Cleavable linkers for porous silicon-based mass spectrometry

Article abstract of DOI:10.1002/anie.200352803

Dip, wash, and analyze: Molecules can be attached to tailored porous silicon wafers by using Diels-Alder reactions and efficiently detached and detected in the laser pulse of a MALDI mass spectrometer in a positionally encoded manner (see scheme).

Full text of DOI:10.1002/anie.200352803

Angewandte
Chemie
Mass Spectrometry
stereospecific for cis-annelated systems, which suggests a
concerted mechanism.[11]
Cleavable Linkers for Porous Silicon-Based Mass
Spectrometry**
We have found that Diels–Alder adducts undergo [4+2]
cycloreversion readily upon desorption/ionization by the laser
pulse in DIOS analyses. Thus, the maleimide–furan adduct 1
bearing a 3-quinoline carboxamide moiety showed only a
single intense peak in the DIOSMS spectrum, which corre-
sponds to the protonated diene [3·H]+ (Scheme 1). The
Jun-Cai Meng, Claudia Averbuj, Warren G. Lewis,
Gary Siuzdak, and M. G. Finn*
Desorption/ionization on silicon mass spectrometry
(DIOSMS) uses porous silicon (pSi) to generate gas-phase
ions of small (< 3000 Da) molecules without a matrix by using
standard MALDI (matrix-assisted laser desorption/ioniza-
tion) instrumentation.[1] The unique laser desorption/ioniza-
tion surface properties of DIOSMS allow for the simulta-
neous detection of a broad range of small molecules as their
molecular ions, and the chemical properties of DIOSMS
facilitate the attachment of a variety of organic fragments
[2]
[3]
À
À
through surface Si H and Si OH groups. We aim to
combine these advantages with the proven ability of covalent
immobilization to facilitate combinatorial chemistry and the
extraction of molecular information from complex mixtures.[4]
Previous reports in this vein include the use of self-assembled
monolayers of functionalized alkanethiols to purify mixtures
(usually of proteins) by specific molecular interactions or
properties such as hydrophobicity or charge complementarity
prior to MALDI analysis.[5] DIOS has been similarly
employed to detect small molecules bound to pSi-immobi-
lized proteins.[6] We report herein that Diels–Alder adducts
undergo retro-[4+2] fragmentation in DIOSMS analysis, thus
providing a convenient way to covalently attach and detach
probe structures for chip-based mass analysis.
The most useful covalent linkers for our purposes should
be stable under organic synthetic conditions, yet cleavable
during desorption/ionization. While photocleavable groups
are commonly used in solid-phase organic synthesis,[7] most
(but not all[8]) of these systems require a workup step after
irradiation to complete the cleavage, and are therefore poorly
suited to detachment in the MALDI laser pulse. The retro-
Diels–Alder (rDA) reaction was one of the first dissociation
pathways to be investigated in mass spectrometry,[9] and has
since been characterized with a variety of ionization meth-
ods.[10] Some studies showed that the fragmentation is highly
Scheme 1. Retero Diels–Alder cleavage of a furan–maleimide adduct.
maleimide was not detected because it is poorly ionized, as
verified by control experiments. Hydrogenation of 1 into 2
eliminated fragmentation in the DIOS mass spectrum, thus
supporting the notion that cycloreversion is the cleavage
mechanism. A survey of a set of Diels–Alder adducts
suggested that the ease of thermal cycloreversion in solution
may be correlated with retro-Diels–Alder fragmentation in
DIOSMS (see Supporting Information). Furan–maleimide
adducts were found to be conveniently accessible, of sufficient
chemical stability,[12] and to undergo uniformly clean rDA
cleavage, and were therefore used in subsequent studies.
The use of a Diels–Alder moiety as a connector allows
small molecules to be both covalently attached to the porous
silicon surface and detected by mass spectrometry when
desired. This was illustrated by the derivatization of freshly-
etched pSi by hydrosilylative attachment of N-(4-vinylphe-
nyl)maleimide to give 4 (Scheme 2 and Supporting Informa-
tion). It should be noted that many of the hydrosilylation
reactions described here were performed at room temper-
ature in the absence of air, which is in contrast to the higher
temperatures (ca. 1008C) that appear to be standard in
previous cases.[1d,13] Under our mild conditions, it is assumed
that only the most reactive surface silicon-hydride sites are
derivatized giving rise to incomplete surface coverage, but a
sufficient density for DIOSMS detection was invariably
achieved. 1,3-Diphenylisobenzofuran was then introduced in
a CH2Cl2 solution, followed by extensive washing with organic
solvent. DIOSMS analysis showed a strong signal for the
expected rDA peak at m/z 270. Control samples, involving
noncovalently deposited diene and dienophile, as well as the
diene alone, showed no signal after they had been washed
(Scheme 2).
[*] Prof. G. Siuzdak
Department of Molecular Biology
The Scripps Research Institute
10550 N. Torrey Pines Rd., La Jolla, CA 92037 (USA)
Dr. J.-C. Meng, Dr. C. Averbuj, W. G. Lewis, Prof. M. G. Finn
Department of Chemistry
The Scripps Research Institute
10550 N. Torrey Pines Rd., La Jolla, CA 92037 (USA)
Fax: (+1)858-784-8850
E-mail: mgfinn@scripps.edu
[**] We thank the National Institutes of Health (RR-15066) and The
Skaggs Institute for Chemical Biology for support of this work; WGL
is a Skaggs Predoctoral Fellow. We are grateful to Dr. Zhouxin Shen
for helpful discussions.
As in all mass-spectrometry techniques, the strength of
the DIOS signal is affected by many factors, especially the
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2004, 43, 1255 –1255
DOI: 10.1002/anie.200352803
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1255

Communications
Furfurylamine addition to cyanuric chloride at
low temperature afforded 5, which was subsequently
converted into the fully substituted triazine–diene 6.
Compounds 7 and 8 were prepared analogously.
Diels–Alder attachment to N-phenethylmaleimide
(giving 9a) and to the maleimide-decorated porous
silicon surface 4 (giving 9b) occurred under mild
conditions. Both 9a, deposited on a DIOS chip, and
modified pSi 9b gave a single dominant MS peak at
m/z 387, indicative of clean cycloreversion to 6 upon
laser desorption/ionization. Furthermore, immersion
of a pSi-maleimide chip in a toluene solution of
equimolar amounts of 6, 7, and 8 (10 mmol each in
2 mL), or deposition of a drop of this solution on the
chip, showed all three species in approximately equal
intensity. When the immersed chip was subsequently
Scheme 2. Attachment and detachment of isobenzofuran by using a maleimide-derivatized
porous-silicon surface.
ionization efficiency of the analyte. The combination of a
retro-Diels–Alder linkage with an easily ionizable spacer
should provide a useful platform for the analysis of a number
of chip-based phenomena. The 1,3,5-triazine unit proved to
be an effective scaffold, since it provides a strong DIOSMS
signal regardless of the attached species and can be addressed
sequentially in three positions from trihalide (cyanuric)
derivatives.[14] The general design of such a system is shown
in Scheme 3, and Figure 1 depicts an example of the synthetic
manipulations performed.
rinsed thoroughly with toluene and ethanol, only 6 appeared,
presumably held by covalent attachment to the surface,
whereas the other species were washed away. A nonfunction-
alized pSi wafer retained none of these species after identical
analyte deposition and washing procedures.
The sequence shown in Figure 2 illustrates the application
of an alternative pSi-attachment strategy that incorporates a
preformed oxanorbornene structure. Hydrosilyation of 1,6-
heptadiyne with freshly prepared pSi afforded the terminal
alkyne surface 10. In spite of the presence of unconverted
À
surface Si H sites, at least some terminal alkyne residues
remained available, as established by IR spectroscopy and
subsequent reactivity; samples prepared by thermal or
photochemical[15] hydrosilylation methods behaved very sim-
ilarly. Attachment of 10 to azide 11 was then performed by
using the CuI-catalyzed procedure recently reported by Fokin
and Sharpless,[16] in a mixture of acetonitrile and pH 8 buffer
at room temperature. The capture of azide at 10 mm concen-
Scheme 3. Trifunctional 1,3,5-triazine core.
Figure 1. Detection of deposited and covalently attached compounds by DIOSMS. a) Furfurylamine (1 equiv), NaHCO3, 2:3 acetone/H2O, 93%.
b) Benzylamine (10 equiv), THF, reflux, 90%. c) Toluene (258C) or benzene (reflux), 75–80%. d) Toluene, 258C, 12 h.
1256
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 1255 –1255

Angewandte
Chemie
Figure 2. a) (I) 10 mm 11, 1.0 mm CuSO4, 2.0 mm l-ascorbic acid; 1:1 (v:v) MeCN:Tris buffer (pH 8.0), room temperature, 8 h. (II) Wash with THF
and ethanol.
tration by the surface-attached alkyne by using 1 mm copper
catalyst is among the most impressive examples of this “click
reaction”[17] so far reported,[18] thus illustrating the extraordi-
nary ability of the process to join appropriate pieces at low
concentration.[19] After the product had been washed, DIOS
analysis showed the expected rDA product 6 as the dominant
signal. The corresponding control compound 9a gave no MS
signal after incubation with 10 and washing of the product
under identical conditions.
The capture of a set of azides by an immobilized alkyne
was demonstrated as shown in Figure 3. In this case, the
alkyne was brought to the pSi surface as part of the triazenyl
À
Figure 3. Capture and detection of solution-phase azides by DIOSMS. a) Toluene, 258C, 12 h. b) Mixture of indicated R N3 (10 mm each), 3.0 mm
CuSO4, 6.0 mm l-ascorbic acid, 1:1 (v/v) MeCN:Tris buffer (pH 8.0), RT, 8 h. c) Wash with THF, then ethanol.
Angew. Chem. Int. Ed. 2004, 43, 1255 –1255
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1257

Communications
furan dienophile 12. The resulting wafer, 13,
was then exposed to an equimolar mixture of
three azides, chosen to differ in mass by
successive methylene units. Washing of the
product followed by DIOSMS analysis gave a
clean set of signals for the three [M+H]+
species, each showing the expected bromine
isotope pattern arising from the Br label
introduced to the triazine core. As before,
the negative controls (allyl in place of prop-
argyl in 12, nonderivatized pSi in place of 4)
gave no MS signals, thus demonstrating that
both the covalent connections (Diels–Alder
and [3+2] cycloaddition) were made.
To show that the covalent attachment and
detachment events enabled by the Diels–
Alder reaction can be useful in catalyst
screening, we explored the enantioselective
hydrolysis of acetate esters by two lipase
Scheme 4. Lipase-mediated hydrolysis of soluble and pSi-bound acetate esters.
enzymes.[20] Triazines 14–19 comprise three pairs of pseudoe-
nantiomers (Scheme 4). Each member of a pair has the
opposite absolute configuration of the chiral component and
was tagged with a bromide or chloride substituent to encode
this information into the MS signal.[21]
Pseudoracemic mixtures of triazine–acetates (equimolar
amounts of 14 + 15, 16 + 17, and 18 + 19) were subjected to
hydrolysis by lipases from Pseudomonas cepacia (PCL) and
Candida rugosa (CRL) in aqueous buffer and the results were
analyzed by reversed-phase HPLC (Figure 4). A range of
outcomes was observed. In two cases, the enzymes differed
dramatically in their reactivity: for the 14/15 pair, PCL gave
excellent resolution (krel > 50) and CRL was nonselective
(krel = 1), whereas for 18/19, CRL catalyzed a moderately
successful resolution (krel = 16) and PCL was inactive. In the
case of 16/17, the lipases were more similar in their activity,
with PCL being somewhat more selective (krel = 5.1 versus
1.1). The good performance of PCL with 14/15 and the
absolute configuration of the faster-reacting compound are
consistent with previous results involving structures that are
related to the substrates used here.[22] All the other cases are
sufficiently far removed from prior reports as to constitute
new results, although it was expected that the primary
acetates 16/17 would undergo less efficient kinetic resolution
than secondary acetates, which have the chiral center of
interest closer to the reactive bond.[23] Few lipase-catalyzed
transformations of alcohols and esters bearing nitrogen
substituents in the a position have been reported.[24]
Figure 4. HPLC (top; I=relative intensity arbitrary units) and DIOSMS (bottom; %I=percentage base peak intensity) analyses of hydrolysis reac-
tions using CRL and PCL on acetates 14–19.
1258
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 1255 –1255

Angewandte
Chemie
The pseudoenantiomeric mixtures of acetates were also
separately incubated with maleimide-functionalized porous
silicon as described above and the chips were then thoroughly
rinsed with toluene and ethanol. Alternatively, each pair of
acetates was attached to a different spot of the same pSi chip.
The resulting wafers were swirled together in an aqueous
buffer solution containing either CRL or PCL (12 h at room
temperature), washed with water and ethanol, and then
analyzed directly by DIOSMS. The resulting spectra, shown in
Figure 4, reproduced the solution-phase/HPLC findings quite
well. For example, with PCL, only the alcohol derived from 14
was observed, and no hydrolysis product from 15 was evident,
whereas both alcohols were detected in comparable amounts
with CRL. Similarly, in all other cases the relative reactivity
observed by HPLC were clearly reproduced in the chip-based
format. As a control, undecorated porous silicon was
incubated with the same mixtures of substrates 14–19 and
lipase in phosphate buffer under identical conditions. The
substrates and the hydrolyzed products were poorly detected
by DIOS, and washing of the products removed the signals
entirely.
Analytical methods based on mass spectrometry have the
general advantage over optical and radiolabel assays in that
the installation of chromophoric or radioactive tags is not
required and that detection of mass provides a general means
to monitor chemical changes in the analytes of interest. The
combination of the ability to detect analytes of different mass
in a single spectrum with the rapid data acquisition provided
by the chip-based format makes both DIOSMS and MALDI
well suited to high-throughput screening. Covalent attach-
ment of analytes further enhances these approaches by
enabling spatially addressable and multistep synthesis on
the chip. Most importantly, the making of a cleavable covalent
link between the analyte/probe and the surface allows the
user to wash the chips vigorously and thus overcome
problems of nonspecific adsorption and signal suppression.
Mrksich and Su have described the use of MALDI in this
fashion to analyze the conversion of surface-bound substrates
by a galactosyltransferase enzyme by using functionalized
alkanethiols, which relies on the ability of the MALDI laser
pulse to dissociate the Au–thiol bond.[5a]
[1] a) J. Wei, J. Buriak, G. Siuzdak, Nature 1999, 399, 243 – 246; b) Z.
Shen, J. J. Thomas, C. Averbuj, K. M. Broo, M. Engelhard, J. E.
Crowell, M. G. Finn, G. Siuzdak, Anal. Chem. 2001, 73, 612 –
619; c) J. J. Thomas, Z. Shen, J. E. Crowell, M. G. Finn, G.
Siuzdak, Proc. Natl. Acad. Sci. USA 2001, 98, 4932 – 4937; d) S.
Tuomikowki, K. Huikko, K. Grigoras, P. Östman, R. Kostiani-
nen, M. Baumann, J. Abian, T. Kotiaho, S. Franssila, Lab Chip
2002, 2, 247 – 253.
[2] a) M. P. Stewart, J. M. Buriak, Comments Inorg. Chem. 2002, 23,
179 – 203; b) J. M. Schmeltzer, L. A. Porter, Jr., M. P. Stewart,
C. M. Lopez, J. M. Buriak, Mater. Res. Soc. Symp. Proc. 2003,
737, 561 – 566; c) J. M. Buriak, M. J. Allen, J. Am. Chem. Soc.
1998, 120, 1339 – 1340; d) B. R. Hart, S. E. LØtant, S. R. Kane,
M. Z. Hadi, S. J. Shields, J. G. Reynolds, Chem. Commun. 2003,
322 – 323; e) F. Effenberger, G. Götz, B. Bidlingmaier, M.
Wezstein, Angew. Chem. 1998, 110, 2651 – 2654; Angew. Chem.
Int. Ed. 1998, 37, 2462 – 2464; f) A. R. Pike, L. H. Lie, R. A.
Eagling, L. C. Ryder, S. N. Patole, B. A. Connolly, B. R. Hor-
rocks, A. Houlton, Angew. Chem. 2002, 114, 637 – 639; Angew.
Chem. Int. Ed. 2002, 41, 615 – 617; g) A. Janshoff, K.-P. S. Dancil,
C. Steinem, D. P. Greiner, V. S.-Y. Lin, C. Gurtner, K. Mote-
sharei, M. J. Sailor, M. R. Ghadiri, J. Am. Chem. Soc. 1998, 120,
12108 – 12116; h) C. Gurtner, A. W. Wun, M. J. Sailor, Angew.
Chem. 1999, 111, 2132 – 2135; Angew. Chem. Int. Ed. 1999, 38,
1966 – 1968; i) N. Y. Kim, P. E. Laibinis, J. Am. Chem. Soc. 1997,
119, 2297 – 2298; j) N. Y. Kim, P. E. Laibinis, J. Am. Chem. Soc.
1998, 120, 4516 – 4517; k) L. H. Lie, S. N. Patole, E. R. Hart, A.
Houlton, B. R. Horrocks, J. Phys. Chem. B 2002, 106, 113 – 120.
[3] K.-P. S. Dancil, D. P. Greiner, M. J. Sailor, J. Am. Chem. Soc.
1999, 121, 7925 – 7930.
[4] a) M. C. Pirrung, Chem. Rev. 1997, 97, 473 – 488; b) R. W.
Nelson, D. Nedelkov, K. A. Tubbs, Electrophoresis 2000, 21,
1155 – 1163; c) G. MacBeath, S. L. Schreiber, Science 2000, 289,
1760 – 1763.
[5] a) J. Su, M. Mrksich, Angew. Chem. 2002, 114, 4909 – 4912;
Angew. Chem. Int. Ed. 2002, 41, 4715 – 4718; b) M. Merchant,
S. R. Weinberger, Electrophoresis 2000, 21, 1164 – 1167; c) R. W.
Nelson, J. R. Krone, O. Jansson, Anal. Chem. 1997, 69, 4363 –
4368; d) J. L. Bundy, C. Fenselau, Anal. Chem. 2001, 73, 751 –
757.
[6] H. Zou, Q. Zhang, Z. Guo, B. Guo, Q. Zhang, X. Chen, Angew.
Chem. 2002, 114, 668 – 670; Angew. Chem. Int. Ed. 2002, 41, 646 –
648.
[7] a) V. N. R. Pillai, Synthesis 1980, 1 – 26; b) V. N. R. Pillai in
Organic Photochemistry, Vol. 9 (Ed.: A. Padwa), Marcel
Dekker, New York, 1987, pp. 225 – 323; c) F. Guiller, D. Orain,
M. Bradley, Chem Rev. 2000, 100, 2091 – 2157; d) R. Glatthar, B.
Giese, Org. Lett. 2000, 2, 2315 – 2317.
[8] a) M. C. Fitzgerald, K. Harris, C. G. Shevlin, G. Siuzdak, Bioorg.
Med. Chem. Lett. 1996, 6, 979 – 982; b) J. M. Gerdes, H.
Waldmann, J. Comb. Chem. 2003, 5, 814 – 820.
[9] a) H. Budzikiewicz, J. I. Brauman, C. Djerassi, Tetrahedron 1965,
21, 1855 – 1879; b) H. Kwart, K. King, Chem. Rev. 1968, 68, 415 –
447.
Here we have shown Diels–Alder adducts to be cleavable
covalent linkers compatible with the DIOSMS technique,
useful in probing the reactivity of species attached to the
porous silicon surface. The triazine unit is a well-ionized
tripodal spacer, which allows for the attachment of tagging
residues or other components. Enzyme-catalyzed transforma-
tions on pSi-immobilized substrates were found to proceed
with similar relative activity and selectivity as those observed
in solution. Our methodology allows covalent attachment and
detachment to be implemented in the mass spectrometer
without added matrix material, and should be applicable to a
wide variety of analytes, chemical transformations, and
washing conditions.
[10] a) J. H. Bowie, A. H. Ho, J. Chem. Soc. Perkin Trans. 2 1975,
724 – 728; b) D. J. Burinsky, R. Dunphy, J. D. Alves-Santana,
M. L. Cotter, Org. Mass Spectrom. 1991, 26, 669 – 670; c) A.
Etinger, A. Mandelbaum, Org. Mass Spectrom. 1992, 27, 761 –
762; d) K. P. Madhusudanan, T. S. Dhami, A. Rani, S. N.
Suryawanshi, Rapid Commun. Mass Spectrom. 1993, 7, 92 – 94;
e) A. Lucas, J. Fernꢀndez-Gadea, N. Martin, R. Martꢁnez, C.
Seoane, Rapid Commun. Mass Spectrom. 2000, 14, 1783 – 1786;
f) N. Martin, R. Martꢁnez-Alvarez, C. Seoane, M. Suꢀrez, E.
Salfran, Y. Verdecia, N. K. Sayadi, Rapid Commun. Mass
Spectrom. 2001, 15, 20 – 24.
Received: September 5, 2003 [Z52803]
Keywords: cleavage reactions · combinatorial chemistry ·
.
lipases · mass spectrometry · porous silicon
Angew. Chem. Int. Ed. 2004, 43, 1255 –1255
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1259

Communications
[11] A. Mandelbaum in Applications of Mass Spectrometry to
Stereochemical Problems (Eds.: J. Splitter, F. Turecek), VCH
Publishers, New York, 1992, and references therein.
[12] D. Tobia, R. Harrison, B. Phillips, T. L. White, M. DiMare, B.
Rickborn, J. Org. Chem. 1993, 58, 6701 – 6706.
[13] J. T. C. Wojtyk, K. A. Morin, R. Boukherroub, D. D. M. Wayner,
Langmuir 2002, 18, 6081 – 6087.
[14] Our use of triazines was inspired by their elegant adaptation to
dendrimer synthesis described by Simanek and co-workers: W.
Zhang, E. E. Simanek, Org. Lett. 2000, 2, 843 – 845; W. Zhang,
D. T. Nowlan III, L. M. Thomson, W. M. Lackowski, E. E.
Simanek, J. Am. Chem. Soc. 2001, 123, 8914 – 8922.
[15] M. P. Stewart, J. Buriak, J. Am. Chem. Soc. 2001, 123, 7821 –
7830.
[16] V. V. Rostovtsev, L. G. Green, V. V. Fokin, K. B. Sharpless,
Angew. Chem. 2002, 114, 2708 – 2711; Angew. Chem. Int. Ed.
2002, 41, 2596 – 2599.
[17] H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew. Chem. 2001, 113,
2056 – 2075; Angew. Chem. Int. Ed. 2001, 40, 2004 – 2021.
[18] a) Q. Wang, T. R. Chan, R. Hilgraf, V. V. Fokin, K. B. Sharpless,
M. G. Finn, J. Am. Chem. Soc. 2003, 125, 3192 – 3193; b) A. E.
Speers, G. C. Adam, B. F. Cravatt, J. Am. Chem. Soc. 2003, 125,
4686 – 4687; c) A. J. Link, D. A. Tirrell, J. Am. Chem. Soc. 2003,
125, 11164 – 11165d) A. Deiters, T. A. Cropp, M. Mukherji, J. W.
Chin, J. C. Anderson, P. G. Schultz, J. Am. Chem. Soc. 2003, 125,
11782 – 11783.
[19] The azide–alkyne cycloaddition reactions described in this paper
were performed in the absence of the tris(triazolyl)amine ligand
previously shown to be helpful to the solution-phase process.[18a–c]
While the rate of the ligand-free process may be great enough to
effect the necessary degree of coupling to surface-tethered
alkynes, we suspect that the pSi surface can play an important
role in the reaction mechanism,[16] perhaps by providing active
I
À
hydride to preserve the Cu oxidation state or assist in Cu C
bond cleavage.
[20] S. Servi, Top. Curr. Chem. 1999, 200, 127 – 158.
[21] M. T. Reetz, M. H. Becker, H.-W. Klein, D. Stöckigt, Angew.
Chem. 1999, 111, 1872 – 1875; Angew. Chem. Int. Ed. 1999, 38,
1758 – 1761.
[22] R. J. Kazlauskas, A. N. E. Weissfloch, A. T. Rappaport, L. A.
Cuccia, J. Org. Chem. 1991, 56, 2656 – 2665.
[23] a) A. N. E. Weissfloch, R. J. Kazlauskas, J. Org. Chem. 1995, 60,
6959 – 6969; b) B.-V. Nguyen, O. Nordin, C. Voerde, E. Heden-
stroem, H.-E. Hoegberg, Tetrahedron: Asymmetry 1997, 8, 983 –
986.
[24] a) T. Izumi, K. Fukaya, Bull. Chem. Soc. Jpn. 1993, 66, 1216 –
1221; b) K. Kundell, L. T. Kanerva, Tetrahedron: Asymmetry
1995, 6, 2281 – 2286.
1260
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 1255 –1260