Taking control of P1, P1′ and double bond stereochemistry in the synthesis of Phe-Phe (E)-alkene amide isostere dipeptidomimetics

Article abstract of DOI:10.1039/b616906f

A protocol for the stereocontrolled independent preparation of Phe-Phe trans-vinyl amide isostere dipeptideomimetics was developed based on a Wittig-type reaction. In the reaction two chiral building blocks were joined with excellent E-selectivity to give

Full text of DOI:10.1039/b616906f

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/obc | Organic & Biomolecular Chemistry
Taking control of P1, P1ꢀ and double bond stereochemistry in the synthesis of
Phe-Phe (E)-alkene amide isostere dipeptidomimetics†
Daniel Wiktelius and Kristina Luthman*
Received 20th November 2006, Accepted 1st December 2006
First published as an Advance Article on the web 14th December 2006
DOI: 10.1039/b616906f
A protocol for the stereocontrolled independent preparation
of both C-2 epimers of Phe-Phe trans-vinyl amide isostere
dipeptidomimetics has been devised based on a Wittig-type
reaction, in which two chiral building blocks were joined
with excellent E-selectivity to give compounds of the type
chain stereocentres (P1 and P1ꢀ attachment points) as well as
of the double bond. We wanted to find conditions to provide
good control of all three stereocentres, which could be used
also for the synthesis of similar derivatives. For versatility, we
desired a convergent approach employing chiral amino acid-
like building blocks to install the remote stereogenic centres
(Scheme 1).5 For simplicity, a Wittig-based reaction was chosen
for alkene formation and the P1 fragment was chosen to contain
the ylide. Full deprotection of the isosteres is rarely reported,1e
but we also wanted to ensure that mimetics without protective
=
PheW[(E)-CH CH]-PheOH.
(E)-Alkene dipeptidomimetics (e.g. 1, Scheme 1) are of interest as
building blocks in artificial pseudopeptides. They can be used to
probe the involvement of amide bond geometry and polarity in
biomolecular structure, recognition and function,1,2 since the (E)-
alkene closely resembles the natural amide with regard to bond
lengths, angles and rigidity but is devoid of hydrogen bonding
capability.2 We have for some time been interested in the chemical
and pharmacological properties of (E)-alkene dipeptidomimetics,
mainly Phe-Phe and Phe-Gly derivatives.3 In spite of their simple
structure, the synthesis of these (E)-5-aminopent-3-enoic acids can
be challenging. Over the years, several interesting routes to (E)-
alkene dipeptide isosteres have been devised.1e,3a,4 While many of
them display excellent stereoselectivity in double bond formation
and/or side chain installation steps, these syntheses are usually
quite complex, have not been applied to disubstituted alkene
derivatives, or have stereochemical weak spots. We have previously
used sulfonium ylide-based chemistry to synthesise derivatives of
1, where the key Julia olefination step installs the alkene with
good E/Z-selectivity;3,4c however, the procedure is laborious and
our data indicate loss of the stereochemical integrity at C-2.3a
An ideal synthetic route to disubstituted (E)-alkene isosteres
like 1 would confer stereocontrol of both C-5 and C-2 side
=
groups could be obtained. With PheW[(E)-CH CH]-PheOH
dipeptidomimetics as model compounds, we wanted to explore
the full utility and stereocontrol of the method by independently
synthesising both C-2 epimers of the target compounds. Thus,
we set out to prepare chiral phosphonium salt 5 (Scheme 2)
and chiral aldehydes 6 (vide infra). b-Amino phosphonium salts
have been prepared previously,6 but we were surprised to find
that carbamate-protected derivatives were scarce in the literature.
For convenient deprotection of the dipeptidomimetic products, we
wanted to use Boc-protection; however, the tert-butyl carbamate
was found to be unstable in the reaction between a Boc-protected
amino bromide derivative of phenylalaninol (i.e. Boc-analogue
of 4, Scheme 2) with triphenylphosphine using a variety of
solvents and temperatures. Similarly, the methylcarbamate- and
benzylcarbamate-protected amino bromide analogues afforded
only trace amounts of the desired product. In all reactions, the
carbamate was cleaved, and a phosphorous-containing adduct
was formed according to NMR spectroscopy, probably due to
the nucleophilic nature of the carbamate carbonyl oxygen. Instead,
trifluoroacetyl protection was tried since its oxygen is less electron-
rich. Bromide 4 was found to be perfectly stable in the reaction with
triphenylphosphine, giving phosphonium salt 5 in quantitative
yield (Scheme 2).
The P1ꢀ building blocks, the chiral aldehydes (R)-6 (82% ee)
and (S)-6 (84% ee), were synthesised in five steps from dimethyl
Scheme 1 Retrosynthesis of an (E)-alkene dipeptidomimetic (1) for a
Wittig-based synthetic route.
Department of Chemistry, Medicinal Chemistry, Go¨teborg University, SE-
412 96, Go¨teborg, Sweden. E-mail: luthman@chem.gu.se; Fax: +46 31
7723840; Tel: +46 31 7722894
† Electronic supplementary information (ESI) available: Experimental
procedures, characterisation data and NMR spectra for all new com-
pounds. See DOI: 10.1039/b616906f
Scheme 2 Synthesis of the phenylalanine-like P1 synthon; trifluoro-
acetyl-protected amino phosphonium salt 5.
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 603–605 | 603
©

View Article Online
Scheme 3 Independent assembly of Phe-Phe (E)-alkene dipeptidomimetics 10a–b from phosphonium salt 5 and aldehyde (S)-6 or (R)-6. Key
(for compounds 7–10), a: R1 = H, R2 = Bn; b: R1 = Bn, R2 = H.
benzyl malonate by a protocol employing lipase-catalysed desym-
metrisation or resolution as key steps, respectively. The details of
this chemoenzymatic synthesis are described elsewere.7 With both
enantiomers of aldehyde 6 at hand, we hoped to gain control of the
C-2 stereochemistry if the following reactions proceeded without
epimerisation.
In earlier work with compounds like 7–10,3c we have found that
the double bond may rearrange under forcing conditions, and this
prompted us to complete the synthesis using the mildest reagents
possible. The silyl protecting group was removed to yield alcohols
8, which were converted to carboxylic acids 9 by tandem Dess–
Martin/NaClO2 oxidation.
For the mimetics to be useful in the synthesis of pseudopeptides,
the N-protecting group should be interchangeable, and for this
reason we wanted to fully deprotect the acids 9 to give the salts 10.
We were surprised to find that the trifluoroacetamide functions
of acids 9 were exceptionally stable toward alkaline hydrolysis;
in the case of 9a the reaction took 30 days to complete with a
large excess of K2CO3 in H2O–MeOH. A variety of conditions
were investigated on a small scale to obtain a higher reaction
rate. While none of the alkali hydroxides or carbonates afforded
any appreciable amount of product in methanol or THF with
varying amounts water, barium hydroxide was found to cleave the
trifluoroacetamide of 9a efficiently.11 Ammonia in methanol was
somewhat less effective than barium hydroxide, but was the reagent
of choice due to the simplified work-up. With suitable conditions
for deprotection identified, the corresponding ammonium salts
could be isolated. After checking the diastereomeric ratio by
HPLC (vide supra), compounds 10a–b were purified by preparative
HPLC to full stereochemical purity.
The diastereomeric purity of the alkene derivatives 7–10
(Scheme 3) is of course dependent on the enantiomeric purity of
the aldehyde reactant. We and others have reported the prepara-
tion of a-chiral aldehydes like 6 in high enantiomeric purity,7,12
and thus, the current method may be an important addition
to stereocontrolled preparation of (E)-alkene dipeptidomimetics
in high de. Further studies of the generality of this method
involving different ylide and aldehyde coupling partners are
ongoing. The convergent building block approach presented in
this communication can offer versatility in that the peptidomimetic
product is assembled from two chiral fragments, which are easily
prepared. The key Wittig reaction proceeded with excellent E/Z-
selectivity and with full control of both side-chain stereocentres,
and the complexity of the overall sequence is low.
The assembly of dipeptidomimetics 10a–b commenced with a
Wittig reaction involving either enantiomer of aldehyde 6 and the
ylide derived from phosphonium salt 5 (Scheme 3). A standard
Wittig reaction between an a-branched aliphatic aldehyde and a
b-branched non-stabilised ylide can be expected to proceed with
Z-selectivity;8,9 however, the presence of the trifluoroacetamide
in 5 gives the possibility of double deprotonation, which may
influence the outcome of the reaction. Ylides containing anionic
groups (i.e. lithiated heteroatoms) in proximity to the reacting
centre are prone to E-selelectivity,9,10 if suitable reaction conditions
are provided. While this has been well established for a-, b- and
c-oxido-, carboxy- and amido-ylides in reactions with aromatic
aldehydes,10 we have found no account of reactions involving
lithiated amides and aliphatic aldehydes. Treatment of 5 with 2
equivalents of n-butyllithium at −78 ◦C, followed by addition of
6 and slow warming to room temperature followed by an aqueous
quench afforded the alkene products 7a–b in 55% and 57% yield,
respectively. To our delight, no Z-isomer could be detected in the
1H-NMR spectra.
Racemisation of the coupling partners or epimerisation of either
stereocentre of the product was of concern in this transformation,
and for this reason the reactants were thoroughly purified to
ensure a precise stoichiometry in the deprotonation. A less than
full enantiomeric purity of aldehydes 6 was advantageous in this
model study, since this allowed the detection of stereochemical
scrambling by NMR more conveniently. No racemisation or
epimerisation could be detected at any point in the synthesis of
10a (Scheme 3), and this was further affirmed by HPLC, since the
de of the final product 10a matched the ee of the starting aldehyde
(S)-6. In the case of 10b, at most 1% stereochemical scrambling
was detected.
Acknowledgements
Financial support was obtained from the Knut and Alice Wallen-
berg Foundation, and the Swedish Research Council. The authors
604 | Org. Biomol. Chem., 2007, 5, 603–605
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
thank Dr Lene Hyldtoft and ACADIA Pharmaceuticals AB,
Malmo¨, Sweden, for advice and the loan of equipment, and Prof.
Stig Allenmark for his interest in this work.
A. Nadzan, J. Org. Chem., 1991, 56, 2107; (e) A. C. Bohnstedt, J. V. N. V.
Prasad and D. H. Rich, Tetrahedron Lett., 1993, 34, 5217; (f) P. Wipf
and P. C. Fritch, J. Org. Chem., 1994, 59, 4875; (g) P. Wipf and T. C.
Henninger, J. Org. Chem., 1997, 62, 1586; (h) C. E. Masse, B. S. Knight,
P. Stavropoulos and J. S. Panek, J. Am. Chem. Soc., 1997, 119, 6040;
(i) M. M. Vasbinder and S. J. Miller, J. Org. Chem., 2002, 67, 6240; (j) S.
Oishi, A. Niida, T. Kamano, Y. Odagaki, H. Tamamura, A. Otaka, N.
Hamanaka and N. Fujii, Org. Lett., 2002, 4, 1055; (k) P. Wipf and J. B.
Xiao, Org. Lett., 2005, 7, 103.
Notes and references
1 (a) P. Wipf, T. C. Henninger and S. J. Geib, J. Org. Chem., 1998, 63,
6088; (b) S. Oishi, T. Kamano, A. Niida, Y. Odagaki, N. Hamanaka,
M. Yamamoto, K. Ajito, H. Tamamura, A. Otaka and N. Fujii, J. Org.
Chem., 2002, 67, 6162; (c) C. L. Jenkins, M. M. Vasbinder, S. J. Miller
and R. T. Raines, Org. Lett., 2005, 7, 2619; (d) S. Oishi, K. Miyamoto, A.
Niida, M. Yamamoto, K. Ajito, H. Tamamura, A. Otaka, Y. Kuroda,
A. Asai and N. Fujii, Tetrahedron, 2006, 62, 1416; (e) A. Niida, K.
Tomita, M. Mizumoto, H. Tanigaki, T. Terada, S. Oishi, A. Otaka, K.
Inui and N. Fujii, Org. Lett., 2006, 8, 613.
2 (a) J.-M. Ahn, N. A. Boyle, M. T. MacDonald and K. D. Janda, Mini-
Rev. Med. Chem., 2002, 2, 463; (b) N. Venkatesan and B. H. Kim, Curr.
Med. Chem., 2002, 9, 2243.
3 (a) A. Jenmalm, W. Berts, Y. L. Li, K. Luthman, I. Cso¨regh and U.
Hacksell, J. Org. Chem., 1994, 59, 1139; (b) A. Jenmalm, W. Berts, K.
Luthman, I. Cso¨regh and U. Hacksell, J. Org. Chem., 1995, 60, 1026;
(c) D. Wiktelius, W. Berts, A. Jenmalm Jensen, J. Gullbo, S. Saitton, I.
Cso¨regh and K. Luthman, Tetrahedron, 2006, 62, 3600; (d) D. Wiktelius
and K. Luthman, Abstracts of Papers, 232nd ACS National Meeting,
San Francisco, 2006.
5 For a general review, see: H. J. Mitchell, A. Nelson and S. Warren,
J. Chem. Soc., Perkin Trans. 1, 1999, 1899.
6 (a) T. Itaya and A. Mizutani, Tetrahedron Lett., 1985, 26, 347; (b) R.
Wolin, M. Connolly, A. Afonso, J. A. Hey, H. She, M. A. Rivelli,
S. M. Willams and R. E. West, Bioorg. Med. Chem. Lett., 1998,
8, 2157; (c) M. P. Sibi, D. Rutherford, P. A. Renhowe and B. Q.
Li, J. Am. Chem. Soc., 1999, 121, 7509; (d) F. Meyer, A. Laaziri,
A. M. Papini, J. Uziel and S. Juge´, Tetrahedron: Asymmetry, 2003, 14,
2229.
7 D. Wiktelius, E. K. Larsson and K. Luthman, Tetrahedron: Asymmetry,
2006, 17, 2088.
8 B. E. Maryanoff and A. B. Reitz, Chem. Rev., 1989, 89, 863.
9 E. Vedejs and M. J. Peterson, Top. Stereochem., 1994, 21, 1.
10 B. E. Maryanoff, A. B. Reitz and B. A. Duhl-Emswiler, J. Am. Chem.
Soc., 1985, 107, 217.
11 Similar results have been reported by others: J. S. Clark, P. B. Hodgson,
M. D. Goldsmith and L. J. Street, J. Chem. Soc., Perkin Trans. 1, 2001,
3312.
12 (a) D. L. Boger and J. Y. Hong, J. Am. Chem. Soc., 2001, 123, 8515;
(b) L. A. Paquette, R. Guevel, S. Sakamoto, I. H. Kim and J. Crawford,
J. Org. Chem., 2003, 68, 6096; (c) I. Paterson, M. E. Di Francesco and
T. Ku¨hn, Org. Lett., 2003, 5, 599.
4 (a) M. T. Cox, D. W. Heaton and J. Horbury, J. Chem. Soc., Chem.
Commun., 1980, 799; (b) N. J. Miles, P. G. Sammes, P. D. Kennewell
and R. Westwood, J. Chem. Soc., Perkin Trans. 1, 1985, 2299; (c) A.
Spaltenstein, P. A. Carpino, F. Miyake and P. B. Hopkins, J. Org.
Chem., 1987, 52, 3759; (d) Y. K. Shue, G. M. Carrera, M. D. Tufano and
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 603–605 | 605
©