One unifying issue impacting all antibody driven diagnostic formats is the inevitable signal generation interference effects that sample matrix components impose on assay signal generation processes. The generic term, “Matrix Effect” was assigned to represent the sum of the effects of “all” sample composition components impacting the results of the final assay reading1,2. The severity of the signal interference impact can, depending upon the nature of the interfering agent, be greatly reduced by sample dilution factors > 20X. With respect to the sample viscosity mediator of signal inhibition/interference that is the subject of this article, the degree of signal suppression varies with the amount by which the bio-physiological sample type (serum, plasma, urine etc.) can be diluted out prior to assay. In their native undiluted form, these sample types present an environment known to suppress antibody-antigen binding kinetics. Obviously, the level of binding kinetics suppression is by no means, complete. But, when quantitatively accurate results are needed, the resulting signal inhibition can often result in underreporting of actual analyte concentrations in bio-physiological sample types. The probability of underreporting target analyte concentrations is the greatest when analyte detection sensitivity requirements limit your ability to dilute your samples by more than three-fold using your sample diluent formulation. Sample matrix complexity differentials between the e.g. serum sample and the calibrator diluent used for standard curve construction, can be exacerbated when the calibrator diluent is something simple like PBS or PBS–1% BSA. It is a generally accepted fact that the more complex the physical composition of the sample matrix, the greater potential there is for assay signal suppression leading to underreporting actual analyte concentrations in bio-physiological sample-types. In the discussion topic sections below, we lay out a logical explanation of the chemical basis behind this frustrating sample matrix inhibition phenomenon.
In order to devote a majority of the discussion content in this article to matrix viscosity inhibition issues, we chose to bypass the more well documented immune driven matrix-inhibition mechanisms. These include, low binding affinity heterophilic antibodies, IgM-rheumatoid factors, high binding affinity human anti-mouse antibodies (HAMA) and human anti-animal antibodies (HAAA). Complement, an additional potentiator of signal inhibition is not elaborated on to any great degree as it too does not involve a matrix viscosity driven mechanism. We do include a brief explanation about dealing with complement inhibition issues within the ICT sample diluent product options section at the end of this article. So, if one were able to selectively remove most of the above immune mechanism-based inhibitor proteins from your serum/plasma samples, you would still be left to contend with the remaining signal inhibition problems arising from sample viscosity differentials. So, the question that we would like to answer going forward is, how does a physical feature of the serum/plasma solvent environment like sample matrix viscosity, negatively impact antibody-antigen binding kinetics within complex bio-physiological samples?
After reviewing publications describing the various molecular entities known to cause sample matrix inhibition, it becomes readily apparent that sample viscosity, as a common source of immunoassay signal suppression, is a largely overlooked subject. That said, what really elevates the importance of sample matrix viscosity interference is its potential to occur to a variable degree in “any” bio-physiological sample solution. To begin to address the chemical basis for and consequences of, sample viscosity driven immunoassay or enzyme immunoassay (EIA) signal suppression, we needed to draw upon some published conclusions from the few publications willing to address this matrix viscosity inhibition subject3-7.
Chemical equilibrium constants as useful indicators of matrix inhibition activity
Defining the physicochemical terminology associated with assessing antibody-antigen binding kinetics
As a prelude to our sample matrix discussion, a brief review of several equilibrium constant parameter terms would be helpful. The binding affinity by which a designated antibody binds to its respective target analyte can be defined by three different but related terms. These consist of, the equilibrium dissociation constant (KD), the equilibrium association constant (KA), and the general equilibrium constant for the reaction (Keq) when equilibrium for the overall reaction is achieved. Only once equilibrium conditions have been established does KA = Keq. Putting this equilibrium constant terminology into an antibody (A) + antigen (B) à AB complex formation reaction context, [A] and [B] = concentration of the reactants A (antibody) and concentration of reactants B (antigen) respectively. Correspondingly, [AB] = concentration of the antibody antigen complex. The antibody dissociation constant is KD = [A][B] / [AB] and the antibody association constant is KA = [AB] / [A][B] which is also Keq once the antibody-antigen binding reaction has reached equilibrium, are all chemical representations of antibody binding affinity at equilibrium (especially KD) as well as a chemical representation of the relative amount (concentration) of reactants in the complexed versus the free state (Keq) at equilibrium. The chemical association constant KA provides a picture of the rate at which antigen can bind to antibody over a particular time point or reach equilibrium state. Ideally, the greatest assay detection capabilities are achieved when the concentration ratio of complex over unbound A and B are as large as possible8.
Assessing the Influence of matrix viscosity on antibody-antigen complex formation
Published investigators studying the impact of viscosity on antibody-antigen binding kinetics concluded that sample viscosity levels play a substantial role in the incidence of under reported target analyte concentration in serum/plasma sample assays. Viscosity factor elements responsible for creating an inhibitory antibody-antigen binding environment arise from molecular interactions occurring between the various serum/plasma protein components which define the sample matrix. With respect to the proteins themselves, higher molecular weight proteins with asymmetrical shapes have the greatest ability to create high sample viscosity environments. An additional factor as one would expect, is the ability of a particular protein to aggregate with itself or other proteins. The greater the aggregating properties of these proteins, the higher the sample viscosity potentials will be9.
Within serum/plasma matrix environments, as sample matrix viscosity increases, there appear to be very consistent reductions in both the KA and KD equilibrium constant values. These changes are a direct reflection of diminished antibody-antigen binding complex formation reactions3,4,6,7. Decreases in both KA and KD values were attributed to a slowing in the molecular diffusion rates affecting both the association (on) rate (KA) as well as the dissociation (off) rate (KD), values4. Higher dissociation rates equate with larger KD values. Higher binding affinity antibodies have low KD values and vice versa. But with respect to KA (on rate) values, higher binding affinity antibodies have larger KA values.
To explain the observation of decreasing KD values with increasing viscosity levels, we will use a simple low viscosity environment versus high viscosity environment example. In a low solvent viscosity environment (e.g. PBS), antigens are unhindered from diffusing away from their prior location attached to an antibody binding site. In an elevated viscosity environment (e.g. serum, plasma), the opposite condition exists. Dissociated antigens are hindered from freely diffusing away from the antibody binding site and can potentially re-associate with proximal antibody binding sites. This situation equates to lower KD values.
Looking at increasing sample viscosity from a KA association rate perspective, using the low versus high viscosity example again, in low viscosity environments (e.g. PBS), antigen molecules are unencumbered from diffusing toward antibody binding sites. When an elevated viscosity parameter is factored in, antigen diffusion kinetics are once again hindered. This leads to lower rates of fruitful collision between antigen and antibody binding sites. These events can only lead to slower/lower antibody-antigen binding kinetics = lower KA values.
Since KA values represent the ratio of antibody-antigen complex versus how much free antibody and antigen remain in solution at equilibrium, high KA value = high binding affinity antibody clones would exist in the complexed state during a higher % of time compared to in a noncomplexed state. This means that the complexed form concentration, [AB], is present at a much higher concentration than the concentration of the un-complexed forms [A] and [B]. Since antibody-antigen binding association values (KA) are expressed as the inverse molar (M-1) concentration of [AB] / [A][B] = [A][B] / [AB] high binding affinity antibodies would have a large negative exponent sign. Thus, it is easy to infer that when elevated viscosity levels increase the time to which antibody-antigen complex formation equilibrium is attained, much more antigen can be captured by a fixed amount of capture antibody within a finite time period within a low matrix viscosity sample solution well than could ever occur within a corresponding high matrix viscosity sample solution well. Within an antibody sandwich ELISA format, lower amounts of captured antigen translates into lower plate well chromogenic signal generation. And for all practical purposes, this defines the mechanism by which complex serum/plasma samples suffer from under-reporting of target analyte concentration. Or, more commonly known as matrix inhibition to 99% of research population that don’t get paid to worry about signal inhibition mechanisms.
Resolving sample matrix viscosity inhibition issues
Matrix inhibition problems not associated with the presence of HAMA or HAAA interference effectors but rather the result of viscosity-based signal inhibition mechanisms, can usually be dealt with using one of two different strategies. One of the simplest routes for resolving matrix viscosity-based interference is to simply dilute it out your problem with your calibrator diluent that was used to prepare your assay standard curve. Matrix viscosity interference can greatly be reduced by a simple 1:10 or greater dilution of your complex matrix samples into your diluent used to prepare your standard curve (calibrator diluent). The success level of this process is improved if your sample diluent contains something more than just PBS. In many cases where assay sensitivity limitations are relaxed, diluting out serum/plasma samples by > 10X in a sample diluent consisting of PBS-1% BSA will greatly alleviate much of the matrix viscosity differential problems.
When assay sensitivity requirements are a central priority during the assay development process, you don’t have the luxury of simply diluting out your matrix inhibition problem. This is where a logical strategy is necessary when determining the composition of your sample/calibrator diluent. You will need to determine which additives must be included within your sample/calibrator diluent that will create the required matrix viscosity levels to more closely match the viscosity levels present in routine serum/plasma samples. Completion of this task will take on the role of a make or break significance when it comes to the likelihood of successful completion of your assay. The process of finding an effective sample diluent formulation to match up calibrator diluent viscosity with serum/plasma sample viscosity, is a lengthy process involving multiple trial and error assay runs. Finding a resolution for approximating sample/calibrator diluent matrix viscosity with serum/plasma viscosity is a mandatory step in most antibody sandwich ELISA development ventures.
Complement inhibition of immunoassay signal generation
A quick note on this potential problem
Complement has been proven to bind to Fc fragment regions of immunoglobulins, sterically blocking the analyte-specific binding sites of antibodies2,10. Complement is composed of multiple sub-components of which C1 is the first. C1 formation is calcium dependent. That being the case, in the presence of calcium chelating agents, C1 formation is inhibited. This eliminates any immunoassay signal inhibitory problems coming from complement binding to Fc regions of solid phase adsorbed antibodies10.
Immunochemistry Technologies (ICT) assay development products
Over the years, ICT has made a deliberative effort to provide the full gamut of ELISA development solution products. These products were intended to assist novice and advanced developers alike in developing their various forms of immunoassay formats. Given the theme of this article, it is appropriate to mention ICT’s sample and assay diluent products. Both the sample and assay diluent products provide opportunities for assay development people to begin to address matrix viscosity disparities. This is the universal challenge of all assay developers that, due to the need to maximize assay sensitivity, must contend with only being able to dilute their serum/plasma samples no more than two or three-fold.
ICT offers an Assay Diluent Optimization Pack (Cat. No. 958) containing four different assay diluent formulations. We also offer a Sample Diluent Optimization Pack (Cat. No. 959) containing three different sample diluent formulations. Employment of both an assay diluent as well as a sample diluent in your assay wells allows you to employ two different matrix complexity equalization reagents within a single sample well. A key feature of all our assay diluent products is their ability to inhibit complement interference. Identifying the ideal sample diluent composition to equalize the sample/calibrator diluent matrix viscosity with the serum/plasma sample matrix viscosity, may require additional additives to be included within the sample diluent product(s). ICT’s sample diluent products provide an ideal sample diluent platform for the further addition of other matrix viscosity enhancing components. This can simplify the trial and error process to obtain a “perfect match” between sample/calibrator diluent matrix and the serum/plasma sample matrix. We stand ready to assist you with any matrix equalization questions that you may have going forward.
1. Wood, W. G. “Matrix effects” in immunoassays. Scand J Clin Lab Invest Suppl 205, 105-112 (1991).
2. Weber, T. H., Kapyaho, K. I. & Tanner, P. Endogenous interference in immunoassays in clinical chemistry. A review. Scand J Clin Lab Invest Suppl 201, 77-82 (1990).
3. Xavier, K. A. & Willson, R. C. Association and dissociation kinetics of anti-hen egg lysozyme monoclonal antibodies HyHEL-5 and HyHEL-10. Biophys J 74, 2036-2045 (1998).
4. Morgan, C. L., Newman, D. J., Burrin, J. M. & Price, C. P. The matrix effects on kinetic rate constants of antibody-antigen interactions reflect solvent viscosity. J Immunol Methods 217, 51-60 (1998).
5. May L. Chiu, W. L., Steven T. Snyder, Pak Kin Wong, Joseph C. Liao, and Vincent Gau. Matrix effects – A challenge towards automation of molecular analysis. JALA 15, 233-242, doi:10.1016 (2010).
6. Wienken, C. J., Baaske, P., Rothbauer, U., Braun, D. & Duhr, S. Protein-binding assays in biological liquids using microscale thermophoresis. Nat Commun 1, 100, doi:10.1038/ncomms1093 (2010).
7. Papaneophytou, C. P., Grigoroudis, A. I., McInnes, C. & Kontopidis, G. Quantification of the effects of ionic strength, viscosity, and hydrophobicity on protein-ligand binding affinity. ACS Med Chem Lett 5, 931-936, doi:10.1021/ml500204e (2014).
8. Reverberi, R. & Reverberi, L. Factors affecting the antigen-antibody reaction. Blood Transfus 5, 227-240, doi:10.2450/2007.0047-07 (2007).
9. Kesmarky, G., Kenyeres, P., Rabai, M. & Toth, K. Plasma viscosity: a forgotten variable. Clin Hemorheol Microcirc 39, 243-246 (2008).
10. Zapf, S. & Loos, M. Effect of EDTA and citrate on the functional activity of the first component of complement, C1, and the C1q subcomponent. Immunobiology 170, 123-132, doi:10.1016/S0171-2985(85)80085-1 (1985).