Efficient affinity maturation of antibodies in an engineered chicken B cell line DT40-SW by increasing point mutation

Journal of Bioscience and Bioengineering VOL. 110 No. 3, 351 – 358, 2010 www.elsevier.com/locate/jbiosc

Efficient affinity maturation of antibodies in an engineered chicken B cell line DT40-SW by increasing point mutation Masamichi Kajita, Takahiro Okazawa, Mika Ikeda, Kagefumi Todo, Masaki Magari, Naoki Kanayama,⁎ and Hitoshi Ohmori Department of Bioscience and Biotechnology, Graduate School of Natural Science and Technology, Okayama University, 3-1-1 Tsushima-Naka, Okayama 700-8530, Japan Received 19 January 2010; accepted 11 March 2010 Available online 11 April 2010

The chicken B cell line DT40 undergoes hypermutation of immunoglobulin variable region (IgV) genes during culture, thereby constituting an antibody (Ab) library. We previously established an in vitro Ab generation system using an engineered line DT40-SW whose hypermutation machinery can be switched on and off. Abs for various antigens (Ags) can be obtained from the DT40-SW library and the specificity of the Ag-specific clones can be stabilized by stopping hypermutation. Furthermore, the affinity of obtained monoclonal Abs (mAbs) can be improved through further mutation followed by selection, a process analogous to “affinity maturation” that occurs in vivo. Although gene conversion dominantly diversifies the IgV genes in DT40 cells, point mutation is considered to be more favorable for fine-tuning Ab properties during affinity maturation. Here, we examined whether affinity maturation occurs more efficiently when the hypermutation pattern was transformed from gene conversion into point mutation in DT40-SW cells. To this end, we disrupted the XRCC3 gene that is essential for gene conversion. It was found that hemizygous disruption of the XRCC3 gene was sufficient to increase the point mutation frequency. Since hemizygous disruption is conducted more easily, we tested whether the XRCC3 (+/−) mutant generates high-affinity Abs through affinity maturation more efficiently than the wild type. Using this affinity maturation technique, we generated an improved 4-hydroxy-3-nitrophenylacetyl-specific mAb with ∼600-fold lower KD than that of the original mAb. Taken together, hemizygous disruption of the XRCC3 gene is considered to be useful for obtaining high-affinity mAbs from DT40-SW cells though affinity maturation. © 2010, The Society for Biotechnology, Japan. All rights reserved. [Key words: Antibody; Affinity maturation; Point mutation; Gene conversion; XRCC3; Chicken B cell line DT40]

In vitro screening systems have been developed to obtain desired monoclonal antibodies (mAbs) efficiently with avoiding the limitation of immunologic tolerance, which is a mechanism restricting generation of self-reactive Abs in vivo. For instance, a variety of single-chain Fv Abs have been selected from phage-display libraries (1–3). However, in immunized animals, antigen (Ag)-stimulated B cells undergo somatic hypermutation in the immunoglobulin variable region (IgV) genes and only B cells producing mutated Abs with high affinity are positively selected, a process termed affinity maturation (4–6), and thereby the in vivo affinity maturation is efficiently accomplished. On the other hand, the affinity of the primarily selected Abs from phage-display libraries can be also improved by diversifying the Ab clones with error-prone PCR, followed by expression in appropriate host cells to assess their biological activities. Generally, these processes of the in vitro affinity maturation are laborious and time-consuming. In this context, a chicken B cell line DT40 is considered to be useful to mimic the in vivo affinity maturation in in vitro culture. DT40 cells produce full-length Abs both

⁎ Corresponding author. Tel./fax: +81 86 251 8198. E-mail address: [email protected] (N. Kanayama).

in secreted and membrane-bound forms, and spontaneously introduce the activation-induced cytidine deaminase (AID)-dependent hypermutation in the IgV genes during culture (7–11), while the phage display system expresses Ab fragments and is not equipped with an intrinsic mutation machinery. Consequently, only a longterm culture of DT40 cells results in the formation of an Abdisplaying cell library, which would hold a wide variety of Ag specificities, in a cell-autonomous manner (7,8), whereas the other in vitro systems require separate processes for constructing libraries (e.g. error-prone PCR for mutation and phage DNA packaging for expression and display). It has been reported that several Ag-specific clones were successfully selected from the DT40 cell library (12–14). Thus, an Ab selection system using DT40 cells could be an alternative to conventional in vitro methods. However, clones that are selected from the DT40 library may alter their Ag-specificity during expansion culture unless the hypermutation machinery is shut down. To overcome this problem, we have established an engineered DT40 line named DT40-SW, in which AID expression can be reversibly switched on and off using the Cre-loxP system driven by an exogenous estrogen derivative, 4-hydroxytamoxifen (4-OHT) (15–18). We have shown that Ag-specific mAbproducing clones were efficiently isolated from the DT40-SW library,

1389-1723/$ - see front matter © 2010, The Society for Biotechnology, Japan. All rights reserved. doi:10.1016/j.jbiosc.2010.03.006

352

KAJITA ET AL.

and their Ag-specificity was genetically stabilized by switching off AID expression (14). Another advantage of the DT40-SW system is that selected mAbs can be improved in their affinity for target Ags through further rounds of cell-based mutation and selection (14). Basically, this procedure mimics the affinity maturation that occurs in vivo. In the chicken, the IgV genes in Ag-stimulated B cells are diversified by gene conversion and point mutation, both of which have been shown to be AIDdependent (19). Gene conversion is a mutation mechanism that introduces partial sequences from upstream pseudo-V genes into the homologous regions of expressed IgV genes (20). In contrast, point mutation does not depend on pseudo-V genes (11). In young chickens, expansion of B cell repertoire during B cell development occurs predominantly through gene conversion in the specified organ called the bursa of Fabricius (21). On the other hand, during immune responses, Ag-stimulated B cells undergo hypermutation preferentially by point mutation in germinal centers (22). Thus it is suggested that gene conversion is effective in generating a wide variety of B cell repertoire, while point mutation that may enable to tune Agspecificity finely is more favorable in mutating the IgV genes for affinity maturation (19, 22). Gene conversion has been shown to be a dominant mutation mechanism in cultured DT40 cells (7, 8). It has been reported that Rad51 paralogues, including XRCC2 and XRCC3, which are involved in homologous recombination during DNA damage repair, are essential in gene conversion (23). Interestingly, it has been shown that disruption of one of these genes led to the

J. BIOSCI. BIOENG., transformation of the mutation pattern from gene conversion into point mutation in DT40 cells (23, 24). To explore whether point mutation is more favorable for affinity maturation of Abs in the DT40-SW system, we generated DT40-SW mutants in which the XRCC3 gene was disrupted by gene targeting hemizygously (XRCC3 (+/−)) or homozygously (XRCC3 (−/−)). Interestingly, we found that the XRCC3 (+/−) mutant that can be easily generated by a single gene targeting operation accumulated a comparable level of point mutations during culture with that observed in the XRCC3 (−/−) mutant. Here, we examine whether the XRCC3 (+/−) mutant is more useful than the wild type in obtaining highaffinity mAbs through affinity maturation in the DT40-SW system. MATERIALS AND METHODS Cell culture and transfection The DT40 cell line was obtained from RIKEN Cell Bank (Tsukuba, Japan). An engineered DT40 line, DT40-SW, whose AID expression can be switched on or off by Cre-mediated recombination, was established as reported previously (17, 18). DT40 cells were cultured in RPMI-1640 medium (MP Biomedicals, Irvine, CA, USA) supplemented with 10% fetal bovine serum (JRH Biosciences, Lenexa, Kansas, USA), 1% chicken serum (Sigma, St. Louis, MO, USA), 50 µM 2-mercaptethanol, 2 mM glutamine, 1 mM pyruvic acid, 100 µg/ml penicillin G, and 50 µg/ml streptomycin at 40 °C in 5% CO2 and 95% air. In transfection experiments, DT40-SW cells were suspended in 250 µl of phosphatebuffered saline (PBS) at 2 × 107 cells/ml and transfected with 15 µg of a linearized targeting vector by electroporation using Gene Pulser Xcell (Bio-Rad Laboratories, Hercules, CA, USA) at 550 V and 25 µF in 4 mm cuvette. The XRCC3 alleles were disrupted by pXRCC3-his and pXRCC3-bsr vectors containing the histidinol dehydrogenase (his-D)

FIG. 1. Mutation pattern in the IgVL genes of the wild type and XRCC3-knockout DT40-SW cells. (A) Hemizygous (+/−) and homozygous (−/−) disruption of the XRCC3 gene in DT40-SW were confirmed by genomic PCR. Primer pairs used are shown as arrowheads. The genomic DNA of the IgVL gene was amplified using primes CVLF-6 and CVLR-3 as a control. (B) Transcription of the XRCC3 gene in the wild-type, XRCC3 (+/−), and XRCC3 (−/−) cells. The level of the XRCC3 transcript analyzed by quantitative RT-PCR is expressed as the normalized value (ΔCt/ΔCt) for that of the control β-actin transcript. Data are indicated as means ± standard deviations from triplicate assays. (C) Comparison of mutation pattern in the IgVL genes from XRCC3 wild type (+/+), hemizygous (+/−), and homozygous (−/−) DT40-SW cells. Thin horizontal lines represent the IgVL genes with mutations. Point mutation, gene conversion tract, and deletion are indicated as lollipop shape, thick horizontal bar above line, and hollow box, respectively. An asterisk indicates a common mutation found in 6 out of 9 mutated XRCC3 (+/−) clones.

VOL. 110, 2010

EFFICIENT AB AFFINITY MATURATION IN THE CHICKEN B LINE

gene or the blasticidin S resistant (bsr) gene, respectively. XRCC3 knockout vectors, pXRCC3-his and pXRCC3-bsr, were constructed as described previously (25). Transfected DT40 cells were selected in the medium containing 50 µg/ml blasticidin S (Invitrogen, Carlsbad, CA, USA) or 1 mg/ml L-histidinol dihydrochloride (Sigma). Switching of AID expression by 4-OHT (Sigma) in DT40-SW cells was performed as described previously (17). Genomic PCR Genomic PCR was performed by using KOD plus DNA polymerase (TOYOBO, Osaka, Japan). The successful targeted disruption of the XRCC3 gene using pXRCC3-his or pXRCC3-bsr was confirmed by genomic PCR using primers HIS-1 5'-GCCGTGACCCTGCGCGTAAACGCCCTCAAGG-3' and XRCC3-R2 5'-AGTCCTCATGCTTGCAGGGTGGTGT-3' for pXRCC3-his (40 cycles of 15 s at 94 °C, 30 s at 65 °C, and 3 min at 68 °C) or BSR-3 5'-CTGTGGTGTGACATAATTGGACAAACTACCTACAGAG-3' and XRCC3-R2 for pXRCC3-bsr (40 cycles of 15 s at 94 °C, 30 s at 65 °C, and 2 min at 68 °C), respectively (25). The IgL gene was amplified using primes CVLF-6 5'-CAGGAGCTCGCGGGGCCGTCACTGATTGCCG-3' and CVLR-3 5'-GCGCAAGCTTCCCCAGCCTGCCGCCAAGTCCAAG-3' (40 cycles of 15 s at 94 °C, 30 s at 65 °C, and 1 min at 68 °C) as a control. Quantitative RT-PCR Total mRNA was isolated from DT40-SW cells using Trizol reagent (Invitrogen) and then the first strand cDNA was synthesized using the oligo-dT primer and Superscript II reverse transcriptase (Invitrogen) as described previously (26). XRCC3 mRNA levels were determined by quantitative RT-PCR using iQ SYBR Green Supermix on the iQ5 system (Bio-Rad Laboratories). PCR primers used were as follows: XRCC3-rt-f 5'-TTGACCTGAACCCGAAGGTG-3' and XRCC3-rt-r 5'-TGGCACTGGGAGGAGAAACAATCC-3' for the XRCC3 mRNA; CACT-rt1 5'-TTGTTGACAATGGCTCCGGTATGTG-3' and CACT-rt2 5'-GGGCTTCATCACCAACGTAGCTGTC-3' for the β-actin mRNA. The cycling conditions were 40 cycles of 10 s at 95 °C, 30 s at 60 °C, and 30 s at 72 °C. Results were normalized for the β-actin mRNA level. Flow cytometry and fluorescence-activated cell sorting of Ag-specific clones from the DT40-SW library The DT40-SW library was formed by maintaining ∼108 “AID-ON” DT40-SW cells in continuous culture for more than one year. To isolate 4-hydroxy-3-nitrophenylacetyl (NP)-specific clones from DT40-SW library, 107 cells from the DT40-SW library were incubated on ice for 30 min with 25 µg/ml biotinylated bovine serum albumin (BSA) conjugated with ∼15 NP moieties (bio-NP-BSA), followed by staining with phycoerythrin (PE)-conjugated anti-chicken IgM (M1-PE) (Southern Biotechnology, Birmingham, CA, USA). After washing to remove free bio-NP-BSA and M1-PE, these cells were incubated with 0.1 µg/ml of streptavidin conjugated with cychrome (cyc-SA) (BD Biosciences, Mountain View, CA, USA) in PBS(−). Cell sorting of NP-specific clones was performed with FACS Aria (BD Biosciences) (14). The NP-binding activity of surface IgM (sIgM) on DT40-SW cells was estimated as the mean fluorescence intensity (MFI) of NP-BSA-binding by flow cytometry using FACS Calibur (BD Biosciences). ELISA ELISA for chicken anti-NP or anti-BSA Abs was performed using 96-well microplates coated with 10 µg/ml NP-BSA or BSA, respectively. Anti-NP IgM mAbs bound to the plates were detected by horseradish peroxidase-conjugated goat antichicken IgM Ab (Rockland Immunochemical, Gilbertsville, PA, USA). The amount of an anti-NP IgM mAb present in each sample was also determined by using goat antichicken IgM Ab (Rockland Immunochemical)-coated microplates. Relative affinity of anti-NP mAbs secreted from isolated clones was estimated as the ratio of chicken IgM bound to NP-BSA over the amount of chicken IgM that was contained in the samples. Estimation of the dissociation constant (KD) of anti-NP mAbs To determine KD values of anti-NP IgM mAbs, culture supernatants were concentrated using Amicon Ultra-15 100 K (Millipore, Bedford, MA, USA) and then the concentration of these mAbs was determined by ELISA using 96-well microplates coated with anti-chicken IgM Abs (14). KD values of anti-NP IgM mAbs were estimated by ELISA using the serially diluted anti-NP IgM mAb (0.1 - 550nM) and NP-BSA-coated microplates. Binding of anti-NP IgM mAbs was measured with a microplate reader Model 680 (Bio-Rad Laboratories) and these data were fitted to a binding curve using GraphPad Prism (GraphPad Software, La Jolla, CA, USA). Apparent KD values of mAbs were calculated from the scatchard plots of the OD values for the binding to NP-BSA. Sequence analysis cDNA was synthesized from total RNA of DT40-SW cells as described previously (17). The rearranged IgV alleles in the light chain (IgVL) and heavy chain (IgVH) genes were amplified with PCR using KOD plus DNA polymerase. Primers used were as follows: cLL5 5'-CGGCGTGGGGATCCACAGCTGCTGGGATT-3' and CCMVCVLR 5'-GGAGCCATCGATCACCCAATCCAC-3' for the IgVL gene and cVH1F2 5'-GGCGGCTCCGTCAGCGCTCTCT-3' and cJH1 5'- GGGACTAGTACTCACCGGAGGAGACGATGACTTCGGTC-3' for the IgVH gene, respectively. The cycling conditions were 40 cycles of 15 s at 94 °C, 30 s at 60 °C, and 1 min at 68 °C. PCR products were cloned into the pCR-Blunt vector (Invitrogen), and sequenced using the BigDye1.1 terminator cycle sequencing kit and ABI PRISM 310 Genetic Analyzer (Applied Biosystems, Foster City, CA, USA). Sequence changes in the IgVL gene were assigned to gene conversion or point mutation by comparing mutated sequences with the published pseudo-Vλ gene sequences that could act as donors for gene conversion (20, 25).

RESULTS Hemizygous disruption is sufficient to increase the frequency of point mutation To transform hypermutation pattern from gene conversion to point mutation, we generated XRCC3 (+/−) and XRCC3 (−/−) mutants from DT40-SW cells by disrupting the exon 6 of the

353

XRCC3 gene using targeting vectors bearing the bsr or the his-D gene (Fig. 1A, upper panel). The successful gene targeting of the XRCC3 gene with each targeting vector was confirmed by genomic PCR (Fig. 1A, lower panel) (25). The level of the XRCC3 transcript in the XRCC3 (+/−) cells was decreased to less than 1/5 of that in the wildtype XRCC3 (+/+) cells and no transcript was observed in the XRCC3 (−/−) cells (Fig. 1B). After continuous culture of each clone for 2 months, mutation pattern in the IgVL genes was analyzed (Fig. 1C). The gene conversion tracts were predominantly observed in the wildtype DT40-SW cells. In contrast, a substantial shift of the mutation pattern to point mutation was seen in the IgVL sequences examined in

FIG. 2. Isolation of clones secreting anti-NP mAbs from the DT40-SW library by fluorescence-activated cell sorting. (A) Flow cytometric analysis of anti-NP clones in the DT40-SW library. Surface IgM (sIgM)high, NP-BSAhigh area (indicated as a rectangle) was gated (∼0.1% of total cells), and sorted into microplates at a single cell per well. (B) Anti-NP-BSA and anti-BSA Ab titers in culture supernatants of single cell-derived colonies were examined by ELISA. Five representative clones that showed relatively high anti-NP titer were selected. The culture supernatant from the DT40-SW library was used as a negative control. (C) Flow cytometric analysis of the clone 5-27 for NP binding. Cells were stained with bio-NP-BSA and cyc-SA. Cells from the unselected DT40-SW library were used as a negative control.

354

KAJITA ET AL.

the XRCC3 (−/−) mutant. Interestingly, the XRCC3 (+/−) mutant accumulated a comparable level of point mutations to that observed in the XRCC3 (−/−) mutant. A conserved point mutation was found in the CDR2 of 6 out of 9 mutated IgVL genes from XRCC3 (+/−) cells, but it is possible that the mutation occurred within a few cell cycles after single cell sorting, resulting in a high proportion of clones with the same mutation. No mutations were observed in the wild-type and XRCC3 knockout cells when AID expression was off, indicating these mutations are dependent on the activity of AID (data not shown). Although the XRCC3 (−/−) mutant grew more slowly and exhibited a greater loss of surface IgM during culture than the wild type counterpart, the XRCC3 (+/−) mutant showed no abnormality in these respects (data not shown). In addition, the XRCC3 (+/−) mutant can be easily generated by a single gene targeting operation. Thus, we examined whether the XRCC3 (+/−) mutant is more useful

J. BIOSCI. BIOENG., than the wild-type cells in obtaining high-affinity Abs through affinity maturation. Isolation of NP-specific clones by fluorescence-activated cell sorting In this study, we attempted to analyze affinity maturation of a mAb against NP, a hapten widely used in analyzing immune responses. First, we isolated NP-reactive clones from ∼107 DT40-SW library cells by fluorescence-activated cell sorting, and the sorted cells were seeded into each well of a 96-well microplate at a single cell per well (Fig. 2A). After culture for 7–10 days, colonies appeared in 40 wells. Anti-NP and anti-BSA Ab titers in each supernatant from colony-positive wells were assayed by ELISA using NP-BSA- or BSAcoated plates (Fig. 2B). Of the 40 isolated clones, 5 were found to produce mAbs that were reactive with NP-BSA, but not BSA. We selected a clone, 5-27 that showed highest anti-NP titer and stably secreted the mAb. The NP-binding of the surface IgM (sIgM) in 5-27

FIG. 3. Affinity maturation of the anti-NP mAb in the clone 5-27 whose XRCC3 gene was disrupted hemizygously. (A) Quantitative RT-PCR analysis of XRCC3 expression in XRCC3 (+/+) or XRCC3 (+/−) cells of the clone 5-27. The level of the XRCC3 transcript is expressed as the normalized value (ΔCt/ΔCt) for that of the control β-actin transcript. Data are indicated as means ± standard deviations from triplicate assays. (B) Scheme for screening high-affinity anti-NP clones from the secondary library. After continuous culture of XRCC3 (+/+) and XRCC3 (+/−) clones for 6 weeks, single cells were sorted from the gated population in the same fashion as described in Fig. 2. (C) Relative affinity of improved anti-NP mAbs secreted from 5-27-derived clones. Anti-NP mAbs in culture supernatants of affinity-maturated clones were examined in terms of binding avidity to NP-BSA by ELISA. The relative affinity of each anti-NP mAb was expressed as the OD value for the binding of each sample to NP-BSA that was divided by the level of IgM Ab in the culture supernatant. The mean relative affinity of maturated clones and the value of the original 5-27 are indicated as solid and broken horizontal bar, respectively. The clone showing the highest relative affinity was designated as 5-27-1 and indicated as closed circle in the panel. Statistical significance was examined by two-tailed unpaired t test. *p b 0.001.

VOL. 110, 2010 cells was confirmed by flow cytometry (Fig. 2C). The anti-NP clone 5-27 was used in the following experiments. Effect of XRCC3 disruption on affinity maturation of anti-NP mAb Next, we generated the XRCC3 (+/−) mutant from the antiNP clone 5-27 and tested whether the hemizygous disruption is effective to achieve affinity maturation of the anti-NP mAb more efficiently. The expression level of the XRCC3 gene in the XRCC3 (+/−) 5-27 cells was decreased to ∼ 1/10 of that in the wild-type XRCC3 (+/+) 5-27 cells (Fig. 3A). The strategy for isolating higheraffinity clones from the 5-27 library is shown schematically in Fig. 3B. The secondary cell library derived from each of XRCC3 (+/+) and XRCC3 (+/−) clones was generated by maintaining ∼ 107 “AID-ON” cells of each clone in culture for 6 weeks. The cells that appeared to show higher affinity for NP were gated by the same region in each library, and sorted into a 96-well microplate at a single cell per well by using the fluorescence-activated cell sorter (Fig. 3B). The percentage of the gated region for XRCC3 (+/−) cells (0.11%) was not different from that for XRCC3 (+/+) cells, but the mean fluorescence intensity of the gated region for XRCC3 (+/−) cells (615) was higher that that for XRCC3 (+/+) cells (245), suggesting that the XRCC3 (+/−) library contained clones with higher NP binding. After culture for 7–10 days, 47 and 46 colonies developed from the XRCC3 (+/+) and the XRCC3 (+/−) libraries, respectively. Relative affinity of anti-NP

EFFICIENT AB AFFINITY MATURATION IN THE CHICKEN B LINE

355

mAbs secreted from the grown colonies was assessed by ELISA (Fig. 3C). Interestingly, results indicated that clones selected from the XRCC3 (+/−) library exhibited higher affinity than those from the wild-type library. These results were also confirmed by flow cytometric analysis of NP binding to surface IgM on the selected clones (data not shown). Thus, the XRCC3 (+/−) library was more excellent in generating high-affinity mAbs than the XRCC3 (+/+) library. We assessed binding avidity of the anti-NP mAb produced by a clone, 5-27-1 that was selected from the XRCC3 (+/−) library and considered to show the highest relative affinity in comparison with that of the original 5-27 clone (Fig. 4). While an apparent KD value of the original 5-27 clone was 185 nM, that of the improved mAb secreted from the clone 5-27-1 was estimated to be 0.3 nM (Fig. 4). These results demonstrated that the XRCC3 (+/−) mutant is more useful in generating high-affinity Abs through affinity maturation. Mutation analysis of improved anti-NP mAbs Finally, we analyzed the nucleotide sequences of the IgV regions of the top 12 clones with higher relative affinity that were selected from the XRCC3 (+/−) library, and compared with the original sequence of the clone 5-27. In the IgVL gene, a common point mutation (C to G) in the frame work region was found in 9 of higher-affinity clones (Fig. 5A). Sequence alterations in the IgVH region were also detected. Common nucleotide substitutions were identified in the complementarity determining region (CDR) 2 and CDR3 in the IgVH gene (Fig. 5B). In addition, common mutations also accumulated throughout the IgVH sequences other than CDRs. It should be noted that these conserved mutations resulted in replacement of coding amino acids. Thus, it is suggested that affinity maturation, which is the result of random mutation followed by selection, proceeded successfully in the culture of the XRCC3 (+/−) mutant. DISCUSSION

FIG. 4. Estimation of apparent KD values of anti-NP mAbs. Serially diluted anti-NP mAb from the original 5-27 (A) or affinity-maturated 5-27-1 clone (B) was examined for binding to NP-BSA by ELISA. KD values were calculated as described in Materials and methods.

In the development of therapeutic mAbs, primarily screened mAbs are often manipulated genetically to improve or optimize their biological activities. To improve affinity of mAbs, mimicking affinity maturation that occurs in vivo during immune responses is considered to be one of the basic technologies for generating practically useful mAbs. In this context, the use of DT40-SW, a hypermutating B cell line, is advantageous in obtaining high-affinity mAbs through an in vitro affinity maturation that mimics the in vivo system. Here, we demonstrated that switching of the mutation pattern from gene conversion to point mutation by hemizygous disruption of the XRCC3 gene was effective to achieve affinity maturation efficiently. In the chicken immune system, gene conversion that occurs in the specified organ called the bursa of Fabricius is known to generate a wide variety of B cell repertoire during B cell development (19). In the gene conversion, sequences (8 to around 200 bp) derived from upstream pseudo-V genes are copied into homologous regions within the IgV gene. Gene conversion is also responsible for the diversification of Agstimulated B cells during the early phase of Ab responses. On the other hand, it has been shown that point mutation contributes to affinity maturation of Abs that occurs in germinal centers in the later phase (19, 22). Point mutation that can result in the replacement of a single amino acid in each event is considered to be favorable to tune Agspecificity of Abs more finely. Our present results suggest that gene conversion is effective in constructing a primary Ab library, and in contrast, point mutation is more suitable for improving the affinity of selected Ag-specific clones. Disruption of the XRCC3 gene led to the transformation of the mutation pattern from gene conversion to point mutation in DT40SW cells. Similar transformation of the mutation pattern in DT40 cells has been reported to occur by homozygous deletion of the BRCA2 gene or one of Rad51 paralogue genes including the XRCC2 and

356

KAJITA ET AL.

J. BIOSCI. BIOENG.,

VOL. 110, 2010 RAD51B genes (23, 24). These molecules have been shown to be critical for homologous DNA recombination (reviewed in (27, 28)). Gene conversion can be attributed to a recombination-dependent DNA damage repair where AID-dependent DNA breaks in the IgV gene are fixed by repairing the damaged regions using pseudo-V genes as templates (19, 27). On the other hand, gene conversion and point mutation have been shown to occur through a common intermediate (11). Thus, defect of homologous recombination by suppressing the expression of essential genes such as the XRCC3 gene leads to a decrease in gene conversion and a reciprocal increase in nontemplated point mutation in DT40 cells. It has been reported that homozygous disruption of the XRCC2 or XRCC3 gene resulted in a decreased growth rate (24, 29) and an increased generation of sIgM-deficient population (23). Because XRCC2 and XRCC3 molecules are essential for a DNA repair mechanism mediated by homologous recombination, the complete loss of these expression would lead to several layers of genomic instability that affect cell cycle progression. In addition, point mutation could cause nonsense and frame-shift mutations in the IgV gene, leading to the loss of Ig expression when the repair of these destructive mutations is severely impaired in the XRCC3 (−/−). Thus, the abnormalities of the XRCC3 (−/−) cells could hinder efficient generation of the cell library and specific selection of Ag-binding clones from the library. In this study, we found, however, hemizygous disruption of the XRCC3 gene was sufficient to increase the frequency of point mutation without affecting the growth rate and inducing sIgM loss. The XRCC3 (+/−) mutants expressed the XRCC3 gene at less than 1/5 of the level of the wild type (Figs. 1B and 3A). This might reflect a positive feedback regulatory mechanism of the XRCC3 expression. Although point mutation frequency increased in the XRCC3 (+/−) mutants to a comparable level to that in the wild type, gene conversion was not totally lost in these mutants, suggesting that a decreased level of XRCC3 could support functions of DNA repair and gene conversion in the XRCC3 (+/−) cells. The remaining activities of DNA repair and gene conversion might contribute to maintenance of genomic integrity and to restoration of the loss of sIgM that is often caused by point mutation, respectively. Thus, hemizygous knockout of the XRCC3 gene is advantageous in enhancing the frequency of point mutation without affecting the growth rate and sIgM expression. In the present experiments, we isolated Ag-specific clones from the DT40-SW library using the fluorescence-activated cell sorter. Flow cytometry has been shown to be advantageous for selecting singlechain Fv Abs from yeast display libraries in terms of Ag specificity and affinity (30–32). We have previously shown that panning with Agconjugated magnetic beads is useful in screening Ag-specific clones from the DT40-SW library (14). Here, we show that fluorescenceactivated cell sorting was excellent in selecting Ag-specific and highaffinity clones from the library on the basis of differential binding affinity during affinity maturation. We have recently generated a novel DT40 line, DT40-SW-hXR, in which XRCC3 expression can be conditionally shut down using the tetracycline-repressible promoter system (25). The hypermutation pattern in DT-40-SW-hXR cells can be clearly transformed from gene conversion into point mutation when the cells are cultured in the presence of doxycycline. The cell line will be also a useful tool for promoting affinity maturation of mAbs by altering the mutation pattern when it is necessary. Taken together, manipulation of mutation pattern during Ab generation will be effective for obtaining practically valuable mAbs in the DT40-SW system.

EFFICIENT AB AFFINITY MATURATION IN THE CHICKEN B LINE

357

ACKNOWLEDGMENTS This work was partially supported in part by Grants from The Ministry of Education, Culture, Sports, Science and Technology of Japan, New Energy and Industrial Technology Organization (NEDO) of Japan, and Japan Livestock Technology Organization. References 1. McCafferty, J., Griffiths, A. D., Winter, G., and Chiswell, D. J.: Phage antibodies: filamentous phage displaying antibody variable domains, Nature, 348, 552–554 (1990). 2. Winter, G., Griffiths, A. D., Hawkins, R. E., and Hoogenboom, H. R.: Making antibodies by phage display technology, Annu. Rev. Immunol., 12, 433–455 (1994). 3. Griffiths, A. D. and Duncan, A. R.: Strategies for selection of antibodies by phage display, Curr. Opin. Biotechnol., 9, 102–108 (1998). 4. Rajewsky, K.: Clonal selection and learning in the antibody system, Nature, 381, 751–758 (1996). 5. Wagner, S. D. and Neuberger, M. S.: Somatic hypermutation of immunoglobulin genes, Annu. Rev. Immunol., 14, 441–457 (1996). 6. Kanayama, N., Kimoto, T., Todo, K., Nishikawa, Y., Hikida, M., Magari, M., Cascalho, M., and Ohmori, H.: B cell selection and affinity maturation during an antibody response in the mouse with limited B cell diversity, J. Immunol., 169, 6865–6874 (2002). 7. Buerstedde, J. M., Reynaud, C. A., Humphries, E. H., Olson, W., Ewert, D. L., and Weill, J. C.: Light chain gene conversion continues at high rate in an ALV-induced cell line, EMBO J., 9, 921–927 (1990). 8. Kim, S., Humphries, E. H., Tjoelker, L., Carlson, L., and Thompson, C. B.: Ongoing diversification of the rearranged immunoglobulin light-chain gene in a bursal lymphoma cell line, Mol. Cell. Biol., 10, 3224–3231 (1990). 9. Arakawa, H., Hauschild, J., and Buerstedde, J. M.: Requirement of the activationinduced deaminase (AID) gene for immunoglobulin gene conversion, Science, 295, 1301–1306 (2002). 10. Harris, R. S., Sale, J. E., Petersen-Mahrt, S. K., and Neuberger, M. S.: AID is essential for immunoglobulin V gene conversion in a cultured B cell line, Curr. Biol., 12, 435–438 (2002). 11. Arakawa, H., Saribasak, H., and Buerstedde, J. M.: Activation-induced cytidine deaminase initiates immunoglobulin gene conversion and hypermutation by a common intermediate, PLoS Biol., 2, E179 (2004). 12. Cumbers, S. J., Williams, G. T., Davies, S. L., Grenfell, R. L., Takeda, S., Batista, F. D., Sale, J. E., and Neuberger, M. S.: Generation and iterative affinity maturation of antibodies in vitro using hypermutating B-cell lines, Nat. Biotechnol., 20, 1129–1134 (2002). 13. Seo, H., Masuoka, M., Murofushi, H., Takeda, S., Shibata, T., and Ohta, K.: Rapid generation of specific antibodies by enhanced homologous recombination, Nat. Biotechnol., 23, 731–735 (2005). 14. Todo, K., Miyake, K., Magari, M., Kanayama, N., and Ohmori, H.: Novel in vitro screening system for monoclonal antibodies using hypermutating chicken B cell library, J. Biosci. Bioeng., 102, 478–481 (2006). 15. Zhang, Y., Riesterer, C., Ayrall, A. M., Sablitzky, F., Littlewood, T. D., and Reth, M.: Inducible site-directed recombination in mouse embryonic stem cells, Nucleic Acids Res., 24, 543–548 (1996). 16. Lam, K. P. and Rajewsky, K.: Rapid elimination of mature autoreactive B cells demonstrated by Cre-induced change in B cell antigen receptor specificity in vivo, Proc. Natl. Acad. Sci. U. S. A., 95, 13171–13175 (1998). 17. Kanayama, N., Todo, K., Reth, M., and Ohmori, H.: Reversible switching of immunoglobulin hypermutation machinery in a chicken B cell line, Biochem. Biophys. Res. Commun., 327, 70–75 (2005). 18. Kanayama, N., Todo, K., Takahashi, S., Magari, M., and Ohmori, H.: Genetic manipulation of an exogenous non-immunoglobulin protein by gene conversion machinery in a chicken B cell line, Nucleic Acids Res., 34, e10 (2006). 19. Arakawa, H. and Buerstedde, J. M.: Immunoglobulin gene conversion: insights from bursal B cells and the DT40 cell line, Dev. Dyn., 229, 458–464 (2004). 20. Reynaud, C. A., Anquez, V., Grimal, H., and Weill, J. C.: A hyperconversion mechanism generates the chicken light chain preimmune repertoire, Cell, 48, 379–388 (1987). 21. Ratcliffe, M. J.: Antibodies, immunoglobulin genes and the bursa of Fabricius in chicken B cell development, Dev. Comp. Immunol., 30, 101–118 (2006). 22. Arakawa, H., Kuma, K., Yasuda, M., Furusawa, S., Ekino, S., and Yamagishi, H.: Oligoclonal development of B cells bearing discrete Ig chains in chicken single germinal centers, J. Immunol., 160, 4232–4241 (1998). 23. Sale, J. E., Calandrini, D. M., Takata, M., Takeda, S., and Neuberger, M. S.: Ablation of XRCC2/3 transforms immunoglobulin V gene conversion into somatic hypermutation, Nature, 412, 921–926 (2001).

FIG. 5. Mutation analysis of the IgVL gene (A) and the IgVH gene (B) of affinity-maturated anti-NP mAbs. Twelve 5-27-derived clones with higher affinity for NP that were obtained from the XRCC3 (+/−) library were analyzed for the nucleotide sequences, and compared with the original 5-27 clone. Sequences are arranged in descending order of the relative affinity shown in Fig. 3C. Hyphens indicate nucleotides identical with the unmutated 5-27.

358

KAJITA ET AL.

24. Hatanaka, A., Yamazoe, M., Sale, J. E., Takata, M., Yamamoto, K., Kitao, H., Sonoda, E., Kikuchi, K., Yonetani, Y., and Takeda, S.: Similar effects of Brca2 truncation and Rad51 paralog deficiency on immunoglobulin V gene diversification in DT40 cells support an early role for Rad51 paralogs in homologous recombination, Mol. Cell. Biol., 25, 1124–1134 (2005). 25 Kajita, M., Magari, M., Todo, K., Kanayama, N., Kanayama, N., and Ohmori, H.: Conditional transformation of immunoglobulin gene conversion into point mutation by controlling XRCC3 expression in the DT40 B cell line, J. Biosci. Bioeng., 109, 407–410 (2010). 26. Magari, M., Aya, T., Ikeda, M., Todo, K., Kanayama, N., and Ohmori, H.: Enhancement of antibody production from a chicken B cell line DT40 by reducing Pax5 expression, J. Biosci. Bioeng., 107, 206–209 (2009). 27. Sale, J. E.: Immunoglobulin diversification in DT40: a model for vertebrate DNA damage tolerance, DNA Repair (Amst), 3, 693–702 (2004).

J. BIOSCI. BIOENG., 28. Sonoda, E., Takata, M., Yamashita, Y. M., Morrison, C., and Takeda, S.: Homologous DNA recombination in vertebrate cells, Proc. Natl. Acad. Sci. U. S. A., 98, 8388–8394 (2001). 29. Takata, M., Sasaki, M. S., Tachiiri, S., Fukushima, T., Sonoda, E., Schild, D., Thompson, L. H., and Takeda, S.: Chromosome instability and defective recombinational repair in knockout mutants of the five Rad51 paralogs, Mol. Cell. Biol., 21, 2858–2866 (2001). 30. Boder, E. T. and Wittrup, K. D.: Yeast surface display for screening combinatorial polypeptide libraries, Nat. Biotechnol., 15, 553–557 (1997). 31. VanAntwerp, J. J. and Wittrup, K. D.: Fine affinity discrimination by yeast surface display and flow cytometry, Biotechnol. Prog., 16, 31–37 (2000). 32. Feldhaus, M. J., Siegel, R. W., Opresko, L. K., Coleman, J. R., Feldhaus, J. M., Yeung, Y. A., Cochran, J. R., Heinzelman, P., Colby, D., and Swers, J., et al.: Flowcytometric isolation of human antibodies from a nonimmune Saccharomyces cerevisiae surface display library, Nat. Biotechnol., 21, 163–170 (2003).