An optical immunosensor based on
surface plasmon resonance for determination of B factor
Mu Ying , Song Daqian , Zhao Xiaojun , Zhang Hanqi , Jin Qinhan
(Department of Chemistry, Jilin University, Changchun, Jilin 130023, China)
Received Oct. 24, 1999; Supported by the National Natural Science
Foundation of China. (Grant No. 29875010)
Abstract A novel optical
immunosensor based on surface plasmon resonance (SPR) has been developed for
identification of antigen. The sensor is designed on the basis of fixing angle of
incidence and measuring the reflected intensities in the wavelength ranges of 400-800 nm in real-time. Molecular self-assembling in solution is used to
form the sensing membrane on gold substrate. The kinetic processes of sensing monolayer
formation were studied. Human B factor (Bf), an activator of
complement C3, was detected in the concentration range of 1-80µg/mL. Under selected
experimental conditions, the sensor has a good repeatability, reversibility and
selectivity.
Keywords surface plasmon resonance, immunosensor, B factor
Many kinds of surface plasmon resonance
(SPR) sensors have been proposed rapidly in recent years [1-7]. This technique
has two obvious advantages over traditional methods. Firstly, labeling of molecules is
dispensable in SPR sensor. Secondly, simultaneous observation of the interaction among
macromolecules can be realized; therefore, kinetic constants of reactions can be obtained.
This kind of instruments is used in fields such as environmental and medical science.
SPR is a simple and direct sensing technique used to probe refractive index changes in the
close vicinity of a thin metal film surface. If the wavelength of incident light keeps the
same, the reflective intensity is the least at certain angle when SPR happens. Most of SPR
sensors developed so far are based on fixing the excitation wavelength and modulating the
angle of incident light [1-3]. There are two methods to change the angle. One
is to use a huge machinery rotate shelf. They make use of a coupling prism coated with a
thin gold film and perform the sensing by varying the incidence angle with a goniometer.
The best angular accuracy of the goniometer is about 0.001o which corresponds
to a shift in optical wavelength of 0.6 nm if a fixed angle-of-incidence system is used [3].
The other method is to use the diffuse effect of a spotlight measuring the intensity of
the refractive light in a detector matrix [3]. The shortage of it is the
limited measuring angle.
On the other hand, if the angle of the incident light is fixed, the
reflective intensity will have a minimum at certain wavelength. Based on the latter
theorem, simultaneous multi-wavelength SPR sensors were put forward [4-7].
Halogen tungsten lamp with wide range of wavelength was used as the light source. When the
incident angle of the light is fixed, changes in sensing membrane will give rise to a
shift of resonant wavelength. The reflective intensity of SPR spectra can be observed
simultaneously for the wavelength from 400 to 800 nm. It is the changes in concentrations
or kinds of analytes in the cell that lead to the changes in the refractive index of
sensing membrane. In some range of concentration, the shifting value of resonant
wavelength is in linear relationship with the concentration. So the concentration of an
analyte can be determined according to the shifting value of resonant wavelength. There
are three obvious advantages of this SPR sensor. Firstly, there is no mobile part.
Secondly, the variable is the wavelength not the incident angle. In theory, the
sensitivity of measuring the resonance angle is lower than that of measuring the resonance
wavelength. And finally, the sensor was detected simultaneously for multi-wavelength, not
the scan of multi-wavelength. The scan is not a real time monitoring because the scan
costs time. The system is suited for determination of some chemical species [5,6],
antibody-antigen [7], and DNA.
Over the past decade, the technique of preparing self-assembled
monolayers (SAMs) from solutions containing alkyl thiols and w-functionalized alkyl thiols has been
developed for modifying noble metal (Au and Ag) surfaces. Gold surfaces
modified using SAMs with various terminal groups provide surfaces with a wide range of
different characteristic [8]. Moreover, these surfaces are at least
quasicrystalline, stable, and reproducible. It has been shown that the SAM surface is a
well-defined system for examining the interaction between proteins and surfaces and
enables one the test ideas regarding the mechanisms of these interactions [9].
This paper presents an immunosensor for determination of human
properdin B factor. Properdin B factor is related to some diseases, such as diabetes,
pregnancy induced hypertension (PIH), elbow synovitis, pulmonary inflammation and host
defense . The determination of Bf is useful for the diagnosis of these diseases. The
original methods of determination of Bf are ELISA, SDS-PAGE and Western blot. All of these
methods are too time-consuming, but SPR method present now is rapid response, simplicity
and cost effectiveness.
1. EXPERIMENTAL
1.1 Materials
Staphylococcal protein A (SPA), Anti-human Bf antiserum (rabbit IgG fraction to human Bf,
1:40), Human Bf, Human fibrin, Human serum albumin, Human g-globulin were purchased from Shanghai Biology Product Research
Institute. All other chemicals were of analytical reagent grade. Bovine serum albumin
(BSA) was purchased from Sino-American Biotechnology Company (SABC). All
solutions were prepared with ultrapure water (>18 MW·cm) supplied by an
EASYpure RF compact ultrapure water system (Barrnstead Thermdyme Inc., USA).
The following were employed:0.01 mol/L phosphate-buffered saline
solution (PBS, pH=7.4): 0.2 g KCL, 8.0 g NaCl, 0.24 g KH2PO4, 1.44 g
NaHPO4, dissolved in 1000 mL ultrapure water. 0.3 mol/L citrate buffer
(pH=2.7): 21 g C6H8O7·H20,
11 g Na2HPO4 dissolved in 350 mL ultrapure water. 10 mg/mL BSA
solution: 100 mg BSA dissolved in 10 mL 0.01 mol/L PBS buffer (pH=7.4).
1.2 Apparatus
The schematic diagram of SPR optical immunosensor is shown in Figure 1. The light source
is a halogen tungsten lamp in conjunction with a constant voltage transformer. The light
from this source passes through a polarizer and becomes TM polarized light. In order to
make the light be parallel one, two lens are employed. The focal length of lens 1 and lens
2 are 30 mm and the working distance from lens 1 to lens 2 is 60 mm. Then the output light
from the prism is guided into the optical fiber and into the Fullwave Spectrophotometer
(Ocean Optics, Inc. USA). The focal length of lens 3 is 75 mm and the distance from lens 3
to optical fiber is 75 mm. The prism was mounted in an electron-beam evaporator system in
the arrangement that the flux of evaporated metal was perpendicular to the prism. The
deposition process was monitored using quartz crystal detector. The prism was
vacuum-deposited with a 50 nm gold film. The incident angle is fixed at a suitable angle
to ensure the surface plasmon resonance phenomenon to occur. The reflected light intensity
is the minimum at the resonant wavelength, a smaller increase in refractive index of the
analyzed solution would cause a clear shift in SPR reflected spectra towards longer
resonant wavelength [5-7]. A small volume flow cell (0.2 mL) was used for
reaction.
Fig.1 Schematic
diagram of the optical immunosensor based on surface plasmon resonance
1.3 Self-assembled monolayers
SPA was dissolved in 0.01 mol/L phosphate-buffered saline solution (PBS, pH=7.4) to a
final concentration of 0.1 mg/mL. SPA was injected into the flow cell. The process of the
monolayer formation was monitored through the computer. After the SPA monolayer formation,
10 mL of 0.01 mol/L PBS (pH=7.4) was used to wash off the unbound SPA for several times.
10 mg/mL of BSA solution was used to block the nonspecific protein binding sites.
The diluted anti-human Bf antiserum (1:10) was injected into the flow
cell after the SPA monolayer formation. The process of the monolayer formation was
monitored through the computer. The unbound antibody was rinsed several times with PBS.
All the analyses were performed at room temperature, 20±1°C.
2 RESULTS AND DISCUSSION
2.1 Monolayer formation
The assembly of self-assembling monolayer (SAM) on gold surface is relatively easy and the
molecules are closely packed, well ordered and stable in ambient conditions. Thus, SAM is
a good technique for immobilizing biomolecules on gold or sensing membrane. This technique
depends mostly on the use of the thiol-gold interaction [11]. It was found that
SPA can spread over the air/water interface to form a homogeneous monolayer. SPA is a
polypeptide isolated from Staphylococcus aureus that binds specifically to the Fc region
of immunoglobulin molecules without interacting at the antigen binding sites. This
property permits the formation of tertiary complexes, consisting of SPA, antibody and
antigen . SPA binds the Fc region of antibodies at near neutral pH, leaving the
antigen-binding sites free. The bound antibodies can be eluted from SPA at lower pH [10].
SPA is also easily attached to gold. The immobilization of SPA is a chemisorption
[11-13]. Gold-SPA complexes are highly stable, having an association constant of 108
mol-1[12]. In this paper, we take the advantage of these phenomena to develop
SPR sensor membrane having SPA bound to the gold on a prism surface (Fig. 2).
|
Fig.2 Composition of the
self-assembling monolayer |
Fig.3 The adsorption curve of 0.1 mg/mL SPA
on gold film |
To observe the SPA
assembling on gold substrate, SPA was injected into the flow cell. The changes of resonant
wavelength were determined in real time. The adsorption curve of SPA at the surface of
gold was shown as Figure 3. The shift of the resonant wavelength reaches about 99% of its
total shift within 0.5 h in the 0.1 mg/L SPA solution. Further increasing the assembling
time, the l keeps almost constant. That means the self-assembly has finished and the
monolayer formed. The large excess of BSA (10 mg/mL, final concentration) blocked the
nonspecific binding sites on the sensor surface.
Then the antiserum of Bf was injected in to the flow cell. The antibody
assembling was monitored again in real time. SPR spectra of Bf antiserum on the surface of
SPA monolayer at different time were shown as Figure 4. The best dilution titer of Bf
antiserum is 1 to 10. If the antibody titer is too low that the change of the resonant
wavelength will be too small. After the antibody titer comes up to 1:10, the change of the
resonant wavelength is stable at 10.19 nm. The sensor surface is well assembled under this
antibody titer. The antiserum was diluted with PBS at different pH values. The result
showed that the antibody assembling was good in the pH range from 7.0 to 7.6. The protein
activity is higher in this range. In this study, pH 7.4 was used. Figure 5 shows the time
dependence of SPA monolayer assembled with rabbit anti-human B factor antibody at 20°C.
The assembling of the antibody (1:10) was carried out for 2 h to organize the processing
antibody molecular on the solid surface and such the sensor membrane was stable.
|
Fig.4 SPR
spectra of B factor antibody on the surface of SPA monolayer at different combination
time(a. 1min.; b. 20min ) |
|
Fig.5 The
adsorption curve of B factor antibody on the SPA monolayer B factor antiserum titer (1:10) |
2.2 Determination of Bf
B factor plays a crucial role in the activation of the alternative pathway of complement
on the surfaces of biomaterials during extracorporeal procedures. It influences the host
defense ability. A series of standard Bf solutions was prepared. Figure 6 shows the
relationship between Bf concentration and resonance wavelength change. The determination
conditions for this sensor are in PBS buffer (pH 7.4) at 20ºC for
30 min. Repeating the determination of 4 µg/mL Bf solution for 11 times, a relative
standard deviation of 1.6% is obtained.
|
Fig.6 The relationship
between B factor concentration and resonance wavelength change |
2.3 Selectivity
Under the optimum conditions, 3.5 mg/mL human serum albumin (HSA) and 1.0 mg/mL human g-globulin, 5 µg/mL human fibrin were
added to flow cell. It was found that these components of human serum did not interfere
the determination of Bf when Bf concentration is over 5µg/mL. The immunosensor has good
selectivity.
2.4 Regeneration
The bound antibody can be eluted from SPA with acidic solution [14,15]. After
being rinsed with 0.3 mol/L citrate buffer (pH=2.7), the sensor can be used repeatedly.
The antigen-antibody complex is eluted from the SPA monolayer within 10 minutes.
2.5 Improvement
In this study, an ordinary spectrograph is used, the sensitivity is limited by its
spectral resolution (0.54 nm). Improving the spectral resolution can make a more sensitive
sensor.
ACKNOWLEDGMENTS
The authors wish to thank the National Natural Science Foundation of China and Research
Laboratory of SEDC of Analytical Science for Material and Life Chemistry, Xiamen
University for funding this research.
REFERENCES
[1] Liedberg B, Nylander C, Lundstrom I. Sensors Actuators, 1983, 4: 299-304.
[2] Nilsson P, Persson B, Uhlen M et al. Anal. Biochem., 1995, 224: 400-408.
[3] Liedberg B, Yang Z, Engquist I et al. J. Phys. Chem. B., 1997, 101: 5951-5954.
[4] Jorgenson R C, Yee S S, Johnston K S et al. SPIE., 1993, 188: 35-48.
[5] Zhao X J, Wang Z, Xu H Y et al. Chem. J. Chinese
Universities (Gaodeng Xuexiao Huaxue Xuebao), 1998, 19 (8):
1214-1218.
[6] Zhao X J, Wang Z, Liang F et al. Chinese J. Anal. Chem. (Fenxi Huaxue)., 1998, 26:
1320-1323.
[7] Zhao X J, Mu Y, Wang Z et al. Chem. J. Chinese Universities (Gaodeng Xuexiao Huaxue
Xuebao), 1999, 20 (5): 704-708.
[8] Silin V, Weetall H, Vanderah D J. J. Colloid Interface Sci., 1997, 185: 94-103.
[9] Fredrikson G N, Truedsson L, Sjoholm A G. J. Immunol Methods., 1993, 166 (2):
263-270.
[10] Duan C, Meyerhoff M E. Anal. Chem., 1994, 66: 1369-1377.
[11] Owaku K, Goto M, Ikariyama Y et al. Anal. Chem., 1995, 67: 1613-1616.
[12] Ren X Z, Miller J N. Anal. Lett., 1994, 27: 1067-1074.
[13] Davis K A, Leary T R. Anal. Chem., 1989, 61: 1227-1230.
[14] Attili B S, Suleiman A A. Anal. Lett., 1995, 28: 2149-2159.
[15] Prusak S E, Luong J H. Anal. Lett., 1990, 23: 183-194.
|