EX VIVO ANTIBODY PRODUCTION

Abstract

The present invention provides means and methods for producing improved ex vivo B cell cultures with a short doubling time.

Claims

1. A method for obtaining antibodies comprising: inducing, enhancing and/or maintaining expression of Bcl-6, or a rabbit homologue thereof, in a rabbit B-cell; inducing, enhancing and/or maintaining expression of at least one anti-apoptotic nucleic acid molecule in said B-cell, culturing said B cell ex vivo; and harvesting antibodies produced by said B cell within 7-14 days.

2-24. (canceled)

25. The method of claim 1, wherein said rabbit B cell is provided with: a nucleic acid molecule encoding Bcl-6 or a functional part or a functional derivative thereof, and/or at least one anti-apoptotic nucleic acid molecule.

26. The method of claim 1, wherein said rabbit B cell is provided with: a nucleic acid molecule encoding a non-rabbit Bcl-6 or a functional part or a functional derivative thereof, and/or at least one non-rabbit anti-apoptotic nucleic acid molecule.

27. The method of claim 1, wherein said rabbit B cell is provided with: a nucleic acid molecule encoding a rabbit Bcl-6 or a functional part or a functional derivative thereof, and/or at least one rabbit anti-apoptotic nucleic acid molecule.

28. The method of claim 1, wherein said rabbit B cell is provided with: a nucleic acid molecule encoding a non-rabbit Bcl-6 or a functional part or a functional derivative thereof, and at least one non-rabbit anti-apoptotic nucleic acid molecule.

29. The method of claim 1, wherein said rabbit B cell is provided with: a nucleic acid molecule encoding a human or murine Bcl-6 or a functional part or a functional derivative thereof, and/or at least one human or murine anti-apoptotic nucleic acid molecule.

30. The method of claim 1, wherein said rabbit B cell is provided with: a nucleic acid molecule encoding a human Bcl-6 or a functional part or a functional derivative thereof, and at least one human anti-apoptotic nucleic acid molecule.

31. The method of claim 3, wherein said non-rabbit nucleic acid molecules are human nucleic acid molecules.

32. The method of claim 1, wherein said rabbit B cell is provided with: a nucleic acid molecule encoding a non-rabbit Bcl-6 or a functional part or a functional derivative thereof, and at least one rabbit anti-apoptotic nucleic acid molecule.

33. The method of claim 1, wherein said rabbit B cell is provided with: a nucleic acid molecule encoding a rabbit Bcl-6 or a functional part or a functional derivative thereof, and at least one non-rabbit anti-apoptotic nucleic acid molecule.

34. The method of claim 1, wherein said at least one anti-apoptotic nucleic acid molecule comprises a gene of the Bcl2 family.

35. The method of claim 1, wherein said at least one anti-apoptotic nucleic acid molecule comprises a gene of the Bcl2 family selected from the group consisting of Bcl-xL, Mcl 1, Bcl-2, A1, Bcl-w, Bcl2L10, and rabbit homologues thereof and functional parts thereof and functional derivatives thereof.

36. The method of claim 1 further comprising: inducing, enhancing and/or maintaining expression of Blimp 1, or a rabbit homologue thereof, in said rabbit B cell.

37. The method of claim 1, further comprising providing said rabbit B cell with 1L21 and CD40L.

38. The method of claim 14, wherein said 1L21 is mouse or human 1L21 and/or wherein said CD40L is mouse or human CD40L.

39. The method of claim 1, comprising: providing said rabbit B cell with a compound capable of directly or indirectly enhancing expression of Bcl-6, or expression of a rabbit homologue thereof; and/or culturing said rabbit B cell in the presence of a compound capable of directly or indirectly enhancing expression of Bcl-6, or expression of a rabbit homologue thereof.

40. The method of claim 1, comprising: providing said rabbit B cell with at least one compound capable of directly or indirectly enhancing expression of Bcl-xL and/or Mcl-1 and/or Bcl-2 and/or A1 and/or Bcl w and/or Bcl2L10 and/or or a rabbit homologue thereof; and/or culturing said rabbit B cell in the presence of at least one compound capable of directly or indirectly enhancing expression of Bcl-xL and/or Mcl-1 and/or Bcl-2 and/or A1 and/or Bcl w and/or Bcl2L10 and/or or a rabbit homologue thereof.

41. The method of claim 1, further comprising: providing said rabbit B cell with at least one compound capable of directly or indirectly increasing expression of Blimp 1, or expression of a rabbit homologue of Blimp 1; and/or culturing said rabbit B cell in the presence of at least one compound capable of directly or indirectly increasing expression of Blimp 1, or expression of a rabbit homologue of Blimp 1.

42. The method of claim 1, wherein antibodies produced by said B cell are harvested within 9-12 days.

43. The method of claim 1, wherein antibodies produced by said B cell are harvested within 9-10 days.

44. The method of claim 1, wherein antibodies produced by said B cell are harvested in an amount of about 30-100 ng/ml.

Description

EXAMPLES

Example 1

Transduction of B Cells

[0107] Gene transfer into lymphocytes by traditional methods like calcium phosphate precipitation, liposome formation or electroporation is inefficient but more importantly stable gene integration is generally absent. Viral transduction however leads directly to stable gene integration into the genome of the target cell and can be very efficient if the proper virus envelope is chosen. Both retroviral and lentiviral transductions are suitable for efficient gene transfer. While retroviral integration is dependent on cell division, lentiviral transduction can also be applied to non-dividing cells like plasma B cells. Large-scale preparation of recombinant retrovirus can easily be achieved by using stable producer cell lines such as the Phoenix expression platform (Kinsella and Nolan, 1996). Production of high titer lentivirus tends to be more cumbersome mainly because of the toxicity of the expressed virus proteins and envelopes.

[0108] For the current Examples, we used a Moloney Murine Leukemia Virus (MMLV) based platform, using either amphotropic or Gibbon Ape Leukemia Virus (GALV) envelope expressing producer cells (Wilson et al., 1995). In our GALV-based vector, the transmembrane domain of the GALV strain SEATO envelope protein was fused to the cytoplasmic domain of an ampho envelope protein (FIG. 9).

[0109] The transfer vector is set-up such that Bcl-6, Bcl-xL and the green fluorescent protein (GFP) marker protein are simultaneously translated from the same viral RNA (FIG. 8). This multicistronic approach is achieved by placing a self-cleaving 2A peptide sequence (Szymczak et al., 2004) between the BCL-6 and BCL-xL coding regions and an Internal Ribosomal Entry Sequence (IRES) upstream of the GFP reporter gene. Viral transduction efficiencies are high and unbiased.

Generation of Immortalized Rabbit B Cells

[0110] Human memory B cells were immortalized using the BCL-6/Bcl-xL technology described by Kwakkenbos et al., 2010 and patent application WO 2007/067046. In brief, PBMC's from rabbit blood were isolated using a ficoll density gradient and stained for Ig expression using an antibody that recognizes Ig (IgG H+L: IgG heavy chain and kappa and lambda light chains) sometimes in combination with an IgM specific antibody. B cells were isolated (Ig positive, or Ig positive+IgM negative) using a FACS sorter and stimulated on irradiated (50 Gy) mouse L cell fibroblasts stably expressing CD40L (CD40L-L cells, 10.sup.5 cells ml.sup.1) together with recombinant mouse interleukin (IL)-21 for 36-48 hours. Cells were harvested and washed with medium without FCS and cells were then transferred to Retronectin (Takara, Shiga, Japan)-coated tissue culture plates where they were transduced with a retroviral vector containing BCL-6, Bcl-xL, and GFP as a reporter protein. Alternatively cells were transduced with a retroviral vector containing BCL-6, Mcl-1 and GFP. Transduced B cells were maintained in culture with CD40 Ligand expressing L-cells and IL-21. In FIG. 1 the transduction efficiency is compared for GALV and amphotropic type retroviruses at 4 days after transduction. Four days after transduction with the amphotropic type retrovirus 0.8% of the cells was transduced compared to 80% of cells after transduction with a GALV type retrovirus. Clearly the GALV type retrovirus is superior to the amphotropic type retrovirus for transducing rabbit B cells.

Example 2

Cell Culture.

[0111] We maintained B cells (210.sup.5 cells ml.sup.1) in Iscove's modified Dulbecco's medium (Gibco) containing 8% FBS and penicillin-streptomycin (Roche) supplemented with recombinant mouse interleukin 21 (IL-21) (50 ng ml.sup.1) and cultured them on irradiated (50 Gy) mouse L cell fibroblasts stably expressing CD40L (CD40L-L cells, 10 cells ml.sup.1). To determine cell doubling time cells were cultured in 24-well plates at 50-100.000 cells/well together with CD40L-L cells and IL-21. Every 3-4 days cell were counted and 50-100.000 cells transferred to a new well. In FIG. 2 growth curves are depicted for B cells from two human donors (89 and 93), one llama B-cell sample (Llama) and one rabbit B-cell sample which was transduced with a GALV type retrovirus carrying a nucleic acid molecule containing a human Bcl-6 sequence and a human Bcl-xL (Rb 6XL). Also a growth curve is depicted for one rabbit sample that was transduced with a GALV type retrovirus carrying a nucleic acid molecule containing a human Bcl-6 sequence and a human Mcl-1 (Rb 6M). The transduced rabbit B cells have an average doubling time of 19 hours and thus grow faster than the human or llama B cells that have doubling times between 26 and 32 hours. These average doubling times were originally calculated by determining the increase of B cells during several 3-4 days time intervals, and averaging the obtained results. Subsequently, the overall average doubling time during the whole culturing period was calculated. This resulted in an average doubling time of the transduced rabbit B cells of 18 hours, an average doubling time of the transduced human B cells of 25-29 hours and an average doubling time of the transduced llama B cells of 27 hours. This confirms our observations that our methods yield rabbit B-cell cultures with a mean doubling time of 20 hours or less, whereas human, murine and llama B cells typically have a doubling time of between 25 and 36 hours.

Example 3

B-Cell Receptor Expression and Antigen-Specific Staining

[0112] Immortalized human B cells express the B-cell receptor. This quality enables antigen-specific staining and sorting of B cells. To determine whether the B-cell receptor is also expressed on transduced rabbit B cells, B-cell clones are stained with fluorescently labeled antibodies reacting specifically with either rabbit IgG, rabbit IgM or rabbit IgA. B cells were washed in cold (4 C.) cell culture medium and incubated on ice in the dark with cell culture medium containing immunofluorescently labelled antibodies that are specific for either rabbit IgG, IgM, IgA or labelled antigen. Afterwards excess of labelled antibodies or antigen was washed away and B-cell receptor expression analysed on a FACS analyser; the Guava easycyte (Millipore) or FACS Aria3 (BD).

[0113] In FIG. 3 three different B-cell clones of different isotypes were stained with fluorescently labelled antibodies specifically recognizing rabbit antibody isotype IgG, IgA or IgM. Clearly the B-cell receptor can be efficiently stained for the different rabbit antibody isotypes. We therefore conclude that immortalized rabbit B cells also express the B-cell receptor.

[0114] In addition, also fluorescently labeled influenza proteins were used to stain for influenza-specific B-cells from rabbits that had been immunized with a human influenza vaccine or untreated control rabbits (FIG. 4). Rabbit B cells were stained with fluorescently labeled H1, H3 or influenza B and sorted 1 cell per well using a FACS sorter.

Example 4

Development of Single-Cell Derived, Clonal Rabbit B Cell Cultures.

[0115] Transduced B cells were sorted one cell per well using a FACS sorter and cultured in the presence of irradiated (50 Gy) mouse L cell fibroblasts stably expressing CD40L (CD40L-L cells, 105 cells ml.sup.1) together with recombinant mouse IL-21. Every 3-4 days fresh CD40L-L cells and IL-21 were added. Starting 9 days after seeding the cells (one cell per well), the supernatants were analyzed in ELISA for the production of rabbit immunoglobulin G (IgG). For comparison also the human IgG in the supernatant of human B-cell clones was analyzed in parallel.

[0116] In FIG. 7 the antibody concentration in the supernatant is depicted over time starting at 9 days after the initiation of the single cell cultures. The antibody concentration was determined for two human donors and one rabbit B-cell sample that were transduced with a GALV type retrovirus carrying a nucleic acid molecule containing a human Bcl-6 sequence and a human Bcl-xL and for one rabbit B-cell sample that was transduced with a GALV type retrovirus carrying a nucleic acid molecule containing a human Bcl-6 sequence and a human Mcl-1. B cell clones from rabbits produce IgG concentrations of 30 ng/ml and 100 ng/ml within a shorter time period (9-10 days and 11-12 days, respectively) than do the human B-cell clones (13-18 and 15-20 days, respectively). This provides the important advantage that it allows for earlier screening for antibodies of interest of rabbit B cell clones, compared to human B cell clones.

Example 5

Immunization of Rabbits.

[0117] 2 New Zealand White rabbits were immunized with a human influenza vaccine containing 15 ug H1N1, 15 ug H3N2 and 15 ug infl B in complete Freunds adjuvans. After 3 weeks rabbits were boosted with the same vaccine in incomplete Freunds adjuvans. Five days after the boost rabbits were bled, B-cells were isolated from the blood and transduced with a GALV type retrovirus (containing the extracellular domain and transmembrane domain of the GALV strain SEATO envelope protein, fused to the cytoplasmic domain of an ampho envelope protein) carrying a nucleic acid molecule containing a human Bcl-6 sequence and a human Bcl-xL. Transduced B cells were seeded at different cell densities into culture plates and cultured as described in Example 4. Also, transduced B cells were labeled with fluorescently labeled components of the vaccine; H1, H3 or influenza B and sorted 1 cell per well using a FACS sorter and cultured as described in Example 4. The supernatants of the cultured cells were analyzed for binding to the complete vaccine or to its individual components. The results are depicted in FIGS. 4-6 and show that antigen-specific B cells can be identified in the B-cell pool from vaccinated rabbits by seeding cells at different density (FIG. 5) and also very efficiently by sorting cells using the labeled antigens (FIG. 4 and FIG. 6).

Example 6

Rabbit B Cells are Immortalized by the Introduction of the Genes Bcl-6 and Bcl-xl Using an Amphotropic Type Retrovirus.

[0118] Immortalization of rabbit B cells by introduction of the genes Bcl-6 and Bcl-xl can be achieved by using different types of vectors, such as for instance GALV and amphotrophic type retroviruses as is shown in Example 1. The growth of B cells transduced with the amphotrophic type retrovirus was further pursued to confirm that introduction of Bcl-6 and Bcl-xl by amphotrophic retrovirus also leads to immortalization of rabbit B cells. Four days after transduction with the amphotropic type retrovirus 0.8% of the cells was transduced compared to 80% of cells after transduction with a GALV type retrovirus (FIG. 1 and FIG. 10). Ten days after transduction 94% of the cell population transduced with the amphotrophic retrovirus was GFP positive demonstrating that the transduced cells overgrow the non-transduced cells (FIG. 10).

[0119] To determine cell doubling time cells were cultured as done in Example 2 in 24-well plates at 50-100.000 cells/well together with CD40L-L cells and IL-21. Every 3-4 days cell were counted and 50-100.000 cells transferred to a new well. In FIG. 11 the growth curve is depicted for rabbit B cells transduced with amphotrophic virus. The calculated doubling time is 19 hours, which is comparable to rabbit B cells transduced with GALV type retrovirus (18 hours). In conclusion, introduction of Bcl-6 and Bcl-xl into rabbit B cells by amphotrophic retrovirus also results in immortalization of rabbit B cells, although the transduction efficiency is much lower as compared to a GALV based vector.

BRIEF DESCRIPTION OF THE DRAWINGS

[0120] FIG. 1.

[0121] Transduction of rabbit memory B cells. Rabbit B cells were isolated from PBMCs based on Ig expression. Cells were activated for 36-40 hrs on CD40L L-cells with rm-IL-21. Cells were transduced with a retroviral vector containing BCL6 and Bcl-xL. Both GALV and amphotropic type retroviruses were tested. Transduced cells are then cultured on CD40L-L cells in the presence of recombinant mouse IL-21. After four days of culture the transduction efficiency was determined based on GFP expression. GALV typed retrovirus showed superior (80%) transduction efficiency compared to amphotropic (0.8%) typed retrovirus.

[0122] FIG. 2.

[0123] Growth curves were analyzed for rabbit B cells transduced with a retroviral vector containing BCL6 and Bcl-xL or a retroviral vector containing BCL6 and Mcl-1. For comparison growth curves were analysed in parallel B cells from llama cells and human cells from two different donors that were transduced with an identical retroviral vector containing BCL6 and Bcl-xL.

[0124] FIG. 3.

[0125] IgG, IgM and IgA surface immunoglobulin expression was detected using FACS on three different Bcl-6 Bcl-xL transduced rabbit B-cell clones.

[0126] FIG. 4.

[0127] Identification of antigen-specific rabbit B-cells within a pool of rabbit B cells with different specificities.

[0128] FIG. 5.

[0129] Antigen-specific rabbit antibodies were obtained against the different components of a human influenza vaccine containing 15 ug H1N1, 15 ug H3N2 and 15 ug infl B. Rabbits were immunized and boosted with the human influenza vaccine. B cells were immortalized and seeded at different densities in 384-well plates on CD40L-L cells in the presence of recombinant mouse IL-21. Antibodies present in the rabbit B cell culture supernatants were screened in ELISA for influenza-specificity.

[0130] Antigen-specific antibodies were observed for all the components of the vaccine.

[0131] FIG. 6.

[0132] Immortalized B cells from rabbits immunized with a human influenza vaccine containing 15 ug H1N1, 15 ug H3N2 and 15 ug infl B were stained with fluorescently labelled influenza proteins. B cells showing binding to the influenza proteins were sorted 1 cell per well in 384-well plates on CD40L-L cells in the presence of recombinant mouse IL-21 using a FACSAria sorter. Supernatants were screened in ELISA for influenza-specific antibodies. Antigen-specific antibodies were observed with a high frequency for the components of the vaccine that were used for antigen-specific sorting.

[0133] FIG. 7.

[0134] Antibody concentration in the supernatant of clonal B cells at different time points. Human, llama and rabbit transduced B cells were seeded 1 cell per well in the presence of irradiated CD40L-L cells and supplemented with mouse IL-21. Every 3-4 days CD40L-L cells and IL-21 were replenished. The IgG concentration was analyzed in ELISA for individual wells at different time points during culture. Each measurement was done on different wells. The rabbit B cells were either transduced with a retroviral vector containing BCL6 and Bcl-xL or a retroviral vector containing BCL6 and Mcl-1. All other cells (human and llama) were transduced with BCL6 and Bcl-xL.

[0135] FIG. 8.

[0136] Schematic representation of the vector used to transduce the rabbit and human B cells

[0137] FIG. 9.

[0138] Sequence of the extracellular domain of GALV SEATO envelope protein (bold) and the transmembrane domain of the GALV SEATO envelope protein (underlined), fused to the cytoplasmic domain of ampho envelope protein (italics+dotted-underlined).

[0139] FIG. 10.

[0140] Transduction of rabbit memory B cells and outgrowth of rabbit B cells transduced with amphotrophic type retrovirus. Rabbit B cells were isolated from PBMCs based on Ig expression. Cells were activated for 36-40 hrs on CD40L L-cells with rm-IL-21. Cells were transduced with a retroviral vector containing BCL6 and Bcl-xL. Both GALV and amphotropic type retroviruses were tested. Transduced cells were then cultured on CD40L-L cells in the presence of recombinant mouse IL-21. After four days of culture the transduction efficiency was determined based on GFP expression. GALV typed retrovirus showed superior (80%) transduction efficiency compared to amphotropic (0.8%) typed retrovirus. After 10 days 94% of rabbit B cells transduced with amphotrophic type retrovirus were immortalized based on GFP expression showing outgrowth of transduced cells over non-transduced cells.

[0141] FIG. 11.

[0142] A growth curve was analyzed for rabbit B cells transduced with a amphotrophic type retroviral vector containing BCL6 and Bcl-xL

REFERENCES

[0143] Christopherson, K. S. et al. PNAS 89, 6314-8 (1992) [0144] Guzman, L. M. et al. Bacteriol 177, 4121-4130 (1995) [0145] T. M. Kinsella, G. P. Nolan, Hum Gene Ther 7 (1996) 1405. [0146] Kwakkenbos et al. Generation of stable monoclonal antibody-producing B cell receptor-positive human memory B cells by genetic programming. Nature Medicine (2010) vol. 16 (1) pp. 123-8 [0147] Lam et al. Improved gene transfer into human lymphocytes using retroviruses with the gibbon ape leukemia virus envelope. Human gene therapy 7 (1996) 1415-1422 [0148] A. L. Szymczak, C. J. Workman, Y. Wang, K. M. Vignali, S. Dilioglou, E. F. Vanin, D. A. A. Vignali, Nat Biotechnol 22 (2004) 589. [0149] C. A. Wilson, M. V. Eiden, W. B. Anderson, C. Lehel, Z. Olah, J Virol 69 (1995) 534. [0150] WO 2007/067046