Biosensors and Bioelectronics ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Contents lists available at ScienceDirect

Biosensors and Bioelectronics

Enzyme-controlled dissolution of MnO2 nanoflakes with enzyme

art ic l e i nf o a b s t r a c t

Article history: A new colorimetric immunosensing platform accompanying enzyme cascade amplification strategy was
Received 19 November 2015 fabricated for quantitative screening of small-molecular mycotoxins (aflatoxin B1, AFB1 used in this case)
Received in revised form coupling with enzyme-controlled dissolution of MnO2 nanoflakes. The visual colored assay was executed
29 November 2015
by high-efficient MnO2–3,3′,5,5′-tetramethylbenzidine (TMB) system (blue). In the presence of ascorbic
Accepted 14 December 2015
acid, MnO2 nanoflakes were dissolved into Mn2 þ ions, thus resulting in a perceptible color change from
blue to colorless. The reaction could be weakened through ascorbate oxidase to catalyze ascorbic acid
Keywords: into dehydroascorbic acid, which indirectly depended on the concentration of ascorbate oxidase. By
MnO2 nanoflakes using ascorbate oxidase/ anti-AFB1 antibody-labeled gold nanoparticles, a novel competitive-type col-
Colorimetric immunoassay
orimetric enzyme immunoassay was developed for detection of AFB1 on AFB1-bovine serum albumin
Aflatoxin B1
(BSA)-conjugated magnetic beads. Upon addition of target AFB1, the analyte competed with the con-
Enzyme cascade amplification
jugated AFB1-BSA on the magnetic beads for the labeled anti-AFB1 antibody on the gold nanoparticles.
Under optimal conditions, the absorbance decreased with increasing target AFB1 within the dynamic
range of 0.05–150 ng mL  1 with a detection limit of 6.5 pg mL  1 at the 3Sblank level. The precision and
specificity of the MnO2–TMB-based immunosensing system were acceptable. In addition, method ac-
curacy was further validated for monitoring spiked peanut samples, giving results matched well with
those obtained from commercialized AFB1 ELISA kit.
& 2015 Elsevier B.V. All rights reserved.

1. Introduction added H2O2 was often unstable and easily decomposed in air (Ray
et al., 2014). Recent research has looked to develop innovative and
Sensitive and specific determination of biotoxins including marine powerful nanocatalysts with peroxidase-like activity (e.g., Fe3O4, Pt,
toxins and mycotoxins has been attracting great attention in the en- Co3O4, graphene oxide and bimetallic nano-alloys), controlling and
vironmental pollution and food monitoring (Gawley et al., 2002; Bo- tailoring their properties in a very predictable manner to meet the
vee et al., 2011; Samdal et al., 2015; Nathanail et al., 2015; Lin et al., requirements of specific applications (Liang et al., 2013; Qin et al.,
2015a). A huge effort has been expanded in the field of simple and 2014; Wu et al., 2014; Wang et al., 2014; Gao et al., 2015). A more
user-friendly assay development. Based on different signal-generation preferable approach that can in-situ generate H2O2 and does not need
principles, colorimetric detection methods can be utilized to realize the addition of H2O2 would be advantageous. Wang et al. utilized
this purpose because of their simplicity, practicality and high-speed glucose oxidase to catalyze glucose for generation of H2O2 (Wang et al.,
operation (Soh et al., 2015). Enzyme-based labeling strategies are 2015). Our group reported a simple colorimetric immunoassay by
usually used during this measurement (Berg et al., 2015; Geng et al., coupling with the TMB-Ag(I) system, in which Ag(I) could oxidize TMB
2015; Qu et al., 2014). For example, horseradish peroxidase (HRP) directly in the absence of H2O2 (Lai et al., 2015). However, we recently
could catalyze 3,3′,5,5′-tetramethylbenzidine (TMB) for development found that the 1:1 oxidation in the TMB-Ag(I) system could be im-
of blue color in the presence of hydrogen peroxide (H2O2) (Chen et al., proved by using enzyme mimics, e.g., nanocatalysts (Kumari et al.,
2014). However, the detection sensitive was always limited due to the 2014; Su et al., 2012; Zhang et al., 2014a, 2014b; Gao et al., 2014).
limitation of the labeling amount on each antibody. Moreover, the To acquire a high sensitivity and a low detection limit, the signal
amplification would be very crucial. Wang et al. utilized folate-poly-
n oxometalate hybrid spheres as the oxidase mimics for the colorimetric
Corresponding authors.
E-mail addresses: dianping.tang@fzu.edu.cn (Q. Wei), immunoassay of human hepatocellular liver carcinoma cell line (Wang
dianping.tang@fzu.edu.cn (D. Tang). et al., 2011). Perez's group employed ceria nanoparticles with oxidase-

http://dx.doi.org/10.1016/j.bios.2015.12.035
0956-5663/& 2015 Elsevier B.V. All rights reserved.

Please cite this article as: Lai, W., et al., Biosensors and Bioelectronics (2015), http://dx.doi.org/10.1016/j.bios.2015.12.035i
2 W. Lai et al. / Biosensors and Bioelectronics ∎ (∎∎∎∎) ∎∎∎–∎∎∎

like activity as the labeled agents for colorimetric immunoassay (Asati water with the aid of 0.1 M HCl. 0.1 M citrate acid and 0.2 M Na2HPO4
et al., 2009). Manganese dioxide (MnO2) is a material that has been were mixed and adjusted to pH 4.0 as the TMB substrate solution. A
applied in dry cell batteries, supercapacitors, ceramics, biosensors and pH 7.4 PBS was prepared by using 2.9 g Na2HPO4, 0.24 g KH2PO4,
organic synthesis (Wan et al., 2012). Moreover, MnO2 exhibits huge 0.2 g KCl and 8.0 g NaCl in 1000 mL distilled water. The washing and
potential prospect because of the abundance of manganese in the blocking buffer were obtained by adding 0.05% Tween 20 (v/v) and
earth, low cost, low toxicity and the relative high activity. As is well 0.1 wt % BSA in PBS, respectively.
known, MnO2 nanostructures could catalyze TMB to produce a blue
color in the absence of H2O2. It has been reported that various-shaped 2.2. Synthesis of two-dimensional MnO2 nanoflakes
nanostructures often exhibited different catalytic efficiencies (Wan
et al., 2012; Liu et al., 2012). Inspiringly, the emergence of MnO2 na- MnO2 nanoflakes were synthesized by using a typical hot-injec-
noflakes (o100 nm, a kind of two-dimensional nanomaterials) opens tion method according to this literature with minor modification (Wei
a new horizon for advanced development of colorimetric assay (Wei et al., 2014). Before synthesis, two solutions including 0.3644 g CTAB
et al., 2014). Meanwhile, we also noticed that some chemicals (e.g., and 0.0316 g KMnO4 in 10 mL ultrapure water were prepared, re-
ascorbic acid and glutathione) could reduce MnO2 to Mn2þ ion, which spectively. And then, CTAB aqueous solution was heated to 140 °C in
lost the peroxidase-like activity or oxidase-like activity (Zhai et al., the oil bath under vigorous stirring. Following that, the above-pre-
2014; Khan et al., 2005; Luo et al., 2004; Zhang et al., 2014a, 2014b), pared KMnO4 aqueous solution was quickly added into CTAB solution,
thereby resulting in the change in the color from blue to colorless in and the temperature was maintained at 140 °C for 2 h. Finally, the
the presence of TMB. More favorably, ascorbic acid (AA) can be oxi- resulting suspension was centrifuged (three times, 12,000 g, 10 min)
dized to dehydroascorbic acid (DAA) by ascorbate oxidase (AOx). To and washed with acetone. The obtained MnO2 nanoflakes were re-
this end, our motivation in this work is to construct a new colorimetric dispersed into 1.0-mL distilled water for further use.
detection system through enzyme-controlled MnO2–TMB system for
development of colorimetric immunoassay with high sensitivity. 2.3. Monitoring of AOx activity based on MnO2-to-Mn(II) conversion
Aflatoxins, the highly toxic secondary metabolites produced by a with MnO2–TMB system
number of different fungi, are present in a wide range of food and
feed commodities, and are assumed significant because of their de- To monitor the AOx activity, 10 μL of AOx standard with dif-
leterious effects on human beings, poultry and livestock (Elzupir and ferent concentrations from 0 to 2000 mU mL  1 was initially mixed
Alamer, 2014; Arduini et al., 2010; Lin et al., 2015b). The major afla- with 50 μL of 300 μM AA in pH 6.5 Tris–HCl buffer, and incubated
toxins of interest are designated B1, B2, G1 and G2, however, aflatoxin for 15 min at room temperature. Following that, 50 μL of the
B1 (AFB1) is usually predominant and most toxic. By using AFB1 as a above-prepared MnO2 nanoflakes was injected into the mixture
model analyte, we herein design a new MnO2–TMB-based colori- and reacted for 2 min. Afterwards, 100 μL of TMB substrate solu-
metric immunoassay for sensitive monitoring of AFB1 on bovine tion (1.0 mM, ethanol: pH 4.0 Tris–HCl buffer ¼1:9) was added to
serum albumin-AFB1 (BSA-AFB1)-modified magnetic beads by using the cell and reacted for 1.0 min at room temperature. Finally, a
AOx and anti-AFB1 antibody-labeled nanogold particles as the com- microplate reader (Tecan Infinite 200 Pro, TECAN, Switzerland)
petitors. With a competitive-type assay format, the carried AOx ac- was used for monitoring of the resultant solution, and the absor-
companying gold nanoparticles initially catalyzes AA to DAA, and bance was recorded at λ ¼652 nm.
decreases the dissolution of MnO2 nanoflakes, thus resulting in a
strong blue color and a high absorbance. The color and absorbance 2.4. MnO2–TMB-based colorimetric immunoassay toward target
decreases with the increasing target AFB1 in the sample. MnO2 na- AFB1
noflakes are not only used as the enzymatic mimics, but also utilized
for progression of enzyme cascade reaction. The aim of this study is to Scheme 1 gives schematic illustration of MnO2–TMB-ba-
set up a high-efficient colorimetric immunoassay without the parti- sed colorimetric immunoassay. The preparation procedure of
cipation of H2O2 during the measurement. AFB1-BSA-conjugated magnetic beads (designated as AFB1-MB)
and AOx/anti-AFB1-labeled gold nanoparticles (designated as Ab-
AuNP-AOx) were described in detail in our previous report with
2. Experimental minor modification (Lai et al., 2015). The assay toward target AFB1
was carried out as follows. Initially, 25 μL of AFB1 standard or
2.1. Material and chemicals sample and 50 μL of Ab-AuNP-AOx (C[Au] E38.8 nM) were added
in turn into 25 μL of AFB1-MB (10 mg mL  1) in a 200-μL PCR tube.
Polyclonal anti-AFB1 antibody produced in rabbit (product no.: Then the mixture was incubated for 60 min at 37 °C with gentle
A8679, fractionated antiserum buffered aqueous solution) (Im- shaking. After being separated by an external magnet and washed
munogen: aflatoxin B1-KLH; Physical form: 0.01 M pH 7.4 phosphate- with the washing buffer, 50 μL of AA (300 μM) in pH 6.5 Tris–HCl
buffered saline solution containing 15 mM sodium azide), AFB1 buffer was injected into the precipitate and incubated for 15 min
standards from Aspergillus flavus (product no.: A6636), 3,3′,5,5′-tet- at room temperature. After that, 50 μL of the above-prepared
ramethylbenzidine (TMB), ascorbate oxidase (AOx), and bovine ser- MnO2 nanoflakes was thrown to the mixture and reacted for
um albumin (BSA) were purchased from Sigma-Aldrich (Shanghai, 2 min. Finally, TMB (100 μL, 1.0 mM) substrate solution was added
China). AFB1–BSA conjugate was obtained from the School of Food in the mixture, and the absorbance was recorded as above.
Science and Technology, Jiangnan University (Wuxi, China). Cetyl-
trimethylammonium bromide (CTAB) was achieved from Genview
(USA). L-Ascorbic acid (AA) was acquired from Beijing Dingguo Bio- 3. Results and discussion
technol. Inc. (Beijing, China). Potassium permanganate (KMnO4) was
gotten from Fuchen Chemicals (Tianjin, China). All other reagents 3.1. Characterization of MnO2 nanoflakes and MnO2–TMB-based
were used as received without further purification. All water used colorimetric assay
was the Milli-Q ultrapure grade (EMD Millipore). A pH 9.6 carbonate
buffer solution was prepared by using 1.59 g Na2CO3 and 2.93 g In this work, MnO2 nanoflakes were synthesized by a typical
NaHCO3 in 1000 mL water. A pH 6.5 Tris–HCl buffer (4.0 mM) was hot-injection method on the basis of reaction between KMnO4 and
prepared by adding the corresponding chemicals in 100 mL distilled CTAB at 140 °C. During the experiment, we observed that MnO2

Please cite this article as: Lai, W., et al., Biosensors and Bioelectronics (2015), http://dx.doi.org/10.1016/j.bios.2015.12.035i
 W. Lai et al. / Biosensors and Bioelectronics ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 3

Scheme 1. Schematic illustration of MnO2 nanoflakes-mediated colorimetric immunoassay on bovine serum albumin-AFB1 (BSA-AFB1)-modified magnetic beads by using
AOx/anti-AFB1 antibody-labeled nanogold particles with enzyme cascade amplification strategy and the MnO2–TMB system.

nanoflakes were formed immediately upon addition of KMnO4 the morphology of the nanomaterials was planar sheet-like and
into CTAB solution. Meanwhile, the color of the mixture was ultrathin (Fig. 1B), indicating that the nanostructures were neither
changed from violet to brown. Fig. 1A and B shows typical trans- nanoparticles nor nanotubes. Such a two-dimensional (2D) nano-
mission electron microscope (TEM, FEI Titan 80-300, Hillsboro, flakes structure could provide a large surface area for the reaction
USA) images of the synthesized MnO2 nanostructures. Obviously, with TMB. As seen from Fig. 1A, the average size of the as-prepared

Fig. 1. (A,B) TEM images of the as-synthesized MnO2 nanoflakes; (C) dependence of 1.0 mM TMB in the absorbance intensity on different-concentration MnO2 nanoflakes;
and (D) comparison of 1.0 mM TMB in the absorbance intensity between MnO2–TMB system and AgNO3-TMB system towards different-concentration (a) MnO2 nanoflakes
and (b) AgNO3.

Please cite this article as: Lai, W., et al., Biosensors and Bioelectronics (2015), http://dx.doi.org/10.1016/j.bios.2015.12.035i
4 W. Lai et al. / Biosensors and Bioelectronics ∎ (∎∎∎∎) ∎∎∎–∎∎∎

MnO2 nanoflakes was  70 nm. Logically, one important question added into the solution or not, no absorption peaks were achieved
arises to whether the as-synthesized MnO2 nanoflakes could be (Fig. 2B, curve “a” vs. curve “b”). Meanwhile, the color of TMB
used as enzyme mimics to catalyze TMB in the absence of H2O2. To solution was not changed (Fig. 2B, photographs “a–b”). The result
demonstrate this issue, 1.0 mM TMB (used as an example) was indicated that Mn(II) could not catalyze the TMB for the devel-
employed for direct incubation with MnO2 nanoflakes with dif- opment of colorimetric assay.
ferent concentrations. The obtained product was monitored by Furthermore, we also investigated the effect of ascorbic acid on
using a microplate reader. As shown in Fig. 1C, the absorbance the MnO2–TMB system, i.e., whether the added ascorbic acid could
increased with the increasing concentration of MnO2 nanoflakes. weaken the MnO2–TMB system to generate the visual color. Two
The results indicated that the as-prepared MnO2 nanoflakes could strong characteristic peaks (curve “a”) and a thick blue color
catalyze the TMB to produce a visual color and cause the change in (photograph “a”) could be achieved when mixture of TMB with
the absorbance in the absence of H2O2. MnO2 nanoflakes (Fig. 2C). When MnO2 nanoflakes were first re-
To further elucidate the advantages of the as-synthesized MnO2 acted with excess ascorbic acid, however, the subsequently added
nanoflakes in the MnO2–TMB system, a comparative study with TMB did not cause appearance of blue color (Fig. 2C, photograph
our previously reported system (i.e., Ag þ –TMB system) (Lai et al., “b”). Meanwhile, two absorption peaks for the MnO2–TMB system
2015) was carried out by assaying the same-concentration MnO2 were disappeared (Fig. 2C, curve “b”). The reason might be as-
or AgNO3 toward the change of 1.0 mM TMB in the absorbance cribed to the fact that MnO2 nanoflakes in the detection solution
intensity (Fig. 1D). As seen from Fig. 1D, the MnO2–TMB system were dissolved into Mn(II) by AA. Hence, the results revealed that
could exhibit a higher sensitivity and stronger absorbance in- the MnO2–TMB colorimetric detection system could be weakened
tensity relative to the Ag þ -TMB system. That is to say, introduction by AA. Significantly, the residual amount of AA in the same could
of MnO2 nanoflakes could display strong catalytic efficiency to- be controlled through AOx toward the oxidation of AA. In other
ward the same-concentration TMB, which provided a precondition words, the MnO2–TMB system could be used for the monitoring of
for the development of highly sensitive colorimetric assay. AOx activity.
To ensure the feasibility of MnO2 nanoflakes-based colorimetric
assay, some control tests was carried out. Recent reports have 3.2. Optimization of experimental conditions
found that MnO2 could catalyze TMB, and produce a blue color
with an absorption peak at 652 nm. As shown from curves “a–b” in To acquire an optimal analytical performance, some experi-
Fig. 2A, no characteristic absorption peaks were observed at MnO2 mental conditions influencing the analytical properties of the
nanoflakes (curve “a”) or TMB (curve “b”) alone. After mixture of MnO2–TMB system should be studied. In this work, the visual
TMB with MnO2 nanoflakes, two strong characteristic peaks were color derived from the reaction between TMB and MnO2 nano-
appeared (Fig. 2A, curve “c”). Moreover, the color of the solution flakes. Fig. 3A shows the dynamic response curve of the MnO2–
was changed from colorless to blue (Fig. 2A, photograph “c”), in- TMB reaction relative reaction time. The reaction was very fast,
dicating the catalytic ability of MnO2 nanoflakes toward TMB. and reached to the steady-state equilibrium within one minute.
Maybe, we might be asked whether Mn(II) could catalyze TMB. To avoid possible error resulting from different additions of TMB
Fig. 2B shows the UV–vis adsorption spectra of TMB solution in the and deduct the responses induced by various-substrate reaction,
absence and presence of Mn(II). No matter whether Mn(II) was the signal was recorded from 30 s (after the addition of sample)

Fig. 2. UV–vis absorption spectra of (A-a) MnO2 nanoflakes, (A-b) TMB, (A-c) MnO2 nanoflakesþ TMB, (B-a) TMB, (B-b) TMB þMn(II), (C-a) MnO2 nanoflakes þTMB, and (C-b)
AA þMnO2 nanoflakesþ TMB (Insets: the corresponding photographs) (100 μL of 1.0 mM TMB and 50 μL of the above-prepared MnO2 nanoflakes used in the case).

Please cite this article as: Lai, W., et al., Biosensors and Bioelectronics (2015), http://dx.doi.org/10.1016/j.bios.2015.12.035i
 W. Lai et al. / Biosensors and Bioelectronics ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 5

Fig. 3. Effects of (A) reaction time between TMB and MnO2 nanoflakes and (B) concentration of ascorbic acid on the absorbance of the MnO2–TMB system (Inset B: Linear
plots); (C) UV–vis absorption spectra of (a) AOx þAA þ MnO2 þ TMB, (b) AOxþ MnO2 þTMB, (c) AA þMnO2 þTMB, (d) AOx þMnO2 and (e) AOxþ TMB, respectively (Insets: The
corresponding photographs for figure “C”); and (D) the responses of the MnO2–TMB system toward different-activity AOx (100 μL of 1.0 mM TMB, 50 μL of the above-
prepared MnO2 nanoflakes and 300 μM AA used in the case).

until equilibrium was reached, 1.0 min, for the colorimetric introduction of AOx and AA in the MnO2–TMB system was not
development. almost changed the characteristic absorption peaks and visual
Normally, ascorbic acid can reduce MnO2 nanoflakes to Mn(II) color. When only AA was present in the MnO2–TMB system,
ions, thereby inhibiting the development of the MnO2–TMB sys- however, the absorption peaks were disappeared (curve “c”) and
tem in the color and decreasing the absorbance of the system. To the solution became colorless (photograph “c”) in Fig. 3C. The re-
this end, we investigated the responses of the MnO2–TMB system sults indicated that the added AA could be consumed by AOx, thus
(100 μL of 1.0 mM TMB and 50 μL of the above-prepared MnO2 displaying the inherent color of the MnO2–TMB system. Moreover,
nanoflakes used in the case) toward ascorbic acid standards with the presence of AOx could not cause the significant change in the
different concentrations. As shown in Fig. 3B, the absorbance de- absorbance and visual color of the MnO2–TMB system (Fig. 3C,
creased with the increasing AA concentration, and tended to level curve “b” and photograph “b”). Meanwhile, we also observed that
off after 300 μM. An obvious jumping plot could be observed at MnO2 nanoflakes (curve “d” and photograph “d”) or TMB (curve
300 μM AA. That it to say, the change in the absorbance was more “e” and photograph “e”) alone could not react with AOx to induce
obvious with the decreasing AA concentration at r300 μM than development of the absorbance and blue (Fig. 3C). Comparison
that of high-concentration ascorbic acid. Considering development curve/photograph “a” with curve/photograph “c”, the MnO2–TMB
of the following MnO2–TMB-based immunoassay, 300 μM AA system could be controlled by AOx in the presence of AA. Fur-
should be preferable in this work. thermore, we utilized the MnO2–TMB system for the detection of
AOx activity by using AA as the enzymatic substrate. The assay
3.3. Control tests and monitoring of enzymatic activity with the principle was schematically illustrated in Scheme 1. As seen from
MnO2–TMB system Fig. 3D, the absorbance increased with the increasing AOx activity,
indicating that the system could be employed for quantitative
To further monitoring whether the signal of the MnO2–TMB monitoring of AOx activity. Favorably, a low detection limit could
system could be really realized by AOx-controlling AA amount, be estimated to  12 mU mL  1 AOx at the 3sblank criterion. The
several control tests were carried out under different conditions results also suggested that the absorbance of using the MnO2–TMB
(Fig. 3C). As indicated from curve “c” in Fig. 2A, it has been de- system could be significantly changed only if the level of AOx in
monstrated that two characteristic absorption peaks and a strong the sample was lower than 14 mU mL  1, which also provide a vital
blue color could be observed in the presence of MnO2 nanoflakes precondition for the development of high-efficient MnO2–TMB-
and TMB. As seen from curve “a” and photograph “a” in Fig. 3C, based colorimetric immunoassay.

Please cite this article as: Lai, W., et al., Biosensors and Bioelectronics (2015), http://dx.doi.org/10.1016/j.bios.2015.12.035i
6 W. Lai et al. / Biosensors and Bioelectronics ∎ (∎∎∎∎) ∎∎∎–∎∎∎

3.4. Analytical performance of MnO2–TMB-based colorimetric results clearly revealed that the specificity of MnO2–TMB-based
immunoassay colorimetric immunoassay is acceptable.

To investigate the applicable potential of the developed MnO2–TMB 3.5. Monitoring of real peanut samples
visual system on the colorimetric immunoassay, AOx/anti-AFB1-labeled
gold nanoparticles (Ab-AuNP-AOx) was utilized for the detection of The feasibility of applying the developed colorimetric im-
target AFB1 on AFB1-BSA-conjugated magnetic beads (AFB1-MB) with a munoassay to evaluate AFB1 levels in a complex matrix was also
competitive-type immunoassay format (Scheme 1). In the presence of monitored by spiking AFB1 standards in the peanut samples. The
target AFB1, the analyte competed with the conjugated AFB1-BSA on extraction and spiking procedure were described in detail in our
magnetic beads for the labeled anti-AFB1 antibody on gold nano- previous work (Lin et al., 2015b). To verify the accuracy of MnO2–
particles. The conjugated amount of Ab-AuNP-AOx on the AFB1-MB TMB-based colorimetric immunoassay, the assayed results toward
decreased with the increasing target AFB1 in the sample, which was these peanut samples were compared with those obtained from
collected by using an external magnet. In this case, the carried AOx commercialized AFB1 ELISA kit (Diagnostic, Automation Inc.). The
accompanying gold nanoparticles decreased, which catalyzed relatively data are listed in Table 1. Statistical comparison between two
less AA molecules, thus decreasing the reaction efficiency of ascorbic methods was performed by using a t-test for comparison of means
acid toward the MnO2–TMB system. Under optimal conditions, the
preceded by the application of an F-test. The t-test statistics for
absorbance decreased with the increasing target AFB1 in the sample
each sample was calculated by using independent two-sample t-
(Fig. 4A). A good linear dependence between the absorbance and AFB1
test with equal sample sizes and equal variance. As seen from
concentration was achieved within the dynamic working range from
Table 1, no significant differences were encountered between the
0.05 to 150 ng mL  1 with a detection limit (LOD) of 6.5 pg mL  1 on
two methods at the 0.05 significance level because the texp were in
the basis of 3sblank criterion (Fig. 4A, inset). The linear regression
all cases below than tcrit (tcrit[4,0.05] ¼2.77). the regression equation
equation could be fitted to y¼  0.0065  C[AFB1] þ1.0561 (ng mL  1,
between two methods could be fitted to y¼ 1.0017x þ0.3017
R2 ¼0.9973, n¼21). Each data point represented the average value
(R2 ¼0.9978, n ¼30). The slope of the equation for the average
obtained from three different measurements. The maximum relative
values obtained between two methods was close to the ideal.
standard deviation (RSD) was 9.4%. Further, we studied the batch-to-
Thus, the MnO2–TMB-based colorimetric immunoassay could be
batch reproducibility of MnO2–TMB-based immunoassay toward three
utilized for quantitative monitoring of target AFB1 in the complex
level including 0.1, 20 and 100 ng mL  1 AFB1, and the RSDs were 6.1%,
systems.
9.2% and 7.9% (n¼3), respectively. These results suggested a good re-
producibility and precision of the colorimetric immunoassay. Inspir-
ingly, the LOD of our design was obviously even lower compared to the
existed commercialized AFB1 ELISA kits (e.g., Quicking Biotech: 100 ppt; 4. Conclusions
MaxSignals: 50 pg mL  1; MyBioSource: 250 pg mL  1; Diagnostic
Automation Inc.: 5.0 pg mL  1). Owing to the legal limit of AFB1 in In this work, we found a new colorimetric detection system on
foodstuff (o2.0 ng mL  1), the developed colorimetric immunoassay the basis of conventional TMB substrate by using MnO2 nanoflakes
coupling with the MnO2–TMB system can completely meet the need of as the enzymatic mimics. The system was carried out by enzyme-
AFB1 detection in foodstuff. catalyzed ascorbic acid to weaken the development of MnO2–TMB
To further monitor the specificity of MnO2–TMB-based colori- in the visual color. The color and density in the absorbance could
metric immunoassay toward target AFB1, the system was em- be controlled by conversion of MnO2 nanoflakes to Mn(II). Com-
ployed for challenging other mycotoxins [e.g., AFB2, AFG1, AFG2 pared with classical TMB-H2O2-based detection system, highlights
and ochratoxin A (OTA)] and marine toxins [e.g., okadaic acid of our strategy can be simply summarized as follows: (i) The
(OA)]. As shown from Fig. 4B, the MnO2–TMB-based colorimetric method can efficiently avoid the participation of H2O2 during this
immunoassay displayed a high cross-reactivity toward AFB2, measurement; (ii) natural enzyme (i.e., ascorbate oxidase) and
whereas no false compliant results were achieved for AFG1, AFG2 enzymatic mimic (i.e., MnO2 nanoflakes) were utilized for the
OTA and OA. The reason might be attributed to the fact that amplification of detectable signal coupling with enzyme cascade
anti-AFB1 antibody used in this work has a high cross-reactivity reaction-based amplification strategy; and (iii) Two-dimensional
with AFB2, thus resulting in the change in the absorbance. These MnO2 nanoflakes can exhibit higher surface area for the oxidation

Fig. 4. (A) UV–vis absorption spectra of MnO2–TMB-based colorimetric immunoassay toward AFB1 standards with different concentrations (ng mL  1) (Inset: The corre-
sponding linear curve), and (B) the specificity of the developed immunoassay against target AFB1, AFB2, AFG1, AFG2, OTA and OA (50 ng mL  1 used in this case).

Please cite this article as: Lai, W., et al., Biosensors and Bioelectronics (2015), http://dx.doi.org/10.1016/j.bios.2015.12.035i
 W. Lai et al. / Biosensors and Bioelectronics ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 7

Table 1 References
Results obtained by the MnO2–TMB-based colorimetric immunoassay and com-
mercial AFB1 ELISA Kit (Diagnostic Automation Inc., LOD ¼ 5.0 pg mL  1) for spiked
Arduini, F., Amine, A., Moscone, D., Palleschi, G., 2010. Microchim. Acta 170,
peanut samples.
193–214.
Asati, A., Santra, S., Kaittanis, C., Nath, S., Perez, J., 2009. Angew. Chem. Int. Ed. 48,
Sample no. Method; concentration (ng mL  1, mean 7 SD, n ¼3)a texp 2308–2312.
Berg, B., Cortazar, B., Tseng, D., Ozkan, H., Feng, S., Wei, Q., Chan, R., Burbano, J.,
a
MnO2–TMB-based immunoassay AFB1 ELISA kit Farooqui, Q., Lewinski, M., Carlo, D., Garner, O., Ozcan, A., 2015. ACS Nano 9,
7857–7866.
1 1.7 7 0.1 1.8 7 0.2 0.77 Bovee, T., Hendriksen, P., Portier, L., Wang, S., Elliott, C., Egmond, H., Nielen, M.,
2 13.2 7 0.9 14.17 0.6 1.44 Peijnenburg, A., Hoogenboom, L., 2011. Environ. Sci. Technol. 45, 8965–8973.
3 60.17 2.1 62.7 7 1.5 1.74 Chen, F., Hou, S., Li, Q., Fan, H., Fan, R., Xu, Z., Zhala, G., Mai, X., Chen, X., Chen, X., Liu,
4 93.7 7 2.1 90.17 1.9 2.20 Y., 2014. Anal. Chem. 86, 10021–10024.
5 124.27 1.9 127.5 7 2.3 1.92 Elzupir, A., Alamer, A., 2014. Toxin Rev. 33, 202–205.
6 74.5 7 1.5 73.2 7 0.9 1.29 Gao, Z., Hou, L., Xu, M., Tang, D., 2014. Sci. Rep. 4, 3966.
7 0.89 7 0.1 0.93 7 0.09 0.51 Gao, Z., Xu, M., Lu, M., Chen, G., Tang, D., 2015. Biosens. Bioelectron. 70, 194–201.
Gawley, R., Pinet, S., Cardona, C., Datta, P., Ren, T., Guida, W., Nydick, J., Leblanc, R.,
8 23.5 7 1.1 22.3 7 0.8 1.53
2002. J. Am. Chem. Soc. 124, 13448–13453.
9 51.2 7 1.8 52.7 7 1.2 1.20
Geng, T., Liu, K., Frazier, R., Shi, L., Bell, E., Glenn, K., Ward, J., 2015. J. Agric. Food
10 5.4 7 0.9 7.17 0.4 2.40 Chem. 63, 4947–4953.
Khan, Z., Kumar, P., Din, K., 2005. J. Colloid Interf. Sci. 290, 184–189.
a
Each sample was determined in triplicate, and the high-concentration AFB1 Kumari, S., Dhar, B., Panda, C., Meena, A., Gupta, S., 2014. ACS Appl. Mater. Interfaces
sample was assayed with an appropriate dilution. The linear regression equation 6, 13866–13873.
for these usable mean data between two methods is as follows: y ¼1.0017x þ0.3071 Lai, W., Zhuang, J., Tang, D., 2015. J. Agric. Food Chem. 63, 1982–1989.
(R2 ¼ 0.9978, n¼ 30); x axis: by MnO2–TMB-based colorimetric immunoassay; y Liang, M., Fan, K., Pan, Y., Jiang, H., Wang, F., Yang, D., Lu, D., Feng, J., Zhao, J., Yang, L.,
axis: by AFB1 ELISA kit. Yan, X., 2013. Anal. Chem. 85, 308–312.
Liu, X., Wang, Q., Zhao, H., Zhang, L., Su, Y., Lv, Y., 2012. Analyst 137, 4552–4558.
Lin, Y., Zhou, Q., Lin, Y., Tang, D., Chen, G., Tang, D., 2015a. Biosens. Bioelectron. 74,
of TMB. Importantly, the MnO2–TMB system could be utilized for 680–686.
the development of colorimetric immunoassay with high sensi- Lin, Y., Zhou, Q., Lin, Y., Tang, D., Niessner, R., Knopp, D., 2015b. Anal. Chem. 87,
8531–8540.
tivity, good reproducibility and acceptable method accuracy with Luo, X., Xu, J., Zhao, W., Chen, H., 2004. Anal. Chim. Acta 512, 57–61.
commercialized ELISA kit. Nevertheless, only one disadvantage of Nathanail, A., Varga, E., Meng-Reiterer, J., Bueschl, C., Michlmayr, H., Malachova, A.,
the developed strategy involved in the multi-step operations, Fruhmann, P., Jestoi, M., Peltonen, K., Adam, G., Lemments, M., Schuhmacher, R.,
Berthiller, F., 2015. J. Agric. Food Chem. 63, 7862–7872.
which might make the immunoassay complicated and time-con- Qin, W., Su, L., Yang, C., Ma, Y., Zhang, H., Chen, X., 2014. J. Agric. Food Chem. 62,
suming though MnO2 nanoflakes could simplify the colorimetric 5827–5834.
Qu, Z., Xu, H., Xu, P., Chen, K., Mu, R., Fu, J., Gu, H., 2014. Anal. Chem. 86, 9367–9371.
reaction system (i.e., the MnO2–TMB system). To fulfill the po- Ray, C., Dutta, S., Sarkar, S., Sahoo, R., Roy, A., Pal, T., 2014. J. Mater. Chem. B 2,
tential application for point-of-care testing, future work should 6097–6105.
Samdal, I., Lovberg, K., Briggs, L., Kilcoyne, J., Xu, J., Forsyth, C., Miles, C., 2015. J.
focus on the improvement of operation steps. Agric. Food Chem. 63, 7855–7861.
Soh, J., Lin, Y., Rana, S., Ying, J., Stevens, M., 2015. Anal. Chem. 87, 7644–7652.
Su, L., Feng, J., Zhou, X., Ren, C., Li, H., Chen, X., 2012. Anal. Chem. 84, 5753–5758.
Wan, Y., Qi, P., Zhang, D., Wu, J., Wang, Y., 2012. Biosens. Bioelectron. 33, 69–74.
Wang, G., Xu, X., Wu, X., Cao, G., Dong, Y., Li, Z., 2014. J. Phys. C.hem. C. 118,
Acknowledgments 28109–28117.
Wang, J., Mi, X., Guan, H., Wang, X., Wu, Y., 2011. Chem. Commun. 47, 2940–2942.
Wang, Q., Song, Y., Xie, H., Chai, Y., Yuan, Y., Yuan, R., 2015. Chem. Commun. 51,
This work was financially supported by the National Natural
1255–1258.
Science Foundation of China (Grant nos. 41176079 and 21475025), Wei, C., Yu, L., Cui, C., Lin, J., Wei, Chen, Mathews, N., Huo, F., Sritharan, T., Xu, Z.,
the National Science Foundation of Fujian Province (Grant no. 2014. Chem. Commun. 50, 7885–7888.
Wu, G., He, S., Peng, H., Deng, H., Liu, A., Lin, X., Xia, X., Chen, W., 2014. Anal. Chem.
2014J07001), Key Science Project (type A) of the Fujian Provincial 86, 10955–10960.
Department of Education, China (Grant no. JA12021) and the Zhai, W., Wang, C., Yu, P., Wang, Y., Mao, L., 2014. Anal. Chem. 86, 12206–12213.
Zhang, T., Lu, Y., Luo, G., 2014a. ACS Appl. Mater. Interfaces 6, 14433–14438.
Program for Changjiang Scholars and Innovative Research Team in Zhang, X., Zheng, C., Guo, S., Li, J., Yang, H., Chen, G., 2014b. Anal. Chem. 86,
University (Grant no. IRT1116). 3426–3434.

Please cite this article as: Lai, W., et al., Biosensors and Bioelectronics (2015), http://dx.doi.org/10.1016/j.bios.2015.12.035i