A versatile synthetic route to quinoxaline, pyrazine and 1,2,4-triazine substituted α-amino acids from vicinal tricarbonyls

Article abstract of DOI:10.1039/a909722h

A range of novel heterocyclic α-amino acids has been synthesised by the reaction of diamines and amidrazones with α-amino acid vicinal tricarbonyl reactive substrates. The Royal Society of Chemistry 2000.

Full text of DOI:10.1039/a909722h

View Article Online / Journal Homepage / Table of Contents for this issue
A versatile synthetic route to quinoxaline, pyrazine and 1,2,4-
triazine substituted ꢀ-amino acids from vicinal tricarbonyls
Robert M. Adlington, Jack E. Baldwin,* David Catterick and Gareth J. Pritchard
1
The Dyson Perrins Laboratory, University of Oxford, South Parks Road, Oxford,
UK OX1 3QY. E-mail: jack.baldwin@chem.ox.ac.uk
Received (in Cambridge, UK) 9th December 1999, Accepted 21st December 1999
A range of novel heterocyclic ꢀ-amino acids has been
synthesised by the reaction of diamines and amidrazones
with ꢀ-amino acid vicinal tricarbonyl reactive substrates.
Conversion of 1 to the ketophosphorane 3 was thus carried
out, in good yield, by a DCC–DMAP coupling reaction
with (tert-butoxycarbonylmethylene)triphenylphosphorane in
DCM. Ozonolysis of 3 however generated a product 4 which
existed as a mixture of two species at equilibrium, the vicinal
tricarbonyl α-amino acid 4a and the 5-membered ring closed
species 4b, which resulted from attack of the α-amino group at
the highly electrophilic central carbonyl (Scheme 1).
Initially unperturbed by the equilibrium existence of 4, trial
cyclocondensations with ethylenediamine were carried out, in
order to attempt to generate a pyrazine product. These proved
to be unsuccessful at a range of temperatures and in a range
of solvents, with rapid consumption of 4 being accompanied
by the formation of complex unidentifiable products. Conden-
sation reactions with o-phenylenediamine were next attemp-
ted,13 however instead of the desired quinoxaline substituted
protected amino acid being obtained the benzimidazole 5
was isolated. It is believed that this product is formed by initial
condensation of the diamine onto the ketonic carbonyl of 4b
followed by ring cleavage and aromatisation to the benz-
imidazole. This results in opening of the 5-membered ring, with
subsequent elimination of oxoacetic acid tert-butyl ester to
generate 5 (Scheme 2).
In an effort to effect pyrazine formation with the substrate 4
it was decided to simply prevent the formation of the equi-
librium species 4b by α-amino di-Boc protection.14,15 N-Di-Boc
protection of the orthogonally protected aspartate 6, a pre-
cursor in the synthesis of 1, was thus carried out by reaction
of a concentrated acetonitrile solution of 6 with an excess of
di-tert-butyl dicarbonate and DMAP, which generated 7 in
almost quantitative yield. The β-benzyl ester was removed by
hydrogenation over Pd–C to give 8, then acid activation and
reaction with (tert-butoxycarbonylmethylene)triphenylphos-
phorane generated 9. Subsequent ozonolysis of 9 gave the
Amino acids play a significant role in the synthesis of novel
drug candidates, with the use of non-proteinogenic and
unnatural amino acids becoming increasingly important.1–5 In
view of this and of the biological and toxicological properties
displayed by many heterocyclic substituted non-proteinogenic
α-amino acids we have recently developed versatile synthetic
routes towards compounds of this type.6–8 This was achieved
by the incorporation of alkynyl ketone reactive cores into the
side chains of suitable amino acids allowing the formation
of a range of α-amino acid alkynyl ketones capable of varied
heterocyclic construction.
Many groups however have the potential to be suitable
reactive cores for introduction into amino acid side chains and
subsequent facile heterocyclic construction. It was therefore
decided to attempt the formation of new amino acid reactive
substrates by incorporation of a vicinal tricarbonyl moiety into
our aspartate and glutamate side chains. The vicinal tricarbonyl
is a bis-acceptor reactive building block which behaves as a
potent electrophile. As such it has been used in the synthesis of
many different heterocyclic structures, with Wasserman and
co-workers being instrumental in this field.9–12 One route to
vicinal tricarbonyls, developed by Wasserman, involves con-
version of carboxylic acids into ketophosphoranes followed
by ozonolysis oxidation.9 Facile formation of the required
functionalised α-amino acid reactive substrates thus appeared
possible by exploiting the selectively protected amino acids
α-tert-butyl ester N-(tert-butoxycarbonyl)--aspartate 1 and α-
tert-butyl ester N-(tert-butoxycarbonyl)--glutamate 2 which
we had utilised in the formation of alkynyl ketone α-amino
acids.6–8
Scheme 1 Reagents and conditions: i, (tert-butoxycarbonylmethylene)triphenylphosphorane, DCC, DMAP, DCM, 0 ЊC, 70% (3), 63% (17); ii, O3,
DCM, Ϫ78 ЊC, 80% (4), 75% (19).
J. Chem. Soc., Perkin Trans. 1, 2000, 299–302
This journal is © The Royal Society of Chemistry 2000
299

View Article Online
Scheme 2
Next, in an attempt to diversify our methodology towards
homologous amino acids, it was decided to expand to the
glutamate system. Initially γ-benzyl α-tert-butyl diester N-(tert-
butoxycarbonyl)--glutamate 14 underwent N-di-Boc protec-
tion, to give 15, followed by hydrogenolysis to 16, in excellent
overall yield (Scheme 3). Both the mono-Boc and di-Boc,
glutamates 2 and 16, then underwent DCC–DMAP coupling
reactions with (tert-butoxycarbonylmethylene)triphenylphos-
phorane to give the ketophosphoranes 17 and 18, which with
subsequent ozonolysis oxidations generated the reactive sub-
strates 19 and 20. As with the aspartate system the mono-Boc
substrate 19 was found to exist as an equilibrium mixture of
ring opened vicinal tricarbonyl 19a and cyclic ketone species
19b, and the di-Boc system 20 was found to exist exclusively as
the vicinal tricarbonyl (Schemes 2 and 3).
Analogous reactions of 19 with ethylenediamine (followed by
Pd/C oxidation), phenylenediamine and S-methyl isothiosemi-
carbazide now resulted in high yielding generation of the
pyrazine, quinoxaline and 1,2,4-triazine substituted protected
amino acids 21, 22 and 23a/b. These results indicated that the
equilibrium existence of 19 is not problematic (19a > 19b), as in
the aspartate system, which we believed was a consequence
of preference for the ring closed form in the aspartate series,
i.e. 4b > 4a. The reaction of 20 with phenylenediamine then
allowed the quinoxaline amino acid 24 to be isolated in quan-
titative yield and cyclocondensation with S-methyl isothio-
semicarbazide generated the triazine amino acids 25a/b in high
yield (Scheme 4).
Deprotection of the protected amino acids 5, 12 and 24 was
carried out by treatment with TFA–anisole, with the free amino
acids 26 and 27 being obtained by ion-exchange chrom-
atography whilst 28 was isolated, after trituration with diethyl
ether, as its TFA salt (Scheme 5). Deprotection of the protected
amino acids 11, 13a/b, 21 and 25a/b was then carried out by
their azeotropic distillation with 1.5 equivalents of TsOHؒH2O–
PhMe, as TFA–anisole conditions had led to decomposition.
The amino acids 29, 30a/b, 31 and 32a/b were thus isolated as
their tosylate salts with an addtional 0.5 equivalents of TsOH,
in high yields, as air and moisture sensitive species (Scheme 5).
In conclusion, by variation of the reactive core to the vicinal
tricarbonyl we have been able to access a range of novel hetero-
cyclic substituted non-proteinogenic amino acids.17 The equi-
librium existence of 4, although initially problematic, allowed
an interesting formation of the benzimidazolyl substituted β-
alanine 5, further investigation of which will be reported in due
course. This problem was then shown to be easily circumvented,
by simple addition of a second N-Boc protecting group, or not
to be a consideration upon expanding to the glutamate system,
from which the desired pyrazine, quinoxaline and 1,2,4-triazine
amino acids were readily obtained. Further investigations
towards novel heterocyclic systems will be reported.
Scheme 3 Reagents and conditions: i, Boc2O, DMAP, MeCN; ii, H2,
(10%) Pd–C, EtOH–H2O (19:1); iii, (tert-butoxycarbonylmethylene)-
triphenylphosphorane, DCC, DMAP, DCM, 0 ЊC; iv, O3, DCM,
Ϫ78 ЊC.
vicinal tricarbonyl reactive substrate 10 in good overall yield
with no ring closed equilibrium species being observed
(Scheme 3).
Reaction of 10, in ethanol, with ethylenediamine now
resulted in a rapid consumption of starting material along with
the appearance of a single product, which was not isolated
but expected to be the dihydropyrazine. Oxidation was then
achieved by the addition of Pd/C, coupled with heating the
reaction to reflux and the desired pyrazine substituted amino
acid 11 was obtained in high yield (Scheme 4). A cyclocon-
densation of 10 with o-phenylenediamine, in refluxing ethanol,
then resulted in quantitative conversion to the quinoxaline
substituted amino acid 12 (Scheme 4).
In order to investigate the possibility of 1,2,4-triazine form-
ation,16 a condensaton reaction of 10 with S-methyl iso-
thiosemicarbazide was then attempted. This reaction also
proved successful with the desired triazines 13a/b being isolated,
in high yield, as a partially separable 1:1 mixture of regio-
isomers (Scheme 4).
300
J. Chem. Soc., Perkin Trans. 1, 2000, 299–302

View Article Online
Scheme 4 Reagents and conditions: i, H2NCH2CH2NH2, EtOH, rt; ii, (10%) Pd–C, EtOH, reflux; iii, o-C6H4(NH2)2, EtOH, reflux; iv, H2NHN-
(CSMe)NHؒHI, Pri2NEt, DCM, reflux.
J. Chem. Soc., Perkin Trans. 1, 2000, 299–302
301

View Article Online
Acknowledgements
We thank the EPSRC for a studentship to D. C. and the EPSRC
mass spectrometry service (Swansea) for high resolution mass
spectra.
References
1 R. O. Duthaler, Tetrahedron, 1994, 50, 1539.
2 J. Jones, The Chemical Synthesis of Peptides, Oxford University
Press, Oxford, 1991.
3 D. Ward, Peptide Pharmaceuticals, Open University Press, Milton
Keynes, 1991.
4 Organic Synthesis Highlights III, eds. J. Mulzer and H. Waldman,
Wiley-VCH, New York, 1998, p. 366–373.
5 G. A. Rosenthal, Plant Nonprotein Amino and Imino Acids
Biological, Biochemical and Toxicological Properties, Academic
Press, New York, 1982; G. A. Rosenthal and E. A. Bell, “Naturally
Occurring Toxic Nonprotein Amino Acids”, in Herbivores – Their
Interaction with Secondary Plant Metabolites, Academic Press,
New York, 1979, p. 353.
6 R. M. Adlington, J. E. Baldwin, D. Catterick and G. J. Pritchard,
Chem. Commun., 1997, 1757.
7 R. M. Adlington, J. E. Baldwin, D. Catterick and G. J. Pritchard,
J. Chem. Soc., Perkin Trans. 1, 1999, 855.
8 R. M. Adlington, J. E. Baldwin, D. Catterick, G. J. Pritchard
and Lam T. Tang, J. Chem. Soc., Perkin Trans. 1, following
paper.
9 H. H. Wasserman, D. S. Ennis, C. A. Blum and V. M. Rotello,
Tetrahedron Lett., 1992, 33, 6003.
10 H. H. Wasserman, R. Amici, R. Frechette and J. H. van Duzer,
Tetrahedron Lett., 1989, 30, 869.
11 H. H. Wasserman and G. M. Lee, Tetrahedron Lett., 1994, 35,
9783.
12 H. H. Wasserman, J. H. van Duzer and C. B. Vu, Tetrahedron Lett.,
1990, 31, 1609.
13 M. Murakami, H. Masuda, T. Kawano, H. Nakamura and Y. Ito,
J. Org. Chem., 1991, 56, 1.
14 L. Grehn and U. Ragnarsson, Angew. Chem., Int. Ed. Engl., 1985,
24, 510.
15 K. Gunnarsson, L. Grehn and U. Ragnarsson, Angew. Chem.,
Int. Ed. Engl., 1988, 27, 400.
16 H. Neunhoeffer, D. Reichel, B. Cullman and I. Rehn, Liebigs Ann.
Chem., 1990, 631.
17 The enantiomeric purity of the amino acids was determined by
Mosher’s amide formation (di-Boc deprotection was carried out
by azeotropic distillation with TsOHؒH2O–PhMe), with subsequent
19F NMR analysis of the resulting diastereoisomers indicating an
enantiomeric excess of greater than 98%; J. A. Dale, D. L. Dull and
H. S. Mosher, J. Org. Chem., 1969, 34, 2543.
Communication a909722h
Scheme 5 Reagents and conditions: i, TFA, anisole; ii, Dowex 50X8-
100 ion-exchange resin; iii, TsOHؒH2O (1.5 equiv.), toluene, azeotropic
distillation.
302
J. Chem. Soc., Perkin Trans. 1, 2000, 299–302