Rapid and numerous advances in health care are a major driving force for the development of biosensors. In health care it is imperative to obtain precise and accurate quantitative diagnostics, and nonspecific binding (NSB) plays a deleterious role in such analysis. Thus, an elimination of, or at least a reduction in, NSB is critical not only in the health care field, but also in areas such as agriculture, horticulture, veterinary medicine, physics, chemistry, oceanography, aviation, and environmental control. Areas for the application of biosensors are continuously increasing, and thus it is imperative to obtain a better understanding of NSB in a general sense, and then further fine-tune it for a particular application. Mathewson and Finley (1992) emphasize that the exponential growth in the interest in biosensor development is due to the biosensor's potential for the convenient detection of an almost limitless number of analytes in a wide variety of surroundings. It is anticipated that the combination of biotechnology with microelectronics will result in a variety of inexpensive, disposable biosensors (Dambrot, 1993; Wise and Wingard, 1991). Sensors should be reliable, simple, rapid in their measurement, and able to detect low levels of analytes, often in a mixture of similar substances. Understandably, NSB becomes more and more critical with a decrease in (or the dilution of) the analyte of interest.
Biosensors, as the name indicates, use biologically derived molecules as sensing elements. The main feature of the biosensor is the spatial unity of the biomolecules with a signal transducer (Lowe, 1985). Scheller et al. (1991) emphasize the importance of providing a better understanding of the mode of operation of biosensors to improve their sensitivity, stability, specificity, and speed of response. NSB plays a critical role in such analysis but, unfortunately, very little is known about NSB. Thus, its influence on biosensor performance parameters is difficult to estimate using a scientific basis.
The solid-phase immunoassay technique provides a convenient means for the separation of reactants (for example, antigen) in solution. Such a separation is possible because of the high specificity of the analyte for the immobilized antibody. As indicated in earlier chapters, external diffusional limitations play a role in the analysis of such assays. We need to emphasize that in most biosensor analysis, nonspecific binding has not been seriously considered. However, according to Place et al. (1985) there is a balance inherent in the practical utility of biosensor systems. These authors estimate that although acceptable response times of minutes or less should be obtainable at /uM concentration levels, inconveniently lengthy response times will be found at nM or lower concentration levels. The presence of high degrees of nonspecific binding will necessarily increase these response times. Nonspecific binding of the antigen may occur directly on the biosensor surface area where no antibody is present. Also, during the immobilization procedure on the biosensor surface, nonspecific binding of the antibody may occur. The latter case can lead to heterogeneities on the biosensor surface. Thus there is a need for analyzing nonspecific binding and for providing some factor that relates the extent of nonspecific binding to the total binding (specific and nonspecific) (Scheller et al., 1991; Sadana and Sii, 1992).
We will now analyze some examples wherein nonspecific binding is presumed present. Walczak et al. (1992) have developed an evanescent biosensor to analyze the human enzyme creatine kinase (CK; EC 22.214.171.124) isoenzyme MB form (CK MB) with a molecular weight of 84,000. Custom [1-phycoeythin CK-MB antibody conjugates were immobilized on fused silica fiber-optic sensors. There is considerable interest in the development of a biosensor for the detection of this cardiac isoenzyme creatine kinase since it would permit early detection of myocardial infarctions. Walczak et al. indicate that it is unlikely that nonspecific adsorption of the antibody conjugate accounts for a large fraction of the signal observed. These authors estimate that, for their case, nonspecific adsorption mechanisms at best can account for 25% of the accumulated signal for the lowest CK-MB
concentrations used. As expected, the relative amount of this nonspecific adsorption mechanism when compared to the specific adsorption mechanism is proportionately smaller when higher CK-MB isoenzyme samples are analyzed.
Betts et al. (1991) utilized dansylated F(ab') antibody fragments in the fabrication of a selective, sensitive, and regenerable biosensor. These authors indicate that one of the negative points is the nonspecific adsorption of the immobilized F(ab') onto the quartz substrate. The authors planned to use different substrata to help alleviate this problem.
Byfield and Abuknesha (1994) indicate that antibodies or immunoglobulins (Ig) consist of four polypeptide subunits. There are two identical small, or light, chains (L chains) and two identical large, or heavy, chains (H chains) held together by noncovalent forces. See Fig. 8.1. Also, covalent interchain disulfide bonds are involved. These authors indicate that the Fab fragments contain the key portions of the antibody molecule that contain the binding sites for the antigen. The Fab fragment itself contains a light chain and a heavy
chain. The advantage of using Fab fragments in biosensor applications is very clear: it minimizes the nonspecific binding. This is because the entire Ig molecule contains Fc sections to which nonantigen components may stick. This is different from the binding of the actual antigen to anything (such as directly to the biosensor surface), from which no signal is obtained. Extending the definition of nonspecific binding to include distortion of the measured signal, Byfield and Abuknesha indicate that a single binding site on a Fab as compared to two binding sites on an Ig molecule will minimize or prevent the formation of antigen-bridged complexes. These complexes distort the quantitative aspects of conventional immunoassays, which could lead to disastrous results, especially in the health care field.
Since not much is currently known about the science of nonspecific binding and how it influences the amount of specific binding, various attempts have been made to minimize NSB. It is certain that only under rare or unusual circumstances will NSB actually be useful in immunoassay analysis. Thus, it is useful to examine and analyze at least some examples where the influence of NSB has been minimized. Owaku et al. (1993) indicate that employing the Langmuir-Blodgett (LB) film technique produces a highly ordered and alligned antibody layer on a biosensor surface, which should help minimize this type of nonspecific binding.
Ahluwalia et al. (1991) have utilized the LB film technique to deposit oriented antibodies on a surface; the antibodies form reproducible expanded films at a gas-water interface. These films are then transferred to a solid surface to produce high-density surface coatings of the material of interest. These authors indicate that for an efficient surface one requires high-surface density of active molecules, the absence of nonspecific binding, and stability and durability.
Figure 8.2 shows the procedure for the immobilization of the antibody using hydrophilic adsorption, hydrophobic adsorption, and amino silaniza-tion. The monoclonal antiprolactin (IgGi) forms a complex with the conjugated antibody (IgG2-HRP). Here HRP is a horse radish peroxidase. The antigen is human prolactin in solution with the salts and the preservatives. Ahluwalia et al. (1991) indicate that the nonspecific binding here is due to the adsorption of salts or preservatives. Out of the three immobilization procedures utilized, the APTS (5% aminopropyltriethoxysi-lane) method yielded the smallest nonspecific binding (0.1) for the highest surface density of antibody (p = 320ng/cm2). In comparison, the adsorption on a hydrophilic surface gave p and NSB values of 41ng/cm2 and 0.67, respectively. Similarly, p and NSB values of 217ng/cm2 and 0.21, respectively, were obtained for adsorption on a hydrophobic surface. It appears that NSB is inversely proportional to the surface densities of the antibody, at least for this case.
Was this article helpful?