Biomacromolecules 2008, 9, 403–407
403
Bioconjugated Phospholipid Polymer Biointerface for Enzyme-Linked Immunosorbent Assay Kazuki Nishizawa, Tomohiro Konno, Madoka Takai, and Kazuhiko Ishihara* Department of Materials Engineering and Center for NanoBio Integration, The University of Tokyo, 7-3-1, Hongo, Bunkyo-ku, Tokyo, 113-8656, Japan Received September 13, 2007; Revised Manuscript Received November 2, 2007
This study aimed to develop a sensitive and reliable immunoassay by applying a highly functional phospholipid polymer biointerface. We synthesized a phospholipid polymer—poly[2-methacryloyloxyethyl phosphorylcholine (MPC)-co-n-butyl methacrylate (BMA)-co-p-nitrophenyloxycarbonyl poly(ethylene glycol) methacrylate (MEONP)] (PMBN). MEONP contains active ester groups on the side chains for immobilization of antibodies via oxyethylene. PMBN with different compositions and oxyethylene chain lengths were synthesized; their effects on nonspecific and specific values in the immunoassay were evaluated. MPC units reduce the background by preventing nonspecific protein adsorption. MEONP units could conjugate antibodies and enhance the specific signal. The specific signal was independent of the oxyethylene chain length, but long oxyethylene chains increased the background. Specific signals corresponding to the antigen were observed with the PMBN coating, and a liner standard curve was obtained. The PMBN-coated surface maintained residual activity after long-term storage. This surface affords a low background without requiring blocking treatment and is suitable for immobilized antibodies.
1. Introduction Immunoassays are currently an indispensable analytical method in the fields of biochemistry and clinical diagnosis. Among several immunoassays, enzyme-linked immunosorbent assay (ELISA) with a relatively high sensitivity has been the most widely used method1,2 for clinical examination. However, the nonspecific adsorption of proteins on the solid phase or denaturation of immobilized antibodies leads to a decrease in its reliability and sensitivity. It is therefore necessary to achieve highly reliable and extremely sensitive ELISA systems not only for quantifying a minute amount of hormone or tumor marker for accurate diagnosis but also for understanding and elucidating the pathophysiology of diseases. A major cause of an unreliable assay is the nonspecific adsorption of an analyte or labeled antibody on a primary antibody-immobilized solid phase, protein-based blocking reagents such as bovine serum albumin (BSA), and casein that are commonly used in laboratories worldwide. However, protein-based blocking reagents denaturate easily and transform their properties after a long preservation period. Further, cross reactions between detection reagents and blocking reagents persist as one of the main causes of a high background and low signal-to-noise ratio. On the basis of the concept of proteinfree blocking reagent, several types of artificial blocking reagents such as Tween203 and poly(ethylene glycol)4,5 have been studied and reported. The 2-methacryloyloxyethyl phosphorylcholine (MPC) polymer, which is based on the surface structure of cell membranes with phosphorylcholine groups, is reported as an excellent blocking reagent for ELISA.6,7 To develop a highly sensitive ELISA system, both enhancement of specific signals and reduction of nonspecific signals should be achieved. The following factors are important for the enhancement of specific signals: (1) enhancement of the antigen–antibody reaction efficiency8–11 and (2) maintenance of the activity of * To whom correspondence may be addressed. Tel: +81-3-5841-7124. Fax: +81-3-5841-8647. E-mail:
[email protected].
Figure 1. The chemical structure of PMBN.
immobilized antibodies on the solid surface.6,12 In fact, the use of biotinylated antibodies and streptavidin application are wellknown methods for amplifying signals to enhance the antigen–antibody reaction efficiency.8,9 It was reported that when a primary antibody is supported by some ligand via a spacer, the reaction efficiency is enhanced due to the high mobility of the antibody.10,11 The immobilization methods of biomolecules are an important issue not only for ELISA but for developing highly sensitive enzyme, DNA, and other protein sensing devices.13,14 In the case of ELISA, the denaturation of the physically adsorbed primary antibodies onto a solid surface, which is a significant problem concerning the immunoreactivity and reliability of the assay,15 may be ignored. Our group evaluated the residual primary antibody activity by performing a long-term storage test using an artificial blocking reagent of the MPC polymer.6,12 The MPC polymer exhibits a high resistivity to protein adsorption and can effectively decrease the denaturation of biomolecules because of its high number of free water-binding sites in the phosphorylcholine group.16 From these results, it is suggested that the MPC polymer biointerface may possibly enhance the specific signal as well as decrease the background. In this study, we developed a solid biointerface for ELISA by using an MPC polymer, namely, poly[MPC-co-n-butyl methacrylate (BMA)-co-p-nitrophenyloxycarbonyl poly(ethylene glycol) methacrylate (MEONP)] (PMBN). Figure 1 shows the chemical structure of PMBN. The MEONP unit contains active ester groups for the conjugation of antibodies via the oxyethylene chain. The PMBN comprises units to prevent nonspecific
10.1021/bm7010246 CCC: $40.75 2008 American Chemical Society Published on Web 12/20/2007
404 Biomacromolecules, Vol. 9, No. 1, 2008
adsorption as well as to conjugate antibodies in a single polymer chain. Furthermore, this polymer is expected to serve as a suitable material for easy construction of a biointerface in heterogeneous ELISA that does not require a blocking step. Additionally, chemical binding via the oxyethylene chain in the case of PMBN can enhance the efficiency of an antigen–antibody reaction. In this report, PMBN with various compositions and oxyethylene chain lengths were synthesized, and their effects on nonspecific and specific signals were investigated. Furthermore, the validity of PMBN for ELISA was discussed by comparing it with the conventional ELISA process that uses a protein-blocking buffer.
2. Experimental Section 2.1. Synthesis of PMBN for Bioconjugation. PMBN, the phospholipid polymer for bioconjugation, was prepared using MPC, BMA, and MEONP. MPC and MEONP were synthesized as described previously in detail.17–19 BMA was purchased from Nacalai Tesque, Inc. (Kyoto, Japan). PMBN was synthesized by conventional radical polymerization of MPC, BMA, and MEONP by using R,R′-azobisisobutyronitrile (AIBN) as an initiator. After polymerization, the reaction mixture was precipitated using a mixture of chloroform and diethyl ether (2:8) as a solvent. The ratio of monomer unit composition in PMBN was determined by 1H-NMR (JEOL JNM-NR30, Tokyo, Japan). The molecular weight of PMBN was evaluated by gelpermeation chromatography (OHpak SB-804 HQ column, Shodex, Tokyo, Japan) with poly(ethylene oxide) (PEO) standards. 2.2. Coating with Bioconjugated Phospholipid Polymer. Coating was performed using a solvent evaporation method. A volume of 300 µL of 0.2 wt % PMBN in ethanol was pipetted into a 96-well polystyrene (PS) microtiter plate (Nunc Maxisorp F8, Nunc, New York) for 10 min. The PMBN solution was removed by inverting the microtiter plate, and the sample was put in the same solvent vapor atmosphere at room temperature for a night and then dried under vacuum overnight. Surface analysis of the PMBN-coated surface was carried out using X-ray photoelectron spectroscopy (XPS; AXIS-Hsi, Shimadzu/Kratos, Kyoto, Japan). The takeoff angle of the photoelectrons was 90°. 2.3. Conjugation and Immobilization of Antibody. The wells of the microtiter plate were coated with each type of PMBN solution, as described above. A volume of 150 µL of 10 µg/mL anti[human thyroid stimulating hormone (hTSH)] mouse IgG (Bioclone Australia Pty Ltd., Sydney, Australia) in phosphate buffer solution (pH 8) was pipetted into the wells of the microtiter plate and allowed to react with the active esters of PMBN for 24 h at 25 °C. After incubation, the wells were washed three times with phosphate-buffered saline (PBS; pH 7.1). To compare the assay performed using the PMBN-coated surface and the conventional method that uses BSA as a blocking reagent, the following solid phases were prepared. For preparation of the BSAblocked and uncoated surfaces, 150 µL of 10 µg/mL anti-hTSH mouse IgG in phosphate buffer solution (pH 8) was pipetted into the wells of a naked microtiter plate and the antibodies were allowed to immobilize directly on the microtiter plate. After incubation for 24 h at 4 °C, the wells were washed three times with PBS. Next, 300 µL of 1 wt % BSA (Sigma-Aldrich, Corp., St. Louis, MO) in PBS solution for BSAblocked surface and PBS solution for uncoated surface were added into separate wells. After incubation for 24 h at 4 °C, the wells were again washed three times with PBS. 2.4. Immunoassay with Various Types of PMBN and Solid Surfaces. To confirm the effect of different compositions of PMBN on ELISA, the wells were coated with PMBN of various compositions. BSA-blocked and uncoated surfaces were also used in this assay to compare conventional ELISA with ELISA using the PMBN-coated surface. The following ELISA was carried out. To perform the antigen–antibody reaction, 150 µL of 0–10 µIU/mL hTSH (Biogenesis Ltd., England, UK) in PBS solution was pipetted into each well. After incubation for 2 h at 25 °C, the wells were washed
Nishizawa et al. six times with PBS solution containing 0.1 wt % Tween20. Next, 150 µL of 0.03 µg/mL anti-hTSH IgG biotinylated (Bioclone Australia Pty Ltd.) in PBS solution containing 1 wt % BSA was pipetted into the wells. After incubation for 1 h at 25 °C, the wells were washed six times with PBS solution containing 0.1 wt % Tween20. Subsequently, 150 µL of 0.16 µg/mL streptavidin-horseradish peroxidase (HRP) (Zymed Laboratories Inc., CA) in PBS solution containing 1 wt % BSA was pipetted into the wells. After incubation for 10 min at 25 °C, the wells were washed six times with PBS solution containing 0.1 wt % Tween20. Next, 100 µL of tetramethylbenzidine solution (SUMILON peroxidase chromogenic substrate T, Sumitomo Bakelite Co., Ltd., Tokyo, Japan) was pipetted into the wells as a substrate for HRP and incubated for 20 min at 25 °C. After incubation, 100 µL of 2 N H2SO4 solution was added into the wells, and the absorbance at 450 and 620 nm was measured using a multilabel counter (Wallac ARVOsx1420, Perkin-Elmer. Inc. Waltham, USA). 2.5. Assay for Testing the Stability of the Immobilized Antibodies on PMBN. In order to confirm the stability of antibodies immobilized on the solid surfaces, the following assay was carried out. The PMBN-coated microtiter plate was prepared as described above: 150 µL of 10 µg/mL anti-hTSH IgG in phosphate buffer solution (pH 8) was pipetted into the wells of the microtiter plate and allowed to react with the active ester of PMBN for 24 h at 25 °C. After incubation, the wells were washed three times with PBS. To compare the stability of the antibodies, we also used BSA as a conventional blocking reagent and the uncoated surface. To obtain BSA-blocked and uncoated surfaces, 150 µL of 10 µg/mL anti-hTSH IgG in phosphate buffer solution (pH 8) was pipetted into the wells of the microtiter plate and the antibodies were allowed to immobilize on the microtiter plate. After incubation for 24 h at 25 °C, the wells were washed three times with PBS, and then 300 µL of 1 wt % BSA in PBS solution for obtaining the BSA-coated surface and PBS solution for obtaining the uncoated surface were added into the wells. After incubation for 24 h at 4 °C, the wells were washed three times with PBS. These surfaces (PMBN-coated, BSA-blocked, and uncoated) with immobilized antibodies were dried, sealed with an adhesive strip, and stored at 37 °C. The antigen–antibody reaction was carried out after incubation for 5, 10, and 15 days by using the same protocol as described in section 2.4.
3. Results and Discussion 3.1. Synthesis Results and Surface Analysis of PMBN. The synthesis results for PMBN are shown in Table 1. The monomer unit composition was quantitatively related to the feeding ratio. The solubility of the polymer in water depends on the MPC unit composition. PMBN containing more than 40 unit mol % of MPC (PMBN451-n2 and PMBN631-n2) could be dissolved in water; however, PMBN containing less than 20 unit mol % of MPC could not be dissolved in water. All PMBN synthesized in this experiment could be dissolved in ethanol; therefore, PMBN-ethanol solution was used for the coating process. Surface analysis of PMBN262-n2 that was coated on the PS microtiter plate was carried out using XPS. The XPS spectra of carbon (C 1s), oxygen (O 1s), phosphorus (P 2p), and nitrogen (N 1s) are shown in Figure 2. The C 1s spectrum was attributed to the C–N (286.3 eV), C–O (286.7 eV), and CdO (288.8 eV) components from the MPC unit. The spectrum was also attributed to the aromatic carbons based on the p-nitrophenyl ester group (284.5 eV, 285.8 eV for binding to nitro group, and 286.3 eV for ester group). Furthermore, a phosphorus peak and a nitrogen peak attributed to the phosphorylcholine group were also observed. These results indicated that the surface of the PMBN-coated PS microtiter plate was covered with the phosphorylcholine groups of the MPC unit and with the p-nitrophenyl ester groups of the MEONP unit.
A Solid Biointerface for ELISA
Biomacromolecules, Vol. 9, No. 1, 2008 405
Table 1. Synthetic Result of PMBNa monomer unit composition (mol %) in compositionc
in feed b
solubilitye
sample no.
n
MPC
BMA
MEONP
MPC
BMA
MEONP
PMBN631-n2 PMBN451-n2 PMBN271-n2 PMBN262-n2 PMBN172-n2 PMBN262-n4 PMBN262-n8
2 2 2 2 2 4 8
60 40 20 20 10 20 20
30 50 70 60 70 60 60
10 10 10 20 20 20 20
55 38 18 22 16 19
42 59 78 65 74 72
3 3 4 13 10 9
-4
10
Mw
d
4.8 4.3 3.0 3.0 1.6 3.4 2.7
% yield
in water
in ethanol
77 80 72 58 34 72 57
+ + -
+ + + + + + +
a [monomer] ) 1.0 mol L-1; [AIBN] ) 10 mmol L-1 in ethanol; temperature, 60 °C; reaction time, 6 h. b Number of repeating units in the oxyethylene chain of MEONP. c Determined by 1H-NMR spectrum. d Determined by GPC in water/methanol ) 3:7, PEO standards. e Solubility was determined with 10 mg mL-1 of each polymer sample and described as soluble (+) and insoluble (-) at 25 °C.
Figure 2. X-ray photoelectron spectra of the PMBN-coated PS surface.
3.2. The Effect of Different Compositions of PMBN on ELISA. After PMBN was coated on the microtiter plate, primary antibodies were immobilized and ELISA was carried out. Figure 3 shows the results of ELISA performed using wells coated with PMBN of various compositions. The oxyethylene chain of MEONP in the PMBN used in this test contained two repeating units. The MPC unit composition affected the background (the absorbance value at hTSH ) 0 µIU/mL) as shown in Figure 3a. An increase in the proportion of the MPC unit decreased the background levels. It was considered that hydrophobic units in PMBN could adsorb on the PS microtiter plate and that the MPC units reduced the nonspecific adsorption of labeled IgG. It was also clear that the PMBN-coated layer persisted throughout the assay without detaching from the PS surface even when water-soluble PMBN was used. On the other hand, Figure 3b indicates that an increase in the proportion of the MEONP unit enhances the specific signal (the absorbance value at hTSH ) 10 µIU/mL). Due to the increase in the binding sites for primary antibodies, the amount of immobilized antibodies increased, leading to the enhancement of the specific signal. This result indicates that the active ester group in the MEONP unit could react with the amino group of the antibody, resulting in antibody immobilization on the surface. From these results, bioconjugated phospholipid polymer was successfully synthesized. It was confirmed that the MEONP unit could conjugate antibodies and enhance the specific signal and that the MPC unit could reduce the background by preventing the nonspecific adsorption of proteins. A schematic image of antibody-conjugated PMBN surface is shown in Figure 4. We presumed that the residual active ester groups in PMBN would affect the background level because the labeled antibodies
Figure 3. Absorbance values (450–620 nm) obtained in ELISA performed using various PMBN polymers. (a) Changes in the composition of the MPC unit (PMBN631-n2, PMBN451-n2, and PMBN271-n2). (b) Changes in the composition of the MEONP unit (PMBN271-n2, PMBN262-n2, and PMBN172-n2); there were two repeating units in the oxyethylene chain of MEONP. The squares and circles indicate the absorbance values of hTSH ) 10 and 0 µIU/mL, respectively.
can react with the active ester groups and immobilize on the solid phase. To confirm the effect of the residual active ester groups on ELISA, 0.01 M aminoethanol in phosphate buffer solution (pH 8) was pipetted into the wells and incubated for 24 h at 4 °C and allowed to hydrolyze the residual active ester groups after the immobilization of the primary antibodies on the PMBN. The results of ELISA performed using the hydrolyzed PMBN surface revealed that the background level was almost same as with the nonhydrolyzed PMBN surface, but the specific signal had decreased (data not shown). This result indicates that labeled antibodies cannot access the residual active ester groups because the already immobilized primary antibodies and the residual active ester groups exist only in a small space between the primary antibodies or inside the coated polymer.
406 Biomacromolecules, Vol. 9, No. 1, 2008
Nishizawa et al.
Figure 4. Schematic illustration of the PMBN surface antibodies that were immobilized via an oxyethylene chain.
Figure 6. Standard curves of hTSH obtained by ELISA. Absorbance values (450–620 nm) obtained in ELISA were measured. The squares, circles, and triangles indicate the results with PMBN262-n2 coating, no coating, and BSA blocking, respectively. Each symbol indicates the mean of three samples.
Figure 5. Absorbance values (450–620 nm) obtained in ELISA performed using various PMBN polymers. Changes in the number of repeating units in the oxyethylene chain of the PMBN polymers. The sample used was PMBN262-n2, PMBN262-n4, and PMBN262n8. The squares and circles indicate the absorbance values of hTSH ) 10 and 0 µIU/mL, respectively.
The decreased specific signal indicated lowering of the activity of the immobilized primary antibodies during the 24 h incubation. On the basis of this result, we believe that the residual active ester group does not have an adverse effect, and the process of hydrolysis is not required. Accordingly, hydrolysis was not carried out in ELISA presented in Figure 3 and the subsequent experiments. Figure 5 shows the results of ELISA performed using various oxyethylene chain lengths of PMBN. The samples used in this assay were PMBN262-n2, PMBN262-n4, and PMBN262-n8 (the number beside n indicates the number of repeating units in the oxyethylene of the MEONP). The specific signal was independent of the oxyethylene chain length, but a long oxyethylene chain increased the background by the labeled IgG. It was considered that the shortest oxyethylene chain is adequate for efficient antigen–antibody reaction, and the extremely long oxyethylene chain increased the nonspecific binding site. From these results, PMBN262-n2 is the best composition of PMBN and was used for the subsequent experiments. 3.3. Standard Curves in ELISA Performed Using Various Solid Surfaces. The result of ELISA with an antigen concentration ranging from 0 to 10 µIU/mL is shown in Figure 6. To compare the PMBN ELISA method with the conventional ELISA method, BSA was used as a blocking reagent. With regard to nonspecific signals (hTSH ) 0 µIU/mL), PMBN262n2-coated and BSA-blocked surfaces exhibited a lower absorbance than the uncoated surface. Without a blocking treatment, PMBN262-n2-coated surface reduced the nonspecific adsorption of the labeled IgG to levels similar to that achieved by BSA blocking. Specific signals corresponding to hTSH were observed with PMBN262-n2 coating, and a linear standard curve was obtained.
Figure 7. Long-term stability of the immobilized antibodies on various surfaces. Absorbance values (450–620 nm) obtained in ELISA were measured after the immobilized antibodies were stored at 37 °C under dry conditions. The residual activity was calculated as (absorbance values of hTSH ) 10 µIU/mL) - (absorbance values of hTSH ) 0 µIU/mL). The squares, circles, and triangles indicate the results with PMBN262-n2 coating, no coating, and BSA blocking, respectively. Each symbol indicates the mean of three samples.
3.4. Long-Term Stability of Immobilized Antibodies on PMBN. In order to confirm the stability of the antibodies immobilized on these PMBN surfaces, they were incubated under dry conditions at 37 °C for several days, and the assay was then carried out as an accelerated test. The results are summarized in Figure 7. The surfaces coated with BSA as a blocking reagent and the uncoated surface exhibited a low stability of the immobilized antibodies. Their residual activities decreased drastically in 5 days. This indicates that the physical adsorption of the antibody on the PS microtiter plate easily induces the denaturation of the antibody. When compared with BSA-blocked surface, the PMBN-coated surface had maintained 50% of the residual activity even after 15 days. This indicates that the PMBN surface provides a stable condition for the antibodies due to the presence of the MPC unit and the chemical binding to the polymer via the oxyethylene chain. The reduction in the denaturation of immobilized antibodies on the PMBN surface results in a high specific signal and realizes a high signalto-noise ratio. Additionally, a long-term stability of immobilized antibodies is required for heterogeneous immunoassays used for diagnosis.
A Solid Biointerface for ELISA
A commercial ELISA kit is freeze-dried after primary antibody immobilization on the plate, and the kit is used after storage in a refrigerator. The denaturation of the immobilized primary antibodies on the stored plate is a significant problem with regard to reliability of the assay. Therefore, the PMBN surface affords great advantages in terms of the practical use of the immunoassay.
Biomacromolecules, Vol. 9, No. 1, 2008 407
References and Notes (1) (2) (3) (4) (5) (6)
4. Conclusion PMBN, the bioconjugated phospholipid polymer, was successfully synthesized for ELISA. PMBN polymers of different compositions and oxyethylene chain lengths were synthesized, and their effects on nonspecific and specific values in the immunoassay were evaluated. The increase in the proportion of the MPC unit decreased the background levels, and the increase in the proportion of the MEONP unit enhanced the specific signal. Therefore, it was confirmed that the MEONP unit can conjugate antibodies and enhance the specific signal, and the MPC unit can reduce the background level by preventing the nonspecific adsorption of proteins. The shortest oxyethylene chain length was suitable for decreasing the background. A linear standard curve corresponding to the antigen was obtained. The PMBN-coated surface showed a low background without a blocking treatment and maintained the activity of the immobilized antibodies. A PMBN surface showing low background and maintaining antibody activity is a basic bioconjugated polymer biointerface for developing high sensitive, reliable biosensing devices not only for ELISA.
(7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19)
Farr, A. G.; Nakane, P. K. J. Immunol. Method 1981, 47, 129–144. Orci, L. Diabetologia 1985, 28, 528–546. Engvall, E.; Perlmann, P. J. Immunol. 1972, 109, 129–135. Nagasaki, Y.; Kobayashi, H.; Katsuyama, Y.; Jomura, T.; Sakura, T. J. Colloid Interface Sci. 2007, 309, 524–530. Uchida, K.; Otsuka, H.; Kaneko, M.; Kataoka, K.; Nagasaki, Y. Anal. Chem. 2005, 77, 1075–1080. Sakaki, S.; Nakabayashi, N.; Ishihara, K. J. Biomed. Mater. Res. 1999, 47, 523–528. Sakaki, S.; Iwasaki, Y.; Nakabayashi, N.; Ishihara, K. Polym. J. 2000, 32, 637–641. Guesdon, J.; Ternyck, T.; Avrameas, S. J. Histochem. Cytochem. 1979, 27, 1131–1139. Alevy, Y. G.; Blyrm, C. M. J. Immunol. Methods 1986, 87, 273–281. Weimer, B. C.; Walsh, M. K.; Wang, X. J. Biochem. Biophys. Methods 2000, 45, 211–219. Yakovleva, J.; Davidsson, R.; Lobanova, A.; Bengtsson, M.; Eremin, S.; Laurell, T.; Emneus, J. Anal. Chem. 2002, 74, 2994–3004. Park, J.; Kurosawa, S.; Takai, M.; Ishihara, K. Colloids Surf., B 2007, 55, 164–172. Hervas` Per`ez, J. P.; Lop`ez-Cabarcos, E.; Lop`ez-Ruiz, B. Biomol. Eng. 2006, 23, 233–245. Kusnezow, W.; Hoheisel, J. D. J. Mol. Recognit. 2003, 16, 165–176. Butler, J. E. Methods 2000, 22, 4–23. Ishihara, K.; Iwasaki, Y. J. Biomed. Mater. Res. 1998, 39, 323–330. Ishihara, K.; Ueda, T.; Nakabayashi, N. Polym. J. 1990, 22, 355– 360. Konno, T.; Watanabe, J.; Ishihara, K. Biomacromolecules 2004, 5, 342–347. Park, J.; Kurosawa, S.; Watanabe, J.; Ishihara, K. Anal. Chem. 2004, 76, 2649–2655.
BM7010246
