Quantum dot-containing polymer particles with
Alla N. Generalovaa, Vladimir A. Oleinikova,b, Alyona Sukhanovab,c, Mikhail V. Artemyevb,
Vitaly P. Zubova, Igor Nabievb,c,n
aShemyakin–Ovchinnikov Institute of Bioorganic Chemistry, Russian Academy of Sciences, 117997 Moscow, Russian Federation
bLaboratory of Nano-Bioengineering, Moscow Engineering Physics Institute, 31 Kashirskoe sh., 115409 Moscow, Russian Federation
cEuropean Technological Platform Semiconductor Nanocrystals, Institute of Molecular Medicine, Trinity College Dublin, James’s Street, Dublin 8, Ireland
a r t i c l e i n f o
Received 6 June 2012
Received in revised form
17 July 2012
Accepted 18 July 2012
Available online 25 July 2012
a b s t r a c t
Composite polymer particles consisting of a solid poly(acrolein-co-styrene) core and a poly
(N-vinylcaprolactam) (PVCL) polymer shell doped with CdSe/ZnS semiconductor quantum dots (QDs)
were fabricated. The temperature response of the composite particles was observed as a decrease in
their hydrodynamic diameter upon heating above the lower critical solution temperature of the
thermosensitive PVCL polymer. Embedding QDs in the PVCL shell yields particles whose fluorescence is
sensitive to temperature changes. This sensitivity was determined by the dependence of the QD
fluorescence intensity on the distances between them in the PVCL shell, which reversibly change as a
result of the temperature-driven conformational changes in the polymer. The QD-containing thermo-
sensitive particles were assembled with protein molecules in such a way that they retained their
thermosensitive properties, including the completely reversible temperature dependence of their
fluorescence response. The composite particles developed can be used as local temperature sensors,
as carriers for biomolecules, as well as in biosensing and various bioassays employing optical detection
& 2012 Elsevier B.V. All rights reserved.
Dispersions of polymer particles are characterized by a large
specific surface area and are easy to produce and functionalize;
therefore, they are widely used in analytical chemistry, biosen-
sing, and clinical diagnosis (Bangs, 1996). In the past decades,
increasing attention has been paid to the preparation of ‘‘smart’’
functionalized polymer particles reversibly responding to slight
environmental changes, such as variations of temperature, pH,
and ionic strength (Takata et al., 2003). In particular, unique
properties of thermosensitive polymer particles (TPPs) make
them suitable for many applications, especially in biology, such
as the measurement of the local temperature in a single cell or in
volumes smaller than 10?18l (Duracher et al., 2000; Snowden
et al., 1994; Vihola et al., 2007).
At present, optical detection schemes become a technological
frontier in biosensing (Bachmann et al., 2008). In this connection,
the development of thermosensitive polymer particles containing
an optical label with temperature-dependent properties attracts
much attention. For example, photoluminescent (PL) nano-
crystals, such as CdSe/ZnS quantum dots (QDs), are considered
promising as labels for optical detection based on changes in the
fluorescence intensity and/or peak position (Sukhanova et al.,
2002, 2004). QDs are characterized by a high quantum yield and
exceptional resistance to both chemical degradation and photo-
degradation. Another advantage is that QDs of different sizes can
be excited at the same wavelength while emitting PL with a
narrow symmetrical spectrum at distinctly different wavelengths
in the visible or near-IR region. This allows multicolor detection
using nanoprobes (Oleinikov et al., 2007; Nabiev et al., 2008) and
multiple optical encoding of microparticles with nanocrystals
(Sukhanova et al., 2007; Sukhanova and Nabiev, 2008). QDs may
also be included in complex superstructures (Sukhanova et al.,
2006) and used in fluorescent resonance energy transfer (FRET)
schemes (Wargnier et al., 2004).
The QD fluorescence intensity exhibits a linear temperature
response (Liu et al., 2006) that is sensitive to QD local environ-
ment (Kalyuzhny and Murray, 2005). Interestingly, QD-capping
agents are known to make QD fluorescence temperature-insensi-
tive. For example, QD fluorescence intensity becomes indepen-
dent of temperature both when denatured ovalbumin is used as a
capping agent (Wang et al., 2008) and when QDs are embedded in
polymer particles (Stsiapura et al., 2004; Joumaa et al., 2006).
Contents lists available at SciVerse ScienceDirect
journal homepage: www.elsevier.com/locate/bios
Biosensors and Bioelectronics
0956-5663/$-see front matter & 2012 Elsevier B.V. All rights reserved.
nCorresponding author at: the Institute of Molecular Medicine, Trinity College
Dublin, James’s Street, Dublin 8, Ireland. Tel.: þ33 631 259 180; fax: þ33 326 918 127.
E-mail address: firstname.lastname@example.org (I. Nabiev).
Biosensors and Bioelectronics 39 (2013) 187–193
The excellent temperature-detection properties of QDs were
explored in superstructures engineered from ensembles of QDs
connected with molecular springs (Lee and Kotov, 2007). Ensem-
bles of QDs can be obtained by modifying their surface and linking
them via surface functional groups, with polymer molecules
serving as linkers. An example of such a superstructure whose PL
intensity undergoes reversible changes in response to temperature
variation has been described by Lee et al. (2005). These authors
have developed a reversible nanothermometer based on a dynamic
superstructure of two types of nanoparticles—a central gold core
and corona-like superstructures of CdTe QDs. These are linked
via a polyethylene glycol (PEG) spacer acting as a temperature-
dependent molecular spring in the aqueous medium. However, not
only is the gold core expansible, but its diameter is practically
We have attempted to design a superstructure based on an
advanced architecture consisting of a colloidal polymer core and a
shell containing QDs. The size of the synthetic polymer core can
be controlled by varying the conditions of its synthesis. In order to
make the distance between QDs, which determines the fluores-
cence intensity, changeable, we covered the core with a shell of a
temperature-sensitive ‘‘smart’’ polymer. Such polymers exhibit
critical phenomena, such as phase transitions, in response to
external stimuli, including changes in temperature (Cole et al.,
2009).They undergo reversible
separation changes at temperatures above the so-called lower
critical solution temperature (LCST) (Kirsh, 1998). We chose poly-
(acrolein-co-styrene) as the material for this core, because its
outer layer bears double bonds, which make it easy to form a shell
over it. The shell consisted of the temperature-sensitive polymer
Fluorescent TPP can be obtained by embedding QDs in the
PVCL shell. Temperature variation around LCST causes conforma-
tional changes in PVCL and, hence, changes in the distances
between QDs, which, in turn, results in fluorescence changes
(Fig. 1). These fluorescence changes may serve as a basis for
optical detection methods in bioassays.
Thus, the purpose of this study was to design polymer particles
with thermosensitive fluorescence and study their properties, in
particular, as carriers of biologically active compounds (exemplified
in this study by bovine serum albumin, BSA) and as nanotherm-
ometers for measuring changes in the local temperature (exemplified
by monitoring the temperature in the course of chemical reactions).
2. Materials and methods
Acrolein (H2CQCH–CHO) was purchased from Fluka, Germany.
It was distilled three times at the atmospheric pressure, and the
fraction with a boiling point of 56 1C, r4
1.40 was used. Styrene was also purchased from Fluka, Germany,
20¼0.806 g/cm3, and nD
purified with a 5% sodium hydroxide aqueous solution to remove
the stabilizer, rinsed with water until pH became neutral, dried
over calcium chloride, and distilled twice in vacuum. The fraction
with a boiling point of 51 1C (2.1 kPa), r4
The following materials were purchased from Sigma–Aldrich
and used without further purification: N-vinylcaprolactam (VCL),
potassium persulfate (PP), a,a0-azo-isobutyronitrile (AIBN), sodium
chloride, sodium borate buffer, bovine serum albumin (BSA), and
sodium azide. Ethanol, methanol, propanol-2, and chloroform
(Aldrich) were of analytical grade.
Semiconductor CdSe/ZnS core–shell nanocrystals were synthe-
sized as described earlier (Wargnier et al., 2004). In the present
study, hydrophobic nanocrystals with diameters of 3.5 nm (with a
PL emission peak at 554 nm), and 6 nm (with a PL emission peak
at 610 nm) were used. Their PL was excited at lex¼480 nm in
20¼0.906 g/sm3, and
20¼1.54 was used.
Optical and fluorescent characteristics were measured using
an UV/VIS Beckman DU–700 spectrophotometer, a Shimadzu
RF–551 spectrofluorimeter, and a BioDoc-IT System UV-Transil-
luminator. The FT-IR spectra were recorded using a Varian 3100
2.2.1. Synthesis of poly(acrolein-co-styrene) core particles
Emulsifier-free radical copolymerization was carried out in
distilled water at a comonomer-to-water ratio of 1:9 and an
acrolein-to-styrene molar ratio of 10:1. A homogeneous styrene–
water mixture (at a monomer-to-water ratio of 1:10 v/v) was
prepared, acrolein was added into the reactor. The reaction
mixture was deoxygenated by purging with N2for 30 min, and
PP (0.5 wt% relative to the monomer mixture) was added as an
initiator. The temperature of the polymerizing mixture was
adjusted at 65 1C. Polymerization was carried out under nitrogen
for 12 h while stirring.
2.2.2. Radical polymerization of N-vinylcaprolactam in the presence
of core particles
The dispersion of poly(acrolein-co-styrene) core particles to be
used as seeds was purified by centrifugation in the presence of
excess water. VCL (0.5 ml) at a seed particle to VCL ratio of 1:0.1,
1:0.2, 1:0.5, 1:1, or 1:1.5 (w/w) in a 20:1 water–propanol-2
mixture and 0.15 ml of PP or AIBN (0.2 wt% relative to VCL) in
the same solvent was added to 1 ml of a 1-wt% dispersion of seed
particles. The mixture was purged with nitrogen and stirred for
0.5 h. Then, the flask was placed to a water bath, and the mixture
was stirred under nitrogen at 70 1C for 1, 2, or 3 h.
2.2.3. Radical seed polymerization of N-vinylcaprolactam
VCL (0.5 ml) at a seed particle to VCL ratio of 1:0.1, 1:0.2, 1:0.5,
1:1, or 1:1.5 (w/w) in a 20:1 water–propanol-2 mixture was
added to 1 ml of a 1-wt% dispersion of seed particles and left for
swelling at 4 1C for 12 h. Then, 0.15 ml of PP or AIBN (0.2 wt%
relative to VCL) in a 20:1 water–propanol-2 mixture was added,
and the temperature of the polymerizing mixture was adjusted
to 70 1C. Polymerization was carried out under nitrogen for 1, 2,
or 3 h while stirring.
2.2.4. Measurement of the hydrodynamic radius of polymer particles
The hydrodynamic radius (R) of the polymer particles was
measured using the dynamic light scattering technique. The
dispersion was diluted with water to obtain the concentration
required for the light scattering experiments according to the
Fig. 1. Engineering of thermosensitive polymer particles. Poly(acrolein-co-styr-
ene) particles were used as a solid core (the red sphere). A thermosensitive shell
(green) around the solid core was obtained via radical polymerization of
vinylcaprolactam (VCL). Embedding QDs (pink) in this thermosensitive shell
resulted in fluorescent particles whose fluorescence changed due to variations of
the distance between QDs as a result of changes in the PVCL conformation at the
lower critical solution temperature. (For interpretation of the references to color in
this figure caption, the reader is referred to the web version of this article.)
A.N. Generalova et al. / Biosensors and Bioelectronics 39 (2013) 187–193
manufacturer’s recommendations and then poured into a cuvette
(Lines, 1985). The cuvette holder was kept at the desired tem-
perature between 20 and 45 1C. The particle size was measured
using a Coulter N4-MD sub-micron particle analyzer.
2.2.5. Measurement of the acrolein oligomer concentration
The acrolein oligomer concentration in the supernatant obtained
after centrifugation of the polymer suspension was measured
against water at lmax¼273 nm using a Beckman DU-70 spectro-
photometer (Margel and Rembaum, 1980). The results obtained
(in absorbance units) were represented as the oligomer mass using
a calibration graph of the optical absorption of known quantities of
oligomer dissolved in water.
2.2.6. Incorporation of quantum dots into thermosensitive polymer
Solvents for QD incorporation into TPPs were selected among
water, methanol, ethanol, propanol, propanol-2, butanol, hexane,
chloroform, and their mixtures at ratios of 1:1, 5:1, and 10:1.
It was required that the solvent do not affect the size of TPPs, their
aggregation, or colloid formation upon incubation.
QDs (0.2 mg) were purified from TOP/TOPO by dispersing in
chloroform and precipitating with methanol (at a chloroform-to-
methanol ratio of 1:3). The purified QDs were dispersed in 1 ml of
propanol-2 and added to 0.5 ml of a 1 wt% TPP dispersion in a
20:1 water–propanol-2 mixture. The mixture was stirred vigor-
ously, sonicated for 2 min, incubated for 20 min while stirring
(this procedure was repeated three times), shaken for 1 h at room
temperature, and centrifuged at 7000 rpm for 10 min with addi-
tion of water (this procedure was repeated five times to remove
free QDs). The pellet was then dispersed in 0.5 ml of water. To
remove propanol-2, the obtained TPPs embedded with QDs were
dialyzed against the water–propanol-2 mixture.
2.2.7. Bovine serum albumin immobilization on polymer particles
An aliquot (0.125 ml) of a 1-wt% dispersion of TPPs containing
QDs was incubated with BSA (1.6–15 mg/g polymer solids) in a
0.1 M sodium borate buffer solution pH 8.2 at 20 1C for 2 h and in
a water bath at 40 1C for 0.5, 1, or 2 h. To block the groups that
had not reacted, 0.5 ml of a glycine solution in 0.1 M sodium
borate (10 mg/ml buffer) was added. Then, the reaction mixture
was washed by centrifugation–dispersion three times to remove
the excess protein, and the pellet was dispersed in 1 ml of a
glycine solution buffered with 0.1 M sodium borate (10 mg/ml).
The concentration of unbound BSA was determined by Bradford’s
method at l¼595 nm, with allowance for dilution during the
3. Results and discussions
3.1. Preparation of thermosensitive polymer particles
Thermosensitive composite particles were obtained using a
two-stage reaction: first, core particles were synthesized via
emulsifier-free radical copolymerization; then, the particles were
modified with the thermosensitive polymer (Fig. 1).
The first step was the synthesis of core particles based on the
copolymer of styrene and acrolein. This type of cores possessed
the properties of polystyrene particles (Yen et al., 1976); in
addition, the polyacrolein component provided hydrophilicity of
the surface and contained double bonds due to the specific
characteristics of acrolein polymerization (Slomkowski, 1998).
The particle size can be easily varied by changing the ratio of
the polymerized monomers (Generalova et al., 2007). We used
emulsifier-free radical copolymerization of acrolein and styrene
in water with an acrolein-to-styrene monomer ratio of 10:1 in the
presence of K2S2O8(PP) for preparing polymer particles with a
hydrodynamic diameter of 185715 nm. This diameter was suffi-
ciently small to preclude spontaneous sedimentation during
The second step was the formation of the PVCL thermosensi-
tive shell. We used two approaches:
(1) Radical polymerization of VCL in the presence of core parti-
cles, which served as seed particles. This seed polymerization
procedure made it possible to graft PVCL polymer chains on
the surface of seed particles owing to the initiator inducing
VCL polymerization in the dispersion. The resultant PVCL was
water-insoluble, because it was formed at 70 1C and adsorbed
on the surface of the seed particles, after which it formed
bonds with them, with the activated double bonds of poly-
acrolein involved in the process (Eliseeva, 1988).
(2) Radical seed polymerization of VCL. This polymerization was
carried out after swelling of seed particles with VCL in the
presence of the initiator. Under these conditions, the grafting
of the formed PVCL molecules was mainly due to the double
bonds of polyacrolein. This type of polymerization allowed
the formation of composite particles, with few, if any, new
particles been generated (Gardon, 1973).
The effects of different factors, such as the ratio between the
seed-particle and VCL concentrations (1:0.1, 1:0.2, 1:0.5, 1:1, or
1:1.5, w/w), duration of polymerization (1, 2, or 3 h), dispersion
medium (water, water–methanol, or water–propanol-2), type of
the initiator (the water-soluble PP or oil-soluble AIBN), were
estimated in terms of optimizing the conditions of TPP preparation.
The thermosensitive properties of the obtained particles were
evaluated by measuring the dynamic light scattering. It is note-
worthy that the hydrodynamic particle sizes of the TPPs decreased
with increasing temperature above the LCST due to conformational
changes of PVCL from a hydrated coil to a collapsed hydrophobic
globule (Yi and Xu, 2005). TPPs collapsed remarkably at 32 1C,
which was the LCST of PVCL. It was found that the preferable
medium for the second step was a 20:1 water–propanol-2 mixture.
The results (Supplementary Table S1) show that the TPPs with
optimal properties were prepared using seed polymerization by
method (1) in the presence of the water-soluble initiator PP (TPP I)
and seed polymerization by method (2) in the presence of the
oil-soluble initiator AIBN (TPP II). Both approaches to obtaining
TPPs could be used under identical conditions, namely, a 20:1
water–propanol-2 mixture as a dispersion medium, a seed particle
to VCL ratio of 1:0.5, and a polymerization duration of 3 h.
3.2. Characterization of thermosensitive polymer particles
The FT-IR technique was used to control the desired surface
modification of TPPs I and II (Supplementary Fig. S1). In these
spectra, one can see an adsorption peak at 1650 cm?1, which is
characteristic of CQO amide groups (Silverstein and Webster,
1998). This peak was more intense for particles modified by
method (2). These results provide evidence for the grafting of
PVCL onto poly(acrolein-co-styrene) cores.
The obtained particles remained stable for a long time and
were unaffected by electrolyte (0.15 M NaCl, physiological saline).
This stability was also preserved at high temperatures, when the
PVCL particles were shrunken. It may be concluded that methods
(1) and (2) of seed polymerization are methods of choice for
obtaining TPPs that do not coagulate or precipitate in solutions
with a high ionic strength.
The modification of copolymer particles with PVCL made it
possible to decrease the amount of low-molecular-weight products
A.N. Generalova et al. / Biosensors and Bioelectronics 39 (2013) 187–193
in dispersion media. These products are formed because of partial
degradation of polyacrolein during storage, with oligomers
released into the dispersion medium (Rembaum et al., 1984). The
amounts of low-molecular-weight products in the cases of TPPs I
(0.21 mg/ml) and II (0.11 mg/ml) were found to be, respectively,
half as much and quarter as much compared to unmodified
copolymer particles (0.46 mg/ml).
Fig. 2A shows the temperature dependence of the hydro-
dynamic radius of TTPs of types I and II. With increasing
temperature, the radii of both types of TPPs gradually decreased
when heated to 29 1C and then decreased abruptly as the
temperature further increased to 32 1C. This sharp decrease was
due to hydrophobic aggregation of PVCL chains. The results
indicated that the particles drastically collapsed at 32 1C, which
was the LCST for the PVCL polymer. Thus, colloidally and chemi-
cally stable TPPs can be obtained by methods 1 (TPP I) and 2
(TPP II). The following part of the study was aimed at obtaining
fluorescent TPPs and studying their properties.
3.3. Thermosensitive polymer particles embedded with QDs
Semiconductor QDs emitting light at about 555 nm were incorpo-
rated into the PVCL shells of TPPs from a 1:20 chloroform–propanol-
2 mixture after removal of TOPO as described earlier (Generalova
et al., 2009). The resultant TPPs doped with QDs displayed intense
green fluorescence. Efficient incorporation of QDs into TPPs was
proved by the absence of free QDs in the dispersion medium after
centrifugation of the TPP suspension: the fluorescence intensity
of the supernatant fraction after centrifugation was found to be
Microfluorescence and transmission electron microscopic photo-
graphs of the resultant thermosensitive polymer particles doped
with semiconductor QDs are shown in Supplementary Fig. S2.
Note that QDs had almost no effect on the thermosensitive
properties of TPPs. Fig. 2B shows that both types of TPPs contain-
ing QDs responded to heating in about the same way as TPPs
without QDs. However, the embedding of QDs in both TPP types
resulted in a slight LCTS shift towards lower temperatures.
This phenomenon can be explained in terms of the effect of QDs
as a hydrophobic component of the composite particle on the
conformation of the aqueous associates of PVCL, which results in
breakage of cross-linking hydrogen bonds (Kirsh, 1998).
In Fig. 3A, the fluorescence spectra of the obtained TPPs at
various temperatures are compared with the fluorescence spectra
of seed poly(acrolein-co-styrene) particles doped with the same
amount of QDs. The swelling procedure used for doping was
described earlier (Generalova et al., 2007). Note that heating
of TPPs I and II reduced their fluorescence intensity, but the
fluorescence intensity of seed copolymer particles was almost
unchanged at higher temperatures (Fig. 3A). Moreover, the
intensity of TPP fluorescence recorded at 20 1C was almost four-
fold higher compared to that of seed particles doped with QDs by
swelling. Thus, the inclusion of QDs into TPPs seems to be
preferable over their inclusion into seed copolymer particles. As
noted by Nida et al. (2008), the ZnS shell may be damaged by a
solvent (e.g., chloroform), which results in coordinative unsatura-
tion (the surface emitting state) of QDs and, consequently,
fluorescence quenching. PVCL around each QD is likely to occupy
the vacant coordinate sites on the QD surface and efficiently
passivate the surface emitting state (as compared to the copoly-
mer chains of seed particles), which results in an increase in the
The temperature effect on the fluorescence intensity of TPPs is
confirmed by the fact that only the peak corresponding to TPPs
was decreased (Fig. 3B) upon heating the mixture containing
QD-embedded TPPs (lem¼550 nm) and QD-embedded unmodified
seed copolymer particles (lem¼610 nm).
The maximum variation of the fluorescence intensity corre-
sponds to the largest changes in the TPPs radius at temperatures
between 26 and 32 1C for TPPI and between 27 and 31 1C for
TPPII (Supplementary Table S1). The sensitivity of temperature
measurement within this range has been found to be about 0.1 1C.
Note that the region of the maximum sensitivity of temperature
measurement may be varied by changing the type of the thermo-
sensitive polymer used. For example, the use of the copolymer of
PVCL and poly-N-vinylpyrrolidone allows shifting the range of the
maximum sensitivity toward higher temperatures, whereas the
use of the copolymer of PVCL and vinyl alcohols results in a
downshift of the region of the maximum sensitivity.
The fluorescence intensities of TPPs I and II were also mea-
sured in heating–cooling cycles with the temperature varying
between 25 and 40 1C (Fig. 4). Note that the process was totally
reversible for both TPP I and TPP II, showing negligible photo-
degradation in every temperature cycle (20 min).
Fig. 4B shows the dependence of the fluorescence intensity in a
heating–cooling cycle on the time as a dynamic characteristic of
TPPs. The maximum decrease in the fluorescence intensity upon
heating a 1-ml sample containing 0.1 mg of TPPs from 25 to 40 1C
was observed for both TPP types within 5 min of heating. This
level of fluorescence remained unchanged at 40 1C for at least
30 min. Upon cooling, TPPs I and II behaved somewhat differ-
ently: TPPs I recovered their fluorescence intensity to the initial
value within 5 min, whereas this took almost 10 min in the case
of TPPs II.
The fluorescence of TPP supernatants after centrifugation was
at a vanishingly low level, which confirmed that the QDs were not
20 25 303540 4550
202530 3540 45 50
core (seed particle)
core (seed particle)
Fig. 2. The effect of temperature on the hydrodynamic radii of (A) thermosensitive polymer particles and (B) thermosensitive polymer particles doped with QDs.
A.N. Generalova et al. / Biosensors and Bioelectronics 39 (2013) 187–193
lost from the TPPs during heating–cooling cycles. The sensors
developed were very stable, with no more than 10% of the
fluorescence lost during 10 heating–cooling cycles, 10 min per
cycle. The shelf life of the TPPs was found to be more than 2 years
without any effect on their fluorescence properties.
The phenomenon of reversible temperature-dependent fluor-
escence of TPPs can be explained as follows. The rise of tempera-
ture altered the conformation of PVCL on the surface of seed
copolymer particles, resulting in the formation of hydrophobic
globules (Songa et al., 2011). As mentioned above, this may be
observed as shrinkage of the PVCL layer and, hence, a decrease in
the particle size. The shrinkage of the PVCL layer seems to
decrease the distances between embedded QDs, which is the
crucial factor in quenching the QD fluorescence (Zaharchenko
et al., 2005). In addition, the peaks were slightly red-shifted at
temperatures above the LCST, which also indicated that QDs were
located close to one another.
Our calculations of the distance between QDs before and after
heating confirmed this suggestion. Since our previous data
showed that QDs could not be incorporated into seed copolymer
particles without preliminarily swelling for at least 1 h, the
distance was calculated on the assumption on predominant QD
penetration into the PVCL shell (Generalova et al., 2007). UV–vis
measurements show that about 104QDs could be incorporated
into each TPP, which gives a mean distance between QD centers
of about 9.5 nm at 20 1C. According to Chistyakov et al. (2008),
this distance is typical of films with a relatively low QD density.
Heating to 40 1C induced shrinkage of the PVCL layer, and the
distance between QDs was decreased to 4.2 nm. This agrees with
the data obtained by Chistyakov et al. (2008) for films of densely
packed QDs (4.0–4.1 nm). Thus, the PVCL shells on copolymer
particles analyzed at different temperatures may be regarded as
films with different QD densities. The films containing QDs at a
low density were characterized by narrow fluorescence spectral
bands and a relatively intense fluorescence that can be quenched
by increasing the density of QDs accompanied by a red shift of the
emission peak (Murray et al., 2000; Sukhanova et al., 2006). This
quenching resulted from nonradiative excitation transfer between
QDs (Murray et al., 2000) and interaction of the dipole moments
related to the QD asymmetry (Colvin et al., 1994). To summarize,
we may conclude that the above analogy between PVCL shells
containing QDs and QD films supports our strategy of the
formation of TPPs with temperature-dependent fluorescence.
This strategy is based on variations of the distances between
QDs caused by temperature-dependent conformational changes
The cooling procedure resulted in expansion of PVCL chains,
and the distance between QDs probably returned to the initial
value, which resulted in fluorescence recovery. This behavior
ensured reversibility of the fluorescence intensity changes. The
difference between TPPs I and II in the duration of fluorescence
recovery may be accounted for by differences in the morphology
of the PVCL layer (chain length, flexibility, conformation, etc.) and
the characteristics of its grafting onto the particle surface owing
to the specificity of seed polymerization used in methods (1) and (2).
Thus, incorporation of fluorescence labels, including QDs, into
TPPs facilitates the production of optically sensitive polymer
fl. intensity, a.u.
t ~ 20 C
t > 40 C
t < 40 C
fl. intensity, a.u.
Fig. 3. The fluorescence spectra of poly(acrolein-co-styrene) particles and mixtures of thermosensitive polymer particles embedded with QDs and seed copolymer particles
prepared by the swelling procedure. (A) The fluorescence spectra of poly(acrolein-co-styrene) particles at (1) 20 1C and (2) 40 1C and thermosensitive polymer particles at
(3) 20 1C and (4) 40 1C. (B) Temperature dependence of the fluorescence spectra of a mixture of thermosensitive polymer particles embedded with QDs (lem¼550 nm) and
unmodified seed copolymer particles embedded with QDs (lem¼610 nm) using the swelling procedure described by Generalova et al. (2007).
Fig. 4. Cyclic heating (40 1C)–cooling (25 1C) temperature variation in solutions of
thermosensitive polymer particles (TPPs) of types I and II (A) and corresponding
changes in the fluorescence of these solutions (B). In Panel C, the dependence of
the fluorescence intensities on the time of heating (40 1C) and cooling is
represented as the ratio of the TPP fluorescence intensity at the given moment
(It) to the initial level at 20 1C (I20). dl is the region of the maximum temperature
sensitivity of TPPs of both types.
A.N. Generalova et al. / Biosensors and Bioelectronics 39 (2013) 187–193
particles with temperature-dependent fluorescence. In addition,
the fluorescence intensity of these particles is reversible during a
heating–cooling cycle, which is promising in terms of the devel-
opment of optical detection methods for bioassays.
3.4. Bovine serum albumin immobilization on thermosensitive
For all bioanalytical applications, particles should be conju-
gated with a specific bioligand, preferably protein or peptide. It is
known that PVCL is capable of complexing with various com-
pounds. In the case of its interaction with proteins, PVCL amide
groups form hydrogen bonds with carboxyl or amino groups of
proteins. The conditions and efficiency of this complexing were
estimated using the model of BSA immobilized at concentrations
from 1.6 to 15 mg/g polymer (Kirsh, 1998).
It can be seen in Fig. 5A that the amount of the adsorbed
protein increased with increasing protein concentration in the
solution until signal saturation is reached at a concentration of
10 mg/g polymer. Apparently, the plateau corresponded to the
situation where the surfaces of the polymer particles were
completely covered with the attached protein macromolecules,
and there was no free space left for more protein. Note that the
increase in the amount of immobilized BSA to 10 mg/g caused a
decrease in the TPP fluorescence intensity (Fig. 5A), which
remained practically unchanged as the BSA concentration further
increased. The adsorption processes for TPPs I and II were almost
identical. BSA at the saturating surface concentration seemed to
form a complex with PVCL, which gave rise to conformational
changes and decreased the PVCL capacity for passivating the
surface emitting states of QDs. After saturation of the surface
with BSA (at a concentration of 10 mg/g), the PVCL shell probably
underwent no further conformational changes, and the fluores-
cence remained practically unvaried. In addition, the BSA adsorp-
tion had almost no effect on the thermosensitive properties of
TPPs embedded with QDs: in the vicinity of 32 1C, the hydro-
dynamic radii of TPPs I and II decreased by about 45 and 48 nm,
respectively. Therefore, we estimated the amount of BSA added
that corresponded to saturation.
We studied different conditions of BSA immobilization on
TPPs, including incubation at 20 1C for 1 h and incubation at
40 1C for 0.5, 1, and 2 h. The maximum amount of adsorbed BSA
(?80% of the amount added) on both types of TPPs was found in
the case of incubation at 40 1C for 1 h. With increasing tempera-
ture, the hydrophobic interactions between PVCL and BSA became
stronger. It is known that deformation of protein molecules due to
their interaction with PVCL, which is facilitated by the rise of
temperature, promotes mechanical entrapment of BSA during the
shell shrinkage (Songa et al., 2011). This effect increased the
amount of adsorbed BSA at 40 1C as compared to that at 20 1C
Although the BSA adsorption at higher temperatures decreased
the fluorescence intensity of TPPs, this decrease was smaller than
in the case of BSA absorption at 20 1C (Fig. 5B). In addition, the
smallest change in the TPP fluorescence after BSA adsorption was
observed in the case of BSA immobilization on TPP II at 40 1C for
1 h. Fig. 5B also shows the reversible fluorescence dependence on
temperature during a cooling–heating cycle for both TPPs I and II.
Thus, QD-containing TPPs I and II could be efficiently assembled
with protein molecules (as exemplified by BSA) in such a manner
that TPPs retained their thermosensitive properties, including
the reversible dependence on temperature, with a relatively
small loss of fluorescence intensity under the optimal protein
We have developed an approach to obtaining QD-based
reversibly temperature-sensitive superstructures with a synthetic
colloidal polymer core of desirable functionality, whose diameter
is easily controllable by the synthesis procedure. The super-
structure is based on an advanced architecture consisting of the
polymer core and a shell containing QDs.
In order to make the distance between QDs, which determines
the fluorescence intensity, changeable, we made the shell
from a temperature-sensitive ‘‘smart’’ polymer undergoing rever-
sible conformational changes at temperatures above the LCST.
The solid polymer core is composed of poly(acrolein-co-styrene)
particles, which have double bonds in the outer layer due to the
characteristics of acrolein polymerization. These double bonds
allow the fabrication of a shell of the thermosensitive polymer
PVCL, whose polymer chain undergoes conformational changes
from a hydrated coil to a collapsed globule at LCST (Songa et al.,
2011). The coil-to-globule transition of PVCL, which entails inter-
and intra-chain bonding resulting in a loss of solubility and
hydrophobic aggregation, has been detected at a temperature of
about 32 1C. This temperature is assumed to be the LCST of this
polymer in water (Lau and Wu, 1999). It is close to the physio-
logically normal temperature in higher mammals. In addition,
PVCL can adsorb, and form complexes with, protein molecules
(Kirsh, 1998). Thus, PVCL forms a functionalized layer binding
protein molecules on the surface of polymer particles, which
makes them promising for the use in bioassays.
The use of particles consisting of a thin thermosensitive shell
over a solid core instead of bulk thermosensitive particles makes
Fig. 5. Dependences of the amount of bovine serum albumin (BSA) adsorbed on thermosensitive polymer particles (TTPs) and their fluorescence intensity on the amount of
BSA added (A) and changes in the intensities of fluorescence of TPPs of types I and II during a heating–cooling cycle after BSA immobilization (B). BSA was immobilized at
20 1C or 40 1C for 1 h. IBSAis the fluorescence intensity of TPPs after BSA immobilization. I0is the initial fluorescence intensity of TPPs.
A.N. Generalova et al. / Biosensors and Bioelectronics 39 (2013) 187–193
it possible to accelerate the response to temperature changes:
heating above the LCST induces a globule conformation of PVCL,
shrinkage of the TPP shell, and decrease in the distances between
QDs resulting in fluorescence quenching.
The potential for TPP application as carriers of biopolymers has
been studied using BSA as a model. The effects of the temperature,
time of BSA immobilization, and BSA concentration on the TPP
fluorescence have been evaluated. The optimal adsorption condi-
tions allowed us to obtain BSA-tagged particles with a reversibly
Finally, bioanalytical applications of developed TPPs were illu-
strated by two examples of their use for real-time remote mon-
itoring the local temperature of a reaction mixture in the course of
exothermic chemical reactions: enzymatic hydrolysis of BSA and
cross-linking of chitosan (see Supplementary Information). These
data show that the TPPs developed may be used for measurement
of the local temperature, as carriers for biomolecules, and in
bioassays employing optical detection schemes.
This study was partly supported by the European Commission
through the FP7 Cooperation Program (grant no. NMP-2009-4.0-
3-246479 NAMDIATREAM) and the Ministry of Higher Education
and Science of the Russian Federation (grant no. 11.G34.31.0050
to I.N). V.A.O. and V.P.Z. acknowledge the support of the Russian
Foundation for Basic Research (RFBR, grant nos. 10-04-00393 and
12-04-00779) and the Ministry of Higher Education and Science
of the Russian Federation (grant 11.519.11.2005).
Appendix A. Supporting information
Supplementary data associated with this article can be found in
the online version at http://dx.doi.org/10.1016/j.bios.2012.07.030.
Bachmann, P.K., Hummel, H., et al., 2008. Journal of Nanophotonics 2, 021920.
Bangs, L.B., 1996. Pure and Applied Chemistry 68, 1873–1877.
Chistyakov, A.A., Martynov, I.L., et al., 2008. Laser Physics 18 (8), 925–938.
Cole, M.A., Voelcker, N.H., et al., 2009. Biomaterials 30, 1827–1850.
Colvin, V.L., Cunningham, K.L., Alivisators, P., 1994. Journal of Chemical Physics
101 (8), 7122–7138.
Duracher, D., Elaissari, A., Pichot, C., 2000. Macromolecular Symposia 150,
Eliseeva, V.I., 1988. Russian Chemical Reviews 57 (2), 307–331.
Gardon, J.L., 1973. Journal of Polymer Science - Polymer Chemistry Edition 11,
Generalova, A.N., Sizova, S.V., et al., 2007. Nanotechnology Russia 2 (7–8),
Generalova, A.N., Sizova, S.V., et al., 2009. Colloids Surface, A 342 (1–3), 59–64.
Joumaa, N., Lansalot, M., et al., 2006. Langmuir 22, 1810–1816.
Kalyuzhny, G., Murray, R.W., 2005. Journal of Physical Chemistry B. 109 (15),
Kirsh, Yu.E., 1998. Poly-N-vinylpyrrolidone and Other Poly-n-Vinylamides: Synth-
esis and Physico-Chemical Properties. Nauka, Moscow.
Lau, A.C.W., Wu, C., 1999. Macromolecules 32, 581–584.
Lee, J., Govorov, A.O., Kotov, N.A., 2005. Angewandte Chemie International Edition
Lee, J., Kotov, N.A., 2007. Nanotoday 2 (1), 48–51.
Lines, R.W., 1985. Polymer Science 45, 235–237.
Liu, T.C., Huang, Z.L., et al., 2006. Analytica Chimica Acta 559, 120–123.
Margel, S., Rembaum, A., 1980. Macromolecular Chemistry Suppl. 13, 19–24.
Murray, C.B., Kagan, C.R., Bawendi, M.G., 2000. Annual Review of Materials Science
Nabiev, I., Sukhanova, A., et al., 2008. Colloidal Nanoparticles in Biotechnology.
In: Elaissari, A. (Ed.), Wiley&Sons Inc, pp. 68–133, pp. 376.
Nida, D.L., Nitin, N., et al., 2008. Nanotechnology 19, 1–6.
Oleinikov, V.A., Sukhanova, A.V., Nabiev, I., 2007. Nanotechnologies in Russia 2
Rembaum, A., Chang, M., et al., 1984. Journal of Polymer Science Part A: Polymer
Chemistry 22, 609–617.
Silverstein, R.M., Webster, F.X., 1998. Spectrometric Identification of Organic
Compounds, sixth ed. John Wiley & Sons, New York.
Slomkowski, S., 1998. Progress in Polymer Science 23, 815–878.
Snowden, M.J., Marston, N.J., Vincent, B., 1994. Colloid and Polymer Science 272
Songa, M.M., Branford-White, C., et al., 2011. Colloids Surface B 84, 477–483.
Sukhanova, A., Baranov, A.V., et al., 2006. Angewandte Chemie International
Edition 45, 2048–2052.
Sukhanova, A., Venteo, L., et al., 2002. Laboratory Investigation 82, 1259–1262.
Sukhanova, A., Devy, J., et al., 2004. Analytical Biochemistry 324, 60–67.
Sukhanova, A., Nabiev, I., 2008. Critical Reviews in Oncology/Hematology 68,
Sukhanova, A., Susha, A.S., et al., 2007. Nano Letters 7, 2322–2327.
Stsiapura, V., Sukhanova, A., et al., 2004. Analytical Biochemistry 334 (2), 257–265.
Takata, S., Shibayama, M., et al., 2003. Polymer 44, 495–501.
Vihola, H., Marttila, A., et al., 2007. International Journal of Pharmaceutics 343,
Wang, J.H., Wang, H.Q., et al., 2008. Talanta 74, 724–729.
Wargnier, R., Baranov, A., et al., 2004. Nano Letters 4, 451–457.
Yen, S.P., Rembaum, A., et al., 1976. In: Emulsion Polymerization. ACS Symposium
Series, Washington, pp. 236–257.
Yi, C., Xu, Z., 2005. Journal of Applied Polymer Science 96, 824–828.
Zaharchenko, K.V., Obraztcova, E.A., et al., 2005. Laser Physics 15 (8), 1150–1153.
A.N. Generalova et al. / Biosensors and Bioelectronics 39 (2013) 187–193