2′-(1H-phenanthro[9,10-d]imidazol-2-yl)-phenyl-4-carboxylic acid N-hydroxysuccinimide ester: A new phenanthroimidazole derivative as a fluorescent probe for medical imaging applications

Article abstract of DOI:10.1016/j.saa.2019.117757

In this study a new probe (2′-(1H-phenanthro[9,10-d]imidazol-2-yl)-phenyl-4-carboxylic acid N-hydroxysuccinimide ester, PB1-1) was synthesized and presented, containing the ester group as reactive group for medical imaging applications. The tests included

Full text of DOI:10.1016/j.saa.2019.117757

Journal Pre-proof
2′-(1H-phenanthro[9,10-d]imidazol-2-yl)-phenyl-4-carboxylic
acid N-hydroxysuccinimide ester: A new phenanthroimidazole
derivative as a fluorescent probe for medical imaging applications
Przemysław Krawczyk, Tomasz Wybranowski, Łukasz
Kaźmierski, Iga Hołyńska-Iwan, Magdalena Bratkowska, Piotr
Cysewski, Beata Jędrzejewska
PII:
S1386-1425(19)31147-3
https://doi.org/10.1016/j.saa.2019.117757
SAA 117757
DOI:
Reference:
Spectrochimica Acta Part A: Molecular and Biomolecular
Spectroscopy
To appear in:
Received date: 17 July 2019
Revised date:
24 October 2019
3 November 2019
Accepted
date:
Please cite this article as: P. Krawczyk, T. Wybranowski, Ł. Kaźmierski, et al., 2′-(1H-
phenanthro[9,10-d]imidazol-2-yl)-phenyl-4-carboxylic acid N-hydroxysuccinimide ester:
A new phenanthroimidazole derivative as a fluorescent probe for medical imaging
applications, Spectrochimica Acta Part A: Molecular and Biomolecular
Spectroscopy(2018), https://doi.org/10.1016/j.saa.2019.117757
This is a PDF file of an article that has undergone enhancements after acceptance, such
as the addition of a cover page and metadata, and formatting for readability, but it is
not yet the definitive version of record. This version will undergo additional copyediting,
typesetting and review before it is published in its final form, but we are providing this
version to give early visibility of the article. Please note that, during the production
process, errors may be discovered which could affect the content, and all legal disclaimers
that apply to the journal pertain.

© 2018 Published by Elsevier.

Journal Pre-proof
2'-(1H-phenanthro[9,10-d]imidazol-2-yl)-phenyl-4-carboxylic acid N-
hydroxysuccinimide ester: a new phenanthroimidazole derivative as a fluorescent probe
for medical imaging applications
Przemysław Krawczyk1*, Tomasz Wybranowski2, Łukasz Kaźmierski3,4, Iga Hołyńska-Iwan5,
Magdalena Bratkowska6, Piotr Cysewski1, Beata Jędrzejewska6
1Nicolaus Copernicus University, Collegium Medicum, Faculty of Pharmacy, Department of
Physical Chemistry, Kurpińskiego 5, 85-950 Bydgoszcz, Poland
2Nicolaus Copernicus University, Collegium Medicum, Faculty of Pharmacy, Medical
Physics Division, Biophysics Department, Jagiellońska 13, 85-067 Bydgoszcz, Poland
3Nicolaus Copernicus University, Collegium Medicum, Faculty of Medicine, Department of
Tissue Engineering, Karłowicza 24, 85-092 Bydgoszcz, Poland
4Nicolaus Copernicus University, Collegium Medicum, Faculty of Medicine, Department of
Oncology, Radiotherapy and Oncological, M. Curie Skłodowskiej 9, 85-094 Bydgoszcz,
Poland
5 Nicolaus Copernicus University, Collegium Medicum, Faculty of Pharmacy, Laboratory of
Electrophysiology of Epithelial Tissue and Skin, Department of Pathobiochemistry and
Clinical Chemistry, M. Curie Skłodowskiej 9, 85-094 Bydgoszcz, Poland
6UTP University of Science and Technology, Faculty of Chemical Technology and
Engineering, Seminaryjna 3, 85-326 Bydgoszcz, Poland
*Corresponding author: e-mail address: przemekk@cm.umk.pl (P. Krawczyk)
Graphical abstract
Keywords: fluorescent probe, reactive group, phenanthroimidazole, spectroscopy of
conjugate, cell staining
1

Journal Pre-proof
Abstract
In this study a new probe (2'-(1H-phenanthro[9,10-d]imidazol-2-yl)-phenyl-4-carboxylic acid
N-hydroxysuccinimide ester, PB1-1) was synthesized and presented, containing the ester
group as reactive group for medical imaging applications. The tests included a comparison to
the PB1 probe with the aldehyde group described earlier. Also, the photophysics of PB1 and
PB1-1 when conjugated to albumin (HSA) and concanavalin A (Con A) was studied. The
fluorescence anisotropy measurements and the method of fluorescence quenching of protein
were used to examine these interactions. The results showed that both dyes are highly bound
to the studied proteins, especially PB1-1. In the present study we also compared the stability
of prepared conjugates. The in vitro study have shown that all tested compounds presented to
be usable in the case of fixated cell staining. PB1-1-ConA and PB1-1-HSA were
characterized with the lowest cytotoxicity during the MTT assay, and thus should be more
suitable for live imaging applications than PB1-ConA and PB1-HSA. The results obtained in
this work confirmed the theses presented in in silico studies as to the correctness of the choice
of ester group as actively binding to the protein. At the same time, we have experimentally
demonstrated the significant influence of a probe-protein linker on the spectral properties of
conjugates used in medical imaging. We have clearly indicated that a detailed analysis of
derivatives with different reactive group allows for proper probe selection. We also pointed
out that based on the geometric skeleton of one dye, a whole range of fluorescent probes with
different absorption and fluorescence spectra can be obtained for in vitro tests in medical
imaging
1. Introduction
Molecular imaging (bioimaging) has become a powerful and widely used tool in
medical research, due to the unique possibilities of visualizing anatomical features and
morphological details at the cellular and subcellular levels. Visualization of these phenomena
is possible due to introduction of properly constructed fluorescent markers into the living
organism. Their binding to specific elements of the tissue structure allows for spatial
assessment of these structures and for monitoring the processes occurring in them. Of
particular interest is the possibility of using these techniques in the analysis of interactions
that are responsible for the development of pathogenic processes, as well as tracking
pharmaceuticals during their interaction with biomolecules. In addition, these methods are of
great importance for the monitoring of new therapies that operate, or are controlled, at the
cellular and molecular level1-3. Many fluorescent probes with a wide range of fluorescence
2

Journal Pre-proof
colors and a wide spectrum of activity are available on the pharmaceutical market. Despite
their diversity, fluorescent dyes are still a very important field of research for applications in
live microscopy, due to the possibility of early diagnosis of neoplastic lesions and reduction
of resection of healthy tissues during surgical procedures; diabetes; Alzheimer's disease;
monitoring of intercellular interactions in a selected organ and many others. The search for
new markers is therefore an important scientific current, because their use in clinical medicine
is invaluable and not fully exploited. The newly designed fluorescent probes, in addition to a
suitable reactive group enabling easy coupling with biomolecules, should be characterized by
many photophysical properties. The dye should also have good solubility; high chemical and
environmental stability; lack of toxicity; non-invasiveness; high bioavailability and no
bioaccumulation4-6. It should also selectively bind to a particular type of cell, tissue or
organelle, showing compatibility with the source of excitation allowing for easy detection.
In an earlier study, the phenanthroimidazole derivatives with different reactive groups
as active fluorochromes were characterized as potential biomedical imaging probes7. It was
concluded that in the case of phenanthroimidazole derivatives the factor limiting the
possibility of use in medical imaging would be the presence of a pyridyl disulfide group,
while the introduction of the N-hydroxysuccinimide ester may be used to create stable and
valuable fluorescent probes with a wide spectrum for applications in biomedical imaging. For
this reason, such derivative was synthesized and in the presented work a description of its
photophysical properties was described in detail. For the best cognitive aspect, the results
obtained for PB1-1 were compared with the results obtained for PB18. These compounds
differ only in the reactive group, and in the case of PB1 an aldehyde group was used. The
work focuses not only on the marker, but also its conjugation with Concanavalin A and
describes the optical properties of the biocomplex. The tests were also supplemented with
applied in vitro study.
Quantitation of dye-protein conjugation ratio is essential for application of the probe
for experiments and for controlling fluorescence properties. The degree of labelling may be
calculated on the basis of protein and dye extinction molar coefficients after dialysis process.
However, the study of fluorescent dye-protein conjugates requires determining whether the
dye is bound to protein after removing the free dye by membrane. In addition, it is also
necessary to examine conjugates stability during time.
Presently, there are multiple techniques for the determination of interaction between
dye and protein including equilibrium dialyses, ultrafiltration, affinity chromatography,
magnetic beads, surface plasmon resonance, circular dichroism, electrochemistry, capillary
3

Journal Pre-proof
electrophoresis, analysis of absorption or fluorescence spectra and quenching fluorescence of
protein. Each method has both advantages and disadvantages. If the dye is highly fluorescent,
the best and very simple method is to measure the anisotropy fluorescence. It allows
determine if the molecules are bound or not to big molecules (e.g. albumin or concanavalin).
For the characterization of depolarization degree, the fluorescence anisotropy r is usually
used. As a consequence of the random orientation of small molecules, their rotational
movements and limited lifetime of the excited (electronic) states of fluorophores, the
fluorescence light emitted by the solution of dyes is always depolarized (r ≈ 0). The
anisotropy of fluorescence light emitted by dyes bound to the big molecules (albumin) is
close to fundamental anisotropy (r ≈ 0.3). By measuring the anisotropy fluorescence of
prepared solution it is easy to assess the degree the binding of the fluorescent dye to protein.
Fluorescence anisotropy was first applied to examine the interactions between dye and
albumin 7 decades ago 9,10. Fluorescence anisotropy measurements are now extremely
widespread for the study of the behaviour of molecules in solutions containing protein and
well documented 11-13. This method presents several advantages over other methods as it is a
direct, non-time-consuming, repeatable and non-invasive technique. The tested systems
interact only with low-intensity UV-VIS radiation. This method does not require additional
reagents. It eliminates the need for dissolution of the tested solution samples in mobile phase
solvents used in high-performance liquid chromatography so measurement conditions may
become close to physiological ones. Compared to other methods, the presented one provides a
more accurate measurement of bound fraction of dye when the dye has a strong affinity to
protein. Another advantage of this method is that it lets determine changes in solution during
the time without affecting the structure of the complex. This is an extraordinary benefit.
Determination of the stability of dye-protein conjugates is essential to apply this probe as an
excellent candidate for further bioanalytical studies.
Also recently, mass spectrometric (MS) has been found to be a tool to quantify dye-
protein binding. However, this method is very expensive, long time and very difficult to
apply. Under the conditions of conventional MS, biological substances dissociate into
individual molecules during ionization or/and desolvation. It is difficult to perform
measurements without disruption of non-covalent interactions between dye and protein. So,
protein-dye interactions are rarely observed. In fact, several studies have demonstrated that
MS can be used to determine the association constants for protein-dye interactions (MS under
non-denaturing conditions or native ESI MS) 14,15. However, the results are not quite
unambiguous, require few assumptions and data analysis and interpretation are not so easy. In
4

Journal Pre-proof
the case of hydrogen-deuterium exchange mass spectrometry (HDX MS), protein-dye
16
complexes are dissolved in a solution of D2O . But, it is not still clear if a change in HDX
rate is a result of direct interaction. These are a relatively new method. So far, these methods
have not been popularized to examine dye-protein conjugates. These methods also often
require that the complexes be placed in some solvent or gas phase (hydrophobic interaction
are weakened and electrostatic interaction are enhanced), which may change the degree of
binding 17. Currently, the MS methods are limited to a small number of studied samples but
they are very promising.
The quenching of protein fluorescence upon binding of dye has become a very simple
method for studying these interactions. The protein intrinsic fluorescence and quenching of
fluorescence in the presence of dye could be used since albumin and concanavalin A contain
tryptophan residues which generate autofluorescence. The quenching of fluorescence is
associated with two most important processes: static and dynamic quenching. It is generally
assumed that the fluorescence intensity is only reduced by the formation of the protein-dye
complex in the basal state (static quenching). However, dynamic quenching occurs when the
excited-state of protein collides with dye molecule and returns to ground state non-radiatively.
Only in the case of dynamic quenching the florescence lifetime of the sample is reduced.
The aim was to experimentally indicate the effect of conjugation mechanisms in the
probe - protein system on linear and nonlinear optical properties of bioconjugates. As it is
known, fluorescent probes can be conjugated to a protein by attaching active groups to the
dye such as :N-hydroxysuccinimide esters, isothiocyanates group, sulfonyl halide group,
pyridyl disulfides group and many others. In this way, many new probes are created, however,
there are practically no publications indicating differences in the desired optical properties,
and influence on the mechanism of bioconjugation. Studies on new fluorescent probes are
usually limited to the easiest attachment of the reactive group, without thinking about the
impact of the other active group and produced the same linker. Meanwhile, such differences
are emphasized in our work. It clearly indicates that the change of the reactive group allows
for obtaining, within one skeleton of the dye, the whole spectrum of fluorescent probes with
different activities after conjugation with the protein 17.
Abbreviations: ConA - concanavalin A, HSA - human serum albumin, TMS -
tetramethylsilane, 1,4-Dx - 1,4-dioxane, Bu2O - dibutylether, Et2O - diethylether, EtOAc -
ethyl acetate, THF - tetrahydrofuran, MeAc - acetone, DMF - N,N-dimethylformamide,
DMSO - dimethylsulfoxide, MeCN - acetonitrile
5

Journal Pre-proof
2. Methodology
2.1. Materials and synthesis
All reagents and solvents were obtained from Aldrich Chemical Co.
The 2'-(1H-phenanthro[9,10-d]imidazol-2-yl)-phenyl-4-carboxylic acid was synthesized
based on the procedures taken from literature18. The phenanthraquinone (10 mmol), 4-
formylbenzoic acid (30 mmol), ammonium acetate (50 mmol) and acetic acid (10 mL) were
magnetically stirred and refluxed for 3 h. Then, the mixture was cooled to room temperature,
poured into ice water. The crude product was filtered out, washed with water and purified by
flash chromatography on silica gel eluted with ethyl acetate/n-hexane (3:2 v/v) to give the
product.
C22H14N2O2; 338.36 g/mol; 53.7% yield; Rf = 0.5 for ethyl acetate/n-hexane 3:2 v/v.
1H NMR (DMSO-d6)  (ppm): 13.67 (s, 1H, NH), 12.85 (s, 1H, COOH), 8.90-8.85 (t, 2H, Ar), 8.62-8.57 (t, 2H,
Ar), 8.45-8.43 (d, J=8.0 Hz, 2H, Ar), 8.18-8.16 (d, 2H, J=8.0 Hz, Ar), 7.78-7.73
(m, 2H, Ar), 7.67-7.65 (d, J=8.0 Hz, 2H, Ar)
IR (KBr) (cm-1): 3234, 3065, 1695, 1668, 1612, 1553, 1530, 1481, 1458, 1385, 1285, 1246, 1121, 1061, 1016,
996, 966, 867, 752, 720
2'-(1H-phenanthro[9,10-d]imidazol-2-yl)-phenyl-4-carboxylic acid N-hydroxysuccinimide
ester
The
2'-(1H-phenanthro[9,10-d]imidazol-2-yl)-phenyl-4-carboxylic
acid
N-
hydroxysuccinimide ester, was obtained according to literature19. The 2'-(1H-
phenanthro[9,10-d]imidazol-2-yl)-phenyl-4-carboxylic acid (0.98g, 2.89 mmol) was dissolved
in 50 mL of dry tetrahydrofuran (THF). Then, the N-hydroxysuccinimide (0.4 g, 3.47 mmol)
and N,N'-dicyclohexylcarbodiimide (DCC; 0.776g, 3.76 mmol) were added. The mixture was
stirred at room temperature for 2 days. The precipitate was filtered off and washed with THF.
The organic layers were combined and the solvent was removed under reduced pressure. The
residue was treated with diethyl ether and the crude product was filtered off. The white solid
was purified by flash chromatography on silica gel eluted with ethyl acetate/n-hexane (3:2
v/v).
6

Journal Pre-proof
C26H17N3O4; 435.12 g/mol; 87.3% yield; Rf = 0.37 for ethyl acetate/n-hexane (3:2 v/v)
1H NMR (DMSO-d6)  (ppm): 13.82 (s, 1H, NH), 8.90-8.85 (t, 2H, Ar), 8.63-8.58 (t, 2H, Ar),
8.57-8.55 (d, J=8.0 Hz, 2H, Ar), 8.33-8.31 (d, 2H, J=8.0 Hz, Ar), 7.79-7.76 (m, 2H, Ar), 7.69-
7.67 (m, 2H, Ar), 2.94 (s, 4H, CH2)
IR (KBr) (cm-1): 3417, 3244, 3073, 1799, 1732, 1610, 1427, 1384, 1238, 1204, 1074, 997,
860, 753, 722, 614
2.2. Spectroscopic measurements
1
The H NMR (400 MHz) spectra were recorded on a Bruker AscendTM 400 NMR
spectrometers. DMSO-d6 was used as the solvent and TMS as internal standard.
The IR spectra of the synthesized salts were recorded using a Bruker Vector 22 FT-IR
spectrophotometer (Germany) in the range 400-4500 cm-1, by KBr pellet technique.
Absorption and emission spectra in different polarity solvents were recorded at room
temperature using a Shimadzu UV-vis Multispec-1501 spectrophotometer and a Hitachi F-
7100 spectrofluorimeter, respectively.
The fluorescence quantum yield of the tested compounds () was calculated using
equation (1):
IAref n2
  ref

(1)
n2ref
Iref A
where:
ref is the fluorescence quantum yield of the reference (Coumarin 1 in ethanol ref = 0.6420,21
)
sample in ethanol, A and Aref are the absorbances of the compound under the study and
reference sample at the excitation wavelength (366 nm), I and Iref are the integrated emission
intensity for the tested compound and reference sample, n and nref are the refractive indexes of
the solvents used for the compound and the reference, respectively.
The fluorescence lifetimes were measured using an Edinburgh Instruments single-
photon counting system (FLS920P Spectrometers). The excitation was provided by a
picosecond diode laser generating pulses of about 55 ps at 375 nm. The compound was
7

Journal Pre-proof
studied at concentration needed to provide absorbance of 0.10.3 in a 10 mm cell at 375 nm.
The fluorescence decays were deconvoluted by FAST program.
2.3. Measurement of skin resistance
The measurement of skin resistance can be used to confirm the vitality and tightness of
examined tissue sample22. In the experiment, an Ussing apparatus was used for the
23
measurements of transepithelial resistance of isolated skin specimens . The skin samples
were taken from the inner layer of the ears. Every skin specimen was subdivided into 2 cm2
pieces, which were submerged and incubated in the PB1 or PB1-1 solution. After 30 minutes
of incubation in RH (used as control fluid, which does not affect skin properties) or in PB1 or
PB1-1, the specimens were mounted in the ussing apparatus and filled with RH solution. The
electrical resistance of tissue (R) was determined by applying the current intensity of ± 10 μA
to the tissue stimulus and by measuring the corresponding voltage change according to the
Ohm's law.
Statistical analysis was performed using STATISTICA 13.1 (Statsoft, Poland). Values
of R for each group were compared to control group using the U-Mann-Whitney test, with p
value < 0.05.
2.4. Conjugation of fluorochrome with protein
2.4.1 General protein labeling procedures for PB1
Concanavalin A (ConA) and albumin (HSA) were conjugated with a PB1-fluorophore
in phosphate-buffered saline (PBS, 10mM phosphate, 150mM NaCl, pH 7.4). Briefly, two
milligrams of Con A or HSA was resuspended in 1 mL of buffer solution and 50 μl of 1
mg/ml PB1 in DMSO was added during vigorous mixing. The Schiff bases formed by amine-
aldehyde coupling were reduced by adding 10 µl of 5M sodium cyanoborohydride per 1 ml
reaction volume, and the reaction was allowed to proceed overnight at room temperature.
Then this solution was dialyzed with membrane against 2 l of PBS for 1 day in the dark at 4
°C to remove any free dye. The conjugated PB1-HSA/Con A solution retained in the dialysis
tube was collected and diluted in PBS to a protein concentration of 1 mg/ml.
2.4.2 General protein labeling procedures for PB1-1
In order to conjugate PB1-1 with HSA and Con A, the same preparation procedure as
above was used. However, the conjunction was reached without adding sodium
cyanoborohydride to the solution.
2.4.3. Quantitation of dye-protein conjugation
The concentration of dye and protein was determined by measuring the absorbance at
280 nm for Concanavalin A or albumin and at the maximum absorption wavelength of the
8

Journal Pre-proof
studied dye. A UV-Vis JASCO V-550 spectrophotometer was used to measure absorbance.
The resultant product was stored at 4 °C.
2.3.4. Fluorescence measurements
A PTI (Photon Technology International, Birmingham, NJ, USA) spectrofluorometer
was used for the measurement of steady-state fluorescence anisotropy. Measurements of
fluorescence anisotropy of dyes in protein solution were performed with the instrument in the
“L-format”, using excitation at 370 nm and 420 nm long-pass filters on the emission channel.
The use of long pass filter on emission channel ensures the separation of fluorescence from
scattered light. Both excitation and emission channels were equipped with computer-driven
motorized polarising filters. Triple fluorescence anisotropy measurements were conducted.
The anisotropy value applied to the below equation is the average of these measurements.
For determining the bound fraction (FB) of dye to protein, the following formula was
used15-17
:
푟−푟
퐹
퐹퐵 = 푟−푟 +푔(푟 −푟)
,
(2)
퐹
퐵
where rF and rB are the fluorescence anisotropies of the free and bound dye, r is anisotropy of
dye-protein conjugate and g is the ratio of maximum fluorescence intensities of dye upon
binding to protein and free forms of dye (quantum yield enhancement factor). On the basis of
Eq. (2) it is possible to estimate the fraction of bound dye also in the case of changing
quantum yield. Presented method was described in detail in our previous paper24.
The time-resolved spectrofluorometer Life Spec II (Edinburgh Instruments Ltd,
United Kingdom) with the sub-nanosecond pulsed EPLED® diode emitting light at a
wavelength of 280 nm was used in order to measure the fluorescence intensity and lifetime of
protein. The exposure time of samples was 30 seconds. The emission of protein was collected
from 300 to 400 nm. The fluorescence lifetime of protein was measured at a wavelength of
355 nm.
2.5. In vitro studies
2.5.1. Cell culture
3T3 (mouse fibroblasts) were cultured in DMEM/F12K medium (Dulbeco’s Modified
of Eagle’s Medium/F12, Corning,) supplemented with 10% FBS (Fetal Bovine Serum
GOOD, Pan-Biotech, Germany) and 1% antibiotics (amphotericin B Corning,
penicillin/streptomycin HyClone). Prior to any assays cells were in culture for at least 3
passages after storage in liquid nitrogen. Cells were cultured in 37°C, 5% CO2 and 99%
humidity. Passages including those prior to any assays were performed with the use of 0,25%
9

Journal Pre-proof
tripsin EDTA (Corning), PBS. During the passage procedure cells were centrifuged using
350xg/rcf (AFI LISA). Cell counting for passage needs was performed with a trypan blue
assay (Corning) and a standard Neubauer hemocytometer. All procedures were performed
under a II level laminar flow cabinet (Telstar bio advanced II).
2.5.2. Cytotoxicity assay
A standard MTT assay was used for the determination of compound cytotoxicity
(performed according to ISO 10993-5:2009 guidelines). Cells were seeded on 96 well flat
bottom plates (Sterile, Falcon® 96-well Clear Flat Bottom TC-treated Culture Microplates) at
a density 15625 cells/cm2 of growth area. After 24h of culture, using the same medium as
described in the section “Cell Culture”, the medium was discarded and all wells were rinsed
with PBS. 100μl tested solutions diluted in the growth medium were added (concentrations
tested: 10%, 1%, 0,1%, 0,01% v/v) and cells were incubated for 24h. After the incubation all
wells were rinsed twice with PBS and 50μl of filtered MTT solution (Thiazolyl Blue
Tetrazolium Bromide, Sigma,) was added at a concentration of 1mg/ml and incubated for 2h.
After the incubation the MTT solution was discarded and the formazan crystals were
dissolved in 100μl of DMSO (ACS reagent, ≥99.9%, Sigma). Absorbance of samples at
570nm (together with a reference absorbance of 655nm) was determined using Varioskan Sky
(Thermo). Blank absorbance – pure DMSO – was subtracted from all other measured
absorbance values before any further calculations. Cell survivability in tested samples was
calculated as a % of the average absorbance of the control. Statistical analysis was performed
using ANOVA test with Tukey`s post-hoc (Cl of 95% was used). The assay was performed at
least twice at different time points for every tested compound and each tested compound
concentration had at least 6 well replications on each test plate including the control. Samples
were randomized during the assay and decoded after the analysis.
2.5.3. Fluorescence spectrum analysis
Fluorescence excitation and emission maxima were determined using Varioscan Lux
and a μDrop plate to ensure correct transmission when using UV excitation. Excitation
wavelengths between 200nm and 400nm were used in conjunction with emission analysis
between 270nm to 700nm. Blank sample intensity – PBS – was subtracted from all results.
2.5.4. Fluorescence microscopy
The usefulness as potential fluorescence stains was determined using 3T3 cells. Tests
were performed both on fixated cells and on live cells. Olympus CX53FL inverted
fluorescence microscope was used to visualize and analyze the staining. Images were acquired
with a dedicated Olympus UC90 camera in a monochromatic mode using Olympus Cellsens
10

Journal Pre-proof
standard 2.2 software. A software level background subtraction was carried out on all
acquired images to ensure a low background level. Since the camera was used in a
monochromatic mode, pseudo-color was applied to the images and the final color is not
representative of the actual emission spectrum.
2.5.5. Staining fixated cells
All staining procedures were performed using optimal low-light conditions to ensure
minimal photo-bleaching. 3T3 cells were cultured on dedicated rounded coverslips in 24 well
places (Sterile, Falcon® 24-well Clear Flat Bottom TC-treated Culture Microplates) at a
density 15625 cells/cm2 of growth area (cell culture conditions identical to described in
section 1). After 24h of culture, the medium was discarded. After each step cells were rinsed
twice with a sufficient amount of PBS (Phosphate Buffered Saline, Corning). Cell fixation
was performed using 100μl 4% methanol-free buffered formaldehyde solution with a
incubation time of 15 min. Cell membrane permeabilization was performed using a 0,1%
Triton-X solution in PBS prior to staining. F-actin cytoskeleton counterstaining was
performed using 100μl solution of commercially available Phalloidin-iFluor™ 488
(Phalloidin-iFluor™ 488 Conjugate, Cayman Chemical) with the concentration and
incubation time recommended by the manufacturer and an incubation time of 15 min. 100μl
of stock tested solutions were added onto the separated coverslips and incubated for 15min. F-
actin and tested solutions staining was also performed in a different order to ensure that the
staining sequence does not affect the final outcome. After the final staining the coverslips
were mounted onto glass slides (Thermo, Superfrost Plus) with Aqua-Poly/Mount
(Polysciences, Inc) and VECTASHIELD® Hardset™ (Vector Laboratories) to check if
different mounting media affect the staining performance and stability over time.
2.5.6. Staining live cells
All staining procedures were performed using optimal low-light conditions to ensure
minimal photo-bleaching. Cells were seeded on 96 well plates (Sterile, Falcon® 96-well Clear
Flat Bottom TC-treated Culture Microplates) at a density 15625 cells/cm2 of growth area. Cell
culture conditions were identical to described in section 2.3.1. After each step cells were
rinsed twice with a sufficient amount of PBS (Phosphate Buffered Saline, Corning). After 24h
of culture, the medium was discarded and tested compounds were added diluted in growth
medium (concentrations tested: 10%, 1%, 0,1%, 0,01% v/v). For control purposes, a medium
without the addition of tested compounds was used. After 24h cell fixation was performed
using 100μl 4% methanol-free buffered formaldehyde solution with the incubation time of
15min. F-actin cytoskeleton counterstaining was performed using 100μl solution of
11

Journal Pre-proof
commercially available Phalloidin-iFluor™ 488 (Phalloidin-iFluor™ 488 Conjugate, Cayman
Chemical) with the concentration and incubation time recommended by the manufacturer and
an incubation time of 15 min.
2.6. Theoretical calculation
All geometrical parameters of investigated molecules in their ground (Sg) and excited
states (SCT) were calculated using density functional theory (DFT) approach implemented in
25
Gaussian 09 program package with TIGHT threshold option and PBE0/6-311++G(d,p)
basis set. In order to verify that all the structures correspond to the minima on the potential
energy surface, the analysis of Hessians was performed. The electronic properties,
characterized by computations of the vertical absorption and emission spectra, were obtained
using the time-dependent density functional theory (TDDFT/PBE0) 26 and by including the
state-specific (SS) corrected linear response (cLR) approach 27. In all spectroscopic
calculations the solvent effects were taken into account using the Integral Equation Formalism
for the Polarisable Continuum Model (IEF–PCM) 28,29
.
The binding properties of considered dyes were studied by performing series of
AutoDock 4.2 30-32 and AutoDock Vina 33 simulations. As in the case of studies regarding the
PB1 molecule, for each active complex comprising ConA and HSA, their crystal structures
were taken from PDB ID: 2a7a34 and 1AO635, respectivelly. The cubic grid box with the
dimensions of 16 Å and a grid spacing of 1 Å was set up in such way that the reactive -NH2
groups of Lysine and -SH of Cysteine were at its centre. In order to identify appropriate
binding energy and conformation of compounds, the Lamarkian genetic algorithm was
employed. For each atom of the receptor molecule Gasteiger charges were calculated. The
investigation of the binding site was performed using a united-atom scoring function. For
each amino acid the docking simulations were performed tenfold.
The biological activities were simulated using a combination of the 3D/4D QSAR
BiS/MC and CoCon algorithms 36,37
.
3. Results and discussion
3.1. Spectroscopic measurements
During our previous work7 on developing new fluorophores for bioimaging applications,
compounds based on phenathoimidazole skeleton possessing different functional groups were
designed and studied by means of first principle methods. All tested probes were
characterized by high fluorescent quantum yield and good chemical stability, and should be
considered as soft molecules with very high reactivity, intramolecular charge transfer
transition and a strongly polar low-lying CT excited state7. Additionally, the quantum
12

Journal Pre-proof
chemistry calculations showed that a key compound for binding to biomaterials is a
phenanthroimiazole derivative containing N-hydroxysuccinimide residue. Thus, to verified
the theoretical results the 2'-(1H-phenanthro[9,10-d]imidazol-2-yl)-phenyl-4-carboxylic acid
N-hydroxysuccinimide ester was synthesized, in a straightforward way, through a direct
reaction between 2'-(1H-phenanthro[9,10-d]imidazol-2-yl)-phenyl-4-carboxylic acid and the
N-hydroxysuccinimide, and tested as a potential biomedical imaging marker. The expected
1
structure of the probe was proved by H NMR and IR spectroscopy. For example, the
formation of the ester was confirmed by strong bands observed in IR spectra at ca. 1700–1800
cm-1 due to C=O lactone and a disappearance of the signal from the carboxylic group at ca.
12.8 ppm and the appearance of new signals of the N-hydroxysuccinimide at 2.94 ppm in the
1H NMR spectra.
The optical properties of the synthesized compound were studied in solvents of different
polarity. The electronic absorption and the fluorescence spectra for the tested compound are
shown in Figures 1 and 2, whereas, the spectral data such as the values of the absorption and
fluorescence maximum positions, molar absorption coefficients, Stokes shifts and
fluorescence quantum yields for the phenanthroimidazole derivative are collected in Table 1.
The shape and the intensity of the electronic absorption and fluorescence spectra of
PB1-1 is similar to the previously describe compounds8,38. Independent on solvent the spectra
display a broad band with maximum in the UV region (at ca. 375 nm) and high molar
absorption coefficient ranging from 28 700 M-1cm-1 to 37 800 M-1cm-1, while the emission
wavelengths are located in the visible region (425-478 nm). In consequence, large Stokes
shifts are observed (3137-6034 cm-1), which suggests significant differences (vibrational,
electronic, geometric) between the Franck-Condon state and the excited state39. Furthermore,
the compound is high fluorescent in all the solvents with a quantum yield higher than 0.57. If
we compare the tested PB1-1 with previously described compound containing aldehyde group
in the para position of the phenyl ring, we can notice that the ester displays lower Stokes
shifts and higher quantum yields.
Considering the solvent effect on the photophysical properties of the fluorophore, we
noted that emission band is red-shifted with increasing solvent polarity. This exemplified by
the increase of emission maximum from 436 nm to 478 nm on going from 1,4-dioxane to
acetonitrile. On the contrary, the position of the -* absorption is not significantly shifted.
We observed shift of about 4 nm for the same solvents. This indicates less polar character of
the compound in the ground state. Moreover, such a solvatochromic behavior is characteristic
for photoinduced intramolecular charge transfer in the lowest excited state 40,41. Additionally,
13

Journal Pre-proof
the substantial red shift of the emission spectra and Stokes shift in polar solvents suggest an
increase of the dipole moment in the excited state in respect to that of the ground state.
To explain the solvation of the probe by the solvents, four different parameters of Catalán
scale were used. According to the Catalán theory the change of the properties (Y) including,
e.g. the shift of the absorption (Ab) or emission band (Fl) maxima or Stokes shifts (St) or
lifetimes (Fl) are a linear combination of four solvent-dependent parameters as shown in Eq.
(3).
y  y0  aSPSP  bSdPSdP  cSASA  dSBSB
(3)
where: y0 is the property of substance in interest in the absence of solvent e.g. in the gas-
phase, SP, SdP, SA and SB are the solvent parameters that characterize the polarizability,
dipolarity, acidity and basicity of the solvent, respectively. The aSP, bSdP, cSA, and dSB are the
corresponding coefficients.
The Catalán coefficients for the spectra properties were obtained by multilinear
regression analysis. Table 3 summarizes the estimated values of the coefficients y0, aSP, bSdP,
cSA, and dSB and the regression coefficient values (R-Square).
It should be noted that reasonable correlation has been found between spectral parameters
and the four parameters scale proposed by Catalán. The absorption maximum is red-shifted
with increasing both polarizability and basicity of the environment and also is blue-shifted
with increasing solvent dipolarity. However, low values of the solvent acidity and basicity
coefficients and large standard errors indicate that these solvent parameters do not exactly
affect the absorption maximum. Hence, the results indicate that the absorption arises from
polarized π-π* transition in the molecules.
The fluorescence band position undergoes a marked red shift with increasing
polarizability, and dipolarity of the solvents and also a weak shift with increasing solvent
acidity. Moreover, the basicity of the medium causes a blue shift in fluorescence maximum.
However, the coefficients of the polarizability (SP) and acidity (SA) terms in eq. (3) are
subject to a large error so the fluorescence spectra are mainly affected by dipolarity and
basicity of the environment.
Besides fluorescence band position also Stokes shift is affected by solvent dipolarity and
basicity to a great extend. These parameters have the higher values with smaller error but
contrary to fluorescence band position, the bSdP coefficient value has positive sign indicating
decrease in the shift with increasing solvent dipolarity while the dSB coefficient is negative
indicating increase in the Stokes shift.
14

Journal Pre-proof
The change of the solvent polarity influences also the fluorescence quantum yield and
fluorescence lifetimes. The fluorescence decays of PB1-1 in tested solvents are shown in
Supplementary Information. They were reconvoluted based on exponential component
analysis using FAST program. The decays of the PB1-1 in all solvents consist of one, quit
long component, ranging from 1.8 ps to 2.3 ps (Table 3). Mono-exponential fluorescence
intensity decay allows correlation of the fluorescence lifetime with the solvent parameters
using eq. (3). The multiparametric fits were mainly dependent on SdP and in some extent on
SA and SB. The statistic was very bad for solvent dipolarity. The positive sign in the case of
SdP and SA indicates decrease in the fluorescence lifetime with increasing solvent dipolarity
and acidity whereas a weak elongation of the decay time is observed with increasing solvent
basicity as indicated by negative sign of the coefficient.
In general, the fluorescence quantum yield calculated from the area of total fluorescence
emission over the whole spectral range shows a decrease in quantum yield with the increase in
solvent polarity which better stabilize the ICT. However, unexpectedly, the trend is not
followed for dibutyl ether, acetone and acetonitrile. In acetonitrile and acetone PB1-1 shows
an increase in Fl while the fluorescence intensity is reduced in dibutyl ether. This inverse
phenomenon could be explained by a reduction of twisting of the single bonds involved in the
charge transfer in the excited state. On the other hand, a significant decrease of the quantum
yield in butyl ether may arise form an aggregation effect.
Based on the lifetimes and quantum yields of fluorescence, the values of radiative (kr) and
nonradiative (knr) rate constants were calculated using the following relations:
Fl

(4)
(5)
kr 

1 Fl

knr 

and the parameters obtained are tabulated in Table 3. Clearly, there is a decrease in the kr/knr
ratio with increasing polarity of the medium pointing to the increasing role of radiative
transition which is consistent with previous results described for phenanthroimidazole
derivative with aldehyde group in para position of phenyl ring.
3.2. Measurement of skin resistance
The median value of electrical resistance of the examined rabbit ear skin specimens
was 7632 Ω/cm2 (25 quartile = 955 Ω/cm2, 75 quartile = 22 644 Ω/cm2) for the control group
(n=23), 19 410 Ω/cm2 (25 quartile = 8009 Ω/cm2, 75 quartile = 27 870 Ω/cm2) for the PB1
15

Journal Pre-proof
group, and 4449 Ω/cm2 (25 quartile = 1647 Ω/cm2, 75 quartile = 9270 Ω/cm2) for the PB1-1
group. There was a statistically difference between the control and PB1 groups, with p value
0.043093. The values were not statistically different for the PB1-1 group compared to the
control group, with p value 0.463618.
The electrical resistance of the skin fragments studied in vitro seems to depend on the
vitality of the tissue, its structure, functions and also the ability to transport ions through the
cells and between them 12-13. In the experiment, the highest resistance of mammalian skin was
not changed by incubation in the PB1-1 solution, but it raised after PB-1 solution application.
Thus, it could be concluded that PB1-1 does not change the electrical properties of
mammalian skin, as well as does not have an impact on the skin tightness. But the contact
with PB-1 solution changes the ion transport on the skin surface and may influence the skin
cell adhesion.
3.3. Protein labelling
The PB1 probes with reactive aldehyde groups formed Schiff bases with the primary
amines of HSA and Con A. The reaction efficiency and bond stability between aldehydes
(PB1) and primary amines was increased by the addition of sodium cyanoborohydride. The
emission maxima of unbound PB1 and PB1-1 in PBS were at 495 and 448 nm, respectively.
After binding of the PB1 to Con A and HSA, a 10 nm blue shift in wavelength of maximum
emission was observed. Spectral changes were the result of the reduction of Schiff bases after
the addition of sodium cyanoborohydride. In the case of PB1-1, the changes in maximum of
fluorescence were not revealed, as this dye was linked with the protein without using sodium
cyanoborohydride as the reducing agent. The intensity of the emission of HSA/Con A
conjugate of PB1-1 was by a factor of 3 higher compared to PB1 conjugate.
To determine the interaction between dyes and proteins, the method of fluorescence
quenching of protein in the presence of dyes was used, which is a practical and simple
method. The intensity of HSA and Con A fluorescence decreased rapidly with the addition of
the studied dyes. PB1-1 and PB1 was found to yield about 96 and 93% albumin fluorescence
quenching for HSA, respectively, while for Con A, quenching is slightly smaller and is equal
to 94 and 91%, respectively. In order to clarify the nature of the quenching of fluorescence
(static or dynamic), time-resolved fluorescence was applied. A change in the fluorescence
lifetime of proteins after adding the dyes has not been observed, which indicates that the
decrease in fluorescence is related to static quenching, responsible for binding. The analysis
of spectroscopic data for proteins leads to the conclusion that PB1 and PB1-1 are strongly
bound with HSA and Con A.
16

Journal Pre-proof
In order to verify that dyes are bound to studied proteins after dialysis, fluorescence
anisotropy was also measured. Fluorescence anisotropy measurements are particularly useful
in determining bound fraction of a dye which possesses high affinity. As a result of binding of
the dye and the protein, the possibility of rotational motion of molecules significantly
decreases and the value of anisotropy increases. This increase is due to the large size of the
conjugate relative to the free dye. This method has been described in detail in our earlier
paper 13.To calculate the factor and anisotropy of the bound dye, the intensity and anisotropy
of fluorescence of dye at saturation regime were also measured. Then, the bound fraction of
these dyes was calculated according to Eq.1. The fraction of bound dyes with HSA and Con A
and the degree of labelling were presented in Table 4. The degree of labelling was estimated
on the basis of the absorbance of the dye and protein. One can see that the fraction of PB1-1
bound to HSA and Con A was larger than in the case of PB1. It can be concluded that dyes
are bound to proteins almost completely and the degree of labelling is slightly below 1. These
degrees of protein labelling seem sufficient for further experiments. Of course, high level of
labelling is desirable because of high sensitivity, however over-labelling above 1 can often
cause quenching as a result of energy migration between dyes (homoRET process)41 making it
difficult for controlling fluorescence intensity between experiments.
In order to examine the stability of dye-protein conjugate, anisotropy of fluorescence
was measured for 4 weeks with 2 day intervals. Then, the degree of binding (Fb) of dye to
protein was calculated according equation 1 (Fig. 3).
Experimental data revealed that there was no difference in binding for PB1-1 in time,
what is a promising sign for a serious application of this compound as a fluorophore for
biomolecules. The bound fraction of PB1 decreased a little bit over time up to 89% and 87 %
for HSA and Con A, respectively. The Schiff base formed with ordinary amines rapidly
hydrolyzes in PBS solution and must be reduced to an alkylamine (secondary amine) linkage
in order to stabilize. Our results showed that as a consequence of addition of the reducing
agents (NaCNBH3) adduct is formed. However, some molecules of PB1 are disconnected
from protein or some not-reduced Schiff bases form other products during time.
3.4. Cytotoxicity
All tested compounds did not exhibit cytotoxic properties towards 3T3 cells in
concentrations of 0,1-0,01% (one exception being PB1-HSA which did exhibit statistically
significant lower survivability at the 0,01% concentration). PB1-1-HSA did not exhibit
significantly lower survivability than the control even at the highest tested concentration (89%
survivability +/- SD 10,92). PB1-1 did achieve similar survivability results to PB1-1-HSA but
17

Journal Pre-proof
statistical analysis showed a significant difference at the 10% concentration (p=0,122). PB1
and PB1-HSA at 10% concentration level have shown survivability results below 70%, which
indicates high cytotoxic potential of the compounds at this concentration towards 3T3 cells.
Survivability of treated cells was also lowered significantly when using 1% concentrations of
those compounds.
3.5. Fluorescence spectrum analysis
All tested compounds exhibited at least two emission maxima. PB1-HSA presented
two emission maxima at 310-315nm and 420-425nm for excitation at 270nm. PB1-ConA
presented 320-325nm and 400-410nm emission maxima for 270nm excitation wavelength.
PB1-1HSA has shown 325-330nm emission maxima for 350nm excitation and 400-410nm
maxima emission for 270nm excitation. PB1-1-ConA has shown 320-325nm emission
maxima for 230nm excitation and 430-435nm maxima emission for 250nm excitation.
3.6 Staining fixated cells
All tested compounds were most likely incorporated into the outer cell membrane and
did not stain the F-actin cytoskeleton. Fluorescence signals of PB1-HSA and PB1-ConA were
unstable and severe diffusion to the mounting medium over a short period of time (overnight
mounting medium hardening) was observed when Aqua-Poly/Mount (Polysciences, Inc) and
VECTASHIELD® Hardset™ (Vector Laboratories) were used as mounting medium (increase
of background fluorescence and decrease of sample fluorescence). Both PB1-HSA and PB1-
ConA caused Phalloidin to loose most or all of its activity directly after incubation. Usable
images form PB1-HSA and PB1-ConA staining were obtained only when acquired from the
edge of the cover slip (Fig. 4). PB1-1-ConA and PB1-1-HSA presented similar performance,
however they did not affect F-actin staining to the same extend as PB1-HSA and PB1-ConA.
In their case a moderate stain migration to the mounting medium was also observed over time
(increase of background fluorescence and decrease of sample fluorescence).
3.7. Staining live cells
Compared to control, no detectable fluorescence was determined using U-FUW mirror
unit (UV excitation) for cells incubated with 0,1% and 0,01% v/v solutions of tested
compounds (results not shown due to lack of detectable fluorescence). Cells incubated with
10% solutions of PB1-ConA and PB1-HSA have shown the best results out of all tested
compounds, although fluorescence intensity was significantly lower compared to the control
stain. All 1% tested solutions, as well as 10% PB1-1-ConA and PB1-1-HSA have shown an
uneven, low but detectable fluorescence. No major morphological changes were observed
18

Journal Pre-proof
after the use of tested compounds, F-actin cytoskeleton was comparable with the one
presented with the control for all compounds (Fig. 5).
3.8. Molecular docking studies
In an earlier work7, focusing only on conjugation with ConA, it was shown that for
PB1 the lowest binding free energy (ΔGb= -5.84 kcal/mol) characterizes the sites with
LYS116 with an inhibition constant (Ki) of 0.79 mM. In the case of PB1-1, the site of active
binding to the protein is LYS30 with ΔGb= -5.44 kcal/mol and Ki = 0.77 mM. Despite the
close neighbourhood of other amino acids forming the active cavity, in both cases the
presence of H-bonds is not observed and the binding is not stabilized by the π-π stacking
interaction. However, molecular docking simulations have shown that after docking to the
ConA the planar PB1 molecule becomes non-planar, undergoing slight twisting on the C3-C4
(Table S2). The PB1-1 molecule undergoes a larger twist on this bond and the fracture of
planarity is enhanced by changes in dihedral angles within the ester part. Although the
geometric changes did not alter the length of PB1, in the case of PB1-1 the marker was
shortened in the axis of symmetry and the N1-N13 distance was reduced by 0.7 Å. Docking
with HSA revealed that LYS195 is the active site for both compounds (Table S3). PB1 is
characterized by a lower affinity for lysine (ΔGb= -9.2 kcal/mol), relative to PB1-1 (ΔGb= -
11.0 kcal/mol). At the same time, the affinity of both compounds for this protein is almost
twice as high as for ConA. This is also translated into other quantities, since the inhibition
constant decreases from 790.92 M and 770.99 M to 14.11 M and 3.35 M, for PB1 and
PB1-1, respectively. Moreover, the binding energy is significantly changed, decreasing from -
4.23 kcal/mol to -6.62 kcal/mol and from -4.25 kcal/mol to -7.47 kcal/mol. In both cases the
binding is not stabilized by the π-π stacking interaction. However, while no H-bonds are
observed for PB1, for PB1-1 they are formed between the N2 molecules and ARG222, as well
as between O11 and ARG257. By adapting to the active binding site, both molecules also
undergo significant geometric changes. When compared to ConA, the structure of PB1
undergoes identical structural changes. On the other hand, for PB1, the changes in dihedral
angles are smaller and in addition there is a significant straightening of the ester bridge. For
this reason, a slight elongation of the structure from the original one is observed (by 0.03Å).
For the deeper cognitive aspect, theoretical studies also included a phenanthimidazole
derivative containing a maleimide group as being actively bound to the protein. During
conjugation with the protein, this marker binds to the sulphur of cysteine HSA42. Molecular
docking studies indicated that the PB1-1 probe is characterized by an affinity to the protein
equal to -8.4 kcal/mol. It is the lowest among the considered cases for albumin, but
19

Journal Pre-proof
significantly higher than ConA. At the same time, the conjugate is characterized by a higher
binding energy (-7.51 kcal/mol) and a lower Ki value (3.1 M) relative to the PB1-HSA and
PB1-1-HSA conjugates. In the active binding site the presence of H-bonds is not observed and
the binding is not stabilized by the π-π stacking interaction. Because the similar structure of
PB4 (1(-4-(1H-phenanthro[9,10-d]imidazol-2-yl)phenyl)pyrrolidine-2,5-dione) differs from
PB1-1 by the lack of an ester bridge between the maleimide group and the probe (Table S1), it
should be assumed that the presence of an ester bridge slightly affects the increased affinity to
the protein, while reducing the binding energy in the probe-protein biocomplex. During
adaptation to the active cavity, the benzene ring rotates on the C3-C4 and C9-N10 bonds.
However, the position of the maleimide group remains unchanged relative to the initial
structure.
The approach of the fluorochrome particle to the surface of the protein causes an
increase in internal energy. The smallest ΔE difference between the initial structure and the
one obtained after molecular docking is obtained for the PB1 dye, for which the value is 0.52
kcal/mol, both for ConA and HSA. The highest value is obtained for PB1-1, where for both
proteins ΔE is 0.69 kcal/mol. In the case of PB4, this value is 0.60 kcal/mol. Conjugation with
the protein results in a significant reduction in the energy of the biocomplex. The largest
reduction in the energy of the system is observed for the PB1-concanavalin A conjugate,
where ΔE is 8.01 kcal / mol and for PB4-HSA, where ΔE is 7.58 kcal/mol. In turn, the lowest
ΔE is shown by the PB1-1-HSA conjugate (4.26 kcal/mol) and PB1-1-ConA (5.75 kcal/mol).
3.9. Biological properties
For a dye to act as a fluorescent probe, it must meet many specific physicochemical
features which affect the bioavailability. One of them is LogP which determines the
absorption and the permeability across cell membranes. Of the compounds discussed here, the
highest LogP value is shown by PB1-1 (5.22 ± 0.36). According to the Lipinski "5"43, it is
characterized by relatively poor bioavailability (LogP>5). In turn, the substitution of N-
hydroxysuccinimide ester with a maleimide group reduces LogP to 4.86, which significantly
improves permeability through cell membranes. For PB1 this value is close to PB4 and
amounts to 4.71. All markers are described by a high Log BCF value, which is -3.32, -4.76
and -4.50 (±0.25) for PB1, PB1-1 and PB4, respectively. This indicates unambiguously that
the tested probes will not be bioaccumulative in living organisms and should be excreted in
the urine. Replacement of an aldehyde group as the reactive one with an ester or maleimide
group does not affect the high metabolism by CYP450 3A4 and for each molecule and the
20

Journal Pre-proof
likelihood of metabolism is 99%. However, it raises significantly the likelihood of CYP450
2D6 metabolism, i.e. from 67% to over 92%.
Investigated markers are classified as class 4 or 5 in terms of the degree of toxicity and
can be considered as practically nontoxic (Table 5). In addition, only PB4 should not exhibit
hepatotoxicity, caricinogenicity, immunotoxicity, mutagenicity and cytotoxicity (Fig. 6).
Substitution of a maleimide group with an N-hydroxysuccinimide ester group results in the
visualization of the possibility of hepatotoxicity and mutagenicity. In the case of PB1, only
hepatotoxicity can be expected.
4. Conclusion
The aim of the present study was to prepare the HSA/ConA conjugates of PB1 and
PB1-1 which can provide important insight into the interactions between dyes and proteins.
The strong binding of PB1 and PB1-1 to HSA and ConA after performing the dialysis
process, indicated by the fluorescence quenching of protein, was confirmed by fluorescence
anisotropy spectroscopy. These results suggest that both PB1 and PB1-1 are bound to HSA
and Con A after performing the dialysis process.
Summarizing, the results presented in this paper allow drawing a conclusion that the
examined probes are very promising, especially the HSA/ConA conjugate of PB1-1. The
results obtained by fluorescence anisotropy measurements revealed such conjugation as a very
stable one. Moreover, these conjugates are characterized by strong and stable fluorescence. It
means that these probes can become excellent candidates for further pharmacological studies.
As it is demonstrated in the case of PB1-1-protein conjugate, significant stability can be
achieved without the use of reducing agents. It was also demonstrated that sodium
cyanoborohydride markedly enhances PB1-protein conjugate stability. However, a small
decrease in binding during time was noted.
Although the tested fluorescence compounds absorption/emision spectra are
overlapping with those of most commonly used stains like DAPI and Hoechst 33342 there is a
growing demand for stains possible to use in this channel that are not nuclei stains. Recently
introduced nuclear stains such as NucSpot® 488 or NucSpot® 650 by Biotium are a great
example of the market demand for this kind of product.
All tested compounds presented to be usable in the case of fixated cell staining. Using
higher compound concentrations and performing extra wash steps after staining could yield
better fluorescence staining performance and decrease noise levels. PB11-ConA and PB1-1-
HSA were characterized with the lowest cytotoxicity during the MTT assay, and thus should
be more suitable for live imaging applications than PB1A and PB1K. Unfortunately, imaging
21

Journal Pre-proof
of live stained cells of all samples shown low fluorescence levels. It might be caused by the
U-FUW dichroic mirror used during live sample imaging and the usage of a plastic F-bottom
plate absorbing light at the excitation spectrum. The fluorescence intensity could probably be
improved by using glass bottom plates and a tailored, dedicated fluorescence mirror unit.
Moreover, some of our researched compounds exhibited not only good performance
while staining fixed cells but also good performance during the live imaging tests. Days that
enable both visualization of fixed cells and live imaging are especially welcome in the
community. There is a growing interest in live imaging cells with compounds that enable
fluorescence visualization, companies such as Thermo scientific, BioTek, NanoEntek, and
Etluma present new, high performance and high throughput screening devices that base on
this premise and require a growing range of such compounds. Those compounds do not need
to exhibit high levels of specificity, since simple proliferation assays, invasion assays and
growth kinetic do not require such specificity but rather simplicity in the assay itself. The
researched compounds are also surprisingly easy to use since a standard DAPI fluorescence
cube enables visualization and is present in most devices for imaging as a basic and most
common element. Furthermore expanding our knowledge base about potential fluorescence
compounds is a necessary step in enabling this field of science to grow
Supporting Information Available: Analytical data, molecular structure and spectroscopic
properties for investigated compound.
Submission declaration
Submitting the article "2'-(1H-phenanthro[9,10-d]imidazol-2-yl)-phenyl-4-carboxylic acid N-
hydroxysuccinimide ester: a new phenanthroimidazole derivative as a fluorescent probe for
medical imaging applications" to the Spectrochimica Acta, Part A: Molecular and
Biomolecular Spectroscopyhas not been published previously, is not under consideration for
publication elsewhere, publication is approved by all authors and it will not be published
elsewhere including electronically in the same form, in English or in any other language,
without the written consent of the copyright-holder.
Conflict of interest
The authors declare that there are no conflicts of interest.
Declaration of interest
Submitting the article "2'-(1H-phenanthro[9,10-d]imidazol-2-yl)-phenyl-4-carboxylic acid N-
hydroxysuccinimide ester: a new phenanthroimidazole derivative as a fluorescent probe for
medical imaging applications" to the Spectrochimica Acta, Part A: Molecular and
22

Journal Pre-proof
Biomolecular Spectroscopyhas does not have any potential conflicts of financial and non-
financial interests.
Acknowledgements
This research was supported in part by PLGrid Infrastructure.
References
1. C.S. Levin, Primer on molecular imaging technology, Eur. J. Nucl. Med. Mol. I., 32 (2005)
325-345. https://doi.org/10.1007/s00259-005-1973-y
2. J. Bil, M. Winiarska, Postępy biologii komórki, 2007, 34 (2), 335-359.
3. T. Sołtysiński; Targets and strategies for imaging dynamics of molecular-cell interactions
in autoimmune disorders; Molecular Imaging 6, 2007, s. 365
4. D. Janczewski, Y. Zhang, G.K. Das, D.K. Yi, P. Padmanabhan, K.K Bhakoo, T.T Tan, S.T.
Selvan, Bimodal magnetic-fluorescent probes for bioimaging, Microsc. Res. Tech. 74 (2011)
563-576. https://doi.org/10.1002/jemt.20912
5. M. Tsuji, S. Ueda, T. Hirayama, K. Okuda, Y. Sakaguchi, A. Isono, H. Nagasawa, FRET-
based imaging of transbilayer movement of pepducin in living cells by novel intracellular
bioreductively activatable fluorescent probes, Org. Biomol. Chem. 11(18) (2013) 3030-3037.
https://doi.org/10.1039/c3ob27445d
6. F. Auzel, Upconversion and Anti-Stokes Processes with f and d Ions in Solids, Chem. Rev.
104(1) (2004) 139–174. https://doi.org/10.1021/cr020357g
7. P. Krawczyk, P. Czeleń, P. Cysewski, Reactive group effects on the photophysical and
biological properties of 2-phenyl-1H-phenanthro[9,10-d]imidazole derivatives as fluorescent
markers, Org. Biomol. Chem., 16 (2018) 3788-3800. https://doi.org/10.1039/c8ob00729b
8. M. Tian, C. Wang, L. Wang, K. Luo, A. Zhao, et al., Study on the synthesis and structure
effect relationship of multi-aryl imidazoles with their fluorescence properties, Luminescence,
29 (2014) 540-548. https://doi.org/10.1002/bio.2580
9. G. Weber, Polarization of the fluorescence of macromolecules. 2. Fluorescent conjugates of
ovalbumin and bovine serum albumin, Biochem J. 51 (1952) 155–167. DOI:
10.1042/bj0510155
10. G. Weber, Polarization of the fluorescence of macromolecules. 1. Theory and
experimental method, Biochem J. 51 (1952)145–155. DOI: 10.1042/bj0510145
11. J.R. Lakowic, Principles of Fluorescence Spectroscopy, Kluwer Academic/Plenum
Publishers, New York, 1999
12. D.M. Jameson, J.A. Ross, Chem. Rev. 2010, 110, 2685–2708. DOI: 10.1021/cr900267p
23

Journal Pre-proof
13. T. Wybranowski, B. Ziomkowska, M. Cyrankiewicz, S. Kruszewski, The impact of
oxidative stress on binding of drugs with plasma proteins studied by fluorescence anisotropy
methods, Gen. Physiol. Biophys. 37 (2018) 647-655. https://doi.org/10.4149/gpb_2018017
14. K. Ishii, M. Noda, S. Uchiyama, Mass spectrometric analysis of protein–ligand
interactions. Biophys Physicobiol.; 13 (2016) 87–95. DOI: 0.2142/biophysico.13.0_87
15. H. Hernandez, C.V. Robinson, Determining the stoichiometry and interactions of
macromolecular assemblies from mass spectrometry, Nat Protoc. 2 (2007) 715–726. DOI:
10.1038/nprot.2007.73
16. M.J. Chalmers, S.A. Busby, B.D. Pascal, G.M. West, P.R. Griffin, Differential
hydrogen/deuterium exchange mass spectrometry analysis of protein–ligand interactions.
Expert Rev Proteomics, 8 (2011) 43–59. DOI: 10.1586/epr.10.109.
17. S. Sanglier, C. Atmanene, G. Chevreux, A.V. Dorsselaer, Mass spectrometry to study
noncovalent protein/protein and protein/ligand complexes: technical aspects and application
to the determination of binding stoichiometries, Methods Mol Biol. 484 (2008) 217-243. DOI:
10.1007/978-1-59745-398-1_15.
18. M. Zabet-Moghaddam, A. L. Shaikh, L. B. Jones, S. Niwayama, Synthesis of d-labeled
and unlabeled benzoyloxysuccinimides and application to quantitative analysis of peptides
and a protein by isotope differential mass spectrometry, Bioorg. Med. Chem. Lett., 21 (2011)
4629-4632. https://doi.org/10.1016/j.bmcl.2011.05.090
19. M. Brouwer A., Standards for photoluminescence quantum yield measurements in
Solution (IUPAC Technical Report), Pure Appl. Chem., 83 (2011) 2213.
https://doi.org/10.1351/pac-rep-10-09-31
20. J. Olmsted, Calorimetric determinations of absolute fluorescence quantum yields, Journal
Of Physical Chemistry, 83 (1979) 2581-2584. https://doi.org/10.1021/j100483a006
21. R. Abdayem, S. Cellejon, P. Portes, P. Kiriloy, F. Demame, F. Pirot, V. Jannin, M.
Haftek, Modulation of transepithelial electric resistance (TEER) in reconstructed human
epidermis by excipients known to permeate intestinal tight junctions, Exp. Dermatol. 24
(2015) 686-691. https://doi.org/10.1111/exd.12750
22. I. Hołyńska-Iwan, J. Bogusiewicz, D. Chajdas, K. Szewczyk-Golec, M. Lampka, D.
Olszewska-Słonina, The immediate influence of deltamethrin on ion transport through rabbit
skin. An in vitro study, Pest. Biochem. Physiol. 148 (2018) 144-150.
https://doi.org/10.1016/j.pestbp.2018.04.011
23. T. Wybranowski, M. Cyrankiewicz, B. Ziomkowska, S. Kruszewski, The HSA affinity of
warfarin and flurbiprofen determined by fluorescence anisotropy measurements of
24

Journal Pre-proof
camptothecin,
Biosystems
94
(2008)
258–262.
https://doi.org/10.1016/j.biosystems.2008.05.034
24. M.J. Frisch, G.W. Trucks, G.B. Schlegel et al. Gaussian 09, Revision A.1, Gaussian, Inc.,
Wallingford CT, 2009.
25. C. Adamo, G.E. Scuseria, V. Barone, Accurate excitation energies from time-dependent
density functional theory: Assessing the PBE0 model, J. Chem. Phys. 111 (1999) 2889-2899.
https://doi.org/10.1063/1.479571
26. C. Guido, S. Caprasecca, How to perform corrected Linear Response calculations in G09,
https://www1.dcci.unipi.it/molecolab/tools/white-papers/pisalr/;
10.13140/RG.2.1.1903.7845.
2016.
DOI:
27. M.T. Cancés, B. Mennucci, J. Tomasi, A new integral equation formalism for the
polarizable continuum model: Theoretical background and applications to isotropic and
anisotropic
dielectrics,
J.
Chem.
Phys.
107
(1997)
3032-3041.
https://doi.org/10.1063/1.474659
28. M. Arivazhagan, P. Muniappan, R. Meenakshi, G. Rajavel, PCM/TD-DFT analysis of 1-
bromo-2,3-dichlorobenzene – A combined study of experimental (FT-IR and FT-Raman) and
theoretical calculations, Spectrochim. Acta Part A, 105 (2013) 497-508. DOI:
https://doi.org/10.1016/j.saa.2012.11.033
29. G.M. Morris, R. Huey, W. Lindstrom, M.F. Sanner, R.K. Belew, D.S. Goodsell, A.J.
Olson, AutoDock4 and AutoDockTools4: Automated docking with selective receptor
flexibility, J. Comput. Chem., 30 (2009) 2785-2791. https://doi.org/10.1002/jcc.21256
30. S. Cosconati, S. Forli, A.L. Perryman, R. Harris, D.S. Goodsell, A.J. Olson, Virtual
screening with AutoDock: theory and practice, Expert Opin. Drug Discovery, 5 (2010) 597-
607. https://doi.org/10.1517/17460441.2010.484460
31. S. Forli, A.J. Olson, A Force Field with Discrete Displaceable Waters and Desolvation
Entropy for Hydrated Ligand Docking, J. Med. Chem., 55 (2012) 623-638.
https://doi.org/10.1021/jm2005145
32. O. Trott, A.J. Olson, AutoDock Vina: Improving the speed and accuracy of docking with
a new scoring function, efficient optimization, and multithreading, J. Comp. Chem., 31 (2010)
455-461. https://doi.org/10.1002/jcc.21334
33. C. S. Panjikar, P.A. Tucker, M.S. Weiss, On the Routine Use of Soft X-Rays in
Macromolecular Crystallography, Part III- The Optimal Data Collection Wavelength, Acta
Crystallogr. D Biol. Crystallogr., 61 (2005) 1263-1272. https://doi.org/10.2210/pdb2a7e/pdb
25

Journal Pre-proof
34. S. Sugio, S. Mochizuki, M. Noda, A. Kashima, Crystal structure of human serum
albumin, Protein Data Bank, Rutgers University, 1998. https://doi.org/10.2210/pdb1ao6/pdb
35. V. Potemkin, M. Grishina, A new paradigm for pattern recognition of drugs, J. Comput.
Aided. Mol. Des., 22(6-7) (2008) 489-505. https://doi.org/10.1007/s10822-008-9203-x
36. V. Potemkin, A.A. Pogrebnoy, M.A. Grishina, Technique for Energy Decomposition in
the Study of “Receptor-Ligand” Complexes, J. Chem. Inf. Model., 49(6) (2009) 1389-406.
https://doi.org/10.1021/ci800405n
37. P. Krawczyk, B. Jędrzejewska, M. Pietrzak, T. Janek, Synthesis, spectroscopic,
physicochemical properties and binding site analysis of 4-(1H-phenanthro[9,10-d]-imidazol 2
yl) benzaldehyde fluorescent probe for imaging in cell biology: Experimental and theoretical
study J. Photochem. Photobiol. B, 164 (2016) 112-122.
https://doi.org/10.1016/j.jphotobiol.2016.07.044
38. P. Krawczyk, B. Jędrzejewska, M. Pietrzak, T. Janek, Synthesis, photophysical properties
and systematic evaluations of new phenanthroimidazole fluorescent probe for bioimaging:
Experimental and theoretical study, J. Photochem. Photobiol. B, 166 (2017) 74-85.
https://doi.org/10.1016/j.jphotobiol.2016.11.008
39. R. Vabre, M. Legraverend, S. Piguel, Synthesis and evaluation of spectroscopic properties
of newly synthesized push–pull 6-amino-8-styryl purines, Dyes Pigm., 105 (2014) 145-151.
https://doi.org/10.1016/j.dyepig.2014.01.025
40. K. Panthi, R. M. Adhikari, T. H. Kinstle, Aromatic Fumaronitrile Core-Based
Donor−Linker−Acceptor−Linker−Donor (D-π-A-π-D) Compounds: Synthesis and
Photophysical Properties, J. Phys. Chem. A, 114 (2010) 4542-4549.
https://doi.org/10.1021/jp9115589
41. M. Brinkley, A brief survey of methods for preparing protein conjugates with dyes,
haptens and crosslinking reagents, Bioconjugate Chem., 3 (1992) 2-13.
https://doi.org/10.1021/bc00013a001
42. G. Hungerford, J. Benesch, J. F. Mano, .L. Rui, Effect of the labelling ratio on the
photophysics of fluorescein isothiocyanate (FITC) conjugated to bovine serum albumin,
Photochem. Photobiol. Sci. 6(2) (2007) 152-158. https://doi.org/10.1039/b612870j
43. Ch. Lipinski, F. Lombardo,B.W Dominy, P.J. Feeney, Experimental and computational
approaches to estimate solubility and permeability in drug discovery and development settings
1PII of original article: S0169-409X(96)00423-1, Adv. Drug. Deliv. Rev. 23 (1997) 3-37.
ttps://doi.org/10.1016/s0169-409x(00)00129-0
26

Journal Pre-proof
Tables
Table 1. The values of the absorption and fluorescence maximum positions, molar absorption
coefficients, Stokes shifts and fluorescence quantum yields for the PB1-1
ε [M-1cm-1]
Fl
max
mAbax [nm]

[nm] Fl [nm]
 St [cm-1]
Solvent
PB1-1
436
425
429
447
447
466
472
472
478
1,4-Dx
Bu2O
375
375
373
373
374
372
373
375
371
35300
-
0.761
0.657
0.851
0.731
0.718
0.754
0.577
0.574
0.743
3731
3137
3500
4438
4367
5422
5623
5480
6034
Et2O
37800
30600
32500
32200
32600
28700
32000
EtOAc
THF
MeAc
DMF
DMSO
MeCN
Table 2. Estimated from eq. (2), coefficients (y0, aSP, bSdP, cSA, and dSB), their standard errors and
correlation coefficients (R-Square) for the multiple linear regression analysis of Ab, Fl, St and Fl of
PB1-1 in 8 solvents as a function of the Catalán four-parameter solvent scale
Intercept
SP
SdP
SA
SB
Statistics
Standard
Error
Standard
Error
Standard
Error
Standard
Standard
Error
Adj. R-
Square
Value
Value
Value
Value
Value
Error
582.6
3614.4
3737.0
1.445
λAb
λFl
St
27491.2
24439.5
3051.7
2.294
121.7
754.7
780.3
0.302
-1135.4
-1901.8
766.4
204.0
1265.4
1308.3
0.506
249.3
40.5
251.2
259.7
0.100
-158.8
-514.8
356.0
2.090
-156.5
1625.9
95.4
592.1
612.2
0.237
0.940
0.975
0.976
0.911
-2759.2
3008.5
0.474
-1782.4
-0.462
-0.374

27

Journal Pre-proof
Table 3. Fluorescence lifetimes (), radiative (kr) and non-radiative (knr) rate constants of PB1-1 in
different solvents
 2
kr (108 s-1)
knr (108 s-1) kr / knr
 (ns)
Solvent
PB1-1
3.83
1.28
4.15
3.63
3.56
3.32
2.50
2.50
1,4-Dx 1.986
1.134
1.588
1.149
1.242
1.311
1.359
1.632
1.228
1.20
1.90
0.73
1.33
1.40
1.08
1.83
1.86
3.2
0.7
5.7
2.7
2.5
3.1
1.4
1.3
Bu2O
Et2O
1.809
2.053
EtOAc 2.015
THF
2.018
2.273
DMF
DMSO 2.311
MeCN 2.294
Table 4. The bound fraction and degree of labelling for studied conjugates after dialyses process
Conjugates
PB1-1-HSA
PB1-1-Con A
PB1-HSA
Bound fraction [%]
99,3 %
Degree of labelling
0.95
0.94
0.94
0.92
96.2%
95.1%
93.8 %
PB1-1-Con A
Table 5. Rat acute toxicity. In brackets there is given the class in terms of the degree of toxicity
LD50 (mg/kg)
Intraperitoneal
route of
administration
Subcutaneous
Intravenous route
of administration
Oral route of
administration
route of
administration
676 (4)
PB1
PB1-1
PB4
508 (5)
408 (4)
826 (5)
99 (4)
201 (4)
155 (4)
2855 (5)
1707 (4)
2087 (5)
965 (4)
899 (4)
28