A human myeloma cell line suitable for the generation of human monoclonal antibodies

Development of Cell Lines.

Derivation of Human–Human Hybridomas.

The first fusion was between approximately 2 × 10⁶ of the PEG-resistant myeloma cells and 5 × 10⁶ 164 cells, with the standard PEG fusion protocol (9). The cell suspension was seeded in 96-well and 24-well plates and cultured in the presence of HAT and ouabain (0.3 μg/ml). Tissue culture fluid collected from wells with active cell growth was assayed in the AIDS cell test for anti-viral antibodies (10).

Thereafter we used unfractionated WBC from 20 ml of peripheral blood of adults and tonsil cells from two children. In each of the cell mixtures we used approximately 5 × 10⁶ myeloma cells and 10⁷ WBC. The cell suspensions obtained after the PEG fusion were seeded in two 96-well tissue culture plates and grown in RPMI medium 1640 supplemented with 20% FBS and HAT for the first 12 days. Thereafter, the growth medium was supplemented with hypoxanthine and thymidine for 1 week. Tissue culture fluid collected from wells with active cell growth was assayed by ELISA for the presence of IgG, with the use of an ELISA test with protein A on the plastic wells to capture the IgG monoclonal antibodies produced by the newly formed hybridomas. The tissue culture fluids of 20 hybridomas that produce Ig have been screened with commercial ELISA tests (DiaSorin, Saluggia, Italy) for antibodies against EBV, cytomegalovirus, herpes simplex virus, varicella zoster virus, and mumps and measles viruses.

Karyotype Analysis and Electron Microscopy.

Chromosome analysis by the G-banding method was as described by Czepulkowski et al. (11). For the ultrastructural examination the cells were fixed in 3% glutaraldehyde, and ultrathin sections were stained in uranyl acetate and lead citrate (12).

Human–Human Hybridoma Formation.

The first successful hybridoma was with the 164 cell. To monitor for antibody production we have been using the AIDS cell test (10).

Further studies of the IgG produced by the 707H/164 hybridoma, with the use of several polypeptides of the gp41, enabled us to establish that the monoclonal antibody reacted against the peptide LAVERYLKDQQLLGIWG, but it failed to react against four other peptides of the gp41 of HIV-1 (results not shown). The epitope that this monoclonal antibody recognizes seems not to be present in the African NDK strain of HIV-1 (13), inasmuch as the antibody failed to react with NDK-infected T cells but reacted not only with the Cambridge isolate of HIV-1 (14) but also with the early French LAV isolate (15). A database search (July 2000) confirms that the French (LAV) and British (CAM1) isolates of HIV-1 share the above amino acid sequence, whereas the African isolate has a lysine-to-arginine mutation.

After fusion of the myeloma with fresh lymphocytes, colony formation was observed during the 2–3 weeks after the removal of hypoxanthine and thymidine in approximately 28% of the wells when the myeloma cells were fused with tonsil cells and in about 18% the wells when the myeloma cells were fused with WBC (Table 1).

The growth pattern of the clones was variable. The parent myeloma cells grew in a uniform pattern of adherent cells, forming an irregular mosaic that covered the plastic surface, with continuous release of cells into the culture fluid. The growth of some hybrids resembled the parent myeloma cells, but many others became more refractile and acquired a distinct morphology and growth forms. Some hybridomas adhered to the plastic surfaces and displayed numerous thick but short spindles and rectangular and triangular shapes. Others grew mainly in suspension as single cells or clusters.

Tests for the secretion of Ig confirm that the parental cells produced a small quantity of light chain. As can be seen in Fig. 1, the fusions with the tonsil cells and blood lymphocytes produced clones that secreted IgG as well as IgM. We have now developed a large number of independent Ig-secreting hybrids by fusion with peripheral blood and tonsils lymphocytes. Most of the 40 hybridomas maintained continuously in culture have been secreting Ig during the past 5 months.

For the quantitation of IgG production the nephelometric assay kit (Dade Behring, Marburg, Germany) was used. A comparative assay for the secretion of antibodies to the gp41 of HIV-1 by the 164 cells and the hybrid (707H/164) revealed that the hybridoma produced far more antibody than the lymphoblasts after culture of an identical number of cells (10⁶ per ml) for 4 days under the same conditions. The hybrid 707H/164 secreted 470 μg/ml of IgG, whereas the 164 cells secreted less than the assay detection level, 70 μg/ml. Thus the derived hybridomas are far more effective in antibody production. Hybridomas derived with the 164 cells and with WBC and kept over 10 days in culture without medium change revealed a gradual increase in IgG in the tissue culture fluid that continued despite the fact that the cells stopped dividing after 4 days and thereafter gradually died off (Fig. 2). This increase in IgG is similar to that in murine hybridomas that secreted IgG even after the cultures became stationary (9). The yield of IgG produced by the hybridoma secreting antibodies against the gp41 of HIV-1 and the hybridomas formed after fusion with blood lymphocytes indicate that the hybridomas that are formed secrete at least as much or even more antibody than the murine hybridomas. It correlates with the extensive rough endoplasmic reticulum (RER) in the human hybridomas (see below) as compared with the less developed RER of the murine hybrids.

By now, we have maintained in culture over the past 10 months the hybridoma cells from the well that had only one colony that had been giving a strong positive signal for the anti-HIV-1 antibodies. To test for the stability of antibody production we measured the titer of the anti-HIV-1 antibodies after 1, 5, and 9 months in culture. Using the AIDS cell test, we were able to show that the titer remained stable at 1:40. Furthermore, we also cloned the hybrids after 4 months in culture, plating 100 cells in a 96-well plate. In 27 wells colonies grew out and when tested 2 months later all produced the anti-HIV antibodies.

Tissue culture fluids from 5 of the 13 hybridomas formed with tonsil cells and 3 of the 7 formed with WBC contained antibodies against the viral capsid antigen of EBV (Table 2). Two of these hybridomas that were kept in culture for 3 months kept producing the anti-EBV antibodies.

Karoytype Analysis.

Initially the cultured cells, like the fresh patient cells, were found to be hypodiploid, with 45 chromosomes containing several chromosomal abnormalities (7). However, Karpas 707H is near-tetraploid, containing numerous and an unusually high number of gross chromosomal abnormalities (Fig. 3A). Karyotype analysis of the hybridoma 707H/164 revealed that the cells are near-hexaploid, containing the normal chromosomes from the lymphoblasts, which in the case of chromosomes no. 1 are in sharp contrast to the five abnormal chromosomes no. 1 of the myeloma (Fig. 3B).

It is notable that in the 707H parental myeloma many of the chromosomes were abnormal (Fig. 4). To establish whether the repeated cycles of PEG treatment induced these karyotype abnormalities, the chromosomes of cells that were frozen before the PEG treatment were also analyzed. The analysis revealed that all of these changes were present before the PEG treatment. Hence it appears that during the continuous in vitro culture over many months and treatment with 8-azaguanine, 6-thioguanine, HAT, and ouabain, a near-tetraploid cell line emerged that had a growth advantage over the hypodiploid myeloma cells. It is interesting to note that the chromosomes 2, 14, and 22 that contain the Ig genes are normal and diploid (Fig. 3A).

Electron Microscopy.

Ultrastructural examination of the cultured myeloma revealed that in contrast to the cells that contained large amounts RER during the first year in culture (7), after the long-term in vitro growth hardly any RER could be seen (Fig. 5A). Not surprisingly, the cytoplasm of the 164 cells contained only a few strands of RER (Fig. 5B). In contrast, examination of two hybridomas, 707H/164 (Fig. 5C) and 707H/105 (not shown), revealed that in both cases the hybrids resembled plasma cells. A large part of the cytoplasm was taken up by parallel and concentric RER. A large number of mitochondria could be seen in most sections, whereas sections through the Golgi area revealed a highly developed apparatus. Inasmuch as the ultrastructural examination of the EBV-immortalized 164 B cells revealed only a small number of short and thin strands of RER, it would not be unreasonable to suggest that the fused lymphoblast somehow activates the latent 707H myeloma RER for the increased production of Ig.