Synthesis of cyclic endiamino peptides

Article abstract of DOI:10.1021/ja050299r

Several building blocks for endiamino peptides, as well as several cyclic endiamino peptides themselves, and pyrazin-6-one, which embodies the endiamino group, were prepared. The variety of synthesized compounds shows the potential of this synthesis in th

Full text of DOI:10.1021/ja050299r

Published on Web 05/06/2005
Synthesis of Cyclic Endiamino Peptides
Doron Pappo, Maida Vartanian, Steffen Lang, and Yoel Kashman*
School of Chemistry, Tel-AViV UniVersity, Ramat AViV 69978, Israel
Received January 17, 2005; E-mail: kashman@post.tau.ac.il
Cyclic endiamino peptides, represented by callynormine A
Like dipeptides, which have a high tendency to cyclize to diketo-
piperazines, enol-tosylates of FGly amino acid methyl esters (e.g.
compound 2, Scheme 1) may also give the corresponding diketo-
(Figure 1),1 are a new class of heterodetic peptides embodying the
R-amino-â-aminoacrylamide functionality (instead of the ester
group of depsipeptides). It is suggested that the endiamino group
is derived from the condensation of the formyl group of FGly2,3
and the amino group of another amino acid (Ile in the case of
compound 1).1 R-Formylglycine (FGly) was recently reported for
both eukaryotic and prokaryotic sulfatasesslocated within the
catalytic site of the enzyme.2,3 It was shown that the formylglycine
is generated by oxidation of cysteine or serine and, furthermore,
that the FGly hydrate is covalently sulfated2 or covalently phos-
phorylated3 during catalysis. To the best of our knowledge, there
are no reports of natural compounds embodying the endiamino
group. Indeed, synthetic linear compounds with this group are
known.4,5 The endiamino group is of special interest for the
synthesis of biomimetic cyclic peptides, as it is expected to
introduce additional rigidity into their structure. Hereafter we report
the first synthesis of several cyclic endiamino peptides, including
2-(1H)-pyrazinone, which, formally, is the smallest member of this
new group. We also demonstrate the preparation of endiamino-
containing building blocks for biomimetic peptides. FGly is very
unstable. However, its enol-tosylate derivative, prepared from serine,
is stable and acts with amino groups as an aldehyde to produce the
R-amido-â-aminoacrylamide functionality.4
Scheme 1
piperazine, rather than the cyclic six-membered endiamino peptide,
carrying the enol-tosylate functionality, an anchor for the connection
of additional amino acids. In this event, deprotection of the Boc
group from the enol-tosylate derivative of N-Boc-Ile-FGly methyl
ester (2) indeed afforded the diketo-piperazine (3), rather than the
endiamino cyclic peptide (Scheme 1). Compound 3 can potentially
react further with amines to give interesting synthons.
In fact, a six-membered endiamino peptide (e.g. 6) could be
obtained by first preparing the endiamino functionality followed
by internal amidation, from the reaction of the deprotected amino
group of FGly with the ester group of the second amino acid. An
example is the synthesis of compound 6 from 4 via 5 (Scheme 1).
As expected, compound 6 readily oxidizes to the 2-(1H)-pyrazinone
7.7 Elimination of the chiral center of the leucine of 6 brought about
the expected collapse of the two doublets of the leucine methyls
(δH 0.92 and 0.95 ppm) into a single doublet (δH 0.96 ppm) in 7.
The latter synthesis is a new route to the preparation of substituted
2-(1H)-pyrazinones,8 known as secondary metabolites, produced
by Aspergillus and Streptomyces and was recently carried out in a
different way.9
All-L cyclotripeptides with all R-amino acids are almost impos-
sible to construct, as the corresponding cyclodimers, the cyclo-
hexapeptides, are the main products.6 This was also the case with
the enol-tosylate of tripeptide 8, which gave the cyclic dimeric
endiamino hexapeptide 9 together with the corresponding trimeric
nonapeptide 1010 (Scheme 2). Similarly, tetrapeptide 11 afforded
the symmetric dimeric cyclic endiamino octapeptide 1211 (Scheme
2). From hexapeptide 13, it was already possible to obtain the
Figure 1. Callynormine A (1).
Cyclic peptides and depsipeptides have been characterized in
many organisms and show a wide spectrum of biological activities.
Hence, the cyclic peptides are promising lead compounds for
potential drugs. The special quality of these cyclic peptides stems,
inter alia, from the reduction in conformational freedom brought
about by the cyclization, which is expected to result in higher
receptor binding affinities. Replacing an amide bond, or the ester
group of depsipeptides, with the endiamino functionality is
anticipated to introduce additional rigidity in the cyclic endiamino
peptides.
The macrocyclization step, which is known to be the yield-
determining step for cyclic peptides, can, in the case of the cyclic
endiamino peptides, be achieved by the formation of either an amide
or the endiamino functionality.1 Examples of both routes follow.
It could also be expected that the tendency to cyclize will change
with the size of the ring, as is known for cyclic peptides.6
9
7682
J. AM. CHEM. SOC. 2005, 127, 7682-7683
10.1021/ja050299r CCC: $30.25 © 2005 American Chemical Society

C O M M U N I C A T I O N S
Scheme 2
Furthermore, a Z stereochemistry was established for the endiamino
group in all synthesized cyclic compounds, according to the low-
field resonance of the endiamino proton (δH 7.2-7.3 ppm) and an
NOE between the latter proton and the neighboring ester OCH3
group. Suitable for compound 14, and for the other cyclic endiamino
peptide, were the observed 1H13C- and 1H15N-HMBC correlations,
as shown for 14 in Scheme 3.12
For the reaction of the enol-tosylate of the FGly group in the
various peptides with amines in MeOH, we suggest an addition-
elimination mechanism rather than the mechanism reported earlier;4
experimental results will be reported in detail elsewhere.
In summary, the above results describe the first synthesis of
cyclic endiamino peptides and endiamino building blocks, where
the preparations of compounds 9, 10, 12, and 14 demonstrate the
possibility of incorporation of the endiamino functionality in cyclic
peptides, with the endiamino group replacing the lactone of
depsipeptides, further reducing the conformational freedom. In
addition, the applied methodology enabled the synthesis of interest-
ing endiamino building blocks and pyrazin-6-ones.
Acknowledgment. We thank Dr. Amira Rudi for her help with
NMR measurements and Dr. Ayelet Sacher (of the Maiman Institute
for Proteome Research, Tel Aviv University) for performing the
electrospray mass spectra measurements.
Supporting Information Available: Experimental procedures and
spectral characterization data for all compounds and 1H NMR and 13
C
NMR spectra for selected compounds. This material is available free
of charge via the Internet at http://pubs.acs.org.
References
(1) Berer, N.; Rudi, A.; Goldberg, I.; Benayahu, Y.; Kashman, Y. Org. Lett.
2004, 6, 2543-2545.
(2) Dierks, T.; Miech, C.; Hummerjohann, J.; Schmidt, B.; Kertesz, M. A.;
von Figura, K. J. Biol. Chem. 1998, 273, 25560-25564.
(3) Chruszcz, M.; Laidler, P.; Monkiewicz, M.; Ortlund, E.; Lebioda, L.;
Lewinski, K.; J. Inorg. Biochem. 2003, 96, 386-392.
desired monomeric cyclic endiamino peptide 1412 (Scheme 3).
Characteristic of the endiamino group in 1 and all synthesized cyclic
compounds are their NMR resonances: e.g. δH 7.29 ppm (d, J )
13.5 Hz) and δC 145.2 ppm (d) for the dCHNH- group in 14.
(4) Nakazawa, T.; Suzuki, T.; Ishii, M. Tetrahedron Lett. 1997, 38, 8951-
8954.
(5) Singh, R.; Zhou, N.; Reddy, A. V. N.; Thomas, G.; Ding, Q. Kaleta, J.;
Micetich, R. G.; Whittaker, M. U.S. Patent 6,635,621 B1, 2003.
(6) Davis, J. S. J. Peptide Sci. 2003, 9, 471-501.
Scheme 3
(7) All new structures were confirmed by HRMS and 1D and 2D NMR
experiments.
(8) MacDonald, J. C.; Bishop, G. G.; Mazurek, M. Tetrahedron 1976, 32,
655-660.
(9) Matsushita, H.; Lee, S.-H.; Yoshida, K.; Clapham, B.; Koch, G.;
Zimmermann, J.; Janda, K. D. Org. Lett. 2004, 6, 4627-4629.
(10) According to the NMR data both 9 and 10 are symmetric, exhibiting a
single tripeptide moiety. Both give the suitable HRESMS data for the
suggested structure (m/z 561.2890 (M + Na)+ and 830.4018 (M + Na)+
for 9 and 10, respectively). The characteristic vinyl endiamine proton
(originating from the FGly acid) resonances at δ 7.33 (d, J ) 14 Hz) and
7.26 (d, J ) 14 Hz) for 9 and 10, respectively.
(11) Compound 12 exhibited the proper HRESMS (m/z 563.2086 (M + Na)+)
and 1H NMR data, confirming a symmetric dimeric structure.
(12) The structure of 14 was established from the 1H,13C NMR and HRESMS
(m/z 573.3002 (MH)+) data as well from the δN values and their NH
correlations derived from NH-HMBC and NH-HSQC-TOCSY experiments.
JA050299R
9
J. AM. CHEM. SOC. VOL. 127, NO. 21, 2005 7683