High performance immunoassay using immobilized enzyme in nanoporous carbon.

Abstract

A highly stable immunoassay format was constructed using signal-generating enzyme immobilized in nanoporous carbon. A mesocellular carbon foam, called MSU-F-C, was loaded with horseradish peroxidase (HRP), followed by cross-linking of the enzyme using glutaraldehyde (GA) and modification of the surface with anti-human IgG through EDC/sulfo-NHS chemistry. The resulting MSU-F-C/HRP/anti-human IgG stably retained immobilized enzymes and antibodies, showing higher thermal stability. The MSU-F-C/HRP/anti-human IgG retained about 80 % of initial enzyme activity at 40 degrees C after a 5 h incubation, while the HRP/anti-human IgG conjugate resulted in almost 90% loss of initial activity in the same condition. In bead-based immunoassays, the signal amplification using MSU-F-C/HRP/anti-human IgG enabled the sensitive colorimetric detection of a target analyte, human IgG, in a detection limit of approximately 33 pM, with negligible cross-reactivity against rabbit and chicken IgGs.

Full text

Page 1

High performance immunoassay using immobilized enzyme
in nanoporous carbon†
Yunxian Piao,aDohoon Lee,bJungbae Kim,*cJaeyun Kim,dTaeghwan Hyeondand Hak-Sung Kim*a
Received 4th August 2008, Accepted 26th January 2009
First published as an Advance Article on the web 27th February 2009
DOI: 10.1039/b813451k
A highly stable immunoassay format was constructed using signal-generating enzyme immobilized in
nanoporous carbon. A mesocellular carbon foam, called MSU-F-C, was loaded with horseradish
peroxidase (HRP), followed by cross-linking of the enzyme using glutaraldehyde (GA) and
modification of the surface with anti-human IgG through EDC/sulfo-NHS chemistry. The resulting
MSU-F-C/HRP/anti-human IgG stably retained immobilized enzymes and antibodies, showing higher
thermal stability. The MSU-F-C/HRP/anti-human IgG retained about 80 % of initial enzyme activity
at 40?C after a 5 h incubation, while the HRP/anti-human IgG conjugate resulted in almost 90% loss of
initial activity in the same condition. In bead-based immunoassays, the signal amplification using
MSU-F-C/HRP/anti-human IgG enabled the sensitive colorimetric detection of a target analyte,
human IgG, in a detection limit of ?33 pM, with negligible cross-reactivity against rabbit and
chicken IgGs.
Introduction
Immunoassay is a popular analytical technique based on the
specific interaction between antibody and antigen. Fairly
sophisticated immunoassay systems are currently used for
molecular diagnosis in modern clinical laboratories, and their
full performance requires special reagents, a stable power supply,
and trained personnel. However, where the technology is of
urgent need, such as in disease control in the developing world1
and warfare detection,2access to such resources is generally
limited, and the reliability and ruggedness of detection are of
comparable importance to the sensitivity and accuracy.
In conventional immunoassays, antibody–enzyme conjugates
are generally used to translate the molecular recognition events
to measurable signals generated by enzymatic catalysis. Varia-
tions of this technique have been pursued to further enhance the
detection sensitivity. One of them is to use enzymes immobilized
on proper carriers3like carbon nanotubes,4sol–gel composites,
silica, or some other nano-structured materials,5which resulted
in improved signal amplification based on a high enzyme :
antibody ratio. These techniques have been reported to be
successful in improving sensitivity, but the stability of sensing
systems has not been assessed in detail.
In an effort to develop a highly stable immunoassay system
with a comparable sensitivity, we immobilized a signal-gener-
ating enzyme in mesocellular carbon foam, MSU-F-C. In our
previous report, the MSU-F-C was used as a host matrix for
enzyme immobilization due to its large surface area and pore
volume.6Immobilization of enzymes in tailor-made nanoscale
structures was revealed to significantly improve the perfor-
mance of biocatalytic processes,7–9since they can enhance the
loading of bio-molecules and stability of the biocatalysts as well
as mass transport. In addition, the MSU-F-C can be func-
tionalized with carboxylate groups by proper acid treatment,10
and the carbon surface can be easily labeled with various
bio-molecules.
Here we report on construction of a highly stable immuno-
assay format based on immobilization of signal-generating
enzyme in nanoporous carbon. Horseradish peroxidase (HRP)
was immobilized in the pores of MSU-F-C in a simple way and
cross-linked using glutaraldehydes (GA) as reported elsewhere.11
The surface of MSU-F-C was covalently labeled with anti-
human IgG using EDC as zero-length linker (Scheme 1A). The
antigen(humanIgG), captured
magnetic beads covalently coated with antibodies, is probed by
the stabilized enzyme activity of MSU-F-C/HRP/anti-human
IgG as depicted in Scheme 1B. Following magnetic separation,
the amount of captured IgG is quantified by colorimetric assay
using TMB (3,30,5,50-tetramethylbenzidine) as a HRP substrate.
The signal generated by MSU-F-C/HRP/anti-human IgG was
found to be linearly proportional to the target analyte concen-
tration, and the detection limit was estimated to be 33 pM. The
MSU-F-C/HRP/anti-human IgG showed high thermal stability
compared to the conventional HRP/anti-human IgG conjugate.
Details are reported herein.
by epoxy-functionalized
aDepartment of Biological Sciences, Korea Advanced Institute of Science
and Technology, Daejeon, 305-701, Korea. E-mail: hskim76@kaist.ac.kr
bKorea Institute of Industrial Technology, 35-3, Hongcheon-ri, Ipjang-
myun, Cheonan, 330-825, Korea
cDepartment of Chemical & Biological Engineering, Korea University,
Seongbuk-Gu Anam-Dong 5-Ga 1-Bunji, Seoul, 136-701, Korea. E-mail:
jbkim3@korea.ac.kr
dNational Creative Research Initiative Center for Oxide Nanocrystalline
Materials and School of Chemical and Biological Engineering, Seoul
National University, Seoul, 151-744, Korea
† Electronic supplementary information (ESI) available: Loading and
activity of HRP in MSU-F-C treated with acid under different
conditions, dependence of antibody loading on initial concentration
during immobilization on MSU-F-C/HRP, calibration plot for the
magnetic-bead basedimmunoassay
MSU-F-C/HRP/anti-human IgG and MB/anti-human IgG. See DOI:
10.1039/b813451k
ofhumanIgG using
926 | Analyst, 2009, 134, 926–932This journal is ª The Royal Society of Chemistry 2009
PAPERwww.rsc.org/analyst | Analyst

Page 2

Experimental
Chemicals and materials
Horseradish peroxidase (HRP, 190 U mg?1; U: enzyme unit, RZ
¼ 1.9), glutaraldehyde (GA, 25%), anti-human IgG (Fab
specific), anti-human IgG (Fc specific), HRP-conjugated anti-
human IgG (Fc specific), and IgGs from human, rabbit, and
chicken sera were purchased from Sigma. 3,30,5,50-Tetrame-
thylbenzidine (TMB) as HRP substrate (supplied as a ready-to-
use form),
N-(3-dimethylaminopropyl)-N0-ethylcarbodiimide
HCl (EDC), 4-morpholineethanesulfonic acid (MES), ammo-
nium sulfate, and skimmed milk were also purchased from
Sigma.
N-hydroxysulfosuccinimide
Pierce.DynabeadM-270-epoxy
purchased from Dynal Biotech, Invitrogen. Other reagents used
were of the highest quality available and purchased from the
regular sources. Distilled and deionized water was used
throughout the work.
For the buffer solutions, 0.1 M sodium phosphate buffer (pH
7.0) was used as a reaction buffer, and 0.1 M Tris-HCl buffer (pH
8.0) was used as blocking buffer when immobilizing HRP and
anti-human IgG to MSU-F-C. Phosphate buffered saline (PBS,
10 mM phosphate, 138 mM NaCl, and 2.7 mM KCl, pH 7.4)
containing 0.5 % skimmed milk and 0.05 % Tween 20 was
used as wash buffer, and PBS containing 0.5 % skimmed milk
was used for blocking buffer. PBS containing 1% skimmed milk
was used when storing and blocking MSU-F-C/HRP/anti-
human IgG. All buffer solutions were used after filtration using
a membrane (0.2 mm cut-off).
(sulfo-NHS)
(2.8
mm
wasfrom
wasdiameter)
Synthesis and functionalization of MSU-F-C
MSU-F-C was synthesized using MSU-F silica12as a template as
previously described.6First, nanometer-sized MSU-F silica
particles were synthesized by means of a modified version of
the original synthetic procedure, employing hydrothermal
post-treatment at 100?C.12After 4 h calcination at 550?C,
alumination (Si/Al ¼ 20) was performed using the impregnation
method to generate acidic catalytic sites for the polymerization of
furfuryl alcohol inside the mesopores (ALMSU-F). Typically,
the MSU-F was impregnated with furfuryl alcohol (1 mL)
through an incipient-wetness method. The furfuryl alcohol was
polymerized at 85?C under vacuum for more than 12 h. The
resulting composite was heat-treated under N2at 850?C. Finally,
etching with HF or NaOH generated mesocellular carbon foam,
designated as MSU-F-C. Structure of the MSU-F-C was
analyzed using transmission electron microscopy (TEM) on
a JEOL JEM-2010 electron microscope. The micropore and
mesopore size distributions were calculated from the adsorption
branches of Ar isotherms using the Horvath–Kawazoe and the
BJH formalisms,13respectively.
In order to generate carboxylate groups on the carbon surface
and improve its dispersion in aqueous solution, the MSU-F-C
(10 mg) was treated with a mixture (2.2 mL) of concentrated
H2SO4and HNO3by incubating it at 25?C with mild shaking,
followed by centrifugation and thorough washing with DDW.
Three different conditions were used: (1) 30 min treatment in the
mixture of concentrated H2SO4, HNO3and filtered DDW (2 : 1 :
8, v/v), (2) 2 h treatment in the same kind of mixture as the first
one, (3) 30 min treatment in the mixture of concentrated H2SO4
and HNO3(3 : 1, v/v). Surface chemical functionality of the
treated MSU-F-C was analyzed by Fourier transform infrared
(FT-IR) spectrophotometer (Thermo Nicolet Nexus).
Immobilization of HRP in MSU-F-C
After washing with 0.1 M sodium phosphate buffer (pH 7.0)
three times, the MSU-F-C (6 mg) was incubated in HRP solution
(2 mg mL?1in sodium phosphate buffer) at 16?C with shaking
(250 rpm) for 20 min. By centrifugation at 4?C and 13 000 rpm
for 5 min, the MSU-F-C/HRP was collected and treated with GA
(0.5% in sodium phosphate buffer) at 25?C for 1 h (without
Scheme 1
IgG). (A) The MSU-F-C/HRP/Ab was prepared by two-step modification of the nanoporous carbons, MSU-F-C. First, HRP was immobilized inside
the carbon particles accompanied by GA treatment, and then the resulting particles were further modified with antibodies through EDC/Sulfo-NHS
chemistry. (B) The MB/Ab was prepared by covalent attachment of antibodies on the epoxy-modified magnetic bead surface. After sandwich-type
immunocomplex formation, the target antigen (Ag) was quantified by measuring the enzymatic turnover of TMB at 655 nm using a spectrophotometer.
Schematic diagram of immunoassay using MSU-F-C/HRP/Ab (Fc specific) and MB/Ab (Fab specific) (MB: magnetic bead, Ab: anti-human
This journal is ª The Royal Society of Chemistry 2009Analyst, 2009, 134, 926–932 | 927

Page 3

shaking for the first 30 min, and with shaking at 200 rpm for the
next 30 min). After briefly washing with Tris buffer (0.1 M, pH
8.0), the MSU-F-C/HRP was incubated in the Tris buffer for 1 h
at 25
aldehyde groups. Finally, it was washed three times with MES
buffer (0.1 M MES, 0.5 M NaCl, pH 6.0) until no enzyme was
detected in the buffer. The enzyme concentration in the buffers
was measured by the Bradford method (Bio-Rad) and used for
the calculation of enzyme loading within MSU-F-C.
?C with shaking (250 rpm) to block remaining active
Immobilization of anti-human IgG on to MSU-F-C/HRP
particle surface
In order to covalently immobilize anti-human IgG, EDC (86.8
mM) and sulfo-NHS (5.32 mM) were added to MSU-F-C/HRP
(2 mg) resuspended in 0.1 M MES (pH 6.0, 2 mL). To form labile
intermediates on the carbon surface, the mixture was incubated
at 25?C for 30 min with shaking (250 rpm) accompanied by
sonication every 10 min. The excess EDC and sulfo-NHS
reagents were removed thoroughly by three-time washing using
MES with shaking (250 rpm) at 25?C and centrifugation at
13 000 rpm (10 min). The activated MSU-F-C/HRP (2 mg) was
resuspended in anti-human IgG (Fc-specific) solution (in sodium
phosphate buffer). After brief sonication, the mixture was incu-
bated in Intelli Mixer (MyLab? SLRM-2M) with mild shaking
(F/60, RPM/5) for 2 h at 25?C and overnight at 4?C. To cap the
remaining sulfo-NHS activated intermediates, the MSU-F-C/
HRP/anti-human IgG was incubated in 0.1 M Tris buffer (pH
7.2) for 1 h at room temperature with shaking (250 rpm). Finally,
the MSU-F-C-/HRP/anti-human IgG was washed thoroughly
with PBS (pH 7.4) three times, followed by centrifugation. The
final product was resuspended in PBS (pH 7.4) containing
skimmed milk (1%) and stored at 4?C. The antibody concen-
tration in buffer solution was determined by the Bradford
method (Bio-Rad) and used for the calculation of antibody
loading on MSU-F-C/HRP/anti-human IgG.
Assay of HRP activity
HRP activity was assayed, using TMB as a substrate, by
measuring the absorbance change at 655 nm s?1. Briefly, sample
solutions containing HRP were diluted to proper concentrations
in 0.1 M sodium phosphate buffer (pH 7.0). The diluted sample
solutions (100 mL each) were loaded into acryl cuvettes, followed
by addition of 900 mL of TMB liquid substrate. Then the cuvette
was incubated in the cell of a spectrophotometer (UV-2550,
Shimadzu) for 30 s at room temperature in the dark. Absorbance
at 655 nm was measured every 10 s. By preparing a plot of A655
vs. time, absorbance change (DA) over 30 s was calculated,
during which the absorbance should increase linearly. The rela-
tive specific activity of HRP was expressed as (DA s?1) mg?1of
HRP or (DA s?1) mg?1of MSU-F-C. The activity of HRP
immobilized in MSU-F-C or conjugated to antibodies were also
determined following the same procedure after proper dilution.
In this dilution range, effect of the suspended MSU-F-C particles
on the absorbance measurement was negligible. During the
absorbance measurement, the solution was stirred regularly by
mild pipetting to prevent the particles from settling down and to
facilitate the mass transport.
Modification of magnetic beads with anti-human IgG
Magnetic beads were modified with anti-human IgG (Fab-
specific) according to the manufacturer’s instruction. Briefly, 2
mg of epoxy-coated magnetic bead (Dynal Biotech) was sus-
pended in 133 mL of 0.1 M sodium phosphate buffer (pH 7.4) to
give a concentration of approximately 109beads mL?1by vor-
texing. The beads were washed three times for 10 min each by
incubating in Intelli mixer. After magnetic separation and
resuspension in sodium phosphate buffer (40 mL), the magnetic
beads were mixed with 1 mg mL?1of anti-human IgG (40 mL)
prepared in PBS (pH 7.4) and 3 M ammonium sulfate stock
solution (40 mL). After incubation for 24 h at room temperature
with mild shaking, the antibody-coated magnetic beads were
washed twice with PBS, and twice with PBS containing 0.5%
skimmed milk. Finally, the MB/anti-human IgG (MB: magnetic
bead) was resuspended in PBS containing 0.5% skimmed milk to
make 108beads mL?1stock solution, and stored at 4?C.
Immunoassay using MSU-F-C/HRP/anti-human IgG and MB/
anti-human IgG
Prior to the assay, the MB/anti-human IgG was washed three
times with wash buffer (PBS containing 0.05% Tween 20 and
0.5% skimmed milk) accompanied by mild sonication. The
washed MB/anti-human IgG (5 ? 105beads) were incubated with
different concentrations of human IgG (50 mL, in wash buffer)
for 1 h with mild shaking. Then, they were washed three times
with wash buffer and collected by magnetic separation (Scheme
1B). To probe the captured human IgGs, the MB/anti-human
IgG/human IgG was incubated with 100 mg mL?1MSU-F-C/
HRP/anti-human IgG (50 mL in wash buffer) for 2 h with
shaking. After washing three times with wash buffer followed by
magnetic separation, the immunocomplex was incubated in 100
mL TMB for 30 min with shaking. After magnetic separation of
the immunocomplex, the solution was transferred to 96-well
plates, and the absorbance (at 655 nm) was measured immedi-
ately using a microplate reader (Infinite M200 Megellan, Tecan).
The whole immunoassay process was carried out at room
temperature and all the shaking steps were performed in Intelli
Mixer (MyLab? SLRM-2M). A magnetic separator (Magna
Rack?, Invitrogen) was used for the bead separation and it takes
generally less than 1 min to completely collect the magnetic
beads. As a cross-reactivity test, rabbit IgG and chicken IgG
were used separately, and the immunoassay was performed with
the same experimental procedures.
Results and discussion
Characterization of MSU-F-C
It was found that MSU-F-C (particle size < 500 nm) has large
pore size (1.63 cm3g?1) and large surface area (671 m2g?1) from
BET analysis. TEM images and pore-size-distribution analyses
revealed that MSU-F-C possesses two types of pores with
distinct sizes. Large cellular pores (D z 31 nm), interconnected
by smaller windows (D z 21 nm), are surrounded by small pores
928 | Analyst, 2009, 134, 926–932This journal is ª The Royal Society of Chemistry 2009

Page 4

(D z 5.6 nm), those are empty spaces formed after removal of
the silica template (Fig. 1b). These unique properties of the
MSU-F-C offer several favorable features as a host matrix for
enzyme immobilization. The horseradish peroxidase (HRP) with
molecular dimensions of 4.3 nm ? 3.7 nm ? 6.4 nm,14which is
much smaller than the pore window, can readily enter the large
pores and be kept stably within the pore by cross-linking with
glutaraldehyde (GA) (Fig. 1b). Small pores can play a role as
a substrate transporting channel, facilitating the enzyme
reaction.
Functionalization of MSU-F-C
Prior to the enzyme immobilization, MSU-F-C was treated with
diluted acid to improve wettability in aqueous solution and to
provide carboxylate functional groups for covalent attachment
of antibodies. The functionalization of MSU-F-C by carboxylic
acid groups was confirmed by IR spectroscopy. Fig. 2 shows IR
spectra of the MSU-F-C treated with acid under different
conditions. Comparison with control (untreated sample) shows
appearance of a peak at 1740 cm?1(carboxylate C]O stretch-
ing15) after acid treatment. In addition, this peak became sharper
as the acid treatment condition became stronger, indicating that
more carboxylate groups were created by oxidation. Although
presence of large numbers of functional groups should be
favorable for the immobilization of antibodies, significant
reductions in enzyme loading and specific activity recovery was
observed when stronger acid treatment conditions were used
(Fig. S1 in ESI†). This implies that the pores of MSU-F-C got
somehow damaged by over oxidation and could not keep
enzymes effectively. Or, the dense carboxylate groups on MSU-
F-C’s surface might hinder enzymes from entering the pores,
possibly by charge repulsion. Therefore, MSU-F-C treated under
the mildest condition (condition 1, see Experimental) was used in
the following experiments.
Immobilization of HRP in MSU-F-C
By incubating the MSU-F-C in HRP solution with shaking and
carefully treating it with GA, the enzymes were immobilized
within the MSU-F-C. To determine the appropriate initial
enzyme concentration for the immobilization, the MSU-F-C was
incubated with different concentrations of HRP and the loadings
were assessed. The enzyme loading increased almost linearly with
the enzyme’s initial concentrations (Fig. 3a), which means that
the mesopores of MSU-F-C can hold large amounts of small
enzymes like HRP (40 kDa). The HRP activity per mass of
MSU-F-C increased almost linearly with enzyme loading either,
up to 41.7 ? 1.7% (w/w carbon, n ¼ 3, n ¼ number of indepen-
dent measurements) when incubated in 2 mg mL?1HRP solution
(Fig. 3b). The activity, however, levels off above the value,
maybe due to increased mass transfer resistance by over-crowded
enzymes. Therefore, in the following experiments, 2 mg mL?1
HRP was used for the immobilization. Specific activity of HRP
in MSU-F-C was about 19% of that of free HRP in solution.
Immobilization of anti-human IgG
For covalent immobilization of anti-human IgG onto the MSU-
F-C/HRP surface, the zero length linker EDC and sulfo-NHS16
were used to give rise to active ester functional groups by reacting
with the carboxylic acid groups on the surface of MSU-F-C/
HRP. The ester groups are easily reacted with amines of the IgG.
As a negative control, immobilization of IgG was tested without
EDC and sulfo-NHS. Assessment by protein assay showed that
Fig. 1 TEM image of MSU-F-C (a) and schematic representation of the MSU-F-C/HRP showing immobilized enzymes in large cellular mesopores (b).
Fig. 2
different conditions.
FT-IR spectra of MSU-F-C after being treated with acid under
This journal is ª The Royal Society of Chemistry 2009 Analyst, 2009, 134, 926–932 | 929

Page 5

the amounts of the bound anti-human IgG through covalent or
linker-free immobilization were almost the same (z12%), which
suggests that anti-human IgG can also bind to the MSU-F-C
surface by adsorption through Van der Waals forces or charge
interaction between the negatively charged carboxylate groups
on the particle and the positively charged amine groups on the
IgG. However, immunoassays using the two types of probes,
MSU-F-C-/HRP/EDC/anti-human IgG (by covalent immobili-
zation) and MSU-F-C-/HRP/anti-human IgG (by adsorption)
revealed that the former generated a detection signal about five
times higher than the latter (Fig. 4). This result also confirms that
carboxylate groups were formed by acid treatment and contrib-
uted to the stable immobilization of anti-human IgG by covalent
linkage. A comparable amount of anti-human IgGs can be
immobilized by simple adsorption without linker, but they
are not likely to beimmobilized effectively to favor binding of the
target analyte17and might be easily washed away during the
immunoassay process.
The amount of immobilized antibody via covalent attachment
was increased by increasing the antibody concentration from
0.008 to 2.4 mg mL?1(Fig. S2 in ESI†). However, the detection
signal reached a maximum level when the initial antibody
concentration was 0.8 mg mL?1(Fig. 5), implying that the
molecular recognition can be retarded by steric hindrance in the
over-crowded state.18Acid treatment condition was also found
to have an effect on the performance of the immunoassay. A
prominent reduction in the detection signal was observed
(Fig. S3 in ESI†) when the MSU-F-C was treated under the
severest condition (condition 3, see Experimental). This result
seems to be due to the low recovery of enzyme activity after
enzyme loading (Fig. S1 B in ESI†).
Fig. 3
of MSU-F-C/HRP with different enzyme loadings (b). The error bars
indicate the standard deviation in triplet experiments. MSU-F-C (6 mg)
was incubated with HRP solution of specified concentration under
shaking (250 rpm) at 16?C for 20 min, followed by GA treatment. To
determine the appropriate initial enzyme concentration for immobiliza-
tion,the enzymatic activity per massof MSU-F-C particles was measured
using a TMB based activity assay.
Immobilization isotherm of HRP in MSU-F-C (a) and activities
Fig. 4
using MSU-F-C/HRP/anti-human IgG. Black bars represent the signals
when anti-human IgG was covalently attached onto the MSU-F-C/HRP.
Gray bars represent the signals from the negative control, in which
no chemical linker was used and the antibody was immobilized via
adsorption. The error bars indicate the standard deviations in duplicate
experiments.
Magnetic bead-based immunoassay of human IgG (hIgG)
Fig. 5
antibody concentrations for the immobilization on MSU-F-C/HRP were
varied from 0.008 to 2.4 mg mL?1. The error bars indicate the standard
deviations in duplicate experiments.
Effect of antibody loading on detection performance. The initial
930 | Analyst, 2009, 134, 926–932This journal is ª The Royal Society of Chemistry 2009

Page 6

When EDC and sulfo-NHS were used, the activity of immo-
bilized HRP was reduced to about 11.6% of that of free HRP,
which was lower than the activity of HRP immobilized without
linker (z15.7%), indicating that the enzyme activity might be
affected by EDC/sulfo-NHS chemistry. One of the reasons for
activity loss might be that EDC and sulfo-NHS can enter the
pores of the MSU-F-C and react with the carboxylic acid groups
on the enzymes, resulting in further polymerization between
the enzymes. Another possibility is that the anti-human IgG
(150 kDa) may also enter the main pores of the MSU-F-C and be
covalently linked to the enzymes or just stay there retarding the
mass transfer.
Bead-based immunoassay using MSU-F-C/HRP/anti-human
IgG
As described in the Experimental section, magnetic beads
modified with capture antibodies (MB/anti-human IgG) were
used to capture model antigens, human IgGs. After washing, the
human IgGs captured were probed by MSU-F-C/HRP/anti-
human IgG, forming sandwiched immunocomplexes (Scheme
1B). The more human IgGs were captured by the MB/anti-
human IgG, the more MSU-F-C/HRP/anti-human IgGs can be
included in the immunocomplex, generating larger signals by
enzymatic reactions.
In the calibration plot (Fig. 6) of the assay, a linear correlation
between the analyte concentration and detection signal (after
30 min colorimetric reaction) was observed over the concentra-
tion range of 5?1000 ng mL?1in the semi-log plot. In the linear
plot, the dynamic range was estimated to be 5?100 ng mL?1, and
the linear regression equation was Abs ¼ 0.0893 + 0.0042c (ng
mL?1) (R2¼ 0.9996). The detection limit was estimated to be 5 ng
mL?1or 33 pM for human IgG (150 kDa). To evaluate the
specificity of the developed method, the same immunoassay was
also performed using rabbit IgG and chicken IgG as negative
controls, respectively. As shown in Fig. 6, there was a negligible
cross-reactivity against rabbit and chicken IgGs. This result
suggests that the developed assay system taking advantage of the
easy separation by a magnet and the proper blocking method has
a reliable specificity to its target analyte.
With MSU-F-C/HRP/anti-human IgG as an immuno-probe,
the ratio of enzyme to antibody could be significantly improved
compared to conventional methods using HRP/antibody conju-
gates, and consequently the detection signal of molecular inter-
action events is expected to increase. Furthermore, with stably
immobilized enzymes in porous carbon, other enzymes, which
may not be stable when exists as a conjugated form with anti-
bodies, can also be used as a signal generator. The detection
sensitivity of this technique was not as high as those of conven-
tional immunoassays using conjugate of HRP/anti-human IgG
(data not shown) and other similar techniques.3,19However, it is
expected that the developed method as a proof-of-concept can be
further improved by tuning of morphology and structure of the
carbon carrier, and more rigorous control of non-specific
binding.
Stability of the MSU-F-C/HRP/anti-human IgG
When the MSU-F-C/HRP/anti-human IgG particles were stored
in 4?C, the initial activity of HRP was revealed to maintain even
after 10 days (data not shown) without leaching of any enzymes.
To assess the thermostability of the immobilized enzymes, the
MSU-F-C/HRP/anti-human IgG (5 mg mL?1) was incubated in
a 40?C chamber. At intervals, small aliquots containing MSU-F-
C/HRP/anti-human IgG were sampled and assayed for HRP
activity. For comparison, stability of the HRP/anti-human IgG
conjugate was also traced under the same condition. As shown in
Fig. 7, the enzyme in MSU-F-C/HRP/anti-human IgG exhibited
Fig. 6
human IgG (black squares) using MSU-F-C/HRP/anti-human IgG and
MB/anti-humanIgG.TherabbitIgG(rIgG,whitecircles)andchickenIgG
(cIgG, white squares) were used as negative controls. The concentrations
ofMSU-F-C/HRP/anti-humanIgGandMB/anti-humanIgGusedforthe
assay were 100 mg mL?1and 1 ? 107beads mL?1, respectively. Volume of
the assay solution was 50 mL. The enzyme reaction time was 30 min.
The error bars indicate standard deviations in triplicate experiments.
Calibration plot for the magnetic-bead based immunoassay of
Fig. 7
Ab at 40?C (Ab: anti-humanIgG). The diluted MSU-F-C/HRP/Ab (5 mg
mL?1) (black circles) or the HRP/Ab conjugate (1 : 10000) (white circles)
were incubated at 40?C for 24 h. At intervals, small aliquots (100 mL)
were sampled and assayed for residual HRP activity using TMB as
substrate.
Thermostability of the HRP/Ab conjugate and MSU-F-C/HRP/
This journal is ª The Royal Society of Chemistry 2009Analyst, 2009, 134, 926–932 | 931

Page 7

much higher thermal stability compared to that of HRP/anti-
human IgG. After 5 h incubation, HRPs immobilized in meso-
porous carbon still maintained about 80% of the initial activity
and about55% evenafter 24 h. However, the conventionally used
HRP/anti-human IgG conjugates already lost almost 90% of its
initial activity in 5 h, and no remaining activity was observed
after 12 h under the same conditions. This result indicates that
the enzyme immobilized in mesoporous carbons can be protected
from harsh environmental condition. In addition, by measuring
the HRP activity of the solution, we confirmed that leaching out
of enzymes from the mesoporous carbon during the incubation
at high temperature was negligible. It is likely that cross-linking
of the enzymes with glutaraldehyde (GA) and bottleneck-like
structure of MSU-F-C with smaller windows (Fig. 1b) would
prevent the enzymes from leaching out.20
Conclusions
Wehave demonstrateda high-performance immunoassay format
based on immobilized enzymes in nanoporous carbon. With the
unique structural characteristics of MSU-F-C including large
pore volume, surface area, and mesopores, the immobilized HRP
inside MSU-F-C showed much higher stability at high temper-
atures compared to HRP directly conjugated to anti-human IgG.
After further modification with anti-human IgG, the resulting
MSU-F-C/HRP/anti-human IgG probe was effectively used in
a magnetic bead-based immunoassay. The developed assay
format enabled detection of pico molar human IgG with
a detection limit of 33 pM, showing a negligible cross-reactivity
against rabbit and chicken IgGs. When considering the high
stability of the immobilized enzyme, the enzyme-containing
MSU-F-C can also be a valuable component for immunoassay in
field applications requiring robustness of the assay system.
Acknowledgements
This work was supported by the Nano/Bio Science & Technology
Program (M1053609000205N3609-00210), Pioneer Research
Program for Converging Technology (M10711300001-08M1130-
00110) from Ministry of Education, Science, and Technology
(MEST), and the Korea Health 21 R&D Project (0405-MN01-
0604-0007) of the Ministry of Health & Welfare (MOHW). We
also thank I. S. Choi and S. M. Kang of the Department of
Chemistry, KAIST for technical assistance of FT-IR analyses.
References
1 P. Yager, T. Edwards, E. Fu, K. Helton, K. Nelson, M. R. Tam and
B. H. Weigl, Nature, 2006, 442, 412.
2 A. H. Peruski, L. H. Johnson, 3rd. and L. F. Peruski, Jr., J. Immunol.
Methods, 2002, 263, 35; A. H. Peruski and L. F. Peruski, Jr., Clin.
Diagn. Lab. Immunol., 2003, 10, 506; Y.-H. Oh, Y.-P. Kim and H.-
S. Kim, Biochip J., 2007, 1, 28; E. Kim, B. S. Lee, H.-B. Pyo, H.-
W. Song, Y.-P. Kim, I. S. Choi and H.-S. Kim, Biochip J., 2008, 2,
103.
3 X. Yu, B. Munge, V. Patel, G. Jensen, A. Bhirde, J. D. Gong,
S. N. Kim, J. Gillespie, J. S. Gutkind, F. Papadimitrakopoulos and
J. F. Rusling, J. Am. Chem. Soc., 2006, 128, 11199; Z. Wang,
Y. Yang, J. Gong, G. Shen and R. Yu, Talanta, 2006, 69, 686;
A. Ambrosi, M. T. Castaneda, A. J. Killard, M. R. Smyth,
S. Alegret and A. Merkoci, Anal. Chem., 2007, 79, 5232.
4 J. Wang, G. Liu and M. R. Jan, J. Am. Chem. Soc., 2004, 126, 3010;
B. Munge, G. Liu, G. Collins and J. Wang, Anal. Chem., 2005, 77,
4662.
5 B. Y. Won, H. G. Choi, K. H. Kim, S. Y. Byun, H. S. Kim and
H. C. Yoon, Biotechnol. Bioeng., 2005, 89, 815.
6 D. Lee, J. Lee, J. Kim, J. Kim, H. B. Na, B. Kim, C. H. Shin,
T. Hyeon and H. S. Kim, Adv. Mater., 2005, 17, 2828.
7 C. Lei, Y. Shin, J. Liu and E. J. Ackerman, J. Am. Chem. Soc., 2002,
124, 11242.
8 Y. J. Han, J. T. Watson, G. D. Stucky and A. Butler, J. Mol. Catal. B:
Enzym., 2002, 17, 1.
9 Y. Wang and F. Caruso, Chem. Commun., 2004, 1528.
10 Y. Yang, H. Zou, B. Wu, Q. Li, J. Zhang, Z. Liu, X. Guo and Z. Du,
J. Phys. Chem. B, 2002, 106, 7160.
11 J. Lee, J. Kim, J. Kim, H. Jia, M. I. Kim, J. H. Kwak, S. Jin,
A. Dohnalkova, H. G. Park, H. N. Chang, P. Wang, J. W. Grate
and T. Hyeon, Small, 2005, 1, 744.
12 S. S. Kim, T. R. Pauly and T. J. Pinnavaia, Chem. Commun., 2000,
1661; P. Schmidt-Winkel, W. W. Lukens, Jr., D. Zhao, P. Yang,
B. F. Chmelka and G. D. Stucky, J. Am. Chem. Soc., 1999, 121,
254; P.Schmidt-Winkel, W.
D. L. Margolese, J. S. Lettow, J. Y. Ying and G. D. Studky, Chem.
Mater., 2000, 12, 686.
13 E. P. Barrett, L. G. Joyner and P. P. Halenda, J. Am. Chem. Soc.,
1951, 73, 373.
14 H. Takahashi, B. Li, T. Saaki, C. Miyazaki, T. Kajino and S. Inagaki,
Chem. Mater., 2000, 12, 3301.
15 K. R. Yoon, W. J. Kim and I. S. Choi, Macromol. Chem. Phys., 2004,
205, 1218; X. J. Huang, H. S. Im, D. H. Lee, H. S. Kim and
Y. K. Choi, J. Phys. Chem. C, 2007, 111, 1200; C. Ping, F. Li,
Z. Jian and J. Wei, Propell. Explos. Pyrot., 2006, 31, 452.
16 G. T. Hermanson, Bioconjugate Techniques, Academic Press, San
Diego, 1996, pp. 169–176.
17 H. Zhang and M. E. Meyerhoff, Anal. Chem., 2006, 78, 609.
18 C. Duschl, A. F. S? evin-Landais and H. Vogel, Biophys. J., 1996, 70,
1985; Y. Luo, M. Chen, Q. Wen, M. Zhao, B. Zhang, X. Li,
F. Wang, Q. Huang, C. Yao, T. Jiang, G. Cai and W. Fu, Clin.
Chem., 2006, 52, 2273.
19 T. Chen, J. D. Lei and A. J. Tong, Luminescence, 2005, 20, 256.
20 J. Lee, D. Lee, E. Oh, J. Kim, Y. P. Kim, S. Jin, H. S. Kim, Y. Hwang,
J. H. Kwak, J. G. Park, C. H. Shin, J. Kim and T. Hyeon, Angew.
Chem., Int. Ed., 2005, 44, 7427.
W.Lukens, Jr., P. Yang,
932 | Analyst, 2009, 134, 926–932This journal is ª The Royal Society of Chemistry 2009