Dual-color control of nucleotide polymerization sensed by a fluorescence actuator

Article abstract of DOI:10.1039/c3pp50438g

Spatial and temporal control of molecular mechanisms can be achieved using photolabile bonds that connect biomolecules to protective caging groups, which can be cleaved upon irradiation of a specific wavelength, releasing the biomolecule ready-to-use. Here we apply and improve a previously reported strategy to tightly control in vitro transcription reactions. The strategy involves two caging molecules that block both ATP and GTP nucleotides. Additionally, we designed a molecular beacon complementary to the synthesized mRNA to infer its presence through a light signal. Upon release of both nucleotides through a specific monochromatic light (390 and 325 nm) we attain a light signal indicative of a successful in vitro transcription reaction. Similarly, in the absence of irradiation, no intense fluorescence signal was obtained. We believe this strategy could further be applied to DNA synthesis or the development of logic gates. This journal is

Full text of DOI:10.1039/c3pp50438g

Photochemical &
Photobiological Sciences
View Article Online
View Journal | View Issue
PAPER
Dual-color control of nucleotide polymerization
sensed by a fluorescence actuator†
Cite this: Photochem. Photobiol. Sci.,
014, 13, 751
2
a
a,b
a
Madalena M. Reimão-Pinto,‡ Ana Cordeiro,‡ Carina Almeida,
a,b
b
b
a
André V. Pinheiro, Artur Moro, João C. Lima* and Pedro V. Baptista*
Spatial and temporal control of molecular mechanisms can be achieved using photolabile bonds that
connect biomolecules to protective caging groups, which can be cleaved upon irradiation of a specific
wavelength, releasing the biomolecule ready-to-use. Here we apply and improve a previously reported
strategy to tightly control in vitro transcription reactions. The strategy involves two caging molecules that
block both ATP and GTP nucleotides. Additionally, we designed a molecular beacon complementary to
the synthesized mRNA to infer its presence through a light signal. Upon release of both nucleotides
through a specific monochromatic light (390 and 325 nm) we attain a light signal indicative of a successful
in vitro transcription reaction. Similarly, in the absence of irradiation, no intense fluorescence signal was
obtained. We believe this strategy could further be applied to DNA synthesis or the development of logic
gates.
Received 23rd December 2013,
Accepted 10th February 2014
DOI: 10.1039/c3pp50438g
www.rsc.org/pps
relatively high extinction coefficients and photochemical
quantum yields. They also absorb in the visible region
1
. Introduction
Caged molecules are photosensitive molecules that have (<420 nm) and have the advantage of fast photocleavage kine-
6
,7
shown great potential for the study and control of molecular tics. We have previously reported the use of caged nucleo-
5
mechanisms involved in cellular and physiological processes. tides for controlled polymerization of nucleic acids. The
Caging can be achieved by linking a biomolecule of interest system is based on ATP caged with [7-(diethylamino)coumarin-
through a photolabile bond to a protecting group, thus render- 4-yl]methyl (DEACM), which can be released by light
1
ing it temporarily inactive. Upon specific wavelength irradiation and, thus, allowing for the controlled triggering of
irradiation, release of the bioactive molecule occurs, allowing in vitro RNA polymerization. The earlier developments of cou-
for a time-controlled burst in concentration with tight spatial marin derivatives to cage ATP were achieved by Geibler and co-
8
control of the biological process under study, without sub- workers in 2003. Here, we take one step further and introduce
sequent addition of mass to the system. Many biological appli- GTP caged with (7-methoxycoumarin-4-yl)methyl (MCM) for
cations for caging molecules have been proposed, such as the development of a dual color setup that allows for the
caged calcium and caged ATP to control physiological pro- control of integration of either nucleotide: caging renders both
cesses, caged neurotransmitters like glutamate and caged nucleotides biologically inactive and release of either ATP or
2
nucleic acids to regulate gene expression. In this way it is GTP (or both simultaneously) is attained via irradiation using
possible to switch biological processes on/off via a light/ a monochromatic light pulse of the respective wavelength
3
,4
irradiation input. However, the choice of the photoremova- (Fig. 1). Once the free nucleotides are present in the reaction
ble protecting group must fulfill several requirements mainly mixture, RNA polymerization can proceed. Additionally, we
related to the molecule concentration and velocity of photo- have introduced a molecular beacon (MB) to detect the for-
1
,5
9
deprotection.
Coumarin derivatives stand out, presenting mation of the RNA product. MBs have been widely used as
1
0,11
nucleic acid reporters.
The MB consists of a single-
stranded oligonucleotide probe labeled with a fluorophore (5′-
FAM) at one extremity and a quencher molecule (3′-DABCYL)
at the other end. In the absence of a complementary sequence,
the MB assumes a stem–loop conformation, leading to an
energy transfer from the excited fluorophore to the quencher,
and no fluorescence signal is observed. However, upon hybrid-
ization to the complementary RNA target, the stem–loop con-
formation is disrupted, which in turn causes the fluorophore
a
CIGMH, Departamento Ciências da Vida, Faculdade de Ciências e Tecnologia,
Universidade Nova de Lisboa, Campus de Caparica, 2829-516 Caparica, Portugal.
E-mail: pmvb@fct.unl.pt; Tel: +351 212948530
b
REQUIMTE, Departamento de Química, Faculdade de Ciências e Tecnologia,
Universidade Nova de Lisboa, Campus de Caparica, 2829-516 Caparica, Portugal.
E-mail: lima@fct.unl.pt
†
Electronic supplementary information (ESI) available. See DOI: 10.1039/
c3pp50438g
These authors contributed equally to the work.
‡
This journal is © The Royal Society of Chemistry and Owner Societies 2014
Photochem. Photobiol. Sci., 2014, 13, 751–756 | 751

View Article Online
Paper
Photochemical & Photobiological Sciences
Fig. 1 Light induced release of ATP and GTP from its precursors DEACM-ATP and MCM-GTP. (A) DEACM-ATP and MCM-GTP structure photolysis.
Upon irradiation at specific wavelengths (390 nm for DEACM-ATP and 325 nm for MCM-GTP), the coumarin–nucleotide ester bond is cleaved and
free ribonucleotides are produced. (B) Representation of dual color ribonucleotide activation prior to in vitro transcription. A mixture of two caging
molecules is subjected to light irradiation at 390 (yellow) and 325 nm (purple), with subsequent release of nucleotides, available for downstream
applications. The absorption spectra of both coumarins, under the experiment conditions, are represented on the right (yellow and purple areas
highlight irradiation bandwidth).
and the quencher to be moved apart from each other, increas- either DEACM-ATP or MCM-GTP, and the other three remain-
1
2,13
ing the fluorescence signal.
As such, transcription is trans- ing free nucleotides were irradiated at the selected wavelengths
lated into a fluorescent light signal. This work represents one (390 and 325 nm, respectively). The release of ATP and GTP
more step towards controlling nucleic acid synthesis.
upon irradiation (in each case), and concomitant activation of
transcription, was detected by measuring the fluorescence
intensity resulting from the hybridization of the MB with the
complementary target (see Fig. 2). The MB performance and
behavior were fully characterized and calibrated for optimal
sequence identification and signal output (Fig. S1 and S2†)
using negative and positive controls of transcription. We
also set up a fluorescence intensity threshold (approximately
2
. Results
2
.1. Photochemical characterization
Both caging molecules selected for this study were fully charac-
terized prior to use. The photochemical properties of DEACM-
5
,14
ATP have been described elsewhere.
The absorption spec-
25 a.u.) limited by the non-irradiated control above which we
trum of MCM-GTP is characterized by three bands, with
maxima at 250 nm, 268 nm and 325 nm (Fig. 1B). The band
with the maximum at 268 nm corresponds to the guanine ring
absorption, while the other two correspond to MCM ring
absorption. The DEACM-ATP (λmax = 390 nm, εmax = 15 000
consider that the reaction provided a full length mRNA.
Results show the detection of a clear fluorescence signal in
both cases where ATP and GTP were substituted respectively
with irradiated DEACM-ATP and MCM-GTP, indicating the pro-
duction of a full length RNA transcript (Fig. 3). However, there
is residual fluorescence signal of the MB even when there is no
irradiation of the coumarin. Moreover, the fluorescence signal
from the transcription reaction where the nucleotides were
activated by light was lower than the positive control where all
nucleotides were free in solution. The obtained fluorescence
results are supported by agarose gel electrophoresis.
−1
−1
M
1
cm , Φchem = 0.0125) and MCM-GTP (λmax = 325 nm, εmax
=
3 500 M− cm−1, Φchem = 0.0100) solutions in 10 mM phos-
1
phate buffer (pH 8.0) present absorption maxima in different
regions (Fig. 1B), therefore 390 nm light was used to excite the
DEACM caging group in the DEACM-ATP molecule to release
specifically the nucleotide functionalized to it. Likewise, for
MCM-GTP, 325 nm light was chosen, leading to the formation
of GTP and MCM-OH (Fig. 1A).
2.3. Dual color activation of transcription
2
.2. Molecular beacon detection of a RNA transcript
Following demonstration that both coumarins, when used
To ensure maximum efficiency of the caging and photorelease, independently, can break down and release ready-to-use
different reactions containing solely one of the caging agents, nucleotides, we tested their performance when put together in
752 | Photochem. Photobiol. Sci., 2014, 13, 751–756
This journal is © The Royal Society of Chemistry and Owner Societies 2014

View Article Online
Photochemical & Photobiological Sciences
Paper
Fig. 2 Representation of a molecular beacon target detection. Upon hybridization to the complementary target (mRNA transcript), the molecular
beacon hairpin structure opens and binds to the target, resulting in an intense fluorescence signal. In the absence of target (no transcript), the mole-
cular beacon maintains its hairpin conformation bringing the fluorophore (left) and quencher (right) in close proximity resulting in a residual emission
signal.
Fig. 3 Fluorescence actuator for sensing light-controlled transcription. (A) Irradiation of transcription reactions at specific wavelengths containing
either DEACM-ATP or MCM-GTP and the three remaining free NTPs, followed by addition of the MB, results in a clear increase of fluorescence
signal. Dashed line represents the established threshold for the absence of transcript. (B) The fluorescence signal correlates to the generation of a
full length RNA product.
the same mixture. First we observed whether the small spectral Data depicted in Fig. 2 show that 50 and 37.5 µM of DEAC-
absorption overlap between DEACM and MCM at 325 nm M-ATP are sufficient to produce mRNA visible by agarose gel
would produce significant ATP leakage (Fig. 4). We fixed the electrophoresis, thus confirming the existence of a small
MCM-GTP concentration at 75 µM and varied DEACM-ATP. leakage effect at 325 nm. Considering these results, we
Fig. 4 Effect of DEACM-ATP leakage at 325 nm. Agarose gel electrophoresis of T7 RNA polymerase in vitro transcription reaction with DEACM-ATP
and MCM-GTP molecules. Solutions were irradiated at 325 nm. Extra time of exposure for image acquisition was used to improve visualization of the
leakage effect.
This journal is © The Royal Society of Chemistry and Owner Societies 2014
Photochem. Photobiol. Sci., 2014, 13, 751–756 | 753

View Article Online
Paper
Photochemical & Photobiological Sciences
designed a MB that in the presence of its complementary
target (mRNA) adopts an open conformation and fluorescence
is detected; in the absence of transcription, the MB retains the
native closed hairpin conformation, keeping the fluorophore
in close proximity with the quencher and no fluorescence is
observed – Fig. 2 (see also ESI Fig. S2†).
When DEACM-ATP and MCM-GTP are used as caging
agents there is an increase in MB fluorescence that can be cor-
related with a successful transcription reaction (Fig. 3).
Despite the efficiency of this approach, fluorescence is lower
than that attained when the four free nucleotides are directly
added to the reaction. The inhibitory effect of coumarin–
alcohol derivatives may be responsible for hampering polymer-
5
,15
ization, as previously described.
The DEACM-OH photolysis
by-product formed during DEACM-ATP irradiation has been
Fig. 5 Control of transcription via light-input/RNA-output “AND” logic shown to interfere with transcription, due to its poor solubility
gate. DEACM-ATP is used as a source of ATP and MCM-GTP as a source in water and consequent partition into T7 RNA polymerase,
of GTP. Only in the presence of both wavelength light inputs (390 and
25 nm) formation of mRNA product occurs.
5
inhibiting its activity. To assess whether in vitro transcription
3
reaction was inhibited by the MCM-GTP photolysis by-product,
increasing MCM-OH concentrations were used in samples con-
taining a mixture of the four natural nucleotides. MCM-OH
concentrations above 100 μM eventually lead to decreasing
transcription yields (Fig. S3†). The inhibition of transcription
by the photolysis by-product limits the concentration range of
the MCM-caged nucleotide that can be used for light-con-
trolled in vitro transcription reactions. However one should
also consider that not all of the irradiated molecules were able
to generate free triphosphate nucleotides ready to be incorpor-
ated by RNA polymerase, which resulted in a lesser amount of
mRNA synthesis for the irradiated sample and therefore a
lower MB fluorescence signal. By establishing a threshold from
which we consider that there is an effective transcription
product, we can compare results from independent
experiments.
performed the dual color activation of transcription where a
mixture of both coumarins was irradiated at the selected wave-
lengths. We denote that, in general, whenever a mixture con-
taining irradiated DEACM-ATP and MCM-GTP is used for the
transcription reaction (Fig. 5), the resulting fluorescence signal
is diminished, regarding the use of only one coumarin (data in
Fig. 3). The established threshold made this comparison poss-
ible between experiments. Taking a closer look at Fig. 5, we
can see an increase in MB fluorescence intensity when the cou-
marin mixture was irradiated at both wavelengths, versus
single wavelength irradiation at either 325 or 390 nm. Conver-
sely it still presents a higher signal than the fluorescence
threshold, establishing the formation of mRNA during in vitro
transcription. Once more we can correlate the fluorescence
results with the ones obtained in agarose gel electrophoresis.
Following an experimental design where ATP and GTP are
replaced respectively by DEACM-ATP and MCM-GTP within a
regular NTP pool, both 390 nm and 325 nm light irradiation
are needed for the release of both missing ribonucleotides and
for RNA transcription to resume – dual-color system. However,
3
. Discussion
The possibility of controlling RNA polymerization by means of when 325 nm light alone is used as excitation, due to the
direct light irradiation of caged-ATP for the release of the func- residual photochemistry of the DEACM-ATP derivative (circa
tional nucleotide moiety was previously demonstrated by our 15% of the maximum absorption), there is an ATP leakage
5
group. Here, we have taken one step further in what the effect (Fig. 4). Therefore, MCM-GTP relative concentration was
control of RNA polymerization by light concerns – in addition increased in order for excitation light competition to occur
to DEACM-ATP we simultaneously used MGM-GTP, thus and DEACM-caged nucleotide cleavage to be reduced upon
having the possibility to control the release of two substrates 325 nm irradiation, thus minimizing transcription leakage in
required by the enzyme (see Fig. 1A). The choice of caging moi- the absence of the second input. For the non-irradiated
eties (DEACM and MCM) was based on minimizing spectral sample, no RNA product was observed, indicating efficient
overlap while maintaining similar chemical structure and DEACM and MCM caging of ATP and GTP, respectively, render-
photorelease behavior, so as to reduce bias between the two ing the ribonucleotides biologically inert and impairing tran-
nucleotides being released. In this way, by irradiating a RNA scription (Fig. 5). When only 390 nm light irradiation is
polymerization mixture without enzyme at 390 nm, one is applied, only ATP, but no GTP, is released, hence no transcrip-
capable of releasing functional ATP into the mixture, whereas tion product is synthesized. If both wavelengths are used to
irradiation at 325 nm will release GTP (Fig. 1B).
irradiate the sample, both ATP and GTP are released and tran-
To visualize RNA production directly in solution and scription resumes. When compared to the usage of a single
without the need for further sample manipulation, we caging molecule, a lower increase in fluorescence is denoted,
754 | Photochem. Photobiol. Sci., 2014, 13, 751–756
This journal is © The Royal Society of Chemistry and Owner Societies 2014

View Article Online
Photochemical & Photobiological Sciences
Paper
which may be due to joint inhibitory effect of both coumarin– MCM-GTP = 0.0100, DEACM-ATP = 0.0025. For irradiation with
alcohol derivatives formed after irradiation. 390 nm light, 10 mM phosphate buffer: DEACM-ATP = 0.0125.
4.3. Dual-color in vitro transcription
4
. Experimental section
Standard in vitro transcription using 250 ng of template was
performed with 20 U of T7 RNA polymerase according to the
manufacturer’s protocol. Reactions were incubated for 1 h at
All chemicals were purchased from Sigma-Aldrich (St. Louis,
MO, USA), were of the highest purity available and used
without further purification. T7 RNA polymerase was pur-
chased from Fermentas (Vilnius, Lithuania). All oligonucleo-
tides were purchased from STAB Vida (Caparica, Portugal).
37 °C, followed by heat inactivation of the enzyme for 15 min
at 75 °C. In assays involving caged-ATP and caged-GTP, ATP
was substituted by 50 µM of DEACM-ATP and GTP was substi-
tuted by 75 µM of MCM-GTP. Coumarin-derivatives were irra-
diated prior to addition to the reaction mixture. For irradiation
experiments, the remaining free triphosphate nucleotides were
present at concentrations of 100 µM each. Products were ana-
lyzed on 3% agarose gel electrophoresis in 1 × TBE buffer.
Results were visualized in a Gel Doc XR+ Molecular Imager
system (Bio-Rad, USA) following staining with GelRed
4
.1. MCM-GTP synthesis and characterization
DEACM-ATP synthesis and purification is described else-
where. [7-Methoxycoumarin-4-yl]methyl guanine 5′-triphos-
phate (MCM-GTP) was synthesized following the same
procedure, but using [7-methoxycoumarin-4-yl]phosphate
5
(MCM-P) and GDP as starting materials. A Hitachi-Merck
(
Biotium, USA). Transcription template cloning and purifi-
5
HPLC L6200A Pump with an L-4500 Diode Array Detector
using a polystyrene-divinylbenzene (PLRP-S, Polymer Labs)
cation is described elsewhere.
semi-preparative column (7.4 mm × 15 mm, 8 μm, 300 Å) was 4.4. DEACM-ATP leakage at 325 nm and MCM-OH inhibition
employed for separation and purification of MCM-GTP. Eluent
DEACM-ATP leakage assay was performed using standard
A was triethylammonium acetate buffer in water, 20 mM, pH
in vitro transcription conditions, 100 µM of free CTP and UTP,
6
.9; eluent B was methanol. The gradient started with 25% of
75 µM of MCM-GTP and increasing concentrations of DEAC-
B in A and isocratic separation after 4 min, with an increase to
00% B after 5 min, and finished after 16 min at 100% of
M-ATP (12.5, 25, 37.5, 50 µM). Prior to the addition of T7 RNA
polymerase the solutions were irradiated at 325 nm for 1 hour.
Products were analyzed on 3% agarose gel electrophoresis in 1
1
−
1
B. Separations were run at a flow rate of 0.9 mL min and the
column temperature was 35 °C. A final MCM-GTP yield of circa
×
TBE buffer. For MCM-OH inhibition assay, the super-satu-
2
0% was estimated. After peak separation and collection,
rated 868 μM MCM-OH water solution was prepared by dissol-
ving pure MCM-OH powder (1.79 mg) in 10 mL acetone, and
evaporated under vacuum forming a thin film on the volu-
metric flask walls. Hot sterile water (10 mL) was added to the
flask and the solution was put in an ultra-sound bath for
samples were lyophilized, resuspended in water and stored in
the dark at −20 °C. A purity of >95% was determined by HPLC.
All solutions were protected from light and manipulations
were made under red-light illumination.
1
0 minutes. No precipitation was observed after cooling to
4
.2. Irradiation assays
room temperature.
All spectroscopic measurements and irradiation were per-
formed in a 60 µL quartz fluorescence cuvette (10.00 mm 4.5. Molecular beacon for detection of transcription
optical path) at 21 °C. The DEACM-ATP absorption spectrum
was acquired with 1.5 nm slit bandwidth for excitation and
emission, with correction files. The MCM-GTP spectrum was
collected with 2.5 nm slit bandwidth for excitation and 5.0 nm
Exon 7 of the human p53 gene was used. It is labelled with a
bandwidth for emission, with correction files. Coumarin
In order to rapidly evaluate the efficiency of the reaction, a 25
nt DNA molecular beacon (FAM-5′-TCGATAACAGTTCCTGC-
ATGATCGA-3′-D) complementary to the RNA transcript of the
6-carboxyfluorescein (FAM) at the 5′ end and with [4-((4-(di-
derivatives irradiation were carried out on a Spex Fluorolog
.22 m spectrometer with a 150 W xenon arc lamp, monochro-
methylamino)phenyl)azo)benzoic acid] (DABCYL) at the 3′ end.
For hybridization assays, 1.5 µM of molecular beacon was added
to the reaction mix before 1 hour incubation at 37 °C. Then, the
reaction mixture was brought up to a total volume of 50 µL with
DEPC treated water. Fluorescence measurements were per-
formed at 4 °C, to ensure a closed conformation in the hairpin
region by default, using 490 nm as the excitation wavelength.
0
mated for the desired excitation wavelength (35 min of
irradiation at 390 nm, 4.5 nm slit bandwidth for DEACM-ATP;
3
5 min of irradiation at 325 nm, 9 nm bandwidth for
MCM-GTP). The actinometry of the irradiation setup was per-
formed with the concentrated potassium ferrioxalate actin-
1
6
−8
−8
ometer and intensities of 9.5 × 10 and 4.2 × 10 Einstein
−
1
min were measured, for 325 nm and 390 nm monochromatic
irradiation, respectively. Photochemical quantum yields for
the formation of alcohol were calculated both for MCM-GTP
5
. Conclusions
and DEACM-ATP as described by Pinheiro et al., 2008, which We developed a universally applicable light controlled binary
in turn translate to quantum yields of nucleotide release. For system as a proof-of-concept for multi-color selective release of
irradiation with 325 nm light, 10 mM phosphate buffer: nucleotides. Light allows activation of a biologically active
This journal is © The Royal Society of Chemistry and Owner Societies 2014
Photochem. Photobiol. Sci., 2014, 13, 751–756 | 755

View Article Online
Paper
Photochemical & Photobiological Sciences
molecule with defined temporal accuracy without the need for
further substrate addition into the system. Ultimately, having
each of the four nucleotides functionalized with a suitable
caging group with different non-overlapping absorption pro-
files would allow for the construction of a four letter code
instead of a binary one. By coupling the output to a light
signal through the use of a MB, we are able to immediately
detect and quantify the target molecule without the need for
separation of the hybridized MB from the unbound probe in
excess. By using light as a trigger, we demonstrate a new uni-
versally applicable means to surpass the necessity of continu-
ously adding mass to the system. Furthermore, this approach
could be extendable to light-controlled DNA synthesis, another
attractive biomolecule for devising computational networks
3 A. Deiters, ChemBioChem, 2010, 11, 47–53.
4 H. Lee, D. R. Larson and D. S. Lawrence, ACS Chem. Biol.,
2009, 4, 409–427.
5 A. V. Pinheiro, P. Baptista and J. C. Lima, Nucleic Acids Res.,
2008, 36, e90.
6 R. Schmidt, D. Geiβler, V. Hagen and J. Bendig, J. Phys.
Chem. A, 2007, 111, 5768–5774.
7 T. Eckardt, V. Hagen, B. Schade, R. Schmidt, C. Schweitzer
and J. Bendig, J. Org. Chem., 2002, 67, 703–710.
8 D. Geibler, W. Kresse, B. Wiesner, J. Bendig,
H. Kettenmann and V. Hagen, ChemBioChem, 2003, 4, 162–
170.
9 W. Tan, K. Wang and T. J. Drake, Curr. Opin. Chem. Biol.,
2004, 8, 547–553.
and logic gates involved in functioning and regulatory mecha- 10 K. Wang, Z. Tang, C. J. Yang, Y. Kim, X. Fang,
nisms in living cells.
W. Li, Y. Wu, C. D. Medley, Z. Cao, J. Li, P. Colon,
H. Lin and W. Tan, Angew. Chem., Int. Ed., 2009, 48,
856–870.
1
1 H. W. Lim, S. H. Lee, K. A. Yang, S. I. Yoo, T. H. Park and
B. T. Zhang, BioSystems, 2013, 111, 11–17.
Acknowledgements
The authors acknowledge Fundação para a Ciência e Tecnolo- 12 K. Abravaya, J. Huff, R. Marshall, B. Merchant, C. Mullen,
gia (FCT/MEC) for funding: CIGMH (PEst-OE/SAU/UI0009/
011); REQUIMTE (PEst-C/EQB/LA0006/2011); PTDC/
G. Schneider and J. Robinson, Clin. Chem. Lab. Med., 2003,
41, 468–474.
2
BBB-NAN/1812/2012; PTDC/QUI-QUI/112597/2009; PTDC/ 13 A. P. Silverman and E. T. Kool, Trends Biotechnol., 2005, 23,
CTM-NAN/109877/2009; AM for SFRH/BPD/69210/2010.
225–230.
1
4 A. V. Pinheiro, A. J. Parola, P. V. Baptista and J. C. Lima,
J. Phys. Chem. A, 2010a, 114, 12795–12803.
5 A. V. Pinheiro, J. Conde, A. J. Parola, J. C. Lima and
P. V. Baptista, J. Photochem. Photobiol. A, 2010b, 213, 147–
151.
1
References
1
A. Specht, F. Bolze, Z. Omran, J. F. Nicoud and
M. Goeldner, HFSP J., 2009, 3, 255–264.
16 M. Montalti, A. Credi, L. Prodi and M. T. Gandolfi, Hand-
book of Photochemistry, CRC Press, Taylor and Francis
Group, Boca Raton, 3rd edn, 2006.
2
H. Yu, J. Li, D. Wu, Z. Qiu and Y. Zhang, Chem. Soc. Rev.,
2
009, 39, 464–473.
756 | Photochem. Photobiol. Sci., 2014, 13, 751–756
This journal is © The Royal Society of Chemistry and Owner Societies 2014