Synthesis of a Gln-Phe hydroxy-ethylene dipeptide isostere

Article abstract of DOI:10.1021/ol047879y

(Chemical Equation Presented) The protected Gln-Phe hydroxyethylene dipeptide isostere 1 was synthesized as a precursor for preparation of potential inhibitors of Botulinum neurotoxin B metalloprotease. The method allows for the synthesis of additional hydroxyethylene dipeptide isosteres such as 2 with functionalized P₁ side chains. The isosteres prepared were coupled with a dipeptide to produce protected pseudotetrapeptide derivatives.

Full text of DOI:10.1021/ol047879y

ORGANIC
LETTERS
2004
Vol. 6, No. 25
4783-4786
Synthesis of a Gln-Phe
Hydroxy-ethylene Dipeptide Isostere
Bengt Erik Haug†,‡ and Daniel H. Rich*,†,§
Department of Chemistry and School of Pharmacy, UniVersity of Wisconsin-Madison,
Madison, Wisconsin 53706
dhrich@wisc.edu
Received October 13, 2004
ABSTRACT
The protected Gln-Phe hydroxyethylene dipeptide isostere 1 was synthesized as a precursor for preparation of potential inhibitors of Botulinum
neurotoxin B metalloprotease. The method allows for the synthesis of additional hydroxyethylene dipeptide isosteres such as 2 with functionalized
P1 side chains. The isosteres prepared were coupled with a dipeptide to produce protected pseudotetrapeptide derivatives.
Botulinum neurotoxins (BoNTs), produced by the anaerobic
bacteria Clostridium botulinum are among the most potent
toxins known. The seven BoNT proteases (serotypes A-G)
are produced as 150 kDa inactive polypeptide chains, which
are further converted to 50 kDa zinc metalloproteases inside
cells. The BoNT metalloproteases act by cleaving peptides
in the neuroexocytosis apparatus, causing muscle paralysis
and the symptoms of botulism.1 Our target toxin, Botulinum
neurotoxin B (BoNT/B) endoprotease, cleaves the human
vesicle-associated membrane protein (VAMP-2) between
residues Gln76 and Phe77 and has an extraordinary substrate
specificity, with a 35-mer peptide as its minimal substrate.2
Only a few X-ray structures of the 150 kDa holotoxins have
been published,3 and although a number of inhibitors have
been reported4 no structure of the active 50 kDa proteases
with an inhibitor bound has been reported to this date. Our
goal was to develop a synthesis of hydroxyethylene (HE)
dipeptide isostere 1 for the scissile Gln76-Phe77 bond in order
to prepare transition state analogue pseudopeptide inhibitors
of BoNT/B.
To date, most HE isosteres reported have unfunctionalized
side chains, and most synthetic procedures for HE isosteres
are focused on the assembly of the isostere backbone and
chiral centers. Herein we report the synthesis of the func-
tionalized Gln-Phe HE dipeptide isostere. In initial attempts
on synthesizing HE 1, the Horner-Wadsworth-Emmons
reaction between a Gln-derived keto phosphonate5 and 2-oxo-
3-phenyl propionic acid methyl ester proved to be low
yielding and difficult to reproduce. To overcome these
difficulties, and to allow for a diversification of the Gln P1
(4) (a) Schmidt, J. J.; Stafford, R. G.; Bostian, K. A. FEBS Lett. 1998,
435, 61-64. (b) Schmidt, J. J.; Stafford, R. G. FEBS Lett. 2002, 532, 423-
426. (c) Schmidt, J. J.; Stafford, R. G. Appl. EnViron. Microbiol. 2003, 69,
297-303. (d) Burnett, J. C.; Schmidt, J. J.; Stafford, R. G.; Panchal, R. G.;
Nguyen, T. L.; Hermone, A. R.; Vennerstrom, J. L.; McGrath, C. F.; Lane,
D. J.; Sausville, E. A.; Zaharevitz, D. W.; Gussio, R.; Bavari, S. Biochem.
Biophys. Res. Commun. 2003, 310, 84-93. (e) Hayden, J.; Pires, J.; Roy,
S.; Hamilton, M.; Moore, G. J. J. Appl. Toxicol. 2003, 23, 1-7. (f) Anne,
C.; Turcaud, S.; Quancard, J.; Teffo, F.; Meudal, H.; Fournie-Zaluski, M.-
C.; Roques, B. P. J. Med. Chem. 2003, 46, 4648-4656. (g) Anne, C.;
Blommaert, A.; Turcaud, S.; Martin, A.-S.; Meudal, H.; Roques, B. P.
Bioorg. Med. Chem. 2003, 11, 4655-4660. (h) Sukonpan, C.; Oost, T.;
Goodnough, M.; Tepp, W.; Johnson, E. A.; Rich, D. H. J. Peptide Res.
2004, 63, 181-193. (i) Oost, T.; Sukonpan, C.; Brewer, M.; Goodnough,
M.; Tepp, W.; Johnson, E. A.; Rich, D. H. Biopolymers (Peptide Sci.) 2003,
71, 602-619.
* To whom correspondence should be addressed.
† School of Pharmacy, University of Wisconsin-Madison.
‡ Present address: Department of Chemisty, University of Tromsø,
Norway.
§ Department of Chemistry, University of Wisconsin-Madison.
(1) Montecucco, C.; Schiavo, G. Q. ReV. Biophys. 1995, 28, 423-472.
(2) (a) Shone, C. C.; Quinn, C. P.; Wait, R.; Hallis, B.; Fooks, S. G.;
Hambleton, P. Eur. J. Biochem. 1993, 217, 965-971. (b) Shone, C. C.;
Roberts, A. K. Eur. J. Biochem. 1994, 225, 263-270.
(3) (a) Lacy, D. B.; Tepp, W.; Cohen, A. C.; DasGupta, B. R.; Stevens,
R. C. Nat. Struct. Biol. 1998, 5, 898-902. (b) Swaminathan, S.; Eswar-
amoorthy, S. Nat. Struct. Biol. 2000, 7, 693-699.
(5) Compound 4 in the preceding manuscript of this journal.
10.1021/ol047879y CCC: $27.50
© 2004 American Chemical Society
Published on Web 11/12/2004

side chain of HE 1, we envisioned introduction of the amide
functionality of the target Gln-Phe isostere at a late stage in
the synthesis. We chose to mask the Gln side chain as a
TBS silyl ether, which could be oxidized to the carboxylic
acid and functionalized to an amide at an appropriate point
in the synthesis. This gave lactone 3 as a key intermediate
(Scheme 1).
Scheme 2
Scheme 1
column chromatography. In recent years, several methods
for the stereoselective reduction of R-amino γ-keto esters
have been reported.10 In this case, using LiAlH(O-t-Bu)3 as
the reducing agent (Scheme 3) under chelation control
Our synthetic approach was designed to accommodate the
synthesis of both the (2R,4R,5S)-HE 1 and the diastereo-
isomer with opposite chirality at the C-4. Both diastereo-
isomers would be valuable synthetic targets, as it is difficult
to predict which isomer would bind more strongly to the
BoNT/B metalloprotease. Stereoselective benzylation of
either (4R,5S)-lactone 4 or the (4S,5S)-isomer would provide
the backbone of the HE isosteres. The two isomers of lactone
4 could both be obtained from keto ester 5 after stereo-
selective reduction followed by cyclization. Herein we report
the synthesis of the (2R,4R,5S)-HE 1.
Scheme 3
In the forward sense, our synthetic route started with the
conversion of Boc-Glu-OBn to keto phosphonate 9 (Scheme
2). Reduction of succinimide ester6 6 followed by TBS
protection gave TBS silyl ether 8, which was allowed to react
with lithio dimethyl methylphosphonate.7 Keto phosphonate
9 was utilized in a HWE reaction with freshly prepared
methyl glyoxolate8,9 to give a mixture of cis- and trans-
alkenes in a 2:3 ratio (Scheme 3). Subsequent hydrogenation
of the crude product mixture gave the saturated keto ester 5
in 81% yield over the two steps (Scheme 3).
Reduction of keto ester 5 using NaBH4 in methanol at
-30 °C gave an inseparable mixture of the syn- and anti-
amino alcohols in a 2:3 ratio, which after conversion to the
corresponding diastereomeric lactones could be separated by
(6) (a) Nikawa, J.-i.; Shiba, T. Chem. Lett. 1979, 981-982. (b) Jackson,
R. F. W.; Moore, R. J.; Dexter, C. S. J. Org. Chem. 1998, 63, 7875-7884.
(c) Co´rdova, A.; Reed, N. N.; Ashley, J. A.; Janda, K. D. Bioorg. Med.
Chem. Lett. 1999, 9, 3119-3122.
conditions10a gave almost exclusively the anti-amino alcohol
10. Subsequent lactonization by treatment with acetic acid
in refluxing toluene11 gave lactone 4 contaminated with only
a trace of its diastereoisomer. Separation by flash chroma-
(7) Shikata, K.; Azuma, H.; Tachibana, T.; Ogino, K. Tetrahedron 2002,
58, 5803-5809.
(8) (a) Kelly, T. R.; Schmidt, T. E.; Haggerty, J. G. Synthesis 1972, 544-
545. (b) Schuda, P. F.; Ebner, C. B.; Potlock, S. J. Synthesis 1987, 1055-
1057.
(9) We found that to ensure full consumption of the keto phosphonate
and reproducibility of the reaction, extensive drying of the crude aldehyde
with anhydrous magnesium sulfate was required. See experimental part.
(10) (a) Hoffman, R. V.; Maslouh, N.; Cervantes-Lee, F. J. Org. Chem.
2002, 67, 1045-1056. (b) Va˚benø, J.; Brisander, M.; Lejon, T.; Luthman,
K. J. Org. Chem. 2002, 67, 9186-9191.
(11) Litera, J.; Budesinsky, M.; Urban, J.; Soucek, M. Collect. Czech.
Chem. Commun 1998, 63, 231-244.
4784
Org. Lett., Vol. 6, No. 25, 2004

tography gave 4 as a single diastereoisomer in 87% yield
over the two steps (Scheme 3). Alkylation of lactones with
(4R,5S) configurations such as 4 using, e.g., LDA and alkyl
halides, results in formation of the undesired (2S)-diastereo-
isomer,12 which has the opposite chirality compared to the
side chain of naturally occurring phenylalanine. However,
Nadin and co-workers13 have shown that the desired (2R)-
diastereomer can be obtained through an Aldol-elimination-
hydrogenation sequence. The hydrogenation occurs from the
side opposite of the large substituent at the C-4 carbon atom
of the unsaturated lactone (Scheme 3)14 and gives rise to
the desired (2R) stereochemistry.
Scheme 5
Applying this methodology (Scheme 3) provided (2R)-
lactone 3, in 59% yield over the three steps. It should be
noted that at this point all the stereocenters of target molecule
HE 1 have been set. Lactone 3 showed an NOE between
the H-2 and H-4 (see Scheme 3), which indicates that the
desired (2R,4R) stereochemistry was obtained.
The relative stereochemistry of N-Boc-protected amino
alcohol 10 was confirmed by conversion of lactone 4 to
oxazolidinone 12 (Scheme 4) following the procedure of
butyl ester analogue of primary alcohol 7 with CrO3-pyridine
in an effort to prepare the corresponding aldehyde.19 It was
hypothesized that once the aldehyde was formed, it rapidly
underwent cyclization to the aminal, which was further
oxidized to yield the pyroglutamate.19
Pyroglutamate formation could be avoided by introduction
of a second N-Boc protecting group on the amine (Scheme
6) before submission to the Jones oxidation conditions. In
Scheme 4
Scheme 6
Litera et al.11 Oxazolidinone 12 showed a coupling constant
of 7.7 Hz between H-4 and H-5, which is typical for cis-
oxazolidinones reported in the literature.10a,11,15
Attempted removal of the TBS group from lactone 3 using
TBAF resulted in epimerization of the C-2 position to give
an inseparable mixture of diastereomeric alcohols 13 in a
2:1 ratio in 85% yield. Alternatively, the TBS group could
be removed with acetic acid in THF-water16 to give alcohol
13 (Scheme 5) as a single diastereoisomer in 76% yield.
Oxidation of primary alcohol 13 to the carboxylic acid
proved to be difficult, and treatment of 13 with RuCl3/NaIO4
or TEMPO/PhI(OAc)2 resulted in formation of pyroglutamate
14 (Scheme 5) with only trace amounts of the desired
carboxylic acid. The cyclized side product was also obtained
after direct oxidation of the TBS silyl ether17 3 using the
Jones reagent.18 Pyroglutamate formation had previously
been observed by Olsen et al.19 upon oxidation of a tert-
this case, oxidation of alcohol 15 proceeded uneventfully to
provide acid 16 in good yield. The amide functionality of
the Gln P1 side chain was introduced through coupling with
2,4,6-trimethoxy benzylamine20 to give 17, which has the
(12) Henning, R. In Organic Synthesis Highlights II; Waldmann, H., Ed.;
VCH Verlagsgesellchaft mbH: Weinheim, 1995; pp 251-259.
(13) Nadin, A.; Lope´z, J. M. S.; Neduvelil, J. G.; Thomas, S. R.
Tetrahedron 2001, 57, 1861-1864.
(14) Moritani, Y.; Fukushima, C.; Miyagishima, T.; Ohmizu, H.; Iwasaki,
T. Bull. Chem. Soc. Jpn. 1996, 69, 2281-2286.
(18) Bowden, K.; Heilbron, I. M.; Jones, E. R. H.; Weedon, B. C. L. J.
Chem. Soc. 1946, 39-45.
(15) Yoo, D.; Oh, J. S.; Kim, Y. G. Org. Lett. 2002, 4, 1213-1215.
(16) Kawai, A.; Hara, O.; Hamada, Y.; Shioiri, T. Tetrahedron Lett. 1988,
29, 6331-6334.
(19) Olsen, R. K.; Ramasamy, K.; Emery, T. J. Org. Chem. 1984, 49,
3527-3534.
(17) Evans, P. A.; Roseman, J. D.; Garber, L. T. Synth. Commun. 1996,
26, 4685-4692.
(20) Weygand, F.; Steglich, W.; Bjarnason, J. Chem. Ber. 1968, 101,
3642-3648.
Org. Lett., Vol. 6, No. 25, 2004
4785

desired protection of the amide side chain. Further elabora-
tion required the selective removal of one N-Boc protecting
group, as attempted lactone hydrolysis of 17 with lithium
hydroxide gave a complex mixture in which some Boc
cleavage was noted as well as possible epimerization at the
R-carbon atom.21
Scheme 7
Hernandez et al.22 recently discovered that one N-Boc
group can be selectively cleaved from N,N-di-Boc-protected
amino compounds by treatment with LiBr. Stirring lactone
17 with an excess of LiBr in MeCN at 65 °C overnight
provided mono-N-Boc-protected lactone 18 in near quantita-
tive yield.
Both HE derivatives 1 and 2 were coupled with H2N-Glu-
(OtBu)-Thr(tBu)-NHBn to give the fully protected pseudo-
tetrapeptides 19 and 20 (Figure 1) as short protected
Upon hydrolysis of the lactone to give the hydroxy acid,
we found that relactonization occurs very readily when an
acidic aqueous workup procedure was followed, much in
agreement with what was initially discovered in an earlier
synthesis of a Gln-Arg HE isostere.23 We were able to obtain
protected Gln-Phe HE isostere 1 without any trace of
relactonization (Scheme 6) by using the same optimized cold
workup under basic conditions developed for the Gln-Arg
system.23 The secondary alcohol of the free hydroxy acid
obtained after treatment of N-Boc-protected lactone 18 with
lithium hydroxide was protected as the TBS silyl ether using
TBSCl and imidazole after extensive drying to give 1 in 82%
yield over the two steps.
Figure 1. Protected pseudotetrapeptides.
Most reported HE syntheses contain aliphatic or aromatic
side chains corresponding to the P1 and P1′ positions of the
native dipeptide, whereas our route allows for incorporation
of functionalized side chains in the P1 position. The present
synthesis is particularly nice as it provides late-stage synthetic
intermediates (in which all of the chiral centers have been
set), with functionality that can be easily manipulated to
provide an array of HE analogues differing at the P1 residue.
This is highlighted by the late-stage conversion of the side-
chain alcohol in lactone 15 into the Gln-amide side chain.
analogues of the BoNT/B VAMP [60-94] substrate in 75
and 72% yields, respectively. The biological activity of
pseudopeptides containing the HE isosteres will be reported
elsewhere.
Acknowledgment. This work was supported by The
Norwegian Research Council (Grant 154105/V30 to B.E.H.).
This work was sponsored by the NIH/NIAID Regional
Center of Excellence for Biodefense and Emerging Infectious
Diseases Research (RCE) Program. D.H.R. wishes to ac-
knowledge membership within and support from the Region
V “Great Lakes” RCE (NIH Award 1-U54-AI-057153). The
authors thank Dr. Martha Vestling and Dr. Gary Girdaukas
for obtaining mass spectra and Dr. Thomas Stringfellow for
his assistance in obtaining NMR spectra.
One derivative was obtained by direct hydrolysis of lactone
3 (Scheme 7) followed by TBS protection of the secondary
alcohol to give protected HE isostere 2 in 84% yield over
the two steps. The cold workup under basic conditions23 also
was required in order to prevent relactonization of this
isostere.
Supporting Information Available: Full experimental
details and characterization of compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
(21) Kokotos, G.; Padro´n, J. M.; Mart´ın, T.; Gibbons, W. A.; Mart´ın,
V. S. J. Org. Chem. 1998, 63, 3741-3744.
(22) Herna´ndez, J. N.; Ram´ırez, M. A.; Mart´ın, V. S. J. Org. Chem.
2003, 68, 743-746.
(23) Brewer, M.; Rich, D. H. Org. Lett. 2004, 6, 4779-4782.
OL047879Y
4786
Org. Lett., Vol. 6, No. 25, 2004