Immune Response to Pneumococcal Polysaccharides 4 and 14 in Elderly and Young Adults: Analysis of the Variable Light Chain Repertoire
http://www.100md.com
感染与免疫杂志 2005年第11期
Department of Medicine, Medical University of Ohio, Toledo, Ohio
ABSTRACT
Streptococcus pneumoniae is a human bacterial pathogen responsible for serious infections including pneumonia. The currently licensed polysaccharide vaccine provides 60 to 80% protection in young adults, but in the elderly the vaccine efficacy is drastically reduced despite normal antibody levels. We hypothesized that the reduced vaccine efficacy in the elderly results from altered variable gene family usage. We have analyzed the light chain gene usage in 20 young (20 to 30 years of age) and 20 elderly (65 to 86 years of age) adults in response to pneumococcal polysaccharide 4 (PPS4) and PPS14. We generated a variable light chain library using B cells specific for PPS4 and PPS14 from each vaccinated individual. We determined complete sequences and somatic mutation frequencies in all isolated variable light chain fragments. Six gene families, 1, 2, 3, 4, 1, and 3, were identified in response to PPS4 and PPS14 in both age groups. Comparison of young and elderly adults demonstrated significant differences in 4, 1, and 3 gene usage in response to PPS4 and PPS14. With aging, there was a significant increase in 4 gene usage and a significant decrease in 1 and 3 gene usage in response to both PPS4 and PPS14. Although both V1 and V3 gene products demonstrated extensive mutations, there was no age-related difference in mutational frequency per gene family. These findings suggest an age-related change in light chain gene usage in response to PPS4 and PPS14.
INTRODUCTION
The incidence of pneumococcal pneumonia is significantly increased in individuals >65 years of age (13). Several studies (8, 18) have demonstrated a significant impairment in the immune response to pneumococcal polysaccharides in the elderly, particularly those >77 years of age, associated with a markedly decreased efficacy of the pneumococcal vaccine (6, 23).
The molecular mechanisms responsible for the decreased immune response in the elderly remain poorly understood. Nicoletti et al. (15) studied the immune response to phosphorylcholine (PC) in aged mice vaccinated with Streptococcus pneumoniae. These studies demonstrated a significant functional impairment in the immune response to PC despite normal antibody concentrations. Molecular analysis of the anti-PC-specific antibody repertoire in aged mice revealed a significantly altered V gene usage compared to young immunized mice (14).
To date, studies examining the structure-function relationship of antipneumococcal antibodies have been performed using human monoclonal antibodies (1, 22, 26), combinatorial libraries (12, 17, 27, 28), and monoclonal antibodies derived from a transgenic mouse strain with human immunoglobulin loci (2, 19). All studies, with the exception of those using the transgenic mouse, have analyzed the V repertoire using peripheral blood mononuclear cells obtained from young high responders. Although these studies have enriched our knowledge concerning the human immune response to pneumococcal polysaccharide (PPS), they were not designed to represent the in vivo antibody repertoire response to PPS in the elderly or the adult population at large, by excluding intermediate and low responders.
We have studied the immune response to PPS4 and PPS14 in 20 young (20 to 30 years of age) and 20 elderly (>65 years of age) volunteers. In this study, we have compared the antibody concentrations and opsonophagocytic antibody responses as well as levels of gene family usage in an effort to increase our understanding of the aging immune response to pneumococcal polysaccharides. We have previously reported a significant decrease in opsonophagocytic activity in elderly persons >77 years of age, as reported by others (8, 18). Comparative analysis of the variable heavy chain (VH) gene usage in the young and elderly demonstrated significant age-related differences in VH gene usage in response to both PPS4 and PPS14. Furthermore, a significant loss of oligoclonality, a characteristic of the immune response to polysaccharide antigens, was noted in the elderly (7). The results of the present study demonstrates a wide variety of variable light chain (VL) gene usage in response to both polysaccharides, although the response was dominated by six light chain gene families, 1, 3, 1, 2, 3, and 4. Despite common VL gene family usage, significant differences between young and elderly volunteers were noted in the distribution of these VL gene families.
MATERIALS AND METHODS
Human volunteers and vaccination. Only healthy young adults (20 to 30 years old) and healthy elderly adults (>65 years old) were asked to participate in this study. The age of the elderly ranged from 65 to 86 years, with 14 volunteers in the 65- to 77-year age group and 6 volunteers in the 78- to 86-year age group. Elderly volunteers were recruited from the community as well as from the general internal medicine clinic at the Medical University of Ohio (MUOT). Young volunteers were recruited from the student population at MUOT. Each individual was questioned concerning prior pneumococcal vaccination, medications, previous illness, and present health. In addition, we obtained complete blood count (CBC); a comprehensive chemistry profile including renal and liver functions and serum albumin; total B cells; T-cell subsets; and total immunoglobulin G (IgG), IgM, and IgA levels in all study candidates. Individuals previously immunized with the pneumococcal vaccine and any individual considered to be immunocompromised on the basis of medication (chemotherapy, steroid preparations, immunosuppressive agents including anti-tumor necrosis factor alpha agents), previous or present illness (previous pneumococcal disease, splenectomy, autoimmune disease, end stage renal or liver disease, human immunodeficiency virus positivity, organ transplant, or cancer) and those with abnormal CBC, chemistries, B or T cells, or immunoglobulin levels did not qualify. Informed consent was obtained from all participants using protocols reviewed and approved by the Institutional Review Board at MUOT.
Purification of PPS-specific B cells. Biotinylated polysaccharides were used to select PPS4- and PPS14-specific B cells as described previously (7). Briefly PPS4 and PPS14 were biotinylated by the method described by Lucas et al. (12). These polysaccharides were then reacted with streptavidin-coated immunomagnetic beads (Dynal, Lake Success, NY) and washed. Peripheral blood lymphocytes collected 6 weeks postvaccination were obtained by Ficoll Histopaque gradient (Sigma, St. Louis, MO). Cross-reactive B cells were depleted from the population by incubation with beads coated with PPS23F and cell wall polysaccharide (CWPS) and discarded. Further depletion of cross-reactive B cells was achieved by cross incubating the B cells; cells for PPS4-specific isolation were first depleted with PPS14-coated beads, and likewise cells for PPS14 isolation were depleted with PPS4-coated beads. Isolation of PPS-specific B cells was achieved by incubation with 25 μg of PPS4-coated immunomagnetic beads or PPS14-coated beads with 10 μg/ml CWPS added to the buffer to block nonspecific binding. These steps were taken to eliminate binding to cross-reactive antigens between PPS. Polysaccharide-specific B cells were isolated by magnetic separation and washed with RPMI 1640-5% newborn calf serum (NBCS). cDNA was prepared from the isolated B cells using a Dynabeads mRNA direct kit by following the manufacturer's instructions and stored in aliquots at –70°C for further use. Validation of the selection method for PPS-specific B cells was performed as previously described (7).
Production of light chain libraries. The cDNA obtained from PPS-selected B cells was used as a template in the PCR to generate light chain libraries. PCR controls consisted of cDNA obtained from unselected B cells. Primer sets described by Welschof et al. (25) and Taq polymerase (Fisher Scientific, Pittsburgh, PA) were used to generate light chain products. The PCR amplification conditions consisted of 32 cycles of 94°C for 45 s, 65°C for 30 s, and 72°C for 45 s. Amplification products were purified using the GeneClean gel extraction kit (Bio101, La Jolla, CA) and ligated into the TA cloning vector system (Invitrogen, San Diego, CA). Ligated plasmids were transformed into Top 10 Escherichia coli cells by chemical transformation. The light chain libraries were plated on Luria broth (LB)-kanamycin-X-Gal (5-bromo-4-chloro-3-indolyl--D-galactopyranoside) plates at low density and grown overnight at 37°C.
Selection and sequence analysis of positive PPS4 and PPS14 clones. Individual E. coli clones were selected and streaked onto a LB-kanamycin master plate and grown overnight at 37°C. These clones were lifted onto nylon filters and fixed by UV exposure for 5 min. The nylon filters were probed with a [-32P]ATP-labeled oligonucleotide with specificity for the and framework III region (25). Sequence analysis was performed on selected clones (MWG Biotech, High Point, NC) using primers complementary to the vector. The resultant sequences were compared to germ line sequences using VBASE DNAPLOT (http://vbase.mrc-cpe.cam.ac.uk).
Statistical analysis. Percentages of light chain gene usage against PPS4 and PPS14 were calculated for each group and volunteer. Fisher exact test and the Pearson chi square value were used to determine significance between gene usage and age groups. Student's t test was used to determine significance between mutational frequencies. A P value equal or less than 0.05 was considered to be significant. Statistical calculations were performed with the use of SPSS software 11.5.1.
Nucleotide sequence accession numbers. CDR sequences are available from GenBank under accession numbers AY928106 and AY928172.
RESULTS
Light chain analysis. The cDNA obtained from PPS-selected B cells was used to generate VL libraries. The cDNA from each individual's cDNA was amplified twice on separate occasions to control for PCR artifacts. Those samples that failed to yield consistent results were eliminated from this analysis. The accuracy of the B-cell selection method has been verified as described previously (7).
Successful sequence analysis was performed on a total of 175 (average of 10.9 VLs/volunteer) light chains with specificity for PPS4 and 202 (average of 12.6 VLs/volunteer) light chains with specificity for PPS14 derived from 17 immunized young volunteers. Sequence analysis was obtained for 222 (average of 13.9 VLs/volunteer) in response to PPS4 and 235 (average of 13.8 VLs/volunteer) with specificity for PPS14 from 16 and 17 elderly volunteers, respectively. All isolated sequences were productively rearranged. Variable light chain gene family complementarity-determining region 3 (CDR3) length and composition, J chain, percent identity to germ line sequence, and somatic mutation frequencies were determined. PCR amplification was performed on unselected B cells on a monthly basis throughout the study and compared to the expected VL gene usage in unselected B cells (3, 4) to ensure unbiased amplification of all gene families.
VL gene response to PPS4 and PPS14 in young adults. All immunized young adults utilized one or more light chains belonging to 6 of the 16 light chain gene families. These six light chain gene families (1, 3, 1, 2, 3, and 4) accounted for 99.9% of the repertoire used in response to PPS4. As shown in Fig. 1, the percent utilization of 1 (27.4%) was significantly greater (P = 0.001) than that reported (9.6%) in unselected IgM+ B cells (3). We utilized IgM+ B cells for comparison as this was the most comprehensive study of the light chain usage in the circulating-B-cell repertoire available. Members of the 1 gene family were also commonly used and represented 21.7% of the VL sequences isolated, but this was not significantly different from peripheral-B-cell expression. Together these two light chain gene families dominated the L chain gene expression in young adults at 49.1% while the 3 (16.6%), 2 (7.4%), 3 (13.7%), and 4 (13.1%) gene families comprised the remainder of the response.
Analysis of the light chain variable gene family response to PPS14 (Fig. 1) revealed that the young volunteers utilized the same six VL gene families specific for PPS4, namely, 1, 3, 1, 2, 3, and 4. The 3 gene family was dominant, expressed in 23.3% of isolated sequence, and significantly overexpressed (P < 0.001) in comparison to peripheral-IgM+-B-cell expression of 5.2% (3). The gene families 1 and 1 were also highly expressed at 22.8% of total sequences obtained from young adult B cells specific for PPS14. The remaining light chains, 2, 3, and 4, accounted for 31.2% of the L chain sequences obtained from young adults.
There was wide variability in the specific VL locus utilized in response to both PPS4 and PPS14 within and between donors (Tables 1 and 2). Overall, 10 loci were identified as being the most commonly expressed VL loci in the 17 donors in response to PPS4 and PPS14, although loci were not identical between the two groups. However, in response to both PPS4 and PPS14, 9 of 17 young donors were found to express a predominant (>50% utilization) gene locus, with the remainder of the donors showing increased variability.
VL gene response to PPS4 and PPS14 in elderly adults. Analysis of L chain gene family usage in immunized elderly adults in response to PPS4 and PPS14 demonstrated similar levels of usage of the 1, 3, 1, 2, 3, and 4 light chain gene families. As shown in Fig. 1, the predominant light chain gene family used in response to both PPS4 and PPS14 was the 4 gene family, 27.9% and 27.2%, respectively. This is in sharp contrast (P < 0.001) to the 3.3% 4 representation found in unselected IgM+ B cells (4). In response to PPS4, the 1 gene family and the 3 gene family were commonly expressed at 20.3% and 21.1%, respectively, with 1, 2, and 3 representing a further 30.6% of L chain gene usage. In response to PPS14, as shown in Fig. 1, in addition to the 4 gene product, the 1 (18.7%) and 3 (19.6%) gene products were commonly expressed L chains by the elderly.
As demonstrated in Tables 1 and 2, specific VL locus usage varied in the elderly. The single most common locus, B3 (4) was represented as the most common locus in 7 of 17 elderly donors against PPS4 and 6 of 17 elderly donors against PPS14. In the elderly, 8 of 17 donors demonstrated an oligoclonal response with expression of a predominant (>50% utilization) locus against PPS4 and 7 of 17 against PPS14. Some volunteers utilized the same gene locus, but not identical sequences, in response to both PPS4 and PPS14. However, in most volunteers, the VL gene families used in response to PPS4 were different than those used in response to PPS14 (Tables 1 and 2).
VL gene response to PPS4: comparison of young versus elderly adults. In response to PPS4, comparison of the young and elderly generally revealed similar responses in gene family expression with the exception of 3 and 4 gene usage. As shown in Fig. 1, there was a significant difference (P = 0.01) between the elderly (8.1%) and young (16.6%) in the expression of the 3 gene family. In contrast, 4 gene usage was significantly higher in the elderly versus the young (P < 0.01). The elderly group displayed predominant expression of 4, 27.9%, in contrast to 13.1% of the VL repertoire isolated from young adults. Although the group size of the oldest population, those >77 years of age (donors 34 to 40), was inadequate to perform statistical analysis, there was a definite trend towards increased use of the 4 gene family and a decreased use of 3 and 1 gene family with increasing age.
VL gene response to PPS14: comparison of young versus elderly adults. As shown in Fig. 1, the immune response to PPS14 in the young consisted predominantly of 1, 3, and 1 gene families. However, with aging a significant shift in the use of VL gene families occurred, with an increase in 4 gene usage (P < 0.001) and a decrease in both 1 (P = 0.025) and 3 (P = 0.001) usage. As shown for PPS4, this was an age-dependent phenomenon, with a trend towards increasing differences with increasing age. Nine percent of young volunteers and 27.2% of elderly volunteers expressed 4 as part of their PPS14-specific antibody repertoire. There was no statistical difference in 1, 2, or 3 gene usage between the young and the elderly in response to PPS14.
CDR composition and germ line identity. Although a wide variety of VL gene products were used, the CDR3 length of the isolated sequences was well conserved in the young-adult group, with an average of 9.38 amino acids (aa) and range of 5 to 13 aa. There was no significant difference between age groups or polysaccharides in regard to CDR3 length. All VL sequences were analyzed, and the sequences of the CDRs of commonly occurring VL gene products obtained from both young and elderly volunteers are shown in Tables 3 and 4. Within 4 sequences several mutations occurred in the CDR1 with limited replacement mutations in CDR2. Most mutations occurred at the RGYW mutational hot spots. In addition, all demonstrated a threonine residue at position 95. Overall, the 4 sequences were 95 to 100% homologous to germ line with diversity at the V-J junction.
The 1 gene family, represented by L12 gene products, was used by a variety of donors but more commonly in young adults. Most L12 sequences were heavily mutated in all CDRs in response to both polysaccharides and showed 88 to 97% identity to germ line. The CDR3 regions demonstrated either deletion with N insertions or extensive replacement mutations and varied in length from 8 to 9 aa, with an average of 8.9 aa. The 3 gene family, consisting of L6 and A27 gene products, was either unmutated or demonstrated 1 or 2 aa replacements in the CDR1/2 region. Diversity, generated by N and P insertions in the V-J junction, yielded CDR3s of 8 to 11 aa and resulted in unique sequences in response to PPS4 and PPS14 with no overlap between individuals.
A variety of lambda family genes were also used in response to both PPS4 and PPS14. Representative sequences of the most common 1 gene product, 1e, are shown in Tables 3 and 4 and were 95 to 100% homologous to germ line. In response to PPS4, 44% of sequences were mutated in the CDR1 and -2, but less than 10% were mutated in response to PPS14. Unique to young adults and present in the response to both polysaccharides was the CDR3 motif QSYDSSLSG. Diversity between polysaccharides was maintained by N and P insertions at the V-J junction. The elderly demonstrated similar unmutated sequences with highly hydrophobic N insertions at the V-J junction. In contrast to the 1e sequences, the 3 3h products were highly mutated in all CDRs with 88 to 95% identity to germ line sequence. A number of the 3h gene products were present in response to both polysaccharides and recognized in subjects from all age groups. However, the CDR3 region demonstrated polysaccharide-related differences in the number of hydrophobic amino acid substitutions. This gene was more commonly isolated from the young than the elderly in response to PPS14, but not in response to PPS4.
Overall, despite the use of similar VL sequences between individuals and polysaccharides, the VH gene analysis, performed on identical aliquots of cDNA, demonstrated that the VH gene usage was unique between polysaccharides and between subjects (Tables 1 and 2).
J-chain usage. Due to restricted usage, percent J chain was calculated in relation to utilization of V and V (Table 5). In the V response to PPS4 both the young and elderly utilized the same J chains (J1, J2, and J3), with predominant use of J1 (85% and 66%, respectively). However, there was a significant reduction in J1 usage with age and a consequent shift towards usage of J2 and J3. In the V response to PPS4 both young and elderly used predominantly J2 genes; however, the elderly demonstrated a significant shift from J2 to J3b usage. Utilization of J1 was not affected by age.
The response to PPS14 consisted of the same J genes (J1, J2, and J3); however, unlike the anti-PPS4 response there was no significant difference in J usage between age groups (Table 5). In contrast, within the lambda response there were significant differences. In the young J2 continued to predominate whereas in the elderly J3b was dominant. There was consequently a significant increase in J3b expression and a decrease in J2 expression with age. Similar to the anti-PPS4 response, there was no alteration in utilization of J1. In contrast to the findings in unselected B cells described by Farner et al. (3), we did not detect the presence of J7. Similarly, in response to PPS antigens, Zhou et al. (28) and Lucas et al. (12) did not report the presence of J7.
Mutational frequencies in light chain response to PPS4 and PPS14. All light chain sequences were pooled within their respective gene family and age group and analyzed for mutational frequencies in the CDR and framework regions (FR). Mutational frequencies were compared between age groups and in response to PPS4 and PPS14 (data not shown). Overall, the mutational frequencies of all light chain gene families were similar between age groups and in response to both PPS4 and PPS14. Within these gene families, mutations occurred more frequently in the CDRs than in the FR in all age groups and in response to both polysaccharides. As expected many mutations in both young and elderly adults were found in mutation hotspots (RGYW and WRCY) as shown in Tables 3 and 4.
Comparison of young adults to the elderly demonstrated no significant difference in replacement mutations and R/S mutation ratio in the FR or CDR in response to either polysaccharide, although there was a trend toward higher R/S ratio in the elderly (PPS4: young, 1.1; elderly, 2.2; P = 0.21; PPS14: young, 1.0; elderly, 2.9; P = 0.09).
VL gene usage and functional antibody avidity. PPS-specific antibody concentrations, and avidity were measured in all pre- and postimmunization sera (8). We compared levels of VL gene family usage in individuals demonstrating high (>0.5 M concentration NaSCN resulting in a 50% reduction of enzyme-linked immunosorbent assay [ELISA] optical density [OD])- and low (<0.3 M concentration NaSCN resulting in a 50% reduction of ELISA OD)-avidity serum antibody responses. There was no clear correlation between antibody avidity and particular VL gene usage, as most predominant VL gene families were associated with low, medium, and high antibody avidity (Table 2 and 3). The lack of association between VL gene usage and avidity is likely related to the complexity of the anti-PPS antibody repertoire.
DISCUSSION
We hypothesized that the observed loss of antibody avidity in the elderly is associated with age-related changes in V region repertoire and/or somatic mutation. Our previous analysis of the VH region repertoire in response to both PPS4 and PPS14 demonstrated significant changes in VH locus usage in the elderly compared to young adults (7). In addition, a significant loss of oligoclonality was noted in elderly volunteers. The present study was specifically designed to address potential changes in light chain gene usage and mutational frequency as a function of age. VL libraries were generated using peripheral blood lymphocytes from vaccinated individuals enriched for PPS4- and PPS14-specific B cells with the use of PPS-coated paramagnetic beads. The specificity of the selection system was subsequently verified by testing culture supernatants for PPS specificity following in vitro expansion of selected B cells.
Overall, the VL gene usage in response to both PPS4 and PPS14 demonstrated the use of a wide variety of VL gene families within the study population, in contrast to the highly restricted VL repertoire in response to PPS23F and Haemophilus influenzae group b capsular polysaccharide (HibPS) (20, 27). The repertoire, however, was dominated by six gene families and restricted within individuals, as previously described for the VL repertoire in the response to other pneumococcal polysaccharides (12, 28). Several individuals utilized the same gene family and locus in response to both PPS4 and PPS14 although paired with different VH gene segments, illustrating the promiscuity of these VL gene families. Despite the use of similar VL gene families, significant differences in VL gene family usage were noted between age groups. The elderly demonstrated a predominant use of V4 in response to both PPS4 and PPS14, while young adults predominantly expressed 3 and/or 1 gene products.
Previous sequence analysis of VL region gene utilization in response to PPS4 and PPS14 is limited to nine sequences. Lucas et al. used both a combinatorial approach (12) and sequence analysis of purified PPS14-specific antibodies (9) to determine variable light chain gene usage in the response to PPS14. A total of five PPS14-specific VLs were isolated and utilized the 3 (n = 3) and 1 (n = 2) gene families. These results concur with our findings, as the 3 and 1 gene families together represented 34.1% (elderly) and 36.2% (young) of the VL sequences in our study. In addition, Baxendale et al. (1) generated human PPS-specific heterohybridomas with specificity for PPS4 from three volunteers. Three of the four heterohybridomas expressed the 2 VL gene family, while the other expressed 3 A27. The 2 gene family, although present in both age groups, represented <10% of the total sequences obtained in response to PPS4. In contrast, the 3 gene products were commonly represented in our study, ranging from 13.5% (young) to 21.1% (elderly). The differences between this study and our results may be based on a number of profound differences in study design including vaccine preparation used (conjugate versus PPS), timing of the isolation of B cells (7 days versus 42 days), study population (high versus all responders), and methodology (heterohybridomas versus PPS-specific B-cell isolation).
To our knowledge, research regarding the circulating VL repertoire in aging humans is limited to analysis of the V4 gene family (24). Troutaud et al. (24) analyzed V4-J rearrangements and V4 mutational frequencies in the peripheral blood mononuclear cells isolated from young and elderly (mean age = 83 years) individuals. These authors demonstrated significantly lower levels of somatic mutation, specifically replacement mutations, with aging. Furthermore, comparison of CDR3 regions of V4 revealed significant changes in light chain junctional diversity that correlated with age. In our study, we found no significant age-related difference in either silent or replacement mutations in either V4, predominantly expressed by the elderly, or other VL gene families. However, we did observe significant age-related differences in JL usage in response to both PPS. Analysis of the V gene families demonstrated a significantly lower use of J1 in elderly compared to young adults, although this J gene was overrepresented in both age groups compared to unselected IgM+ B cells (4). Similarly, in the V gene family analysis, a significant reduction in J2 usage with concomitant increase in J3 was noted in the elderly compared to the young. Despite differences in JL gene usage, the overall lengths of the CDR3 regions were not significantly different between age groups. All VL gene families demonstrated nontemplated insertions at the V-J junction commonly generating a 9- to 11-aa CDR3, in line with the expected length of productive rearrangements (3, 4). These N or P insertions are commonly observed in the response to other polysaccharides (10, 11, 20, 27, 28). The V-J junctional deletions and insertions presumably occur at the time of V-J joining, prior to antigenic stimulation.
The L12 (V1) and 3 h (V3) gene products demonstrated extensive modifications in the CDRs. It has been suggested that prior stimulation of the immune response, through either colonization or exposure to cross-reactive antigens, is responsible for the high mutation frequency noted in anti-PPS antibodies (1, 16, 27, 28). These findings are consistent with the significant concentrations of PPS-specific antibodies found in preimmune sera obtained from many of our volunteers (8) and the highly mutated VH sequences isolated (7) from this cDNA population. It is not possible to discern preexisting versus vaccine-induced somatic mutations without in-depth analysis of the preimmune antibody repertoire. It should be noted, however, that most of the VL sequences belonging to the V3, V4, and V1 gene families closely resemble germ line sequences. Despite the notion that somatic mutations result in improved antibody avidity, this may not necessarily be the case, as illustrated by the canonical, minimally mutated VL A2 sequence isolated in the response to Hib (10, 21). It has been demonstrated that a single amino acid substitution in a canonical VL A2 sequence can have a significant impact on avidity and bactericidal activity (10). It has therefore been proposed that canonical antibodies, resembling germ line sequences, are optimally fit for binding polysaccharide antigens without further improvement by somatic mutation (5). This phenomenon may be responsible for the low mutational frequency observed in the V3, V4 and V1 gene families.
The work presented here represents the most comprehensive study of antigen-driven light chain gene usage in the elderly immune response. Moreover, the characterization of a large number of sequences specific for two PPS antigens, isolated from young and elderly, as well as the inclusion of high- and low-responding donors, makes this study unique. We have isolated anti-PPS antibody-producing cells at 6 weeks postimmunization. The isolated repertoire may therefore not reflect the PPS-specific repertoire expressed by antibody-producing B cells isolated at 7 days postimmunization. Potential differences in the antibody repertoire expressed by PPS-specific circulating B cells over time remain to be elucidated. Should there be significant repertoire differences in B cells circulating early versus late postvaccination, it will be important to define which B cells (early versus late) generate the majority of serum anti-PPS antibodies.
It should be recognized, however, that adults possess considerable PPS-specific antibody concentrations prior to vaccination (8), and thus this study as well as others (1, 12, 28) most likely represents analysis of a recall, not a de novo, response to pneumococcal polysaccharides. Several differences were noted in the aged immune response to PPS. There was a significant shift in VL gene usage, similar to the previously noted age associated shift in VH gene usage (7). Light chain gene usage was oligoclonal in many but not all donors, which is in contrast with the predominantly oligoclonal response noted previously in the heavy chain gene usage in both elderly and young donors. In addition, there was no detectable change in mutational frequency with aging for any of the light chain genes observed. Overall, VL sequences demonstrated low mutational frequencies that could explain the lack of detectable differences between the young and elderly. We are currently undertaking experiments to further define the precise role of gene usage in age-related changes of the anti-PPS response. Specifically, these studies are aimed at discerning the relationship between anti-PPS antibody structure and function and investigating the role of preexisting immunity in the anti-PPS vaccine response in adults.
ACKNOWLEDGMENTS
This work was supported by Public Service Grant AG15978 from the National Institute of Aging.
REFERENCES
1. 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.
2. Chang, Q., Z. Zhong, A. Lees, M. Pekna, and L. Pirofski. 2002. Structure-function relationships for human antibodies to pneumococcal capsular polysaccharide from transgenic mice with human immunoglobulin loci. Infect. Immun. 70:4977-4986.
3. Farner, N. L., T. Dorner, and P. E. Lipsky. 1999. Molecular mechanisms and selection influence the generation of the human VJ repertoire. J. Immunol. 162:2137-2145.
4. Foster, S. J., H. P. Brezinschek, R. I. Brezinschek, and P. E. Lipsky. 1997. Molecular mechanisms and selective influences that shape the kappa gene repertoire of IgM+ B cells. J. Clin. Investig. 99:1614-1627.
5. Hougs, L., T. Barington, H. O. Madsen, L. P. Ryder, and A. Svejgaard. 1993. Rapid analysis of rearranged kappa light chain genes of circulating polysaccharide-specific B lymphocytes by means of immunomagnetic beads and the polymerase chain reaction. Exp. Clin. Immunogenet. 10:141-151.
6. Jackson, L. A., K. M. Neuzil, O. Yu, P. Benson, W. E. Barlow, A. L. Adams, C. A. Hanson, L. D. Mahoney, D. K. Shay, and W. W. Thompson. 2003. Effectiveness of pneumococcal polysaccharide vaccine in older adults. N. Engl. J. Med. 348:1747-1755.
7. Kolibab, K., S. L. Smithson, B. Rabquer, S. Khuder, and M. A. J. Westerink. 2005. Immune response to pneumococcal polysaccharides 4 and 14 in elderly and young adults: analysis of the variable heavy chain repertoire. Infect. Immun. 73:7465-7476.
8. Kolibab, K., S. L. Smithson, A. K. Shriner, S. Khuder, S. Romero-Steiner, G. M. Carlone, and M. A. Westerink. 2005. Immune response to pneumococcal polysaccharides 4 and 14 in elderly and young adults: I. Antibody concentrations, avidity and functional activity. Immun. Ageing 2:10.
9. Lucas, A. H., D. M. Granoff, R. E. Mandrell, C. C. Connolly, A. S. Shan, and D. C. Powers. 1997. Oligoclonality of serum immunoglobulin G antibody responses to Streptococcus pneumoniae capsular polysaccharide serotypes 6B, 14, and 23F. Infect. Immun. 65:5103-5109.
10. Lucas, A. H., G. R. McLean, D. C. Reason, A. P. O'Connor, M. C. Felton, and K. D. Moulton. 2003. Molecular ontogeny of the human antibody repertoire to the Haemophilus influenzae type B polysaccharide: expression of canonical variable regions and their variants in vaccinated infants. Clin. Immunol. 108:119-127.
11. Lucas, A. H., K. D. Moulton, and D. C. Reason. 1998. Role of kappa II-A2 light chain CDR-3 junctional residues in human antibody binding to the Haemophilus influenzae type b polysaccharide. J. Immunol. 161:3776-3780.
12. Lucas, A. H., K. D. Moulton, V. R. Tang, and D. C. Reason. 2001. Combinatorial library cloning of human antibodies to Streptococcus pneumoniae capsular polysaccharides: variable region primary structures and evidence for somatic mutation of Fab fragments specific for capsular serotypes 6B, 14, and 23F. Infect. Immun. 69:853-864.
13. Musher, D. M. 1992. Infections caused by Streptococcus pneumoniae: clinical spectrum, pathogenesis, immunity, and treatment. Clin. Infect. Dis. 14:801-807.
14. Nicoletti, C., C. Borghesi-Nicoletti, X. Yang, D. H. Schulze, and J. Cerny. 1991. Repertoire diversity of antibody response to bacterial antigens in aged mice. II. Phosphorylcholine-antibody in young and aged mice differ in both VH/VL gene repertoire and in specificity. J. Immunol. 147:2750-2755.
15. Nicoletti, C., X. Yang, and J. Cerny. 1993. Repertoire diversity of antibody response to bacterial antigens in aged mice. III. Phosphorylcholine antibody from young and aged mice differ in structure and protective activity against infection with Streptococcus pneumoniae. J. Immunol. 150:543-549.
16. Reason, D. C., T. C. Wagner, and A. H. Lucas. 1997. Human Fab fragments specific for the Haemophilus influenzae b polysaccharide isolated from a bacteriophage combinatorial library use variable region gene combinations and express an idiotype that mirrors in vivo expression. Infect. Immun. 65:261-266.
17. Reason, D. C., and J. Zhou. 2004. Correlation of antigenic epitope and antibody gene usage in the human immune response to Streptococcus pneumoniae type 23F capsular polysaccharide. Clin. Immunol. 111:132-136.
18. Romero-Steiner, S., D. M. Musher, M. S. Cetron, L. B. Pais, J. E. Groover, A. E. Fiore, B. D. Plikaytis, and G. M. Carlone. 1999. Reduction in functional antibody activity against Streptococcus pneumoniae in vaccinated elderly individuals highly correlates with decreased IgG antibody avidity. Clin. Infect. Dis. 29:281-288.
19. Russell, N. D., J. R. Corvalan, M. L. Gallo, C. G. Davis, and L. Pirofski. 2000. Production of protective human antipneumococcal antibodies by transgenic mice with human immunoglobulin loci. Infect. Immun. 68:1820-1826.
20. Scott, M., and M. H. Nahm. 1992. Characterization of the human IgG antibody VL repertoire to Haemophilus influenzae type b polysaccharide. J. Infect. Dis. 165(Suppl. 1):S53-S56.
21. Scott, M. G., D. L. Crimmins, D. W. McCourt, I. Zocher, R. Thiebe, H. G. Zachau, and M. H. Nahm. 1989. Clonal characterization of the human IgG antibody repertoire to Haemophilus influenzae type b polysaccharide. III. A single VKII gene and one of several JK genes are joined by an invariant arginine to form the most common L chain V region. J. Immunol. 143:4110-4116.
22. Shaw, D. R., P. Kirkham, H. W. Schroeder, Jr., P. Roben, and G. J. Silverman. 1995. Structure-function studies of human monoclonal antibodies to pneumococcus type 3 polysaccharide. Ann. N. Y. Acad. Sci. 764:370-373.
23. Simberkoff, M. S., A. P. Cross, M. Al-Ibrahim, A. L. Baltch, P. J. Geiseler, J. Nadler, A. S. Richmond, R. P. Smith, G. Schiffman, D. S. Shepard, et al. 1986. Efficacy of pneumococcal vaccine in high-risk patients. Results of a Veterans Administration cooperative study. N. Engl. J. Med. 315:1318-1327.
24. Troutaud, D., M. Drouet, C. Decourt, C. Le Morvan, and M. Cogne. 1999. Age-related alterations of somatic hypermutation and CDR3 lengths in human V4-expressing B lymphocytes. Immunology 97:197-203.
25. Welschof, M., P. Terness, F. Kolbinger, M. Zewe, S. Dubel, H. Dorsam, C. Hain, M. Finger, M. Jung, G. Moldenhauer, N. Hayashi, M. Little, and G. Opelz. 1995. Amino acid sequence based PCR primers for amplification of rearranged human heavy and light chain immunoglobulin variable region genes. J. Immunol. Methods 179:203-214.
26. Zhong, Z., T. Burns, Q. Chang, M. Carroll, and L. Pirofski. 1999. Molecular and functional characteristics of a protective human monoclonal antibody to serotype 8 Streptococcus pneumoniae capsular polysaccharide. Infect. Immun. 67:4119-4127.
27. Zhou, J., K. R. Lottenbach, S. J. Barenkamp, A. H. Lucas, and D. C. Reason. 2002. Recurrent variable region gene usage and somatic mutation in the human antibody response to the capsular polysaccharide of Streptococcus pneumoniae type 23F. Infect. Immun. 70:4083-4091.
28. Zhou, J., K. R. Lottenbach, S. J. Barenkamp, and D. C. Reason. 2004. Somatic hypermutation and diverse immunoglobulin gene usage in the human antibody response to the capsular polysaccharide of Streptococcus pneumoniae type 6B. Infect. Immun. 72:3505-3514.(S. Louise Smithson, Kris )
ABSTRACT
Streptococcus pneumoniae is a human bacterial pathogen responsible for serious infections including pneumonia. The currently licensed polysaccharide vaccine provides 60 to 80% protection in young adults, but in the elderly the vaccine efficacy is drastically reduced despite normal antibody levels. We hypothesized that the reduced vaccine efficacy in the elderly results from altered variable gene family usage. We have analyzed the light chain gene usage in 20 young (20 to 30 years of age) and 20 elderly (65 to 86 years of age) adults in response to pneumococcal polysaccharide 4 (PPS4) and PPS14. We generated a variable light chain library using B cells specific for PPS4 and PPS14 from each vaccinated individual. We determined complete sequences and somatic mutation frequencies in all isolated variable light chain fragments. Six gene families, 1, 2, 3, 4, 1, and 3, were identified in response to PPS4 and PPS14 in both age groups. Comparison of young and elderly adults demonstrated significant differences in 4, 1, and 3 gene usage in response to PPS4 and PPS14. With aging, there was a significant increase in 4 gene usage and a significant decrease in 1 and 3 gene usage in response to both PPS4 and PPS14. Although both V1 and V3 gene products demonstrated extensive mutations, there was no age-related difference in mutational frequency per gene family. These findings suggest an age-related change in light chain gene usage in response to PPS4 and PPS14.
INTRODUCTION
The incidence of pneumococcal pneumonia is significantly increased in individuals >65 years of age (13). Several studies (8, 18) have demonstrated a significant impairment in the immune response to pneumococcal polysaccharides in the elderly, particularly those >77 years of age, associated with a markedly decreased efficacy of the pneumococcal vaccine (6, 23).
The molecular mechanisms responsible for the decreased immune response in the elderly remain poorly understood. Nicoletti et al. (15) studied the immune response to phosphorylcholine (PC) in aged mice vaccinated with Streptococcus pneumoniae. These studies demonstrated a significant functional impairment in the immune response to PC despite normal antibody concentrations. Molecular analysis of the anti-PC-specific antibody repertoire in aged mice revealed a significantly altered V gene usage compared to young immunized mice (14).
To date, studies examining the structure-function relationship of antipneumococcal antibodies have been performed using human monoclonal antibodies (1, 22, 26), combinatorial libraries (12, 17, 27, 28), and monoclonal antibodies derived from a transgenic mouse strain with human immunoglobulin loci (2, 19). All studies, with the exception of those using the transgenic mouse, have analyzed the V repertoire using peripheral blood mononuclear cells obtained from young high responders. Although these studies have enriched our knowledge concerning the human immune response to pneumococcal polysaccharide (PPS), they were not designed to represent the in vivo antibody repertoire response to PPS in the elderly or the adult population at large, by excluding intermediate and low responders.
We have studied the immune response to PPS4 and PPS14 in 20 young (20 to 30 years of age) and 20 elderly (>65 years of age) volunteers. In this study, we have compared the antibody concentrations and opsonophagocytic antibody responses as well as levels of gene family usage in an effort to increase our understanding of the aging immune response to pneumococcal polysaccharides. We have previously reported a significant decrease in opsonophagocytic activity in elderly persons >77 years of age, as reported by others (8, 18). Comparative analysis of the variable heavy chain (VH) gene usage in the young and elderly demonstrated significant age-related differences in VH gene usage in response to both PPS4 and PPS14. Furthermore, a significant loss of oligoclonality, a characteristic of the immune response to polysaccharide antigens, was noted in the elderly (7). The results of the present study demonstrates a wide variety of variable light chain (VL) gene usage in response to both polysaccharides, although the response was dominated by six light chain gene families, 1, 3, 1, 2, 3, and 4. Despite common VL gene family usage, significant differences between young and elderly volunteers were noted in the distribution of these VL gene families.
MATERIALS AND METHODS
Human volunteers and vaccination. Only healthy young adults (20 to 30 years old) and healthy elderly adults (>65 years old) were asked to participate in this study. The age of the elderly ranged from 65 to 86 years, with 14 volunteers in the 65- to 77-year age group and 6 volunteers in the 78- to 86-year age group. Elderly volunteers were recruited from the community as well as from the general internal medicine clinic at the Medical University of Ohio (MUOT). Young volunteers were recruited from the student population at MUOT. Each individual was questioned concerning prior pneumococcal vaccination, medications, previous illness, and present health. In addition, we obtained complete blood count (CBC); a comprehensive chemistry profile including renal and liver functions and serum albumin; total B cells; T-cell subsets; and total immunoglobulin G (IgG), IgM, and IgA levels in all study candidates. Individuals previously immunized with the pneumococcal vaccine and any individual considered to be immunocompromised on the basis of medication (chemotherapy, steroid preparations, immunosuppressive agents including anti-tumor necrosis factor alpha agents), previous or present illness (previous pneumococcal disease, splenectomy, autoimmune disease, end stage renal or liver disease, human immunodeficiency virus positivity, organ transplant, or cancer) and those with abnormal CBC, chemistries, B or T cells, or immunoglobulin levels did not qualify. Informed consent was obtained from all participants using protocols reviewed and approved by the Institutional Review Board at MUOT.
Purification of PPS-specific B cells. Biotinylated polysaccharides were used to select PPS4- and PPS14-specific B cells as described previously (7). Briefly PPS4 and PPS14 were biotinylated by the method described by Lucas et al. (12). These polysaccharides were then reacted with streptavidin-coated immunomagnetic beads (Dynal, Lake Success, NY) and washed. Peripheral blood lymphocytes collected 6 weeks postvaccination were obtained by Ficoll Histopaque gradient (Sigma, St. Louis, MO). Cross-reactive B cells were depleted from the population by incubation with beads coated with PPS23F and cell wall polysaccharide (CWPS) and discarded. Further depletion of cross-reactive B cells was achieved by cross incubating the B cells; cells for PPS4-specific isolation were first depleted with PPS14-coated beads, and likewise cells for PPS14 isolation were depleted with PPS4-coated beads. Isolation of PPS-specific B cells was achieved by incubation with 25 μg of PPS4-coated immunomagnetic beads or PPS14-coated beads with 10 μg/ml CWPS added to the buffer to block nonspecific binding. These steps were taken to eliminate binding to cross-reactive antigens between PPS. Polysaccharide-specific B cells were isolated by magnetic separation and washed with RPMI 1640-5% newborn calf serum (NBCS). cDNA was prepared from the isolated B cells using a Dynabeads mRNA direct kit by following the manufacturer's instructions and stored in aliquots at –70°C for further use. Validation of the selection method for PPS-specific B cells was performed as previously described (7).
Production of light chain libraries. The cDNA obtained from PPS-selected B cells was used as a template in the PCR to generate light chain libraries. PCR controls consisted of cDNA obtained from unselected B cells. Primer sets described by Welschof et al. (25) and Taq polymerase (Fisher Scientific, Pittsburgh, PA) were used to generate light chain products. The PCR amplification conditions consisted of 32 cycles of 94°C for 45 s, 65°C for 30 s, and 72°C for 45 s. Amplification products were purified using the GeneClean gel extraction kit (Bio101, La Jolla, CA) and ligated into the TA cloning vector system (Invitrogen, San Diego, CA). Ligated plasmids were transformed into Top 10 Escherichia coli cells by chemical transformation. The light chain libraries were plated on Luria broth (LB)-kanamycin-X-Gal (5-bromo-4-chloro-3-indolyl--D-galactopyranoside) plates at low density and grown overnight at 37°C.
Selection and sequence analysis of positive PPS4 and PPS14 clones. Individual E. coli clones were selected and streaked onto a LB-kanamycin master plate and grown overnight at 37°C. These clones were lifted onto nylon filters and fixed by UV exposure for 5 min. The nylon filters were probed with a [-32P]ATP-labeled oligonucleotide with specificity for the and framework III region (25). Sequence analysis was performed on selected clones (MWG Biotech, High Point, NC) using primers complementary to the vector. The resultant sequences were compared to germ line sequences using VBASE DNAPLOT (http://vbase.mrc-cpe.cam.ac.uk).
Statistical analysis. Percentages of light chain gene usage against PPS4 and PPS14 were calculated for each group and volunteer. Fisher exact test and the Pearson chi square value were used to determine significance between gene usage and age groups. Student's t test was used to determine significance between mutational frequencies. A P value equal or less than 0.05 was considered to be significant. Statistical calculations were performed with the use of SPSS software 11.5.1.
Nucleotide sequence accession numbers. CDR sequences are available from GenBank under accession numbers AY928106 and AY928172.
RESULTS
Light chain analysis. The cDNA obtained from PPS-selected B cells was used to generate VL libraries. The cDNA from each individual's cDNA was amplified twice on separate occasions to control for PCR artifacts. Those samples that failed to yield consistent results were eliminated from this analysis. The accuracy of the B-cell selection method has been verified as described previously (7).
Successful sequence analysis was performed on a total of 175 (average of 10.9 VLs/volunteer) light chains with specificity for PPS4 and 202 (average of 12.6 VLs/volunteer) light chains with specificity for PPS14 derived from 17 immunized young volunteers. Sequence analysis was obtained for 222 (average of 13.9 VLs/volunteer) in response to PPS4 and 235 (average of 13.8 VLs/volunteer) with specificity for PPS14 from 16 and 17 elderly volunteers, respectively. All isolated sequences were productively rearranged. Variable light chain gene family complementarity-determining region 3 (CDR3) length and composition, J chain, percent identity to germ line sequence, and somatic mutation frequencies were determined. PCR amplification was performed on unselected B cells on a monthly basis throughout the study and compared to the expected VL gene usage in unselected B cells (3, 4) to ensure unbiased amplification of all gene families.
VL gene response to PPS4 and PPS14 in young adults. All immunized young adults utilized one or more light chains belonging to 6 of the 16 light chain gene families. These six light chain gene families (1, 3, 1, 2, 3, and 4) accounted for 99.9% of the repertoire used in response to PPS4. As shown in Fig. 1, the percent utilization of 1 (27.4%) was significantly greater (P = 0.001) than that reported (9.6%) in unselected IgM+ B cells (3). We utilized IgM+ B cells for comparison as this was the most comprehensive study of the light chain usage in the circulating-B-cell repertoire available. Members of the 1 gene family were also commonly used and represented 21.7% of the VL sequences isolated, but this was not significantly different from peripheral-B-cell expression. Together these two light chain gene families dominated the L chain gene expression in young adults at 49.1% while the 3 (16.6%), 2 (7.4%), 3 (13.7%), and 4 (13.1%) gene families comprised the remainder of the response.
Analysis of the light chain variable gene family response to PPS14 (Fig. 1) revealed that the young volunteers utilized the same six VL gene families specific for PPS4, namely, 1, 3, 1, 2, 3, and 4. The 3 gene family was dominant, expressed in 23.3% of isolated sequence, and significantly overexpressed (P < 0.001) in comparison to peripheral-IgM+-B-cell expression of 5.2% (3). The gene families 1 and 1 were also highly expressed at 22.8% of total sequences obtained from young adult B cells specific for PPS14. The remaining light chains, 2, 3, and 4, accounted for 31.2% of the L chain sequences obtained from young adults.
There was wide variability in the specific VL locus utilized in response to both PPS4 and PPS14 within and between donors (Tables 1 and 2). Overall, 10 loci were identified as being the most commonly expressed VL loci in the 17 donors in response to PPS4 and PPS14, although loci were not identical between the two groups. However, in response to both PPS4 and PPS14, 9 of 17 young donors were found to express a predominant (>50% utilization) gene locus, with the remainder of the donors showing increased variability.
VL gene response to PPS4 and PPS14 in elderly adults. Analysis of L chain gene family usage in immunized elderly adults in response to PPS4 and PPS14 demonstrated similar levels of usage of the 1, 3, 1, 2, 3, and 4 light chain gene families. As shown in Fig. 1, the predominant light chain gene family used in response to both PPS4 and PPS14 was the 4 gene family, 27.9% and 27.2%, respectively. This is in sharp contrast (P < 0.001) to the 3.3% 4 representation found in unselected IgM+ B cells (4). In response to PPS4, the 1 gene family and the 3 gene family were commonly expressed at 20.3% and 21.1%, respectively, with 1, 2, and 3 representing a further 30.6% of L chain gene usage. In response to PPS14, as shown in Fig. 1, in addition to the 4 gene product, the 1 (18.7%) and 3 (19.6%) gene products were commonly expressed L chains by the elderly.
As demonstrated in Tables 1 and 2, specific VL locus usage varied in the elderly. The single most common locus, B3 (4) was represented as the most common locus in 7 of 17 elderly donors against PPS4 and 6 of 17 elderly donors against PPS14. In the elderly, 8 of 17 donors demonstrated an oligoclonal response with expression of a predominant (>50% utilization) locus against PPS4 and 7 of 17 against PPS14. Some volunteers utilized the same gene locus, but not identical sequences, in response to both PPS4 and PPS14. However, in most volunteers, the VL gene families used in response to PPS4 were different than those used in response to PPS14 (Tables 1 and 2).
VL gene response to PPS4: comparison of young versus elderly adults. In response to PPS4, comparison of the young and elderly generally revealed similar responses in gene family expression with the exception of 3 and 4 gene usage. As shown in Fig. 1, there was a significant difference (P = 0.01) between the elderly (8.1%) and young (16.6%) in the expression of the 3 gene family. In contrast, 4 gene usage was significantly higher in the elderly versus the young (P < 0.01). The elderly group displayed predominant expression of 4, 27.9%, in contrast to 13.1% of the VL repertoire isolated from young adults. Although the group size of the oldest population, those >77 years of age (donors 34 to 40), was inadequate to perform statistical analysis, there was a definite trend towards increased use of the 4 gene family and a decreased use of 3 and 1 gene family with increasing age.
VL gene response to PPS14: comparison of young versus elderly adults. As shown in Fig. 1, the immune response to PPS14 in the young consisted predominantly of 1, 3, and 1 gene families. However, with aging a significant shift in the use of VL gene families occurred, with an increase in 4 gene usage (P < 0.001) and a decrease in both 1 (P = 0.025) and 3 (P = 0.001) usage. As shown for PPS4, this was an age-dependent phenomenon, with a trend towards increasing differences with increasing age. Nine percent of young volunteers and 27.2% of elderly volunteers expressed 4 as part of their PPS14-specific antibody repertoire. There was no statistical difference in 1, 2, or 3 gene usage between the young and the elderly in response to PPS14.
CDR composition and germ line identity. Although a wide variety of VL gene products were used, the CDR3 length of the isolated sequences was well conserved in the young-adult group, with an average of 9.38 amino acids (aa) and range of 5 to 13 aa. There was no significant difference between age groups or polysaccharides in regard to CDR3 length. All VL sequences were analyzed, and the sequences of the CDRs of commonly occurring VL gene products obtained from both young and elderly volunteers are shown in Tables 3 and 4. Within 4 sequences several mutations occurred in the CDR1 with limited replacement mutations in CDR2. Most mutations occurred at the RGYW mutational hot spots. In addition, all demonstrated a threonine residue at position 95. Overall, the 4 sequences were 95 to 100% homologous to germ line with diversity at the V-J junction.
The 1 gene family, represented by L12 gene products, was used by a variety of donors but more commonly in young adults. Most L12 sequences were heavily mutated in all CDRs in response to both polysaccharides and showed 88 to 97% identity to germ line. The CDR3 regions demonstrated either deletion with N insertions or extensive replacement mutations and varied in length from 8 to 9 aa, with an average of 8.9 aa. The 3 gene family, consisting of L6 and A27 gene products, was either unmutated or demonstrated 1 or 2 aa replacements in the CDR1/2 region. Diversity, generated by N and P insertions in the V-J junction, yielded CDR3s of 8 to 11 aa and resulted in unique sequences in response to PPS4 and PPS14 with no overlap between individuals.
A variety of lambda family genes were also used in response to both PPS4 and PPS14. Representative sequences of the most common 1 gene product, 1e, are shown in Tables 3 and 4 and were 95 to 100% homologous to germ line. In response to PPS4, 44% of sequences were mutated in the CDR1 and -2, but less than 10% were mutated in response to PPS14. Unique to young adults and present in the response to both polysaccharides was the CDR3 motif QSYDSSLSG. Diversity between polysaccharides was maintained by N and P insertions at the V-J junction. The elderly demonstrated similar unmutated sequences with highly hydrophobic N insertions at the V-J junction. In contrast to the 1e sequences, the 3 3h products were highly mutated in all CDRs with 88 to 95% identity to germ line sequence. A number of the 3h gene products were present in response to both polysaccharides and recognized in subjects from all age groups. However, the CDR3 region demonstrated polysaccharide-related differences in the number of hydrophobic amino acid substitutions. This gene was more commonly isolated from the young than the elderly in response to PPS14, but not in response to PPS4.
Overall, despite the use of similar VL sequences between individuals and polysaccharides, the VH gene analysis, performed on identical aliquots of cDNA, demonstrated that the VH gene usage was unique between polysaccharides and between subjects (Tables 1 and 2).
J-chain usage. Due to restricted usage, percent J chain was calculated in relation to utilization of V and V (Table 5). In the V response to PPS4 both the young and elderly utilized the same J chains (J1, J2, and J3), with predominant use of J1 (85% and 66%, respectively). However, there was a significant reduction in J1 usage with age and a consequent shift towards usage of J2 and J3. In the V response to PPS4 both young and elderly used predominantly J2 genes; however, the elderly demonstrated a significant shift from J2 to J3b usage. Utilization of J1 was not affected by age.
The response to PPS14 consisted of the same J genes (J1, J2, and J3); however, unlike the anti-PPS4 response there was no significant difference in J usage between age groups (Table 5). In contrast, within the lambda response there were significant differences. In the young J2 continued to predominate whereas in the elderly J3b was dominant. There was consequently a significant increase in J3b expression and a decrease in J2 expression with age. Similar to the anti-PPS4 response, there was no alteration in utilization of J1. In contrast to the findings in unselected B cells described by Farner et al. (3), we did not detect the presence of J7. Similarly, in response to PPS antigens, Zhou et al. (28) and Lucas et al. (12) did not report the presence of J7.
Mutational frequencies in light chain response to PPS4 and PPS14. All light chain sequences were pooled within their respective gene family and age group and analyzed for mutational frequencies in the CDR and framework regions (FR). Mutational frequencies were compared between age groups and in response to PPS4 and PPS14 (data not shown). Overall, the mutational frequencies of all light chain gene families were similar between age groups and in response to both PPS4 and PPS14. Within these gene families, mutations occurred more frequently in the CDRs than in the FR in all age groups and in response to both polysaccharides. As expected many mutations in both young and elderly adults were found in mutation hotspots (RGYW and WRCY) as shown in Tables 3 and 4.
Comparison of young adults to the elderly demonstrated no significant difference in replacement mutations and R/S mutation ratio in the FR or CDR in response to either polysaccharide, although there was a trend toward higher R/S ratio in the elderly (PPS4: young, 1.1; elderly, 2.2; P = 0.21; PPS14: young, 1.0; elderly, 2.9; P = 0.09).
VL gene usage and functional antibody avidity. PPS-specific antibody concentrations, and avidity were measured in all pre- and postimmunization sera (8). We compared levels of VL gene family usage in individuals demonstrating high (>0.5 M concentration NaSCN resulting in a 50% reduction of enzyme-linked immunosorbent assay [ELISA] optical density [OD])- and low (<0.3 M concentration NaSCN resulting in a 50% reduction of ELISA OD)-avidity serum antibody responses. There was no clear correlation between antibody avidity and particular VL gene usage, as most predominant VL gene families were associated with low, medium, and high antibody avidity (Table 2 and 3). The lack of association between VL gene usage and avidity is likely related to the complexity of the anti-PPS antibody repertoire.
DISCUSSION
We hypothesized that the observed loss of antibody avidity in the elderly is associated with age-related changes in V region repertoire and/or somatic mutation. Our previous analysis of the VH region repertoire in response to both PPS4 and PPS14 demonstrated significant changes in VH locus usage in the elderly compared to young adults (7). In addition, a significant loss of oligoclonality was noted in elderly volunteers. The present study was specifically designed to address potential changes in light chain gene usage and mutational frequency as a function of age. VL libraries were generated using peripheral blood lymphocytes from vaccinated individuals enriched for PPS4- and PPS14-specific B cells with the use of PPS-coated paramagnetic beads. The specificity of the selection system was subsequently verified by testing culture supernatants for PPS specificity following in vitro expansion of selected B cells.
Overall, the VL gene usage in response to both PPS4 and PPS14 demonstrated the use of a wide variety of VL gene families within the study population, in contrast to the highly restricted VL repertoire in response to PPS23F and Haemophilus influenzae group b capsular polysaccharide (HibPS) (20, 27). The repertoire, however, was dominated by six gene families and restricted within individuals, as previously described for the VL repertoire in the response to other pneumococcal polysaccharides (12, 28). Several individuals utilized the same gene family and locus in response to both PPS4 and PPS14 although paired with different VH gene segments, illustrating the promiscuity of these VL gene families. Despite the use of similar VL gene families, significant differences in VL gene family usage were noted between age groups. The elderly demonstrated a predominant use of V4 in response to both PPS4 and PPS14, while young adults predominantly expressed 3 and/or 1 gene products.
Previous sequence analysis of VL region gene utilization in response to PPS4 and PPS14 is limited to nine sequences. Lucas et al. used both a combinatorial approach (12) and sequence analysis of purified PPS14-specific antibodies (9) to determine variable light chain gene usage in the response to PPS14. A total of five PPS14-specific VLs were isolated and utilized the 3 (n = 3) and 1 (n = 2) gene families. These results concur with our findings, as the 3 and 1 gene families together represented 34.1% (elderly) and 36.2% (young) of the VL sequences in our study. In addition, Baxendale et al. (1) generated human PPS-specific heterohybridomas with specificity for PPS4 from three volunteers. Three of the four heterohybridomas expressed the 2 VL gene family, while the other expressed 3 A27. The 2 gene family, although present in both age groups, represented <10% of the total sequences obtained in response to PPS4. In contrast, the 3 gene products were commonly represented in our study, ranging from 13.5% (young) to 21.1% (elderly). The differences between this study and our results may be based on a number of profound differences in study design including vaccine preparation used (conjugate versus PPS), timing of the isolation of B cells (7 days versus 42 days), study population (high versus all responders), and methodology (heterohybridomas versus PPS-specific B-cell isolation).
To our knowledge, research regarding the circulating VL repertoire in aging humans is limited to analysis of the V4 gene family (24). Troutaud et al. (24) analyzed V4-J rearrangements and V4 mutational frequencies in the peripheral blood mononuclear cells isolated from young and elderly (mean age = 83 years) individuals. These authors demonstrated significantly lower levels of somatic mutation, specifically replacement mutations, with aging. Furthermore, comparison of CDR3 regions of V4 revealed significant changes in light chain junctional diversity that correlated with age. In our study, we found no significant age-related difference in either silent or replacement mutations in either V4, predominantly expressed by the elderly, or other VL gene families. However, we did observe significant age-related differences in JL usage in response to both PPS. Analysis of the V gene families demonstrated a significantly lower use of J1 in elderly compared to young adults, although this J gene was overrepresented in both age groups compared to unselected IgM+ B cells (4). Similarly, in the V gene family analysis, a significant reduction in J2 usage with concomitant increase in J3 was noted in the elderly compared to the young. Despite differences in JL gene usage, the overall lengths of the CDR3 regions were not significantly different between age groups. All VL gene families demonstrated nontemplated insertions at the V-J junction commonly generating a 9- to 11-aa CDR3, in line with the expected length of productive rearrangements (3, 4). These N or P insertions are commonly observed in the response to other polysaccharides (10, 11, 20, 27, 28). The V-J junctional deletions and insertions presumably occur at the time of V-J joining, prior to antigenic stimulation.
The L12 (V1) and 3 h (V3) gene products demonstrated extensive modifications in the CDRs. It has been suggested that prior stimulation of the immune response, through either colonization or exposure to cross-reactive antigens, is responsible for the high mutation frequency noted in anti-PPS antibodies (1, 16, 27, 28). These findings are consistent with the significant concentrations of PPS-specific antibodies found in preimmune sera obtained from many of our volunteers (8) and the highly mutated VH sequences isolated (7) from this cDNA population. It is not possible to discern preexisting versus vaccine-induced somatic mutations without in-depth analysis of the preimmune antibody repertoire. It should be noted, however, that most of the VL sequences belonging to the V3, V4, and V1 gene families closely resemble germ line sequences. Despite the notion that somatic mutations result in improved antibody avidity, this may not necessarily be the case, as illustrated by the canonical, minimally mutated VL A2 sequence isolated in the response to Hib (10, 21). It has been demonstrated that a single amino acid substitution in a canonical VL A2 sequence can have a significant impact on avidity and bactericidal activity (10). It has therefore been proposed that canonical antibodies, resembling germ line sequences, are optimally fit for binding polysaccharide antigens without further improvement by somatic mutation (5). This phenomenon may be responsible for the low mutational frequency observed in the V3, V4 and V1 gene families.
The work presented here represents the most comprehensive study of antigen-driven light chain gene usage in the elderly immune response. Moreover, the characterization of a large number of sequences specific for two PPS antigens, isolated from young and elderly, as well as the inclusion of high- and low-responding donors, makes this study unique. We have isolated anti-PPS antibody-producing cells at 6 weeks postimmunization. The isolated repertoire may therefore not reflect the PPS-specific repertoire expressed by antibody-producing B cells isolated at 7 days postimmunization. Potential differences in the antibody repertoire expressed by PPS-specific circulating B cells over time remain to be elucidated. Should there be significant repertoire differences in B cells circulating early versus late postvaccination, it will be important to define which B cells (early versus late) generate the majority of serum anti-PPS antibodies.
It should be recognized, however, that adults possess considerable PPS-specific antibody concentrations prior to vaccination (8), and thus this study as well as others (1, 12, 28) most likely represents analysis of a recall, not a de novo, response to pneumococcal polysaccharides. Several differences were noted in the aged immune response to PPS. There was a significant shift in VL gene usage, similar to the previously noted age associated shift in VH gene usage (7). Light chain gene usage was oligoclonal in many but not all donors, which is in contrast with the predominantly oligoclonal response noted previously in the heavy chain gene usage in both elderly and young donors. In addition, there was no detectable change in mutational frequency with aging for any of the light chain genes observed. Overall, VL sequences demonstrated low mutational frequencies that could explain the lack of detectable differences between the young and elderly. We are currently undertaking experiments to further define the precise role of gene usage in age-related changes of the anti-PPS response. Specifically, these studies are aimed at discerning the relationship between anti-PPS antibody structure and function and investigating the role of preexisting immunity in the anti-PPS vaccine response in adults.
ACKNOWLEDGMENTS
This work was supported by Public Service Grant AG15978 from the National Institute of Aging.
REFERENCES
1. 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.
2. Chang, Q., Z. Zhong, A. Lees, M. Pekna, and L. Pirofski. 2002. Structure-function relationships for human antibodies to pneumococcal capsular polysaccharide from transgenic mice with human immunoglobulin loci. Infect. Immun. 70:4977-4986.
3. Farner, N. L., T. Dorner, and P. E. Lipsky. 1999. Molecular mechanisms and selection influence the generation of the human VJ repertoire. J. Immunol. 162:2137-2145.
4. Foster, S. J., H. P. Brezinschek, R. I. Brezinschek, and P. E. Lipsky. 1997. Molecular mechanisms and selective influences that shape the kappa gene repertoire of IgM+ B cells. J. Clin. Investig. 99:1614-1627.
5. Hougs, L., T. Barington, H. O. Madsen, L. P. Ryder, and A. Svejgaard. 1993. Rapid analysis of rearranged kappa light chain genes of circulating polysaccharide-specific B lymphocytes by means of immunomagnetic beads and the polymerase chain reaction. Exp. Clin. Immunogenet. 10:141-151.
6. Jackson, L. A., K. M. Neuzil, O. Yu, P. Benson, W. E. Barlow, A. L. Adams, C. A. Hanson, L. D. Mahoney, D. K. Shay, and W. W. Thompson. 2003. Effectiveness of pneumococcal polysaccharide vaccine in older adults. N. Engl. J. Med. 348:1747-1755.
7. Kolibab, K., S. L. Smithson, B. Rabquer, S. Khuder, and M. A. J. Westerink. 2005. Immune response to pneumococcal polysaccharides 4 and 14 in elderly and young adults: analysis of the variable heavy chain repertoire. Infect. Immun. 73:7465-7476.
8. Kolibab, K., S. L. Smithson, A. K. Shriner, S. Khuder, S. Romero-Steiner, G. M. Carlone, and M. A. Westerink. 2005. Immune response to pneumococcal polysaccharides 4 and 14 in elderly and young adults: I. Antibody concentrations, avidity and functional activity. Immun. Ageing 2:10.
9. Lucas, A. H., D. M. Granoff, R. E. Mandrell, C. C. Connolly, A. S. Shan, and D. C. Powers. 1997. Oligoclonality of serum immunoglobulin G antibody responses to Streptococcus pneumoniae capsular polysaccharide serotypes 6B, 14, and 23F. Infect. Immun. 65:5103-5109.
10. Lucas, A. H., G. R. McLean, D. C. Reason, A. P. O'Connor, M. C. Felton, and K. D. Moulton. 2003. Molecular ontogeny of the human antibody repertoire to the Haemophilus influenzae type B polysaccharide: expression of canonical variable regions and their variants in vaccinated infants. Clin. Immunol. 108:119-127.
11. Lucas, A. H., K. D. Moulton, and D. C. Reason. 1998. Role of kappa II-A2 light chain CDR-3 junctional residues in human antibody binding to the Haemophilus influenzae type b polysaccharide. J. Immunol. 161:3776-3780.
12. Lucas, A. H., K. D. Moulton, V. R. Tang, and D. C. Reason. 2001. Combinatorial library cloning of human antibodies to Streptococcus pneumoniae capsular polysaccharides: variable region primary structures and evidence for somatic mutation of Fab fragments specific for capsular serotypes 6B, 14, and 23F. Infect. Immun. 69:853-864.
13. Musher, D. M. 1992. Infections caused by Streptococcus pneumoniae: clinical spectrum, pathogenesis, immunity, and treatment. Clin. Infect. Dis. 14:801-807.
14. Nicoletti, C., C. Borghesi-Nicoletti, X. Yang, D. H. Schulze, and J. Cerny. 1991. Repertoire diversity of antibody response to bacterial antigens in aged mice. II. Phosphorylcholine-antibody in young and aged mice differ in both VH/VL gene repertoire and in specificity. J. Immunol. 147:2750-2755.
15. Nicoletti, C., X. Yang, and J. Cerny. 1993. Repertoire diversity of antibody response to bacterial antigens in aged mice. III. Phosphorylcholine antibody from young and aged mice differ in structure and protective activity against infection with Streptococcus pneumoniae. J. Immunol. 150:543-549.
16. Reason, D. C., T. C. Wagner, and A. H. Lucas. 1997. Human Fab fragments specific for the Haemophilus influenzae b polysaccharide isolated from a bacteriophage combinatorial library use variable region gene combinations and express an idiotype that mirrors in vivo expression. Infect. Immun. 65:261-266.
17. Reason, D. C., and J. Zhou. 2004. Correlation of antigenic epitope and antibody gene usage in the human immune response to Streptococcus pneumoniae type 23F capsular polysaccharide. Clin. Immunol. 111:132-136.
18. Romero-Steiner, S., D. M. Musher, M. S. Cetron, L. B. Pais, J. E. Groover, A. E. Fiore, B. D. Plikaytis, and G. M. Carlone. 1999. Reduction in functional antibody activity against Streptococcus pneumoniae in vaccinated elderly individuals highly correlates with decreased IgG antibody avidity. Clin. Infect. Dis. 29:281-288.
19. Russell, N. D., J. R. Corvalan, M. L. Gallo, C. G. Davis, and L. Pirofski. 2000. Production of protective human antipneumococcal antibodies by transgenic mice with human immunoglobulin loci. Infect. Immun. 68:1820-1826.
20. Scott, M., and M. H. Nahm. 1992. Characterization of the human IgG antibody VL repertoire to Haemophilus influenzae type b polysaccharide. J. Infect. Dis. 165(Suppl. 1):S53-S56.
21. Scott, M. G., D. L. Crimmins, D. W. McCourt, I. Zocher, R. Thiebe, H. G. Zachau, and M. H. Nahm. 1989. Clonal characterization of the human IgG antibody repertoire to Haemophilus influenzae type b polysaccharide. III. A single VKII gene and one of several JK genes are joined by an invariant arginine to form the most common L chain V region. J. Immunol. 143:4110-4116.
22. Shaw, D. R., P. Kirkham, H. W. Schroeder, Jr., P. Roben, and G. J. Silverman. 1995. Structure-function studies of human monoclonal antibodies to pneumococcus type 3 polysaccharide. Ann. N. Y. Acad. Sci. 764:370-373.
23. Simberkoff, M. S., A. P. Cross, M. Al-Ibrahim, A. L. Baltch, P. J. Geiseler, J. Nadler, A. S. Richmond, R. P. Smith, G. Schiffman, D. S. Shepard, et al. 1986. Efficacy of pneumococcal vaccine in high-risk patients. Results of a Veterans Administration cooperative study. N. Engl. J. Med. 315:1318-1327.
24. Troutaud, D., M. Drouet, C. Decourt, C. Le Morvan, and M. Cogne. 1999. Age-related alterations of somatic hypermutation and CDR3 lengths in human V4-expressing B lymphocytes. Immunology 97:197-203.
25. Welschof, M., P. Terness, F. Kolbinger, M. Zewe, S. Dubel, H. Dorsam, C. Hain, M. Finger, M. Jung, G. Moldenhauer, N. Hayashi, M. Little, and G. Opelz. 1995. Amino acid sequence based PCR primers for amplification of rearranged human heavy and light chain immunoglobulin variable region genes. J. Immunol. Methods 179:203-214.
26. Zhong, Z., T. Burns, Q. Chang, M. Carroll, and L. Pirofski. 1999. Molecular and functional characteristics of a protective human monoclonal antibody to serotype 8 Streptococcus pneumoniae capsular polysaccharide. Infect. Immun. 67:4119-4127.
27. Zhou, J., K. R. Lottenbach, S. J. Barenkamp, A. H. Lucas, and D. C. Reason. 2002. Recurrent variable region gene usage and somatic mutation in the human antibody response to the capsular polysaccharide of Streptococcus pneumoniae type 23F. Infect. Immun. 70:4083-4091.
28. Zhou, J., K. R. Lottenbach, S. J. Barenkamp, and D. C. Reason. 2004. Somatic hypermutation and diverse immunoglobulin gene usage in the human antibody response to the capsular polysaccharide of Streptococcus pneumoniae type 6B. Infect. Immun. 72:3505-3514.(S. Louise Smithson, Kris )