Current Drug Discovery Technologies - Volume 11, Issue 1, 2014

Historically, antibody variable domains for therapeutic antibodies have been sourced primarily from the mouse IgG repertoire, and typically either chimerized or humanized. More recently, human antibodies from transgenic mice producing human IgG, phage display libraries, and directly from human B lymphocytes have been used more broadly as sources of antibody variable domains for therapeutic antibodies. Of the total 36 antibodies approved by major maket regulatory agencies, the variable domain sequences of 26 originate from the mouse. Of these, four are marketed as murine antibodies (of which one is a mouse-rat hybrid IgG antibody), six are mouse-human chimeric antibodies, and 16 are humanized. Ten marketed antibodies have originated from human antibody genes, three isolated from phage libraries of human antibody genes and seven from transgenic mice producing human antibodies. Five antibodies currently in clinical trials have been sourced from camelids, as well as two from non-human primates, one from rat, and one from rabbit. Additional sources of antibody variable domains that may soon find their way into the clinic are potential antibodies from sharks and chickens. Finally, the various methods for retrieval of antibodies from humans, mouse and other sources, including various display technologies and amplification directly from B cells, are described.

Antibodies intended for clinical use have been isolated from non-human primates (NHP), chimpanzees (Pan troglodytes) and macaques (Macaca fascicularis and Macaca mulatta), essentially with the use of the phage-display technology. All studies presenting such isolations have been reviewed and presented here, following the main steps of this technology, and advantages and disadvantages of NHP species were analyzed. Optimization of the tolerance of chimeric NHP-human antibodies by germline humanization was mentioned, and the recent alleviation of legal constraints was revealed. The methodology combining the use of phage-displayed libraries built from immunised NHP with germline humanization should be chosen more frequently to develop well-tolerated IgGs, directed against infectious or human antigens.

The smallest independently folded antibody fragments, the domains, are emerging as promising scaffolds for candidate therapeutics and diagnostics that bind specifically targets of interest. The discovery of such binders is based on several technologies including structure-based design and generation of libraries of mutants displayed on phage or yeast, next-generation sequencing for diversity analysis, panning and screening of the libraries, affinity maturation of selected binders, and their expression, purification, and characterization for specific binding, function, and aggregation propensity. In this review, we describe these technologies as applied for the generation of engineered antibody domains (eAds), especially those derived from the human immunoglobulin heavy chain variable region (VH) and the second domain of IgG1 heavy chain constant region (CH2) as potential candidate therapeutics and diagnostics, and discuss examples of eAds against HIV-1 and cancer-related proteins.

Antibodies have emerged as powerful therapeutics effective for treating a number of human conditions and diseases. While early successes utilized small animals to generate therapeutic antibodies, human antibodies are now preferred in order to limit anti-antibody immune responses. Antibodies with human amino acid sequences can be generated in a number of ways, such as humanizing antibodies from other species or expressing human antibodies in transgenic animals. This review focuses on methods for obtaining antibodies directly from human B cells. These methods use both antigen exposed and non-exposed (“naive”) humans as B cell sources, and apply various technologies to isolate desired antibodies; including cell line generation, single cell isolation, display technologies, and B cell library generation.

Utilizing a vaccinia virus based library technology, we previously developed an antibody discovery platform that enabled efficient selection of fully functional IgG antibodies from highly diverse immunoglobulin gene libraries expressed on the surface of mammalian cells. Recently, we have further modified this platform to enable efficient expression of a library of fully human antibodies on the surface of vaccinia virus; an enveloped mammalian virus. Similar in concept to phage display, conditions are utilized under which each vaccinia virion expresses a single antibody specificity on its surface. Various panning and magnetic bead based methods have been developed to allow screening of a library of vaccinia- MAb virions and selection of recombinant vaccinia virus encoding specific antibodies. Upon infection of mammalian cells the antibody is not only incorporated into newly produced virus, it is also displayed on the surface of the host cell. Similar to methods utilized in yeast display, the cells displaying vaccinia encoded antibody can also be selected using a combination of magnetic beads and cell sorting, and the virus encoding the specific antibody heavy and light chains readily recovered and analyzed. This technology allows for rapid high throughput selection of vaccinia-MAb virions in a cell free panning system, followed by cell based screening for high specificity and fine selection of optimal antibodies.

Human therapeutic antibody discovery has utilized a variety of systems, from in vivo immunization of human immunoglobulin-expressing mice, to in vitro display of antibody libraries. Of the in vitro antibody display technologies, mammalian cell display provides a number of advantages with the ability to co-select immunoglobulin molecules for high expression level in mammalian cells, native folding, and biophysical properties appropriate for drug development. Mammalian cell display has been achieved using either transient or stable expression systems, using a number of alternate transmembrane domains to present antibody on the cell surface. The unique capability of mammalian cells to present IgG in its fully post-translationally modified format also allows selection of antibodies for functional properties. One limitation of mammalian cell based systems, however, has been the smaller library size that can be presented compared to phage display approaches. Until recently, this has necessitated the use of libraries biased toward a particular antigen, such as libraries derived from immunized donors, to achieve success. An alternative approach has now been developed which recapitulates key aspects of the in vivo immune system through reproducing somatic hypermutation (SHM) in vitro. Libraries representing a naïve human B lymphocyte antibody repertoire are created by PCR amplification of the rearranged (D)J segments of heavy and light chain variable regions from human donors and incorporating the resulting sequence diversity into panels of human germline VH and VL genes. The resulting antibodies are presented as full length IgG on the surface of HEK293 cells. After isolation of antibodies binding to individual target antigens, subsequent affinity maturation using in vitro SHM is induced by expression of activation-induced cytidine deaminase (AID). Selection of antibodies from naïve fully human libraries using mammalian cell display coupled with in vitro SHM is an efficient methodology for the generation of high affinity human antibodies with excellent properties for drug development.

Since Kohler and Milstein developed the process of generating hybridomas by fusing antibody secreting B cells with an immortal myeloma cell line, the techniques used to develop monoclonal antibodies for use as human therapeutics have progressed significantly. Here, we will briefly review hybridoma technology and the evolution of therapeutic antibodies for the treatment of human disease. We will focus on the evolution of humanized mouse models for the generation of therapeutic human antibodies, comparing the early models, such as severe combined immunodeficient (SCID) mice which do not engraft human leukocytes well due to residual innate immunity, to the more recently developed models such as non-obese diabetic (NOD)/SCID IL-2Rγ-deficient mice in which numerous human hematopoietic lineages can be cultivated. Building on the identification of suitable host strains for the reconstitution of human immune cells, focus has now shifted onto humanizing the murine microenvironment in order to support human immune cell function. Although several recent studies have shown that the provision of human soluble factors can support maturation and function of human immune cells, particularly within the myeloid compartment, this does not appear to impact antibody production significantly. Moreover, models in which grafting of human tissues is performed to provide human microenvironments which support human leukocyte maturation do show improved humoral immune function, but require several surgical manipulations for generation of the model. Ultimately the most desirable scenario is to generate transgenic models that can be bred efficiently and express a sufficient number of human molecules to support functional human immune cells and several groups have made progress in making this idea a reality. These studies in the context of the generation of human antibodies will be discussed.

Transgenic mice have yielded seven of the ten currently-approved human antibody drugs, making them the most successful platform for the discovery of fully human antibody therapeutics. The use of the in vivo immune system helps drive this success by taking advantage of the natural selection process that produces antibodies with desirable characteristics. Appropriately genetically-engineered mice act as robust engines for the generation of diverse repertoires of affinity- matured fully human variable regions with intrinsic properties necessary for successful antibody drug development including high potency, specificity, manufacturability, solubility and low risk of immunogenicity. A broad range of mAb drug targets are addressable in these mice, comprising both secreted and transmembrane targets, including membrane multi-spanning targets, as well as human target antigens that share high sequence identity with their mouse orthologue. Transgenic mice can routinely yield antibodies with sub-nanomolar binding affinity for their antigen, with lead candidate mAbs frequently possessing affinities for binding to their target of less than 100 picomolar, without requiring any ex vivo affinity optimization. While the originator transgenic mice platforms are no longer broadly available, a new generation of transgenic platforms is in development for discovery of the next wave of human therapeutic antibodies.

Antibodies have become one of the dominant therapeutic platforms due to their safety, specificity, and efficacy. Owing to their massive potential diversity intrinsic to their structure, the number of possible different molecules that could be generated and analyzed from natural or synthetic systems is almost limitless. However, even with vast improvements in automation, classic antibody generation and analysis systems are severely limited in the number of molecules that can be interrogated during a typical discovery campaign. When one considers more complex target systems, along with the desire to isolate antibodies with very unique characteristics, the chances are very low that these systems will be successful. Next generation sequencing technologies (also referred to as “deep sequencing”) allow for the analysis of single molecules in millions in a very short period of time. By applying these deep sequencing technologies to antibody discovery, we now have the ability to look for very specific molecules with very unique properties and activities, further our understanding of species and strain specific repertoires, and can now begin to use sequence information to identify function. The application of these technologies is opening the door to the discovery of next generation antibody therapeutics.