Figure 8.7. Real-time SPRI curves representing the sequence specific hybridization adsorption of (a) a 500 nM solution of complementary RNA onto an ssDNA microarray and (b) a 500 nM solution of complementary DNA onto an ssRNA microarray. The surface hydrolysis of both DNA-RNA heteroduplexes (a) and (b) was monitored at an RNase H concentration of 8 nM. The figure insets are representative SPR difference images obtained by subtracting images acquired before and after RNase H hydrolysis of heteroduplexes formed in parts a and b.

a variety of transport and energetic factors. We have recently shown (Fang et al., 2005; Lee et al., 2005b) that a simple model can be used to describe surface enzyme reaction dynamics, such as the surface RNase H hydrolysis process that forms the foundation of our enzymatically amplified SPRI measurements.

A general reaction scheme for the enzymatic processing of a biopolymer microarray that couples diffusion, adsorption, desorption and surface catalysis is shown in Figure 8.8.

Figure 8.8. A reaction scheme outlining the surface enzymatic processing of a biopolymer microarray.

Figure 8.8. A reaction scheme outlining the surface enzymatic processing of a biopolymer microarray.

The enzyme (E) first adsorbs from solution onto the surface-bound substrate (S) to create a 1:1 surface complex (ES). The surface complex then reacts to form the surface-bound product (S*) and releases the enzyme back into solution. This reaction scheme differs from the typical ELISA-based enzymatic method for detecting species on surfaces, which uses a sandwich assay in which an enzyme-protein conjugate binds to an adsorbed molecule and then reacts with a substrate in solution to create an amplified detection signal (Crowther, 1995). For the case of surface enzyme kinetics, the substrate is attached to the surface, and upon reaction of the surface complex, the enzyme is released back into solution.

The model shown in Figure 8.8 can be quantified with three rate constants (ka, kd, and kcat) and an additional parameter (j3) which describes the steady-state diffusion of enzyme to the surface. We rewrite the model for the surface enzyme reaction here in Equations (8.3)-(8.5):

where E(x=m) and E(x=0) are the bulk and surface enzyme species respectively, km is the steady-state mass transport coefficient, S is the RNA-DNA surface-bound substrate (the RNA-DNA heteroduplex), ES is the surface enzyme-substrate complex (the RNase H-heteroduplex complex), ka and kd are the Langmuir adsorption and desorption rate constants, S* is the surface product (ssDNA), and kcat is the surface reaction rate for the enzyme complex (Fang et al., 2005). The ratio ka/kd is the Langmuir adsorption coefficient KAds. The steady-state mass transport coefficient (km) can also be written as D/S, where D is the diffusion coefficient for the enzyme and S is the steady-state diffusion layer thickness (Bourdillon et al., 1999).

If the surface coverages for the three surface species S, ES, and S* are denoted as TS, rES, and rS*, respectively, and the total number of surface sites is rtot, then the surface kinetics equations can be expressed in terms of the relative surface coverages 0x = Tx/rtot, where x = S, ES, or S*:

does ka[E]b(1 - oes - 0s*) - (kd + kcat) oes dt 1 + ß(1 - Oes - Os*)

In Eq. (8.7), [E]b is the bulk enzyme concentration and ß is the dimensionless diffusion parameter (Schuck and Minton, 1996; Bourdillon et al., 1999) mentioned previously and defined by Eq. (8.9):

km D

These equations were derived in a series of recent papers (Fang et al., 2005; Lee et al., 2005b), and can be solved by simple Euler integration methods with the initial conditions 6S = 1 and 6ES = 6S* = 0 to yield three time-dependent surface coverages 6ES (t), 6s* (t), and 6S (t). We have used these computer-generated solutions to Eqs. (8.6)-(8.8) to fit our SPRI measurements and obtain the four constants ka, kd, kcat, and ß (Fang et al., 2005; Lee et al., 2005b).

SPRI measurements are just one of few possible experimental methods that have been employed to monitor surface enzyme processes in real-time. While most research efforts have focused on the use of fluorescence based detection methods (Gaspers et al., 1994, 1995; Jervis et al., 1997; Tachi-iri et al., 2000; Bosma et al., 2003; Tawa and Knoll, 2004), SPR-based techniques are an alternative tool for measuring time-dependent surface coverages of untagged adsorbing species that provides excellent discrimination against possible bulk signal contributions (Peterson et al., 2000, 2002; Goodrich et al., 2004b; Kanda et al., 2004; Shumaker-Parry and Campbell, 2004; Shumaker-Parry et al., 2004; Wegner et al., 2004b; Fang et al., 2005; Lee et al., 2005b).

Perhaps the most effective experimental method for the quantification of surface enzyme processes is to use a combination of SPR and fluorescence experiments. For example, Knoll and co-workers have recently used the combination of SPR and surface plasmon fluorescence spectroscopy (SPFS), which is a very sensitive fluorescence method for detecting labeled surface biochemical species (Yu et al., 2003; Tawa and Knoll, 2004; Yao et al., 2004; Stengel and Knoll, 2005). Kim et al. (2002) have also employed a combination of SPR and SPFS to create separate profiles of the enzyme adsorption and substrate cleavage steps. Moreover, we have recently demonstrated that the combination of time-resolved SPRI and SPFS measurements can be used for the study of RNase H hydrolysis of heteroduplexes on DNA microarrays (Fang et al., 2005).

Figure 8.9 plots the theoretical curves for 6ES (t), 6S (t), and 6S* (t) for the reaction of a 2 nM RNase H solution with a monolayer of RNA-DNA heteroduplexes formed previously by the hybridization adsorption of ssRNA onto a ssDNA microarray element. These curves were obtained by fitting SPRI data to the solutions of Eqs. (8.6)-(8.8) (Fang et al., 2005).

Figure 8.9. Theoretical analysis for the RNase H hydrolysis reaction of a surface immobilized RNA-DNA heteroduplex. The solid lines represent the simulated curves for 9es (t), 9s (t), and 9s* (t) obtained using Eqs. (8.6)-(8.8) with the parameters ka = 3.4 x 106 M-1 s-1, [E]b = 2nM, kd = 0.1 s-1, kcat = 1.0 s-1, and p = 180.

The values of the four reaction parameters in the model obtained from this fit are ka = 3.4 x 106 M-1 s-1, kd = 0.1 s-1, kcat = 1.0 s-1, and a diffusion parameter @ = 180 for the SPRI measurements (Fang et al., 2005). Note that the enzyme-substrate complex coverage (0ES) remains very small throughout the course of the surface reaction. This is because the surface catalysis rate constant (kcat) is much larger than the surface adsorption rate (ka [E]b), so that the enzyme reacts very quickly and is released from the surface. In fact, if we look at the Langmuir adsorption coefficient obtained from ka and kd (KAds = 3.4 x 107 M-1), we find that C0.5 is approximately 30 nM, so that even the equilibrium enzyme surface coverage would be low at a bulk target DNA concentration of 1 nM. However, the kcat value of 1 s-1 for RNase H is approximately 10 times slower than that observed for the RNase H reaction in solution (Hogrefe et al., 1990; Keck et al., 1998), so improvements in the surface attachment chemistry are still possible.

0 0

Post a comment