Rapidly photoactivatable ATP probes for specific labeling of tropomyosin within the actomyosin protein complex

Article abstract of DOI:10.1016/j.bmcl.2011.02.113

Tropomyosin-specific photoaffinity adenosine triphosphate (ATP) probes have been first developed, in which a diazirine moiety is incorporated into the γ-phosphate group as a rapidly carbene-generating photophore. These probes clearly labeled tropomyosin i

Full text of DOI:10.1016/j.bmcl.2011.02.113

Bioorganic & Medicinal Chemistry Letters 21 (2011) 2252–2254
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Rapidly photoactivatable ATP probes for specific labeling of tropomyosin
within the actomyosin protein complex
⇑
Souta Masuda, Takenori Tomohiro, Yasumaru Hatanaka
Laboratory of Biorecognition Chemistry, Graduate School of Medicine and Pharmaceutical Sciences, University of Toyama, 2630 Sugitani, Toyama 930-0194, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Tropomyosin-specific photoaffinity adenosine triphosphate (ATP) probes have been first developed, in
Received 22 November 2010
Revised 24 February 2011
Accepted 26 February 2011
Available online 4 March 2011
which a diazirine moiety is incorporated into the
c-phosphate group as a rapidly carbene-generating
photophore. These probes clearly labeled tropomyosin in the presence of other actomyosin components,
that is, myosin, actin, and troponins. The specific labeling of tropomyosin was easily identified by selec-
tive trapping of the photo-incorporated ATP probe on Fe3+-immobilized metal ion affinity chromatogra-
phy (IMAC) beads. The characteristic nature of tropomyosin-specific photocross-linking was further
confirmed with a biotin-carrying derivative of the ATP probe. These data suggest that the tropomyosin
on the actin filament assembly is located in close proximity to the ATP binding cavity of myosin.
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
Actomyosin
ATP
Diazirine
Photoaffinity labeling
Tropomyosin
Adenosine triphosphate (ATP) is one of the most important mol-
ecules in cell biology; it plays roles in various cell functions, includ-
ing extra- and intracellular signaling, biosynthesis, active
membrane transport of macromolecules, and protein-based molec-
ular motors. Myosins, a family of molecular motors, are ATPases
that transduce the energy from ATP hydrolysis to mechanical
movement on actin thin filaments in muscle contraction. Despite
the recent developments in crystallography, it is difficult to ana-
lyze the moving conformation of the actin–myosin system,1 and
chemical cross-linking is required for crystallization of the ATP
binding state.2 Thus, various site-specific probes have been devel-
oped to investigate the dynamic nature of the actin system;3 of
these, ATP photoprobes are especially useful for cross-linking
protein–protein interactions as well as mapping the ATP binding
sites.4
kinases.7 Although benzophenone and arylazide have been used
in the ATP probe synthesis as photoactivatable species, the use of
diazirine-based ATP analogs has not yet been reported, primarily
because of the inconvenience experienced in probe synthesis. Tri-
fluoromethyl phenyldiazirine, however, has become increasingly
important because it rapidly produces very reactive carbene upon
photolysis at 360 nm.11 The generated carbene introduces specific
and stable cross-links on the target protein, and the availability of
the diazirine and its probes is greatly improved.12 We previously
used new diazirine-carrying guanosine triphosphate analogs for
the rapid proteomic analysis of G-coupled proteins.13 In this Letter,
we describe an efficient method for detection of tropomyosin with-
in the actomyosin complex using novel diazirinyl ATP probes.
4-[3-(Trifluoromethyl)-3H-diazirin-3-yl]benzylamine 2a was
derived from the corresponding bromide14 in 2 steps via synthesis
of phthalimide derivative 1a based on the Gabriel method and was
then coupled with ATP using 1-ethyl-3-(3-dimethylaminopropyl)
carbodiimide (EDCI) by a method of modifying the previous work15
to give the corresponding phosphoramidate probe 3a (Fig. 1). This
product was then purified by reverse-phase HPLC on ODS with a
liner gradient of 0–50% acetonitrile-water containing 50 mM tri-
ethylammonium acetate (pH 7.0) over 20 min at a flow rate of
1 mL/min. The 16.0 min peak was collected and freeze-dried. The
probe 3b bearing a biotin moiety was also prepared in a similar
way. The structure of probes 3a and 3b was confirmed by NMR
and high-resolution MS.16
The ATP probes include a photoreactive moiety within the li-
gand structure, in which azido- or benzophenone groups are com-
monly used. These photophores can be incorporated into the
adenine ring,5 within the ribose structure,6 or on the phosphate
linkage.7 Among of these three types of ATP probes,
c-phos-
phate-modified probes can be a powerful tool for analyzing the
force generation process of the actin–myosin motor derived from
ATP hydrolysis.2,8 Because a phosphoramide9 or thiophosphate10
group resists to hydrolysis under physiological conditions, several
photoactivatable ATP analogs modified at the
c-phosphate have
been used as cross-linking reagents for binding analysis of
Photolabeling of actomyosin obtained from chicken skeletal
muscle with the probe 3a was carried out under UVA (ꢀ360 nm)
irradiation for 10 min at 0 °C in the absence of Ca2+ by addition
⇑
Corresponding author. Tel.: +81 76 434 7515; fax: +81 76 434 5063.
E-mail address: yasu@pha.u-toyama.ac.jp (Y. Hatanaka).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.02.113

S. Masuda et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2252–2254
2253
Figure 1. Synthesis of photoactivatable probes 3: (a) 40% MeNH2/MeOH, rt, (b) ATP, EDCI, pH 6.8, 25 °C.
of ethylene glycol tetraacetic acid (EGTA). The sample was treated
with biotin N-hydroxysuccinimide for chemiluminescent detection
of protein bands using avidin-horseradish peroxidase (HRP) conju-
gate. The products were subjected to Fe3+-IMAC, and the photola-
beled proteins were isolated by chelating a photo-incorporated
ATP tag with Fe3+ on the resin.13 Coomassie brilliant blue (CBB)
staining of actomyosin after SDS–PAGE showed that actomyosin
is mainly made up of a mix of proteins such as myosin, actin,
and tropomyosin (lane 8). A band around 37 kDa was clearly de-
tected as a major photolabeled product (Fig. 2, lane 2), which cor-
responds to tropomyosin among the actomyosin components. It
was detected by Western blot analysis using anti-tropomyosin
(lane 6) as the same position of tropomyosin in actomyosin (lane
7). The binding specificity of the probe to each of these proteins
lated proteins. These proteins are also able to bind to Fe-resin and
may still adhere to the resin after washing. To avoid both nonspe-
cific biotin labeling and the following IMAC purification steps, the
biotinylated probe 3b was used for the detection of labeled prod-
ucts in the actomyosin motor system. The biotinylated products
in this case should be only those molecules cross-linked with the
probe. In addition, this procedure would simplify the analytical
processes for detecting or purifying labeled products. Photoaffinity
labeling with probe 3b and inhibition assays showed similar re-
sults to those obtained by the procedure using probe 3a (Fig. 3),
regardless of the presence of the biotin moiety. Tropomyosin was
primarily detected in UV-irradiated samples (Fig. 3, lane 2), and
the labeled amount decreased depending on the concentration of
c-thio-ATP. This result indicates that the biotin tag on probe 3b
was evaluated on the basis of the inhibitory effects of
c
-thio-ATP
does not have a large effect on the myosin–actin filament interac-
tion, and the binding of these photoprobes within the actomyosin
complex are similarly allocated to that of natural ATP, directing the
phosphate terminal to the actin filament, around which the tropo-
myosin is twisted.19
on the photoaffinity labeling because
c-thio-ATP is an ATP analog
that is stable against hydrolysis.17 As shown in Figure 2, the labeled
amount of 37 kDa protein was reduced by increasing the concen-
tration of c-thio-ATP (lanes 3–5).
The labeled 37-kDa protein was identified by peptide mass fin-
gerprinting analysis of the trypsin-digested sample by using MAL-
DI-TOF MS. Some peptides were detected at m/z = 1157, 1285,
1400, and 1612, and MS/MS analysis of the peak at m/z 1285 indi-
cated that this peptide consisted of the amino acids KLVIIEGDLER.
This sequence exists in part of tropomyosin obtained from chicken
skeletal muscle (NCBI gi|833617).18 Consequently, we concluded
that 3a specifically labeled tropomyosin in the actomyosin com-
plex. In binding analysis of the myosin motor protein complex,
tropomyosin was labeled by photoactivatable ATP analogs for the
first time.
Tropomyosin is composed of 2 a-helixes that form an elongated
coiled-coil structure and lie along the helical strand of the actin
thin filament. Tropomyosin induces molecular switching in re-
sponse to Ca2+ concentration changes to regulate actomyosin
motility. At low Ca2+ concentrations, tropomyosin blocks the myo-
sin-binding site of the actin filament in cooperation with troponin,
which is a tropomyosin-binding protein consisting of 3 subunits
(TnT, TnI, and TnC). Upon Ca2+ influx, calcium-bound TnC causes
TnI to release the hold on actin. This allows tropomyosin to slide
and thus open the myosin-binding site,20 and myosin can bind
strongly to the actin filament in the ATP hydrolysis cycle. Since
we have confirmed specific labeling of tropomyosin in the actomy-
However, the procedures for detection and purification of prod-
ucts labeled with 3a retain the possibility of including phosphory-
osin complex by the ATP analogs bearing a diazirine at the c-phos-
phate 3a or 3b in the absence of Ca2+, the probe 3a was used for
photolabeling actomyosin under conditions that included 1 mM
CaCl2. Unexpectedly, tropomyosin was mainly labeled by the ATP
probe, which is independent of Ca2+ concentration (Fig. 4). Little la-
beled actin or myosin was observed.
Figure 2. Photoaffinity labeling of actomyosin with probe 3a. Actomyosin (3.9
lg)
was incubated with 3a (125 M) in a solution containing 30 mM Tris–HCl (pH 7.2),
l
5 mM MgCl2, and 1 mM EGTA for 1 h followed by irradiation with 360 nm light. The
biotinylated samples isolated by Fe3+-IMAC were subjected to 10% SDS–PAGE and
then blotted onto a polyvinylidene difluoride (PVDF) membrane. Lane 1: size
markers; lanes 2–5: chemiluminescent detection of photolabeled samples using
Figure 3. Photoaffinity labeling of actomyosin with biotinylated probe 3b. Phot-
olabeling was carried out under the same condition as that of 3a in the absence
(lane 2) or presence (lanes 3–6) of c-thio-ATP. Photoproducts were separated by
avidin-HRP conjugate in the presence of 0, 0.125, 1.25, and 12.5 mM
c
-thio-ATP,
10% SDS–PAGE and then blotted on a PVDF membrane. The labeled products were
detected by a chemiluminescent method using avidin-HRP conjugate. Lane 1: size
markers; lanes 2–5: photolabeled samples in the presence of 0, 0.05, 5, and
respectively; lanes 6 and 7: Western blot analysis of photolabeled sample isolated
by Fe3+-IMAC and actomyosin (78 ng) using anti-tropomyosin, respectively; lane 8:
CBB-stained PAGE separation of actomyosin components.
12.5 mM c-thio-ATP, respectively; lane 6: sample without irradiation.

2254
S. Masuda et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2252–2254
Acknowledgements
We are grateful to Professor Kazuo Harada of the Department of
Life Sciences, Tokyo Gakugei University, for affording the facilities
for analysis of protein complexes. This research was supported by
Grants-in-Aid for Scientific Research (20390032, 21310138) from
the Ministry of Education, Culture, Sports, Science and Technology
of Japan.
Figure 4. Effect of calcium on labeling of tropomyosin. Actomyosin (3.9 lg) was
incubated with probe 3a (125 lM) in a solution containing 30 mM Tris–HCl (pH
References and notes
7.2), 1 mM CaCl2 or 1 mM EGTA, and 5 mM MgCl2. Photolabeling and detection
were performed by using the same method described in Figure 2. Lane 1: size
markers; lanes 2 and 3: chemiluminescent detection of photolabeled samples in the
presence of 1 mM CaCl2 (lane 2) and in the absence of Ca2+ (lane 3).
1. Spudich, J. A.; Sivaramakrishnan, S. Nat. Rev. Mol. Cell Biol. 2010, 11, 128.
2. Himmel, D. M.; Gourinath, S.; Reshetnikova, L.; Shen, Y.; Szent-Györgyi, A. G.;
Cohen, C. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 12645.
3. Thomas, D. D.; Kast, D.; Korman, V. L. Annu. Rev. Biophys. 2009, 38, 347.
4. Yount, R. G.; Cremo, C. R.; Grammer, J. C.; Kerwin, B. A. Philos. Trans. R. Soc.
Lond., B, Biol. Sci. 1992, 336, 55.
5. (a) Wong, Y. S.; Lagarias, J. C. Proc. Natl. Acad. Sci. U.S.A. 1989, 86, 3469; (b)
Zhang, Y. Y.; Hammarberg, T.; Radmark, O.; Samuelsson, B.; Ng, C. F.; Funk, C.
D.; Loscalzo, J. Biochem. J. 2000, 351, 697; (c) Maruta, S.; Ohki, T.; Kambara, T.;
Ikebe, M. Eur. J. Biochem. 1998, 256, 229.
6. (a) Williams, N.; Coleman, P. S. J. Biol. Chem. 1982, 257, 2834; (b) Oliveira, R. G.;
Coan, C.; Verjovski-Almeida, S. J. Biol. Chem. 1988, 263, 1861; (c) Williams, N.;
Ackerman, S. H.; Coleman, P. S. Methods Enzymol. 1986, 126, 667.
7. (a) Thompson, A. A.; Zou, A.; Yan, J.; Duggal, R.; Hao, W.; Molina, D.; Cronin, C.
N.; Wells, P. A. Biochemistry 2009, 48, 906; (b) Vandest, P.; Labbe, J. P.; Kassab,
R. Eur. J. Biochem. 1980, 104, 433.
8. Geeves, M. A.; Holmes, K. C. Adv. Protein Chem. 2005, 71, 161.
9. Jahn, W. Biochemistry 2007, 46, 9654.
10. Allen, J. J.; Li, M.; Brinkworth, C. S.; Paulson, J. L.; Wang, D.; Hübner, A.; Chou,
W. H.; Davis, R. J.; Burlingame, A. L.; Messing, R. O.; Katayama, C. D.; Hedrick, S.
M.; Shokat, K. M. Nat. Methods 2007, 4, 511.
Figure 5. Docking model of the photoprobe 3a binding to chicken myosin. The
probe 3a is shown as a CPK model, in which the nitrogen (blue) of the diazirine is
indicated by a blue arrow. The probe binds to the ATP binding site and directs its
photoreactive diazirine to the 50-kDa cleft (actin-binding interface) indicated by
the red arrow. Original data refers to PDB: 1W7I.
11. (a) Nassal, M. Liebigs Ann. Chem. 1983, 1519; (b) Wang, J.; Kubicki, J.; Peng, H.;
Platz, M. S. J. Am. Chem. Soc. 2008, 130, 6604.
12. (a) Hashimoto, M.; Hatanaka, Y. Eur. J. Org. Chem. 2008, 2008, 2513; (b)
Blencowe, A.; Hayes, W. Soft Matter 2005, 1, 178; (c) Moss, R. A. Acc. Chem. Res.
2006, 39, 267.
13. Kaneda, M.; Masuda, S.; Tomohiro, T.; Hatanaka, Y. ChemBioChem 2007, 8, 595.
14. (a) Nakashima, H.; Hashimoto, M.; Sadakane, Y.; Tomohiro, T.; Hatanaka, Y. J.
Am. Chem. Soc. 2006, 128, 15092; (b) Hatanaka, Y.; Kempin, U.; Park, J.-J. J. Org.
Chem. 2000, 65, 5639.
Our ATP probes clearly showed the tropomyosin-specific nature
of labeling in actomyosin. Previous reports showed that photoacti-
vatable ATP analogs labeled myosin or myosin S1, in which the
photophore was attached on the adenine ring5c or in the ribose
ring,21 respectively. We examined the molecular docking of
photoprobe 3a with the reported X-ray crystallographic structure
of myosin. Since the ATP-bound crystal structure has only been re-
vealed for scallop myosin,2 we used the chicken myosin–ADP com-
plex from the Protein Data Bank (PDB) 1W7I dataset which reflects
the ATP binding state.22 The probe 3a manually docked into 1W7I
by replacing bound ADP (Fig. 5). The probe binds to the ATP binding
site of myosin by directing the photoactivatable diazirine to actin
within the actomyosin complex, and the distance from the diazi-
rine to the actin filament could be estimated at a few nm.23 Since
the short-lived carbene intermediate immediately cross-links to a
spatially closed molecule,11 our result of probing the tropomyosin
on the actin filament suggests that the diazirine moiety of 3a and
3b should exist near to the tropomyosin coil on the actin filament
through binding with myosin. This suggests the hypothesis that
the tropomyosin coil remains in the 50-kDa cleft throughout the
15. Rajagopalan, K.; Chavan, A. J.; Haley, B. E.; Watt, D. S. J. Biol. Chem. 1993, 268,
14230.
16. Compound 3a: 1H NMR (500 MHz, D2O, Me3SiCD2CD2CO2Na (TSP-d4)): d 8.41
(1H, s), 8.18 (1H, s), 7.13 (2H, d, J = 8.3 Hz), 6.84 (2H, d, J = 8.3 Hz), 5.97 (1H, d,
J = 5.6 Hz), 4.59 (1H, t, J = 5.3 Hz), 4.48 (1H, t, J = 4.3 Hz), 4.36–4.42 (2H, m),
4.24–4.30 (1H, m), 3.92 ppm (2H, d, J = 8.1 Hz); 19F NMR (376 MHz, D2O-
CD3OD, CFCl3): d À64.2 ppm (3F, s); HRMS (FAB+): Calcd for C19H22O12N8F3P3
[M+H]+ 705.0601. Found 705.0624; 3b: 1H NMR (500 MHz, CD2OD): d 8.66 (1H,
br s), 8.16 (1H, s), 7.61 (1H, br s), 6.79 (1H, d, J = 7.0 Hz), 6.63 (1H, S), 6.09 (1H,
d, J = 5.6 Hz), 4.57 (1H, br s), 4.44 (1H, dd, J = 4.7, 8.1 Hz), 4.18–4.30 (5H, m),
4.11 (2H, t, J = 4.7 Hz), 3.87 (2H, t, J = 4.7 Hz), 3.71–3.73 (2H, m), 3.63–3.66 (2H,
m), 3.55 (2H, t, J = 5.1 Hz), 3.34 (2H, t, J = 5.1 Hz), 3.10–3.15 (1H, m), 2.88 (1H,
dd, J = 5.1, 12.8 Hz), 2.67 (1H, d, J = 12.8 Hz), 2.16–2.20 (2H, m), 1.38–1.62 (2H,
m), 1.35–1.40 (2H, m), 1.28–1.31 ppm (2H, m); 19F NMR (367 MHz, CD3OD,
CFCl3): d À65.2 ppm (s, 3F); HRMS (FAB+): Calcd for C35H52O17N11F3P3S [M+H]+
1080.2428. Found 1080.2465.
17. Prochniewicz, E.; Walseth, T. F.; Thomas, D. D. Biochemistry 2004, 43, 10642.
18. Lemonnier, M.; Balvay, L.; Mouly, V.; Libri, D.; Fiszman, M. Y. Gene 1991, 107,
229.
19. Wu, S.; Liu, J.; Reedy, M. C.; Winkler, H.; Reedy, M. K.; Taylor, K. A. J. Struct. Biol.
2009, 168, 485.
ATP binding stage regardless of the presence of Ca2+
In conclusion, we have synthesized diazirine-based photoacti-
vatable ATP probes modified at the -phosphate, and these probes
.
´
20. Lehman, W.; Galinska-Rakoczy, A.; Hatch, V.; Tobacman, L. S.; Craig, R. J. Mol.
Biol. 2009, 388, 673.
c
21. (a) Cole, D. G.; Yount, R. G. J. Biol. Chem. 1990, 265, 22537; (b) Sutoh, K.;
Yamamoto, K.; Wakabayashi, T. Proc. Natl. Acad. Sci. U.S.A. 1986, 83, 212.
22. Coureux, P.-D.; Sweeney, H. L.; Houdusse, A. EMBO J. 2004, 23, 4527.
23. (a) Pfaendtner, J.; Lyman, E.; Pollard, T. D.; Voth, G. A. J. Mol. Biol. 2010, 396,
252; (b) Houdusse, A.; Kalabokis, V. N.; Himmel, D.; Szent-Györgyi, A. G.;
Cohen, C. Cell 1999, 97, 459.
provided the first case showing that tropomyosin is specifically la-
beled by ATP analogs in the actomyosin complex. Our probe could
therefore be a useful tool to analyze the role of tropomyosin during
the conformational change of the actomyosin complex.