Medical Device Link.

Originally Published IVD Technology March 2002

Assay Development

Immunogold conjugation for IVD applications

To create viable immunogold complexes for use in diagnostic applications, manufacturers must be capable of overcoming a variety of technical challenges.

Nikki Robinson

Immunogold conjugates created by linking proteins to gold colloid (above) are used as detector reagents in a wide array of IVD test systems.

Proteins for Conjugation

For IVD applications, the types of proteins that are most commonly used to create immunogold conjugates are antibodies and antigens. One factor that can affect antibody conjugations is whether monoclonal or polyclonal antibodies are used.

Although most antibody conjugates for diagnostics are created using immunoglobulin G (IgG) antibodies, IgM, IgE, and IgA can be conjugated successfully to colloidal gold. Further consideration should also be given to the subclass of antibody that is used for conjugation. Among the IgG subclasses of mouse antibodies, for example, IgG1 has the best success rate, while IgG3 has proven to be technically challenging.

Preparing conjugates using antigens can also be technically challenging. Conjugators typically encounter no difficulties in creating bonds to link the antigen to the colloid. But because of the ways that antigens bind to gold colloid, the success rate for creating a conjugate in which the gold label does not mask the working reactive epitopes of the protein can be as little as 50%. For this reason, it is important for conjugate developers to give consideration to the ways in which proteins bind to gold.³

Binding Mechanisms

Researchers generally accept the theory that mechanisms related to the residues of three particular amino acids play an important role in binding proteins to gold particles. Each of these amino acids—lysine, tryptophan, and cysteine—operates by a different mechanism to bring about conjugation.⁴

• Lysine is highly positively charged and therefore naturally attracted to a negatively charged gold particle.
  
• Tryptophan works through hydrophobic interactions.
  
• Cysteine creates dative bonds through the formation of sulphur bridges with the gold surface, such that the antibody and gold particles share electrons. Of all the forces controlling the attachment of antibodies to gold particles, this is the strongest, most permanent, and most difficult to break.

The success of the conjugation process depends to a large extent upon the location of these amino acid residues in the protein that is being conjugated. In some cases, poorly located amino acids can bind to gold molecules in such a way that the gold actually interferes with the binding capacity of the protein. Termed steric hindrance, this type of interference can occur when the amino acids are located in the antigen-binding (Fab) region of antibodies, or among the working reactive epitopes of antigens (that is, epitopes recognized by specific antibodies). Such interference is almost impossible to overcome without compromising the integrity of the protein molecule being conjugated.

In order for an immunoassay to achieve optimal sensitivity, it is therefore important that the three amino acids involved in conjugation be located appropriately. For antibodies, they should be located in the Fc region. For antigens, they should be physically isolated from the working reactive epitopes.

Optimizing Conjugation

In order to develop and optimize a single gold conjugate, it is not unusual for researchers to produce small-scale batches of as many as 60 different conjugates. Each of these conjugates differs from the others in one key way, so that each can be tested to determine which conjugate is likely to give the best response in the final assay for which it is intended.

Such testing should encompass all the parameters that moderate the sensitivity, cross-reactivity, and potential stability of the conjugate. Characteristics to be tested include, but are not limited to, antibody vehicle buffer, preservative type, salt ratio, surfactant content, gold colloid size, blocking agent type, total protein concentration, final conjugate buffer, and conjugate concentration. To determine the effects of different blocking agents, for instance, tests for the following should also be considered.

• Type of blocking agent (e.g., casein, bovine serum albumin, ovalbumin, human albumin, PEG, nonspecific IgG).
• Purification grade of the blocking agent.
• Concentration of the blocking agent.
• Compatibility of the blocking agent with the completed assay.
  

Figure 1. Diagrammatic representation of a half dipstick used for screening conjugates.

Developers should conduct experiments to determine the optimal values for all such factors prior to scaling up for prototype production. Ultimately, however, the only way to determine which conjugate should be scaled up for pilot production is to develop a mini-assay that mimics the final test conditions as closely as possible. For membrane-based assays, the easiest way to do this is to use a half-dipstick format that requires only a capture antibody and a purified antigen (see Figure 1). By using such a simplified test format, many small-scale conjugates can be assessed to determine which is most suitable without encountering common problems such as those related to drying techniques or sample buffer preparation.

In order to carry out all of this testing, conjugators usually require 3 mg of antibody or 5 mg of antigen. Using these small samples, the conjugator can then optimize the manufacturing parameters for the conjugate and provide a preliminary sample for assessment. Once a quality antibody or antigen has been conjugated, and its manufacturing parameters have been set, it is usually straightforward for the manufacturer to faithfully reproduce and scale up the chosen conjugate.

Quality of Materials for Conjugation

The suitability of a particular antibody for use in creating an immunogold conjugate depends largely on the technical specifications of the assay in question.

Figure 2. The important aspects of antibody pairs include steric separation of epitopes, adequate titer of stocks, high affinity, high specificity, high avidity, and purity.

For lateral-flow assays, the best types of antibodies to use are generally monoclonal pairs. In this case, one monoclonal antibody is labeled with the gold colloid as the detector reagent, and the other is immobilized on the nitrocellulose membrane as the capture reagent. Each antibody should be specific to a different reactive epitope on the antigen surface, and these epitopes in turn should be physically isolated from one another (see Figure 2).

Polyclonal antibodies can also be used, but they should be at least protein-A purified. In most cases, affinity purification provides the most sensitive and specific conjugate. This depends of course on the level of acceptable cross-reactivity in the assay specifications.

Affinity Purification

Ammonium sulphate precipitation, or DEAE purified serum, contains a mass of protein that will compete for binding sites on the surface of the gold colloid. Those proteins with the highest levels of the three amino acid residues that control protein binding to the colloid will conjugate most readily to the naked, negatively charged colloid. However, these may not be the proteins required to drive the assay. When this occurs, the detector reagent of the assay can have a decreased degree of sensitivity.

Similarly, if the antibody preparation is protein-A or protein-G purified, then the IgG fraction may contain high levels of interfering IgG, and not just the specific IgG required. All of the IgG will attempt to bind to the gold colloid, and a low level of sensitivity may result.

Affinity purification—with careful elution of the highest-affinity antibodies from the column—should yield the most-specific antibodies for antibody preparation and therefore produce the most-specific gold conjugate. These antibodies may require further absorption in order to modify cross-reactivity characteristics. Although affinity purification can be costly in terms of time, money, and serum, it produces a quality conjugate which should be considered a key raw material for any assay.

Antigen Conjugation

Successful creation of antigen conjugates depends on two factors: size and the situation of the three amino acid residues that control the binding of the antigen to the colloid.

Most antigen conjugates are requested for use in rapid tests, such as serological sandwich or competition assays. For such assays, a 40-nm colloid is often chosen. Until recently, it was considered that the smallest particle that could be conjugated to a 40-nm colloid had a molecular cutoff weight of 30 kDa. Under some conditions, however, advances in conjugation techniques can cut this limit in half to 15 kDa. The primary restriction for conjugates using such small particles is that the protein to be conjugated must contain sufficient amounts of lysine, tryptophan, and cysteine residues. Antigens frequently do not contain enough cysteine, so the strength of the bond between the antigen and the colloid is relatively weak and easy to break, particularly in the presence of high-affinity antibodies in the sample.

For antigens with a molecular weight of less than 30 kDa, other techniques may be applied, such as conjugation to smaller gold particles. This too may result in a loss of sensitivity due to the reduced visibility of smaller gold colloids at the capture site.

For antigens containing little or none of the three binding residues, an efficient solution is to preconjugate the antigen to a carrier molecule, such as bovine serum albumin (BSA) or keyhole limpet hemocyanin (KLH). However, this technique is not as simple as it sounds and is not something that the amateur protein chemist should attempt. The type of linker used, the length of the linker used, the molar ratio of the hapten to the BSA, and the type of carrier molecule used are only some of the factors that must be carefully considered in order to carry out a reproducible conjugation that exposes the working reactive epitopes of the antigen—and therefore maximizes the potential sensitivity of the protein-carrier-gold conjugate in the assay.

Size Matters

Selection of the size of the colloid depends upon the application for which the conjugate is intended. The following sections describe relevant issues for two of the most common applications for gold conjugates: microscopy and rapid-test devices.

Microscopy. Any gold conjugate that is 1 to 40 nm may be used for electron microscopy and light microscopy. However, it may not be possible to view 1- or 2-nm colloids even under a transmission electron microscope that provides 250,000´ magnification. To make it possible for end-users to visualize the gold-labeled antigenic sites on such small particles, the gold colloid can be enhanced with silver.

Silver enhancement involves the precipitation of silver salts on the gold surface, which allows the gold particle to grow in size until it can be seen under the microscope. This process is most effective and easily controlled when all of the gold particles are the same size and shape, with a low coefficient of variation. Using small particles, such as 1- and 2-nm gold particles, facilitates access to reactive epitopes that larger gold particles would not be able to label due to steric hindrance. Therefore, the gold conjugate will bind easily and is grown by silver enhancement once the
antigen has been located.

For electron microscopy without silver enhancement, several different reactive epitopes may be labeled on the same section of the conjugate. In this case, colloids with distinct size separations are used to label antibodies specific to each of the different reactive epitopes.

Rapid tests (lateral-flow and flow-through devices). The most common size of gold colloid used for this application is 40 nm.⁴

A 40-nm colloid offers maximum visibility with the least steric hindrance in the case of IgG conjugations (see Figure 3). An IgG antibody with a molecular weight of 160 kDa is approximately 8 nm long.

Figure 3. The relationship between particle size, visibility, and steric hindrance. Note that as the particle size increases, signal visibility increases, but steric hindrance causes an overall decrease in visual sensitivity.

Gold particles between 40 and 100 nm can also be conjugated successfully to IgG antibodies. Such larger particles are certainly more visible than the 40 nm particles, but there are fewer particles in a 1-ml solution at an optical density of 520 nm, so fewer of them can be packed onto the capture line. Overall, this usually expresses itself as a loss of signal at lower levels of analyte (between 1 and 10 ng/ml), when compared with the same antibody conjugated to a 40- or 60-nm colloid.

A 60-nm colloid is not routinely employed but can be useful for rapid-test assays. Its color is slightly different compared with a 40-nm colloid. Good quality 40-nm colloid should be cherry red, while 60-nm colloid is deep pink.4 For some applications in which the sample type may cause background staining of the membrane, this subtle color difference may make a signal easier to read. For example, samples containing bilirubin will leave a brown background stain, and it is easier to visualize pink on brown than red on brown.

If the molecule to be conjugated has a molecular weight less than 160 kDa, then a 20-nm colloid may be more appropriate. Colloids of 20 nm are commonly used for streptavidin, protein A, and antigen conjugations in which the protein that is being presented for conjugation has a molecular weight of less than 60 kDa.

Sensitivity

Advances in rapid-test technologies may increase the sensitivity limits of gold conjugates. At present, the sensitivity limit in most assays is 1 ng/ml, with high-quality antibodies reaching as low as 10 pg/ml. However, the development of silver enhancement techniques may increase sensitivity still further, perhaps from 10 to 100 times.⁴

Conclusion

Conjugation of proteins to colloidal gold is a process that should not be undertaken lightly. While it is relatively straightforward to make an immunogold conjugate, it is difficult to make a quality gold conjugate, and many inexperienced manufacturers encounter problems in batch reproducibility and scale-up.

References

1. S Brunelle, "Electroimmunoassay Technology for Food-Borne Pathogen Detection," IVD Technology 7, no. 5 (2001): 55–66.

2. S Wallace, "Bioterrorism Tests in Demand," IVD Technology 7, no. 6 (2001): 14.

3. J Chandler, N Robinson, and K Whiting, "Handling False Signals in Gold-Based Rapid Tests," IVD Technology 7, no. 2 (2001): 34–45.

4. J Chandler, T Gurmin, and N Robinson, "“The Place of Gold in Rapid Tests," IVD Technology 6, no. 2 (2000): 37–49.

Nikki Robinson, BSc, is custom conjugation manager at British BioCell International (Cardiff, Wales, UK). She can be reached via nikkirobinson@bbigold.com.

Copyright ©2002 IVD Technology