A squaric acid-stimulated electrocatalytic reaction for sensing biomolecules with cycling signal amplification

Article abstract of DOI:10.1039/c3cc41708e

Squaric acid, a 2-dimensional planar structure of squarate C ₄O₄ units linked by protons in a layered sheet, was utilized for the first time as a catalytic substrate for ultrasensitive electronic determination of low-abundance proteins by coupling a target-induced electrocatalytic reaction with the in situ cycling signal amplification strategy.

Full text of DOI:10.1039/c3cc41708e

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
A squaric acid-stimulated electrocatalytic reaction
for sensing biomolecules with cycling signal
amplification†
Cite this: Chem. Commun., 2013,
49, 4761
Received 6th March 2013,
Accepted 3rd April 2013
Wenqiang Lai, Dianping Tang,* Libing Fu, Xiaohua Que, Junyang Zhuang and
Guonan Chen*
DOI: 10.1039/c3cc41708e
www.rsc.org/chemcomm
Squaric acid, a 2-dimensional planar structure of squarate C4O4 units of SQA or the unaffordable conjugation with biomolecules. Signifi-
linked by protons in a layered sheet, was utilized for the first time as a cantly, Das et al. found that C–OH in the p-aminophenol could be
catalytic substrate for ultrasensitive electronic determination of low- initially electrochemically oxidized to CQO through the ferrocene
abundance proteins by coupling a target-induced electrocatalytic mediator, and then reduced back to C–OH by NaBH4.7 The catalytic
reaction with the in situ cycling signal amplification strategy.
recycling process between C–OH and CQO caused the amplification of
the electrochemical signal. Unfortunately, the amount of the hydroxyl
Squaric acid (SQA), one of the most studied hydrogen-bonded anti- groups in the p-aminophenol molecule was limited. To improve this,
ferroelectrics in the KH2PO4 family, is a diacid that exhibits two acidic Akanda et al. constructed another redox cycling protocol by using
hydroxyl groups with pKa values of 0.54 and 3.48 as well as two highly L-ascorbic acid with two available hydroxyl groups.8 The signal was
polarized carbonyl groups.1 The high acidity with high pKa = 0.54 for amplified through two –OH groups based on the redox cycling between
the first proton and pKa = 3.48 for the second is attributable to L-ascorbic acid and dehydroascorbic acid. In this regard, our motive was
resonance stabilization of the anion. At room temperature, SQA has a to utilize SQA with two C–OH and two CQO groups as the catalytic
2-dimensional planar structure of squarate C2O2 units linked by substrate for the amplification of electrocatalytic signal through the
protons in a layered sheet. Upon the dielectric phase transition, the cycling between squaric acid, cyclobutaneoctol and cyclobutanetetrone.
crystal structure changes from the monoclinic (AFE phase; P21/m,
Herein, we report a novel and enzyme-free electrochemical
Z = 2) to the tetragonal form (PE phase, I4/m, Z = 1).2 More inspiringly, immunoassay for sensing a low-abundance protein (alpha-fetoprotein,
the four-membered ring of SQA remains intact with oxidizing agents AFP, used as a model) by coupling with the SQA-stimulated redox
under mild conditions, and produces cyclobutaneoctol as the product, recycling strategy. The assay was carried out in pH 7.4 PBS containing
the tetrahydrate of cyclobutanetetrone.3 Upon reduction with reducing 1.0 mM squaric acid and 1.0 mM tris(2-carboxyethyl)phosphine (TCEP,
agents, the produced cyclobutaneoctol is converted into SQA.
as a reducing agent, which can reduce electroactive organic materials
The unique structure provides not only versatile proton acceptor such as dehydroascorbic acid and quinone at a fast rate) by using
sites at the carbonyl function for hydrogen bonding but also binding ferrocenecarboxylic acid-labelled anti-AFP polyclonal antibody (FC-pAb2)
sites to metal ions.4 With univalent metals, e.g. potassium, squaric acid as trace tags with a sandwich-type immunoassay mode (Scheme 1).
forms salts which are readily soluble in water, whereas the squarates of Mouse monoclonal anti-AFP capture antibody (mAb1) was directly
bivalent and tervalent metals are insoluble.5 Another, quantum immobilized on the surface of glassy carbon electrode (GCE) via the
mechanical, way of describing the dianion is to assume that p electrons electrochemically deposited nanogold–graphene hybrid nanostructures
of the two double-bonded oxygen atoms are shifted to the latter, so that (AuNP–GP). In the presence of the target analyte, the immobilized
all four oxygens become single-bonded –OÀ groups and a double mAb1 and Fc-pAb2 sandwich the target AFP. During the electrochemical
positive electric charge is left in the ring of carbon atoms.
measurement, the added squaric acid (SQA) was reduced to cyclo-
Nowadays, squaric acid is mainly applied for the synthesis of its butaneoctol (CBO) by TCEP, then the generated CBO was oxidized
derivatives and in immunotherapy.6 To the best of our knowledge, step by step to cyclobutanetetrone (CBT) by the labelled ferrocene on
however, there is no report focusing on squaric acid in electroanalytical the FC-pAb2, and then the resulting CBT was reduced again back to
chemistry. The reason might be attributed to the weak redox properties CBO by TCEP. The self-produced CBO reactant was catalytically recycled
between the CBO and CBT, thereby resulting in the amplification of the
Ministry of Education & Fujian Provincial Key Laboratory of Analysis and Detection
electrochemical signal (please see experimental details in the ESI†).
Under the fixed SQA and TCEP, the signal directly depends on the
labelled ferrocene relative to the target AFP level.
of Food Safety, Department of Chemistry, Fuzhou University, Fuzhou 350108,
P.R. China. E-mail: dianping.tang@fzu.edu.cn; Fax: +86 591 22866135;
Tel: +86 591 22866125
In this work, mAb1 was immobilized onto the sensor through the
strong interaction between AuNP–GP and mAb1. To verify the
† Electronic supplementary information (ESI) available: Experimental procedures
and characterization. See DOI: 10.1039/c3cc41708e
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 4761--4763 4761

View Article Online
Communication
ChemComm
Fig. 2 (A) DPV responses of (a) the newly prepared mAb1/AuNP–GP/GCE in
pH 7.4 PBS, (b) electrode ‘a’ after 0.1 ng mLÀ1 AFP and Fc-pAb2 in pH 7.4 PBS,
(c) electrode ‘b’ in pH 7.4 PBS + 1.0 mM SQA, (d) electrode ‘b’ in pH 7.4 PBS + 1.0 mM
CBO, and (e) electrode ‘b’ in pH 7.4 PBS + 1.0 mM SQA + 1.0 mM TCEP. (B) DPV
responses of (a) the mAb1/AuNP–GP/GCE, and (b) electrode ‘a’ after 0.1 ng mLÀ1 AFP
and Fc-pAb2 in pH 7.4 PBS containing 1.0 mM SQA and 1.0 mM TCEP.
Scheme 1 Schematic illustration of the SQA-stimulated electrocatalytic reaction for
sensing low-abundance protein based on the cycling signal amplification strategy.
formation of AuNP–GP on the sensor by the electrochemically
deposited method, SEM images of variously modified electrodes were
investigated. As seen from Fig. 1A and B, nanogold particles and
graphene nanosheets could be simultaneously deposited onto the
sensor with homogeneous dispersion. Logically, an important ques-
tion to be answered was whether the electrodeposited graphene
nanosheets were the reduced graphene or graphene oxide owing to
the weak conductivity of graphene oxide. As seen from curve ‘a’ in the
inset of Fig. 1B, a strong D peak and a relatively weak G peak could be
observed from the Raman spectra of the newly prepared graphene
oxide by the Hummers method.9 In contrast, a weak D peak and a
relatively strong G peak appeared at the electrodeposited graphene
nanosheets (curve ‘b’), indicating the presence of the reduced
graphene.10 The successful preparation of AuNP–GP could provide
a precondition for the development of the immunoassay.
To realize our design, the newly prepared immunosensor was used
for the detection of 0.1 ng mLÀ1 AFP (used as an example), and the
characteristics were investigated in pH 7.4 PBS by using differential
pulse voltammetry (DPV) in the different supporting electrolytes after
each step (Fig. 2A). No peak was observed at the immunosensor (curve
‘a’). When the immunosensor was incubated with the target AFP and
Fc-pAb2, however, an obvious peak current was achieved (curve ‘b’),
which was mainly derived from the labelled ferrocene on the Fc-pAb2.
Upon the addition of 1.0 mM SQA in PBS, the current increased (curve
‘c’ in Fig. 2A, Di = 1.7 mA vs. curve ‘b’). The reason might be due to the
fact that the added SQA was oxidized to CBT by the labelled ferrocene.
For comparison, CBO was directly used in pH 7.4 PBS. As seen from
curve ‘d’ in Fig. 2A, the peak current was obviously higher than that of
SQA (curve ‘c’), which might be attributed to the fact that the added
CBO was oxidized to SQA/CBT by the conjugated ferrocene. More
inspiringly, the peak current further increased when 1.0 mM TCEP
was injected into the PBS–SQA system (curve ‘e’ in Fig. 2A). The added
SQA and TCEP could cause 412 Æ 11.7% signal increases in the
current (vs. curve ‘b’), which was higher than those of SQA and CBO
alone. As the control test, the as-prepared immunosensor was
monitored before and after incubation with 0.1 ng mLÀ1 AFP and
Fc-pAb2 in pH 7.4 PBS containing 1.0 mM SQA and 1.0 mM TCEP
(Fig. 1B). As indicated from curve ‘a’, almost no peak current was
acquired at the immunosensor. In contrast, a strong peak current was
obtained after the formation of the sandwiched immunocomplex on
the immunosensor (curve ‘b’). These results indicated that (i) the non-
specific adsorption of the immunosensor was negligible, (ii) the peak
current mainly derived from the interaction of the labelled ferrocene
with the SQA/TCEP, and (iii) the detectable signal could be further
amplified by TCEP. Hence, SQA could be preliminarily utilized for the
detection of target AFP with the help of TCEP by the designed routine.
To further clarify the in situ amplified efficiency by using SQA in the
developed immunoassay, L-ascorbic acid, as a similar catalytic substrate
to SQA, was also employed as a substitute of the SQA for the deter-
mination of AFP standards with various concentrations on immuno-
sensors of the same batch. The obtained results were compared with
those of the SQA-based immunoassay system. As seen from Fig. 3, the
electrochemical immunoassay by using SQA as a catalytic substrate
could exhibit higher current responses than that using L-ascorbic acid.
Meanwhile, we also found that use of SQA could enhance the sensitivity
(i.e. the slope) of the electrochemical immunoassay. The reason might
most likely be a consequence of the fact that L-ascorbic acid has only
two available –OH groups for the development of the catalytic recycling,
whereas SQA has four available groups. Within the same scanning
time, the SQA could transfer more electrons than L-ascorbic acid, thus
resulting in the strong electrochemical signal and high sensitivity.
Using Fc-pAb2 as trace tags, the sensitivity and dynamic range of the
electrochemical immunoassay were investigated toward AFP standards
in pH 7.4 PBS containing 1.0 mM SQA and 1.0 mM TCEP. As seen from
the inset in Fig. 4a, DPV peak currents increased with the increasing
AFP concentrations. A linear dependence between the peak currents
and the logarithm of AFP was obtained in the range from 1.0 pg mLÀ1
to 200 ng mLÀ1 with a detection limit (LOD) of 0.6 pg mLÀ1 estimated
at a signal-to-noise ratio of 3 (n = 18) (Fig. 4a). The LOD was partially
lower than those of other AFP detection method strategies (Table S1,
ESI†). Since the threshold of AFP in normal humans is about
10 ng mLÀ1, the electrochemical immunoassay can completely
Fig. 1 SEM images of (A) the electrodeposited graphene nanosheets and (B)
AuNP–GP nanostructures (inset: typical Raman spectra of (a) graphene oxide and
(b) the electrodeposited graphene nanosheets).
c
4762 Chem. Commun., 2013, 49, 4761--4763
This journal is The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
respectively. Hence, both intra-assay and inter-assay verified
acceptable reproducibility. In addition, the immunosensor
displayed satisfactory stability. As much as 90% of the initial
current was preserved after storage of the immunosensor and
Fc-pAb2 at 4 1C for 30 days.
Finally, the accuracy of the electrochemical immunoassay
method was evaluated for testing 10 clinical serum specimens,
which were collected from Fujian Provincial Hospital of China
according to the rules of the local ethical committee. The obtained
results were compared with those of the commercialized electro-
chemiluminescent immunoassay (ECLIA) method. As shown in
Table S2 (ESI†), no significant differences were encountered between
the two methods at the 0.05 significance level because all the texp
values in the case were less than tcrit (tcrit[4,0.05] = 2.77). Hence, the
SQA-stimulated electrocatalytic reaction could be regarded as an
optional scheme for detecting AFP in real samples.
Fig. 3 Comparison of electrochemical responses of the as-prepared immuno-
sensor toward AFP standards with various concentrations using different catalytic
substrates in pH 7.4 PBS containing (a) 1.0 mM SQA and 1.0 mM TCEP, (b) 1.0 mM
L-ascorbic acid and 1.0 mM TCEP.
In summary, this work demonstrates a novel and enzyme-free
amplified immunoassay strategy for sensitive electronic detection of
low-abundance protein, using an SQA-stimulated electrocatalytic
reaction and the cycling signal amplification protocol. Compared
with conventional amplified strategies based on enzyme labels or
nanolabels, the method is simple, low-cost and highly efficient.
Further, SQA with a unique crystal structure can be utilized for the
in situ amplification of the electrochemical signal. Although the
present assay system focuses on the determination of target AFP, it
can be easily extended for sensing other biomolecules.
Fig. 4 (a) Calibration plots of the electrochemical immunoassay toward AFP
standards in pH 7.4 PBS containing 1.0 mM SQA and 1.0 mM TCEP (inset: the
corresponding DPV curves). (b) The specificity of the electrochemical immuno-
assay (0.1 ng mLÀ1 AFP and 100 ng mLÀ1 interfering agents used in this case).
Support from the ‘‘973’’ National Basic Research Program of
China (2010CB732403), NSFC of China (21075019, 41176079), the
Research Fund for the Doctoral Program of Higher Education of
China (20103514120003), the China–Russia Bilateral Scientific
meet the requirement of clinical diagnostics. Although the system Cooperation Research Program (21211120157), and the National
has not yet been optimized for maximum efficiency, importantly, Science Foundation of Fujian Province (2011J06003) is gratefully
the sensitivity of using SQA as the enhancer was over 1000-fold acknowledged.
lower than that of commercially available AFP ELISA kits
(1.0 ng mLÀ1, Genway Biotech. Inc.).11
Notes and references
To evaluate the specificity of the developed immunoassay for AFP
detection, we challenged the system with other interfering sub-
stances, e.g. carcinoma embryonic antigen (CEA), thyroid-stimulating
hormone (TSH), luteinizing hormone (LH), rabbit IgG (RIgG), uric
acid, ascorbic acid, acetaminophen, cholesterol, creatinine, para-
cetamol, bilirubin, ibuprofen, dopamine, salicylate, tolazamide and
tolbutamide. The assay was carried out with the same experimental
procedures. As indicated in Fig. 4b and Fig. S4 (ESI†), a significantly
higher current was observed with the target AFP than with other
proteins. These results indicated that the components coexisting in
the sample matrix did not interfere in the determination of AFP, i.e.,
the electrochemical immunoassay was shown to be sufficiently
selective for detection of AFP.
Next, the precision of the developed immunoassay was monitored
by assaying 0.01, 1.0 and 100 ng mLÀ1 AFP as examples, using
identical batches of immunosensor and Fc-pAb2. Experimental results
indicated that the coefficients of variation (CVs, n = 3) of the intra-
assay were 7.8, 5.8 and 6.9% for 0.01, 1.0 and 100 ng mLÀ1 AFP,
respectively, whereas the CVs of the inter-assay with various batches
1 (a) D. Basak, C. Versek, D. Toscano, S. Christensen, M. Tuominen
and D. Venkataraman, Chem. Commun., 2012, 48, 5922; (b) L. Swartz
and L. Howard, J. Phys. Chem., 1971, 75, 1798.
2 (a) D. Semmingse, F. Hollander and T. Koetzle, J. Chem. Phys., 1977,
66, 4405; (b) Z. Dega-Szafran, A. Katrusiak and M. Szafran, J. Mol.
Struct., 2012, 1030, 184.
3 (a) A. Roedig and G. Markl, Justus Liebigs Ann. Chem., 1960, 636, 1;
(b) Y. Huang, Y. Lai and S. Wang, Chem.–Eur. J., 2012, 18, 8614.
4 (a) H. Terao, T. Sugawara, Y. Kita, E. Kaho and S. Takeda, J. Am.
Chem. Soc., 2001, 123, 10468; (b) R. West and H. Niu, J. Am. Chem.
Soc., 1963, 85, 2589.
5 (a) G. Maahs and P. Hegenberg, Angew. Chem., Int. Ed. Engl., 1966,
5, 888; (b) K. Lee, J. Kweon, I. Oh and C. Lee, J. Phys. Chem. Solids,
2012, 73, 890.
6 (a) D. Basak, C. Versek, D. Toscano, S. Christensen, M. Tuominen and
D. Venkataraman, Chem. Commun., 2012, 48, 5922; (b) C. Dingels,
F. Wurm, M. Wagner, H. Klok and H. Frey, Chem.–Eur. J., 2012,
18, 16828.
7 J. Das, M. Aziz and H. Yang, J. Am. Chem. Soc., 2006, 128, 16022.
8 M. Akanda, M. Aziz, K. Jo, V. Tamilavan, M. Hyun, S. Kim and
H. Yang, Anal. Chem., 2011, 83, 3926.
9 W. Hummers and R. Offeman, J. Am. Chem. Soc., 1958, 80, 1339.
10 Y. Guo, B. Wu, H. Liu, Y. Ma, Y. Yang, J. Zheng, G. Yu and Y. Liu,
Adv. Mater., 2011, 23, 4626.
were 8.9, 6.8 and 8.3% toward the above-mentioned concentrations, 11 www.genwaybio.com/media/custom/upload/File-1313516446.pdf.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 4761--4763 4763