Use of Human Hybridoma Technology To Isolate Human Monoclonal Antibodies

Abstract

The human hybridoma technique offers an important approach for isolation of human monoclonal antibodies. A diversity of approaches can be used with varying success. Recent technical advances in expanding the starting number of human antigen-specific B cells, improving fusion efficiency, and isolating new myeloma partners and new cell cloning methods have enabled the development of protocols that make the isolation of human monoclonal antibodies from blood samples feasible. Undoubtedly, additional innovations that could improve efficiency are possible.

HISTORY OF HYBRIDOMAS

Monoclonal antibodies (mAbs) have revolutionized the conduct of science since their first description in 1975 (1). The use of these specific reagents also has made possible improved clinical diagnostics in the medical arena, and many antibodies have found their way to clinical use as prophylactic or therapeutic agents. Nevertheless, the potential of mAbs derived specifically from technology based on human hybridomas remains largely unfulfilled. The principal reason for the lack of a large number of hybridoma-derived mAb therapeutics has simply been the technical difficulty in generating stable hybridomas that secrete human mAbs of high affinity and functional activity. This chapter reviews recent efforts to develop and employ novel methods for the efficient generation of human hybridomas secreting human mAbs for clinical use.

The principal advantage of the use of human hybridoma technology for mAb generation is that this approach preserves the authentic sequence and pairing of antibody DNA from a natural B cell for the expression of a naturally occurring full-length human mAb. There are significant theoretical advantages for expressing cDNAs encoding authentic heavy and light chains with chains that are paired using the coding sequence as it was generated naturally through B cell selection, class switch, and affinity maturation. No genetic modification of these sequences is required. Since antibody expression is typically very stable in hybridomas, sequence amplification of antibody variable genes is achieved easily if recombinant production or manipulation is desired. The resulting recombinant mAb retains most features of naturally occurring human antibodies, as the clone retains the native amino acid sequence and heavy/light chain pairing. Since the native constant region of the antibody in the original human B cell is retained in the mAb expressed by the resulting hybridoma, the functional properties of the particular Fc region can be studied for Fc-mediated activities, such as antibody-dependent cellular cytotoxicity.

Despite the advantages inherent to making natural human mAbs, the low efficiency of the hybridoma isolation process historically was a technical drawback that was too great to overcome, and many other methods of producing human mAbs have been used instead in recent years to meet the demand for generation of therapeutic antibodies. The principal disadvantage of the hybridoma method is low fusion efficiency. Over the last several decades, however, this problem has been overcome slowly through improvements in several technical features of the process. These improvements will be discussed in more detail throughout this article. Currently, human hybridomas can be generated with great efficiency and throughput. Panels of antigen-specific human hybridomas secreting full-length naturally occurring mAbs to a large number of targets have been developed recently using human peripheral blood mononuclear cells as the starting material.

FIRST HUMAN HYBRIDOMAS

The prospect of using human mAbs for the prevention or treatment of human diseases was evident early on and was the driving force behind intense effort put into the development of human hybridoma methods. Initial studies were done in the early 1970s using mouse myeloma cells, fusing them with primary human B cells (2). A human tetraploid hybridoma also was made through the fusion of two human lymphocyte lines (3). It was not until 1980, however, that the first successful human mAb was produced (4). This feat was achieved by fusing lymphoid cells harvested from spleens obtained during staging laparotomy from patients with Hodgkin’s lymphoma with the human myeloma cell line U266. This accomplishment was a large step toward the use of human hybridomas to make mAbs, as it proved the feasibility of the method. The large numbers of lymphoid cells used in that fusion process were sufficient to overcome the low fusion efficiency at the time. The challenge of identifying antigen-specific cells and expanding them to numbers that enabled researchers to overcome the barrier of low fusion efficiency would, however, require several more decades of investigation.

One significant obstacle to the generation of human hybridomas over the years was the inability to consistently expand desired populations of antigen-specific B cells. Antigen-specific memory B cells generally circulate at low frequenciesinthe peripheral blood, typically in a range that centers around 1 in 10,000 B cells or lower. For example, precursor frequencies of anti-tetanus toxoid-specific B cells were initially reported following immunization as being approximately 1 per 10,000 peripheral blood mononuclear cells (PBMCs) (5). Fusion efficiencies typically have not been sufficient to immortalize enough cells from the number of B cells that can be obtained from humans by routine phlebotomy. When combined with a fusion efficiency of 1 in 100,000 B cells in the suspension, as is typical of polyethylene glycol (PEG)-mediated cell fusion, one would need approximately 1 liter of blood to obtain a single tetanus toxoid-specific human hybridoma under these circumstances. Consequently, the generation of human hybridoma cells secreting desirable human mAbs without expanding the frequency or number of antigen-specific B cell populations was difficult.

Human B cells can be immortalized by methods using Epstein-Barr virus (EBV) transformation (6, 7). EBV transformation is often carried out on samples containing large numbers of B cells, either from PBMCs or cells obtained from lymphatic tissues such as tonsil cells. These EBV-transformed B cell lines, which contain many clones of differing specificities, must be grown for a considerable period of time to allow for the emergence of immortalized lines. Since immortalization does not occur with every clone of transformed B cells, production of antigen-specific clones was historically very difficult. Moreover, EBV-transformed B cells generally grow poorly, often secrete low amounts of antibodies, and can exhibit chromosomal instability (6, 8–10). Recently, the addition of CpG oligonucleotide a toll-like receptor (TLR) 9 agonist during B cell transformation was shown to greatly facilitate the transformation process (11–13). This stimulation significantly improved the efficiency of transformation and has been used to expand antigen-specific B cells for the clonal amplification of their antibody variable sequences by reverse transcriptase PCR (13).

The prospect of using EBV-transformed B cells for the generation of human hybridomas was first realized in the early 1980s. This method was achieved initially by fusing a human EBV-transformed B cell line producing antigen-specific mAb with a murine myeloma to produce a stable heterohybridoma secreting mAb (14, 15). This procedure resulted in a more stable, higher-producing clone. Hypoxanthine phosphoribosyl transferase-deficient variant EBV-transformed B cell lines also were developed for use as human fusion partners (16). The use of an immortalized EBV-transformed B cell clone to create a stable fully human hybridoma was described in 1984 (17). A fully human hybridoma was generated successfully by fusing a cytomegalovirus-specific EBV-immortalized B cell line with a hypoxanthine phosphoribosyl transferase-deficient human lymphoblastoid partner created previously. Many groups have now used variations of these techniques to employ EBV transformation to expand populations of antigen-specific B cells. These methods have improved mAb yield and hybridoma generation throughput considerably. In fact, EBV transformation continues to be used to this day to expand B cells prior to human hybridoma generation.

B CELL SOURCE

One of the critical factors underlying the difficulty in generating human hybridomas is the low frequency of antigen-specific B cells in peripheral blood under normal circumstances. Typically, antigen-specific B cells are present in peripheral blood at a frequency of ≤0.1%. Therefore, large numbers of PBMCs are required from individuals to obtain a significant number of antigen-specific B cells. For example, if a desired antigen-specific B cell occurs at a frequency of 1 in 1,000, and PBMCs contain 10% B cells, then 1 million PBMCs contain only 100 antigen-specific B cells. Several strategies have been devised to enrich and/or expand antigen-specific B cell populations prior to fusion.

The most readily available source of antigen-specific B cells is peripheral blood. For the most part, directly fusing PBMCs with a myeloma partner is not productive in generating a desired antigen-specific human hybridoma, due to the limited number of antigen-specific cells in circulation. Enrichment of antigen-specific B cells from large samples of PBMCs using either fluorescence activated cell sorting or other cell isolation techniques (such as magnetic bead separation) often results in only modest improvement in the yield of functional cells.

The use of various cytokines, or cell lines made to express B cell cytokines, to amplify single B cells in culture has been described in the B cell field, principally using CD40-ligand (CD40L or CD154) expressing cells and recombinant human cytokines (interleukin 2 [IL-2], IL-4, and IL-10) (18). Synergy was also noted with the combination of CD40L and IL-21 (19). Feeder cell lines have also been used, based on their ability to support primary human B cell growth, frequently following gamma irradiation to prevent them from dividing. Several feeder cell lines have been described for this purpose: a fibroblast cell line transfected with the human CD40L molecule (20), an immortalized cell line FDC-H1 with features of human follicular dendritic cells (21), the U-937 human macrophage-like cell line (ATCC-CRL-1593.2) (22), and the EL4-B5 mutant thymoma cell line (23).

In addition to human CD40L, B lymphocyte stimulating factor (BLyS), also known as BAFF (for B cell activating factor belonging to the tumor necrosis factor family) (24), has been used to support primary B cells in culture providing costimulatory signals. BAFF is a relatively new member of the tumor necrosis factor family that was simultaneously identified by four different laboratories and named BAFF, BLyS, TALL-1, THANK, and zTNF4. The cytokine is expressed abundantly in monocytes and macrophages and is upregulated by interferon gamma. Endogenous BAFF is processed intra-cellulary by a protease of the furin family of proprotein convertases. Secreted BAFF acts as a potent B cell growth factor. BAFF plays an important role as costimulator of B cell proliferation and function. Using soluble cytokines or feeder cells able to provide costimulator function and/or cytokines, single B cells can be expanded to clones of a few hundred cells in several weeks. Antigen-specific B cell clones can be confirmed by testing cell supernatant specificity using enzyme-linked immunosorbent assay. The expanded antigen-specific cells then can be immortalized by fusion with a myeloma to form a hybridoma.

The simplest and most common strategy for the amplification of B cell populations using PBMCs is to use EBV transformation (25). Freshly isolated or cryopreserved PBMCs, obtained from an individual or individuals who have been identified as possessing B cells of the desired antigen specificity, can be expanded in an oligoclonal manner using EBV transformation. Using culture supernatant of the marmoset cell line B95.8, which contains high titers of EBV, B cells can be infected via complement receptor 2 (CD21), and transformants emerge over the course of 1 to 2 weeks, forming large colonies of cells known as lymphoblastoid cells (often abbreviated LCLs) (26). Cultures can be supplemented with cyclosporin A to inhibit EBV-specific T cells, which are present in many individuals, from killing B cell transformants (27). The overall efficiency of EBV transformation was improved dramatically through the use of CpG oligonucleotide, increasing transformation of B cell populations from <10% to >30% (13)(see Fig. 1).

FIGURE 1.

Lymphoblastoid cell formation from PBMCs. doi:10.1128/microbiolspec.AID-0027-2014.f1

Recently it was shown that by adding an inhibitor of EBV-associated apoptosis, a pharmacologic inhibitor of the serine/threonine-protein kinase 2, which is required for checkpoint-mediated cell cycle arrest, results in further improvement in lymphoblastoid cell survival and expansion (28, 29). In the end, increased numbers of lymphoblastoid cells result in a greater probability of successfully producing hybridomas secreting antibodies of interest. Next, EBV-transformed B cell cultures are screened for antigen-specific antibody production using enzyme-linked immunosorbent assay or other multiwall plate assays. Cultures producing the desired antibody then are fused with a myeloma cell line in an oligoclonal manner. The resulting oligoclonal hybridoma lines are then biologically cloned by a physical method to isolate single cells, using one of the methods discussed later in this article, to produce a human hybridoma secreting an antigen-specific mAb.

FUSION METHODS

The most difficult and critical step in the production of human hybridomas is the ability to efficiently fuse desired populations of lymphocytes with a myeloma partner. There are three basic techniques used to generate hybridomas for the purpose of mAb production: (i) the use of chemical agents such as PEG, (ii) fusogenic viruses, and (iii) electrical cytofusion. The first method used to study cell fusion took advantage of viruses, principally Sendai virus, capable of fusing together the membranes of two cells. The most common technique, however, employed in the 1975 seminal paper by Kohler and Milstein (1), with variations of the theme still commonly used today, involves chemical fusion using PEG. Finally, protocols based on the use of electrical currents to align and fuse the membrane of cells for the efficient generation of hybridomas were developed and optimized.

Viruses were first used as a means to fuse cells in the 1960s and early 1970s (30). There are several viruses, or their fusion proteins, which have been used successfully for the purpose of hybridoma generation. The two most commonly used are Sendai virus and vesicular stomatitis virus. Sendai virus, also referred to as murine parainfluenza virus type 1 or hemagglutinating virus of Japan, is a paramyxovirus which was the first animal virus whose mode of infection was elucidated (31). The viral fusion protein, or F protein, makes up part of the exposed spikes on the envelope of the virus and inserts into the cell membrane, allowing for introduction of the viral nucleocapsid into the cytoplasm of the host cell. The use of Sendai virus for cell fusion was described thoroughly by Okada in 1993 (32). One advantage of using Sendai virus is its broad tropism, which is mediated by the hemagglutinin-neuraminidase protein that recognizes sialic acid as a receptor on the host cell membrane. Kits containing inactivated Sendai virus are commercially available for the generation of murine hybridomas. The second virus that has been used for the fusion of cells to generate hybridomas for mAb secretion is vesicular stomatitis virus (33). In a study performed by Nagata et al., the efficiencies of hybridoma generation were compared using Sendai virus, vesicular stomatitis virus, and PEG-mediated cell fusion (34). Interestingly, their results suggested that the isotypes of antibodies obtained were influenced by the fusion method, as vesicular stomatitis virus-mediated fusion resulted in greater production of IgG mAbs (34). A related method that has been used is to transduce cells with viral fusogenic proteins. Unfortunately, the resulting cells often continue to fuse for as long as the proteins are expressed, which makes this method less optimal.

The most common method used to produce hybridomas takes advantage of the fusogenic properties of PEG, discovered by Kao and Michayluk in 1974 (35). The exact mechanism by which PEG-mediated cell fusion occurs is not known. Most experts believe that the mechanism primarily involves volume exclusion, driving adjacent cell membranes together (36). Unfortunately PEG is quite toxic to cells, as it can fuse multiple cells, resulting in giant polykaryons, and can even result in the fusion of intracellular membrane structures (37). The fusion process, however, is relatively simple and cheap. B cells are mixed with a myeloma fusion partner and suspended in a solution containing PEG in a drop-wise fashion. After a short incubation, fusion solution is removed and the cells are washed and resuspended in selection medium. Different PEG preparations have been shown to result in different efficiencies of cell fusion (38). The main problem with using PEG for the generation of human hybridomas is its low efficiency of fusion. Estimates of average fusion efficiencies are in the order of 1 hybridoma generated per 100,000 starting B cells. When fusing murine splenocytes from hyper-immunized animals (which can possess about 108 B cells), an extremely low inefficiency can be tolerated if one simply aims to find one or several mAbs that bind to an antigen. However, when using human samples such as blood (yielding about 1 × 10⁶ PBMCs or about 50,000 B cells per ml of blood), or more precious samples such as tumor infiltrating lymphocytes, such inefficiencies make success unlikely.

The most efficient method of cell fusion and hybridoma generation is electrical cytofusion. This method of cell fusion is an essential step in the most innovative techniques in modern human hybridoma protocols and has been optimized for the purpose of human hybridoma generation (39). At the center of this membrane fusion technique is the concept of electroporation. The application of high-intensity electric field pulses to cells causes transient membrane permeabilization. The extent of permeabilization is thought to depend on several physical parameters associated with the technique such as pulse intensity, number, duration, shape, and interval. Electric field intensity is the most critical parameter in the induction of permeabilization. The intensity must exceed a critical threshold for membrane permeabilization to be induced. The electric field intensity delivered depends on the cell size. The extent of permeabilization (which correlates experimentally with the flow rate across the membrane) is controlled by both pulse number and duration (40). Increasing the electric field intensity above the critical threshold needed for permeabilization results in an increase in the area of the membrane that is involved. Permeabilization is transient and disappears with time after delivery of the electric field pulses. The half-life of permeabilization may be under the control of the electric field parameters. The rate of resealing of the pore in the membrane may be influenced by both pulse duration and number but is independent of the electric field intensity that creates the permeabilization (40). Therefore, optimal evaluation of critical parameters for membrane permeabilization requires flexible control over each aspect of the electrical field. One of the more recent innovations in the electroporation hardware used for this purpose is that many of these parameters can be controlled and varied independently. Previously designed electroporators did not have such precise control of all variables, thereby likely introducing uncontrolled variation in experiments.

Membrane effects of applied electrical fields in electrofusion are similar to those in electroporation except that membranes in close contact can fuse together during the process of pore formation. Therefore, electric field intensities used in electrofusion are similar to those used in electroporation. Electrofusion is performed in a sequence of steps. First, cells are brought into contact with other cells. Cell-cell contact can be achieved by several methods, although none has been sufficiently optimized in our estimation. Chemical methods such as avidinbiotin bridging have been used to bring together two cell types specifically (41–44). Chemical methods require more manipulation than other methods but can be useful in certain circumstances. Physical methods such as centrifugation can bring cells into contact prior to (or after) the fusion pulse (45). Electroacoustic fusion of cells in sugar solutions and of cells brought into contact in an ultrasonic standing wave field has been described (46). In fact, simple centrifugation followed by electroporation can be used to create cell fusion. The most commonly described effective method for bringing cells into contact prior to electrofusion, however, is termed dielectrophoresis. Dielectrophoresis is achieved within a suspension of cells using an alternating current electrical field.

In any fusion method, sufficient force must be applied to each cell to overcome the negative surface charge. If electricity is used, merely applying a uniform electric field will not move a cell because the net charge of the cell is zero. Thus, from the definition of electric field there is no force applied. However, a nonuniform field moves the positive ions inside each cell to one side and the negative ions to the opposite side, producing a dipole. Once the dipole is induced, a net force is exerted on the cell because the intensity of the field is greater on one side than the other. The movement of cells in one direction causes the cells to concentrate in an area. Since the cells are now dipoles, the negative side of one cell will attract the positive side of another cell, overcoming the negative surface charge. A photomicrograph illustrating pearl chain formation is shown in Fig. 2.

FIGURE 2.

(A) Pre- and (B) post-pearl chain formation. doi:10.1128/microbiolspec.AID-0027-2014.f2

The second step in electrofusion is to apply one or more high-voltage pulses to the cells, inducing membrane fusion. The voltage required must be above a threshold to induce membrane breakdown and below a maximum voltage that causes cell death. The threshold voltage is approximately one volt across the cell membrane or two volts across two membranes (47). The voltage across a cell is equal to 1.5 times the cell radius times the electric field strength times the cosine of the angle of the membrane in relation of the direction of the field. This is the same formula used for electroporation. Multiple fusion pulses may be more efficient than a single pulse.

The last step in the electrofusion process (when using dielectrophoresis as an alignment tool) is postfusion alignment using alternating current fields. Electrofusion is a process that continues to occur over some time after the fusion pulse is applied. Reapplying dielectrophoresis after the fusion pulse allows maturation of the fusion process by holding cells in optimal alignment and contact.

Optimizing of dielectrophoresis and electrofusion parameters has been achieved partly via observable events through the use of a microscope slide or coaxial electrode. “Pearl chain length” (length of aligned cells) can be increased by increasing the voltage or increasing the time of prefusion sine wave application. Increased pearl chain length may not always be advantageous, however. It is unclear at this point whether long pearl chains are advantageous or not. According to Zimmermann (47), the fusion of adjacent cell membranes of cells in a pearl chain is a stochastic process. Thus, the electroporation conditions generate a probability of fusion of adjacent cell membranes. This means that it is possible to select electrofusion conditions that generate predominantly two cell fusions, even if pearl chain length is long in contrast to two-cell-only pearl chains.

The goal of electrofusion, at least as applied to hybridoma formation, is to generate the most B cell-myeloma hybridomas. Through the use of selective medium preparations, fusions of like cells (either B cells or myeloma cells) are not productive. Likewise, unfused B cells and myeloma cells are also unable to survive. This selection is achieved through the use of medium containingHAT (hypoxanthine-aminopterin-thymidine) and ouabain. Unfused myeloma cells or fused myeloma cells that do not contain a B cell nucleus are killed by aminopterin. Without being fused to a B cell, the myeloma fusion partner is not able to perform de novo synthesis of DNA and is forced to undergo apoptosis. The B cell possesses the ability to overcome this blockade by performing the salvage pathway, for which hypoxanthine and thymidine are used as raw material for building DNA. Unfused B cells or fused B cells that do not contain a myeloma cell nucleus are killed by ouabain. Without being fused to the ouabain-resistant myeloma cell, the B cell (particularly transformed B cells) is highly susceptible to the plasma membrane sodium pump inhibiting effects of ouabain, and apoptosis is induced. The only products of fusion to survive in the selective medium are the B cell-myeloma hybridomas.

FUSION PARTNERS

Murine myeloma cells, originally developed for the production of mouse mAbs, were not suitable as fusion partners for human B cells, as they produced unstable heterospecific hybrids, often resulting in rejection of the relevant human chromosomes. The development and refinement of suitable fusion partners has been an ongoing and instrumental part of the human hybridoma technology. Investigators have been developing myeloma cell lines suitable for fusion with human B cells since the early 1980s. In general, these lines can be divided into two major types: fully human and heterohybridomas constructed by fusing murine myeloma cells with human cells. Several of the myeloma cell lines suitable for fusion with human B cells are available at the American Type Culture Collection.

A number of fully human myeloma cell lines have been used successfully in fusion for the generation of human hybridomas. The most studied of these is the human myeloma U266 and its derived lines. U266 was originally isolated from a patient with multiple myeloma and found to secrete IgE (48). This myeloma was made HAT-sensitive and used as the first human fusion partner to generate fully human hybridomas (4, 49). The U266-derived SKO-007 line, made to be 8-azaguanine sensitive, was also used to generate some of the first fully human hybridomas (50–52). The human line LICR-LON-HMy2, also derived from malignant patient cells, was compared to SKO-007 and found to be superior (51, 53). LICR-LON-HMy2 was used to generate a hybridoma secreting mAb through the fusion of lymph node lymphocytes from a patient with breast cancer (54). The human myeloma line RPMI 8226, which expressed free light chain, was also used in the initial attempts to make fully human hybridomas (55, 56). The human myeloma-like cell line KR12 was generated by fusion of the human plasmacytoma line RPMI8226 with the human lymphoblastoid cell line KR-4 (14, 57). A more recently isolated human myeloma fusion partner, designated Karpas 707H, was developed for the purpose of making human hybridomas (58). Like U266, Karpas 707H was established from a patient with multiple myeloma. It was made to be both HAT-sensitive and ouabain-resistant, so it could be used with EBV-transformed B cells. Unfortunately, many laboratories attempting to make human hybridomas using these myeloma lines met with only limited success.

A larger group of fusion partners, themselves being heterohybridomas, have been developed by fusing murine myeloma cell lines with human cells. These fusion partners are nonsecreting mouse-human hybrids that are often made resistant to ouabain so they can produce stable hybridomas upon fusion with human EBV-transformed B cells. Since the heterohybridoma partner does not secrete an antibody, antibody secreted from the resulting hybridoma is encoded by the nucleus of the fused B cell. Ostberg and Pursch described the produc-tionof heterohybridomasmaderesistantto 8-azaguanine and used to generate human hybridomas secreting anti-influenza mAbs (59). One of these heteromyeloma fusion lines, SPAZ-4, was used later to generate human mAbs with specificities to various malignancies by fusion with lymph node lymphocytes from patients with carcinoma (60, 61). Two heterohybridoma cell lines, SHM-D33 and HMMA 2.5, continue to be used for production of human mAbs. SHM-D33 was generated by fusing the mouse myeloma cell line P3X63Ag8 with human myeloma cell line FU-266 (62). This fusion partner has been used very successfully to generate large panels of naturally occurring human anti-HIV mAbs (63–65). Another heterohybridoma fusion partner that has been very productive is HMMA 2.5 (66), which was generated using a multistep process. The parental line, HMMA2.11TG/O, was made by first fusing bone marrow mononuclear cells from a patient with IgA myeloma with the mouse myeloma cell line P3X63Ag8.653. HMMA 2.5 is a subclone that then was selected for optimal fusion efficiency after passaging in 6-thioguanosine. Cell line HMMA 2.5 was found to achieve the highest fusion efficiency when it was compared to six other myeloma cell lines using electrofusion (39). This heterohybridoma has been used to generate large panels of naturally occurring fully human mAbs to many viruses, including influenza and dengue viruses (29, 67–69).

Another heterohybridoma line used to successfully generate human hybridomas is K6H6/B5 (70–72). This line was developed by fusing malignant lymphoid cells isolated from a patient with nodular lymphoma with the mouse myeloma cell line NS-1-Ag4. K6H6/B5 was used to successfully create human hybridomas that secrete mAbs to hepatitis C virus (73). A very unique heterohybridoma fusion partner denoted SPYMEG was developed by fusion of Sp2 murine myeloma cells and MEG-01 human megakaryoblastic leukemia cells (74). Recently, SPYMEG was employed for the production of anti-influenza mAbs, generated by fusing PBMCs from influenza-vaccinated and naturally infected volunteers (75, 76). It was also used successfully to make uman hybridomas secreting anti-HIV neutralizing mAbs by fusion with PBMCs obtained from HIV-1-infected individuals (77). A more recently created line, MFP-2, is a trioma that was generated by fusing a murine myeloma cell line with a human myeloma cell line, yielding the intermediate heteromyeloma B6B11. This heterohybridoma was then fused with a human lymphocyte to generate the MFP-2 trioma line (78). The authors used this partner to develop mAbs with specificity to breast cancer tissue and cell lines (78, 79). The MFP-2 fusion partner was also used to produce several anti-West Nile virus mAbs (80).

Recently, a heterohybridoma fusion partner cell line was derived from the murine myeloma cell line Sp2 and modified to coexpress genes encoding murine interleukin-6 and human telomerase catalytic subunit (TERT) (81). The expression of murine IL-6 directly stimulates proliferation as well as immunoglobulin production from the resulting hybridoma. Human TERT can lengthen telomeres, thereby providing cells with unlimited replication capability and promoting karyotypic stability. The Sp2/mIL-6/hTERT heterohybridoma line then was demonstrated to produce stable hybridomas able to secrete fully human mAbs by fusing splenic B cells from a patient immunized with a Streptococcus pneumoniae vaccine. The authors succeeded in creating hybridomas that secrete human mAbs specific for S. pneumoniae antigens (81).

BIOLOGICAL CLONING METHODS

The final step in the generation of a human hybridoma is to isolate successful fusion products as clones derived from single cells, a process often referred to as biological cloning. There are several advantages to isolating individual hybridomas early and growing them in a clonal manner. Frequently, mixed populations of mAb-producing hybridomas are created following fusion. Depending on the experiment, there may be hundreds or even thousands of different clones produced. Cell cloning is an essential step in ensuring that a monoclonal antibody is ultimately generated. Also, hybridoma clones that produce the greatest quantity of mAbs are relatively rare following fusion and consume a greater amount of energy and nutrients. Because of this, high-producing clones often grow slower and can be overgrown easily by low- or nonproducing hybridoma clones (82). There are now several methods that can be used to perform biological cloning of human hybridomas. Traditionally, this was accomplished by limiting dilution plating. More recently, advances in flow cytometric automated single-cell sorting, with indexing capabilities, have allowed for fast, accurate, and versatile single-cell plating. Finally, semisolid medium preparations can be used to grow single hybridoma cells as isolated, suspended colonies. With special clone-picking devices, this process can be highly automated and can also be performed in an antigen-specific and semi-quantitative fashion for selection and biological cloning of high-producing human hybridomas (83) (Fig. 3).

FIGURE 3.

Cloning in (A) semi-solid medium and (B) final human hybridoma. doi:10.1128/microbiolspec.AID-0027-2014.f3

The most common method to generate biological clones of hybridomas is limiting dilution cloning. This technique, and variations thereof, is based on the Poisson distribution (84). The Poisson distribution is a discrete frequency distribution that describes the probability of a number of independent events occurring in a fixed interval of time. If a known number of viable cells are added to a plate with a given number of wells, this model can be used to provide the probability that any given well would contain a given number of cells, for example, if one were to plate 0.8 cells/well by counting the total number of cells to be added to a given number of wells, such as adding approximately 77 cells to a 96-well culture plate. By using the Poisson distribution, this dilution provides that approximately 36% of wells will contain 1 cell/well. By the same probability distribution, approximately 45% will contain 0 cells, 14% will contain 2 cells, 4% will contain 3 cells, and approximately 1% of wells will contain 4 cells. Thus, by using this method, one would expect that 19% of wells would contain >1 cell. Since hybridomas vary greatly in their ability to survive and make antibody, this calculation is often an overestimation of final growth and mAb expression. Limiting dilution plating is often performed using multiple plates andrepeatedforseveralroundsto improve the likelihood of isolating a true clone that is both stable and produces a high level of mAb. If greater cloning stringency is desired, hybridomas are plated at 0.3 cells/well, resulting in the percent of wells with >1 cell being only 3.3%, and this process can be repeated in a serial fashion two or three times. This method has the advantage of being simple and relatively inexpensive in terms of equipment. However, it is very time consuming and labor intensive, it has low throughput, and obtaining clonality is never guaranteed (85, 86). One cannot be certain that the starting population of hybridomas is in a homogenous single-cell suspension, as some clones may stick together. Additionally, since only a few hundred to a few thousand clones can be interrogated, the prospect of identifying a high-producing clone is low.

An alternative related approach used by many laboratories is to perform serial dilutions such that the cell concentration is diluted down and across the cell culture plate. By adding a known number of cells to the A1 position well, diluting down the A column, then diluting the A column across the plate with a multichannel pipette, one can achieve adequate separation of cells. This method has the advantage of near infinite variability, which may be important in isolating poor-growing or low-viability clones. The clear disadvantage is that only a handful of wells that are likely to contain isolated clones can be assessed in each dilution plate. Multiple rounds of dilution plating are often used as an attempt to ensure clonality.

Increasingly, laboratories have turned to sterile single-cell sorting of hybridomas using a flow cytometer outfitted with an automated single-cell deposition unit to accomplish the task of biological cloning. Methods that use flow cytometry and cell sorting greatly increase the number of cells that can be screened and can almost ensure that clonality is achieved. This technique of biological cloning is in many ways superior to limiting dilution methods but requires access to an expensive instrument, which may not be available. For those with access to cell sorting, the added expense of using the instrument is often offset by saving time and resources needed to perform multiple rounds of limiting dilution cloning. By staining for viability and using forward and side scatter to eliminate doublets and clumps, a single cell can be accurately and consistently placed into culture medium, even within a 384-well plate platform. This approach results in a considerable increase in throughput, and thus the quantity of hybridomas that can be effectively interrogated is dramatically amplified. Ideally, single-cell cloning could be coupled to selection of hybridomas based on antigen specificity and antibody quantity.

Methods for using flow cytometry to single-cell sort hybridomas on the basis of their antigen specificity have been developed and in some cases are commercially available. The use of flow cytometric sorting to select high-producing Chinese hamster ovary (CHO) cell clones was described initially (87) and then applied shortly thereafter to hybridomas (88). Cell sorting to identify and isolate clones also has been used for bispecific hybridomas and isotype-switch variants, and even for identifying variants with higher antibody avidity (89–91). One important limitation in the ability to use flow cytometric sorting for isolating antigen-specific hybridomas is the differential expression levels of native surface antibodies. The B cell receptor is not expressed at high levels on the surface of all hybridomas, making it difficult to selectively sort antigen-specific hybridomas after fusion. For this reason, several strategies have been developed to sort cells for antigen specificity or level of production by employing the secreted antibody.

One method uses an artificial affinity matrix to capture antibody, which has been secreted from the cell and is directly retained on the surface of the cell (92). The secreted molecules are bound to the secreting cell and can be subsequently labeled for flow cytometric analysis. This method is accomplished by first directly biotinylating the secreting cells, adding an avidin conjugated capture antibody, and then isolating cells within a protein-impermeable matrix to prevent cross-contamination of the secreted protein. The captured antibody then can be labeled fluorescently. This method has been used successfully in the isolation of hybridoma cells by flow cytometric sorting (93).

A related method, which is commercially available for use, is the gel microdroplet encapsulation technique. This strategy uses an agarose droplet to encapsulate the secreting cells. The agarose matrix itself is biotinylated so that an avidin bridge forms with biotinylated anti-immunoglobulin antibody, which can capture the antibody secreted from the cell (94, 95). This technique has been employed successfully to use flow cytometric sorting and sort hybridomas based on their secreted mAb (95–97). The agarose matrix is able to prevent secreted antibody from cross-contaminating neighboring encapsulated cells. Unlike the direct biotinylation method described previously, the droplet encapsulation method requires removal of the agarose matrix after the cell(s) is/are selected. Unfortunately, both methods are challenging and require considerable optimization.

Another method used for biological cloning and growth of hybridomas involves embedding in semisolid medium. When cells are plated, often following fusion, in semisolid medium containing methylcellulose and the HAT selection components, hybridomas grow in suspended isolated colonies (83, 98). Hybridoma colonies can be picked into growth medium with a high probability of being clonal. Building on this method, manufacturers have developed instruments (for example, ClonePix [Molecular Devices LLC]) that isolate, detect, and measure relative secretion of monoclonal antibody from hybridomas. With methylcellulose semisolid HAT medium and the ability to pick fluorescent colonies in a fully automated manner, high-producing clones can be obtained in many cases by means of a single culture step following fusion. A complete hybridoma generation procedure that incorporates methylcellulose embedding, fluorescent antigen labeling, and clone picking was described recently (99).