Catalytic enantio- and diastereoselective alkylations with cyclic sulfamidates

Article abstract of DOI:10.1002/anie.200905329

(Figure Presented) Open for business: The enantio- and diastereoselective nucleophilic ring opening of five-membered and six-membered cyclic sulfamidates under asymmetric phase-transfer catalysis is presented. A range of pro-nucleophiles have been successfully alkylated in good yields and in good to excellent enantioselectivites.

Full text of DOI:10.1002/anie.200905329

Communications
DOI: 10.1002/anie.200905329
Phase-Transfer Catalysis
Catalytic Enantio- and Diastereoselective Alkylations with Cyclic
Sulfamidates**
Thomas A. Moss, Beatriz Alonso, David R. Fenwick, and Darren J. Dixon*
The enantioselective construction of derivatives of g-amino
butyric acid and d-amino pentanoic acid from simple starting
materials using asymmetric catalysis provides convenient
access to a range of structurally diverse natural products,
pharmaceutical compounds, and potential building blocks for
g-peptides and foldamer chemistry.[1] Several natural products
containing the aminoethylene and aminopropylene scaffolds
attached to a quaternary stereocenter have been isolated.
Developments in the field of enantioselective Michael
group could be cleaved under mild conditions in some cases,
we believed that a method which encompassed a wider range
of nitrogen-protecting groups would be more desirable.
Furthermore, access to the aminopropylene unit
(CH2CH2CH2NHP) remained elusive. Anticipating that aze-
tidines would lack the required reactivity to be used as three-
carbon electrophiles,[5] we considered cyclic sulfamidates as
potential two- or three-carbon electrophile candidates.[6,7]
Seminal work by Lubell and Wei[8] and extensive studies by
Gallagher and co-workers[9,10] have found that five-membered
and six-membered cyclic sulfamidates are useful precursors
for the synthesis of pyrrolidine and piperidinone alkaloids. In
those studies, methylene carbon acids were typically used as
the nucleophile, with chiral enantiopure electrophiles. To the
best of our knowledge, there have been no reports of a
catalytic enantioselective nucleophilic ring-opening of cyclic
sulfamidates with carbon-centered nucleophiles, despite the
synthetic advantages of such an approach. We reasoned that a
base-catalyzed reaction would be challenging, owing to the
low basicity of sulfamic acid salts that can be formed from the
ring opening of cyclic sulfamidates. Accordingly, we believed
that an enantioselective ring-opening of cyclic sulfamidates
could be realized using asymmetric phase-transfer catalysis
with a stoichiometric base.[11,12] Attracted by the simplicity of
the approach, and the synthetic potential of the methodology,
we began our investigations. Herein, we report our findings
into the direct enantioselective catalytic alkylation reaction of
methine pro-nucleophiles with N-protected five-membered
and six-membered cyclic sulfamidates. Extension of the
procedure to include diastereoselective variants is also
described.
Preliminary studies were carried out using N-Boc-pro-
tected cyclic sulfamidate 1a and tert-butyl-1-methyl-2,5-
dioxopiperidine-3-carboxylate (2a) as a representative pro-
nucleophile (Boc = tert-butoxycarbonyl). Pleasingly, com-
plete consumption of the pro-nucleophile was observed
after 24 h at room temperature when powdered Cs2CO3 was
used as the base and tetrabutylammonium bromide (TBAB)
as the catalyst (Table 1, entry 1). Lower conversions were
generally observed when aqueous base mixtures were
employed (for example, Table 1, entry 2); in these cases a
larger excess of electrophile was required for completion of
the reaction, which is presumably due to partial hydrolysis of
the electrophile in the aqueous phase. As expected, when the
phase-transfer catalyst was omitted, reactivity was signifi-
cantly reduced (Table 1, entry 3). Interestingly, very low
conversions were observed when solid K2CO3 or K3PO4
were used as the base (Table 1, entries 4 and 5), highlighting
the importance of cesium carbonate in this reaction. Cin-
chona-derived phase-transfer catalyst 4a,[13] which we have
additions of carbonyl compounds to nitroolefins and acryl-
onitriles have been significant, with highly enantioselective
examples reported in both cases.[2,3]
We recognized that structures containing aminoethylene
or aminopropylene moieties could be accessed rapidly and
stereoselectively if suitable two-carbon or three-carbon nitro-
gen-containing electrophiles could be utilized. To this end, we
recently described both the base-catalyzed,[4a] and the phase-
transfer catalyzed enantio- and diastereoselective[4b] ring-
opening reactions of nitrogen-protected aziridines as a
method for the direct construction of g-amino butyric acid
derivatives. During the course of that study, we found that
sulfonyl protection of the nitrogen atom was necessary to
achieve acceptable levels of reactivity. Although the sulfonyl
[*] B. Alonso, Prof. Dr. D. J. Dixon
Department of Chemistry, Chemistry Research Laboratory
University of Oxford
Mansfield Road, Oxford, OX1 3TA (UK)
E-mail: darren.dixon@chem.ox.ac.uk
T. A. Moss
School of Chemistry, The University of Manchester (UK)
Dr. D. R. Fenwick
Pfizer Global Research & Development, Sandwich (UK)
[**] We gratefully acknowledge the EPSRC and Pfizer Global Research
and Development for a studentship (T.A.M.). We thank Andrew Kyle
and Katherine Bogle for X-ray structure determination, and the
Oxford Chemical Crystallography Service for the use of the
instrumentation.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200905329.
568
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 568 –571

Angewandte
Chemie
Table 1: Optimization studies on 2a using 1a
Table 2: Enantioselective ring-opening of various N-protected cyclic
sulfamidates 1a–1e with tert-butylindanone carboxylate 2b.
Entry Solvent
T
Conversion
Base
Catalyst
(mol%)
ee
[%]
[8C] 48 h
Entry
PG
n
3
Yield [%][a]
ee [%][b]
1
2
9:1 Tol/
CHCl3
9:1 Tol/
CHCl3
25
25
100
80
Cs2CO3 TBAB (10)
0
0
1
2
Boc
CbZ
o-CF3C6H4SO2
PO(OEt)2
Boc
1 (1a)
1 (1b)
1 (1c)
1 (1d)
2 (1e)
b
c
d
e
f
75 (18)
54 (31)
80
55 (33)
65 (20)
93
50%
aq
Cs2CO3
Cs2CO3
TBAB (10)
-
94
3[c]
4
96[d]
45
3
4
5
6
9:1 Tol/
CHCl3
9:1 Tol/
CHCl3
9:1 Tol/
CHCl3
9:1 Tol/
CHCl3
25
25
25
0
35
<10
<10
60
0
5
92
[a] Yield of isolated product (yield of isolated oxygen alkylation adduct
shown in parentheses). [b] Determined by HPLC analysis. [c] Reaction
performed at 08C due to low solubility of the electrophile. [d] >99% ee
after recrystallization; absolute stereochemistry determined by single
crystal X-ray crystallography (see the Supporting Information).[14] PG=
protecting group, CbZ=carboxybenzyl.
K2CO3 TBAB (10)
K3PO4 TBAB (10)
Cs2CO3 4a (10)
0
0
78
7
8
Toluene
9:1 Xy/
CHCl3
0
0
57
58
Cs2CO3 4a (10)
Cs2CO3 4a (10)
81
80
1a on a gram-quantity scale and at reduced catalyst loading
(4 mol%). Pleasingly, after 24 h at 08C the alkylation adduct
was obtained in 65% yield and in 82% ee (see the Supporting
Information for details).
9
10
11
Xylene
Xylene
Xylene
0
0
0
54
41
30
Cs2CO3 4a (10)
Cs2CO3 4a (5)
Cs2CO3 4a (2.5)
85
84
79
[a] Reactions were performed on a 0.4 mmol scale in 2 mL of solvent.
Conversion determined by H NMR spectroscopy. Enantiomeric excess
determined by HPLC analysis using a Chiralpak AD column.
The scope of the reaction with respect to the pro-
nucleophile was then investigated. Previous reports[4b,c,13]
have shown that cinchona-derived phase-transfer catalyst 4a
reacts well with cyclic b-keto esters bearing a tert-butyl ester
substituent. Wanting to expand the synthetic utility of the
reaction, we subjected various cyclic[15] pro-nucleophiles to
our optimal reaction conditions. Pleasingly, good to high
enantioselectivities[16] were observed for a range of both five-
membered and six-membered cyclic systems, several of which
are novel substrates in asymmetric phase-transfer catalysis.
Substituents were introduced onto the indanone scaffold
(Scheme 1, 3g and 3h) without considerably affecting stereo-
induction (90 and 85% ee, respectively; cf. 3b, 93% ee).
Succinimide (3i and 3m) and lactone (3j and 3l) pro-
nucleophiles performed well, giving the alkylation products in
high yields (87–91%) and in up to 86% ee. As has been
previously demonstrated, glutarimide-derived pro-nucleo-
phile 2a was an effective substrate, affording high levels of
stereoinduction; when 1e was used as the electrophile, the
product 3k was obtained in good yield and in an impressive
93% ee. The variety of ring sizes tolerated for both the
nucleophile and electrophile reactants, along with the mild
reaction conditions, allows for the facile synthesis of a diverse
range of products, with generally good stereocontrol
(Scheme 1).
1
previously reported as giving high levels of enantiocontrol in
aziridine ring-opening reactions,[4b] performed well, giving
alkylation adduct 3a following acidic workup in 78% ee
(Table 1, entry 6). Less-polar solvents were more successful
for high levels of enantiocontrol; for example, xylene gave an
85% ee (Table 1, entry 9). Catalyst loading could be dropped
to 2.5 mol% without considerably affecting enantioinduction
(Table 1, entry 11), although reaction rates were decreased.
Accordingly, and for convenience, further reactions were
conducted at 10 mol% catalyst loading.
With the optimal reaction conditions established, we
sought to probe the scope of nitrogen-protecting groups that
are amenable to this procedure (Table 2). In this case,
representative pro-nucleophile tert-butylindanone carboxyl-
ate (2b) was treated under mild phase-transfer conditions
with five-membered and six-membered cyclic sulfamidates
that bore a range of nitrogen-protecting groups (1a–1 e).
Carbamate-protected (Table 2, entries 1 and 2), sulfonyl-
protected (Table 2, entry 3) and phosphonate-protected
(Table 2, entry 4) electrophiles were screened, and in every
case but one, a highly stereoselective alkylation was observed
(up to 96% ee) following mild acid hydrolysis. In some cases,
competitive oxygen alkylation also occured (Table 2,
entries 1,2,4,5). We were pleased to observe that the six-
membered electrophile 1e (Table 2, entry 5) gave an alkyla-
tion adduct containing the protected aminopropylene scaffold
in 65% yield and in an excellent 92% ee; this opens up a
potential route to a wide spectrum of potential synthetic
targets. To test the scale-up potential of our methodology, tert-
butyl indanone carboxylate 2b was reacted with sulfamidate
To extend the procedure to diastereoselective reactions,
we employed enantiopure cyclic sulfamidates as electrophiles.
Using alanine as a convenient source of chirality, both
enantiomers of sulfamidate 1 f were synthesized in the hope
of observing “matched” and “mismatched” combinations of
substrate and catalyst control. Using achiral phase-transfer
catalyst TBAB, substrate control from (S)-alanine-derived
sulfamidate electrophile (S)-1 f was found to be highly
nucleophile dependent,[17,18] varying from poor with lactone
Angew. Chem. Int. Ed. 2010, 49, 568 –571
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
569

Communications
uously assigned by single crystal X-ray diffraction (see the
Supporting Information for details).[14]
The ring-opening reaction of substituted cyclic sulfami-
dates is known to be completely regioselective.[6] Exploiting
this, alkylation products which have 1,2- or 1,3-substitution
patterns could be accessed depending on the position of the
substituent on the sulfamidate ring. Having established that
C2-substituted cyclic sulfamidates were reactive, C1-substi-
tuted analogues were then investigated (Scheme 3).
Scheme 3. Divergent synthetic strategies available from cyclic sulfami-
date electrophiles. EWG=electron withdrawing group.
Scheme 1. The enantioselective ring opening of unsubstituted cyclic
sulfamidates by asymmetric phase-transfer catalysis. Reactions per-
formed on a 0.4 mmol scale in 2 mL xylene. ee determined by HPLC
analysis. For 3h, the reaction was performed at À208C in 2 mL of a
9:1 Xylene/CHCl3 solution (see the Supporting Information).
Accordingly, chiral sulfamidate 1g was synthesized both
as a racemate and as the single (S)-enantiomer. We chose
chiral pro-nucleophile (R)-2g as a model system, with the
intention of obtaining a measure of substrate control from the
nucleophile. Pleasingly, treatment of (R)-2g with one equiv-
alent of sulfamidate (S)-1g in the presence of a catalytic
amount of TBAB and cesium carbonate for 24 h at 408C,
followed by acid hydrolysis and thermal lactamization (as
demonstrated by Gallagher and co-workers),[9,10] afforded the
deprotected spirolactam 6 in good yield
2e (1:1 d.r.), to good with succinimide 2 f (9:1 d.r.; Scheme 2).
Lactone-derived pro-nucleophile 2e, which had displayed no
substrate control with TBAB, afforded diastereomeric alky-
and in 20:1 d.r. Pleasingly, the use of
catalyst 4a improved the diastereoselectiv-
ity of the reaction to 40:1, this suggested a
“matched” pairing of nucleophile and cat-
alyst (Scheme 4). When the pseudo-enan-
tiomer of catalyst 4a was used, lactam 6 was
obtained in a lower 10:1 d.r. (see the
Supporting Information).[19]
In conclusion, we have developed the
first enantioselective phase-transfer-cata-
lyzed ring-opening of five-membered and
six-membered cyclic sulfamidates. Under
mild conditions, good to excellent selectiv-
ities have been obtained for a range of
Scheme 2. The diastereoselective ring opening of chiral cyclic sulfamidates by phase-
transfer catalysis. Reactions were performed on a 0.4 mmol scale in 2 mL xylene. For the
complete catalyst screen, see the Supporting Information. The relative stereochemistry of
5c was determined by X-ray crystallography.
electrophiles and several novel pro-nucle-
ophile systems. Using single enantiomer
cyclic sulfamidates, moderate to high cata-
lyst controlled diastereoselectivities can be
observed. Finally, functionalized spirolac-
lation adducts 5a and 5b with catalyst 4a in high yields, and in
a moderate 5:1 d.r. in both cases. This level of diastereose-
lectivity was anticipated in light of the enantioselective
alkylation result (Scheme 1). With succinimide pro-nucleo-
phile 2 f, the matched pair of 4a with sulfamidate (S)-1 f
resulted in excellent diastereoselection towards 5c (45:1 d.r.)
and a minor amount of O-alkylation side-product, whereas,
interestingly, the mismatched pair of 4a with sulfamidate (R)-
1 f afforded the O-alkylation product 5d almost exclusively
(Scheme 2). The relative stereochemistry of 5c was unambig-
Scheme 4. Nucleophile-controlled and phase-transfer-catalyst-con-
trolled alkylation/lactamization.
570
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 568 –571

Angewandte
Chemie
[11] For reviews, see: a) T. Ooi, K. Maruoka, Angew. Chem. 2007,
tams have been synthesized using this procedure. Further
work to probe the scope of this method and its application in
total synthesis is currently ongoing and will be reported in due
course.
119, 4300; Angew. Chem. Int. Ed. 2007, 46, 4222; b) T.
Hashimoto, K. Maruoka, Chem. Rev. 2007, 107, 5656; c) B.
Lygo, B. I. Andrews, Acc. Chem. Res. 2004, 37, 518; d) M. J.
OꢀDonnell, Acc. Chem. Res. 2004, 37, 506.
[12] For selected reports of phase-transfer-catalyzed alkylations
forming a quaternary stereocenter, see: a) U. H. Dolling, P.
Davis, E. J. J. Grabowski, J. Am. Chem. Soc. 1984, 106, 466;
b) T. B. K. Lee, G. S. K. Wong, J. Org. Chem. 1991, 56, 872; c) S.-
S. Jew, Y.-J. Lee, J. Lee, M. J. Kang, B.-S. Jeong, J.-H. Lee, M-S.
Yoo, M-J. Kim, S.-h. Choi, J.-M. Ku, H.-g. Park, Angew. Chem.
2004, 116, 2436; Angew. Chem. Int. Ed. 2004, 43, 2382; d) T. Ooi,
T. Miki, M. Taniguchi, M. Shiraishi, M. Takeuchi, K. Maruoka,
Angew. Chem. 2003, 115, 3926; Angew. Chem. Int. Ed. 2003, 42,
3796; e) E. J. Park, M. H. Kim, D. Y. Kim, J. Org. Chem. 2004,
69, 6897; f) T. Ooi, K. Fukumoto, K. Maruoka, Angew. Chem.
2006, 118, 3923; Angew. Chem. Int. Ed. 2006, 45, 3839; g) D.
Uraguchi, Y. Asai, T. Ooi, Angew. Chem. 2009, 121, 747; Angew.
Chem. Int. Ed. 2009, 48, 733.
[13] a) T. B. Poulsen, L. Bernardi, J. Aleman, J. Overgaard, K. A.
Jørgensen, J. Am. Chem. Soc. 2007, 129, 441; b) L. Bernardi, J. L.
Cantarero, B. Neiss, K. A. Jørgensen, J. Am. Chem. Soc. 2007,
129, 5772; c) T. B. Poulsen, L. Bernardi, M. Bell, K. A. Jørgen-
sen, Angew. Chem. 2006, 118, 6701; Angew. Chem. Int. Ed. 2006,
45, 6551.
[14] CCDC 753701 (3d) and 753702 (5c) contain the supplementary
crystallographic data for this paper. These data can be obtained
free of charge from The Cambridge Crystallographic Data
Center via www.ccdc.cam.ac.uk/data_request/cif.
[15] Subjection of acyclic pro-nucleophile tert-butyl-2-methylacetoa-
cetate to our optimal conditions resulted in a 2:3 inseparable
mixture of O- and C-alkylated products, respectively.
[16] The stereochemistry of indanone adducts 3b, 3c, 3e–3h were
assigned by analogy to compound 3d (absolute configuration
determined by X-ray crystallography). Compounds 3j and 3l
were assigned by analogy to the cyclopentanone system (see
Ref. [13c]). Succinimide products 3i and 3m were assigned by
analogy to compound 5c in the matched system with (S)-1 f and
catalyst 4a (relative configuration determined by X-ray crystal-
lography). Glutarimide adducts 3a and 3k were assigned by
analogy to the succinimide, and cyclohexanone systems
(Ref. [13]).
Received: September 23, 2009
Published online: December 15, 2009
Keywords: alkylation · contiguous stereocenters ·
.
phase-transfer catalysis · ring-opening reactions ·
sulfur heterocycles
[1] For a review, see: C. M. Goodman, S. Choi, S. Shandler, W. F.
DeGrado, Nat. Chem. Biol. 2007, 3, 252.
[2] For a review, see: M. Santanu, J. Yang, S. Hoffmann, B. List,
Chem. Rev. 2007, 107, 5471.
[3] B. Wang, F. Wu, Y. Wang, X. Liu, L. Deng, J. Am. Chem. Soc.
2007, 129, 768.
[4] a) T. A. Moss, A. Alba, D. Hepworth, D. J. Dixon, Chem.
Commun. 2008, 2474; b) T. A. Moss, D. R. Fenwick, D. J.
Dixon, J. Am. Chem. Soc. 2008, 130, 10076. For a related study,
see: c) M. W. Paix¼o, M. Nielsen, C. B. Jacobsen, K. A. Jørgen-
sen, Org. Biomol. Chem. 2008, 6, 3467.
[5] For examples of azetidinium ions used as electrophiles with
carbon nucleophiles, see: F. Couty, O. David, B. Drouillat,
Tetrahedron Lett. 2007, 48, 9180.
[6] For a comprehensive review on the synthesis and reactions of
cyclic sulfamidates, see: R. E. Melendez, W. D. Lubell, Tetrahe-
dron 2003, 59, 2581.
[7] For seminal work in this area, see: a) J. E. Baldwin, A. C. Spivey,
C. J. Schofield, Tetrahedron: Asymmetry 1990, 1, 881; b) G. F.
Cooper, K. E. McCarthy, M. G. Martin, Tetrahedron Lett. 1992,
33, 5895; c) G. J. White, M. E. Garst, J. Org. Chem. 1991, 56,
3178; d) L. T. Boulton, H. T. Stock, J. Raphy, D. C. Horwell, J.
Chem. Soc. Perkin Trans. 1 1999, 1421; e) J. J. Posakony, T. J.
Tewson, Synthesis 2002, 859.
[8] L. Wei, W. D. Lubell, Can. J. Chem. 2001, 79, 94.
[9] For synthetic procedures, see: a) J. F. Bower, J. Svenda, A. J.
Williams, J. P. H. Charmant, R. M. Lawrence, P. Szeto, T.
Gallagher, Org. Lett. 2004, 6, 4727; b) J. F. Bower, P. Szeto, T.
Gallagher, Org. Lett. 2007, 9, 4909; c) J. F. Bower, S. Chakthong,
J. Svenda, A. J. Williams, R. M. Lawrence, P. Szeto, T. Gallagher,
Org. Biomol. Chem. 2006, 4, 1868; d) J. F. Bower, A. J. Williams,
H. L. Woodward, P. Szeto, R. M. Lawrence, T. Gallagher, Org.
Biomol. Chem. 2007, 5, 2636.
[10] For synthetic applications, see: a) J. F. Bower, T. Riis-Johannes-
sen, P. Szeto, A. J. Whitehead, T. Gallagher, Chem. Commun.
2007, 728; b) J. F. Bower, P. Szeto, T. Gallagher, Chem. Commun.
2005, 5793.
[17] Acyclic ethyl phenylcyanoacetate gave product 5e with (S)-1 f
and 4a (10 mol%) in an 84% yield and 3:2 d.r. (see the
Supporting Information).
[18] An electrophile-directed diastereoselective alkylation has
recently been described: S. P. Marsden, R. Newton, J. Am.
Chem. Soc. 2007, 129, 12600.
[19] When the reaction was conducted with racemic 1g, a 1:1 mixture
of diastereomers was obtained (see the Supporting Information).
Angew. Chem. Int. Ed. 2010, 49, 568 –571
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
571