Stereocontrolled synthesis of highly functionalized quaternary carbon centers: A route to α-substituted serines

Full text of DOI:10.1002/anie.200900609

Communications
DOI: 10.1002/anie.200900609
Asymmetric Synthesis
Stereocontrolled Synthesis of Highly Functionalized Quaternary
Carbon Centers: A Route to a-Substituted Serines**
Xavier Ariza,* Josep Cornellꢀ, Mercꢁ Font-Bardia, Jordi Garcia,* Jordi Ortiz, Carolina Sꢂnchez,
and Xavier Solans†
In memory of Xavier Solans
The development of efficient stereoselective methods for the
be obtained by the stereoselective addition of a substituted
allylic organoborane II to an aldehyde (Scheme 1). Despite
the extensive use of various terpene- and tartrate-based chiral
À
formation of C C bonds involving the construction of a
quaternary carbon centre has attracted much attention.[1]
These processes become more useful and challenging if
additional neighboring stereogenic centers and polar func-
tionalities are also formed stereoselectively. In this context,
our research group is interested in developing methodologies
for the construction of substructures bearing a densely
functionalized quaternary center; like those found in a-
substituted threonines 1 or serines 2,[2] as well as in more
complex natural products of biological relevance such as the
proteasome inhibitor lactacystin (3)[3] or the immunosuppres-
sant myriocin (4).[4]
Scheme 1. Retrosynthetic analysis of I.
allylborane as well as crotylborane reagents,[5] the required
stereoselective addition of g,g-disubstituted allylboranes to
aldehydes has been much less explored.[6] Examples of the
stereoselective creation of quaternary carbon centers by
addition of g-heteroatom allylboranes such as II are still
scarce.[7] The limited use of such g,g-disubstituted allylboranes
in organic synthesis is probably due to the difficulty involved
in their stereoselective preparation. To overcome this draw-
back, we anticipated that II could be prepared in a straight-
forward manner by hydroboration of allene III.
Our initial proposal for III was allene 5, since it can be
easily obtained in two steps from but-2-yn-1,4-diol
(Scheme 2).[8] We reasoned that its hydroboration at the less
hindered face of the terminal double bond would generate an
unsymmetrical Z allylborane, which could be stereoselec-
tively added to an aldehyde to generate the amino diols 6 that
are protected at the quaternary center. To our delight, we
found that the treatment of 5 with dicyclohexylborane in
We envisaged that quaternary amino acids such as 1 or 4
could arise from a homoallylic alcohol I which, in turn, could
[*] Dr. X. Ariza, J. Cornellꢀ, Dr. J. Garcia, Dr. J. Ortiz, C. Sꢁnchez
Departamento de Quꢂmica Orgꢀnica
Fac. de Quꢂmica, Institut de Biomedicina de la UB (IBUB)
Universitat de Barcelona
C/Martꢂ i Franquꢃs 1–11, 08028-Barcelona (Spain)
Fax: (+34)93-339-7878
E-mail: xariza@ub.edu
jordigarciagomez@ub.edu
Dr. M. Font-Bardia, Prof. X. Solans
Departamento de Cristalografꢂa
Mineralogꢂa y Depꢄsitos Minerales
Fac. de Geologia, Universitat de Barcelona
C/Martꢂ i Franquꢃs, s/n, 08028-Barcelona (Spain)
[†] Deceased.
[**] This work has been supported by the Ministerio de Educaciꢄn y
Ciencia (BQU2003-01269 and CTQ2006-13249). We thank the
Generalitat de Catalunya for a doctorate studentship to C.S. and J.O.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200900609.
Scheme 2. Synthesis of 6a by addition of 5 to isobutyraldehyde.
Cy =cyclohexyl, Ts=4-toluenesulfonyl.
4202
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 4202 –4205

Angewandte
Chemie
Table 1: Tandem hydroboration of 5 and addition of chiral aldehyde to
give the enantioenriched, protected amino diols 6i–l.
CH2Cl2 at room temperature, and subsequent addition of a
slight excess of isobutyraldehyde at À788C afforded 6a as a
single stereoisomer in 74% yield.[9,10]
Interestingly, in the presence of a catalytic amount of
DBU in CH2Cl2 at room temperature, 6a exhibited an
equilibrium largely shifted towards its regioisomer 7a,
which was readily isolated in 85% yield after column
chromatography.[11] One-dimensional nuclear Overhauser
enhancement experiments on 7a enabled us to determine its
relative configuration, which is shown in Scheme 3.
Entry
1[a]
Aldehyde
Adduct
Yield [%][d]
95
2[b]
90
3[b]
80
74
Scheme 3. Isomerization of 6a to 7a. DBU=1,8-diazabicyclo-
[5.4.0]undec-7-ene.
4[c]
These remarkable results prompted us to extend this
reaction to a series of representative aldehydes. As shown in
Figure 1, good to excellent yields of 6b–h were obtained with
[a] 1 equivalent of aldehyde was used. [b] 2 equivalents of aldehyde were
used. [c] 1.4 equivalents of aldehyde were used. [d] Yield of isolated
product. TBDPS=tert-butyldiphenylsilyl.
The relative configuration of 6k was determined by
1H NMR experiments of derivatives 7k and 12 (Scheme 4).
Thus, 6k was easily transformed into the 1,3-diol acetonide
12. The syn relationship of the new stereocenter at C3,
relative to the R configuration of C4 in the starting aldehyde,
was ascertained by analyzing the vicinal proton coupling
constants of the dioxolane ring.[14] Meanwhile, NOESY
experiments on compound 7k, which was readily obtained
by isomerization of 6k in basic media, allowed us to establish
the configuration shown in Scheme 4.
Figure 1. Compounds 6 prepared from achiral aldehydes and allene 5.
Bn=benzyl.
greater than 95% diastereomeric purity.[12] Only small
amounts of compounds 7 (arising from 6 by migration of
the carbamate group in the work-up and/or chromatographic
purification) were observed. NOESYexperiments on samples
of 7b–g confirmed the relative configuration assigned to 6a,
which can be rationalized though a transition state similar to
IV in the addition step (see Scheme 2).
To determine the absolute configuration of adducts 6i and
6j we sought to transform them into crystalline derivatives.
Thus, the silicon protecting group was selectively removed
Seeking to obtain enantioenriched compounds by this
process, we reasoned that obvious methods to use would be
either a chiral dialkylborane in the hydroboration step or to
use an enantiopure aldehyde. As a series of experiments using
(Ipc)2BH (Ipc = isopinocampheyl) or dilongifolylborane led
to moderate yields and low stereoselectivities we therefore
turned our attention to using chiral aldehydes. Once more,
excellent yields were obtained with representatives chiral
aldehydes derived from (S)-lactaldehyde (8; Table 1, entry 1),
(R)-glyceraldehyde (9; entry 2),[13a–c] (R)-3-hydroxy-2-meth-
ylpropanal (10; entry 3), and (S)-phenylalaninal (11;
entry 4),[13d,e] even when a 1:1 ratio of allene to aldehyde
was used (entry 1). Notably, the stereofacial selectivity for
aldehydes 8 and 9 was very high as only one stereoisomer was
detected by 1H NMR spectroscopy. The stereoselectivity
observed for aldehydes 10 and 11 was also high (10:1 and
8:1, respectively).
Scheme 4. Preparation of derivatives 7k and 12 from 6k. PPTS=pyr-
idinium 4-toluenesulfonate, TBAF=tetra-n-butylammonium fluoride,
THF= tetrahydrofuran.
Angew. Chem. Int. Ed. 2009, 48, 4202 –4205
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4203

Communications
using HF or TBAF and the carbamate rings were hydrolyzed
Keywords: allenes · amino alcohols · asymmetric synthesis ·
boranes · substituted serines
.
in basic media. The resulting polyols 13 and 14 were easily
crystallized and their corresponding single-crystal X-ray
analysis agreed with the configurations shown in
Scheme 5.[15,16]
[1] For recent reviews on the stereoselective formation of quater-
nary carbon centers, see: a) I. Denissova, L. Barriault, Tetrahe-
dron 2003, 59, 10105 – 10146; b) J. Christoffers, A. Baro, Adv.
Synth. Catal. 2005, 347, 1473 – 1482; c) B. M. Trost, C. Jiang,
Synthesis 2006, 369 – 396; d) P. G. Cozzi, R. Hilgraf, N. Zimmer-
mann, Eur. J. Org. Chem. 2007, 5969 – 5994.
[2] For recent reviews on the stereoselective construction of a,a-
disubstituted a-amino acids, see: a) C. Cativiela, M. D. Dꢀaz-de-
Villegas, Tetrahedron: Asymmetry 1998, 9, 3517 – 3599; b) C.
Cativiela, M. D. Dꢀaz-de-Villegas, Tetrahedron: Asymmetry
2000, 11, 645 – 732; c) Y. Ohfune, T. Shinada, Eur. J. Org.
Chem. 2005, 5127 – 5143; d) S. H. Kang, S. Y. Kang, H.-S. Lee,
A. J. Buglass, Chem. Rev. 2005, 105, 4537 – 4558; e) C. Cativiela,
M. D. Dꢀaz-de-Villegas, Tetrahedron: Asymmetry 2007, 18, 569 –
623.
Scheme 5. Preparation of derivatives 13 and 14 from 6i and 6j.
[3] C. E. Masse, A. J. Morgan, J. Adams, J. S. Panek, Eur. J. Org.
Chem. 2000, 2513 – 2528.
Indeed, we envisaged that enantiopure 13 (or its enantio-
mer arising from (R)-lactaldehyde) could be a versatile
starting material for the synthesis of quaternary a-amino-b-
hydroxyacids. Remarkably, the three carbon substituents (a–
c) in 15 that are attached to the quaternary centre are
amenable to transformation into either a carboxylic acid or a
hydroxymethyl group (Scheme 6). In particular, selective
protection of the primary alcohol with a tert-butyldiphenyl-
silyl group, and subsequent oxidative cleavage of the 1,2-diol
moiety afforded aldehyde (R)-16, which was easily trans-
[4] a) D. Kluepfel, J. Bagli, H. Baker, M.-P. Charest, A. Kudelski,
S. N. Sehgal, C. Vezina, J. Antibiot. 1972, 25, 109 – 115; b) T.
Fujita, K. Inoue, S. Yamamoto, T. Ikumoto, S. Sasaki, R.
Toyama, K. Chiba, Y. Hoshino, T. Okumoto, J. Antibiot. 1994, 47,
208 – 215.
[5] For a review, see: a) R. W. Hoffmann, Pure Appl. Chem. 1988,
60, 123 – 130; b) Y. Yamamoto, N. Asao, Chem. Rev. 1993, 93,
2207 – 2293; c) W. R. Roush in Methods of Organic Chemistry
(Houben-Weyl), Vol. E21 (Eds.: G. Helmchen, R. W. Hoffmann,
J. Mulzer, E. Schaumann), Thieme, Stuttgart, 1995, pp. 1410 –
1461; d) P. V. Ramachandran, Aldrichimica Acta 2002, 35, 23 –
35; e) S. E. Denmark, J. Fu, Chem. Rev. 2003, 103, 2763 – 2794.
[6] For a recent review, see: I. Marek, G. Sklute, Chem. Commun.
2007, 48, 1683 – 1691.
[7] For their use in the stereoselective preparation of 1,3-butadienes,
see: a) C. Liu, K. K. Wang, J. Org. Chem. 1986, 51, 4733 – 4734;
b) W. H. Pearson, K.-C. Lin, Y.-F. Poon, J. Org. Chem. 1989, 54,
5814 – 5819; c) K. K. Wang, C. Liu, Y. G. Gu, F. N. Burnett, P. D.
Sattsangi, J. Org. Chem. 1991, 56, 1914 – 1922; see also: d) R. W.
Hoffmann, R. Metternich, Leibigs. Ann. Chem. 1985, 2390 –
2402; e) P. V. Ramachandran, A. Chaterjee, Org. Lett. 2008, 10,
1195 – 1198.
[8] Y. Horino, M. Kimura, S. Tanaka, T. Okajima, Y. Tamaru, Chem.
Eur. J. 2003, 9, 2419 – 2438.
[9] A set of experiments featuring different hydroborating agents
(9-BBN (9-borabicyclo[3.3.1]nonane), (Ipc)2BH, catecholbor-
ane), solvents (diethyl ether or THF) and temperatures (from
À788C to 408C) gave lower yields and/or selectivities.
[10] As expected, Z allyldicyclohexylborane does not seem to
undergo isomerization to its E isomer at room temperature. In
this case, the known boratropic allylic rearrangement that causes
Z/E isomerisation in crotyldialkylboranes seems to be inhibited
by steric hindrance. In fact, when the hydroboration step was
performed in CH2Cl2 at reflux, the same single stereoisomer was
obtained, albeit in lower yield.
[11] Isomerization of 6a to 7a was not surprising as related base-
catalyzed processes that involve N-acyl or N-alkyl cyclic
carbamates have been reported in the literature: a) W. R.
Roush, M. A. Adam, J. Org. Chem. 1985, 50, 3752 – 3757;
b) D. L. Boger, M. W. Ledeboer, M. Kume, J. Am. Chem. Soc.
1999, 121, 1098 – 1099; c) S. P. Bew, S. D. Bull, S. G. Davies, E. D.
Savory, D. J. Watkin, Tetrahedron 2002, 58, 9387 – 9401.
[12] However, the more hindered 2,2-dimethylpropanal as well as
acetophenone were resistant to such additions.
Scheme 6. Stereoselective synthesis of protected a-vinylserine.
DMAP=4-dimethylaminopyridine, py=pyridine.
formed to the corresponding protected a-vinylserine (R)-17.
It is worth noting that a-vinylserines are competitive inhib-
itors of serine hydroxymethyl transferase.[17,18]
In conclusion, we have established a new approach to
afford highly functionalized quaternary aminopolyols in
which two adjacent stereocenters are formed with high
stereoselective control. The use of chiral a-substituted
aldehydes provided the highly functionalized enantiopure
building blocks 16 and 17 in excellent yields. These adducts
are expected to have important applications in the synthesis
of quaternary a-amino-b-hydroxyacids.
Received: February 2, 2009
Revised: March 26, 2009
Published online: May 4, 2009
[13] a) P. Michel, S. V. Ley, Angew. Chem. 2002, 114, 4054 – 4057;
Angew. Chem. Int. Ed. 2002, 41, 3898 – 3901; b) S. V. Ley, P.
4204 www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 4202 –4205

Angewandte
Chemie
Michel, Synthesis 2004, 147 – 150; for a recent review on butane-
1,2-diacetales, see: c) E. Lence, L. Castedo, C. Gonzꢁlez-Bello,
Chem. Soc. Rev. 2008, 37, 1689 – 1708; d) L. A. Paquette, T. M.
Mitzel, M. B. Isaac, C. F. Crasto, W. W. Schomer, J. Org. Chem.
1997, 62, 4293 – 4301; for a review on N,N-dibenzylamino
aldehydes, see: e) M. T. Reetz, Chem. Rev. 1999, 99, 1121 – 1162.
[14] The pattern of the proton coupling constant for both cis and
trans 4-substituted 5-methyl-dioxane rings are well established:
a) D. A. Evans, B. D. Allison, M. G. Yang, C. E. Masse, J. Am.
Chem. Soc. 2001, 123, 10840 – 10852; see also: b) G. Bifulco, P.
Dambruoso, L. Gomez-Paloma, R. Riccio, Chem. Rev. 2007, 107,
3744 – 3779.
10353; no. independent reflections = 3261 (R(int) = 0.0574);
restraints/parameters = 11/207; full-matrix least-squares refine-
ment on F2; no. absorption correction; m = 0.230 mmÀ1: final R
indices (I > 2s(I)) are R1 = 0.0375 and wR2 = 0.0988: largest diff.
Peak and hole = 0.291 and À0.328 eꢄÀ3
. Compound 14:
(C20H31NO8S): 0.19 ꢃ 0.18 ꢃ 0.16 mm: monoclinic P21; a =
11.879(7), b = 7.824(4), c = 12.470(4), b = 109.55(2)8; V=
1092.2(9) ꢄ3 (Z = 2); 1calcd = 1.355 MgmÀ3; 2qmax = 30.008; 17 <
h < 17, À9 < k < 10, À17 < l < 18; l = 0.71073 ꢄ; T= 293 K; no.
reflections = 12395; no. independent reflections = 5757 (R(int) =
0.0526); restraints/parameters = 7/277; full-matrix least-squares
refinement on F2; no. absorption correction; m = 0.194 mmÀ1
;
[15] The observed configuration agrees with that expected from the
transition state IV shown in Figure 1, in which the chiral
aldehyde is arranged according to the Felkin–Anh model. For a
review on the addition of achiral allylboron reagents to chiral
aldehydes, see: W. R. Roush in Methods of Organic Chemistry
(Houben-Weyl), Vol. E21 (Eds.: G. Helmchen, R. W. Hoffmann,
J. Mulzer, E. Schaumann), Thieme, Stuttgart, 1995, pp. 1430 –
1439; for a review on Cramꢂs and Felkin–Anh models, see: A.
Mengel, O. Reiser, Chem. Rev. 1999, 99, 1191 – 1223.
[16] X-ray quality crystals of 13 and 14 were grown by slow
evaporation of a dichloromethane solution. Compound 13:
(C14H21NO5S): 0.2 ꢃ 0.1 ꢃ 0.1 mm; orthorhombic P212121; a =
6.118(3), b = 13.304(4), c = 18.982(6) ꢄ; V= 1545.0(10) ꢄ3 (Z =
4); 1calcd = 1.356 MgmÀ3; 2qmax = 28.378; À7 < h < 8, À16 < k <
16, À24 < l < 25; l = 0.71073 ꢄ; T= 293 K; no. reflections =
final R indices (I > 2s(I)) are R1 = 0.0558 and wR2 = 0.1393:
largest diff. Peak and hole = 0.370 and À0.314 eꢄÀ3
.
CCDC 707178 (13) and 707179 (14) contain the supplementary
crystallographic data for this paper. These data can be obtained
free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
[17] a) S. J. B. Tendler, C. H. Schwalbe, M. D. Threadgill, M. J.
Tisdale, L. Schirch, J. Chem. Soc. Perkin Trans. 2 1989, 431 –
436; b) D. B. Berkowitz, J. M. McFadden, E. Chisowa, C. L.
Semerad, J. Am. Chem. Soc. 2000, 122, 11031 – 11032.
[18] A sample of (R)-17 was converted into (R)-a-vinylserine by
sequential removal of the TBDPS (TBAF·3H2O, THF) and tosyl
(naphthalene, Na, DME(1,2-dimethoxyethane)/THF 1:1) pro-
tecting groups. The spectroscopic data of the amino acid
obtained were identical to those reported in Ref. [17b].
Angew. Chem. Int. Ed. 2009, 48, 4202 –4205
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4205