Means and methods for producing high affinity antibodies

Abstract

The invention provides means and methods for modulating the occurrence of somatic hypermutations in antibody producing plasmablast-like B-cells.

Claims

1. A method of antibody production, the method comprising: introducing an exogenous nucleic acid molecule encoding Bcl 6 into a B cell; introducing an exogenous anti-apoptotic nucleic acid molecule into said B cell, thus generating an antibody producing plasmablast-like B cell with functional AID activity; selecting an antibody producing plasmablast-like B cell that produces an antibody of interest; and reducing expression and/or activity of AID in said B cell thereby reducing the occurrence of somatic hypermutations in said B-cell.

2. The method according to claim 1, wherein the expression and/or activity of AID is reduced in said B cell by introduction of a molecule that interferes with the homo- or heterodimerization of E47 and/or E12.

3. The method according to claim 1, wherein the expression and/or activity of AID is reduced in said B cell by introduction of an exogenous nucleic acid molecule encoding an ID protein.

4. The method according to claim 1, wherein the expression and/or activity of AID is reduced in said B cell by introduction of an antisense nucleic acid or ribozyme directed against E12 and/or E47.

5. The method according to claim 1, wherein the expression and/or activity of AID is reduced in said B cell by introduction of a dsRNA molecule that induces E12 or E47 mRNA degradation.

6. The method according to claim 1, wherein the expression and/or activity of AID is reduced in said B cell by introduction of a zinc-finger protein that has been modified to be able to bind to the promoter region of AID and which is coupled to a transcriptional repressor domain.

7. The method according to claim 1, wherein the anti-apoptotic nucleic acid comprises a gene of the BCL2 family.

8. The method according to claim 7, wherein the gene of the BCL2 family is Bcl-xL or Mcl 1, or a functional part thereof.

9. The method according to claim 1, wherein the B-cell is cultured in the presence of IL21 and CD40L.

Description

FIGURE LEGENDS

(1) FIG. 1a and FIG. 1b.

(2) FIG. 1a and FIG. 1b illustrate CD CD27.sup.+ memory peripheral blood cells acquire a stable GC-like phenotype following transduction with BCL6 and Bcl-xL and subsequent culture (throughout the patent application these cells are named antibody producing plasmablast-like cells). FIG. 1a shows phenotype of BCL6+Bcl-xL transduced CD27+ memory B-cells (6XL, black histogram line) compared to tonsil GC cells (GC, CD38.sup.+ CD20[[.sup.30]].sup.+, shaded grey), tonsil naive and memory cells (N/M, CD.sup. CD20.sup.10w, light gray histogram line), and tonsil plasma cells (PC, CD38.sup.++ CD20.sup.low, dark gray histogram line). BCL6+Bcl-xL transduced monoclonal cell lines show an identical phenotype (not shown). FIG. 1b shows relative mRNA levels of AICDA (encoding AID) in CD19.sup.+ IgG.sup.+ CD27.sup.+ PB memory B-cells and CD19.sup.+ CD38.sup.+ CD20.sup.+ IgD.sup. tonsillar GC B-cells (value set as 1) compared to BCL6+Bcl-xL transduced bulk CD27.sup.+ memory cells using quantitative RT-PCR.

(3) FIG. 2a, FIG. 2b, FIG. 2c, FIG. 2d, and FIG. 2e

(4) FIG. 2a, FIG. 2b, FIG. 2c, FIG. 2d, and FIG. 2e show expression and activity of AID in BCL6+Bcl-xL transduced cells. FIG. 2a shows an overview of all D25 subclones that had one or more amino acid substitutions, additional silent mutations are shown in italic. FIG. 2b shows mRNA levels of AICDA (encoding AID) in CD19.sup.+ CD38.sup.+ CD20.sup.+ IgD.sup. tonsillar GC B-cells and CD19.sup.+ IgG.sup.+ CD27.sup.+ PB memory B-cells compared to 23 BCL6+Bcl-xL transduced monoclonal cell lines and monoclonal anti-RSV specific cell line D25, using quantitative RT-PCR. FIG. 2c shows percentage of subclones with indicated number of VH mutations, as percentage of total number of subclones sequenced. FIG. 2d shows location of VH mutations, percentage of mutations per VH region per basepair. FIG. 2e shows binding of D25-subclone Ig to RSV infected HEp2 cells. Squares are individual D25 subclones, black circles rD25. Grey circles indicate clones with deviating affinity.

(5) FIG. 3a and FIG. 3b

(6) FIG. 3a and FIG. 3b show Id regulates AID expression in BCL6 Bcl-xL transduced B-cells. FIG. 3a shows mRNA levels of AICDA in the monoclonal anti-RSV specific cell line D25 transduced with Control-YFP, AID-YFP, ID2-YFP or ID3-YFP using quantitative RT-PCR. FIG. 3b shows growth curves of the monoclonal anti-RSV specific cell line D25 transduced with Control-YFP, AID-YFP, ID2-YFP or ID3-YFP.

(7) FIG. 4a and FIG. 4b

(8) FIG. 4a and FIG. 4b show enhanced binding and competition of B-cell supernatant derived and recombinant protein of mutated D25 clones to RSV infected HEp2 cells. FIG. 4a shows titration of B-cell culture supernatant of 2 clones that were selected based on two identical amino acid substitutions and one additional mutation compared to the original D25 sequence (#29: S83Y/V111I/V112L and #189: G63D/V111I/V112L), one clone that lost binding to RSV infected cell (#77; E107K) and an unmutated D25 clone (#54). The mutated clones were selected retrospectively after 1) two tests in which both culture IgG levels were determined and binding to RSV infected HEp2 was performed. The Igs that showed enhanced binding were ranked in a top 25. When an Ig was in the top 25 twice it was selected and 2) then tested in a competition experiment were they had to compete out the original D25 clone and 3) had mutations that could explain the phenotype seen in the above-mentioned experiments. Replacement was performed as follows: saturate RSV infected Hep2 cells with PE labeled D25, wash 3 and incubate with 250 ng ml.sup.1 D25-PE+ rising amount of competing D25 subclone (0-800 ng ml.sup.1) incubate for 3 hrs at 37. C., wash 2 and analyze on FACS. FIG. 4b shows D25 competition by same clones. Incubate RSV infected Hep2 cells with 250 ng ml.sup.1 D25-PE+ rising amount of competing D25 subclone (0-800 ng ml.sup.1) (same mix as was used for replacement assay), incubate for 30 minutes at 4. C., wash 2 and analyze onFACS.

(9) FIG. 5

(10) Transduction of human IgG+ memory B cells with BCL6, Bcl-xL and MCL-1. Shown are FACS analysis of the percentage of GFP (BCL6 or BCL6 and Bcl-xL) and NGFR (Bcl-xL or MCL-1) expressing cells 4 days and 21 days after transduction.

(11) FIG. 6a, FIG. 6b, and FIG. 6c

(12) FIG. 6a, FIG. 6b, and FIG. 6c show expression of MCL1 in combination with BCL6 and/or Bcl-xl results in outgrowth of human memory B cells. In FIG. 6a activated CD19.sup.+ CD27.sup.+ IgM.sup.IgA.sup. memory cells were transduced with BCL6-GFP and MCL-1-NGFR. In FIG. 6b activated CD19.sup.+ CD27.sup.+ IgM.sup.IgA.sup. memory cells were transduced with BCL6-GFP and Bcl-xL-NGFR. In FIG. 6c activated CD19.sup.+ CD27.sup.+ IgM.sup.IgA.sup. memory cells were transduced with BCL6-Bcl-xL-GFP and MCL-1-NGFR. Transduction markers were followed in time by FACS analysis. Depicted are the percentage of transduced cells in a bulk culture that were maintained under standard culture conditions, e.g. with irradiated CD40L-L cells and rmIL-21. Cells transduced with BCL6 in combination with MCL-1 or Bcl-xL showed growth advantage compared to cells lacking one of these transgenes and ultimately dominated the cultures (>90%).

(13) FIG. 7a, FIG. 7b, and FIG. 7c

(14) FIG. 7a, FIG. 7b, and FIG. 7c show relative increase in cell numbers of human memory B cells expressing MCL1 in combination with BCL6 and/or Bcl-xl. In FIG. 7a activated CD19.sup.+ CD27.sup.+ IgM.sup.IgA.sup.memory cells were transduced with BCL6-GFP and MCL-1-NGFR. In FIG. 7b activated CD19.sup.+ CD27.sup.+ IgM.sup.IgA.sup.memory cells were transduced with BCL6-GFP and Bcl-xL-NGFR. In FIG. 7c activated CD19.sup.+ CD27.sup.+ IgM.sup.IgA.sup.memory cells were transduced with BCL6-Bcl-xL-GFP and MCL-1-NGFR. Cell numbers were followed in time by counting. Relative cell numbers were determined by correction for the percentage of cells carrying the transgenes of interest as shown in FIG. 6. Cells expressing BCL6 and MCL-1 did grow out and showed enhanced proliferation compared to cells with BCL6 alone. However, survival and proliferation was most profound for the combination of BCL6 and Bcl-xL alone or in combination with MCL-1. Interestingly, in FIG. 7c the BCL6 and Bcl-xL cells did not outgrow cells also containing MCL-1 indicating that addition of MCL-1 is an advantage.

(15) FIG. 8

(16) Relative expansion of number of double transduced cells. Cultures of BCL6, Bcl-xL and MCL1 co-transduced cells were taken 21 days post transduction and were started with equal cell numbers and were set to 1. BCL6/Bcl-xL and BCL6/MCL-1 culture are 100% double transduced at day 21. BCL6/Bcl-xL and MCL-1 was 86% double transduced at day 21 and approximately 96% at day 43. Expansion of BCL6/Bcl-xL and MCL-1 transduced cells is equal to the expansion of BCL6 Bcl-xL transduced cells. Numbers of cells expand at least till day 43 post transduction.

(17) FIG. 9a and FIG. 9b

(18) FIG. 9a and FIG. 9b show the phenotype of BCL6, Bcl-xL and MCL-1 transduced cells based on CD20 and CD38 expression on day 4 and 21 after transduction. FIG. 9a shows cells expressing BCL6 in combination with MCL-1 and/or Bcl-xL acquire a Germinal center-cell like phenotype. FIG. 9b shows that these cells are cell surface IgG (BCR) positive, similar to BCL6 Bcl-xL co-transduced cells. Cells lacking BCL6 differentiate toward plasmablast like cells with high CD38 and reduced CD20 expression.

EXAMPLES

(19) Methods

(20) B-cell isolation. We obtained B-cells from peripheral blood (huffy coats from Sanquin) by Ficoll separation and CD22 MACS microbeads (Miltenyi Biotech). Subsequently we sorted these cells for CD19.sup.+ CD3.sup.CD27.sup.+ IgM.sup.IgA.sup. (IgG memory cells) or CD19.sup.+ CD3.sup.CD27.sup.+ IgG.sup.IgA.sup. (IgM memory cells) on a FACSAria (Becton Dickinson).

(21) Tonsil B-cell sorting. We obtained Tonsil B-cells from routine tonsillectomies performed at the Department of Otolaryngology at the Academic Medical Center, Amsterdam, The Netherlands. We separated B-cells by Ficoll and sorted the CD19+CD3CD44IgD GC population. The use of these tissues was approved by the medical ethical committees of the institution.

(22) Cell culture. We maintained B-cells (210.sup.5 cells ml.sup.1) in IMDM (Gibco) culture medium containing 8% FBS (Gibco), penicillin/streptomycin (Roche) supplemented with recombinant mouse IL 21 (25 ng ml.sup.1, R&D systems) and co-cultured them on .gamma.-irradiated (50Gy) mouse L cell fibroblasts stably expressing CD40L (CD40L-L cells, [[0.10.sup.5]]10.sup.5 cells ml.sup.1). We tested cells routinely by PCR for the presence of mycoplasma and EBV.

(23) Retroviral transduction. The BCL6, ID2 and ID3 retroviral constructs were described previously (Shvarts, A. et al. Genes Dev 16, 681-6 (2002); Jaleco, A. C. et al. Blood 94, 2637-46 (1999); Spits H. et al. J Exp Med 192, 1775-83 (2000)). In brief, we have constructed bicistronic vectors with a gene of interest linked to a downstream internal ribosomal entry site (IRES) and a marker gene that allow independent translation of the products of both genes in the transduced target cells. The ID3 coding sequence was cloned from the pCDNA-Id3 plasmid (gift of Dr C. Murre, University of California at San Diego, San Diego, Calif.). The product was ligated between the Xho I and SnaBI site of the polylinker from our plasmid LZRS-linker-IRES-GFP and/or YFP to obtain the retroviral vector LZRS-Id3-IRES-GFP.

(24) The coding sequence of human ID2 was cut from the pSG5-Id2 vector (a gift of Dr. R. de Groot, University of Utrecht, Utrecht, Netherlands) with Not1 and was ligated in the Not1 site of polylinker from our plasmid LZRS-linker-IRES-GFP and/or YFP. A codon optimized sequence of the phosphorylation-site mutant (S159A) of human MCL-1 (Maurer, U. et al. Mol Cell 21, 749-760 (2006)) was ordered from GeneArt (Regensburg Germany) and cloned into the LZRS retroviral vector. Human Bcl-xL cDNA was cloned into the LZRS retroviral vector and helper-free recombinant retroviruses were produced after transfection into a 293T-based amphotropic retroviral packaging cell line, Phoenix (Kinsella, T. M. et al Hum Gene Ther 7, 1405-1413 (1996)). Human memory B-cells were co-transduced by the different retroviruses after activation on CD40L-L cells in the presence of rmIL-21 for 36 hrs as before (Diehl, S. A. et al. J Immunol 180, 4805-15 (2008)), but in addition cells and virus were centrifuged at room temperature for 60 min at 360g (1800 RPM).

(25) Flow cytometry. We analyzed stained cells on an LSRII (BD) and processed flow cytometry data with FlowJo software (Tree Star). We purchased the following mAbs against the human molecules from BD-Pharmingen unless otherwise indicated CD3 (SK7), CD10 (HI10a), CD19 (SJ25C1), CD20 (B9E9; Beckman Coulter), CD21 (B-ly4), CD22 (B-ly8; IQ Products), CD25 (BC96; eBioscience), CD27 (0323; eBioscience), CD30 (BerH8), CD38 (HB7), CD40 (MAB89; Beckman Coulter), CD70 (Ki24), CD71 (YDJ1.2.2; Beckman Coulter), CD80 (L307.4), CD86 (2331), CD95 (DX2), CD132 (TUGh4), CD184 (CXCR4, 12G5), CD271 (LNGFR; ME20.4-1.H4; Miltenyi Biotech), CD275 (MIH12; eBioscience), HLA-DR (L243), IgA (F(ab)2; DAKO), IgD (IA6-2), IgG (G18-145), IgM (G20-127) (BD), IL-21R (152512; R&D systems), Ig-kappa (F(ab)2; DAKO, G20-193), and Ig-lambda (F(ab)2; JDC12, DAKO).

(26) RT-PCR. We carried out quantitative RT-PCR with a BioRad iCycler and used the 2.sup.(AACT) method to calculate relative mRNA expression levels normalized to ACTIN. Primers for AICDA (encoding AID) are described (Smit, L. A. et al. Cancer Res 63, 3894-8 (2003)).

(27) ELISA. We coated plates with anti-human IgG Fc-fragment (Jackson ImmunoResearch Laboratories) at 5.mu.g ml.sup.1 in PBS for 1 h at 37. C. or o/n at 4. C. and washed them in ELISA wash buffer (PBS, 0.5% Tween-20). 4% milk in PBS was used as blocking agent, before we added serial dilutions of cell culture supernatants and HRP-conjugated detection Abs (dilutions 1:2500 for HRP-conjugated IgG antibody (Jackson). We used TMB substrate solution (Biosource) for development of the ELISAs.

(28) Cloning and sequencing of mutants of D25. We isolated total RNA using the RNeasy mini kit (Qiagen), generated cDNA and performed the VH1-69 PCR to determine the sequence of the D25 subclones. From interesting clones (#29, #59, #77, #189) the heavy chain variable region was cloned into the pCR2.1 TA cloning vector (Invitrogen). To rule out reverse transcriptase or DNA polymerase induced mutations, we performed several independent cloning experiments. To produce recombinant D25 mAb we cloned D25 mutated heavy and original light variable regions in frame with human IgG1 and Kappa constant regions into a pcDNA3.1 (Invitrogen) based vector and transiently transfected 293T cells. VL sequences of D25 subclones were also determined but did not harbor any mutations compared to the original D25 light chain sequence. We purified recombinant D25 from the culture supernatant with Protein A.

(29) Results

(30) Human Peripheral Memory B-Cells Transduced with BCL6 and Bcl-xL Resemble GC-Like B-Cells

(31) We showed that overexpression of BCL6 and Bcl-xL (which are expressed in Germinal center (GC) B-cells and are under control of STATS (Scheeren, F. A. et al. Nat Immunol 6, 303-13 (2005); PCT/NL2008/050333)) synergize to increase the proliferative and survival potential of human memory B-cells in vitro when cultured on irradiated CD40L expressing L cells (CD40L-L cells) in the presence of IL 21. Normal human B-cells rapidly differentiate to antibody-producing plasma cells when cultured on CD40L-L cells in the presence of IL 21 (Ettinger, R. et al. J Immunol 175, 7867-79 (2005)) which is accompanied by a decrease in expression of surface BCR and MHC class II and an increase in expression of CD38 (Liu, Y. J. & Arpin, C. Immunol Rev 156, 111-26 (1997)). In contrast to non-transduced cells or cells expressing Bcl-xL only in the same culture, BCL6-only and BCL6+Bcl-xL cells both retained BCR expression and were HLA-DRhighCD38 intermediate (FIG. 1a) confirming that BCL6 inhibits B-cell differentiation (Scheeren, F. A. et al. Nat Immunol 6, 303-13 (2005); Diehl, S. A. et al., J Immunol 180, 4805-15 (2008)).

(32) BCL6+Bcl-xL positive B-cells expanded with CD40L and IL 21 express CD19, CD20, CD21 and CD22, the activation markers CD25, CD30, CD70, CD80, CD86, CD95, ICOSL, and the cytokine receptors CD132c) and IL-21R. These cells express CD38 and CD20 at levels equivalent to tonsil GC cells. The expression of CD27, CXCR4, CD71, CD10 and HLA-DR on transduced cells is consistently higher compared to freshly isolated tonsil GC cells (FIG. 1a), which may be due to the activation status and increased size of the cell.

(33) BCL6+Bcl-xL transduced cells expressed AICDA (encoding the enzyme AID) at levels comparable to those expressed by freshly isolated GC B-cells (FIG. 1b). As shown below, AID is functional in these cells as SHM in the Ig genes of the expanded B-cells were induced. Since AID is not expressed in peripheral blood memory cells (FIG. 1b) or plasma cells (Muramatsu, M. et al. J Biol Chem 274, 18470-6 (1999)), and transduced cells showed expression of typical GC cell surface markers, our results demonstrate that BCL6 and Bcl-xL expression in combination with CD40L and IL 21 signaling conferred GC-like characteristics to human CD27.sup.+memory B-cells.

(34) Anti-RSV Antibody D25 and D25 Mutants

(35) BCL6+Bcl-xL transduced B-cells secrete relatively high amounts of antibodies. Thereby we could select antigen-specific B-cells on the basis of secretion of specific antibody. We chose the pathogenic Respiratory Syncytial virus (RSV) as the antigenic moiety. RSV is the most common cause of bronchiolitis and pneumonia among infants and children under 1 year of age and is a serious health problem for elderly people (Thompson, W. W. et al. JAMA 289, 179-86 (2003); Hall C. B. et al NEMJ 360, 588-599 (2009)). BCL6+Bcl-xL transduced memory B-cells of a healthy donor were expanded with CD40L-L cells and IL 21 and screened for the presence of RSV-neutralizing antibodies in a microneutralization experiment (Johnson, S. et al. J Infect Dis 180, 35-40 (1999)). One of the antibodies, D25, with highest neutralizing activity was cloned by limiting dilution and further characterized (PCT/NL2008/050333). We observed median half maximum inhibitory concentrations (IC50) against the RSV-A2 virus in the range of 2.1 ng ml.sup.1.

(36) Before we studied the functional activity of AID in the D25 cell line we analyzed the AICDA expression in 23 monoclonal cell lines by real-time PCR. AID expression in these cell lines was variable. Several clones expressed levels similar to GC tonsil B-cells, whereas others were AICDA low (FIG. 2b). To determine whether AID is functional in the B-cell clones, we subcloned the RSV specific monoclonal B-cell line D25 by single cell sorting and analyzed their VH genes for the presence of mutations. Sixty three percent of the wells that were seeded with one cell showed robust expansion, showing that expression of AID does not lead to massive genetic instability leading to growth arrest and cell death. After 3 weeks culture supernatant of 108 subclones was harvested and the RNA from the cells isolated. Subsequently cDNA was generated and the VH region sequenced. Sequence analysis revealed a total of 184 VH mutations (107 unique mutations, FIG. 2a), resulting in an estimated mutation rate between 8.8510.sup.5 and 5.1410.sup.5 mutations per by per cell division, which is at the lower end of the estimated AID-mediated mutation rate in vivo (10.sup.3 to 10.sup.5) (Peled, J. U. et al. Annu Rev Immunol 26, 481-511 (2008)). The VH genes of the individual subclones show a variable number of mutations with 65% of subclones harboring one to three mutations in their VH region (of which 23% are silent mutations) and 11% harboring more than three VH mutations. Twenty four percent of the clones had no VH mutations (FIG. 2c). The 372-basepair VH gene of D25 contains 26 AID mutational hotspots (RGYW/WRCY) that account for 30% of the total mutations. Mutations were predominantly observed in the CDR regions and FR3 (FIG. 2d).

(37) While the supernatants of the majority of D25 subclones bound to RSV-infected HEp2 cells similar to recombinant D25, some of those clones bound either less or better than D25 (FIG. 2e). The differences in binding activities were associated with mutations in the VH regions.

(38) Regulation of AID Expression in Bcl6+Bcl-xL Transduced B Cells

(39) For some applications it is desirable to inhibit AID to prevent accumulation of mutations in the Ig genes of BCL6+Bcl-xL transduced B-cell clones. To achieve this we took advantage of the fact that AID is regulated by the basic Helix Loop Helix transcription factor E47 (Sayegh, C. E., et al. Nat Immunol 4, 586-93 (2003)). We overexpressed the Helix loop Helix factor inhibitors of DNA binding ID2 and ID3 which are known to form transcriptionally inactive complexes with E47, thereby inhibiting AID expression (Sayegh, C. E., et al. Nat Immunol 4, 586-93 (2003)). As shown in FIG. 3a, overexpression of both ID2 and ID3 strongly reduced AICDA levels in the D25 cell line. Proliferation of the cell line was reduced when ID2 but not when ID3 was expressed (FIG. 3b). Overexpression of AID also reduced the proliferative capacity of the D25 cells. As expected, overexpression of AID strongly enhanced the expression of AID (FIG. 3a). Thus modulation of ID2 and/or ID3 levels provides a method to modulate AID induced mutations in BCL6+Bcl-xL transduced B-cells.

(40) Function of D25 is Altered by Amino Acid Substitution Due to AID Activity

(41) We wondered if we might find new subclones of D25 that show an altered function. However, since the affinity of D25 for it putative target, the RSV Fusion (F) protein is already high and D25 neutralizes RSV already at low concentration it was difficult to find clones that would be better than D25 itself. Nevertheless we tested the B-cell culture supernatant of the D25 subclones twice for binding to RSV infected HEp2 cells and once for competition with PE-labeled D25. These experiments gave variable results (data not shown) but when all the top 25 antibodies from each data set were compared with the VH sequences some very interesting amino acid position appeared. Several D25 subclones were produced as recombinant protein of which clone #59 still had the same configuration as the original D25 clone, #77 contained a very typical mutation at position 107 (E107K), that seem to cause a reduced binding to RSV infected HEp2 cells since other clones that lost binding to infected HEp2 cells also gathered this mutation. Furthermore clone#29 (S83Y/V111I/V112L) and #189 (G63D/V111I/V112L) share a mutation at position 111 and 112 (V.fwdarw.I and V.fwdarw.L). These two clones replace the original D25 antibody from binding to RSV infected HEp2 cells (FIG. 4a) in a dose dependent manner, while D25 itself and the nonbinding antibody cannot replace the PE-labeled D25 antibody. In addition in a direct competition the two clones #29 and #189 again reduce the binding of the original immunoglobulin D25 (FIG. 4b). Thus by expression of AID in our BCL6+Bcl-xL B-cell lines we perform in vitro antibody affinity maturation within the original B-cell.

(42) Expression of the Anti-Apoptotic Gene MCL-1 Alone or in Combination with BCL6 and/or Bcl-xL Increases Survival and Proliferation of Human B Cells

(43) Since B cell cultures with BCL6 alone and co-stimulated with CD40L and 11-21 show a block in differentiation toward plasmablasts but lack the signals for enhancing survival and proliferation, we studied besides Bcl-xL also the role of the anti-apoptotic molecule MCL-1 in this process. When activated CD19.sup.+ CD27.sup.+ IgM.sup.IgA.sup.memory cells were transduced with: (i) BCL6-GFP and MCL-1-NGFR, or (ii) BCL6-GFP and Bcl-xL-NGFR or (iii) BCL6-Bcl-xL-GFP and MCL-1-NGFR (FIG. 5), they all showed enhanced survival. Cells containing BCL6 MCL-1 proliferated 3 times slower compared to cells containing BCL6 and Bcl-xL or BCL6, Bcl-xL and MCL-1 (FIGS. 6, 7 and 8). Like Bcl-xL, MCL-1 does not interfere with the function of BCL6, namely the block in differentiation towards plasmablast and these co-transduced cells indeed still possess a Germinal center phenotype (CD20+CD38+) (FIG. 9A) and also still express a surface expressed immunoglobulin (FIG. 9B). The expression of AID in MCL1 BCL6 and/or Bcl-xL transduced cells is expected not to be altered.

(44) Regulation of AID Expression in Bcl6+Mcl-1 Transduced B Cells

(45) As shown above for the BCL6+Bcl-xL transduced B-cell clones, ID2 and ID3 are also overexpressed in Bcl6+Mcl-1 transduced B cells, thereby inhibiting AID expression (Sayegh, C. E., et al. Nat Immunol 4, 586-93 (2003)).

(46) Subsequently, it is demonstrated, for instance by using quantitative RT-PCR, overexpression of both ID2 and ID3 reduce AICDA levels in these cells. Thus increasing ID2 and/or ID3 levels provides a method to prevent AID

(47) induced mutations in BCL6+Mcl-1 transduced B-cells. The procedures are described in the Methods section above.

REFERENCES

(48) Adams J. M. et al. Current Opinion in Immunology 19, 488-496 (2007) Banchereau J et al. Science 251, 70-72 (1991) Baron, U. and Bujard, H. Methods Enzymol 327, 401-21 (2000) Boise, L. H. et al. Cell 74, 597-608 (1993) Chipuk, J. E. et al. Trends in Cell Biol 18, 157-163 (2007) Christopherson, K. S. et al. PNAS 89, 6314-8 (1992) Close, P. M., et al. J Pathol 162, 209-16 (1990) Diehl, S. A. et al. J Immunol 180, 4805-15 (2008) Ettinger, R. et al. J Immunol 175, 7867-79 (2005) Good, K. L et al. J Immunol 177, 5236-47 (2006) Gossen, M. and Bujard, H. Proc Natl Acad Sci USA 89, 5547-51 (1992) Gossen, M. et al. Science 268, 1766-9 (1995) Guzman, L. M. et al. Bacteriol 177, 4121-4130 (1995) Hall C. B. et al. NEMJ 360, 588-599 (2009) Ichikawa H. T. et al. J Immunol 177, 355-361 (2006) Johnson, S. et al. J Infect Dis 180, 35-40 (1999) Kee B. L. Nat Rev Immunol 9, 175-84 (2009) Kinsella, T. M. et al Hum Gene Ther 7, 1405-1413 (1996) Kuo, T. C. et al. J Exp Med 204, 819-830 (2007) Liu, Y. J. & Arpin, C. Immunol Rev 156, 111-26 (1997) Maurer, U. et al. Mol Cell 21, 749-760 (2006) Muramatsu, M. et al. J Biol Chem 274, 18470-6 (1999) Peled, J. U. et al. Annu Rev Immunol 26, 481-511 (2008) Rousset F et al. PNAS 89, 1890-1893 (1992) Sayegh, C. E., et al. Nat Immunol 4, 586-93 (2003) Scheeren, F. A. et al. Nat Immunol 6, 303-13 (2005) Sidwell, R. W. & Barnard, D. L. Antiviral Res 71, 379-90 (2006) Smit, L. A. et al. Cancer Res 63, 3894-8 (2003) Spits H. et al. J Exp Med 192, 1775-83 (2000) Tan S et al. PNAS 100, 11997-12002 (2003) Thompson, W. W. et al. JAMA 289, 179-86 (2003) Zamore et al. Cell 101, 25-33 (2000)