Modifying the Surface Chemistry of Silica Nano-Shells for Immunoassays

Journal of Young Investigators
Undergraduate, Peer-Reviewed Science Journal

Volume Six

RESEARCH ARTICLE

RECENT ISSUES | ARCHIVES | RESOURCES | JYI NEWS | ABOUT JYI

Issue 1, July 2002

Engineering & Applied Sciences
Modifying the Surface Chemistry of Silica Nano-Shells for Immunoassays

Cintyu Wong, Joel P. Burgess
University of Rochester
Advisor: Agnes E. Ostafin, Ph.D.
Department of Chemical Engineering, Northwestern University

Abstract

Silica shells are useful in many applications, such as sensitive optical interferometric biosensors and drug delivery. In this experiment, the binding of sheep anti-cytochrome C to silica gel and silica nano-shells through various linkers was investigated. Treatments of linkers included (1) silanisation using 3-aminopropyl trimethoxy silane (APS), (2) silanisation with APS followed by the cross-linker N-5-azido-2-nitrobenzoyloxysuccinimide (ANB-NOS), and (3) no silanisation. A two-step sandwich immunoassay was used to measure the extent of antibody attachment. Antibody activity was determined by assaying the proportion of attached antibody that bound antigen. For the silica nano-shells, levels of 0.5% specific binding for APS and 1.5% specific binding for ANB-NOS were detected. Approximately 10% of the bound antibodies on the APS and 12% on the ANB-NOS surface were active. However, the APS surface bound more antibodies on silica gel. For silica gel, there was 10% specific binding for APS and 3.5% specific binding for ANB-NOS-coated surfaces. About 12% of the bound antibodies on the APS and only 1% on the ANB-NOS-coated surfaces were active. ANB-NOS was found to be the preferred linker for attachment of antibody to silica nano-shells. Interestingly, the curvatives of the surfaces may partially explain the results. These results are important because the covalent linkage of antibodies to silica nano-shells may someday be used for targeted drug delivery.

Introduction

The coupling of particles to immunoreagents is fairly common and has many applications. Current research, however, is exploring many novel immunoreagent-particle systems. For example, Hidalgo-Alvarez's group has studied the stability of polystyrene particles when coated with antibodies (Davalos-Pantoja 2001). Another group has studied the binding of antibodies to latex microspheres and used these to detect antigen in the lachrimal fluid of cattle (Bruning 1999). One new, particularly important particle system is the silica nano-shell.

Silica nano-shells can be fabricated by synthesizing a layer of silica on tiny spheres of colloidal gold. With appropriate reagents, the gold interior can be dissolved away, leaving a hollow silica shell. Antibodies can be bound to the silica shell, and the silica shell-antibody complex can be used to bind to a specific antigen in fluid systems.

Many studies have examined antibody attachment to silica surfaces - for example, glass slides (Lin 1988, Nashat 1998) and silica gel (Nilsson 1989). However, attaching antibodies onto silica nano-shells has received little attention despite its numerous applications. One of the most important of these is drug delivery to specific, targeted locations in the body. Since these shells are hollow, they can be filled with substances (e.g., enzymes and drugs). Targeted drug delivery is then achieved by binding of the antibody attached to the shell to a specific antigen on a particular population of cells. This would be valuable in medical diagnostics and in the future development of medicinal treatments.

The success of targeted drug delivery is contingent upon two issues: (1) sufficient attachment of antibody to the silica surface and (2) the ability of the antibodies to bind their specific antigen once they are attached to the shell.

The first of these issues can be resolved with the use of organosilanes - compounds with a silanol (-Si-OH) group on one end and an organic reactive functional group on the other. Organosilanes are widely used to introduce an organic functional group to an inorganic surface. For example, silanation is used to reduce the microleakage of the acrylic resin-metal framework (Sharp 2000), and silanized surfaces can increase the binding efficiency of magnetic particles to DNA by 14-fold (Yoza 2002). Silanes are also good chelating ligands because of their reactivity with hydroxyls, and they can couple organic groups to virtually any oxide surface (Nashat 1998).

3-aminopropyl trimethoxy silane (APS) is a widely-used organosilane that can be used to attach antibodies to silica surfaces. However, APS does not contain any reactive functional groups that permit covalent bonding. Since covalent bonds are substantially stronger and more secure than non-covalent bonds, antibody attachment via APS would be less than optimal. A stronger linkage can nonetheless be achieved by attaching a heterobifunctional crosslinker, such as N-5-azido-2-nitrobenzoyloxysuccinimide (ANB-NOS), to APS and then binding that complex to the antibody. ANB-NOS can covalently bind antibody through its aryl azide group following activation with ultraviolet light.

The second of these issues is likely to depend on the first and can serve as the ultimate test of in vitro success. Therefore, the first important point of consideration is effective attachment of antibody to the silica nano-shell.

To investigate whether silica nano-shells are good candidates for targeted drug delivery, the effect of the two different surface modifications on the attachment and specific activity of antibodies was examined. In this study, the antibody anti-cytochrome C was attached via silanation with APS alone or with APS followed by the attachment of the cross-linker ANB-NOS. Binding to the native surface alone served as the control. Previous studies have demonstrated that the attachment of antibodies to native silica surfaces is achievable; however, the antibody undergoes partial denaturation and washes off easily (Bhatia 1989). The experiment was also performed for silica gels, which served as a control for the nano-shells.

Materials & Methods

Synthesis of Silica Nano-Shells

All of the silica nano-shells used in this study were graciously provided by coworker Joel P. Burgess, and were synthesized by the method described in Brown (2000). The shells were synthesized on 60 nm colloidal gold, and their diameter, including silica coating, was about 90 nm.

Determining The Surface Area of Silica Nano-Shells

A sample of 0.5 ml of silica particles was diluted with carbon dioxide-free water to 2-3% weight per volume particles in solution. Hydrochloric acid was used to adjust the particle solution to pH 3.0. After the solution volume was diluted to 2.25 ml, 0.5 g of pure crystalline sodium chloride was added. The solution was mixed by vortexing to dissolve the sodium chloride, and the pH was adjusted to 4.0 with 0.1 N sodium hydroxide. Finally, the solution was titrated to pH 9.0 with 0.1 N sodium hydroxide. The volume of sodium hydroxide used when the mixture remained at pH 9.0 ± 0.05 for 1 min was noted and used to calculate the surface area of the silica particles (Sears 1956 and Iler 1979).

Since not all silica gel particles or nano-shells are identical in size, knowing the surface area of these particles is important in the data analysis of the extent of antibody binding. The titration method used in this study gives a rapid and relatively accurate number for the particle's surface area. The surface area is determined by the equation: A = 26.4(V_t-V_b), where A is the square meter per gram or milliliter determined by the method, V_t is the number of milliliters of 0.1 N sodium hydroxide required to titrate the sample from pH 4.0 to pH. 9.0, and V_b is the titration blank in the absence of silica.

Cleaning of Silica Gel

The size of the silica gel (mesh size 200-400, Sigma) samples was 10 mg, and all reactions were done in Eppendorf tubes. The silica gel samples were treated twice with 2 N hydrochloric acid for 5 min and vortexed at room temperature. The samples were then rinsed three times with deionized water. Silica nano-shells did not require cleaning prior to use.

	Figure 1. Schematic of silanisation of APS to particle surface.

Silanization of Cleaned Silica Gel and Silica Nano Shells

Silica gel and silica nano-shells were treated with solutions of 0.3% and 5.0´10-5% of APS (Aldrich) in water, respectively, for 1 hour each at room temperature. The silica gel required occasional resuspension during treatment. Unreacted APS was removed by three washes with deionized water. The Kaiser test (Kaiser 1970) was used to detect the success of silanization by examining the presence of free amines.

Treatment of Silanized Surface with a Heterobifunctional Crosslinker

ANB-NOS (Pierce) was used as heterobifunctional crosslinker for the antibodies. 500 ml and 200 ml of 1 mM ANB-NOS in pH 8.5 carbonate solution were allowed to react with the APS-treated silica gel and silica nano-shells, respectively, for 2 hours at room temperature. The amount of ANB-NOS added was calculated based on the surface area of silica gel and silica shells, and the time of incubation was found experimentally in order to effectively coat the entire surface with ANB-NOS. ANB-NOS is sensitive to ultraviolet light; therefore, all reactions involving ANB-NOS were done in the dark. The Kaiser test was again used to detect the success of ANB-NOS treatment by confirming the absence of free amines.

Figure 2. Schematic of ANB-NOS treatment to silanized surface.

Kaiser Test for Free Amines (Kaiser 1970)

The sample was washed to remove excess APS or ANB-NOS. The supernatant was removed after centrifugation. Solutions of 50 ml of ninhydrin (0.5 g in 10 ml ethanol), phenol (40 g in 10 ml ethanol), and pyridine (1 ml of 0.001 M potassium cyanide in 49 ml pyridine) were added to the shells. The shell-solution mixture was immersed in boiling water for 2 minutes. A blue color indicated the presence of free amine, whereas a yellow color indicated the absence of free amine.

Immunoassay

The modified silica gel surfaces were saturated with the sheep anti-cytochrome C antibody (Sigma) by incubation at 37°C for 30 minutes. The antibody was attached to the nano-shell surfaces by exposing them to 302 nm light for 5 minutes at room temperature. Unbound antibodies were removed by washing three times with 1X phosphate buffer (pH 7.4) containing 0.05% Tween 20 (Sigma). Nano-shells were then treated with 1 ml of skim milk as blocking reagent at 37°C for 30 minutes. After being washed three times with phosphate buffer containing Tween 20, the particles were incubated with donkey anti-sheep IgG conjugated with alkaline phosphatase (Sigma) at 37°C for 30 minutes. Excess antibodies were removed by washing three times with phosphate buffer containing Tween 20. The colorimetric assay was developed with 300 m of 0.1 M borate buffer (pH 9.0, Fisher Chemical) and 60 m of 0.01 M p-nitrophenyl-phosphate disodium hexahydrate (PNPP, Sigma) and incubated at 37°C for 75 min. A₄₀₅ was measured using a Varian Carry UV-VIS Spectrophotometer.

The amount of non-specific binding for the capture antibody was determined by the amount of antibody which bound the control ("naked") silica shells or gel. The level of non-specific binding for the secondary antibody was determined by the amount of antibody bound on a control (APS-treated but no captured antibody) surface. For each sample, subtracting the amount of non-specific binding from the total yielded the amount of specific binding.

Figure 3. Schematic of antibody binding onto modified particle surface.

Detecting The Activity of The Bound Antibodies

After the primary antibody was attached to the silica shells or silica gel beads, 100 ng of cytochrome C was added. Following incubation at 37° C for 30 min, the silica shells were spun at 10,000 rpm for 2 minutes and the supernatant, which only contained the unbound cytochrome C, was retrieved. The amount of unbound cytochrome C in the supernatant was measured by the SLM 8100 fluorescence with excitation at 410 nm. The amount of bound cytochrome C was determined by subtracting the amount of unbound cytochrome C from the amount of cytochrome C added. This amount represents the number of active antibodies.

Results and Interpretations

Determining The Surface Area of Silica Particles

The surface area of silica gel was determined to be 6.336 m²/g. Two different batches of silica nano-shell particles were used. Each batch of silica nano-shell made about 4 ml of solution in total. The two batches had slightly different surface areas; one batch of shell particles had 0.74 m²/ml and the other batch had 1.21 m²/ml.

According to the calculation on the synthesis of silica nano-shells, one batch of shells should have 11 mg of shells if there is 100% yield in synthesizing the shells. Based on these calculations, the silica shell would have about 2.96 m²/11mg to 4.84 m²/11 mg, a much wider surface area than silica gel. This large difference in surface areas may influence the surface chemistry of the two.

Primary Antibody Binding On Silica Nano-Shells

The results illustrated in Figure 4 show that the ANB-NOS-treated surface elicited the greatest antibody binding, while the APS-treated surface bound fewer antibodies. The control, which had no surface chemistry altered, bound the least amount of antibody.

Figure 4. PNP absorbance at 405 nm as a function of different amounts of sheep anti-cyt C (1:500 dilution). A mixture of 250 ml of secondary antibody (1:500 dilution of ALP-conjugated anti-sheep IgG) were added, followed by 0.6 nmoles of PNPP for color development. All three data sets were fitted using second-order polynomials. The control was 84.2% reproducible, the APS was 86.9% reproducible, and the ANB-NOS was 96.8% reproducible.

Both the control shells and the APS-treated shells are hydrophilic. The control shells have mostly hydroxyl groups on the surface, whereas the APS-treated shells have mostly amine groups, which should be only partially charged at pH 8.5. Since neither the control shells nor the APS-treated shells have the ability to covalently link to the antibody, any binding to these surfaces should be the result of hydrogen bonds, electrostatic interactions, or ionic bonding. It is believed the control shells adsorb the antibodies mostly through hydrogen bonds, whereas the APS-treated shells bind the antibodies mostly through electrostatic and ionic interactions. For the APS-treated surface, ionic bonds can form between the positively charged amine groups on the APS and the negatively charged carboxylic groups on the antibodies. Since electrostatic interactions and ionic bonds are stronger than hydrogen bonds, the APS-treated shells should adsorb more antibody than control shells. Indeed, the results shown in Figure 4 support this hypothesis.

Even though non-covalent bonds allowed the adsorption of antibodies, the amount of antibody binding was not high. The ANB-NOS-treated surface bound substantially more antibodies because the aryl azide group on the ANB-NOS can covalently link the antibody to the silica shells. The cross-linking of antibody is initiated by irradiation with ultra-violet light, which activates the aryl azide group to a nitrene. Through the nitrene, some amino acid side chains on the antibody can be cross-linked to form amides. Since nitrene is a very reactive functional group, the lifetime of the cross-linking process takes less than 10^-4 second (Lewis 1977). The short lifetime of the cross-linking process is beneficial because it minimizes nonspecific binding.

Secondary Antibody Binding On Silica Nano-Shell

Figure 5 shows the extent of secondary antibody binding on control, APS- and ANB-NOS-treated surfaces as a function of antibody concentration. Again, more antibodies were bound to the ANB-NOS-treated shells because of the strength of covalent attachment. The control was used to determine the amount of non-specific binding of secondary antibody to silica shell, which was significant when low concentrations of secondary antibody were used. The average percentage of bound antibody on APS- and ANB-NOS-treated surface was about 0.5% and 1.5%, respectively. About 10% of the bound antibodies on the APS- and 12% on ANB-NOS-treated surfaces were active. The amount of active antibodies on the control surface was too low to be detected.

Figure 5. PNP absorbance at 405 nm as a function of different amounts of secondary antibody (1:500 dilution) for 10 ml of a 1:50 dilution of sheep anti-cyt C. The best-fit line was determined for all three sets of data as shown. The control was 83.7% reproducible, the APS was 94.4% reproducible, and the ANB-NOS was 98.4% reproducible.

Unfortunately, the percentage of bound antibody was relatively low, which poses a problem if large quantities of antibody are to be attached to silica shells. However, the percentage of bound antibodies might be increased by using a more "pure" antibody, such as a monoclonal antibody. This is one avenue for further research.

Non-specific binding of secondary antibody to silica shell decreased as the concentration of secondary antibody increased. This was because only a small, limited area was available for non-specific binding. Therefore, at high concentrations of secondary antibody, the amount of non-specific binding became insignificant.

Secondary Antibody Binding On Silica Gel

Figure 6 shows the binding of the secondary antibody to the primary antibody on APS- and ANB-NOS-treated surfaces, as well as on plain silica gel. The average percentage of antibody bound on APS- and ANB-NOS-treated surfaces was about 10% and 3.5%, respectively. On silica gel, in contrast to the nano-shells, the antibody bound best to the APS-treated surface. The percentage of antibody bound on plain silica gel decreased as the concentration of antibody increased. About 12% of the bound antibodies on the APS- and only 1% on the ANB-NOS-treated surface were active as determined by the amount of cytochrome C bound. The amount of active antibodies on control surface was too low to be detected.

Figure 6. PNP absorbance at 405 nm as a function of different amounts of the secondary antibody (1:5000 dilution) for 50 ml of a 1:500 dilution of sheep anti-cyt C. The best-fit line was determined for all three sets of data as shown. The silica gel only was 87.4% reproducible, the APS was 90.8% reproducible, and the ANB-NOS was 93.3% reproducible.

In theory, silica gel has the same surface chemistry properties as the silica nano-shell. However, there is at least one major difference between the two silica surfaces. The gel consisted of visible particles, with a large internal surface area. Silica gel was formed by the aggregation of very small colloidal silica particles into larger particles (Fig. 7A). The interior initially consisted of chains of the smaller particles interconnected to form a network and could be described by mostly positively curved surfaces. As the gel aged, the individual particles became less discernable. Finally, the interior can be described as a network of pores with negatively curved surfaces that interpenetrate the larger particles (Fig. 7B).

Figure 7. [A] represents a cross section of the aggregate of smaller colloidal silica particles at the time of formation of the silica gel. [B] represents the cross section of the same aggregate showing the smoothing and filling in that occurs as the silica gel is aged. Note that the total surface area is somewhat decreased, and more importantly, there is an increase in negatively curved surface area at the expense of regions of positive curvature (Synder 1968).

The curvature of the surface has an effect on the surface chemistry that can be described as a strain on the surface siloxyl groups. Figure 8A shows the surface curvature of a silica nano-shell, which has a positive curvature, while Figure 8B shows the surface curvature of silica gel, which has mostly negative curvature (Iler 1968).

Figure 8. [A] shows the surface curvature of silica shell, which has a positive curvature. [B] shows the surface curvature of silica gel, which has mostly negative curvature (Iler 1979).

Since ANB-NOS is relatively "bulky," it is hard to saturate ANB-NOS on the negatively curved surface. Even if there is ANB-NOS on the surface, it is almost impossible for each ANB-NOS to bind antibody independently in a limited space. In this case, the APS surface is also favored in that when an antibody fits in a negatively curved area, it can have multiple electrostatic interactions to strengthen the binding. Therefore, APS-treated surfaces can bind more antibody than ANB-NOS surfaces. It has also been shown that groups present on a positively curved silica surface tend to be more easily removed than on a negatively curved surface (Snyder 1968). This should affect how strongly the antibody is adsorbed onto an APS-treated surface, since the interaction between the positively charged primary amine groups and the antibodies is electrostatic and not covalent. In contrast, the ANB-NOS-treated surface involves a covalent interaction, and so once attached, antibodies should remain attached. At the same time, the accessibility of the surface to ANB-NOS is expected to be greater for the positively curved surface since the addition of ANB-NOS requires the chemical modification of the surface.

The nano-shell surface consists of the outside of a sphere with an approximate diameter of 70-80 nm. Figure 9A shows antibodies attached to the outside of a curved surface. These antibodies tend to have greater freedom of movement, compared with molecules that would be lining the interior of a pore of similar or smaller internal diameter, as shown in Figure 9B.

Figure 9. [A] represents antibodies attached to the outside of the silica shell. [B] shows the same sized antibodies attached to a concave surface. Note that there is more freedom of motion for molecules attached to a positively curved surface.

This effect should be greatest for small pores that are the size of the antibody, although for the smallest pores, the antibody is unlikely to be able to penetrate into the space. The washing steps may also have been affected by the inclusion of antibody inside pores, where they are protected from fluid shear, and therefore, more antibodies remained despite the relatively weaker attachment to the silica surface.

It is also possible that the smaller pores present in the interior of the silica gel favor the use of a linker molecule with a shorter side chain. APS alone is less bulky than the combination of APS and ANB-NOS, and this restricts the effective surface that can be functionalized with the larger linker molecules to those pores above a certain radius. The entire outside surface of the nanosphere should be accessible, whereas a significant portion of the silica gel interior may consist of smaller pores. To further investigate these possibilities would require the use of a silica gel material with a carefully defined internal porosity. Our primary interest, however, was to find an effective means of attachment to the silica nano-shells.

Binding antibody to plain silica gel served as a control to detect the level of non-specific binding. As expected, like the non-specific binding on silica shells, the level of non-specific binding on silica gel decreases as the concentration of antibody increases.

Table 1. Summary of antibody activity on different surfaces and particles.

The percentage of antibody bound on silica shell and silica gel APS surface improved from 0.5% to 10%, respectively. This showed antibody binding on APS surface was favored on silica gel by 20-fold, compared to silica shell. The percent of active antibodies bound on APS surfaces also increased from 10% to 12%. Since silica gel binds more antibody and its bound antibodies have higher activity on APS-treated surfaces, silanisation of APS alone would be the preferred method for binding antibody onto silica gel.

The percentage of antibody bound on silica shell and silica gel ANB-NOS surface improved less dramatically, from 1.5% to 3.5%. However, the amount of active antibody dropped from 12% on silica shells to 1% on silica gel. This dramatic decrease in active antibody is again believed to be an effect of the surface curvature of the silica gel. As previously explained, the negative curvature on silica gel has limited space to fit ANB-NOS. Therefore, as the antibody is forced into the negatively curved areas, it puts angular strain on the covalent bond between the antibody and the ANB-NOS. This strain might have disrupted the antibody's ability to bind antigen. Since silica shells bind more antibody and have higher antibody activity when coated with ANB-NOS, using the ANB-NOS cross-linker would be the preferred method for binding antibody onto silica shells.

Conclusion

It appears that, as expected, the covalent method using ANB-NOS for attaching antibodies to silica nano-shells was superior to the non-covalent method using APS, which in turn was more effective than non-specific adsorption by an untreated surface. The effectiveness of attachment was rated by measuring the amount of antibody that remained bound after washing and also by examining the activity of the bound antibody. These results were not surprising, as the covalent bond between the silica surface and the antibody formed should be very resistant to disruption.

Overall, greater antibody activity was expected for the antibodies attached to the silica shells than those attached to the silica gel. The antibodies should have greater freedom of motion on the outside of a small sphere, as compared to antibodies located on the interior of a pore. The finding that the actual amount of antibody bound on APS-treated surfaces is greater in the case of the silica gel than the nano-shells was therefore unexpected, as the opposite is true for the ANB-NOS treatment. This finding was attributed to differences in surface curvature. The negatively curved surfaces on the silica gel differ in chemical activity from the positive surfaces of the nano-shell. Also, physical effects such as the pore size may favor the smaller size of the APS-only linker. It may be that in areas of negative curvature, multiple attachments between the silica surfaces and the walls of the pore are likely. If these were of a covalent nature, as with ANB-NOS, it would be likely that these interactions limit the mobility, and therefore binding ability, of the antibody. Finally, it would be expected that antibody bound in a small pore is relatively protected from the effects of physical washing methods involving sonication and vortexing.

References
H. Aizawa, S. Kurosawa, K. Ogawa, M. Yoshimoto, J. Miyake and H. Tanaka. (2001) Conventional Diagnosis of C-reactive Protein In Serum Using Latex Piezoelectric Immunoassay. Sensors and Actuators B: Chemical. 76: 173-76.

R. Axen and S. Ernback. (1971) Chemical Fixation of Enzymes to Cyanogen Halide Activated Polysaccharide Carriers. Er. J. Biochem. 18: 351-60.

R. Axen and P. Vretblad. (1971) Binding of Protens to Polysaccharides by Means of Cyanogen Halides, Studies of Cyanogen Bromide Treated Sephadex. Acta Chem. Scand. 25: 2711-16.

S.K. Bhatia, L.C. Shriver-Lake, K.J. Prior, J.H. Georger, J.M. Calvert, R. Bredehorst and F.S. Ligler. (1989) Use of Thiol-terminal Silanes And Heterobifunctional Crosslinkers For Immobilization of Antibodies On Silica Surfaces. Anal. Biochem. 178: 408-13.

M. Bradfort. (1976) A Rapid And Sensitive Method for the Quantitiation of Microgram Quantities of Protein Utilizing the Principle of Protein Dye Binding. Anal. Biochem. 72: 248-54.

K.R. Brown, D.G. Walter and M.J. Natan. (2000) Seeding of Colloidal Au Nanoparticle Solutions. 2. Improved Control of Particle Size And Shape. Chem. Mater. 12: 306-13.

A. Bruning, K. Bellamy, D. Talbot and J. Anderson. (1999) A Rapid Chromatographic Strip Test For The Pen-side Diagnosis of Rinderpest Virus. J. of Virological Meth. 81: 143-54.

J.B. Brzoska, I.B. Azouz and F. Rondelez. (1994) Silanization of Solid Substrates: A Step Toward Reproducibility. Langmuir. 10: 4367-73.

P.A. Buining, B.M. Humbel, A.P. Philipse and A.J. Verkleij. (1997) Preparation of Functional Silane-stabilized Gold Colloids In The (Sub)nanometer Size Range. Langmuir. 13: 3921-26.

L. Davalos-Pantoja, J.L. Ortega-Vinuesa, D. Bastos-Gonzales and R. Hidalgo-Alvarez. (2001) Colloidal Stability of IgG- And IgY-coated Latex Microspheres. Colloids and Surfaces B: Biointerfaces. 20: 165-75.

D. Gabel and R. Exen. (1976) Characterization of Immobilized Enzymes by Chemical Methods. Meth. Enzymol. 44: 383-93.

R.K. Iler. (1979) The Chemistry of Silica. John Wiley & Sons, New York. 203-206, 243.

E.T. Kaiser, R.L Colescott, C.D. Bossinger, P.I. Cook (1970) Color Test For Detection of Free Terminal Amino Groups In the Solid-Phase Synthesis of Peptides. Anal. Biochem. 34:595-8.

Y. Levin, M. Pecht, L. Goldstein and E. Katchalski. (1964) A Water-Insoluble Polyanionic Derivative of Trypsin. Biochemistry. 3: 1905-13.

R.V. Lewis, M.F. Roberts, E.A. Dennis and W.S. Allison. (1977) Photoactivated Heterobifunctional Cross-linking Reagents Which Demonstrate The Aggregation State of Phospholipase A2. Biochemistry 16: 5650-54.

J.N. Lin, J. Herron and J.D. Andrade. (1988) Characterization of Immobilized Antibodies On Silica Surfaces. IEEE Trans. Biomed. Eng. 35: 466-71.

O. Lowry, N. Rosebrough, A. Farr and R. Randall. (1951) Protein Measurement With the Folin Phenol Reagent. J. Biol. Chem. 193: 265-275.

C.A. Mirkin, R.L. Letsinger, R.C. Mucic and J.J. Storhoff. (1996) DNA-based Method Assembles Nanoparticles In Macroscopic Materials. Nature. 382: 607-09.

A.H. Nashat, M. Moronne and M. Ferrari. (1998) Detection of Functional Groups And Antibodies On Microfabricated Surfaces By Confocal Microscopy. Biotech. And Bioeng. 60: 137-46.

K.G.I. Nilsson. (1989) Preparation of Nanoparticles Conjugated With Enzyme And Antibody And Their Use In Heterogeneous Enzyme Immunoassays. J. of Immuno. Meth. 122: 273-77.

G.W. Sears. (1956) Determination of Specific Surface Area of Colloidal Silica By Titration With Sodium Hydroxide. Anal. Chem., 28: 1981-3.

L.R. Synder. (1968) Priciples of Adsorption Chromatography. M. Dekker, New York. 156.

B. Sharp, D. Morton and A.E. Clark. (2000) Effectiveness of Metal Surface Treatments In Controlling Microleakage of the Acrylic Resin-Metal Framework Interface. J. Prosthetic Dentistry. 84: 617-22.

P. Smith, R. Krohn, G. Hermanson, A. Mallia, F. Gartner, M. Provenzano, E. Fujimoto, N. Goeke, B. Olson and D. Klenk. (1985) Measurement of Protein Using Bicinchoninic Acid. Anal. Biochem. 150: 76-85.

A.M. Tinsley-Bown et al. (2000) Tuning The Pore Size And Surface Chemistry of Porous Silicon For Immunoassays. Phys. Stat. Sol. 182: 547-53.

B. Yoza, M. Matsumoto and T. Matsunaga. (2002) DNA Extraction Using Modified Bacterial Magnetic Particles In the Presence of Amino Silane Compound. J. Biotech. 94: 217-24.

SEARCH | SITE MAP | RECENT WEB SITE ADDITIONS PRIVACY POLICY | CONTACT US

JYI is supported by: The National Science Foundation, The Burroughs Wellcome Fund, Glaxo Wellcome Inc., Science Magazine, Science's Next Wave, Swarthmore College, Duke University, Georgetown University, and many others.