Which one is preferred: Myers-Saito cyclization of ene-yne-allene or Garratt-Braverman cyclization of conjugated bisallenic sulfone? A theoretical and experimental study

Article abstract of DOI:10.1021/ja9023644

A competitive scenario between Myers-Saito (MS) and Garratt-Braverman (GB) cyclization has been created in a molecule. High-level computations indicate a preference for GB over MS cyclization. The activation energies for the rate-determining steps of the

Full text of DOI:10.1021/ja9023644

Published on Web 10/13/2009
Which One Is Preferred: Myers-Saito Cyclization of
Ene-Yne-Allene or Garratt-Braverman Cyclization of
Conjugated Bisallenic Sulfone? A Theoretical and
Experimental Study
Amit Basak,*,† Sanket Das,† Dibyendu Mallick,‡ and Eluvathingal D. Jemmis*,‡,§
Department of Chemistry, Indian Institute of Technology, Kharagpur 721 302, India, Department
of Inorganic and Physical Chemistry, Indian Institute of Science, Bangalore 560 012, India, and
Indian Institute of Science Education and Research ThiruVananthapuram, CET Campus,
ThiruVananthapuram, 695 016 Kerala, India
Received April 3, 2009; E-mail: absk@chem.iitkgp.ernet.in
Abstract: A competitive scenario between Myers-Saito (MS) and Garratt-Braverman (GB) cyclization
has been created in a molecule. High-level computations indicate a preference for GB over MS cyclization.
The activation energies for the rate-determining steps of the GB and MS cyclizations were found to be the
same (24.4 kcal/mol) at the B3LYP/6-31G* level of theory; thus, from the kinetic point of view, both reactions
are feasible. However, the main biradical intermediate GB2 of the GB reaction is 6.2 kcal/mol lower in
energy than the biradical MS2, which is the main intermediate of MS reaction, so GB cyclization is
thermodynamically favored over MS cyclization. To verify the prediction by computational techniques,
bisenediynyl sulfones 1-4 and bisenediynyl sulfoxide 17 were synthesized. Under basic conditions, these
molecules isomerized to a system possessing both the ene-yne-allene and the bisallenic sulfone. The
isolation of only one product, identified as the corresponding naphthalene- or benzene-fused sulfone 8-11,
indicated the occurrence of GB cyclization as the sole reaction pathway. No product corresponding to the
MS cyclization pathway could be isolated. Though the theoretical prediction showed a preference for the
GB pathway over the MS pathway, the exclusive preference for GB over MS cyclization is very striking.
Further analysis showed that the intramolecular self-quenching nature of the GB pathway may play an
important role in the complete preference for this reaction. Apart from the mechanistic studies, these sulfones
showed DNA cleavage activity that had an inverse relation with the reactivity order. Our findings are important
for the design of artificial DNA-cleaving agents.
Ever since the discovery of enediyne antibiotics in the 1980s,1
the chemistry of spontaneous generation of diradicals has
attracted quite unprecedented attention2 because of their role
in organic synthesis,3 in the preparation of new materials,4 and
especially in pharmacology.1 Acting as natural antibiotics, these
molecules destroy the DNA of bacteria and viruses, and their
cytotoxic activity has been exploited in the development of
anticancer agents.5 The DNA-damaging activities of these
molecules depend upon their ability to abstract hydrogen atoms
from various locations of the deoxyribose moiety (Scheme 1).
However, not all diradicals are generated spontaneously under
ambient conditions. Also, not all of them have the same
efficiency of H abstraction. It is those diradicals that do not
have any mechanism of self-quenching that become diamagnetic
via abstraction of atoms from external sources. Bergman
cyclization (BC)6 and related reactions such as Myers-Saito
(MS) cyclization7 belong to this category. On the other hand,
reactions such as Garratt-Braverman (GB) cyclization8 involv-
ing conjugated bisallenes (Scheme 2) have a self-quenching
mechanism, and hence, the chances of interaction with external
† Indian Institute of Technology Kharagpur.
‡ Indian Institute of Science.
§ Indian Institute of Science Education and Research Thiruvanan-
thapuram.
(1) (a) Enediyne Antibiotics as Antitumor Agents; Border, D. B., Doyle,
T. W., Eds.; Marcel Dekker: New York, 1995. (b) Cancer Chemo-
therapeutic Agents; Foye, W. O., Ed.; American Chemical Society:
Washington, DC, 1995.
(2) (a) Basak, A.; Mandal, S.; Bag, S. S. Chem. ReV. 2003, 103, 4077.
(b) Krohn, K. Angew. Chem., Int. Ed. 2006, 45, 536. (c) Saito, I.;
Nakatani, K. Bull. Chem. Soc. Jpn. 1996, 69, 3007. (d) Wenk, H. H.;
Winkler, M.; Sander, W. Angew. Chem., Int. Ed. 2003, 42, 502. (e)
Maier, M. E.; Bosse, F.; Niestroj, A. J. Eur. J. Org. Chem. 1999, 1.
(f) Nicolaou, K. C.; Smith, A. L. Modern Acetylene Chemistry; Stang,
P. J., Diederich, F., Eds.; VCH: Weinheim, Germany, 1995; pp 203-
283; (g) Maier, M. E. Synlett 1995, 13. (h) DNA and RNA CleaVers
and Chemotherapy of Cancer and Viral Diseases; Meunier, B., Ed.;
Kluwer: Dordrecht, The Netherlands, 1996; p 372.
(5) Bross, P. F.; Beitz, J.; Chen, G.; Chen, X. H.; Duffy, E.; Kieffer, L.;
Roy, S.; Sridhara, R.; Rahman, A.; Williams, G.; Pazdur, R. Clin.
Cancer Res. 2001, 7, 1490.
(3) (a) Grissom, J. W.; Gunawardena, G. U.; Klinberg, D.; Huang, D.
Tetrahedron 1996, 52, 6453. (b) Schreiner, P. R.; Prall, M.; Lutz, V.
Angew. Chem., Int. Ed. 2003, 42, 5757.
(6) (a) Jones, R. G.; Bergman, R. G. J. Am. Chem. Soc. 1972, 94, 660.
(b) Bergman, R. G. Acc. Chem. Res. 1973, 6, 25. (c) Lockhart, T. P.;
Bergman, R. G. J. Am. Chem. Soc. 1981, 103, 4091. (d) Nicolaou,
K. C.; Zuccarello, G.; Oogawa, Y.; Schweiger, E. J.; Kumazawa, T.
J. Am. Chem. Soc. 1988, 110, 4866. (e) Gaffney, S. H.; Capitani, J. F.;
Castado, L.; Mitra, A. Int. J. Quantum Chem. 2003, 95, 706.
(4) (a) Perpall, M. W.; Perera, K. P. U.; DiMaio, J.; Ballato, J.; Foulger,
S. H.; Smith, D. W., Jr. Langmuir 2003, 19, 7153. (b) John, J. A.;
Tour, J. M. Tetrahedron 1997, 53, 15515.
9
10.1021/ja9023644 CCC: $40.75  2009 American Chemical Society
J. AM. CHEM. SOC. 2009, 131, 15695–15704 15695

A R T I C L E S
Basak et al.
Scheme 1. Mechanisms of Phosphodiester Bond Cleavage Induced by Radicals: Damage via 5′-H and 4′-H Abstraction Are Shown, and
Other Abstraction Processes Are Also Possible
Scheme 2. Generation of a Biradical through (a) Bergman Cyclization (BC), (b) Garratt-Braverman Cyclization (GBC), and (c) Myers-Saito
Cyclization (MSC)
sources are slim. BC is usually a slow process with a high
activation energy unless the enediyne is constrained in a ring
system of the proper size, as in the natural product calicheami-
cin.9 On the other hand, MS cyclization involving ene-yne-
allenes7 takes place spontaneously under ambient conditions,
even for acyclic systems. The latter chemistry is shown by the
neocarzinostatin (NCS) chromophore10 when it acts as an
antitumor agent. Nicolaou et al.11 first attempted to utilize GB
cyclization chemistry in bisallenic sulfones. However, the DNA
cleavage exhibited by these molecules mostly involved Michael
addition of a DNA base followed by Maxam-Gilbert-type12
cleavage rather than proceeding via H abstraction. Nonetheless,
this observation provided an alternative way of inducing DNA
cleavage. GB rearrangement generally takes place at higher
temperature unless the allenes are further conjugated, in which
case the reaction takes place under ambient conditions.13 The
biological relevance of all these reactions, along with their
challenging electronic structures, demanded the attention of
computational chemists to apply theoretical methodologies to
these systems and, if possible, take a leading role in predicting
the outcome in case of substrates endowed with various possible
pathways. Though many theoretical studies have been devoted
to BC processes,14 computational studies of MS cyclization are
(7) (a) Nagata, R.; Yamanaka, H.; Okazaki, E.; Saito, I. Tetrahedron Lett.
1989, 30, 4995. (b) Myers, A. G.; Kuo, E. Y.; Finney, N. S. J. Am.
Chem. Soc. 1989, 111, 8057. (c) Myers, A. G.; Dragovich, P. S. J. Am.
Chem. Soc. 1989, 111, 9130. (d) Nagata, R.; Yamanaka, H.; Murahashi,
E.; Saito, I. Tetrahedron Lett. 1990, 31, 2907.
(8) (a) Braverman, S.; Segev, D. J. Am. Chem. Soc. 1974, 96, 1245. (b)
Garratt, P. J.; Neoh, S. B. J. Org. Chem. 1979, 44, 2667. (c) Cheng,
Y. S. P.; Garratt, P. J.; Neoh, S. B.; Rumjanek, V. H. Isr. J. Chem
1985, 26, 101. (d) Braverman, S.; Duar, Y.; Segev, D. Tetrahedron
Lett. 1976, 17, 3181. (e) Zafrani, Y.; Gottlieb, H. E.; Sprecher, M.;
Braverman, S. J. Org. Chem. 2005, 70, 10166.
(12) (a) Maxam, A. M.; Gilbert, W. Methods Enzymol. 1980, 65, 499. (b)
Maxam, A. M.; Gilbert, W. Proc. Natl. Acad. Sci. U.S.A. 1977, 74,
560.
(9) (a) Lee, M. D.; Dunne, T. S.; Chang, C. C.; Ellestad, G. A.; Siegel,
M. M.; Morton, G. O.; McGahren, W. J.; Borders, D. B. J. Am. Chem.
Soc. 1987, 109, 3466. (b) Zein, N.; Sinha, A. M.; McGahren, W. J.;
Ellestad, G. A. Science 1988, 240, 1198. (c) Zein, N.; Poncin, M.;
Nilakantan, R.; Ellestad, G. A. Science 1989, 244, 697. (d) Ellestad,
G. A.; Hammann, P. R.; Zein, N.; Morton, G. O.; Siegel, M. M.; Patel,
M.; Borders, D. B.; McGahren, W. J. Tetrahedron Lett. 1989, 30, 3033.
(e) Magnus, P.; Carter, P. J. Am. Chem. Soc. 1989, 111, 7630. (f)
Mantlo, N. B.; Danishefsky, S. J. Org. Chem. 1989, 54, 2781. (g)
Chatterjee, M.; Cramer, K. D.; Townsend, C. A. J. Am. Chem. Soc.
1993, 115, 3374.
(13) Braverman, S.; Zafrani, Y.; Gottlieb, H. E. Tetrahedron Lett. 2000,
41, 2675.
(14) (a) Lindh, R.; Persson, B. J. J. Am. Chem. Soc. 1994, 116, 4963. (b)
Lindh, R.; Lee, T. J.; Bernhardsson, A.; Persson, B. J.; Karlstro¨m, G.
J. Am. Chem. Soc. 1995, 117, 7186. (c) Kraka, E.; Cremer, D.; Bucher,
G.; Wandel, H.; Sander, W. Chem. Phys. Lett. 1997, 268, 313. (d)
Kraka, E.; Cremer, D. J. Am. Chem. Soc. 1994, 116, 4929. (e)
Schreiner, P. R. J. Am. Chem. Soc. 1998, 120, 4184. (f) Nicolaides,
A.; Borden, W. T. J. Am. Chem. Soc. 1993, 115, 11951. (g) Wierschke,
S. G.; Nash, J. J.; Squires, R. R. J. Am. Chem. Soc. 1993, 115, 11958.
(h) Cramer, C. J.; Squires, R. R. J. Phys. Chem. A 1997, 101, 9191.
(i) Cramer, C. J. J. Am. Chem. Soc. 1998, 120, 6261. (j) Gra¨fenstein,
J.; Hjerpe, A. M.; Kraka, E.; Cremer, D. J. Phys. Chem. A 2000, 104,
1748.
(10) (a) Myers, A. G. Tetrahedron Lett. 1987, 28, 4493. (b) Myers, A. G.;
Proteau, P. J. J. Am. Chem. Soc. 1988, 110, 1146.
(11) Nicolaou, K. C.; Wendeborn, S.; Maligres, P.; Isshiki, K.; Zein, N.;
Ellestad, G. Angew. Chem., Int. Ed. Engl. 1991, 30, 418.
9
15696 J. AM. CHEM. SOC. VOL. 131, NO. 43, 2009

Myers-Saito or Garratt-Braverman Cyclization?
A R T I C L E S
Figure 2. Optimized structures of the two isomers MS1 and GB1.
Figure 1. Optimized structure of the model system, bisenediynyl sulfone 1.
a π/π benzylic biradical (GB2) as an intermediate. The energy
barrier for the formation of biradical GB2 from GB1 is 24.4
kcal/mol at the B3LYP/6-31G* level of theory (Figure 3), and
biradical GB2 is 7.2 kcal/mol more stable than the reactant GB1.
GB2 undergoes two successive rotations, first along one of the
C-C bonds (shown in red in Scheme 3) and then along another
C-C bond (shown in green in Scheme 3), to form GB4. The
energy barrier for the first rotation (along the C-C bond shown
in red) is 6.4 kcal/mol. Biradical GB3, which is a rotamer of
GB2, is almost equal in energy with GB2. The barrier for the
second rotation (from GB3 to GB4) is 9.0 kcal/mol. Cyclization
of GB4 leads to the formation of GB5. The structure GB4 is
13.0 kcal/mol higher in energy than GB5, and the activation
energy for this conversion (GB4 to GB5) is 8.4 kcal/mol.
Finally, GB5 isomerizes to the product GB6 via base-catalyzed
tautomeric proton shifts.8e,13 A similar kind of highly exothermic
H-atom shift accompanied by aromatization was extensively
studied by Lewis and co-workers.17 The structure GB6 is 63.2
kcal/mol more stable than GB5. The large thermodynamic
stability of the product GB6 is the main driving force for the
reaction.
The schematic representation and the energetics of the MS
pathway are shown in Scheme 4 and Figure 4, respectively. In
this case also, the first step is the formation of a biradical, MS2,
via a ring-closure mechanism. One of the key differences
between the GB and MS mechanisms is the nature of the
biradical species formed. In contrast to the GB reaction, the
MS reaction produces a σ/π biradical intermediate. Biradical
MS2 is slightly more stable (∆EMS2-MS1 ) -0.3 kcal/mol) than
the reactant MS1 at the B3LYP/6-31G* level of theory. The
energy barrier for the conversion of MS1 to MS2 is 24.4 kcal/
mol. Here, in order to compare the energetics of the different
intermediates, which differ in the number of hydrogens, we have
added suitable numbers of hydrogen molecules with the different
intermediates. Unlike the GB pathway, where the biradical is
self-quenched through intramolecular rearrangement processes,
biradical MS2 takes up two hydrogen atoms from the solvent
to form MS3. In order to calculate the H-abstraction barrier in
the MS pathway, we used 1,4-cyclohexadiene (1,4-CHD) as the
H donor. The H-abstraction barrier at the σ radical position is
4.2 kcal/mol, whereas at the π radical position it is 10.2 kcal/
mol (see the Supporting Information). A similar kind of ring
closure takes place in the other part of the structure MS3 to
produce σ/π biradical MS4 as an intermediate. The energy
barrier for this cyclization is 27.0 kcal/mol. Finally, biradical
MS4 takes up two more hydrogen atoms from the solvent to
produce the MS product MS5.
comparatively rare.15 No computational study of GB cyclization
has been reported to date. As both the MS and GB (for
conjugated allene) cyclizations take place under ambient condi-
tions, it is interesting as well as important to determine which
process is more favored in a system equipped with structural
features that permit both cyclization processes. Before embark-
ing upon any experimental endeavor toward resolving the issue,
we undertook a computational approach to see whether there is
a definite preference for one over the other in a bisenediynyl
sulfone system. This exercise pointed out a preference for GB
over MS cyclization that was nicely proved by subsequent
experiments.
Computational Study
We employed bisenediynyl sulfone 1 (Figure 1) as a model
system in carrying out the theoretical calculation. We calculated
the kinetic and thermodynamic parameters for both the MS and
GB cyclizations for this system. It was expected that 1 under
basic conditions would isomerize to form a system possessing
the two moieties required for the MS and GB cyclizations,
namely, an ene-yne-allene and a bisallenic sulfone, respec-
tively.16
We first carried out a thorough conformational search for the
ene-yne-allene system. Of the several conformational isomers, we
chose only those two isomers that occupy minima in the potential
energy surface and have the proper conformations for cyclization.
These two isomers, MS1 and GB1, are shown in Figure 2.
There is very little difference in energy (∆EGB1-MS1 ) 0.7
kcal/mol) between these two isomers at the B3LYP/6-31G* level
of theory. The conformation of isomer GB1 is such that it can
be considered as a bisallenic sulfone system, and hence, it is a
potential candidate to undergo the GB reaction under suitable
conditions. On the other hand, isomer MS1 has a suitable
conformation (ene-yne-allene) to undergo the MS reaction. In
the present study, first we will discuss the complete mechanism
of the GB reaction and then discuss the MS pathway.
The schematic representation of the GB pathway is shown
in Scheme 3. The first step of the reaction is the formation of
(15) (a) Schreiner, P. R.; Prall, M. J. Am. Chem. Soc. 1999, 121, 8615. (b)
de Visser, S. P.; Filatov, M.; Shaik, S. Phys. Chem. Chem. Phys. 2001,
3, 1242.
(16) Although in principle the ene-yne-allene can also undergo Schmittel
cyclization (SC) to form the fulvene diradical, we did not consider
such a process in our theoretical calculations. Our chosen system is
most likely to refrain from undergoing such a process, as Schmittel
and co-workers have shown that only ene-yne-allenes with an aryl,
silyl, or tert-butyl group attached to the alkyne terminus undergo SC.
See: (a) Schmittel, M.; Strittmatter, M.; Mahajan, A. A.; Vavilala,
C.; Cinar, M. E.; Maywald, M. ARKIVOC 2007, 2007 (viii), 66. (b)
Schmittel, M.; Strittmatter, M.; Stiffen, J.-P.; Maywald, M.; Engels,
B.; Helten, H.; Musch, P. J. Chem. Soc., Perkin Trans. 2 2001, 1331.
(c) Schmittel, M.; Steffen, J.-P.; Angel, M. A. W.; Engels, B.; Lennartz,
C.; Hanrath, M. Angew. Chem., Int. Ed. 1998, 37, 1562. (d) Bekele,
T.; Christian, C. F.; Lipton, M. A.; Singleton, D. A. J. Am. Chem.
Soc. 2005, 127, 9216.
From the above mechanistic studies for the GB and MS
processes, we found that the rate-determining step (rds) for the
GB pathway is the formation of the π/π biradical GB2 from
(17) Alabugin, I. V.; Manoharan, M.; Breiner, B.; Lewis, F. D. J. Am. Chem.
Soc. 2003, 125, 9329.
9
J. AM. CHEM. SOC. VOL. 131, NO. 43, 2009 15697

A R T I C L E S
Basak et al.
Scheme 3. Garratt-Braverman Pathway for GB1a
a The relative energies and the number of imaginary frequencies (in parentheses) at B3LYP/6-31G* level of theory (values at BLYP/6-31G* level of
theory are in italics) are given for different species involved in the reaction pathway.
pathway, since a π radical on the benzylic methylene group is
more stable than the σ radical of the MS intermediate MS2.
The higher stability of the π radical is partly due to conjugation
with the aromatic ring and partly due to the fact that the alkyl-H
bond is intrinsically weaker than the aryl-H bond.18 Thus, GB
cyclization is thermodynamically favored over MS cyclization.
If we compare the reverse of the reactions for the formation of
the biradicals for the GB and MS processes, we find that the
activation energy for the conversion of GB2 to GB1 is 31.6
kcal/mol whereas that for the conversion of MS2 to MS1 is
24.7 kcal/mol. Thus, biradical GB2 is also kinetically more
stable than biradical MS2. One may argue that the presence of
the electron-withdrawing sulfonyl group disfavors the MS
pathway. However, calculations done on a series of ene-yne-
allenes with different electron-withdrawing substituents at the
allene terminus showed that there is actually a decrease in the
activation energy barrier for the biradical formation step (see
the Supporting Information).
Figure 3. Reaction profile for the Garratt-Braverman mechanism.
Computational Details
GB1, whereas the rds for the MS pathway is the conversion of
All of the computations were performed with the Gaussian 03
software package.19 All optimizations of ground-state geometries
were done using the hybrid Hartree-Fock (HF)-density functional
MS1 to the σ/π biradical MS2. The activation energies for these
two reactions are same (24.4 kcal/mol). Thus, from the kinetic
point of view, both reactions are feasible. However, the main
intermediate of the GB reaction (i.e., the π/π biradical GB2) is
6.2 kcal/mol lower in energy than the σ/π biradical MS2 that
is the main intermediate of MS reaction. This biradical formation
step of the GB reaction is less endothermic than that of MS
(18) Koga, N.; Morokuma, K. J. Am. Chem. Soc. 1991, 113, 1907.
(19) Frisch, M. J.; et al. Gaussian 03, revision C.02; Gaussian, Inc.:
Wallingford, CT, 2004.
9
15698 J. AM. CHEM. SOC. VOL. 131, NO. 43, 2009

Myers-Saito or Garratt-Braverman Cyclization?
Scheme 4. Myers-Saito Pathway for MS1a
A R T I C L E S
a The relative energies and the number of imaginary frequencies (in parentheses) at B3LYP/6-31G* level of theory (values at BLYP/6-31G* level of
theory are in italics) are given for different species involved in the reaction pathway.
suggested by Lee et al. The 6-31G* basis set was used for all of
the calculations.19 A restricted approach was used in the compu-
tational analysis for the closed-shell structures, whereas an unre-
stricted broken-spin-symmetry approach (UBS-B3LYP) was used
for the open-shell singlet-state transition states and intermediates.
To get the broken-spin-symmetry solutions, we fed the SCF
computation with a 50:50 mix (singlet-triplet) initial guess of the
HOMO and LUMO orbitals. This approach worked acceptably well
for the biradical intermediates and the transition states of the GB
reaction, but for the MS biradicals, this approach was problematic,
as the HOMO and LUMO for these biradicals do not correspond
to the orbitals required for proper mixing. In these cases, we used
permuted orbitals DFT (PO-DFT), as suggested by Cremer and co-
workers,22 to calculate the open-shell singlet state. Schreiner et al.23
had suggested that gradient-corrected generalized gradient ap-
proximation (GGA) functionals24 such as BLYP are superior to
hybrid functionals such as B3LYP for calculations on biradicals.
We optimized all of the structures using the BLYP method with
Figure 4. Reaction profile for the Myers-Saito mechanism.
(21) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (b) Becke, A. D.
Phys. ReV. A 1988, 38, 3098. (c) Lee, C.; Yang, W.; Parr, R. G. Phys.
theory (DFT) method B3LYP,20,21 based on Becke’s three-
parameter functional including the HF exchange contribution with
a nonlocal correction for the exchange potential proposed by Becke
together with the nonlocal correction for the correlation energy
ReV. B 1988, 37, 785. (d) Vosko, S. H.; Wilk, L.; Nusair, M. Can J.
Phys. 1980, 58, 1200.
(22) Gra¨fenstein, J.; Kraka, E.; Filatov, M.; Cremer, D. Int. J. Mol. Sci.
2002, 3, 360.
(23) Schreiner, P. R.; Vazquez, A. N.; Prall, M. Acc. Chem. Res. 2005,
38, 29.
(20) Hehre, W.; Radom, L.; Schleyer, P. v. R.; Pople, J. A. Ab Initio
Molecular Orbital Theory; Wiley: New York, 1986.
(24) Koch, W.; Holthausen, M. C. A Chemist’s Guide to Density Functional
Theory; Wiley-VCH: Weinheim, Germany, 2000.
9
J. AM. CHEM. SOC. VOL. 131, NO. 43, 2009 15699

A R T I C L E S
Basak et al.
Figure 5. Target bispropargyl sulfones 1-4.
Scheme 5. Synthesis of Sulfonesa
a (i) Na2S, dry MeOH; (ii) PPTS (catalytic), EtOH; (iii) m-CPBA, dry CH2Cl2; (iv) DHP, PPTS (catalytic), dry CH2Cl2.
1
the same basis set, and the results were very much comparable
with the earlier method. The nature of the stationary points was
characterized by vibrational frequency calculations.
and H NMR spectra were recorded at different time points.
The concentration of the starting material decreased with time
while new peaks started to appear. The transformation was
complete within 2 h. The reaction was repeated in benzene and
triethylamine, and the same product, characterized as 8, was
isolated in 90% yield (Scheme 6).
Structure elucidation was done on the basis of NMR and mass
spectral data. For example, compound 8 showed the presence
of four acetylenic carbons and new aromatic protons indicating
the occurrence of cycloaromatization. The mass spectrum of 8
showed a molecular ion peak at m/z 403. Final confirmation of
the structure was obtained from single-crystal X-ray analysis25
(the ORTEP diagram is shown in Figure 6). Similar results were
obtained for the nitro-substituted and aliphatic enediynes, which
gave products 9 and 10, respectively, in quantitative yield
(Scheme 6). Of these propargyl sulfones, the reaction with the
nitro-substituted one was the fastest, with the conversion over
within 30 min.
Experimental Work
With the computational exercise favoring GB over MS
cyclization, we set out to prove the experimental validity of
the theoretical prediction. For that task, we designed and
subsequently synthesized bis(enediyne)sulfones 1-4 (Figure 5).
Sulfones 1-4 were synthesized in a straightforward manner,
as shown in Scheme 5. The key step was the formation of the
tetrahydropyran (THP)-protected thiobisenediynyl diol 6a by
the reaction of bromide 5a with Na2S in a 2:1 ratio in
methanol. Deprotection followed by oxidation of sulfide 7a
led to the formation of sulfone 1. The bis-THP-protected
sulfone 4 was prepared via protection of diol 1. Direct
oxidation of THP-protected sulfide 6a with m-chloroperoxy-
benzoic acid (m-CPBA) led to the formation of several products,
possibly due to partial deprotection. Nitro-substituted sulfone
2 and nonbenzenoid sulfone 3 were prepared in a similar way
from bromides 5b and 5c, respectively. All of the sulfones were
characterized by NMR and mass spectral data.
The mechanism of formation of the product is shown in
Scheme 7. This is according to what has been proposed by
(25) The crystal structure has been deposited at the Cambridge Crystal-
lographic Data Centre and has been allocated the deposition number
CCDC 722767.
In an initial NMR experiment, sulfone 1 (10 mg) dissolved
in CDCl3 was treated with a catalytic amount of triethylamine,
9
15700 J. AM. CHEM. SOC. VOL. 131, NO. 43, 2009

Myers-Saito or Garratt-Braverman Cyclization?
A R T I C L E S
Scheme 6. Reactivity of Various Sulfones with Et3Na
a (i) CDCl3, Et3N; (ii) m-CPBA, dry CH2Cl2.
pathway involves an intramolecular radical addition in which
the effective molarities of the reacting parts of the molecule
are very high, but the subsequent steps in MS pathway are
intermolecular hydrogen abstractions for which the actual
molarities of the H donors can never approach the effective
molarities of the components in the GB pathway. Thus, the
intramolecular self-quenching nature of the GB pathway may
also play an important role in the exclusive preference for this
reaction even in the presence of a large excess of external H
donor.
Having established that the GB cyclization is the sole path
followed by these bisenediynyl sulfones, the reason for such a
preference was then investigated from a chemical standpoint.
First, we established the fact that there was no role of the
hydroxyl groups in forcing the molecules to adopt a conforma-
tion suitable for GB cyclization by protecting the hydroxyl
groups as THP ethers. Thus, bis-THP-protected sulfone 4
(actually a mixture of diastereomers), when subjected to
triethylamine treatment, underwent GB cyclization to yield
Figure 6. X-ray structure of product 8.
1
product 11 (Scheme 6). The H NMR spectrum of the latter
Braverman.8e,13 The initial bisallenes 12 undergo GB cyclization
to yield the dihydronaphthalene derivatives 14, which isomerize
to the product 8 via a base-catalyzed tautomeric shift (pathway
a). The alternate MS cyclization (pathway b) was not followed
at all. It may be argued that the outcome of the reaction might
change in the presence of an external H donor. We repeated
the experiment in presence of up to 50 equiv of 1,4-CHD, but
the outcome remained the same. We could detect the formation
of only the GB product in the 1H NMR spectra (see the
Supporting Information). Though the computational study
indicated a preference for the GB pathway over the MS pathway,
the exclusive preference for the GB pathway, as found
experimentally, demands a more concrete explanation. It is true
that the higher stability of biradical GB2 over biradical MS2
definitely favors the GB pathway, but this may not be the sole
reason that can explain the complete preference for the GB
pathway. To analyze this issue, we once again carefully looked
at both mechanisms. After the formation of GB2, the GB
was very complicated because of the presence of various
diastereomers. Hence, the THP groups were removed to produce
a diol identical to 8 isolated earlier from reaction of 1. Since
bispropargyl sulfoxides are also reported to undergo GB
cyclization upon base treatment,26 we also prepared bisenediyne
sulfoxide 17 by carrying out the m-CPBA oxidation at 0 °C.
The sulfoxide in CDCl3, upon treatment with triethylamine
(catalytic amount), again followed the same GB pathway
(Scheme 6), although much more slowly (the time for comple-
tion was 24 h). In this case also, the structure was confirmed
by oxidizing the product, 18, to sulfone 1.
1
A low-temperature H NMR experiment was carried out to
identify the intermediate (Figure 7). There was practically no
reaction until the reaction temperature was raised to 15 °C. At
that point, new peaks at δ ∼4.4 and 4.3 started to appear. With
time, these peaks started to disappear with concomitant appear-
(26) Braverman, S.; Zafrani, Y.; Gottlieb, H. E. Tetrahedron 2001, 57, 9177.
9
J. AM. CHEM. SOC. VOL. 131, NO. 43, 2009 15701

A R T I C L E S
Basak et al.
Scheme 7. Mechanisms of Formation of Products
ance of the peaks corresponding to the product. This intermedi-
ate, we believe, is the cyclization product GB5 that is rapidly
converted to the final product via H shifts.
cleavage potential for the latter. In this case, the incubations
were carried out for 24 h because of slow rearrangement. Thus,
although the cleavage potentials are low, there is a fairly good
correlation between the ease of rearrangement and the cleavage
efficiency.27
In conclusion, computational techniques were used to deter-
mine the preference between MS and GB cyclizations The
theoretical prediction that GB cyclization is favored over MS
cyclization was nicely demonstrated by studying the reactivity
of various bisenediyne sulfones. Currently we are exploring
ways to change such a preference as well as the synthetic
potential of the GB pathway.
The end product of the GB biradical is a self-quenched
diamagnetic species; hence, the possibility that the biradical will
abstract an external hydrogen is slim. It is possible that biradicals
that have an appreciable half-life will have enough time to
abstract a hydrogen from an external source such as DNA and
hence result in its damage. It may be mentioned that Braverman
has reported26 DNA cleavage by bispropargyl sulfones under
basic conditions. We expected an inverse relation between the
kinetics of cyclization of the sulfones and the kinetics of
cleavage efficiency. Thus, sulfones 1-3 were incubated with
pBR322 DNA at pH 8.5. Aliquots were taken at 10 and 60 min
and subjected to agarose gel electrophoresis. The results are
shown in Figure 8. It is evident that the benzene-fused
bispropargyl sulfone has the highest cleavage efficiency, and
the order followed is 1 > 3 > 2 (compound 2 seemed to have
negligible activity). The faster collapse of the biradicals formed
via GB cyclization lowered the prospect of DNA cleavage by
H abstraction from the sugar moiety. The incubation results for
THP-protected sulfone 4 and sulfoxide 17 showed higher
Experimental Section
1
All of the H and 13C NMR spectra were recorded at 400 and
100 MHz, respectively, in CDCl3. The X-ray crystal data were
recorded on Bruker AXS Smart Apex-II. Electrospray ionization
(ESI) and high-resolution (HR) mass spectrometry (MS) were
(27) Sulfone 4 and sulfoxide 17 can inflict DNA cleavage even at a
concentration of 50 µM (the gel picture is included in the Supporting
Information).
9
15702 J. AM. CHEM. SOC. VOL. 131, NO. 43, 2009

Myers-Saito or Garratt-Braverman Cyclization?
A R T I C L E S
Figure 7. Low-temperature NMR spectra during the GB reaction of sulfone 1.
Figure 8. DNA cleavage studies with the sulfones. Lane 1: DNA (7 µL) in TAE buffer (pH 8.5) (5 µL) + DMSO (5 µL) at 23 °C. Lane 2: DNA (7 µL)
in TAE buffer (pH 8.5) (5 µL) + compound 1 in DMSO (1000 µM) at 23 °C, 10 min. Lane 3: DNA (7 µL) in TAE buffer (pH 8.5) (5 µL) + compound
3 in DMSO (1000 µM) at 23 °C, 10 min. Lane 4: DNA (7 µL) in TAE buffer (pH 8.5) (5 µL) + compound 2 in DMSO (1000 µM) at 23 °C, 10 min. Lane
5: DNA (7 µL) in TAE buffer (pH 8.5) (5 µL) + DMSO (5 µL) at 23 °C. Lane 6: DNA (7 µL) in TAE buffer (pH 8.5) (5 µL) + compound 4 in DMSO
(1000 µM, 24 h) at 23 °C. Lane 7: DNA (7 µL) in TAE buffer (pH 8.5) (5 µL) + compound 17 in DMSO (1000 µM, 24 h) at 23 °C.
performed on a Waters LCT mass spectrometer; the solutions of
the compounds were injected directly into the spectrometer via a
Rheodyne injector equipped with 10 µL loop. A Phoenix 20 micro-
LC syringe pump delivered the solution to the vaporization nozzle
(12H, m), 3.52-3.55 (2H, m), 3.69 (4H, s), 3.81-3.86 (2H, m),
4.46 (4H, ABq, J ) 16.0 Hz), 4.86 (2H, bs), 5.84 (4H, bs). 13C
NMR: δ 19.0, 20.0, 25.3, 30.2, 54.7, 61.9, 80.7, 83.0, 92.5, 93.0,
96.7, 119.5, 119.6. EI-MS: m/z 439.2 (MH+). HRMS: calcd for
C26H30O4S + H+, 439.1944; found, 439.1947.
of the electrospray ion source at a flow rate of 10 µL min-1
.
Nitrogen was used both as a drying gas and for nebulization with
flow rates of ∼3 L min-1 and 100 mL min-1, respectively. The
pressure in the analyzer region was usually ∼3 × 10-5 torr.
Bis(3-(2-(3-(tetrahydro-2H-pyran-2-yloxy)prop-1-ynyl)phe-
nyl)prop-2-ynyl)sulfane (6a). To a methanolic solution (5 mL) of
bromide 5a (50 mg, 0.15 mmol), Na2S (6 mg, 0.07 mmol) was
added and the mixture was stirred for 30 mins at room tempera-
ture.28 The mixture was evaporated to dryness and then worked
up in the usual procedure. The title compound 6a was isolated by
column chromatography (Si-gel, 7:1 PE/EA). State: yellow liquid.
3,3′-(2,2′-(3,3′-Thiobis(prop-1-yne-3,1-diyl))bis(4-nitro-2,1-
phenylene))diprop-2-yn-1-ol (6b). This was prepared as described
for 6a using 5b (50 mg, 0.17 mmol) and Na2S (7 mg, 0.08 mmol).
The crude material was purified by column chromatography (Si-
gel, 2:1 PE/EA). State: yellow solid. Yield: 57%. Mp: 163 °C. 1H
NMR: δ 3.86 (4H, s), 4.59 (4H, s), 7.55 (2H, d, J ) 8.8 Hz), 8.09
(2H, dd, J ) 2.8, 8.8 Hz), 8.26 (2H, d, J ) 2.0 Hz). 13C NMR: δ
20.4, 51.4, 80.5, 82.8, 91.2, 96.6, 122.8, 126.7, 131.4, 132.6, 133.0,
146.8. EI-MS: m/z 461 (MH+). HRMS: calcd for C24H16N2O6S +
H+, 461.0808; found, 461.0810.
3,3′-(2,2′-(3,3′-Thiobis(prop-1-yne-3,1-diyl))bis(2,1-phenylene))-
diprop-2-yn-1-ol (7a). State: yellow liquid. Yield: 87%. 1H NMR:
δ 3.85 (4H, s), 4.55 (4H, s), 7.25 (4H, bs), 7.42 (4H, bs). 13C NMR:
δ 20.2, 51.4, 82.2, 84.1, 88.6, 91.5, 125.2, 125.3, 128.1, 128.2,
131.9, 132.0. EI-MS: m/z 371.3 (MH+). HRMS: calcd for
C24H18O2S + H+, 371.1107; found, 371.1111.
1
Yield: 85%. H NMR: δ 1.62-1.84 (12H, m) 3.5-3.53 (2H, m),
3.89-3.87 (6H, m), 4.54 (4H, ABq, J ) 8.8 Hz), 5.29 (2H, bs),
7.23-7.26 (4H, m), 7.41-7.46 (4H, m). 13C NMR: δ 19.0, 20.1,
25.3, 30.2, 54.8, 61.9, 81.9, 84.4, 88.9, 89.1, 96.6, 125.4, 125.6,
127.8, 128.0, 131.9, 132.2. EI-MS: m/z 539.4 (MH+). HRMS: calcd
for C34H34O4S + H+, 539.2258; found, 539.2262.
2,2′-(4Z,4′Z)-8,8′-Sulfonylbis(octa-4-en-2,6-diyne-8,1-diyl)bis-
(oxy)bis(tetrahydro-2H-pyran) (6c). Compound 6c was prepared
from bromide 5c in essentially the same manner as described for
(4Z,4′Z)-8,8′-Thiodiocta-4-en-2,6-diyn-1-ol (7c). State: pale-
yellow liquid. Yield: 73%. 1H NMR: δ 3.73 (4H, s), 4.48 (4H, s),
5.87 (4H, s). 13C NMR: δ 20.0, 51.5, 81.0, 82.8, 92.4, 96.0, 119.6,
119.8. EI-MS: m/z 171.2 (MH+). HRMS: calcd for C16H14O2S +
H+, 271.0793; found, 271.0796.
1
6a. State: pale-yellow liquid. Yield: 83%. H NMR: δ 1.54-1.86
General Procedure for the Synthesis of Sulfones. To a CH2Cl2
(10 mL) solution of sulfide (0.1 mmol) at 0 °C, m-CPBA (2 equiv)
(28) Cao, X.; Yang, Y.; Wang, X. J. Chem. Soc., Perkin Trans. 1 2002,
2485.
9
J. AM. CHEM. SOC. VOL. 131, NO. 43, 2009 15703

A R T I C L E S
Basak et al.
was added and the solution was stirred for 2 h at room tempera-
ture.29 The reaction mixture was then diluted with CH2Cl2 (20 mL),
and the organic layer was washed successively with aqueous
saturated NaHCO3, saturated Na2SO3, and saturated Na2CO3 (30
mL each). The organic layer was evaporated to dryness, and the
residue upon chromatography (Si-gel, 2:1 PE/EA) furnished the
desired compound.
3,3′-(2,2′-(3,3′-Sulfonylbis(prop-1-yne-3,1-diyl))bis(2,1-phe-
nylene))diprop-2-yn-1-ol (1). State: yellow viscous liquid. Yield:
62%. IR (neat) νmax (cm-1): 9449, 2925, 2068, 1638, 1327, 1219,
1126, 1024, 769. 1H NMR: δ 4.50 (4H, s), 4.51 (4H, s), 7.25-7.32
(4H, m), 7.43-7.44 (4H, m). 13C NMR: δ 44.5, 51.3, 79.3, 83.3,
86.9, 92.3, 123.8, 125.9, 128.1, 129.0, 131.8, 131.9. ESI-MS: m/z
403 (MH+). HRMS: calcd for C24H18O4S + H+, 403.1005; found,
403.1009.
3,3′-(2,2′-(3,3′-Sulfonylbis(prop-1-yne-3,1-diyl))bis(4-nitro-2,1-
phenylene))diprop-2-yn-1-ol (2). State: yellow solid. Yield: 64%.
Mp: 212-216 °C. IR (KBr) νmax (cm-1): 3426, 2365, 2345, 1637,
1517, 1346, 1220, 1028, 900, 873, 836, 772, 745. 1H NMR: δ 4.51
(4H, s), 4.56 (4H, s), 7.57 (2H, d, J ) 8.8 Hz), 8.15 (2H, dd, J )
2.0, 8.8 Hz), 8.29 (2H, d, J ) 2.0 Hz). 13C NMR: δ 45.0, 51.3,
81.3, 82.0, 85.1, 97.9, 123.8, 125.0, 126.8, 132.1, 132.6. ESI-MS:
m/z 493.06 (MH+). HRMS: calcd for C24H16N2O8S + H+, 493.0706;
found, 493.0710.
was evaporated to dryness, and the residue upon chromatography
(Si-gel, 1:2 PE/EA) furnished the desired compound. State: yellow
solid. Mp: 260-262 °C (decomposition temperature). Yield: 87%.
IR (neat) νmax (cm-1): 3376, 2944, 2246, 1481, 1444, 1223, 1043,
1
1028, 952, 912, 759. H NMR: δ 4.22 (4H, ABq, J ) 16.4 Hz),
4.51 (4H, s), 7.23-7.3 (4H, m), 7.41-7.44 (4H, m). 13C NMR: δ
42.0, 52.2, 80.9, 83.6, 87.1, 92.6, 124.4, 125.8, 128.0, 128.7, 131.7,
131.8. ESI-MS: m/z 387.2 (MH+). HRMS: calcd for C24H18O3S +
H+, 387.1056; found, 387.1060.
General Procedure for Garratt-Braverman Reaction. Sul-
fone (10 mg) was taken in an NMR tube and dissolved in CDCl3
(600 µL). A catalytic amount of Et3N (20 mol %) was added, and
reaction was monitored by recording 1H NMR spectra at different
times. The reaction mixture was worked up using chloroform/water,
and the final product was isolated by column filtration (Si-gel, 1:1
PE/EA).
Naphthosulfolene 8. State: yellow solid. Yield: 90%. Mp:
236-240 °C (decomposition temperature). IR (KBr) νmax (cm-1):
1
3450, 2367, 2345, 2079, 1638, 1310, 1220, 1132, 1016, 770. H
NMR: δ 4.07, 4.32 (2 × 2H, ABq, J ) 16.4 Hz), 4.63, 4.67 (2 ×
2H, s), 7.26-7.27 (1H, m), 7.36-7.42 (2H, m), 7.47-7.48 (2H,
m), 7.63-7.66 (1H, m), 7.71 (1H, d, J ) 6.8 Hz), 8.38 (1H, s). 13
C
NMR: δ 51.0, 51.8, 55.7, 56.8, 82.6, 83.2, 92.2, 92.8, 120.4, 122.4,
123.8, 126.3, 126.8, 128.6, 128.7, 129.1, 129.2, 130.2, 131.5, 131.8,
132.9, 133.1, 136.6, 139.3. ESI-MS: m/z 403 (MH+), 425 (MNa+).
HRMS: calcd for C24H18O4S + H+, 403.1005; found, 403.1008.
Nitronaphthosulfolene 9. State: light-yellow solid. Yield: 92%.
Mp: 310-322 °C (decomposition temperature). IR (KBr) νmax
(cm-1): 3436, 2918, 2850, 2345, 2367, 2067, 1637, 1342, 1028.
1H NMR: δ 4.24, 4.30 (2 × 2H, ABq, J ) 16.0, 16.8 Hz), 4.68,
4.71 (2 × 2H, s), 7.68 (1H, d, J ) 7.6 Hz), 7.78 (2H, d, J ) 8.8
Hz), 7.9 (1H, d, J ) 2.0 Hz), 8.3 (1H, dd, J ) 2.0, 8.4 Hz), 8.61
(1H, s). 13C NMR: δ 51.0, 51.6, 55.6, 56.4, 81.5, 81.6, 96.9, 98.1,
123.7, 124.1, 124.2, 125.3, 125.6, 129.1, 130.3, 131.4, 131.5, 133.6,
134.0, 134.4, 139.1, 146.4, 147.7. ESI-MS: m/z 493 (MH+). HRMS:
calcd for C24H16N2O8S + Na+, 515.0525; found, 515.0533.
Benzosulfolene 10. State: white solid. Yield: 95%. Mp: 96-102
°C. IR (KBr) νmax (cm-1): 3408, 2923, 2851, 2370, 1607, 1308,
(4Z,4′Z)-8,8′-Sulfonyldiocta-4-en-2,6-diyn-1-ol (3). State: pale-
yellow liquid. Yield: 65%. IR (neat) νmax (cm-1): 3441, 2365, 2345,
1
2066, 1637, 1126, 1016, 771. H NMR: δ 4.37 (4H, d, J ) 1.2
Hz), 4.48 (4H, d, J ) 1.6 Hz), 5.92 (4H, ABq, J ) 10.8 Hz). 13C
NMR: δ 44.4, 51.3, 82.3, 82.8, 85.5, 96.5, 118.2, 122.2. ESI-MS:
m/z 303.1 (MH+). HRMS: calcd for C16H14O4S + H+, 303.0692;
found, 303.0696.
2,2′-(3,3′-(2,2′-(3,3′-Sulfonylbis(prop-1-yne-3,1-diyl))bis(2,1-
phenylene))bis(prop-2-yne-3,1-diyl))bis(oxy)bis(tetrahydro-2H-
pyran) (4). A catalytic amount of pyridinium p-toluenesulfonate
(PPTS) was added to a solution of sulfone 1 (22 mg, 0.05 mmol)
in dry CH2Cl2 (5 mL). Dihydropyran (DHP) (11 µL, 0.12 mmol)
was then added dropwise, and the mixture was stirred at room
temperature for 12 h.30 Evaporation of the solvent in vacuum gave
an oil from which title compound 4 was isolated by column
chromatography (Si-gel, 4:1 PE/EA). State: yellow viscous liquid.
Yield: 82%. IR (neat) νmax (cm-1): 3450, 2946, 2369, 1653, 1637,
1
1218, 1133, 1013, 772. H NMR: δ 4.29, 4.39 (2 × 2H, s), 4.51,
4.55 (2 × 2H, s), 5.94 (1H, d, J ) 11.6 Hz), 6.93 (1H, d, J ) 11.6
Hz), 7.23 (1H, d, J ) 8.0 Hz), 7.48 (1H, d, J ) 8.0 Hz). 13C NMR:
δ 51.3, 51.6, 56.1, 56.8, 82.1, 83.1, 93.6, 96.2, 113.4, 122.7, 125.2,
130.6, 131.8, 132.6, 136.6, 137.0. ESI-MS: m/z 303 (MH+). HRMS:
calcd for C16H14O4S + Na+, 325.0511; found, 325.0483.
1
1338, 1128, 1077, 1024, 964, 942, 901, 813, 768. H NMR: δ
1.53-1.83 (12H, m), 3.50-3.53 (2H, m), 3.8-3.86 (2H, m),
4.48-4.58 (8H, m), 4.87 (2H, t, J ) 3.2 Hz), 7.25-7.32 (4H, m),
7.45 (4H, dd, J ) 1.6, 7.2 Hz). 13C NMR: δ 19.0, 25.3, 30.2, 44.1,
54.7, 61.9, 80.1, 83.9, 86.2, 89.7, 96.8, 124.1, 125.8, 128.1, 128.8,
132.1, 132.4. ESI-MS: m/z 570.3 (MH+). HRMS: calcd for
C34H34O6S + H+, 571.2156; found, 571.2161.
Acknowledgment. D.M. and E.D.J. thank the SERC, IISc, for
computational resources. DST is acknowledged for the J. C. Bose
Fellowship Grant to E.D.J. and for a SERC Grant to A.B. that
supported this research. S.D. is grateful to the CSIR, Government
of India, for a senior research fellowship. The NMR and X-ray
facilities were provided at IIT Kharagpur by DST under the IRPHA
and FIST Programmes, respectively.
3,3′-(2,2′-(3,3′-Sulfonylbis(prop-1-yne-3,1-diyl))bis(2,1-phe-
nylene))diprop-2-yn-1-ol (17). To a solution of sulfide 1 (40 mg,
0.1 mmol) in CH2Cl2 (10 mL) at 0 °C was added m-CPBA (20
mg, 0.12 mmol), and the solution was stirred for 1 h at 0 °C.28
The reaction mixture was then diluted with CH2Cl2 (20 mL), and
the organic layer was washed successively with aqueous saturated
NaHCO3, Na2SO3, and Na2CO3 (30 mL each). The organic layer
Supporting Information Available: Total energies, optimized
1
1
Cartesian coordinates, various H, 13C, and H NMR kinetics
spectra, a DNA-gel picture, complete ref 19, and crystallographic
data for 8 (CIF). This material is available free of charge via
the Internet at http://pubs.acs.org.
(29) (a) Nicolaou, K. C.; Zuccarello, G.; Riemer, C.; Estevez, V. A.; Dai,
W. M. J. Am. Chem. Soc. 1992, 114, 7360. (b) Kar, M.; Basak, A.
Chem. Commun. 2006, 3818.
(30) Miyashita, M.; Yoshikoshi, A.; Grieco, A. P. J. Org. Chem. 1977,
42, 3772.
JA9023644
9
15704 J. AM. CHEM. SOC. VOL. 131, NO. 43, 2009