Journal of Colloid and Interface Science 480 (2016) 232–239

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science

journal homepage: www.elsevier.com/locate/jcis

Synthesis of lab-in-a-pipette-tip extraction using hydrophilic nano-sized

dummy molecularly imprinted polymer for purification and analysis of
prednisolone
Maryam Arabi a, Mehrorang Ghaedi a,⇑, Abbas Ostovan b, Shaobin Wang c
a
Chemistry Department, Yasouj University, Yasouj 75914-35, Iran
b
Department of Chemistry, Kerman Branch, Islamic Azad University, Kerman, Iran
c
Department of Chemical Engineering, Curtin University, GPO Box U1987, Perth WA 6845, Australia

g r a p h i c a l a b s t r a c t
The PT-DMIP-HPLC-UV extraction procedure.

a r t i c l e i n f o a b s t r a c t

Article history: A novel pipette-tip based on nano-sized dummy molecularly imprinted polymer (PT-DMIP) assisted by
Received 21 May 2016 ultrasonication for the effective enrichment and analysis of prednisolone from urine samples was devel-
Accepted 11 July 2016 oped. The PT-DMIP cartridge was prepared by packing the dummy molecularly imprinted polymer at the
Available online 12 July 2016
tip of the micropipette. The polymerization used betamethasone (BM) as the dummy template,
3-aminopropyltrimethoxysilane (APTMS) as the functionalized monomer, tetraethyl orthosilicate
Keywords: (TEOS) as the cross-linker and aluminum ion (Al3+) as a dopant to produce Lewis acid sites in the silica
Molecularly imprinted polymer
matrix for metal coordinative interactions with the analyte. Compared to conventional solid phase
Dummy template
Pipette-tip solid phase extraction
extraction (SPE), the PT-DMIP is cost-effective, fast, and easy to handle, while the system is very
Prednisolone approachable and reduces the consumption of toxic organic solvent. HPLC-UV analysis revealed success-
ful applicability of the sorbent for highly efficient extraction of perdnisolone from urine matrices. The
extraction recovery was investigated and optimum conditions were obtained using central composite
design. Good linearity for prednisolone in the range of 0.22–220 lg L
1 with regression coefficients of
0.99 reveals high applicability of the method for trace analysis. Under the optimized conditions, the
recoveries are 89.0–96.1 with relative standard deviations (RSD) of less than 9.0%.
Ó 2016 Elsevier Inc. All rights reserved.

⇑ Corresponding author.
E-mail address: m_ghaedi@mail.yu.ac.ir (M. Ghaedi).

http://dx.doi.org/10.1016/j.jcis.2016.07.017
0021-9797/Ó 2016 Elsevier Inc. All rights reserved.
 M. Arabi et al. / Journal of Colloid and Interface Science 480 (2016) 232–239 233

1. Introduction molecule was removed and functional binding sites were achieved
[17]. Unfortunately, this method is time-consuming and the
Prednisolone (11,17-dihydroxy-17-(2-hydroxyacetyl)-10,13-di grinding and sieving are also labor intensive with consumption
methyl6,7,8,9,10,11,12,13,14,15,16,17-dodecahydrocyclopenta [a] of large amount of polymers and incomplete template removal.
phenanthren3-one) (Fig. 1a) is one of synthetic glucocorticoids Additionally, the irregular shape particles lead to reduction of
with extensive usage in the treatment of acute rejection episodes, extraction efficiency especially using MIPs as SPE sorbents [18].
inflammatory conditions such as arthritis, colitis, asthma, certain Nano-structured MIP materials with extremely higher surface-to-
skin rashes and bronchitis [1,2]. Moreover, prednisolone is also volume ratio provide more complete removal of templates and
employed for organ transplants to decrease the risk of organ injec- better site accessibility [19–21]. This technique can produce
tion [3,4]. The metabolism of corticosteroids is highly dependent homogeneous particles with the ideal porous structure decreasing
on individual situation [5–7]. The immunotherapy of acute injec- consumption of toxic organic solvents, grinding and sieving.
tion is done by a fixed drug regimen intravenous on the first, third Moreover, inorganic silica-based polymer provides high mechani-
and fifth days of the treatment. This regime leads to a risk of sub- cal stability, good solvent resistance and high efficiency [22–24].
therapeutic concentrations and toxic effects caused by overdosing Compared to acrylic-based MIP, sol–gel organically modified has
[8]. These indicate the importance of measuring prednisolone in been verified to supply more specific toward the target species
different matrices. Scientists have reported different methods for and allow for faster diffusion of analytes at mild reaction temper-
the pretreatment and determination of prednisolone based on ature [25–28]. One of the MIPs disadvantages while using target
liquid-liquid extraction combined with liquid chromatography molecules as templates to synthesize MIPs is leakage of residual
[9], electrochemical technique [10], solid phase extraction com- template molecules after solvent extraction which lead to error
bined liquid chromatography tandom mass spectrometry [11], in results which simply can be overcome using dummy template
micellar electrokinetic chromatography [12] and solid phase for the MIPs synthesis. As a pre-condition, the dummy molecule
extraction combined liquid chromatography UV spectrometry must resemble the target analyte in terms of shape, size and
[13]. Although these methods were simple and could just solve functionalities without interference in analytical determination
some certain matrix interferences problem, while they suffer from [29,30].
low acceptable selectivity and usage of high amount of organic sol- Li and co-workers [31] reported the molecular imprinting using
vent in pretreatment. Urine samples contain a very low level of aluminum ion (Al3+) to supply Lewis acid sites in the silica matrix.
prednisolone a useful technique should be employed for precon- Metal coordinative interactions provide higher strength and
centration of prednisolone. In this work, highly selective extraction selectivity compared to traditional interactions such as polar,
of prednisolone was under taken by synthesis of a molecularly hydrogen-bond and van der Waals forces [32].
imprinted polymer (MIP) according to a sol-gel method. MIPs This paper describes a novel and inexpensive strategy for
attain great attentions due to their favorable characteristics such synthesis of nano-sized dummy molecularly imprinted amine
as mechanical and chemical robustness, high selectivity with functionalized silica using a small amount of organic solvent. The
target molecule, high capacity and low cost of preparation [14]. prepared MIPs possessed good recognition ability toward pred-
Commonly, MIPs are synthesized by bulk polymerization in a nisolone. Additionally, the leakages of residual templates were per-
mixture solution containing a template molecule, porogen formed using betamethasone as the dummy template. Moreover,
reagents, functional monomer, cross-linker and initiator [15,16]. ultrasonic assisted extraction was employed for desorption of ana-
After the polymerization, ground and sieving, the template lyte. To the best of our knowledge, ultrasonic assisted extraction by
PT-DMIP sorbent based on functionalized silica with metal doping
has not been reported for prednisolone analysis. The attempt
focused on development of simplified prednisolone enrichment
method to enhance the selectivity of prednisolone analysis by
HPLC-UV. The different parameters affecting the extraction effi-
ciency have been optimized using experimental design methodol-
ogy. Finally, the assay was fully validated and applied to the
quantification of prednisolone in different urine samples.

2. Experimental

2.1. Chemicals and reagents

Tetraethyl orthosilicate (TEOS), 3-aminopropyl trimethoxysi-

lane (APTMS) and aluminum chloride hexaydrate (AlCl36H2O)
were obtained from Merck (Darmstadt, Germany). Prednisolone
and betamethasone were obtained from Ministry of Health and
Medical Education (Tehran, Iran), and their chemical structures
are shown in Fig. 1. Other chemicals including ethylene glycol,
methanol, ethanol, DMSO and acetic acid were purchased from
Merck (Darmstadt, Germany). Urine samples were obtained from
Healthy volunteers and stored at 4 °C until use.

2.2. Instrumentation and HPLC analysis

HPLC was performed with an Agilent 1100 liquid chromatogra-

phy (USA), Quaternary Pump (model G1311A), Multiple
Wavelength Detector (model G13658, setting at 254 nm for
Fig. 1. The structures of (a) prednisolone and (b) betamethasone. prednisolone), a sample injection valve with a 20 lL sample loop,
234 M. Arabi et al. / Journal of Colloid and Interface Science 480 (2016) 232–239

and a Knauer C18 column (4.6 mm i.d. 250 mm, 5 lm). Data were sonications (3 min) the caps were removed (final extract volume
collected and analyzed using the Agilent Chemstation software. was 100 lL) and 20 lL the extract was analyzed by HPLC.
The mobile phase is consisted of water – acetonitrile (60:40, v:v)
at flow rate of 1 mL min
1. The mobile phase was filtered through
3. Results and discussion
a 0.45 lm filter and degassed under vacuum before use. The sys-
tem was operated at ambient temperature. The DMIPs were char-
3.1. Preparation and evaluation of nano-sized dummy molecularly
acterized by FE-SEM (VEGA TESCAN USA) and FT-IR (FTIR-8300,
imprinted particles
Shimadzu).
Miniaturized SPE method was based on highly porous silica par-
2.3. Molecularly imprinted and non-imprinted polymer synthesis ticles due to their high surface area and fast mass transfer as a solid
phase. Generally, organic solvents such as ethanol and methanol
APTMS (0.23 mL) and TEOS (1.31 mL) were mixed in vortex were used during sol-gel procedure. Experimental results showed
with the mixtures of 7 mL solution of betamethasone in ethylene that addition of ethylene glycol can apparently increase the pore
glycol (containing approximately 0.5 mol dummy template) and size of the gels while organic solvent such as ethanol would pre-
20 mL H2O. 0.08 mmol aluminum chloride hexaydrate and 0.5 mL vent the reaction progress [33]. In this work, to obtain highly por-
of HCl were added as catalysts immediately, and the mixture was ous structure with a large pore size, ethylene glycol was used in the
stirred vigorously and kept at 45–50 °C for 24 h. Subsequently, sol-gel process. Additionally, ethylene glycol could completely dis-
for removing dummy template, the processed particles were puri- solve dummy template to provide monomer-dummy template
fied by several cycles of centrifugation, decantation and re- pre-polymer complex.
suspension in methanol/acetic acid (9:1 v:v) by sonication until The schematic procedure of preparing DMIP is shown in Fig. 2.
the dummy template was not detected by a UV–VIS Spectropho- Prednisolone has several reactive centers and functional groups
tometer. Finally, the as-prepared silica nano-sized particles were like oxygen and hydroxyl in cooperation to amine group of APTMS
dried at room temperature under vacuum for further use. None for formation of strong electrostatic interactions and hydrogen
imprinted silica polymer (NIP) was prepared according to the same bonds between prednisolone and APTMS. In addition to hydrogen
procedure in the absence of the template. bonds, Al3+ doping gives Lewis acid sites to polymer matrix which
can increase the interaction between polymer and ion pair electron
2.4. Preparation of stock solutions and real samples of analyte via coordination bonding.
Using dummy template avoids the interference of residual tem-
A standard stock solution of prednisolone (1.0 mg mL
1) was plate during extraction. Moreover, dummy template selection is a
prepared by dissolving 10.0 mg of prednisolone in 10 mL of metha- vital factor affecting the selectivity of the sorbent. Because pred-
nol. Urine samples were prepared by spiking a prednisolone solu- nisolone and betamethasone have similar chemical structures,
tion into blank urine to study extraction efficiency under different while their peaks were separated easily in analytical HPLC column,
conditions. All real samples were filtered through a 0.45 lm filter betamethasone was selected as the template of DMIP.
before instrumental analysis. Since, the ratio of functional monomer to cross-linker has an
influence on binding capacity of a polymer we examined the effect
2.5. Procedure of PT-DMIP of different monomer:cross-linker molar ratio in a range of 1:1–
1:6. The obtained polymers in the different molar ratios were used
The prepared imprinted silica particles were applied as sorbents to extract prednisolone from working solution, and the results
for extraction and enrichment of prednisolone in urine sample. showed that 1:5 was the optimum molar ratio of monomer to
General procedures were as follows: 10.0 mg DMIP or NIP were cross-linker. To optimize the quantity of the dummy template, dif-
packed into a 100 lL pipette tip using two small pieces of degrease ferent pre-polymer solutions containing various amounts (0.1–
cotton to avoid adsorbent loss. Firstly, the PT-DMIP was washed by 1 mmol) of betamethasone were prepared. At a low quantity of
methanol (1.0 mL) and deionized water (1.0 mL) to eliminate pos- dummy template the numbers of specific binding sites were
sible adsorbed materials. Then 4.5 mL urine sample was filling into decreased so the binding capacity decreased as well. At a high
a syringe and loaded on the PT-DMIP followed by hexane (100 lL) quantity of dummy template the binding capacity was decreased
washing to remove matrix interferences. After that, 110 lL MeOH presumably because of too low inter space of the specific binding
as an eluent solvent was loaded in the pipette tip and the both ends sites and target molecule couldn’t adsorb to specific cavities
of the pipette tip were sealed with polypropylene caps and the because of steric hindrance. So, the extraction capability of the pre-
miniaturized column was immersed in an ultrasonic bath. After pared DMIP implied that 0.5 mmol was the optimum quantity of

Fig. 2. Schematic representation of the DMIP synthesis.

 M. Arabi et al. / Journal of Colloid and Interface Science 480 (2016) 232–239 235

the dummy template. Additionally, different ratios of cross-linker: absorption stretching of NAH bond at 2940 cm
1, the SiAO
dopant including (1:0.04, 1:0.08, 1:0.12, 1:0.16, 1:0.2) and dopant stretching at 1075 cm
1 and CAH bonds at 2938 and 790 cm
1
free polymer was also investigated and the results revealed1:0.08 are observed, while the peak at 970 cm
1 is corresponding to
ratio gave the highest adsorption capacity. SiAOH. The major band spectra of DMIP and NIP have the same
locations which could prove complete removal of dummy
3.2. Characterization of the nano-sized molecularly imprinted polymer template from DMIPs.
The morphology of nano-sized DMIP under optimization condi-
FT-IR was performed to check successful preparation of tions correspond to synthesis stage was followed by scanning elec-
dummy molecularly imprinted and non imprinted polymers. tron microscopy (Fig. 4). The results confirm that agglomerated
Fig. 3 shows the FT-IR of non imprinted and imprinted polymers particles had nano-sized which provide suitable structure for SPE
after dummy template removal. A broad adsorption band was adsorbents.
observed at 3300 cm
1 which was characteristic OAH stretching, Al doping into the polymer matrix was confirmed by energy
representing adsorbed water on/into the MIP and NIP. The dispersive X-ray spectroscopy (EDS) (Fig. 5).

Fig. 3. FT-IR spectra (a) DMIP after dummy template removal and (b) NIP.

Fig. 4. FE-SEM image of DMIPs.

236 M. Arabi et al. / Journal of Colloid and Interface Science 480 (2016) 232–239

Element Weight% Atomic%

CK 29.87 41

NK 5.88 6.92

OK 32.15 33.13

Al K 4.16 2.54

Si K 27.94 16.41

Totals 100.00

Fig. 5. EDS image of DMIPs.

The adsorption capacity of the sorbent was studied by contact- 120

ing 5 lg mL
1 of prednisolone standard solution continuously with MIP
PT-DMIP at a constant flow rate. The adsorption capacity of sorbent 100 NIP
Extraction efficiency (%)

was calculated on Eq. (1):

80
ðC 0
C r ÞV
Q¼  100 ð1Þ
M
60
While C0 and Cr are the analyte concentrations before and after
adsorption, V is the volume of prednisolone solution (L), and M is 40
the mass of DMIP or NIP (g). According to the results, 0.42 mg of
prednisolone could have saturated of the accessible specific cavities 20
which is proportional to 42.0 mg g
1 of sorbent. Also it was found
that the adsorption capacity of DMIP toward prednisolone was at 0
least three times higher than NIP (11.0 mg g
1). In order to evaluate 0.5 100 220
the presence of selective sites on the prepared imprinted polymer, Concentration (µg L-1)
pipette-tip micro extraction experiments were carried out working
Fig. 6. Extraction efficiencies of prednisolone on PT-DMIP or PT-NIP with different
in parallel with the corresponding NIP. Three urine spiked samples
concentration.
with concentrations of 0.5, 100 and 220 mg L
1 were used for the
SPE tests. As shown in Fig. 6 the extraction efficiencies of pred- lyzed by HPLC. The loss of prednisolone due to the degreased cot-
nisolone on pipette-tip-imprinted polymer were much higher than ton and pipette-tip was very slight and negligible. Optimization
those on pipette-tip-non-imprinted polymer. This result further was carried out by extraction onto blank urine spiked with various
verified the success of the present specific cavities in polymer amounts of prednisolone. To ensure the full description of the opti-
structure. mization approach, the types of eluent and washing solvent effect
have been primarily investigated through a univariate method.
3.3. Optimization of PT-DMIP conditions

To consider the influence of the degreased cotton on analyte 3.3.1. Eluent composition
adsorption the spiked sample solution (100 lg L
1) was loaded Eluent conditions (composition and volume) are important
on blank pipette-tip without any sorbent and the effluent was ana- parameters in the present study and different organic solvents at
 M. Arabi et al. / Journal of Colloid and Interface Science 480 (2016) 232–239 237

various volumes such as ethanol, DMSO, methanol, MeOH:DMSO Table 2

(50:50 v:v) were tested. It was found that methanol exhibited ANOVA for the response surface model.

the highest recovery. Methanol hinders the formation of hydrogen Source Sun of Degree of Mean F-value P-value
bonding between prednisolone and functional monomer, and squares freedom square
prednisolone desorption was easier and thus we apply methanol Model 1453.29 14 103.81 73.12 <0.0001
as eluent agent. A-pH 865.20 1 865.20 609.42 <0.0001
Experiments showed that ultrasound irradiation reduces the B-Sonication time 95.60 1 95.60 67.34 <0.0001
C-Sample volume 136.80 1 136.80 96.36 <0.0001
consumption of eluent solvent. Ultrasound can induce hydrogen D-Sample flow rate 0.84 1 0.84 0.59 0.4527
bonds between MIP and analyte broken and mass transfer process AB 0.23 1 0.23 0.16 0.6958
can be accelerated. Additionally, tests of different eluent volumes AC 6.38 1 6.38 4.49 <0.0512
reveal that 110 lL of methanol was enough to desorbed the ana- AD 0.016 1 0.016 0.011 <0.9178
BC 3.15 1 3.15 2.22 0.1570
lyte with a highly recovery efficiency (96%).
BD 36.91 1 36.91 25.99 0.0001
CD 10.08 1 10.08 7.10 0.0177
3.3.2. Effect of washing solvent A2 248.40 1 248.40 174.97 <0.0001
B2 2.84 1 2.84 2.00 0.1776
A washing solvent was used to remove the matrix interferences
C2 5.18 1 5.18 3.65 0.0755
from the sample that were retented in the solid phase. The effects D2 21.76 1 21.76 15.32 0.0014
of various washing solvents with different volumes (ultra-pure Lack of fit 15.22 10 1.52 0.4246
water, acetone, ethyl acetate, and hexane) were investigated, Pure error 6.07 5 1.21
showing that hexane at 100 lL produced the best recoveries. Total error 1474.59 29

3.3.3. Screening of the effective parameters using central composite

design quite close to the actual experiment. RSM for plotting ER (%) versus
Variable parameters such as sonication time, pH, sample flow significant variables was investigated and shown in Fig. 2S. pH of
rate and volume affect the efficiency of the extraction and purity solution is an important parameter that affects sorption of the drug
of the final extract. A central composite design (CCD) was carried molecules. Hence, investigation of its effect over pH 2.5–8.5
out for process optimization of the variables and statistical signif- revealed that maximum sorption was obtained at the pH of 4.0.
icance of the factors. The effect of critical factors, interactions and The hydrogen bonds between prednisolone and sorbent as an
model efficiency were evaluated by ANOVA analysis. The center essential interaction is highly required for construction of binding
points are used to determine the experimental error and the sites. At very low pHs the acidic moiety of the polymer from pro-
reproducibility of the data. The independent variables are coded tonation of the nitrogen atoms of the sorbent and the hydroxyl
to the (
1 and +1) interval where the low and high levels are groups of prednisolone lead to significant reduction in pred-
coded as
1 and +1, respectively (Table 1). The axial points are nisolone accumulation, which reduced extraction efficiency. In this
located at distance from center and make the design rotatable. pH, the nitrogen atoms of sorbent are in their protonated form
Response surface methodology (RSM) is a combination of mathe- while prednisolone has neutral structure and hydrogen bonds
matical and statistical techniques used to developing, improving has main contribution on accumulation of analyte on polymer
and optimizing the processes and allows for the determination and template. Also the enhanced adsorption of prednisolone by
and evaluation of the relative significance and interaction of all polymer could be attributed to the addition of Al3+ ions, some of
variables on the process even in the presence of complex interac- Al3+ was incorporated into silica matrix and become Lewis acid
tions. RSM has a modeling step followed by optimal region deter- sites during the sol-gel process. The carbonyl groups of pred-
mination. The modeling is performed adjusting first or second nisolone in their molecular form lead to enhanced interaction of
order polynomials equations to the experimental responses template and Lewis acid sites in polymer and the recovery. Addi-
obtained in the experimental design followed by analysis of vari- tion of Al3+ ion during the preparation of MIP rendered much larger
ance (ANOVA). The validated model can be plotted in tridimen- surface area allowing the polymer to accumulate and take up
sional graph and generate surface response that corresponds to higher amount of prednisolone. The effect of ultrasonic time in
response function used for determination of the best operating the range of 0.5–5.5 min (Fig. 2Sa and c) revealed that the best
conditions of a process. recovery was obtained at 3 min. Creating a binding of the analyte
Table 1S lists 30 random experiments selected to minimize the by polymer is also dependent on the sample volume. A solution
effect of uncontrolled variables. As it is shown in Table 2, P values of prednisolone with various volumes (1.5–7.5 mL) was loaded
of significant variables were less than 0.05, while their F values on to the pipette tip. The results revealed that until using sample
were more than 0.05 of all terms in the polynomial equation volume 4.5 lead to achievement of maximum analyte sorption.
within 95% confidence level. Coefficient of regression (R2 = 0.986) As can be seen from Table 2 and Fig. 1Sa (Pareto chart) p-values
indicate a good relationship between the experimental data and of sample flow rate is more than 0.05 which indicates that this
the fitted model. Fig. 1Sb showed that the predicted values are variable is not significant. At a low flow rate the mass transfer of
the analyte from sample solution to DMIPs was increased, to
Table 1
Experimental design summary.
Table 3
Factor Low High Low High Mean Std. Intra and inter-day precision data for the assays of prednisolone in spiked urine
actual actual coded coded dev. samples.

pH 4.0 7.0
1.0 1.0 5.5 1.3 Samples Add (lg L
1) Recovery (%) RSD (%) n = 4
Sonication time 2.0 4.0
1.0 1.0 3.0 0.8
Intra-day Inter-day
(min)
Sample volume 3.0 6.0
1.0 1.0 4.5 1.3 1 89.0 5.1 8.8
(mL) Urine 10 90.8 4.9 8.4
Sample flow rate 1.0 2.0
1.0 1.0 1.5 0.4 50 94.0 3.2 6.6
(mL min
1) 100 96.1 2.2 3.6
238 M. Arabi et al. / Journal of Colloid and Interface Science 480 (2016) 232–239

Fig. 7. The chromatograms of the urine samples (a) the blank urine and (b) spiked sample of prednisolone.

achieve satisfactory extraction efficiency within acceptable 4. Conclusion

extraction time 1 mL min
1 was chosen as sample flow rate.
Based on the data analysis semi-empirical expression for A durable and inexpensive hydrophilic dummy molecular
evaluation of ER (%) was obtained as follow: imprinting polymer based on a sol-gel method was synthesized
for the extraction of prednisolone from urine samples. The results
ER ð%Þ ¼ þ85:67
6:00X1 þ 2:00X2 þ 2:39X3
1:52X3 X4 demonstrate that present method is unique in term of simplicity,

3:01X21 þ 0:89X24 ð2Þ facility and low consumption of toxic organic solvent by ultrasonic
assisted. In additional, the effect of template bleeding was com-
As it is shown in Fig. 3S the desirability option was examined pletely expiring by employing betamethasone as dummy template.
and achieved by the STATISTICA 7.0 software. The highest ER (%) Variable parameters including pH, sonication time, sample flow
was achieved at sonication time (3 min), low pH (4) and sample rate and sample volume were evaluated by experimental design
volume (4.5 mL). The final optimized conditions of the variables methodology. Finally, different method for determination pred-
to obtain the maximum extraction efficiency were calculated and nisolone with purposed method was compared in Table 4
their significant interaction was investigated by RSM (Fig. 2Sa-c). [11,12,34].

3.3.4. Validation of the PT-DMIP procedure and real sample analysis Acknowledgements
A calibration curves was established for prednisolone in urine
spiked samples over a concentration range of 0.22–220 lg L
1 with The authors express their appreciation to the Graduate School
regression coefficients R2 = 0.99 under optimized conditions. The and Research Council of the University of Yasouj for financial sup-
applicability of the PT-DMIP-HPLC-UV procedure was checked port of this work.
using the spiked urine samples at different concentrations and
the inter- and intra-assay precisions are shown in Table 3. As it Appendix A. Supplementary material
can be seen in Fig. 7 no peak detected in blank urine that interferes
with prednisolone. The limits of detection (LOD) and quantification Supplementary data associated with this article can be found, in
(LOQ) were determined based on the signal to noise (S/N) ratio of the online version, at http://dx.doi.org/10.1016/j.jcis.2016.07.017.
3–10. The LOD and LOQ for prednisolone standard were 0.085 and
0.2 lg L
1, respectively.
References

[1] J.A. Miller, D.D. Munro, Topical corticosteroids: clinical pharmacology and
Table 4 therapeutic use, Drugs 19 (1980) 119–134.
[2] M.S. Shaefer, Current topics in immunotherapy and the role of the pharmacist
Comparison of different methods for preconcentration of prednisolone with proposed
on a solid organ transplant service, Pharmacotherapy 11 (1991) 136S–141S.
method.
[3] R.N. Goyal, M. Oyama, N. Bachheti, S.P. Singh, Fullerene C60 modified gold
Method LOD Linear range Recovery Ref. electrode and nanogold modified indium tin oxide electrode for prednisolone
(lg L
1) (lg L
1) (%) determination, Bioelectrochemistry 74 (2009) 272–277.
[4] L. Wennberg, H. Genberg, G. Tydén, Preapheresis immunosuppressive
LLE-HPLC-UV 25 25–400 81.4–88.2 [34] induction: necessary or harmful?, Transplantation 84 (2007) S37–S39
SPE-UPLC-MS 0.5 2–1000 98.7–102.3 [11] [5] W.J. Jusko, J.Q. Rose, Monitoring prednisone and prednisolone, Ther. Drug
SPE-HPLC-UV 6.25 25–250 89.29–95.50 [12] Monit. 2 (1980) 169–176.
PT-DMIP-HPLC-UV 0.085 0.2–220 89.0–96.1 Present work [6] V. Garg, W.J. Jusko, Bioavailability and reversible metabolism of prednisone
and prednisolone in man, Biopharm. Drug Dispos. 15 (1994) 163–172.
 M. Arabi et al. / Journal of Colloid and Interface Science 480 (2016) 232–239 239

[7] R. Venkataramanan, K. Habucky, G.J. Burckart, R.J. Ptachcinski, Clinical surface using carbodiimide coupling for chemical sensing, J. Colloid Interface
pharmacokinetics in organ transplant patients, Clin. Pharmacokinet. 16 Sci. 461 (2016) 1–8.
(1989) 134–161. [21] T. Kamra, S. Chaudhary, C. Xu, N. Johansson, L. Montelius, J. Schnadt, L. Ye,
[8] S. Noe, J. Böhler, E. Keller, A.W. Frahm, Determination of prednisolone Covalent immobilization of molecularly imprinted polymer nanoparticles
in serum: method development using solid-phase extraction and using an epoxy silane, J. Colloid Interface Sci. 445 (2015) 277–284.
micellar electrokinetic chromatography, J. Pharm. Biomed. Anal. 18 [22] M. Wu, Y. Chen, R.a. Wu, R. Li, H. Zou, B. Chen, S. Yaob, The synthesis of
(1998) 471–476. chloropropyl-functionalized silica hybrid monolithic column with
[9] Y. Liu, K. Zhu, J. Wang, X. Huang, G. Wang, C. Li, J. Cao, S. Ding, Simultaneous modification of N,N-dimethyl-N-dodecylamine for capillary
detection and comparative pharmacokinetics of amoxicillin, clavulanic acid electrochromatography separation, J. Chromatogr. A 1217 (2010) 4389–4394.
and prednisolone in cows’ milk by UPLC–MS/MS, J. Chromatogr. B 1008 (2016) [23] M. Arabi, M. Ghaedi, A. Ostovan, Development of dummy molecularly
74–80. imprinted based on functionalized silica nanoparticles for determination of
[10] J. Smajdor, R. Piech, B.P. Bator, A novel method of high sensitive determination acrylamide in processed food by matrix solid phase dispersion, Food Chem.
of prednisolone on renewable mercury film silver based electrode, 210 (2016) 78–84.
Electroanalysis 28 (2015) 394–400. [24] J.E. Lofgreen, G.A. Ozin, Controlling morphology and porosity to improve
[11] H. Li, X. Xia, Y. Xue, S. Tang, X. Xiao, J. Li, J. Shen, Simultaneous determination performance of molecularly imprinted sol–gel silica, Chem. Soc. Rev. 43 (3)
of amoxicillin and prednisolone in bovine milk using ultra-high performance (2014) 911–933.
liquid chromatography tandem mass spectrometry, J. Chromatogr. B 900 [25] R.G.d.C. Silva, F. Augusto, Sol–gel molecular imprinted ormosil for solid-phase
(2012) 59–63. extraction of methylxanthines, J. Chromatogr. A 1114 (2006) 216–223.
[12] J.M.L. Gallego, J.P. Arroyo, Determination of prednisolone, naphazoline, and [26] W. Cummins, P. Duggan, P. McLoughlin, A comparative study of the potential
phenylephrine in local pharmaceutical preparations by micellar electrokinetic of acrylic and sol–gel polymers for molecular imprinting, Anal. Chim. Acta 542
chromatography, J. Sep. Sci. 26 (2003) 947–952. (2005) 52–60.
[13] H. Hashem, T. Jira, Chromatographic applications on monolithic columns: [27] Y.-L. Wang, Y.-L. Gao, P.-P. Wang, H. Shang, S.-Y. Pan, X.-J. Li, Sol–gel
determination of triamcinolone, prednisolone and dexamethasone in molecularly imprinted polymer for selective solid phase microextraction of
pharmaceutical tablet formulations using a solid phase extraction and a organophosphorous pesticides, Talanta 115 (2013) (2013) 920–927.
monolithic column, Chromatographia 61 (2005) 133–136. [28] R. Gao, X. Mu, J. Zhang, Y. Tang, Specific recognition of bovine serum albumin
[14] Y. Hao, R. Gao, D. Liu, B. Zhang, Y. Tang, Z. Guo, Preparation of biocompatible using superparamagnetic molecularly imprinted nanomaterials prepared by
molecularly imprinted shell on superparamagnetic iron oxide nanoparticles two-stage core–shell sol–gel polymerization, J. Mater. Chem. B 2 (7) (2014)
for selective depletion of bovine hemoglobin in biological sample, J. Colloid 783–792.
Interface Sci. 470 (2016) 100–107. [29] X. Hu, X. Wu, F. Yang, Q. Wang, C. He, S. Liu, Novel surface dummy molecularly
[15] L. Zhu, G. Xu, F. Wei, J. Yang, Q. Hu, Determination of melamine in powdered imprinted silica as sorbent for solid-phase extraction of bisphenol A from
milk by molecularly imprinted stir bar sorptive extraction coupled with HPLC, water samples, Talanta 148 (2016) 29–36.
J. Colloid Interface Sci. 454 (2015) 8–13. [30] M. Arabi, A. Ostovan, M. Ghaedi, M.K. Purkait, Novel strategy for synthesis of
[16] B.B. Prasad, A. Srivastava, M.P. Tiwari, Highly selective and sensitive analysis of magnetic dummy molecularly imprinted nanoparticles based on
dopamine by molecularly imprinted stir bar sorptive extraction technique functionalized silica as an efficient sorbent for the determination of
coupled with complementary molecularly imprinted polymer sensor, J. Colloid acrylamide in potato chips: optimization by experimental design
Interface Sci. 396 (2013) 234–241. methodology, Talanta 154 (2016) 526–532.
[17] Z. Li, C. Qin, D. Li, Y. Hou, S. Li, J. Sun, Molecularly imprinted polymer for [31] T.-J. Li, P.-Y. Chen, P.-C. Nien, C.-Y. Lin, R. Vittal, T.-R. Ling, K.-C. Ho, Preparation
specific extraction of hypericin from Hypericum perforatum L. herbal extract, J. of a novel molecularly imprinted polymer by the sol–gel process for sensing
Pharm. Biomed. Anal. 98 (2014) 210–220. creatinine, Anal. Chim. Acta 711 (2012) 83–90.
[18] L. Ye, K. Mosbach, Molecularly imprinted materials: towards the next [32] J. Huang, Y. Hu, Y. Hu, G. Li, Development of metal complex imprinted solid-
generation, Mater. Res. Soc. Symp. Proc. 723 (2002) 51–59. phase microextraction fiber for 2,20 -dipyridine recognition in aqueous
[19] M. Arabi, M. Ghaedi, A. Ostovan, J. Tashkhourian, H. Asadallahzadeh, Synthesis medium, Talanta 83 (2011) 1721–1729.
and application of molecularly imprinted nanoparticles combined ultrasonic [33] L. Siqiang, T. Kairong, Low temperature synthesis of monolithic silica glass
assisted for highly selective solid phase extraction trace amount of celecoxib from the system Si(OC2H5)4–H2O-HCI-HOCH2CH2OH by the sol-gel method, J.
from human plasma samples using design expert (DXB) software, Ultrason. Non-Cryst. Solids 100 (1988) 254–262.
Sonochem. 33 (2016) 67–76. [34] N.R. Scott, J. Chakraborty, V. Marks, Determination of prednisolone,
[20] T. Kamra, S. Chaudhary, C. Xu, L. Montelius, J. Schnadt, L. Ye, Covalent prednisone, and cortisol in human plasma by high-performance liquid
immobilization of molecularly imprinted polymer nanoparticles on a gold chromatography, Anal. Biochem. 108 (1980) 266–268.