Accessibility of the Fluorescent Reporter Group in Native, Silica

Sep 15, 1996 - Fluorescence quenching techniques are used to investigate the accessibility of a model biorecognition element−reporter group system w...
0 downloads 5 Views 172KB Size
Download PDF
Anal. Chem. 1996, 68, 3194-3198

Accessibility of the Fluorescent Reporter Group in Native, Silica-Adsorbed, and Covalently Attached Acrylodan-Labeled Serum Albumins Christine M. Ingersoll, Jeffrey D. Jordan, and Frank V. Bright*

Department of Chemistry, Natural Sciences and Mathematics Complex, State University of New York at Buffalo, Buffalo, New York 14260-3000

Fluorescence quenching techniques are used to investigate the accessibility of a model biorecognition elementreporter group system when in buffer, surface-adsorbed, and covalently attached to a silica surface. The siteselective fluorescent reporter group, 6-acryloyl(dimethylamino)naphthalene (acrylodan, Ac), is attached covalently (at cysteine-34) to bovine and human serum albumin (BSA and HSA, respectively) and serves as a surrogate recognition element-reporter group system. Molecular oxygen is used to quench the Ac fluorescence and the accessibility, in the form of bimolecular rate constants (kq), in each model system is quantified. Although one might expect these systems to exhibit similar behavior, differences in quenching characteristics are observed, such as wavelength dependency of the Stern-Volmer quenching constant (KSV) for the native proteins in buffer. BSA-Ac exhibits wavelength dependent KSV values as well as a blue-shifted emission spectrum on O2 addition. Physisorption of BSA-Ac onto a fused-silica optical fiber lowers the accessibility of Ac to O2, whereas covalent attachment of BSA-Ac to APTES/glutaraldehyde-modified silica enhances the accessibility of the Ac reporter group to O2. Over the past several years, the development of new chemical and biosensing schemes has been a topic of growing interest. A portion of our research1-7 has centered on the development of new chemical biosensors. In the course of this work, we have discovered the advantage of using surrogate biorecognition (1) (a) Bright, F. V.; Litwiler, K. S.; Vargo, T. G.; Gardella, J. A., Jr. Anal. Chim. Acta 1992, 262, 323-30. (b) Betts, T. A.; Catena, G. C.; Huang, J.; Litwiler, K. S.; Zhang, J.; Zagrobelny, J.; Bright, F. V. Anal. Chim. Acta 1991, 246, 55-63. (c) Bright, F. V.; Betts, T. A.; Litwiler, K. S. Anal. Chem. 1990, 62, 1065-9. (2) Bright, F. V.; Wang, R.; Li, M.; Dunbar, R. A. Immunomethods 1993, 3, 104-11. (3) (a) Narang, U.; Wang, R.; Prasad, P. N.; Bright, F. V. Anal. Chem. 1993, 65, 2671-5. (b) Narang, U.; Prasad, P. N.; Bright, F. V.; Kumar, A.; Kumar, N. D.; Malhotra, B. D.; Kamalasanan, M. N.; Chandra, S. Anal. Chem. 1994, 66, 3139-44. (c) Narang, U.; Prasad, P. N.; Bright, F. V.; Kumar, A.; Kumar, N. D.; Malhotra, B. D.; Kamalasanan, M. N.; Chandra, S. Chem. Mater. 1994, 6, 1596-8. (4) Lundgren, J. S.; Heitz, M. P.; Bright, F. V. Anal. Chem. 1995, 67, 377581. (5) (a) Wang, R.; Bright, F. V. J. Phys. Chem. 1993, 97, 4231-8. (b) Wang, R.; Bright, F. V. J. Phys. Chem. 1993, 97, 10872-8. (c) Wang, R.; Bright, F. V. Appl. Spectrosc. 1993, 47, 800-6. (d) Wang, R.; Bright, F. V. Appl. Spectrosc. 1993, 47, 792-9. (6) Wang, R.; Sun, S.; Bekos, E. J.; Bright, F. V. Anal. Chem. 1995, 67, 14959. (7) Jordan, J. D.; Dunbar, R. A.; Bright, F. V. Anal. Chem. 1995, 67, 2436-43.

3194 Analytical Chemistry, Vol. 68, No. 18, September 15, 1996

element-reporter group systems as tools to understand the “biosensing” process. As examples, our work to date has shown how the protein dynamics itself and those of the solvent molecules affect the reporter group response2,5-7 and how the extent of protein hydration4 can govern the overall system response and dynamics. Together this work is providing a link between the local dynamics surrounding the biorecognition-reporter group pair and the factors controlling the analytical figures of merit for a given biosensor. Bovine and human serum albumins (BSA and HSA, respectively), labeled with the fluorescent probe acrylodan (BSA- and HSA-Ac), have been our most common model systems. BSA and HSA were chosen because they represent the most abundant of all plasma proteins and they have been extensively studied.8-12 Serum albumins are also large, multidomain globular proteins and their biological functions include maintaining proper osmotic pressure in the blood as well as assisting in transport and regulatory processes.8,9,13 Serum albumins consist of ∼580 amino acid residues, dimensions on the order of 40 × 140 Å, and a molecular weight close to 67 000.9,13 BSA and HSA also contain a free thiol group at cysteine-34.13 Many studies have also used BSA as a model to investigate surface-induced conformational changes in proteins at interfaces.14-19 The fluorescent reporter group used in the current experiments is 6-acryloyl(dimethylamino)naphthalene, commonly known as acrylodan (Ac).20 The reasons for using this particular fluorescent probe are 3-fold. First, the photophysics of Ac are (8) Brown, J. R.; Shockley, P. In Serum Albumins: Structure and Characterization of Its Ligand Binding Sites in Lipid Protein Interactions; Jost, P. C., Griffith, O. H., Eds.; John Wiley & Sons: New York, 1982. (9) Garrison, M. D.; Iuliano, D. J.; Saavadra, S. S.; Truskey, G. A.; Reichert, W. M. J. Colloid Interface Sci. 1992, 148, 415-24. (10) Thomas, M. P.; Nelson, G.; Patonay, G.; Warner, I. M. Spectrochim. Acta 1988, 43B, 651-60. (11) Sudlow, G.; Birkett, D. J.; Wade, D. N. Mol. Pharmacol. 1975, 11, 824-32. (12) Sudlow, G.; Birkett, D. J.; Wade, D. N. Mol. Pharmacol. 1976, 12, 105261. (13) Peters, T.; Sjoholm, I. Albumin: Structure, Biosynthesis and Function; Pergammon Press: New York, 1978. (14) Horbett, T. A.; Brash, J. L. Proteins at Interfaces: Current Issues and Future Prospects. In Proteins at Interfaces, Physicochemical and Biochemical Studies; Brash, J. L., Horbett, T. A., Eds.; ACS Symposium Series, 343; American Chemical Society: Washington, DC, 1987. (15) Hlady, V.; Andrade, J. D. Colloid Surf. 1988, 32, 359-69. (16) Hlady, V.; Andrade, J. D. Colloid Surf. 1989, 42, 85-96. (17) Rainbow, M. R.; Atherton, S.; Eberhart, R. C. J. Biomed. Mater. Res. 1987, 21, 539-55. (18) Crystal, B.; Rumbles, G.; Smith, T. A.; Phillips, D. J. Colloid Interface Sci. 1993, 155, 247-50. (19) Iuliano, D. J.; Saavedra, S. S.; Truskey, G. A. J. Biomed. Mater. Res. 1993, 27, 1103-13. S0003-2700(96)00315-0 CCC: $12.00

© 1996 American Chemical Society

well-known.19,21 Second, Ac is very sensitive to changes in the physicochemical properties of its local environment.20-27 Third, at or above physiological pH, Ac reacts selectively only with free thiol groups;20-27 therefore, in BSA and HSA, the Ac emission reports from a single microenvironment (i.e., cybotactic region) within the protein (i.e., the region surrounding cysteine-34). Biomolecules are especially attractive recognition elements because their inherent function in nature includes selective binding to, conversion of, and/or transportation of chemical species. Therefore, biomolecules generally do not require any modifications to serve as recognition elements. However, to use a biorecognition element in a biosensor scheme, the biomolecule is generally immobilized on/in a selected substrate via covalent attachment, physisorption, or entrapment procedures. Because biosensor development generally involves immobilizing the recognition element on or in a substrate, it becomes critical to determine how the accessibility of the reporter group (within the biorecognition element) is affected upon surface immobilization. In this paper, our aim is to determine the accessibility of the reporter group (i.e., Ac) within a model biomolecule as a function of surface immobilization chemistry. Specifically, we report on (1) the accessibility of Ac in native BSA- and HSA-Ac in buffer, (2) how physisorption of BSA-Ac to silica affects the Ac accessibility, and (3) how the Ac accessibility is influenced by covalent attachment of BSA-Ac to silica. THEORY To determine the accessibility of the Ac reporter group within BSA- and HSA-Ac, we use fluorescence-quenching techniques.10,28-38 This particular approach is very useful because it provides considerable information about the local environment surrounding the fluorescent reporter group, it can sense structural changes within the cybotactic region, and it reports on the accessibility of the reporter group to the quencher.28 In this work, fluorescence quenching provides a convenient means to compare the accessibility of the Ac residue within native BSA- and HSA-Ac as well as BSA-Ac in two different surface-immobilized forms. The SternVolmer relationship10,28-30,35,36 provides the link between the (20) Prendergast, F. G.; Meyer, M.; Carlson, G. L.; Iida, S.; Potter, J. D. J. Biol. Chem. 1983, 258, 7541-4. (21) Ilich, P.; Prendergast, F. G. J. Phys. Chem. 1989, 93, 4441-7. (22) Sommer, A.; Gorges, R.; Kostner, G. M.; Paltauf, F.; Hermetter, A. Biochemistry 1991, 30, 11245-9. (23) Clark, I. D.; Burtnick, L. D. Arch. Biochem. Biophys. 1988, 260, 595-600. (24) Yem, A. W.; Epps, D. E.; Mathews, W. R.; Guido, D. M.; Richard, K. A.; Staite, N. D.; Deibel, M. R., Jr. J. Biol. Chem. 1992, 267, 3122-8. (25) Epps, D. E.; Yem, A. W.; Fisher, J. F.; McGee, J. E.; Paslay, J. W.; Deibel, M. R., Jr. J. Biol. Chem. 1992, 267, 3129-35. (26) Lehrer, S. S.; Ishii, Y. Biochemistry 1988, 27, 5899-906. (27) Reid, S. W.; Koepf, E. K.; Burtnick, L. D. Arch. Biochem. Biophys. 1993, 302, 31-6. (28) Eftink, M. R. In Topics in Fluorescence Spectroscopy, Vol. 2: Principles; Lakowicz, J. R., Ed.; Plenum Press: New York, 1991; Chapter 2. (29) Molski, A.; Keizer, J. J. Phys. Chem. 1993, 97, 8707-12. (30) Lakowicz, J. R. Principles of Fluorescence Spectroscopy; Plenum Press: New York, 1991. (31) Lakowicz, J. R.; Kus´ba, J.; Gryczynski, I.; Wiczk, W.; Szmacinski, H.; Johnson, M. L. J. Phys. Chem. 1991, 95, 9654-60. (32) Edmiston, P. L.; Wambolt, C. L.; Smith, M. K.; Saavadra, S. S. J. Colloid Interface Sci. 1994, 163, 395-406. (33) Blatt, E.; Husain, A.; Sawyer, W. H. Biochim. Biophys. Acta 1986, 871, 6-13. (34) Eftink, M. R.; Ghiron, C. A. Biochim. Biophys. Acta 1987, 916, 343-9. (35) Lakowicz, J. R.; Weber, G. Biochemistry 1973, 12, 4161-70. (36) Lakowicz, J. R.; Weber, G. Biochemistry 1973, 12, 4171-9. (37) Lakowicz, J. R.; Cherek, H.; Gryczynski, I.; Joshi, N.; Johnson, M. L. Biophys. J. 1987, 51, 755-68. (38) Wong, A. L.; Harris, J. M. J. Phys. Chem. 1991, 95, 5895-901.

experimental measurables and the “accessibility” terms:

Fo/F ) 1 + KSV[Q] ) 1 + kq〈τo〉[Q] ) 〈τo〉/〈τ〉

(1)

In this expression, Fo and F are the fluorescence intensities in the absence and in the presence of quencher, respectively; τo and τ are the average excited-state fluorescence lifetimes in the absence and in the presence of quencher, respectively; KSV is the Stern-Volmer quenching constant (M-1); kq is the bimolecular quenching rate constant (M-1 s-1); and [Q] is the analytical quencher concentration (M). Under Stern-Volmer quenching, for a single fluorophore in a homogeneous microdomain, a plot of Fo/F vs [Q] yields a straight line with a unity y-intercept. The slope equals KSV and one can compute kq with the average excitedstate lifetime in the absence of quencher (〈τo〉). The fluorescence quencher used for these experiments is molecular oxygen (O2). There are several key advantages to using O2 as opposed to other common quenchers like iodide32,38 and acrylamide.33,34,37 First, O2 is an effective fluorescence quencher.28,30,35,36 Second, O2 is neutral; thus electrostatic interactions between protein and quencher (as is common when ionic quenchers are used) are minimized.35,36 Third, there is no need to continuously control variables such as ionic strength and system pH when working with O2. This is especially important because both are known to influence protein conformation.35,36 Fourth, quenchers such as acrylamide are known to actually bind to proteins and affect their function/behavior.33 Fifth, the relatively small size of O2 allows it to diffuse easily within the protein matrix.35,36 Finally, the exact same sample is used throughout the entire experiment; therefore there is no repetitive sample preparation errors like are associated with typical quenching experiments. The main drawback to using O2 as a fluorescence quencher is associated with its relatively low solubility in aqueous solution (99%) (Sigma Chemical Co.); 1,4-bis(4-methyl-5-phenyl-2-oxazolyl)benzene (Me2POPOP) (Aldrich Chemical Co.); Na2HPO4‚7H2O, NaH2PO4‚2H2O, hydrofluoric acid (50% in water), and acetonitrile (J. T. Baker); toluene, chloroform, acetone, methanol (all >99.9%), and glutaraldehyde (50% solution in water) (Fisher Scientific); and ethanol (200 proof, Quantum Chemical Corp.). All reagents were used as received without further purification. Aqueous solutions were prepared in doubly distilled deionized water, and stock solutions were kept refrigerated in the dark at 4 °C. Preparation of BSA-Ac and HSA-Ac. The protocol used to prepare BSA- and HSA-Ac has been described elsewhere.6,9 (39) Chang, R. Chemistry, 3rd ed.; Random House: New York, 1988; p 468.

Analytical Chemistry, Vol. 68, No. 18, September 15, 1996

3195

Adsorption of BSA-Ac to Silica.6,40 An all-silica optical fiber (1000-µm core diameter, General Fiber Optics) was treated as follows: the polyimide protective coating was mechanically stripped from a 1-cm segment at the distal end of the optical fiber. The stripped segment was then soaked in HF for 10 min to remove the cladding (yielding a bare fused-silica surface), it was rinsed with water and allowed to air-dry. The bare fiber segment was soaked in a 50 µM stock solution of BSA-Ac for several hours, permitting the BSA-Ac to physisorb onto the silica surface. The fiber was subsequently rinsed and soaked in phosphate buffer (0.1 M, pH 7.0) until no Ac fluorescence was detected in the rinse buffer. Covalent Attachment of BSA-Ac to Silica.41 The optical fiber used for these experiments was prepared as above except the silica surface was derivatized with APTES and glutaraldehyde following a procedure similar to that of Lu and co-workers.41 For APTES modification of the silica surface, the optical fiber was soaked in 20 mL of dry toluene containing 0.4 mL of APTES and 0.2 mL of triethylamine for 6 h at 20 °C. The solution was decanted and successive washings with chloroform, acetone, and methanol were repeated. The APTES-modified silica fiber was then reacted for 90 min with 2.5% glutaraldehyde (v/v) in phosphate buffer (0.1 M, pH 7.0). The APTES/glutaraldehyde-modified silica surface was rinsed well with water and then with phosphate buffer. Finally, the optical fiber was soaked in a 50 µM stock solution of BSA-Ac for several hours and rinsed with buffer until no Ac fluorescence was detected in the wash solution. Instrumentation/Special Equipment. The instrumentation used for all steady-state and frequency-domain fluorescence measurements and the software used for data analysis have been described previously.6 For the native protein in solution studies, a Xe arc lamp was used as the excitation source. The stainless steel high-pressure O2 cell is similar to one described by Lakowicz and Weber.35 (Note: These experiments require the use of moderately high pressures. One must exercise extreme caution in order to ensure against catastrophic decompression.) A temperature-regulated fluid is pumped directly through a coil within the cell to ensure temperature control (20 ( 0.1 °C). O2 from a standard gas cylinder is introduced directly into the cell through stainless steel tubing and a valve. The internal cell pressure is monitored ((1 psia) with a standard digital pressure transducer. The solution is constantly stirred throughout the experiment. The concentration of B/HSA-Ac within the cell is 2 µM in 0.1 M, pH 7 phosphate buffer. Surface Experiments. The surface-immobilized BSA-Ac quenching experiments were carried out in a modified form of the basic high-pressure cell that accepted a 1000-µm optical fiber through the top port. Figure 1 is a schematic of the high-pressure cell system as used for the interfacial O2 quenching experiments. For the physisorbed BSA-Ac experiment, an argon ion laser operating at 351.1 nm was used as the excitation source and the emission was collected through a 385-nm long-pass filter. For the covalently attached BSA-Ac experiments, a Xe arc lamp was used as the excitation source (λex ) 351 nm, bandpass 16 nm). RESULTS AND DISCUSSION All KSV, kq, and values for BSA- and HSA-Ac are presented in Table 1. The actual Stern-Volmer plots from which some of (40) Newby, K.; Andrade, J. D.; Benner, R. E.; Reichert, W. M. J. Colloid Interface Sci. 1986, 111, 280-2. (41) Lu B.; Xie, J.; Lu, C.; Wei, Y. Anal. Chem. 1995, 67, 83-7.

3196 Analytical Chemistry, Vol. 68, No. 18, September 15, 1996

Figure 1. Schematic of the high-pressure cell, including modifications to accept optical fiber, used for all oxygen quenching experiments. Table 1. Stern-Volmer Quenching Constants (KSV), Average Fluorescence Lifetime in the Absence of Quencher (〈τo〉), and Bimolecular Rate Constants (kq) for Acrylodan in Solution and in the Cybotactic Region of Serum Albumins under Various Conditions system

KSV (M-1)a

〈τo〉 (ns)b

10-9 kq (M-1 s-1)

acrylodan HSA-Ac τo/τ (HSA-Ac) BSA-Ac (native) τo/τ (BSA-Ac) BSA-Ac (physisorbed) BSA-Ac (covalently attached)

4.6 ( 0.2 8.2 ( 0.4 8.2 ( 0.4 7.1 ( 0.3 7.1 ( 0.3 3.6 ( 0.3 9.0 ( 0.7

0.80 3.2 3.2 2.9 2.9 2.5 2.8

5.8 ( 0.3 2.6 ( 0.2 2.6 ( 0.2 2.4 ( 0.2 2.4 ( 0.2 1.4 ( 0.2 3.2 ( 0.3

a Determined by integration of the entire emission spectra. b The uncertainty in the excited-state lifetime is e0.01 ns.

the data are derived are available in the supporting information. Native BSA-Ac and HSA-Ac. The Ac fluorescence from native BSA- and HSA-Ac is quenched to ∼50% at 0.14 M O2 (∼1500 psia). The intensity and lifetime Stern-Volmer plots for BSA-Ac and HSA-Ac quenched with O2 provides information about the accessibility of the reporter group (Ac) and the nature of the quenching process.28,30,35 The fact that the intensity Stern-Volmer relationship is linear and that the intensity- and lifetime-based plots are superimposeable (see supporting information) demonstrates that the quenching process is purely dynamic28,30,35 (collisional). That is, there is no static phenomenon (complex formation between the quencher and fluorophore) taking place in this system. The KSV for BSA- and HSA-Ac quenched with O2 are 7.1 and 8.2 M-1, respectively. At first glance, this might lead one to speculate that the Ac residue in HSA is more accessible than in BSA; however, when the average Ac excited-state fluorescence lifetimes (2.9 ns for BSA-Ac, 3.2 ns for HSA-Ac) are considered, one finds the kq values are indistinguishable at the 99% confidence level. The KSV for free Ac is only 4.6 M-1 (slightly more than half of the value found when attached to the proteins), but when one takes into account the free Ac excited-state fluorescence

Figure 2. Stern-Volmer quenching constant as a function of emission wavelength (interval 20 nm) for native BSA-Ac (A) and native HSA-Ac (B) in buffer quenched with O2.

lifetime (0.80 ns), the kq value for free Ac is significantly greater than kq for the Ac residue within native BSA- and HSA-Ac in buffer. These results demonstrate that the accessibility of the Ac residue in native BSA- and HSA-Ac is equal, but only about half that of free Ac. To better understand the effects of O2 on these protein systems, we investigated how O2 quenching affected the Ac emission spectral profiles. Toward this end, we dissected the steady-state BSA- and HSA-Ac emission spectra into 20-nm wavelength segments as a function of O2 concentration, and the area within the wavelength segment was integrated to generate a Stern-Volmer plot for each wavelength interval. The recovered Stern-Volmer quenching constants vs emission wavelength are shown in Figure 2. These results demonstrate that BSA-Ac (A) and HSA-Ac (B) behave differently when quenched by O2. Specifically, for native BSA-Ac, the red edge of the emission profile (associated with the “relaxed” form of Ac)4,6,7 is more effectively quenched compared to the emission blue edge (associated with “unrelaxed” Ac).4,6,7 In contrast, O2 quenching of native HSA-Ac does not lead to any detectable change in KSV across the emission profile. Together these results suggest that the dynamics within

the cybotactic region surrounding the Ac reporter group4,6,7 are different in BSA- and HSA-Ac. This is consistent with previous picosecond and nanosecond dipolar relaxation experiments.7 Thus, the BSA-Ac emission spectra shift to the blue on addition of O2, but no detectable spectral shift is evident in the HSA-Ac system (see supporting information). Surface-Adsorbed BSA-Ac. O2 quenching experiments on BSA-Ac physisorbed to a silica surface are summarized in Figure 3. Panel A illustrates a typical raw Stern-Volmer plot for silicaadsorbed BSA-Ac where a clear upward deviation is apparent. Curvature toward the y-axis is generally evidence of two types of quenching (static and dynamic) occurring on the same fluorophore population.28,30,35 However, attempts to fit these data to a “modified” Stern-Volmer expression30,35 led to poor fits and imaginary (i.e., negative) quenching constants. We hypothesized that desorption of BSA-Ac from the silica surface during the time course of our O2 quenching experiments (up to 24 h) was the cause of the nonlinear Stern-Volmer plots (i.e., Figure 3A). (Control experiments confirmed that photobleaching was not a cause of the fluorescence decrease.) Figure 3B presents the percentage of the total BSA-Ac that desorbed from the surface during a typical 24-h experiment. These results clearly demonstrate that protein desorption contributed to the upward deviation seen in Figure 3A. The Figure 3B data also provide a convenient means to correct for protein desorption. Figure 3C illustrates the corrected Stern-Volmer plot for silica-adsorbed BSAAc (r2 ) 0.9995). The kq value recovered from these experiments for physisorbed BSA-Ac is significantly smaller (∼40%) than that for the native BSA-Ac. This result suggests that the Ac residue is much less accessible to O2 when BSA-Ac is physisorbed to silica. Covalently Attached BSA-Ac. When BSA-Ac is covalently attached to the APTES/glutaraldehyde-modified silica, the SternVolmer plots are linear (r2 g 0.99) and there is no evidence of protein desorption. The kq for the covalently attached BSA-Ac is 3.2 × 109 M-1 s-1 compared to 2.4 × 109 M-1 s-1 for native BSAAc in buffer. Assuming that the analytical concentrations of O2 in the bulk and at the interface are equal, this increase in kq suggests a substantial increase in the accessibility of the Ac residue when BSA-Ac is covalently attached to the silica surface using the APTES/glutaraldehyde procedure. In turn this implies either that the protein-reporter group system is bonded to the silica surface in such a way that the pocket in which the Ac residue is sequestered becomes more exposed to the bulk solution or that there is a conformational change in the BSA that leads to an opening up of the cybotactic region upon chemical attachment of the protein to the silica surface.

Figure 3. Typical raw Stern-Volmer plot for BSA-Ac physisorbed to silica when quenched with O2 (A), percent of BSA-Ac that desorbed from the silica surface of the optical fiber as a function of time (B), and Stern-Volmer plot for silica-adsorbed BSA-Ac quenched with O2 corrected for time-dependent desorption of BSA-Ac from the surface (C).

Analytical Chemistry, Vol. 68, No. 18, September 15, 1996

3197

biosensor systems (Ac-labeled bovine and human serum albumins), and the effects of surface immobilization on the accessibility of Ac in BSA-Ac were investigated. Stern-Volmer quenching constants and bimolecular rate constants determined in these experiments for free Ac in buffer, native HSA- and BSA-Ac in buffer, and surface-adsorbed and covalently attached BSA-Ac are summarized in Table 1. These results demonstrate that the O2/ Ac quenching process is purely dynamical and there is no complex formation. In the native proteins in buffer, the Ac reporter group is equally accessible to O2 in terms of kq. O2 quenching of BSAAc causes a shift in the emission spectrum. The HSA-Ac emission spectra do not shift with O2. Upon physisorption of BSA-Ac to a silica surface, the acrylodan reporter group becomes less accessible, suggesting that the domain surrounding the Ac residue is protected in some way. The results for BSA-Ac covalently attached to the APTES/glutaraldehyde-modified silica demonstrate that the Ac accessibility to O2 increases due to covalent immobilization. Figure 4. Behavior of BSA-Ac quenched with O2 in the native form in buffer (A), when physisorbed to a silica surface (B), and when covalently attached to an APTES/glutaraldehyde-modified silica surface (C). No three-dimensional protein structure is implied nor are all proteins at the same conformation at the interface.

Figure 4 presents a model consistent with the behavior of BSAAc in solution (A), physisorbed to a silica surface (B), and covalently attached to a silica surface (C). CONCLUSIONS In this paper, we have developed a convenient tool that allows one to study fluorescent reporter group accessibility and dynamics within surface-immobilized biomolecule-reporter group systems without the need to alter the system ionic strength, pH, and/or protein structure. O2 quenching techniques were used to measure the accessibility of a fluorescent reporter group in surrogate

3198 Analytical Chemistry, Vol. 68, No. 18, September 15, 1996

ACKNOWLEDGMENT This research was generously supported by the National Science Foundation (CHE-9300694 and CHE-9626636). Assistance by Jeffrey S. Lundgren on the silica surface modification with APTES/glutaraldehyde is greatly appreciated. SUPPORTING INFORMATION AVAILABLE Supporting information (three figures) demonstrating the linearity of the Stern-Volmer plots and the effects of O2 quenching on the emission spectra (5 pages). Ordering information is given on any current masthead page. Received for review April 1, 1996. Accepted June 27, 1996.X AC960315Q X

Abstract published in Advance ACS Abstracts, August 1, 1996.