November 1, 2006 (Vol. 26, No. 19)
Addressing a Host of Immune Response Challenges
Antibody engineering is moving into its fourth decade, a stage of maturity that emphasizes both the success and the challenges of the discipline. One problem that antibody-engineering companies have long struggled with is the host immune response to biologic therapeutics.
Researchers will meet at IBC’s Antibody Engineering Conference in San Diego next month to talk about potential strategies to address this issue.
Typically, low molecular weight drugs are not recognized by the immune system, whereas protein molecules of much higher molecular weight invariably are. Even molecules not usually recognized as foreign (such as humanized antibodies), when pumped into the circulation in quantities large enough to perform their therapeutic function, can activate an autoimmune response.
Antitope (www.antitope.co.uk), which is scheduled to present at the IBC meeting, is one of a number of companies working on this frustrating problem. It has been shown that within the variable regions of even fully human antibodies, somatic mutations and other alterations in sequence information can stimulate a powerful reaction on the part of the host.
According to Matthew Baker, Ph.D., CEO and CSO at Antitope, therapeutic proteins, in addition to antibodies, can be extremely troublesome. The antihemophilic protein, Factor VIII, has been used to treat Hemophilia A patients, using either protein purified from pooled plasma or recombinant Factor VIII. But in at least 50% of cases, the patients produce neutralizing antibodies, rendering therapy with these proteins ineffective. This is due to the fact that certain amino acid sequences, so-called T-cell epitopes, when presented by antigen-presenting cells and recognized by helper T cells, will stimulate a powerful immune response.
It is known that all individuals possess classes of T cells that can recognize these forbidden epitopes. By comparing responses to different amino acid sequences, a T-cell epitope map can be built up which allows the scientist to predict what modification can be designed into the Factor VIII molecule to make it less immunogenic.
Dr. Baker and his colleagues were successful in synthesizing a peptide containing two mutations, which provoked a greatly reduced T-cell reaction in healthy individuals. When these mutations were incorporated into a cellular clone producing Factor VIII, the resultant protein still performed its role in the clotting cascade without raising an anti-Factor VIII immune response.
Such peptides are believed to be “promiscuous” T-cell epitopes that bind to the HLA supertypes on the surface of antigen-presenting cells that are broadly represented in the world’s population. T-cell and CD4 receptors on the surface of T cells bind to the HLA protein/T-cell epitope complex in order to generate an immune response. Their analysis is critical to determining the therapeutic proteins that will fail to induce immune responses while still performing their biological function.
Antitope offers several screening technologies that are carried out at their facility under GLP guidelines. These include EpiScreen™, a T-cell epitope mapping technology that screens peptides for their immunostimulatory properties, and EpiScreen DC:T cell assay, which screens biologics for their immunogenicity. These and other technologies are available from the company on a contractual basis.
“We do contractual screening for a number of big pharma companies,” says Dr. Baker. “With the volume of repeat business we’re seeing, we know the approach is paying dividends.” The ability of EpiScreen to detect possible autoimmune responses early on in testing is now well documented. In particular, by identifying problem antibodies, companies can now avoid moving inappropriate antibodies into the clinic—with the potential for large savings.
“Many antibodies, including humanized and fully human antibodies, did not present serious immune dysfunction issues until they had been approved and a large body of documentation concerning their performance had accumulated,” Dr. Baker continues. “We feel the ability of EpiScreen to recognize these problematic antibodies early in their development is an important tool for insuring highly successful biologics. If our technology had been available during the early developmental phases of Remicade and Humira, their immunogenicity could have been noted and abrogated.”
According to Dr. Baker, Antitope researchers are investigating the potential of their platforms to delineate the critical amino acid sequences in enzymes used as various therapeutic tools. These include asparaginase (used in the treatment of aspargine synthetase-deficient cancer cells), urate oxidase (for breaking down excess uric acid), and carboxypeptidase G2 (used to eliminate methotrexate following chemotherapy). All of these valuable treatment options are compromised by their immunogenicity. Antitope’s new strategy involves a redesigning of enzymes to active, but less immunogenic forms.
Today the immunogenicity of biological drugs is counteracted with immunosuppressive medication, a highly unsatisfactory option. The development of non-immunogenic therapeutic proteins may represent an appealing solution to this quandary.
Exploiting the in vivo genetic evolution of hybridomas is an alternative method for generating fully human antibodies of high affinity, as Luigi Grasso, Ph.D., senior vp for research and development at Morphotek (www.morphotek.com), explains. “We use two platforms, Morphodoma™ and Libradoma™, to generate fully human antibodies to target antigens,” says Dr. Grasso, another scheduled speaker for the IBC conference.
In the human Morphodoma technology, human B cells obtained from selected volunteers, proven to be free of infectious viruses, are immunized ex vivo to human antigens related to selected pathogens, and then immortalized by fusion to a myeloma cell line. In some cases, the B cells are obtained from patients who have been exposed to the target antigen, whereas in other instances, naive B cells are employed for de novo immunizations. To stimulate the immunization process, human CD4-positive T cells and antigen-presenting cells are added to the cultures along with an optimized cocktail of cytokines.
In addition to human IgG, some other antibodies produced are of the IgM isotype, which may be restricted in their usefulness. Dr. Grasso and his colleagues were able to switch from IgM to IgG by growing hybridomas under defined conditions and screening for subclones producing IgG. The screening steps were performed robotically so a large number of clones could be screened at each step.
In Morphotek’s Libradoma (libraries of hybridomas), many antibodies of unknown specificity are generated from B cells obtained from selected patients, followed by screening and improvement.
“A good candidate for this approach would be a patient that shows some initial spontaneous remission and then relapses,” says Dr. Grasso. “These would be individuals that we suspect of generating their own antitumor response, so they should be a rich source of B lymphocytes producing antitumor antibodies, including those against targets that have been overlooked in the past.”
A second means of generating Libradomas is to use B cells immunized ex vivo with a mixture of antigens, while a third strategy involves a form of vaccination, such as injection of tumor antigen-presenting cells that can elicit humoral responses. The Morphotek robotic capabilities can rapidly process libraries of greater than 10,000 independent hybridoma clones, which can be screened for production of desired antibodies, according to the company. The lead clones are then validated and optimized to meet manufacturing standards.
Dr. Grasso and his coworkers have combined their antibody-discovery technology with morphogenics, a whole-genome evolution platform, which increases the genetic diversity of host cell lines. By using a DNA mismatch repair chemical inhibitor, this process can be turned on and off so that the cell lines can be restabilized after the desired phenotype is obtained. Thus stable subclones that secrete antibody with enhanced properties, improved growth characteristics, and high-titer production can be identified using robotic high-throughput screenings.
These cell lines are not burdened with the aneuploidy and chromosomal instability that accompanies the use of chemical mutagens. Through these subtle changes in the host genome, Dr. Grasso and his colleagues were able to generate clones that produced antibodies with as much as a 34-fold improvement in affinity and enhanced antibody titers of up to 10-fold.
Ex vivo immunization protocols have traditionally been challenging and lacked repeatability, and for this reason the Morphotek team spent several years refining the technology, according to Dr. Grasso.
Automated Antibody Screening Tools
Applied Biosystems’ (www.appliedbiosystems.com) 8200 Cellular Detection System allows automated screening of binding activity more rapidly than previously possible, according to Carol Khodier, senior scientist. The instrument is a fluorescence event analyzer that detects antibody and ligand binding without multiple rinsing and incubation steps, as is the case with conventional ELISAs.
The instrument is an optical-detection platform enabling homogeneous assay formats and is capable of performing quantitative bead- and cell-based assays. It uses a red, 633-nm helium-neon laser and quantifies fluorescence in real time with proprietary algorithms through macroconfocal technology, collecting data from two different dye channels run concurrently.
The detection principle is based on the settling of fluor-antibody conjugates bound to cells or beads on the bottom of a microtiter plate well. The complexes are activated with a laser and scanned with approximately 100-µm depth of focus so that only fluorophores that are in the focal plane are detected. Free, unbound antibody will remain in solution, out of the plane of focus. Because of the use of red laser there is low auto-fluorescence from cells and almost no interference from compounds that autofluoresce in wavelengths close to the emission maxima of fluorescein.
In a conventional ELISA, antibodies are reacted with antigens bound to plates, and the unreacted material is washed away in a series of rinses and incubation steps, which may run from minutes to hours depending on the protocol. In contrast, the 8200 system offers a mix-and-read binding event analysis, since the unreacted fluors remain undetected.
The 8200 system is sensitive, detecting as few as 10,000 binding events, and is compatible with 96-, 384-, and 1,536-well microtiter plates, according to the company. Its versatility allows adaptation to multiple applications.
The system operates rapidly, screening as many as 30,000 wells per day. It can be used for chemokine assays, apoptosis, and cytotoxicity analysis. A variety of bead-based assays can be executed including hybridoma/Ab screening, kinase assessments, protein-protein interactions, receptor-ligand binding, and IgG quantitation.
Lately a number of newly developed solutions for label-free detection of protein-protein interactions have been introduced to the market place. The ForteBio (www.fortebio.com) system uses biolayer interferometry.
The Octet system uses biosensor probes through which visible light is passed. The tip of the biosensor is flat and possesses a coated surface that binds various proteins. When the biosensors carrying a bound antigen are inserted in a microplate well they can bind antibody molecules, changing the optical thickness at the tip of the biosensor. This action will cause a shift in the wavelength spectrum of light reflected back through the fiber optics of the apparatus. This change can be measured as a nm shift in real time with high sensitivity and reproducibility.
Because the Octet System only measures light reflected back from proteins bound to the biosensor tip, media containing crude supernatants can be used.
The system may lend itself to diagnostic tests in which serum and other crude samples are evaluated. Moreover, the measurements are non-destructive, and in the case of hybridoma supernatants, hybridoma cells from candidate wells can subsequently be withdrawn, expanded, and evaluated further.
“Currently, we offer biosensors with four surface chemistries—an amine-reactive surface, streptavidin, anti-human IgG, and Protein A. These provide a wide range of binding capabilities for the Octet System to measure kinetics and protein quantitation,” states Krista Witte, Ph.D., director of applications for ForteBio.
Crossing the Cell Membrane
MSM Protein Technologies (www.msmprotein.com) is attacking a major challenge for antibody development—proteins that reach across the cell membrane. These multispanners are integral membrane proteins, whose polypeptide chains weave back and forth through the lipid bilayer in an intricate series of twists and turns. Multispanners are the major class of cell surface proteins responsible for cellular homeostasis and intercellular communications.
“We have developed proprietary technologies for efficient discovery that significantly increase the likelihood of the generation of therapeutic antibodies,” states Tajib Mirzabekov, Ph.D., MSM president and CSO.
MSM Protein Technologies’ methodology consists of a suite of tools for handling purified multispanners in their native state. Here the extracellular portion is comprised of constrained, post-translationally modified small loops. These loops cannot be expressed and maintained separately as folded polypeptides. Since antibodies against peptides or unfolded proteins fail to recognize native antigen, the native state of the multispanner target is essential in order to generate useful antibodies.
The MSM technology employs proprietary vectors that carry a human multispanner gene and compatible cell lines designed for their efficient expression. In addition the immunization protocols employ technologies that allow the concentrated multispanning membrane proteins to be properly orientated in their native state.
This platform permits efficient development of antibodies with a substantial probability of selection of those with therapeutic activities. In addition, characterization of generated antibodies, including their specificity and their binding coefficients, is greatly facilitated through the MPL technology.
Most recombinant antibodies are developed for therapeutic purposes, but Biosite (www.biosite.com) is pursuing a substantial effort to generate diagnostic recombinant antibodies using its patented Omniclonal® antibody technology. This approach depends on the use of immunized mice whose spleens are harvested and used to build focused libraries for screening by phage display. Since the antibodies are already high affinity it is unnecessary to use phage-display maturation protocols to further hone their performance.
“As part of a diagnostics discovery program we have generated hundreds of antibodies against hundreds of biomarkers that are being used to build tests for cancer, sepsis, kidney injury, cardiovascular disease, and a number of other conditions,” said Gunars E. Valkirs, Ph.D., senior vp, Biosite .
Because the libraries are not isolated from hybridomas, they are essentially polyclonal libraries, reacting against numerous epitopes on the target.
Monoclonal antibodies are easily obtained since the libraries are collections of cells, each expressing a monoclonal antibody.
The range of new technologies illustrates the quantum leap in sophistication available to companies and academic researchers working in antibody engineering. The level of speed, sensitivity, and amount of data that can be processed through new instrumentation is truly remarkable. Yet the challenges, including autoimmune diseases, aging, failures of the central nervous system, cancer, and cardiovascular ailments, are formidable. All are extremely complex, poorly understood conditions that have proven resistance to 100 years of attack. It will take immense resources, skill, and imagination to produce the antibody therapies that will move medicine to the next level.
K. John Morrow, Jr., Ph.D., is president of Newport Biotech. Web: newportbiotech.com. Phone: (513) 237-3303. E-mail: [email protected].
