2H), 3.95 (m, 1H),
4.43 (d, 1H, J AB ) 11.7 Hz), 4.53 (d, 1H, J AB ) 11.7 Hz), 7.27-
7.35 (m, 5H); 13C NMR (CDCl3) δ -4.69, 6.21, 7.81, 18.94,
26.06, 26.57, 30.97, 33.67, 37.03, 37.36, 38.38, 40.59, 43.34,
55.58, 62.35, 70.42, 73.29, 73.74, 128.12, 128.48, 128.92,
139.06; MS m/z (relative intensity) 486 (3), 216 (12), 91 (20),
58 (100).
1
0.27; IR (neat) 3415, 2953, 1454 cm-1; H NMR (CDCl3, 500
MHz,) δ 0.88 (s, 9H), 0.92 (s, 3H), 0.95 (s, 3H), 1.02 (m, 1H),
1.05 (m, 1H), 1.30 (dd, 1H, J ) 12.8 and 12.8 Hz), 1.52 (dd,
1H, J ) 13.2 and 13.2 Hz), 1.68 (m, 1H), 1.74 (m, 1H), 1.79
(m, 1H), 2.22 (s, 3H), 2.31 (dd, 1H, J ) 12.1 and 6.5 Hz), 2.41
(dd, 1H, J ) 12.1 and 7.6 Hz), 2.53-2.60 (m, 2H), 3.50-3.55
(m, 2H), 3.63 (dd, 1H, J ) 10.2 and 10.2 Hz), 3.92 (m, 1H),
To a solution of amine rac-21c (214 mg, 0.37 mmol) in THF
(2 mL) was added a 1 M solution of TBAF in THF (1.85 mL,
1.85 mmol, 5 equiv). After stirring for 15 h at room temper-
ature, 1 mL of water was added, and the mixture was extracted
3.93 (m, 1H), 4.48 (d, 1H, J AB ) 12.0 Hz), 4.53 (d, 1H, J AB
)
12.0 Hz), 7.28-7.36 (m, 5H); 13C NMR (CDCl3): 25.02, 27.95,
30.53, 31.22, 35.80, 36.17, 37.48, 38.47, 41.22, 42.33, 44.97,

2558 J . Org. Chem., Vol. 63, No. 8, 1998
Madder et al.
62.30, 66.40, 68.55, 73.44, 73.82, 127.65, 128.43, 138.12; MS
m/z (relative intensity) 390 (M•+ - CH3, 2), 158 (12), 91 (34),
58 (100).
nitrogen. Recrystallization was performed in a mixture of
isooctane and CH2Cl2.
(1R,2R,6S)-2-[(Ben zyloxy)oxy]-6-[[N-[(2R)-2-[[(ter t-bu -
t y ld im e t h y ls ily l)o x y ]m e t h y l]-3,3-d im e t h y lb u t y l]-N -
m eth yla m in o]m eth yl]-4,4-d im eth ylcycloh exa n -1-ol (24c).
To a solution of amino diol 5c (76.7 mg, 0.189 mmol) in dry
DMF (3 mL) was added triethylamine (0.058 mL, 0.416 mmol),
followed by a few mg of DMAP. Subsequently TBDMSCl (32
mg, 0.208 mmol) was added in one portion. After stirring for
20 h at room temperature brine was added and the mixture
extracted with Et2O. After drying (MgSO4), filtration and
concentration in vacuo the residue was purified by flash silica
gel chromatography with isooctane:acetone:25% ammonia
solution (90:9.8:0.2). This yielded 77 mg (78%) of silyl ether
24c next to 12 mg (16%) of starting material: [R]23D ) -23.72°,
[R]23365 ) -69.03° (c ) 1.13, CHCl3); IR (neat) 3437, 2955, 1471
(1S ,2R ,6S )-2-[(B e n z y lo x y )o x y ]-6-[[N -[(2R )-2-(h y -
dr oxym eth yl)-3,3-dim eth ylbu tyl]-N-m eth ylam in o]m eth yl]-
4,4-d im eth ylcycloh exa n -1-ol (5t) a n d (1R,2S,6R)-2-[(Ben -
zyloxy)oxy]-6-[[N-[(2R)-2-(h yd r oxym eth yl)-3,3-d im eth yl-
bu tyl]-N-m eth yla m in o]m eth yl]-4,4-d im eth ylcycloh exa n -
1-ol (6t). The synthesis of the all-trans amino diols is
analogous to the synthesis of the corresponding all-cis amino
diols and is fully detailed in the Supporting Information.
Spectral data for 5t: Rf (isooctane:acetone:25% ammonia
solution (40:9.8:0.2)) ) 0.50; IR (neat) 3410, 2952, 2855, 1365
cm-1; 1H NMR (CDCl3, 500 MHz) δ 0.81 (dd, 1H, J ) 12.9 and
12.9 Hz), 0.89 (s, 12H), 0.95 (s, 3H), 1.02 (dd, 1H, J ) 13.1
and 13.1 Hz), 1.36 (ddd, 1H, J ) 13.3, 3.3 and 3.3 Hz), 1.44
(ddd, 1H, J ) 13.4, 3.5 and 3.3 Hz), 1.46 (m, 1H), 1.73-1.79
(m, 2H), 2.20 (s, 3H), 2.19 (m, 1H), 2.37 (ddd, 1H, J ) 12.4,
2.9 and 1.0 Hz), 2.52 (dd, 1H, J ) 12.2 and 7.3 Hz), 3.02 (dd,
1H, J ) 9.8 and 9.8 Hz), 3.42 (dd, 1H, J ) 9.0 and 6.2 Hz),
3.50 (dd, 1H, J ) 10.7 and 7.5 Hz), 3.57 (dd, 1H, J ) 9.0 and
4.7 Hz), 3.68 (ddd, 1H, J ) 10.7, 3.8 and 1.1 Hz), 4.47 (s, 2H),
7.26-7.37 (m, 5H); 13C NMR (CDCl3) δ 25.03, 28.00, 30.26,
31.80, 32.53, 37.41, 39.81, 40.05, 41.59, 43.38, 45.42, 60.91,
61.56, 65.55, 73.52, 76.06, 78.94, 127.60, 127.72, 128.40,
1
cm-1; H NMR (CDCl3, 500 MHz) δ 0.04 (s, 6H), 0.88 (s, 9H),
0.91 (s, 3H), 0.93 (s, 9H), 0.95 (s, 3H), 1.07-1.12 (m, 2H), 1.25
(m, 1H), 1.40 (dd, 1H, J ) 13.1 and 13.1 Hz), 1.47 (dd, 1H, J
) 13.3 and 13.3 Hz), 1.68 (m, 1H), 1.83 (m, 1H), 2.18 (s, 3H),
2.19 (dd, 1H, J ) 12.3 and 3.1 Hz), 2.28 (dd, 1H, J ) 12.7 and
6.0 Hz), 2.34 (dd, 1H, J ) 12.4 and 9.3 Hz), 2.38 (dd, 1H, J )
12.6 and 6.7 Hz), 3.48 (dd, 1H, J ) 9.0 and 5.1 Hz), 3.58 (dd,
1H, J ) 9.0 and 6.3 Hz), 3.69 (dd, 1H, J ) 10.1 and 3.5 Hz),
3.73 (dd, 1H, J ) 10.2 and 4.4 Hz), 4.02 (m, 1H), 4.50 (d, 1H,
J AB ) 12.0 Hz), 4.53 (d, 1H, J AB ) 12.0 Hz), 7.28-7.38 (m,
5H); 13C NMR (CDCl3) δ -5.52, -5.47, 18.13, 24.98, 25.88,
25.96, 28.45, 30.65, 32.31, 33.09, 36.33, 38.16, 38.59, 43.96,
48.12, 57.62, 61.78, 62.11, 69.19, 73.40, 73.43, 127.53, 127.61,
128.35, 138.43; MS m/z (relative intensity) 519 (M•+, <0.5),
504 (M•+ - CH3, 0.5), 462 (M•+ - tBu, 4), 58 (100).
137.63; MS m/z (relative intensity) 405 (M•+, 1), 404 (M•+
-
H, 2), 390 (M•+ - CH3, 3), 58 (100). Spectral data for 6t: Rf
(isooctane:acetone:25% ammonia solution (40:9.8:0.2)) ) 0.50;
1
IR (neat) 3410, 2951, 1464 cm-1; H NMR (CDCl3, 500 MHz)
δ 0.85 (dd, 1H, J ) 13.1 and 13.1 Hz), 0.90 (s, 12H), 0.93 (m,
1H), 0.96 (s, 3H), 1.33-1.37 (m, 2H), 1.57 (m, 1H), 1.81 (m,
1H), 1.92 (m, 1H), 2.17 (m, 1H), 2.20 (s, 3H), 2.39 (m, 1H),
2.52 (dd, 1H, J ) 11.9 and 11.9 Hz), 2.67 (dd, 1H, J ) 12.2
and 6.9 Hz), 3.15 (dd, 1H, J ) 9.7 and 9.7 Hz), 3.50-3.51 (m,
2H), 3.63 (dd, 1H, J ) 10.9 and 7.1), 3.82 (ddd, 1H, J ) 10.9,
3.9 and 1.2 Hz), 4.51 (s, 2H), 7.25-7.34 (m, 5H); 13C NMR
(CDCl3) δ 25.26, 28.06, 30.40, 31.63, 32.64, 36.78, 40.09, 40.32,
41.54, 41.93, 46.43, 60.30, 63.78, 64.42, 73.44, 74.71, 78.92,
127.62, 128.39, 138.04; MS m/z (relative intensity) 404 (M•+
- H, 1), 390 (M•+ - CH3, 2), 91 (22), 58 (100).
(1S ,2R ,6S )-2-(H y d r o x y m e t h y l)-6-[[N -[(2R )-2-(h y -
dr oxym eth yl)-3,3-dim eth ylbu tyl]-N-m eth ylam in o]m eth yl]-
4,4-d im eth ylcycloh exa n -1-ol (7t). To a solution of alcohol
24c (11 mg, 0.021 mmol) in CH2Cl2 (0.2 mL) was added Dess-
Martin reagent (44 mg, 0.105 mmol). After complete conver-
sion (TLC control) the reaction mixture was diluted with Et2O,
and 1 mL of a saturated sodium bicarbonate solution was
added followed by 2 mL of a 5% sodium thiosulfate solution.
The organic layer was separated and dried (Na2SO4). The
excess solvent was removed under reduced pressure and the
residue chromatographed on silica gel using isooctane:acetone:
25% ammonia solution (95:4.9:0.01). This yielded 6 mg (55%)
of the desired ketone (IR 1711 cm-1) which was immediately
taken into the next reaction. A solution of the ketone (10 mg,
0.019 mmol) in THF was transferred to a reaction flask
containing 3 mL of predried liquid ammonia and 0.3 mL of
dry methanol. Subsequently, lithium wire was added until
the solution turned blue. After stirring for 15 min, solid
ammonium chloride was added to destroy the excess lithium.
The ammonia was evaporated, the residue was dissolved in
CH2Cl2, and the inorganic salts were filtered. The filtrate was
concentrated in vacuo yielding an oil which was purified by
flash silica gel chromatography with isooctane:acetone:25%
ammonia solution (90:9.8:0.2). This yielded 5 mg (61%) of the
desired amino diol 25t which was not further characterized
but immediately converted to the known amino triol 7t.
To a solution of amino diol 25t (2.5 mg, 0.0058 mmol) in
THF (0.1 mL) was added a 1 M solution of TBAF in THF (12
mL, 0.012 mmol). The reaction mixture was stirred for 20 h
at room temperature. The excess solvent was removed under
reduced pressure and the residue purified by flash silica gel
chromatography with isooctane:acetone:25% ammonia solution
(20:9.8:0.2). Thus 1.5 mg of amino triol 7t (81%) was ob-
tained: IR (neat) 3381, 2954, 1472 cm-1; 1H NMR (CDCl3, 500
MHz) δ 0.87 (dd, 1H, J ) 13.4 and 13.4 Hz), 0.89 (m, 1H),
0.90 (s, 3H), 0.92 (s, 9H), 0.98 (s, 3H), 1.21 (ddd, 1H, J ) 13.4,
3.5 and 3.0 Hz), 1.28 (ddd, 1H, J ) 13.4, 3.3 and 3.1 Hz), 1.47
(m, 1H), 1.79-1.89 (m, 2H), 2.24 (s, 3H), 2.31 (dd, 1H, J )
12.4 and 5.0 Hz), 2.39 (dd, 1H, J ) 12.3 and 2.9 Hz), 2.53 (dd,
1H, J ) 12.2 and 9.5 Hz), 2.55 (dd, 1H, J ) 12.0 and 10.9 Hz),
3.27 (dd, 1H, J ) 9.7 and 9.7 Hz), 3.58 (dd, 1H, J ) 10.7 and
7.7 Hz), 3.60 (dd, 1H, J ) 10.8 and 4.1 Hz), 3.65 (dd, 1H, J )
10.9 and 6.4 Hz), 3.80 (dd, 1H, J ) 10.9 and 3.9 Hz); 13C NMR
(CDCl3) δ 25.29, 28.04, 30.41, 31.92, 32.45, 36.10, 39.63, 41.23,
(1R ,2R ,6S )-2-(H y d r o x y m e t h y l)-6-[[N -[(2R )-2-(h y -
dr oxym eth yl)-3,3-dim eth ylbu tyl]-N-m eth ylam in o]m eth yl]-
4,4-d im eth ylcycloh exa n -1-ol (7c). Sodium metal was added
in small portions to a solution of amino diol 5c (95.6 mg, 0.236
mmol) in THF (1 mL) and 15 mL of freshly distilled ammonia
until the solution remained deep blue. After stirring for 4 h
at -33 °C, the excess sodium was destroyed by addition of solid
ammonium chloride. The ammonia was allowed to evaporate,
the residue was dissolved in CH2Cl2, and the inorganic
precipitate was filtered off. The solvent was removed under
reduced pressure, and the residue was purified via flash silica
gel chromatography with isooctane:acetone:25% ammonia
solution (70:29.4:0.6). This yielded 51.3 mg (69%) of amino
diol 7c: IR (neat) 3380, 2953, 1469, 1367 cm-1; 1H NMR (CD3-
CN, 500 MHz) δ 0.88 (s, 6H), 0.92 (s, 9H), 1.02 (ddd, 1H, J )
12.9, 3.1 and 2.4 Hz), 1.09 (ddd, 1H, J ) 13.0, 3.6 and 2.2 Hz),
1.20 (dd, 1H, J ) 13.7 and 12.7 Hz), 1.31 (dd, 1H, J ) 13.1
and 13.1 Hz), 1.58-1.67 (m, 2H), 1.75 (m, 1H), 2.19 (s, 3H),
2.29-2.30 (m, 2H), 2.49 (ddd, 1H, J ) 12.3, 2.6 and 2.6 Hz),
2.58 (dd, 1H, J ) 12.1 and 11.7 Hz), 3.45 (dd, 1H, J ) 10.6
and 5.0 Hz), 3.51 (dd, 1H, J ) 10.6 and 6.7 Hz), 3.56 (dd, 1H,
J ) 9.9 and 9.9 Hz), 3.78 (ddd, 1H, J ) 10.1, 3.4 and 2.6 Hz),
3.81 (m, 1H); 13C NMR (CDCl3) δ 25.11, 27.90, 30.46, 31.18,
32.98, 35.30, 35.62, 37.75, 39.84, 41.26, 44.70, 61.92, 62.73,
65.37, 66.37, 66.62; MS m/z (relative intensity) 315 (M•+, 1),
300 (M•+ - CH3, 3), 58 (100), 44 (28). Anal. Calcd for C18H37
-
NO3: C, 68.53; H, 11.82; N, 4.44. Found: C, 68.04; H, 11.77;
N, 4.28.
The structure of 7c was further proven via X-ray diffraction
analysis of the ammonium salt obtained in the following way:
to a solution of (S)-(+)-camphorsulfonic acid (9 mg, 0.039
mmol) in CH2Cl2 (0.7 mL) was added a solution of amino triol
7c (12.2 mg, 0.039 mmol) in CH2Cl2 (1 mL). After stirring for
30 min, the excess CH2Cl2 was removed under a stream of

First Generation Nonenzymatic Hydrolases
J . Org. Chem., Vol. 63, No. 8, 1998 2559
41.52, 41.94, 46.79, 59.49, 63.08, 63.15, 68.72, 82.29; MS m/z
(relative intensity) 300 (M•+ - CH3, 1), 258 (2), 58 (100). Anal.
Calcd for C18H37NO3: C, 68.53; H, 11.82; N, 4.44. Found: C,
67.54; H, 11.69; N, 4.41.
67.46, 70.75, 72.91, 95.52, 127.56, 127.65, 128.37, 138.42; MS
m/z (relative intensity) 492 (M•+ - H, <0.5), 478 (M•+ - CH3,
1), 436 (M•+ - tBu, 1), 58 (100).
Deprotection of the benzyl ether was performed using
lithium in liquid ammonia as described for the deprotection
of 5c (reflux time at -33 °C: 4 h). The obtained crude product
was purified by flash silica gel chromatography with isooctane:
acetone:25% ammonia solution (60:39.2:0.8). This yielded 23.5
mg (93%) of amino triol 8c: 1H NMR (CDCl3, 500 MHz) δ 0.88
(s, 9H), 0.90 (s, 3H), 0.93 (s, 3H), 1.03 (m, 2H), 1.21 (dd, 1H,
J ) 12. 8 and 12.8 Hz), 1.25 (dd, 1H, J ) 13.0 and 13.0 Hz),
1.62 (dddd, 1H, J ) 10.5, 10.5, 3.0 and 2.9 Hz), 1.72 (tt, 2H, J
) 6.2 and 6.2 Hz), 1.78-1.83 (m, 2H), 2.18 (s, 3H), 2.19 (m,
1H), 2.38 (dd, 1H, J ) 12.1 and 8.0 Hz), 2.50 (ddd, 1H, J )
12.3, 2.5 and 2.5 Hz), 2.58 (dd, 1H, J ) 12.1 and 12.1 Hz),
3.26 (dd, 1H, J ) 9.2 and 4.9 Hz), 3.48 (ddd, 1H, J ) 9.5, 5.9
and 5.9 Hz), 3.53-3.61 (m, 4H), 3.67 (ddd, 1H, J ) 9.5, 6.7
and 6.7 Hz), 3.77 (m, 1H), 3.78 (m, 1H), 4.59 (s, 2H); 13C NMR
(CDCl3) δ 25.70, 28.16, 30.79, 31.44, 32.35, 33.19, 35.30, 36.15,
38.08, 39.30, 40.73, 44.42, 58.88, 63.53, 63.39, 63.57, 64.30,
66.66, 69.66, 94.78.
(1S,2R,6S)-6-[[N-[(2R)-2-(Hyd r oxym eth yl)-3,3-d im eth -
ylb u t yl]-N -m e t h yla m in o]m e t h yl]-2-[[(3-h yd r oxyp r o-
p oxy)m eth oxy]m eth yl]-4,4-d im eth ylcycloh exa n -1-ol (8t).
Amino triol 8t was prepared in an analogous way which is
fully detailed in the Supporting Information. Spectral data
for 8t: Rf (isooctane:acetone:25% ammonia solution (65:34.2:
0.8) ) 0.2; 1H NMR(CDCl3, 500 MHz) δ 0.82 (dd, 1H, J ) 12.9
and 12.9 Hz), 0.90 (s, 12H), 0.96 (s, 3H), 1.10 (dd, 1H, J )
13.2 and 13.2 Hz), 1.32 (ddd, 1H, J ) 13.3, 3.3 and 3.3 Hz),
1.40 (ddd, 1H, J ) 13.1, 3.2 and 3.2 Hz), 1.40-1.43 (m, 1H),
1.69 (tt, 2H, J ) 6.5 and 6.5 Hz), 1.71-1.81 (m, 2H), 2.20 (s,
3H), 2.21 (m, 1H), 2.34 (dd, 1H, J ) 12.5 and 2.8 Hz), 2.49
(dd, 1H, J ) 12.4 and 10.1 Hz), 2.52 (dd, 1H, J ) 12.3 and 8.3
Hz), 3.09 (dd, 1H, J ) 9.9 and 9.9 Hz), 3.48-3.64 (m, 8H),
4.58 (s, 2H); 13C NMR (CDCl3) δ 24.99, 28.00, 30.29, 31.59,
31.85, 32.22, 35.52, 39.89, 40.17, 40.76, 41.35, 46.52, 58.01,
58.71, 61.17, 62.47, 66.11, 68.14, 77.79, 94.30.
(1R,2R,6S)-6-[[N-[(2R)-2-[[(ter t-Bu t yld im e t h ylsilyl)-
oxy]m eth yl]-3,3-dim eth ylbu tyl]-N-m eth yla m in o]m eth yl]-
2-(h ydr oxym eth yl)-4,4-dim eth ylcycloh exan -1-ol (25c). The
benzyl deprotection was performed under standard conditions
with lithium in liquid ammonia as described above for the
preparation of 7c. The procedure led to 52 mg of the desired
amino diol 25c (100%): [R]23 ) -26.29°, [R]23 ) -73.71° (c
D
365
) 0.62, CHCl3); Rf (isooctane:acetone:25% ammonia solution
(85:14.7:0.3)) ) 0.24; IR (neat) 3628, 3478, 2956, 2857, 1472
1
cm-1; H NMR (CDCl3, 500 MHz) δ 0.04 (s, 3H), 0.05 (s, 3H),
0.88 (s, 9H), 0.90 (s, 3H), 0.93 (s, 9H), 0.98 (s, 3H), 1.03 (m,
1H), 1.30 (m, 1H), 1.48 (dd, 1H, J ) 13.0 and 13.0 Hz), 1.63
(m, 2H), 1.68 (m, 2H), 2.25 (m, 4H), 2.38 (m, 1H), 2.44 (m,
1H), 2.49 (dd, 1H, J ) 12.9 and 5.4 Hz), 2.94 (m, 1H), 3.64 (m,
1H), 3.70 (dd, 1H, J ) 10.3 and 3.5 Hz), 3.74 (m, 1H), 3.84 (m,
1H), 4.14 (m, 1H); 13C NMR (CDCl3) δ -5.56, 18.13, 24.62,
25.96, 28.49, 30.69, 32.32, 33.09, 35.62, 36.09, 38.46, 38.86,
44.23, 48.16, 58.15, 61.93, 62.25, 67.10, 71.70; MS m/z (relative
intensity) 429 (M•+, 1), 414 (M•+ - CH3, 2), 372 (M•+ - tBu,
4), 58 (100).
(1R,2R,6S)-6-[[N-[(2R)-2-(Hyd r oxym eth yl)-3,3-d im eth -
ylb u t yl]-N -m e t h yla m in o]m e t h yl]-2-[[(3-h yd r oxyp r o-
p oxy)m eth oxy]m eth yl]-4,4-d im eth ylcycloh exa n -1-ol (8c).
To a solution of amino diol 25c (50 mg, 0.116 mmol) in THF
(1 mL) at -78 °C was added n-butyllithium (1.5 M in hexanes,
88 mL, 0.140 mmol). After stirring for 10 min at -78 °C,
stirring was continued at -30 °C for 30 min. The reaction
mixture was cooled again to -78 °C, and chloromethyl ether
26 (37 mg, 0.174 mmol) was added. Stirring was continued
overnight while the solution was allowed to slowly reach rt.
After dilution with Et2O the reaction mixture was concentrated
in vacuo. The residue was dissolved in CH2Cl2. After the
addition of a saturated ammonium chloride solution, the
mixture was extracted with CH2Cl2. The organic layer was
dried (Na2SO4) and concentrated in vacuo. The residue was
purified by flash silica gel chromatography with isooctane:
acetone:25% ammonia solution (93:6.9:0.1). This yielded 48
mg (68%) of the desired product which was immediately
further deprotected.
Ack n ow led gm en t. A. Madder is indebted to the
“Fonds voor Wetenschappelijk Onderzoek - Vlaanderen”
for a position as Research Assistant. The FWO is
thanked for financial support to the laboratory.
To a solution of the amino alcohol (47 mg, 0.077 mmol) in
THF (0.8 mL) was added a 1 M solution of TBAF in THF (0.232
mL, 0.232 mmol). The reaction mixture was stirred for 12 h
at room temperature. After removal of the solvent under
reduced pressure, the residue was purified by flash silica gel
chromatography with isooctane:acetone:25% ammonia solution
(85:14.7:0.3). This yielded 30.3 mg (80%) of amino diol 27c:
[R]23D ) -28.07°, [R]23365 ) -82.62° (c ) 1.45, CHCl3); IR (neat)
Su p p or tin g In for m a tion Ava ila ble: Schemes describing
the synthesis of derivatives 19, (R)-20, rac-4t, 5t, 6t, 8t and
experimental procedures including spectroscopic data for the
preparation of these compounds. Detailed procedure for
determination of half-live values. X-ray structure of the
camphor sulfonate of amine 7c: atomic coordinates and
stereoscopic diagram. X-ray data are deposited with the
Cambridge Crystallographic Data Centre. 1H NMR and 13C
NMR spectra for compounds 4c, 4t, 5c, 5t, 6c, 6t, 7c, 7t, 8c,
8t, rac-11c, rac-12c, rac-12t, rac-15c, rac-18t and 24c (44
pages). This material is contained in libraries on microfiche,
immediately follows this article in the microfilm version of the
journal, and can be ordered from the ACS; see any current
masthead page for ordering information.
1
3384, 2924, 1466 cm-1; H NMR (CDCl3, 500 MHz) δ 0.87 (s,
9H), 0.92 (s, 3H), 0.95 (s, 3H), 1.04 (m, 1H), 1.11 (m, 1H), 1.25
(br s, 1H), 1.29 (dd, 1H, J ) 13.0 and 13.0 Hz), 1.41 (dd, 1H,
J ) 13.2 and 13.2 Hz), 1.67 (m, 1H), 1.74 (m, 1H), 1.79 (m,
1H), 1.90 (quintet, 2H, J ) 6.3 Hz), 2.20 (s, 3H), 2.29 (m, 1H),
2.39 (dd, 1H, J ) 11.9 and 7.8 Hz), 2.55 (m, 2H), 3.50 (dd, 1H,
J ) 9.5 and 4.7 Hz), 3.54-3.69 (m, 6H), 3.88 (m, 1H), 3.91 (m,
1H), 4.51 (s, 2H), 4.65 (s, 2H), 7.27-7.36 (m, 5H); 13C NMR
(CDCl3) δ 25.04, 27.93, 29.97, 30.48, 31.20, 32.94, 35.64, 36.07,
37.59, 38.32, 42.15, 44.95, 61.95, 62.34, 64.77, 66.40, 67.11,
J O971935+