Horseradish Peroxidase Adsorption on Silica Surfaces as an Oscillatory Dynamical Behavior Ewa S. Kirkor † and Alexander Scheeline* ,† Department of Chemistry, UniVersity of Illinois at Urbana-Champaign, Urbana, Illinois 61801 ReceiVed: March 27, 2001; In Final Form: May 23, 2001 Oxidation of Horseradish Peroxidase (HRP) to Compound I (CpI) induces release of the protein from adsorbing silica surfaces. Adsorption and desorption can be modulated by changes in concentration of hydrogen peroxide in solution in contact with the adsorber. Coupling of the HRP adsorption-desorption cycles to the chemistry of the peroxidase in bulk solution by oscillations in concentration of hydrogen peroxide is postulated as a plausible explanation for a qualitative change in the dynamics of the peroxidase oscillator at low concentrations of the enzyme. We hypothesize that the relative change in the HRP concentration in solution will be most pronounced at low concentrations of the enzyme and high reactor surface-to-volume ratio. Introduction Our interest in retention of horseradish peroxidase (HRP, native state ferriperoxidase, Per 3+ ) on silica (quartz) surfaces stems from a potential effect of adsorption on the dynamics of an oscillatory oxidation of NADH catalyzed by peroxidases (the PO reaction 1 ). A prototype for biological oscillatory reaction, the PO reaction is a much-studied model of nonlinear systems. 2-4 In general, proteins are known to bind to various oxides, silica among them. Adsorption of proteins to these surfaces is nonspecific and practically irreversible. 5 As a rule, an ability to manipulate adsorption of a protein requires information on its properties, or at least its class. 6 HRP adsorption properties are important for its multitude of analytical applications. Often, HRP is used adsorbed on silica particles, glass, or metal oxides. Compound I (CpI), formed in a reaction of native enzyme with hydrogen peroxide, initiates oxidations catalyzed by HRP. The dissolved protein activity depends on its pK a ,pI, and the acidity of the solvent. 7 Complexation or protonation of the native enzyme, which decreases its reactivity toward H 2 O 2 , lowers the observed rate constant for CpI formation. At pH values at or below the pK a for protonation of the imidazole group of the distal histidine of the enzyme cavity (pK a ) 4.1), the rate constant for CpI formation significantly decreases. In the presence of acetate, the rate constant of CpI formation (2 × 10 7 M -1 s -1 ) remains stable in the pH 5-8 range. 7 A decline of the activity of adsorbed enzyme is interpreted in terms of conformational relaxation induced by adsorption. 8 The magni- tude of the decline is disputed, and the enzyme regains its full activity upon reentering solution. A slow permanent loss of activity attributed to structural distortion progresses over tens of hours. 9 Adsorption on hydrophilic silicas (silanol termination) is presumably different from hydrophobic ones (siloxane termination). The extent of adsorption of the native enzyme may reach ten-monolayer coverage on some silica-containing sur- faces. 10 Current work has been conducted under mildly acidic conditions (0.1 M sodium acetate buffer, pH 5.1, our oscillatory reaction medium) on HRP nominally containing mainly isoform C1, and on hydrophilic silicas. Concentration of the enzyme in solution is an important experimental parameter controlling reaction dynamics. Existing models of the PO oscillator have difficulty in consistently simulating reaction dynamics for the whole range of experi- mentally accessible concentration of HRP, although significant progress in explaining low concentration behavior has recently been made. 4 Bistability and bursting that develop at low enzyme concentrations (below 1 μM) are difficult to simulate. Obviously, adsorption of HRP on the walls of the reactor can diminish the enzyme concentration in solution and so affect the reaction dynamics. A selection of the reactor material (polypropylene vs quartz) influences the frequency, amplitude, and duration of oscillations in the PO reaction. 11,12 Neither coupling of HRP reactivity on the walls of the reactor to the chemistry of the bulk of solution nor heterogeneity in PO reactions has been established. The presently described observation of reactive desorption of HRP from silica surfaces provides qualitative information about a mechanism through which chemistry of HRP on the walls of a quartz reactor could influence the dynamics of the chemistry in solution. Experimental Section To minimize HRP losses from solution, polypropylene containers were used. Adsorption of HRP on hydrophilic silica surfaces (NP2 chips and quartz) was monitored in SELDI-TOF- MS (surface enhanced laser desorption ionization-time-of-flight- mass spectrometry) experiments carried out with a Protein Chip Reader Series PBS II (Ciphergen Biosystems, Inc., Fremont, CA). The m/z calibration was based on the manufacturer’s protocol and verified using carbonic anhydrase (Sigma, M w 29 024). HRP (Boehringer-Mannheim) 1-10 μM solutions were prepared in 0.1 M sodium acetate at pH 5.1. To assess HRP losses due to adsorption on polypropylene or quartz surfaces, a comparative solution depletion method was used. Quartz was rendered hydrophilic by a minimum 10 min soak in solution of Nochromix (Aldrich) in concentrated sulfuric acid, followed by extensive rinsing with water. The same treatment was given to polypropylene containers. SELDI samples were prepared by deposition of 0.5-5 μL aliquots of HRP on the chip surface for 2-10 min. The spots were subsequently washed twice with 5 μL of water. (HPLC quality solvents were used throughout this work.) The treated area was covered with 0.5 μL of a saturated solution of matrix forming 3-(4-hydroxy-3,5-dimethoxy- † E-mail: kirkor@scs.uiuc.edu and scheelin@scs.uiuc.edu. 6278 J. Phys. Chem. B 2001, 105, 6278-6280 10.1021/jp011148n CCC: $20.00 © 2001 American Chemical Society Published on Web 06/15/2001