Photoresponsive 5'-cap for the reversible photoregulation of gene expression

Article abstract of DOI:10.1016/j.bmcl.2011.06.119

Photoresponsive 5'-caps that can be reversibly cis-trans isomerized by light irradiation were developed for the reversible photoregulation of gene expression. The 8-naphthylvinyl cap (8NV-cap) in the trans form completely inhibited translation of mRNA, wh

Full text of DOI:10.1016/j.bmcl.2011.06.119

Bioorganic & Medicinal Chemistry Letters 21 (2011) 5457–5459
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Photoresponsive 50-cap for the reversible photoregulation of gene expression
Shinzi Ogasawara a,b, , Mizuo Maeda b
⇑
a Innovative use of Light and Materials/Life, JST-PRESTO, Sanbancho, Chiyoda-ku, Tokyo 102-0075, Japan
b Bioengineering Laboratory, RIKEN, 2-1 Hirosawa, Wako, Saitama 351-0198, Japan
a r t i c l e i n f o
a b s t r a c t
Photoresponsive 50-caps that can be reversibly cis–trans isomerized by light irradiation were developed
for the reversible photoregulation of gene expression. The 8-naphthylvinyl cap (8NV-cap) in the trans
form completely inhibited translation of mRNA, whereas the cis form yielded protein with the same effi-
ciency as mRNA capped with the normal-cap, a 26-fold higher efficiency than that of the trans form. The
8NV-capped mRNA could be switched between a translating state (ON state) and a non-translating state
(OFF state) in a reversible fashion by alternately irradiating with monochromatic 410 nm or 310 nm light.
Ó 2011 Elsevier Ltd. All rights reserved.
Article history:
Received 9 May 2011
Revised 24 June 2011
Accepted 28 June 2011
Available online 7 July 2011
Keywords:
mRNA
Protein
Translation
Photoregulation
Photoisomerization
Living processes are managed through the precise control of
‘when, where, and how long (or how frequently)’ genes are ex-
pressed. For example, expression of the bicoid protein over a period
of several hours at the anterior tip of an oocyte syncytium orga-
nizes the anterior development of drosophila embryos.1 The cyclic
expression of the Hes7 protein in the presomitic mesoderm (PSM)
occurs every 2 h during somitogenesis and accompanies formation
of the periodic structure of the somite.2 Oscillatory expression of
the Hes1 protein contributes to diverse differentiation responses
in embryonic stem (ES) cells.3 To control such important biological
events using external stimuli is a long-cherished dream for biolo-
gist. The most promising external trigger is photoirradiation be-
cause it allows accurate and easy control of the location and time
at which an event occurs. One common approach to the photoregu-
lation of gene expression involves installation of a photoprotecting
group that can be removed by photoirradiation.4 Okamoto et al. de-
scribed a mRNA caging system that used a caging compound
6-bromo-4-diazomethyl-7-hydroxycoumarine (Bhc-diazo).5 They
introduced the Bhc group into mRNA at a rate of 30 phosphate sites
per 1 kb mRNA by mixing in DMSO. Bhc-caged mRNA displayed
severely reduced translational activity, whereas illumination of
Bhc-caged mRNA with ultraviolet light enabled recovery of the
translational activity. However, the recovery rate of gene expres-
sion in the in vitro experiment was low, 23% compared with intact
mRNA, due to RNA cleavage by Bhc-mediated caging. Furthermore,
this approach required cumbersome processes after in vitro tran-
scription, such as introduction of Bhc group and purification. More-
over, this strategy allowed for only a single off-to-on regulation
event because the uncaged mRNA produced upon photoirradiation
could not be re-caged; once the Bhc group was removed by phot-
oirradiation, the uncaged mRNA continued to be translated until it
was degraded by nucleases. These problems prohibited control
over the magnitude and period of gene expression. Additional ef-
forts are, therefore, needed to develop reversible photoregulation
methods that allow precise control over ‘when, where, and how
long (or how frequently)’ gene expression occurs. Here, we report
a reversible method for photoregulating translation using the cis–
trans photoisomerization of a photoresponsive 50-cap. Recently,
we developed several photochromic nucleosides (PCNs) that could
reversibly change their photochemical and physical properties
upon cis–trans photoisomerization under external photostimulus.6
In our strategy, we focused attention on the translation initiation
mechanism. Translation begins with binding between 7-methylgu-
anosine (50-cap) at the 50-end of mRNA and the eukaryotic initia-
tion factor 4E (eIF4E).7 Several eukaryotic initiation factors then
associate continuously and cooperatively. Finally, the ribosome
assembles onto the mRNA complex and translation begins. Trans-
lation efficiency depends on the affinity of interaction between
the 50-cap and eIF4E.8 Therefore, if we substitute the photorespon-
sive 50-cap (7-methylated PCN) for the native 50-cap, translation
can be reversibly controlled by photoregulation of the interaction
between the mRNA and eIF4E via cis–trans photoisomerization of
the photoresponsive 50-cap (Fig. 1).
We designed three photoresponsive 50-caps, 8-styryl cap
(8ST-cap), 8-naphthylvinyl cap (8NV-cap), and 8-fluorenylvinyl
cap (8FV-cap) such that the affinity for eIF4E could be modulated
by steric hindrance. The photoresponsive 50-caps were synthesized
⇑
Corresponding author. Tel.: +81 48 467 9312; fax: +81 48 462 4658.
E-mail address: o_shinji@riken.jp (S. Ogasawara).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.06.119

5458
S. Ogasawara, M. Maeda / Bioorg. Med. Chem. Lett. 21 (2011) 5457–5459
Figure 1. Reversible photoregulation of the translation by cis–trans photoisomerization of photoresponsive 50-cap.
according to Scheme 1. The starting compounds, PCNs (1a–c), were
prepared as described previously.6a PCN was converted to the cor-
responding 50-monophosphate by stirring PCN with phosphorus
oxychloride and trimethyl phosphate at 2 °C. After quenching with
1 M TEAB, the precipitate was filtered. Treatment of the residue
with methyl iodide in DMSO afforded the corresponding N7-meth-
ylated 50-monophosphate. The methylated 50-monophosphate was
converted to the corresponding imidazolide salt in the presence of
imidazole, triphenylphosphine, triethylamine, and aldrithiol. Final-
ly, the coupling reaction between the resulting imidazolide salt
and GDP in the presence of the zinc chloride catalyst gave the cor-
responding photoresponsive 50-cap (2a–c). Pure 8ST-cap, 8NV-cap
trans = 1:99, 8:92 and 2:8, respectively, as determined by HPLC
peak area). The structures of the photoresponsive 50-caps were
characterized by 1H and 31P NMR and mass spectroscopy.
As reported previously,6a PCNs show rapid and highly efficient
reversible cis–trans photoisomerization upon alternate irradiation
with monochromatic light. In these PCNs, both cis and trans iso-
mers were thermally stable. They showed no thermal isomeriza-
tion even at 80 °C for 20 h. Methylation at the N7 position of PCN
produced only small changes in the photoisomerization properties.
Photoisomerization of the photoresponsive 50-caps was performed
at 25 °C by using a 300 W Xenon lamp (MAX-302, Asahi spectra
Co., Ltd). The reaction mixture was held from the light source to
2 cm and irradiated with 410 nm (103 mW/cm2) for 2 min for cis
to trans photoisomerization and 310 nm (81 mW/cm2) for 2 min
for trans to cis photoisomerization. The trans forms of 8ST-cap,
8NV-cap, and 8FV-cap were photoisomerized to the cis forms by
irradiation at 410 nm, with 89%, 88%, and 95% conversion, respec-
tively, as determined by HPLC peak area. Subsequent irradiation at
310 nm yielded the trans forms with 64%, 81%, and 92% conversion,
respectively. To estimate the capping efficiency, short capped RNA
(50-cap-GGGAGA-30) was synthesized using the MAXIscriptTM kit
(Ambion, Inc.). 8ST-cap produced optimum efficiency as compared
to the normal-cap. However, the efficiency decreased as the size of
R-moiety become large. Introduction of the photoresponsive 50-cap
to the 50-end of the green fluorescence protein (GFP) mRNA was
accomplished by in vitro transcription using the MEGAscript kit
(Ambion, Inc.). The reaction solution, containing four ribonucleo-
tide triphosphates, the photoresponsive 50-cap, and T7 RNA poly-
merase, was incubated at 37 °C for 4 h in the presence of the
DNA template prepared by PCR amplification of the plasmid
DNA, which contained a 30-base poly(A) tail and a T7 promoter
site. After DNase digestion of the DNA template, the transcribed
products with 50-caps were purified using the MEGAclearTM kit
(Ambion, Inc.).
and 8FV-cap were obtained as
a mixture of isomers (cis:-
The difference in mRNA translational efficiency for mRNA con-
taining either the trans or cis forms of the 8ST-cap, 8NV-cap, or
8FV-cap, was investigated by in vitro translation of the capped
GFP-mRNA using Transdirect insect cells (Shimadzu, Co.). Transla-
tion of mRNA was more efficient for photoresponsive 50-caps in the
cis form than in the trans form, as shown in Figure 2. From the
molecular modeling, this effect was most likely due to different de-
grees of steric hindrance between the R-moiety and the active site
Scheme 1. Reagents and conditions: (a) POCl3, (OMe)3P, 2 °C, 16 h; (b) MeI, DMSO,
20 h; (c) imidazole, aldrithiol, PPh3, Et3N, DMF, 15 h; (d) GDP (TEA-salt), ZnCl2, DMF,
70 h.

S. Ogasawara, M. Maeda / Bioorg. Med. Chem. Lett. 21 (2011) 5457–5459
5459
Figure 3. The time course of fluorescence intensity of GFP (excitation: 480 nm,
emission: 509 nm). Blue line: normal-capped mRNA, red line: 8NV-capped mRNA.
The reaction mixtures were irradiated with 310 nm for 2 min at 0 min and 130 min
time points, and with 410 nm for 2 min at 60 min time point.
Figure 2. Relative fluorescence intensity of translational GFP after 10 h from
photoirradiation (excitation: 480 nm, emission: 509 nm). The reaction mixture was
irradiated with 410 nm for 2 min for cis to trans photoisomerization and 310 nm for
2 min for trans to cis photoisomerization at room temperature. Inset: fluorescence
image of GFP translated from 8NV-capped mRNA.
state (ON state) and a non-translating state (OFF state) in a revers-
ible fashion by alternately irradiating with monochromatic 410 nm
or 310 nm light. Therefore, our method enable us to control ‘when,
where, and how long (or how frequently)’ gene expression occurs.
For the application to in vivo photoregulation of important biolog-
ical events, introduction of additional techniques such as multi-
photon exitation and photochromic nucleobases photoisomerized
by longer wavelength will be needed and is now in progress.6
of eIF4E (see Supplementary data). Although the cis forms of the
photoresponsive 50-caps could interact with eIF4E, the bulky sub-
stituents, such as fluorene in the 8FV-cap, introduced steric hin-
drance in the cavity of eIF4E, even in the cis form. Thus,
translation was suppressed regardless of the isomer adopted.
Translation efficiency of 8ST-capped mRNA in cis form was 2.2
times higher than the normal-capped mRNA. This is probably
because photoresponsive 50-caps cannot be incorporated in
the reverse orientation as anti-reverse cap analogs (ARCAs).9 The
8NV-cap exhibited the highest photomodulation efficiency. The
8NV-cap in the trans form completely inhibited translation of
mRNA, whereas the cis form yielded GFP translation with the same
efficiency as mRNA capped with the normal-cap, a 26-fold higher
efficiency than that of the trans form.
Acknowledgments
The present work was supported by Precursory Research for
Embryonic Science and Technology (PRESTO), Japan Science and
Technology Agency (JST).
Supplementary data
Next, we reversibly photoregulated translation using the 8NV-
capped mRNA. Figure 3 shows the time course of fluorescence
intensity during in vitro translation. The reaction solution contain-
ing mRNA capped with the normal cap or the 8NV-cap was irradi-
ated with 310 nm light for 2 min, at the 0 min and 130 min time
points, and with 410 nm for 2 min at the 60 min time point. The
fluorescence intensity of the sample containing normal-capped
mRNA increased at a constant rate irrespective of photoirradiation.
In contrast, translation of 8NV-capped mRNA was not observed be-
tween 0 min and 60 min, during which time the 8NV-cap was in
the trans form due to illumination with 310 nm at 0 min (OFF
state). After photoisomerization to the cis form by 410 nm illumi-
nation, the fluorescence intensity gradually increased, indicating
that the mRNA was translated (ON state). Subsequent illumination
with 310 nm light displayed a low increase rate of fluorescence and
no increase finally, suggesting that translation was again inhibited
(OFF state). The increase in fluorescence observed in the second
OFF state derived from GFP synthesized before 310 nm illumina-
tion at 130 min time point because maturation of GFP need tens
of minutes.10
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2011.06.119.
References and notes
1. Driever, W.; Volhard, C. N. Cell 1988, 54, 83.
2. Hirata, H.; Bessho, Y.; Kokubu, H.; Masamizu, Y.; Yamada, S.; Lewis, J.;
Kageyama, R. Nat. Genet. 2004, 36, 750.
3. Kobayashi, T.; Mizuno, H.; Imayashi, I.; Furusawa, C.; Shirahige, K.; Kageyama,
R. Gene Dev. 2009, 23, 1870.
4. (a) Tang, X.; Swaminathan, J.; Gewirtz, A. M.; Dmochowski, I. J. Nucleic Acids Res.
2008, 36, 559; (b) Tang, X.; Maegawa, S.; Weinberg, E. S.; Dmochowski, I. J. J.
Am. Chem. Soc. 2007, 129, 11000; (c) Tang, X.; Dmochowski, I. J. Angew. Chem.,
Int. Ed. 2006, 45, 3523; (d) Tang, X.; Dmochowski, I. J. Nat. Protocols 2006, 1,
3041; (e) Deiters, A.; Garner, R. A.; Lusic, H.; Govan, J. M.; Dush, M.; Yoder, N. M.
N.; Yoder, J. A. J. Am. Chem. Soc. 2010, 132, 15644; (f) Shestopalov, I. A.; Sinha, S.;
Chen, J. K. Nat. Chem. Biol. 2007, 3, 650.
5. (a) Ando, H.; Furuta, T.; Tsien, R. Y.; Okamoto, H. Nat. Genet. 2001, 28, 317; (b)
Ando, H.; Okamoto, H. Methods Cell Sci. 2003, 25, 25; (c) Ando, H.; Furuta, T.;
Okamoto, H. Methods Cell Biol. 2004, 77, 159; (d) Ando, H.; Kobayashi, M.;
Tsubokawa, T.; Uyemura, K.; Furuta, T.; Okamoto, H. Dev. Biol. 2005, 287, 456.
6. (a) Ogasawara, S.; Maeda, M. Angew. Chem., Int. Ed. 2008, 47, 8839; (b)
Ogasawara, S.; Maeda, M. Angew. Chem., Int. Ed. 2009, 48, 6671; (c) Ogasawara,
S.; Ito, S.; Miyasaka, H.; Maeda, M. Chem. Lett. 2010, 39, 956.
7. Marcotrigiano, J.; Gingras, A. C.; Sonenberg, N.; Burley, S. K. Cell 1997, 89, 951.
8. (a) Kore, A. R.; Shanmugasundaram, M.; Charlres, I.; Alassov, A. V.; Barta, T. J. J.
Am. Chem. Soc. 2009, 131, 6364; (b) Nogalska, E. G.; Jemielity, J.; Kowalska, J.;
Darzynkiewicz, E.; Rhoads, R. E. RNA 2007, 13, 1745; (c) Grudzien, E.; Stepinski,
J.; Anyszka, M. J.; Stolarski, R.; Darzynkiewicz, E.; Rhoads, R. E. RNA 2004, 10,
1479.
In summary, we synthesized three photoresponsive 50-caps and
successfully developed a method for the reversible photoregula-
tion of translation using the cis–trans photoisomerization of the
photoresponsive 50-cap. The photoresponsive 50-caps showed rapid
and highly efficient reversible cis–trans photoisomerization upon
illumination at specific wavelengths. The mRNA containing the
8NV-cap at the 50-end could be switched between a translating
9. Stepinski, J.; Waddell, C.; Stolarski, R.; Darzynkiewicz, E.; Rhoada, R. E. RNA
2001, 7, 1486.
10. Tsien, R. Y. Ann. Rev. Biochem. 1998, 67, 509.