The mouse monoclonal antibody has been a dominant tool in research and diagnostic and therapeutic areas. However, a new technology for generating rabbit monoclonal antibodies offers an improved alternative for the scientific community and novel opportunities for diagnostics and therapeutics. The advantages of this new technology are manifold: higher affinity, improved specificity, and recognition of antigens and epitopes that are nonimmunogenic to rodents. While the immunoassay is the most popular analytical technique applied in clinical and nonclinical diagnoses, there is still a gap between the antibody developer and IVD manufacturer. This is caused, to a certain extent, by a lack of high-quality antibodies in immunoassays and by increased demand for better or novel antibodies along with the emergence of genomics, proteomics, and epitomics. Antibody performance related to sensitivity, specificity, linear dynamic range, matrix effect, and immunoassay standardization sets the primary determinant for the quality of IVD medical devices. This article discusses some characteristics of the rabbit immune system that result in better antibody generation, reviews rabbit monoclonal antibody technology, and considers its applications and potential in diagnostic areas.

A Better Immune System

Although rabbits are well-known for mounting a strong immune response against foreign antigens and rabbit polyclonal antibodies are the oldest and most widely used research tools in immunology, why rabbits generally produce better antibodies to many antigens than rodents is still not fully understood. Recent research advancements in rabbit immunology have shed some light on this question: after primary and subsequent immunizations, naïve B cells are stimulated to differentiate into antibody-secreting plasma or memory cells. T-cell epitopes presented by MHC class I and class II molecules are typically peptides, whereas nonclassical MHC molecules, CD1 family molecules, present nonpeptidic epitopes such as lipid or glycolipid. The CD1 family molecules can be classified into three groups: group 1 (CD1a, CD1b, and CD1c); group 2 (CD1d); and group 3 (CD1e). Interestingly, rabbits express all three groups, whereas mice only express group 2, but not groups 1 and 3. Group 1 molecules mainly present a vast range of lipid antigens to clonally diverse T cells that mediate adaptive immunity. By contrast, group 2 molecules present lipid antigens to natural killer T (NKT) cells.¹ Thus, the difference in the presentation of nonpeptidic epitopes between rabbit and mouse may explain why rabbits make better antibodies to lipids or glycolipids than mice in general.

Primary and secondary antibody repertoire developments in rabbits possess several unique characteristics², which also may account for the quality of the antibodies produced. First, unlike human and mouse, rabbit B cell development continues in gut-associated lymphoid tissues (GALT), including the appendix. That is where B cell follicles form and extensive B cell expansion occurs in response to intestinal microflora for nonimmunogen-specific antibody repertoire amplification, resulting in

a larger naïve antibody library. The primary high copy number repertoire is developed by somatic diversification of Ig genes mainly through gene conversion and, to a lesser extent, somatic hypermutation. Gene conversion—a unique IgG gene diversification mechanism for rabbits—is a process involving clusters of nucleotide changes from upstream genes to amplify the diversity of antibody repertoire.

Second, B cells in the rabbit GALT are selected by both positive and negative selective events. The cells that survive the selection process exit to the periphery to participate in further engagement with foreign antigens. Both gene conversion and somatic hypermutation occur in secondary lymphoid tissues such as spleen, lymph nodes, and Peyer’s patches for secondary antibody repertoire diversifying and affinity maturation. Indeed, the features of B cell follicle expansion in GALT and the gene conversion mechanism for antibody repertoire diversification provide greater rabbit antibody selection.

Third, the rabbit antibody is somewhat simpler than mouse and human antibodies. Rabbit IgG has no subclass because it has only one Cγ gene, and the majority (90 to 95%) of light chains is derived from isotype Cκ1. Only 5% to 10% of total IgG light chains are isotype l. These characteristics of rabbit IgG make gene cloning much easier than with mouse or human IgG. Rabbit IgG has extra disulfide bonds in the variable region of the heavy chain; an extra disulfide bond also exists between Vκ and Cκ in commonly used b4 and b5 allotype rabbits. These extra disulfide bonds may result in the great stability and long shelf-life of rabbit antibodies (Figure 1). It is noted that rabbits display less immunodominance and less IgM occurrence when polymers with repeated domains such as lipids or carbohydrates are immunized as immunogens. Although the rabbit immune system generally sets the foundation for better antibody selection, in order to develop the most suitable antibody for a particular application one must also consider the immunogen quality, its structure and format; the conjugation of an immunogen with a carrier, if necessary; the immunization scheme; antibody screening assays; and the screening strategy.

Rabbit Monoclonal Antibody Technology

Rabbit monoclonal antibodies combine the best properties of monoclonal antibodies with the most desirable attributes of rabbit polyclonal antibodies.^3-7 In 1995, Katherine Knight and her colleagues at Loyola University of Chicago developed a plasmacytoma cell line, termed 240E-1, from a double transgenic rabbit overexpressing the oncogenes v-abl and c-myc. Fusion of 240E-1 cells with rabbit lymphocytes produced hybridomas that secreted rabbit monoclonal antibodies.⁸ However, the stability of 240E-1 derived hybridomas was a concern. The resulting hybridomas were not genetically stable and antibody activity was often lost after multiple passages of cells. In 1996, Weimin Zhu and Robert Pytela, then at University of

California, San Francisco, put forth great effort into improving 240E-1 and developed a better fusion partner cell line, 240E-W, with high fusion efficiency and reliable hybridoma stability. A number of characteristics that distinguish 240E-W from its parent 240E-1 were observed.⁹ This new fusion partner cell line has made rabbit hybridoma production a routine lab technique, making large-scale development of rabbit monoclonal antibodies a reality. Similar to the original mouse fusion partner cell line, the 240E-W cell line inherited the undesirable traits of endogenous IgG chains.⁸ To solve this problem, Weimin Zhu and his colleagues at Epitomics developed new derivative lines of 240E-W, termed 240E-W2, which does not carry an endogenous heavy chain gene, and 240E-W3, which lacks both endogenous heavy and light chain genes. Table I summarizes the features of different rabbit fusion partner cell lines.

The rabbit hybridoma generation procedure is similar to that of mouse hybridomas (Fig 2). So far, with the newer fusion partner cell lines (240E-W2 and 240E-W3), thousands of hybridoma cell lines have been generated for life science, diagnostic, and therapeutic applications. Among them, several thousand research reagents and more than 150 IVD immunohistochemistry (IHC) products have been used in the market, and three rabbit monoclonal antibodies (RabMAb) have been approved by FDA as Class II or Class III IVD products. Epitomics has developed a humanized RabMAb drug candidate, which was demonstrated to have better in vitro and in vivo efficacies than a commercial benchmark.¹⁰ This drug candidate is currently in clinical trials.

Advantages of Rabbit Monoclonal Antibody

High sensitivity and specificity. Typically, RabMAbs have 10 to 100 times higher affinity than mouse monoclonal antibodies. Our recent Kd measurement data for 1000 RabMAbs (not shown) has substantiated this. Identical or similar epitopes sometimes may be found on apparently unrelated molecules. As a result, antibodies directed against one antigen may react

unexpectedly with an unrelated antigen. Thus, the specificity of an antibody is extremely important to avoid a false positive or signal-to-noise ratio problem in diagnosis. With rabbit’s larger antibody repertoire and stringent antibody screening, a highly specific RabMAb can be developed, which has been well-demonstrated in the IHC diagnostic application. In comparison with mouse monoclonal antibodies to the same targets, RabMAbs detect with higher sensitivity and specificity, giving cleaner staining patterns with lower or no background on paraffin-embedded tissue.¹¹ In addition, each immunized rabbit spleen contains as much as 50 times more lymphocytes than a mouse spleen. Rabbit lymph nodes, bone marrow, and other lymphatic organs also provide numerous sources of B cells for generating a larger hybridoma pool for the selection of higher affinity and specific hybridoma clones. There are various techniques to enhance the detection sensitivity for immunoassays; however, the signal-to-noise ratio (specificity) typically does not improve. We believe a better quality antibody is the best solution to address the dual issues of sensitivity and specificity.

Novel epitope recognition. Rabbit antibodies have been shown to recognize epitopes that are not recognized by mouse antibodies. This phenomenon may be caused by reduced immunodominance in the rabbit, which allows the recognition of low abundant proteins or epitopes that usually are not visible to the mouse immune system.⁹ It has also been shown that rabbits have higher success rates than mice in generating antibodies against these small epitopes, such as protein modification or cleavage sites, or conformational epitopes. The ability to develop rabbit monoclonal antibodies against novel epitopes offers a new opportunity for immunodiagnostics since many disease targets are in low abundance or form complexes with other proteins. RabMAb to estrogen receptor (ER), SP-1 clone, an FDA approved antibody for pathological assessment of breast cancer prognosis and treatment, is one such example. Because of the novel epitope recognized by this antibody, the sensitivity and specificity for the pathological assessment are improved significantly.¹² Another example is RabMAb to ERG, clone EP110 or EPR3864, which demonstrates exquisite concordance between ERG protein expression and the presence of ERG gene rearrangements in prostate cancer because of its ability to recognize a rare epitope.¹³

Easy to recognize small molecules or haptens. It is observed that rabbits develop antibodies to small molecules more easily than mice—the nonpeptidic epitope presenting mechanism described previously may provide an explanation.

Small molecules such as drugs, steroid hormones, lipids or glycolipids, and food or environmental contaminants with a molecular weight less than 1 kDa are too small to be appropriately processed and presented to the immune system. As a result, they are not immunogenic. However, if the small molecule is chemically linked to a large protein molecule, a specific antibody to small molecules or chemical groups can be generated. Studies have shown that any alteration in the shape, size, or charge of a hapten changes its ability to bind to antibodies; thus, it is essential to have a large diversifying antibody repertoire combined with comprehensive nonpeptidic epitope presenting molecules as a basis for developing high-quality antibodies to these haptens. Apparently, the rabbit immune system is developed to fulfill this purpose. A lot of surface antigens from pathogenic microbes are carbohydrates, lipids, or glycolipids. Seemingly, rabbits possess a well-suited immune system to fight these microbes and are able to produce high quality antibodies for IVD applications.

Applications in Diagnostics

Anatomic pathology. One of the most widely used applications for RabMAbs is IHC for anatomic pathology. RabMAbs are well-recognized in pathology labs as best-in-class reagents for IHC, and have taken the majority of market share from mouse monoclonal antibodies for many important markers. Two RabMAbs received FDA approval as companion diagnostic antibodies for IHC tests of Her-2 and c-Kit for cancer drugs Herceptin (Genentech) and Gleevec (Novartis), respectively. One RabMAb to estrogen receptor (ER) received FDA 510K approval as a Class II diagnostic reagent for breast cancer prognosis. Major players in anatomic pathology diagnosis have adapted RabMAbs in their automated immunostainer systems. EP Clones, a new generation of RabMAbs provided by Epitomics, are recognized as premium antibodies for anatomic pathology. For example, the RabMAb against ETS TMPRSS2-ERG fusion protein is a gold-standard clone for prostate cancer diagnosis and prognosis.

Companion diagnostics. Rab-MAbs have been demonstrated to be superior reagents for detecting subtle protein modifications such as phosphorylation, methylation, acetylation, and glycosylation. The posttranslational modifications of proteins play central roles in cell-signal transduction and other cellular activities, which could reflect a change of disease status as both a disease target and/or biomarker. In addition, the pathogeneses originates not only from changes in protein sequence or expression level but also from changes in protein conformational structure or even abnormal interactions among proteins such as prion disease. Therapeutic companies have discovered or identified a large number of disease targets in these categories and are developing companion diagnostic antibodies for patient selection and therapy efficacy monitoring as they develop therapeutic products. We believe RabMAb technology will allow the development of antibodies that quantitatively measure subtle changes in biomarkers with better specificity and sensitivity than current techniques. Indeed, RabMAbs are becoming the first choice of reagent for companion diagnosis, and more products are expected to enter the market in the years ahead.

Immunoassays for small molecules. Immunodiagnostics in body fluids such as blood, urine, and saliva is the largest IVD segment. Most point-of-care tests belong to this category. From an antigen perspective, immunodiagnostic tests include microbial antigens such as viral, bacterial, fungal, and parasitic antigens; nonmicrobial antigens such as tumor antigens and auto-antigens; and nonclinically relevant antigens such as food toxins or environmental contaminants. Antibodies to small molecules have vast diagnostic applications such as steroid hormone test, food safety assessment, drug abuse detection, and drug metabolite concentration monitoring. A set of RabMAbs to steroid hormones such as estradiol, estriol, and progesterone and food toxic substances such as chinolone, enrofloxacin, flumequine and neomycin have been developed. Typically these RabMAbs demonstrate better sensitivity and specificity in comparison with rabbit polyclonal antibodies or mouse monoclonal antibodies. We have developed a RabMAb to THC (a marijuana substance) for a drug test with 1000 times the sensitivity that can be achieved through commercial mouse monoclonal antibodies (data not shown). RabMAbs to multiple epitopes of polyethylene glycol (PEG) have allowed us to develop unique detection kits that are widely used in the pharmaceutical field. The RabMAb to thiophosphate esters has advanced scientific research.¹⁴

The diagnoses of many bacterial infectious organisms such as strep A, tuberculosis, legionella, H. pylori, and chlamydia detect glycolipids, polysaccharides, or lipids on bacterial surfaces. Normally, mouse is not the ideal species for generating antibodies to this type of antigen, and RabMAb technology presents a great opportunity for developing better quality antibodies. Several RabMAbs to these targets are under development through the collaboration between Epitomics and IVD companies. In addition, glycomics and lipidomics are emerging scientific disciplines for understanding the biological and pathological functions of lipids, glycolipids, and carbohydrates at the cellular and organismal levels. RabMAb will be an important research tool, and some RabMAbs may have diagnostic applications.

Viruses usually undergo mutation and selection and gradually change their structure to avoid the surveillance of the host immune system. These changes lead to alterations in the antigenicity of the virus, called antigenic drift. This, in turn, makes clinical diagnosis more difficult. Parasitic protozoa can also manage to survive within their host by altering surface antigens rapidly and repeatedly, as well. The larger antibody repertoire derived from rabbit offers a better chance to zoom in on such subtle changes in epitopes and develop better diagnostic reagents in comparison with mouse antibody technology. This has been demonstrated by a few cases (data not shown).

Immunoassays for cancer and cardiac markers. In real biological samples, such as blood, serum or urine, target antigens are more complicated than purified immunogens. These antigens could exist as free forms, chemically modified variants, complexes with other proteins, or degraded fragments. In addition, the challenge involving immunoassay standardization is mainly caused by epitope variability, the matrix effect, and the lack of dynamic linear range of antibody and antigen interaction. A significant number of cancer antigens are glycoproteins on tumor cell surfaces such as CA-50, CA-153, CA-199, and CA72-4, and many antibodies actually recognize the carbohydrate epitopes on the glycoproteins instead of protein. Developing more-sensitive or specific antibodies to these carbohydrate epitopes will lead to better or novel diagnoses. In the cardiac disease diagnostic area, more-sensitive and specific antibodies to Troponin I, BNP (B-type natriuretic peptide), and D-dimer are needed for better or early detection of acute coronary syndrome, congestive heart failure, and thrombotic disorders. Typically, antigens isolated from patient body fluids are a better source for developing high-quality antibodies, and clinical sample screening is also necessary to identify the appropriate clones. We believe the new form of antibody, RabMAb technology, will change the diagnostic landscape when more best-in-class antibodies are developed and incorporated into medical devices.

Conclusion

The rabbit immune system generates antibody diversity and optimizes affinity through mechanisms that are more efficient than those of mice and other rodents. This increases the possibility of obtaining better quality or novel antibodies for diagnostic applications. The simplicity of rabbit IgG gene cloning makes recombinant RabMAb production easy and scalable. Future advances in antibody-based diagnostics will continue to rely on the availability of high-quality or novel antibodies. Significant improvements in antibody affinity and specificity should make diagnosis, prognosis, and companion diagnosis more sensitive, precise, and accurate. Indeed, there are a number of diseases that are under-diagnosed because of a lack of high-quality antibodies. We believe immunodiagnostics with RabMAbs offers a unique opportunity to address these unmet medical needs. Furthermore, new RabMAbs to targets/epitopes that are not immunogenic or weakly immunogenic in rodents will significantly enhance the power of immunoassays and ultimately expand the scope of immunodiagnostics.

References
1.    D Barral and M Brenner, “CD1 antigen presentation: how it works,”  Nature Reviews / Immunology 7 (2007): 929-941.
2.    RG Mage, D Lanning, KL Knight, “B cell and antibody repertoire development in rabbits: The requirement of gut-associated lymphoid tissues,” Development & Comparative Immunology 30 (2006): 137–153.
3.    RM Krause, “The search for antibodies with molecular uniformity,” Adv Immunol. 12 (1970): 1-56.
4.    RM Krause, “Experimental approaches to homogenous antibody populations: factors controlling the occurrence of antibodies with uniform properties,” Fed Proc. 29 (1970): 59-65.
5.    JC Bystryn, JS Jacobsen, P Liu et al., “Comparison of cell surface human melanoma-associated antigens identified by rabbit and murine antibodies,” Hybridoma 1 (1982): 465-472.
6.    A Weller, J Meek, ED Adamson, “Preparation and properties of monoclonal and polyclonal antibodies to mouse epidermal growth factor (EGF) receptors: evidence for cryptic EGF receptors in embryonal carcinoma cells,” Development 100 (1987): 351-363.
7.    TJ Raybould, M Takahashi, “Production of stable rabbit-mouse hybridomas that secrete rabbit mAb of defined specificity,” Science  240 (1988): 1788-1790.
8.    H Spieker-Polet, P Sethupathi, PC Yam et al., “Rabbit monoclonal antibodies: generating a fusion partner to produce rabbit-rabbit hybridomas,” Proc Natl Acad Sci USA,  92 (1995): 9348-9352.
9.    W Zhu and G-L Yu,  “Rabbit Hybridoma,” in Therapeutic Monoclonal Antibodies: from bench to clinic, (Wiley, 2009) 151-168
10.    Y Yu et al., “A Humanized Anti-VEGF Rabbit Monoclonal Antibody Inhibits Angiogenesis and Blocks Tumor Growth in Xenograft Models,” PLoS ONE 5 (2010): 2-12  
11.    S Rossi et al.,  “A Comparative study between a novel category of immunoreagents and the corresponding mouse monoclonal antibodies,”  Am J Clin. Pathol. 124 (2005): 295-302.
12.    Z Huang et al., “Development of new rabbit monoclonal antibody to estrogen receptor: immunohistochemical assessment on formalin-fixed, paraffin-embedded tissue sections,”  Appl Immunohistochem Mol Morphol. 13 (2005): 91-95.
13.    K Park et al., “Antibody-based detection of ERG rearrangement-positive prostate cancer,” Neoplasia 12 (2010): 590-598.
14.    J Allen, “A semisynthetic epitope for kinase substrates,” Nature Methods 4 (2007): 511-516.

Weimin Zhu is cofounder and senior vice president, Antibody Technology, at Epitomics, an Abcam Co., in Burlingame, CA. He can be reached at [email protected].