Unusual peptides containing the 2,6-diaminopimelic acid framework: Stereocontrolled synthesis, X-ray analysis, and computational modelling. Part 2

Article abstract of DOI:10.1016/S0957-4166(03)00541-X

The stereocontrolled synthesis of peptides 6, 9 and 14, structural variants of 2,6-diaminopimelic acid, was carried out starting from the chiral synthon 1, easily obtained from L-valine. The configuration of the introduced stereogenic centres has been ass

Full text of DOI:10.1016/S0957-4166(03)00541-X

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 14 (2003) 2639–2649
Unusual peptides containing the 2,6-diaminopimelic acid
framework: Stereocontrolled synthesis, X-ray analysis, and
computational modelling. Part 2ꢀ
R. Galeazzi,a M. Garavelli,b A. Grandi,b M. Monari,b G. Porzib,* and S. Sandrib,*
aDipartimento di Scienze dei Materiali e della Terra, Universita` di Ancona, Via Brecce Bianche, 60131 Ancona, Italy
bDipartimento di Chimica ‘G. Ciamician’, Universita` di Bologna, Via Selmi 2, 40126 Bologna, Italy
Received 11 June 2003; accepted 7 July 2003
Abstract—The stereocontrolled synthesis of peptides 6, 9 and 14, structural variants of 2,6-diaminopimelic acid, was carried out
starting from the chiral synthon 1, easily obtained from L-valine. The configuration of the introduced stereogenic centres has been
1
assigned on the basis of H NMR spectroscopic data. X-Ray crystal structure and conformation analysis of 5 are also reported.
© 2003 Elsevier Ltd. All rights reserved.
1. Introduction
exhibit a wide range of biological activity (for instance,
immunomodulators, antitumors, antibiotics).
In a recent paper1 we described a new stereoselective
approach to unusual tripeptides C-termini at both ends
of the chain, containing the 2,6-diaminoipimelic acid
(2,6-DAP) framework. We are interested in these struc-
tural derivatives of 2,6-DAP, which is biosynthesized in
bacteria and higher plants, for their potential antibacte-
rial and herbicide activity.2 In fact, both (2R,6R)-DAP
and (2R,6S)-DAP are involved in the biosynthetic con-
This study has been complemented by X-ray analysis
and extended molecular modelling conformational
searches in solution and DFT computational
investigations.
2. Synthesis and stereochemical assignments
version of pyruvate and
-lysine, necessary for the growth of Gram-(+) and
many Gram-(−) bacteria. Both the -lysine and
L-aspartate to the amino acid
L
Herein we have followed a strategy which makes use of
L
the chiral synthon 1, a mono-lactim ether easily synthe-
(2R,6S)-DAP function as cross-linking constituents in
the cell wall peptidoglycan of bacteria.3 The structural
analogues of 2,6-DAP can then function as inhibitors
of biosynthetic formation or metabolism of this com-
pound offering promising biological activity as antibac-
terial and herbicide agents. Furthermore, it is
interesting to note that some tri- and tetrapeptides
incorporating the 2,6-DAP skeleton conjugated with
lauric or palmitic acid show biological activity as
immuno-adjuvants.4 Therefore, we undertook the
asymmetric synthesis of peptides more complex than
the 2,6-DAP derivatives previously described,1 i.e. con-
taining a proline residue fused to a diketopiperazine
ring. The reason for our interest in the stereocontrolled
synthesis of such peptide-like structures is because some
natural products, containing a 6,5-fused ring system,5
sized from
-valine.1 Deprotonation of 1 with LHMDS
L
at −78°C, followed by alkylation with 0.5 equiv. of
1,3-diiodopropane, afforded, in 87% yield, the
diastereomer 2 which was purified by silica gel chro-
matography, as previously described.1 The largely
prevalent diastereomer 2 was then converted into the
corresponding enolate and then alkylated with 1,3-
dichloropropane in good yield with a total 1,4-trans
induction (>98%) with respect to the isopropyl group,
as previously observed for similar sustrates.1,6 By sub-
mitting the intermediate 3 to the Finkelstein reaction
(NaI in refluxing acetone), the bicyclic derivative 4 was
also isolated in good yield (Scheme 1).
The removal of the benzyl group from the intermediate
4 was achieved through the Birch reduction. However,
by performing the reaction under the usual conditions
(i.e. the addition of the substrate to an excess of Li
ꢀ For Part 1, see Reference 1.
* Corresponding authors. E-mail: gianni.porzi@unibo.it
0957-4166/$ - see front matter © 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0957-4166(03)00541-X

2640
R. Galeazzi et al. / Tetrahedron: Asymmetry 14 (2003) 2639–2649
The intermediate 4, alkylated with iodomethane or
benzylbromide, gave 7 in a good yield and with total
regio- and stereoselectivity (Scheme 2). The 1,4-trans-
induction was demonstrated by comparing the 1H
NMR spectra of derivatives 7a and 7b, on the basis of
the upfield shift induced on the H-6¦ by the shielding
effect of the phenyl ring of the benzyl group introduced
on C-3¦. Indeed, as already observed in analogous
substrates,7 the phenyl ring preferentially lies internal to
the heterocyclic monolactim ether (i.e. the preferred
‘aryl inside’ arrangement is adopted) causing significant
shielding on the cis-substituent at C-6¦. This shielding
of about 0.7 ppm seen for 7b, with respect to the
derivative 7a, provided clear evidence of the cis-rela-
tionship between (C-3¦)-CH2Ph and (C-6¦)-H. The
absolute configuration of C-3¦ follows on from the note
(S)-configuration of the stereocentre at C-6¦.
dissolved in liquid ammonia) the reaction gave in large
prevalence compound 5%. After some attempts complete
transformation to product 5 was achieved by changing
the usual Birch reaction methodology. The modifica-
tion consisted of the addition of only 1 equiv. of Li to
a solution of the substrate. It is noteworthy that even a
slight excess of Li caused the formation of 5% due to the
further reduction of CꢀO group followed by the base
catalysed dehydration (Scheme 1).
The proposed structure of the unexpected compound 5%
1
was ascertained by means of 13C NMR, H NMR and
HPLC–MS spectra. Indeed, the following analytical
data was available: a) In the low field region of the 13C
NMR spectrum, four signals were observed (158.5,
161.2, 167.5 and 171.8 ppm) attributable to either a
CꢀO or a CꢀN group. The Attached Proton Test (APT)
also showed that the signal at 161.2 ppm is due to a
carbon atom bonded to an odd number of protons; b)
The monolactim ethers 7a,b were submitted to the
debenzylation affording 8a,b in good yields. The final
products 9a,b were then obtained after acid cleavage in
mild condition. It is noteworthy that while 9a was
isolated in good yield, the acid hydrolysis of 8b fur-
nished 9b together with a not too negligible amount of
9%b (Scheme 2).
1
the H NMR showed a doublet proton signal at 7.9
ppm, split (J=1.8 Hz) by the H-6% resonating at 3.99
ppm with the small J value being attributable to an
allylic coupling constant; c) the HPLC–MS exhibited a
molecular ion peak at mass 404, i.e. 16 unities lesser
than the molecular ion showed by 5. All these findings
are consistent with the proposed structure of 5%.
Diastereomeric derivatives with an opposite configura-
tion at the stereocentre C-3%, in comparison to the
compounds described above, were obtained by using
the following alternative strategy. While still starting
from synthon 1, the protocol consisted of the formation
of the fused ring 11, followed by the alkylation with
The acid cleavage of 5, performed under mild condi-
tions and in good yield, afforded the final product 6
which contained a bicyclic moiety analogous to that
synthesized by Davies et al.5 starting from a bis-lactim
ethylether (Schollkopf’s auxiliary).
Scheme 1. Reagents and conditions: (i) LHMDS, THF, at −78°C, then Cl-(CH2)3-Cl; (ii) NaI in refluxing acetone; (iii) Birch
reduction by using a large excess of Li; (iv) Birch reduction by adding an equimolar amount of Li; (v) aqueous HCl 0.5 M, EtOH
at room temperature.

R. Galeazzi et al. / Tetrahedron: Asymmetry 14 (2003) 2639–2649
2641
Scheme 2. Reagents and conditions: (i) LHMDS, THF, at −78°C, then CH3I or PhCH2Br; (ii) Birch reduction using an equimolar
amount of Li; (iii) aqueous HCl 0.5N, EtOH at room temperature.
Scheme 3. Reagents and conditions: (i) LHMDS, THF, at −78°C, then I-(CH2)3-Cl; (ii) NaI in refluxing acetone; (iii) Li-enolate
of 1, THF, at −78°C.
3-chloroiodopropane, which furnished 12. This interme-
diate, after conversion into 13, was reacted with the
Li-enolate of 1 to give 14, it being the epimer of 4 at the
C-3% stereocentre (Scheme 3).
that the anion 11%A was 1.5 kcal/mol more stable than
the 11%B one (Fig. 1).
1
3. H NMR and IR studies
The alkylation of bicyclic system 11 occurred with total
regio- and diastereoselectivity giving the intermediate
12 in a good yield. The observed regioselection was due
to the greater acidity of the (C-3)-H proton in compari-
son to the (C-6)-H one. In fact, computational investi-
gation,8 performed on the simple model 11%, showed
Compound 5 showed a linear relationship between the
chemical shift of the amide protons (lNH) and the
temperature: the temperature coefficients (DlNH/DT)
calculated for both amide protons in 1 mM CDCl3
solution were substantially coincident, i.e. −12.8 and
Figure 1.

2642
R. Galeazzi et al. / Tetrahedron: Asymmetry 14 (2003) 2639–2649
−12.45 ppb/°C (see Fig. 2). The large temperature
coefficients, the chemical shift values lower than 7 ppm
for both amide protons in diluted solution at rt, and the
decrease in chemical shifts observed from going from
concentrated (30 mM, lNH=8.38 and 8.42 ppm) to
diluted (1 mM, lNH=6.1 and 6.17 ppm) CDCl3 solu-
tions all suggested the existence of an equilibrium
between a non hydrogen bonded and a hydrogen
bonded state.9 Thus, taking into account all the 1H
NMR parameters, it is reasonable to assume that the
intermolecular aggregation, involving the amide proton
and the carbonyl oxygen, is preferred to the intramolec-
ular hydrogen bond which could produce a ten-mem-
bered ring, i.e. a non-peptide mimetic b-turn structure.
The absence of an intramolecular hydrogen bond was
supported by the IR spectra of 5 in CHCl3. In fact, a 30
mM solution displayed two different adsorptions for
the NH stretching, indicating an equilibrium between a
non hydrogen bonded and a hydrogen bonded struc-
ture: a narrow band at 3394 cm−1, typical of a free
hydrogen bond structure, and a broad band at 3197
cm−1 attributable to a hydrogen bonded stretching
vibration. As the concentration was reduced to 3 mM,
the band at 3197 cm−1 disappeared, thus showing that
the hydrogen bonding was intermolecular rather than
intramolecular.9
4. X-Ray analysis
The solid state molecular structure of 5 is shown in
Figure 3 with the relevant bond lengths reported in
Table 1. In this structure the diketopiperazine ring
adopts an almost planar conformation (maximum devi-
,
ation from planarity ca. 0.1 A) which is different to
what has been observed, for example, in the dipeptide
cyclo(L-Phe-L
-Pro)10 in which the diketopiperazine ring
is in a boat conformation, as it has been found in the
majority of these heterocyclic systems. The proline ring
adopts a slightly distorted envelope conformation. The
main feature of compound 5 is the orientation of the
six-membered rings that are facing each other being
almost parallel (the angle between the rings being ca.
7°) despite the great degree of flexibility of the aliphatic
bridge. The molecular conformation seems dictated by
the minimisation of the molecular volume. The absolute
configuration of the stereogenic centres at C(4) and
C(19) is S, while at C(9) and C(13), it is R (see Section
2).
Figure 2. Linear relationship of the chemical shift of one
amide proton (lNH) versus the temperature.
In the crystal packing each molecule is engaged in four
intermolecular NꢁH···O hydrogen bonds N(1)ꢁH(1N)
0.86(2), H(1N)···O(4) 2.06(2), N(1)···O(4) 2.907(2),
N(4)ꢁH(4N) 0.84(2), H(4N)···O(1) 2.04(2), N(1)···O(4)
,
2.871(2) A, using the amidic hydrogen and the oxygen
of the CO group (Fig. 3).
,
Table 1. Comparison between bond lengths (A) of 5
obtained from the X-ray study and calculated (in square
bracket)
N(1)ꢁC(8)
N(1)ꢁC(4)
C(3)ꢁC(4)
C(4)ꢁC(5)
N(2)ꢁC(3)
N(2)ꢁC(9)
C(8)ꢁC(9)
C(8)ꢁO(1)
C(3)ꢁO(2)
O(2)ꢁC(2)
C(2)ꢁC(1)
C(13)ꢁC(14)
C(12)ꢁC(13)
C(11)ꢁC(12)
C(5)ꢁC(7)
C(20)ꢁC(21)
1.333(2) [1.37] C(13)ꢁN(3)
1.456(2) [1.46] N(3)ꢁC(18)
1.502(2) [1.52] C(18)ꢁO(3)
1.548(3) [1.56] C(18)ꢁC(19)
1.252(2) [1.27] C(19)ꢁN(4)
1.457(2) [1.46] C(17)ꢁN(4)
1.513(2) [1.53] C(17)ꢁO(4)
1.242(2) [1.23] C(17)ꢁC(13)
1.357(2) [1.35] C(16)ꢁN(3)
1.437(3) [1.44] C(15)ꢁC(16)
1.470(3) [1.52] C(14)ꢁC(15)
1.545(3) [1.55] C(10)ꢁC(11)
1.549(2) [1.55] C(9)ꢁC(10)
1.510(2) [1.53] C(5)ꢁC(6)
1.537(3) [1.54] C(19)ꢁC(20)
1.508(3) [1.54] C(20)ꢁC(22)
1.459(2) [1.48]
1.338(2) [1.36]
1.227(2) [1.23]
1.523(3) [1.53]
1.460(2) [1.46]
1.329(2) [1.36]
1.235(2) [1.23]
1.507(2) [1.53]
1.470(3) [1.47]
1.475(4) [1.55]
1.499(4) [1.54]
1.531(2) [1.53]
1.530(3) [1.54]
1.506(4) [1.54]
1.554(3) [1.56]
1.511(3) [1.54]
Figure 3. ORTEP drawing of 5 (thermal ellipsoids at 30%
probability) showing the intermolecular NꢁH···O hydrogen
bonds that each molecule establishes with two neighbours.

R. Galeazzi et al. / Tetrahedron: Asymmetry 14 (2003) 2639–2649
2643
5. Molecular modelling: conformational analysis
of a classical b or g turn. For these reasons, all these
structures can be viewed as open turns. The most
significant conformers found for compound 5 (i.e. all
the conformers within DE=1.5 kcal/mol) are reported
in Table 2. Among these the conformers 1, 3 and 10
represent open turns.
A complete, extensive, unconstrained conformational
analysis of compound 5 was performed by using an
AMBER* force field11 and the Monte Carlo12 confor-
mational search (MC/EM) varying all the degrees of
freedom, including CHCl3 as the solvent. A relatively
non-polar solvent was chosen in order not to interfere
with the hydrogen bonding of the amide protons. All
the conformers within the energy gap of 6 kcal/mol
were kept, and subsequently only those that lie below
3.6 kcal/mol were fully analysed (a total number of 123
conformers). The turn propensity of the minimum
energy conformations was assessed by computing and
analyzing geometric parameters, specifically the dis-
tance between the capping groups on the N- and C-ter-
The low energy conformers were then compared to the
X-ray structures observed. Most remarkably, the latter
is very close to the modelled lowest energy structure
(conformer 1 in Table 2), thus confirming the preference
for this molecule to intermolecular hydrogen bonding.
The geometry of this conformer was then refined via
optimisation at the B3LYP/6-31G* level of theory in
order to obtain a more accurate structure. The similar-
ity with the observed X-ray structure is evident by the
superposition plot of the calculated and X-ray deter-
mined geometry of 5 depicted in Figure 4 and by
comparison of the bond distances (Table 1) in the
calculated model and in the solid state structure. This
agreement validates the joint MM/DFT computational
approach used here for conformational investigation.
mini
(da),
the
virtual
torsion
angles
i
(CaiꢁCai+1ꢁCai+2ꢁNi+3), and all the parameters indicative
,
of hydrogen bonding (b- or g-turns involve da <7 A and
−60°<i<60°). Only 23 conformers (22.7%) satisfied
these structural requisites, but none of them showed any
evidence of intramolecular hydrogen bonds indicative
Table 2. Low energy conformations for compound 5
,
,
,
Conformer
E (kcal/mol)
Populationa (%)
da(1-4) (A)
da(1-3) (A)
da(2-4) (A)
i
i1
i2
−23.6
38.2
−24.9
−73.6
38.6
38.6
−143.2
73.0
−73.4
−21.8
140.0
25.9
−108.8
22.2
112.7
112.7
−71.0
−89.0
93.4
1
2
3
4
5
6
7
8
0.00
0.40
0.43
0.60
0.69
0.80
0.92
1.01
1.04
1.06
1.08
1.08
1.10
1.10
1.17
1.18
1.21
1.25
1.27
1.29
15.7
7.99
7.59
5.70
4.89
4.06
3.31
2.84
2.71
2.62
2.53
2.53
2.45
2.45
2.18
2.18
2.04
1.90
1.84
1.77
5.513
7.739
6.216
9.048
8.320
9.820
8.605
8.586
9.287
5.127
10.062
7.155
8.322
7.463
8.825
9.221
5.097
8.031
7.310
8.540
5.369
5.750
6.612
6.964
7.737
8.977
6.973
7.035
7.180
5.625
7.229
5.563
6.175
5.892
7.284
6.452
5.122
6.903
6.201
6.435
6.446
7.895
6.080
7.747
6.139
7.774
6.678
6.724
7.758
5.676
7.789
6.782
7.030
6.793
6.747
7.173
5.832
6.219
5.256
7.062
−53.2
36.4
136.6
35.7
47.8
−54.1
−64.0
48.1
148.1
−171.8
43.0
−60.7
−53.6
148.2
67.3
−61.0
62.6
22.0
137.0
−122.7
−81.1
133.6
25.7
9
10
11
12
13
14
15
16
17
18
19
20
36.2
−121.9
57.2
−86.2
54.0
101.6
101.6
−15.8
−52.3
153.6
−74.8
42.8
169.8
−102.5
−46.7
62.8
−55.6
−103.9
a The population percentage was calculated according to the Boltzamn distribution pop2/pop1=e−DE/RT considering all the conformers which lie
under 2.0 kcal/mol at T=298 K.

2644
R. Galeazzi et al. / Tetrahedron: Asymmetry 14 (2003) 2639–2649
Figure 4. Superimposition of the DFT/optimized lowest energy conformer (dotted line) and X-ray (solid line) structure of 5.
Table 3. Calculated and experimental (CDCl3 solution) chemical shifts (l in ppm) for the isolated monomer and the X-ray
resolved trimer 5. For atom labels, see Figure 3
Protons bonded to
Monomer
exper.a
Trimer
calcd
calcd
exper.b
C1c
1.25
4.08
3.86
1.3
4.15
3.88
1.21
4.03
3.93
1.27
4.1
3.78
C2c
C4c
C5 and C20
C6,C7,C21,C22c
C9
2.25 and 1.94
0.95–1.04
4.08
1.33–1.90
3.59
2.2
0.87–1.12
4.03
1.4–2.0
3.7
2.23 and 1.74
1.12–1.41
4.09
1.37–1.81
3.57
2.2
0.82–1.13
4.03
1.5–2.1
3.68
C10,C11,C12,C14,C15c
C16c
C19
3.88
3.66
3.15
3.61
N1 and N4
4.23 and 4.37
6.1 and 6.17
9.01 and 8.33
8.38 and 8.42
a 1 mM CDCl3 solution.
b 30 mM CDCl3 solution.
c Mean value for CH3 or CH2 protons.
To validate further the computed data and structures
and provide a solid ground for the interpretation of the
experiments, H NMR chemical shift values were also
On the other hand, the monomer displays much higher-
field values for lNH (4.37 and 4.23 ppm) due to the lack
of hydrogen bond interactions, which deshield the NH
proton’s signals. The same trend (i.e. decreasing in the
lNH values) has been observed experimentally on going
from concentrated (30 mM, lNH :8.4 ppm) to dilute (1
mM, lNH :6.1 ppm) solutions (see Section 3), support-
ing a loosening in intermolecular hydrogen bond inter-
actions or, alternatively, suggesting an equilibrium
between a hydrogen bonded and a non-hydrogen
bonded structure, which shifts progressively towards
the latter state as the solution is made more and more
dilute. This hypothesis is in agreement with the IR
spectra (see Section 3).
1
simulated13,14 for the B3LYP/6-31G* optimized struc-
ture of compound 5 (i.e. the lowest energy conformer)
and compared to those observed in CDCl3 solution at
different concentrations. Proton chemical shifts were
computed for both the isolated monomer (which ideal-
istically represents the situation in dilute solution) and
the trimer as resolved by X-ray crystallography (see
Fig. 3, which approximately models the solid-phase
environment or concentrate solution).15 This data are
reported in Table 3. Interestingly, the computed chemi-
cal shifts are in good agreement with the experimental
chemical shift. In particular, the lNH values for the
trimer show anomalous low-field values (8.33 and 9.01
ppm) in agreement with the observed ones (:8.4 ppm),
supporting the hypothesis for intermolecular aggrega-
tion in concentrated solution with the absence of
intramolecular hydrogen bonds.
6. Conclusion
We have synthesized new optically active peptides con-
taining the 2,6-DAP skeleton and a system based on a

R. Galeazzi et al. / Tetrahedron: Asymmetry 14 (2003) 2639–2649
2645
proline residue fused to a diketopiperazine ring. This
approach, accomplished by starting from -valine,
l: 0.88 (d, 3H, J=7); 0.92 (d, 3H, J=7); 1.05 (d, 3H,
J=6.8); 1.08 (d, 3H, J=6.8); 1.22 (t, 3H, J=7.2); 1.25
(t, 3H, J=7.2); 1.5–2.3 (m, 12H); 3.45 (m, 2H); 3.67
(dd, 1H, J=1.6, 3.8); 3.76 (d, 1H, J=3); 3.9 (d, 1H,
J=15); 3.98 (d, 1H, J=15); 3.95–4.2 (m, 5H); 5.46 (d,
1H, J=15); 5.48 (d, 1H, J=15); 7.3 (m, 10ArH). 13C
NMR l: 14.1, 17.3, 17.5, 19.6, 19.9, 20.7, 28.0, 29.9,
31.5, 34.1, 37.6, 41.2, 44.9, 46.9, 47.1, 57.5, 60.6, 60.9,
61.1, 61.7, 62.9, 127.3, 127.4, 127.7, 128.2, 128.5, 136.1,
136.2, 156.0, 158.6, 170.3, 171.6. Anal. calcd for
C38H53N4O4Cl: C, 68.6; H, 8.03; N, 8.42; Cl, 5.33.
Found: C, 68.85; H, 8.05; N, 8.4; Cl, 5.31.
L
might also represent a new and simple synthetic path to
brevianamides which are fungal metabolites, many of
which show interesting biological activities.16 The struc-
tures of these natural products incorporate a bicyclic
system containing the diketopiperazine ring, already
accomplished by us,17 and diketopiperazine proline
fused rings described herein. In the crystal state the
compound 5 appears to be aggregated by the formation
of intermolecular hydrogen bonds between the carbonyl
oxygen and the amide proton. The computational con-
formational analysis is in good agreement with the
1
X-ray analysis and also the H NMR experiments in
The doubly alkylated 1,3-bis[(3%R,6%S)-1%-benzyl-3%-(3-
chloropropyl)-6%-hydro-5%-ethoxy-6%-isopropyl-2%-pirazi-
non-3%-yl]propane was isolated as a by-product in about
CDCl3 (lNH and DlNH/DT values) and IR spectra are
coherent with the formation of an intermolecular
hydrogen bond.
1
10% yield. H NMR l: 0.91 (d, 6H, J=7); 1.09 (d, 6H,
J=6.8); 1.26 (t, 6H, J=7.4); 1.4–2.3 (m, 16H); 3.4 (m,
4H); 3.77 (d, 2H, J=3); 3.9–4.2 (m, 6H); 5.43 (d, 2H,
J=15); 7.3 (m, 10ArH). 13C NMR l: 13.9, 17.0, 18.0,
20.5, 26.6, 27.7, 29.6, 37.7, 41.6, 44.6, 46.7, 60.4, 61.0,
62.7, 127.2, 128.0, 128.3, 135.8, 155.9, 171.2.
We conclude that the experimental data (1H NMR
analysis, IR spectra, X-ray experiments), when comple-
mented and supported by computational modelling,
represents a valid tool to elucidate the structure and the
intermolecular interactions of these molecules.
7.3. 1-[(3%S,6%R)-4%-Benzyl-2%,5%-diketo-3%-isopropyl-1%,4%-
diazabicyclo[4,3,0]-6%-nonyl]-3-[(3¦S,6¦R)-4¦-benzyl-2¦-
ethoxy-3¦,6¦-dihydro-3¦-isopropyl-5¦-pirazinon-6¦-yl]-
propane, 4
7. Experimental
7.1. General information
A solution of the intermediate 3 (13.3 g, 20 mmol) and
NaI (6 g, 40 mmol), dissolved in 50 mL of acetone, was
refluxed for 36 hours. The organic solvent was evapo-
rated, water added to the residue and the reaction
product extracted with ethylacetate. The organic phase,
having been dried over Na2SO4, was evaporated in
vacuo to dryness and the residue purified by silica gel
chromatography eluting with hexane/ethylacetate. The
product was obtained pure in practically quantitative
Melting points are uncorrected. 1H and 13C NMR
spectra were recorded on a Gemini spectrometer at 300
MHz using CDCl3 as the solvent, unless otherwise
stated. Chemical shifts are reported in ppm relative to
CDCl3. The coupling constants (J) are in Hz. IR spec-
tra were recorded on a Nicolet 210 spectrometer. Opti-
cal rotation values were measured at 25°C on a
Perkin–Elmer 343 polarimeter. HPLC–MS spectra were
recorded on a HP LC-MSD 1100 single quadrupole
(with interface APCI-ES) spectrometer.
1
yield. [h]D=+44.6 (c 2.6, CHCl3); H NMR l: 0.92 (d,
3H, J=6.8); 1.06 (d, 3H, J=6.8); 1.11 (d, 3H, J=6.8);
1.18 (d, 3H, J=6.8); 1.26 (t, 3H, J=7); 1.6–2.4 (m,
12H); 3.4 (m, 1H); 3.6 (d, 1H, J=7.6); 3.69 (dd, 1H,
J=1.6, 4.2); 3.92 (d, 1H, J=15); 3.93 (d, 1H, J=15);
3.95–4.2 (m, 4H); 5.45 (d, 1H, J=15); 5.46 (d, 1H,
J=15); 7.3 (m, 10ArH). 13C NMR l: 14.0, 17.4, 19.8,
19.9, 20.2, 20.5, 20.9, 31.5, 33.1, 33.7, 34.7, 38.2, 45.0,
47.3, 49.2, 57.3, 60.9, 61.9, 66.9, 67.1, 127.3, 127.4,
127.9, 128.5, 135.9, 136.0, 159.0, 164.7, 169.7, 170.0.
Anal. calcd for C36H48N4O4: C, 71.97; H, 8.05; N, 9.33.
Found: C, 71.9; H, 8.08; N, 9.36.
Dry THF was distilled from sodium benzophenone
ketyl. Chromatographic separations were performed
with silica gel 60 (230–400 mesh).
7.2. 1-[(3%R,6%S)-1%-Benzyl-3%-(3-chloropropyl)-6%-hydro-
5%-ethoxy-6%-isopropyl-2-pirazinon-3%-yl]-3-[(3’’R,6’’S)-
1’’-benzyl-3’’,6’’-dihydro-5’’-ethoxy-6’’-isopropyl-2¦-
pirazinon-3’’-yl]propane, 3
To a solution of 21 (5.9 g, 10 mmol) in dry THF (250
mL) cooled at −78°C was added 10 mmol of LHMDS
(1 M solution in THF). After about one hour under
stirring, 1-chloro-3-iodopropane (1.3 mL, 12 mmol)
dissolved in 50 mL of dry THF was added and the
reaction monitored by TLC. When the reaction was
complete, water and ethylacetate was then added. The
organic extract was dried over Na2SO4 and then evapo-
rated to dryness in vacuo. The crude reaction product
was purified by silica gel chromatography eluting with
hexane/ethylacetate with the pure product isolated in
7.4. Birch reduction of 4
7.4.1. Usual procedure. Li (0.35 gr, 50 mmol) was
dissolved in liquid ammonia (80 mL) cooled at −50°C
under strirring. After one hour, the substrate 4 (2.1 g,
3.5 mmol), dissolved in 30 mL of dry THF/t-butanol
9:1, was added and the reaction mixture stirred for 5
minutes. The reaction was then quenced by the addition
of NH4Cl (2.7 gr, 50 mmol). After evaporation of the
ammonia, to the crude reaction product was added
water and ethylacetate. The organic extract was dried
over Na2SO4 and evaporated under vacuum to dryness.
1
about 80% yield. [h]D=+2.8 (c 0.7, CHCl3); H NMR

2646
R. Galeazzi et al. / Tetrahedron: Asymmetry 14 (2003) 2639–2649
7.4.2. Modified procedure. A solution of the substrate 4
(2.1 g, 3.5 mmol) in 30 mL of dry THF/t-butanol 9:1
was added to about 80 mL of liquid ammonia cooled at
−50°C. Li (0.05 gr, 7 mmol) was then added to the
substrate under stirring. The addition of Li in small
pieces, was controlled by the monitoring of the TLC in
the presence of the substrate (starting material) and was
stopped as soon as the reaction mixture became blue.
The reaction was then rapidly quenced by the addition
of NH4Cl (0.43 gr, 8 mmol) and, after evaporation of
ammonia, the residue was worked up as reported.
19.7, 19.9, 20.2, 20.5, 21.3, 31.7, 32.9, 34.5, 34.9, 38.9,
46.3, 54.1, 59.3, 62.3, 64.0, 68.1, 167.3, 170.1, 172.4.
Anal. calcd for C22H39N4O5Cl: C, 55.63; H, 8.28; N,
11.79; Cl, 7.46. Found: C, 55.44; H, 8.25; N, 11.77; Cl,
7.44.
7.8. 1-[(3%S,6%R)-4%-Benzyl-2%,5%-diketo-3%-isopropyl-1%,4%-
diazabicyclo[4,3,0]non-6%-yl]-3-[(3¦S,6¦S)-4¦-benzyl-2¦-
ethoxy-3¦-hydro-6¦-methyl-3¦-isopropyl-5¦-pirazinon-6¦-
yl]propane, 7a
Compound 7a was obtained in about 90% yield by
alkylating 4 with methyl iodide and following the pro-
cedure reported in Section 7.2. [h]D=−27.6 (c 0.7,
7.5. 1-[(3%S,6%R)-2%,5%-Diketo-3%-isopropyl-1%,4%-diazabicy-
clo[4,3,0]-6%-nonyl]-3-[(3¦S,6¦R)-2¦-ethoxy-3¦,6¦-dihydro-
3¦-isopropyl-5¦-pirazinon-6¦-yl]propane, 5
1
CHCl3); H NMR l: 0.91 (d, 3H, J=6.6); 1.07 (d, 3H,
J=6.6); 1.11 (d, 3H, J=6.6); 1.18 (d, 3H, J=6.6); 1.26
(t, 3H, J=7.4); 1.41 (s, 3H); 1.5–2.4 (m, 12H); 3.4 (m,
1H); 3.6 (d, 2H, J=8); 3.75 (d, 1H, J=2.6); 3.9–4.2 (m,
5H); 5.44 (d, 1H, J=15); 5.49 (d, 1H, J=15); 7.3 (m,
10ArH). 13C NMR l: 14.0, 17.0, 19.4, 19.9, 20.3, 20.4,
20.8, 29.3, 29.6, 33.0, 35.0, 39.0, 41.2, 45.2, 46.5, 49.2,
60.0, 60.4, 61.0, 66.9, 67.2, 127.5, 127.8, 127.4, 128.5,
128.6, 136.0, 136.2, 155.1, 165.0, 169.8, 172.7. Anal.
calcd for C37H50N4O4: C, 72.28; H, 8.2; N, 9.11. Found:
C, 72.02; H, 8.18; N, 9.15.
Compound 5 was obtained with a 90% yield submitting
the intermediate 4 to the modified Birch reaction. The
product was a white solid melting at 169–171°C. In
Section 7.19 is reported the X-ray resolved structure.
1H NMR l: 0.82 (d, 3H, J=6.6); 0.98 (d, 3H, J=6.6);
0.99 (d, 3H, J=6.6); 1.13 (d, 3H, J=7); 1.27 (t, 3H,
J=7.2); 1.5–2.3 (m, 12H); 3.3 (m, 1H); 3.61 (dd, 1H,
J=3.4, 8); 3.78 (m, 1H); 4–4.2 (m, 4H); 8.38 (d, 1H,
J=2); 8.42 (d, 1H, J=3). 13C NMR l: 14.3, 16.0, 18.4,
19.1, 19.5, 19.9, 20.1, 32.6, 33.7, 33.9, 36.0, 36.6, 44.1,
57.1, 58.3, 61.2, 62.6, 67.5, 158.8, 164.9, 171.5, 172.4.
[h]D=130.3 (c 1.1, CHCl3).
7.9. 1-[(3%S,6%R)-4%-Benzyl-2%,5%-diketo-3%-isopropyl-1%,4%-
diazabicyclo[4,3,0]non-6%-yl]-3-[(3¦S,6¦R)-4¦,6¦-dibenzyl-
2¦-ethoxy-3¦-hydro-3¦-isopropyl-5¦-pirazinon-6¦-ylone]-
propane, 7b
7.6. 1-[(3%S,6%R)-2%-Keto-3%-isopropyl-1%,4%-diazabicy-
clo[4,3,0]-4%-nonen-6%-yl]-3-[(3¦S,6¦R)-2¦-ethoxy-3¦,6¦-
dihydro-3¦-isopropyl-5¦-pirazinon-6¦-yl]propane, 5%
Compound 7b was obtained in about 90% yield by
alkylating 4 with benzylbromide and following the pro-
cedure reported in Section 7.2. [h]D=−40.2 (c 1.0,
1
CHCl3); H NMR l: 0.82 (d, 3H, J=6.8); 0.91 (d, 3H,
Compound 5% was obtained in a practically quantitative
yield starting from 4 and following the usual Birch
procedure. [h]D=+78.6 (c 0.9, CHCl3); 1H NMR l:
0.86 (d, 3H, J=6.6); 1.11 (d, 6H, J=7); 1.17 (d, 3H,
J=6.6); 1.29 (t, 3H, J=7.2); 1.5–2.3 (m, 12H); 3.3 (m,
1H); 3.95 (m, 1H); 3.99 (dd, 1H, J=1.8, 6.6); 3.9–4.2
(m, 4H); 6.3 (bs, 1H); 7.9 (d, 1H, J=1.8). 13C NMR l:
14.2, 16.1, 18.2, 19.4, 20.3, 32.2, 33.2, 33.3, 33.9, 37.2,
43.1, 56.9, 58.6, 61.3, 66.6, 68.1, 158.5, 161.2, 167.5,
171.8. Anal. calcd for C22H36N4O3: C, 65.32; H, 8.97;
N, 13.85. Found: C, 65.48; H, 9.01; N, 13.9.
J=7.2); 1.16 (d, 3H, J=6.8); 1.22 (d, 3H, J=6.8); 1.33
(t, 3H, J=7); 1.7–2.4 (m, 12H); 3.01 (d, 1H, J=12.6);
3.17 (d, 1H, J=2.4); 3.31 (d, 1H, J=12.6); 3.4 (m, 1H);
3.63 (d, 1H, J=12.6); 3.96 (d, 1H, J=15); 3.98 (m, 1H);
4.02 (d, 1H, J=15); 4.1–4.4 (m, 2H); 5.06 (d, 1H,
J=15); 5.48 (d, 1H, J=15); 6.6 (m, 2ArH); 7.3 (m,
13ArH). 13C NMR l: 14.2, 16.5, 19.6, 20.1, 20.4, 21.0,
29.9, 33.2, 35.2, 39.1, 42.0, 45.3, 46.7, 46.8, 49.4, 60.6,
60.9, 65.0, 67.1, 67.3, 126.3, 127.1, 127.7, 127.8, 127.9,
128.4, 128.8, 130.7, 135.3, 136.2, 137.2, 156.9, 165.2,
170.0, 170.4. Anal. calcd for C43H54N4O4: C, 74.75; H,
7.88; N, 9.26. Found: C, 74.55; H, 7.9; N, 9.3.
7.7. 8-[(3%S,6%R)-2%,5%-Diketo-3%-isopropyl-1%,4%-diazabicy-
clo[4,3,0]non-6%-yl]-(2S,5R)-3-aza-5-amino-4-keto-2-iso-
propylethyloctanoate hydrochloride, 6
7.10. 1-[(3%S,6%R)-2%,5%-Diketo-3%-isopropyl-1%,4%-diazabi-
cyclo[4,3,0]-6%-nonyl]-3-[(3¦S,6¦S)-2¦-ethoxy-3¦,6¦-dihy-
dro-3¦-isopropyl-6¦-methyl-5¦-pirazinon-6¦-yl]propane, 8a
The intermediate 5 (0.42 g, 1 mmol) was dissolved in 10
mL of 50% EtOH/0.5N HCl and the solution stirred at
room temperature. The reaction was monitored by
TLC or 1H NMR and when the starting material
totally disappeared, the reaction mixture was evapo-
rated in vacuo to dryness and the product recovered in
practically quantitative yield. [h]D=−24 (c 0.8, 1 M
Compound 8a was obtained in 90% yield by submitting
7a to the Birch reaction. [h]D=+12.7 (c 1.2, CHCl3); 1H
NMR l: 0.86 (d, 3H, J=7); 1 (d, 3H, J=7); 1.02 (d,
3H, J=6.8); 1.1 (d, 3H, J=6.8); 1.27 (t, 3H, J=7.4);
1.32 (s, 3H); 1.4–2.4 (m, 12H); 3.35 (m, 1H); 3.62 (dd,
1H, J=3.8, 7.8); 3.9–4.2 (m, 4H); 6.14 (bs, 1H); 6.58 (d,
1H, J=3.8). 13C NMR l: 14.2, 16.1, 18.3, 19.1, 19.7,
19.8, 20, 28.4, 30.4, 33.2, 34.2, 39.2, 40.9, 45.1, 58, 59.9,
61.1, 63.1, 66.9, 156.1, 165, 171.4, 174.3. Anal. calcd for
C23H38N4O4: C, 63.57; H, 8.81; N, 12.89. Found: C,
63.75; H, 8.85; N, 12.86.
1
HCl); H NMR (CD3OD) l: 0.99 (d, 6H, J=7); 1.04
(d, 3H, J=6.8); 1.09 (d, 3H, J=6.8); 1.3 (t, 3H, J=
7.4); 1.4–2.3 (m, 12H); 3.45 (m, 1H); 3.56 (d, 1H, J=8);
3.85 (m, 1H); 3.98 (t, 1H, J=6); 4.22 (q, 2H, J=7.4);
4.41 (d, 1H, J=5.4). 13C NMR (CD3OD) l: 14.6, 18.6,

R. Galeazzi et al. / Tetrahedron: Asymmetry 14 (2003) 2639–2649
2647
7.11. 1-[(3%S,6%R)-2%,5%-Diketo-3%-isopropyl-1%,4%-diazabi-
cyclo[4,3,0]-6%-nonyl]-3-[(3¦S,6¦R)-6¦-benzyl-2¦-ethoxy-
3¦,6¦-dihydro-3¦-isopropyl-5¦-pirazinon-6¦-yl]propane, 8b
7.14. (3R,6S)-1-Benzyl-5-ethoxy-6-isopropyl-3-(3-chloro-
propyl)-6-hydro-2-pirazinone, 10
Compound 10 was obtained with an 85% yield by
alkylating 1 with 1-chloro-3-iodopropane and following
the procedure as described in Section 7.2. [h]D=+50.5
Compound 8b was obtained in 85% yield by submitting
1
7b to the Birch reaction. [h]D=−25.2 (c 1, CHCl3); H
1
NMR l: 0.75 (d, 3H, J=7); 0.78 (d, 3H, J=7); 1.05 (d,
3H, J=7); 1.12 (d, 3H, J=6.8); 1.3 (t, 3H, J=7);
1.35–2.3 (m, 12H); 2.7 (bs, 1H); 2.74 (d, 1H, J=14.6);
3.18 (d, 1H, J=14.6); 3.38 (m, 1H); 3.62 (dd, 1H, J=4,
7.6); 3.95 (m, 1H); 4.1 (m, 2H); 5.6 (bs, 1H); 6.3 (d, 1H,
J=3.2); 7.25 (m, 5ArH). 13C NMR l: 14.2, 15.7, 17.9,
19.1, 19.6, 19.8, 19.9, 29.6, 33.2, 33.9, 39.3, 40.6, 45.2,
47.2, 57.3, 61.1, 63.2, 65.1, 66.8, 126.5, 127.6, 130.3,
136.3, 157.7, 165.2, 171.5, 172.4. Anal. calcd for
C29H42N4O4: C, 68.21; H, 8.29; N, 10.97. Found: C,
67.98; H, 8.31; N, 10.95.
(c 2.0, CHCl3); H NMR l: 0.91 (d, 3H, J=7); 1.03 (d,
3H, J=6.8); 1.23 (t, 3H, J=6.8); 1.8–2.3 (m, 5H); 3.58
(t, 2H, J=6.6); 3.68 (dd, 1H, J=1.6, 4); 3.91 (d, 1H,
J=15.2); 4.1 (m, 3H); 5.46 (d, 1H, J=15.2); 7.3 (m,
5ArH). 13C NMR l: 14.0; 17.4; 19.8; 28.7; 31.0; 31.5;
44.9; 47.2; 57.1; 61.0; 61.8; 127.3; 127.5; 128.4; 136.0;
159.2; 169.7. Anal. calcd for C19H27N2O2Cl: C, 65.04;
H, 7.76; N, 7.98; Cl, 10.1. Found: C, 65.3; H, 7.77; N,
7.95; Cl, 10.12.
7.15. (3S,6R)-4-Benzyl-2,5-diketo-3-isopropyl-1,4-diaza-
7.12. 8-[(3%S,6%R)-2%,5%-Diketo-3%-isopropyl-1%,4%-diazabi-
cyclo[4,3,0]non-6%-yl]-(2S,5S)-3-aza-5-amino-4-keto-2-
isopropyl-5-methylethyloctanoate hydrochloride, 9a
bicyclo[4,3,0]nonane, 11
Compound 11 was obtained in practically quantitative
yield from 10 following the procedure as described in
Section 7.3. The product was not obtained in a suffi-
Compound 9a was obtained in practically quantitative
yield starting from 8a and following the procedure as
1
ciently pure form to measure the specific rotation. H
1
described in Section 7.7. [h]D=+10 (c 1.4, CHCl3); H
NMR l: 1.07 (d, 3H, J=6.9); 1.15 (d, 3H, J=6.9);
1.8–2.4 (m, 5H); 3.52 (m, 1H); 3.6 (m, 1H); 3.71 (d, 1H,
J=6.2); 3.72 (m, 1H); 3.98 (d, 2H, J=15); 4.2 (m, 2H);
5.4 (d, 2H, J=15); 7.3 (m, 5ArH). 13C NMR l: 18.1;
19.7; 22.0; 29.7; 31.4; 45.2; 48.3; 58.5; 67.1; 127.4; 127.5;
128.5; 135.7; 164.2; 167.5.
NMR l: 0.93 (d, 3H, J=6.6); 0.96 (d, 3H, J=6.6); 1.00
(d, 3H, J=7); 1.17 (d, 3H, J=7); 1.3 (t, 3H, J=7); 1.34
(s, 3H); 1.4–2.4 (m, 12H); 3.3 (m, 1H); 3.99 (dd, 1H,
J=3.8, 7.6); 4.2 (m, 3H); 6.4 (bs, 1H); 4.5 (dd, 1H,
J=4.8, 8.8); 7.87 (d, 1H, J=1.8); 8.02 (d, 1H, J=8.8).
13C NMR l: 14.2, 18.2, 19, 19.1, 19.4, 19.6, 19.7, 23.2,
30.8, 33.1, 33.7, 37.2, 37.9, 44.9, 58, 60.5, 61.6, 63.3,
66.6, 165, 170.8, 171.1, 171.7. Anal. calcd for
C23H41N4O5Cl: C, 56.49; H, 8.45; N, 11.46; Cl, 7.25.
Found: C, 56.68; H, 8.43; N, 11.5; Cl, 7.27.
7.16. (3S,6R)-4-Benzyl-2,5-diketo-3-isopropyl-6-(3-
chloropropyl)-1,4-diazabicyclo[4,3,0]nonane, 12
Compound 12 was obtained in 80% yield by alkylating
11 with 1-chloro-3-iodopropane and following the pro-
cedure as described in Section 7.2. [h]D=−55.3 (c 1.5,
7.13. 8-[(3%S,6%R)-2%,5%-Diketo-3%-isopropyl-1%,4%-diazabi-
cyclo[4,3,0]non-6%-yl]-(2S,5R)-3-aza-5-amino-5-benzyl-4-
keto-2-isopropylethyloctanoate hydrochloride, 9b
1
CHCl3); H NMR l: 0.83 (d, 3H, J=6.9); 1.18 (d, 3H,
J=6.9); 1.5–2.4 (m, 9H); 3.4 (m, 3H); 3.87 (d, 1H,
J=2.4); 3.9 (m, 1H); 3.99 (d, 1H, J=14.7); 5.4 (d, 1H,
J=14.7); 7.3 (m, 5ArH). 13C NMR l: 15.5; 19.7; 19.9;
26.8; 30.4; 35.1; 35.4; 43.3; 44.1; 47.1; 64.0; 66.5; 128.0;
128.7; 135.4; 163.5; 168.6. Anal. calcd for
C20H27N2O2Cl: C, 66.19; H, 7.5; N, 7.72; Cl, 9.77.
Found: C, 66.38; H, 7.53; N, 7.75; Cl, 9.8.
Compound 9b was obtained from 8b following the
procedure described in Section 7.7. After chromato-
graphic purification by elution with hexane/ethylac-
etate, the product was isolated with a 50% yield.
However it was not obtained in a sufficiently pure form
1
to measure the specific rotation. H NMR (CD3OD) l:
1.04 (d, 3H, J=7); 1.06 (d, 3H, J=7);1.08 (d, 3H,
J=7); 1.09 (d, 3H, J=7); 1.34 (t, 3H, J=7.4); 1.4–2.4
(m, 12H); 3.3 (qAB, 1H, J=14.4); 3.46 (m, 1H); 3.58 (d,
1H, J=8.4); 3.9 (m, 1H); 4.28 (m, 2H); 4.4 (d, 1H,
J=6.6); 7.4 (m, 5ArH). 13C NMR (CD3OD) l: 14.7,
19.4, 19.7, 20.0, 20.7, 31.4, 34.6, 35.0, 36.4, 39.2, 43.1,
46.4, 60.2, 62.4, 64.2, 65.2, 68.2, 129, 129.9, 131.6, 134,
167.4, 171.5, 172.6, 172.7.
7.17. (3S,6R)-4-Benzyl-2,5-diketo-3-isopropyl-6-(3-iodo-
propyl)-1,4-diazabicyclo[4,3,0]nonane, 13
Compound 13 was obtained in practically quantitative
yield from 12 and following the procedure as described
in Section 7.3. The product was not obtained in a
sufficiently pure form to measure the specific rotation.
1H NMR l: 0.82 (d, 3H, J=6.8); 1.16 (d, 3H, J=6.8);
1.5–2.4 (m, 9H); 3 (m, 2H); 3.35 (m, 1H); 3.85 (d, 1H,
J=1.8); 3.9 (m, 1H); 3.97 (d, 1H, J=14.4); 5.4 (d, 2H,
J=14.4); 7.3 (m, 5ArH). 13C NMR l: 5.0; 15.1; 19.2;
19.5; 27.3; 29.9; 34.7; 38.4; 42.8; 46.6; 63.4; 65.8; 127.3;
128.2; 134.9; 162.7; 167.8.
From the reaction bis-diketopiperazine 9%b was also
obtained in about a 50% yield. 1H NMR l: 0.77 (d, 3H,
J=7); 0.81 (d, 3H, J=6.8); 0.98 (d, 3H, J=7); 1.1 (d,
3H, J=6.8); 1.3–2.3 (m, 12H); 2.4 (d, 1H, J=1.2); 2.78
(d, 1H, J=13.2); 3.2 (d, 1H, J=13.2); 3.33 (m, 1H);
3.68 (dd, 1H, J=3.4, 7.8); 4.1 (m, 1H); 6.1 (bs, 1H); 7.2
(m, 5ArH); 8.2 (bs, 1H); 8.6 (d, 1H, J=3.4).

2648
R. Galeazzi et al. / Tetrahedron: Asymmetry 14 (2003) 2639–2649
7.18. 1-[(3%S,6%S)-4%-Benzyl-2%,5%-diketo-3%-isopropyl-1%,4%-
diazabicyclo[4,3,0]-6%-nonyl]-3-[(3¦S,6¦R)-4¦-benzyl-2¦-
ethoxy-3¦,6¦-dihydro-3¦-isopropyl-5¦-pirazinon-6¦-yl]-
propane, 14
the data can be obtained free of charge, on applica-
tion to CCDC, 12 Union Road, Cambridge, CB2 1EZ,
UK [fax: +44(0)1223336033 or e-mail: deposit@ccdc.
cam.ac.uk].
To a solution of 1 (2.74 g, 10 mmol) in dry THF (100
mL) cooled at −78°C was added 10 mmol of LHMDS
(1 M solution in THF). After about one hour under
stirring, the intermediate 13 (4.5 g, 10 mmol) was added
and the reaction monitored by TLC. When the starting
material disappeared, water and ethylacetate were then
added. The organic extract was dried over Na2SO4 and
then evaporated to dryness in vacuo. The crude reac-
tion product after purification by silica gel chromatog-
raphy (eluting with hexane/ethylacetate) was obtained
in about a 70% yield. However, the product was not
recovered in a sufficiently pure form to measure the
7.20. Computational methods
All calculations were carried out on SGI IRIX 6.5
workstations. Molecular mechanic calculations were
performed using the implementation of the Amber all-
atom force field (AMBER*) within the framework of
Macromodel version 5.5.22 The AMBER* force field in
MMOD 5.5 contained a new set of parameters for
proline containing peptides, recently developed on the
basis of high-level ab initio calculations.23 The solvent
effect was included by the implicit chloroform GB/SA
solvation model of Still et al.24 The torsional space of
each molecule was randomly varied with the usage-
directed Monte Carlo conformational search of Chang-
Guida-Still. For each search, at least 1000 starting
structures for each variable torsion angle was generated
and minimized until the gradient was less than 0.05
kJ/A mol. The cyclic moieties containing the amide
bonds were also included into the search. Duplicate
conformations and those with an energy in excess of 5
kcal/mol above the global minimum were discarded.
1
specific rotation. H NMR l: 0.78 (d, 3H, J=6.6); 0.91
(d, 3H, J=6.6); 1.04 (d, 3H, J=7); 1.09 (d, 3H, J=7);
1.12 (t, 3H, J=7); 1.4–2.4 (m, 12H); 3.35 (m, 1H); 3.67
(dd, 1H, J=1.4, 4); 3.74 (d, 1H, J=2.6); 3.8–4.1 (m,
6H); 5.38 (d, 1H, J=15); 5.45 (d, 1H, J=15); 7.3 (m,
10ArH). 13C NMR l: 13.7, 13.8, 14.9, 17.3, 19.4, 19.6,
29.5, 31.2, 32.9, 35.1, 37.7, 42.9, 46.6, 46.8, 57.1, 60.8,
61.3, 63.0, 66.7, 127.1, 127.2, 127.4, 127.6, 128.1, 128.3,
134.8, 135.8, 158.8, 163.0, 168.7, 169.5.
All DFT calculations (i.e. geometry optimizations and
chemical shift simulations) were carried out using the
standard tools available in the Gaussian 98 package,25
with the DFT/B3LYP functional (i.e. Becke’s three
parameter hybrid functional with the Lee-Yang-Parr
correlation functional)26 and the 6-31G(d) basis set.
This functional and basis set have been shown to
properly describe both standard hydrogen bonds,27 as
well as non-classical, weakly bound hydrogen bonds
(such as CꢁH···OꢀC interactions),28–30 and to provide
reliable results for the protons chemical shifts.13,14,29
However, the computed data does not directly yield the
chemical shift value, but only a value for the isotropic
magnetic tensor. The chemical shift value was obtained
from the equation lH=32.18−|H, where 32.18 is the
calculated isotropic magnetic tensor for the protons in
tetramethylsilane and |H is the calculated isotropic
magnetic tensor for the investigated proton. This proce-
dure has been recently validated and applied in other
molecular systems.29,30
7.19. X-Ray crystallography of 5
The diffraction experiments were carried out at room
temperature on a Bruker AXS SMART 2000 CCD
based diffractometer using graphite monochromated
,
Mo–Ka radiation (u=0.71073 A). Intensity data were
measured over full diffraction spheres using 0.3° wide ꢀ
scans, crystal-to-detector distance 5.0 cm. The software
SMART18 was used for collecting frames of data,
indexing reflections and determination of lattice
parameters. The collected frames were then processed
for integration by software SAINT18 and an empirical
absorption correction applied with SADABS.19 The
structure was solved by direct methods (SIR97)20 and
subsequent Fourier syntheses, and refined by full-
matrix least-squares calculations on F2 (SHELXTL),21
attributing anisotropic thermal parameters to the non-
hydrogen atoms. The methyl and methylene hydrogen
atoms were placed in calculated positions and refined
with idealized geometry, whereas the other H atoms
were located in the Fourier map and refined
isotropically.
Acknowledgements
7.19.1. Crystallographic data. C22H36N4O4, orthorhom-
bic, P212121 (No. 19), a=9.2158(7), b=9.6257(7), c=
3
Thanks are due to the MIUR (Rome) (COFIN 2002)
and to the University of Bologna for financial support.
,
,
26.205(2) A, V=2324.6(3) A , Z=4, dcalcd=1.202 Mg
m−3, v=0.083 mm−1. 30453 reflections were collected,
6781 unique, 5062 observed for I>2|(I), which were
used in all calculations. Final R factors: R1=0.0486
[I>2|(I)], wR2=0.1516 (all data).
References
Crystallographic data (excluding structure factors) for
the structure in this paper have been deposited with the
Cambridge Crystallographic Data Centre as supple-
mentary publication number CCDC 212112. Copies of
1. Paradisi, F.; Porzi, G.; Sandri, S. Tetrahedron: Asymme-
try 2001, 12, 3319–3324.
2. (a) Gelb, M. H.; Lin, Y.; Pickard, M. A.; Song, Y.;
Vederas, J. C. J. Am. Chem. Soc. 1990, 112, 4932–4942;

R. Galeazzi et al. / Tetrahedron: Asymmetry 14 (2003) 2639–2649
2649
(b) Williams, R. M.; Fegley, G. J.; Gallegos, R.; Schaefer,
F.; Pruess, D. L. Tetrahedron 1996, 52, 1149–1164; (c)
Bull, S. D.; Chernega, A. N.; Davies, S. G.; Moss, W. O.;
Parkin, R. M. Tetrahedron 1998, 54, 10379–10388.
3. (a) Williams, R. M.; Myeong-Nyeo Im; Cao, J. J. Am.
Chem. Soc 1991, 113, 6976–6981; (b) Satyanarayana, S.;
Grossert, J. S.; Lee, S. F.; White, R. L. Can. Amino Acids
2001, 21, 221–235; (c) Kubasch, N.; Schmidt, R. R. Eur.
J. Org. Chem. 2002, 2710–2726.
1187; (d) Paradisi, F.; Porzi, G.; Rinaldi, S.; Sandri, S.
Tetrahedron: Asymmetry 2000, 11, 4617–4622; (e) Para-
disi, F.; Porzi, G.; Rinaldi, S.; Sandri, S. Tetrahedron:
Asymmetry 2000, 11, 1259–1262.
18. SMART & SAINT Software Reference Manuals, version
5.051 (Windows NT Version), Bruker Analytical X-ray
Instruments Inc.: Madison, WI, 1998.
19. Sheldrick, G. M. SADABS: Program for Empirical
Absorption Correction; University of Go¨ttingen, Ger-
many, 1996.
20. Altomare, A.; Burla, M. C.; Cavalli, M.; Cascarano, G.
L.; Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.;
Polidori, G.; Spagna, R. J. Appl. Crystallogr. 1999, 32,
115–119.
21. Sheldrick, G. M. SHELXTLplus Version 5.1 (Windows
NT version) Structure Determination Package; Bruker
Analytical X-ray Instruments Inc.: Madison, WI, 1998.
22. Mohamadi, F.; Richards, N. G. J.; Guida, W. C.;
Liskamp, R.; Lipton, M.; Caulfield, C.; Chang, G.; Hen-
drickson, T.; Still, W. C. J. Comput. Chem. 1990, 11,
440–467.
4. Bouchaudon, J.; Dutruc-Rosset, G.; Farge, D.; James, C.
J. Chem. Soc., Perkin Trans. 1 1989, 695–701.
5. Bull, S. D.; Davies, S. G.; Parkin, R. M.; Sanchez-San-
cho, F. J. Chem. Soc., Perkin Trans. 1 1998, 2313–2320.
6. (a) Favero, V.; Porzi, G.; Sandri, S. Tetrahedron: Asym-
metry 1997, 8, 599–612; (b) Di Felice, P.; Porzi, G.;
Sandri, S. Tetrahedron: Asymmetry 1999, 10, 2191–2201.
7. Porzi, G.; Sandri, S.; Verrocchio, P. Tetrahedron: Asym-
metry 1998, 9, 119–132.
8. Ab initio calculations were carried out by using a Gaus-
sian-98 W2.1 program and full geometry optimizations
were performed at the RB3LYP/6-311G(d) level.
9. (a) Trabocchi, A.; Occhiato, E. G.; Potenza, D.; Guarna,
A. J. Org. Chem. 2002, 67, 7483–7492; (b) Fernandez, M.
M.; Diez, A.; Rubiralta, M.; Montenegro, E.; Casamit-
jana, N. J. Org. Chem. 2002, 67, 7587–7599; (c) Belvisi,
L.; Bernardi, A.; Manzoni, L.; Potenza, D.; Scolastico, C.
Eur. J. Org. Chem. 2000, 2563–2569; (d) Belvisi, L.;
Gennai, C.; Mielgo, A.; Potenza, D.; Scolastico, C. Eur.
J. Org. Chem. 1999, 389–400.
23. McDonald, D. Q.; Still, W. C. J. Org. Chem. 1996, 61,
1385–1391.
24. Still, W. C.; Tempczyk, A.; Hawley, R. C.; Hendrickson,
T. J. Am. Chem. Soc. 1990, 12, 6127–6129.
25. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria,
G. E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V.
G.; Montgomery, J. A., Jr.; Stratmann, R. E.; Burant, J.
C.; Dapprich, S.; Millan, J. M.; Daniels, A. D.; Kudin,
K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.;
Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.;
Adamo, C.; Clifford, S.; Ochterski, J.; Patersson, G. A.;
Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malik, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.;
Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov, B.
B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.;
Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challa-
combe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.;
Wong, M. W.; Andres, J. L.; Gonzales, C.; Head-Gor-
don, M.; Replogle, E. S.; Pople, J. Gaussian 98, Revision
A.9, Gaussian Inc., Pittsburgh, PA, 1998.
26. Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652.
27. (a) Gonzalez, L.; Mo, O.; Yanez, M. J. Comput. Chem.
1997, 18, 1124–1135; (b) Klein, R. A. J. Comput. Chem.
2002, 23, 585–599; (c) Novoa, J. J.; Sosa, C. J. Phys.
Chem. 1995, 99, 15837–15845.
28. Brunel, L.; Carre’, F.; Dutremez, S. G.; Guerin, C.;
Dahan, F.; Eisenstein, O.; Sini, G. Organometallics 2001,
20, 47–54.
10. Mazza, F.; Lucente, G.; Pinnen, F.; Zanotti, G. Acta
Crystallogr. 1984, C40, 1974–1976.
11. Weiner, S. J.; Kollman, P. A.; Nguyen, D. T.; Case, D.
A. J. Comput. Chem. 1986, 4, 230–252.
12. Chang, G.; Guida, W. C.; Still, W. C. J. Am. Chem. Soc.
1989, 111, 4379–4386.
13. (a) Wolinski, K.; Hilton, J. F.; Pulay, P. J. Am. Chem.
Soc. 1990, 112, 8251–8260; (b) Wolinski, K.; Sadlej, A.
Mol. Phys. 1980, 41, 1419–1430; (c) Ditchfield, R. Mol.
Phys. 1974, 27, 789–807; (d) McWeeny, R. Phys. Rev.
1962, 126, 1024–1028; (e) London, F. J. Phys. Radium
1937, 8, 397–409.
14. Bagno, A. Chem. Eur. J. 2001, 7, 1652–1660.
15. 1H NMR chemical shift DFT computations for the
trimer involve a molecular system where the DFT opti-
mized monomer is surrounded by two simple amidic
groups (NH2-CHO) placed at crystallographic positions
and simulating the intermolecular hydrogen bonding net-
work around a monomer as seen in the X-ray resolved
trimer (see Fig. 3).
16. Sanz-Cervera, J. F.; Williams, R. M.; Alberto Marco, J.;
Lopez-Sanchez, J. M.; Gonzalez, F.; Martinez, M. E.;
Sancenon, F. Tetrahedron 2000, 56, 6345–6358.
17. (a) Piccinelli, F.; Porzi, G.; Sandri, M.; Sandri, S. Tetra-
hedron: Asymmetry 2003, 14, 393–398; (b) Paradisi, F.;
Piccinelli, F.; Porzi, G.; Sandri, S. Tetrahedron: Asymme-
try 2002, 13, 497–502; (c) Ferioli, F.; Piccinelli, F.; Porzi,
G.; Sandri, S. Tetrahedron: Asymmetry 2002, 13, 1181–
29. Bernardi, F.; Garavelli, M.; Scatizzi, M.; Tomasini, C.;
Trigari, V.; Crisma, M.; Formaggio, F.; Peggion, C.;
Toniolo, C. Chem. Eur. J. 2002, 8, 2516–2525.
30. Tomasini, C.; Trigari, V.; Lucarini, S.; Bernardi, F.;
Garavelli, M.; Peggion, C.; Formaggio, F.; Toniolo, C.
Eur. J. Org. Chem. 2003, 259–267.