Novel bifunctional probe for radioisotope-free photoaffinity labeling: Compact structure comprised of photospecific ligand ligation and detectable tag anchoring units

Article abstract of DOI:10.1039/b316221d

A novel method for radioisotope-free photoaffinity labeling was developed, in which a bifunctional ligand is connected to a target protein by activation of a photoreactive group, such as an aromatic azido or 3-trifluoromethyl-3H-diazirin-3-yl group, and identification of the ligated product is achieved by anchoring of a detectable tag through the Staudinger-Bertozzi reaction with an alkyl azido moiety that survives photolysis. The chemical ground of this method was confirmed using model compounds with the bifunctional group under photoirradiation in the presence of trapping agents for reactive intermediates. The utility of the method has been demonstrated by specific labeling of the catalytic portion of human HMG-CoA reductase.

Full text of DOI:10.1039/b316221d

View Article Online / Journal Homepage / Table of Contents for this issue
Novel bifunctional probe for radioisotope-free photoaffinity
labeling: compact structure comprised of photospecific ligand
ligation and detectable tag anchoring units†
Takamitsu Hosoya,a,b Toshiyuki Hiramatsu,b Takaaki Ikemoto,c Masayuki Nakanishi,b
Hiroshi Aoyama,b Ayako Hosoya,a Tomoya Iwata,a Kei Maruyama,c Makoto Endoc and
Masaaki Suzuki*a,b
a Division of Regeneration and Advanced Medical Science,
Gifu University Graduate School of Medicine, Yanagido 1-1, Gifu 501-1193, Japan.
E-mail: suzukims@biomol.gifu-u.ac.jp
b Department of Biomolecular Science, Faculty of Engineering, Gifu University, Yanagido 1-1,
Gifu 501-1193, Japan
c Department of Pharmacology, Saitama Medical School, Moroyama-machi,
Saitama 350-0495, Japan
Received 12th December 2003, Accepted 23rd January 2004
First published as an Advance Article on the web 4th February 2004
A novel method for radioisotope-free photoaffinity label-
ing was developed, in which a bifunctional ligand is con-
nected to a target protein by activation of a photoreactive
group, such as an aromatic azido or 3-trifluoromethyl-3H-
diazirin-3-yl group, and identification of the ligated prod-
uct is achieved by anchoring of a detectable tag through
the Staudinger–Bertozzi reaction with an alkyl azido
moiety that survives photolysis. The chemical ground of
this method was confirmed using model compounds with
the bifunctional group under photoirradiation in the pres-
ence of trapping agents for reactive intermediates. The
utility of the method has been demonstrated by specific
labeling of the catalytic portion of human HMG-CoA
reductase.
on a compact bifunctional probe unit capable of differentiating
between ligand ligation and labeled protein detection.
The concept of our method based on the design of a bifunc-
tional photoaffinity probe equipped with a highly photo-
sensitive group, such as an aromatic azido or 3-trifluoromethyl-
3H-diazirin-3-yl group, and a relatively photostable alkyl azido
group is illustrated in Fig. 1.5 We anticipated that an alkyl azido
group could be left intact under conditions for photoactivation
of an aryl azido or 3-trifluoromethyl-3H-diazirin-3-yl group
(Step 1), and could therefore be used as a tag for consecutive
detection of the covalently cross-linked protein.6,7 Thus, we
envisaged that the remaining alkyl azido group could be labeled
chemoselectively by Staudinger–Bertozzi ligation8 using a
fluorescent marker-9 or biotin-anchored8 triarylphosphine
derivative (Step 2).10 We also considered that the introduction
of a small azidomethyl group would be preferable to minimize
steric effects and changes in polarity that may adversely influ-
ence the bioactivity. Furthermore, the additional azido and
azidomethyl groups would be placed in meta positions to avoid
unnecessary interactions with the three neighboring functional
groups on the aromatic ring.11
First, we conducted model experiments using azido- and tri-
fluoromethyldiazirinyl-functionalized compounds, 1 and 5, to
verify the tolerance of the alkyl azido group under the condi-
tions of photoactivation adaptable to the aryl azido and
diazirinyl photo-functions (Scheme 1). Thus, a C2-symmetrical
triazido-functionalized compound 1 (20 mM in cyclohexane-
d12)12 and a 300-fold excess of diethylamine were put in a quartz
NMR tube and exposed to UV with a wavelength of 254 nm
(Asahi Spectra, LAX-101, 100 W) at 22 ЊC (Scheme 1a). The
course of the reaction was monitored by 1H-NMR. Dis-
appearance of 1 was accompanied by a significant set of new
spectra. Consequently, irradiation for 8 h afforded 4 in 56%
yield, which is equivalent to 64% yield based on the consump-
tion of 1. The product was isolated and characterized as the di-
ethylamino-substituted 3H-azepine 4, obtained by nucleophilic
addition of diethylamine to didehydroazepine intermediate 3
produced by ring expansion of photo-generated singlet nitrene
species 2.2b,5,13 The above observations indicated that most of
the alkyl azido groups remained intact under the conditions for
photoactivation of the aromatic azido group.14 Similarly, the
diazirinyl-type compound 5 (3.0 mg/0.75 mL CD3OD) under-
went photoreaction in a quartz NMR tube by continuous
irradiation with UV at 365 nm (UVP, UVL-56, 6 W) for 10 min
and 302 nm (UVP, UVM-57, 6 W) for 8 min at 22 ЊC (Scheme
1b).15 On 19F-NMR, the peak corresponding to 5 (δ 10.8 ppm)
The discovery of new proteins with specific functions is a cen-
tral theme in post-genomic science. Photoaffinity labeling
(PAL)1,2 has become a powerful method in this area as con-
firmed by the identification of target proteins of biologically
active molecules and elucidation of their binding sites. There-
fore, PAL offers good potential for discovery and develop-
ment of drugs.3 PAL studies focus on how to design an efficient
photoaffinity probe satisfying the following three requirements:
(1) strong and specific binding with a target protein, (2) high
photospecificity to allow the efficient formation of a covalent
bond between the ligand and its target protein, and (3) high
sensitivity of detection. Photoreactive functional groups, such
as azido, diazo, diazirinyl, and benzophenone, and radioiso-
topes (RIs), such as 3H, 14C, 32P, 35S, and 125I, are frequently used
for the second and third purposes. With the increase in import-
ance of PAL, novel strategies that do not require the use of RIs
are urgently needed. Photoaffinity biotinylation is a represent-
ative non-RI-based labeling protocol.4 This method utilizes
strong avidin–biotin complex formation to allow harmless
chemiluminescent detection and direct affinity purification of
a labeled protein for sequencing. However, the introduction of a
highly polar and sterically congested biotin-anchored tag to
an affinity compound often results in marked impairment of
intrinsic biological activity in the crucial probe design step.
This paper presents a novel protocol for RI-free PAL based
† Electronic supplementary information (ESI) available: Photo-
decomposition study of phenyl and benzyl azides. Experimental details
for photoreactions of 1 and 5. Synthetic procedures and characteriz-
ation of new compounds. See http://www.rsc.org/suppdata/ob/b3/
b316221d/.
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 6 3 7 – 6 4 1
637

View Article Online
Fig. 1 Concept for radioisotope-free photoaffinity labeling using a specific ligand with bifunctional groups. The photoreaction realizes formation
of a covalent bond between the ligand and target protein (step 1) and consecutive Staudinger–Bertozzi ligation of the detectable tag allows
visualization of the captured protein (step 2).
Scheme 1 Photoreactions in model compounds to confirm functional discrimination.
disappeared completely and a new peak corresponding to the
adduct 7 appeared at δ Ϫ0.4 ppm. Adduct 7 was produced by
insertion of the photo-generated carbene species 6 into CD3OD
and was almost the sole product (>95% according to inte-
gration).4f,16 Adduct 7 was isolated by preparative TLC in 82%
yield. Thus, we showed that the aryl azido and diazirinyl units
are preferably photoactivated to the alkyl azido group under
the above conditions, and therefore, the latter alkyl azido
functional group could be utilized as trigger component for
Staudinger–Bertozzi ligation in both cases.
Encouraged by the positive verification in the model reaction,
we set up an actual PAL protocol to confirm our concept. We
chose to photolabel the catalytic portion of human 3-hydroxy-
3-methylglutaryl coenzyme A (HMG-CoA) reductase using a
diazido-functionalized cerivastatin derivative, referred to as
photovastatin CAA1 (9), and then fluorescein-anchored tri-
arylphosphine derivative, GIF-0373 (10).
CoA to mevalonate.17 Cerivastatin18 and other therapeutically
useful statins inhibit HMGR activity by binding to its active
site19 to decrease cholesterol biosynthesis, thus reducing the
risk of coronary heart disease.20 In contrast to their clinical
efficacy, statins occasionally trigger harmful side effects, such as
rhabdomyolysis.21 A high incidence of this fatal myopathy was
induced by treatment with cerivastatin, prompting removal of
this statin from the global market.22 The molecular mechanism
underlying statin-induced myopathy is still not clear.20,23 Prom-
ising pleiotropic effects of statins on other diseases, such as
osteoporosis, dementia, and multiple sclerosis, have attracted a
great deal of attention in recent years,24 leading to the demand
for elucidation of the mechanism of their severe adverse effects
in association with the expansion of the application of these
molecules and new drug discovery. Our photoaffinity probe 9
was also designed with the aim of identifying the key molecule
that triggers the myopathy.
HMG-CoA reductase (HMGR) is the rate-limiting enzyme
in the cholesterol biosynthetic pathway, which converts HMG-
In designing 9, we replaced the methoxy moiety of ceriva-
statin, which is the most distinctive structural unit among the
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 6 3 7 – 6 4 1
638

View Article Online
Scheme 2 Synthesis of photovastatin CAA1 (9). (a) NaNO2, AcOH/H2O (9 : 1), 0 ЊC, 3 min, then NaN3, 0 ЊC, 15 min, 88%; (b) CBr4, PPh3, DMF,
0 ЊC, 2 h, 53%; (c) NaN3, DMF, 0 ЊC to room temperature, 6.5 h, 96%; (d) NaH, DMF, 0 ЊC, 15 min, then 15, 0 ЊC, 3 h, 59%; (e) 10% aq.
HCl/THF (1:4), room temperature, 14 h, 91%; (f ) aq. NaOH, THF, room temperature, 11.5 h, 93%; (g) Ph2P(OEt), 150 ЊC, 3 h, 99%; (h) nBuLi, THF,
0 ЊC, 20 min, then 14, 0 ЊC, 4 h, 66%; (i) nBu4NF, THF, room temperature, 5 h, 98%; (j) CBr4, PPh3, CH2Cl2, room temperature, 1 h, 71%. TBDMS =
tert-butyldimethylsilyl.
HMGR426–888 in the reaction mixture for 1–2 min at 95 ЊC in
0.1 M Tris-Cl (pH 6.8) containing 1% sodium dodecylsulfate
(SDS) and 2.5% β-mercaptoethanol, and (5) SDS-polyacryl-
amide gel electrophoresis (SDS-PAGE) of the mixtures using
12.5% polyacrylamide gels in the Laemmli buffer system.30 As
shown in Fig. 2, the gels were directly visualized by laser-
scanning in a fluorescent imaging analyzer.31 As can be seen
from lanes 1, 3, 5, and 7, HMGR426–888, with a calculated
molecular mass of 51 kDa, was successfully photolabeled by 9
in a dose-dependent manner. It is worth noting that the labeled
signal was detectable even when 9 was employed at a concen-
tration of 2 nM (lane 1), which is close to its IC50 value. All of
the fluorescent signals were decreased by coincubation with
statins, by the photoactivatable benzyloxy group. We chose the
sterically less congested diazido derivative as a representative
to minimize the change of the biological function of the origin-
al compound. As shown in Scheme 2, the synthesis of 9 was
accomplished by coupling the diazido-functionalized benzyl
alcohol 12 with bromide 15, derived from known components
1125 and 13,26 respectively, under conventional conditions,
followed by successive deprotections of acetal and ester por-
tions. The inhibitory effect of synthesized 9 on the catalytic
portion (residues 426–888) of recombinant human HMGR
(HMGR426–888) was assayed.27 Under our assay conditions,28 the
IC50 values for 9 and cerivastatin were determined to be 0.95
nM and 0.93 nM, respectively. Thus, we judged that 9 could be
used as a photoaffinity probe.
Using the promising photoaffinity probe 9, we next
carried out the actual PAL experiment according to the proto-
col shown in Fig. 1 as follows: (1) incubation of 9 in various
Fig. 2 SDS-PAGE analysis of HMGR426–888 (51 kDa) photolabeled in
various concentrations of 9 in the absence (Ϫ) or presence (ϩ) of
concentrations (final concentration, 2, 5, 20, 50 nM) with
HMGR426–888 for 5–10 min at room temperature in the absence
or presence of cerivastatin (final concentration, 5 µM),29 (2)
photoirradiation of the mixtures using a portable UV-lamp
(UVP, UVG-54, 254 nm, 6 W, 60 s × 2, from a distance of 1 cm,
room temperature), (3) Staudinger–Bertozzi ligation by add-
ing an excess amount of 10 (100 µM) to the mixtures and
incubation for 4 h at room temperature, (4) denaturation of
cerivastatin (5 µM) followed by chemoselective ligation with 10
(100 µM). The fluorescent signal was visualized by laser-scanning of the
electrophoresed gel with a fluorescent imaging analyzer (FluorImager
SI, Molecular Dynamics, 488 nm excitation, 530 15 nm detection
filter). For comparison, Coomassie brilliant blue (CBB)-stained
HMGR426–888, which was transferred onto a polyvinylidene difluoride
(PVDF) membrane after SDS-PAGE, is shown in the leftmost lane. The
rightmost lane shows the fluorescent high molecular weight marker
(F-HMWM, Sigma).
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 6 3 7 – 6 4 1
639

View Article Online
excess amounts of cerivastatin, indicating that the cross-
linking proceeded specifically at the active binding site (lanes 2,
4, 6, 8).32
from kinetic and structural studies that NADPH does not
share its binding site on HMGR with statins,38 this co-enzyme
would be invoking some conformational change on stain-
photolabeled HMGR complex to confer some resistance to
proteolysis, although the precise effect of NADPH is not yet
clear. These results demonstrated the specific cross-linking
of 9 at the C-terminal domain of HMGR, as expected based
on the X-ray co-crystallographic structures of HMGR and
cerivastatin.19,39
In the present study, we developed a novel chemical method-
ology for RI-free PAL. This protocol may also be utilized not
only for other types of photoreactive functions, such as nitro-
or perfluoro-substituted aryl azides and benzophenone deriv-
atives to modulate the efficiency of the ligand–protein ligation,
but may also allow the introduction of an alkyl azido group
anywhere in a ligand structure to increase structural diversity
for suitable matching of binding and activity.40 Further studies
on PAL using another diazirinyl unit as a photoreactive func-
tion and the exploration of target proteins involved in statin-
associated myopathy based on the versatile PAL method will be
reported in due course.
Next, we identified the site of cross-linkage of 9 on HMGR
by digesting the photolabeled protein.33 Low molecular weight
peptide fragments (<10 kDa) were analyzed by Tris-tricine
SDS-PAGE.34 Among the proteolytic enzymes examined,
lysylendopeptidase (LEP), which specifically cleaves the peptide
bond at the C-terminal of lysine residues, gave the best result.
As shown in Fig. 3a, after transferring proteins from the gel
onto the polyvinylidene difluoride (PVDF) membrane, the
fluorescent signals were visualized using a UV transilluminator.
As can be seen from lanes 1 and 2 in Fig. 3a, two sharp
fluorescent bands, referred to as A and B, were observed.35
After staining with Coomassie brilliant blue (CBB) (Fig. 3b),
the selected peptide fragments, bands A–E in Fig. 3, were
cut out and subjected directly to protein sequence analysis. The
first five N-terminal amino acid residues identifiable from
bands A and C were D-N-P-G-E- and L-S-E-P-S- corre-
sponding to residues 829–833 and 503–507, respectively,
of human HMGR.36 On the other hand, band B contained
an approximately 1:4 mixture of fragments starting with
D-N-P-G-E- and L-S-E-P-S-.37 As band C was derived from a
mixture without photoreaction with 9, both fluorescent bands
were mostly attributed to the same fragment starting with
D-N-P-G-E-. The different lengths of these bands could be
explained by the alternative digestion patterns by LEP. The
fluorescence intensity of these two bands differed depending on
whether the photoreaction was performed in the presence or
absence of NADPH (lanes 1 and 2 in Fig. 3a, respectively), in
which the signal for the fragment with lower molecular weight,
band B in lane 1, relatively decreased. Because it is elucidated
Acknowledgements
We are grateful to Kaneka Co. for providing an optically active
statin side-chain unit. We would like to thank Dr. Y. Sakamoto
and Ms. N. Hirosawa of the Biomedical Research Center,
Saitama Medical School, for their generous assistance with pro-
tein sequence analysis and Mr. Kin-ichi Oyama of Nagoya
University Chemical Instrumental Center for performing mass
spectral analysis. We also thank Professor Dr. K. Nishikawa,
Dr. T. Yokogawa, and Dr. S. Ohno of Gifu University for
helpful discussions. T. I. is supported by a grant from Saitama
Medical School Research Center for Genomic Medicine. This
work was supported in part by a Grant-in-Aid for Creative
Scientific Research (No. 13NP0401) from the Ministry of
Education, Culture, Sports, Science and Technology (MEXT)
of Japan.
Notes and references
1 A. Singh, E. R. Thorton and F. H. Westheimer, J. Biol. Chem., 1962,
237, 3006–3008.
2 (a) J. Brunner, Annu. Rev. Biochem., 1993, 62, 483–514;
(b) F. Kotzyba-Hibert, I. Kapfer and M. Goeldner, Angew. Chem.,
Int. Ed. Engl., 1995, 34, 1296–1312; (c) S. A. Fleming, Tetrahedron,
1995, 51, 12479–12520; (d ) Y. Hatanaka, H. Nakayama
and Y. Kanaoka, Rev. Heteroat. Chem., 1996, 14, 213–243;
(e) G. Dormán, in Bioorganic Chemistry of Biological Signal Trans-
duction (Topics in Current Chemistry 211, Ed. H. Waldmann),
Springer-Verlag, Berlin, 2001, pp. 169–225.
3 (a) G. Dormán and G. D. Prestwich, Trends Biotechnol., 2000, 18,
64–77; (b) Y. Hatanaka and Y. Sadakane, Curr. Top. Med. Chem.,
2002, 2, 271–288.
Fig. 3 Tris-tricine SDS-PAGE analysis of digested HMGR426–888
peptide fragments obtained by photolabeling with 9 (5 µM) followed by
ligation with 10 (100 µM) and then treatment with LEP (Wako). (a) The
photograph was taken with a monochrome CCD camera (ATTO,
Printgraph AE-6914) after transferring proteins from the gel onto a
PVDF membrane, and visualization on a UV transilluminator (UVP,
TLW-20, 365 nm, 8 W × 4). (b) The same membrane after CBB
4 Some examples of photoaffinity biotinylation: (a) F. M. Finn,
C. J. Stehle and K. Hofmann, Biochemistry, 1985, 24, 1960–1965;
(b) Y. Hatanaka, M. Hashimoto and Y. Kanaoka, Bioorg. Med.
Chem., 1994, 2, 1367–1373; (c) P. J. DeLaLuz, M. Golinski, D. S.
Watt and T. C. Vanaman, Bioconjugate Chem., 1995, 6, 558–566;
(d ) B. A. Gilbert and R. R. Rando, J. Am. Chem. Soc., 1995, 117,
8061–8066; (e) Y. Hatanaka, M. Hashimoto and Y. Kanaoka, J. Am.
Chem. Soc., 1998, 120, 453–454; ( f ) K. Fang, M. Hashimoto,
S. Jockusch, N. J. Yurro and K. Nakanishi, J. Am. Chem. Soc., 1998,
120, 8543–8544; (g) S. C. Alley, F. T. Ishmael, A. D. Jones and
S. J. Benkovic, J. Am. Chem. Soc., 2000, 122, 6126–6127; (h) J. Yang,
J. Dowden, A. Tatibouët, Y. Hatanaka and G. D. Holman, Biochem.
J., 2002, 367, 533–539; (i) M. Daghish, L. Hennig, M. Findeisen,
S. Giesa, F. Schumer, H. Hennig, A. G. Beck-Sickinger and
P. Welzel, Angew. Chem., Int. Ed., 2002, 41, 2293–2297; (j) W. Frick,
A. Bauer-Schåfer, J. Bauer, F. Girbig, D. Corsiero, H. Heuer and
W. Kramer, Bioorg. Med. Chem., 2003, 11, 1639–1642.
staining. Lanes
1 and 2 represent peptide fragments from the
photoreaction in the presence or absence of 250 µM NADPH,
respectively. Lane 3 shows the control experiment for proteolysis of
HMGR426–888 without photoreaction with 9. Lanes 4 and 5 indicate
ultra-low range molecular weight marker (Sigma) and fluorescent low
molecular weight marker (Sigma), respectively. The dots in lanes 1 and
2 and the circles in lane 4 in the CBB-stained membrane are pencil
marks made to pinpoint the positions of fluorescent bands and the
molecular weight marker, respectively. Fragments A–E were cut out of
the CBB-stained gel and subjected to direct protein sequence analysis
(Shimadzu, Automated Protein/Peptide Sequencer PPSQ-23A). Bands
F and G are from LEP.
5 (a) The chemistry of the azido group (Ed. S. Patai), John Wiley &
Sons, London, 1971; (b) Azides and Nitrenes: Reactivity and Utility
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 6 3 7 – 6 4 1
640

View Article Online
(Ed. E. F. V. Scriven), Academic Press, Orlando, 1984; (c) E. F. V.
Scriven and K. Turnbull, Chem. Rev., 1988, 88, 351–368; (d ) G. B.
Schuster and M. S. Platz, Adv. Photochem., 1992, 17, 69–143.
6 For the post-PAL modification (PPALM), see, T. Nagase,
E. Nakata, S. Shinkai and I. Hamachi, Chem. Eur. J., 2003, 9, 3660–
3669 and references therein.
7 For PAL on magnetic microspheres (PALMm), see, E. Halbfinger,
K. Gorochesky, S. A. Lévesque, A. R. Beaudoin, L. Sheihet,
S. Margel and B. Fischer, Org. Biomol. Chem., 2003, 1, 2821–2832.
8 (a) E. Saxon and C. R. Bertozzi, Science, 2000, 287, 207–210;
(b) K. L. Kiick, E. Saxon, D. A. Tirrell and C. R. Bertozzi, Proc.
Natl. Acad. Sci. USA, 2002, 99, 19–24; (c) E. Saxon, S. J. Luchansky,
H. C. Hang, C. Yu, S. C. Lee and C. R. Bertozzi, J. Am. Chem. Soc.,
2002, 124, 14893–14902; (d ) D. J. Vocadlo, H. C. Hang, E.-J. Kim,
J. A. Hanover and C. R. Bertozzi, Proc. Natl. Acad. Sci. USA, 2003,
100, 9116–9121.
9 (a) G. A. Lemieux, C. L. de Graffenried and C. R. Bertozzi, J. Am.
Chem. Soc., 2003, 125, 4708–4709; (b) C. C.-Y. Wang, T. S. Seo,
Z. Li, H. Ruparel and J. Ju, Bioconjugate Chem., 2003, 14, 697–
701.
10 For the two-step labeling protocol comprised of irreversible protea-
some inhibition followed by Staudinger–Bertozzi ligation with bio-
tin-anchored reagent, see, H. Ovaa, P. F. van Swieten, B. M. Kessler,
M. A. Leeuwenburgh, E. Fiebiger, A. M. C. H. van den Nieuwendijk,
P. J. Galardy, G. A. van der Marel, H. L. Ploegh and H. S. Overkleeft,
Angew. Chem., Int. Ed., 2003, 42, 3626–3629.
11 T. Hosoya, H. Aoyama, T. Ikemoto, T. Hiramatsu, Y. Kihara,
M. Endo and M. Suzuki, Bioorg. Med. Chem. Lett., 2002, 12, 3263–
3265.
12 We chose the C2-symmetrical molecule to avoid the product
complication.
13 (a) W. von E. Doering and R. A. Odum, Tetrahedron, 1966, 22, 81–
93; (b) Y.-Z. Li, J. P. Kirby, M. W. George, M. Poliakoff and G. B.
Schuster, J. Am. Chem. Soc., 1988, 110, 8092–8098; (c) E. Leyva and
R. Sagredo, Tetrahedron, 1998, 54, 7367–7374.
14 The relative rate of nitrogen evolution by photoirradiation of phenyl
azide was reported to be approximately twice that of benzyl azide,
see, R. M. Moriarty and R. C. Readon, Tetrahedron, 1970, 26, 1379–
1392. We confirmed, however, that >90% of benzyl azide remains
intact under the conditions for complete photo-decomposition of
coexistent phenyl azide. See ESI for the detail†.
27 A 1.4-kb DNA encoding human HMGR426–888 was obtained by
reverse transcription-polymerase chain reaction (RT-PCR) from
total RNA of HepG2 cells and then subcloned into the pQE30
expression plasmid (QIAGEN) to express an N-terminal hexa-
histidine-tagged protein. The recombinant protein was expressed in
E. coli JM109 cells at 37 ЊC followed by purification using a TALON
affinity matrix (Clontech). The nearly homogeneous protein was
stored in a buffer containing 50 mM sodium phosphate (pH 7.0), 0.3
M NaCl, 10 mM dithiothreitol (DTT), and 10% glycerol at Ϫ20 ЊC.
The stocked protein solution was used directly for enzyme assay and
photolabeling studies in its histidine-tagged form without further
modification.
28 The enzyme assays to determine the values of median inhibitory
concentrations (IC50) of 9 and cerivastatin for HMGR were per-
formed by monitoring the (R,S)-HMG-CoA-dependent oxidation
of NADPH at 340 nm in an absorbance microplate reader (Tecan,
Sunrise) at 30 ЊC for 18 min. Standard assay mixtures placed on the
96-well plate contained, in a final volume of 100 µL, 2.5 µL DMSO
solution of 9 or cerivastatin in various concentrations, 24 ng
HMGR426–888, 60 µM (R,S)-HMG-CoA (Sigma), 250 µM NADPH
(Oriental Yeast), 75 mM NaCl, 1 mM EDTA, 10 mM DTT, and 0.1
M sodium phosphate buffer (pH 7.0). The IC50 values were obtained
as the averages of at least three runs.
29 PAL experiments were performed under conditions similar to those
used in the enzyme assay. The reaction mixtures contained, in a final
volume of 20 µL, 0.2 µL DMSO solution of 9 in various concen-
trations with or without 0.2 µL DMSO solution of cerivastatin
(5 µM), 1 µg HMGR426–888, 60 µM (R,S)-HMG-CoA, 250 µM
NADPH, 0.2 M NaCl, and 20 mM PIPES/Tris buffer (pH 7.0).
30 U. K. Laemmli, Nature, 1970, 227, 680–685.
31 Fluorescent signals were similarly observed after transferring pro-
teins from the gel onto PVDF membranes. PAL using >50 nM of 9
allowed detection of the fluorescent signal on
a UV trans-
illuminator.
32 Not significant but slight fluorescent signal was observed even when
there was no UVirradiation. See ESI for the detail.† Staudinger–
Bertozzi ligation of photolabeled protein with a biotin-anchored
triarylphosphine derivative and its chemiluminescent detection by
streptavidin–horseradish peroxidase was also effective.
33 Increasing the amounts of 9 and enzyme used made the analysis
easier. Thus, the reaction mixtures for LEP treatment contained, in a
final volume of 40 µL, 0.4 µL DMSO solution of 9 (5 µM), 0.4 µL
15 Irradiation of 5 for 10 min with 365 nm UV afforded a mixture of
the desired CD3OD adduct 7 and 8, which is a linear diazoisomer of
5 (19F-NMR; δ 18.8 ppm), in ca. 2:1 ratio. Due to its UV absorp-
tion character, 8 was smoothly converted to 7 by exposure to UV
at a wavelength of 302 nm rather than 365 nm. See, M. Nassal,
Liebigs Ann. Chem., 1983, 1510–1523.
DMSO solution of cerivastatin (500 µM), 10 µg HMGR426–888
,
60 µM (R,S)-HMG-CoA, 250 µM NADPH, 0.2 M NaCl, and
20 mM PIPES/Tris buffer (pH 7.0). The mixture was photo-
irradiated for 60 s × 2, followed by addition of 10 (100 µM) and
incubation for 30–60 min. This was then followed by the successive
addition of SDS (a final volume, 0.1%) and 5 µg of LEP. The
mixture was incubated at 37 ЊC for 24 h and then subjected to
Tris-tricine SDS-PAGE analysis.
16 J. Brunner, H. Senn and F. M. Richards, J. Biol. Chem., 1980, 255,
3313–3318.
17 (a) K. Bloch, Science, 1965, 150, 19–28; (b) J. L. Goldstein and
M. S. Brown, Nature, 1990, 343, 425–430.
18 R. Angerbauer, P. Fey, W. Huebsch, T. Philipps, H. Bischoff,
D. Petzinna and D. Schmidt, 1992, DE 4040026.
19 E. S. Istvan and J. Deisenhofer, Science, 2001, 292, 1160–1164.
20 J. A. Tobert, Nat. Rev. Drug Discovery, 2003, 2, 517–526.
21 C. M. Ballantyne, A. Corsini, M. H. Davidson, H. Holdaas,
T. A. Jacobson, E. Leitersdorf, W. März, J. P. D. Reckless and
E. A. Stein, Arch. Intern. Med., 2003, 163, 553–564.
34 H. Schägger and G. von Jagow, Anal. Biochem., 1987, 166, 368–379.
35 The fluorescent intensities of these bands were significantly
decreased by co-incubation with a hundredfold greater amount of
cerivastatin. The lower molecular weight fragments (<4 kDa) were
superimposed with the broad low molecular weight band including
unreacted 10.
36 For the full amino acid sequence of human HMGR consisting of
888 amino acids, see, K. L. Luskey and B. Stevens, J. Biol. Chem.,
1985, 260, 10271–10277.
37 Non-florescent but CBB-stained bands D and E were shown to start
with E-F-Q-V-P- and N-L-V-G-S-, which correspond to residues
550–554 and 736–740, respectively.
38 (a) A. Endo, M. Kuroda and K. Tanzawa, FEBS Lett., 1976, 72,
323–326; (b) E. S. Istvan, M. Palnitkar, S. K. Buchanan and
J. Deisenhofer, EMBO J., 2000, 19, 819–830; (c) E. S. Istvan
and J. Deisenhofer, Biochim. Biophys. Acta, 2000, 1529, 9–18.
39 Protein Data Bank, accession number 1HWJ.
40 Bioconjugation by azide–alkyne [3ϩ2] cycloaddition instead of
Staudinger–Bertozzi ligation would be also applicable, see, (a) A. J.
Link and D. A. Tirrell, J. Am. Chem. Soc., 2003, 125, 11164–11165;
(b) A. Dieters, T. A. Cropp, M. Mukherji, J. W. Chin, C. Anderson
and P. G. Schultz, J. Am. Chem. Soc., 2003, 125, 11782–11783;
(c) R. Breinbauer and M. Köhn, ChemBioChem., 2003, 4, 1147–
1149 and references therein.
22 C. D. Furberg and B. Pitt, Curr. Controlled Trials Cardiovasc. Med.,
2001, 2, 205–207.
23 For cerivastatin-induced Ca2ϩ release from the sarcoplasmic reticu-
lum of mouse/rat skeletal muscle, see, R. Inoue, M. Tanabe,
K. Kono, K. Maruyama, T. Ikemoto and M. Endo, J. Pharmacol.
Sci., 2003, 93, 279–288.
24 (a) S. Youssef, O. Stüve, J. C. Patarroyo, P. J. Ruiz, J. L. Radosevich,
E. M. Hur, M. Bravo, D. J. Mitchell, R. A. Sobel, L. Steinman and
S. S. Zamvil, Nature, 2002, 420, 78–84; (b) A. Waldman and
L. Kritharides, Drugs, 2003, 63, 139–152.
25 C. Behrens, M. Egholm and O. Buchardt, Synthesis, 1992, 1235–
1236.
26 G. Schmidt, R. Angerbauer, A. Brandes, M. Muller-Gliemann,
H. Bischoff, D. Schmidt, S. Wohlfeil, W. R. Schoen, G. H.
Ladouceur, J. H. Cook II, T. G. Lease, D. J. Wolanin, R. H. Kramss,
D. L. Hertzog and M. H. Osterhout, WO 98/04528.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 6 3 7 – 6 4 1
641