By: Claire Moore

Recombinant monoclonal antibodies are identical to traditional monoclonal antibodies in their basic functionality. However, since they are manufactured using fully in vitro processes, they offer greater flexibility during their production and more opportunities for optimization after their creation than typical monoclonal antibodies. Even though recombinant monoclonal antibodies are a newer technology, they have an established track record in the development of human therapeutics and in various research settings, with more than 11,000 such antibodies produced to date and a success rate of greater than 95%.

 

The use of recombinant monoclonal antibodies has recently begun to expand into IVD assay development as more clearly defined replacements for polyclonal antibodies and as calibrators and controls in diagnostic testing. This article describes the production of recombinant monoclonal antibodies using phage display technology, along with examples of screening flexibility, antibody optimization, and applications of the antibodies in IVD assay development.

Traditional monoclonal antibodies are produced by injecting mice, rats, or rabbits with a target immunogen during a period of several weeks. Once the animal displays a sufficient immune response to the antigen, the spleen is harvested, and the antibody producing cells are fused with an immortalized cell line. Following cloning and further testing, individual hybridomas are generated which can provide a long-term supply of individual monoclonal antibodies. After this stage, the monoclonal antibodies created using traditional hybridoma technology are screened for specificity. The entire process of immunization, hybridoma production, and initial screening takes 2-6 months. Later, the individual monoclonal antibodies are tested for their suitability in particular immunoassays.

 

Significant expertise, time, and specialized cell culture and animal facilities are all needed for the successful creation of hybridomas. Moreover, the process can be unpredictable, and will not yield results if the target molecules are either toxic or poorly immunogenic in the mammals. Once an antibody is produced using traditional hybridoma technology, it usually remains fixed in its specificity and isotype. While traditional monoclonal antibodies can be sequenced and manipulated, it is a complex process. In addition, hybridoma output can vary over time, and a loss of antibody-producing cell lines can occur during long-term storage.

 

Recombinant human monoclonal antibodies have been developed to address such issues. Already, well-established, recombinant monoclonal antibodies have been successfully used in various therapeutic, research, and diagnostic applications. The technology for developing such highly specific, high-affinity monoclonal antibodies is a fully in vitro method. This method relies on a complex gene library that represents the human repertoire of potential antibody specificities in a tube. The antibody binding sequences within the library are presented to the target immunogen using phage display, so that the typical antibody binding site is available to the antigen as a part of the phage coat rather than at the ends of the antibody molecule. The binding specificity of the antibody is also tied directly to its genetic sequence, which can be captured along with the phage (see Figure 1).

 

This advanced phage display technology offers the following benefits compared with conventional monoclonal antibody production methods (see Table I):

 

•         The genetically engineered system used to make recombinant monoclonal antibodies is modular in its design with ready access to fully characterized DNA sequences for facilitating antibody manipulation. For example, specific tags for purification, immobilization, and detection can be rapidly added to the antibody sequence, and particular isotypes or antibody fragments can be created by replacing modular sections. Creating genetic fusions by directly joining the antibody to alkaline phosphatase or other reporter proteins is also possible.

•         Production security is improved since the recombinant monoclonal antibody binding information is retained as purified DNA and as an electronic backup of the sequenced plasmid DNA. DNA is extremely stable, and if necessary, the electronic sequence can quickly recreate the antibody binding regions by gene synthesis. Therefore, recombinant antibodies can be rapidly replaced in case of loss, providing an unlimited, consistent, and fully defined reagent supply.

•         Recombinant monoclonal antibodies are produced in bacteria using automated, high throughput systems. This process is easier, faster, and less labor intensive than using animals and tissue culture reagents.

•         Recombinant monoclonal antibody production takes place in vitro, allowing for greater latitude in selection and screening conditions. For example, the antibodies can be identified in the presence of reagents that will eventually be used for the final diagnostic assay. Furthermore, the selection process can be directed to fine tune the specificities achieved toward less similar sequences or modified formats, such as non-phosphorylated versions of a protein. While traditional monoclonal antibodies can also be produced against phosphorylated antigens, they are identified later in the process, and only if they occur among the candidates produced.

•         Recombinant monoclonal antibodies can be further optimized by stringent selection or additional rounds of mutagenesis, thereby increasing sensitivity in a targeted manner that can be tailored to an assay’s requirements.

•         Much smaller quantities of antigen are required, typically 0.5 mg or less.

 

Recombinant monoclonal antibodies were originally manufactured by isolating the relevant genetic material from a hybridoma cell line, or by amplifying via polymerase chain reaction (PCR) a pool of antibody-producing cells obtained from an immunized animal. Once the genetic material was isolated, it was inserted into a vector and expressed in E. coli. Today, recombinant monoclonal antibodies are produced from highly sophisticated libraries of antibody genes that have been specifically created by de novo synthesis to represent the structural diversity of the human antibody repertoire, and have been optimized for expression in E. coli.

 

Two such libraries were developed by MorphoSys AG (Munich, Germany) for producing therapeutic antibodies: HuCAL Gold, with more than 15 billion specificities, and HuCAL Platinum, with more than 45 billion specificities. These libraries were constructed based on sequence information obtained by bioinformatics analysis of the human immune system. They are composed of a modular system of 42 framework genes representing the variations in immunoglobulin heavy and light chains, combined with six complementarity determining regions (CDR) cassettes that represent the hypervariable regions of the antibody (see Figure 2).¹

 

The CDR cassettes in the library have been created to mimic the natural diversity in the antigen binding sites using an optimized trinucleotide strategy to maximize potential diversity.² Transforming with vectors containing the library’s multibillion specificities produces a starting point for isolating antibodies of any given specificity. For these libraries, the antibodies have been created in a monovalent Fab format, since the Fc region is not necessary for antigen binding. The modular construction of the recombinant monoclonal antibody fragments means that they can later be easily adapted to divalent or full-length immunoglobulin molecules.

 

The desired antibodies that bind with certain antigens are identified from such large gene libraries by selection rather than screening. All selection methods to date physically link the binding properties of every antibody in the library (the phenotype) with the genetic information that encodes the given antibody (the genotype). Among the numerous selection techniques, the most commonly used are phage display, ribosome display, and yeast display. Phage display is the most popular and best established selection method.

 

The phage display process begins by infecting antibody gene-containing E. coli host cells with modified phage to create daughter phage, each of which contains a vector encoding one of the antibody library members. Each daughter phage also displays the corresponding antibody molecule on its surface, which is achieved by genetically fusing the antibody gene with one of the phage coat proteins. The phage biology links the antibody binding properties to the corresponding genetic information, which can later generate a monoclonal cell line. The daughter phage are harvested from the supernatant, precipitated, dissolved in small volumes that contain about 10¹³ phage particles per ml, and stored frozen for later use.

 

Desired antibodies are then selected by phage panning, which is similar to solid-phase immunoassays.³ In this process, the antigen of interest is first immobilized, usually on microplate wells, magnetic beads, or a column (see Figure 3). The phage are then incubated with the antigen to allow capture of the few phage that display an antibody with binding affinity to the target molecule.

 

At this step, other substances such as molecules closely related to the antigen can be added to the solution to effectively block undesired binding sites and deplete the library of unwanted specificities. The binding buffers can also be manipulated to mimic more closely the final assay conditions, thereby improving chances that the candidate antibodies will work in the planned immunoassay. After extensive washing to remove all non-specific material, the bound phage are eluted and amplified by replication in new host cells.

 

This selection procedure is repeated several times, resulting in a phage population that is highly enriched with members that express the desired antibodies (i.e., those that specifically bind the antigen of interest). After selection, the desired antibody genes are isolated as a gene pool and inserted into an antibody expression vector. The whole process is accomplished in a few days and is amenable to automation.

 

To produce soluble antibody fragments, the antibody genes cloned into the phagemid must be expressed without the phage coat protein. Cloned individual antibodies are obtained by isolating the genes for light and heavy chain with restriction enzymes or PCR amplification and by religating them into an expression vector that does not contain the phage protein gene. Following introduction into new host cells, the transformed cells are isolated as single colonies, each producing a uniquely defined monoclonal antibody. When automated systems are used, 384 colonies are picked and grown, and antibody expression is induced. Cells are subsequently lysed, and the antibody-containing lysate is tested by an enzyme-linked immunosorbent assay (ELISA) for the presence of antigen-specific antibody material.

 

This selection, expression, and screening process is completed in 4-6 weeks, and results in the identification of multiple unique antigen-specific monoclonal antibody fragments. These antibodies are then grown in larger scale for purification. The entire process from inception to delivery takes approximately eight weeks and represents a substantial time savings compared with traditional antibody production methods.

 

The monoclonal antibody fragments manufactured by this process are fully functional and include the complete antigen binding site with the same intrinsic antigen-binding affinity as their full-length antibody counterparts. They are available in a variety of antibody formats (e.g., monovalent Fab fragments, bivalent Fab fragments) with a wide selection of protein tags, such as 6-His, to simplify purification and detection. Due to the modularity of the system, the monoclonal antibody fragments can also be rapidly converted into full-length immunoglobulin molecules in isotypes IgG1, IgG2, IgG4, IgA, and IgM.

 

For most applications, the bivalent Fab fragments are preferred because they offer increased avidity due to the presence of two antigen binding sites. However, they are considerably smaller than an immunoglobulin molecule and diffuse more readily. Moreover, since they lack the Fc region, there is less chance of non-specific binding to Fc receptors.

 

A fully in vitro process for developing recombinant monoclonal antibodies allows for guided screening protocols that can discriminate between closely related antigens and improve the binding characteristics of candidate antibodies. Examples of several different targeted screening protocols and an example of an optimized binding activity are described below. While carrying out some specific antigen targeting with classical hybridomas is possible, the process is more haphazard since it occurs inside an animal in which conditions cannot be readily manipulated.

 

Modified Amino Acids. To identify antigens against modified amino acids, the library is first incubated with the unmodified form of the peptide in order to remove the antibodies that recognize it. This process is referred to as depleting the library. At this time, other related peptides are often added as well since doing so will improve the final specificity of the antibodies that is achieved (see Figure 4). The depleted library is screened with a peptide displaying the desired modification (e.g., phosphorylated or oxidized amino acids) to generate antibodies that will have a better chance of demonstrating a unique specificity to the desired form of the target antigen rather than closely related molecules.⁴ To confirm success, the antibodies are tested for specificity by a quality control (QC) ELISA against the target and both related and unrelated peptides before purification and scale up.

 

Highly Conserved Antigens. The blocking/subtraction strategy described above is also ideal for selecting specific antibodies when the antigen is highly conserved (e.g., a protein that varies slightly between species or has a single amino acid mutation). Figure 5 compares the ability of three antibodies to detect a wild type versus a single amino acid mutation in the target antigen. In addition, the candidates are screened against different carrier molecules, BSA and transferrin. If both the target and the related protein are available, the library can be depleted of unwanted specificities prior to screening with the desired target antigen.

 

Distinguishing the Parts from the Whole. Sometimes it is desirable to be able to detect only a portion of a target antigen and not the full-length molecule—in the case of a cleavage product, for example. As depicted in Figure 6, the HuCAL library was used to generate antibodies to two cleavage peptides of 10 amino acids each, while not recognizing the 20-amino-acid full-length peptide. As in the examples discussed above, the undesired specificity against the full-length peptide was used to deplete the library before screening with each of the target peptides. This procedure is ideal for producing anti-idiotypic antibodies for use in clinical monitoring.

 

Improving the Binding Affinity of Recombinant Monoclonal Antibodies. One key advantage to producing antibodies in vitro is that once they are created, they can be further manipulated and matured. In the simplest case, such manipulation entails changing a recombinant monoclonal antibody’s structural backbone or valency without modifying the specificity. Since the HuCAL library is modular, optimizing further the binding affinity of candidate antibodies is also possible by doing additional rounds of screening with a mix and match approach to insert new CDR sequences at unique flanking restriction sites in the CDR cassettes (see Figure 7). This strategy improved a blocking antibody’s binding affinity against a rat protease inhibitor by more than 300 fold (see Figure 8).

 

Positive Controls and Calibrators. Detecting human antibodies in patient sera in various formats or isotypes lies at the heart of a broad range of immunodiagnostic assays. For example, IgA and IgG are key indicators for infectious diseases and autoimmunity, IgE is important in tests for allergic reactions, and IgM is used for diagnosing early immune response. The majority of such tests require positive controls and calibrators. Today, while patient sera are currently considered the gold standard, these polyclonal isotope controls need to be repeatedly calibrated due to variations in product quality, limited quantities, and the risk of viral contamination. The HuCAL technology has created recombinant monoclonal antibodies in an IgG, IgA, and IgM format for a major IVD company in Germany for use in autoimmune tests.

 

Matched Pair Development. The in vitro screening process provides an efficient route for identifying matched pairs for ELISA-type immunodiagnostic assays. The process can identify a detection antibody to match an existing capture antibody or vice versa. It can also select a new matched pair against a novel biomarker. If a second binding antibody is required, the library can be screened using the first antibody to present the antigen for screening, and by including the unbound first antibody, the selection is targeted toward the antigen rather than the first antibody.

 

When a new matched pair is desired, there are two strategies. The first strategy is standard panning against the antigen, followed by testing of all possible combinations of the selected antibodies as capture and detection antibodies in a sandwich ELISA. If this strategy is not successful, the second strategy is standard panning followed by screening using a labeled (e.g., biotinylated) antigen in which the antibodies are tested for their suitability as capture molecules in ELISA. The best capture antibody is then used in a new panning to detect those antibodies that bind the antigen when presented by the capture antibody.

 

The technology for producing recombinant monoclonal antibodies offers numerous advantages compared with traditional hybridoma methods for developing monoclonal antibodies for immunodiagnostic assays.⁵ One of the primary advantages is the ease of genetic manipulation that results from a fully defined modular system, which allows for simplified downstream alterations and further optimization to match individual assay requirements. Furthermore, since the process occurs without the use of animals, there is greater flexibility in choosing the conditions used for the initial generation of monoclonal antibodies, including the use of toxins and directed screening strategies. Finally, all the recombinant monoclonal antibodies manufactured are fully defined and secured for a safe, long-term supply since the plasmid DNA sequences are easily stored, rapidly recreated, and fully characterized. As a result, the use of recombinant DNA technology increases the efficiency and flexibility in identifying monoclonal antibodies for diagnostic assay development.

 

1. A Knappik et al., “Fully Synthetic Human Combinatorial Antibody Libraries (HuCAL) Based on Modular Consensus Frameworks and CDRs Randomized with Trinucleotides,” Journal of Molecular Biology 296 (2000): 57-86.

 

2. B Virnekäs et al., “Trinucleotide Phosphoramidites: Ideal Reagents for the Synthesis of Mixed Oligonucleotides for Random Mutagenesis,” Nucleic Acids Research 22 (1994): 5600-5607.

 

3. CF Barbas and RA Lerner, “Combinatorial Immunoglobulin Libraries on the Surface of Phage (Phabs): Rapid Selection of Antigen-Specific Fab,” Methods: A Companion to Methods in Enzymology 2 (1991): 119-124.

 

4. H Ooe et al., “Establishment of Specific Antibodies that Recognize C106-Oxidized DJ-1,” Neuroscience Letters 404, no. 1-2 (2006): 166-169.

 

5. R Ohara et al., “Antibodies for Proteomic Research: Comparison of Traditional Immunization with Recombinant Antibody Technology,” Proteomics 6 (2006): 2638-2646.

 

 Claire Moore is a technical writer at AbD Serotec (Kidlington, Oxford, UK). She can be  reached at claire.moore@abdsserotec.com.