Selective and Sensitive Sensing of Free Bilirubin in Human Serum Using Water-Soluble Polyfluorene as Fluorescent Probe

Article abstract of DOI:10.1021/acs.macromol.5b00043

The adherence of serum protein on conjugated polymer is a major bottleneck in the application of the latter for selective sensing of small biomolecules in blood serum. In this report, we present new polyfluorenes with d-glucuronic acid appendage that is a nonreceptor for any serum protein, thereby providing a platform for selective sensing of free bilirubin in the clinically relevant range of <25 to >50 μmol/L in human blood serum. The appended d-glucuronic acid formed noncovalent interactions with bilirubin, which in conjunction with favorable spectral overlap between the polymers and bilirubin facilitated efficient FRET process in aqueous solutions. Addition of bilirubin resulted in the quenching of the polyfluorene emission with simultaneous appearance of bilirubin emission exhibiting visual emission color change from blue to light green. The polymer remained stable in serum even under severe basic conditions and exhibited high selectivity with visual sensitivity only toward free bilirubin in human serum in the presence of crucial interferences such as hemoglobin, proteins, biliverdin, glucose, cholesterol, and metal ions. Nanomolar sensing of bilirubin could also be demonstrated successfully using one of the d-glucuronic acid appended polymer (PF-Ph-GlcA), which could sense ～150 nm of bilirubin in human serum. The combined role of energy transfer and noncovalent interaction highlights the potential of the new polymer design for highly selective sensing activity in complex biofluids.

Full text of DOI:10.1021/acs.macromol.5b00043

Article
pubs.acs.org/Macromolecules
Selective and Sensitive Sensing of Free Bilirubin in Human Serum
Using Water-Soluble Polyfluorene as Fluorescent Probe
T. Senthilkumar† and S. K. Asha*,†,‡,§
†Polymer Science and Engineering Division, CSIR-National Chemical Laboratory, Dr Homi Bhabha Road, Pune 411008, India
‡Academy of Scientific and Innovative Research, New Delhi, India
§CSIR-Network Institutes of Solar Energy, New Delhi, India
S
* Supporting Information
ABSTRACT: The adherence of serum protein on conjugated polymer
is a major bottleneck in the application of the latter for selective sensing
of small biomolecules in blood serum. In this report, we present new
polyfluorenes with D-glucuronic acid appendage that is a nonreceptor for
any serum protein, thereby providing a platform for selective sensing of
free bilirubin in the clinically relevant range of <25 to >50 μmol/L in
human blood serum. The appended D-glucuronic acid formed
noncovalent interactions with bilirubin, which in conjunction with
favorable spectral overlap between the polymers and bilirubin facilitated
efficient FRET process in aqueous solutions. Addition of bilirubin
resulted in the quenching of the polyfluorene emission with
simultaneous appearance of bilirubin emission exhibiting visual emission
color change from blue to light green. The polymer remained stable in
serum even under severe basic conditions and exhibited high selectivity with visual sensitivity only toward free bilirubin in human
serum in the presence of crucial interferences such as hemoglobin, proteins, biliverdin, glucose, cholesterol, and metal ions.
Nanomolar sensing of bilirubin could also be demonstrated successfully using one of the D-glucuronic acid appended polymer
(PF-Ph-GlcA), which could sense ∼150 nm of bilirubin in human serum. The combined role of energy transfer and noncovalent
interaction highlights the potential of the new polymer design for highly selective sensing activity in complex biofluids.
The detection of free bilirubin in serum is considered as one
of the potential routes for diagnosing liver disorders.28 Liver
disorders result in increment of free bilirubin concentration
from normal level of <25 μmol (<1.2 mg/dL) to >50 μmol/L
(>2.5 mg) in blood indicative of jaundice condition.29 Excess
level of free bilirubin is severely toxic to human body because it
can accumulate in body organs and causes brain hemorrhage by
crossing the blood−brain barrier.30 Considering the drawbacks
in the currently practiced diazo test method for free bilirubin
estimation,31 fluorometric biosensors offering amplified sensi-
tivity and selectivity with a wide window of analyte
concentration range are highly desirable. Reports on fluoro-
metric detection of bilirubin in human blood serum are very
sparse. However, very recently, Au nanoparticle conjugated
HSA protein as fluorometric biosensor for the sensing of
bilirubin in blood serum was reported.32 They made use of
quenching of the emission of human serum albumin (HSA)
stabilized gold nanoclusters upon interaction with bilirubin for
detection of free bilirubin. In our current report we present a
FRET based visual fluorescence color change (from blue to
light green) method of detection of free bilirubin in human
INTRODUCTION
■
Exploration of water-soluble conjugated polymers remains an
active area of research for their applicability in chemo-1−3 and
biosensors,4−10 targeted delivery,11,12 cell imaging,13−15 and
logic gates.16,17 Solubility in aqueous medium is the prime
prerequisite for a conjugated polymer to find desired potential
in bioapplications. Generally, the rigid conjugated polymers are
functionalized with either charged side chains or nonionic
neutral side chains with high polarity to achieve solubility in
aqueous medium. The hydrophilic pendant groups aid in
compensating the hydrophobic nature of the polymer back-
bone, thereby enhancing water solubility and also facilitating
interactions with specific targeted analytes.18 Functionalization
of ionic pendants19−21 induces remarkable water solubility, but
selectivity is compromised in the presence of multiple biological
and nonbiological substances such as proteins22 and nucleic
acids23 due to nonspecific electrostatic interactions.24 Con-
jugated polymers with charged pendants are known to show
differential binding toward serum proteins that are used to
develop sensing array for proteins.25 Substitution of neutral side
chains of sugar units such as glucose, mannose, etc., also act as
receptors toward proteins.26,27 Consequently, selective sensing
of small biomolecules like free bilirubin in the presence of
serum proteins poses significant challenges to a conjugate
polymer based probe.
Received: January 7, 2015
Revised: May 4, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.macromol.5b00043
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
AVENS 200 and 400 MHz spectrometer. Chemical shifts are reported
in ppm at 25 °C using CDCl3 and MeOH-d4 as solvents containing a
trace quantity of tetramethylsilane (TMS) as internal standard. The
purity of the compounds was determined by HR-MS or MALDI-TOF
in combination with elemental analysis. The MALDI-TOF analysis
was done on Voyager-De-STR MALDI-TOF (Applied Biosystems,
Framingham, MA) equipped with 337 nm pulsed nitrogen laser for the
purpose of desorption and ionization. A 2 μM solution of sample was
premixed with DHB (2,5-dihydroxybenzoic acid) matrix in THF and
mixed well before spotting on 96-well stainless steel MALDI plate
through dried droplet method for the analysis. Mass spectra were
measured with ESI ionization in MSQ LCMS mass spectrometer.
Elemental analysis was done by Thermofinniganflash EA 1112 series
CHN analyzer. HRMS (ESI) were recorded on ORBITRAP mass
analyzer (Thermo Scientific, Q Exactive). The molecular weights of
the polymers were determined by size exclusion chromatography
(SEC), which was performed using a Viscotek VE 1122 pump, a
Viscotek VE 3580 RI detector, and a Viscotek VE 3210 UV/vis
detector in DMF using polystyrene as standards. DLS measurements
were carried out on a Zetasizer ZS-90 apparatus, utilizing 633 nm red
laser (at 90° angle) from Malvern Instruments. The reproducibility of
the data was checked at least three times using independent polymer
solutions.
Photophysical Studies. Absorption spectra were measured using
PerkinElmer Lambda 35 UV−vis spectrophotometer. Steady-state
fluorescence studies and time-resolved fluorescence lifetime measure-
ments were recorded using Horiba Jobin Yvon Fluorolog 3
spectrophotometer having a 450 W xenon lamp for steady-state
fluorescence, and fluorescence lifetime decays were collected by a
time-correlated single photon counting (TCSPC) setup from IBH
Horiba Jobin Yvon (U.S.) using a diode laser (IBH, U.K., NanoLED-
320 and 375 L, with a λmax = 320 and 375 nm) having a full width at
half-maximum of 89 ps as a sample excitation source. The emission
and excitation slit width was maintained at 1 nm throughout the
experiments, and the data were obtained in “S1c/R1” mode (to
account for the variations in lamp intensity). All the absorption and
emission studies were conducted in dim light at the temperature of 15
°C in order to avoid structural isomerization of bilirubin, and the stock
solutions were stored below the temperature of 10 °C. Because of the
insolubility of bilirubin in water at pH = 7.4, all the bilirubin stock
solutions were prepared at pH = 10 using PBS buffer.
Isothermal Titration Calorimeter Experiment. ITC experi-
ments were carried out in Micro cal iTC-200 instrument. The titration
was carried out in PBS buffer in alkaline medium at 25 °C using an
isothermal titration calorimeter (Microcal iTC-200) with stirring at
1000 rpm. Polymer (200 μL with concentration of 2 mM) was taken
in the sample cell, and bilirubin (∼20 mM) was taken in the syringe. A
typical titration experiment consisting of 19 consecutive injections
each of 2 μL volume in the duration of 20 s at an interval of 180 s was
conducted. Heat of dilution (T) upon addition of bilirubin in solvent
was determined by performing a blank titration by injecting the
bilirubin solution into the solvent. To obtain the heat of binding
between polymer and bilirubin, the value of heat of dilution was
subtracted from the total value of heat of binding. A single set of
binding model was fitted with the binding isotherm which was used to
extract binding constant (K), binding stoichiometry (N), change of
enthalpy (ΔH) and the change of entropy (ΔS) for the binding.
Sensing of Free Bilirubin in Water. The stock solutions of the
polymers were prepared at 1 μM in water at pH = 10 using PBS buffer.
Bilirubin solutions of concentrations ranging from 1 × 10−6 to 1 ×
10−4 M in PBS buffer at pH = 10 were prepared and kept in dark at 15
°C. The bilirubin solution was added to the 1 μM solution of polymer
in water (pH = 10). The changes in the absorption and emission of the
polymer on addition of different concentrations of bilirubin were
recorded at 15 °C with the slit width of 1 nm.
blood serum sample. In our previous work, pentadecylphenol
(PDP)-functionalized polyfluorenes were employed as a model
for fluorometric detection of bilirubin in organic (THF)
medium.33 Our preliminary work highlighted the potential of
developing a water-soluble polyfluorene which could take
advantage of the favorable photophysical properties of
polyfluorene−bilirubin combination for the detection of
bilirubin. The advantage is expected to be twofold if the
moiety incorporated into polyfluorene backbone could also
additionally act as an interaction site for bilirubin.
D-Glucuronic acid is a vital water-soluble carbohydrate that
performs the key function of removing several insoluble toxic
substances through glucuronidation and subsequent urinal
excretion as glucuronides in human metabolism.34 In addition,
D-glucuronic acid is also a nonreceptor carbohydrate for the
serum proteins, thereby providing a platform for receptor-free
biosensing applications. In the liver, glucuronidation of bilirubin
(insoluble in water) plays a vital role in forming bilirubin
glucuronides (soluble in water) to avoid accumulation of toxic
unconjugated (free) bilirubin in body fluids.35 Thus, incorpo-
ration of D-glucuronic acid into polyfluorenes is expected to
serve the dual purpose of imparting water solubility to the
otherwise fully hydrophobic polyfluorenes while additionally
acting as an interacting site for bilirubin. As the competitive
interference of proteins can be eliminated via D-glucuronic acid
appendage, selective sensing of bilirubin in serum is expected to
be feasible. A polymer appended with the sugar unit D-glucose
was also designed as a nonbilirubin interacting reference
polymer.
In this report, new water-soluble conjugated polyfluorene
appended with D-glucuronic acid and D-glucose were
successfully designed and demonstrated for biosensing
application in human blood serum for the first time. The
structure and photophysical properties of the polymers were
characterized by relevant techniques. The appendage of
glucuronic acid imparted water solubility and amphiphilic
nature that could aid in the self-organization of the polymers.
Unfortunately, the water solubility of the glucose appended
polyfluorene was not very good. The polymers exhibited good
spectral overlap with bilirubin, thus fulfilling the criteria for
energy transfer process. Fluorescence sensing experiments were
carried out in human serum which contained hemoglobin,
proteins, cholesterol, glucose, and metal ions as crucial
interferences. The glucuronic acid appended polymer exhibited
high degree of selectivity toward bilirubin that was verified from
the poor quenching efficiencies of competing interferences such
as biliverdin, cholesterol, sugars, and metal ions.
EXPERIMENTAL SECTION
■
Materials and Methods. 2,7-Dibromo-9H-fluorene, D-glucuronic
acid, D-glucose pentaacetate, trifluoroacetic acid (TFA), bilirubin,
biliverdin, phosphate buffered saline (PBS), human serum, azido-
trimethylsilane, SnCl4, propargyl alcohol, Pd(PPh3)4, 1,4-benzene-
diboronic bis(pinacol ester), sodium methoxide, sodium hydride,
Pd2(dba)3, CsF, quinine sulfate, tetrabutylammonium bromide, 1,6-
dibromohexane, uridine diphosphate glucuronic acid sodium salt, and
bis(pinacolato)diboron were purchased from Aldrich Company Ltd.
and were used as received. Copper sulfate, sodium ascorbate, HCl,
acetic anhydride, iodine, sodium thiosulfate, potassium hydroxide,
glucose, sucrose, sodium bicarbonate, K2CO3, and ethanol were
purchased from Merck Chemicals Ltd. Tetrahydrofuran (THF),
acetone, petroleum ether, ethyl acetate, DCM, and methanol were
Sensing of Free Bilirubin in Human Serum. Human blood
serum was purchased from Sigma-Aldrich (Heat Inactivated from male
AB clotted whole blood, USA origin and sterile-filtered, CAS no-
H5667) and used as such without any further treatment. Human
serum is the preferred part of blood containing vital components such
purchased locally and were purified using standard procedures. 1H, 13
NMR and 1H−1H COSY spectra were analyzed using a Bruker-
C
B
DOI: 10.1021/acs.macromol.5b00043
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Scheme 1. Synthesis of Glucuronic Acid- and Glucose-Functionalized Polyfluorenes
as hemoglobin (≤25 mg/dL), total proteins (4−9 g/dL), iron (40%
UG), cholesterol (80−200 mg/dL), glucose (50−180 mg/dL), and
sodium (100−160 mg/dL). The absorption and fluorescence
measurements were conducted using a fixed amount of 100 μL of
human blood serum. The analyte, free bilirubin was added directly to
serum in the concentration range of 1 × 10−6 to 1 × 10−4 M, and this
mixture was added to the 1 μM solution of polymer in water (pH =
10). The final volume of test solution was adjusted to the fixed volume
of 2.5 mL by using PBS buffer at pH = 10. A control experiment was
performed in the same volume of serum and polymer without the
addition of bilirubin. The changes in the absorption and emission of
the polymer on addition of different concentrations of bilirubin in the
presence of serum were recorded at 15 °C with the slit width of 1 nm.
Preparation of Conjugated Bilirubin. An aliquot of the freshly
prepared liver homogenate was mixed thoroughly but gently with
stock solutions of free bilirubin and uridine diphosphateglucuronide
C
DOI: 10.1021/acs.macromol.5b00043
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
1
Figure 1. H NMR spectra of PF-Ph-GlcA and PF-GlcA recorded in MeOH-D4.
(UDPG1cUA) to achieve final concentrations of 50−500 mg/L (0.86
to 8.6 × 10 mol/L) bilirubin and 10 to 5 × 10 mol/L UDPGIcUA.
The acceptable pH range for the incubation mixture was 7.0−7.8. For
RESULTS AND DISCUSSION
■
Synthesis and Structural Characterization. Water-
soluble polyfluorenes functionalized with glucuronic acid
(GlcA) and glucose (Glu) side chains were synthesized via
click chemistry followed by palladium-catalyzed cross-coupling
which is shown in Scheme 1a,b. 2,7-Dibromofluorene (1) was
reacted with 1,6-dibromohexane in the presence of NaH to
obtain 2,7-dibromo-9,9-bis(6-bromohexyl)-9H-fluorene (2).
Compound 2 was etherified with propargyl alcohol to obtain
2,7-dibromo-9,9-bis(6-(prop-2-yn-1-yloxy)hexyl)-9H-fluorene
(3). The synthesis of azide-functionalized glucuronic acid and
glucose was followed from a literature report, and the scheme
routine incubations, 10−15 g/L Triton X-100 surfactant and 10 mol/L
MgC12 were also present in the medium. The entire reaction mixture
was kept in a nitrogen-rich atmosphere, shielded from strong light, and
at 37 °C (with constant, gentle swirling) during the 3−4 h incubation.
After incubation the content was extracted with chilled chloroform to
remove unconjugated bilirubin from the medium. The contents were
then centrifuged to remove the solid materials. The aqueous layer was
separated, and the isolation of conjugated bilirubin was done using the
Lucassen procedure.36
D
DOI: 10.1021/acs.macromol.5b00043
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
of synthesis is shown in Figure S1.37 In short, D-glucuronic
acid/D-glucose was initially protected with acetic anhydride in
the presence of iodine to give the penta-acetylated product. In
the case of D-glucuronic acid an additional step involving
conversion of the acetylated ester (COOAc) to its methyl ester
derivative (COOMe) was also carried out by refluxing the
penta-acetylated product in dry methanol for 24 h. The esters
were further reacted with TMS-N3 and SnCl4 to obtain the
corresponding azido derivatives 4 and 6, respectively, as the
major products which were purified by column chromatog-
raphy. Click reaction between the azide-functionalized
glucuronic acid/glucose and propargyl-functionalized fluorene
was carried out using copper-catalyzed azide−alkyne chemistry
to get the desired glucuronic acid- and glucose-functionalized
fluorene monomers 5 and 7, respectively. The monomers were
copolymerized with 1,4-benzenediboronic bis(pinacol ester) via
Suzuki coupling reaction to obtain the glucuronic acid-
functionalized fluorene copolymer PF-Ph-GlcA(OAc) and
glucose-functionalized fluorene copolymer PF-Ph-Glu(OAc),
respectively. A homopolymer of the glucuronic acid-function-
alized polyfluorene PF-GlcA(OAc) was also synthesized by
using bis(pinacolato)diboron as the coupling reagent. The
polymers were deprotected using NaOMe in MeOH:DCM
(1:1) and further treated with dilute HCl or TFA (trifluoro-
acetic acid) to produce PF-Ph-Glu, PF-Ph-GlcA, and PF-GlcA.
The polymers were purified by dialyzing the powdered sample
against Milli-Q water using 2 kDa molecular weight cut-off
dialysis membrane for 2 days. The deprotected glucuronic acid-
functionalized polymers were readily soluble in water, DMF,
DMSO, and MeOH compared to its precursor protected
polymer which was insoluble in water. The glucose-function-
alized polyfluorene had poor water solubility compared to the
glucuronic acid-functionalized counterpart. The detailed
synthetic procedures for the monomers and polymers are
given in the Supporting Information (S2). Both protected and
deprotected polymers were characterized by NMR spectrosco-
py for chemical structure analysis and size exclusion
chromatography (SEC) (using DMF as solvent) for molecular
weight analysis. Broadening of proton NMR peaks and broad
distribution in SEC indicated high molecular weight for the
Table 1. Polymer Designation, Molecular Weight,
Polydispersity Index, and Yield of Polymers before and after
Deprotection
polydispersity index
yield
(%)
a
a
a
name
Mn
Mw
(ĐM)
PF-GlcA(OAc)
16800
20200
25800
38300
1.54
1.89
90
85
PF-Ph-
GlcA(OAc)
PF-Ph-Glu(OAc)
PF-GlcA
17300
15500
17900
14200
35800
24200
35900
25600
2.07
1.56
1.94
1.8
87
78
80
85
PF-Ph-GlcA
PF-Ph-Glu
a
Molecular weights as determined by size exclusion chromatography
using DMF as eluent at 30 °C and polystyrene standards for
calibration.
of the deprotected polymers due to the elimination of
protecting functional groups during deprotection.
Photophysical Properties of the Water-Soluble Poly-
fluorenes. The normalized absorption and emission spectra of
the synthesized polymers were recorded in PBS buffer at pH =
7.4 and 10. Supporting Information Figure S5 compares the
normalized emission and excitation spectra at pH = 10 and at
the neutral pH = 7.4 for the three polymers. It should be
mentioned here that due to poor solubility of the glucose-
functionalized reference polymer in water at both the pH
studied, its solution remained turbid. Red-shift and broadening
of the emission were observed upon changing the pH from 7.4
to 10 for the glucuronic acid appended polyfluorene; however,
the absorption spectra remained unaffected. The excitation
spectra were comparable with that of the absorption spectra of
the polymers. The sensing studies to be discussed later on were
all carried out at pH = 10; therefore, the photophysical studies
carried out at pH = 10 is discussed in detail here. Figure 2
compares the absorption (top) and emission (bottom) spectra
of the three polymers in PBS buffer at pH = 10. PF-Ph-GlcA
copolymer and PF-GlcA homopolymer showed absorption
centered at 315 and 360 nm, respectively. The absorption
spectrum of PF-Ph-Glu was broad (maximum ∼360 nm), and
the baseline was high due to scattering. The emission maximum
of 0.1 OD aqueous solutions of the polymers PF-Ph-Glu
excited at 315 nm and PF-Ph-GlcA and PF-GlcA excited at 360
nm was observed around 410−420 nm. The peak appearing at
350 nm in the emission spectra of PF-Ph-GlcA corresponded to
the water Raman peak.39 The emission spectrum of the
polymers overlapped with the absorption maximum of bilirubin
(target analyte) at 455 nm (denoted in dotted line), thereby
favoring energy transfer from polymer to bilirubin. The
fluorescence quantum yield of the polymers were determined
in water at neutral pH using quinine sulfate (Φfl = 0.546 in 0.1
M H2SO4) as reference. The calculated quantum yield for PF-
Ph-GlcA polymer was 0.43 and for PF-GlcA was 0.48. The
quantum yield of the homopolymer (PF-GlcA) was slightly
higher than PF-Ph-GlcA because the solubility of PF-GlcA (5
mg/mL) was better in water compared to that of PF-Ph-GlcA
(2 mg/mL). The quantum yield of PF-Ph-Glu was not
determined due to the turbid nature of the solution.
1
polymers. The H and 13C NMR spectra of monomers and all
polymers are shown in Supporting Information S3. The proton
NMR spectra of deprotected polymers (PF-GlcA and PF-Ph-
GlcA) are shown in Figure 1. The spectrum of the glucose-
functionalized polymer PF-Ph-Glu is given in the Supporting
Information S3n,o,p. The peaks observed at δ 3.7 ppm for
methyl ester and δ 2.15−2.0 for acetyl protons in the NMR
spectra for the protected polymers were completely absent in
the case of deprotected polymers represented in Figure 1.
Supporting Information S3-m shows the 1H−1H COSY
spectrum of the PF-GlcA(OAc) polymer in CDCl3, which
was recorded to detect the coupling partners of the various
peaks. The molecular weights (Mn and Mw) and polydispersity
index (ĐM) of the protected and deprotected polymers were
determined by size exclusion chromatography (SEC) using
DMF as solvent, and the corresponding values are tabulated in
Table 1. The molecular weights of the new polymers were
comparable with that of polyfluorenes functionalized with
mannose reported in the literature.38 Corresponding SEC
chromatograms of the polymers are shown in Supporting
Information Figure S4. As expected, the observed molecular
weights of the protected polymers were larger compared to that
Fluorescence Sensing of Free Bilirubin in Water.
Sensing of free bilirubin in water was targeted taking the key
advantage of spectral overlap between emission of polymers
and absorption of bilirubin which is the necessary condition for
feasible Forster resonance energy transfer (FRET) to occur.40
̈
The glucuronic acid functionalization imparted water solubility
E
DOI: 10.1021/acs.macromol.5b00043
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
kept constant at 1 × 10−6 M (1 μM). When the concentration
of bilirubin was increased, an increase in the absorption of
bilirubin at 455 nm was observed. Figure 3b shows the
corresponding emission spectra upon excitation at 360 nm. The
notable observation was the quenching of polymer fluorescence
with concurrent increase in the emission of bilirubin at 520 nm.
Supporting Information Figure S6 shows emission spectra of
PF-GlcA, normalized at polymer emission maxima highlighting
the FRET-induced bilirubin emission. A clear visual color
change from blue fluorescence to light green was observed at
higher concentrations of bilirubin, which is shown in the inset
image of Figure S6. Bilirubin does not emit in any solvent
(including water) due to inherent property of poor quantum
yield. However, upon interaction with the polymer, mediated
by the glucuronic acid moiety, enhanced green emission of
bilirubin was observed. The intensity of the bilirubin emission
observed in this PF−bilirubin system was much higher
compared to that of bilirubin complexed with HSA protein
reported in the literature.43 Direct excitation at the bilirubin
absorption maxima at 455 nm showed a steady increase in the
bilirubin emission intensity with increase in the concentration
of bilirubin as shown in Supporting Information Figure S7.
FRET-induced bilirubin emission at 520 nm that was observed
upon excitation of the polymer in water highlighted the
potential of the polymer as a biosensor.
Similar sensing experiments were conducted for the
copolymers (PF-Ph-GlcA and PF-Ph-Glu) in water as shown
in Figure 3c−f. Similar conditions of addition of bilirubin to
polymer in water were followed, and fluorescence quenching of
the polymer was monitored. The changes in the fluorescence
intensity of the polymers upon addition of bilirubin were
plotted against the concentration of bilirubin, and the result is
given in Figure S8a. From the plot it could be seen that the
homopolymer PF-GlcA exhibited a sharper drop in the
polyfluorene emission upon addition of bilirubin with almost
complete quenching of the blue polymer fluorescence at 20 μM
concentration of bilirubin. The copolymer exhibited complete
quenching of the polymer fluorescence at 40 μM concentration
of bilirubin. On the other hand, the glucose appended polymer
PF-Ph-Glu exhibited very poor quenching of polymer
fluorescence with only 40% quenching observed even for the
highest addition of bilirubin (1 × 10−4 M). In fact, beyond 10
μM of bilirubin, the polymer fluorescence remained unaffected.
Figure S8b compares the intensity of FRET-induced bilirubin
emission upon excitation of polymers at the polymer
absorption maximum, which also showed that the homopol-
ymer exhibited higher FRET-induced bilirubin emission
compared to the copolymers at higher bilirubin concentrations
in water. The emission from bilirubin was almost non-
observable in the case of the glucose appended polymer PF-
Ph-Glu. The better sensing efficiency of the homopolymer
compared to the copolymers could be attributed to the greater
number of glucuronic acid appendage per repeat unit present in
the former compared to the latter, which facilitated better
polymer−analyte interaction. Isothermal titration calorimetric
(ITC) experiments (to be discussed later on) pointed toward a
higher binding constant for the homopolymer with bilirubin,
which also supported this observation of better sensing
efficiency for the homopolymer. Thus, the above experiment
stressed the importance of interaction between the polyfluorene
and free bilirubin in facilitating the biosensing property of the
polymers.
Figure 2. Absorption and emission spectra of PF-Ph-Glu, PF-Ph-GlcA,
and PF-GlcA polymers along with absorption of bilirubin (dotted line,
bottom Figure) in water at pH = 10.
to the polymer, facilitating sensing studies to be carried out in
water. The bilirubin molecule is a tetrapyrrole moiety, which
exists in the ridge tile conformation due to several intra-
molecular hydrogen bonds. It has been variously reported in
the literature as a porphyrin-like structure or linear
representation. Single-crystal X-ray structure of bilirubin has
confirmed the Z configuration at the C4−C5 and C15−C16
positions.41 The gross structure of bilirubin is shown in Scheme
2 in its linear representation. Bilirubin is reported to have a very
Scheme 2. Gross Structure of Bilirubin in Its Linear
Representation
low solubility of ∼7 nmol in water at pH = 7.4.42 A pH of ∼10
is required to induce solubility in water by breaking the
hydrogen bonding in bilirubin. Therefore, studies involving
bilirubin are usually reported at pH = 10 in the literature.42
Thus, stock solutions of bilirubin and polyfluorene were
prepared at pH = 10. Figure 3a shows the absorption spectrum
of the homopolymer PF-GlcA upon addition of various
concentration of bilirubin ranging from 1 × 10−6 to 1 × 10−4
M in water at pH = 10. The concentration of the polymer was
F
DOI: 10.1021/acs.macromol.5b00043
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 3. (a, c, e) Absorption and (b, d, f) emission spectra of PF-GlcA, PF-Ph-GlcA, and PF-Ph-Glu (1 μM) in water at pH = 10 upon addition of
various concentrations of bilirubin from 1 × 10−6 to 1 × 10−4 M.
Sensing of Bilirubin in Human Serum Samples. To
understand the reliability of the new polymers for biosensing
applications, sensing studies were proposed to be carried out in
human serum containing interferences such as hemoglobin,
proteins, triglyceride, cholesterol, glucose, and metal ions.
Although the homopolymer (PF-GlcA) showed better sensing
activity in water, it could not be used for sensing studies in
human serum due to its tendency to aggregate (discussed in
detail later on). The bilirubin sensing studies in human serum
was conducted with the copolymers PF-Ph-GlcA and PF-Ph-
Glu. 0.1 mL of human serum was mixed with free bilirubin in
the concentration range of 1 × 10−6 to 1 × 10−4 M, and this
mixture was added to 1 μM solution of the polymers. A blank
experiment was performed by recording the emission of
bilirubin in water as well as in serum at pH = 10 in the
absence of polymer. Supporting Information Figure S9
compares the emission intensity from bilirubin upon excitation
at 325 nm (the polymer absorption wavelength which was used
to excite the polymer in the actual bilirubin sensing
experiments) as well as at 455 nm (the bilirubin absorption
maximum). Bilirubin in water at pH = 10 did not exhibit any
fluorescence upon excitation at 325 or 455 nm. In serum at pH
=10, bilirubin exhibited a weak emission centered at 520 nm,
indicating that selective excitation of polymer at 325 nm was
not possible due to the not so negligible absorption of bilirubin
at 325 nm. Direct excitation at 455 nm resulted in sharp
emission from bilirubin at 520 nm. Figure 4a−d demonstrates
the changes in the photophysical properties of the probe for
various concentrations of the analyte. In the absorption
spectrum, human serum exhibited strong absorption of light
below 300 nm, indicating the presence of other analytes. In the
emission spectrum of PF-Ph-GlcA (Figure 4b), blue
fluorescence of the polymer quenched upon addition of
bilirubin with concurrent appearance of light green emission
from bilirubin. The photograph of the visual color change from
blue to green observed in the serum is given as an inset in
Figure 4b, highlighting the sensing ability of PF-Ph-GlcA in
human serum. The emission plot also shows the emission from
bilirubin in serum at pH =10 which indicated that its intensity
was negligible compared to the FRET-induced emission from
bilirubin in the presence of polymer. From Figure 4d it could
be seen that the quenching of fluorescence of polymer upon
addition of bilirubin was much less for the reference glucose
appended polymer PF-Ph-Glu, and also the FRET-induced
bilirubin emission was almost negligible. Figure 5 compares the
efficiency of FRET-induced bilirubin emission for both
polymers for increasing additions of bilirubin. The plot gives
the ratio of the integrated areas of bilirubin emission (486−600
nm) to that of polymer emission (335−486 nm) as a function
of increasing concentration of bilirubin. PF-Ph-GlcA exhibited
more than 5-fold increase in bilirubin emission compared to
PF-Ph-Glu at the highest added bilirubin concentration of 10−4
mol. Supporting Information Figure S10 compares the
reduction in fluorescence emission of the polymers (area
under the emission plot) as a function of increasing bilirubin
concentration. The drop was much steeper for the glucuronic
acid-functionalized polymer, thereby once again establishing its
superiority as a bilirubin sensor. Although the photophysical
criteria for FRET-based energy transfer from polymer to
bilirubin was favorable for both polymers PF-Ph-GlcA and PF-
Ph-Glu, the higher tendency of the glucuronic acid−bilirubin
interaction in the former compared to the low or non-
G
DOI: 10.1021/acs.macromol.5b00043
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 4. (a, c) Absorption and (b, d) emission spectra of PF-Ph-GlcA and PF-Ph-Glu, respectively, in human serum upon addition of various
concentrations of bilirubin from 1 × 10−6 to 1 × 10−4 M in phosphate buffered saline at pH = 10. The emission from 1 × 10−4 M bilirubin in serum
upon excitation at the polymer absorption wavelength is also included in the emission spectra.
An experiment was conducted to study the sensing efficiency
of PF-Ph-GlcA in the presence of fixed concentration of
bilirubin (100 μm) in serum at various pH. Figure 6 compares
Figure 5. Ratio of integrated area of FRET induced bilirubin emission
(486−600 nm) to that of the polymer emission (335−486 nm) for
PF-Ph-GlcA and PF-Ph-Glu in serum as a function of bilirubin
concentration.
Figure 6. Effect of varying pH (7−10) on the fluorescence emission
spectra of 0.1 OD solution (∼1.25 × 10−6 M) of PF-Ph-GlcA in the
interacting glucose−bilirubin combination in the latter acted in
presence of 100 μM of bilirubin in human serum.
favor of PF-Ph-GlcA to behave as a better sensor for bilirubin in
human serum. Additionally, the low solubility of PF-Ph-Glu
also could have resulted in the poor performance of PF-Ph-Glu
as an efficient bilirubin sensor. Thus, among the three polymers
examined for sensing bilirubin, only the glucuronic acid
appended copolymer, i.e. PF-Ph-GlcA, was found to be effective
for sensing bilirubin in human serum. Therefore, only PF-Ph-
GlcA was taken up for further detailed studies to better
understand its biosensing characteristics.
the intensity of emission from a 0.1 OD solution (1.3 μM) of
PF-Ph-GlcA in serum at pH varying from 7.0 to 10. No
quenching of polymer fluorescence was observed at pH = 7,
since the bilirubin was not soluble at this pH. As the pH was
slowly increased, one could visually observe the bilirubin
solution turning yellow due to its dissolution. Complete
dissolution of bilirubin with absence of any undissolved
H
DOI: 10.1021/acs.macromol.5b00043
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 7. DLS plot of (a) serum, (b) polymers in water, (c) PF-Ph-GlcA, and (d) PF-GlcA in PBS buffer at pH = 10 upon addition of serum. The
concentration of polymers was 1 μM, and serum of 100 μL was added to the polymer solution.
Figure 8. Isothermal calorimetric thermograms (a, c) for the injection of 2 μL of 20 mM bilirubin into 2 mM PF-GlcA and PF-Ph-GlcA at an interval
of 3 min at pH = 12 and (b, d) the respective sigmoidal fit.
I
DOI: 10.1021/acs.macromol.5b00043
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
particles was observed between pH = 9 and 10 (Supporting
Information photograph S11). The observation from the
fluorescence measurements also correlated by exhibiting the
highest quenching of polymer emission at pH = 10. This study
also validated the conduction of the fluorescence based
bilirubin sensing experiments at pH = 10.
Although the homopolymer PF-GlcA had exhibited better
sensing efficiency for bilirubin in water compared to the
copolymer, its employment for sensing studies in serum could
not be carried out. This was due to the considerable quenching
of fluorescence as well as aggregate emission at 550 nm, which
hampered observation of the bilirubin emission beyond 520 nm
in serum. Supporting Information Figure S12 shows the
reduction in fluorescence intensity ∼410 nm and aggregate
emission at 550 nm exhibited by the homopolymer (PF-GlcA)
in serum.
the polymer. The free energy change ΔG for the binding of
bilirubin with PF-GlcA was −6.1 kcal/mol while that for
binding with PF-Ph-GlcA polymer was −4.94 kcal/mol.
Generally, the Gibbs free energy change for electrostatic type
of interactions and hydrogen bonding interactions falls in the
range of −(15−30) and −(2−10) kcal/mol, respectively.45
From the values of ΔG of the polymer−bilirubin complex, the
type of interaction between the polymer and bilirubin was
identified to be that of hydrogen-bonding interactions. The OH
and COOH groups of glucuronic acid appendage on the
polyfluorene could be expected to play a key role in making
hydrogen bond interactions with COOH and NH moieties
present in the pyrrole ring of the bilirubin.
The binding constants for the interaction between bilirubin
and polyfluorenes were 4.57 × 103 M−1 for PF-GlcA polymer
and 4.07 × 103 M−1 for PF-Ph-GlcA. The stoichiometry of the
interaction was obtained as 2.25 for homopolymer (PF-GlcA)
and 1.13 for copolymer (PF-Ph-GlcA). As the homopolymer
PF-GlcA contained relatively more glucuronic acid side chains
per mole than the copolymer PF-Ph-GlcA, the PF-GlcA has
higher possibility for noncovalent interactions, resulting in a
higher binding constant. Therefore, the homopolymer showed
relatively higher sensing activity in water compared to the
copolymer. Unfortunately, the sensing activity of homopolymer
in serum was hindered due to its severe aggregation in serum.
Fluorescence Lifetime. For a better understanding of the
energy transfer processes taking place between the polymer and
bilirubin, time-correlated single photon counting (TCSPC)
analysis of the fluorescence lifetime was utilized. Time-resolved
fluorescence decays were collected for the polymers PF-Ph-
GlcA and PF-GlcA at 420 and 410 nm with excitation from
nano LED at 320 and 375 nm in PBS buffer at pH = 10 at 25
°C. The decay curve for the polymer PF-Ph-GlcA is shown in
Figure S13a. The fluorescence decays were fitted to a
biexponential fit, and the lifetime values are given in Supporting
Information Table 1. PF-Ph-GlcA exhibited a lifetime of τ1 =
457 ps and τ2 = 2.42 ns with α1 = 0.87 and α2 = 0.13. Upon
addition of various concentrations of bilirubin, the lifetime of
the polymer decreased drastically to a minimum value of τ1 =
21 ps and τ2 = 2.56 ns with α1 = 1 and α2 = 0. Time-resolved
fluorescence decays were also collected at 520 nm (bilirubin
emission) for 320 nm excitation. Figure S13c reveals the decay
curve for excitation at 320 nm and collection at 520 nm upon
various additions of bilirubin to PF-Ph-GlcA polymer. This
graph confirmed the increase in the lifetime of bilirubin in the
polymer−bilirubin complex, giving evidence for the FRET
process. The increase in the lifetime values of bilirubin are given
in the Supporting Information (Table S2). Similar behavior was
observed for PF-GlcA homopolymer that showed lifetime of τ1
= 470 ps and τ2 =3.54 ns with α1 = 0.89 and α2 = 0.11. The
lifetime value of the polymer decreased continuously with the
addition of bilirubin, and on final addition of analyte the value
reached τ1 = 52 ps and τ2 = 1.54 ns with α1 = 0.97 and α2 =
0.03. The corresponding decay curve for polymer (PF-GlcA) is
shown in Figure S13b,d. The decay curve for excitation at 375
nm and observation at 520 nm is given in Figure S13d. The rise
in the lifetime of the bilirubin values is tabulated in Supporting
Information (Table S2). Simultaneous decrement in fluores-
cence lifetime of the polymer and increase in the lifetime of the
bilirubin strongly supported the occurrence of FRET from
polymer to bilirubin.
Dynamic light scattering (DLS) studies conducted both in
water and in serum also supported the tendency of the
homopolymer to form large size aggregates in serum. The DLS
spectrum of human serum is shown in Figure 7a. In water, PF-
GlcA exhibited particle size of 330 nm whereas PF-Ph-GlcA
formed smaller particles with size of 110 nm (Figure 7b).
Figure 7a−d compares the size versus intensity plot from the
DLS studies for the three polymers in water and serum. Figure
7a shows that the serum itself exhibited multimodal distribution
of particle size due to presence of proteins and other
ingredients. From Figure 7c,d it could be seen that while the
copolymer did not exhibit any changes in the inherent size of
the particles upon addition of serum (Figure 7c), the
homopolymer exhibited both broadening of its inherent particle
size and the appearance of new broad peak at higher particle
size (Figure 7d). It could be inferred from the above results that
the homopolymer PF-GlcA with higher volume ratio of
glucuronic acid tend to adhere to the hydrophilic patches of
proteins44 in serum, resulting in polymer aggregation in serum.
Therefore, it is essential to design a polymer with optimum
content of glucuronic acid appendage so as to avoid adherence
to proteins.
Identification of the Interactions between Polymer
and Bilirubin. The nature of interaction existing between the
new polymers appended with D-glucuronic acid and bilirubin in
water medium was identified using isothermal titration
calorimeter (ITC). ITC measures the heat changes produced
in the course of the addition of bilirubin acting as ligand to the
polymer acting as the macromolecule. High concentration of
bilirubin (20 mM) was required to observe saturation of
polymer−bilirubin interaction in the ITC experiment. Such
high concentrations of bilirubin could be fully solubilized only
in PBS buffer at pH = 12. Thus, 20 mM of bilirubin solution in
PBS buffer at pH = 12 was added in injections of 2 μL volume
to 200 μL polymer solution at a concentration of 2 mM.
Corresponding results of heat changes during the course of
addition for both the polymers PF-GlcA and PF-Ph-GlcA are
given in Figures 8a,b and 8c,d. The outcome of the results
revealed the exothermic nature of the interaction between
polymer and bilirubin. To eliminate the solvent−solvent and
solvent−bilirubin interactions, blank reactions without polymer
was also carried out. The blank heat change value was
subtracted from the heat change value for the polymer−
bilirubin complex. The fitting in Figure 8c,d and calculation
methods are given in the Materials and Methods section. The
free energy change for the reactions obtained in the negative
scale indicated the spontaneity of the binding of bilirubin with
Selectivity and Sensitivity of Bilirubin Sensing in
Human Serum. Sensing of analyte in the presence of crucial
J
DOI: 10.1021/acs.macromol.5b00043
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
interferences is important in order to establish the potential of
the sensor for real world biosensing applications. Human serum
contains components such as hemoglobin, proteins, iron,
cholesterol, glucose, sodium, etc. The D-glucuronic acid
appended PF-Ph-GlcA polymer could selectively sense bilirubin
in the presence of these components of blood. The porphyrin
unit of hemoglobin has structural similarity to bilirubin, but as
demonstrated by a control experiment, where the polymer was
added to human serum sample in the absence of added
bilirubin (Figure S14), the emission of polymer (PF-Ph-GlcA)
did not quench in the serum. Despite the similarity in chemical
structure, the closed ring structure of hemoglobin prevented
hydrogen-bonding interaction with the polymer. In fact, human
serum albumin (HSA) is known to bind to free bilirubin with
high affinity;32 however, the polymer exhibited competitive
binding with free bilirubin in human serum sample via the
hydrogen-bonding interactions of its D-glucuronic acid
appendage. This was evident from the quenching of polymer
emission in human serum samples containing bilirubin.
Other notable interferences such as biliverdin which is the
structural analogue of bilirubin, rhodamine, cholesterol,
glucose, and sucrose were supplemented in fixed concentration
(1 × 10−4 M) to 100 μL serum in separate experiments. The
polymer PF-Ph-GlcA concentration was fixed at 1 μM, and the
total volume of the experimental solution was made up to 2.5
mL with PBS buffer. The absorption spectrum of biliverdin and
rhodamine along with emission of PF-Ph-GlcA polymer is given
in Figure S15. Each sensing experiment was performed four
times, and the average fluorescence response of the polymer
was recorded and plotted against respective analytes in Figure
9a as a bar graph representation with error bars (the
corresponding emission spectrum is given in Figure S16). A
huge decrement in fluorescence intensity of polymer was
observed only with the addition of bilirubin, highlighting the
high selectivity of new polymer toward bilirubin in human
blood serum samples. To differentiate the type of bilirubin, the
fluorescence experiment under same conditions were con-
ducted keeping conjugated bilirubin as the analyte. Conjugated
bilirubin (con.BR) was prepared following the reported
literature procedure, and the details are given in the
Experimental Section.36 There was not much change observed
for the emission intensity of the polymer in the presence of
conjugated bilirubin, indicating its poor interaction with the
polymer (Figure S7). Conjugated bilirubin was incapable of
hydrogen-bonding interactions with the glucuronic acid
appendage, thereby disabling the polymer from binding the
conjugated bilirubin. From Figure S15, it could be seen that
both biliverdin and rhodamine exhibited reasonable spectral
overlap with polymer emission; however, a very low (for
biliverdin) or no quenching of polymer emission was observed
when these were added to the polymer in serum. Biliverdin is a
structural analogue of bilirubin, with an extra double bond in
the C10 position, which makes the two dipyrromethene units
conjugated.46 Although this structural change is very small, it
makes biliverdin more rigid compared to bilirubin. The
probability of hydrogen bond interaction between biliverdin
and glucuronic acid is good; however, a probable nonfavorable
conformation of the former could be the reason for the
observed poor quenching of polymer fluorescence.
Figure 9. (a) Bar diagram depicting the effect of various interfering
analytes on the fluorescence intensity of PF-Ph-GlcA in human serum.
The analyte concentration was kept constant at 1 × 10−4 M. (b)
Nanomolar sensing by PF-Ph-GlcA in human serum; [polymer] = 1 ×
10−5 M.
in human serum at pH =10. Figure 9b depicts the change in
fluorescence emission from polymer upon addition of various
nanomolar concentration of bilirubin. A 10% reduction in total
area of fluorescence emission was observed upon addition of
150 nm of bilirubin. This was well below the clinically relevant
range of <25 to >50 μmol/L bilirubin observed in the human
blood serum. Thus, the selectivity and sensitivity experiments
demonstrated the superior performance of the glucuronic acid
appended polyfluorene PF-Ph-GlcA as a biosensor for sensing
bilirubin in human serum.
Mechanism of Sensing. The aforementioned experiments
displayed the ability of the glucuronic acid appended
polyfluorene to function as a highly selective and sensitive
visual sensor for sensing bilirubin. The appendage of D-
glucuronic acid performed the dual role of imparting water
solubility to the polymer and interacting with the analyte
through hydrogen bonding. The polymer exhibited good extent
of spectral overlap with bilirubin in water as well as in human
serum, which was the necessary condition to facilitate energy
transfer from polymer to bilirubin. When bilirubin was added to
the polymer in aqueous medium or in serum, the appendage of
D-glucuronic acid interacted with bilirubin enabling the binding
of bilirubin with the polymer (highlighted schematically in
Figure 10). The binding of bilirubin with polymer, in turn,
facilitated energy transfer from polymer to bilirubin, which was
manifested as a visual fluorescence color change from blue to
Besides selective sensing, sensitivity of detecting nanomolar
amounts of analyte is also an important requirement for an
efficient sensor. Therefore, nanomolar amounts of bilirubin (1−
150 nm) were added to a 2 × 10−5 M solution of PF-Ph-GlcA
K
DOI: 10.1021/acs.macromol.5b00043
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 10. Schematic diagram showing mechanism of FRET-induced sensing of bilirubin in human serum using PF-Ph-GlcA.
light green. In contrast, a reference polymer designed with D-
glucose as the appendage in place of D-glucuronic acid (PF-Ph-
Glu) could not perform as effectively as PF-Ph-GlcA even
though its photophysical properties were favorable for FRET-
based energy transfer. The glucose appendage was not
successful in affording the high water solubility that was
made feasible by the glucuronic acid appendage. PF-Ph-GlcA
was highly selective toward free bilirubin compared to other
interferences and biological substances present in the serum
due to the feasible hydrogen-bonded interactions and favorable
spectral overlap existing only for the bilirubin-polymer pair.
Though competition for binding sites could be expected to be
present in the pool of interferences, binding of bilirubin in
minimum sites was sufficient for FRET-based energy transfer to
occur due to the well-known molecular wire effect47 that is
inherent for the conjugated polymer backbone. It also
successfully demonstrated its efficiency in sensing nanomolar
amounts (∼150 nm) of bilirubin in human serum. Consolidat-
ing all the observations, the combined role of spectral overlap
and noncovalent interactions afforded high selectivity and
nanomolar sensitivity for the sensing of free bilirubin in water
as well as in human blood serum.
bilirubin. Isothermal titration calorimetric (ITC) experiments
revealed hydrogen-bonding interaction between the polymers
and bilirubin with a strong binding constant of ∼4.5 × 103 M−1.
Although both polymers exhibited similar sensing activity
toward bilirubin in water, their sensing activity in human serum
was entirely different. The homopolymer PF-GlcA aggregated
in serum with aggregate emission observed at 550 nm, which
was also supported by DLS experiments conducted in both
water and serum media. DLS experiments showed that PF-
GlcA formed particles with size of 330 nm in water, which
increased considerably in serum. The copolymer PF-Ph-GlcA
was successful as an efficient sensor for bilirubin in human
serum with visual change of emission color from blue to green
for FRET-induced bilirubin emission at 520 nm. The sensing of
bilirubin in human blood serum was found to be sensitive even
in the presence of crucial interferences such as hemoglobin,
biliverdin, proteins, triglyceride, cholesterol, metal ions, and
sugars. Nanomolar sensing of bilirubin could also be
demonstrated successfully in human serum using PF-Ph-GlcA
as the conjugated polymer sensor. Thus, we have demonstrated
for the first time the successful application of water-soluble
conjugated polymers based on polyfluorenes in the selective
and sensitive (sensing ∼150 nm bilirubin in human serum)
sensing of free bilirubin in the clinically relevant range of <25 to
>50 μmol/L in the human blood serum. These novel
polyfluorenes appended with glucuronic acid could thus
function as a highly selective and sensitive material for the
visual sensing of free bilirubin in human blood serum via the
combined operation of energy transfer and noncovalent
interactions.
CONCLUSION
■
Novel water-soluble polyfluorenes were designed and success-
fully synthesized by using click chemistry followed by Suzuki
polymerization. The polymers were structurally characterized
by NMR spectroscopy and size exclusion chromatography. The
appendage of D-glucuronic acid to polyfluorene imparted water
solubility through hydrogen-bonded interactions of polar −OH
and −COOH groups with the aqueous medium. In analogy to
the key role of enabling easy excretion of free bilirubin from the
body played by D-glucuronic acid in the body metabolism, the
appendage of D-glucuronic acid to polyfluorene afforded the
added advantage of interactive site for the analyte bilirubin.
Taking advantage of the water solubility and spectral overlap of
the polymer emission with bilirubin absorption, the sensing of
free bilirubin in water was targeted. A complete quenching of
the blue polymer fluorescence was observed at 20 μM
concentration of bilirubin addition for PF-GlcA and 40 μM
for PF-Ph-GlcA. Energy transfer from polymer to bilirubin was
proved by the decrease in polymer fluorescence lifetime with
concurrent increase in the emission as well as the lifetime of
ASSOCIATED CONTENT
* Supporting Information
■
S
Detailed synthetic procedures and characterizations for all
monomers and polymers. The Supporting Information is
available free of charge on the ACS Publications website at
DOI: 10.1021/acs.macromol.5b00043.
AUTHOR INFORMATION
Corresponding Author
*(S.K.A.) E-mail sk.asha@ncl.res.in; Fax 0091-20-25902615.
■
Notes
The authors declare no competing financial interest.
L
DOI: 10.1021/acs.macromol.5b00043
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
(28) Boyer, T. D.; Manns, P. P.; Sanyal, A. J. Hepatology: A Textbook
of Liver Disease, 6th ed.; Saunders: London, 2011; p 1079.
(29) Silbernagl, S.; Despopoulos, A. Color Atlas of Physiology, 6th ed.;
Thieme: Stuttgart, 2009; p 252.
(30) Hansen, T. W. R. J. Perinatol. 2001, 21, S48−S51.
(31) Rand, R. N.; di Pasqua, A. Clin. Chem. 1962, 8, 570−578.
(32) Santhosh, M.; Chinnadayyala, S. R.; Kakoti, A.; Goswami, P.
Biosens. Bioelectron. 2014, 59, 370−376.
(33) Senthilkumar, T.; Asha, S. K. Macromolecules 2013, 46, 2159−
2171.
ACKNOWLEDGMENTS
■
This work has been financially supported by the Science &
Engineering Research Board funded project SB/S1/OC-66/
2013. T.S. thanks CSIR for senior research fellowship. The
authors thank Dr. Partha Hazra, IISER Pune, for picosecond
fluorescence lifetime instrument facility. The authors thank Dr.
Neha Gupta, NCL, Pune, for preparation of conjugated
bilirubin. The authors also thank Dr. Amitavadas and Vadde
Ramu from NCL, Pune, for ITC measurement.
(34) Zheng, Z.; Fang, J. L.; Lazarus, P. Drug Metab. Dispos. 2002, 30,
397−403.
REFERENCES
(35) Fevery, J. Liver Int. 2008, 592−605.
■
(36) Wu, T. W.; Zumbulyadis, N.; Gross, S.; Gohlke, R. S. Clin.
Chem. 1980, 26, 1323−1335.
(1) McQuade, D. T.; Pullen, A. E.; Swager, T. M. Chem. Rev. 2000,
100, 2537−2574.
(37) Tosin, M.; Murphy, P. V. Org. Lett. 2002, 4, 3675−3678.
(38) Pu, K. Y.; Shi, J.; Wang, L.; Cai, L.; Wang, G.; Liu, B.
Macromolecules 2010, 43, 9690−9697.
(2) Qin, C.; Cheng, Y.; Wang, L.; Jing, X.; Wang, F. Macromolecules
2008, 41, 7798−7804.
(3) Thomas, S. W., III; Joly, G. D.; Swager, T. M. Chem. Rev. 2007,
107, 1339−1386.
(39) Lakowicz, J. Principles of Fluorescence Spectroscopy, 3rd ed.;
Springer: Berlin, 2006.
(4) Feng, X.; Liu, L.; Wang, S.; Zhu, D. Chem. Soc. Rev. 2010, 39,
2411−2419.
(40) Sindbert, S.; Kalinin, S.; Nguyen, H.; Kienzler, A.; Clima, L.;
Bannwarth, W.; Appel, B.; Muller, S.; Seidel, C. A. M. J. Am. Chem. Soc.
2011, 133, 2463−2480.
(41) Bonnett, R.; Davies, J. E.; Hursthouse, M. B. Nature 1976, 262,
326−328.
(42) Brodersen, R. J. Biol. Chem. 1979, 254, 2364−2369.
(43) Lamola, A. A. Effects of Environment of Photophysical
Processes of Bilirubin. In Optical Properties and Structure of
Tetrapyrroles; Blauer, G., Sund, H., Eds.; Walter de Gruyter & Co.:
Berlin, 1985; p 311.
(44) Dutton, G. J. Glucuronic acid Free and Combined Chemistry,
Biochemistry, Pharmacology, and Medicine; Academic Press: New York,
1966.
(5) Feng, F.; Tang, Y.; Wang, S.; Li, Y.; Zhu, D. Angew. Chem., Int. Ed.
2007, 46, 7882−7886.
(6) He, F.; Tang, Y. L.; Yu, M. H.; Feng, F.; An, L. L.; Sun, H.; Wang,
S.; Li, Y. L.; Zhu, D. B.; Bazan, G. C. J. Am. Chem. Soc. 2006, 128,
6764−6765.
(7) Ho, H. A.; Dore, K.; Boissinot, M.; Bergeron, M. G.; Tanguay, R.
M.; Boudreau, D.; Leclerc, M. J. Am. Chem. Soc. 2005, 127, 12673−
12676.
(8) Bajaj, A.; Miranda, O. R.; Kim, Ik-B.; Phillips, R. L.; Jerry, D. J.;
Bunz, U. H. F.; Rotello, V. M. Proc. Natl. Acad. Sci. U. S. A. 2009, 106,
10912−10916.
(9) Disney, M. D.; Zheng, J.; Swager, T. M.; Seeberger, P. H. J. Am.
Chem. Soc. 2004, 126, 13343−13346.
(45) Moreno-Pirajan, J. C. Thermodynamics - Interaction Studies,
Solids, Liquids and Gases; InTech: Rijeka, Croatia, 2011.
(10) Pu, K. Y.; Liu, B. Adv. Funct. Mater. 2009, 19, 1371−1378.
(11) Tang, H. W.; Duan, X. R.; Feng, X. L.; Liu, L. B.; Wang, S.; Li,
Y. L.; Zhu, D. B. Chem. Commun. 2009, 6, 641−643.
(46) Blauer, G.; Wagnier
1954.
̀
e, G. J. Am. Chem. Soc. 1975, 97, 1949−
(47) Zhou, Q.; Swager, T. M. J. Am. Chem. Soc. 1995, 117, 7017−
7018.
̈
(12) Pennakalathil, J.; Jahja, E.; Ozdemir, E. S.; Konu, O.; Tuncel, D.
Biomacromolecules 2014, 15, 3366−3374.
(13) Zhu, C.; Yang, Q.; Liu, L.; Wang, S. Chem. Commun. 2011, 47,
5524−5526.
(14) Moon, J. H.; McDaniel, W.; MacLean, P.; Hancock, L. F. Angew.
Chem., Int. Ed. 2007, 46, 8223−8225.
(15) Pu, Kan-Y.; Shi, J.; Cai, L.; Li, K.; Liu, B. Biomacromolecules
2011, 12, 2966−2974.
(16) Tang, Y.; He, F.; Wang, S.; Li, Y.; Zhu, D.; Bazan, G. C. Adv.
Mater. 2006, 18, 2105−2110.
(17) Feng, X.; Duan, X.; Liu, L.; Feng, F.; Wang, S.; Li, Y.; Zhu, D.
Angew. Chem., Int. Ed. 2009, 48, 5316−5321.
(18) Zhu, C.; Liu, L.; Yang, Q.; Lv, F.; Wang, S. Chem. Rev. 2012,
112, 4687−4735.
(19) Jiang, H.; Taranekar, P.; Reynolds, J. R.; Schanze, K. S. Angew.
Chem., Int. Ed. 2009, 48, 4300−4316.
(20) Liu, B.; Yu, Wang-L.; Lai, Yee-H.; Huang, W. Macromolecules
2002, 35, 4975−4982.
(21) Huang, F.; Hou, L.; Wu, H.; Wang, X.; Shen, H.; Cao, W.; Yang,
W.; Cao, Y. J. Am. Chem. Soc. 2004, 126, 9845−9853.
(22) Jose, M.; Tome, M.; Esquembre, R.; Mallavia, R.; Mateo, C. R.
Biomacromolecules 2010, 11, 1494−1501.
(23) Gaylord, B. S.; Heeger, A. J.; Bazan, G. C. J. Am. Chem. Soc.
2003, 125, 896−900.
(24) Kim, I. B.; Dunkhorst, A.; Bunz, U. H. F. Langmuir 2005, 21,
7985−7989.
(25) Miranda, O. R.; You, C. C.; Phillips, R.; Kim, I. K.; Ghosh, P. S.;
Bunz, U. H. F.; Rotello, V. J. Am. Chem. Soc. 2007, 129, 9856−9857.
(26) Xue, C.; Donuru, V. R. R.; Liu, H. Macromolecules 2006, 39,
5747−5752.
(27) Takasu, A.; Iso, K.; Dohmae, T.; Hirabayashi, T. Biomacromo-
lecules 2006, 7, 411−414.
M
DOI: 10.1021/acs.macromol.5b00043
Macromolecules XXXX, XXX, XXX−XXX