A new access to 3,5-disubstituted piperazinones via Pd(0)-catalyzed amination

Article abstract of DOI:10.1016/S0040-4039(03)00885-2

An original synthetic route toward 3,5-disubstituted piperazinones has been developed. The method relies upon a 6-exo intramolecular process between a sulfonylated nitrogen atom of amino acid derivation and an η³-allyl-palladium moiety. This cyclization process generates the two possible (cis and trans) diastereoisomers whose ratio depends on the amino acid employed. The bulkier the amino acid residue, the higher the observed cis:trans ratio. Convincing evidence for reversible intramolecular addition of the nitrogen nucleophile to the η³-allyl-palladium complex is put forward.

Full text of DOI:10.1016/S0040-4039(03)00885-2

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 4213–4216
A new access to 3,5-disubstituted piperazinones via
Pd(0)-catalyzed amination
Benoit Ferber,a Se´bastien Lemaire,a Mary M. Mader,b Guillaume Prestata and Giovanni Polia,*
aLaboratoire de Chimie Organique, UMR 7611 CNRS, Universite´ Pierre et Marie Curie, Tour 44-45, Boˆıte 183,
4, Place Jussieu, F-75252 Paris, Cedex 05, France
bEli Lilly and Company, Lilly Corporate Center, Drop Code 1523, Indianapolis, IN 46285, USA
Received 13 March 2003; revised 4 April 2003; accepted 4 April 2003
Abstract—An original synthetic route toward 3,5-disubstituted piperazinones has been developed. The method relies upon a 6-exo
intramolecular process between a sulfonylated nitrogen atom of amino acid derivation and an h3-allyl-palladium moiety. This
cyclization process generates the two possible (cis and trans) diastereoisomers whose ratio depends on the amino acid employed.
The bulkier the amino acid residue, the higher the observed cis:trans ratio. Convincing evidence for reversible intramolecular
addition of the nitrogen nucleophile to the h3-allyl-palladium complex is put forward. © 2003 Elsevier Science Ltd. All rights
reserved.
We recently described a novel access to 3,4-disubsti-
tuted pyrrolidones1 based on the intramolecular inter-
action between a stabilized acetamide enolate anion
and an h3-allyl-palladium appendage (Scheme 1).
envisaged the formal replacement of the active methyl-
ene moiety for a suitably N-protected amino acid
residue in the nucleophilic left-side arm of the precur-
sor. This set the stage for a new 6-exo allylic amination
process allowing the generation of 3,5-disubstituted
piperazinones. Of course, the role of the pre-existing
stereogenic center on the diastereoselectivity of the
cyclization represented a point of potential interest
(Scheme 2).
We next thought that an appropriate modification of
the above strategy might have enabled the construction
of piperazinone derivatives, a family of molecules
endowed of a particular pharmacological interest. The
piperazine or piperazinone structure is in fact present in
compounds possessing antifungal,2 antidepressant,3
antimigraine,4 antithrombotic,5 antihistaminic,6 anti-
aggregating,7 or nootropic activity.8 Accordingly, we
According to the above reasoning the known allylic
amine 11 has been acylated with three differently N-
protected
L
-alanine derivatives 2–4. N-Boc -alanine 2
L
was commercially available. N-Cbz-3 and N-Ts-4 were
instead prepared from -alanine according to known
and/or standard procedures. N-Bz- -alanine precursor
L
L
8 was prepared from 5 by Boc-to-Bz exchange (Scheme
3).
A preliminary set of experiments was performed in
DMF using Pd(OAc)2 (5%), 1,2-bis(diphenylphos-
phino)ethane (dppe) (10%) and NaH (1 equiv.) as
deprotonating agent.9 Under these conditions N-Boc,
N-Cbz and N-Bz precursors were recovered unchanged.
Conversely and quite gratifyingly, the same reaction
conditions smoothly and quantitatively converted the
N-tosyl derivative 7 into the expected piperazinone 11
as a 1:1 mixture of the two possible C5-epimers. Further
optimization experiments confirmed the couple
Pd(OAc)2/dppe as the optimal catalytic system, and
DMF as the best solvent. Furthermore, it was soon
realized that upon thermal activation (80°C), NaH was
Scheme 1. Reagents and conditions: Pd2(dba)3, (0.05 equiv.),
PPh3, (0.5 equiv.), BSA (1.2 equiv.) AcOK (0.1 equiv.), THF,
reflux, 12 h, 61–93%. EWG: CO2Me, COMe, CN, SO2Ph,
PO(OEt)2.
* Corresponding author. Tel.: +33-(0)1-4427-5572; fax: +33-(0)1-
4427-7567; e-mail: poli@ccr.jussieu.fr
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4039(03)00885-2

4214
B. Ferber et al. / Tetrahedron Letters 44 (2003) 4213–4216
Having established the tosyl group as the optimal N-
derivatization of the cyclization precursor, -leucine,
-isoleucine, -phenylglycine, -phenylalanine, and L-
L
L
L
L
tryptophan were similarly converted into the corre-
sponding allylamides 13–18 and tested for cyclization
(Scheme 5, Table 2).
Scheme 2. Retrosynthetic disconnection of the 3,5-disubsti-
tuted piperazinone heterocycle.
Despite the excellent chemical yields, the diastereoselec-
tivities of the cyclization process were variable, span-
ning from totally random, in the case of the
derivative (entry 1), to a 95:5 ratio, in the case of the
-isoleucine derivative (entry 3). A careful inspection of
L-alanine
L
1
the H NMR spectra as well as of the chromatographic
behavior of the diastereoisomeric piperazinones
revealed that specific features constantly discriminated
the set of the major diastereomers from that of the
minor ones. In fact, in each series studied the major and
more apolar isomer regularly featured
a
more
deshielded H5% as well as a smaller Dl (H6−H6%). Such
indications were thus suggestive of a precise, though
variable, diastereoselective trend. Although NOE differ-
ence analysis of the resulting piperazinones turned out
to be inconclusive, an X-ray diffraction structure of the
crystalline major isomer of 21 ([h]2D0 (c 0.4, CHCl3)
−39)14,15 unambiguously showed its cis 3,5-configura-
tion (Fig. 1).16 Accordingly, we confidently assigned the
same cis configuration to the other major isomers.
Scheme 3. Reagents and conditions: (i) Cbz-Cl, NaOH, Et2O/
H2O (3, 91%); (ii) TsCl, NaOH, Et2O/H2O (4, 76%); (iii)
DCC, DMAP (5%), THF (5: 78%; 6: 65%, 7: 90%); (iv) (1) 5,
TMSCl, NaI, CH3CN, then MeOH, (2) PhCOCl, NEt3,
DMAP (5%), CH2Cl2 (8, 65% over the two steps).
unnecessary for the success of the cyclization. Under
these conditions 1110 was obtained in quantitative yield
within 3 h with the same diasteromeric ratio as
observed at room temperature in the presence of NaH.
Moreover, these conditions allowed the cyclization of
the N-Boc and N-Cbz derivatives, albeit in limited yield
and longer reaction time (Scheme 4, Table 1).
Having established the sense of the diastereoselection,
rationalization of its extent as a function of the incor-
porated amino acid moiety became a tractable task.
From a qualitative viewpoint, it is interesting to note
that diastereoselectivity augments with increasing of the
steric demand of the incorporated amino acid moiety.
Although such a result was not unexpected, the thermo-
dynamics of the present cyclization process remained to
be ascertained. Accordingly, an equilibration test was
performed by re-submission of an equimolar mixture of
the isolated piperazinone trans-21 and its precursor 15
to the above cyclization conditions (Scheme 6).17
The reasons for the better cyclization behavior of the
N-tosyl derivative 7 with respect to the carbamates 5
and 6, or the amide derivative 8 are still unclear. It
should be noted, however, that precedents of Pd(0)-cat-
alyzed aminations using carbamate-protected11 or sul-
fonamide-protected12 amines are rather rare, and
examples of failure have been reported.13
After complete conversion of the precursor to the piper-
azinone, a 50:50 epimeric mixture cis- and trans-21 was
obtained. This result implies that 25% of the trans
epimer has been converted into the cis one. We thus
conclude that the amination reaction is a reversible
process, the equilibration between the two epimers tak-
ing place very likely through the corresponding zwitter-
ionic h3-allylpalladium species 22 (Scheme 7). It is
interesting to note that the above behavior strictly
parallels that found by Ibuka et al. in the Pd(0)-cata-
lyzed equilibration of N-methane- and N-arenesulfonyl-
Scheme 4. Reagents and conditions: (i) 5% Pd(OAc)2, 10%
dppe, DMF, 80°C.
Table 1. Cyclization outcomes for allylic alanine deriva-
tives
Precursor
EWG
Piperazinone
Yield (%)
5
6
7
8
Boc
Cbz
Ts
9
85
57
]98
0
10
11
12
Bz
Scheme 5. Reagents and conditions: (i) 5% Pd(OAc)2, 10%
dppe, DMF, 80°C, 3 h.

B. Ferber et al. / Tetrahedron Letters 44 (2003) 4213–4216
Table 2. Synthesis of 3,5-disubstituted N-tosyl-piperazinones
4215
Entry
R
Precursor
Piperazinone
(%)
cis/trans
1
2
3
4
5
6
7
Me
7
13
14
15
16
17
18
11
19
20
21
22
23
24
]98
]98
]98
]98
]98
]98
]98
50:50
60:40
95:5
74:26
65:35
69:31
–
i-Bu
s-Bu
Ph
Bn
3-CH2-ind
H
epimers of all the minimized piperazinones showed
nitrogen pyramidalization22 and the same boat-type
conformation as seen in the X-ray structure of cis-21.23
More importantly, the cis isomers were consistently
found to be more stable than the corresponding trans
ones.24 Although the calculated values do not quantita-
tively match the experimental observed cis/trans ratios,
the qualitative trend is respected, with piperazinones 20
featuring the highest energy difference.25 The overall cis
preference is expected to be essentially due to the
tetrahedral geometry of the sulfonamide nitrogen,
which favors the equatorial disposition of the bulky
tosyl group and forces the lone pair of electrons on
nitrogen in the axial position. As a result, only the cis
epimer can arrange both the amino acid residue and
vinyl group trans to the bulky sulfonyl group.
Figure 1. ORTEP drawing of cis-21.
In summary, we have developed a new route to 3,5-di-
substituted piperazinones based on an intramolecular
Pd(0)-catalyzed allylic amination of N-tosylated amino
acid derivatives. Piperazinones are obtained in high
yields as a mixture of the two possible diastereoisomers.
Such a cyclization has been shown to be a reversible
process, constantly favoring the thermodynamically
more stable cis isomer with cis/trans ratios increasing
with the size of the amino acid residue.
Scheme 6. Reagents and conditions: (i) 5% Pd(OAc)2, 10%
dppe, DMF, 80°C, 3 h.
Acknowledgements
We gratefully acknowledge the University Pierre et
Marie Curie, the CNRS and COST action D-24 of the
European Community for funding of this research. We
also thank Carine Guyard-Duhayon (Laboratoire de
Chimie Inorganique et Materiaux Moleculaires, Uni-
versite´ Pierre et Marie Curie) for the X-ray structure
and the Department of Organic Chemistry at the Dan-
ish Technical University for computing time.
Scheme 7. Reversibility of the Pd(0)-catalyzed amination
reaction.
2-alkyl-3-vinylaziridines. Indeed, in these substrates
too, the cis isomers are constantly found to be more
stable than the trans ones.18,19
References
1. Giambastiani, G.; Pacini, B.; Porcelloni, M.; Poli, G. J.
Org. Chem. 1998, 63, 804–807.
Verification of the higher stability of the cis piperazi-
nones over the trans ones was judged essential to
corroborate our rationale. Accordingly, we carried out
a molecular mechanics computational analysis of the
cyclized products. Conformational searches20 in Macro-
Model v. 7.221 were performed on both epimers of 11,
19–23, and the energy differences between the global
minima for each pair was calculated. The cis and trans
2. Gala, D. Tetrahedron Lett. 1999, 40, 5655–5659.
3. Giannangeli, M. J. Med. Chem. 1999, 42, 336–345.
4. Chambers, M. S.; Street, L. J. J. Med. Chem. 1999, 42,
691–705.
5. Matsoukas, J.; Alexopulos, K. Bioorg. Med. Chem. 1999,
7, 1033–1041.
6. Park Choo, H.-Y. Bioorg. Med. Chem. 1999, 9, 2727–
2730.

4216
B. Ferber et al. / Tetrahedron Letters 44 (2003) 4213–4216
7. Sugihara, H. J. Med. Chem. 1998, 41, 489–502.
integrity in the resulting piperazinones has not been
specifically addressed. Nevertheless, related literature
precedents clearly indicate that the reaction conditions are
not expected to be racemizing. See for example: (a)
Verhelst, S. H. L.; Wiednhof, W.; Ovaa, H.; van der Marel,
G. A.; Overkleeft, H. S.; van Boeckel, C. A. A.; van Boom,
J. H. Tetrahedron Lett. 2002, 43, 6451–6455; (b) Kinder-
man, S. S.; Doodeman, R.; van Beijma, J. W.; Russcher,
J. C.; Tjen, K. C. M. F.; Kooistra, T. M.; Mohaselzadeh,
H.; van Maarseveen, J. H.; Hiemstra, H.; Schoemaker, H.
E.; Rutjes, F. P. J. T. Adv. Synth. Catal. 2002, 344, 736–748;
(c) Zorn, C.; Gnad, F.; Salmen, S.; Herpin, T.; Reiser, O.
Tetrahedron Lett. 2001, 42, 7049–7053.
8. (a) Manetti, D.; Bartolini, F.; Gualtieri, F. Bioorg. Med.
Chem. 1999, 7, 457–465; (b) Manetti, D.; Ghelardini, C.;
Bartolini, A.; Bellucci, C.; Dei, S.; Galeotti, N.; Gualtieri,
F.; Romanelli, M. N.; Scapecchi, S.; Teodori, E. J. Med.
Chem. 2000, 43, 1969–1974; (c) Manetti, D.; Ghelardini,
C.; Bartlini, A.; Dei, S.; Galeotti, N.; Gualtieri, F.;
Romanelli, M. N.; Teodori, E. J. Med. Chem. 2000, 43,
4499–4507.
9. These cyclization conditions are optimized for the synthesis
of pyrrolidones and differ with respect to those used in Ref.
1 (Pd2(dba)3, PPh3, BSA, AcOK, THF). See: Poli, G.;
Giambastiani, G. J. Org. Chem. 2002, 67, 9456–9459.
10. Piperazinone 11: dppe (20 mg, 0.1 mmol) was added to a
solution of Pd(OAc)2 (5.6 mg, 0.025 mmol) in dry DMF
(1 mL) under N2 atmosphere. After 15 min of stirring,
cyclization precursor 7 (0.5 mmol, 222 mg) in DMF (1 mL)
was added through cannula to the thus formed Pd(0)
complex. The resulting mixture was stirred at 80°C for 3
h before it was cooled to rt. Et2O was then added and the
organic phase was washed with brine (3×20 mL) dried over
MgSO4 and concentrated under reduced pressure. Flash
chromatography gave the pure piperazinone as a pale
16. Crystallographic data (excluding structure factors) for the
structures in this paper, have been deposited with the
Cambridge Crystallographic Data Centre as supplemen-
tary publication numbers CCDC 202406. Copies of the
data can be obtained, free of charge, on application to
CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (Fax:
+44(0)-1223-336033 or e-mail: deposit@ccdc.cam.ac.uk).
17. An equimolar mixture of the isolated piperazinone trans-
21 and its precursor 15 was deliberately chosen for this test
in order to reproduce the original reaction conditions. In
fact, a potential interaction between the starting acetate
and the catalytic system, which would affect the exact
nature of the real catalyst, cannot be completely ruled out.
18. (a) Ibuka, T.; Mimura, N.; Aoyama, H.; Akaji, M.; Ohno,
H.; Miwa, Y.; Taga, T.; Nakai, K.; Tamamura, H.; Fujii,
N.; Yamamoto, Y. J. Org. Chem. 1997, 62, 999–1015; (b)
See also: Ohno, H.; Anzai, M.; Toda, A.; Ohishi, S.; Fujii,
N.; Tanaka, T.; Takemoto, Y.; Ibuka, T. J. Org. Chem.
2001, 66, 4904–4914; (c) Mimura, N.; Miwa, Y.; Ibuka, T.;
Yamamoto, Y. J. Org. Chem. 2002, 67, 5796–5801.
19. For a related trans-to-cis equilibration of N-Boc-5-alkyl-
pipecolaldehydes, see: (a) Wilkinson, T. J.; Stehle, N. W.;
Beak, P. Org. Lett. 2000, 2, 155–158; (b) Beak, P.; Lee, W.
K. J. Org. Chem. 1993, 58, 1109–1117; (c) Kotsuki, H.;
Kusumi, T.; Inoue, M.; Ushio, Y.; Ochi, M. Tetrahedron
Lett. 1991, 32, 4159–4162.
20. Conformational searches were performed by the Monte
Carlo method, generating 1000–4000 structures within 10
kJ/mol energy window. The lowest energy structure in each
search was found more than 20 times and converged to a
low gradient, typically approx. 2 kJ/mol lower in energy
than the next most frequently found structure, confirming
it as a global energy minimum.
21. http://www.schrodinger.com
22. Belvisi, L.; Carugo, O.; Poli, G. J. Mol. Struct. 1994, 318,
189–202.
23. The RMS between the calculated global minimum of cis-21
and the X-ray structure is 0.5752, the major difference
being the orientation of the phenyl rings.
24. Found cis-trans relative steric energy differences (kJ/mol)
11: −14.5; 19: −12.3; 20: −22.0; 21: −14.9; 22: −9.9; 23:
−14.5.
25. Precipitation of black palladium was constantly observed
a few hours after completion of the reaction. As a
consequence, further experiments at longer reaction times,
in order to reach the final thermodynamic equilibrium,
were not attempted.
1
yellow oil (188 mg, 98%). H NMR (CDCl3, 400 MHz):
l cis compound 1.60 (d, 3J=7 Hz, 3 H), 2.44 (s, 3H), 3.17
(dd, 3J=4.5 Hz, 2J=13 Hz, 2H), 4.36 (d, 2J=14 Hz, 1H),
2
3
4.39 (m, 1H), 4.54 (d, J=14 Hz, 1H), 4.59 (q, J=7 Hz,
1H), 5.06 (dd, 2J=1.5 Hz, 3J=17 Hz, 1H), 5.17 (dd,
2J=1.5 Hz, 3J=10.5 Hz, 1H), 5.76 (ddd, 3J=5 Hz,
3
3J=10.5 Hz, J=17 Hz, 1H), 6.98 (m, 2H), 7.24–7.32 (m,
3
5H), 7.70 (m, 2H); trans compound: 1.63 (d, J=6.5 Hz,
3
2
3H), 2.41 (s, 3H), 3.13 (dd, J=3.5 Hz, J=13 Hz, 1H),
3.60 (dd, 3J=3.5 Hz, 2J=13 Hz, 1H), 4.28 (d, 2J=14.5 Hz,
1H), 4.37 (q, 3J=6.5 Hz, 1H), 4.57 (m, 1H), 4.71 (d,
3
3
2J=14.5, 1H), 5.04 (d, J=10 Hz, 1H), 5.15 (d, J=16.5
Hz, 1H), 5.47 (dd, 3J=7 Hz, 3J=10 Hz, 3J=16.5 Hz, 1 H),
7.18 (m, 2H), 7.23–7.32 (m, 5 H), 7.67 (m, 2H); 13C NMR
(CDCl3, 100 MHz): l cis compound 21.0, 21.5, 46.9, 49.9,
52.8, 53.9, 117.8, 127.1, 127.7, 127.9, 128.4, 128.6, 128.7,
129.5, 129.9, 133.7, 135.7, 137.4, 143.6, 168.8; l trans
compound 21.3, 21.4, 47.3, 50.4, 53.2, 54.3, 118.3, 127.5,
128.2, 128.6, 129.0, 130.4, 136.1, 136.7, 144.3, 168.5.
11. (a) Inoue, Y.; Taguchi, M.; Hashimoto, H. Bull. Chem.
Soc. Jpn. 1985, 58, 2721–2722; (b) Takao, K.; Nigawara,
Y.; Nishino, E.; Takagi, I.; Maeda, K.; Tadano, K.;
Ogawa, S. Tetrahedron 1994, 50, 5681–5704.
12. (a) Bystrom, S. A.; Aslanian, R.; Ba¨ckvall, J. E. Tetra-
hedron Lett. 1985, 26, 1749; (b) Cerezo, S.; Core´s, J.;
Moreno-Man˜as, M.; Pleixats, R.; Roglans, A. Tetrahedron
1998, 54, 14869–14884; (c) Jumnah, R.; Williams, J. M. J.;
Williams, A. C. Tetrahedron Lett. 1993, 34, 6619–6622.
,
13. Connell, R. D.; Rein, T.; Akermark, B.; Helquist, P. J. Org.
Chem. 1988, 53, 3845–3849.
14. Sharp cis/trans isomer separation of the resulting piperazi-
nones turned out to be rather hard, and traces of the
undesired isomer often contaminated the selected one. As
a consequence, measurement of the optical rotations values
of these compounds was not attempted at this stage, except
for the crystalline cis-21.
15. Check for the full preservation of the enantiomeric