Stereoselective synthesis of functionalized cyclic amino acid derivatives via a [2,3]-stevens rearrangement and ring-closing metathesis

Article abstract of DOI:10.1021/ol402129h

Unnatural cyclic amino acids are valuable tools in biomedical research and drug discovery. A two-step stereoselective strategy for converting simple glycine-derived aminoesters into unnatural cyclic amino acid derivatives has been developed. The process includes a palladium-catalyzed tandem allylic amination/[2,3]-Stevens rearrangement followed by a ruthenium-catalyzed ring-closing metathesis. The [2,3]-rearrangement proceeds with high diastereoselectivity through an exo transition state. Oppolzer's chiral auxiliary was utilized to access an enantiopure cyclic amino acid by this approach, which will enable future biological applications.

Full text of DOI:10.1021/ol402129h

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Stereoselective Synthesis
of Functionalized Cyclic Amino Acid
Derivatives via a [2,3]-Stevens
Rearrangement and Ring-Closing
Metathesis
Aaron Nash, Arash Soheili, and Uttam K. Tambar*
Department of Biochemistry, The University of Texas Southwestern Medical Center at
Dallas, 5323 Harry Hines Boulevard, Dallas, Texas 75390-9038, United States
Uttam.Tambar@utsouthwestern.edu
Received July 26, 2013
ABSTRACT
Unnatural cyclic amino acids are valuable tools in biomedical research and drug discovery. A two-step stereoselective strategy for converting
simple glycine-derived aminoesters into unnatural cyclic amino acid derivatives has been developed. The process includes a palladium-catalyzed
tandem allylic amination/[2,3]-Stevens rearrangement followed by a ruthenium-catalyzed ring-closing metathesis. The [2,3]-rearrangement
proceeds with high diastereoselectivity through an exo transition state. Oppolzer’s chiral auxiliary was utilized to access an enantiopure cyclic
amino acid by this approach, which will enable future biological applications.
Cyclic amino acids are an important class of compounds
for drug discovery and biomedical research, because they
adoptuniqueamidebondconformationsthatinfluencethe
three-dimensional structure of peptides and proteins.1 For
example, while acyclic amino acids preferentially form
amide bonds in the trans conformation, this preference is
less pronounced for cyclic amino acids such as proline
(1, Figure 1).2 The subtle changes in relative equilibria of cis
and trans amide bonds for acyclic and cyclic amino acids can
have profound effects on the structure and function of
(4) (a) Brandl, C. J.; Deber, C. M. Proc. Natl. Acad. Sci. U.S.A. 1986,
83, 917. (b) An, S. S. A.; Lester, C. C.; Peng, J.-L.; Li, Y.-J.; Rothwarf,
D. M.; Welker, E.; Thannhauser, T. W.; Zhang, L. S.; Tam, J. P.;
Scheraga, H. A. J. Am. Chem. Soc. 1999, 121, 11558. (c) Lummis,
S. C. R.; Beene, D. L.; Lee, L. W.; Lester, H. A.; Broadhurst, R. W.;
(1) (a) Stryer, L., Ed. Biochemistry, 4th Revised Edition; Spektrum:
1996. (b) MacArthur, M. W.; Thornton, J. M. J. Mol. Biol. 1991, 218, 397.
(2) (a) Kern, D.; Schutkowski, M.; Drakenberg, T. J. Am. Chem. Soc.
1997, 119, 8403. (b) Hur, S.; Bruice, T. C. J. Am. Chem. Soc. 2002, 124,
7303. (c) Wedemeyer, W. J.; Welker, E.; Scheraga, H. A. Biochemistry
2002, 41, 14637.
ꢀ
ꢀ
ꢀ
Dougherty, D. A. Nature 2005, 438, 248. (d) Zakova, L.; Kazdova, L.;
ꢀ
ꢀ
ꢀ
Hanclova, I.; Protivınska, E.; Sanda, M.; Budesınski, M.; Jiracek, J. i.
ꢀ
ꢀ
ꢀ
ꢀ
(3) (a) Moretto, V.; Formaggio, F.; Crisma, M.; Bonora, G. M.;
Toniolo, C.; Benedetti, E.; Santini, A.; Saviano, M.; Di Blasio, B.;
Pedone, C. J. Pept. Sci. 1996, 2, 14. (b) Lavrador, K.; Guillerm, D.;
Guillerm, G. Bioorg. Med. Chem. Lett. 1998, 8, 1629. (c) Sisto, A.;
Altamura, M.; Cardinali, F.; D’Andrea, P.; Rossi, C.; Fattori, D.
Bioorg. Med. Chem. Lett. 2007, 17, 4841. (d) Cini, N.; Trabocchi, A.;
Menchi, G.; Bottoncetti, A.; Raspanti, S.; Pupi, A.; Guarna, A. Biorg.
Med. Chem. 2009, 17, 1542. (e) Baeza, J. L.; de la Torre, B. G.; Santiveri,
Biochemistry 2008, 47, 5858. (e) Jiracek, J.; Zakova, L.; Antolıkova, E.;
Watson, C. J.; Turkenburg, J. P.; Dodson, G. G.; Brzozowski, A. M.
Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 1966. (f) Miller, K. H.; Karr,
J. R.; Marqusee, S. J. Mol. Biol. 2010, 400, 567. (g) Spyranti, Z.;
Fragiadaki, M.; Magafa, V.; Borovickova, L.; Spyroulias, G.; Cordopatis,
P.; Slaninova, J. Amino Acids 2010, 39, 539. (h) Kopacz, M. M.; Rovida, S.;
van Duijn, E.; Fraaije, M. W.; Mattevi, A. Biochemistry 2011, 50, 4209. (i)
Das, M.; Basu, G. J. Am. Chem. Soc. 2012, 134, 16536. (j) Song, B.; Bomar,
M. G.; Kibler, P.; Kodukula, K.; Galande, A. K. Org. Lett. 2012, 14, 732.
(k) Pierson, N. A.; Chen, L.; Russell, D. H.; Clemmer, D. E. J. Am. Chem.
Soc. 2013, 135, 3186.
ꢀ
C. M.; Almeida, R. D.; Garcıa-Lopez, M. T.; Gerona-Navarro, G.;
ꢀ
Jaffrey, S. R.; Jimenez, M. A.; Andreu, D.; Gonzalez-Muniz, R.;
ꢀ
ꢀ
~
Martın-Martınez, M. Bioorg. Med. Chem. Lett. 2012, 22, 444.
r
10.1021/ol402129h
XXXX American Chemical Society

peptides and proteins. As a result, polypeptide chains
with unnatural cyclic amino acids have served as leads in
medicinal chemistry programs3 and are useful chemical
tools for elucidating novel biochemical pathways.4 Although
unnatural cyclic amino acids exhibit the potential for
important biological applications, their stereoselective
synthesis with functionally diverse side chains and multiple
stereocenters remains a challenging problem.5
allylic carbonates 4 in this approach, we expected to
directly access polyunsaturated amino acid derivatives 5
through a diastereoselective [2,3]-rearrangement (Scheme 1).9
These rearrangement products could be subjected to a sub-
sequent ring-closing metathesis to obtain novel cyclic amino
acid derivatives 6 with multiple stereocenters.10
Scheme 1. Two-Step Strategy for Synthesizing Cyclic Amino
Acid Derivatives
Figure 1. Natural and unnatural cyclic amino acids.
Our investigation began by an examination of the
reactivity and stereoselectivity of homoallylic glycine deri-
vatives 3a and 3b in the tandem palladium-catalyzed allylic
amination/[2,3]-Stevens rearrangement (Scheme 2). Treat-
ment of aminoesters 3a and 3b with cinnamyl ethyl carbo-
We became interested in developing a stereoselective
strategy to construct functionalized cyclic amino acid
derivatives with medium-sized rings and multiple stereo-
centers (2, Figure 1). We reasoned that amino acid deriva-
tives with seven- and eight-membered rings would provide a
unique opportunity for generating structures with multiple
stereocenters and diverse functional groups. Moreover,
these larger rings would maintain the conformational ben-
efits of proline and other cyclic amino acids.6 Herein, we
describe a two-step diastereoselective protocol for synthe-
sizing cyclic amino acid derivatives with multiple stereo-
centers that meets these criteria. The mild conditions of this
strategy are compatible with a wide range of chemical
functionality and are utilized to generate an enantioenriched
cyclic amino acid.
nate 7, 1 mol % Pd2dba3 CHCl3, 3 mol % P(2-Furyl)3,
3
and Cs2CO3 as an exogenous base resulted in the forma-
tion of rearrangement products 11a and 11b. To our
delight, polyunsaturated aminoesters 11a and 11b were
both generated in acceptable yields. However, the less
sterically encumbered methyl ester derivative was gener-
ated in a 3:1 diastereomeric ratio, while the tert-butyl ester
derivative was isolated as a more desirable 8:1 mixture of
diastereomers. The major diastereomer was presumably
formed through the exo transition state 9 of the [2,3]-
rearrangement, with the large tert-butyl ester and the
phenyl ring oriented in an anti configuration.7,9
Our approach for synthesizing functionalized cyclic
amino acid derivatives was based on a tandem palladium-
catalyzed allylic amination/[2,3]-Stevens rearrangement
previously developed by our group.7 This work repre-
sented the first example of using tertiary amines as inter-
molecular nucleophiles in metal-catalyzed allylic substitu-
tion chemistry.8 By utilizing homoallylic aminoesters 3 and
Scheme 2. Diastereoselective Palladium-Catalyzed Tandem
Allylic Amination/[2,3]-Stevens Rearrangement
(5) (a) Park, K.-H.; Kurth, M. J. Tetrahedron 2002, 58, 8629. (b)
ꢀ~
Cativiela, C.; Ordonez, M. Tetrahedron: Asymmetry 2009, 20, 1. (c) Cini,
E.; Bifulco, G.; Menchi, G.; Rodriquez, M.; Taddei, M. Eur. J. Org.
Chem. 2012, 2012, 2133. (d) Kubyshkin, V. S.; Mykhailiuk, P. K.;
Afonin, S.; Ulrich, A. S.; Komarov, I. V. Org. Lett. 2012, 14, 5254. (e)
ꢀ
Singh, S.; Martinez, C.-M.; Calvet-Vitale, S.; Prasad, A. K.; Prange, T.;
Dalko, P. I.; Dhimane, H. Synlett 2012, 23, 2421.
(6) Wu, W.-J.; Raleigh, D. P. J. Org. Chem. 1998, 63, 6689.
(7) Soheili, A.; Tambar, U. K. J. Am. Chem. Soc. 2011, 133, 12956.
(8) For an example of intramolecular allylic amination with tertiary
amines, see: van der Schaaf, P. A.; Sutter, J.-P.; Grellier, M.; van Mier,
G. P. M.; Spek, A. L.; van Koten, G.; Pfeffer, M. J. Am. Chem. Soc.
1994, 116, 5134.
Once we identified an efficient approach to polyunsatu-
rated amino acid derivative 11b, we optimized the ring-
closing metathesis step of our approach to cyclic amino
ꢀ
(9) (a) Marko, I. E. In Comprehensive Organic Synthesis; Trost, B. M.,
Fleming, I., Eds.; Pergamon Press: New York, 1991; Vol. 3, pp 913À974. (b)
€
Bruckner, R. In Comprehensive Organic Synthesis; Trost, B. M., Fleming,
I., Eds.; Pergamon Press: New York, 1991; Vol. 6, pp 873À908. (c) Sweeney,
J. B. Chem. Soc. Rev. 2009, 38, 1027. (d) Hoffmann, R. W. Angew.
Chem., Int. Ed. Engl. 1979, 18, 563. (e) Blid, J.; Panknin, O.; Somfai, P.
J. Am. Chem. Soc. 2005, 127, 9352. (f) Workman, J. A.; Garrido, N. P.;
Sanc-on, J.; Roberts, E.; Wessel, H. P.; Sweeney, J. B. J. Am. Chem. Soc.
2005, 127, 1066.
(10) (a) Maier, M. E. Angew. Chem., Int. Ed. 2000, 39, 2073. (b)
Monfette, S.; Fogg, D. In Green Metathesis Chemistry; Dragutan, V.,
Demonceau, A., Dragutan, I., Finkelshtein, E., Eds.; Springer: Netherlands,
2010; pp 129À156.
B
Org. Lett., Vol. XX, No. XX, XXXX

Table 1. Optimization of Ring-Closing Metathesis Step
Table 2. Substrate Scope of Allylic Carbonatea
entry
catalyst (mol %)
additive (1.5 equiv)
yield (%)a
1b
2b
3b
4b
5
13a (5)
13b (5)
13a (5)
13b (5)
13a (5)
13b (5)
13b (5)
13b (10)
13b (10)d
À
<5
<5
<5
<5
<5
50
65
58
85
À
Ti(Oi-Pr)4
Ti(Oi-Pr)4
p-TsOH H2O
3
3
3
3
6
p-TsOH H2O
p-TsOH H2Oc
7
8
p-TsOH H2O
9
p-TsOH H2O
3
a Isolated yield. b Reaction also resulted in no product in refluxing
CH2Cl2. c 3 equiv of p-TsOH H2O. d Catalyst was added as two separate
amounts of 5 mol % at the beginning and 6 h into the reaction.
3
acid derivatives with commercially available ruthenium
catalysts (Table 1).11 The Grubbs second generation catalyst
13a and the HoveydaÀGrubbs second generation catalyst
13b were both ineffective in facilitating the desired ring-
closing event (entries 1 and 2), presumably because of catalyst
inhibition by the Lewis basic amine starting material.12
Pretreatment of basic amine substrates with Lewis acid
additives, such as Ti(Oi-Pr)4, is a well-known strategy for
engaging unsaturated amines in ring-closing metatheses.13
Unfortunately, this approach proved unsuccessful for
aminoester 11b (entries 3 and 4).
We also examined the formation of the ammonium salt
of aminoester 11b in the presence of Brønsted acids to
mitigatethe basicity ofthe nitrogen in thesubstrate(entries
5À9).14 While the ring-closing metathesis of aminoester
11b was unsuccessful with 5 mol % Grubbs second gen-
eration catalyst 13a and p-toluenesulfonic acid monohy-
drate (entry 5), we were delighted to obtain a 50% isolated
yield for cyclic aminoester 12 in the presence of 5 mol %
HoveydaÀGrubbs second generation catalyst 13b and
p-toluenesulfonic acid monohydrate (entry 6). We suspected
that the reaction was not proceeding to complete conversion
because of slow inhibition or decomposition of the 5 mol %
loading of catalyst 13b. To alleviate this problem, we
increased the amount of Brønsted acid (entry 7) and catalyst
a Step 1: 1.25 equiv of aminoester 3b, 1 equiv of allylcarbonate 4, 1
mol % Pd2dba3 CHCl3, 4 mol % P(2-Furyl)3, 3 equiv of Cs2CO3, 0.2 M
3
in MeCN. Step 2: 1.5 equiv of p-TsOH H2O, 0.01 M CH2Cl2; 10 mol %
3
catalyst 13b (added in two batches). b Isolated yield for 2 steps. c Dia-
stereomeric ratios as determined by NMR analysis. d Ru catalyst13a was
used.
13b (entry 8), which only increased the yield to 65% and
58%, respectively. Gratifyingly, when we added a second
5 mol % loading of catalyst 13b after 6 h, we obtained
desired product 12 in 85% yield (entry 9).
With optimized conditions in hand for the two-step
conversion of homoallylic glycine derivative 3b to cyclic
aminoesters, we surveyed the scope of functionalized
allylic carbonates 4 that were compatible with this process
(Table 2). To streamline the procedure, we performed both
steps in a telescoped sequence by simply filtering and con-
centrating the reaction mixture after completion of the first
step. We then redissolved the rearrangement product in
CH2Cl2 and subjected it to ring-closing metathesis conditions.
(11) Vougioukalakis, G. C.; Grubbs, R. H. Chem. Rev. 2009, 110,
1746.
(12) Compain, P. Adv. Synth. Catal. 2007, 349, 1829.
(13) (a) Ghosh, A. K.; Cappiello, J.; Shin, D. Tetrahedron Lett. 1998,
39, 4651. (b) Yang, Q.; Xiao, W.-J.; Yu, Z. Org. Lett. 2005, 7, 871.
(14) (a) Fu, G. C.; Nguyen, S. T.; Grubbs, R. H. J. Am. Chem. Soc.
1993, 115, 9856. (b) Wright, D. L.; Schulte, J. P.; Page, M. A. Org. Lett.
2000, 2, 1847.
Org. Lett., Vol. XX, No. XX, XXXX
C

(entry 12), and the azide product can be useful for click
chemistry applications (entry 13).
Scheme 3. Synthesis of Eight-Membered Ring and Enantiopure
Amino Acid
We explored the scope of this method further by examin-
ing aminoester 14 (Scheme 3a). This bis-homoallylic sub-
strate underwent the palladium-catalyzed tandem allylic
amination/[2,3]-Stevens rearrangement followed by a ruthe-
nium-catalyzed ring-closing metathesis at 40 °C to furnish
the eight-membered ring cyclic aminoester 15 in 75% yield
and 4:1 dr. The relative stereochemistry of the major diaste-
reomer was confirmed by X-ray crystallography.
Access to enantiopure unnatural cyclic amino acids is
imperative for the integration of our method into biologi-
cal applications. In this context, we subjected aminoester
16, which was functionalized with Oppolzer’s chiral
auxiliary,15,9f to our two-step protocol (Scheme 3b). Grat-
ifyingly, we isolated cyclic aminoester 17 as a single stereo-
isomer.16 The absolute and relative stereochemistry of this
ring-closing metathesis product was confirmed by X-ray
crystallography. Moreover, aminoester 17 was converted
in two steps to enantiopure cyclic amino acid 18. Access to
this unnatural cyclic amino acid will provide new oppor-
tunities for controlling the conformation of synthetic
peptides.
a ACE-Cl = 2-chloroethyl chloroformate.
The product of the ring-closing metathesis was purified by
flash chromatography to obtain a two-step isolated yield of
cyclic aminoesters.
In conclusion, we have developed a mild and stereoselective
method to access unnatural cyclic amino acid derivatives via a
palladium-catalyzed tandem allylic amination/[2,3]-Stevens
rearrangement followed by a ring-closing metathesis. This
method allows for the incorporation of multiple functiona-
lized side chains, including substituted aromatic rings, nucleo-
philic groups, a benzophenone, and an azide. Utilization of
Oppolzer’s chiral auxiliary provides access to enantioenriched
cyclic amino acids. We are exploring the incorporation of
these cyclic amino acids into peptide chains via automated
peptide synthesis for specific biological applications.
The method was tolerant of both linear and branched
allylic carbonates (entries 1 and 2À13, respectively). Aryl
rings with diverse substitution patterns could be incorpo-
rated, including electron-donating and -withdrawing groups
(entries2À6). Aliphatic allylic carbonates with hydrocarbon
substituents, such as isopropyl, pentyl, and benzyl, furn-
ished the desired cyclic aminoester products as predomi-
nantly single diastereomers (entries 7À9). Cyclic aminoester
products with heteroatomic functional groups were also
generated, including a phthalimide (entry 10) and a thioether
(entry 11). We were interested in these heteroatom-bearing
products because of their potential to be transformed into
unnatural cyclic amino acids with nucleophilic side chains.
Interestingly, ring-closing metathesis with the sulfur-
containing substrate was only successful in the presence of
the Grubbs second generation catalyst 13a. Finally, cyclic
aminoesters with benzophenone and azide substituents were
generated in 63% and 60% yields over two steps, respec-
tively (entries 12À13). We anticipate that the benzophenone
product can be converted into a photo-cross-linking probe
Acknowledgment. Financial support was provided by
the W. W. Caruth, Jr. Endowed Scholarship, the RobertA.
Welch Foundation (Grant I-1748), the National Science
Foundation (NSF CAREER 1150875), and the Sloan
Research Fellowship.
Supporting Information Available. Full experimental
1
procedures, characterization data, and H and 13C NMR
spectra for all synthesized compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
(15) Oppolzer, W. Tetrahedron 1987, 43, 1969.
(16) NMR analysis of the unpurified product from the allylic
amination/[2,3]-rearrangement indicated an 8:2:1:0 dr.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX