A practical synthesis of trans -3-substituted proline derivatives through 1,4-addition

Article abstract of DOI:10.1021/ol102613z

A practical four-step synthesis of 3-alkyl-, vinyl-, and aryl-substituted proline derivatives, which are important building blocks for conformationally restrained peptide analogs, was developed. The method relies on a Cu-catalyzed 1,4-addition of Grignard reagents to N-protected 2,3-dehydroproline esters, efficiently prepared in a new one-pot protocol. The 1,4-addition products are obtained with good trans-selectivity (dr 5:1 to 25:1). A nonracemic sample of N-Cbz-3-vinylproline (74% ee) was obtained using Evans oxazolidinone as a chiral auxiliary.

Full text of DOI:10.1021/ol102613z

ORGANIC
LETTERS
2011
Vol. 13, No. 2
216-219
A Practical Synthesis of
Trans-3-Substituted Proline Derivatives
through 1,4-Addition
Peter Huy, Jo¨rg-Martin Neudo¨rfl, and Hans-Gu¨nther Schmalz*
Department of Chemistry, UniVersity of Cologne, Greinstr. 4, 50939 Ko¨ln, Germany
schmalz@uni-koeln.de
Received October 28, 2010
ABSTRACT
A practical four-step synthesis of 3-alkyl-, vinyl-, and aryl-substituted proline derivatives, which are important building blocks for conformationally
restrained peptide analogs, was developed. The method relies on a Cu-catalyzed 1,4-addition of Grignard reagents to N-protected
2,3-dehydroproline esters, efficiently prepared in a new one-pot protocol. The 1,4-addition products are obtained with good trans-selectivity
(dr 5:1 to 25:1). A nonracemic sample of N-Cbz-3-vinylproline (74% ee) was obtained using Evans oxazolidinone as a chiral auxiliary.
Trans-3-substituted proline derivatives of type 2 (Scheme
1) are widely used in the synthesis of conformationally
constrained peptide mimetics (e.g., of type 1),1,2 as modified
amino acid building blocks in peptide synthesis3 and as
intermediates in the synthesis of alkaloids.4 The pharmaco-
logical relevance of such compounds was demonstrated in
the development of potent hepatitis C virus protease
inhibitors3a and farnesyl transferase inhibitors.5
protected products as racemates (trans/cis ) ca. 2:1) in six
steps in up to 40% overall yield but is not suitable for the
synthesis of 3-vinyl-proline derivatives.6,7 (2) An alternative
method, introduced by Herdeis, utilizes a Cu-mediated 1,4-
addition to R,ꢀ-unsaturated pyroglutamic acid derivatives
followed by deoxygenation.3a,4 This strategy provides the
products in high stereochemical purity; however, it requires
10-11 steps and, accordingly, proceeds only with low overall
yield.8
In the course of our research program aiming at the
development of Pro-Pro dipeptide mimetics locked in a PPII-
helix conformation,1,9 we initially applied a (modified)
Herdeis sequence for the synthesis of trans-3-vinyl-proline
(2, PG ) Boc, R ) vinyl).1 However, the large number of
steps and the low overall yield prompted us to search for a
So far, two approaches have mainly been used for the
synthesis of compounds of type 2: (1) Michael-addition of
an amino-malonate to an R,ꢀ-unsaturated aldehyde with
subsequent reduction and decarboxylation affords the N-
(1) Zaminer, J.; Brockmann, C.; Huy, P.; Opitz, R.; Reuter, C.;
Beyermann, M.; Freund, C.; Mu¨ller, M.; Oschkinat, H.; Ku¨hne, R.; Schmalz,
H.-G. Angew. Chem., Int. Ed. 2010, 49, 7111–7115
.
(2) (a) Tong, Y.; Fobian, Y. M.; Wu, M.; Boyd, N. D.; Moeller, K. D.
Bioorg. Med. Chem. Lett. 1998, 8, 1679–l682. (b) Einsiedel, J.; Lanig, H.;
Waibel, R.; Gmeiner, P. J. Org. Chem. 2007, 72, 9102–9113
.
(3) (a) Perni, R. B.; et al. Bioorg. Med. Chem. Lett. 2004, 14, 1939–
1942. (b) Tong, Y.; Fobian, Y. M.; Wu, M.; Boyd, N. D.; Moeller, K. D.
Bioorg. Med. Chem. Lett. 2008, 18, 1931–1938. (c) Jackson, D. Y.; Quan,
C.; Artis, D. R.; Rawson, T.; Blackburn, B.; Struble, M.; Fitzgerald, G.;
Chan, K.; Mullins, S.; Burnier, J. P.; Fairbrother, W. J.; Clark, K.; Berisini,
M.; Chui, H.; Renz, M.; Jones, S.; Fong, S. J. Med. Chem. 1997, 40, 3359–
Scheme 1. A Strategy for the Synthesis of 3-Substituted Proline
Derivatives (2) through 1,4-Addition
3368
.
(4) Hereides, C.; Hubmann, H. P.; Lotter, H. Tetrahedron: Asymmetry
1994, 5, 351–354.
(5) Latest example: DeSolms, S. J. U.S. Patent 5,872,135, 1999; CAN
130:196955.
10.1021/ol102613z  2011 American Chemical Society
Published on Web 12/15/2010

more practical alternative. Herein, we report an efficient,
flexible, and operationally reliable four-step synthesis of
trans-3-substituted prolines of type 2 following the strategy
shown in Scheme 1.
lines 3a/b in an efficient and reliable fashion on a 50 g scale
(Scheme 2).
Our concept is based on the consideration that the
substituent R could be introduced to a N-protected 2,3-
dehydroproline ester of type 3 by means of a Cu-catalyzed
1,4-addition. The trans-configuration should result from a
diastereoselective protonation of the initially formed eno-
late.10 As a precondition for the overall success, an efficient
access to the required 2,3-dehydroprolines of type 3 from
proline (4) is required. In principle, this method also allows
the preparation of nonracemic compounds by either employ-
ing a chirally modified substrate or by performing the 1,4-
addition in a catalytic enantioselective fashion.11
Scheme 2
.
One-Pot Synthesis of 3a/b (Moc ) Methoxy
Carbonyl)
Initially, we synthesized the dehydroprolines 3a/b
according to the most common literature protocol. Thus,
esterification of 4 (MeOH, SOCl2, 100%)12 and workup
of the crude hydrochloride 5 with NEt3 in ether gave rise
to the methyl ester 6 in 75% yield after distillation.
Oxidation of 6 with tBuOCl13 with concomitant elimina-
tion of HCl according to Ha¨usler14a then afforded the 1,2-
dehydroproline 8 in 81% yield, again after distillation.
Finally, treatment of 8 with Cbz-Cl or Moc-Cl in the
presence of pyridine in CH2Cl2 delivered the 2,3-dehy-
As a major change, we replaced the light-sensitive and
dangerous tBuOCl by inexpensive NCS.15 Due to its lower
oxidation power this reagent allowed us to perform the
reaction in the presence of NEt3, which is a sufficiently strong
base to directly induce HCl elimination (7 to 8). For this
purpose, the crude hydrochloride 5 was dissolved in CH2Cl2
and treated with NEt3 (2 equiv) prior to the portionwise
addition of NCS. The resulting imine 8 was then converted
to the N-protected enamine (3a/3b) simply by addition of
Moc-Cl (or Cbz-Cl, respectively) and pyridine.16 It proved
to be crucial to use a 2-fold excess of both the chloroformiate
(PG-Cl) and the base (pyridine) because the succinimide
(formed from NCS) also consumes 1 equiv of the chloro-
formiate to give 9.17 In comparison to the literature method14
our one-pot protocol (Scheme 2) affords the dehydroprolines
3a/b in greatly improved overall yield from proline (77%
of 3a, 90% of 3b after chromatography) and could be
performed in the air using technical-grade solvents and
reagents.18
droprolines 3a/b in typical yields of 50 to 70%.12b
c
However, despite reasonable high overall yields from
proline (30% for 3a, and 54% for 3b) the method did not
satisfy our demands for practicability.
Fortunately, we succeeded in developing a greatly im-
proved one-pot procedure (see below), which allows direct
conversion of the crude hydrochloride 5, obtained in
quantitative yield from 4 (see above), into the dehydropro-
(6) (a) Chung, J. Y. L.; Wasicak, J. T.; Arnold, W. A.; May, C. S.;
Nadzan, A. M.; Holladay, M. W. J. Org. Chem. 1990, 55, 270-275. Further
examples: ref 3b-c.
(7) Recent variants of this method exploit an organocatalytic key step:
(a) Rios, H. R.; Ibrahem, I.; Vesely, J.; Sunde´n, H.; Co´rdova, A. Tetrahedron
Lett. 2007, 48, 8695–8699. (b) Merck & Co., Emerson, K. M.; Ho, G.-J.
Chem. Abstr. 2000, 134, 86538.
(8) For other syntheses of compounds of type 2, see: (a) Soloshonok,
V. A.; Ueki, H.; Tiwari, R.; Cai, C.; Hruby, V. J. J. Org. Chem. 2004, 69,
4984–4990. (b) Kanemasa, S.; Tatsukawa, A.; Wada, E. J. Org. Chem. 1991,
56, 2875–2874. (c) Damour, D.; Pulicani, J.-P.; Vuilhorgne, M.; Mignani,
S. Synlett 1999, 786–788.
Having developed a comfortable access to the dehydro-
prolines 3a/b, we next investigated the Cu-catalyzed 1,4-
addition of Grignard reagents to these compounds. Following
an established protocol19 we treated 3a with 1.5 equiv of
vinyl-MgBr in the presence of 0.3 equiv of CuBr·SMe2 at
-30 °C in THF. Under these conditions the product rac-
(9) (a) Ball, L.; Ku¨hne, R.; Schneider-Mergner, J.; Oschkinat, H. Angew.
Chem. Int. Ed. 2005, 44, 2852–2869. (b) Freund, C.; Schmalz, H.-G.; Sticht,
J.; Ku¨hne, R. Handb. Exp. Pharmacol. 2008, 186, 407–429.
(10) For noncatalytic 1,4-additions to substituted dehydroprolines, see:
(a) Ezquerra, J.; Escribano, A.; Rubio, A.; Remuififin, M. J.; Vaquero, J. J.
Tetrahedron: Asymmetry 1996, 7, 2613–2626. (b) Toyooka, N.; Okumura,
M.; Himiyama, T.; Nakazawa, A.; Nemoto, H. Synlett 2003, 55–58.
(11) Reviews: (a) Schmalz, H.-G. In ComprehensiVe Organic Synthesis,
Vol. 4; Trost, M., Fleming, I., Eds.; Pergamon: Oxford, 1991; pp 199-236.
(b) Rossiter, B. E.; Swingle, N. M. Chem. ReV. 1992, 92, 711–806. (c)
Alexakis, A.; Ba¨ckvall, J. E.; Krause, N.; Pamies, O.; Die´guez, M. Chem.
ReV. 2008, 108, 2796–2823. (d) Lopez, F.; Minnaard, A. J.; Feringa, B. L.
Acc. Chem. Res. 2007, 40, 179–188. For a recent contribution from this
laboratory, see: (e) Robert, T.; Velder, J.; Schmalz, H.-G. Angew. Chem.,
Int. Ed. 2008, 47, 7718–7721.
(15) For the NCS-oxidation of amines to N-chloroamines, see: (a)
Furneaux, R. H.; Limberg, G.; Tyler, P. C.; Schramm, V. L. Tetrahedron
1997, 53, 2915–2930. (b) Haeberli, A.; Leumann, C. J. Org. Lett. 2001, 3,
489–492.
(16) Neither NEt3 nor pyridine could be replaced by each other. Pyridine
proved not to be basic enough to induce HCl elimination (7 to 8) while we
assume pyridine to act as a catalyst in the acylation of 8.
(17) With 1.5 equiv of Cbz-Cl/pyridine the yield of 3a dropped to
30%.
(12) (a) Webb, R. G.; Haskell, M. W.; Stammer, C. H. J. Org. Chem.
1969, 34, 576–580. (b) Guttmann, S. HelV. Chim. Acta 1961, 85, 721–744.
(13) Peparation: Mintz, M. J.; Walling, C. Org. Synth. 1969, 49, 9–12.
(14) (a) Ha¨usler, J. Liebigs Ann. Chem. 1981, 1073–1088. (b) Shin, C.;
Takahashi, N.; Yonezawa, Y. Chem. Pharm. Bull. 1990, 38, 2020–2023.
(c) Purvis, M. B.; LeFevre, J. W.; Jones, V. L.; Kingston, D. G. I.; Biot,
A. M.; Gossole´, F. J. Am. Chem. Soc. 1989, 111, 5931–5935.
(18) For other (operationally less attractive) entries to dehydroproline
analogs of type 3 involving the oxidation of N-protected proline esters, see
ref 10a (LiHMDS, PhSeCl) or Kublitskii, V. S.; Stepanov, A. E.; Trukhan,
V. M. Russian J. Org. Chem. 2008, 44, 933–934 (LiHMDS, Br2).
(19) Quinkert, G.; Schmalz, H.-G.; Walzer, E.; Gross, S.; Kowa1czyk-
Przewloka, T.; Schierloh, C.; Du¨rner, G.; Bats, J. W.; Kessler, H. Liebigs.
Ann. Chem. 1988, 283–315.
Org. Lett., Vol. 13, No. 2, 2011
217

10a was formed in 66% isolated yield as a (inseparable) 10:1
mixture of the trans- and cis-diastereomers (Scheme 3).20
To further convert the 1,4-addition products (of type 10)
into the deprotected target compounds of type 2 (Scheme 1)
the methylester function had to be saponified. Treatment of
rac-10a with an excess of LiOH in H2O/MeOH/THF21
afforded rac-2a as a mixture of diastereomers in 87% yield.
While initial attempts to achieve a selective hydrolysis of
the sterically less hindered trans-diastereomer using NaOH
in H2O/MeOH22 were not successful, we succeeded in
achieving an efficient kinetic discrimination of the diaster-
eomers by using less than 1 equiv of LiOH (0.8 to 0.95 equiv
in H2O/MeOH) and running the reactions at low temperature
(slowly warming from -30 to 20 °C). In all four cases
investigated (Table 2), the acids rac-2 were isolated in good
Scheme 3. Cu-Catalyzed 1,4-Addition of Vinyl-MgBr to 3a/b
In a similar experiment, the Moc-protected substrate 3b
afforded rac-10b in a slightly lower yield (52%; trans/cis
) 6:1).
Table 2. Trans-Selective Saponification of 10
In the course of extensive experimentation it became
obvious that a slow addition (by means of a syringe pump)
of a solution of the substrate (3) to the Grignard/Cu(I)
mixture is essential to ensure reproducible yields. A further
improvement was achieved by changing the temperature of
the initial cuprate formation from -10 to -35 °C and by
lowering the amount of the Cu(I) catalyst to 15-20 mol %.
Under the optimized conditions the product rac-10a was
obtained in 80% yield, even on a 20 g (75 mmol) scale (Table
1). The generality of the protocol was demonstrated by also
entry
R
substrate(dr)
product(yield)
dr g
1
2
3
4
vinyl
Me
iPr
10a (11:1)
10c (7:1)
10d (25:1)
10e (5:1)
2a (74%)
2c (71%)
2d (69%)
2e (69%)
99:1
99:1
99:1
99:1
Ph
Table 1. 1,4-Addition of Different Grignard Reagents to 3a
yield as analytically pure trans-diastereomers (dr g 99:1)
after extractive workup and without the need for additional
purification.23 The synthesis of the (yet racemic) compounds
of type 2 was thus achieved in only four steps (29-46%
overall yield) from 4.
The trans-configuration of the acids of type rac-2 was
assigned by NOE-NMR spectroscopy (see Supporting In-
formation) and additionally secured by an X-ray crystal
structure analysis of rac-2d and 2f (Figure 1).
product
(isol. yield)
dr20
(trans/cis)
byproduct
(isol. yield)
entry
R-MgX
1
2
3
4
Vinyl-MgCl
Me-MgBr
iPr-MgCl
Ph-MgBr
10a (80%)
10c (53%)
10d (55%)
10e (59%)
11:1
7:1
25:1
5:1
-
11c (13%)
11d (11%)
11e (8%)
employing other Grignard reagents (Me, iPr, and Ph) to give
the products of type rac-10 in 53-59% yield. Not unex-
pected, the product (rac-10d) with the most bulky substituent
R was formed with the highest diastereoselectivity (trans/
cis ) 25:1).
As byproduct of the 1,4-addition, ketones of type rac-11
were formed (notably as virtually pure diastereomers).
Interestingly, such a byproduct was not observed in the
synthesis of the vinyl derivative rac-10a (and rac-10b). We
suppose that the ketones of type 11 arise from an initial attack
of the Grignard reagent at the ester function of 3a and Cu-
catalyzed 1,4-addition to the resulting ketone (as the more
reactive Michael acceptor).
Figure 1. Structures of 2d and 2f in the crystalline state.
To exploit the developed method also for the synthesis of
nonracemic compounds, we decided to probe the 1,4-addition
(21) Boger, D. L.; Yohannes, D.; Zhou, J.; Patane, M. A. J. Am. Chem.
Soc. 1993, 115, 3420–3430.
(22) Not unexpected, the diastereomers could not be separated by column
chromatography; see ref 6 and Sarges, R.; Tretter, J. R. J. Org. Chem. 1974,
39, 1710–1716.
(20) Due to signal overlap in the crude product, the trans/cis ratio was
1
determined by means of H NMR of the purified sample.
218
Org. Lett., Vol. 13, No. 2, 2011

using a chirally modified dehydroproline derivative as a
substrate. For this purpose compound 14 was prepared from
3a by ester hydrolysis and subsequent coupling24 of the acid
12 with Evans oxazolidinone 1325 (Scheme 4). Under the
An interesting observation was made when we tried to
optimize the access to the dehydroproline 14. In an attempt
to deprotect the chiral proline derivative 16 by treatment with
2 N HCl in dioxane31 (in order to prepare for its oxidation
to 14) we isolated the novel hydantoine 18 in 75% yield
(Scheme 5).
Scheme 4. Synthesis of Nonracemic 3-Vinyl-proline Derivatives
through Diastereoselective 1,4-Addition (Piv ) Pivaloyl)
Scheme 5. Unexpected Formation of the Hydantoine 18
The structure of 18 was proven through X-ray crystal
structure analysis (Figure 2). We assume that after Boc-
proven conditions, the key 1,4-addition then proceeded
smoothly to give 15 with a useful degree of diastereoselectity
(dr ) 7:1).26,27
After cleaving off the chiral auxiliary (LiOH/H2O2)28 the
trans-vinyl-proline derivative 2a was isolated in 54% yield,
supposably as a 7:1 mixture of enantiomers.29 To determine
the absolute configuration of the major enantiomer, the Cbz-
group was replaced by Boc through treating 2a with TMSI30
and subsequently with Boc2O (Scheme 4). The comparison
of the optical rotation of the resulting compound 2f ([R]589
) +28.3; [R]546 ) +33.9; c ) 1.125 in CHCl3) with
literature values1 clearly indicated that the (2R,3S)-enanti-
omer of 2f was formed with an enantiomeric excess of 74%
ee. Within the experimental limits, this perfectly matched
the value predicted from the dr (7:1) determined by 1H NMR
at the stage of 15.
Figure 2. Structure of hydantoine 18 in the crystalline state.
cleavage the pyrrolidine nitrogen attacks the carbamate
carbon atom to give intermediate 17, from which 18 is readily
generated as the more stable isomer. The formation of the
hydantoine 18 was also observed even under neutral condi-
tions (65% yield), i.e. on hydrogenolytic deprotection of the
Cbz-protected analog of 16 (H2, Pd/C, MeOH or EtOAc).
In conclusion we have developed a short and practical
synthesis of trans-3-substituted proline derivatives (rac-2)
allowing a late and flexible introduction of the substituent
R. The dehydroproline derivatives (3) used as substrates for
the Cu-catalyzed key step were efficiently prepared in a new
one-pot protocol. While the nonracemic vinyl-proline 2a was
obtained with 74% ee through a chiral auxiliary approach,
future research in this laboratory will focus on the developent
of a catalytic-enantioselective variant of the method devel-
oped.
(23) In all cases, the diastereomeric purity of the products (trans-rac-
1
2) was determined by means of H NMR (400 MHz). The nonconverted
starting material (rac-10) was re-isolated as a mixture of trans and cis
diastereomers.
(24) Dobarro, A.; Velasco, D. Tetrahedron 1996, 52, 13733–13738.
(25) For reviews, see: (a) Evans, D. A. Aldrichimica Acta 1982, 15,
20–43. (b) Ager, D. J.; Prakash, I.; Schaad, D. R. Chem. ReV. 1996, 96,
835–875. (c) Ager, D. J.; Prakash, I.; Schaad, D. R. Aldrichimica Acta 1997,
30, 1–24. (d) Gacs-Baitz, E.; Delle Monache, G.; Misiti, D.; Nevolam, L.;
Botta, B.; Zappia, G.; Cancelliere, G. Curr. Org. Synth. 2007, 4, 238–307.
1
(26) The ratio of the inseparable diastereomers was determined by H
NMR. The assumed trans-configuration of both major isomers was
confirmed after hydrolysis to 2a. The formation of minor amounts of cis-
isomers cannot be excluded because of the complexity of the NMR spectra
on the stage of 15.
Acknowledgment. For financial support we thank the
DFG (FOR 806) and the Fonds der chemischen Industrie
(doctoral fellowship to P.H.).
(27) The BF3-mediated 1,4-addition of a vinyl-Cu reagent to 14
(according to Yamamoto, Y. Angew. Chem., Int. Ed. Engl. 1986, 25, 947–
959) proceeded in 82% yield, however, with lower selectivity (dr ) 3.2:
1).
Supporting Information Available: Experimental pro-
cedures and analytical data. This material is available free
of charge via the Internet at http://pubs.acs.org.
(28) Evans, D. A.; Britton, T. C.; Ellman, J. A. Tetrahedron Lett. 1987,
28, 6141–6144.
(29) The auxiliary 13 was reisolated in 57% yield. The dr of 15 was
1
OL102613Z
determined by H NMR (integration of the H-2 signals at ca. 5.5 ppm) .
(30) (a) Jung, M. E.; Lyster, M. A. J. Chem. Soc., Chem. Commun.
1978, 315–316. (b) Lott, R. S.; Chauhan, V. S.; Stammer, C. H. J. Chem.
Soc., Chem. Commun. 1979, 495–496.
(31) Dragovich, P. S.; Webber, S. E.; Babine, R. E.; Fuhrman, S. A.;
Patick, A. K. J. Med. Chem. 1998, 41, 2813–2818.
Org. Lett., Vol. 13, No. 2, 2011
219