Synthesis of Cα methylated carboxylic acids: Isosteres of arginine and lysine for use as N-terminal capping residues in polypeptides

Article abstract of DOI:10.1016/j.tetlet.2005.08.056

Replacement of the N-terminal α-amine with the isosteric methyl functionality in bioactive peptides can influence various pharmacokinetic parameters, including hydrophobicity and stability. C^α methylated amino acid analogues are thus of great interest to expand the repertoire of nonnatural synthons available as N-terminal 'capping' residues for peptide-based drug design. Several novel arginine and lysine analogues stereoselectively modified in the C^α position with a methyl group in place of the α-amine were prepared.

Full text of DOI:10.1016/j.tetlet.2005.08.056

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 7007–7009
Synthesis of Ca methylated carboxylic acids: isosteres
of arginine and lysine for use as N-terminal
capping residues in polypeptides
Kevin S. Orwiga and Thomas A. Dixa,b,
*
aDepartment of Pharmaceutical Sciences, Medical University of South Carolina, 280 Calhoun Street, Charleston, SC 29425, USA
bArgolyn Bioscience Inc., 2741 Speissegger Dr., Suite 202, North Charleston, SC 29405, USA
Received 18 July2005; revised 9 August 2005; accepted 11 August 2005
Abstract—Replacement of the N-terminal a-amine with the isosteric methyl functionality in bioactive peptides can influence various
pharmacokinetic parameters, including hydrophobicity and stability. Ca methylated amino acid analogues are thus of great interest
to expand the repertoire of nonnatural synthons available as N-terminal ÔcappingÕ residues for peptide-based drug design. Several
novel arginine and lysine analogues stereoselectively modified in the Ca position with a methyl group in place of the a-amine were
prepared.
Ó 2005 Elsevier Ltd. All rights reserved.
Lack of stabilityin biological matrices is a keyobstacle
in developing peptides as therapeutic entities. One ap-
proach used to address this issue is derivatization of
peptide N-termini, for example byN a-acetylation or
Na-methylation.1,2 Such modification provides resis-
tance to cleavage byaminopeptidases, the predominant
route of degradation for peptides containing an unpro-
tected N-terminal amine.3–5 Previous work in this labo-
ratoryhas produced a libraryof side-chain alkylated
analogues of the cationic amino acids L-arginine (Arg)
and L-lysine (Lys).6–9 Recently, a subset of these ana-
logues featuring an a-methyl group substituted for the
a-amine has been synthesized10 and incorporated at
the N-terminus of peptides.11 The Ca methyl group is
virtuallyisosteric to the N a ammonium ion, and its sub-
stitution removes the positive charge, therebyincreasing
the overall hydrophobicity of the peptide in which the
replacement is made. The resulting compounds shared
dramaticallyincreased biological stabilityand, in certain
cases, increased central nervous system activity follow-
ing oral administration.11,12 Demonstration of the
substantial biological effects resulting from stereochem-
tion of enantiomericallypure a-methyl acids of great
interest. Accordingly, synthesis of the Arg analogue
(2S)-5-guanidino-2-methylvaleric acid (1) and the orni-
thine (Orn), Lys, and homolysine (HLys) analogues
(2S)-5-amino-2-methylvaleric acid (12a), (2S)-6-amino-
2-methylhexanoic acid (12b), and (2S)-7-amino-2-hepta-
noic acid (12c), respectively, are reported herein.
The initial attempt at the synthesis of 1 followed the
method of Hadden et al.10 This procedure utilizes the
Evans chiral auxiliary, (S)-(À)-4-benzyl-2-oxazolidinone
(XP), to effect stereoselective introduction (>95% enan-
tiomeric purity)10 of the methyl group at the a-position
on the appropriate x-bromo acid (Scheme 1). However,
in contrast to previous reports,10,13 removal of the chiral
auxiliaryfrom 2a following generation of the stereocen-
ter resulted in unacceptablylow yields (10–20%). Upon
further investigation, the cyclic lactone side product (4)
was identified following saponification along with 3a
and the free chiral auxiliary.
It appears that base-induced cyclization to produce 4 is
highlyfavorable and competes with formation of the
protonated form of the acid following carboximide
hydrolysis, thus the unprotected carboxylate of 3a may
have limited utilityas a substrate in this reaction. While
x-bromo acids work well with respect to the preparation
of a-azido acids,13 the a-methyl group employed here
seems to destabilize the carboxylate during the ensuing
a
icallydefined C methyl incorporation makes produc-
Keywords: Amino acid analogues; Isostere; Arginine; Lysine; N-termi-
nal capping.
*
Corresponding author. Tel.: +1 843 876 5092; fax: +1 843 792
1617; e-mail: dixta@musc.edu
0040-4039/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.08.056

7008
K. S. Orwig, T. A. Dix / Tetrahedron Letters 46 (2005) 7007–7009
Scheme 1.
removal of the chiral auxiliary. Further, similar prob-
lems were observed in the subsequent derivatization of
isolated 3a where a clean product in acceptable yield
was not obtained under a varietyof conditions.
iary. Substitution of the bromine can be accomplished at
anystep prior to removal of the chiral auxiliarywithout
loss of efficiency, having been performed on the 5-bromo
counterparts of 9a and 10a with equal effectiveness (96%
and 95% yields, respectively). The use of this azide
substitution strategycircumvents the problems of
cyclization noted here and observed with similar
compounds.8,10 With the azide in the x-position, trans-
formation from 10a–c to 11a–c proceeded as expected.
Azide reduction then afforded the Orn, Lys, and HLys
analogues (12a–c) in excellent yields from 11a–c, allow-
ing guanylation with S-methylthiourea to give 1 from
12a in 65% overall yield (Scheme 3).
Replacing the x-bromo moietyfrom Scheme 1 with a
Boc-protected amine was also ineffective (Scheme 2). In-
stead of the desired enolate formation, the amide proton
of 5 is abstracted under the conditions of the methyl-
ation reaction, allowing displacement of the chiral auxil-
iarywith concomitant formation of the six-membered
lactam, 6. Methylation then occurs on the free auxiliary
upon methyl iodide addition to generate 7 as the major
product.
Peptides featuring N-terminal (S)-Ca-azido homolysine
and (S)-Ca-methyl homolysine residues have been pre-
pared.12,11 However, in these cases either (2S)-2-azido-
7-bromoheptanoic acid or (2S)-7-bromo-2-methylhepta-
noic acid was coupled to the peptidyl-resin directly and
the bromine was substituted byammonia in approxi-
mately70% iyeld following cleavage from the solid
phase. It is demonstrated herein that substitution of
the bromine with the azide group is both more efficient
and widelyuseful than the previous approach. The pres-
ent work constitutes the first report of (S)-Ca methyl
amino acid analogues of Arg, Orn, Lys, and HLys as
monomers for solid-phase peptide synthesis. The HCl
The synthesis strategy was revised further by substitut-
ing the bromo moietyof the x-bromo acids 2a–c with
the azide ion as the initial step in the synthesis (Scheme
3). Manyprocedures for the introduction of azides onto
various backbone structures require complex workups,
extensive reaction times, elevated temperatures, and
result in low yields.14–16 However, the straightforward
method of Alvarez and Alvarez17 was extended to pro-
duce the primaryazido acids 8a–c, usuallyin quantita-
tive yield, at ambient temperature. This approach
effectivelymasks the amine during construction of the
stereocenter and subsequent removal of the chiral auxil-
Scheme 2.
Scheme 3.

K. S. Orwig, T. A. Dix / Tetrahedron Letters 46 (2005) 7007–7009
7009
salt of 1 and Boc-protected 12a–c have been coupled to
peptidyl-resin by standard methods without difficulty
and the resulting peptides and their corresponding prop-
erties will be reported elsewhere. Scheme 3 also provides
an efficient route to the production of 2(S)-methyl ver-
sions of a number of other side-chain alkylated Arg ana-
logues via reaction of 12a with the requisite isothiourea
salts.
this article can be found, in the online version at
doi:10.1016/j.tetlet.2005.08.056.
References and notes
1. Brown, J. L.; Roberts, W. K. J. Biol. Chem. 1976, 251,
1009–1014.
2. Machida, R.; Tokumura, T.; Sasaki, A.; Tsuchiya, Y.;
Abe, K. J. Chromatogr. A 1990, 534, 190–195.
3. Delange, R. J. The Enzymes, 3rd ed.; Academic Press:
New York, 1971; Vol. 3.
4. Lugrin, D.; Vecchini, F.; Doulut, S.; Rodriguez, M.;
Martinez, J.; Kitabgi, P. Eur. J. Pharmacol. 1991, 205,
191–198.
5. Kokko, K. P.; Dix, T. A. Anal. Biochem. 2002, 308, 34–
41.
6. Kennedy, K. J.; Lundquist, I.; Joseph, T.; Simandan, T.
L.; Beeson, C. C.; Dix, T. A. Bioorg. Med. Chem. Lett.
1997, 7, 1937–1940.
7. Kennedy, K. J.; Simandan, T. L.; Dix, T. A. Synth.
Commun. 1998, 28, 741–746.
8. Lundquist, J. T.; Orwig, K. S.; Dix, T. A. J. Org. Chem.
1999, 64, 9265–9267.
9. Kennedy, K. J.; Lundquist, J. T. I.; Simandan, T. L.;
Kokko, K. P.; Beeson, C. C.; Dix, T. A. J. Pept. Res. 2000,
55, 348–358.
The novel enantiomericallypure ( S)-Ca methylated ana-
logues of Arg and Lys reported here lack the a-amine
and therefore will resist aminopeptidase degradation
when incorporated at the N-terminus of peptides.3,18
These compounds are useful as N-terminal capping res-
idues that simultaneouslyincrease peptide stabilityand
hydrophobicity resulting from the loss of the charged
amine. Such a strategyimproves peptide biostability
without drasticallyaltering the structure of the mole-
cule, leaving the backbone and amino acid side-chain
functional groups free to mediate the interactions re-
quired for full biological activityof the peptide. This
Ca-methyl substitution should find widespread use in
peptide drug design since it is essentiallyisosteric with
the Na-ammonium ion and thus without the adverse
steric effects possible with Na-acetylation and Na-
methylation.
10. Hadden, M.; Kokko, K.; Dix, T. Synth. Commun. 2005,
35, 1675–1680.
11. Hadden, M. K.; Orwig, K. S.; Mazella, J.; Dix, T. A.
Neuropharmacology, in press.
12. Kokko, K. P.; Hadden, M. K.; Orwig, K. S.; Mazella, J.;
Dix, T. A. J. Med. Chem. 2003, 46, 4141–4148.
13. Lundquist, J. T. I. V.; Dix, T. A. Tetrahedron Lett. 1998,
39, 775–778.
Acknowledgements
This research was supported byNIH grant MH-65088.
K.S.O. is an American Foundation for Pharmaceutical
Education predoctoral scholar. The authors thank
14. Hassner, A.; Fibiger, R.; Andisik, D. J. Org. Chem. 1984,
49, 4237–4244.
Drs. Craig Beeson, Erika Bullesbach, and John Oatis
¨
for useful discussions and technical assistance. The
MUSC NMR Facilitywas used in support of this work.
15. Dyke, J. M.; Groves, A. P.; Morris, A.; Ogden, J. S.; Dias,
A. A.; Oliveira, A. M. S.; Costa, M. L.; Barros, M. T.;
Cabral, M. H.; Moutinho, A. M. C. J. Am. Chem. Soc.
1997, 119, 6883–6887.
16. Tornoe, C. W.; Davis, P.; Porreca, F.; Meldal, M. J. Pept.
Sci. 2000, 6, 594–602.
Supplementary data
17. Alvarez, S. G.; Alvarez, M. T. Synthesis 1997, 413–
414.
18. Luciani, N.; Marie-Claire, C.; Ruffet, E.; Beaumont, A.;
Roques, B. P.; Fournie-Zaluski, M. C. Biochemistry 1998,
37, 686–692.
1
Supplementarydata: 400 MHz H NMR for the mixture
of 3a and 4, 1H NMR. HSQC and HMBC spectra for 7,
8a–c, 9a–c, 10a–c, 11a–c, 12a–c, and 1 as well as exper-
imental procedures. Supplementarydata associated with