Laboratory Diagnosis in Neurology, 1 Ed.

4 Principles of Analytical Methods

H. Reiber


Antigen–Antibody Binding

Immunochemical analysis is modeled on the biological process of an immune reaction in which antibodies are produced against foreign antigens and the antigens are then eliminated as immune complexes. The formation of an immune complex, which is very specific evidence of a macromolecule, is now used as the basis for a multitude of different procedures for demonstrating the presence of molecules in body fluids and tissues. The range of techniques for quantitative detection of these immune complexes has become extremely wide; the following selection describes only the most commonly used.

Properties of Antigen–Antibody Binding

Immune complex formation. The binding of an antibody molecule to its specific antigenic determinant (epitope, see below) with the corresponding antigen creates a very stable immune complex due to the often extremely high affinity between the two molecules (Gharavi and Reiber, 1996) (Fig. 4.1).

Reaction equilibrium. Two molecules that can react with each other or that can bind to each other, such as an antigen (Ag) and an antibody (Ab), form an equilibrium between their bound state (Ag-Ab complex) and their dissociated state (free Ag and Ab molecules), as described by the following equation:


The probability with which the molecule is found in one of the two states (bound or unbound) depends on the equilibrium constant (K), which equals the association rate constant (k1) divided by the dissociation rate constant (k2). Equilibrium is reached if k1 × A × B = k2 × AB.

The concentration of an immune complex (AB) in equilibrium depends on both the equilibrium constant (affinity) and the concentrations of antigen and antibody.

The practical consequence is that raising the antibody concentration in the solution (e. g., at low antigen concentration) has the effect that more immune complex is formed: the velocity (vk1 × Ag × Ab) rises as the antibody concentration rises, and the equilibrium shifts in the direction of the immune complex (Ag-Ab), increasing the sensitivity of the test.

Avidity. Since both viral epitopes and antibody molecules are multivalent, their association calls for a more general presentation of the equilibrium constant, i. e., one that incorporates the effective antibody valency and the effective antigen valency. The maximum valency of the IgG molecule is 2, that of the IgA molecule is 4, and that of the IgM molecule is 5. The effective valency of viruses (with up to 1000 identical antigenic subunits) depends very much on the steric hindrance between the antibody molecules on adjacent epitopes. This physically somewhat imprecise overall attraction between an antibody and a complex antigen is called avidity (rather than affinity). For methods of determining avidity, see Gharavi and Reiber (1996).


Fig. 4.1 a–d Formation of immune complexes.

a Antibody excess.

b Equivalence (optimal for precipitation).

c Antigen excess (impedes cross-linking).

d Heidelberger-Kendall curve for the intensity of scattered light as a function of antigen concentration at constant antibody concentration. Methodologically, the measuring method needs to avoid the zone of antigen excess because of the falsely low concentration measurements that result: the signal measured in this zone (right-hand point) mimics a 10-fold lower antigen concentration (left-hand point).

Specificity. The binding strength (avidity) of two macroglobulins grows with increasing numbers of complementary binding sites. This capacity of the macromolecules to reduce the dissociation constant significantly is a newly emerging property known as the binding specificity between two molecules. This means that an antibody's specificity (a quality) for a given antigen is defined by its higher affinity (a quantity) for it, compared to its affinities for other antigens.

Cross-reactivity is therefore only a question of avidity: i. e., the chances of cross-reactivity increase with decreasing avidity of the antibody.

Epitopes and Antigen–Antibody Complex

Epitopes. Proteins carry several antigenic determinants (epitopes) on their surface. These epitopes are limited structures formed by 6–8 amino acids or by carbohydrate residues. The protein–protein interaction is based on:

• Formation of noncovalent hydrogen bonds between amino acids.

• Electrostatic and hydrophobic properties.

• Van der Waals forces between the reaction components.

Antigen–antibody complex. Antisera used for protein determination usually contain different antibodies directed against many different epitopes. Furthermore, since every antibody molecule (IgG class) has two identical antigen binding sites, a three-dimensional molecular aggregate is created during formation of the immune complex with a protein antigen (Fig. 4.1 b). This is the basic principle of agglutination and precipitation methods. The classic precipitation reaction, corresponding to the equivalence zone of the Heidelberger-Kendall curve (Fig. 4.1 d) (Heidelberger and Kendall, 1935) is also the basis of immunodiffusion methods (see below). Nephelometry and turbidimetry, including particle-enhanced reactions, are used in the zone of antibody excess (Fig. 4.1 a,d).

Methods of Immune Complex Analysis

Many physical, optical, and chemical methods have been developed for determining the amount of immune complexes formed. Agglutination tests are widely used for semiquantitative analysis (screening, titer determination). The most common methods for quantitative detection of immune complexes include measuring the distances of precipitation lines, photometry of the scattered light in solution, or indirect enzymatic color reactions of the solid phase-bound complex. Qualitative immunochemical detection of specific proteins is also possible in electrophoresis gels.

The following methods are described in more detail:

• Agglutination.

• Radial immunodiffusion and electroimmunodiffusion.

• Nephelometry and turbidimetry.

• Immunoassay.

• Immune detection using electrophoresis techniques.

• Immunohistochemistry.


Relevance. The oldest, fastest, and least complicated tests for immune reactions use agglutination methods; they have stood the test of time primarily for screening and semiquantitative analysis (Zane, 2001). Many of the commonly used rapid tests in microbiology are based on this method (see Chap. 5, “Rapid Antigen Tests”).

Principle. Bacteria or cells with their specific surface antigens are aggregated by bivalent antibodies (Fig. 4.1 b) into large complexes which become immediately visible as agglutination. Such reactions are carried out on test strips, in test tubes, or on microtiter plates. Instead of antigens present on biological material, any antigen can be bound to a large carrier particle and thus may also be used for the detection of antibodies by agglutination.

Advantage. These semiquantitative methods (titer determination) have the advantage that they permit fast screening and, above all, require little laboratory equipment.

Precipitation Methods

Relevance. Radial immunodiffusion (Mancini et al., 1965) and electroimmunodiffusion (Laurell, 1966) are still commonly used worldwide.


• Radial immunodiffusion (Fig. 4.2 a): The antigen-containing sample (e. g., albumin) diffuses into the gel containing the corresponding antiserum. The size of the precipitation ring formed by antigen and antibodies in the equivalence zone depends on the concentration of the antigen. The diameter of the ring is measured, e. g., using a caliper.

• Electroimmunodiffusion (Laurell's rocket electrophoresis, Fig. 4.2 b): As in radial immunodiffusion, the antigen-containing sample diffuses into a gel containing the corresponding antiserum. When an electrical field is applied, the molecules move according to their electrophoretic mobility in the direction of the field, as they do during electrophoresis (see below). The length of the rocket-shaped precipitation arc is measured, and the amount of antigen is determined by reference to a standard curve.


Fig. 4.2 a, b Measuring techniques used in the equivalence zone of the antigen–antibody concentration ratios (precipitation methods).

a Radial immunodiffusion with a central well for sample application. The maximum precipitation ring is formed after about 12–18 hours; its diameter depends on both the constant antibody concentration within the gel and the concentration of the antigen in the sample.

b Electroimmunodiffusion (Laurell's rocket electrophoresis). The antigen migrates into the antibody-containing gel in the direction of an applied electrical field until it reaches the equivalence point; the height of the resulting precipitation arch correlates with the concentration of the antigen.

Advantages. Both procedures have the advantage that they do not require major equipment. Radial immunodiffusion needs no electrical equipment, whereas electroimmunodiffusion requires a power supply. The agarose gels in which the reaction takes place are easy to produce in the laboratory.

Nephelometry and Turbidimetry


Definition of Nephelometry and Turbidimetry

The formation of antigen–antibody complexes creates large molecules that scatter light (Fig. 4.3):

• Measuring the intensity of the scattered light is nephelometry. The set-up corresponds to that of fluorescence photometry.

• Measuring the reduction of incoming light caused by scattering is turbidimetry. The set-up corresponds to that of absorption photometry.

A light beam hitting an antigen–antibody aggregate in solution is scattered. Depending on the relationship between the size of these particles and the wavelength of the light (Rayleigh, 1871; Debye, 1947; Reiber 1983), the intensity distribution of the scattered light in space will vary (Fig. 4.4). The strongest scattered light signal is obtained when the particles are relatively large and forward scatter is measured. If the measuring angle of the photodetector is 70°–90° (Beckman Nephelometer), particle size is still important for the total intensity of the scattered light, but changing the particle size has less effect on the distribution of the intensity of the scattered light. As a particular application for total protein analysis (Reiber, 1983) the dynamics of protein aggregation in a highly viscous solution (TCA) allow detection of how a nephelometric signal increases to a maximum and decreases again as a consequence of slowly changing particle size without flocculation (slowly growing particles induce the change from Rayleigh to Rayleigh-Debye scattering).


Fig. 4.3 Comparison of the principles of turbidimetry and nephelometry. Turbidimetry records the decrease in light intensity (I T) caused by particles that scatter the incoming light (I0). In nephelometry, the scattered light (IN) is measured directly. The different sensitivities of these two methods are explained by the fact that, at low concentrations, turbidimetry determines the difference between two large signals (I0 – IT), while nephelometry determines the difference between two small signals (IN – IB), with a smaller degree of imprecision. (Blank value, IB= 0.)


Physical measuring set-up. The concepts of turbidimetry and nephelometry follow those of absorption photometry and fluorescence photometry—with the exception that scattered light (same wavelength) is measured, rather than the absorption and emission of light (different wavelengths) (Fig. 4.3). In both absorption photometry and turbidimetry, the difference between two large light signals at low sample concentrations is determined; these procedures are therefore less sensitive than fluorescence photometry or nephelometry, where the difference between small light signals at low sample concentrations is determined, with a correspondingly smaller imprecision (see legend to Fig. 4.3).


Fig. 4.4 a–c Light scattering as a function of particle size. Double arrows represent the incoming beam of light, and single arrows show the direction and intensity (length of arrows) of the scattered light; d, diameter of particle; λ, wavelength of incoming and scattered light.

a Rayleigh scattering: When the particle size is significantly smaller than the wavelength (d<< λ), the light is scattered symmetrically in the direction of the incoming light as well as against the incoming light.

b Rayleigh-Debye scattering: When the particle size is only a little smaller than, or is equal to, the wavelength (d ≤ λ), asymmetric forward scattering is observed.

c Mie scattering: Asymmetric forward scattering increases significantly when the particle size is larger than the wavelength (d> λ).


Fig. 4.5 Measuring technique used in nephelometry. The higher the protein concentration, the steeper the curve, i. e., the faster the scattered light reaches maximum intensity. The endpoint method determines the maximum intensity of the scattered light (I max). The two-point method (fixed-time method) measures the difference in intensity between two preset time points (t). Kinetic nephelometry (rate method) measures the maximum rate of complex formation. This corresponds to the ascent at the inflection point of the reaction curve and to the maximum of the differential curve ([d I/d t]max).

Nephelometric analysis. In addition to the optical conditions of light scattering (Figs. 4.34.4), there are various ways of analyzing the dynamics of light scattering due to the dynamics of the immunochemical reaction (Fig. 4.5):

• Endpoint method: The immunochemical reaction gives rise to a steady increase of the light scatter signal I with time (Fig. 4.5), which is recorded at the maximum of light intensity, I max, reached after 0.5–4 hours, depending on the antigen and its concentration. This maximum of light intensity is stable for different lengths of time depending on the stability of the homogeneous suspension of the immune complex in solution.

• Two-point method (fixed-time method): The intensity I of scattered light is determined by measuring at two points in time, for example, 10 seconds and 6 minutes after the start of the reaction (Fig. 4.5). The difference I between the two values is correlated with the antigen concentration.

• Kinetic nephelometry (rate method; Sternberg, 1977): This method determines the maximum reaction rate of immune complex formation by differentiating the curve I/t to get dI/dt (lower curve in Fig. 4.5). The maximum value of the reaction rate thus obtained (i. e., the ascent at the inflection point of the reaction curve) is then correlated with the antigen concentration.

Advantages and disadvantages. Endpoint determination has the disadvantage of unstable endpoint values and varying measuring times, as the time until the endpoint is reached depends on the concentration of the antigen. The two-point method uses a fixed measuring time period, which must be shorter than the time taken to reach the endpoint—a limitation for fast reactions at high concentrations with a steep curve and an early maximum value. Because of its shorter measuring time and independence of initial values, the rate method is more robust, but it is basically less sensitive than the two-point and endpoint methods since the inflection point is more difficult to determine at lower concentrations.

Particle-Enhanced Nephelometry

The sensitivity of the nephelometric detection method can be increased by coating latex particles with immunoglobulins. This results in much larger complexes during the immunochemical reaction (Fig. 4.6), and the intensity of the scattered light increases (“particle-enhanced nephelometric test”). This is particularly important when testing at low concentrations, such as with CSF. These methods are used especially in CSF analysis for IgA and IgM.

Heidelberger-Kendall Curve

A special problem in immunochemical nephelometry is the possibility that, unnoticed, falsely low values are measured in the sample when it contains unexpectedly high amounts of antigen, i. e., an antigen excess (Heidelberger and Kendall, 1935).

Dependence of the detection signal on the relative amounts of the reaction components. The size of the antigen–antibody complex (immune complex) depends on the relative amounts of the reaction components (Fig. 4.1). When there is excess of antibody (Fig. 4.1 a), every increase in the amount of antigen leads to further cross-linking, which is most complete in the zone of equivalence (roughly equal concentrations, Fig. 4.1 b). The solubility of the antigen–antibody aggregates is lowest in this zone. When there is excess of antigen (Fig. 4.1 c), not all antigens become complexed. Furthermore, because the antibodies are saturated with antigen, thus preventing the formation of large complexes, the aggregates of molecules are smaller and more soluble, and the intensity of scattered light decreases accordingly (Fig. 4.1 d).

To take the example of an IgM paraprotein present at an unexpectedly high concentration in the sample, this situation results in a low signal, which is falsely interpreted as a low concentration (as indicated in Fig. 4.1 d by the two points on the curve, which demonstrate how in this case the true concentration is 10 times higher than indicated by the falsely low signal measured in the zone of antigen excess). Particularly in CSF analysis, there is often considerable variation between protein values related to intrathecal synthesis and those related to the solely blood-derived, normal fraction in CSF (sometimes more than 100-fold). This makes development of automated methods for CSF analysis a much more demanding challenge than for serum analysis.

Preventing Misinterpretations

image A falsely low value in the zone of antigen excess can be recognized if the measuring signal increases when more antibodies are added to the reaction mixture.

image A further possibility of avoiding errors is to start the analysis of immunoglobulins generally with diluted samples (based on the total protein content of CSF samples).

image In CSF analysis, the evaluation with quotient diagrams may point to antigen excess because of implausible data combinations.


Fig. 4.6 Particle-enhanced immune reaction. The detection signal of an immunochemical reaction is considerably enhanced when a reaction component is coupled to larger particles. This creates larger immune complexes with a higher intensity of the scattered light and thus increases the sensitivity of the test.


Relevance. Immunoassays permit very sensitive measurements with detection limits in the ng/L range (10−9 g/L). This makes immunoassays the most sensitive quantitative protein detection methods for routine analysis in the clinical chemistry laboratory.

Principle. In an immunoassay (Crowther, 1995), an antigen or antibody is bound to the wall of a test vessel, such as the well of a microtiter plate (Fig. 4.7). When the body fluid to be analyzed is added, the corresponding complementary molecule (antibody or antigen) is bound. The resulting antigen–antibody complex on the wall of the test vessel does not require the formation of large aggregates, but for sensitive detection needs a signal-producing indicator conjugated with a second antibody. Depending on the indicator, the different kinds of immunoassay are (Zane, 2001):

• Radioimmunoassay (RIA).

• Enzyme immunoassay (EIA or ELISA, Fig. 4.7).

• Chemiluminescence immunoassay (LIA).

• Fluorescence immunoassay (FIA).


In the past, the isotope 125I has been used as the indicator of choice (Fig. 4.7 d), and the detection of spontaneous decay of the radioisotope was measured in a scintillation counter. Because of the health hazards associated with gamma rays, the radioimmunoassay, which is highly sensitive, has now been largely abandoned.


Fig. 4.7 a–d Principle of an immunoassay.

a The wall of the well (e. g., of a microtiter plate) is coated with the antibody specific to the antigen to be detected.

b Incubation with the sample leads to binding of the specific antigen (image).

c After washing, another specific antibody (conjugate) is added, which recognizes the antigen–antibody complex (sandwich principle).

d This second antibody is conjugated with an indicator (*), e. g., an enzyme, an isotope, or a fluorescence label. The different detection reactions in these immunoassays (EIA, RIA, or FIA, respectively) are described in the text. The concentration of the antigen is calculated by reference to a standard curve analyzed in the same analytical run.

Enzyme Immunoassay

Of the various detection methods, enzyme immunoassays (EIA, also called ELISA—“enzyme-linked immunoassay”) have gained wide acceptance because their detection signal is stable and they employ simple photometric detection. In CSF analysis, the ELISA is primarily used for analysis of specific antibodies (e. g., against measles, rubella, herpes zoster, Borrelia spp.) in CSF and serum. Figure 4.7illustrates the basic principle of a “sandwich” ELISA. The surface of the microtiter plate is coated with antigen, the sample is added, and an anti-IgG (or anti-IgA, or anti-IgM) antibody conjugated with horseradish peroxidase (HRP) is added. Occasionally, alkaline phosphatase (AP) is used as a conjugate. These marker enzymes, in particular HRP, are popular for the following reasons:

• They do not naturally occur in patient samples.

• Their activity is highly specific and is not changed by binding to an antigen or antibody.

• They are stable during the tests.

• They are inexpensive and easy to adapt to automated procedures.

Best Practice in CSF/Serum Analysis with ELISA

image For analysis of antibodies, immunoglobulins, or antigens in the CSF, the matched CSF and serum samples should always be examined in the same analytical run. CSF and serum are used in appropriate dilutions. The optimum dilution ratio of CSF to serum for the analysis of IgG classes has proved to be 1:200. The following sample dilutions are used for the determination of a variety of antibodies: CSF, 1:15 and 1:45; serum, 1:3000 and 1:9000.

image Evaluation should, if possible, be by reference to a standard curve with arbitrary concentration units (Reiber and Lange, 1991). A comparison of optical density (OD) values or titers is not suitable for CSF analysis. For CSF analysis, aliquots of a self-made serum pool containing an adequate concentration of the antibody in question suffice as a standard. The serum pool is prepared in such a way that an absorption of about 2.0 OD is measured for the highest standard value in the test system. For this OD, a concentration of 100 arbitrary units (AU) is established, and 5 standard values up to 2.2 AU are obtained by serial dilution (factor 1:2.8). OD values of less than 0.1 or greater than 2.0 are not used. With two sample dilutions, mean values are not used; instead, CSF and serum values are identified that are as close together as possible. If the concentration difference between the two values from two dilutions of a sample is more than 15–20%, the run must be repeated. (Incorrect standard curve? Standard curve does not fit sample dilution series?)

image Evaluation using the antibody index (see Chap. 5, “Antibody Index”) does not usually require absolute values, since the arbitrary concentration units are eliminated by formation of the dimensionless CSF/serum quotient. To quantify pathogen-specific antibodies (antibody concentration in mg/L), a special procedure (Conrad et al., 1994; Quentin and Reiber, 2004) is available and may be used generally for every antibody species.

image For CSF analysis, the occasionally offered “capture” ELISA (binding competition) is often not sensitive enough, because extension of the incubation time, unlike with the “sandwich” assay, can not result in improved sensitivity.

Chemiluminescence Assay

Some chemical compounds emit light when oxidized. This feature is exploited as a marker in the chemiluminescence assay (LIA). The most common compounds are luminol, acridinium ester, oxalate, ruthenium complexes, and dioxetane derivatives. Acridinium ester and luminol derivatives are used preferentially; they change with H2O2 into an excited intermediate state, from which they return to the ground state by emission of light. Detection requires a luminometer. Chemiluminescent molecules are used as direct and indirect markers:

• Direct markers are bound directly to the reagents (antibodies, antigens, or DNA samples).

• Indirect markers are used as a substrate for enzymes, such as alkaline phosphatase, HRP, and β-galactosidase.


Fig. 4.8 a, b Comparison of electrophoresis and isoelectric focusing (Westermeier, 1990). A and B are the sample components to be separated.

a Electrophoresis: The different electrophoretic mobilities of molecules A and B (m RA and m RB) lead to their separation in the electric field as they migrate in the same direction. If the electric current were applied indefinitely, the bands would continue to migrate, becoming wider, until eventually they would disappear into the electrode buffer.

b Isoelectric focusing: Molecules A and B with different isoelectric points (pIA and pIB) migrate through a pH gradient made up of ampholytes in an electric field. When they reach the pH in the gel at which their overall charge is neutral (depending on their isoelectric points), they come to a stop and form two sharp bands. For further explanation, see text and Fig. 4.10.

Fluorescence Immunoassay

Fluorescent labels, such as fluorescein isothiocyanate (FITC), are molecules that can be easily linked to other molecules (Fig. 4.7 d). Excitation by light causes these molecules to change into an energetically higher state, from which they return to the ground state by emitting photons of a longer wavelength. In fluorescence immunoassays, the excitation is caused by light, whereas in chemiluminescence immunoassays it is caused by a chemical reaction.

Qualitative Electrophoretic Methods with or without Immunodetection

Relevance and Principle of Electrophoresis

Relevance. Electrophoretic separation of proteins by means of electrophoresis, immunoelectrophoresis, and isoelectric focusing has, through combination with immunochemical detection reactions such as immunofixation and immunoblot, been the means by which these qualitative procedures have become standard procedures in clinical chemistry. While serum electrophoresis is important for identifying changes in protein patterns in the blood (Figs. 4.8 a4.9), isoelectric focusing (Fig. 4.8 b) is important in CSF analysis for identifying intrathecal IgG synthesis.

Principle. In a field of direct electrical current, charged molecules migrate towards the electrode with the opposite polarity to their own (Fig. 4.8) (Westermeier, 1990). The electrophoretic mobility, i. e., the rate of migration, depends on the following factors:

• Isoelectric point (pI) of the charged groups.

• Molecular size.

• Type, concentration, and pH of the buffer.


Fig. 4.9 a, b Densitograms of the serum electrophoresis of CSF samples.

a Normal CSF sample. Representative levels are: transthyretin (TT, 3.1%, a typically high level for CSF relative to serum), albumin (61.7%), globulins α1 (2.6%), α2 (7.2%), β (9.1%), and γ (16.3%).

b Pathological CSF sample. The high peaks in the γ region indicate an IgG reaction, but they do not justify the term “oligoclonal IgG.”


Fig. 4.10 Formation of a carrier ampholyte pH gradient in an electric field (from Westermeier, 1990). The figure shows the distribution of ampholytes before the start (no pH gradient), after an intermediate period of time, and when equilibrium is reached (linear pH gradient). The pH gradient is created by an electric field applied to the isoelectric focusing gel at the commonly used concentration of 2–2.5% carrier ampholytes. This initially results in a uniform average pH. Almost all carrier ampholytes are charged; those with higher isoelectric points (pI) have a positive charge, those with lower pI have a negative charge. When an electric current is applied, the negatively charged carrier ampholytes migrate to the anode while the positively charged ones migrate to the cathode, with their migration rates depending on the net charge. Carrier ampholytes of low molecular weight have a high diffusion rate in the gel. This means that they continuously and quickly diffuse away from their isoelectric point and migrate back to it upon electrophoresis. It is therefore necessary to focus sufficiently long in the electric field to create a linear pH gradient (right-hand diagram).

• Temperature.

• Field strength.

• Nature of the carrier medium.

Protein Electrophoresis

Serum electrophoresis. Serum proteins are separated into groups by subjecting them to an electric field on a carrier material. A buffer with a pH of 8.6 serves as a separating agent in which all serum proteins have a negative overall charge (Fig. 4.8 a). Depending on the amount of charge, they migrate at different rates, i. e., different distances toward the anode. Once the separation is finished, the proteins are stained. The stained electrophoretogram is evaluated visually, or scanned with a densitometer and evaluated quantitatively (Fig. 4.9).

For the purposes of CSF analysis, serum electrophoresis (Fig. 4.9) is completely outdated.

Immunoelectrophoresis. Immunoelectrophoresis is a special type of protein electrophoresis. It is primarily used for analyzing immunoglobulin status in serum (paraproteinemia) and for the detection of immunoglobulin light chains (Bence Jones proteins) in the urine. For CSF analysis this method was introduced in rocket electrophoresis (see above, “Methods of Immune Complex Analysis”; Fig. 4.2 b).

SDS Gel Electrophoresis (One-Dimensional)

As an important method for molecular-size-dependent separation of biological protein mixtures, SDS gel electrophoresis has stood the test of time and is increasingly used for routine diagnosis.

For SDS gel electrophoresis, the proteins are treated with sodium dodecyl sulfonate (SDS). Depending on the size of the molecules, considerable amounts of SDS are bound, thus increasing the electrophoretic mobility of the protein molecule. By blotting (see Fig. 4.11), test strips can be produced that contain, for example, various antigens of a microorganism (e. g., Borrelia burgdorferi). These test strips can then be used for qualitative immunochemical detection of specific antibodies in body fluids (see the example of serum and CSF in Fig. 4.12).

Isoelectric Focusing

Principle. Isoelectric focusing (IEF) (Delmotte and Gonsette, 1977) takes place in a pH gradient (Fig. 4.8 b) and is performed exclusively with amphoteric substances, such as peptides or proteins. The method uses polyacrylamide or agarose gels in which a pH gradient is formed by mixing polyelectrolytes (ampholytes) (Fig. 4.10). The protein molecules migrate, depending on their charge, in the direction of the anode or cathode until they arrive at the pH in the gradient at which their charge becomes neutral (Fig. 4.8 b). For each protein, the pH at which this occurs is its isoelectric point (pI). Since protein molecules are not charged at their isoelectric point, the electric field ceases to have an effect on them. If they move away from this place, due to diffusion, they regain a net charge and are transported back to their isoelectric point by the electric field. This has a concentrating effect, hence the name isoelectric focusing. Examples for oligoclonal IgG are shown in Figs. 4.134.14, and 4.15.


Fig. 4.11 Most important steps in the blotting of electrophoresis gels (from Westermeier, 1990). Example: qualitative antibody detection in the CSF or serum. Pathogen-specific antigens with bound SDS are separated according to molecular size in a polyacrylamide gel. After electrophoretic transfer onto a carrier membrane (nitrocellulose), the nonoccupied binding sites of the membrane are saturated with a blocking reagent (e. g., low-fat milk solution, bovine serum albumin). This blocks nonspecific binding during the rest of the analytical run (Fig. 4.12). The membrane is cut into strips which are then used for incubation with the sample (CSF or serum) to detect the particular ligands (e. g., different antibody species against a single microorganism). The detection of immune complexes on the strips follows the usual procedures (Fig. 4.12).

Interpretation. Isoelectric focusing has a very high separation efficiency; it even separates protein isoforms with different charges, e. g., because they have different carbohydrate residues. The method thus permits detection of different IgGs in the CSF, i. e., it separates IgG molecules produced in different plasma cell clones (oligoclonal IgGs), or the monoclonal IgG fractions with different post-translational processing produced in a single plasma cell clone (Delmotte and Gonsette, 1977; Wurster, 2005).

pH Gradients with Free Carrier Ampholytes

To achieve high resolution with reproducible separation, a stable and continuous pH gradient with constant conductivity and buffer capacity is required (Fig. 4.10). The theoretical basis for the preparation of natural pH gradients originates from Svensson (1961) and its practical realization from Vesterberg (1969).

Ampholytes. Synthesis of a heterogeneous mixture of isomeric, aliphatic oligoamino/oligocarbonic acids yields a spectrum of low-molecular-weight ampholytes whose isoelectric points are very close together. Because the spectrum of ampholytes varies from manufacturer to manufacturer, it is advisable to use a mixture of ampholytes from different manufacturers, in order to obtain a pH gradient as homogeneous as possible.

Pitfalls with Ampholytes

Using ampholytes from a single manufacturer can result in severe inhomogeneity of the pH gradient, with loss of sensitivity and interpretation problems due to band artifacts (Wurster, 2005).

Fixation. When the electrical field is turned off, the focused proteins immediately diffuse, leading to widening of the initially sharp bands. It is therefore important for the result of isoelectric focusing that the proteins are fixed as quickly as possible, which is easily done when a simple protein stain (e. g., silver staining) is used. When immunodetection methods are used, fixation is necessarily delayed, thus yielding wider bands than are obtained with immediate silver staining. However, this loss of sensitivity is compensated by the immunochemical method, in which amplification by enzymatic staining increases the sensitivity of the test.

Electrophoresis Gels

The choice between agarose and polyacrylamide gels is a question of pore size and the intended further processing. For isoelectric focusing (see above), the combination of agarose gel and immunoblot is often used (Figs. 4.134.14), or else the combination of polyacrylamide gel with immunofixation or direct protein staining (Fig. 4.15).

• Polyacrylamide gel (PAG). The pore size of polyacrylamide can be adjusted precisely with its components acrylamide and methylene bisacrylamide. The gels are chemically and mechanically very stable (Raymond and Weintraub, 1959). Prefabricated gels polymerized on a carrier membrane are commercially available.

• Agarose gel. Agarose gels (Serwer, 1980) are the choice when large pores are required for analyzing molecules more than 10 nm in diameter. Agarose is a polysaccharide manufactured from marine red algae. It is dissolved by boiling in water, and it gels when cooling. During the sol–gel transition, double helices form and aggregate laterally to form relatively thick fibers. By removing the agaropectin, different degrees of electro-osmosis and purity are obtained. The gels are characterized by their melting temperature (35–95°C) and the degree of electro-osmosis. The pore size depends on the agarose concentration. The weight of agarose and the water volume are registered for general orientation: since the unavoidable loss of vapor during boiling varies from batch to batch, an intended ratio can never be absolutely exact. Generally speaking, gels used have pore sizes between 150 nm at 1% (w/v) and 500 nm at 0.16% agarose.


Fig. 4.12 Western blot for detection of Borrelia antigens (Wilske, 2004). This commonly used method determines the antibody spectrum in CSF and serum of patients infected with Borrelia spp. It uses strips of a prefabricated blotting membrane (see Fig. 4.11). These strips are incubated for 2 hours with the sample (CSF or serum) and then incubated for 45 minutes with peroxidase-conjugated antihuman antibody (IgG- or IgM-class). After 5–10 minutes following the addition of a staining solution (enzymatic reaction), insoluble color bands form on the test strip at the sites occupied by antibodies. The figure shows the bands of a patient's serum and CSF. The sample dilutions are chosen in a such way that similar total IgG or IgM concentrations are used in CSF and serum. The table shows very different results for the antibody indices (AI) obtained in two different ELISAs, one using total lysate (with p41, flagellin, but no VlsE) and the other using recombinant antigens (no p41, but VlsE). The AI results obviously depend on the manufacturer of the ELISA! (Since most manufacturers have now begun to include VlsE antigens in their assays, the discrepancies have diminished.) This result also indicates that, despite similar total IgG concentrations in CSF and serum samples, it is impossible to obtain a matching specific antibody ratio in the blot. It is in this greater sensitivity for detection of intrathecal synthesis that the decisive advantage of ELISA over blot lies.

Electrophoretic Methods with Immunodetection

For immunodetection of proteins in gel electrophoresis, two techniques are mainly used:

• Immunofixation.

• Immunoblotting.

The immunochemical reaction during immunofixation takes place directly in the gel (preferably polyacrylamide), whereas immunodetection in the immunoblot takes place after transfer of the proteins from the gel (preferably agarose) to a blotting membrane (Fig. 4.11).


Samples from patients are applied to several lanes of a thin agarose or polyacrylamide gel, and their proteins are separated in parallel by electrophoresis. Following separation, each lane is overlaid with a different specific antiserum. The antibodies diffuse within a short time into the gel and form insoluble complexes with their corresponding antigens. These complexes remain in the gel during subsequent washing steps, whereas other proteins in the sample and unbound antisera are washed out. The final step involves staining of the remaining bands with a protein stain.

Laboratories that perform isoelectric focusing in polyacrylamide gels often use immunofixation. A special example is immunoelectrophoresis of serum, where agarose gel is usually used as the carrier material. Because resolution and reproducibility of the separation is better in agarose and polyacrylamide gels, electrophoresis with cellulose acetate membranes has increasingly been replaced by gel electrophoresis. After electrophoretic separation, antiserum against a single protein or against groups of proteins is placed between patient and reference samples into a groove parallel to the direction of migration. Antibodies and separated proteins diffuse toward each other. Within about 24 hours, sharp arched precipitation lines form in the equivalence zone. Immunoelectrophoresis is evaluated visually by comparing the precipitation lines of paired samples.


Fig. 4.13 Isoelectric focusing in agarose gel with immunoblot. The parallel analysis of cerebrospinal fluid (CSF) and serum (S) is illustrated for the classic types 1–5 according to the international consensus (Andersson et al., 1994).

Type 1: No bands in CSF and serum.

Type 2: Oligoclonal IgG bands in CSF, none in serum → intrathecal IgG synthesis.

Type 3: Oligoclonal bands in CSF (like type 2) plus identical oligoclonal bands in serum and CSF (like type 4) → intrathecal IgG synthesis.

Type 4: Identical patterns of oligoclonal bands in CSF and serum → no intrathecal IgG synthesis, but systemic immune reaction.

Type 5: Identical patterns of monoclonal bands in CSF and serum → systemic paraproteinemia.


Fig. 4.14 a–d Isoelectric focusing in agarose gel with immunoblot. These four cases may help in resolving practical difficulties of interpretation of the types in Fig. 4.13. CSF, cerebrospinal fluid; S, serum.

a A few oligoclonal bands in the CSF, no bands in the serum → type 2, intrathecal IgG synthesis. Note: The number and location of bands are uninterpretable.

b A single IgG band in the CSF, which we do not consider to be oligo clonal IgG → type 1, no intrathecal IgG synthesis. Note: A study of 100 cases with a single band in the CSF (D. Mehwald, unpublished) showed that only 16% of these were referable to inflammatory processes, most of them identified by increased cell counts or intrathecal antibody synthesis (antibody index). In 50% of cases with a single band, there was blood contamination of the CSF (artificial or subarachnoid hemorrhage). Depending on the inhomogeneity of the ampholyte pattern in some laboratories, three bands may be regarded as necessary for definite identification of type 2.

c A rare case of oligoclonal IgG pattern in CSF (like type 2) with additional identical monoclonal bands in CSF and serum (like type 5) → similar to type 3 (Fig. 4.13): intrathecal IgG synthesis and, in this case, systemic paraproteinemia.

d Identical patterns of monoclonal bands in CSF and serum, with the bands clearly derived from two different monoclonal paraproteins → type 5: systemic and, in this case, biclonal paraproteinemia.


Fig. 4.15 Isoelectric focusing in polyacrylamide gel with silver stain. In contrast to immunodetection, this presentation shows the albumin region (together with the site of sample application) at pH < 5.0. The single band in the alkaline region at pH 9.3 from cystatin C (g-trace protein) can be used for discrimination of the CSF sample from serum sample. The pH range of the gradient starts on the right at pH 3.5 with the anode (+) and ends at pH 10.5 with the cathode (–). In contrast to the direct staining of the gel shown here, the pattern of immunoglobulin bands in the immunoblots (Figs. 4.13 and 4.14) is presented as a mirror image. The pH range in the blots is restricted as the nitrocellulose membrane for the blot is laid out starting at pH 6.5.


Blotting is the transfer of large molecules to the surface of an immobilizing membrane (Fig. 4.11). This method extends the detection possibilities for electrophoretically separated fractions, since the molecules absorbed onto the membrane surface are freely accessible to high-molecular ligands, such as antigens, antibodies, lectins, and nucleic acids. Prior to the specific test, unoccupied binding sites must be blocked by substances that do not participate in the detection reaction (Fig. 4.11).

Blotting membranes. Nitrocellulose is the most commonly used membrane material. The pore size is a measure of the specific surface quality: the smaller the pores, the higher the binding capacity. A disadvantage of nitrocellulose is its limited binding capacity and poor mechanical stability. Nitrocellulose membranes are often used in combination with isoelectric focusing in agarose gel]s.

Other types of membranes include:

• Teflon-based polyvinylidene difluoride (PVDF) membranes (with high binding capacity and high mechanical stability).

• Diazobenzyloxymethyl paper.

• Nylon membranes.

• Ion exchange membranes.

• Activated glass fiber membranes.

Since there are no universal membranes suitable for all purposes, in each case the membrane should be chosen to fit the molecules being studied (Westermeier, 1990).

Transfer methods. There are various ways of transferring gel to the blotting membrane: diffusion blotting, capillary blotting, vacuum blotting, and electrophoretic blotting. The simplest is diffusion blotting, which is predominantly used for immunoblots of proteins. Capillary blotting is primarily used for DNA separation (Southern blot: Southern, 1975) and RNA transfer (Northern blot: Alwine et al., 1977).

The Western blot (Towbin et al., 1979) is frequently used in protein research, mainly in the combination of SDS gel electrophoresis (see above) with electrotransfer onto a nitrocellulose membrane and using an antibody–enzyme conjugate for the detection reaction. This method allows apparent molecular weight and post-translational modifications to be determined.

Detection on the blotting membrane. In the two examples of isoelectric focusing and immunoblot (Fig. 4.13Fig. 4.14), the blotting membrane was incubated with peroxidaseconjugated anti-human IgG antibodies for detecting oligoclonal bands. In the example of a Western blot for detecting Borrelia-specific antibodies (Fig. 4.12; Schulte-Spechtel et al., 2003), the blotting membranes obtained after SDS gel electrophoresis were incubated first with the patient's samples (for binding of specific antibodies) and then with enzyme-conjugated anti-human IgG antibodies (for detection of the resulting immune complexes).

Detection of Oligoclonal IgG Bands

After isoelectric focusing (see above) of CSF and serum, the most common detection methods for oligoclonal IgG are immunofixation, immunoblot (Figs. 4.13 and 4.14), and silver staining (Fig. 4.15) (see above, “Electrophoretic Methods with Immunodetection”).

Direct protein detection with silver stain is generally more sensitive than immunodetection, but also more demanding in terms of handling. The bands are sharper with this method (no additional diffusion, unlike with blotting). On the other hand, immunoblotting requires less material because of the better enhancement provided by the enzymatic reaction of the conjugate. In veterinary medicine, silver staining offers the advantage that it does not require species-specific antibodies.

The sensitivity of a test method can be demonstrated by serial dilution of a CSF sample, which contains a large intrathecal fraction of oligoclonal IgG (e. g., IgGIF = 80%, as in some MS or neurosyphilis samples). The limit of detectability is reached when the bands are no longer visible. If the distribution of the bands is highly homogenous, the detection sensitivity for intrathecal IgG is lower than if there are some bands with a higher intensity.

Procedure for isoelectric focusing in agarose gel followed by immunoblot:

• Pour agarose gels (0.5 mm) onto GelBond film between silanized vertical glass plates.

• After boiling the mixture of agarose, glycerol, and sorbit (stir slowly, degas) and cooling to 60°C, add the ampholyte mixture (Servalyte, pH 3.0–10, and Pharmalyte, pH 3.0–10).

• Electrode buffers: 1 N NaOH, 0.05 N sulfuric acid.

• Isoelectric focusing: 1.75 h at 10°C and 1500 V, 150 mA, 10 W.

• Blot onto nitrocellulose membrane.

• Block.

• Incubate with HRP-conjugated anti-IgG.

• Perform detection reaction with 3-amino-9-ethylcarbazol in acetate buffer and H2O2.

Antigen-driven immunoblot. Antigen-driven immunoblotting is a particular application of specific antibody detection. After isoelectric focusing and blotting, the specificity of oligoclonal bands in the CSF is demonstrated by incubation with a single specific antigen (e. g., measles virus) (Sindic et al., 1994).


As an immunochemical detection method for specific molecules, immunohistochemistry has an important analytical function in pathology. The test is carried out directly on tissue sections and thus also provides information about the localization of the molecule within the tissue. Its use for the detection of paraneoplastic antibodies, an important example of tumor diagnosis in neurology, is shown in Chap. 7, “Antineural Antibodies.”


Alwine JC, Kemp DJ, Stark GR, et al. Method for detection of specific RNAs in agarose gels by transfer to diazobenzyloxymethyl-paper and hybridization with DNA probes. Proc Natl Acad Sci USA 1977; 74:5350–5354

Andersson M, Alvarez-Cermeno J, Bernardi G, et al. Cerebrospinal fluid in the diagnosis of MS: a consensus report. J Neurol Neurosurg Psychiatry 1994;57:897–902

Conrad AJ, Chiang EY, Andeen LE, et al. Quantitation of intrathecal measles virus IgG antibody synthesis rate: subacute sclerosing panencephalitis and multiple sclerosis. J Neuroimmunol 1994;54:99–108

Crowther JR. ELISA—theory and practice. In: Methods in molecular biology. Vol 42. Totowa, NJ: Humana Press; 1995

Debye P. Molecular-weight determination by light scattering. J Phys Chem 1947;51:18–32

Delmotte P, Gonsette R. Biochemical findings in multiple sclerosis: isoelectric focusing of the gamma globulins in multiple sclerosis and other neurological diseases. J Neurol 1977;215:27–37

Gharavi AE, Reiber H. Affinity and avidity of autoantibodies. In: Peter JB, Schoenfeld Y, eds. Autoantibodies. New York: Elsevier Science; 1996: 13–23

Heidelberger M, Kendall FE. A quantitative theory of the precipitation reaction. J Exp Med 1935;62:697–720

Laurell CB. Quantitative estimation of proteins in agarose gel containing antibodies. Anal Biochem 1966;15:45–62

Mancini GA, Carbonara A, Heremans J. Immunochemical quantitation of antigens by single radial immunodiffusion. Immunochemistry 1965;2:235–254

Metzmann E. Plasmaproteinbestimmungen. In: Fresenius W, Günzler H, Huber W, et al. (eds). Analytiker-Taschenbuch. Berlin, Heidelberg, New York: Springer; 1988:p.190

Quentin CD, Reiber H. Fuchs heterochromic cyclitis: rubella virus antibodies and genome in aqueous humor. Am J Ophthalmol 2004;138:46–54.

Rayleigh, Lord (JW Strutt). On the light from the sky, its polarization and colour. Philos Mag 1871;41:107–120, 274–279, 447–454

Raymond S, Weintraub L. Acrylamide gel as a supporting medium for zone electrophoresis. Science 1959;130:711

Reiber H. Kinetics of protein agglomeration. A nephelometric method for the determination of total protein in biological samples. J Biochem Biophys Meth 1983;7:153–160

Reiber H, Lange P. Quantification of virus-specific antibodies in cerebrospinal fluid and serum: sensitive and specific detection of antibody synthesis in brain. Clin Chem 1991;37:1153–1160

Schulte-Spechtel U, Lehner G, Liegl G, et al. Significant improvement of the recombinant Borrelia-specific immunoglobulin G immunoblot test by addition of VIsE and a DbpA homologue derived from Borrelia garinii for diagnosis of early neuroborreliosis. J Clin Microbiol 2003; 41:1299–1303

Serwer P. Electrophoresis of duplex deoxyribonucleic acid in multipleconcentration agarose gels: fractionation of molecules with molecular weights between 2 × 10(6) and 110 × 10(6). Biochemistry 1980;19: 3001–3004

Sindic CJM, Monteyne P, Laterre E. The intrathecal synthesis of virusspecific oligoclonal IgG in MS. J Neuroimmunol 1994;54:75–80

Southern EM. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J Mol Biol 1975;98:503–517

Sternberg JC. A rate nephelometer for measuring specific proteins by immunoprecipitation reactions. Clin Chem 1977;23:1456–1464

Svensson H. Isoelectric fractionation analyses and characterization of ampholytes in neutral pH gradients. Acta Chem Scand 1961;15:325–341

Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Biotechnology 1979;24:145–149

Vesterberg O. Synthesis and isoelectric fractionation of carrier ampholytes. Acta Chem Scand 1969;23:2653–2666

Westermeier R. Elektrophoresepraktikum. Weinheim: VCH; 1990

Wilske B. Diagnosis of Lyme borreliosis in Europe. Vector Borne Zoonotic Dis 2004;3:215–227

Wurster U. Elektrophoreseverfahren: Nachweis und Bedeutung von oligoklonalen Banden. In: Zettl UK, Lehmitz R, Mix E, eds. Klinische Liquordiagnostik. 2nd edn. Berlin: Walter de Gruyter; 2005:208–238

Zane HD, ed. Immunology. Theoretical and practical concepts in laboratory medicine. Philadelphia: Saunders; 2001