Absolute stereochemical assignment and fluorescence tuning of the small molecule tool, (-)-blebbistatin

Article abstract of DOI:10.1002/ejoc.200500103

(-)-Blebbistatin (1), a recently discovered small molecule inhibitor of the ATPase activity of non-muscle myosin II has been prepared from methyl 5-methylanthranilate (6) in three steps. This flexible synthetic route has also been used to prepare a nitro

Full text of DOI:10.1002/ejoc.200500103

SHORT COMMUNICATION
Absolute Stereochemical Assignment and Fluorescence Tuning of the Small
Molecule Tool, (–)-Blebbistatin
Cristina Lucas-Lopez,[a][‡] Stephen Patterson,[a][‡] Till Blum,[a] Aaron F. Straight,[b,c]
Judit Toth,[d] Alexandra M. Z. Slawin,[a] Timothy J. Mitchison,[b] James R. Sellers,[d] and
Nicholas J. Westwood*[a]
Keywords: Inhibitors / Asymmetric synthesis / Fluorescence
(–)-Blebbistatin (1), a recently discovered small molecule in-
hibitor of the ATPase activity of non-muscle myosin II has
been prepared from methyl 5-methylanthranilate (6) in three
steps. This flexible synthetic route has also been used to pre-
pare a nitro group-containing analogue 12 that has modified
fluorescence properties and improved stability under micro-
scope illumination. The key step in the synthesis of 1 and its
analogues was the asymmetric hydroxylation of the quino-
lone intermediate 3 using the Davis oxaziridine methodology.
The absolute stereochemistry of (–)-blebbistatin (1) was
shown to be S by X-ray crystal structure analysis of a heavy
atom (bromine) containing analogue 11, which was sub-
sequently reduced and shown to be identical to 1.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2005)
Recent developments at the chemistry–biology interface of its reduced activity against nmMII, it can function as a
have revitalised the search for specific small molecule mod- control compound in these types of studies.[4] To date, no
ulators of protein function.[1] These molecules, once discov- experimental evidence that supports the assignment of the
ered and with their activity optimized, play an important absolute stereochemistry of 1 has been reported despite its
role in improving our understanding of complex biological prominent role as a novel molecular tool. Unambiguous as-
processes.[2] For example, a programme to identify selective signment of the absolute stereochemistry of (–)-blebbistatin
inhibitors of the myosin protein subfamilies[3] has led to the (1) is also essential for the refinement of computational
discovery of several novel biological reagents including (–)-
blebbistatin (1) (Figure 1).[4,5] In the two years since its dis-
covery 1 has had a significant impact on biomedical re-
search, with reports of its use in dissecting the mechanism
of cancerous cell migration,[6] severing-induced axon retrac-
tion[7] and cell cycle control.[8] In a recent application,[4]
1
was used to study the role of non-muscle Myosin II
(nmMII) in cytokinesis, the final and critical stage in the
complex process that leads to the formation of two daugh-
ter cells from one mother cell. (+)-Blebbistatin (2) (the en-
antiomer of 1) is also an important compound. As a result
[a] School of Chemistry and Centre for Biomolecular Sciences,
University of St. Andrews,
North Haugh, St. Andrews, Fife, KY16 9ST, UK
Fax: +44-1334-462595
E-mail: njw3@st-andrews.ac.uk
[b] Institute of Chemistry and Cell Biology and the Department of
Systems Biology, Harvard Medical School,
240 Longwood Avenue, Boston, MA 02115, USA
[c] Current address: Department of Biochemistry, Stanford Uni-
versity, School of Medicine,
297 Campus Dr., B400, Stanford, CA 94305, USA
[d] Laboratory of Molecular Cardiology, National Heart, Lung
and Blood Institute, National Institutes of Health,
Bethesda, Maryland 20892-1762, USA
[‡] These authors contributed equally to this work.
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Figure 1. A flexible approach to (–)-blebbistatin (1) and its ana-
logues.
1736
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejoc.200500103
Eur. J. Org. Chem. 2005, 1736–1740

Absolute Stereochemical Assignment and Fluorescence Tuning of (–)-Blebbistatin
SHORT COMMUNICATION
models of the interaction of 1 and 2 with myosins.[9] These to be stable in the absence of air and light, and was readily
studies are designed to identify putative binding sites. prepared and stored for extended periods of time in multi-
Here we clarify several chemical issues relating to (–)- gramme quantities. However, it was observed that a DMSO
blebbistatin (1) and its analogues. We describe an efficient solution of 3 slowly decomposed in air in the presence of
and flexible synthetic approach to highly optically enriched light to give ( )-blebbistatin. The rate and yield of this pro-
(–)-blebbistatin (1) and prove, for the first time, that the cess was increased by irradiation of 3 in DMSO or THF
absolute stereochemistry of (–)-blebbistatin (1) is S (Fig- using a medium-pressure mercury lamp (400 W, unfiltered)
ure 1). In addition, we report that incorporation of a nitro or upon irradiation (365 nm) of 3 supported on silica gel.
functional group into the blebbistatin core structure im- Interestingly, ( )-blebbistatin was initially discovered fol-
proves the physical properties of blebbistatin, extending the lowing biological assays carried out with an “aged” DMSO
utility of these reagents. The nitro analogue of 1 is not fluo- solution of 8 (Figure 2). The azatacrine[14] analogue 8 is not
rescent (Figure 5) and is also stable for extended periods of an inhibitor of nmMII and the observed activity of the
time at wavelengths typically used for the fluorescence- “aged” sample results from degradation of 8 to ( )-blebbis-
based imaging of live cells, unlike (–)-blebbistatin 1 (Fig- tatin.
ure 6).[10]
As a result of the above observations care was taken to
Our approach to (–)-blebbistatin (1) relies on late stage exclude any oxygen from the subsequent oxidation reac-
asymmetric oxidation of the quinolone 3 to the previously tions. In initial experiments, treatment of the anion gener-
unknown heterocyclic core structure in 1 (Figure 1). It was ated from 3 using LiHMDS or LDA with oxaziridine 9 or
envisaged that conversion of 3 to 1 could be carried out 10 at –10 °C or 0 °C, respectively, gave high yields of op-
using Davis oxaziridine methodology.[11] 3 can be accessed tically enriched (–)-blebbistatin (1) (Figure 3, Entries 1–
rapidly from commercially available starting materials via 3).[11] However, little or no reaction was observed when
the corresponding amidine 4.
these transformations were carried out at temperatures of
Reaction of the pyrrolidinone 5 with the anthranilate 6 –40 °C or below. Generation of the sodium enolate of 3
in the presence of phosphorus oxychloride gave the desired using NaHMDS resulted in a more reactive enolate that
amidine 4 in 41% yield (Figure 2). 1H NMR analysis of the reacted with 10 at –78 °C to give highly optically enriched
reaction between 5 and POCl3 did not support the efficient (–)-blebbistatin (1) in 69% yield and 90% ee as judged by
formation of an imonium intermediate.[12] Reaction of 5 chiral HPLC analysis of the crude reaction mixture (Fig-
and 6 under microwave irradiation in the absence of POCl3 ure 3, Entry 4). A single recrystallisation from acetonitrile
did not produce the desired amidine 4 even in the presence of the crude reaction mixture prepared according to Entry
of a dehydrating agent. Attempts to optimize this reaction 3 (Figure 3) provided a route to 1 in Ͼ99% ee ([α]2D6 = –464
by addition of base[12] or by the use of alternative reagents (c = 0.2 in dichloromethane).[15,16] (S)-(–)-Blebbistatin (1)
(SOCl2, MeOTf, PCl5) proved unsuccessful. The pyrrolidin- inhibits nmMIIa with an IC50 of 7.1 μM (see supporting
one dimer 7 was also isolated on a number of occasions.[13] information; for supporting information see also the foot-
Cyclisation of 4 to the desired quinolone 3 was carried out note on the first page of this article), whilst 2 is significantly
using excess lithium bis(trimethylsilyl)amide (LiHMDS) in less active than 1. This result clarifies the conflicting litera-
high yield. The quinolone 3, an off-white solid, was found ture reports.[4,9]
Figure 2. Preparation of ( )-blebbistatin. Reagents and conditions: a) i) POCl3, CH2Cl2, 25 °C, 3 h; ii) 40 °C, 16 h, 41%; b) LiHMDS
(3 equiv.), –78 °C to 0 °C, 3 h, 90%; c) O2, hν, DMSO, 25 °C, 3 h, 29%; or O2, hν, 25 °C, 3 h, THF, 26%. LiHMDS = lithium bis(trimethyl-
silyl)amide.
Eur. J. Org. Chem. 2005, 1736–1740
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1737

N. J. Westwood et al.
SHORT COMMUNICATION
chlorides gave the desired products, but did not yield suffic-
iently high quality crystals for X-ray analysis. Attempts to
prepare an analogue of 1 with a bromine atom in place of
the methyl substituent also proved unsuccessful due to the
low yield of the reaction between methyl 5-bromoanthra-
nilate and pyrrolidinone 5. X-ray quality crystals of the bro-
mine-containing analogue 11 were successfully prepared
using similar methods to those described previously (see
Figure 4 legend for further details). A highly enantiomer-
ically enriched sample of 11 (Ͼ99% ee as judged by chiral
HPLC analysis) was then reduced using hydrogen and 1%
palladium on carbon in the presence of triethylamine.[17]
The resulting sample of (–)-blebbistatin (1) was shown by
1
chiral HPLC and H NMR analysis to be identical to a
sample of 1 prepared as described in Figure 3, hence con-
firming that the absolute stereochemistry of (–)-blebbistatin
(1) is S (as drawn).[15,16] Formation of 1 presumably pro-
ceeds through an analogous transition state to the one pro-
posed by Davis.[11,16] Blebbistatin analogue 11 can also
function as a useful precursor for the synthesis of a radio-
labelled derivative of 1 (by palladium-catalysed tritiation)
and for the preparation of (S)-(–)-blebbistatin (1) analogues
functionalised at C4Ј. Preliminary data on the activity of
compounds substituted at C4Ј shows that they retain
nmMII ATPase inhibitory activity.[4]
An informative method of studying protein function in a
cell uses fluorescence microscopy techniques on live cells
that are expressing a green fluorescent protein (GFP)-lab-
eled version of the protein of interest.[18] Unfortunately, the
fluorescence of 1 is one factor that limits its application in
this type of experiment.[19] In order to observe GFP-labeled
proteins it is necessary to irradiate live cells with light at a
wavelength between 420 and 490 nm (488 nm in confocal
microscopy applications). The light that is subsequently
emitted is collected using pass filters with a typical wave-
length range of between 520 and 570 nm. As shown in Fig-
Figure 3. Preparation of enantiomerically enriched blebbistatin
using the Davis oxaziridine methodology. Reagents and conditions:
a) LiHMDS or LDA or NaHMDS (1.2 equiv.), THF, oxaziridine
9 or 10 (2.4 equiv.). The Table summarizes the yield and enantio-
meric excess (ee) obtained as a function of base, temperature and
reagent. All ee values were determined using chiral HPLC analysis
of the crude reaction mixture (see supporting information). LDA =
Lithium diisopropylamide, NaHMDS = sodium bis(trimethylsilyl)-
amide.
In order to assign by crystallographic analysis the abso-
lute stereochemistry of (–)-blebbistatin (1), we decided to
prepare an analogue that contained a heavy atom (bro-
mine). Purification of the products resulting from direct
bromination of 1 (or intermediates en route to 1) with NBS
proved difficult. Additionally, reactions to acylate the terti-
ary alcohol functionality in 1 with various bromobenzoyl
Figure 4. Assignment of the absolute stereochemistry of (–)-blebbistatin (1). Reagents and conditions: a) NBS, DMF, 25 °C, 2 days, 50%;
b) i) POCl3, CH2Cl2, 25 °C, 3 h; ii) 6, 40 °C, 16 h, 26%; c) LiHMDS (3 equiv.), –78 °C to 0 °C, 3 h, 60%; d) i) LiHMDS (1.2 equiv.),
THF, –78 °C, 30 min; ii) 10 (2.4 equiv.), –10 °C, 16 h, 68%, 88% ee; e) recrystallisation from acetonitrile Ͼ99% ee. [α]2D6 = –526 (c = 0.1,
CH2Cl2); f) H2, 1% Pd/C, Et3N, DMF/MeOH, 25 °C, 24 h, 99%. NBS = N-bromosuccinimide, DMF = N,N-dimethylformamide.
1738
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 1736–1740

Absolute Stereochemical Assignment and Fluorescence Tuning of (–)-Blebbistatin
SHORT COMMUNICATION
ure 5 (solid line), 1 exhibits significant fluorescence emis-
sion in the GFP emission wavelength range with an emis-
sion maximum at 601 nm. It was proposed that addition of
a nitro functional group to the chromophore within (S)-
(–)-blebbistatin (1) would modify the fluorescence proper-
ties without having a major effect on the biological activity,
hence overcoming this limitation.[20] Optically enriched 12
was prepared (see Figure 5 legend for details[16]) and subse-
quent analysis of the fluorescence properties of 12 showed
that there was a significant reduction in fluorescence emis-
sion in the required wavelength range following excitation
at 440 nm. Biochemical and cell-based assays using 12
showed that it retained the desired biological activity (see
supporting information).
Figure 6. A) UV spectra of (S)-(–)-blebbistatin (1) in PBS (solid
line). The UV spectra of 1 after exposure to filtered light (436 and
510 nm) for 3 hours (dashed line), showing that complete degrada-
tion of 1 occurs under these conditions consistent with literature
reports.[10] B) UV spectra of 12 in PBS (solid line) and 12 after
exposure to filtered light (436 and 510 nm) for 3 hours (dashed
line) indicating that 12 is stable under these conditions. PBS =
phosphate-buffered saline.
Figure 5. Preparation of a blebbistatin analogue containing a nitro
functionality resulting in improved optical properties. Reagents and
conditions: a) MeOH, H2SO4, 65 °C, 96 h, 81%; b) i) POCl3, 5,
CH2Cl2, 25 °C, 3 h; ii) 40 °C, 72 h, 22%; c) LiHMDS (2.5 equiv.),
–78 °C to 0 °C, 96 h, 44%; d) i) LiHMDS (1.2 equiv.), THF, –78 °C,
30 min; ii) 10 (3.1 equiv.), –10 °C, 32 h, 31%, 76% ee. Fluorescence
emission spectrum of (–)-blebbistatin (1) (solid line) and its nitro
analogue 12 (dashed line) in the wavelength range 450–800 nm after
excitation at 440 nm.
In summary, we report that the absolute stereochemistry
of the selective small molecule myosin inhibitor (–)-blebbis-
tatin (1) is S. In addition, we describe a flexible and efficient
route to highly optically enriched 1 (Ͼ99% ee) using the
Davis oxaziridine methodology. (S)-(–)-Blebbistatin (1) has
already been used as a small molecule tool in a number of
biological studies.[6–8,21] The synthetic approach we report
enables rapid access to both enantiomers of blebbistatin (1
and 2) and its analogues in large quantities. We have also
demonstrated that incorporation of a nitro functionality
into the chromophore of 1 tunes its fluorescent properties
and improves its stability, extending the scope of biological
experiments in which analogues of (S)-(–)-blebbistatin (1)
can be used.
A recent microscopy-based study has identified a further
limitation of (S)-(–)-blebbistatin (1) that limits its use in
live-cell imaging experiments. It was shown that prolonged
exposure to filtered light (450–490 nm) results in degrada-
tion of ( )-blebbistatin to an unidentified non-inhibitory
product via cytotoxic intermediates.[10] In analogous experi-
ments performed by us the blebbistatin analogue 12 was
shown to be stable to prolonged irradiation at the same
wavelengths (see B in Figure 6), whereas 1 decomposed (see
A in Figure 6). This coupled with its reduced fluorescence
and retained biological activity suggests that analogue 12
will be an excellent reagent for imaging experiments with
live cells, a situation where (S)-(–)-blebbistatin (1) itself is
sub-optimal.
Acknowledgments
The authors would like to thank the following for funding: EPSRC
(N. J. W., C. L.-L.), The Royal Society (N. J. W. Fellowship), The
Eur. J. Org. Chem. 2005, 1736–1740
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1739

N. J. Westwood et al.
SHORT COMMUNICATION
[12] M. E. Kuehne, P. J. Shannon, J. Org. Chem. 1977, 42, 2082–
2087.
Wellcome Trust (S. P.), Deutscher Akademischer Austauschdienst
(T. B. Fellowship), NIH GM062566 (T. J. M.). For analytical and
microscopy support: Melanja Smith (1H NMR), Carolyn Hores-
burgh (mass spectrometry), Dr. Graham Turnbull, Prof. Ifor
Samuel and Mark Goossens (fluorescence measurements), Dr.
Tony Vaughan (microscopy). For helpful discussions during prepa-
ration of the manuscript: Dr. Stuart Conway, Dr. David Smith and
Prof. Gary Ward.
[13] K. H. Büchel, A. K. Bocz, F. Korte, Chem. Ber. 1966, 99, 724–
735.
[14] H. Fujiwara, S. Nishino, I. Nakamura, H. Tokunaga (Ube In-
dustries Ltd.), EP 0 430 485 A2, 1990. [Chem. Abstr. 1991, 115,
136078 g].
[15] X-ray crystallographic analysis of single crystals of 1, whilst
supporting the atom connectivity, were unable to provide
sufficiently high quality data to allow the absolute stereochemi-
cal assignment of 1. CCDC-238391, -243020, and -238392
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.
[16] For a detailed report of all experimental and analytical data
see supporting information.
[17] G-D. Zhu, V. Schaefer, S. A. Boyd, G. F. Okasinski, J. Org.
Chem. 2002, 67, 943–948.
[1] N. J. Westwood, Philos. Transact. R. Soc. London, Ser. A 2004,
362, 2761–2774, and references cited therein.
[2] T. U. Mayer, T. M. Kapoor, S. J. Haggarty, R. W. King, S. L.
Schreiber, T. J. Mitchison, Science 1999, 286, 971–974.
[3] J. R. Sellers, Myosins, Oxford University Press, Oxford, 1999.
[4] A. F. Straight, A. Cheung, J. Limouze, I. Chen, N. J. West-
wood, J. R. Sellers, T. J. Mitchison, Science 2003, 299, 1743–
1747.
[5] A. Cheung, J. A. Dantzig, S. Hollingworth, S. M. Baylor, Y. E. [18] D. Ehrhardt, Curr. Opin. Plant Biol. 2003, 6, 622–628.
Goldman, T. J. Mitchison, A. F. Straight, Nat. Cell Biol. 2002,
4, 83–88.
[6] M. S. Duxbury, S. W. Ashley, E. E. Whang, Biochem. Biophys.
Res. Commun. 2004, 313, 992–997.
[19] T. Sakamoto, J. Limouze, C. A. Combs, A. F. Straight, J. R.
Sellers, Biochemistry 2005, 44, 584–588.
[20] Attempts to interrupt the chromophore by reduction (either of
the keto or amidine functional groups) had previously led to
inactive analogues of 1 (data not shown).
[21] a) R. M. Pielak, V. A. Gaysinskaya, W. D. Cohen, Dev. Biol.
2004, 269, 421–432; b) J. Rosenblatt, L. P. Cramer, B. Baum,
K. M. McGee, Cell 2004, 117, 361–372; c) A. I. Ivanov, I. C.
McCall, C. A. Parkos, A. Nusrat, Mol. Biol. Cell 2004, 15,
2639–2651; d) M. Abe, C. H. Ho, K. E. Kamm, F. Grinnell, J.
Biol. Chem. 2003, 278, 47707–47712; e) A. Ponti, M. Mach-
acek, S. L. Gupton, C. M. Waterman-Storer, G. Danuser, Sci-
ence 2004, 305, 1782–1786; f) M. A. Griffin, S. Sen, H. L.
Sweeney, D. E. Discher, J. Cell Sci. 2004, 117, 5855–5863.
Received: February 08, 2005
[7] G. Gallo, Exp. Neurol. 2004, 189, 112–121.
[8] Y. Uetake, G. Sluder, J. Cell Biol. 2004, 165, 609–615.
[9] M. Kovacs, J. Toth, C. Hetenyi, A. Malnasi-Csizmadia, J. R.
Sellers, J. Biol. Chem. 2004, 165, 35557–35563.
[10] J. Kolega, Biochem. Biophys. Res. Commun. 2004, 320, 1020–
1025.
[11] a) F. A. Davis, A. C. Sheppard, Tetrahedron 1989, 45, 5703–
5742; b) F. A. Davis, A. C. Sheppard, B-C. Chen, M. S. Haque,
J. Am. Chem. Soc. 1990, 112, 6679–6690; c) F. A. Davis, M. C.
Weismiller, C. K. Murphy, T. Reddy, B-C. Chen, J. Org. Chem.
1992, 57, 7274–7285.
1740
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 1736–1740