Novel photochrome aptamer switch assay (PHASA) for adaptive binding to aptamers

Article abstract of DOI:10.1007/s10895-014-1441-9

A novel Photochrome-Aptamer Switch Assay (PHASA) for the detection and quantification of small environmentally important molecules such as toxins, explosives, drugs and pollutants, which are difficult to detect using antibodies-based assays with high sensitivity and specificity, has been developed. The assay is based on the conjugation of a particular stilbene-analyte derivative to any aptamer of interest. A unique feature of the stilbene molecule is its reporting power via trans-cis photoisomerisation (from fluorescent trans-isomer to non-fluorescent cis-isomer) upon irradiation with the excitation light. The resulting fluorescence decay rate for the trans-isomer of the stilbene-analyte depends on viscosity and spatial freedom to rotate in the surrounding medium and can be used to indicate the presence of the analyte. Quantification of the assay is achieved by calibration of the fluorescence decay rate for the amount of the tested analyte. Two different formats of PHASA have been recently developed: direct conjugation and adaptive binding. New stilbene-maleimide derivatives used in the adaptive binding format have been prepared and characterised. They demonstrate effective binding to the model thiol compound and to the thiolated Malachite Green aptamer.

Full text of DOI:10.1007/s10895-014-1441-9

J Fluoresc
DOI 10.1007/s10895-014-1441-9
ORIGINAL PAPER
Novel Photochrome Aptamer Switch Assay (PHASA)
for Adaptive Binding to Aptamers
Vladislav Papper & Oleksandr Pokholenko &
Yuanyuan Wu & Yubin Zhou & Ping Jianfeng &
Terry W. J. Steele & Robert S. Marks
Received: 5 August 2014 /Accepted: 11 August 2014
#
Springer Science+Business Media New York 2014
. .
Keywords Stilbene Aptamer Fluorescence
.
Abstract A novel Photochrome-Aptamer Switch Assay
(PHASA) for the detection and quantification of small envi-
ronmentally important molecules such as toxins, explosives,
drugs and pollutants, which are difficult to detect using
antibodies-based assays with high sensitivity and specificity,
has been developed. The assay is based on the conjugation of
a particular stilbene-analyte derivative to any aptamer of in-
terest. A unique feature of the stilbene molecule is its reporting
power via trans-cis photoisomerisation (from fluorescent
trans-isomer to non-fluorescent cis-isomer) upon irradiation
with the excitation light. The resulting fluorescence decay rate
for the trans-isomer of the stilbene-analyte depends on vis-
cosity and spatial freedom to rotate in the surrounding medi-
um and can be used to indicate the presence of the analyte.
Quantification of the assay is achieved by calibration of the
fluorescence decay rate for the amount of the tested analyte.
Two different formats of PHASA have been recently devel-
oped: direct conjugation and adaptive binding. New stilbene-
maleimide derivatives used in the adaptive binding format
have been prepared and characterised. They demonstrate ef-
fective binding to the model thiol compound and to the
thiolated Malachite Green aptamer.
.
Photoisomerisation Maleimide
Abbreviations
2b
4-methoxy-4′-stilbene maleimide
2a
4-N,N′-dimethylamino-4′-stilbene maleimide
MG
Malachite green
DMSO Dimethyl sulfoxide
DMF
MCH
EX
Dimethyl formamide
6-mercaptohexanol
Fluorescence excitation
Fluorescence emission
EM
Introduction
As one of the simplest means of converting light into mechan-
ical motion on the angstrom scale, photo-induced trans-cis
isomerisation about double bonds has long been a subject of
thorough research. The mechanisms of the photoisomerisation
of 1,2-diphenylethenes, or stilbenes, and their derivatives have
been intensively discussed in literature [1–4]. The convention-
al picture of this process has been simplified as a two-state,
one-dimensional model, stressing torsional motion as the pri-
mary nuclear coordinate.
Vladislav Papper and Oleksandr Pokholenko contributed equally to this
manuscript.
So far, stilbene compounds have never been successfully
used as the molecular sensing entities in the biosensing de-
vices and bioassays, although they possess all the desired
features of the ideal reporter molecules, for instance, chemical
and biological stability, relatively low toxicity, synthetic avail-
ability, high photochemical sensitivity, rapid response and
easy regeneration. This is mainly because of their strong
tendency to rapidly lose fluorescence upon excitation via the
non-radiative deactivation process connected to the twisted
transition in the excited state [1–4]. The latter is responsible
:
:
:
:
:
V. Papper O. Pokholenko Y. Wu Y. Zhou P. Jianfeng
T. W. J. Steele (*)
School of Materials Science and Engineering, Nanyang
Technological University, Singapore 639798, Singapore
e-mail: wjsteele@ntu.edu.sg
R. S. Marks
Department of Biotechnology Engineering, National Institute for
Biotechnology in the Negev, Ilse Kats Institute for Nanoscale Science
and Technology, Ben Gurion University of the Negev,
Beer Sheva, Israel

J Fluoresc
for the trans-cis photoisomerisation of the molecule and acts
as a quenching funnel on fluorescence emission, thereby
depriving the fluorescence probe of its photochemical
stability.
compounds is very sensitive to medium microviscosity and
can be used in biomedical applications [5–8]. The new
fluorescence-photochrome method, developed by us, was
used for investigation of local medium effects and phase
transitions in biological membranes and on solid surfaces
[5–8]. The method is based on monitoring the trans-cis
photoisomerisation kinetics of stilbenes incorporated into an
object of interest. It has been shown that the apparent rate
constant of the trans-cis photoisomerisation in a relatively
viscous media, for instance, in biological membranes, is con-
trolled by the medium relaxation rate [6]. This method was
also applied to studies of the microviscosity effects on the
trans-cis photoisomerisation of the stilbene molecules
immobilized onto quartz plates coated with lysozyme [7, 8].
The apparent trans-cis photoisomerisation rate constant of the
process was found to be 3–4 times lower for the immobilised
stilbene molecule than for the same free molecule in solution.
That indicated that the surface and the protein sterically hin-
dered the twisting motion of the stilbene molecule in the
excited state.
Thus, as schematically pictured on Fig. 1a, many stilbene
compounds can rapidly change their molecular configuration
upon irradiation at the excitation maximum, if they are not
sterically and electronically restricted to do so. The trans-cis
photoisomerisation process is experimentally observed at the
constant-illumination conditions as the rapid steady-state fluo-
rescence decay to the photostationary equilibrium between the
two isomers. Such photochemical instability could be a huge
problem if the stilbene compound is considered a fluorescence
probe. And this is probably the reason why the stilbene
compounds have never been effectively used in biomedical
applications. However, the actual reporting power of the stil-
benes is not in their fluorescence, but in rapid loss of their
fluorescence via an instant conformational change upon irra-
diation with the excitation light. This makes the stilbene
switches unique in the sense that most fluorescent reporters
(labels or probes) either do not possess this intramolecular
switchable nature or require the separation of adjacent
fluorophores.
The latest studies on molecular dynamics of an antibody
binding site by using the above fluorescence-photochrome
method led to development of the novel fluorochrome immu-
noassay (FCIA), which is based on the conjugation of a
particular trans-stilbene derivative to an antibody of interest
[9, 10]. This trans-stilbene derivative, which is a photochrome
probe and analogue (hapten) to the native antigen of a
In view of the above, one of the most striking features of
stilbene photochemistry is its essentially strong dependence
on medium microviscosity. We have previously provided
clear evidence that the trans-cis photoisomerisation of stilbene
Fig. 1 Stilbene fluorescent
switch. a Stilbene exists in two
molecular configurations; the
fluorescent trans-isomer is
switched to the non-fluorescent
cis-isomer upon irradiation at the
excitation maximum. b Synthesis
of stilbene-maleimides and their
reaction with c) 6-
mercaptohexanol yielding the
conjugates 3a and 3b, and d
Malachite Green (MG) Aptamer,
A9 C6SH label yielding the
conjugate 4, MG Aptamer, A9
Stilbene

J Fluoresc
particular antibody, was completely prevented from the twist-
ing motion upon irradiation with the excitation light within the
antibody binding site. As a result, no change in fluorescence
intensity of the trans-stilbene molecule was observed, and that
was reported as almost zero fluorescence intensity decay with
time compared to that of the free trans-stilbene molecule
measured in solution [10].
commercially non-applicable. Antibodies, although employed
in many biosensor technologies today, have many essential
drawbacks: 1) limited shelf-lives in the dry or hydrated state,
2) denature in organic solvents and aqueous mixtures, 3)
display background fluorescence signals that lower signal-
to-noise ratios, 4) tend to have bulky molecular volumes
several orders of magnitude larger than the binding ligand of
interest, thereby limiting high density and high intensity fluo-
rescence measurements, 5) frequently fail to produce avid
antibodies to hapten sized analytes such as TNT, and 6) the
cross-reactivity for these small size targets is very high, lead-
ing to a lower specificity. Thus, due to their relatively big size,
it is difficult for antibodies to recognize small molecules with
high sensitivity and selectivity. Besides, antibodies suffer
from batch to batch variation in their structure and function-
ality and denaturation and their generation process involves
the use of animals. As a result, any bioassay or biosensing
system based on the FCIA for detection of a certain analyte
would differ in its sensitivity, and might be difficult to cali-
brate and make generic.
The aforementioned limitations of antibodies can be easily
overcome by replacing them with aptamers, which are single
strand DNA or RNA oligonucleotides. Aptamers are different
from antibodies, yet they mimic properties of antibodies in a
variety of diagnostic formats. They are attracting more and
more attention as a replacement for antibodies due to their
ease of manufacturing, strong affinity and excellent specificity
even for small molecular weight targets. The main advantage
of aptamers is their in-vitro selection process. By isolating
aptamers in-vitro, an aptamer can be produced for any target
molecule. Moreover, aptamers have very long shelf lives, can
be rapidly screened by the SELEX method for new designs,
and can be rapidly scaled for manufacturing. Aptamers are
relatively small in size, can be stabilised in organic/aqueous
solvents, and achieve high ligand-specificity. Therefore, they
are a more logical choice for bioassays than antibodies, in-
cluding the fact that higher surface densities can be placed in a
microtiter plate well-the higher the density, the greater the
sensitivity, especially when combined with metal enhanced
fluorescence. Thus, due to their small size, tuneable stability,
and high ligand-specificity, aptamers seem to be a more log-
ical choice than antibodies in our proposed assay, as it will be
described below.
The majority of commonly employed non-radioactive and
non-enzyme immunoassays are based on physical labelling.
The principle of biophysical labelling approach is based on
chemical modification of chosen sites of the object of interest
by special compounds (labels and probes) whose properties
make it possible to trace the state of the biological matrix by
appropriate physical methods. Several types of physical label-
ling immunoassays have been commonly used at present,
such as Free-Radical Assay Techniques (FRAT); Liposome
Immune Lysis Assay in a combination with FRAT (LILA-
FRAT); Fluorescence Polarisation Immunoassay (FPIA);
Fluorescence Resonance Transfer Immunoassay (FRTIA)
and LILA Fluorescence Immunoassay (LILA-FIA). The gen-
eral advantage of all these modern physical labelling immu-
noassays is monitoring the complex formation between anti-
gens and antibodies in solution without preliminary separa-
tion. Each method has however its limitations. FRAT being
monitored by ESR techniques requires relatively high concen-
tration of the tested compound. Direct luminescence methods,
such as FPIA and FRTIA, appear to be more sensitive but they
require solutions of low optical density and cannot be used in
biological or non-transparent media. FRAT and FPIA are not
practically applicable to antigens of higher molecular weight.
FRTIA requires preliminary labelling of both antigen and
antibody and its limited by a molecular distance critical for
the resonance energy transfer. LILA-FRAT and LILA-FIA,
involve a complicated procedure of liposome preparation and
loading with spin and fluorescent probes, and bearing cova-
lently bound antibodies. The most commonly employed im-
munoassay methods for small antigen molecules appear to be
the FPIA and FRTIA. However, they have two main disad-
vantages: they require special polarisation or resonance instru-
ments, and they are insufficiently sensitive for incident and
emitted light to be polarised. Also, the practical sensitivity of
these methods does not make them commercially viable.
Being a non-separation and rapid immunoassay producing
results within a few minutes and based on time-dependent
fluorescence emission in-situ, the FCIA seems to have a
number of essential advantages over all the above techniques.
Also, the FCIA is a quantitative method and can measure the
concentration of an analyte in a sample by a competitive
binding of the analyte molecules to the active site of the
antibodies occupied by the trans-stilbene-conjugates. Never-
theless, with all its advantages over the listed above tech-
niques, it is still an antibody-based immunoassay with many
disadvantages of using antibodies, which make it
Thus, for the first time, we report here the experimental
studies connected to development of the novel photochrome-
aptamer switch assay (PHASA) allowing state-of-the-art ana-
lyte detection and quantification of small molecules such as
toxins, explosives, drugs and pollutants, which are difficult to
detect using antibodies-based assays with high sensitivity and
specificity, in minutes. In the first stage of the PHASA devel-
opment, we have focused on the so called “adaptive binding”
of the stilbene compound to an aptamer of interest, as it will be
explained in the next section. Synthetic preparation and

J Fluoresc
experimental studies of the novel stilbene-maleimide switches
and their adaptive binding to the thiolated Malachite Green
aptamer are described in the present manuscript.
TOF2 in linear positive mode and calibrated against lysozyme
(m/z 14306). Samples were spotted neat or with 2,4,6-
trihydroxyacetophenone (10 mg in 250 μL ACN).
Synthesis
Experimental
The following reactants were prepared by slight modification
of our procedure that we published in [6]: trans-4-N,N′-
dimethylamino-4′-nitrostilbene, trans-4-methoxy-4′-
nitrostilbene, trans-4-N,N′-dimethylamino-4′-N-
aminostilbene and trans-4-methoxy-4′-N-aminostilbene.
Reactants and Solvents
Reactants were either commercially available or freshly pre-
pared as detailed in the “Synthesis” section. Commercially
available organic solvents, namely dimethyl sulfoxide (Sig-
ma-Aldrich ACS spectrophotometric grade, ≥99 %) and eth-
anol (Sigma-Aldrich ACS spectrophotometric grade, ≥99 %),
used for spectroscopy did not require any additional
purification.
Trans-4-N,N′-Dimethylamino-4′-Nitrostilbene (DANS)
An equimolar mixture of 4-nitrophenylacetic acid (Sigma-
Aldrich N20204) (9.1 g) and 4-(dimethylamino)benzaldehyde
(Sigma-Aldrich 156477) (7.5 g) in 3 mL of piperidine was
heated for 24 h under reflux in an oil bath maintained at 140°.
The resulting dark-red crude solid was recrystallized from
100 mL of chlorobenzene. Upon cooling, the product precip-
itated as lustrous red flakes, then was collected by vacuum
filtration on a Buchner funnel, thoroughly washed with petro-
leum ether (60–80°) to remove the adherent piperidine and
dried in a vacuum oven at 60° for about 2 h. Yield of the
lustrous red flakes—7.4 g (53 %). Chemical purity and struc-
tural identity of the product was confirmed by 1H NMR.
Apparatus and Methods
Semi-preparative flash chromatography used for purification
of the compounds was performed with the Büchi Sepacore®
flash chromatography system using 25 g silica gel cartridges
(particle size 40–63 μm) and dichloromethane as an eluting
1
solvent. H NMR spectra were run on 10 % (w/v) sample
solutions in CDCl3 with (CH3)4Si as an internal standard at
room temperature using a 400 MHz Bruker Fourier transform
spectrometer, equipped with a DMX AVANCE I system. UV
absorption spectra were measured using an Agilent Cary 300
spectrophotometer, and the steady-state fluorescence spectra
were recorded with Horiba Jobin Yvon FluoroLog®-3 modu-
lar spectrofluorometer and Agilent Cary Eclipse fluorescence
spectrophotometer. The fluorescence excitation and emission
spectra were corrected for instrumental sensitivity at different
excitation and emission slits using the instrument internal
excitation-emission matrix (EEM) correction [11, 12]. A so-
lution of quinine bisulphate in 0.1 N H2SO4 (Φf=0.52) was
taken as fluorescence standard for the determination of fluo-
rescence quantum yields [13]. Constant-illumination fluores-
cence intensity decay curves at the photostationary steady-
state equilibrium between the trans- and cis-isomers of stil-
benes were recorded with Shimadzu RF-5301 spectrofluo-
rometer equipped with the 150 W Xenon lamp as a radiation
light source. The fluorescence decay at the photostationary
steady-state equilibrium was monitored at the emission max-
imum of the stilbenes after excitation at the excitation maxi-
mum using typically 5-nm slit width for excitation and 5-nm
slit width for emission. Analysis of the experimental data was
performed using Origin® Pro 9.0 for Windows. Fluorescence
intensity decay curves were analysed with a polynomial fit for
calculation of the trans-cis isomerization rate constants using
a self-written routine within the Origin® Pro 9.0 for analysis
of the first-order photochemical reaction rates. Mass spectros-
copy analysis was performed on a Shimadzu Biotech Axima
4-Nitrobenzyltriphenylphosphonium Bromide
Triphenylphosphonium salt was prepared from the corre-
sponding 4-nitrobenzyl bromide (Sigma-Aldrich N13054)
and triphenylphosphine (Fluka 93090) in toluene by slight
modification of Wittig reaction [14]. In short, 2.16 g
(10 mmol) of 4-nitrobenzyl bromide and 3.14 g (12 mmol)
of triphenylphosphine were dissolved in 50 mL of toluene,
and the reaction mixture was heated and stirred at 70 °C
overnight. The precipitated salt was collected by vacuum
filtration on a Buchner funnel and washed twice with acetone
and petroleum ether and then dried in a vacuum oven at 80 °C
for two hours. Yield of the crystalline white powder—4.58 g
(96 %). It was used in the next step without further
purification.
Trans-4-Methoxy-4′-Nitrostilbene
4 . 5 8
g ( 9 . 6 m m o l ) o f t h e p r e p a r e d 4 -
nitrobenzyltriphenylphosphonium salt and 1.50 g (11 mmol)
of p-anisaldehyde (Sigma-Aldrich A88107) were dissolved in
25 mL of absolute methanol. 25 mL of 0.4-M lithium
methoxide solution in absolute methanol was added, and the
dark red reaction mixture was intensively stirred at room
temperature for 20 min. The resulting solution was left to
stand overnight at room temperature to provide the

J Fluoresc
precipitation of the product. The precipitated orange crystals
of the trans-isomer were collected by vacuum filtration on a
Buchner funnel and re-crystallised from ethanol. Yield of the
Trans-4-Methoxy-4′-N-Aminostilbene (1b, Fig. 1b)
Trans-4-methoxy-4′-nitrostilbene was reduced in a similar
manner as DANS with stannous chloride dihydrate in con-
centrated hydrochloric acid to afford the corresponding
aminostilbene. The product was purified with the flash chro-
matography on the silica gel (particle size 40–63 μm) with
dichloromethane as an eluent solvent. The fractions contain-
ing trans-isomer (λabs=330 nm) were collected and evapo-
rated to yield the white amorphous powder, which was re-
crystallised from ethanol to yield white small needles with
the 63 % reaction yield. Sample of the product was submit-
1
bright orange crystals—1.77 g (72 %). H NMR (chemical
shifts are in ppm): δ 3.85 (singlet, 3H, MeO); CH=CH AB
pattern: δ 6.98 (doublet, vinyl 1H), δ 7.23 (doublet, vinyl 1H);
p-MeO-Ar AA‘XX’ pattern: δ 6.98 (doublet, 2H: H3, H5), δ
7.49 (doublet, 2H: H2, H6); p-O2N-Ar AA‘XX’ pattern: δ
7.59 (doublet, 2H: H2′, H6′), δ 8.20 (doublet, 2H: H3′, H5′).
Melting point of 150–151 °C was found to be in a good
agreement with the literature value [14].
1
ted for H NMR, 400 MHz, CDCl3:: 7.60–7.58 (2H; d;
Trans-4-N,N′-Dimethylamino-4′-N-Aminostilbene
(1a, Fig. 1b)
8.53 Hz); 7.50–7.48 (2H; d; 8.72 Hz); 7.36–7.34 (2H; d;
8.46 Hz); 7.12–7.08 (1H; d; 16.23 Hz); 7.02–6.98 (1H; d;
16.29 Hz); 6.94–6.92 (2H; d; 8.65 Hz); 6.88 (2H; s); 3.86
(3H; s).
Solution of 120 g of potassium hydroxide pellets dissolved in
200 mL of ultrapure water was prepared in 500 mL Erlen-
meyer flask and placed in a fridge for storing. This solution is
used for decomposition of the prepared stannous complex salt
of the product. 15 g of stannous chloride dihydrate was
dissolved in 150 mL of concentrated hydrochloric acid at
room temperature with stirring. 4 g of fresh trans-4-
dimethylamino-4′-nitrostilbene prepared in the previous step
was introduced into the stirred solution of SnCl2. The suspen-
sion was refluxed for about 4–5 h until all red flakes and
orange viscous particles of the intermediate stannous complex
salt formed during the reaction have been completely dis-
solved, and the yellow reaction mixture has become transpar-
ent. The reaction mixture was stirred for additional half an
hour and allowed to cool to the point of turbidity. Then it was
poured carefully into a cold stirred solution of potassium
hydroxide taken from the fridge and placed into an ice bath,
in order to decompose the complex stannous salt and to
release the product. Caution: the neutralization reaction is
highly exothermic and vigour! Well-protective glasses, gloves
and clothes should be worn during this step, and all precau-
tions for minimising the reaction mixture splashing should be
taken. The crude yellow product was collected by vacuum
filtration, thoroughly washed with 10 % sodium bicarbonate
solution, ultrapure water and then with petroleum ether (60–
80°). The crude yellow product was dissolved in 10 mL of
dichloromethane, and subjected to the flash chromatography
on the regular silica gel (particle size 40–63 μm). The frac-
tions containing trans-isomer (λabs=360 nm) were collected
and evaporated to yield the yellow amorphous powder
(2.82 g, 77 % reaction yield). Chemical purity was >99 % as
determined by RP-HPLC and structural identity of the product
Trans-4-N,N′-Dimethylamino-4′-Stilbene Maleimide (2a)
Step1: Preparation of the maleamic acid intermediate.
Stilbene-maleimides and the corresponding sulphides
were prepared according to the reaction scheme seen in
Fig. 1b and c. Maleic anhydride (0.98 g, 10 mmol, Fluka
63200) and 50 mL of DCM were introduced into a round-
bottom flask equipped with a condenser and a heated
magnetic stirrer. After maleic anhydride dissolution, 1a
(1.9 g, 8 mmol) was added and the mixture stirred at RT
for 1.5 h to complete the reaction whereby the crude dark
red product was precipitated and collected by vacuum
filtration after cooling in an ice bath. The product was
washed with deionized water, DCM, and vacuum dried.
If desired, the intermediate maleamic acid may be recrys-
tallized from DMF or DMF/water mixture.
Step 2: Preparation of the maleimide (2a).
The maleamic acid from the previous step was
cyclised by dehydration with acetic anhydride and
sodium acetate as described in the literature [15, 16].
Briefly, 7 mmol of the obtained maleamic acid was
mixed with 0.35 g of anhydrous sodium acetate and
10 mL of acetic anhydride. Upon stirring and heating at
100 °C, the suspension was dissolved and the resulting
solution gradually changed colour from dark red to dark
yellow. After 45 min of heating the reaction mixture
was allowed to cool for 1 h and poured on ice. Stirring
the ice mixture for 30 m (until the ice melted) resulted
in a separation of a dark red precipitate, which was
collected by vacuum filtration, washed with ultrapure
water and vacuum dried. The product was purified with
the flash chromatography on the silica gel (particle size
40–63 μm) with DCM eluent. Yield of the purified
maleimide is 71 % and purity is >99 % as assessed
by RP-HPLC. 1H-NMR, 400 MHz, DMSO-d6:
1
was confirmed by H NMR, 400 MHz, CDCl3: 7.41–7.38
(2H; d; 8.78Hz); 7.33–7.31 (2H; d; 8.46 Hz); 6.92–6.88 (1H;
d; 16.36 Hz) 6.87–6.83 (1H; d; 16.36 Hz); 6.75–6.72 (2H; d;
8.78 Hz); 6.70–6.68 (2H; d; 8.40 Hz); 3.71 (broad s; 2H); 2.99
(s; 6H).

J Fluoresc
7.64–7.62 (2H; d; 8.53 Hz); 7.47–7.44 (2H; d;
8.84 Hz); 7.30–7.28 (2H; d; 8.53 Hz); 7.19 (2H; s);
7.20–7.16 (1H; d; 16.29 Hz); 7.03–6.99 (1H; d;
16.42 Hz); 6.75–6.72 (2H, d, 8.91 Hz); 2.95 (6H;
s).13C-NMR, DMSO-d6: 170.44; 150.58; 137.85;
135.16; 130.17; 130.12; 128.14; 127.32; 126.50;
125.17; 123.03; 112.67. MALDI TOF (neat; reflectron
mode, positive) e/z 318.15 [M]+ (calculated 318.14).
Reaction of 2a with thiolated MG Aptamer, A9 C6SH Label
to Yield 4, MG Aptamer, A9 Stilbene conjugate (Fig. 1d)
MG RNA aptamer (NCBI Accession:1Q8N_A) A9 modified
with C6SH label (5 nmol, AITbiotech, Singapore) as solution
in 40 μL of PBS buffer was treated with TCEP (1 μL; 40 mM
solution in PBS buffer) to activate its thiol group during
30 min at r.t. Then 340 μL DMSO and 24 μL of 4′-
dimethylamino-stilbenyl-4-maleinimide stock solution in
DMSO (8.67 mM, i.e. 208 nanomoles of the maleimide) were
added. Vortexing resulted in clear solution, which was left at
room temperature overnight. RNA was precipitated with
90 μL 5 M aqueous ammonium acetate and 1.7 mL
isopropanol. Upon keeping at −20 °C and centrifugation
(13,000 rpm; 30 min) an RNA pellet was formed. The super-
natant was removed and the remaining pellet was washed with
isopropanol (3×1 mL). Then the pellet was re-dissolved in
600 uL of PBS buffer. OD (260 nm) 0.579; calculated con-
centration 1.226 μM. Thus 0.735 nanomoles of the conjugat-
ed aptamer was obtained and the yield was 14 %. According
to RP-HPLC there was no starting maleimide detected and the
purity was assessed >99 %. MALDI-TOF. Theoretical m/z
(12514, [M+H]+); Supplier m/z: A9 modified with C6SH
label, 12515, [M+H]+; Shimadzu MALDI-TOF (THAP ma-
trix, reflection mode, positive), A9 modified with C6SH label,
12615, [M+H+DMSO+H2O]+, Shimadzu MALDI-TOF
(THAP matrix, reflectron mode, positive), 4, 12833, [M+H]+.
Trans-4-Methoxy-4′-Stilbene Maleimide (2b)
2b was prepared in a similar manner as 2a in two steps starting
from 1b and maleic anhydride and further cyclising the ob-
tained corresponding maleamic acid by dehydration with
acetic anhydride and sodium acetate. After purification with
the flash chromatography, light lemon-yellow crystals were
obtained. Purity is >99 % as assessed by RP-HPLC. 1H-NMR,
400 MHz, DMSO-d6: 7.60–7.58 (2H; d; 8.53 Hz); 7.50–7.48
(2H; d; 8.72 Hz); 7.36–7.34 (2H; d; 8.46 Hz); 7.12–7.08 (1H;
d; 16.23 Hz); 7.02–6.98 (1H; d; 16.29 Hz); 6.94–6.92 (2H; d;
8.65 Hz); 6.88 (2H; s); 3.86 (3H; s). 13C-NMR, DMSO-d6,
170.40; 159.60; 137.32; 135.19; 130.72; 129.96; 129.43;
128.40; 127.33; 126.88; 125.65; 114.68; 55.65. MALDI
TOF (neat, reflectron mode, positive) e/z 306.13 [M+H]+
(calculated 306.11).
Reaction of 2a/b with 6-Mercaptohexanol towards 3a/b
(Fig. 1c)
Results and Discussion
Either 2a or 2b (0.272 mmol) was dissolved in 4 mL of
dichloromethane. Isopropanol and triethylamine (0.4 mL
each) were added followed by addition of 6-
mercaptohexanol (70 μL, 0.51 mmol). The mixture was
stirred for 4 h to complete the reaction, the progress of which
was monitored with silica-TLC. The reaction mixture was
then evaporated and the remaining white residue was washed
with isopropanol, methanol and vacuum dried. Reaction
The PHASA bioassay is based on the conjugation of a switch-
able trans-stilbene molecule to any aptamer of interest and
utilizes the numerous advantages of aptamers, such as their
high binding affinity and specificity, stability and easy chem-
ical synthesis. The mechanism of operation behind PHASA
relies on the trans-cis photoisomerisation of a stilbene-ligand
compound having a chemical moiety (ligand) matching the
analyte of interest. As mentioned above, the trans-cis
photoisomerisation rate of the stilbene compound is depended
on the surrounding media, which can be used to indicate the
presence of the analyte and its further quantification. Quanti-
fication of the PHASA is essentially based on the constant-
illumination fluorescence decay of the trans-isomer toward
the photostationary equilibrium that removes cumbersome
and error-prone light intensity calibrations. The signal is mon-
itored under constant-illumination conditions as a time trace
or gradient of the fluorescence emission decay.
1
yields of approximately 75 % were obtained. 3b, H-NMR,
400 MHz, DMSO-d6: 7.69–7.67 (2H; d; 8.46 Hz); 7.59–7.56
(2H; d; 8.78 Hz); 7.29–7.25 (1H; d; 16.67 Hz); 7.26–7.24
(2H; d; 8.46 Hz); 7.17–7.13 (1H; d; 16.48 Hz); 6.98–6.96
(2H; d; 8.78 Hz); 4/37–4.34 (1H; t; 5.15 Hz; −OH); 4.13–4.09
(1H; dd; 9.12 Hz; 4.14 Hz); 3.79 (3H; s); 3.41–3.38 (2H; t;
6.25 Hz); 2.86–2.74 (2H; m); 2.72–2.67 (1H; dd; 18.31 Hz;
4.10 Hz); 1.65–1.53 (2H; m); 1.46–1.28 (6H; m). 13C-NMR,
DMSO-d6, 176.42; 174.78; 159.65; 138.03; 131.36; 129.91;
129.72; 128.45; 127.56; 126.92; 125.57; 114.69; 61.09;
55.65; 36.67; 36.67; 32.86; 31.02; 29.18; 28.56; 25.53.
MALDI-TOF (neat; reflectron mode, positive) m/z 440.46
[M+H]+ (calculated 440.19).
We have recently developed two formats of the PHASA:
competitive and adaptive binding formats. The competitive
binding format will be briefly reviewed below and is a subject
of the subsequent publication, while the experimental studies

J Fluoresc
related to the adaptive binding format will be described in the
present manuscript.
shape after binding to the analyte. In this case, instead of using
the stilbene-ligand molecule, the stilbene molecule can be
directly covalently attached to the aptamer via a crosslinker.
The mechanism works by allowing the aptamer to fold or
unfold around the attached stilbene molecule and can be
classified into two cases:
In Case 1, following the covalent binding of the stilbene
onto an aptamer, the adaptive binding of the aptamer is in-
duced resulting in the stilbene being squeezed inside the
aptamer, and its fluorescence decay becomes slow. This situ-
ation is shown in Fig. 3a. In Case 2, following the covalent
binding of the stilbene molecule onto an aptamer, the aptamer
folds around the stilbene, sterically hindering the twisting
motion of the stilbene molecule in the excited state, resulting
in relatively slow fluorescence decay. When the analyte B is
added, the aptamer folds around the analyte, exposing the
stilbene to solution and thus, the fluorescence decay becomes
fast. This can be illustrated in Fig. 3b.
Competitive Binding
The method for analyte detection and concentration measure-
ment in a sample based on a direct conjugation between the
stilbene-ligand molecule and the analyte-specific aptamer in-
volves a competitive binding of the free analyte molecules
present in the sample to the binding site of the aptamer. This
will competitively replace the stilbene-ligand molecule initial-
ly forming a complex with the analyte-specific aptamer as
shown in Fig. 2. Introduction of a sample containing the
analyte, results in a competitive analyte-aptamer complex
formation. This subsequently releases the fluorescent trans-
stilbene-ligand molecule from the aptamer into solution that is
observed as a significant acceleration of the fluorescence
decay of the trans-isomer upon irradiation with the excitation
light. Analyte concentration is calibrated to the trans-cis
photoisomerisation rate constant to provide a quantification
of the assay.
The crosslinker that we used in our studies of the adaptive
binding format is maleimide, where ‘click chemistry’ allows
for efficient covalent binding of the stilbene to the aptamer.
The stilbene-maleimide was attached directly to the thiol-
modified aptamer by the click-like chemistry as shown in
Fig. 1d.
Adaptive Binding
Besides the above competitive binding format, direct covalent
attachment of the stilbene probe with the modified aptamer
followed by adaptive binding of the analyte molecules can
also lead to the generation of a calibration curve for practical
applications. Thus, the following mechanism has been put
forward, by making use of the adaptive binding property of
aptamers. This refers to the aptamer changing its folding and
The prepared maleimides 2a and 2b were fully
characterised. Figure 4a, b, and c displays the absorption,
fluorescence excitation and emission spectrum of the stil-
bene-maleimides, respectively. Their fluorescence decay ki-
netics to the photostationary state under constant-illumination
conditions is shown in Fig. 4d. There is a strong red-shift of
the absorption and fluorescence emission maximum observed
Fig. 2 PHASA competitive
binding format illustrated with the
detection and quantification of
Analyte A. The stilbene-ligand
switch is initially bound to the
analyte-specific aptamer. Free
analyte replaces a portion of the
stilbene-ligand conjugates. The
fluorescence decay rate is then
measured and calibrated to the
free analyte concentration

J Fluoresc
The measured fluorescence quantum yield 2b is almost
twice as low compared to the stilbene-maleimide-MCH ad-
duct 3b. We assume that part of the additional non-radiative
losses in 2b are caused by the electronic energy transfer onto
the maleimide acceptor fragment, which intrinsically shows
negligible fluorescence, due to the presence of an efficient
conical intersection of the S1 state with the ground state S0
[17]. The effect should be even more pronounced in 2a having
the lowest quantum yield, since the DMA group is a stronger
electronic donor than methoxy group. Drastic increase of the
quantum yield of 3b then should be connected to the destruc-
tion of the charge transfer interaction between the donor
methoxy and maleimide acceptor substituents, as a result of
the chemical reaction with MCH that removes the
conjugation/electronic delocalization in the maleimide ring
and thereby decreases its acceptor power.
Fig. 3 a Case 1 where the adaptive binding of the analyte A results in
retardation of the fluorescence decay. b Case 2 where the adaptive
binding of the analyte A results in acceleration of the fluorescence decay
It is known that the trans-cis photoisomerisation process of
stilbene compounds can be sterically hindered by their sur-
rounding environment [7, 8]. Medium viscosity effects on the
trans-cis photoisomerisation of both stilbene-maleimides are
shown in Fig. 5a and b. In order to study the medium viscosity
effect on the fluorescence decay rate of 2a, 2b and 3b
(Fig. 5a–c), several aqueous glycerol solutions of the com-
pounds having the same concentration were prepared. Viscos-
ity of these solutions was varied by adding in different
amounts of glycerol and measured using the Brookfield vis-
cometer. The fluorescence decays were measured and the
corresponding trans-cis photoisomerisation rate constants
were calculated. The logarithmic dependence of the trans-cis
in the spectra of 2a compared to 2b, and can be explained by
the presence of the strong donor dimethylamino (DMA) group
substituent at the para-position of the aromatic ring. The
DMA group facilitates the charge delocalization in the ground
state 1t of 2a that results in a lower energy and longer wave-
length transition to the Frank-Condon state [3, 2]. It also
1
stabilizes the excited t* state of 2a in comparison with 2b,
thereby increasing the Stock shift. The spectral properties of
the measured compounds are summarised in Table 1, which
brings together the absorption (λabs), fluorescence excitation
and emission (λex and λem) maxima, fluorescence quantum
yields (Φf) and apparent trans-cis photoisomerisation rate
constants (Kapp) for the investigated compounds
Fig. 4 a Absorption spectrum of
2a, 2b, and 3b in DMSO (20 μM
for all, b Fluorescence excitation
spectrum of 2a, 2b, and 3b in
DMSO (2 μM for all), c
Fluorescence emission spectrum
in DMSO (2 μM for all) and d
Fluorescence decay kinetics in
DMSO (2 μM for all)

J Fluoresc
Table 1 Absorption (λabs), fluorescence excitation and emission (λex and λem) maxima, fluorescence quantum yields (Φf) and apparent trans-cis
photoisomerisation rate constants (Kapp) for 2a, 2b and 3b
λmax (nm)
λ
ex (nm)
λ
em (nm)
Quantum yield
Apparent trans-cis photoisomerisation rate constant (s−1
)
2a
2b
3b
368
329
329
362
332
323
468
393
392
0.009 0.001
0.015 0.001
0.032 0.001
0.776
0.562
0.756
photoisomerisation rate constant on the reciprocal medium
viscosity is shown in Fig. 5d.
quantify the method and establish the sensitivity range of the
stilbene switches.
From Fig. 5d, it can be observed that for all the three
compounds, certain linear detection range exists. For 2a, the
equation for analyte quantification in the linear detection
range (solid line) can be described by Kapp=(1.066 0.011)+
(0.309 0.012) log (1/η). For 2b, it is described by
Malachite Green Aptamer Conjugation to Stilbene (4) Retains
Malachite Green Binding Ability
Malachite green (MG)-aptamer, a 38 nt single-stranded RNA
(NCBI #1Q8N), was synthesized by solid-phase synthesis
from a local supplier and was found to have a Kd of 800
200 nM, which is in agreement with that reported in the
literature [18–20]. Four sites of nucleotide C6SH (thiol) mod-
ification (U4, A9, C20, and C38) were chosen based on their
non-essential nature, distance from the binding pocket and
surface exposure to the solvent [18, 20]. The A9 C6SH
modification was found to have the lowest binding affinity
of 700 100 nM to MG, although all 4 nucleotide C6SH
modifications had binding affinities below 1,000 nM. The
MG aptamer A9 C6SH label was chosen for grafting by 2a–
2a happens to be slightly more soluble than 2b. Grafting of 2a
to MG aptamer A9 C6SH label was accomplished by reacting
K
K
app =(0.726 0.009)+(0.083 0.005) log (1/η) and
app=(0.824 0.009)+(0.098 0.005) log (1/η) for the 3b.
In view of the above results, we can conclude that the
fluorescence decay rate is indeed strongly dependent on me-
dium viscosity. With increase of the medium viscosity, retar-
dation of the fluorescence decay is clearly observed, which
can be considered as an indicator of the steric freedom for the
twisted motion of the stilbene molecule in the excited state
within its lifetime. In other words, the change of fluorescence
decay rate constant with medium viscosity confirms that there
is a strong dependence of the trans- cis photoisomerisation
process on the viscosity of the surrounding medium. The
logarithmic curve shown in Fig. 5d makes it possible to
Fig. 5 Viscosity profile of 2 μM
a 2a b 2b and c 3b. d The
logarithmic dependence of the
trans-cis photoisomerization rate
constant on the reciprocal
medium viscosity for 1 μM of 2a,
2b, and 3b

J Fluoresc
in DMSO/10 % PBS and precipitation of 4 in isopropanol as
depicted in Fig. 1d. Reverse phase-HPLC found equal molar
amounts of MG aptamer (ABS @ 260 nm) and the 2a (ABS
@ 340) eluting at the same peak times, with no traces of
unbound 2a as seen in Fig. 6a and b. Binding affinity of the
MG to 4 was subsequently assessed by following the fluores-
cence of MG. When MG is bound within the MG-aptamer
pocket, it exhibits a quantum yield three orders of magnitude
higher than unbound MG due to planarization of the molecule
as a result of the electron delocalization in its excited state [21,
22]. We can take advantages of this unique feature for easy
assessment of binding coefficients, which can be calculated by
ubiquitous fluorescent microplate readers. Figure 6c displays
the fluorescence emission of 4, MG, and 4 + MG—little to no
fluorescence is seen for 4 or free MG. However when mixed, a
high fluorescence intensity signal was seen, indicating that the
4 MG Aptamer, A9 stilbene retained its ability to bind to MG
with affinity similar to the wild-type MG aptamer and uncon-
jugated MG aptamer A9 C6SH label. This shows that in the
presence of the bound stilbene molecule, the MG successfully
enters the binding pocket of the aptamer, which substantiates
the mechanism of adaptive binding format described above.
Due to manuscript constraints, our subsequent publication
will further investigate the adaptive binding photodecay ki-
netics and analyte sensitivity of MG Aptamer modified with
stilbene at the four positions previously mentioned: U4, A9,
C20, and C38.
Conclusions
Aptamers existing on the market are available for over 150
targets, mostly for therapeutic purposes. This leaves much
opportunity for PHASA to be quickly developed towards
numerous applications. Should a particular aptamer be un-
available, it can be easily designed and manufactured in 6–
8 weeks for a reasonable cost, which would be unfeasible (if
not impossible) for antibodies. Aptamers will allow for more
novel designs and continued development of the next gener-
ation PHASA biosensors. A key design feature in the PHASA
platform is based on the ‘no-separation-needed’ concept—
reagents and bound aptamers will require no discrete separa-
tion protocols or washing procedures before or after measure-
ments. It is essentially a one-step procedure that is self-
contained. The bioassay can operate not only in aqueous
solutions but also in a solvent mixtures, heavily contaminated
heterogeneous wastes and biological media collected in a
single tube, without separate purification protocols. Eventual-
ly, the PHASA assay permits not only the instant detection of
an analyte in a sample but also rapid (within a few minutes)
determination of its concentration.
In the first stage of the PHASA development, we have
focused on the adaptive binding of the stilbene compound to
an aptamer of interest, as shown in Figs. 3 and 6. Synthetic
preparation and experimental studies of the novel maleimide-
stilbene compounds and their adaptive binding to the thiolated
Malachite Green aptamer, 4, are described in the present
manuscript.
Acknowledgments The authors wish to thank Dr. Vitali Lipik for his
invaluable help with the MALDI-TOF analysis of the samples. This
research is funded by the Singapore National Research Foundation and
the publication is supported under the Campus for Research Excellence
and Technological Enterprise (CREATE) programme (13-04-00364 А).
The authors thank Huang Ruo Cheng, Tan Chong Yuan and Terence Mak
Fig. 6 RP-HPLC characterization of 4 by two ABS wavelengths of a
260 nm, which allows quantitation of the MG-aptamer and b 365 nm,
which allows quantitation of 2a. c Fluorescence intensity of free MG, 4,
bound MG with the MG-aptamer binding pocket of 4

J Fluoresc
of River Valley High School for their active participation in this project in
the frame of the program for Singapore Science & Engineering Fair. In
addition, the authors thank NTU for providing partial financial support to
this program. Funding was also greatly appreciated from the Ministry of
Education Tier 1 Grant: Photochrome aptamer switch assay: A universal
bioassay device - RG54/13.
photochrome method. Biophys Chem 103(2):139–156. doi:10.
1016/S0301-4622(02)00252-1
10. Chen O, Glaser R, Likhtenshtein GI (2008) The novel FluoroChrome
ImmunoAssay—(FCIA). The role of molecular environment upon
molecular structure exemplified by constriction of a flourescence-
photochrome flanked by two proteins. J Biochem Biophys Methods
70(6):1073–1079. doi:10.1016/j.jbbm.2007.07.007
11. Horiba Scientific, “NIST develops fluorescence standards with
FluoroLog”, Fluorescence Technical Note FL-30. (n.d.). http:www.
horiba.com/fileadmin/uploads/Scientific/Documents/Fluorescence/
FL-30_NIST_Fluorolog.pdf. Accessed 27 July 2014
References
12. Meir A, Stojanovic M, Marks RS (2007) Aptameric Biosensors. In:
Marks RS, Cullen D, Lowe C, Weetall HH and Karube I (eds).
Handbook of Biosensors and Biochips, John Wiley & Sons Ltd
Publishers, Chapter 15, pp 217–232
13. Meech SR, Phillips D (1983) Photophysics of some common fluo-
rescence standards. J Photochem 23(2):193–217. doi:10.1016/0047-
2670(83)80061-6
14. Manecke G, Lüuttke S (1970) Über Synthesen einiger oligomerer
substituierter Arylenvinylene. Chem Ber 103(3):700–707. doi:10.
1002/cber.19701030308
15. Steele TWJ, Shier WT (2010) Dendrimeric alkylated
polyethylenimine nano-carriers with acid-cleavable outer cationic
shells mediate improved transfection efficiency without increas-
ing toxicity. Pharm Res 27(4):683–698. doi:10.1007/s11095-010-
0058-1
1. Görner H, Kuhn HJ (2007) Cis-trans photoisomerization of stilbenes
and stilbene-like molecules. In: Advances in photochemistry. John
Wiley & Sons, Inc, Hoboken, pp 1–117. doi:10.1002/
9780470133507.ch1
2. Papper V, Likhtenshtein GI (2001) Substituted stilbenes: a new view
on well-known systems: new applications in chemistry and biophys-
ics. J Photochem Photobiol A Chem 140(1):39–52. doi:10.1016/
S1010-6030(00)00428-7
3. Papper V, Pines D, Likhtenshtein G, Pines E (1997) Photophysical
characterization of trans-4,4′-disubstituted stilbenes. J Photochem
Photobiol A Chem 111(1–3):87–96. doi:10.1016/S1010-6030(97)
00234-7
4. Waldeck DH (1991) Photoisomerization dynamics of stilbenes.
Chem Rev 91(3):415–436. doi:10.1021/cr00003a007
5. Ahluwalia A, De Rossi D, Giusto G, Chen O, Papper V,
Likhtenshtein GI (2002) A fluorescent-photochrome method for the
quantitative characterization of solid phase antibody orientation.
Anal Biochem 305(2):121–134. doi:10.1006/abio.2002.5601
6. Likhtenshtein GI, Bishara R, Papper V, Uzan B, Fishov I, Gill D,
Parola AH (1996) Novel fluorescence-photochrome labeling method
in the study of biomembrane dynamics. J Biochem Biophys Methods
33(2):117–133. doi:10.1016/S0165-022X(96)00022-X
7. Parkhomyuk-Ben Arye P, Strashnikova N, Likhtenshtein GI (2002)
Stilbene photochrome-fluorescence-spin molecules: covalent immo-
bilization on silica plate and applications as redox and viscosity
probes. J Biochem Biophys Methods 51(1):1–15. doi:10.1016/
S0165-022X(01)00234-2
16. Cava MP, Deana AA, Muth K, Mitchell MJ (1961) N-
Phenylmaleimide. Org Synth 41:93. doi:10.15227/orgsyn.041.0093
17. Braun D, Rettig W (1997) Excitation energy dependence of the
kinetics of charge-transfer formation. Chem Phys Lett 268(1–2):
110–116. doi:10.1016/S0009-2614(97)00156-5
18. Flinders J, DeFina SC, Brackett DM, Baugh C, Wilson C,
Dieckmann T (2004) Recognition of planar and nonplanar ligands
in the malachite green-RNA aptamer complex. Chembiochem Eur J
Chem Biol 5(1):62–72. doi:10.1002/cbic.200300701
19. Grate D, Wilson C (1999) Laser-mediated, site-specific inactivation
of RNA transcripts. Proc Natl Acad Sci 96(11):6131–6136. doi:10.
1073/pnas.96.11.6131
20. Baugh C, Grate D, Wilson C (2000) 2.8 A crystal structure of the
malachite green aptamer. J Mol Biol 301(1):117–128. doi:10.1006/
jmbi.2000.3951
8. Strashnikova N, Papper V, Parkhomyuk P, Likhtenshtein GI, Ratner
V, Marks R (1999) Local medium effects in the photochemical
behavior of substituted stilbenes immobilized on quartz surfaces. J
Photochem Photobiol A Chem 122(2):133–142. doi:10.1016/S1010-
6030(99)00009-X
21. Stojanovic MN, Kolpashchikov DM (2004) Modular aptameric sen-
sors. J Am Chem Soc 126(30):9266–9270. doi:10.1021/ja032013t
22. Babendure JR, Adams SR, Tsien RY (2003) Aptamers switch on
fluorescence of triphenylmethane dyes. J Am Chem Soc 125(48):
14716–14717. doi:10.1021/ja037994o
9. Chen O, Glaser R, Likhtenshtein GI (2003) Molecular dynamics
investigation of an antibody binding site by the fluorescence–