Sensors and Actuators B 260 (2018) 396–399

Contents lists available at ScienceDirect

Sensors and Actuators B: Chemical

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

Direct detection for concentration ratio of HbA1c to total hemoglobin

by using potentiometric immunosensor with simple process of
denaturing HbA1c
Junko Tanaka a,∗ , Yu Ishige a , Riko Iwata b , Baku Maekawa c , Hidehiro Nakamura b ,
Takeshi Sawazaki d , Masao Kamahori a
a
Research & Development Group, Hitachi, Ltd., 1-280, Higashi-koigakubo, Kokubunji-shi, Tokyo, 185-8601 Japan
b
Research & Innovation Promotion Headquarters, Hitachi Chemical, Ltd., 1-9-2, Marunouchi, Chiyoda-ku, Tokyo, 100-6606 Japan
c
Risk Management Center, Hitachi Chemical, Ltd., 1-9-2, Marunouchi, Chiyoda-ku, Tokyo, 100-6606 Japan
d
Life Science Business Headquarters, Hitachi Chemical, Ltd., 1-9-2, Marunouchi, Chiyoda-ku, Tokyo, 100-6606 Japan

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

Article history: A direct measurement method for HbA1c (%), which is the concentration ratio of HbA1c to total
Received 13 August 2017 hemoglobin (Hb), was investigate with a potentiometric immunosensor. The direct measurement method
Received in revised form is based on sandwich immunoassay that combines anti-Hb and enzyme-labeled anti-HbA1c antibodies.
21 November 2017
While the anti-Hb antibodies capture both Hb and HbA1c maintaining HbA1c (%) in blood sample, the
Accepted 22 December 2017
enzyme-labeled anti-HbA1c antibodies bind only to HbA1c in the total captured Hb. Though we tried the
Available online 24 December 2017
direct detection of HbA1c (%) with a combination of four anti-Hb antibodies and two anti-HbA1c antibod-
ies, the obtained result showed lower sensitivity and reproducibility. We presumed that the binding site
Keywords:
HbA1c
of HbA1c for anti-HbA1c antibody hides inside the folding structure of HbA1c and anti-HbA1c antibody
Direct detection is not able to bind to HbA1c efficiently resulting in lower sensitivity. As a result of examination of both
Potentiometric immunosensor the type and concentration of the surfactant, it was found that the process of denaturing HbA1c with 0.2%
Sandwich immunoassay dodecyltrimethylammonium bromide was added to enable it to be accessible to anti-HbA1c antibodies
Denaturing process and could enhance the sensitivity. By adding the simple process of denaturing HbA1c, the direct mea-
surement of HbA1c (%) was successfully performed with this immunosensor. The results showed a good
correlation (correlation efficiency, r2 = 0.975) between the certified and experimental values of HbA1c
(%) in the clinically relevant range (from 5.6% to 10.6%).
© 2017 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).

1. Introduction
HbA1c, which is hemoglobin (Hb) glycated by glucose in blood,
reflects the average blood glucose level of 1–2 months, so it is
considered the gold standard for determining the treatment of dia-
betes. To prevent worsening of diabetes, monitoring of HbA1c levels
for patients in the diabetes mellitus preliminary group is important.
Point-of-care devices for HbA1c will help on-site disease manage-
ment not only at large hospitals that have clinical laboratories but
also at small hospitals. HbA1c (%), which is the concentration ratio
of HbA1c to total Hb, is determined by a high-performance liq-
uid chromatography (HPLC) using ion exchange or affinity columns
to separate HbA1c from other Hb [1,2], and has been widely used
∗ Corresponding author.
E-mail address: junko.tanaka.gw@hitachi.com (J. Tanaka).
as the HbA1c indicator. The HLPC used as a standard method can
determine HbA1c (%) directly, but it requires large and expensive
instruments.
To date, antibody-based immunoassay and enzyme-based assay
methods that do not require large instruments have been increas-
ingly developed. An electrochemical detection of HbA1c using flow
immunoassay system was reported by Tanaka et al. [3]. In this
method, HbA1c reacted with anti-HbA1c antibodies modified with
ferrocene monocarboxylic acid, was separated by boronate-affinity
chromatography, and then electrochemically detected. An enzyme-
based method for detecting HbA1c using fructosyl peptide oxidase
was reported by Hirokawa et al. [4]. These methods do not require
large instruments and are also inexpensive. However, HbA1c (%)
cannot be obtained using only these methods. To obtain HbA1c (%),
a separate measurement of total Hb in blood and a calculation of
the ratio of HbA1c to total Hb is required.

https://doi.org/10.1016/j.snb.2017.12.148
0925-4005/© 2017 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
 J. Tanaka et al. / Sensors and Actuators B 260 (2018) 396–399
Fig. 1. Procedure for direct detection of ratio of HbA1c to total hemoglobin.
An immunoassay using the direct adsorption method without
any capture molecule was developed for directly measuring HbA1c
(%) by Engbaeck et al. [5]. The direct adsorption method is suscep-
tible to other protein in blood and can causes lower reproducibility
and sensitivity. To solve this problem, an immunosensor using
membrane-immobilized haptoglobin as an affinity matrix to cap-
ture Hb and HbA1c was developed to bind Hb selectively from
blood samples [6]. The haptoglobin-based sandwich immunoas-
say demonstrated a better reproducibility and higher sensitivity
than the anti-Hb antibody-based assay. The main reason is that the
complex formation of HbA1c with haptoglobin will cause a partial
unfolding of the HbA1c molecule. This strongly suggests that it is
necessary to denature HbA1c to enhance the interaction between
anti-HbA1c antibodies and HbA1c.
In this study, we focused on a simple sample pretreatment of
a blood sample to enhance the sensitivity of HbA1c (%) detection
using a conventional sandwich immunoassay with both anti-Hb
and anti-HbA1c antibodies. Direct detection of HbA1c (%) in the
clinically relevant range was performed by this immunoassay
method with a disposable potentiometric immunosensor.
2. Materials and methods
2.1. Chemicals and reagents
Screen-printing carbon electrodes on a polyethylene tereph-
thalate (PET) film was made by Advanced U-corporation (Gunma,
Japan). Pasting graphite powder was purchased from BAS (Tokyo,
Japan). Lyophilized streptavidin powder was purchased from
Sigma-Aldrich (MO, USA). Monoclonal anti-Hb capture antibod-
ies (Clone M1709Hg1, M1709Hg2, M1202099, M1202100) were
purchased from Fitzgerald (MA, USA), and monoclonal anti-
HbA1c detection antibodies (Clone B842 M, 75C9) were purchased
from Abcam (Cambridge, UK). Biotin labeling kit and alkaline
phosphatase labeling kit were purchased from Dojindo Molecu-
lar Technologies (Kumamoto, Japan). Certified reference material
(JCCRM423) of HbA1c was obtained from the Health Care Technol-
ogy Foundation (Kanagawa, Japan). Dodecyltrimethylammonium
bromide (DTAB) was obtained from Wako Pure Chemical Indus-
tries (Osaka, Japan). Milk protein solution was purchased from
Kirkegaard & Perry Laboratories (MD, USA).
397
2.2. Fabrication of disposable immunosensor
The disposable cartridge potentiometric immunosensor was
made of a PET film with screen-printed carbon electrodes and
PET-EVA films. Antibodies capturing target molecules were immo-
bilized on the surface of the flow channel as follows. The flow
channel was incubated with 1 mg/ml streptavidin solution in car-
bonate buffer at room temperature for 1 h, and washed with
1 ml Tris-buffered Saline with Tween 20 (TBST). Twenty ␮g/ml of
biotinylated anti-Hb capture antibodies were added to the flow
channel and incubated at room temperature for 1 h. The num-
ber of molecules of the added biotin-labeled anti-Hb antibody is
sufficiently smaller than the number of molecules of streptavidin
adsorbed on the flow path. Since the affinity of streptavidin and
biotin is high (KD = 10−15 mol/l), all of the added biotin-labeled
anti-Hb antibody is captured on the surface of the flow path. From
the above, it is possible to control the number of molecules of the
anti-Hb antibody immobilized on the channel surface by control-
ling the concentration of the biotin-labeled anti-Hb antibody to be
added. The flow channel was washed with 1 ml TBST again. The
flow channel was then blocked with milk protein solution at room
temperature for 1 h, and washed with 1 ml TBST.
2.3. Immunoassay (Measurement procedure)
The proposed HbA1c (%) detection method is based on a sand-
wich immunoassay combining anti-Hb and anti-HbA1c antibodies
(Fig. 1). First, the certified reference material (JCCRM 423) used as
a sample is diluted 50 times with 0.2% DTAB. The pretreated solu-
tion is added to capture antibodies immobilized on the surface. The
sample contains an excess number of Hb and HbA1c against the
number of anti-Hb antibodies. The anti-Hb antibodies can capture
a certain portion of Hb maintaining HbA1c (%) in the sample. Next,
20 ␮g/ml enzyme-labeled anti-HbA1c antibodies are added. Finally,
enzyme activity is measured by potentiometric measurement after
adding 2.5 mM ascorbic acid phosphate (AsA-P) into a borate buffer
(100 mM H3 BO3 , 200 mM NaCl, 1 mM MgCl2 , 4.95 mM ferricyanide,
0.05 mM ferrocyanide, pH9.5). The potentiometric measurement
system was previously described [7,8]. Calibration of the experi-
mental value (HbA1c (%)) is performed using the output potential
data in volts of one sample with certified HbA1c (%) of 5.61%.
398 J. Tanaka et al. / Sensors and Actuators B 260 (2018) 396–399
Fig. 2. Sandwich immunoassay of HbA1c (%) with microtiter plate. (A) Calibration curve of the sandwich immunoassay for HbA1c (%) without pretreatment. (B) Calibration
curve of the sandwich immunoassay for HbA1c (%) with pretreatment using 0.2% DTAB. (C) The relationship between DTAB concentration and the efficiency of the sandwich
immunoassay.
Fig. 3. Schematic process of denaturing HbA1c with a denaturing agent.
3. Results and discussion
3.1. Principle of direct detection of HbA1c (%) by sandwich
immunoassay
Proposed direct detection of HbA1c (%) was carried out by the
sandwich immunoassay with a conventional microtiter plate. The
obtained results showed that lower sensitivity and the absorbance
were correlated poorly with HbA1c (%) (Fig. 2A). Although we
tried the direct detection of HbA1c (%) with a combination of four
anti-Hb antibodies and two anti-HbA1c antibodies, the results did
not change for any combination of antibodies. In the structure of
Hb, the binding site for anti-HbA1c antibodies, which is glycated
N-terminal valine residue of the ␤-chain of HbA1c, is almost com-
pletely buried in the protein structure (PDB ID, 1BBB). Therefore,
anti-HbA1c antibodies are not able to bind to the binding site of
HbA1c efficiently. In a previous report [6], the sandwich immunoas-
say with haptoglobin as a capture molecule for hemoglobin and
HbA1c demonstrated a better reproducibility and higher sensitiv-
ity than the anti-Hb antibody-based assay. The complex formations
of HbA1c with haptoglobin lead to a partial unfolding of the HbA1c
molecule, which results in a strong binding ability. This strongly
suggests that denaturing HbA1c causes to increase the detection
sensitivity of HbA1c. As a result, the process of denaturing HbA1c
with a denaturing agent was added to enable it to expose gly-
cated residue of HbA1c and be accessible to anti-HbA1c antibodies
(Fig. 3). The pretreatment required only the adding of the dena-
turing agent to the sample solution. We selected three kinds of
surfactants, guanidine hydrochloride, sucrose monolaurate and
DTAB, as a denaturing agent. These surfactants are commonly
used in biochemical fields. As a result of sample pretreatment
using three kinds of surfactants, the pretreatment using DTAB was
the most effective for sensitivity. When performing immunoas-
say using pretreated samples, the combination of anti-Hb antibody
(Clone M1709Hg2) and anti-HbA1c antibody (Clone B842 M) was
most suitable for measurement. Then, we optimized the concen-
Fig. 4. Experimental results of direct detection of HbA1c (%) using disposable poten-
tiometric immunosensor.
tration of DTAB. Assuming that the slope of the plot of the certified
value of HbA1c (%) versus the absorbance of product solution rep-
resents the reaction efficiency of the immunoassay, the highest
reaction efficiency was obtained when the concentration of DTAB
was around 0.2% (Fig. 2C). It is estimated that reaction efficiency
decreased in a region where the concentration of DTAB was higher
than 0.2% due to denaturation of the antibody and breakage of the
quaternary structure of Hb and HbA1c. After the pretreatment with
0.2% DTAB, the experimental value of sandwich immunoassay for
HbA1c was correlated highly with a certified value of HbA1c (%),
and the reaction efficiency of the immunoassay improved about
60 times (Fig. 2B). Thus, by adding the simple pretreatment of the
sandwich immunoassay with anti-Hb and anti-HbA1c antibodies,
the direct measurement of HbA1c (%) was successfully performed.
3.2. Direct detection of HbA1c (%) with disposable cartridge
potentiometric immunosensor
Direct detection of HbA1c (%) was performed by using the
disposable potentiometric immunosensor. Anti-Hb antibodies are
immobilized on the surface of the channel in the immunosensor.
The immunoreaction, enzymatic hydrolysis, and potentiometric
measurement were performed in the flow channel. The mea-
surements were made for each concentration of 3 samples. As
a result, by adding the simple process of denaturing HbA1c, the
direct detection of HbA1c (%) was successfully performed with the
immunosensor. The results showed a good correlation (correlation
coefficient; r2 = 0.975) between the certified and measured values
of HbA1c (%) at concentrations from 5.6% to 10.6% (Fig. 4). The error
 J. Tanaka et al. / Sensors and Actuators B 260 (2018) 396–399
bars indicate standard deviation, and the representative data indi-
cated by the circle is the mean value. By combining pretreatment of
specimens with an immunosensor, it has become possible to mea-
sure accurately in the clinically relevant range of HbA1c (%). This is
a great advance for the HbA1c sensor.
Recently, antibody-based immunoassay and enzyme-based
assay with electrochemical detection were more miniaturized
for point-of-care devices [6,9,10]. However, these methods only
measure HbA1c or measure Hb and HbA1c separately, except
for immunosensors that use haptoglobulin [6]. The proposed
immunoassay for HbA1c (%) combined with the process of dena-
turing HbA1c, which showed the direct detection of HbA1c (%)
without any separation steps of Hb and HbA1c, could be achieved
with a disposable potentiometric immunosensor. This direct detec-
tion method of HbA1c (%) could also be applicable to other glycated
proteins, such as glycoalbumin.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at https://doi.org/10.1016/j.snb.2017.12.148.
References
[1] R.R. Little, J.D. England, H.M. Wiedmeyer, E.M. McKenzie, R. Mitra, P.M. Erhart,
J.B. Durham, D.E. Goldstein, Interlaboratory standardization of glycated
hemoglobin determinations, Clin. Chem. 32 (1986) 358–360.
[2] F. Frantzen, K. Grimsrud, D.E. Heggli, E. Sundrehagen, Protein-boronic acid
conjugates and their binding to low-molecular-mass cis-diols and glycated
hemoglobin, J. Chromatogr. B Biomed. Sci. Appl. 670 (1995) 37–45.
[3] T. Tanaka, S. Tsukube, K. Izawa, M. Okochi, T.K. Lim, S. Watanabe, M. Harada, T.
Matsunaga, Electrochemical detection of HbA1c a marker for diabetes, using a
flow immunoassay system, Biosens. Bioelectron. 22 (2007) 2051–2056.
[4] K. Hirokawa, K. Shimoji, N. Kajiyama, An enzymatic method for the
determination of hemoglobinA(1C), Biotechnol. Lett. 27 (2005) 963–968.
[5] F. Engbaek, S.E. Christensen, B. Jespersen, Enzyme immunoassay of
hemoglobin A1c: analytical characteristics and clinical performance for
patients with diabetes mellitus, with and without uremia, Clin. Chem. 35
(1989) 93–97.
[6] D. Stöllner, W. Stöcklein, F. Scheller, A. Warsinke, Membrane-immobilized
haptoglobin as affinity matrix for a hemoglobin-A1c immunosensor, Anal.
Chim. Acta 470 (2002) 111–119.
[7] Y. Ishige, S. Takeda, M. Kamahori, Direct detection of enzyme-catalyzed
products by FET sensor with ferrocene-modified electrode, Biosens.
Bioelectron. 26 (2010) 1366–1372.
[8] M. Kamahori, Y. Ishige, M. Shimoda, Enzyme immunoassay using a reusable
extended-gate field-effect-transistor sensor with a
ferrocenylalkanethiol-modified gold electrode, Anal. Sci. 24 (2008)
1073–1079.
399
[9] C. Bian, J. Tong, J. Sun, H. Zhang, Q. Xue, S. Xia, A field effect transistor
(FET)-based immunosensor for detection of HbA1c and Hb, Biomed.
Microdevices 13 (2011) 345–352.
[10] S.Y. Song, Y.D. Han, Y.M. Park, C.Y. Jeong, Y.J. Yang, M.S. Kim, Y. Ku, H.C. Yoon,
Bioelectrocatalytic detection of glycated hemoglobin (HbA1c) based on the
competitive binding of target and signaling glycoproteins to a
boronate-modified surface, Biosens. Bioelectron. 35 (2012) 355–362.
Biographies
Junko Tanaka received her PhD in Biology from Keio University (Yokohama, Japan)
in 2011. She currently works at Research & Development Group of Hitachi. Her
research interests include biosensors, electrochemical devices and in vitro diagnostic
tests.
Yu Ishige received his PhD in Engineering from Nagoya University (Nagoya, Japan) in
2017. He currently works at Research & Development Group of Hitachi. His research
interests include biosensors, electrochemical devices and in vitro diagnostic instru-
ments.
Riko Iwata received her PhD in Pharmacy from Tokyo University of Science (Tokyo,
Japan) in 1997. She is currently dispatched from Hitachi Chemical to Japan Asso-
ciation for Chemical Innovation. She is working on strategy formulation to make
a roadmap of what the Japanese chemical industry should do taking a look at the
future after thirty years.
Baku Maekawa received his MSc from Waseda University (Tokyo, Japan) in 1986.
He currently works at Risk Management Center of Hitachi Chemical for company
compliance.
Hidehiro Nakamura received his MSc from Kobe University (Kobe, Japan) in 1985.
He currently works at Marketing Center of Hitachi Chemical. He dedicates himself
to open up new markets such as not only electrochemical sensor devices but also
environmental materials including carbon dioxide adsorbent, methanation catalyst
etc.
Takeshi Sawazaki received his MSc in Environmental Science from University of
Tsukuba (Tsukuba, Japan) in 1989. He currently works at Medical Business Unit of
Hitachi Chemical. He belongs to the Planning and Marketing Department, and he is
giving the market information of in vitro diagnostics, mainly allergy and infection
diseases, to Research & Development Group.
Masao Kamahori received the B.S. and M.S. degrees in chemistry from Kyusyu
University (Japan) in 1983, 1985 respectively and PhD degree at School of Pharma-
ceutical Sciences, Showa University (Japan) in 2005. He joined the central research
laboratory (CRL), Hitachi Ltd., in 1985, where he has been engaged in the research
on the high-performance liquid chromatography and its application for medical use.
He then worked on the capillary electrophoresis for DNA separation in the advanced
research laboratory and on the capillary-array DNA sequencer and bioluminescence
DNA detection device in CRL. He is currently working on electrochemical sensors and
spectrophotometer.