The Journal of Immunology
Molecular Dissection of Antibody Responses against Pneumococcal Surface Protein A: Evidence for Diverse DH-Less Heavy Chain Gene Usage and Avidity Maturation1,2 Soma Rohatgi, Debjani Dutta, Suhail Tahir,3 and Devinder Sehgal4 Immunization of human volunteers with a single dose of pneumococcal surface protein A (PspA) stimulates broad cross-reactive Abs to heterologous PspA molecules that, when transferred, protect mice from fatal infection with Streptococcus pneumoniae. In this study, we report the molecular characterization of 36 mouse mAbs generated against the extracellular domain of PspA (PspA3–286) from strain R36A. Abs to PspA3–286 were encoded by diverse VH and V families/genes. The H chain CDR3 and L chain CDR3 lengths were 3–13 (7.8 ⴞ 0.5) and 8 –9 (8.7 ⴞ 0.2) codons, respectively. Unexpectedly, seven hybridomas expressed H chains that lack DH gene-derived amino acids. Nontemplate-encoded addition(s) were observed in the H chain expressed in six of these seven hybridomas; Palindromic addition(s) were absent. Absence of DH gene-derived amino acids did not prevent antiPspA3–286 mAbs from attaining average relative avidity. Avidity maturation occurred during primary IgG anti-PspA3–286 polyclonal Ab response in PspA3–286- and R36A-immunized mice. Compared with PspA3–286-immunized mice, the relative avidity of the primary polyclonal IgG Abs was higher in R36A immunized mice on days 72, 86, and 100. Two pairs of clonally related hybridomas were observed. DH genes expressed in the majority (75.9%) of the hybridomas used reading frame 3. Analysis of replacement/silent mutation ratio in the CDR and framework regions provided evidence for Ag-driven selection in 11 mAbs. Based on epitope localization experiments, the mAbs were classified into 12 independent groups. ELISA additivity assay indicated that members within a group recognized topographically related epitopes. This study provides molecular insights into the biology of DH-less Abs. The Journal of Immunology, 2009, 182: 5570 –5585.
treptococcus pneumoniae, the etiological agent for pneumonia, bacteremia, meningitis, and otitis media, is a major cause of morbidity and mortality worldwide. It is estimated that 1.6 million people die of pneumococcal disease every year, 40 – 60% of which are children ⬍5 years of age (1). Emergence of multidrug-resistant clinical isolates, high cost, limited coverage of the current pneumococcal polysaccharide-based vaccines, and recent reports of strain replacement have highlighted the urgent need for new alternative and cost-effective strategies. Several surface proteins have been investigated as potential vaccine candidates against pneumococcal infections. These proteins include pneumococcal surface protein A (PspA),5 pneumolysin, PsaA, PspC, PppA, neuraminidase, autolysin, PiuA, PiaA, PcsB, and StkP,
National Institute of Immunology, Aruna Asaf Ali Marg, New Delhi, India Received for publication October 6, 2008. Accepted for publication February 12, 2009. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1
This work was supported by a grant from the Department of Biotechnology, Government of India (to D.S.) and a Senior Research Fellowship from the Council of Scientific and Industrial Research, India (to S.R.).
The sequences presented in this article have been submitted to GenBank under accession numbers EU915613-EU915684.
Current address: Graduate School of Medicine, Kyoto University, Kyoto, Japan.
Address correspondence and reprint requests to Dr. Devinder Sehgal, Molecular Immunology Laboratory, National Institute of Immunology, Aruna Asaf Ali Marg, New Delhi 110 067, India. E-mail address: [email protected]
Abbreviations used in this paper: PspA, pneumococcal surface protein A; AI, additivity index; FP, founder progenitor; FR, framework region; HCDR, H chain CDR; HP, hypothetical common precursor; LCDR, L chain CDR; N, nontemplate encoded; P, palindromic; RF, reading frame. Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00
among others (2– 4). PspA has been most extensively studied for its ability to elicit protective immunity in animal models (5–10). PspA has been shown to elicit high Ab levels in humans (11), and human anti-PspA serum Abs can protect mice against challenge with virulent pneumococcal strains when transferred (12). Recently, vaccine-induced human Abs to PspA has been shown to augment complement component C3 deposition on S. pneumoniae (13). PspA is a polymorphic, cell surface, choline-binding protein. PspA interferes with fixation of C3 on the pneumococcal surface (14), and its lactoferrin-binding activity is believed to protect pneumococci from bactericidal activity of apolactoferrin (15). PspA, a virulence factor, is present in all strains of pneumococci and is serologically variable, cross-reactive, and cross-protective (6, 11, 12). PspA proteins from different strains have a similar basic molecular structure. PspA consists of four distinct structural domains: an N-terminal ␣-helical domain that is surface exposed, a proline-rich region, a stretch of highly conserved 20-aa repeats (choline-binding domain), and a slightly hydrophobic sequence at the C terminus (16, 17). Based on the amino acid sequence of C-terminal 100 aa of the surface exposed domain, PspAs have been classified into six clades and three cross-reacting families (⬎55% identity) (18). Family 1 is comprised of clade 1 and 2; family 2 covers clade 3, 4 and 5; and family 3 has only clade 6. Families 1 and 2 account for the majority (94 –99%) of the pneumococcal isolates (18 –20). Anti-PspA mAbs have been shown to be protective, and the protective epitopes have been mapped to the ␣-helical domain of PspA (21–23). However, the Ab response has not been analyzed at the molecular level. PspA is a leading vaccine candidate and therefore it is important to characterize the V genes and their structural characteristics, which are used in response to S. pneumoniae.
The Journal of Immunology This report presents a comprehensive analysis of the molecular characteristics of the Abs induced in response to PspA from strain R36A. The molecular nature of the murine Ab response to PspA was examined using a panel of well-defined anti-PspA3–286 mAbs. Abs to PspA3–286 were encoded by diverse V genes, with at least eight VH and 14 V families represented in these Abs. Remarkably, seven of the anti-PspA3–286 mAbs expressed a H chain that lacked DH gene-derived amino acids. Nontemplate-encoded (N) but not palindromic (P) nucleotide addition(s) were observed in these seven hybridomas. The absence of DH gene-derived amino acids did not prevent anti-PspA3–286 mAbs from attaining a meaningful or average relative avidity. To best of our knowledge, our study is the first experimental demonstration of avidity maturation for a human, bacterial pathogen-derived, highly protective protein Ag in mice immunized with whole heat-killed bacteria. We demonstrate that the relative avidity of the primary IgG anti-PspA3–286 polyclonal Ab elicited by immunization with heat-killed R36A is higher as compared with recombinant PspA3–286. Our data provide evidence for clonal expansion and Ag-driven selection. We show that the anti-PspA3–286 Ab response recognizes at least 12 epitopes. Furthermore, the study provides evidence that B cells bearing surface Ig devoid of DH gene-derived amino acids are present in normal mice, presumably at very low frequency, and can be recruited in significant numbers in an anti-PspA3–286 Ab response.
Materials and Methods PCR amplification and molecular cloning of PspA3–286 The extracellular (surface exposed) domain of PspA from the S. pneumoniae strain R36A corresponds to aa residues 1–288 of the mature peptide. Genomic DNA from R36A was isolated using a commercially available kit following the manufacturer’s instructions (Qiagen). The subfragment of PspA corresponding to aa residues 3–286 of mature peptide (referred to as PspA3–286) was PCR amplified using genomic DNA from R36A and cloned Pfu DNA polymerase (Stratagene). PspA3–286 was amplified using hot start touchdown PCR and 5⬘-TCTCCCGTAGCCAGT CAGTCTAA-3⬘ and 5⬘-CCCCCAAGCTTCTAAACTGCTTTCTTA AGGTCAGCTTC A-3⬘ as sense and antisense primers, respectively. A HindIII site (to facilitate cloning; underlined) and a stop codon were incorporated in the antisense primer. The 50-l reaction mix was comprised of 1⫻ cloned Pfu DNA polymerase buffer, 200 M dNTP mix, 0.2 M each primer, 50 ng of R36A genomic DNA, and 2.5 U of cloned Pfu DNA polymerase. A touchdown PCR was performed using a programmable thermal Cycler (GeneAmp PCR system 2700; Applied Biosystems). The touchdown part of the PCR program started with an initial denaturation step at 95°C for 2 min. This was followed by denaturation at 94°C for 1 min, annealing at 62°C for 1 min, and an extension at 72°C for 1 min. The annealing temperature was decreased from 62°C to 52°C over five cycles at the rate of 2°C per cycle. This was followed by 25 cycles consisting of denaturation at 94°C for 1 min, annealing at 52°C for 1 min, and an extension at 72°C for 1 min. A final extension step was conducted at 72°C for 10 min. The HindII-digested PCR product was ligated into StuI-HindIII digested Escherichia coli expression vector pQE-30 Xa using T4 DNA ligase. The ligated product was transformed in E. coli strain XL1-Blue. Recombinants were identified by restriction analysis and confirmed by sequencing.
Expression and purification of recombinant PspA3–286 For expression purposes the construct was transformed in E. coli expression strain SG13009 (Qiagen). Bacterial cultures were grown in LuriaBertani medium containing ampicillin (100 g/ml) and kanamycin (25 g/ml). Expression of recombinant protein was induced by adding 1 mM isopropyl-␤-D-thiogalactoside to a mid-logarithmic phase culture (A600 ⬃ 0.6) for 2 h. Bacteria were harvested by centrifugation at 4000 ⫻ g for 30 min and resuspended in lysis buffer. Cell lysates were prepared by 10 sonication cycles of 20-s duration with a 20-s time interval between pulses. The lysate was centrifuged at 20,000 ⫻ g for 30 min at 4°C and the recombinant protein was purified from the supernatant using Ni-NTA affinity chromatography following the manufacturer’s instructions (Qiagen). The vector-encoded amino acids, including the N-terminal histidine affinity tag, were cleaved using factor Xa protease following the manufacturer’s in-
5571 structions (Qiagen). The removal of the tag was confirmed by probing Western blots with anti-PspA3–286 polyclonal sera and anti-His tag mAbs. Purified PspA3–286 (devoid of the tag) was used for immunization.
Preparation of heat-killed R36A R36A, a nonencapsulated variant of virulent S. pneumoniae capsular type 2 strain D39, was maintained on tryptic soy agar plates supplemented with 5% sheep blood as described by Wu et al. (24). For liquid cultures, pneumococci were grown in Todd-Hewitt broth supplemented with 0.5% yeast extract (THY) to mid-logarithmic phase and stored at ⫺70°C. One or two characteristic colonies were suspended in 200 ml of THY and the culture was incubated at 37°C with 5% CO2 for 4 – 6 h until it reached mid-logarithmic phase (A650 ⬃ 0.4 – 0.6). An aliquot was kept aside for determining the viable cell count by plating. The remaining culture was heat killed by incubating it at 56°C in a water bath for 10 h (1 h/20 ml of culture). Sterility was confirmed by subculturing and plating. The pneumococcal stock was aliquoted and stored at ⫺70°C until required. Using an ELISA-based assay, 109 CFU of heat-killed R36A was estimated to contain the equivalent of 45 g of PspA3–286.
Mice BALB/c and CBA/J inbred strains of mice were obtained from The Jackson Laboratory and were bred and maintained in the Small Animal Facility of the National Institute of Immunology (Aruna Asaf Ali Marg, New Delhi, India). Mice used for experiments were 6 – 8 wk old. Approval from the Institutional Animal Ethics Committee was obtained for all experimental procedures involving animals.
Generation of anti-PspA3–286 mAbs Primary (three independent fusions) and tertiary (one fusion) antiPspA3–286 mAbs were generated using splenocytes from mice immunized with recombinant PspA3–286- and heat-killed R36A. Similarly, primary (one fusion) and tertiary (two independent fusions) anti-PspA3–286 mAbs were generated from mice immunized with heat-killed R36A. All antiPspA3–286 mAbs were raised in BALB/c mice except tertiary antiPspA3–286 mAbs generated from PspA3–286 immunized mice, which were raised in the CBA/J strain. Anti-PspA3–286 primary mAbs were generated from female BALB/c mice immunized i. p. with the following: 1) 25 g of recombinant PspA3–286 in 50 g of alum; and 2) 109 CFU of heat-killed R36A in PBS. Primary mAbs were generated from splenocytes recovered on day 18 postimmunization, using a standard protocol (25). Anti-PspA3–286 tertiary mAbs were generated from the following: 1) female CBA/J mice immunized i. p. with 25 g of recombinant PspA3–286 in 50 g of alum; and 2) female BALB/c mice immunized i. p. with 109 CFU of heat-killed R36A in PBS. For generating the tertiary mAbs, mice were immunized on days 0, 14, and 28, and splenocytes were obtained 4 days after the last immunization. All culture supernatants were screened for Ab reactivity to PspA3–286 using ELISA. The isotypes of the H and L chains of the antiPspA3–286 mAbs were determined by ELISA using a commercially available isotyping kit (BD Biosciences).
Amplification of the expressed Ig H and L chain genes by RT-PCR Total RNA was extracted from 107 hybridoma cells using a commercially available RNA isolation kit following the manufacturer’s instructions (Qiagen). The purified RNA was quantitated spectrophotometrically. We opted for a nested RT-PCR-based strategy for amplifying the expressed Ig H and L chain genes. Isotype-specific CH and CL antisense primers, along with VH and VL sense primers, were designed in our laboratory (26). The VH and VL external primers bind in the leader or in framework region (FR) 1, whereas the VH and VL internal primers bind in FR1. The designed primer set covers all possible VH and VL genes, families, and isotypes. The first strand cDNA for the H and L chains was synthesized using a commercially available RT-PCR kit (Applied Biosystems). The 20-l reverse transcription reaction mixture consisted of 1⫻ PCR buffer II, 0.75 M antisense RT primer, 10 mM dNTP mix, 5 mM MgCl2, 1 U of RNase inhibitor, 50 U of murine leukemia virus reverse transcriptase (Applied Biosystems), and 10 g of total RNA. The reaction was performed at 42°C for 1 h. Reverse transcriptase was inactivated by incubation at 99°C for 5 min. The 50-l first-round PCR mix consisted of 1⫻ PfuUltra hot start buffer (Stratagene), 200 M dNTP, 0.25 M of each external sense and antisense primer, 5 l of cDNA, and 2.5 U of PfuUltra hot start DNA polymerase (Stratagene). Touchdown PCR was set up on a GeneAmp 2700 PCR system. The PCR cycling parameters used were the same as those described
MOLECULAR CHARACTERIZATION OF ANTIBODY RESPONSE TO PspA
Table I. Primers used for PCR amplification and expression cloning of PspA deletion mutantsa Primers DS-411 DS-412 DS-413 DS-414 DS-415 DS-416
(sense) (antisense) (sense) (antisense) (sense) (antisense)
Sequence of Primer (5⬘ to 3⬘) CCCCGGATCCTCTCCCGTAGCCAGTCAG CCCCCAAGCTTCTATATCATCTTATCTGCTGCGTC CCCCGGATCCGATGAAGCTAAGAAACGCGA CCCCCAAGCTTCTAGAGCTCTTGTTCTAGTCTAT CCCCGGATCCAAAGAGATTGATGAGTCTGAA CCCCCAAGCTTCTAAACTGCTTTCTTAAGGTCAG
a To facilitate directional cloning, BamH1 (G/GATCC) and HindIII (A/AGCTT) sites (underlined) were incorporated in the sense and antisense primers, respectively. A stop codon (TAG, boldface) was incorporated in the antisense primers to prevent the addition of a restriction site and vector-encoded amino acids at the C terminus of the recombinant protein.
for PspA3–286. The composition of the reaction mixture and the cycling parameters for the second-round PCR were the same as those used for the first-round PCR, except that 2 l of the first-round PCR product served as template for the second-round PCR instead of cDNA.
Sequencing of the expressed VH and VL genes The VH and VL genes expressed in the anti-PspA3–286 hybridomas were RT-PCR amplified, purified from agarose gel using a DNA gel extraction kit (Stratagene), and sequenced directly using a C region antisense internal PCR primer. Alternatively, the PCR products were ligated into pPCRScript Amp SK(⫹) cloning vector (Stratagene) and transformed into E. coli strain XL1-Blue. Recombinants were identified by restriction analysis and sequenced using primers that bind in the vector.
Sequence analysis of Ig sequences The nucleotide sequence obtained was analyzed using Sequencher (version 4.8; Gene Codes) and MacVector (with Assembler) software (version 9.5.1; MacVector). Sequence comparison and Ig gene/family assignment for the expressed H and L chain genes was done using the IMGT (ImMunoGeneTics; imgt.cines.fr/) (27) and IgBLAST (www.ncbi.nlm.nih. gov/igblast/) germline databases. Throughout this study, the FR and CDR assignments and codon numbering were according to the Kabat system (28). The DH genes were identified as described below. After the VH and JH genes were assigned, the intervening sequence was aligned to the germline DH gene database to identify the DH gene used in the H chain expressed in the anti-PspA3–286 hybridoma. A minimum five-nucleotide match with the germline DH gene was required for the DH gene assignment. In cases where more than one germline DH gene matched the query sequence, the germline DH gene with the longest stretch of sequence identity was assigned. In cases where two or more germline DH genes showed sequence identities of equal length, it was presumed that they are equally likely to be used.
Generation of deletion mutants of PspA3–286 and localization of B cell epitopes by dot blot analysis To localize the epitopes recognized by the anti-PspA3–286 mAbs, a series of overlapping and nonoverlapping deletion mutants of PspA3–286 were generated. The deletion mutants generated corresponded to aa residues 3–97 (PspA3–97), 98 –192 (PspA98 –192), 193–286 (PspA193–286), 3–192 (PspA3–192), and 98- 286 (PspA98 –286) of the mature PspA peptide. The PCR primers used for generating the deletion mutants are listed in Table I. To facilitate cloning, a restriction site for BamH1 (underlined in Table I) was incorporated in the sense primers DS-411, DS-413, and DS-415. A stop codon (TAG) and a restriction site for HindIII (underlined) were incorporated in the antisense primers DS-412, DS-414, and DS-416. PCR amplification for PspA3–97, PspA98 –192, PspA193–286, PspA3–192, and PspA98 –286 was performed using primer pairs DS-411 and DS-412, DS-413 and DS-414, DS-415 and DS-416, DS-411 and DS-414, and DS-413 and DS-416, respectively (Table I). The PCR cycling parameters were the same as those used for amplifying PspA3–286. The PCR products corresponding to PspA98 –192, PspA193–286, PspA3–192, and PspA98 –286 were digested with BamHI and HindIII and cloned in a BamHI-HindIII-digested pQE-30 expression vector (Qiagen) to generate N-terminal 6⫻ His tagged fusion proteins. PspA3–97 was expressed as a 6⫻ His plus a dihydrofolate reductase-tagged protein in a pQE-40 expression vector (Qiagen). The PspA3–286 deletion mutants were expressed and purified following the same protocol as that mentioned above for PspA3–286.
The purified deletion mutants were used to localize the B cell epitopes recognized by anti-PspA3–286 mAbs using dot blot analysis. One microgram of PspA3–286 and its five deletion mutants were spotted on a nitrocellulose membrane using a dot blot apparatus (Bio-Rad). The membrane was blocked with 1% milk protein in PBS and incubated with culture supernatants from anti-PspA3–286 mAb-secreting hybridomas diluted in PBS containing 1% BSA. The membrane was further incubated with HRPlabeled goat anti-mouse Ig Abs. The blot was washed extensively with PBS containing 0.05% Tween 20 (PBST) and developed using 3, 3⬘-diaminobenzidine and H2O2. Additionally, the 36 mAbs were checked for reactivity with PspA3–286 (3 g) and R36A lysates (15 g) using Western blotting. The membrane was developed as described above.
Determination of relative avidity of anti-PspA3–286 monoclonal and polyclonal Abs The relative avidities of anti-PspA3–286 monoclonal and polyclonal Abs were determined by competition ELISA using protocols described and validated by Rath et al. (29) and Devey et al. (30), respectively. This assay has been shown to rank a panel of mAbs of different affinities to the hapten dinitrophenyl in the same order as equilibrium dialysis, and its use and limitations have previously been discussed in detail (31, 32). Briefly, a 96-well microplate was coated overnight at 4°C with 1 g/ml recombinant PspA3–286 in 100 mM carbonate buffer (pH 9.5) (50 l per well). The plate was washed three times with PBST and blocked with PBS containing 1% BSA (200 l/well) at 37°C for 1 h. The mAbs were titrated using ELISA to give an A450 value of 1. The mAbs were mixed with varying concentrations (ranging from 10⫺4 to 10⫺7 M) of a soluble competitor (recombinant PspA3–286). The mix was added to the well (100 l/well) and the plate was incubated at 37°C for 1 h. The plate was washed and incubated with HRP-conjugated goat anti-mouse Ig diluted 1/1000 in PBST containing 1% BSA at 37°C for 1 h. The plate was washed thoroughly with PBST and color was developed with freshly prepared 3,3⬘,5,5⬘-tetramethylbenzidine and H2O2 (75 l/well). The reaction was stopped by adding 2N H2SO4 at 125 l/well and the absorbance at 450 nm was recorded. The concentration of the competitor that resulted in 50% inhibition was determined for each mAb. For estimating the relative avidities of anti-PspA3–286 polyclonal IgG Abs, two sets of BALB/c mice (n ⫽ 6) were immunized i. p. with 25 g of PspA3–286 in alum and 109 CFU of heat-killed R36A. Serum was taken on days 14, 28, 42, 56, 72, 86, and 100 postimmunization. Competitive inhibition ELISA was performed using PspA3–286 as the competitor as described above.
ELISA additivity assay To test whether any given pair of anti-PspA3–286 mAbs recognized the same/overlapping or different B cell epitopes, ELISA additivity assays were performed as described by Huang et al. (33). Briefly, PspA3–286 (1 g/well) was coated on a 96-well ELISA plate. Two anti-PspA3–286 mAbs were added together or separately, and the amount of bound Ab was quantitated by ELISA. Color reaction was developed as described above for the competition ELISA. An additivity index (AI) was calculated for each mAb pair tested according to the following formula: AI ⫽ ([2 ⫻ A1 ⫹ 2/(A1 ⫹ A2)] ⫺ l) ⫻ 100, where A1 and A2 are the OD values obtained when the mAbs are assayed separately, and A1 ⫹ 2 is the OD value when the two mAbs are pooled in the same well. Provided the concentrations of the mAbs are saturating for the Ag, the AI will tend to be zero if both mAbs recognize the same or overlapping epitope but close to 100 if the two epitopes are topographically unrelated. In other words, two scenarios are possible: 1) when A1 ⫹ 2 ⫽ A1 ⫹ A2 (then AI ⫽ 100); and 2) when A1 ⫹ 2 corresponds to A1 or A2, depending on which is higher (then AI ⫽ 0).
Accession numbers The nucleotide sequences of the genes encoding the mAbs have been deposited in GenBank (www.ncbi.nlm.nih.gov/GenBank/index.html). The following list includes the name of the hybridoma followed by the accession number of the H chain and the accession number of the L chain in parentheses: B5E11 (EU915613/EU915614); C1E7 (EU915615/ EU915616); B4G5 (EU915617/EU915618); CL38 (EU915619/EU915620); C6H7 (EU915621/EU915622); A7A1 (EU915623/EU915624); C3D5 (EU915625/EU915626); F4B1 (EU915627/EU915628); P1C7 (EU915629/ EU915630); P2F9 (EU915631/EU915632); P2G11 (EU915633/ EU915634); P2A4 (EU915635/EU915636); P2B5 (EU915637/ EU915638); P2C2 (EU915639/EU915640); D2F3 (EU915641/ EU915642); E3D1 (EU915643/EU915644); G3B5 (EU915645/ EU915646); I4D3 (EU915647/EU915648); M6E10 (EU915649/
The Journal of Immunology EU915650); P3H11 (EU915651/EU915652); L5F10 EU915654); M4F4 (EU915655/EU915656); O2F8 EU915658); P1E11 (EU915659/EU915660); B3H8 EU915662); J2C11 (EU915663/EU915664); D1A5 EU915666); K1B12 (EU915667/EU915668); M6B2 EU915670); B3D12 (EU915671/EU915672); L5C8 EU915674); J4C1 (EU915675/EU915676); A1D9 EU915678); C4B4 (EU915679/EU915680); F4B6 EU915682); D3H6 (EU915683/EU915684).
5573 (EU915653/ (EU915657/ (EU915661/ (EU915665/ (EU915669/ (EU915673/ (EU915677/ (EU915681/
Results Generation of anti-PspA3–286 mAbs To gain a detailed understanding of the genetic and molecular basis for the anti-PspA Ab response, primary and tertiary antiPspA3–286 mAbs were generated from PspA3–286 and heat-killed S. pneumoniae strain R36A-immunized mice. We chose PspA from R36A (PspA family 1) because it is identical in sequence to PspA from strain Rx1, which has been studied in great detail and used as a prototype of other pneumococcal surface proteins. After two rounds of subcloning by limiting dilution, a total of 116 antiPspA3–286 mAbs were isolated. The isotypes of the H and L chains expressed by the anti-PspA3–286 hybridomas were determined by ELISA. Additionally, total RNA was isolated from the 116 antiPspA3–286 hybridomas and the expressed H chain was amplified by RT-PCR using pooled sense VH primers and the corresponding isotype-specific antisense primers as described. The amplified H chain was sequenced using the internal antisense CH1 PCR primer. The expressed H chain sequences were analyzed for redundancy using MacVector and IgBLAST database, and 36 unique hybridomas were identified. To further characterize the 36 anti-PspA3–286 hybridomas, the expressed H and L chains were RT-PCR amplified, cloned in the pPCR-Script Amp SK(⫹) cloning vector, and sequenced. The sequences were compared with germline sequences in the IMGT and IgBLAST databases. The basic characteristics of the 36 anti-PspA3–286 mAbs were analyzed. Table II summarizes the names of the hybridomas, the Ag used, the nature of the mAbs (primary or tertiary), the isotypes of the expressed H and L chains, the VH and VL family used, the Ig gene segments used, the percentage identity relative to the corresponding germline gene, and the H chain CDR (HCDR) 3 and L chain CDR (LCDR) 3 lengths. Multiple fusions were performed for generating these mAbs. A total of four primary (from four independent fusions) and 32 tertiary (from three independent fusions) mAbs were generated. Isotype analysis indicated that the anti-PspA3–286 Ab response was diverse. The isotypes observed were IgM/, IgM/, IgG1/, IgG2a/, IgG2b/, and IgG3/. The predominant H chain isotypes were IgG1 (38.8%) and IgM (36.1%). A majority (97.3%) of the hybridomas expressed the L chain. The frequencies of other isotypes observed in the antiPspA3–286 Ab response were 13.9% (IgG2a), 8.3% (IgG2b), and 2.7% (IgG3) for the H chain and 2.7% for the L chain. Molecular characterization of VH families/genes in anti-PspA3–286 hybridomas The VH family usage was analyzed in anti-PspA3–286 hybridomas (Table II and Fig. 1A). Of the 16 mouse VH families, eight VH families were used in the anti-PspA3–286 Ab response. The rearranged VH genes present in the 36 hybridomas belonged to the VH1 (27.7%), VH2 (16.6%), VH3 (11.1%), VH5 (11.1%), VH7 (5.5%), VH9 (5.5%), VH10 (2.7%), and VH14 (19.4%) gene families. The 36 hybridomas represent 26 unique VH genes. Seven (19.4%) hybridomas used the VH14 family member VHSM7.a3.93, although the VH14 gene family contributes only 1.4% of the germline VH genes (Table II). Predominance of the VH1 and VH14 families was observed in the anti-PspA3–286 Ab
response (Fig. 1A). Three of the four DH families (DFL16, DSP2, and DQ52) were used in the anti-PspA3–286 Ab response, the DST4 family being the exception (Table II and Fig. 1B). N and P nucleotide additions were observed at the VH to DH and DH to JH junctions in most of the anti-PspA3–286 hybridomas. Analysis of JH sequences in anti-PspA3–286 Abs demonstrated the presence of all four JH gene segments: JH2 (36.1%), JH3 (33.3%), and JH4 (27.7%) were preferred over JH1 (2.7%) (Table II and Fig. 1C). The sequence identity along the entire H chain V region (i.e., FR1 through FR4) was found to be in the range of 91–100% (Table II). Diverse VL families are used in anti-PspA3–286 Ab response VL family usage was found to be more diverse than the VH family usage in the anti-PspA3–286 Ab response. The anti-PspA3–286 Ab response used 14 of the 19 V families (Table II and Fig. 2A). These included the V1 (11.4%), V2 (2.8%), V3 (14.2%), V4 (20%), V5 (5.7%), V6 (2.8%), V8 (5.7%), V9 (5.7%), V10 (2.8%), V12 (11.4%), V13 (8.5%), V15 (2.8%), V18 (2.8%), and V19 (2.8%) families. The 35 -chain-expressing antiPspA3–286 hybridomas represent 23 unique V genes. The lone L chain-expressing hybridoma rearranged V2 and J2 genes. V3 family member 21-10, was the most frequently used (11.4%) V gene in the anti-PspA3–286 mAbs. Four of the five J genes, J1 (20%), J2 (40%), J4 (14.2%), and J5 (25.7%) were used in the anti-PspA3–286 Ab response; J3 was not used in any of the hybridomas (Table II and Fig. 2B). The expressed L chain V region (i.e., FR1 through FR4) exhibited 96 to 100% identity (Table II). These data suggest that diverse VH and V genes encode Abs to PspA3–286. CDR3 analysis The CDR of an Ab molecule is responsible for Ag recognition. Of the three HCDRs and three LCDRs, HCDR3 is known to contribute the most toward Ag recognition. The HCDR3 lengths in the H chain expressed in the 36 anti-PspA3–286 mAbs ranged from 3 to 13 codons (mean ⫾ SD ⫽ 7.8 ⫾ 0.5), HCDR3 lengths of 4 and 9 codons being the most common (Fig. 3A). Nine of the 36 hybridomas had an unusually short HCDR3 (3–5 codons). The VH family, germline VH gene, and sequences of HCDR1, HCDR2, and HCDR3 of the anti-PspA3–286 mAbs are shown in Table III. The corresponding data for the L chain expressed in the anti-PspA3–286 hybridomas are given in Table IV. Although diverse VL gene families were used, the LCDR3 length in the anti-PspA3–286 hybridomas was remarkably restricted and was found to be in the range of 8- 9 (mean ⫾ SD ⫽ 8.7 ⫾ 0.2) codons (Fig. 3B). In one of the hybridomas (E3D1) the LCDR3 length was two codons (Table II). Analysis of the amino acid sequences of CDRs indicated that the majority (77.8%) of the expressed H chain sequences had HCDR1 and HCDR2 with at least one amino acid replacement. In contrast to the H chain, 63.9% of the expressed L chains were germline in sequence. No position-specific recurrent amino acid replacement was observed in HCDR1, HCDR2, LCDR1, or LCDR2 (Tables III and IV). Anti-PspA3–286 Ab response uses IgH chains that lack DH gene-derived amino acids Detailed analysis of the HCDR3 nucleotide sequence revealed that 7 of the 36 (19.4%) anti-PspA3–286 hybridomas expressed a H chain that lacked DH gene-derived amino acids (Fig. 4). The H chain expressed in hybridoma D2F3 was the result of a rearrangement involving the VH gene VHS107.a1.81 and JH3. There was no evidence for the presence of DH gene-derived amino acids in the expressed H chain. There were no N nucleotide addition(s) at the VH to JH junction in D2F3-H. In another hybridoma, I4D3, the H
1 2 3 4 5 5 5 5 6 6 6 6 6 6 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7
PspA3–286 PspA3–286 PspA3–286 R36A PspA3–286 PspA3–286 PspA3–286 PspA3–286 R36A R36A R36A R36A R36A R36A R36A R36A R36A R36A R36A R36A R36A R36A R36A R36A R36A R36A R36A R36A R36A R36A R36A R36A R36A R36A R36A R36A
B5E11 C1E7 B4G5 CL38 C6H7 A7A1 C3D5 F4B1 P1C7 P2F9 P2G11 P2A4 P2B5 P2C2 D2F3 E3D1 G3B5 I4D3 M6E10 P3H11 L5F10 M4F4 O2F8 P1E11 B3H8 J2C11 D1A5 K1B12 M6B2 B3D12 L5C8 J4C1 A1D9 C4B4 F4B6 D3H6
Pri Pri Pri Pri Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter Ter
mAb (Pri/Ter)c Family
VH1 VH14 VH9 VH14 VH1 VH2 VH1 VH5 VH2 VH3 VH1 VH9 VH10 VH1 VH7 VH3 VH1 VH1 VH2 VH7 VH1 VH2 VH3 VH3 VH5 VH5 VH14 VH14 VH14 VH14 VH14 VH2 VH1 VH1 VH5 VH2
IgM/ IgM/ IgG1/ IgM/ IgM/ IgM/ IgM/ IgG1/ IgM/ IgG1/ IgG2a/ IgG2a/ IgG2a/ IgG2a/ IgM/ IgM/ IgM/ IgM/ IgM/ IgM/ IgG1/ IgG1/ IgG1/ IgG1/ IgG1/ IgG1/ IgG1/ IgG1/ IgG1/ IgG1/ IgG1/ IgG2a/ IgG2b/ IgG2b/ IgG2b/ IgG3/ VH108A VHSM7.a3.93 VHVGAM3.8.a5.105 VHSM7.a3.93 J558.3 VHQ52.a12.33 J558.45 VH7183.3b VHQ52.a7.18 VH36-60.a6.114 J558.83.189 VHVGAM3.8.a4.102 Vh10.2a J558.45 VHS107.a1.81 VH36-60.a2.90 J558.1 J558.50 VHQ52.a2.4 VHS107.a3.106 J558.6 VHQ52.a27.79 VH36-60.a5.112 VH36-60.a1.85 VH7183.a13.20 VH7183.a35.57 VHSM7.a3.93 VHSM7.a3.93 VHSM7.a3.93 VHSM7.a3.93 VHSM7.a3.93 VHQ52.a2.4 J558.15 J558.47 VH7183.a35.57 VHQ52.a2.4
DFL16.1 DFL16.2 Absent DSP2.11 DSP2.12 Absent DSP2.X; DQ52-C57BL/6 DFL16.2 DQ52-BALB/c; DQ52-C57BL/6 DSP2.2 Absent DFL16.1 DQ52-BALB/c DSP2.3; DSP2.4; DSP2.9 Absent DFL16.2 DFL16.1 Absent DSP2.10; DSP2.11 DSP2.5; DSP2.7; DSP2.8 DFL16.1 DQ52-BALB/c; DQ52-C57BL/6 DSP2.2 DSP2.x DFL16.1 DFL16.2 DSP2.12; DFL16.1j DSP2.12; DFL16.1j Absent DQ52-BALB/c; DQ52-C57BL/6 Absent DSP2.10; DSP2.11 DSP2.10; DSP2.11 DFL16.1 DFL16.1; DFL16.2 DSP2.10; DSP2.11
JH2 JH2 JH2 JH2 JH3 JH2 JH2 JH2 JH4 JH4 JH3 JH3 JH3 JH4 JH3 JH4 JH4 JH3 JH4 JH2 JH3 JH3 JH4 JH3 JH3 JH2 JH2 JH2 JH4 JH3 JH2 JH4 JH2 JH1 JH3 JH4
97.4 99.4 99.7 97.5 92.8 97.4 98.9 93.0 100.0 98.6 95.8 97.8 97.3 97.7 100.0 98.7 100.0 100.0 96.6 100.0 96.0 94.1 97.3 98.9 97.6 98.6 98.1 94.7 96.9 98.1 94.9 97.2 94.2 98.9 97.5 96.9
9 8 4 7 11 4 7 8 6 13 5 9 6 4 3 13 12 4 9 8 11 9 10 12 9 10 7 7 7 4 4 9 8 11 5 9
V1 V8 V3 V12 V10 V4 V4 V19 V13 V13 V2 V5 V13 V1 V4 V18 V6 V9 V12 V 2 V3 V8 V15 V5 V3 V4 V4 V4 V4 V3 V3 V12 V1 V9 V1 V12
bd2 8-27 21-10 12-41 ce9 kh4 kk4 gj38c gm33 gn33 hf24 23-43 gm33 bd2 aa4 dv-36 19-15 bv9 12-44 VL2 21-12 8-24 gr32 23–39 21-10 kk4 aq4 aq4 aq4 21-10 21-10 12-44 cr1 bv9 bd2 12-44
J2 J1 J2 J4 J 1 J2 J5 J2 J1 J2 J5 J2 J1 J5 J 5 J4 J4 J1 J5 J2 J4 J2 J2 J2 J2 J2 J5 J5 J4 J2 J2 J5 J 1 J2 J1 J5
98.9 99.4 99.3 99.0 96.9 100.0 100.0 98.9 100.0 99.3 98.7 98.6 100.0 98.9 100.0 98.1 100.0 99.6 98.9 99.7 99.0 96.2 100.0 99.3 99.7 99.3 99.6 99.3 97.3 99.0 98.0 98.3 97.5 98.3 99.3 99.3
9 8 9 9 9 9 9 8 9 9 9 9 9 9 9 2 9 8 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9
Four hybridomas (C6H7, A7A1, C3D5, and F4B1) were generated from CBA/J mice while the remaining 32 hybridomas were generated from BALB/c mice. Recombinant PspA3–286 or heat-killed R36A was used as immunogen. The mAbs were screened using PspA3–286 as the capture Ag in ELISA. Pri, Primary; Ter, tertiary. d Nucleotide sequence identity was calculated for the complete V region, i.e., from FR1 through FR4. Mutations were scored relative to the corresponding VH, DH, and JH germline gene segments for the H chain and the VL and JL gene segments for the L chain. N and P nucleotides added at the VH to DH, DH to JH, and VL to JL junctions were excluded for calculating the nucleotide sequence identity. e HCDR3 and LCDR3 are as per the Kabat system. HCDR3 and LCDR3 length is expressed as the number of codons.
Table II. Molecular characteristics of VH and VL genes expressed in anti-PspA3–286 hybridomas
5574 MOLECULAR CHARACTERIZATION OF ANTIBODY RESPONSE TO PspA
The Journal of Immunology
FIGURE 2. V and J gene usage in anti-PspA3–286 Ab response. V (A) and J (B) family usage in the 35 L chain-expressing anti-PspA3–286 hybridomas is presented as a percentage. The lone L chain-expressing hybridoma was not included in this analysis.
FIGURE 1. VH, DH, and JH gene usage in anti-PspA3–286 hybridomas. The VH, DH, and JH gene families used in the 36 anti-PspA3–286 hybridomas are presented (as percentage) in A, B and C, respectively. The DH genes were subgrouped in families according to the IgBLAST database. In cases where the DH gene could be assigned unambiguously or where the assigned DH genes belonged to same family, a weightage of 1 was given. In cases where the assigned DH genes belonged to two or more families, the weightage was prorated. The seven hybridomas that expressed H chains devoid of DH gene-derived amino acids were not included in the analysis of DH gene usage.
chain was the result of a J558.50 to JH3 rearrangement. DH gene-derived amino acids were absent in I4D3-H. The five other hybridomas where the expressed H chain lacked DH gene-derived amino acids were A7A1 (VHQ52.a12.33 to JH2 rearrangement), P2G11 (J558.83.189 to JH3 rearrangement), M6B2 (VHSM7.a3.93 to JH4 rearrangement), L5C8 (VHSM7.a3.93 to JH2 rearrangement), and B4G5 (VHVGAM3.8.a5.105 to JH2 rearrangement). The seven hybridomas used diverse VH and JH genes. N nucleotide addition(s) were observed in the expressed H chains of six of the seven hybridomas, but none showed a P nucleotide addition. The number of N nucleotide(s) observed in I4D3-H, A7A1-H, P2G11-H, M6B2-H, L5C8-H, and B4G5-H were 1, 4, 4, 6, 6, and 9, respectively. To understand the possible mechanism by which these novel H chain rearrangements arose, the stretch of N nucleotides added at the VH to JH junction was analyzed for possible similarities with the recombination signal sequences. No match was found in any of the seven cases. The anti-PspA3–286 hybridomas that expressed a H chain lacking DH gene-derived amino acids were obtained from PspA3–286- and R36A-immunized mice
and four independent fusions. These primary (1/7) and tertiary (6/7) hybridomas secrete the anti-PspA3–286 mAb of the IgM (3/7), IgG1 (3/7), or IgG2a (1/7) isotype. Interestingly, three of the six tertiary mAbs were of the IgM isotype, whereas the lone primary mAb was of the IgG1 isotype. One of the hybridomas (A7A1) that expressed a H chain devoid of DH gene-derived amino acids was generated from CBA/J mice while the rest of the six were from BALB/c mice (Table II). Lack of DH gene-derived amino acids does not prevent anti-PspA3–286 mAbs from attaining average relative avidity To assess whether the lack of DH gene-derived amino acids makes a difference to binding ability, the relative avidities of 13 IgM and 23 IgG anti-PspA3–286 mAbs were determined by competition ELISA using PspA3–286 as a soluble competitor (Fig. 5). For the
FIGURE 3. HCDR3 and LCDR3 length distribution in anti-PspA3–286 mAbs. HCDR3 (A) and LCDR3 (B) lengths are shown in codons.
MOLECULAR CHARACTERIZATION OF ANTIBODY RESPONSE TO PspA
Table III. HCDR sequences of the H chain expressed in anti-PspA3–286 hybridomasa Family
YINPSTGYTEYNQKFKD ----N----------------S----------YIYPYNGGTGYNQKFKS -------D-------QR DINPNNGGTSYNQKFKG R-S-S--D-T------DINPNNGGTIYNQKFKG -----T-T-F---N--AIYPGNSDTSYNQKFKG ----------------QIYPGDGDTNYNGKFKG ---------K-I----YINPYNDGTKYNEKFKG ----D-----------YISCYNGATSYNQKFKG ----------------DIYPGGGYTNYNEKFKG H------D-D-----E-
B3D12 C1E7 CL38 K1B12 D1A5 M6B2 L5C8
DTYMH -----------I---IY ----Y ----D ---VN
SYGVH --DIH-D-N-DISYGVH N-A-SYGVH N---SYGVH -----
VIWSGGSTDYNAAFIS L--T-----------M--------------M--------------VIWSGGSTDYNAAFIS ----D----------VIWAGGSTNYNSALMS ----S-N--------VIWRGGSTDYNAAFMS ----------------
TGNYRWS ------SGYYWN -D---SDYAWN -----SGYSWH ------
YIYYSGTITYNPSLTS --------SD-----YISYDGSNNYNPSLKN -----D--K------YISYSGSTSYNPSLKS ------R---T----YIHYSGSTNYNPSLKS ---ST-I---------
SYTMS ------I-DYYMY ---ISYTMS I-S--
YISNGGGSTYYPDTVKG -----------------------F--------YISNGGGSTYYPDTVKG -----------I----TISGGGGNTYYPDSVKG ---HD--I---------
DFYME ----DYYMS -----
ASRNKANDYTTEYSASVKG ------------------FIRNKANGYTTEYSASVKG -------------------
DYSMH ----NYGMN -F-L-
WINTETGEPTYADDFKG ----------------WINTYTGEPTYADDFKG --S--------------
VH10A B5E11 J558.3 C6H7 J558.6 L5F10 J558.50 I4D3 J558.83.189 P2G11 J558.47 C4B4 J558.1 G3B5 J558.15
VHQ52.a2.4 J4C1 D3H6 M6E10 VHQ52.a12.33 A7A1 VHQ52.a27.79 M4F4 VHQ52.a7.18 P1C7
VH36-60.a5.112 O2F8 VH36-60.a6.114 P2F9 VH36-60.a2.90 E3D1 VH36-60.a1.85 P1E11
VH7183.a35.57 J2C11 F4B6 VH7183.a13.20 B3H8 VH7183.3b F4B1
VHS107.a1.81 D2F3 VHS107.a3.106 P3H11
VHVGAM3.8.a5.105 B4G5 VHVGAM3.8.a4.102
SYWMH ---I----DYNMH ----DYYMK ----N DYNMD ----SYWMH ----SYWMN ----SYVMH --I-GYYMH ----NYWIG -----
AMDY NSNWDDY LGTTVVIDY WGYYTYEGFAY GRNYGSDWFAY WFAY GGFDY GGYGSRRYFDV SGYGSSYEAMDY FRYYRFDY
GWEY SYGYYFDY SWYYFDY WSYSFDY WSYSFDY ASYAMDY GWDV
KNYRYLMDY KNYRYLMDY KNYRYLMDN NWDY GMRGAWLTY WDAMDY
RSYDYGSMDY GRDYGGLYYAMDY EPLYYGDFYAMDY SLSRTYSRVFAY
QRTTSTYFDY PTEGY FSLLVGFPY SSFPIFDY
HCDRs of the H chain expressed in the anti-PspA3–286 hybridomas are according to the Kabat system. The translation of germline HCDRs (boldface) are shown for comparison, and dashes are used to indicate sequence identity with the germline VH gene.
The Journal of Immunology
Table IV. LCDR sequences for the L chain expressed in anti-PspA3–286 hybridomasa Family
RASESVDSYGNSFMH -------------------------------------------T---IN-------RASKSVSTSGYSYMH ---------------
LASNLES ---------------S-----S--LASNLES -------
KSSQSLLDSDGKTYLN -------------F--------N----------------------RSSQSIVHSNGNTYLE ----D-----------
LVSKLDS ---Q--------------KVSNRFS -I-----
RASENIYSYLA ------------------------------RASGNIHNYLA ----------V
NAKTLAE ------------------NAKTLAD -------
KASEDIYNRLA --------------------KASDHINNWLA -----------
GATSLET ------------GATSLET -------
RASQSISNNLH -----S----RASQSISDYLH ----N------
YASQSIS ------YASQSIS -------
KSSQSVLYSSNQKNYLA ----------------KSSQSLLNSSNQKNYLA -------S-R-------
WASTRES ------FASTRES -------
aa4 D2F3 kh4 A7A1 21-10 B3D12 B3H8 B4G5 L5C8 21-12 L5F10 bd2 B5E11 P2C2 F4B6 cr1 A1D9 V12
12-44 M6E10 J4C1 D3H6 12-41 CL38
gm33 P1C7 P2B5 gn33 P2F9
23-43 P2A4 23-39 P1E11
8-27 C1E7 8-24
V 2 V6 V10 V15
hf24 19-15 ce9 gr32
LTSNLAS ------------------DTSKLAS ------------RTSNLAS ------GTSNLAS -------
SASSSVSYMY ---------------------------SASSSVSYMH ----------------S-SASSSVSYMY ---------SVSSSISSSNLH ------------
D1A5 K1B12 M6B2 kk4
QQWSSNPLT QQWSGNPLT HQWSINPFT QQWSSNPPT QQWSSNPYT QQYHSYPPT QQWSSYPLT
QQNNEDPYT QQNNEDPYT QQNNEDPYT QQNNEDPYT QHSRELPPT
WQGTHFPFT WQGTHFPLT WQGTHFPWT FQGSHVPPT
QHHYGTPLT QHHYGTPLT QHHYGTPLT QHFWSTPFT
QQYWSTPRT QQYWSTPRT QQYWSSPYT
LCDRs of the L chain expressed in the anti-PspA hybridomas are according to the Kabat system. The translations of germline LCDRs (boldface) are shown for comparison, and dashes are used to indicate sequence identity with the germline VL gene.
purpose of this analysis, mAbs with a relative avidity ⱕ1 ⫻ 1011 M⫺1 were considered low avidity binders. By this criterion, all 13 anti-PspA IgM mAbs are low avidity binders. Three of these mAbs
lacked DH gene-derived amino acids (I4D3, D2F3, and A7A1; Fig. 5A). There was no significant difference in the average relative avidities of the three IgM mAbs that lacked DH gene-derived
MOLECULAR CHARACTERIZATION OF ANTIBODY RESPONSE TO PspA
FIGURE 4. Novel H chains that lack DH gene-derived amino acids are recruited in an anti-PspA3–286 Ab response. Only HCDR3 and its adjoining sequence are shown. The suffix “-H” has been added to the hybridoma name to indicate the expressed H chain. The sequences of the germline VH and JH genes and the amino acid translation are shown in boldface for reference. There was no P nucleotide addition at the VH to JH junction in these hybridomas. Refer to Table II for the isotype and nature (primary or tertiary) of the anti-PspA3–286 mAbs. Dash indicates nucleotide identity to germline sequence and N denotes N nucleotide addition.
amino acids (7.3 ⫾ 4.6 ⫻ 109 M⫺1) and 10 IgM mAbs with DH gene-derived amino acids (11.1 ⫾ 6.2 ⫻ 109 M⫺1). Of the 23 anti-PspA3–286 IgG mAbs, nine had a relative avidity in the range of 1 ⫻ 1011 M⫺1 to 5 ⫻ 1011 M⫺1; the remaining 14 were low avidity binders (Fig. 5B). The four anti-PspA3–286 IgG mAbs (L5C8, M6B2, P2G11, and B4G5) devoid of DH gene-derived amino acids had low relative avidities. The average relative avidities of the four anti-PspA IgG mAbs that lacked DH gene-derived amino acids (12.7 ⫾ 2.1 ⫻ 109 M⫺1) and 10 low avidity antiPspA3–286 IgG mAbs that possess DH gene-derived amino acids (19.7 ⫾ 5.0 ⫻ 109 M⫺1) were comparable. None of the nine high avidity anti-PspA3–286 IgG mAbs lacked DH gene-derived amino acids. There was no correlation between relative avidity and the somatic hypermutation pattern of the anti-PspA3–286 mAbs (data not shown). Based on these data, it can be inferred that the absence of DH gene-derived amino acids does not prevent anti-PspA3–286 mAbs from attaining a meaningful or average relative avidity. IgG subclass-associated affinity differences of Ag-specific Abs have been documented in humans (31, 34). To explore whether IgG subclasses differ with respect to their relative avidities for PspA3–286, we analyzed the relative avidities of the 23 anti-PspA3–286 mAbs belonging to different IgG subclasses (Fig.
FIGURE 5. Relative avidities of anti-PspA3–286 mAbs. The relative avidities of IgM and IgG anti-PspA3–286 mAbs are shown in A and B, respectively. The IC50 values obtained using PspA3–286 as a competitor are plotted on the x-axis. The name of the hybridoma is indicated on the y-axis. The anti-PspA3–286 mAbs that lack DH gene-derived amino acids are indicated by an asterisk (ⴱ). Comp conc, Competitor concentration.
5B). We focused our attention on the 21 tertiary anti-PspA3–286 mAbs that were raised from heat-killed, R36A-immunized mice (Table I). These 21 mAbs belonged to the IgG1 (n ⫽ 12), IgG2a (n ⫽ 5), IgG2b (n ⫽ 3), and IgG3 (n ⫽ 1) IgG subclasses. The average relative avidity (⫾ SE) for the IgG1, IgG2a, and IgG2b mAbs was found to be 167.60 ⫾ 52.32 ⫻ 109 M⫺1, 39.86 ⫾ 23.17 ⫻ 109 M⫺1, and 58.46 ⫾ 32.41 ⫻ 109 M⫺1, respectively. The average relative avidities of the IgG2a and IgG2b mAbs were found to be comparable but appear to be significantly different from the average relative avidity of IgG1 mAbs. Whether this is also true in vivo was not investigated in our study. Relative avidity of primary anti-PspA3–286 polyclonal IgG Abs induced in response to immunization with heat-killed R36A is higher than that induced in response to PspA3–286 The nature of B cell responses induced is likely to be dependent on the form of the Ag delivered. Therefore, it is likely that the antiPspA3–286 Ab responses might differ when PspA is provided as a recombinant protein (given with alum) or in the context of the pneumococcal cell surface (given as heat-killed R36A). The relative avidities of primary anti-PspA3–286 polyclonal IgG Abs raised in mice immunized with PspA3–286 or heat-killed R36A were determined. The results demonstrate that there is avidity maturation during the primary IgG anti-PspA3–286 polyclonal Ab response in R36A and PspA3–286 immunized mice (Fig. 6). The relative avidity of the primary IgG anti-PspA3–286 polyclonal Abs reached a
The Journal of Immunology
5579 Our study does not address whether PspA3–286-specific IgM and IgG responses differ with respect to their ability to undergo avidity maturation in vivo. The IgM anti-PspA3–286 Ab end point titers had dropped to a low of ⬃1/100 by 2 wks postimmunization in mice immunized with PspA3–286 and heat-killed R36A. The titers continued to drop further till our last time point, i.e., day 60 postimmunization. Devey et al., while analyzing human polyclonal sera for avidity maturation, observed that low-affinity Abs, especially if present in low titer, were affected by the presence of highaffinity Abs, resulting in irregular inhibition curves at limiting concentrations of free Ag (31). Because of low IgM anti-PspA3–286 Ab end-point titers, low relative avidities of IgM anti-PspA3–286 mAbs (Fig. 5A), and the technical difficulty associated with determining the relative avidity of low avidity Abs, we did not attempt to investigate whether the PspA3–286-specific IgM response underwent avidity maturation in vivo. Evidence for Ag-driven selection
FIGURE 6. Maturation of relative avidities during primary anti-PspA3–286 polyclonal IgG Ab response. BALB/c mice (n ⫽ 6) were immunized i. p. with 25 g of PspA3–286 in alum or 109 CFU of heat-killed R36A. Serum was collected on days 14, 28, 42, 56, 72, 86, and 100. Competition ELISA was performed to estimate the relative avidities using PspA3–286 as a competitor. The IC50 value was determined for each serum sample and plotted as a function of time postimmunization. The IC50 value is presented as mean ⫾ SE. Comp conc, Competitor concentration.
peak at day 72, beyond which there was no further increase in the relative avidity for PspA3–286. The relative avidities of primary anti-PspA3–286 polyclonal IgG Abs induced in response to R36A were higher compared with those generated in response to PspA3–286 on days 72, 86, and 100 postimmunization (Fig. 6). Avidity maturation could conceivably arise in part through changes in the proportions of different isotypes over time due to differences in isotype-associated structural features such as segmental flexibility. To test this possibility, we selected two time points: an early time point (day 28) where the relative avidities of the primary anti-PspA3–286 polyclonal IgG Abs induced in response to immunization with PspA3–286 and R36A were comparable, and a later time point (day 72) at which the relative avidities differed the most (Fig. 6). We determined the serum PspA3–286specific end point titers of the various IgG subclasses in the PspA3–286- and heat-killed R36A-immunized mice that were used for determining the relative avidities (Fig. 6). Our data suggest that IgG1 is the predominant IgG subclass and that there is no significant difference in the profile of the serum primary PspA3–286specific Abs belonging to various IgG subclasses induced on days 28 and 72 in R36A-immunized mice (Fig. S1).6 In the PspA3–286immunized mice, the serum primary PspA3–286-specific IgG end point titer was essentially due to the IgG1 subclass; the antiPspA3–286 IgG2a, IgG2b, and IgG3 subclass titers were below the preimmune cut-off (data not shown). Thus, the observed relative avidity maturation of the primary PspA3–286-specific IgG Abs induced in response to immunization with PspA3–286 and R36A does not appear to be due to changes in the proportions of different IgG subclasses over time. This data when viewed in conjunction with the finding that IgG1 is the predominant serum IgG subclass in the primary PspA3–286-specific IgG Abs induced in response to immunization with PspA3–286 and R36A provides indirect evidence that the observed relative avidity maturation is most likely due to an increase in the intrinsic affinities of the Abs over time. This, however, requires formal confirmation. 6
The online version of this article contains supplemental material.
The sequences of the H and L chains expressed in the 36 antiPspA3–286 hybridomas were further analyzed for evidence of Agdriven selection. This was done by analyzing the replacement and silent mutations in the CDR and FR regions of the expressed H chain and L chain genes. The distribution of somatic mutations in the H and L chain genes expressed in the anti-PspA3–286 hybridomas was analyzed using the multinomial distribution method of Lossos et al. (35). The probability that an excess of replacement mutations in CDRs or a scarcity of replacement mutations in FRs resulted from chance alone was calculated for the expressed H and L chains (Table V). In total, 24 H and seven L chains had four or more mutations, and these mutated sequences were from 25 antiPspA3–286 hybridomas. Of the 25 anti-PspA3–286 hybridomas, 11 (44%) H chain sequences showed evidence for Ag-driven selection, i.e., significant accumulation of replacement mutations in CDRs or scarcity of replacement mutations in FRs ( p ⬍ 0.05) (Table V). Of the 11 anti-PspA3–286 hybridomas that showed evidence for Ag-driven selection at the H chain, only two (A1D9 and L5C8) exhibited evidence of Ag-driven selection in their L chain partners and three showed high avidities for PspA3–286 (Table V and Fig. 5). From these data it can be inferred that the H chain is the major contributor toward PspA recognition. This is consistent with the observation that the anti-PspA3–286 Ab response uses a greater diversity of V families compared with VH families. DH gene segments are important sources of HCDR3 diversity. Each DH segment gives the developing B cell access to up to six functional reading frames (RFs) of distinctly different germline sequences. The DH RFs used in the H chain expressed in the antiPspA3–286 hybridomas were analyzed. The seven anti-PspA3–286 mAbs that lacked DH gene-derived amino acids were excluded from the analysis. The distribution of the RFs used in the DH gene expressed in the remaining anti-PspA3–286 mAbs was as follows: RF3, 75.9%; RF1, 13.8%; and RF2, 10.3%. DH RF3 was preferred over RF1 and RF2. The majority (72.2%) of the anti-PspA3–286 mAbs used a DH RF that was rich in tyrosine and glycine codons. Recognition of the Ag by Ag-specific B cells results in clonal expansion. Analysis of the sequences of the H and L chains expressed in the anti-PspA3–286 hybridomas indicated the presence of two pairs of clonally related hybridomas. The clones were identified based on identical gene segment usage and HCDR3 and LCDR3 sequences. Both pairs of clonally related hybridomas were obtained from the same fusion (Table II). In the first case, the anti-PspA3–286 hybridomas D3H6 and M6E10 were found to be clonally related (Fig. 7A). The founder progenitor (FP) B cell that gave rise to this clone rearranged the germline genes VHQ52.a2.4, DSP2.10 and JH4 at the H chain locus and 12-44 and J5 at the
MOLECULAR CHARACTERIZATION OF ANTIBODY RESPONSE TO PspA
Table V. Analysis of replacement (R) and silent (S) mutations in the FR and CDR regions of the H and L chain expressed in the anti-PspA3–286 hybridomasa
Observed R/S Ratio in FR
P for FRb
B5E11 CL38 C6H7
VH VH VH VL VH VH VH VL VH VH VH VL VH VL VH VH VH VL VH VH VH VH VH VH VH VH VH VL VH VH VL
VH108A VHSM7.a3.93 J558.3 ce9 VHQ52.a12.33 VH7183.3b VH36-60.a2.90 dv-36 VHQ52.a2.4 J558.83.189 J558.15 cr1 J558.45 bv9 J558.6 VHQ52.a2.4 VHQ52.a27.79 8 –24 VHQ52.a2.4 VH36-60.a6.114 VH36-60.a5.112 VH7183.a35.57 VHVGAM3.8.a4.102 Vh10.2a VHSM7.a3.93 VHSM7.a3.93 VHSM7.a3.93 aq4 VHSM7.a3.93 VHSM7.a3.93 21-10
3/1 4/1 2/5 4/2 1/2 9/3 1/1 4/1 4/2 2/4 8/2 0/2 1/1 4/0 4/2 4/1 5/3 6/0 5/1 1/0 1/3 0/3 2/2 2/1 5/0 7/2 4/2 4/0 3/1 6/0 0/0
0.20911 0.64656 0.00043 0.28744 0.01141 0.17612 0.12061 0.80910 0.12022 0.01840 0.35497 0.00778 0.11401 0.78360 0.13292 0.19574 0.12039 0.57298 0.41005 0.11831 0.01524 0.00634 0.06116 0.03565 0.87910 0.38055 0.13619 0.77956 0.35228 0.29206 0.01759
A7A1 F4B1 E3D1 M6E10 P2G11 A1D9 P2C2 C4B4 L5F10 J4C1 M4F4 D3H6 P2F9 O2F8 F4B6 P2A4 P2B5 D1A5 K1B12 M6B2 B3D12 L5C8
Selection against R in FR
Yes Yes Yes
Observed R/S Ratio in CDR
P for CDRb
0/0 1/0 6/1 2/0 3/1 5/2 2/0 0/0 4/0 2/1 4/1 2/1 2/0 1/0 4/0 4/0 3/1 2/2 3/0 3/0 2/1 2/0 3/0 4/1 1/0 3/1 3/1 0/1 2/0 5/1 4/0
0.07502 0.49813 0.01985 0.25272 0.05873 0.15780 0.08882 0.78298 0.04452 0.36854 0.19557 0.15102 0.09376 0.37746 0.06066 0.02982 0.22272 0.48674 0.11686 0.01161 0.24276 0.12598 0.06887 0.02871 0.49813 0.31181 0.17778 0.77668 0.18749 0.03096 0.00066
Selection for R in CDR
a Data are shown from only those anti-PspA3–286 hybridomas where the number of mutations in the expressed H or L chain is ⱖ4, as it is difficult to draw meaningful conclusions regarding Ag-driven selection from sequences that are germline or near germline (⬍4 mutations). The L chain partner of a H chain (having ⱖ4 mutations) with ⬍4 mutations and vice versa are not shown. b The probability that an excess of R mutations in CDRs or a scarcity of R mutations in FR occurred by chance was calculated according to Lossos et al. (35). Probability values of ⬍0.05 were considered significant and are shown in boldface. In this table, FR represents FR1, FR2, and FR3, whereas CDR represents CDR1 and CDR2. Inferences regarding evidence for Ag-driven selection drawn from the statistical analysis are provided in the last column.
L chain locus. The FP B cell underwent three somatic mutations at the H chain locus to give rise to a hypothetical common precursor (HP) B cell of the IgM/ isotype (Fig. 7, A and C). M6E10 (IgM/ isotype) was generated by the acquisition of eight and three additional nucleotide changes relative to the H and L chain sequences of the HP B cell, respectively. D3H6 was generated from the HP B cell by a class switch recombination event to the IgG3 isotype, seven somatic mutation events in the H chain, and two nucleotide changes in the L chain. The second clonally related pair of anti-PspA3–286 hybridomas found was D1A5 and K1B12 (Fig. 7B). The FP B cell that gave rise to D1A5 and K1B12 rearranged the germline genes VHSM7.a3.93, DSP2.12, and JH2 at the H chain locus and aq4 and J5 at the L chain locus. The FP B cell underwent a class switch recombination event at the H chain locus to the IgG1 isotype, six mutations in the H chain, and one mutation in the L chain to give rise to a HP B cell (Fig. 7, B and D). The HP B cell gave rise to D1A5 following a point mutation in the H chain. K1B12 originated from the HP B cell as a result of 11 and one somatic mutations at the H chain and L chain loci, respectively. There was preferential use of the VH1 and VH14 families in the anti-PspA3–286 Ab response (Table II and Fig. 1A). Preferential H chain and L chain pairing was seen in three instances: VH1 family member paired with V1 family member (8.33%); VH2 family member paired with V12 family member (8.33%); and VH14 family member paired with V3 family member
(8.33%) (Table II). The mean CDR3 length at the H chain and L chain loci was 7.8 ⫾ 0.5 and 8.7 ⫾ 0.2 codons, respectively. This is in contrast to CDR3 length reported for the mouse preimmune repertoire, which ranges from 5 to 20 codons with an average length of ⬃12.5 codons (36) (Fig. 3). These data collectively provide evidence for Ag-driven selection during the anti-PspA3–286 Ab response. Localization of B cell epitopes recognized by anti-PspA3–286 mAbs To map the B cell epitopes recognized by the anti-PspA3–286 mAbs, a series of overlapping and nonoverlapping deletion mutants of PspA3–286 were generated. The reactivity pattern of antiPspA3–286 mAbs to PspA3–286 and its deletion mutants was analyzed by dot and Western blotting. The subdomain specificity and the location of the “main” B cell epitopes of anti-PspA3–286 mAbs are summarized in Table VI. All of the 36 PspA3–286 mAbs reacted with recombinant PspA3–286 and R36A lysates on Western blots. Based on the reactivity pattern with PspA3–286 and its deletion mutants, the anti-PspA3–286 mAbs were classified into 12 independent groups. Groups 1 and 5 had a lone member (Table VI). Some mAbs exhibited some degree of cross-reactivity between the Nand C-terminal halves of PspA3–286. The epitopes recognized by the mAbs that lack DH gene gene-derived amino acids fell in six of the 12 groups.
The Journal of Immunology
FIGURE 7. Clonally related anti-PspA3–286 hybridomas. A and B, The nucleotide sequence alignment of the H and L chains expressed in the clonally related hybridomas, somatic mutations, and amino acid replacements are summarized. Sequence analyses of the clonally related anti-PspA3–286 hybridomas D3H6 and M6E10 (A) and D1A5 and K1B12 (B). The germline VH, DH, JH, VL, and JL used in the expressed H and L chains are depicted for reference (boldface). The suffixes “-H” and “-L” were added to the name of the hybridoma to indicate H and L chain, respectively. The FR and CDR regions and the codon positions are according to the Kabat system. Only those codon positions are shown where there is at least one nucleotide change (relative to the corresponding germline gene) in one or both the clonally related sequences with the exception of CDR3 and its adjacent codon positions, which are shown in their entirety. The amino acid translation is shown below the nucleotide sequence in single-letter code (in italics). Replacement and silent mutations are shown in upper and lower case, respectively. The P and N nucleotide addition(s) at the VH to DH, DH to JH, and VL to JL junctions are indicated. Nucleotide identity with the corresponding nucleotide in the germline gene is indicated by a dash. C and D, Deduced genealogical tree for the clonally related anti-PspA3–286 hybridomas D3H6 and M6E10 (C) and D1A5 and K1B12 (D) are shown along with number of the nucleotide changes at the H chain (written first) followed by those at the L chain (in italics). The isotypes of the expressed H and L chains are indicated.
The topographic relationship between epitopes recognized by 36 mAbs was analyzed by an ELISA additivity assay. This assay was used to determine whether members within any given group rec-
ognized the same/overlapping or different B cell epitopes. Analysis of the anti-PspA3–286 mAbs indicated that members within a group recognized the same or overlapping epitope with the exception
Table VI. Subdomain specificity of the anti-PspA3–286 mAbsa
Location of ‘Main’ Epitope
1 2 3 4 5 6 7 8 9 10 11 12
A7A1 L5F10, J4C1, P2F9, O2F8b M4F4, D3H6b P1E11, P2A4, D1A5b P2B5 K1B12, M6B2b B4G5, C3D5, F4B1, P1C7, E3D1, G3B5, B5E11d D2F3, C6H7, B3D12, L5C8e I4D3, A1D9, P3H11, B3H8, F4B6, J2C11f P2G11, P2C2b M6E10, C4B4b C1E7, CL38b
⫹⫹ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺
⫺ ⫹⫹ ⫹⫹⫹ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫹⫹⫹ ⫹⫹ ⫹⫹⫹ ⫺ ⫹ ⫺ ⫺ ⫺ ⫹⫹
⫹⫹⫹ ⫹⫹⫹ ⫹⫹⫹ ⫺ ⫺ ⫹⫹ ⫹⫹⫹ ⫹⫹⫹ ⫺ ⫹ ⫹⫹⫹ ⫹⫹⫹
⫺ ⫹⫹⫹ ⫹⫹⫹ ⫹⫹⫹ ⫹⫹⫹ ⫹⫹⫹ ⫺ ⫹ ⫹⫹⫹ ⫹⫹⫹ ⫹⫹⫹ ⫹⫹
PspA3–97 PspA98 –192 PspA98 –192 PspA193–286 PspA193–286c PspA193–286 PspA3–192c PspA3–192 PspA98 –286c PspA98 –286 PspA3–286 PspA3–286
a The seven anti-PspA3–286 hybridomas that express H chains that lack DH gene-derived amino acids are underlined. All 36 mAbs bound PspA3–286 strongly on Western and dot blots and ELISA. ⫹⫹⫹, Strong binding; ⫹⫹, average binding; ⫹, weak but readily detectable; ⫺, no binding. b Based on ELISA additivity assay, the mAbs within the group recognize the same or overlapping B cell epitope. c The mAbs showed some degree of cross-reactivity with the indicated subfragments of PspA3–286. d The ELISA additivity assay data suggest that P1C7 recognizes the same or overlapping B cell epitopes as B4G5, C3D5, F4B1, G3B5, and B5E11. Similarly, E3D1 recognizes the same or overlapping epitopes as B4G5, C3D5, F4B1, G3B5, and B5E11. P1C7 and E3D1 bind topographically unrelated B cell epitopes. e D2F3, B3D12, and L5C8 bind overlapping epitopes. C6H7, B3D12, and L5C8 bind overlapping epitopes. D2F3 and C6H7 bind topographically unrelated B cell epitopes. f F4B6 recognize a B cell epitope that is topographically different from the B cell epitopes recognized by I4D3, A1D9, P3H11, B3H8, and J2C11.
MOLECULAR CHARACTERIZATION OF ANTIBODY RESPONSE TO PspA
FIGURE 8. Model proposed for the location of the B cell epitope recognized by anti-PspA3–286 mAbs. The mAbs P1C7 and E3D1 (A) and D2F3 and C6H7 (B) are shown. Open and filled bars represent PspA3–286 and B cell epitopes recognized by the indicated mAbs, respectively.
of groups 7 and 8 (Table VI). The anti-PspA3–286 mAbs P1C7 and E3D1 recognized topographically unrelated epitopes located within PspA3–192 (Table VI, group 7). P1C7 and E3D1 were found to recognize the same or overlapping B cell epitopes when tested in combination with other mAbs within the group (i.e., B4G5, C3D5, F4B1, G3B5, and B5E11). It can be speculated that the mAbs P1C7 and E3D1 recognize nonoverlapping B cell epitopes, but the B cell epitopes they recognize overlap with the epitopes recognized by B4G5, C3D5, F4B1, G3B5, and B5E11 (Fig. 8A). In the second case, the anti-PspA3–286 mAbs D2F3 and C6H7 recognized two different B cell epitopes within PspA3–192 (Table VI, group 8). These two mAbs, when tested in combination with the remaining members of the group (i.e., B3D12 and L5C8), were found to recognize the same or overlapping B cell epitopes. As in the first case, the mAbs D2F3 and C6H7 recognize nonoverlapping regions of PspA3–192 but recognize B cell epitopes that overlap with the epitopes recognized by B3D12 and L5C8 (Fig. 8B). The anti-PspA3–286 mAbs I4D3, A1D9, P3H11, B3H8, F4B6, and J2C11 belonging to group 9 (Table VI) showed unexpected A1 ⫹ 2 values in ELISA additivity assays. The anti-PspA3–286 mAb F4B6, when assayed in combination with the other five mAbs within group 9, indicated that the B cell epitope recognized by F4B6 was topographically distinct from the one recognized by the other mAbs within the group (Table VI). Interestingly, when pairs from the remaining five mAbs were analyzed, the A1 ⫹ 2 values were found to be closer in OD value to the lower of the two OD values obtained when the mAbs in the test pair were assayed individually. At this point of time, we do not have an explanation for this unexpected observation.
Discussion Advances in molecular biology have enabled molecular characterization of the Ab response. Understanding the molecular diversity of the Ab responses to PspA, a highly protective pneumococcal Ag, has assumed importance as efforts are on to develop a PspAbased pneumococcal vaccine (11, 12, 37, 38). Earlier studies done using a limited number of anti-PspA mAbs have provided some information regarding the location of protection-eliciting B cell epitopes. However, these studies did not investigate the molecular characteristics of the Abs induced in response to immunization with PspA. This study provides a comprehensive analysis of the anti-PspA Ab response at the molecular level. Paratope diversity A diverse set of H and L chains was used in the anti-PspA3–286 Ab response (Tables II–IV). Some preference for the VH1 and VH14 families was observed in the anti-PspA3–286 Ab response (Fig. 1). The V family usage was more diverse than the VH family usage (Fig. 2 and Table II). The VHSM7.a3.93 gene, a VH14 family
member, was expressed in seven (19.4%) of the anti-PspA3–286 mAbs (Table II and III). Restricted IgV family/gene usage has been observed in several anti-hapten and anti-peptide Ab responses (39 – 44). In contrast to these studies, the anti-PspA3–286 response is highly diverse. In this regard, the anti-PspA Ab response also differs from the oligoclonal Ab responses observed against the bacterial Ags phosphorylcholine (45) and capsular polysaccharide (46). Paratope diversity is expected to be high, as the anti-PspA3–286 Ab response recognized at least 12 B cell epitopes in PspA3–286 (Table VI). CDR3 length and composition The broad range of diversity available to HCDR3 and its location at the center of the Ag-binding site permit HCDR3 to often play a significant role in Ag recognition and binding. In the antiPspA3–286 Ab response, the HCDR3 length ranged from three to 13 codons, nine being the most common (Fig. 3A). This is in contrast to the mouse preimmune repertoire, where the HCDR3 length ranges from five to 20 codons (average ⬃12.5 codons) (36). It is well established that upon encountering the Ag the HCDR3 length range typically narrows down, as a restricted set of CDR3 lengths provides optimum Ag binding. The HCDR3 length range observed in the anti-PspA3–286 Ab response is consistent with this notion. Hydrophobic HCDR3s are found in immature B cells, and this population is selectively lost in mature circulating cells. Ig molecules that bind B cell epitopes found on the surface of soluble Ags by necessity tend to be hydrophilic. Various studies have speculated that highly hydrophobic HCDR3 are thus less likely to create optimal paratopes (36, 47, 48). Schroeder and colleagues demonstrated that the mouse mature B cell repertoire uses DH RFs that are rich in the amino acids tyrosine and glycine (49). Despite the differences at the sequence level, the HCDR3 of all anti-PspA3–286 mAbs displayed a predominance of tyrosine and glycine (Table III). Tyrosine and glycine residues constitute 34.7% of the total HCDR3 residues present in 36 anti-PspA3–286 mAbs. Mechanism of generation of novel H chains that lack DH gene-derived amino acids Unlike HCDR1 and HCDR2, which are entirely encoded by the VH gene segment, HCDR3 is created de novo by VHDHJH rearrangement. Imprecision in joining these gene segments permits exonucleolytic loss and/or N/P nucleotide addition(s) at the VH to DH and the DH to JH junctions. The terminal deoxynucleotidyl transferase-catalyzed insertion of N nucleotides at the sites of joining permits the inclusion of N nucleotides into HCDR3. Together, these mechanisms create a HCDR3 repertoire that ranges from unmodified and intact germline-encoded sequence to rearrangements where extensive nibbling and N/P nucleotide addition(s) no longer permit exact identification of the original DH gene. Analysis of the H chain expressed in our anti-PspA3–286 hybridomas revealed unusual features. The H chain expressed in seven of the 36 (19.4%) anti-PspA3–286 hybridomas lacked DH gene-derived amino acids. Antibodies lacking DH gene-derived amino acids are highly unusual and are rarely observed. Maizels and Bothwell reported the nucleotide sequence of a functional gene (22.11) that lacked the DH gene-derived amino acids (50). No N and P nucleotide addition was seen in the VH to JH junction. This is very similar to what we observed for D2F3-H (Fig. 4). Morzycka-Wroblewska et al. proposed a possible mechanism of generation of H chains that lack DH gene-derived amino acids (51). According to their model (termed signal sequence replacement), recombination signal sequence replacement may lead to direct VH to JH joining in keeping with the 12/23 rule (51). Their model proposes that signal sequence replacement may result from the alternative resolution of
The Journal of Immunology an intermediate in the VH to DHJH recombination. This type of rearrangement is thus a means to alter the targeting of Ig gene segments and suggests a mechanism for the occurrence of VH-JH junctions in vivo. Signal sequence replacement is believed to be conducted by the same machinery that normally carries out the joining of the Ig gene segments because the same specific sites are involved, namely the Ig coding segments and their accompanying recombinational signals. The existence of H chain genes that lack DH gene-derived amino acids appears to be consistent with the occurrence of signal sequence replacement in vivo. Direct in vivo VH to JH rearrangement violating the 12/23 rule has been demonstrated by Koralov et al. in a mouse strain lacking canonical DH gene segments in its IgH locus (52). In these DH gene-deficient mice there was no preferential usage of any particular VH gene family or JH element in VHJH junctions, indicating that 23/23-guided recombination is possible but is a low frequency event at the IgH locus in vivo. We observed N nucleotide(s) addition in six of the seven anti-PspA3–286 hybridomas that lacked DH gene-derived amino acids, D2F3 being the exception (Fig. 4). We reanalyzed the IgH V region junction data from B cells of the DH gene-deficient mice reported by Koralov et al. Our analysis of their data indicated that of the 29 H chain sequences that were devoid of DH gene-derived amino acids, 27.6% lacked both N and P nucleotide additions, 69.0% had only N nucleotide additions, and a lone sequence (3.4%) had both N and P nucleotide additions at the junction. It is curious to note that P nucleotide additions are essentially absent in H chain sequences that lack DH gene-derived amino acid residues. We are not certain whether the absence of P nucleotide additions in the H chain sequences that lack DH genederived amino acids is a consequence of the mechanism used to generate these novel H chain sequences. Apart from the two above-mentioned possible mechanisms, H chains devoid of DH gene-derived amino acids can conceivably also arise if the complete DH gene is chewed up by exonuclease activity during VHDHJH rearrangement. Morzycka-Wroblewska et al. have speculated about the possibility that the rearranged H chain genes that lack DH gene-derived amino acids are not always necessarily the result of direct VH to JH joining, in that VH to DHJH joining can be so imprecise as to completely remove the DH sequence (51). The three above-mentioned mechanisms proposed to explain the observed expressed H chains that were devoid of DH gene-derived amino acids operate during VHDHJH rearrangement in bone marrow B cell precursors. We also considered the remote possibility that the DH gene-derived amino acids were lost or deleted as a post-VHDHJH rearrangement event. We considered the possibility that the novel H chain sequences described above might have been mediated by a heptamer-like element that is present within the VH coding segment near its 3⬘ end. H chain receptor editing, however, does not appear to be the possible mechanism involved in the generation of these novel H chain sequences, as the receptor editing site is located upstream of the DH portion and cannot possibly result in the loss or deletion of the DH gene-derived amino acids from the VHDHJH rearrangement. Gene conversion was also explored as a possible mechanism, as it has the potential to bring about the addition or deletion of several nucleotides at a stretch in the rearranged IgV gene. Gene conversion is frequently used in species such as rabbit and chicken as a post-VHDHJH rearrangement mechanism for diversification of the rearranged IgV genes, but it has rarely been reported in the mouse system. In rabbit and chicken, the gene conversion events are restricted to the VH portion of the rearranged IgV genes and have not been documented to occur in the DH portion of the VHDHJH rearrangement. Therefore, gene conversion-like H chain receptor editing is not a likely mech-
5583 anism for the generation of expressed H chains that lack DH genederived amino acids. We believe it is highly likely that this phenomenon is exclusively associated with defective VHDHJH recombination in bone marrow B cell precursors. Interestingly, one could identify the DH gene despite the very short HCDR3 length (four codons) in two anti-PspA3–286 hybridomas. By contrast, the seven anti-PspA3–286 hybridomas that expressed the H chain devoid of DH gene-derived amino acids had HCDR3 lengths in the range of 4 –7 codons (Tables II and III). No significant difference was observed in the relative avidities for PspA3–286 between the mAbs that had DH gene-derived amino acids compared with ones that lacked it, suggesting that the contribution of DH gene-derived amino acids to HCDR3 may not be critical in the recognition of and binding to PspA3–286 (Fig. 5). In general, H chains that lack DH gene-derived amino acids are encountered rarely. The fact that ⬃20% of the anti-PspA3–286 hybridomas expressed an H chain that lacked DH gene-derived amino acids may have something to do with the structure of the PspA in general and the specific B cell epitope(s) being recognized by these mAbs in particular. V gene usage, Ag, and avidity Preferential usage of the VHSM7.a3.93 gene (19.4%), a VH14 family member, was observed in anti-PspA3–286 Ab response, although the family contributes only 1.4% of the germline VH genes. The seven anti-PspA3–286 hybridomas that rearranged the VHSM7.a3.93 gene were derived from three different donors. The VHSM7.a3.93 gene was used in two primary and five tertiary antiPspA3–286 mAbs. In these seven hybridomas, the VHSM7.a3.93 gene recombined with different DH and JH genes, resulting in different HCDR3 lengths (Table II). Thus, although the usage of VHSM7.a3.93 does not share the characteristics of a canonical response, it may contribute significantly to the PspA3–286 Ab response. During the B cell response to the hapten nitrophenyl, evidence for Ag-driven selection was observed at the L chain, indicating that the V domain plays a major role in nitrophenyl binding (39). In contrast, Ag-driven selection was observed at the L chain during B cell response to the hemagglutinin Ag of the influenza virus, indicating that the V domain plays a major role in hemagglutinin binding (53). Our data indicate that the VH domain plays a major role in Ag binding in the anti-PspA3–286 Ab response. Somatic hypermutation events were analyzed for the H and L chains expressed in the anti-PspA3–286 hybridomas. It was observed that the three anti-PspA3–286 tertiary IgM mAbs raised from PspA3–286-immunized mice had an average of 13.3 mutations. Contrastingly, five of the seven anti-PspA3–286 tertiary IgM mAbs raised from heat-killed R36A-immunized mice were essentially germline (average number of mutations ⫽ 0.4). In this study, a mAb was considered to be germline if the H and L chain genes had ⱕ1 mutation relative to the corresponding germline gene. Interestingly, the remaining two anti-PspA3–286 tertiary IgM mAbs raised in heat-killed R36A-immunized mice were mutated (average number of mutations ⫽ 11.5). Naive B cells recruited in the tertiary response can result in germline IgM tertiary mAbs. It can be further speculated that, during evolution, the naive B cells expressing the V genes we observed had been selected due to their intrinsic ability to bind multiple molecules of PspA present on the pneumococcal cell surface, resulting in cross-linking of the surface IgM receptor and secretion of IgM germline Abs. No association was found between the usage of a specific V gene or somatic hypermutation and avidity. Comparison of the amino acid sequence of the expressed H and L chains did not reveal any motif
MOLECULAR CHARACTERIZATION OF ANTIBODY RESPONSE TO PspA
or set of residues that was shared among all or a majority of the 36 anti-PspA3–286 mAbs (Tables III and IV). The relative avidities of the primary anti-PspA3–286 IgG polyclonal Abs induced in response to immunization with heat-killed R36A were found to be higher than the primary polyclonal IgG Abs induced in response to PspA3–286 on days 72, 86, and 100 postimmunization (Fig. 6). One possible explanation could be that, unlike recombinant PspA3–286, heat-killed bacteria have components that can engage TLRs. These TLR ligands affect dendritic cell Ag presentation in ways similar to that of an adjuvant. These effects include improvement in Ag presentation and an increase in costimulatory molecules and cytokine production, leading usually to improved TH1-related responses. Such responses are well suited to defend against bacteria, probably because TLRs have been selected through evolution to respond to these infections and their attendant, intrinsic adjuvants (54). In contrast, alum, a synthetic adjuvant, is not characteristic of the targeted organism and is not recognized by TLRs. In this case evolutionary selection has not had a chance to play its part, and the artificially included adjuvant may not induce the appropriate response. Alum activates innate immune responses in vivo and promotes a Th2-biased response and elevated titers of the Th2-dependent Ab isotypes IgG1 and IgE. Dendritic cells exposed directly to alum do not fully up-regulate costimulatory molecules and do not produce Th1-driving cytokines, the canonical changes to dendritic cells induced by TLR ligands (54). If the higher avidity of the primary anti-PspA3–286 IgG polyclonal Abs raised against heat-killed R36A compared with Abs induced with PspA3–286 is due to the contribution of TLR ligand(s), it would be predicted that adding certain TLR ligand(s) to PspA3–286 should induce better relative avidities. For example, oligodeoxynucleotides containing unmethylated CpG motifs have been shown to enhance avidity in anti-anthrax (55) and human anti-hepatitis B vaccine responses (56). PspA and B cell epitopes The 36 anti-PspA3–286 mAbs were grouped into 12 independent groups based on their reactivity pattern with PspA3–286 and its five deletion mutants (Table VI). The B cell epitope recognized by 18 of the 36 (50%) anti-PspA3–286 mAbs localized to the region corresponding to aa 3–192 of PspA. The B cell epitopes recognized by the remaining 18 mAbs are spread across the surface-exposed portion of PspA (aa residues 1–288), which is consistent with previous studies (21, 23, 37). Anti-PspA mAbs that recognized PspA3–192 did not recognize its subfragments PspA3–97 and PspA98 –192 (Table VI, groups 7 and 8). This result may be due to differences in the structure of PspA3–97 and PspA98 –192 vis-a`-vis the corresponding regions in PspA3–192. McDaniel et al. found that B cell epitopes recognized by nine anti-PspA mAbs localized to two portions of the ␣-helical region of PspA. One region comprised the first 115 amino acids of PspA, and the other was from aa 192–260 (21). In another study, Kolberg et al. located the B cell epitopes to two regions (aa 1– 67 and aa 67–236) that essentially cover the extracellular domain (23). Recently, Miyaji and colleagues demonstrated that aa 32–212 (region A) and aa 213–312 (region B) of PspA are equally immunogenic and that immunization of mice with these subfragments induced high Ab titers. Earlier mapping studies using anti-PspA mAbs have shown that regions A and B of PspA induce protective Ab responses (37). Roche et al. demonstrated that a combination of subfragments from both ends of the ␣-helical region elicited the best protection (9). This result suggests that regions A and B are important not only for the immunogenicity of PspA but also for protection against pneumococcal challenge.
All of the mAbs reacted with PspA3–286 and R36A lysate in Western blotting that was done following transfer from SDSPAGE-resolved samples. We believe that it may be premature to conclude that the mAbs recognize linear epitopes within PspA3–286, as heat treatment of PspA3–286 at 100°C for 2 h did not diminish the reactivity of the four randomly selected antiPspA3–286 mAbs (C6H7, A7A1, C3D5, and F4B1) and polyclonal anti-PspA3–286 anti-serum when tested by ELISA (data not shown). This is most likely due to the very high stability of the coiled-coil structure of PspA; heat-denatured PspA3–286 spontaneously reverts to its native structure as soon as the temperature is brought back to room temperature. The anti-PspA3–286 mAbs were raised from the BALB/c strain of mice with the exception of C6H7, A7A1, C3D5, and F4B1, which were raised from CBA/J mice. Although only four mAbs were of CBA/J mice origin, we did not observe any significant difference in any of the parameters reported in this study. In summary, our molecular and genetic analyses of the PspA3–286 Ab repertoire describe a response that involves the recruitment of B cells that express diverse VH and V genes. An unexpected finding from our study was the observation that seven of the 36 anti-PspA3–286 mAbs expressed a H chain that lacks DH gene-derived amino acids. The absence of DH gene-derived amino acids does not appear to prevent anti-PspA3–286 mAbs from attaining meaningful or average relative avidity. We found that the relative avidities of the primary IgG polyclonal Ab elicited in response to PspA3–286 is higher when the protein is present on the surface of the bacteria as compared with when it is provided as recombinant protein. We observed evidence for clonal expansion and Ag-driven selection in the anti-PspA3–286 mAbs. Based on deletion mapping analysis and an ELISA additivity assay, the antiPspA3–286 Ab response recognizes at least 12 B cell epitopes. The mouse anti-PspA3–286 Ab response presents an opportunity to understand the biology of expressed H chains that lack DH genederived amino acids. Further studies will be needed to fine map the epitopes recognized by these mAbs and to address whether the higher relative avidities of the primary IgG PspA3–286 polyclonal Abs elicited in response to heat-killed R36A correlate with the enhanced protective immunity against lethal pneumococcal challenge.
Acknowledgments We acknowledge K. K. Sarin for expert technical assistance. We thank Drs. Anna George, Ayub Qadri, and Rahul Pal for suggestions and comments on the manuscript.
Disclosures The authors have no financial conflict of interest.
References 1. World Health Organization. 2007. Pneumococcal conjugate vaccine for childhood immunization – WHO position paper. Wkly. Epidemiol. Rec 82: 93–104. 2. Tai, S. S. 2006. Streptococcus pneumoniae protein vaccine candidates: properties, activities and animal studies. Crit. Rev. Microbiol. 32: 139 –153. 3. Giefing, C., A. L. Meinke, M. Hanner, T. Henics, M. D. Bui, D. Gelbmann, U. Lundberg, B. M. Senn, M. Schunn, A. Habel, et al. 2008. Discovery of a novel class of highly conserved vaccine antigens using genomic scale antigenic fingerprinting of pneumococcus with human antibodies. J. Exp. Med. 205: 117–131. 4. Kadioglu, A., J. N. Weiser, J. C. Paton, and P. W. Andrew. 2008. The role of Streptococcus pneumoniae virulence factors in host respiratory colonization and disease. Nat Rev Microbiol 6: 288 –301. 5. Briles, D. E., S. K. Hollingshead, J. C. Paton, E. W. Ades, L. Novak, F. W. van Ginkel, and W. H. Benjamin, Jr. 2003. Immunizations with pneumococcal surface protein A and pneumolysin are protective against pneumonia in a murine model of pulmonary infection with Streptococcus pneumoniae. J. Infect. Dis. 188: 339 –348. 6. Tart, R. C., L. S. McDaniel, B. A. Ralph, and D. E. Briles. 1996. Truncated Streptococcus pneumoniae PspA molecules elicit cross-protective immunity against pneumococcal challenge in mice. J. Infect. Dis. 173: 380 –386.
The Journal of Immunology 7. Briles, D. E., J. D. King, M. A. Gray, L. S. McDaniel, E. Swiatlo, and K. A. Benton. 1996. PspA, a protection-eliciting pneumococcal protein: immunogenicity of isolated native PspA in mice. Vaccine 14: 858 – 867. 8. Briles, D. E., S. K. Hollingshead, G. S. Nabors, J. C. Paton, and A. Brooks-Walter. 2000. The potential for using protein vaccines to protect against otitis media caused by Streptococcus pneumoniae. Vaccine 19(Suppl 1): S87–S95. 9. Roche, H., A. Hakansson, S. K. Hollingshead, and D. E. Briles. 2003. Regions of PspA/EF3296 best able to elicit protection against Streptococcus pneumoniae in a murine infection model. Infect. Immun. 71: 1033–1041. 10. Roche, H., B. Ren, L. S. McDaniel, A. Hakansson, and D. E. Briles. 2003. Relative roles of genetic background and variation in PspA in the ability of antibodies to PspA to protect against capsular type 3 and 4 strains of Streptococcus pneumoniae. Infect. Immun. 71: 4498 – 4505. 11. Nabors, G. S., P. A. Braun, D. J. Herrmann, M. L. Heise, D. J. Pyle, S. Gravenstein, M. Schilling, L. M. Ferguson, S. K. Hollingshead, D. E. Briles, and R. S. Becker. 2000. Immunization of healthy adults with a single recombinant pneumococcal surface protein A (PspA) variant stimulates broadly cross-reactive antibodies to heterologous PspA molecules. Vaccine 18: 1743–1754. 12. Briles, D. E., S. K. Hollingshead, J. King, A. Swift, P. A. Braun, M. K. Park, L. M. Ferguson, M. H. Nahm, and G. S. Nabors. 2000. Immunization of humans with recombinant pneumococcal surface protein A (rPspA) elicits antibodies that passively protect mice from fatal infection with Streptococcus pneumoniae bearing heterologous PspA. J. Infect. Dis. 182: 1694 –1701. 13. Ochs, M. M., W. Bartlett, D. E. Briles, B. Hicks, A. Jurkuvenas, P. Lau, B. Ren, and A. Millar. 2008. Vaccine-induced human antibodies to PspA augment complement C3 deposition on Streptococcus pneumoniae. Microb. Pathog. 44: 204-214. 14. Ren, B., A. J. Szalai, S. K. Hollingshead, and D. E. Briles. 2004. Effects of PspA and antibodies to PspA on activation and deposition of complement on the pneumococcal surface. Infect. Immun. 72: 114 –122. 15. Shaper, M., S. K. Hollingshead, W. H. Benjamin, Jr., and D. E. Briles. 2004. PspA protects Streptococcus pneumoniae from killing by apolactoferrin, and antibody to PspA enhances killing of pneumococci by apolactoferrin (corrected). Infect. Immun. 72: 5031–5040. 16. Yother, J., and J. M. White. 1994. Novel surface attachment mechanism of the Streptococcus pneumoniae protein PspA. J. Bacteriol. 176: 2976 –2985. 17. Yother, J., and D. E. Briles. 1992. Structural properties and evolutionary relationships of PspA, a surface protein of Streptococcus pneumoniae, as revealed by sequence analysis. J. Bacteriol. 174: 601– 609. 18. Hollingshead, S. K., R. Becker, and D. E. Briles. 2000. Diversity of PspA: mosaic genes and evidence for past recombination in Streptococcus pneumoniae. Infect. Immun. 68: 5889 –5900. 19. Mollerach, M., M. Regueira, L. Bonofiglio, R. Callejo, J. Pace, J. L. Di Fabio, S. Hollingshead, and D. Briles. 2004. Invasive Streptococcus pneumoniae isolates from Argentinian children: serotypes, families of pneumococcal surface protein A (PspA) and genetic diversity. Epidemiol. Infect. 132: 177–184. 20. Vela Coral, M. C., N. Fonseca, E. Castaneda, J. L. Di Fabio, S. K. Hollingshead, and D. E. Briles. 2001. Pneumococcal surface protein A of invasive Streptococcus pneumoniae isolates from Colombian children. Emerg. Infect. Dis. 7: 832– 836. 21. McDaniel, L. S., B. A. Ralph, D. O. McDaniel, and D. E. Briles. 1994. Localization of protection-eliciting epitopes on PspA of Streptococcus pneumoniae between amino acid residues 192 and 260. Microb. Pathog. 17: 323–337. 22. Kolberg, J., A. Aase, T. E. Michaelsen, and G. Rodal. 2001. Epitope analyses of pneumococcal surface protein A: a combination of two monoclonal antibodies detects 94% of clinical isolates. FEMS Immunol. Med. Microbiol. 31: 175–180. 23. Kolberg, J., A. Aase, G. Rodal, J. E. Littlejohn, and M. J. Jedrzejas. 2003. Epitope mapping of pneumococcal surface protein A of strain Rx1 using monoclonal antibodies and molecular structure modelling. FEMS Immunol. Med. Microbiol. 39: 265–273. 24. Wu, Z. Q., Y. Shen, A. Q. Khan, C. L. Chu, R. Riese, H. A. Chapman, O. Kanagawa, and C. M. Snapper. 2002. The mechanism underlying T cell help for induction of an antigen-specific in vivo humoral immune response to intact Streptococcus pneumoniae is dependent on the type of antigen. J. Immunol. 168: 5551–5557. 25. Galfre, G., and C. Milstein. 1981. Preparation of monoclonal antibodies: strategies and procedures. Methods Enzymol. 73: 3– 46. 26. Rohatgi, S., P. Ganju, and D. Sehgal. 2008. Systematic design and testing of nested (RT-)PCR primers for specific amplification of mouse rearranged/expressed immunoglobulin variable region genes from small number of B cells. J. Immunol. Methods 339: 205–219. 27. Lefranc, M. P., V. Giudicelli, Q. Kaas, E. Duprat, J. Jabado-Michaloud, D. Scaviner, C. Ginestoux, O. Clement, D. Chaume, and G. Lefranc. 2005. IMGT, the international ImMunoGeneTics information system. Nucleic Acids Res. 33: D593–D597. 28. Kabat, E. A., T. T. Wu, H. M. Perry, K. S. Gotterman, and C. Foeller, C. 1991. Sequences of Proteins of Immunological Interest, 5th Ed. U.S. Department of Health and Human Services, Bethesda, MD. 29. Rath, S., C. M. Stanley, and M. W. Steward. 1988. An inhibition enzyme immunoassay for estimating relative antibody affinity and affinity heterogeneity. J. Immunol. Methods 106: 245–249.
5585 30. Devey, M. E., K. Bleasdale, S. Lee, and S. Rath. 1988. Determination of the functional affinity of IgG1 and IgG4 antibodies to tetanus toxoid by isotypespecific solid-phase assays. J. Immunol. Methods 106: 119 –125. 31. Devey, M. E., K. M. Bleasdale-Barr, P. Bird, and P. L. Amlot. 1990. Antibodies of different human IgG subclasses show distinct patterns of affinity maturation after immunization with keyhole limpet haemocyanin. Immunology 70: 168 –174. 32. van Dam, G. J., A. F. Verheul, G. J. Zigterman, M. J. de Reuver, and H. Snippe. 1989. Estimation of the avidity of antibodies in polyclonal antisera against Streptococcus pneumoniae type 3 by inhibition ELISA. Mol. Immunol. 26: 269 –274. 33. Huang, H., R. Zhou, H. Fan, H. Dan, M. Chen, L. Yan, W. Bei, and H. Chen. 2006. Generation of monoclonal antibodies and epitope mapping of ApxIVA of Actinobacillus pleuropneumoniae. Mol. Immunol. 43: 2130 –2134. 34. Persson, M. A., S. E. Brown, M. W. Steward, L. Hammarstrom, C. I. Smith, C. R. Howard, M. Wahl, B. Rynnel-Dagoo, G. Lefranc, and A. O. Carbonara. 1988. IgG subclass-associated affinity differences of specific antibodies in humans. J. Immunol. 140: 3875-3879. 35. Lossos, I. S., R. Tibshirani, B. Narasimhan, and R. Levy. 2000. The inference of antigen selection on Ig genes. J. Immunol. 165: 5122–5126. 36. Schelonka, R. L., J. Tanner, Y. Zhuang, G. L. Gartland, M. Zemlin, and H. W. Schroeder, Jr. 2007. Categorical selection of the antibody repertoire in splenic B cells. Eur. J. Immunol. 37: 1010 –1021. 37. Darrieux, M., A. T. Moreno, D. M. Ferreira, F. C. Pimenta, A. L. de Andrade, A. P. Lopes, L. C. Leite, and E. N. Miyaji. 2008. Recognition of pneumococcal isolates by antisera raised against PspA fragments from different clades. J. Med. Microbiol. 57: 273–278. 38. Bogaert, D., P. W. Hermans, P. V. Adrian, H. C. Rumke, and R. de Groot. 2004. Pneumococcal vaccines: an update on current strategies. Vaccine 22: 2209 –2220. 39. Tao, W., and A. L. Bothwell. 1990. Development of B cell lineages during a primary anti-hapten immune response. J. Immunol. 145: 3216 –3222. 40. Cumano, A., and K. Rajewsky. 1986. Clonal recruitment and somatic mutation in the generation of immunological memory to the hapten NP. EMBO J. 5: 2459 –2468. 41. Solin, M. L., M. Kaartinen, and O. Makela. 1992. The same few V genes account for a majority of oxazolone antibodies in most mouse strains. Mol. Immunol. 29: 1357-1362. 42. White, H., and D. Gray. 2000. Analysis of immunoglobulin (Ig) isotype diversity and IgM/D memory in the response to phenyl-oxazolone. J. Exp. Med. 191: 2209 –2220. 43. Busto, P., R. Gerstein, L. Dupre, C. A. Giorgetti, E. Selsing, and J. L. Press. 1987. Molecular analysis of heavy and light chains used by primary and secondary anti-(T,G)-A–L antibodies produced by normal and Xid mice. J. Immunol. 139: 608 – 618. 44. Press, J. L., and C. A. Giorgetti. 1986. Clonal analysis of the primary and secondary B cell responses of neonatal, adult, and Xid mice to (T,G)-A–L. J. Immunol. 137: 784 –790. 45. Claflin, J. L., and J. Berry. 1988. Genetics of the phosphocholine-specific antibody response to Streptococcus pneumoniae. Germ-line but not mutated T15 antibodies are dominantly selected. J. Immunol. 141: 4012– 4019. 46. Baxendale, H. E., Z. Davis, H. N. White, M. B. Spellerberg, F. K. Stevenson, and D. Goldblatt. 2000. Immunogenetic analysis of the immune response to pneumococcal polysaccharide. Eur. J. Immunol. 30: 1214-1223. 47. Schroeder, H. W., Jr., G. C. Ippolito, and S. Shiokawa. 1998. Regulation of the antibody repertoire through control of HCDR3 diversity. Vaccine 16: 1383–1390. 48. Raaphorst, F. M., C. S. Raman, B. T. Nall, and J. M. Teale. 1997. Molecular mechanisms governing reading frame choice of immunoglobulin diversity genes. Immunol. Today 18: 37– 43. 49. Ippolito, G. C., R. L. Schelonka, M. Zemlin, I. I. Ivanov, R. Kobayashi, C. Zemlin, G. L. Gartland, L. Nitschke, J. Pelkonen, K. Fujihashi, et al. 2006. Forced usage of positively charged amino acids in immunoglobulin CDR-H3 impairs B cell development and antibody production. J. Exp. Med. 203: 1567–1578. 50. Maizels, N., and A. Bothwell. 1985. The T-cell-independent immune response to the hapten NP uses a large repertoire of heavy chain genes. Cell 43: 715–720. 51. Morzycka-Wroblewska, E., F. E. Lee, and S. V. Desiderio. 1988. Unusual immunoglobulin gene rearrangement leads to replacement of recombinational signal sequences. Science 242: 261–263. 52. Koralov, S. B., T. I. Novobrantseva, K. Hochedlinger, R. Jaenisch, and K. Rajewsky. 2005. Direct in vivo VH to JH rearrangement violating the 12/23 rule. J. Exp. Med. 201: 341–348. 53. Clarke, S. H., L. M. Staudt, J. Kavaler, D. Schwartz, W. U. Gerhard, and M. G. Weigert. 1990. V region gene usage and somatic mutation in the primary and secondary responses to influenza virus hemagglutinin. J. Immunol. 144: 2795–2801. 54. McKee, A. S., M. W. Munks, and P. Marrack. 2007. How do adjuvants work? Important considerations for new generation adjuvants. Immunity 27: 687– 690. 55. Klinman, D. M., H. Xie, S. F. Little, D. Currie, and B. E. Ivins. 2004. CpG oligonucleotides improve the protective immune response induced by the anthrax vaccination of rhesus macaques. Vaccine 22: 2881–2886. 56. Siegrist, C. A., M. Pihlgren, C. Tougne, S. M. Efler, M. L. Morris, M. J. AlAdhami, D. W. Cameron, C. L. Cooper, J. Heathcote, H. L. Davis, and P. H. Lambert. 2004. Co-administration of CpG oligonucleotides enhances the late affinity maturation process of human anti-hepatitis B vaccine response. Vaccine 23: 615– 622.