Amino acid-based reoxidants for aminohydroxylation: Application to the construction of amino acid-amino alcohol conjugates

Article abstract of DOI:10.1002/anie.201103293

A viable nitrogen source for the aminohydroxylation reaction of terminal alkenes: By adding a N-O based reoxidant onto an amino acid acyl carbon atom, compounds were obtained that facilitated catalytic turnover and also promoted the conjugation of an amino acid with an alkene. High levels of regioselectivity were observed, as well as good stereoselectivity induced by catalytic amounts of a chiral ligand.

Full text of DOI:10.1002/anie.201103293

DOI: 10.1002/anie.201103293
Amino Acids as Nitrogen Sources
Amino Acid-Based Reoxidants for Aminohydroxylation: Application
to the Construction of Amino Acid–Amino Alcohol Conjugates**
Timothy J. Donohoe,* Cedric K. A. Callens, Aida Flores, Stefanie Mesch, Darren L. Poole, and
Ishmael A. Roslan
In recent years, there has been a surge in
reports of methodology aimed at synthesiz-
ing the 1,2-amino alcohol motif.[1] Despite
numerous studies, this key unit is still highly
sought after because of its occurrence in
many different organic compounds of value
ranging from natural products to pharma-
ceutical agents.[2] Methods that allow for the
Scheme 1. Amino acids as stable reoxidants and nitrogen sources for aminohydroxylation?
PG: protecting group.
regioselective and stereospecific addition of
nitrogen and oxygen across a double bond
are rare, and continue to be influenced by the
osmium-catalysed aminohydroxylation reaction.[3]
Recent developments of this reaction have wit-
nessed the introduction of N-O based reoxidants
for osmium that have made the process much more
reliable and high-yielding.[4] Herein, we report for
the first time the use of natural amino acids as a
scaffold for such reoxidants, allowing their conju-
gation with alkenes in a regio- and stereocontrolled
process. This sequence enables rapid access to well-
defined and enantiopure amino alcohols using a
transition metal and a chiral ligand in catalytic
amounts.
We set out to explore the role of amino acids 1
Scheme 2. Conversion of amino acids into reoxidants for aminohydroxylation.
as nitrogen sources in the aminohydroxylation
reaction and to search for the factors that control
both the regio- and stereoselectivity of the process
(Scheme 1). We chose two classes of terminal alkenes, namely
substituted styrenes 2 and acrylate esters 3, as the alkene
partners because the product amino acid conjugates are
potentially useful in synthesis, especially in medicinal chemis-
try.[5]
Early work had shown that the nitrogen atom of an amino
acid should be doubly protected in order to give the best
results in aminohydroxylation. Therefore, the synthetic route
began with the conversion of an amino acid into its
phthalimide-protected form (phthalic anhydride, 93–100%
yield[6]) and subsequent transformation of the free carboxylic
acid into an O-acyl hydroxamic acid reoxidant (Scheme 2).
Taking the case of l-valine and l-alanine as examples,
procedure A took place without significant racemization
and the products were formed in 94–98% ee. An exception
to this reactivity involved the use of a-phenylglycine, which
could not be coupled to hydroxylamine derivatives without
significant racemization. An alternative pathway (proce-
dure B) to the activated amino acids involved the separate
formation of O-mesitoyl hydroxylamine (as its HCl salt),[7]
which could be coupled to the free acid using a peptide-
coupling agent (DMTMM);[8] in the case examined (7 f), this
reagent gave no racemization in the product, but was less
convenient to use on a large scale. X-ray analysis proved the
identity of compound 7d.[9]
Armed with these amino acid-based reoxidants, we were
then able to examine their behavior in aminohydroxylation
reactions. In the first instance we chose to examine the
combination of l-valine-derived reoxidant 7 f with p-
methoxystyrene (8; Table 1). A screen of the solvents that
are typically used in aminohydroxylation showed that aque-
ous THF gave the best results in terms of high yield and
[*] Prof. T. J. Donohoe, Dr. C. K. A. Callens, Dr. A. Flores, Dr. S. Mesch,
D. L. Poole, I. A. Roslan
Department of Chemistry, University of Oxford
Chemistry Research Laboratory
Mansfield Road, Oxford, OX1 3TA (UK)
E-mail: timothy.donohoe@chem.ox.ac.uk
[**] We thank the EPSRC, the European Union, the SNSF, and
AstraZeneca for supporting this project. We also thank the Oxford
Chemical Crystallography Service for the use of their instrumenta-
tion.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201103293.
Angew. Chem. Int. Ed. 2011, 50, 10957 –10960
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
10957

Communications
Table 2: The scope of the amino acid-based reoxidant.
Table 1: Addition of l-valine-based reoxidant 7 f to 8.
R
Regioselectivity
d.r. (14:15)
Yield [%]
H (a)
CH3 (b)
CH2CH(CH3)2 (c)
C(CH3)3 (d)
CH2Ph (e)
5:1
–
72
79
81
64
68
14:1
12:1
9:1
6:1
4:1
10:1
6:1
8:1
Entry
Solvent/Ligand
(9+10):11[a]
Yield [%] (d.r., 9:10)
1
2
3
THF
33:1
13:1
6.5:1
91 (2:1)
89 (15:1)
72 (1:1.3)
THF/(DHQD)2AQN[b]
THF/(DHQ)2AQN[b]
matched pairing of ligand and amino acid substrate, the use
of (DHQ)2AQN was less successful and could only overturn
the facial bias of the system to a small degree (Table 1,
entry 3). The identity of the major regio- and diastereomer 9
was proven by X-ray crystallography which revealed that the
configuration at the benzylic center is that to be expected
using a DHQD-based ligand.[9,11] Moreover, the identity of
the alternative diastereomer 10 was confirmed when a 2:1
mixture of 9 and 10 was oxidized to a single benzylic ketone
12 upon treatment with Dess–Martin periodinane (DMP)
(Scheme 3). Finally, in order to prove the synthetic utility of
the sequence, the major diastereomer 9 was deprotected using
hydrazine[12] to furnish 13 without detectable epimerization of
the stereogenic centers (Scheme 3).
Next, we explored the range of amino acids that were
compatible with this sequence, using 8 as a convenient alkene
substrate (Table 2; all amino acids were activated as shown in
Scheme 2). Although a-phenylglycine did give good yields in
this sequence (Table 2, last entry) it is not shown
above because of problems associated with rac-
emization during active-ester formation (see
above). The identity of the major diastereomer
was assumed by analogy to the l-valine example
above; in each case, aminohydroxylation without
a chiral ligand gave high regioselectivity but low
diastereoselectivity. This bias could be re-
enforced with a matching DHQD-based ligand
(only the results employing the DHQD ligand are
shown in Table 2).
[a] The diastereoselectivity within the minor regioisomer 11 was not
determined but the structure of one isomer was proven by X-ray
crystallography.[9] [b] 10 mol% of chiral ligand was used throughout.
Alternative spacer groups such as PHAL and PYR gave inferior
selectivities (see the Supporting Information).
regioselectivity for the benzylic alcohols 9 and 10 (Table 1,
entry 1). This high regioselectivity for benzylic alcohols
makes our reaction complementary to past procedures
which tend to give benzylic amines.[10,11] Moreover, we also
discovered that the low levels of diastereoselectivity for 9
imparted by the chiral amino acid reoxidant could be
augmented by the addition of a catalytic amount of Sharplessꢀ
chiral ligands (the optimal result was obtained with
(DHQD)2AQN giving d.r. = 15:1, Table 1, entry 2).[3d,11]
While this particular combination clearly represents
a
As before, oxidation of a mixture of diaste-
reomers from the reactions shown in Table 2
gave, in each case, a single benzylic ketone thus
1
allowing us to assign the peaks in the H NMR
spectra that were attributable to diastereomers as
opposed to regioisomers. Clearly, the amino acids
glycine, l-alanine, l-leucine, l-tert-butylglycine,
and l-phenylalanine are all compatible with this
methodology.
Next, we examined the range of substituted
styrenes that would react with the l-valine-based
reoxidant 7 f to gauge the generality of the
method for this class of alkenes (Table 3). The
Scheme 3. Proof of regioselectivity and phthalimide deprotection.
10958 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 10957 –10960

Table 3: Aminohydroxylation of substituted styrenes; X-ray analysis
proved the identity of compound 16a.[9]
Changing the terminal alkene substrate from being
electron-rich to being electron-deficient enabled the synthesis
of dipeptides containing valine and isoserine, because the
imido–osmium oxidant placed the bulky amino acid at the
terminus of the alkene during aminohydroxylation (Table 5).
Table 5: Aminohydroxylation of acrylate esters.
Ar
Ligand
–
(DHQD)2AQN 15:1
(DHQ)2AQN
Regioselectivity d.r. (16:17) Yield [%]
C6H5
C6H5
C6H5
20:1
1.5:1
13:1
1:1.4
80
75
68
7:1
p-FC6H4
p-FC6H4
p-FC6H4
–
15:1
2:1
6:1
1:1.3
81
79
71
Ligand
Regioselectivity
d.r. (21:22)
Yield [%]
(DHQD)2AQN 17:1
(DHQ)2AQN
–
9:1
8:1
15:1
12:1
1.5:1
10:1
1:4
83
84
68
82
15:1
(DHQD)2AQN
(DHQ)2AQN
(DHQ)2PHAL
p-AcOC6H4
p-AcOC6H4 (DHQD)2AQN 12:1
p-AcOC6H4 (DHQ)2AQN
–
13:1
1.5:1
8:1
1:2
80
78
69
1:5
6:1
2-naphthyl
2-naphthyl (DHQD)2AQN 13:1
2-naphthyl (DHQ)2AQN 7:1
–
18:1
2:1
12:1
1:1.3
79
77
70
We found that the aminohydroxylation of aryl acrylate ester
20 was especially productive and allowed valine–isoserine
residues 21 and 22 to be prepared. Finding acetonitrile to be
the optimal solvent, we observed matching (DHQD) and
mismatching (DHQ) effects as before, and the selection of a
particular ligand allowed the formation of either l-21 or d-22
amino acids at will.[14] In this case, the DHQ ligand was better
matched with a PHAL spacer rather than with an AQN
spacer, which was not the case with the styrene derivatives. It
is hoped that this new approach to making peptides will have
several uses in organic synthesis.[15]
In conclusion, we have shown that protected amino acids
are viable nitrogen sources and reoxidants for the intermo-
lecular aminohydroxylation reaction, with high levels of
regio- and stereoselectivity being achieved during the oxida-
tion of both electron-rich and electron-deficient terminal
alkenes. The sequence enables the conjugation of amino acids
and alkenes in a concise manner, and will lead to the efficient
and rapid synthesis of amino acid–amino alcohol conjugates
which are expected to have widespread use in organic
synthesis. Further investigations regarding the scope of
substrates with non-terminal alkenes are currently ongoing.
placement of groups at the para-position on the aromatic ring,
or the introduction of alternative aromatic motifs did not
affect the outcome of the sequence, which revealed that:
1) high regioselectivity for the benzylic alcohol was main-
tained throughout; 2) the matching effect of DHQD-based
ligands was supported; 3) the mismatching effect of DHQ-
based ligands was observed. In each case, the subsequent
oxidation of a mixture of diastereomers gave a single benzylic
ketone (see the Supporting Information).
If we used a 1,1-disubstituted alkene, such as a-methyl-
styrene, the aminohydroxylation still proceeded with high
regioselectivity and adequate stereoselectivity with either
diastereomer 18 or 19 being accessible (Table 4). Note that in
this case, the regiochemistry of the major isomers was
assigned by NMR experiments, and the configuration at the
newly formed stereogenic center is assumed to be that as
shown (that is, under the control exerted by the chiral
ligands[13]).
Received: May 13, 2011
Revised: September 7, 2011
Published online: September 26, 2011
Table 4: Aminohydroxylation of a-methylstyrene.
Keywords: amino acids · aminohydroxylation ·
.
homogeneous catalysis · osmium · styrenes
[1] a) T. J. Donohoe, C. K. A. Callens, A. Flores, A. R. Lacy, A. H.
Rath, Chem. Eur. J. 2011, 17, 58; b) E. J. Alexanian, C. Lee, E. J.
Sorensen, J. Am. Chem. Soc. 2005, 127, 7690; c) D. J. Wardrop,
E. G. Bowen, R. E. Forslund, A. D. Sussman, S. L. Weerasekera,
J. Am. Chem. Soc. 2010, 132, 1188; d) K. S. Williamson, T. P.
Yoon, J. Am. Chem. Soc. 2010, 132, 4570; e) T. de Haro, C.
Ligand
Regioselectivity
d.r. (18:19)
Yield [%]
–
25:1
11:1
8:1
2:1
3:1
1:2
64
62
53
(DHQD)2AQN
(DHQ)2AQN
Angew. Chem. Int. Ed. 2011, 50, 10957 –10960
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 10959

Communications
Nevado, Angew. Chem. 2011, 123, 936; Angew. Chem. Int. Ed.
2011, 50, 906.
J. Org. Chem. 1998, 63, 4248; b) A. Falchi, G. Giacomelli, A.
Porcheddu, M. Taddei, Synlett 2000, 275.
[9] CCDC 824647 (7d), 824648 (9), 824649 (11), and 824650 (16a)
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.
[10] a) K. L. Reddy, K. R. Dress, K. B. Sharpless, Tetrahedron Lett.
1998, 39, 3667; b) P. OꢀBrien, S. A. Osborne, D. D. Parker,
Tetrahedron Lett. 1998, 39, 4099; c) D. J. Michaelis, C. J. Shaffer,
T. P. Yoon, J. Am. Chem. Soc. 2007, 129, 1866; however, see:
d) V. Nesterenko, J. T. Byers, P. J. Hergenrother, Org. Lett. 2003,
5, 281; e) P. Wu, R. Higraf, V. V. Fokin, Adv. Synth. Catal. 2006,
348, 1079.
[11] K. L. Reddy, K. B. Sharpless, J. Am. Chem. Soc. 1998, 120, 1207.
[12] a) H. R. Ing, R. H. F. Manske, J. Chem. Soc. 1926, 129, 2348;
b) R. T. Skerlj, G. J. Bridger, A. Kaller, E. J. McEachern, J. B.
Crawford, Y. Zhou, B. Atsma, J. Langille, S. Nan, D. Veale, T.
Wilson, C. Harwig, S. Hatse, K. Princen, E. De Clercq, D. Schols,
J. Med. Chem. 2010, 53, 3376.
[13] While the aminohydroxylation of a-methylstyrene is known
(a) K. B. Sharpless, D. W. Patrick, L. K. Truesdale, S. A. Biller, J.
Am. Chem. Soc. 1975, 97, 2305; b) K. MuÇiz, I. Almodovar, J.
Streuff, M. Nieger, Adv. Synth. Catal. 2006, 348, 1831), the
asymmetric variant is not. Therefore, our assignment is by
analogy with the asymmetric dihydroxylation reaction of a-
methylstryrene, which gives the same sense of enantioselectivity
as for styrene (K. B. Sharpless, W. Amberg, Y. L. Bennani, G. A.
Crispino, J. Hartung, K. S. Jeong, H. L. Kwong, K. Morikawa, Z.
M. Wang, D. Xu, X. L. Zhang, J. Org. Chem. 1992, 57, 2768).
[14] The sense of stereoselectivity was assigned by analogy to related
reactions, see: a) G. Li, H. T. Chang, K. B. Sharpless, Angew.
Chem. 1996, 108, 449; Angew. Chem. Int. Ed. Engl. 1996, 35, 451;
b) K. Lowpetch, D. W. Young, Org. Biomol. Chem. 2005, 3, 3348.
[15] For examples of both natural and nonnatural peptide sequences
that contain isoserine and also exhibit biological activity, see:
a) R. Mariani, G. Granata, S. I. Maffioli, S. Serina, C. Brunati, M.
Sosio, A. Marazzi, A. Vannini, D. Patel, R. White, R. Ciabatti,
Bioorg. Med. Chem. Lett. 2005, 15, 3748; b) B. S. Coller, K. T.
Springer, L. E. Scudder, J. L. Kutok, M. Ceruso, G. D. Prestwich,
J. Biol. Chem. 1993, 268, 20741.
[2] a) M. Y. Kozutsumi, Y. S. Nakamura, T. Fujita, T. Kawasaki,
Biochem. Biophys. Res. Commun. 1995, 211, 396; b) J. C.
Madureira de Freitas, Jr., B. Du Rocher DꢀAguiar Silva, F. Wal-
demir, M. Wallace, E. Saul Furquim, A. Werneck, J. A. Mor-
gado-Dꢁaz, Cancer Chemother. Pharmacol. 2011, in press; c) I.
Paterson, G. J. Naylor, T. Fujita, E. Guzmꢂn, A. E. Wright,
Chem. Commun. 2010, 46, 261; d) S. Yamashita, Exp. Oncol.
2011, 33, 24.
[3] a) H. C. Kolb, K. B. Sharpless, Transition Met. Org. Synth. 1998,
2, 243 – 260; b) D. Nilov, O. Reiser, Adv. Synth. Catal. 2002, 344,
1169; c) P. OꢀBrien, Angew. Chem. 1999, 111, 339; Angew. Chem.
Int. Ed. 1999, 38, 326; d) M. Bruncko, G. Schlingloff, K. B.
Sharpless, Angew. Chem. 1997, 109, 1580; Angew. Chem. Int. Ed.
Engl. 1997, 36, 1483; e) Z. P. Demko, M. Bartsch, K. B. Sharp-
less, Org. Lett. 2000, 2, 2221.
[4] a) T. J. Donohoe, M. J. Chughtai, D. J. Klauber, D. Griffin, A. D.
Campbell, J. Am. Chem. Soc. 2006, 128, 2514; b) T. J. Donohoe,
C. J. R. Bataille, W. Gattrell, J. Kloesges, E. Rossignol, Org. Lett.
2007, 9, 1725; c) T. J. Donohoe, C. K. A. Callens, A. L. Thomp-
son, Org. Lett. 2009, 11, 2305; d) L. Harris, S. P. H. Mee, R. H.
Furneaux, G. J. Gainsford, A. Luxenburger, J. Org. Chem. 2011,
76, 358.
[5] a) R. R. Ruffolo, J. M. Stadel, J. P. Hieble, Med. Res. Rev. 1994,
14, 229; b) C. Bombelli, L. Giansanti, P. Luciani, G. Mancini,
Curr. Med. Chem. 2009, 16, 171; c) W. M. Pardridge, Mol.
Intervention 2003, 3, 90; d) F. Pinne, I. Cacciatore, C. Cornacchia,
A. Mollica, P. Sozio, L. S. Cerasa, A. Iannitelli, A. Fontana, C.
Nasuti, A. Di Stefano, Amino Acids 2011, DOI: 10.1007s00726-
010-0804-z.
[6] a) A. G. Griesbeck, H. Mauder, Angew. Chem. 1992, 104, 97;
Angew. Chem. Int. Ed. Engl. 1992, 31, 73; b) A. G. Griesbeck, H.
Mauder, I. Mꢃller, Chem. Ber. 1992, 125, 2467.
[7] This compound was synthesised from commercially available N-
Boc-hydroxylamine. Acylation and subsequent Boc deprotec-
tion in acidic medium afforded the O-mesitoyl hydroxylamine
salt in 92% yield over the two steps.
[8] DMTMM is 4-(4,6-dimethoxy-1,3,5-triazin-2-yl)-4-methylmor-
´
´
pholinium chloride: a) Z. J. Kaminski, P. Paneth, J. Rudzinski,
10960 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 10957 –10960