Cyclodimerization by ring-closing metathesis: Synthesis, computational, and biological evaluation of novel bis-azetidinyl-macrocycles

Article abstract of DOI:10.1016/j.tet.2010.10.015

During our research on novel, non-traditional, bicyclic β-lactams as potential inhibitors of Penicillin Binding Proteins (PBPs), we focused on the synthesis of 1,3-bridged 2-azetidinones by RCM reaction from 1,3-bis-ω-alkenoyl-3(S)-amino-2-azetidinone pre

Full text of DOI:10.1016/j.tet.2010.10.015

Tetrahedron 66 (2010) 9519e9527
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Cyclodimerization by ring-closing metathesis: synthesis, computational,
and biological evaluation of novel bis-azetidinyl-macrocycles
,
Aline Sliwa a, Georges Dive b, Jean-Louis Habib Jiwan a, Jacqueline Marchand-Brynaert a
*
a
ꢀ
Institute of Condensed Matter and Nanosciences, Universite catholique de Louvain, Place Louis Pasteur 1, B-1348 Louvain-la-Neuve, Belgium
ˇ
ˇ
b
ꢀ
ꢀ
ꢀ
ꢁ
ꢀ
ꢁ
Centre d’ingenierie des Proteines, Universite de Liege, Batiment B6, Allee du 6 Aout, 4000 Sart-Tilman (Liege), Belgium
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 17 August 2010
Received in revised form 28 September 2010
Accepted 8 October 2010
Available online 14 October 2010
During our research on novel, non-traditional, bicyclic
b-lactams as potential inhibitors of Penicillin
Binding Proteins (PBPs), we focused on the synthesis of 1,3-bridged 2-azetidinones by RCM reaction from
1,3-bis- -alkenoyl-3(S)-amino-2-azetidinone precursors. Submitting the precursors to RCM, we faced an
u
unexpected problem: cyclodimerization was the preferred outcome. This peculiar reactivity, explained by
a computational study, led to unprecedented bis-azetidinyl-macrocycles acting as potent inhibitors of
R39 D,D-carboxypeptidase, a bacterial model enzyme for PBPs.
Keywords:
RCM
Ó 2010 Elsevier Ltd. All rights reserved.
b
-Lactams
Cyclodimers
B3LYP calculations
R39 inhibitors
1. Introduction
favor the dimer ring-closing metathesis reaction4 and the com-
petitive pathways leading to the cyclic monomer, and dimer have
Ring-closing metathesis (RCM) of
a
,u
-dienes is nowadays a very
been theoretically studied with 1,7-octadiene as model substrate.5
We have previously applied the RCM reaction for the prepara-
common reaction used for the construction of medium-sized and
macrocyclic organic compounds. Its wide scope and use as the key-
step in numerous total syntheses of complex target molecules have
been reviewed1 recently. The thermodynamic and kinetic aspects
of the RCM and related reactions catalyzed by ruthenium/carbene
complexes have been discussed as well.2
tion of 1,3-bridged bicyclic b-lactams A (i.e., 2-azetidinones) de-
rived from the commercial acetoxy-azetidinone used as the chiral
precursor of carbapenem antibiotics6 (Fig. 1, Eq. (a)). Series of
Despite an increasing amount of technical reports, there are still
no clear rules available for planning a new RCM synthesis: different
catalysts and loading, different substrate concentrations, addition,
and reaction times, different solvents, additives, and temperatures
have to be tested for maximizing the yield of the desired cyclic
product. Beside the experimental conditions, the ring size to be
formed, the substrate structure, its substitution pattern, steric, and
conformational factors will also influence the outcome of the re-
action, i.e., the product E/Z ratio, the formation of isomerized or/and
oligomerized by-products. In general, oligomerization is consid-
ered detrimental to the RCM reaction2 and the corresponding
products are neither isolated nor characterized. But in few cases,
dimers, and in particular cyclic dimers, are highly desired prod-
ucts.3 Recently, a double-centered catalyst has been designed to
* Corresponding author. Tel.: þ32 10 47 27 40; fax: þ32 10 47 41 68; e-mail
address: jacqueline.marchand@uclouvain.be (J. Marchand-Brynaert).
Fig. 1. Two families of 1,3-bridged 2-azetidinones.
0040-4020/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2010.10.015

9520
A. Sliwa et al. / Tetrahedron 66 (2010) 9519e9527
compounds featuring 12 and 13-membered rings (including HC]
CH double bond or the reduced motif) have been obtained and
evaluated as potential inhibitors of Penicillin Binding Proteins
(PBPs). The weak activities recorded were attributed, among other
factors, to the steric effect of the macrocycles, which hinders the so-
Boc-L-serine 1 was converted, in nearly quantitative yield, into
the corresponding hydroxamate 2, employing O-benzylhydroxyl-
amine and DCC.7 Intramolecular cyclization was performed with
the method proposed by Miller et al.,7 in the presence of PPh3 and
CCl4, and afforded the b-lactam 3 in 78% yield. Subsequent hydro-
called
a
-face of the azetidinone ring and thus might prevent
genation in the presence of Raney nickel8 gave the desired chiron 49
in quantitative yield (Scheme 1). In the first attempt to obtain the
corresponding NeMe chiron 8, we considered the direct methyla-
tion10 of 3 into the desired intermediate 7. This reaction proceeded
with 62% yield at the mmol scale, but the scaling-up was not suc-
cessful. So we developed a similar route toward the NeMe de-
nucleophilic attack by the active serine of PBPs. This hypothesis
derives from a theoretical investigation of the reactivity of the
bridged azetidinones A when processed into a model of serine
protease active site.6 To further our research on novel, non-tradi-
tional, bicyclic
b-lactams, we have focused on 1,3-bridged com-
pounds B derived from 3-amino-2-azetidinone featuring the amino
substituent and the chirality of penicillins, and cephalosporins at
the C(3) carbon (Fig. 1, Eq. (b)). This inversion of configuration re-
garding the carbapenem chiron (see Eq. (a)) should position the
rivatives as for the NeH derivatives, starting from Boc-Me-
5. This compound was converted into hydroxamate 6 employing
EDCI in 91% yield. The -lactam 7 was obtained in 82% yield and
L-serine
b
subsequent hydrogenation afforded the chiron 8 in 64% yield.
macrocycle above the b-face of the azetidinone ring and conse-
quently make the serine nucleophilic attack easier on the a-face
during the processing of molecules B by PBPs.
In this article, we describe our synthetic efforts toward the
target molecules B belonging to the bis-acylated family (X¼Y¼O).
From the 3-amino-2-azetidinone chiron, three series of RCM pre-
cursors have been prepared (R0¼Boc, H or Me) varying by the length
of the
u-alkenoyl chain fixed on the N(1) and NeC(3) atoms. Sur-
prisingly, under RCM conditions, the isolated products were all
cases but one the cyclic dimers instead of the expected cyclic
monomers B. These observations stimulated a theoretical in-
vestigation of the possible cyclization reactions. Our study high-
lights the dramatic effect of accessible conformations and flexibility
of precursors on the intra- or intermolecular ring closure, particu-
larly when the substrates possess amide, imide, and carbamate
functions.
Scheme 1. Synthesis of chirons 4 and 8. Reagents and conditions: (a) DCC, NH2OBn,
THF, 0 ꢀC to rt; (b) PPh3, CCl4, TEA, CH3CN, 0 ꢀC to rt; (c) H2, Raney-Ni, MeOH, rt; (d)
EDCI, NH2OBn, CH2Cl2, 0 ꢀC to rt; (e) Me2SO4, LiHMDS, THF, ꢁ78 ꢀC to rt.
The inhibition potential of our compounds (precursors and RCM
products) has been tested against R39 D,D-carboxypeptidase,
which is a commonly used model for bacterial enzymes.
The N-Boc bis-acylated monocyclic azetidinones 9aec were
obtained in one step by treatment of the chiron 4 with 2 equiv of
alkenoyl chlorides in the presence of 2 equiv of lithium hexame-
thyldisilazide (Scheme 2). 4-Butenoyl chloride (n¼0) is commer-
cially available, while 5-hexenoyl chloride (n¼1) and 6-heptenoyl
chloride (n¼2) have been previously synthesized starting from the
corresponding commercial carboxylic acid.11
2. Results and discussion
2.1. Synthesis
The RCM reaction was chosen as the key-step for the synthesis of
1,3-bridgedbicyclesB, astrategy, whichhasalreadybeensuccessfully
used for the preparation of 1,3-bridged derivatives A.6 The precursors
are chiral azetidin-2-ones C equipped with u-alkenoyl chains on the
positions N(1)andC(3)eN(Fig. 2). Thestartingchirons,i.e., (S)-3-(tert-
butyloxycarbonyl)amino-2-azetidinone and (S)-3-N-methyl(tert-
butyloxycarbonyl)amino-2-azetidinone, derived from commercially
available Boc-L-serine and Boc-Me-L-serine, respectively.
Scheme 2. Synthesis of bis-acylated 9. Reagents and conditions: (a) alkenoyl chloride,
LiHMDS, THF, ꢁ78 ꢀC to rt.
For the preparation of NeH and NeMe bis-acylated precursors,
a three-step sequence was used (Scheme 3). The chiron 4 was
mono-acylated at the N(1) position regioselectively under mild
conditions giving the series of compounds 10aec. Then, the Boc
protecting group was removed with trifluoroacetic acid and the
free amine function was acylated with the alkenoyl chlorides
affording the series of compounds 12aec. Similarly, the chiron 8
was mono-acylated (compounds 11aec), N-deprotected and acyl-
ated again to produce the NeMe precursors 13aec.
The first attempts at RCM were performed on bis-acylated
compounds with n¼0 (i.e., 9a, 12a, and 13a) with the view to form
12-membered macrocycles, under standard reaction conditions
(CH2Cl2, 40 ꢀC, 5 mM) in the presence of second generation Grubbs’
Fig. 2. General RCM strategy.

A. Sliwa et al. / Tetrahedron 66 (2010) 9519e9527
9521
spectra were not well resolved,2 most probably due to the presence of
several rotamers, stereoisomers (mixtures of E/Z olefins), regioisom-
ers (HH and HT dimers), and possible contamination by lower ho-
mologs (double bond migration and cyclization with the formal
extrusion of oneortwo CH2 groups). The 13C NMR spectrajust showed
the disappearance of the terminal olefin bonds and the preservation
of the b-lactam ring, with the presence of the typical signals of C(3)
and C(4) (Table 1, entries 2, 4, and 6). In fact, mass spectrometry (MS)
turned out to be the appropriate tool for accurate structural de-
termination. Spectra were recorded in electrospray ionization (ESI)
and atmospheric pressure chemical ionization (APCI) modes for
comparison. All the isolated compounds were cyclic dimers, most
probablyamixtureofheadehead(HH)andheadetail(HT)isomers,as
drawn in the Fig. 3. ESI is a soft ionization method but well-known to
favor the formation of dimeric adducts during the evaporation step.
Moreover, ionization by formation of Naþ or Kþ adducts is also fre-
quently observed. These adducts can be monomeric or dimeric. In our
case, as observed during the analysis of the bis-acylated precursors,
formation of dimeric Naþ adducts was the major ionizationprocess in
ESI. Soforthecompoundsissuedof theRCM, wecouldnotbesurethat
we really observe a cyclodimer compound in complex with Naþ or
two cyclomonomer compounds in complex with Naþ. To confirm, we
applied a somewhat harder method of ionization, APCI not known to
favor formation of adducts, which showed that the compounds are
really the cyclodimers (Table 1). Moreover, the analysis of the ESI
spectra allowed us to detect the presence, in relatively weak abun-
dance, of lower homologs (double bond migration and cyclization
with the formal extrusion of one CH2 group).
Scheme 3. Synthesis of bis-acylated 12, 13. Reagents and conditions: (a) alkenoyl
chloride, pyridine, CH2Cl2, rt; (b) TFA, CH2Cl2, 0 ꢀC to rt; (c) alkenoyl chloride, TEA,
CH2Cl2, 0 ꢀC to rt.
catalyst (5 mol %). Conversion was not observed in any case. To
prevent the possible deactivation of the Ru-catalyst by the forma-
tion of a stable chelated alkylidene ring, we tried to use Ti(OiPr4)12
as an additive, without success. Increasing the temperature (DCE,
80 ꢀC), with or without Ti(OiPr4), only led to the degradation of the
starting material.
Next, RCM reactions were tested on compounds with n¼1 (i.e., 9b,
12b, and 13b) leading in principle to 14-membered macrocycles,
under the same standard conditions previously described. In each
case, monitoring the reaction by TLC showed apparently the pro-
gressive formation of a single product over a time period of 24 h. After
several chromatographies, products were isolated with moderate
yields (35e44%). Products obtained from 9b,12b, and 13b are named
15b, 16b, and 17b (Fig. 3), respectively. Their structural assignment
proved to be an unexpectedly difficult problem: the 1H and 13C NMR
The RCM reactions were finally tested on compounds with n¼2
(i.e., 9c,12c, and 13c) in the same standard conditions as previously.
16-Membered cyclomonomers are expected, but as previously for
compounds with n¼1, the reactions afforded the cyclodimers 15c,
16c, and 17c, respectively, in modest yields (10e30%) (Table 1 and
Fig. 3). Surprisingly, in one case (starting material 9c), the macro-
cyclic monomer 14c was also obtained (Table 1 and Fig. 3), but in
low yield (9%).
The monomer 14c and the series of cyclodimers 15b, c, 16b, c
and 17b, c were subjected to catalytic hydrogenation to produce the
corresponding saturated macrocycles 18c and 19b, c, 20b, c, and
21b, c, respectively (Fig. 3).
Bis-acylated precursors with n¼0 did not cyclize at all: neither
the cyclomonomers nor the cyclodimers could be detected. Bis-ac-
ylated precursors with n¼1 gave cyclodimers but did not cyclize into
the monomers, therefore we suspected that the length of the carbon
chain bearing the olefin was too short to afford the 14-sized mac-
rocycle. On the other hand, the formation of compound 14c, proves
that the chain length is long enough to produce the 16-sized cyclic
product, even if the dimerization remains the preferred process. We
set out to find reaction conditions to promote the formation of
cyclomonomer derived from 12c (n¼2). There are numerous pa-
rameters that could influence the outcome of the RCM, and there are
no pre-established conditions, which would guarantee the suc-
cessful production of a desired compound. We have tested several
parameters, such as the choice of the catalyst, the solvent, the
temperature, the concentration, and the reaction time.2
First and second generation Grubbs or HoveydaeGrubbs, and
Neolyst M2 catalysts were tested but in all cases, they only afforded
the cyclodimer 20c (n¼2). Carrying out the RCM reactions at rt also
produced the cyclodimer. It is known that increasing the reaction
temperature can control the product distribution (monomer vs di-
mer) in RCM macrocyclization.13 But in our case, the RCM reaction in
refluxing DCE afforded only the cyclodimer, just faster than at rt.
Decreasing the concentration of the reaction, from 5 mM to 1 mM, or
using the ‘infinite dilution’ conditions14 led to a lower yield of the
cyclodimer; the desired cyclomonomer was never detected. The
results of these various attempts seem to point toward intrinsic
Fig. 3. Compounds synthesized by RCM.

9522
A. Sliwa et al. / Tetrahedron 66 (2010) 9519e9527
MS ESI m/z Relative abundance
Table 1
Selected structural data of cyclic dimers and monomer
Entry
1
Compound
13C C4; C3; C]C (ppm)
MS APCI m/z [MþH]þ
379.22275
Isolated yielda (%)
9b
14c
44.6
59.7
401 (69) [MþNa]þ
301 (100) (401-CO2 and C4H8)
(C20H31O5N2)
130.6 and 131.1
43.7e43.8
57.5e57.8
2
3
15b
15c
723 (100) [MþNa]þ
701.37449
(C36H53O10N4)
44
709 (13) (723-CH2)
129.9e130.2-130.5e131.6
623 (48) (723-CO2 and C4H8)
523 (68) (723e2(CO2 and C4H8)
779 (100) [MþNa]þ
43.8e43.9
57.2e57.4
757.43682
(C40H61O10N4)
10b
765 (10) (779-CH2)
130.3e130.5-130.6e130.7
679 (37) (779-CO2 and C4H8)
579 (33) (779e2(CO2 and C4H8))
523 (100) [MþNa]þ
4
5
6
7
16b
16c
17b
17c
46.0
55.7
501.26999
(C26H37O6N4)
35
28
43
40
509 (11) (523-CH2)
130.8e131.5
45.1e45.4
55.2e55.6
130.0e130.6-130.8e130.9
43.1
61.8e63.0
130.4e130.7-131.0e131.4
43.1e43.2
62.5e63.1
579 (100) [MþNa]þ
565 (35) (579-CH2)
551 (7) (579-C2H4)
551 (100) [MþNa]þ
537 (14) (551-CH2)
557.33287
(C30H45O6N4)
529.30170
(C28H41O6N4)
607 (100) [MþNa]þ
585.36407
(C32H49O6N4)
593 (38) (607-CH2)
130.2e130.4-130.5e130.6
a
Conversion (80e90%). Crude yields (50e60%). Analysis of crude mixtures by TLC, MS, and NMR: oligomers, isomerized starting materials, and
b
-lactam degradation
compounds were observed as side-products. Several column chromatographies are necessary to clear out the catalyst.
b
Very low isolated yields because major fraction is a mixture of 14c and 15c.
structural parameters governing the outcome of our RCM reactions.
In the next sectionwe describe ab initio calculations that explain our
results.
2.2. Computational chemistry
The almost exclusive formation of macrocyclic dimers instead of
cyclic monomers under RCM conditions is rarely mentioned in
previous literature and thus deserves to be examined from the
theoretical point of view.
The geometry of all the molecules has been fully optimized at
the B3LYP level using 6-31G basis set with added polarization
functions.15 All the calculations have been performed with the
Gaussian 03 program.16 For the Grubbs adducts, the basis sets
LanL2MB and Lan2DZ, minimal and double
z functions for the first
row and Los Alamos effective core potential on the other atoms,
have been used with the same B3LYP functional.17
The precursors have been studied (bis-acylated azetidinones 9
(N-Boc), 12 (NeH), and 13 (NeMe)) taking into account several
factors, which could have an incidence on the formation of the
cyclic monomer (see Supplementary data): the tautomeric amide
forms, the cis or trans geometry of the amide functions, the con-
Fig. 4. Degrees of rotation.
the N-Boc substitution, an additional degree of freedom has to be
taken intoaccount: the conformationwith the two Boc carbonylssyn
and anti noted a and b (Fig. 4 and Table 2). Last, several conforma-
tions of the 14- and 16-membered rings under the
have been located. They are less stable than the up conformations
(i.e., above the -lactam plane) with a mean energy difference of
formations of the
down with respect to the
u
-alkenoyl chains leading to a cycle located up or
-lactam plane in a range of 5e10 kcal/
b
b-lactam plane
mol. In all cases, the OH-tautomers are less stable than the standard
amide in a range of 10e20 kcal/mol. In the same way, the trans
amide and the conformation anti of the two carbonyls of the imide
are preferred. The syn and anti conformations are, respectively,
noted i and ii (Fig. 4 and Table 2). The conformational space is
strongly influenced by the interaction between the two imide
carbonyls. which tend to move off.
The trans/cis rotation barrier depends on the substituent of the
amide nitrogen. The decreasing values going from NeH, NeMe and
N-Boc are 20.49,16.86, and 3.50 kcal/mol. A great number of minima
could be localized in a large energy range (20 kcal/mol). The ring size
also has an incidence on the conformation related to the number of
CH2 on both sides of the double bond. In order to sweep a common
conformational space for all the substituted molecules 9, 12, and 13,
four conformations (amide cis or trans; imide carbonyls in
conformation i or ii) have been searched and located. In the case of
b
4 kcal/mol for 9c and 10 kcal/mol for the 9b.
The heats of formation of the cyclic monomers have been calcu-
lated with respect to the parent conformation of the open precursor:
all the values are positive (Table 2). Fig. 5 summarizes, on the same
graph, for the compound 14c (n¼2) the relative energies of the pre-
cursor 9c/cyclic monomer in the selected conformations and also the
respective heat of formation. Most of the time, the amide trans con-
formations are the most stable ones. Fig. 6 illustrates the conforma-
tional diversity for compound 14c with 4 conformers represented. It
can be noted that the heat of formation of the N-Boc substituted
molecules is higher for the 12-membered ring (9a precursor) than for
the 14- and 16-membered rings (9b and 9c precursors).
In order to investigate the incidence of the conformations on the
cycle formation, the two possible intermediates of the Grubbs

A. Sliwa et al. / Tetrahedron 66 (2010) 9519e9527
9523
Table 2
Relative energies of the precursors/cyclic monomers in the selected conformations and also the respective heat of formation resulting from the cyclization of the 9 (N-Boc),
12 (NeH), and 13 (NeMe) precursors
Precursor
Geometry
Relative energy of open
precursors (kcal/mol)
Relative energy of monomers
(kcal/mol)
Heat of formation of
monomers (kcal/mol)
i
ii
i
ii
i
ii
12a
12b
12c
13a
13b
13c
9a
Amide cis
Amide trans
Amide cis
Amide trans
Amide cis
Amide trans
Amide cis
Amide trans
Amide cis
Amide trans
Amide cis
Amide trans
Amide cis a
Amide cis b
Amide trans a
Amide trans b
Amide cis a
Amide cis b
Amide trans a
Amide trans b
Amide cis a
Amide cis b
Amide trans a
Amide trans b
10.21
5.18
10.33
5.95
10.22
4.88
8.16
6.03
8.59
6.53
0.00
9.05
17.09
13.63
6.19
6.75
8.45
5.04
5.51
5.96
15.32
11.96
5.54
5.03
4.94
0.00
4.77
0.00
4.77
0.00
2.84
0.00
3.01
0.00
6.47
3.66
11.80
8.46
0.88
0.00
3.56
0.26
0.45
0.00
10.81
7.54
1.02
0.00
4.19
9.17
4.95
5.56
0.00
0.89
7.12
0.00
1.88
0.00
0.00
4.30
0.00
2.97
0.21
0.00
9.53
6.18
0.99
0.00
2.99
11.23
0.50
0.00
8.17
4.92
0.42
0.00
8.63
18.65
7.47
12.45
8.53
8.57
22.55
15.90
7.26
12.09
20.70
8.32
0.21
5.57
13.89
13.39
6.47
6.61
4.96
3.87
8.16
9.71
15.54
15.20
12.85
5.56
8.45
9.29
16.43
6.17
12.15
5.78
10.26
5.00
18.58
9.80
6.67
9.45
8.68
5.35
0.99
2.89
3.77
3.83
11.38
8.12
6.93
6.30
15.73
12.40
5.46
5.03
8.36
14.03
14.02
16.42
16.31
2.96
14.50
3.59
3.53
5.11
5.13
7.15
7.75
9a
9b
9b
9ca
9ca
8.20
7.67
7.75
a
Led to compound 14c.
Fig. 5. Relative energies of precursor/cyclic monomer in the selected conformations
and heats of formation for 14c.
catalyst first step addition have been computed for 9c, 12c, 9b, and
12b (see Supplementary data). In NeH series, some adducts gener-
ated with both functions amide and imide in the cis or trans con-
formation have a favorable orientation for the ring closing into
monomer. But, only the cis amide conformation allows the in-
troduction of the N-Boc substituent due to steric hindrance. This
spatial orientation, which isconstrainedbytheN-Bocsubstituentcan
be related to the appearance of 14c as RCM monomer product along
with the dimers. Fig. 7 illustrates the optimized geometry of the 9c
adduct with the NeC(3)-arm linked to the Ru atom.
Inordertoexplainthepreferentialformationofdimers, thesearch
of different conformations has been investigated. For the three sizes
of the ring, two conformations of the b-lactam can be located giving
Fig. 6. Four conformers of compound 14c, in geometry b.
rise to a headehead (HH) or headetail (HT) arrangement in the
macrocycle (see Fig. 3). Moreover, an additional factor has to be
considered, mainly for the NeH substitution, due to the formation of
intramolecular H bonds, which significantly stabilize some confor-
mations. Again, withtheconformationcis/transoftheamideandsyn/
anti imide orientation, a lot of conformations have been found, which
are not necessarily geometrically related going from NeH to N-Boc
substitutions. As an example, Fig. 8 illustrates the most stable con-
former of compounds 15c (HT) and 16c (HH).
The energy range is large and superior to 20 kcal/mol. In most of
the cases, the trans conformation of the two amides is more stable

9524
A. Sliwa et al. / Tetrahedron 66 (2010) 9519e9527
the tested azetidinones (100 m
M) were incubated (1 h, 25 ꢀC) to-
gether. After preincubation, the residual activity (RA) was de-
termined by observing the hydrolysis of the thiolester substrate,19
in the presence of DTNB (a fluoresceine-labeled ampicillin20), cat-
alyzed by the non-inhibited enzyme. Thus, in this protocol, our
compounds do acylate R39, and the RA is determined by the
amount of covalent R39-ampicillin complexes formed, measured
by fluorescence spectroscopy. The results are given in Table 4 as
percentages (%) of initial activity. Low values indicate very active
compounds as the bacterial enzyme has been inhibited by the
tested compound and consequently cannot hydrolyze its reporter
substrate. A tested compound is considered as a ‘hit’ (i.e., potential
inhibitor) for a RA<80%.
Fig. 7. Optimized geometry of the 9c adduct.
Table 4
Evaluation of azetidinones against R39 D,D-carboxypeptidase
Entry
Compound
RA (%)
Entry
Compound
RA (%)
1
2
3
4
5
6
7
8
9a
9b
9c
82ꢂ5
87ꢂ1
96ꢂ3
>100
17ꢂ9
23ꢂ0
>100
103ꢂ3
95ꢂ7
0
13
14
15
16
17
18
19
20
21
22
23
24
15c
16b
16c
17b
17c
19b
19c
20b
20c
21b
21c
PenG
4ꢂ3
4ꢂ8
0
93ꢂ5
43ꢂ3
101ꢂ8
103ꢂ5
2
12a
12b
12c
13a
13b
13c
14c
18c
15b
9
0
10
11
12
88ꢂ7
39ꢂ0
021
14ꢂ10
111ꢂ2
Amongst the precursors (non-cyclized compounds: Table 2,
entries 1e9), two molecules are good inhibitors of R39, namely 12b
and 12c (NeH family, entries 5e6). The cyclomonomers 14c/18c
(entries 10e11) also inhibited efficiently R39 despite the fact that
the molecules are still N-Boc protected. Unfortunately, the corre-
sponding NeH derivatives were not accessible for testing because
the TFA treatment of 14c and 18c (for Boc cleavage) gave a mixture
Fig. 8. Most stable conformer of compounds 15c (HT) and 16c (HH).
of deprotected
b-lactams and hydrolyzed products. Surprisingly,
than other combinations cis/trans or cis/cis. For this spatial disposi-
tion, a complete homogenous set of conformers has been located.
Mostof the time, the HTconformersaremore stablethantheHH ones.
As for the monomers, the heat of formation has been calculated with
respect to the open precursor (Table 3). In many cases, the heat of
formation of dimer is negative or slightly positive with an energy
demand lower than the one required for the cyclic monomer forma-
tion (see Table 2). Remarkably, the heat of formation of 15c corre-
sponds to the highest positive value (9.13 kcal/mol) of the selected
conformers, higher than the one of the cyclic monomer 14c, which is
experimentally also observed (7.75 kcal/mol).
the cyclodimers (entries 12e23) are generally more active than
their respective precursors and some of them showed very good
inhibition potential, i.e., 16b, c (entries 14,15), and 20b, c (entries
20e21); these correspond to the NeH (deprotected) derivatives. It
is worth noting that all these non-traditional inhibitors are quite
lipophilic compounds, devoid of the carboxylic function usually
found in penicillins and related antibiotics.
3. Conclusion
The RCM reaction applied to azetidinone substrates featuring
amide and imide functions led almost exclusively to macrocyclic
dimers instead of the cyclomonomers. These unexpected results
are in fact in good agreement with the computed heats of formation
of the dimers and monomers, respectively. Our theoretical in-
vestigation has highlighted the dramatic importance of the con-
formational flexibility of the substrates and products on the RCM
reaction. Interestingly, in the only case where the dimer (i.e., 15c)
was predicted to be slightly less favored than the corresponding
monomer (i.e., 14c), both products were experimentally observed.
The biological activity results collected with our compounds
using the R39 serine-enzyme are promising. The novel bicyclic
Table 3
Heat of formation of macrocyclic dimers for selected compounds
Precursor Product
DE (HHeHT) (kcal/mol) Heat of formation (kcal/mol)
12a
13a
9a
12b
13b
9b
12c
13c
9c
d
2.23
4.09
ꢁ1.11
1.75
d
d
10.37
9.59
ꢁ1.l3
ꢁ1.74
ꢁ2.43
3.09
6.55
16b
17b
15b
16c
17c
15c
ꢁ5.66
3.76
ꢁ8.06
ꢁ1.81
4.99
5.35
9.13
b-lactams 14c and 18c are good inhibitors of this D,D-peptidase
2.3. Inhibition of R39 D,D-carboxypeptidase
while their precursor 9c is not active. Also, the bis-azetidinyl-mac-
rocycles 15c, 16b, 16c, 20b, and 20c have shown remarkable in-
hibitory effect.
Since the azetidinone moieties are not activated by the tradi-
tional angular strain leading to ‘twisted’ amide functions like in
penicillins,22 the activity should be attributed, amongst other
All bis-acylated compounds, precursors, and cyclic products,
were tested against Actinomadura R39,18 a model serine-enzyme of
low molecular weight D,D-carboxypeptidases, usually considered
for a preliminary screening of penicillin-like compounds. R39 and

A. Sliwa et al. / Tetrahedron 66 (2010) 9519e9527
9525
factors, to the flexibility of the substrates (a lot of conformers are
possible) allowing their adjustment into the enzymic pocket and
their easy conformational rearrangement during the processing by
the active site serine residue. Moreover, the most active compounds
are those with R0¼H (16b, c, 20b, c) comparatively to the corre-
sponding N-substituted derivatives with R0¼Me (17b, c, 21b, c) and
R0¼Boc (15b, 19b, c), most probably for steric reasons, and the ac-
tivity does not seem to be so much dependent on the preferred
headetail or headehead dimeric structures.
Work is in progress for producing preferably bicyclic azetidi-
nones related to the structures 14c/18c, i.e., cyclomonomers. In case
of difficult macrocyclizations, a novel auxiliary has been recently
developed as conformational control element (CCE) favoring the
appropriate conformation toward the desired RCM reaction.23 Since
this elegant strategy is not readily applicable to our precursors 9,12,
and 13, we are now investigating the corresponding bis-alkylated
derivatives, which show an increase of the freedom degrees for the
side-chains rotations.
under reduced pressure, the residue was purified by flash column
chromatography (hexane/EtOAc 4/1), to provide 9c as a pale-yellow
20
oil (366 mg, 56%). Rf¼0.58 (hexane/EtOAc 4/1); [
a]
þ1.4 (c 3.0,
D
CHCl3); 1H NMR (300 MHz, CDCl3):
d
¼5.72e5.86 (m, 3H),
4.92e5.03 (m, 4H), 3.83 (m,1H), 3.61 (dd, J¼4.2, 7.1 Hz,1H), 2.88 (m,
2H), 2.73 (m, 2H), 2.07 (m, 4H), 1.60e1.74 (m, 4H), 1.51 (s, 9H),
1.38e1.48 ppm (m, 4H); 13C (75 MHz, CDCl3):
d
¼175.5, 171.2, 164.5,
151.2, 138.5, 138.4, 114.8 (2C), 85.6, 56.9, 43.9, 38.2, 36.5, 33.6, 33.5,
28.4, 27.9 (2C), 24.4, 23.5 ppm; IR:
n
¼2860e3076, 1798, 1747,
1697 cmꢁ1; HRMS: calcd for C22H34N2O5Na 429.2365, found
429.2382.
4.2.2. (S)-tert-Butyl
(10c). To a stirred solution of
1-hept-6-enoyl-2-oxoazetidin-3-ylcarbamate
-lactam 4 (427 mg, 2.29 mmol) in
b
dry CH2Cl2 (19 mL) were added pyridine (0.37 mL, 4.59 mmol) and
6-heptenoyl chloride (673 mg, 4.59 mmol). The mixture was stirred
for 24 h at rt and then diluted with CH2Cl2 (30 mL) and the organic
layer was washed with HCl 2 M solution (50 mL), a saturated
NaHCO3 solution (50 mL), and brine (50 mL). After drying over
MgSO4 and removing the solvent under reduced pressure, the
residue was purified by flash column chromatography (hexane/
4. Experimental section
4.1. General considerations
EtOAc 4/1), to provide 10c as a colorless solid (523 mg, 77%).
20
Rf¼0.45 (hexane/EtOAc 3/2); mp 92.9e93.6 ꢀC; [
a
]
þ11.3 (c 2.6,
D
Experiments were performed under argon atmosphere in
flame-dried glassware. All solvents, including anhydrous solvents,
and reagents were purchased from Acros Organics, Alfa Aesar,
Fluka, SigmaeAldrich or VWR, and used without any further pu-
rification. TLC analyses were performed on aluminum plates coated
with silica gel 60F254 (Merck) and visualized with a KMnO4 solution
and UV (254 nm) detection, and flash column chromatography was
performed on silica gel (40e60 mesh) purchased from Rocc.
CHCl3); 1H NMR (300 MHz, CDCl3):
d
¼5.78 (m, 1H), 5.17 (d,
J¼7.6 Hz, 1H), 4.92e5.03 (m, 2H), 4.44 and 4.66 (2 br s, 1H,
rotamers), 3.86 (m, 1H), 3.64 (dd, J¼3.9, 7.6 Hz, 1H), 2.72 (m, 2H),
2.07 (m, 2H), 1.67 (m, 2H), 1.40e1.49 ppm (m, 11H); 13C (75 MHz,
CDCl3):
d
¼171.3, 165.2, 154.7, 138.5, 114.8, 81.2, 56.6, 45.6, 36.6, 33.5,
28.4, 28.3, 23.5 ppm; IR:
n
¼3354, 2864e2976, 1796, 1689,
1518 cmꢁ1
319.1636.
; HRMS: calcd for C15H24N2O4Na 319.1634, found
€
Melting points (mp) were determined on a Buchi B-540 apparatus
calibrated with caffeine, vanillin, and phenacetin. [
a
]
was mea-
4.2.3. (S)-tert-Butyl 1-hept-6-enoyl-2-oxoazetidin-3-yl(methyl)car-
bamate (11c). Compound 11c was synthesized according to the
procedure described above for the synthesis of compound 10c from
D
sured on PerkineElmer 241MCpolarimeter, at 20 ꢀC, in CHCl3.
Concentrations are given in percentage (g/100 mL). Nuclear mag-
netic resonance (1H and 13C) spectra were recorded at 300 MHz for
proton and 75 MHz for carbon (Bruker Avance 300) or 500 MHz for
proton and 125 MHz for carbon (Bruker Avance 500) using deu-
terated chloroform (CDCl3) or deuterated methanol (CD3OD).
Chemical shifts are reported in parts per million relative to residual
CHCl3 in CDCl3 (7.26 and 77.16 ppm) or residual CH3OH in CD3OD
(3.31 and 49.00 ppm). NMR coupling constants (J) are reported in
hertz. Infrared (IR) spectra were recorded using FTIR-8400S Shi-
madzu apparatus. Products were analyzed as thin films deposited
on an SeeZn crystal by evaporation from CH2Cl2 solutions. For
precursors, mono-acylated and bis-acylated compounds High
Resolution Mass Spectrometry (HRMS) analyses were performed at
the University of Mons Hainaut (Belgium) or at the University
College London (UK). For compounds issued from RCM, low and
high resolution mass spectrometry were performed at UCL on LTQ-
Orbitrap-XL equipment. For compounds 14e21, only the 13C NMR
spectra are given.
b-lactam 8 (250 mg, 1.24 mmol) and 6-heptenoyl chloride and the
mixture was put under reflux for 24 h. Flash column chromatog-
raphy (hexane/EtOAc 4/1) provided 11c as a colorless oil (336 mg,
20
87%). Rf¼0.53 (hexane/EtOAc 3/2); [
a]
þ11.5 (c 5.0, CHCl3); 1H
D
NMR (300 MHz, CDCl3):
d
¼5.76 (m, 1H), 4.91e5.06 (m, 2Hþ0.5H,
rotamer), 4.60 (br s, 0.5H, rotamer), 3.80 (m, 1H), 3.61 (dd, J¼3.9,
7.5 Hz, 1H), 2.92 (br s, 3H), 2.71 (br s, 2H), 2.05 (m, 2H), 1.66 (m, 2H),
1.43 ppm (br s, 11H); 13C (75 MHz, CDCl3):
d¼171.3, 164.9,
154.3e154.9 (rotamers), 138.4, 114.8, 81.4e82.1 (rotamers),
62.9e63.7 (rotamers), 43.2e44.4 (rotamers), 36.5, 35.2e33.5
(rotamers), 33.5, 28.3, 28.2, 23.5 ppm; IR:
n
¼2864e2976, 1794,
1693, 1639 cmꢁ1; HRMS: calcd for C16H26N2O4Na 333.1790, found
333.1796.
4.2.4. (S)-N-(1-Hept-6-enoyl-2-oxoazetidin-3-yl)hept-6-enamide
(12c). Trifluoroacetic acid (0.75 mL, 10.12 mmol) was added to 10c
(150 mg, 0.51 mmol) dissolved in CH2Cl2 (5 mL) at 0 ꢀC. The mixture
was warmed to rt and stirred for 2 h. Concentration of the reaction
solution afforded the crude trifluoroacetate salt as a viscous oil.
Then 6-heptenoyl chloride (111 mg, 0.76 mmol) was added to
a stirred solution of the crude trifluoroacetate salt and triethyl-
amine (0.21 mL, 1.58 mmol) in CH2Cl2 (5 mL) cooled at 0 ꢀC. The
mixture was then warmed to rt and stirred overnight. The mixture
was then diluted with CH2Cl2 (25 mL), and sequentially washed
with HCl 2 M solution (30 mL), a saturated NaHCO3 solution
(30 mL), and brine (40 mL). After drying over MgSO4 and removing
the solvent under reduced pressure, the residue was purified by
flash column chromatography (hexane/EtOAc 3/2), to provide 12c
as a colorless solid (129 mg, 83%). Rf¼0.34 (hexane/EtOAc 1/1); mp
4.2. Synthesis of precursors 9c, 10c, 11c, 12c, and 13c
4.2.1. (S)-tert-Butyl hept-6-enoyl(1-hept-6-enoyl-2-oxoazetidin-3-
yl)carbamate (9c). A stirred solution of
b-lactam 4 (300 mg,
1.61 mmol) in anhydrous THF (11 mL) was cooled to ꢁ78 ꢀC.
LiHMDS 1 N in hexane (3.54 mL, 3.54 mmol) was slowly added and
the mixture was stirred for 30 min. 6-Heptenoyl chloride (592 mg,
3.54 mmol) was then added by syringe and the mixture was stirred
for another 30 min. The mixture was then warmed to rt over 4 h
and then quenched with saturated NH4Cl (20 mL). The resulting
mixture was extracted with EtOAc (2ꢃ20 mL) and the organic
layers were washed with a saturated NaHCO3 solution (50 mL), and
brine (50 mL). After drying over MgSO4 and removing the solvent
97.2e98.5 ꢀC; [
a
]
20 þ13.4 (c 3.0, CHCl3); 1H NMR (300 MHz, CDCl3):
D
d¼6.21 (d, J¼7.0 Hz, 1H), 5.77 (m, 2H), 4.92e5.03 (m, 4H), 4.70 (td,

9526
A. Sliwa et al. / Tetrahedron 66 (2010) 9519e9527
J¼3.9, 6.9 Hz, 1H), 3.85 (m, 1H), 3.66 (dd, J¼3.9, 7.4 Hz, 1H), 2.72 (m,
2H), 2.24 (t, J¼7.5 Hz, 2H), 2.02e2.10 (m, 4H), 1.59e1.72 (m, 4H),
d
¼174.2, 173.8, 172.4, 171.7, 165.5, 164.9, 130.9, 130.8, 130.6, 130.0,
55.6, 55.2, 45.4, 45.1, 36.8, 36.5, 36.3, 36.1, 31.9, 31.8, 31.7, 28.4, 28.3,
1.36e1.50 ppm (m, 4H); 13C (75 MHz, CDCl3):
d
¼173.6, 171.3, 164.7,
28.2, 28.1, 24.7, 24.4, 23.3, 22.8 ppm; IR:
n
¼3323, 2926, 2854, 1796,
138.4, 138.3, 115.0, 114.8, 55.8, 45.0, 36.5, 35.8, 33.5 (2C), 28.4 (2C),
1697, 1678, 1533 cmꢁ1; HRMS: calcd for C30H44N4O6Na 579.31531,
found 579.31433.
24.8, 23.5 ppm; IR:
n
¼3290, 2852e3080, 1797, 1780, 1693, 1651,
1537 cmꢁ1
329.1846.
; HRMS: calcd for C17H26N2O3Na 329.1841, found
4.3.6. Compound (17b). Yield: 43% (78 mg from 0.69 mmol of 13b).
20
Rf¼0.46 (EtOAc); [
a]
ꢁ10.7 (c 1.1, CHCl3); 13C (125 MHz, CDCl3):
D
4.2.5. (S)-N-(1-Hept-6-enoyl-2-oxoazetidin-3-yl)-N-(methyl)hept-6-
enamide (13c). Compound 13c was synthesized according to the
procedure described above for the synthesis of compound 12c from
d
¼174.1, 171.4, 164.5, 131.4, 130.9, 130.6, 130.4, 63.0, 61.9, 43.1, 36.0,
35.4, 35.2, 35.1, 32.5, 31.9, 31.6, 31.4, 24.1, 24.0, 23.4, 22.8 ppm; IR:
n
¼2851e2976, 1790, 1695, 1651 cmꢁ1
;
HRMS: calcd for
b-lactam 11c (336 mg, 1.08 mmol) and 6-heptenoyl chloride. Flash
C28H40N4O6Na 551.28401, found 551.28341.
column chromatography (hexane/EtOAc 3/2) provided 13c as
20
a colorless oil (263 mg, 76%). Rf¼0.28 (hexane/EtOAc 3/2); [
a
]
4.3.7. Compound (17c). Yield: 40% (20 mg from 0.17 mmol of 13c).
D
20
þ13.3 (c 3.5, CHCl3); 1H NMR (300 MHz, CDCl3):
d
¼5.77 (m, 2H),
Rf¼0.47 (EtOAc); [
a
]
ꢁ9.1 (c 3.0, CHCl3); 13C (125 MHz, CDCl3):
D
4.89e5.02 (m, 5H), 3.81 (m, 1H), 3.63 (dd, J¼3.9, 7.4 Hz, 1H), 3.05 (s,
d
¼174.0, 173.9, 171.6, 164.5, 164.4, 130.6, 130.5, 130.4, 130.2, 63.1,
3H), 2.73 (m, 2H), 2.34 (m, 2H), 2.06 (m, 4H), 1.66 (m, 4H), 1.43 ppm
62.5, 43.2, 43.1, 36.6, 36.5, 33.6, 33.5, 33.4, 32.3, 32.1, 32.0, 29.2,
(m, 4H); 13C (75 MHz, CDCl3):
d
¼173.8, 171.3, 164.4, 138.5, 138.4,
29.0, 28.9, 28.8, 23.9, 23.8, 23.6, 23.5 ppm; IR:
n
¼2851e3003, 1786,
114.9, 114.8, 62.9, 43.1, 36.5, 35.4, 33.6, 33.5, 33.4, 28.6, 28.4, 24.1,
1736, 1716 cmꢁ1; HRMS: calcd for C32H48N4O6Na 607.34661, found
607.34582.
23.5 ppm; IR:
n
¼2858e3074, 1790, 1738, 1703, 1651 cmꢁ1; MS
(APCI): m/z: 320.95 (M); HRMS: calcd for C18H29N2O3 321.21782,
found 321.21822.
Acknowledgements
4.3. General procedure for RCM
This work was supported by the Interuniversity Attraction Pole
(IAP P6/19 PROFUSA), F.R.S.-FNRS, UCL and ULg (computational
facilities). G.D. and J.M.-B. are senior research associates of the
F.R.S.-FNRS (Belgium). J.-L.H.J. thanks the Belgian National Fund for
Scientific Research (FNRS) (FRFC 2.4555.08), the Special Fund for
Research (FSR) and the faculty of medicine of UCL for financial
support on this research (LTQ-Orbitrap). Dr. Astrid Zervosen is ac-
Grubbs catalyst (second generation) (0.05 equiv) was added to
a stirred solution of b-lactam (1 equiv) in dry CH2Cl2 (5 mM) and
the solution was stirred at reflux under argon for 4 h. Then a second
addition of Grubbs catalyst (0.05 equiv) was made and then re-
action was additionally stirred at reflux for 20 h. Then the solvent
was removed under reduced pressure and the crude product was
purified thrice by column chromatography (hexane/EtOAc), to
provide products as pale-brown oil.
ꢀ
knowledged for the R39 testing. Ir. Raoul Rozenberg, Dr. Cecile Le
ꢀ
Duff and Dr. Marie-France Herent have contributed to the structural
analysis.
4.3.1. (S)-tert-Butyl 2,13,16-trioxo-1,14-diazabicyclo[13.1.1]heptadec-
Supplementary data
7-ene-14-carboxylate (14c). Yield: 9% (43 mg from 1.22 mmol of
20
9c). Rf¼0.51 (hexane/EtOAc 7/3); [
a
]
ꢁ32.4 (c 2.8, CHCl3); 13C
D
The Supplementary data for this paper includes experimental
details for compounds 6e8, 18e21, structural data for compounds
6e8, 9a, be13a, b, 18c, 19b, ce21b, c, NMR spectra of all new
compounds, ESI and APCI mass spectra for compounds 14c, 15b,
ce17b, c, generation of conformers, graphics of relative energies of
precursors/cyclic monomers, heat of formation for precursors
9aec, 12aec and 13aec, and testing protocol versus R39. Supple-
mentary data related to this article can be found online at
doi:10.1016/j.tet.2010.10.015.
(125 MHz, CDCl3):
d
¼176.6, 172.6, 164.2, 151.3, 131.1, 130.6, 85.4,
59.7, 44.6, 38.3, 38.2, 32.1, 31.5, 30.5, 28.9, 28.1, 24.8, 22.7 ppm; IR:
n
¼2854e2976, 1799, 1738, 1697 cmꢁ1
;
HRMS: calcd for
C20H30N2O5Na 401.20469, found 401.20442.
4.3.2. Compound (15b). Yield: 44% (229 mg from 1.47 mmol of 9b).
Rf¼0.53 (hexane/EtOAc 3/2); [
a
]
20 ꢁ9.1 (c 3.0, CHCl3); 13C (75 MHz,
D
CDCl3):
d
¼175.7, 175.4, 171.4, 164.1, 163.9, 151.4, 151.3, 85.5, 85.1,
57.8, 57.5, 43.8, 43.7, 37.7, 37.4, 34.8, 31.9, 31.8, 31.7, 31.5, 28.0, 24.5,
22.6 ppm; IR:
for C36H52N4O10Na 723.35756, found 723.35643.
n
¼2853e2970, 1797, 1744, 1701 cmꢁ1; HRMS: calcd
References and notes
ꢀ
4.3.3. Compound (15c). Yield: 10% (46 mg from 1.22 mmol of 9c).
1. (a) Gradillas, A.; Perez-Castells, J. Angew. Chem., Int. Ed. 2006, 45, 6086e6101;
20
Rf¼0.45 (hexane/EtOAc 7/3);
[
a
]
þ5.8 (c 4.4, CHCl3); 13C
(b) Conrad, J. C.; Fogg, D. E. Curr. Org. Chem. 2006, 10, 185e202; (c) Majumdar,
K. C.; Rahaman, H.; Roy, B. Curr. Org. Chem. 2007, 11, 1339e1365; (d) Clavier, H.;
Grela, K.; Kirshning, A.; Mauduit, M.; Nolan, S. P. Angew. Chem., Int. Ed. 2007, 46,
6786e6801; (e) Nicolaou, K. C.; Bulger, P. G.; Sarlah, D. Angew. Chem., Int. Ed.
2005, 44, 4490e4527; (f) Brik, A. Adv. Synth. Catal. 2008, 350, 1661e1675.
2. (a) Monfette, S.; Fogg, D. E. Chem. Rev. 2009, 109, 3783e3816; (b) Monfette, S.;
Crane, A. K.; Duarte Silva, J. A.; Facey, G. A.; dos Santos, E. N.; Araujo, M. H.;
Fogg, D. E. Inorg. Chim. Acta 2010, 363, 481e486.
D
(125 MHz, CDCl3):
d
¼175.6, 175.5, 171.5, 171.4, 164.5, 164.4, 151.4,
151.3, 130.7, 130.6, 130.5, 130.3, 85.5, 57.4, 57.2, 43.9, 43.8, 38.4, 38.2,
36.5, 32.3, 32.2, 32.1, 28.8, 28.7, 28.0, 24.1, 24.0, 23.7, 23.4 ppm; IR:
n
¼2856e2928, 1799, 1742, 1697 cmꢁ1
;
HRMS: calcd for
C40H60N4O10Na 779.42017, found 779.41947.
3. (a) Creighton, C. J.; Leo, G. C.; Du, Y.; Reitz, A. B. Bioorg. Med. Chem. 2004, 12,
4375e4385; (b) Chen, G.; Kirschning, A. Chem.dEur. J. 2002, 8, 2717e2729.
4. (a) Tzur, E.; Ben-Asuly, A.; Diesendruck, C. E.; Goldberg, I.; Lemcoff, N. G. Angew.
Chem., Int. Ed. 2008, 47, 6422e6425; (b) Diesendruck, C. E.; Ben-Asuly, A.;
Goldberg, I.; Lemcoff, N. G. Chim. Oggi 2010, 28, 15e18.
5. Pandian, S.; Hillier, I. H.; Vincent, M. A.; Burton, N. A.; Ashworth, I. W.; Nelson,
D. J.; Percy, J. M.; Rinaudo, G. Chem. Phys. Lett. 2009, 476, 37e40.
6. (a) Urbach, A.; Dive, G.; Marchand-Brynaert, J. Eur. J. Org. Chem. 2009, 11,
1757e1770; (b) Urbach, A.; Dive, G.; Tinant, B.; Duval, V.; Marchand-Brynaert, J.
Eur. J. Med. Chem. 2009, 44, 2071e2080.
4.3.4. Compound (16b). Yield: 35% (30 mg from 0.34 mmol of 12b).
20
Rf¼0.32 (EtOAc); [
a
]
ꢁ15.6 (c 2.2, CHCl3); 13C (75 MHz, CDCl3):
D
d
¼174.4, 171.0, 165.8, 131.5, 130.8, 55.7, 46.0, 35.4, 34.5, 31.8, 31.6,
24.3, 22.0 ppm; IR:
n¼3284e3387, 2906e2930, 1784, 1693, 1676,
1655, 1527 cmꢁ1; HRMS: calcd for C26H36N4O6Na 523.25271, found
523.25171.
7. Miller, M. J.; Mattingly, P. G.; Morrison, M. A.; Kerwin, J. F., Jr. J. Am. Chem. Soc.
1980, 102, 7026e7032.
8. Qabar, M. N.; Kahn, M. Tetrahedron Lett. 1996, 37, 965e968.
4.3.5. Compound (16c). Yield: 28% (46 mg from 0.42 mmol of 12c).
20
Rf¼0.48 (EtOAc); [
a
]
þ13.3 (c 2.4, CHCl3); 13C (125 MHz, CDCl3):
D

A. Sliwa et al. / Tetrahedron 66 (2010) 9519e9527
9527
9. Structural description given by Mattingly, P. G.; Miller, M. J. J. Org. Chem. 1980,
45, 410e415.
10. Prashad, M.; Har, D.; Hu, B.; Kim, H.-Y.; Repic, O.; Blacklock, T. J. Org. Lett. 2003,
5, 125e128.
Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.;
Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe,
M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A.
Gaussian 03, Revision D.02; Gaussian: Wallingford CT, 2004.
11. (a) Ahrendt, K. A.; Williams, R. M. Org. Lett. 2004, 24, 4539e4541; (b) Gaul, C.;
Njardarson, J. T.; Shan, D.; Dorn, D. C.; Wu, K.; Tong, W. P.; Huang, X.; Moore, M.
A. S.; Danishefsky, S. J. J. Am. Chem. Soc. 2004, 126, 11326e11337.
€
12. Furstner, A.; Langemann, K. J. Am. Chem. Soc. 1997, 119, 9130e9136.
17. Dunning, T. H., Jr.; Hay, P. J. In Modern Theoretical Chemistry; Schaefer, H. F., III,
Ed.; PLENUM: New York, NY, 1976; Vol. 3, pp 1e28.
13. Yamamoto, K.; Biswas, K.; Gaul, C.; Danishefsky, S. Tetrahedron Lett. 2003, 44,
3297e3299.
ꢁ
18. Sauvage, E.; Herman, R.; Petrella, S.; Duez, C.; Bouillenne, F.; Frere, J.-M.;
14. (a) Knops, P.; Sendhoff, N.; Mekelburger, B.; Voegtle, F. Top. Curr. Chem. 1991,
161, 1e36; (b) Illumani, G.; Mandolini, L. Acc. Chem. Res. 1981, 14, 95e102.
15. Francl, M. M.; Pietro, W. J.; Hehre, W. J.; Binkley, J. S.; DeFrees, D. J.; Pople, J. A.;
Gordon, M. S. J. Chem. Phys. 1982, 77, 3654e3666.
Charlier, P. J. Biol. Chem. 2005, 280, 31249e31256.
19. (a) Schwyzer, R.; Hurlimann, C.; Coenzyle, C. Helv. Chim. Acta 1954, 37, 155e166;
ꢁ
(b) Adam, M.; Damblon, C.; Plaitin, B.; Christiaens, L.; Frere, J.-M. Biochem. J.
1990, 270, 525e529.
16. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.;
Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.;
Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.;
Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo,
C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi,
R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.;
20. Lakaye, B.; Damblon, C.; Jamin, M.; Galleni, M.; Lepage, S.; Joris, B.; Marchand-
ꢁ
Brynaert, J.; Frydrych, C.; Frere, J.-M. Biochem. J. 1994, 300, 141e145.
21. Zervosen, A.; Lu, W.-P.; Chen, Z.; White, R. E.; Demuth, T. P., Jr.; Frere, J.-M.
ꢁ
Antimicrob. Agents Chemother. 2004, 48, 961e969.
22. (a) Woodward, R. B. Phil. Trans. R. Soc. Lond. (B) 1980, 289, 239e250; (b) Cohen,
N. C. J. Med. Chem. 1983, 26, 259e264; (c) Bruton, G. Bioorg. Med. Chem. Lett.
1993, 3, 2329e2332.
23. El-azizi, Y.; Schmitzer, A.; Collins, S. K. Angew. Chem., Int. Ed. 2006, 45, 968e973.