An unusual functional group interaction and its potential to reproduce steric and electrostatic features of the transition states of peptidolysis

Article abstract of DOI:10.1016/j.bmc.2006.01.031

The donor-acceptor interaction between a tertiary amine and an aldehyde, first observed among a select class of alkaloids, was deliberately established in a peptidomimetic (1a-c) to mimic features of the two principal transition states of peptide hydrolys

Full text of DOI:10.1016/j.bmc.2006.01.031

Bioorganic & Medicinal Chemistry 14 (2006) 3835–3847
An unusual functional group interaction and its potential
to reproduce steric and electrostatic features of the transition
states of peptidolysis
Arnaud Gautier, Delphine Pitrat and Jens Hasserodt*
Laboratoire de Chimie, UMR 5182 ENS/CNRS, Ecole Normale Supe´rieure de Lyon, 46 alle´e d’Italie, 69364 Lyon cedex 07, France
Received 28 October 2005; revised 13 January 2006; accepted 17 January 2006
Available online 7 February 2006
Abstract—The donor–acceptor interaction between a tertiary amine and an aldehyde, first observed among a select class of
alkalo¨ıds, was deliberately established in a peptidomimetic (1a–c) to mimic features of the two principal transition states of peptide
hydrolysis. Compounds 1a–c show preferential adoption in methanol and water of a ‘folded’ conformation displaying the interac-
tion. Proportions of the folded form in MeOH range from 45% to 70% and can reach 84% in buffer. Significantly, three tendencies
for the folded/unfolded equilibrium are observed: increasing solubility and polarity of the medium and decreasing temperature
results in a higher extent of folding. In the absence of any parameter set available for this weak bond, no modeling studies were
conducted to aid in the design of 1a–c. The successful straightforward synthesis of 1 and its folding and inhibition results with
HIV-1 peptidase using FRET technology encourage studies to further preorganize candidate molecules and to screen the structure
space by modeling and parallel combinatorial chemistry.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
sults in the extrusion of the localized water molecule to
be found in crystal structures of ligand-free peptidases
between the two aspartic residues.5 This water molecule
is generally held to be the second substrate molecule of
peptide hydrolysis, itself being a bimolecular reaction.
This has led to the view that hydroxyethylene-based
The imitation of the features of the transition states of
peptide-bond hydrolysis has long been applied to the de-
sign of potent inhibitors of aspartic peptidases as part of
a fundamental teaching of enzymology.1 This strategy
evolved from the discovery that certain naturally occur-
ring peptides, containing a non-proteinogenic amino
acid called statine (pepstatine, Fig. 1), represent a gener-
al inhibitory principle for the majority of aspartic
peptidases.2
P1
O
H
N
Iva-Val-Val
Ala-Sta-OH
N
H
N
N
H
H
O
P1'
OH
O
OH
O
With the advent of the age of structure-based design of
potent HIV-1 peptidase inhibitors at the end of the
1980s, researchers have chosen mainly the hydroxyethyl-
ene moiety (Fig. 1) contained in statine as the point of
departure to elaborate.3 It is now well-established
knowledge that the hydroxyl group therein interacts
with the catalytic machinery of the active site, that is,
with the two aspartic acid residues.4 This interaction re-
Aspartic peptidase
substrate
Statine (Sta)
Pepstatine
P1
O
P1
P1'
P
1OH
H
N
X
N
H
N
H
N
H
N
H
P
P1'
OH
O
O
P1'
Reduced amine
Hydroxyethylamine
Phosph-inate,
-onate, -onamidate
(X=O, CH2, NH)
P1
O
P1
O
P1
O
F
F
N
N
N
N
H
H
H
H
Keywords: Nucleophile–electrophile interactions; Peptidomimetics;
Enzyme inhibitors; Aspartic peptidases; Asymmetric synthesis; Con-
formational analysis.
OH
O
P1'
O
-Hydroxy-
-amino acid
Fluoroketone
Ketoamide
*
Corresponding author. Tel.: +33 472 72 8394; fax: +33 472 72
8860; e-mail: jens.hasserodt@ens-lyon.fr
Figure 1. A selection of transition-state isosteres of peptide hydrolysis.
0968-0896/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2006.01.031

3836
A. Gautier et al. / Bioorg. Med. Chem. 14 (2006) 3835–3847
inhibitors act as ‘collected-substrate’ or ‘multi-substrate’
inhibitors,6 thus benefiting from a considerable entropic
advantage for increased affinity. These highly potent
inhibitors have met with remarkable success in the strug-
gle against related diseases.
two main transition states (TS1, TS2, Fig. 2) separated
by the tetrahedral intermediate (INT).13,14 Some of these
workers find a third barrier leading to product forma-
tion, following a second, fully zwitterionic, intermediate.
According to the main pillar of enzymology, enzymes
have evolved catalytic power by differentially binding
to transition state(s) (here: TS1, TS2) over ground states
(ES, EP).15,16
It is also generally maintained that the tetrahedral nat-
ure of the secondary alcohol function in the hydroxyeth-
ylene moiety reproduces the geometry of the ‘tetrahedral
intermediate,’ or ‘transition state,’ as is sometimes used
in its place, which led to the introduction of the term
‘transition-state isostere.’ Many other such isosteres
have been explored in the wake of the statin-based ap-
proach, such as phosphonic acid- and phosphinic
acid-based moieties,7 a-perfluorinated ketones,8 a-keto-
carboxamides,9 or sulfoxide and sulfone-containing
units (Fig. 1). This listing does not account for the other
penultimate constituent of the transient species on the
reaction coordinate of peptide hydrolysis, namely the
amine leaving group. In fact, many of the above isoster-
es have been combined with substituents on the C-termi-
nal side representing this part, which, if still containing a
nitrogen atom, often necessitate their spatial separation
from the CHOH moiety by at least one carbon atom (see
‘hydroxyethylamine,’ Fig. 1), in effect rendering the
nitrogen a very basic center with the pKa of a tertiary
amine. A more close resemblance with species on the
reaction coordinate can be found in phosphonates and
phosphonamidates;10 which led to the design of numer-
ous potent hydrolase inhibitors. However, their electro-
static properties have been calculated to be different
from those of transition states associated with alkaline
ester hydrolysis.11 They also suffer from the practical
inconvenience of hydrolytic instability in aqueous media
which somewhat limits their pharmaceutical applica-
tions. Sulfonamidates appear to have had little success
in inhibitor design.12
In the context of enzyme inhibition, as well as that of
catalytic-antibody induction, it has thus been found
regrettable that current transition-state isosteres are of-
ten far from faithfully reproducing steric and electrostat-
ic elements of the respective transition state(s)
(Fig. 2),17,18,11 in this case a partial charge on the car-
bonyl oxygen with an associated pKa value distinctly dif-
ferent from that of either hydroxyl groups or
phosphinates, -onates, and -onamidates, a C–O bond
order around 1.5, a pseudo-tetrahedral nature of the
carbon, and, for the 2nd transition state, an additional
partial charge on the nitrogen and an unusually long
C–N single bond (Table 1).
2. Results and discussion
We wish to explore an infrequently observed function-
al group for its capacity to mimic at the same time
some essential geometric and also electrostatic features
of the transition state(s), both being different from
those of the tetrahedral intermediate.19 This molecular
moiety results from the weak interaction of a tertiary
amine with a ketone (or aldehyde) function and has
been observed initially in a particular class of alka-
loids20,21 in the form of trans-annular functional
group interactions, subsequently studied in further de-
tail in nature-inspired compounds22 and eventually
helped in the establishment of the Burgi–Dunitz tra-
¨
jectory as a fundamental element of Physical Organic
Chemistry23 (Fig. 3).
In fact, as many theoretical studies have shown, the
reaction coordinate of carboxamide hydrolysis features
Figure 2. A schematic reaction coordinate of the mechanism of peptide hydrolysis within the active site of aspartic peptidases.

A. Gautier et al. / Bioorg. Med. Chem. 14 (2006) 3835–3847
3837
Table 1. Comparison of bond lengths
carbonyl-containing substrates (being the subject of the
study of Burgi and Dunitz) finds its inverse parallel in
¨
˚
Bond length [A]
Bond nature
Ref.
the final step (TS2) of carboxamide hydrolysis.
Ground state
C–N (MeNH2)
1.47
1.49
1.41
1.19
25,26
25
25
+
By contrast, a forced interaction between a tertiary
amine and a carboxylic acid on a preorganizing matrix,
even though an attractive solution on initial consider-
ation, cannot be envisaged for inhibitor design: while
Dunitz et al. observed a C—N distance of only 2.49 A
˚
(the van der Waals distance is rC + rN = 3.20 A) in the
C–NR3
C–O (MeOH)
C@O (MeCHO)
25
Ground-state peptide bond
C@O
C–N
˚
1.23
1.33
25
25
crystal structure of a naphthalene-based system contain-
ing a tertiary amine group and a carboxylic acid group,32
in aqueous solution at around pH 7 only the classic zwit-
terionic amino acid form would exist.
Transition state peptide bond
C–N calculated in TS 1, TS 2
C–O calculated in TS 1
1.52–2.21
<1.296
27,28
12
GS d+N ! C@Odꢀ (X-ray)
C–N
C@O
1.62–2.91
1.21–1.31
29,35
29,35
Some more recent sporadic observations of the weak
d+N ! C@Odꢀ interaction (C,33 D,34 E,30 and F35) have
helped us in assessing the feasibility of introducing it
into a peptide-derived peptidomimetic.5 Importantly,
detection of the d+N ! C@Odꢀ interaction is not limited
to a trans-annular context, nor does it require the graft-
ing of the interaction partners onto a rigid framework.
Equally important is the fact that the weak interaction
is favored in protic, polar media over aprotic nonpolar
solvents. In logic extension of such findings made in
the 1950s and 1960s, rather recent reports confirm this
effect to be even more pronounced in water.30,34,35 This
may help in designing molecules that adopt in free aque-
ous solution the (cyclic) conformation meant to be com-
plementary to the active site. What has emerged from
these past observations is thus a novel functional group
with rather unique stereoelectronic properties that may
be displayed as the core feature of a peptidomimetic,
provided that its backbone is sufficiently preorganized.
GS R3N+–CHOH (X-ray)
C–N
C–O
1.58
1.37
30,31
30
GS, ground state.
Importantly, the design of such a first example of inhib-
itor candidate in this paper was to be conducted without
knowing the actual structure of the inhibitor (aside from
the configuration of the two asymmetric carbons) nor
having carried out any modeling attempts outside or
within the active site.36
2.1. Design and synthetic strategy
Figure 3. Examples of compounds displaying the tertiary amine-
carbonyl interaction.
The need for incorporation of a tertiary amine function
into the final target has led us to choose as an initial
point of departure a dipeptide mimic (Fig. 4) of the sub-
strate sequence preferentially cleaved by HIV-1 pepti-
dase (HIV1-PR), that is, Phe-Pro. Here, one only
needs to introduce a molecular scaffold (bridge) linking
the a-carbon of Phe with the nitrogen of Pro in order to
The d+N ! C@Odꢀ interaction has been described as a
through-space homoconjugation of the n and p molecu-
lar orbitals of the nitrogen and C@O double bond,
respectively, where electron density is shifted from the
n orbital to the p* orbital, giving rise to an enhanced di-
pole moment and a hypsochromic shift of the carbonyl
UV absorption (see below).24
Its attractive features in the context of this project are: (a)
an oxygen-bearing carbon with a transient configuration
between sp2 and sp3, (b) an oxygen carrying a partial neg-
ative charge, (c) a carbon–oxygen bond of intermediate
length, (d) an unusually long nitrogen–carbon bond dis-
tance, and (e) a partial positive charge on the nitrogen
(see Table 1). In effect, the nucleophilic approach of a ter-
tiary nitrogen on a carbonyl group as a point of reference
for the description of generic nucleophile approach on
Figure 4. A dipeptidic transition-state mimic (1).

3838
A. Gautier et al. / Bioorg. Med. Chem. 14 (2006) 3835–3847
obtain a molecule (1) with increased chances of showing
the desired interaction in aqueous media. The design of
1 was inspired by the observed ring-open/ring-closed
equilibrium of compound D34 in different media, a
behavior largely identical to those of compounds E
and F (Fig. 3).
83% yield by a BF3ÆOEt2-catalyzed exchange reaction
with 1,2-ethanedithiol. Methylation of 6 with trimethyl-
silyldiazomethane in benzene/methanol afforded the es-
ter 7 in 64% yield. Finally, deprotection of 7 with
HgO/BF3ÆOEt2 produced the aldehyde 1a in 47% yield.
Target compound 1a was thus prepared via an efficient
six-step synthesis with 8.7% overall yield (Scheme 1).
Compound 1 has been chosen as a prototype on which
to study the conformational equilibrium between the
open and the closed form in different media, its stereo-
chemical preferences, and its affinity to the model en-
zyme, namely HIV-1 PR. Three separate derivatives of
1 were targeted, 1a–c (Fig. 5). They differ in their periph-
ery around 1, either by carrying, on the C terminus, a
methyl ester, a t-butylated carboxamide terminus, or a
t-butylated valine residue, and, on the N terminus, either
an Alloc-protecting group stemming from synthetic
requirements or a Cbz-protected valine residue.
The synthesis of compounds 1b and 1c started with the
coupling of 6 with tert-butylamine and H-Val-NH-t-Bu,
respectively, using the peptide-coupling cocktail DCC/
HOBt; compounds 8 and 10 were thus obtained in
73% and 67%, respectively (Scheme 2). The Alloc-pro-
tecting group was then selectively removed via
Pd(Ph3)4-catalyzed allyl transfer in the presence of dime-
done.40 Peptide coupling under classic coupling condi-
tions with the thus liberated amine function adjacent
to a quaternary center failed.41 We successfully coupled
the sterically hindered free amine with N-Cbz-L-valine
using the reagent cocktail CIP/HOAt to afford 9 and
11 in 65% and 64% yield, respectively.42 The synthesis
was finished with their deprotection by use of HgO/
BF3ÆOEt2 and aldehydes 1b and 1c were obtained in
61% and 47% yield, respectively.
2.2. Synthesis
Compound 6 is the common synthetic intermediate to
the three targets. The quaternary carbon center in 3
was obtained using the method of ‘self-regeneration of
stereocenters’ developed by Seebach.37 cis-Oxazolidi-
none 2, synthesized from L-phenylalanine by known
procedures,38 was converted to the corresponding eno-
late by treatment with potassium hexamethyldisilazide
in THF at ꢀ78 °C. The diastereoselective alkylation of
the resulting chiral anion with 2-bromoethyl triflate
afforded 3 in 85% yield as a single diastereomer with
retention of configuration at the starting stereocenter.
The stereochemistry of 3 was confirmed by its NOESY
spectrum, which revealed NOE contacts between the t-
Bu group and the benzyl methylene group and between
the methine and the methylene group of the introduced
moiety. The bromide 3 was then reacted with commer-
cially available H-Pro-O-t-Bu in the presence of Et3N
to give 4 in 82% yield.
For the inhibition studies we were in need for a control
inhibitor containing a transition-state isostere of proven
potency while mimicking the Phe-Pro dipeptide targeted
in this study. We therefore synthesized hydroxyethyl-
amine 16 (Scheme 3). The synthetic intermediate 14
was synthesized by regioselective nucleophilic attack of
aminoamide 1343 on chiral a-amino epoxide 1244 in
74% yield. Removal of the Cbz-protecting group from
14 by Pd/C-catalyzed hydrogenolysis afforded 15 in
quantitative yield. Hydroxyethylamine 16 was then ob-
tained in 65% yield by coupling of the amine 15 with
N-Cbz-L-valine using CIP/HOAt.
Careful reduction of 4 with lithium aluminum hydride at
ꢀ78 °C afforded oxazolidinol 5 in 50% yield. To our
knowledge, this reaction constitutes the first example
of the reductive transformation of a sterically hindered
2,5-disubstituted oxazolidinone into an oxazolidinol,
achieved by the use of LAH, without exhaustive reduc-
tion into a ring-opened diol as has been observed for
2-substituted oxazolidinols.39 In parallel to what is fre-
quently seen in carbohydrate chemistry, the oxazolidinol
5 was then directly converted into the dithioketal 6 in
O
O
H
Ph
2
H
COR
Ph
1
N
1
R
N
H
N
R
NH
2
R OC
Scheme 1. Synthesis of 1a. Reagents and conditions: (i) (1) KHMDS
(1 equiv), THF, ꢀ78 °C, 30 min; (2) Br(CH2)2OTf, ꢀ78 °C to rt,
overnight, 85%; (ii) H-Pro-O-t-Bu, Et3N Et3N, DMF, 60 °C, 3 days,
82%; (iii) LAH (2 equiv), THF, ꢀ78 °C, 48 h, 50%; (iv) HS(CH2)2SH
(4 equiv), BF3ÆOEt2 (3 equiv), CH2Cl2, 24 h, 83%; (v) TMSCHN2,
C6H6/MeOH 7:2, 1.5 h, 64%; (vi) HgO/BF3ÆOEt2, THF/H2O 85:15,
2 h, 47%.
1i(O)
1i(C)
i = a, b, c
1a R1= Alloc, R2= OMe; 1b R1= CbzVal, R2= NHtBu;
1c R1= CbzVal, R2= ValNHtBu
Figure 5. The folding equilibrium of the target peptidomimetics 1a–c.

A. Gautier et al. / Bioorg. Med. Chem. 14 (2006) 3835–3847
3839
Scheme 2. Synthesis of compounds 1b and 1c. Reagents and condi-
tions: (i) RNH2, DCC, HOBt, THF, overnight; (ii) Pd(Ph3)4,
dimedone, THF, overnight; (iii) CbzValOH, CIP, HOAt, CH2Cl2, 2
days; (iv) HgO/BF3ÆOEt2, THF/H2O 85:15, 2 h.
O
O
Ph
Ph
NHtBu
NHtBu
i
CbzHN
N
Figure 6. 1H NMR spectra (500 MHz) of 1a in CDCl3 (j signal of the
open-form aldehydic proton) and MeOH-d4 (j signals of the open-
form aldehydic proton and its corresponding methyl ester group; d
signals of the folded-form methine proton and its corresponding
methyl ester group).
CbzHN
HN
O
OH
12
13
14
ii
O
O
Ph
CbzValHN
Ph
well. Thus, the two sets of signals may be attributed to
a slow exchange compared to the NMR time scale of
two forms of 1a. The correspondence of the signal at
3.87 ppm and that at 97.8 ppm in the 13C NMR spec-
trum was proven by a two-dimensional HMQC experi-
NHtBu
NHtBu
iii
N
H N
2
N
OH
OH
16
15
1
ment (1H–13C). A correlation signal due to JH,C
Scheme 3. Synthesis of 16. Reagents and conditions: (i) MeOH,
reflux overnight, 74%; (ii) H2, Pd/C, ethanol, 2 h, quantitative;
(iii) CbzValOH, CIP, HOAt, CH2Cl2, 0 °C, 2 h, 65%.
heteronuclear coupling was found for the postulated
methine group in 1a. Integration of the signals at dH
9.63 ppm and at dH 3.87 ppm allows us to estimate the
conformational equilibrium constant between the closed
and the open form: at 20 °C in MeOH-d4 a 70/30 ratio is
found, corresponding to a Kobs = 2.33.45
2.3. Conformational analysis in solution
Before testing 1a–c for their ability to inhibit peptidolyt-
ic activity, it was of fundamental importance to deter-
mine their actual conformational and configurational
preferences.
In order to potentially obtain thermodynamic data on the
open–close equilibrium of 1a, we conducted a tempera-
1
ture-dependent H NMR experiment (DNMR) at high
When 13C NMR spectra of 1a were recorded in CDCl3
and MeOH-d4, respectively, it was found that the meth-
anol spectrum displayed a splitting for each signal essen-
tially giving rise to two independent sets. The only
significant difference was found for the signal corre-
sponding to the aldehyde carbon that had been shifted
upfield for more than 100 ppm, residing now at
97.8 ppm, right in the middle of the region found for
common organic compounds between the olefinic/aro-
matic and the aliphatic region. It may be deduced that
the carbon corresponding to this signal adopts a hybrid-
ization state between those of sp2 and sp3 as is desired as
part of the fundamental aims of this project. This behav-
temperatures (300–350 K, at 200 MHz), only to realize
that even at elevated temperatures the establishment of
the equilibrium remains slow on the NMR time scale.45
This may not be too surprising in view of the necessary
change of numerous dihedral angles and the movement
of large substituents compared to the process of simple
rotamer exchange where coalescence points are frequent-
ly determined. By integration of the proton signals of the
aldehyde and the methine group, respectively, it was
found that the proportion of the cyclic form decreases
with rising temperature. Over a range of 50 °C the propor-
tion of the cyclic form diminishes by 38%.47 This behavior
is in agreement with those of other reports.35
1
ior found its parallel in the H NMR spectra (Fig. 6)
In view of the above results, a DNMR experiment at
low temperatures (213–293 K, 500 MHz) was also car-
ried out so as to potentially freeze out the cyclic form.45
As was expected from the high-temperature experiment,
the proportion of the cyclic form increased with falling
temperature to attain a value of 90% at 213 K. The slow
appearance of an extra singlet at 3.73 ppm, tentatively
where only one set of signals was observed in chloro-
form, but two sets for the MeOH-d4 solution. The set
displaying higher intensity did not contain a signal in
the aldehydic region (9.35 ppm), but instead a singlet
was found at 3.87 ppm. A weak change in shift was ob-
served for the methyl singlet corresponding to the meth-
yl ester moiety, as is likely the case for all multiplets as

3840
A. Gautier et al. / Bioorg. Med. Chem. 14 (2006) 3835–3847
assigned to the methyl ester group of the cyclic form be-
cause of its integration value, may be explained with the
emergence of two distinct conformers of the spiro-type
structure of 1a.
evidenced by the existence of only an aldehydic signal.
By contrast, the data obtained from the experiment in
methanol are slightly more complicated to interpret than
for 1a. Aside from the typical aldehydic signal, 1b shows
several new signals in the region where the methine pro-
ton of 1a was observed. Compound 1c in methanol also
displays several signals in that same region around
4 ppm being superposed with those stemming from the
a protons of the valine residues. These findings suggest
that compounds 1b and 1c exist at the same time in an
extended conformation and several ring-closed forms
in slow equilibrium on the NMR time scale. As was
pointed out above, the latter may be explained by the
adoption of different configurations of the
d+N ! C@Odꢀ functional group or by different confor-
mations of the five-membered rings of the spiro form or
again by a mixture of both phenomena. An estimation
of the proportion of the extended form can nevertheless
be furnished on the basis of the integration of the corre-
sponding aldehydic signal: 1b and 1c adopt to 45% and
55% the open form, respectively. A higher proportion of
the cyclic form may be observed in water and can no
doubt be obtained by introducing additional elements
of preorganization.
A conformational analysis of 1a alone does not suffice
when taking into account the possibility of observing
four distinct configurational isomers arising from the
formation of two pseudo-tetrahedral asymmetric cen-
ters. In our case, the NMR spectra appear to indicate
the existence of but one diastereomer of the cyclic form.
On this assumption, we conducted a NOESY experi-
ment at 500 MHz. In spite of the difficulty of interpret-
ing the 2D spectrum due to the presence of two
molecular species, three NOE contacts were identified.
One of them is attributed to an interaction between the
methine group and one of the benzyl methylene protons,
another to the one between the methine group and aro-
matic protons, and yet another to the one between the
methyl ester protons and the methylene of the Alloc-
protective group. While one has to be cautious in attrib-
uting the latter due to the closeness of the methyl signals
of the closed and open forms, it is assumed that the open
form will likely exist in an extended conformation and
thus not give an NOE contact. On the grounds of the
NOESY data, we tentatively assign the following config-
uration (figure) to the cyclic form of 1a in MeOH at
room temperature.
UV spectroscopy presents another independent means
of estimating the open-chain/cyclic ratio of a given can-
didate compound.46 UV spectra of 1a–c were thus deter-
mined and found to be similar in appearance (Fig. 7): in
chloroform, an intense band at 245 nm is observed that
may be attributed to the p ! p* absorption; in metha-
nol, a decrease of absorption intensity of this band is
accompanied by a concomitant emergence of a new,
very intense band at around 220 nm that stems from
the newly formed d+N ! C@Odꢀ moiety in accord with
results from previous reports.46
NOE
NOE
O
H
CO CH
2
3
N
HN
O
NOE
O
Aqueous solutions of 1a–c buffered at pH 4.7 (10%
DMSO, acetate buffer), while not being sufficiently con-
centrated for NMR spectroscopy, were subjected to the
same UV analysis. The results47 support the existence of
an equilibrium of two protonated forms, namely the
ring-open (1i(o)-NH+) and the cyclic conformation
The conformational equilibria of compounds 1b and 1c
were also investigated by NMR experiments. As expect-
ed from the above observations with 1a, in chloroform
both compounds exist only in the ring-open form as
Figure 7. UV spectra in chloroform (continuous line) and in methanol (dotted line) of (a) 1a, (b) 1b, and (c) 1c.

A. Gautier et al. / Bioorg. Med. Chem. 14 (2006) 3835–3847
3841
(1i(c)-OH+) (Fig. 5), giving rise to a regular ammonium
ion and a quaternary hemiaminal. It is interesting to
note that the decrease in the folded form in passing from
1a to 1c is matched by a concomitant decrease in solubil-
ity by these compounds in the applied buffer system.
The exact inhibition constants were derived from the
same set of experiments by using several substrate con-
centrations (Table 2).52 The choice of our methodology
has been validated by the results obtained for pepsta-
tine. In fact, its IC50 and Ki literature values determined
for HIV-1 PR inhibition vary between 0.4 and 250 nM
according to experimental conditions (Ki = 17 nM when
using a FRET substrate similar to ours).49 The present
inhibition constants (Table 2) illustrate the additional
importance of interactions between the periphery on
the central moiety and the active site. By progressively
attaching amino acid residues (and/or groups that are
known to also satisfy to a certain degree the recognition
subsites, such as Cbz and t-Bu) to the N-terminal and
C-terminal sides of 1a, a gradual increase in affinity of
an approximate total of two orders of magnitude is
observed.
2.4. Inhibition studies
The inhibitory potency of candidate compounds 1a–c
was tested with HIV-1 peptidase (HIV-1 PR). A
well-established continuous assay based on the FRET
technology (fluorescence resonance energy transfer)
was used48 in order to benefit from minimal substrate,
enzyme and inhibitor consumption while rigorously
determining kinetic data. We employed a commercial-
ly available dodecapeptide substrate specific for HIV-1
PR, ArgGlu(EDANS)SerGlnAsnTyr+ProIleValGln-
Lys(DABCYL)Arg, that is a derivative of a previously
introduced version,49 and that displays excellent
spectroscopic properties for analysis by a microplate
spectrofluorimeter.
3. Conclusion
It has been pointed out by use of numerous examples to
what great extent the optimization of electrostatic inter-
actions can be productive for achieving high-affinity li-
gands.51 The d+N ! C@Odꢀ interaction as part of a
transition-state analogue is proposed here for this par-
ticular reason. An efficient synthesis of a first prototype
has been developed, and its conformational analysis in
polar media demonstrated that the moiety resulting
from this interaction can be deliberately incorporated
into a peptidomimetic, or, for that matter, any molecu-
lar framework allowing for a minimal degree of preorga-
nization. Three tendencies for the folded/unfolded
equilibrium of compounds 1a–c are observed: increasing
solubility and polarity of the medium and decreasing
temperature results in a higher extent of folding.
First, the kinetic parameters for the FRET substrate
were determined in order to evaluate our assay condi-
tions in regard to literature protocols. A mean value
of KM = 27 1 lM and Vmax = 267 10 nM/min was
obtained via different methods. These numbers are
comparable to those published for the original FRET
substrate (KM = 103 lM; Vmax = 164 nM/min at
30 °C).49 The potency of 1a–c was first estimated by
determining their IC50 values.50 Initial rates were mea-
sured for several inhibitor concentrations and IC50
values deduced by a nonlinear fit to m/m0 = 1/(1 + [I]/
IC50) (Table 2). The final reaction mixtures contained
10% DMSO as cosolvent in order to assure complete
solubilization of 1b and 1c (with [1b]max = 1 mM;
[1c]max = 200 lM). As controls we used compound 16
and pepstatine, the latter being the benchmark for
aspartic peptidase inhibitors. Table 2 illustrates to
what extent 1b and 1c can compete with the hydroxy-
ethylene inhibitors.
The conformational constraints imposed by this interac-
tion render a considerable part of the inhibitor rigid
which may have a favorable impact not only on bio-
availability of future inhibitors displaying this moiety,53
but also on affinity, provided a structure truly comple-
mentary to the active site is adopted. It remains to be
seen to what extent candidate molecules displaying the
d+N ! C@Odꢀ interaction are able to reproduce dihe-
dral angles Ca–C–N–Ca0 of the scissile carboxamide
bond found on the reaction coordinate.54,12
Candidate 1a, being the smallest of the candidate inhib-
itors, showed an IC50 value of greater than 1 mM, and
was not analyzed further. Candidate 1b displaying the
same number and the same nature of side chains as
the conformationally flexible hydroxyethylamine 16
gave IC50 = 545 lM, thus being roughly 1000 times less
potent than the latter. Pepstatine, being of molecular
weight comparable to that of 16, surpasses the potency
of 16 by one order of magnitude.
The aim of observing an d+N ! C@Odꢀ interaction as
part of a peptide backbone can only be reached when
a bridging moiety is incorporated, here an ethylene
Table 2. IC50 values obtained by data fitted to m/m0 = 1/(1 + [I]/IC50), Ki values obtained by use of three different methods, and molecular weights for
each compound
Compound
Ki (lM)
Hanes51
IC50 (lM)
MW (g/mol)
Lineweaver–Burk51
Dixon51
1a
1b
—
—
—
>1000
545
204
388
578
677
566
686
574
99
618
96
577
96
1c
16
0.463
0.036
0.497
0.036
0.758
0.036
0.469
0.036
Pepstatine

3842
A. Gautier et al. / Bioorg. Med. Chem. 14 (2006) 3835–3847
bridge, that links both fragments displaying the tertiary
amine and the aldehyde, respectively. Through the dis-
covery of high-affinity inhibitors diverging significantly
from the steric consumption of natural peptidic sub-
strates, aspartic peptidases have been shown to tolerate
extra steric bulk remarkably well, in part due to the flex-
ibility of the domains (‘flaps’) covering the active site.
Bioavailability also critically depends on the molecular
weight of a given inhibitor as has been observed in the
quest for a bioavailable peptide-derived renin inhibitor.5
These points illustrate the need for finding new transi-
tion-state isosteres with superior affinity to their target
site, allowing for reduction of overall inhibitor size.
Glu(EDANS)SerGlnAsnTyrProIleVal-GlnLys-(DAB-
CYL)Arg) was from Sigma–Aldrich.
4.2.2. Methods. Enzymatic activity and inhibition were
determined by a fluorogenic assay using a microplate
spectrofluorimeter (Spectramax Gemini XS, Molecular
Devices) and black 384-well plates (NUNCLONE, Nunc
Inc.). Hydrolysis of the fluorogenic substrate was investi-
gated at pH 4.7, 37 °C, in 0.1 M acetate, 1 M NaCl, 1 mM
EDTA, 1 mM dithiothreitol, and 1 mg/mL bovine serum
albumin. The entire time course was recorded by monitor-
ing the fluorescence at 490 nm with an excitation wave-
length set to 340 nm. Each initial rate was the average
value of five independently performed reactions.
The central transition state-analogue moiety 1 alone
does not show any appreciable inhibitory activity to-
ward HIV-1 PR. When increasing the resemblance to
a natural substrate sequence by attachment of extra ami-
no-acid residues, a successive increase in affinity is ob-
served, reaching a Ki of 97 lM for the best candidate.
This value is still three to four orders of magnitude high-
er than that found in classic hydroxyethylene-based
inhibitors. Structural incongruency with the active site
may have been caused by a counterproductive confor-
mation of the newly formed heterocycle and/or the inca-
pacity of 1a to satisfy the recognition subsites due to a
bad spatial separation of the benzyl group and the pyr-
rolidine moiety. It is also known that subtle differences
in configuration of just one stereocenter can have a huge
impact on affinity, in an extreme case approaching a Ki
difference of seven orders of magnitude.1d,55 Computa-
tional exploration outside and within the active site will
be necessary of molecular matrices promoting
d+N ! C@Odꢀ interactions for their tendencies to adopt
a particular conformation and configuration of the new-
ly formed stereogenic centers.36
KM determination was performed using a range of fluor-
ogenic substrate concentrations from 1 to 10 lM and an
HIV-1 PR concentration of 4 nM. Final solutions con-
tained 5% DMSO.
For determination of IC50 and Ki values, inhibitor was
preincubated with HIV-1 PR for 30 min. IC50 determi-
nations were performed using 10% DMSO solutions
containing a final 5 nM HIV-1 PR and 4 lM fluorogenic
substrate. Initial rates were determined for at least five
inhibitor concentrations. Ki determinations were per-
formed using 10% DMSO solutions containing a final
5 nM HIV-1 PR. Initial rates were determined for at
least five inhibitor concentrations and four fluorogenic
substrate concentrations (2, 3, 4, and 5 lM).
4.3. Bromide 3
To a stirred solution of 2 (11.61 g, 36.5 mmol) in dry
THF (500 mL) at ꢀ78 °C was added dropwise a 0.5 M
solution of KHMDS in toluene (77 mL, 38.4 mmol) un-
der argon atmosphere. The solution was stirred for 1 h
at ꢀ78 °C and then 2-bromoethyltriflate (10.30 g,
40.15 mmol) was added dropwise. The mixture was
maintained at ꢀ78 °C for 3 h and then allowed to warm
to room temperature overnight. The reaction was
quenched by adding saturated NH4Cl solution and then
extracted with ethyl acetate. The combined organic lay-
ers were washed with brine, dried over sodium sulfate
and the solvents were removed by vacuo. The com-
pound 3 was obtained as a yellow oil after flash column
4. Experimental
4.1. General
All reactions were carried out in anhydrous solvents un-
der argon atmosphere in dried glassware. CH2Cl2 and
DMF were distilled under argon on CaH2, THF on
Na/benzophenone. Compounds 2, 12, and 13 were pre-
pared according to literature protocols. 1H and 13C
NMR spectra were recorded on a Bruker DPX 200 spec-
trometer (200 MHz) and a Varian Unity 500 spectrom-
eter (500 MHz). Coupling patterns in the 1H NMR
spectra are designated as s, singlet; d, doublet; dd, dou-
ble doublet; t, triplet; m, multiplet. Mass spectra were
chromatography, using cyclohexane/ethyl acetate 95:5
21
as eluant (13.15 g, 85%, single diastereomer). ½aꢁ +42
D
(c 0.75 in CHCl3); 1H NMR (CDCl3, 200 MHz): d
0.61 (s, 9H), 2.28–2.43 (m, 1H), 2.85–3.37 (m, 5H),
4.65 (d, J = 6 Hz, 2H), 5.27–5.43 (m, 3H), 5.83–6.00
(m, 1H), 7.22 (br s, 5H); 13C NMR (CDCl3, 50 MHz):
d 24.36, 24.94, 37.71, 39.81 (br), 42.22, 66.95, 67.79,
95.22, 120.27, 127.38, 128.30, 130.96, 131.23, 134.83,
154.57, 172.39; HRMS: m/z: calcd for C20H27BrNO4:
424.1124; found: 424.1126 [M++H]; elemental analysis
calcd (%) for C20H26BrNO4: C, 56.61; H, 6.18; N,
3.30; found: C, 57.06; H, 6.36; N 3.25.
´
recorded by the Centre de Spectrometrie de Masse, Uni-
versite de Lyon, France. Elemental analyses were carried
´
out by the Service Central d’Analyses, CNRS. Specific
rotations were measured on a Jasco-P1010 polarimeter.
UV spectra were recorded on a JASCO V550 UV–vis
spectrophotometer.
4.2. Enzyme assays
4.4. Tertiary amine 4
4.2.1. Materials. rec. HIV-1 PR was purchased from
Bachem and HIV protease substrate
A stirred solution of 3 (13 g, 30.6 mmol), H-Pro-O-t-Bu
(6.8 g, 39.8 mmol), and Et3N (5.6 mL, 39.8 mmol) in
1
(Arg-

A. Gautier et al. / Bioorg. Med. Chem. 14 (2006) 3835–3847
3843
DMF (40 mL) was heated at 60 °C for 3 days under ar-
gon atmosphere. The reaction mixture was cooled and
then diluted with water and extracted with ethyl acetate.
The combined organic layers were washed with water
and brine, dried with sodium sulfate and the solvents
were removed by vacuo. The compound 4 was obtained
as a yellow oil after flash column chromatography, using
cyclohexane/ethyl acetate 9:1 then 8:2 as eluant (12.91 g,
chromatography, using ethyl acetate/methanol 6:4 as
22
1
eluant (4.54 g, 83%). ½aꢁ ꢀ14 (c 0.74 in CHCl3); H
D
NMR (CDCl3, 200 MHz): d 1.83–1.90 (m, 2H), 2.11–
2.22 (m, 2H), 2.34–2.45 (m, 2H), 2.73–3.37 (m, 9H),
3.59 (t, J = 7 Hz, 1H), 3.71–3.77 (m, 1H), 4.49 (d,
J = 5.6 Hz, 2H), 4.93 (s, 1H), 5.15–5.31 (m, 2H), 5.61
(br s, 1H), 5.78–5.94 (m, 1H), 7.17–7.29 (m, 5H), 7.90
(br s, 1H); 13C NMR (CDCl3, 50 MHz): d 23.47, 29.10,
31.15, 38.92, 39.23, 42.81, 51.28, 53.99, 61.29, 61.53,
65.44, 69.09, 117.83, 127.16, 128.52, 130.59, 132.70,
135.46, 155.02, 170.82; HRMS (LSIMS): m/z: calcd for
C22H31N2O4S2: 451.1725; found: 451.1724 [M++H].
21
1
82%). ½aꢁ ꢀ16 (c 2.7 in CHCl3); H NMR (CDCl3,
D
500 MHz): d 0.54 (s, 9H), 1.37 (s, 9H,), 1.66 (m, 1H),
1.76 (m, 2H), 1.92 (m, 2H), 2.26 (m, 2H), 2.45–2.58
(m, 2H), 2.98 (m, 2H), 3.21 (m, 2H), 4.49 (dd,
J = 12 Hz, J = 6 Hz, 1H), 4.60 (dd, J = 12 Hz,
J = 6 Hz, 1H), 5.21–5.34 (m, 2H), 5.34 (s, 1H), 5.84
(m, 1H), 7.12–7.20 (m, 5H); 13C NMR (CDCl3,
50 MHz): d 22.43, 24.70, 27.61, 28.42, 37.22, 42.45,
47.91, 52.81, 65.22, 66.29, 66.48, 80.29, 94.70, 119.08,
126.70, 127.78, 130.64, 131.21, 135.21, 154.75 (br),
172.51, 172.88; HRMS: m/z: calcd for C29H43N2O6:
515.3121; found: 515.3119 [M++H]; elemental analysis
calcd (%) for C29H42N2O6: C, 67.68; H, 8.23; N, 5.44;
found: C, 67.23; H, 8.55; N 5.38.
4.7. Methyl ester 7
To a stirred solution of 6 (1 g, 2.22 mmol) in a benzene/
methanol 3.5:1 mixture (20.5 mL) was added a 2 N solu-
tion of trimethylsilyldiazomethane in hexane (1.44 mL,
2.89 mmol). The reaction mixture was stirred for 1.5 h
at room temperature and then the solvent was removed
by vacuo. The compound 7 was obtained as a colorless
oil after flash column chromatography, using pentane/
22
ethyl acetate 8:2 as eluant (656 mg, 64%). ½aꢁ ꢀ75.4
D
4.5. Lactol 5
(c 1.08 in CHCl3); 1H NMR (CDCl3, 200 MHz):
d = 1.70–2.24 (m, 9H), 2.84–3.28 (m, 7H), 3.23 (s, 3H),
3.23–3.66 (m, 1H), 4.56 (d, J = 5 Hz, 2H), 5.14–5.37
(m, 2H), 5.80 (s, 1H), 5.84–5.96 (m, 1H), 7.12–7.25 (m,
5H), 8.21 (s, 1H); 13C NMR (CDCl3, 50 MHz): d
22.99, 29.08, 29.29, 38.50, 39.68, 42.90, 50.17, 51.69,
52.25, 62.34, 62.41, 64.63, 65.72, 116.43, 126.52,
128.04, 130.50, 133.52, 137.32, 155.36, 173.63; HRMS
(LSIMS): m/z: calcd for C23H33N2O4S2: 465.1882;
found: 465.1886 [M++H]; elemental analysis calcd (%)
for C23H32N2O4S2: C, 59.45; H, 6.94; N, 6.03; found:
C, 59.36; H, 6.98; N 5.87.
To a stirred solution of 4 (12.71 g, 24.7 mmol) in dry
THF (120 mL) at ꢀ78 °C was added carefully per por-
tion LAH (1.87 g, 49.4 mmol) under argon atmosphere.
The solution was stirred for 48 h at ꢀ78 °C. The reac-
tion was quenched very carefully at ꢀ78 °C by adding
water (1.9 mL), 15% NaOH (1.9 mL), and water
(5.7 mL). The mixture was allowed to warm to room
temperature and stirred for 15 min. After filtration of
white aluminum salts, the organic layer was dried over
magnesium sulfate and the solvent was removed by vac-
uo. The compound 5 was obtained as a colorless oil after
flash column chromatography, using pentane/ether 7:3
then 5:5 as eluant (6.42 g, 50%). 1H NMR (CDCl3,
200 MHz): d 0.57 (s, 9H), 1.43 (s, 9H), 1.75–2.05 (m,
5H), 2.30–2.69 (m, 4H), 2.94–3.02 (m, 2H), 3.27–3.39
(m, 2H), 4.61–4.70 (m, 2H), 5.06 (s, 1H), 5.19–5.41 (m,
3H), 5.84–6.04 (m, 1H), 7.12–7.24 (m, 5H), 8.90 (s,
1H); 13C NMR (CDCl3, 50 MHz): d 22.79, 26.05,
28.06, 29.41, 38.61, 42.51, 47.27, 52.76, 64.54, 66.10,
69.08, 81.23, 93.71, 101.01, 118.72, 126.66, 128.11,
131.09, 132.37, 136.77, 155.15 (br), 171.95; HRMS
(LSIMS): m/z: calcd for C29H45N2O6: 517.3278; found:
517.3273 [M++H]; elemental analysis calcd (%) for
C29H44N2O6: C, 67.41; H, 8.58; N, 5.42; found: C,
67.18; H, 8.67; N 5.30.
4.8. Target compound 1a
To a stirred solution of HgO (570 mg) in a THF/water
85:15 mixture (10 mL) was added BF3ÆEt2O (3.56 mL)
under argon atmosphere. After discoloration of the or-
ange solution, a solution of 7 (530 mg, 1.14 mmol) in a
THF/water 85:15 mixture (25 mL) was added dropwise.
After the mixture was stirred for 2 h, saturated sodium
bicarbonate solution was added. The aqueous layer
was extracted with CH2Cl2. The combined organic lay-
ers were dried over sodium sulfate and the solvent was
removed by vacuo. The compound 1a was obtained
after flash column chromatography as a colorless oil,
using cyclohexane/ethyl acetate 1:1 as eluant (207 mg,
21
1
47%). ½aꢁ ꢀ5.8 (c 0.63 in CHCl3); H NMR (CDCl3,
D
4.6. Dithiolane 6
500 MHz): d 1.75–1.91 (m, 3H), 2.00–2.17 (m, 2H),
2.19–2.22 (m, 1H), 2.34–2.39 (m, 2H), 2.66–2.71 (m,
1H), 2.99–3.03 (m, 1H), 3.14 (d, J = 14 Hz, 1H), 3.21–
3.24 (m, 1H), 3.41 (d, J = 14 Hz, 1H), 3.69 (s, 3H),
4.56–4.65 (m, 2H), 5.24 (d, J = 10.5 Hz, 1H), 5.33 (d,
J = 17.5 Hz, 1H), 5.91–5.98 (m, 1H), 6.00 (s, 1H),
7.06–7.07 (m, 2H), 7.22–7.28 (m, 3H), 9.35 (s, 1H); 13C
NMR (CDCl3, 50 MHz): d 23.14, 29.14, 33.98, 39.56,
48.87, 51.73, 52.34, 65.14, 65.37, 65.76, 117.37, 126.71,
128.11, 130.37, 132.90, 135.59, 155.06, 174.16, 197.76;
HRMS (LSIMS): m/z: calcd for C21H29N2O5:
389.2076; found: 389.2079 [M++H]; elemental analysis
To a solution of 5 (6.25 g, 12.1 mmol) in dry CH2Cl2
(100 mL) at 0 °C were added 1,2-ethanedithiol
(4.07 mL, 48.4 mmol) and BF3ÆOEt2 (4.6 mL,
36.3 mmol) under argon atmosphere. After 3 h at 0 °C,
the mixture was warmed to room temperature and stir-
red for a night. After adding saturated NH4Cl, the mix-
ture was extracted with CH2Cl2. The combined organic
layers were washed with brine, dried over sodium sulfate
and the solvent was removed by vacuo. The compound 6
was obtained as a colorless oil after flash column

3844
A. Gautier et al. / Bioorg. Med. Chem. 14 (2006) 3835–3847
calcd (%) for C21H28N2O5: C, 64.93; H, 7.27; N, 7.21;
found: C, 65.07; H, 7.33; N 7.08.
3.01–3.19 (m, 6H), 3.38 (d, J = 13.7 Hz, 1H), 3.89 (dd,
J = 9 Hz, J = 6.3 Hz, 1H), 5.02–5.13 (m, 2H), 5.29 (br
s, 1H), 5.50 (m, 1H), 6.72 (br s, 1H), 6.83 (br s, 1H),
7.15–7.36 (m, 10H); 13C NMR (CDCl3, 50 MHz): d
17.74, 19.46, 23.73, 28.75, 30.38, 30.57, 34.99, 38.87,
38.96, 42.42, 50.27, 51.09, 53.68, 61.55, 61.72, 62.15,
66.93, 68.27, 126. 98, 128.03, 128.35, 128.45, 130.64,
135.98, 136.35, 156.28, 170.83, 173.50; HRMS (LSIMS):
m/z: calcd for C35H51N4O4S2: 655.3352; found: 655.3351
[M++H].
4.9. Allyl carbamate 8
To a stirred solution of 6 (425 mg, 0.94 mmol) in dry
THF (5 mL) at 0 °C were added HOBt (254 mg,
1.88 mmol) then DCC (213 mg, 1.03 mmol) under argon
atmosphere. After the mixture was stirred for 10 min,
was added t-BuNH2 (300 lL, 2.82 mmol). The mixture
was maintained for 1 h at 0 °C and then allowed to
warm to room temperature and stirred for a night. After
filtration of DCU, the solution was concentrated in vac-
uo and ethyl acetate was added. The organic layer was
washed with 3 N citric acid solution, saturated sodium
bicarbonate solution and brine, dried over sulfate sodi-
um and then the solvent was removed by vacuo. The
compound 8 was obtained as a colorless oil after flash
4.11. Target compound 1b
To a stirred solution of HgO (110 mg) in a THF/water
85:15 mixture (6 mL) was added BF3ÆEt2O (690 lL) un-
der argon atmosphere. After discoloration of the orange
solution, a solution of 9 (144 mg, 0.22 mmol) in a THF/
water 85:15 mixture (6 mL) was added dropwise. After
the mixture was stirred for 2 h, saturated sodium bicar-
bonate solution was added. The aqueous layer was
extracted with CH2Cl2. The combined organic layers
were dried over sodium sulfate and the solvent was re-
moved by vacuo. The compound 1b was obtained after
flash column chromatography as a colorless oil, using
column chromatography, using pentane/ethyl acetate
21
D
6:4 as eluant (347 mg, 73%). ½aꢁ ꢀ47.1 (c 0.30 in
1
CHCl3); H NMR (CDCl3, 200 MHz): d 1.29 (s, 9H),
1.61–1.68 (m, 3H), 1.99–2.37 (m, 5H), 2.67–2.80 (m,
2H), 3.06–3.29 (m, 7H), 4.47 (d, J = 4.7 Hz, 2H),
5.14–5.30 (m, 3H), 5.70 (s, 1H), 5.80–5.94 (m, 1H),
6.96 (s, 1H), 7.22 (m, 5H); 13C NMR (CDCl3,
50 MHz): d 23.65, 28.62, 30.27, 33.98, 38.69, 39.13,
42.18, 50.12, 51.06, 53.50, 61.42, 62.06, 65.08, 68.89,
117.48, 126.92, 128.32, 130.55, 132.80, 136.00, 154.61,
173.62; HRMS (LSIMS): m/z: calcd for C26H40N3O3S2:
506.2511; found: 506.2510 [M++H].
cyclohexane/ethyl acetate 1:1 as eluant (78 mg, 61%).
21
D
½aꢁ ꢀ1.2 (c 0.25 in CHCl3); 1H NMR (CDCl3,
200 MHz): d = 0.84 (d, J = 6.8 Hz, 3H), 0.93 (d,
J = 6.8 Hz, 3H), 1.33 (s, 9H), 1.58–1.73 (m, 5H), 1.94–
2.42 (m, 6H), 2.63–2.80 (m, 2H), 3.50 (d, J = 13.9 Hz,
1H), 3.92–3.98 (m, 1H), 5.09 (s, 2H), 5.31 (d, J = 8.9,
1H), 6.52 (s, 1H), 6.62 (s, 1H), 6.93–6.96 (m, 2H),
7.13–7.16 (m, 3H), 7.34 (m, 5H), 9.37 (s, 1H); 13C
NMR (CDCl3, 50 MHz): d = 17.43, 19.53, 23.66,
28.79, 30.62, 30.98, 32.18, 39.31, 50.59, 50.77, 54.07,
60.55, 66.83, 67.03, 69.49, 127.18, 128.01, 128.16,
128.44, 128.53, 129.93, 134.38, 136.29, 156.22, 171.15,
173.75, 198.94; HRMS (LSIMS): m/z: calcd for
C33H47N4O5: 579.3546; found: 579.3548 [M++H].
4.10. Benzyl carbamate 9
To a degassed mixture of 8 (230 mg, 0.45 mmol) and
dimedone (318 mg, 2.27 mmol) in dry THF (3 mL) was
added Pd(Ph3)4 (26 mg, 0.02 mmol) under argon atmo-
sphere. The reaction mixture was stirred for 24 h at
room temperature. After the solution was concentrated
in vacuo, ethyl acetate was added and the organic layer
was extracted with 1 N HCl. The aqueous layer was neu-
tralized by adding carefully potassium carbonate and
then extracted with CH2Cl2. The combined organic lay-
ers were dried over sodium sulfate and the solvent was
removed by vacuo. The deprotected amine was obtained
as a yellow oil used without further purification
4.12. Allyl carbamate 10
To a stirred solution of 6 (1.5 g, 3.33 mmol) in dry THF
(7.5 mL) at 0 °C were added HOBt (900 mg, 6.66 mmol)
and then DCC (756 mg, 3.66 mmol) under argon atmo-
sphere. After the mixture was stirred for 10 min was
added dropwise using a cannula a THF solution of H-
Val-NH-t-Bu prepared by adding Et3N (585 lL,
(190 mg). To
a solution of Cbz-Val-OH (90 mg,
0.36 mmol) and HOAT (49 mg, 0.36 mmol) in dry
CH2Cl2 (3 mL) at 0 °C were added Et3N (150 lL,
1.08 mmol), CIP (100 mg, 0.36 mmol) and then the
deprotected amine (150 mg, 0.36 mmol) under argon
atmosphere. The mixture was maintained for 1 h at
0 °C and then allowed to warm to room temperature
and stirred for 2 days. The reaction mixture was then
diluted with CH2Cl2. The organic layer was washed with
3 N citric acid solution, saturated sodium bicarbonate
solution and brine, dried over sulfate sodium and then
the solvent was removed by vacuo. The compound 9
was obtained as a colorless oil after flash column chro-
4.16 mmol) to
a solution of HClÆH-Val-NH-t-Bu
(624 mg, 4.00 mmol) in dry THF (7.5 mL). The mixture
was maintained for 30 min at 0 °C and then allowed to
warm to room temperature and stirred for a night. After
filtration of precipitates, the solution was concentrated
in vacuo and ethyl acetate was added. The organic layer
was washed with 3 N citric acid solution, saturated sodi-
um bicarbonate solution and brine, dried over sulfate
sodium and then the solvent was removed by vacuo.
The compound 10 was obtained as a colorless oil after
flash column chromatography, using pentane/ethyl ace-
21
matography, using pentane/ethyl acetate 1:1 as eluant
tate 1:1 as eluant (1.34 g, 67%). ½aꢁ ꢀ52.7 (c 0.76 in
D
21
(153 mg, 65% on 2 steps). ½aꢁ ꢀ51.3 (c 0.25 in CHCl3);
CHCl3); 1H NMR (CDCl3, 200 MHz): d 0.90 (t,
J = 6.2 Hz 6H), 1.28 (s, 9H), 1.59–1.86 (m, 4H),
2.01–2.16 (m, 4H), 2.26–2.31 (m, 1H), 2.46–2.52 (m,
1H), 2.75–2.77 (m, 1H), 3.00–3.24 (m, 8H), 3.94 (dd,
D
1H NMR (CDCl3, 200 MHz): d 0.89 (d, J = 6.7 Hz, 3H),
0.98 (d, J = 6.7 Hz, 3H), 1.29 (s, 9H), 1.60–1.74 (m, 3H),
1.98–2.37 (m, 6H), 2.58 (m, 1H), 2.74–2.79 (m, 1H),

A. Gautier et al. / Bioorg. Med. Chem. 14 (2006) 3835–3847
3845
J = 8.9 Hz, J = 7.2 Hz, 1H), 4.50 (d, J = 5.5 Hz, 2H),
5.13 (s, 1H), 5.13–5.32 (m, 2H), 5.70 (s, 1H), 5.75 (s,
1H), 5.80–5.99 (m, 1H), 7.12–7.13 (m, 5H), 7.73 (d,
J = 9 Hz, 1H); 13C NMR (CDCl3, 50 MHz): d 18.24,
19.40, 24.16, 28.69, 30.66, 30.80, 34.02, 38.79, 39.12,
41.91, 50.86, 51.30, 53.39, 58.88, 61.51, 62.06, 65.14,
67.46, 117.43, 126.83, 128.28, 130.76, 133.16, 136.39,
154.98, 169.96, 174.97; HRMS (LSIMS): m/z: calcd for
C31H49N4O4S2: 605.3195; found: 605.3196 [M++H]; ele-
mental analysis calcd (%) for C31H48N4O4S2: C, 61.56;
H, 8.00; N, 9.26; found: C, 61.23; H, 8.08; N, 9.04.
sodium bicarbonate solution was added. The aqueous
layer was extracted with CH2Cl2. The combined organic
layers were dried over sodium sulfate and the solvent
was removed by vacuo. The compound 1c was obtained
after flash column chromatography as a colorless oil,
using cyclohexane/ethyl acetate 1:1 as eluant (170 mg,
24
1
47%). ½aꢁ ꢀ8.6 (c 0.42 in CHCl3); H NMR (CDCl3,
D
200 MHz): d 0.83–0.96 (m, 12H), 1.31 (s, 9H), 1.61–
1.80 (m, 5H), 2.01–2.45 (m, 8H), 2.86–3.09 (m, 3H),
3.35 (d, J = 13.9 Hz, 1H), 4.05–4.17 (m, 2H), 5.06 (s,
2H), 5.51 (d, J = 9.1 Hz, 1H), 5.77 (m, 1H), 6.98–7.01
(m, 2H), 7.13–7.17 (m, 3H), 7.32 (m, 5H), 7.65 (d.
J = 9.4 Hz, 1H), 9.38 (s, 1H); 13C NMR (CDCl3,
50 MHz): d 17.44, 18.48, 19.36, 23.90, 28.68, 30.65,
30.98, 31.39, 32.04, 39.65, 51.03, 51.60, 53.65, 58.37,
59.86, 66.17, 66.70, 68.97, 127.00, 127.97, 128.26,
128.40, 130.10, 134.32, 136.45, 156.17, 171.06, 171.55,
174.45, 200.03; HRMS (LSIMS): m/z: calcd for
C38H56N5O6: 678.4231; found: 678.4230 [M++H].
4.13. Benzyl carbamate 11
To a degassed mixture of 10 (604 mg, 1 mmol) and dime-
done (701 mg, 5 mmol) in dry THF (9 mL) was added
Pd(Ph3)4 (58 mg, 0.05 mmol) under argon atmosphere.
The reaction mixture was stirred for 24 h at room temper-
ature. After the solution was concentrated in vacuo, ethyl
acetate was added and the organic layer was extracted
with 1 N HCl. The aqueous layer was neutralized by add-
ing carefully potassium carbonate and then extracted with
CH2Cl2. The combined organic layers were dried over
sodium sulfate and the solvent was removed by vacuo.
The deprotected amine was obtained as a yellow oil used
without further purification (520 mg).
4.15. Hydroxyethylamine 14
A stirred mixture of H-Pro-NH-t-Bu 13 (149 mg,
0.87 mmol) and N-Cbz-phenylalanine epoxyde 12
(260 mg, 0.87 mmol) in dry methanol (5 mL) was re-
fluxed for a night under argon atmosphere. The reaction
mixture was cooled to room temperature, and the sol-
vent was removed by vacuo. The compound 14 was ob-
To a solution of Cbz-Val-OH (251 mg, 1 mmol), and
HOAT (136 mg, 1 mmol) in dry CH2Cl2 (8 mL) at
0 °C were added Et3N (420 lL, 3 mmol), CIP (278 mg,
1 mmol) then the deprotected amine (520 mg, 1 mmol)
under argon atmosphere. The mixture was maintained
1 h at 0 °C and then allowed to warm to room temper-
ature and stirred for 3 days. The reaction mixture was
then diluted with CH2Cl2. The organic layer was washed
with 3 N citric acid solution, saturated sodium bicar-
bonate solution and brine, dried over sulfate sodium
and then the solvent was removed by vacuo. The com-
pound 11 was obtained as a colorless oil after flash col-
tained as
chromatography, using Et2O/methanol 9:1 as eluant
a
colorless oil after flash column
24
D
(302 mg, 74%). ½aꢁ ꢀ47.8 (c 0.30 in CHCl3); 1H
NMR (CDCl3, 200 MHz): d 1.29 (s, 9H), 1.61–1.90
(m, 3H), 2.00–2.20 (m, 1H), 2.38–2.51 (m, 1H), 2.65
(d, J = 6.1 Hz, 1H), 2.81–2.96 (m, 3H), 3.18 (m, 1H),
3.60–3.68 (m, 1H), 3.80–3.91 (m, 1H), 4.93 (s, 1H),
4.99 (s, 2H), 6.96 (s, 1H), 7.12–7.31 (m, 10H); 13C
NMR (CDCl3, 50 MHz): d 24.32, 28.71, 30.94, 35.57,
50.50, 55.58, 56.28, 59.81, 66.78, 68.94, 72.11, 126.60,
127.90, 128.08, 128.47, 128.57, 129.27, 136.31, 137.44,
156.55, 174.26; HRMS (LSIMS): m/z: calcd for
C27H38N3O4: 468.2862; found: 468.2858 [M++H].
umn chromatography, using pentane/ethyl acetate 2:8 as
22
D
eluant (485 mg, 64% over two steps). ½aꢁ ꢀ75.4 (c 0.57
1
in CHCl3); H NMR (CDCl3, 200 MHz): d 0.86–1.00
(m, 12H), 1.26 (s, 9H), 1.63–2.55 (m, 12H), 2.91–3.16
(m, 6H), 3.35 (d,J = 13.8 Hz, 1H), 3.81–3.92 (m, 2H),
5.05 (d,J = 12.3 Hz, 1H), 5.13 (d,J = 12.3 Hz, 1H), 5.38
(br s, 1H), 5.70 (br s, 1H), 5.87 (d, J = 8.7 Hz, 1H),
6.92 (br s, 1H), 7.18–7.36 (m, 10H), 7.46 (d, J = 9 Hz,
1H); 13C NMR (CDCl3, 50 MHz): d 17.30, 17.93,
19.31, 19.56, 23.87, 28.65, 30.38, 30.45, 30.66, 33.83,
38.88, 39.10, 42.25, 50.85, 51.33, 53.39, 59.23, 61.62,
62.59, 63.36, 66.80, 67.45, 126.87, 128.03, 128.08,
128.31, 128.48, 130.77, 136.20, 136.54, 156.39, 170.05,
171.07, 174.5; HRMS (LSIMS): m/z: calcd for
C40H60N5O5S2: 754.4036; found: 754.4039 [M++H].
4.16. Compound 15
A mixture of 14 (246 mg, 0.53 mmol) and 10% Pd/C
(28 mg, 0.026 mmol) in ethanol (3 mL) was stirred under
nitrogen atmosphere (1 atm) for 1 h. After filtration on
Celite, the solvent was removed by vacuo. The compound
15 was obtained as a white solid used without further puri-
24
D
fication (176 mg, quantitative). Mp 144 °C; ½aꢁ ꢀ53.9
(c = 0.55 in CHCl3); 1H NMR (CDCl3, 200 MHz): d
1.31 (s, 9 H), 1.62–1.85 (m, 3H), 2.06–2.25 (m, 1H),
2.54–2.96 (m, 8H), 3.07–3.12 (m, 1H), 3.24–3.28 (m,
1H), 3.59–3.65 (m, 1H), 7.14–7.31 (m, 5H); 13C NMR
(CDCl3, 50 MHz): d 24.27, 28.62, 30.88, 37.37, 50.42,
55.94, 56.15, 58.99, 68.64, 71.93, 126.56, 128.68, 129.17,
138.48, 174.42; HRMS (LSIMS): m/z: calcd for
C19H32N3O2: 334.2495; found: 334.2496 [M++H].
4.14. Target compound 1c
To a stirred solution of HgO (270 mg) in a THF/water
85:15 mixture (15 mL) was added BF3ÆEt2O (1.7 mL)
under argon atmosphere. After discoloration of the or-
ange solution, a solution of 11 (405 mg, 0.54 mmol) in
a THF/water 85:15 mixture (15 mL) was added drop-
wise. After the mixture was stirred for 2 h, saturated
4.17. Target compound 16
To a solution of Cbz-Val-OH (68 mg, 0.27 mmol), and
HOAT (37 mg, 0.27 mmol) in dry CH2Cl2 (1.5 mL) at

3846
A. Gautier et al. / Bioorg. Med. Chem. 14 (2006) 3835–3847
7. Jacobsen, N. E.; Bartlett, P. A. J. Am. Chem. Soc. 1981,
103, 654–657.
0 °C were added Et3N (115 lL, 0.81 mmol), CIP (75 mg,
0.27 mmol) then 15 (90 mg, 0.27 mmol) under argon
atmosphere. The mixture was maintained 2 h at 0 °C.
The reaction mixture was then diluted with CH2Cl2.
The organic layer was washed with 3 N citric acid solu-
tion, saturated sodium bicarbonate solution and brine,
dried over sulfate sodium and then the solvent was re-
moved by vacuo. The compound 16 was obtained as a
colorless oil after flash column chromatography, using
8. (a) Gelb, M. H.; Svaren, J. P.; Abeles, R. H. Biochemistry
1985, 24, 1813–1817; (b) Thairivongs, S.; Pals, D. T.; Kati,
W. M.; Turner, S. R.; Thomasco, L. M. J. Med. Chem.
1985, 28, 1553–1555; (c) Veerapandian, B.; Cooper, J. B.;
Sali, A.; Blundell, T. L.; Rosati, R. L.; Dominy, B. W.;
Damon, D. B.; Hoover, D. J. Protein Sci. 1992, 1, 322–
328.
9. Slee, D. H.; Laslo, K. L.; Elder, J. H.; Ollmann, I. R.;
Gustchina, A.; Kervinen, J.; Zdanov, A.; Wlodawer, A.;
Wong, C.-H. J. Am. Chem. Soc. 1995, 117, 11867–11878.
10. Bartlett, P. A.; Marlowe, C. K. Biochemistry 1983, 22,
4618–4624.
cyclohexane/ethyl acetate 3:7 as eluant (100 mg, 65%).
24
½aꢁ ꢀ59 (c 0.25 in CHCl3); 1H NMR (CDCl3,
D
200 MHz):
d 0.68 (d, J = 6.8 Hz, 3H), 0.83 (d,
J = 6.8 Hz, 3H), 1.29 (s, 9H), 1.60–1.85 (m, 4H), 1.95–
2.16 (m, 2H), 2.41–2.50 (m, 1H), 2.62–2.97 (m, 5H),
3.22 (m, 1H), 3.54–3.68 (m, 2H), 3.83 (dd, J = 5.8 Hz,
J = 8.1 Hz, 1H), 4.10–4.19 (m, 1H), 5.00 (s, 1H), 5.06
(s, 2H), 6.24 (d, J = 7.4 Hz, 1H), 6.99 (s, 1H), 7.10–
7.24 (m, 5H), 7.33 ppm (s, 5H); 13C NMR (CDCl3,
50 MHz): d 17.28, 19.26, 24.30, 28.61, 30.34, 30.96,
34.93, 50.47, 54.19, 56.18, 59.56, 60.77, 67.16, 68.92,
72.01, 126.58, 128.09, 128.29, 128.56, 129.06, 136.06,
137.50, 156.32, 171.57, 174.39; HRMS (LSIMS): m/z:
calcd for C32H47N4O5: 567.3546; found: 567.3548
[M++H].
11. Tantillo, D. J.; Houk, K. N. J. Org. Chem. 1999, 64, 3066–
3076.
12. Radkiewicz, J. L.; McAllister, M. A.; Goldstein, E.; Houk,
K. N. J. Org. Chem. 1998, 63, 1419–1428.
13. For a general amide hydrolysis mechanism see: (a) Gorb,
L.; Asensio, A.; Tunon, I.; Ruiz-Lopez, M. F. Chem. Eur.
J. 2005, 11, 1–12; (b) Strajbl, M.; Florian, J.; Warshel, A.
J. Am. Chem. Soc. 2000, 122, 5354–5366; (c) Bakowies, D.;
Kollman, P. A. J. Am. Chem. Soc. 1999, 121, 5712–5726.
14. For catalytic mechanism by HIV-1 PR see: (a) Okimoto,
N.; Tsukui, T.; Hata, M.; Hoshino, T.; Tsuda, M. J. Am.
Chem. Soc. 1999, 121, 7349–7354; (b) Piana, S.; Carloni,
P.; Parrinello, M. J. Mol. Biol. 2002, 319, 567–583.
15. Hansen, D. E.; Raines, R. T. J. Chem. Ed. 1990, 67, 483–
489.
Acknowledgments
16. But see: Schowen, R. L. Proc. Natl. Acad. Sci. U.S.A.
2003, 100, 11931–11932.
This research was in part financed by an ATIP grant
from the CNRS to J.H., and an EMERGENCE grant
from the Rhone-Alpes region to J.H.
17. Leung, D.; Abbenante, G.; Fairlie, D. P. J. Med. Chem.
2000, 43, 305–341.
18. (a) Mader, M. M.; Bartlett, P. A. Chem. Rev. 1997, 97,
1281–1301; (b) Smithrud, D. B.; Benkovic, S. J. Curr.
Opin. Biotechnol. 1997, 8, 459–466; (c) Hollfelder, F.;
Kirby, A. J.; Tawfik, D. S.; Kikuchi, K.; Hilvert, D. J.
Am. Chem. Soc. 2000, 122, 1022–1029; (d) Barbany, M.;
Gutierrez-de-Teran, H.; Sanz, F.; Villa-Freixa, J.; War-
shel, A. ChemBioChem 2003, 4, 277–285.
Supplementary data
NMR spectra, UV data, and kinetic plots (PDF).
Supplementary data associated with this article can
be found, in the online version, at doi:10.1016/
j.bmc.2006.01.031.
19. Gautier, A. PhD thesis, 2005, Ecole Normale Superieure
de Lyon (France).
20. (a) Gadamer, J. Arch. Chem. 1920, 258, 148; (b) Fermack,
O.; Robinson, R. J. Chem. Soc. 1922, 121, 427–440.
21. See also: (a) Hourcade, S.; Ferdenzi, A.; Retailleau, P.;
Mons, S.; Maranzo, C. Eur. J. Org. Chem. 2005, 1302–
1310; (b) Sisko, J.; Weinreb, S. M. J. Org. Chem. 1991, 56,
3210–3211.
22. (a) Leonard, N. J. Records Chem. Progr. 1956, 17, 243–
257; (b) Bell, M. R.; Archer, S. J. Am. Chem. Soc. 1960,
82, 151–155.
References and notes
1. (a) Pauling, L. Chem. Eng. News 1946, 24, 1375–1377; (b)
Jencks, W. P. In Current Aspects of Biochemical Energet-
ics; Kaplan, N. O., Kennedy, E. P., Eds.; Academic Press:
New York, 1966; pp 273–298; (c) Lienhard, G. E. Science
1973, 180, 149–154; (d) Wolfenden, R. Bioorg. Med.
Chem. 1999, 7, 647–652.
23. Burgi, H. B.; Dunitz, J. D.; Shefter, E. J. Am. Chem. Soc.
¨
1973, 95, 5065–5067.
2. The use of the term ‘peptidase’ over ‘protease’ is recom-
mended by the authors of the Handbook of proteolytic
enzymes: Barret, A. J.; Rawling, N. D.; Woessner, J. F.
Handbook of Proteolytic Enzymes; Academic Press: San
Diego, 1998.
24. (a) Leonard, N. J. Acc. Chem. Res. 1979, 12, 423–429; (b)
Rademacher, P. Chem. Soc. Rev. 1995, 143–150.
25. Allen, F. H.; Kennard, O.; Watson, D. G.; Brammer, L.;
Orpen, A. G.; Taylor, R. J. Chem. Soc., Perkin II 1987,
S1–S19.
3. Erickson, J. W.; Eissenstat, M. A. HIV Proteinase as a
Target for the Design of Antiviral Agents for AIDS. In
Proteinases of Infectious Agents; Dunn, B. M., Ed.;
Academic Press: San Diego, 1999; pp 1–60.
4. (a) Dunn, B. M. Chem. Rev. 2002, 102, 4431–4458; (b)
Brik, A.; Wong, C.-H. Org. Biomol. Chem. 2003, 1, 5–
14.
5. Bursavich, M. G.; Rich, D. H. J. Med. Chem. 2002, 45,
541–558.
6. (a) Rich, D. H. J. Med. Chem. 1985, 28, 263–273; (b)
Krantz, A. Bioorg. Med. Chem. 1992, 2, 1327–1334.
26. Kirby, A. J.; Komarov, I. V.; Wothers, P. D.; Feeder, N.
Angew. Chem. Int. Ed. 1998, 37, 785–786.
27. Okimoto, N.; Tsukui, T.; Hata, M.; Hoshino, T.; Tsuda,
M. J. Am. Chem. Soc. 1999, 121, 7349–7354.
28. Venturini, A.; Lopez-Ortiz, F.; Alvarez, J. M.; Gonzalez,
J. J. Am. Chem. Soc. 1998, 120, 1110–1111.
29. Burgi, H. B.; Dunitz, J. D.; Shefter, E. Nat. New Biol.
¨
1973, 244, 186–188.
30. Caroll, J. D.; Jones, P. R.; Ball, R. G. J. Org. Chem. 1991,
56, 4208–4213.
31. Wunderlich, J. A. Acta Crystallogr. 1967, 23, 846–855.

A. Gautier et al. / Bioorg. Med. Chem. 14 (2006) 3835–3847
3847
32. Schweizer, W. B.; Procter, G.; Kaftory, M.; Dunitz, J. D.
Helv. Chim. Acta 1978, 2783–2808.
33. (a) Hirth, M.; Krapf, H.; Riedl, H.; Sauer, J.; Oeser, E.
Chem. Ber. 1976, 109, 576–596; (b) Eicher, T.; Boehm, S.
Chem. Ber. 1974, 107, 2186–2214.
43. Munegumi, T.; Maruyama, T.; Takasaki, M.; Harada, H.
Bull. Chem. Soc. Jpn. 1990, 63, 1832–1834.
44. Albeck, A.; Persky, R. Tetrahedron 1994, 50, 6333–6346.
45. See table in the supplementary data for DNMR
experiment.
34. McCrindle, R.; McAlees, A. J. Chem. Commun. 1983, 61–
62.
35. Davies, J. E.; Kirby, A. J.; Komarov, I. V. Helv. Chim.
Acta 2003, 86, 1222–1233.
36. For a reliable prediction of the interaction in any given
molecular context, modeling of simpler systems by ab ini-
tio methods are currently in progress, which, in return, will
serve to model their complexes with selected enzymes by
hybrid potentials (QM/MM): Field, M. J. J. Comput.
Chem. 2002, 23, 48–58.
46. Leonard, N. J.; Oki, M. J. Am. Chem. Soc. 1955, 77, 6239–
6240.
47. See supplementary data.
48. Gershkovich, A. A.; Kholodovych, V. V. J. Biochem.
Biophys. Methods 1996, 33, 135–162.
49. Matayoshi, E. D.; Wang, G. T.; Krafft, G. A.; Erickson, J.
Science 1990, 247, 954–958.
50. Burlingham, B. T.; Widlanski, T. S. J. Chem. Ed. 2003, 80,
214–218.
51. Cornish-Bowden, A. Fundamentals in Enzyme Kinetics;
Portland Press Ltd: London, 1995.
52. See, for example: Babine, R. E.; Bender, S. L. Chem. Rev.
1997, 97, 1359–1472, p 1461.
53. (a) Veber, D. F.; Johnson, S. R.; Cheng, H. Y.; Smith, B.
R.; Ward, K. W.; Kopple, K. D. J. Med. Chem. 2002, 45,
2615–2623; (b) Larsen, T.; Link, A. Angew. Chem. Int. Ed.
2005, 44, 4432–4434.
37. Seebach, D.; Sting, A. R.; Hoffmann, M. Angew. Chem.
Int. Ed. 1996, 35, 2708–2748.
38. Smith, A. B.; Guzman, M. C.; Spengeler, P. A.; Keenan,
T. P.; Holcomb, R. C.; Wood, J. L.; Carroll, P. J.;
Hirschmann, R. J. Am. Chem. Soc. 1994, 116, 9947–9962.
39. Yoo, D.; Oh, J. S.; Lee, D. W.; Kim, Y. G. J. Org. Chem.
2003, 68, 2979–2982.
40. Kunz, H.; Unverzagt, C. Angew. Chem. Int. Ed. 1984, 23,
436.
54. See, for example: Jaskolski, M.; Tomasselli, A. G.;
Sawyer, T. K.; Staples, D. G.; Heinrikson, R. L.;
Schneider, J.; Kent, S. B. H.; Wlodawer, A. Biochemistry
1991, 30, 1600–1609.
55. Rich, D. R.; Sun, C.-Q.; Vara Prasad, J. V. N.; Pathiass-
eril, A.; Toth, M. V.; Marshall, G. R.; Clare, M.; Mueller,
R. A.; Houseman, K. J. Med. Chem. 1991, 34, 1222–1225.
41. Coupling reagents DCC/HOBt gave only 14% yield.
42. CIP/HOAt combined coupling reagents were used for
peptide coupling reactions between sterically hindered a,a-
dialkylated amino acids: Akaji, K.; Kuriyama, N.; Kiso,
Y. J. Org. Chem. 1996, 61, 3350–3357.