Means and methods for influencing the stability of antibody producing cells

Abstract

The invention provides a method for influencing the stability of an antibody producing cell, comprising directly or indirectly influencing the amount of BCL6 and/or Blimp 1 expression product within said antibody producing cell. Stable antibody producing cells and cell lines are also provided, as well as methods for producing antibodies using such cells and/or cell lines.

Claims

1. An antibody producing B cell that is capable of replicating for at least nine weeks, comprising: an exogenous nucleic acid sequence encoding a peptide comprising Bcl6, or a functional part thereof, capable of increasing the replicative life span of an antibody producing B cell, and an exogenous anti-apoptotic nucleic acid sequence; wherein the exogenous anti-apoptotic nucleic acid sequence encodes a member of the anti-apoptotic Bcl 2 family.

2. The antibody producing B cell according to claim 1, wherein the exogenous anti-apoptotic nucleic acid sequence encodes a peptide comprising Bcl-xL, or a functional part thereof capable of increasing the replicative life span of an antibody producing B cell.

3. The antibody producing B cell according to claim 1, wherein expression of the nucleic acid sequence encoding a peptide comprising Bc16, or a functional part thereof, is regulated by at least one of an activator and/or a repressor that is inducible by an exogenous compound.

4. A B cell culture, comprising antibody producing B cells according to claim 1.

5. The B cell culture according to claim 4, wherein the culture further comprises at least one of IL 21 and/or Il 10.

6. The antibody producing B cell according to claim 2, wherein expression of the nucleic acid sequence encoding a peptide comprising Bcl6, Bcl-xL or a functional part of Bc16 or Bcl-xL, is regulated by at least one of an activator and/or a repressor that is inducible by an exogenous compound.

Description

EXAMPLES

Example 1

(1) Methods

(2) Human memory B cells are purified from peripheral blood or tonsil by first by positive selection for B cells with CD19 MACS beads (Miltenyi Biotech). Memory B cells are then selected by surface staining and cell sorting for IgG. IgG.sup.+ B cells are then cultured with mouse fibroblast L cells expressing CD40L in the presence of mouse or human IL-21 for 36 to 48 hours. Cells are then transferred to Retronectin (Takara, Shiga, Japan)-coated tissue culture plates where they are transduced with a retrovirus encoding human BCL6-IRES-GFP for 16 h at 37.degree. C. Transduced cells are then cultured on CD40L-L cells in the presence of human IL-2 and human IL-4. After approximately 3-4 weeks the GFP.sup.+ cells (that is, BCL6.sup.+ cells) reach 100% of the culture after which BCL6.sup.+ cells are cultured with IL-2 and IL-4 or with human or mouse IL-21. Using flow cytometry we monitor the expression of GFP, CD19, CD38, CD20, MHC class II, CD27 (BD Biosciences), and other markers using labeled antibodies. We monitor growth by cell counting, and Ig production is monitored by enzyme ELISA detection of Ig in the culture supernatant (Dako, Glostrup, Denmark). Gene expression is monitored by reverse transcriptase polymerase chain reaction (RT-PCR, Invitrogen, Breda, Netherlands).

(3) Results

(4) Introduction of BCL6 into memory B cells results in a greatly extended lifespan over normal B cells in culture (months vs. .about.3 weeks). These cells maintain CD19, surface Ig, MHC class II, and express intermediate levels of CD38 and CD20, suggesting a memory cell phenotype (not shown). Culture of these cells on CD40L-L cells in the presence of IL-21 results in a significant growth advantage (FIG. 1) and acquisition of a plasmablast-like cell surface phenotype (CD38.sup.hiCD20.sup.+, FIG. 2). Importantly, IL-21 cultured cells secrete 300% more IgG compared with cells cultured IL-2 and IL-4. Together these data show that IL-21 culture promotes plasmablast development in an immortalized B cell population, exhibiting enhanced growth and antibody production.

Example 2

(5) A non-limiting model of one embodiment of the present invention is depicted in FIG. 4.

(6) In the human body, differentiation of plasma cells from memory B cells involves downregulation of BCL6 and upregulation of Blimp-1. In memory cells BCL6 is high and Blimp-1 expression is low. Signals that trigger differentiation cause an upregulation of Blimp-1, and this Blimp-1 counteracts the expression of BCL6. This stage is short-lived and is called the plasmablast. With progressively increasing Blimp-1 levels, BCL6 expression is extinguished, resulting in a plasma cell.

(7) In one embodiment of the invention BCL6 expression is locked, for instance because of stable expression mediated by a retroviral expression cassette integrated into the DNA of the B cells. Then, with BCL6 levels maintained, we switch on Blimp-1 expression, for instance by use of a cytokine that activates STATS, such as IL-21 (FIG. 3). This combination, through modulation of key transcription, results in stable growth of cells that secrete antibody and have phenotype characteristics of a plasmablast.

(8) TABLE-US-00001 TABLE 1 Cell surface markers of memory B cells, plasmablasts and plasma cells Memory Plasmablast Plasma Cell CD38 + ++ ++ CD20 + + CD27 + + CD19 ++ + CD138 + proliferation low high none Ig secretion low intermediate high

(9) TABLE-US-00002 TABLE 2 Memory Plasmablast Plasma Cell BCL6 ++ + Blimp-1 + ++

Example 3

(10) Materials and Methods

(11) Maintenance and Isolation of Human B Cells

(12) Using standard procedures, CD19 positive human B cells were isolated from bloodbank derived buffy coat (other sources can be fresh heparin. or ACD blood, or a lymphoid organ for example tonsil or spleen). In brief, total peripheral blood mononuclear cells (PBMC) were isolated using ficoll density separation (Amersham, Buckinghamshire, UK). CD19 labeled beads were used to positively selected B cells by MACS cell sorting technique (Miltenyi, Auburn, Calif., USA). Cells were subsequently stained with appropriate combinations of monoclonal antibodies (mAbs) to CD19, CD27, IgG, IgM, CD3 (Becton Dickinson (BD), Franklin Lakes, N.J., USA) and phycoerythrin (PE) labeled Tetanus Toxoid (provided by A. Radbruch, Berlin, Germany) or any other labeled antigen. Cells were then sorted using the FACSAria (BD). Sorted cells were washed and cultured (1.5 to 2.times.10.sup.5 cells/ml) on irradiated CD40L-expressing L-cells (5.times.10.sup.4 cells/ml; provided by DR. J. Banchereau, Schering Plough France, Dardilly France), in Iscove's Modified D Minimal Essential Medium (IMDM) containing 8% fetal calf serum (FCS) and Penicillin/Streptomycin. Unless mentioned otherwise, these CD40L-expressing L-cells are always present in the cultures.

(13) Transduction and Regulation of Mouse Constitutive Active STAT5b in B Cells

(14) Purified B cells were primed to become transduced with the caSTAT5b gene. Two priming protocols were used: (1) purified B cells were cultured for 3 days with interleukin (IL) 2 (20 U/ml, Chiron, Emeryville, Calif., USA) followed by a 24 hour culture with IL-2 and IL-4 (10 ng/ml, R&D, Minneapolis, Minn., USA) or (2) purified B cells were cultured for 36 hours with recombinant mouse IL-21 (50 ng/ml, R&D). Subsequently, cells were plated on recombinant human fibronectin fragments CH-296 (Hanenberg H., Nat., Med. 1996; RetroNectin, Takara, Japan) and human serum albumin treated plates (Corning Life Sciences, Corning, N.Y., USA) in the absence of L-cells, with the cytokines IL-2/4 or IL-21. At last, cells were transduced with the caSTAT5b gene (described by Ariyoshi K., JBC, 2000 and obtained from T. Kitamura, IMSUT, Tokyo, Japan) fused to the estrogen receptor (ER, provided by H. Kurata, DNAX Institute, Palo Alto, Calif., USA). The activity of the caSTAT5b-ER fusion product can be controlled by the hormone hydroxytamoxifen (4HT, Sigma-Aldrich, St. Louis, Mo., USA). The transduction was performed using a retrovirus as described previously (Heemskerk M. H., JEM, 1997; Heemskerk M. H., Cell Immunol. 1999; Scheeren F. A., Nat Immunol, 2005). Transduction efficiency was determined by antibody staining of a truncated, signaling incompetent mutant of Nerve Growth Factor Receptor (.DELTA.NGFR, provided by C. Bonini, St. Raphael Hospital, Milan, Italy). Thus, outgrowth of B cells that contain the caSTAT5b gene depends on the presence of 4HT and these cells can be detected by antibody staining for NGFR (Chromaprobe, Maryland Heights, Mo., USA).

(15) Development of 100% caSTAT5b Positive B Cell Lines that Secrete Antibodies

(16) We have developed a B cell line that produces monoclonal antibodies and is 100% caSTAT5b (=NGFR) positive. This was achieved by differentiating B cells from a memory into an antibody producing phenotype, and transducing with the caSTA5b-ER-IRES-NGFR construct. The action of caSTAT5b makes the differentiated B cells insensitive to cell death. Differentiation of B cells is induced in the first 2 to 3 weeks after isolation (FIG. 5), using a cytokine mixture (IL-2, 4, 21 or combinations of these cytokines and CD40L). The time point that caSTAT5b is activated by adding 4HT affects the overall phenotype of the cultures. This because caSTAT5b blocks the cell to change its phenotype e.g. blocks further differentiation. Thus, the longer 4HT is withheld the more B cells will differentiate into antibody producing cells or into a type of cell that preferentially grows out under these culture conditions (suggestions for cell types are: naive, follicular, memory, antibody producing, plasma blast, plasma cell, marginal zone, perisinusoidal or transitional B cellsmany of those B cell subsets have only been determined in mice). When 4HT is present in the culture medium, caSTA5b-ER-IRES-NGFR positive B cells can survive for long periods (Table 2).

(17) TABLE-US-00003 TABLE 2 Overview of caSTAT5b-ER-IRES-NGFR transduced human B cell cultures Transduction Isolation Time on Culture Donor Date Subtype Protocol IL-21 Time B12 28 Apr. 2005 CD19+TT+ IL-2 IL-4 4 wks 18 Oct. 2005 B15 17 May 2005 CD19+TT+ IL-21 36 h 5 Dec. 2005 B16 31 May 2005 CD19+TT+ IL-21 20 d 5 Dec. 2005 B18 22 Jun. 2005 CD19+CD27+ IL-2 IL-4 time series 5 Dec. 2005 and TT+ and IL-21 (36 h to 20 d) B19 22 Jun. 2005 CD19+CD27+ IL-2 IL-4 time series 5 Dec. 2005 and TT+ and IL-21 (36 h to 20 d) B20 6 Jul. 2005 CD19+TT+ IL-21 36 h 5 Dec. 2005 B21 6 Jul. 2005 CD19+TT+ IL-21 36 h 6 Sep. 2005 B22/B23/ 6 Sep. 2005 CD19+CD27+ IL-21 5 d 5 Dec. 2005 B24 IgM and TT+ B25/B26 20 Oct. 2005 CD19+CD27+ IL-21 7 d 5 Dec. 2005 IgM B27/B28 10 Nov. 2005 CD19+CD27+ IL-21 7 d 5 Dec. 2005 B29/B30 22 Nov. 2005 CD19+CD27+ IL-21 42 h 5 Dec. 2005 PBMC were obtained after Ficoll gradient isolation of bloodbank derived buffy coats and subsequently sorted by CD19 MACS and CD27 or by FACS Aria cell sorting. Purified B cells were then cultured in the presence of L-cells with indicated cytokines before being transduced with a retrovirus containing the caSTAT5b-ER-IRES-NGFR gene construct.

(18) Development of Single-Cell Derived, Clonal B Cell Cultures

(19) Outgrowth of caSTA5b-ER-IRES-NGFR positive B cells generally takes about 4 weeks, after which clonal cultures can be obtained by performing limiting dilution (LD) cultures or single cell sorting using flow cytometry (the FACSAria). These cultures consist of 2500 to 5000 L-cells, normal concentrations of IL-2 and IL-4 and either 1, 5 or 10 B cell/96-well when the LD is performed with 100% NGFR+ cells and 10, 100 and 1000 cell/96-well when NGFR+ cells are sorted into 96 well using the FACSAria.

(20) Restimulation of Antibody Production of caSTAT5b-ER Positive B Cell Cultures

(21) Poly-, oligo- or monoclonal caSTAT5b-ER-IRES-NGFR positive B cell cultures that were negative or low on antibody production were washed extensively before cultures were (1) deprived of 4HT, IL-2 and IL-4 before being cultured with IL-21, then after 4-10 days of supernatants were tested for IgM and IgG production or (2) deprived of 4HT for 10 days meanwhile cultured with IL-2 and IL-4, and then at day 10 IL-2 and IL-4 are replaced by IL-21. Then at different time points supernatants are tested for IgM and IgG production.

(22) Results

(23) B Cell Differentiation and Proliferation; the IL-2 and IL-4 vs. IL-21 Protocol

(24) IL-21 treated B cell cultures showed enhanced proliferative responses within the first 2-3 weeks compared to IL-2 and IL-4 (FIG. 6a). However, unlike the IL-2 and IL-4 cultures, continuous IL-21 stimulation resulted in decreased proliferation and cell death, even in the presence of active STAT5b (FIG. 6b). Suggesting that IL-21 eventually had to be replaced by IL-2 and IL-4. To study this in more detail, time series experiment were performed with CD19+CD27+ memory B cells, in which IL-21 was replaced by IL-2 and IL-4 after 36 hours or 5, 10, 15 and 20 days. As shown in FIG. 7, most cultures could be maintained after IL-21 withdrawal, even cultures that received IL-21 for 20 days.

(25) Antibody Production by IL-21 Boosted Total Human Memory B Cell Cultures

(26) Interestingly, in contrast to the IL-2 and IL-4 cultures, the IL-21 boosted cultures were able to produce antibodies for a relatively long period (IgG and IgM as measured by ELISA, Dako, Glostrup, Denmark) (FIGS. 8a and 8b, respectively). Importantly, of the polyclonal memory B cell cultures of donors B18 and B19, single-cell clones were obtained by LD culture (table 3).

(27) TABLE-US-00004 TABLE 3 Frequency of clones that were isolated from CD19+ CD27+ NGFR+ sorted B cells B18 B19 positive total # positive total # donor well wells well well 1000 c/w 8 9 8 10 100 c/w 21 48 19 48 10 c/w 6 48 2 48 1 c/w 1 96 6 96

(28) The majority of the clonal cultures produced IgM while only some produced IgG (FIG. 9). In addition, two clonal cultures produced both IgM and IgG clone 7 and clone 8). Whether these clones are indeed clonal or that class switching occurred remains to be determined. In the later case one BCR VDJ region should be found in the IgG or IgM gene fragments in this culture.

(29) Antibody Production of IL-21 Boosted Tetanus Toxoid Specific B Cell Cultures

(30) Next, we tested whether we could isolate B cells producing Tetanus Toxoid (TT) specific antibodies. In brief, the following protocol was carried out:

(31) Part 1

(32) (1) CD27+TT+ B cells were sorted (recovery was donor dependent and ranged from 10000-1000 cells),

(33) (2) cultured with IL-21 for 36 h,

(34) (3) transduced with caSTA5b-ER-IRES-NGFR,

(35) (4) and cultured for variable times with IL-21 (36 h to 3 wks) after which IL-21 was replaced by IL-2, IL-4 and 4HT

(36) Part 2

(37) (5) when cultures were 100% NGFR+ they were cloned by limiting dilution (LD)

(38) After 2 to 3 months of culture, 100% NGFR+, .alpha.-TT-specific polyclonal B cell cultures were obtained from at least 7 different donors (PART 1). All donors were tested positive in a .alpha.-TT-IgG specific antibody ELISA (r-biopharm, Darmstadt, Germany). As shown in FIG. 10a, .alpha.-TT IgG levels were relatively low. Since immortalization of memory B cells resulted in high numbers of IgM producing cultures (FIG. 9), that indicates that the majority of the TT cultures are IgM positive. As shown in FIG. 10b, five out of seven donors were producing IgM, suggesting that the anti-TT antibodies are from IgM origin and thus not detected by our .alpha.-TT IgG ELISA.

(39) This let us to develop an .alpha.-TT IgM ELISA based on the r-biopharm TT IgG ELISA. The only difference is in the final step, now a .alpha.-human IgM-HRP antibody instead of a anti-human IgG-HRP is added.

(40) Next, from the polyclonal TT cultures, .alpha.-TT-specific B cell clones were derived by LD cultures (PART 2). These LD cultures were started with 100% NGFR+ polyclonal .alpha.-TT-specific B cells from four donors (table 4). Clones from donor B16 mainly produced IgG, while B18 and B19 produced IgM and B15 produced both IgG and IgM (not shown). Subsequently, supernatants of these clones were tested in the IgG TT or IgM TT ELISA (FIG. 11). Besides donor 15 all donors showed TT binding, although only 5 clones produced relatively high anti-TT antibody titers.

(41) TABLE-US-00005 TABLE 4 Limiting dilution culture of 100% NGFR+ TT-specific B cells Number of positive clones from 96 well donor Total # from 1 c/w from 5 c/w from 10 c/w B15 12 2 10 B16 14 7 7 B18 10 10 B19 11 1 3 7

(42) Restimulation of Antibody Production of IL-21 Boosted Tetanus Toxoid Specific B Cell Cultures

(43) were able to generate IgM and IgG producing poly- and monoclonal B cell cultures using IL-21 as a stimulus. Nevertheless, antibody production was not stable. To our surprise, however, these IL-21 treated cultures could be restimulated to produce IgG and IgM antibodies (FIGS. 12a and 12b, respectively). This was achieved by 4HT withdrawal and simultaneously stimulation with IL-21. Using this protocol, total antibody production increased 2- to 1000-fold for IgM, and 2- to 25-fold for IgG. Several of the supernatants of restimulated monoclonal cultures were now tested positive in the IgG and IgM Tetanus ELISA. (FIG. 13).

(44) Important to note is that caSTAT5b or caSTAT5b-ER B cell cultures that had not been treated with IL-21 prior to caSTAT5b transduction and subsequent expansion could not be restimulated to produce antibodies under any conditions (see patent application WO 03/052083; not shown here).

Example 4

(45) Materials and Methods

(46) The detailed methods regarding: [0140] Maintenance and isolation of human B cells; [0141] Transduction and regulation of mouse constitutive active STAT5b in B cells; [0142] Development of 100% caSTAT5b positive B cell lines that secrete antibodies; [0143] Development of single-cell derived, clonal B cell cultures; and [0144] Restimulation of antibody production of caSTA5b-ER positive B cell cultures are described in Example 3

(47) Antibody Production in IL-21 Containing B Cell Cultures that are EBV Infected and Express BCL6-IRES-NGFR or caSTAT5-ER-IRES-NGFR

(48) Purified primary CD19+CD27+ B cells (B cells) were stimulated for 36 h with IL-21 and irradiated CD40L expressing L-cells (L-cells) before being transduced with BCL6-IRES-NGFR or caSTAT5b-ER-IRES-NGFR. After transduction, BCL6 transduced cells were cultured with IL-21, and caSTAT5b-ER transduced cells were cultured with IL-2 and IL-4. Transduced cells became NGFR positive within 2 to 3 days and were subsequently sorted on a FACSAria. NGFR sorted cells were then cultured at cell densities of 100 to 5000 cells/96 well (mini bulk cultures, MBC). The BCL6/IL-21 MBC proliferated strongly compared to caSTAT5b-ER cultures. These cultures were tested for antibody production, expanded to 24 well and frozen (viable cells, cell pellet and supernatant). The caSTAT5b-ER cultures were expanded and split in two parallel 96 wells. One well was cultured with IL-2, IL-4 and 4HT and the other well was cultured with IL-21 and without 4HT.

(49) RT-PCR

(50) To test if the strong proliferative response was related to the presence of EBV, an EBV RT-PCR was performed. Total RNA was isolated from thawed pellets using the RNeasy mini kit (Qiagen). RNA was reverse-transcribed in a volume of 20 .mu.l containing 5 first-strand buffer, 500 .mu.M dNTPs, 25 .mu.g/l oligo(dT) and 200 U superscript II RT (Life Technologies). A portion of the cDNA solution (1 .mu.l) was amplified by PCR in a 50 .mu.l solution containing 20 mM Tris-HCl, 50 mM KCl, 1.5 mM MgCl2, 5 mM dNTPs, 2.5 U Taq DNA polymerase (Life Technologies) and 30 pmol of each primer. PCR conditions were as follows: a 7-minute denaturing step at 94.degree. C. followed by 30 cycles of 30 s at 94.degree. C., 30 s at 62.degree. C. (HPRT1), 52.degree. C. (LMP-1) and 58.degree. C. (EBNA1/2) and 30 s at 72.degree. C., and a final 7-minute extension at 72.degree. C. The oligonucleotides used for RT-PCR were as follows: HPRT1 forward (5-TATGGACAGGACTGAACGTCTTGC-3; SEQ ID NO: 1) and HPRT1 reverse (5-GACACAAACATGATTCAAATCCCTGA-3; SEQ ID NO: 2); LMP-1 forward: (5-GCGACTCTGCTGGAAATGAT-3; SEQ ID NO: 3) and LMP-1 reverse (5-GACATGGTAATGCCTAGAAG-3; SEQ ID NO: 4); EBNA1/2 forward (5-AGCAAGAAGAGGAGGTGGTAAG-3; SEQ ID NO: 5) and EBNA1/2 reverse (5-GGCTCAAAGTGGTCTCTAATGC-3; SEQ ID NO: 6).

(51) In addition to the RT-PCR we performed a PCR directly on cell pellet and supernatant DNA that was isolated using the QIAmp isolation kit (Qiagen).

(52) EBV Vultures

(53) To study the role of EBV in our system in more detail, we set up experiments to determine proliferation and antibody production in untransduced, BCL6-IRES-NGFR and caSTAT5b-ER-IRES-NGFR transduced cells that were obtained from cultures with or without natural occurring EBV.

(54) First Experimental Setup

(55) TABLE-US-00006 TABLE 5 Experimental setup BCL6 1 BCL6 IL-21 2 BCL6 IL-21 EBV 3 IL-21 EBV Table 5 in summary: BCL6 MBC with IL-21 and L-cells (cultures Ce2-B9 and Ce2-G9) BCL6 MBC with IL-21 L-cells and EBV (cultures Be3-F3 and Be2 G9) BBV MBC with IL-21 and L-cells (cultures B28 and B29) Cultures with IL-2 and IL-4 were also included. At weekly intervals antibody levels and cell numbers were determined (Table 7), EBV status of the cultures is shown in FIG. 14.

(56) Second Experimental Setup

(57) TABLE-US-00007 TABLE 6 Experimental setup caSTAT5b-ER Table 6, caSTAT5b-ER MBC were cultured on 4HT, L-cells, IL-2 and IL-4 in 96 well at 1000 c/w 1 4HT IL-2 and IL-4 2 no 4HT IL-21 3 no 4HT IL-21 EBV

(58) When cell density was high enough parallel cultures were created. One part was maintained on 4HT, L-cells, IL-2 and IL-4 while the other part was switched to IL-21, L-cells but no 4HT. Antibody production was determined by ELISA (DAKO) EBV infection was determined by LMP1 PCR (FIG. 16)

(59) Results

(60) EBV Infection of BCL6 Cultures, Determined by PCR

(61) It was found that IL-21 induced strong B cell proliferation and differentiation, which we tested in two donors in combination with the transduction of BCL6 (Table 5). Indeed we found strong B cell outgrowth in culture conditions in which we cultured B cells at 1000 and 100 c/w densities (Table 7). However in one donor almost all cultures were suspected to be EBV infected based on the phenotype by light microscopy, color of the culture supernatant and enormous cell expansion. Indeed this donor (B29) turned out to be EBV infected (FIG. 14). The massive outgrowth of EBV infected cells demonstrates that the combination of BCL6 and IL-21 gives a growth advantage for EBV infected cells especially since the frequency of EBV infected B cells in vivo is thought to be relatively low. In contrast to donor B29, donor B30 was EBV negative, except for a weak and relatively small LMP1 band in sample Ce2-F2 (FIG. 14).

(62) Proliferation of BCL6 Cultures in the Presence or Absence of EBV, IL-2 IL-4 and IL-21

(63) To study the role of EBV, BCL6, IL-2, IL-4 and IL-21 on cell growth, proliferation kinetics were compared (see table 5 experimental set up). Samples were selected based on LMP-1 EBV PCR reactions (FIG. 14). EBV status was confirmed by the ability of EBV infected cells to grow in the absence of L-cells. The BCL6 samples cultured with IL-2 and IL-4 displayed a low proliferative capacity and could not be maintained (Table 7). All samples cultured in medium containing IL-21 showed strong proliferation. IL-2 and IL-4 could only induce strong proliferation when cells were infected with EBV alone, not in combination with BCL6.

(64) TABLE-US-00008 TABLE 7 proliferation of BCL6 B cells Table 7, Proliferation of BCL6 transduced and non-transduced B cells with and without EBV or with and without IL-2 IL-4 or IL-21; all samples contained L cells Culture cytokines added condition IL-21 IL-2 IL-4 BCL6 ++ BCL6/EBV +++ + EBV +++ +++

(65) Determine Antibody Production by BCL6 Cultures in the Absence or Presence of EBV, IL-2 IL-4 and IL-21

(66) IgG antibody production was determined of the long term cultures as described in table 5. The indicated antibody production levels are the mean antibody production of six to ten different measurements in time of two donors (FIG. 15). It is clearly shown that the BCL6/EBV/IL-21 cultures produce significantly more IgG than the BCL6/IL-21 cultures. Hence, EBV significantly enhances antibody production of BCL6 transduced B cells.

(67) Thus, the combination of IL-21 and EBV results in high levels of antibody production, in the absence as well as in the presence of BCL6.

(68) Determine the B Cell Receptor (BCR) Expression Alter Long Term Culture of BCL6 Transduced and EBV Infected Cells

(69) It has been described that EBV infected cells lose their BCR expression. Therefore B cells in IL-21 containing cultures (described in table 5) were stained for NGFR, CD19, Kappa en Lambda.

(70) The BCL6 transduced, EBV negative cells remained BCR expression positive as determined by Kappa and Lambda staining. Hence, such cells are particularly suitable for isolating and/or screening after a long period of culture for a desired specificity, for instance using labelled antigen, because such cells will bind said labelled antigen with their BCR. However, when EBV was present BCR expression was lost or diminished. Since the BCL6/IL-21 cultures produced relatively lower amounts of antibody, as compared to the EBV infected cells, but maintain BCR surface expression this demonstrates that these cells remain in a pre-plasmablast phenotype while the EBV infected cells, which produce high amounts of antibody, differentiate or have been differentiated towards a phenotype better described as plasmablast.

(71) In conclusion, in cases where the presence of a B cell receptor on B cells is desired, such as for instance in screening assays, the B cells are preferably not, or at a later stage, infected with EBV.

(72) EBV Infection of caSTAT5b-ER Cultures, Determined by PCR

(73) Parallel to the BCL6 transductions, transductions using caSTAT5b-ER were performed. Strikingly, in contrast to the BCL6 cultures, which all became EBV infected, the caSTAT5b-ER cultures of the same donor (B29) seemed to keep the natural distribution (percentage) of EBV infected cells. Several cultures had clear signs of EBV infection and were checked by LMP1 PCR and indeed were found to be positive. One of the signs caSTAT5b-ER cultures were EBV positive was the ability of the MBC to survive when treated with IL-21 in the absence of 4HT. These 4HT deprived B cells lack the active form of caSTAT5b and normally die within 2 to 3 weeks.

(74) Proliferation and Antibody Production by caSTAT5b-ER-Transduced B Cells Under Different Conditions

(75) To study the role of EBV in the caSTAT5b-ER system experiments were performed as described in table 6. In table 8 a schematic overview is presented of the response of caSTAT5b-ER transduced B cells. Active caSTAT5b-ER by the presence of 4HT blocks B cell differentiation irrespective which cytokines were present or even if B cells were EBV infected. Therefore, caSTAT5b-ER cultures that are EBV infected and maintained in IL-2 IL-4 or IL-21 but with 4HT do not produce antibody. Withdrawal of 4HT results in differentiation and subsequent antibody production. Preferably, IL-21 is added during and/or after withdrawal of 4HT. However, these cells ultimately die since caSTAT5b is inactive, and thus antibodies will only be produced for a restricted period. Hence, if no EBV is present, IL-21 is eventually replaced by at least one other growth stimulating agent, such as for instance Il-2 and IL-4, as already described in Example 3.

(76) EBV and IL-21 together in the absence of 4HT are two strong stimuli which induce long term proliferation and high levels of antibody production (FIG. 18). This combination is therefore preferred.

(77) TABLE-US-00009 TABLE 8 proliferation, survival and antibody production of caSTAT5b-ER B cells cytokines EBV 4HT proliferation survival production IL-2 IL-4 + + pos neg IL-2 IL-4 + + ++ pos neg IL-21 + + ++ pos neg IL-21 ++ neg intermediate IL-21 + ++ pos high

Example 5

(78) Methods

(79) BCL6-IRES-YFP positive cells were transduced with STAT3/ER-IRES-GFP using standard procedures and expanded on L cells with IL-2 and IL-4 as described in the Methods section of Example 1. The expression of STAT3 was regulated through the presence or absence of tamoxifen (4HT). YFP and GFP positive cell were sorted and equal numbers were cultured. BLIMP1 gene expression was monitored by RT-PCR and antibody production was determined in cultures with and without 4HT.

(80) Results

(81) Addition of 4HT to the cells resulted in increased cell numbers, increased Blimp-1 expression. Moreover, enhanced IgG production was measured as is shown in FIGS. 19a-c.

Example 6

(82) CD19 positive B cells were transduced with control YFP-IRES-YFP (cYFP); BCL6-IRES-YFP (BCL6-YFP) or Bcl-xL-GFP (Bcl-xL-GFP). Cells were then maintained on CD40L and IL-4 and the percent YFP and GFP single and double positive cells was determined over time by FACS in unsorted bulk cultures.

(83) Cell division and cumulative expansion was determined in single and double positive cell in the presence of IL-4 or IL-21.

(84) To check for gene expression and EBV co-infection, RT-PCR analysis of Bcl-xL, BCL6, LMP1, and EBNA1 mRNA expression was performed in cultures of single BCL6-transduced and BCL6/Bcl-xL-double transduced bulk cultures.

(85) Results

(86) FIGS. 20-23 show that the double transduced B cells containing BCL6 and Bcl-xL had a higher cumulative expansion rate and indeed also were the dominant cells to grow out in bulk memory B cell cultures. We also show that these double transduced cells divided twice as fast compared to the single transduced cells. Not shown here are the antibody production levels (IgG and IgM), which were equal between the BCL6 and BCL6/Bcl-xL cultures when cultured with IL-4 or IL-21.

Example 7

(87) The Hodgkin cell line L591, which is positive for tyrosine-phophorylated STAT5, was cultured independent of L cells (CD40 stimulation) and cytokines. L591 cells were transduced by lentivirus containing E47-IRES-GFP or control virus with GFP only (methods are described in more detail in example 1). Transduced cells were sorted and cell growth was followed in time.

(88) Results

(89) FIG. 24 shows that L591 cells quickly stop dividing when E47 is expressed. This is suggestive for the effect of E47, which via its downstream targets (besides others Socs1, Socs3, Id2, Eto2 and Xbp1) induces B cell differentiation toward an antibody producing B cell phenotype. The induced differentiation by E47 could also indicate that the effects of STAT5 are abolished or that E47 directly or in directly affects the amount and action of functional STAT5. Thus based on the data with the Hodgkin cell line L591 we state that STAT5b/ER positive B cell cultures that are maintained on L cells with 4HT and IL-2 and IL-4 or IL-21, can be induced to differentiate towards antibody producing cells when the E47 is active.

BRIEF DESCRIPTION OF THE DRAWINGS

(90) FIG. 1 illustrates enhanced growth of BCL6.sup.+ cells cultured with IL-21. 100% pure BCL6.sup.+ memory B cells were cultured in the presence of IL-2 and IL-4 (conventional culture conditions), or with IL-21 alone. The total expansion of live cells over 17 days of culture with IL-21 is shown.

(91) FIG. 2 illustrates: plasmablast immortalization of BCL6-positive cells with IL-21. Memory B cells were transduced With a retrovirus expressing BCL6-GFP and cultured with IL-2 and IL-4 (to prevent differentiation) or with IL-21 for 14 days. The surface staining for CD38 and CD20 of GFP.sup.+ (that is, BCL6.sup.+) cells is shown. IL-21 induces an 8-fold increase in the amount of B cells with a plasmablast phenotype.

(92) FIG. 3 illustrates IL-21 upregulates BLIMP1 in BCL6.sup.+ B cells. 100% pure BCL6-.DELTA.NGFR.sup.+ were cultured with IL-2 and IL-4 or IL-21 for 24 days. cDNA was generated from total RNA and mRNA levels of BLIMP1 and HPRT (loading control) were determined by reverse transcriptase polymerase chain reaction.

(93) FIG. 4 illustrates: non-limiting model of one embodiment according to the present invention.

(94) FIG. 5 illustrates a general overview of ideal culture scheme, see for more details the material and methods section of Example 3.

(95) FIG. 6A. illustrates growth dynamics of IL-2 and IL-4 vs. IL-21 stimulated B cell. Peripheral blood (PB) memory B cells derived from two donors (B18 and B19) were stimulated either with IL-21 or IL-2 and IL-4. Cells were transduced with caSTAT5b-ER-IRES-NGFR at day 2 for the IL-21 and at day 5 for IL-2 and IL-4 treated cultures; 4HT was added at day 13.

(96) FIG. 6B illustrates of 4 donors Tetanus Toxoid specific B cells were sorted from PB (cell numbers ranged from 1000-10.000). Cells were cultured in 96 well with IL-21 and transduced with caSTAT5b-ER-IRES-NGFR on day 2. 4HT was added on day 4 and IL-21 was replaced with IL-2, IL-4 and 4HT after 7 days (B14 and B15) or was replaced after 20 days (B16 and B17). Cells were counted by hand and dead cells were excluded.

(97) FIG. 7 illustrates percentage caSTAT5b-ER-IRES-NGFR transduced cells was determined using the LSR II (BD). Of two donors (B18 and B19) IL-2 and IL-4 vs. IL-21 time series experiment were performed. Of each donor of the cells were transduced using the IL-2 and IL-4 protocol, the remaining was transduced using IL-21. Directly after the IL-21 transduction (36 h) one third of the IL-21 culture was switched to IL-2 and IL-4. This was repeated on day 5, 10 and 20 of the IL-21 culture.

(98) FIG. 8A illustrates total human IgG and IgM antibody production by caSTAT5b-ER-IRES-NGFR transduced PB derived memory B cells, as described in FIGS. 6 and 7. Mean IgG production of donor B18 and B19 is shown. B cells were transduced using the IL-2 and IL-4 vs. the IL-21 protocol. The IgG production indicated with the open symbols represent all cultures that had been treated with IL-21, irrespective when they were switched to IL-2 and IL-4.

(99) FIG. 8B illustrates IgM production in samples as described above, note that the time scale is different.

(100) FIG. 9 illustrates antibody production of B cell clones derived from memory B cells of donors B18 and B19 transduced with caSTAT5b-ER-IRES-NGFR. Ten-day-old cultures that were derived from IL-21 stimulated B cells (stimulated for 36 h) were used for LD culture. Twelve clones were obtained; 5 from and 7 from B19. IgG production is the mean of three time points; IgM production is the mean of two time points.

(101) FIG. 10A. illustrates IgG Tetanus Toxoid ELISA on supernatant of polyclonal, 100% caSTA5b-ER-IRES-NGFR positive, Tetanus Toxoid sorted human B cells. Of 7 donors rapidly proliferating clonal cultures were derived. Shown is the average TT antibody production of at least 3 different measurements per donor. Each time the relative OD was determined (generally a relative increase of >2 to 3 times the background is assumed positive).

(102) FIG. 10B illustrates that to determine if TT IgG ELISA negative cultures could be producing IgM, the same 7 donor samples were tested in a total IgM ELISA.

(103) FIG. 11 illustrates anti-Tetanus Toxoid ELISA. The binding of IgG and IgM .alpha.-TT specific antibodies by ELISA was determined. Supernatants of 100% NGFR positive clonal B cell cultures derived from donors B15, B16, B18 and B19 were tested. Two times the background was set as positive.

(104) FIG. 12A illustrates total IgG and IgM production after restimulation of clonal B cell cultures donor B16 which produces IgG.

(105) FIG. 12B illustrates restimulation of clonal B cell donor B19 which produces IgM. Production was measured in supernatant of cultures that were either cultured with IL-2, IL-4 and in the presence or absence of 4HT or with IL-21 and in the presence or absence of 4HT. Cultures containing IL-2 and IL-4 did not show an increase in antibody secretion (not shown). Only cultures that responded to the restimulation are shown (10 out of 14 IgG and 8 out of 9 IgM clones responded)

(106) FIG. 13A illustrates antibodies secreted by IL-21 restimulated and 4HT deprived cultures, as described in the legend of FIG. 8 were tested for their antigen specificity. The supernatants derived from restimulated donor B16 clonal TT cultures were tested in the .alpha.-TT IgG ELISA.

(107) FIG. 13B illustrates the supernatants derived from donor B19 cultures were tested in the IgM ELISA (b). Shown is the relative increase in antibody binding compared to the negative control, samples B19-10B7 and 10E1 were cut off at 30 for visibility; values were 96 and 121, respectively.

(108) FIG. 14 illustrates LMP1 RT-PCR was performed on RNA isolated from frozen cell pellets of indicated cultures. Shown are 15 cultures which were randomly selected and tested for EBV infection. Sample coding: B indicates donor B29, C indicates donor B30 and both were cultured at 1000 c/w (e3) or 100 c/w (e2). All cultures were transduced with BCL6 except for B28 UTD (untransduced) and B29 UTD; JY cells were used as positive controls.

(109) FIG. 15 illustrates IgG antibody production (ng/ml) by BCL6 and EBV cultures as described in table 5. For each condition the average of two samples is shown each of which consist of a longitudinal follow up of 6 to 10 time points. An asterisks indicates samples are significantly different (p<0.05, unpaired student t-test). The samples cultured with IL-2 and IL-4 are a combination of EBV and BCL6 positive and negative cultures longitudinal followed.

(110) FIG. 16 illustrates Kappa-FITC and Lambda-PE staining on CD19, and NGFR positive cells. One B cell is either positive for Kappa or Lambda. Samples were measured using a LSRII (BD) and analysed using FlowJo software.

(111) FIG. 17 illustrates a representative LMP1 PCR performed on DNA isolated from frozen cell pellets of donor B25 1000 c/w MBC transduced with caSTAT5b and suspected to be EBV positive based on the color of the culture medium, growth kinetics and phenotype as observed by light microscopy. All other cultures were EBV negative.

(112) FIG. 18 illustrates average IgG production (ng/ml) in caSTAT5b-ER B cells cultured without 4HT, with IL-2 IL-4 or IL-21 and with or without EBV. The increase in antibody production in the presence of IL-21 was significant compared to cultures with IL-2 and IL-4 (p<0.05). The increase in antibody production in IL-21 containing, EBV infected cultures was significant compared to cultures without EBV (p<0.05), nonparametric Mann-Whitney)

(113) FIG. 19 (A-C) illustrate BCL6-IRES-YFP.sup.+ cells were transduced with STAT3ER-IRES-GFP and expanded on CD40L L cells with IL-2 and IL-4. BCL6/STAT3ER positive cells were sorted by FACS and equal numbers were cultured in the absence of cytokines, but in the presence or absence of 4HT (1 .mu.M) for 4 days. (A) shows live cell numbers after 4 days. (B) shows semi-quantitative RT-PCR for BLIMP1 and HPRT1 expression in BCL6-YFP.sup.+/STAT3ER-GFP.sup.+ cells. (C) shows IgG production in BCL6-YFP.sup.+STAT3ER-GFP.sup.+ treated for 4 days.+.4HT.

(114) FIG. 20 (A-B) illustrates: (A). CD19.sup.+ B cells were transduced with control YFP-IRES-YFP (cYFP); BCL6-IRES-YFP (BCL6-YFP) or BcIXL-GFP (Bcl-xL-GFP). Cells were then maintained on CD40L and IL-4 and the percent YFP or GFP positive was determined over time by FACS. All data represented in A and B are derived from CD19+CD3.sup. gating. (B). Unsorted bulk cultures of Bcl-xL-IRES-GFP and BCL6-IRES-YFP double transduced B cells on CD40L and IL-4. Individual GFP.sup.+, YFP.sup.+, and GFP/YFP double positive cells was determined by FACS

(115) FIG. 21(A-B) illustrates: IL-21 increased proliferation of B cells transduced with Bcl-xL, BCL6, or Bcl-xL+BCL6 double transduced cells. At day 17 after transduction and maintenance on CD40L and IL-4, cultures were split and cultured on CD40L in the presence of IL-4, or IL-21. Absolute number of transduced cells was determined and the cumulative expansion was calculated in single transduced cells (A) or in double transduced cells (B).

(116) FIG. 22 illustrates Long-Term Cultures are EBV.sup.

(117) RT-PCR analysis of Bcl-xL, BCL6, LMP1, and EBNA1 mRNA expression in Day 66 cultures of BCL6-transduced cells (Lane 1) and BCL6/Bcl-xL-double transduced bulk cultures (Lane 2). 1 .mu.l of a cDNA reaction performed in the absence of reverse transcriptase (RT) reaction was used as a negative control for genomic DNA contamination. Positive controls: Bcl-xL, STATS-ER transduced B cells cultured with 4-HT; BCL6, LMP1, and EBNA1, human Raji B cells.

(118) FIG. 23 illustrates doubling time of Bcl-xL, BCL6, and Bcl-xLBCL6 double transduced cells. Based on the number of transduced (GFP.sup.+,YFP.sup.+) B cells the doubling time between days 51-59 of culture was calculated in Bcl-xL, and BCL6-transduced cells in single transductions as well as in Bcl-xL+BCL6 double-transduced bulk cultures.

(119) FIG. 24 illustrates L591 a Hodgkin cell line, was transduced by lentivirus containing E47-IRES-GFP. GFP positive cells were sorted and cultured independent of L cells (CD40 stimulation) and cytokines. Cell numbers were determined in time.

REFERENCES

(120) Banchereau, J., de Paoli, P., Valle, A., Garcia, E., Rousset, F., (1991). Long term human B cell lines dependent on interleukin-4 and antibody to CD40, Science 251, 70-2. Boise, L. H., M. Gonzalez-Garcia, C. E. Postema, L. Ding, T. Lindsten, L. A. Turka, X. Mao, G. Nunez, and C. B. Thompson. (1993). Bcl-x, a bcl-2-related gene that functions as a dominant regulator of apoptotic cell death. Cell 74:597. Dadgostar, H., Zarnegar, B., Hoffmann, A., Qin, X. F., Truong, U., Rao, G., Baltimore, D., and Cheng, G. (2002). Cooperation of multiple signaling pathways in CD40-regulated gene expression in B lymphocytes. Proc. Natl. Acad. Sci USA 99, 1497-1502. Malisan, F., Briere, F., Bridon, J. M., Harindranath, N., Mills, F. C., Max, E. E., Banchereau, J., Martinez-Valdez, H. (1996). Interleukin-10 induces immunoglobulin G isotype switch recombination in human CD40-activated naive B lymphocytes, J. Exp. Med. 183, 937-47. Mathas S, Janz M, Hummel F, Hummel M, Wollert-Wulf B, Lusatis S, Anagnostopoulos I, Lietz A, Sigvardsson M, Jundt F, Johrens K, Bommert K, Stein H, Dorken B (2006). Intrinsic inhibition of transcription factor E2A by HLH proteins ABF-1 and Id2 mediates reprogramming of neoplastic B cells in Hodgkin lymphoma. Nat Immunol. 7, 207-215. Traggiai, E., Becker, S., Subbarao, K., Kolesnikova, L., Uematsu, Y., Gismondo, M. R., Murphy, B. R., Rappuoli, R., Lanzavecchia, A. (2004). An efficient method to make human monoclonal antibodies from memory B cells: potent neutralization of SARS coronavirus. Nature Medicine Volume 10, No. 8, 871-875. Ye, B. H., Cattoretti, G., Shen, Q., Zhang, J., Hawe, N., de Waard, R., Leung, C., Nouri-Shirazi, M., Orazi, A., Chaganti, R. S., et al. (1997). The BCL-6 proto-oncogene controls germinal-centre formation and Th2-type inflammation. Nat Genet 16, 161-170.