Skip Navigation


International Immunology Advance Access originally published online on March 28, 2006
International Immunology 2006 18(5):661-669; doi:10.1093/intimm/dxl003
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
18/5/661    most recent
dxl003v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (3)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Cassady-Cain, R. L
Right arrow Articles by Kaushik, A. K
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Cassady-Cain, R. L
Right arrow Articles by Kaushik, A. K
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?


© The Japanese Society for Immunology. 2006. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org

Increased negative selection impairs neonatal B cell repertoire but does not directly lead to generation of disease-associated IgM auto-antibodies

Robin L Cassady-Cain1 and Azad K Kaushik1,2,

1 Department of Pathobiology and
2 Department of Molecular and Cellular Biology, University of Guelph, Guelph, Ontario N1G 2W1, Canada

Correspondence to: A. K. Kaushik; E-mail: akaushik{at}uoguelph.ca


   Abstract
Top
 Abstract
Introduction
 Methods
 Results
 Discussion
 Supplementary data
 References
 
To determine if increased negative B cell selection, due to lowered signaling threshold of responsiveness to a ligand as a result of SHP-1 deficiency, during ontogeny leads to the origin of disease-associated IgM auto-antibodies (AAbs), 47 VHJ558+ VDJCµ rearrangements from SHP-1-deficient viable motheaten (mev/mev) and 24 J558+ VDJCµ rearrangements from normal mev/+ neonatal (<24 h post-birth) B cells were examined for their structural properties. None of the J558+ VDJCµ rearrangements from autoimmune-prone mev/mev had the characteristic CDR3H size restriction or arginine residues noted in disease-associated IgM AAbs. However, the MVAR2/10 genes are expressed at a high frequency in mev/mev (31.9%) as compared with mev/+ (16.7%), and pM11 gene expression is exclusively (14.9%) noted in mev/mev B cells. Clearly, there is a trend toward higher expression of pM11 genes (P-value ≤ 0.09) in autoimmune-prone mev/mev strain. The CDR2H region of J558+ VDJCµ recombinations from mev/mev has increased hotspot triplets predisposing to mutations as compared with mev/+ (P-value ≤ 0.01) mice. A higher DFL D-gene expression is noted in J558+ VDJCµ rearrangements from mev/mev (P-value ≤ 0.1) in contrast to mev/+. The sophisticated logistic regression and odds ratio analysis of V-, D- and J-gene expressions in neonatal B cells from mev/mev and mev/+ mice demonstrates differential composition of the germ line IgM repertoire as a result of SHP-1 deficiency. These observations suggest that increased negative B cell selection during ontogeny impairs the developing IgM antibody repertoire but does not directly lead to generation of disease-associated IgM AAbs.

Keywords: B lymphocyte, IgM auto-antibodies, negative B cell selection, SHP-1, VDJ rearrangement, VHJ558 gene


   Introduction
Top
 Abstract
 Introduction
Methods
 Results
 Discussion
 Supplementary data
 References
 
Several studies of disease-associated IgG auto-antibodies (AAbs) in systemic autoimmune disease, such as systemic lupus erythematosus (SLE), have shown that these display characteristics similar to antibodies generated in an exogenous antigen-driven immune response (1). Few studies have focused on pathogenic AAbs of the IgM isotype, which appear before IgG AAb, and are directly capable of initiating autoimmune pathogenesis. The IgMs are also suggested to be involved in the regulation of other antibody isotypes including IgG (2, 3) and are, thus, capable of indirectly influencing AAb-mediated pathogenesis in systemic autoimmune disease. An understanding of the origin of disease-associated IgM AAbs is, therefore, an important step toward identifying the initiating biomolecule aimed at ameliorating AAb-mediated tissue damage observed in systemic autoimmune disease.

Motheaten mice homozygous for the viable recessive mev mutation develop a fatal severe SLE-like systemic autoimmune disorder and immune dysfunction (4). The mev phenotype is linked to a loss of function (80–90%) mutation, involving thymine to adenine transversion, in the SHP-1 gene on chromosome six encoding the cytosolic tyrosine phosphatase (5–7). The SHP-1 deficiency results in severe dysregulation of hematopoietic cell development and function including increased myelopoesis. A subset of TdT+ bone marrow cells are markedly depleted and prothymocytes are developmentally arrested at the pre-TdT+ cell stage (6). The SHP-1 phosphatase regulates negative selection of B lymphocytes by lowering the threshold at which these cells respond to antigen (8–10), leading to extensive B cell deletion during ontogeny (11, 12). The severe autoimmune disease manifested by mev/mev mice includes an over-expansion of the B-1 lymphocytes expressing CD5 (13) that virtually populate the peripheral organs and are known for the production of natural AAbs, especially of IgM isotype (14, 15). Together with hypergammaglobulinaemia and T cell defects (16), these mice develop high titers of IgM and IgG AAbs to thymocyte (17) and double-strand DNA (dsDNA) (13), deposition of immune complexes (IgM and IgG) in the renal glomeruli (18) and die in approximately in 9 weeks from autoimmune pneumonitis. The viable motheaten mouse (mev/mev; 19), therefore, provides a useful model for studying the contribution of increased negative selection of IgM+ B cells during ontogeny as well as the origin of disease-associated IgM AAbs with known structural characteristics, for example, CDR size and amino acid composition (20).

Previously, we reported that expression of VH and V{kappa} gene families and their pairings are significantly skewed in the peripheral B lymphocytes of autoimmune mev/mev mice as compared with background C57/BL6 mice, most notably in VH J558 and Q52 gene families (21). These genetic observations suggested that selection mediated by auto-antigens is probably important in determining the repertoire of IgM AAbs in mev/mev mice. To directly investigate the potential relevance of BCR signal strength-based selection to generation of disease-associated IgM AAbs, we analyzed the structural properties of randomly non-antigen selected VHJ558+ IgM antibodies from autoimmune mev/mev and normal wild-type mice (20). In contrast to a variable CDR3H length (4–12 amino acids) in IgM antibodies from wild-type mice, all the mev/mev J558+ IgMs analyzed had a restricted CDR3H size of exactly 10 amino acids. Thus, disease-associated and natural IgM AAbs are structurally and biologically distinct and demonstrate BCR-based selection of the IgM AAbs involved in autoimmune pathogenesis of mev/mev mice at the terminal disease state. While these studies suggested ligand-mediated positive selection of B-1 cells expressing IgM AAbs in the autoimmune mev/mev mice (20), the possible contribution of negative selection (11, 12) of B cells could not be excluded. To address the question whether increased negative B cell selection early during ontogeny leads to the origin of disease-associated IgM AAbs with restricted CDR3H size, we analyzed VHJ558+ VDJCµ rearrangements from 1-day-old (<24 h post-birth) autoimmune-prone mev/mev and phenotypically normal mev/+ mice. These experiments provide evidence that increased negative selection, due to enhanced responsiveness of BCR for a ligand, during B cell ontogeny does not directly lead to the emergence of disease-associated IgM AAbs but impairs the developing primary antibody repertoire.


   Methods
Top
 Abstract
 Introduction
 Methods
Results
 Discussion
 Supplementary data
 References
 
Mice
Genetically defined homozygous (mev/mev) and heterozygous (mev/+) viable motheaten mice (C57BL/6J-Hcphme-v) were obtained by breeding heterozygous (mev/+) mice (Jackson Laboratories, ME, USA) in the OVC (Ontario Veterinary College) Isolation Unit at the University of Guelph, Ontario, Canada. The mev phenotype has been bred up to 33 generations on C57BL/6J background minimizing intra-strain mouse-to-mouse variation and, thus, permits the required comparative germ line repertoire analysis using a single mouse representative of the population.

PCR genotyping
Viable motheaten mice were genotyped using a PCR assay as described (22). Briefly, genomic tail DNA was prepared by digestion with proteinase K (GIBCO-BRL) in a PCR-compatible buffer (50 mM potassium chloride, 10 mM Tris pH 8.3, 2.5 mM magnesium chloride, 0.1 mg ml–1 gelatin, 0.4% v/v Nonidet P40, 0.45% v/v Tween 20, 1.5 µg ml–1 proteinase K) at 55°C. The genomic DNA was PCR amplified using a set of primers specific for the SHP-1 gene in the vicinity of the T -> A transversion (mevL: 5'-CGTGTCATCGTGATGACT-3'; mevR: 5'-AGGAAGTTGGGGCTTTGCCGT-3') in a 50-µl reaction using Taq polymerase (GIBCO-BRL). The PCR consisted of a hot start at 95°C for 1.5 min followed by 35 cycles of 94°C (denaturation) for 1.5 min, 58°C (annealing) and 72°C (extension) for 1 min each, with a final elongation step of 72°C for 7 min. The RsaI (New England Biolabs)-digested PCR product was analyzed on 6% low melting point agarose. The wild-type allele yields two products of 48 and 21 bp, while the mev allele remains undigested as indicated by a 69-bp product (22). In heterozygous mice, three products of 69, 48 and 21 bp sizes are noted. DNA from two hybridomas of known mev/mev genotype was included as control in each PCR assay.

cDNA Library
RNA prepared from the spleen and liver (source of lymphopoesis in neonate) within 24 h of birth using Trizol (GIBCO-BRL) was used to synthesize cDNA using cDNA First Strand Synthesis kit (Pharmacia-Biotech). The VDJCµ rearrangements were PCR amplified using the high fidelity Expand system (Boehringer-Mannheim) from cDNA using a 5' degenerate primer designed to amplify all mice VH gene families (5'-AGGTCCA(A/G)CT(G/T)CTCGAGTC(A/T)GG-3') and a 3' primer located in the CH2 region of the mouse Cµ gene (5'-ATTGGGACTAGTTTCTGCGACAGCTGGAAT-3'; 17). The PCR included a hot start at 94°C and 38 cycles of denaturation at 94°C for 1.5 min, annealing at 58°C for 1.5 min and extension at 72°C for l min with a final elongation step of 72°C for 7 min. The clean PCR product (Qiaquick kit, Qiagen) was blunt ended with T4 DNA polymerase (Pharmacia-Biotech), gel purified (Qiaex II, Qiagen) and cloned into the pCR-Blunt cloning vector (Invitrogen). The ligate was used to transform TOP10 competent Escherichia coli.

Colony screening and DNA sequencing
Plasmid from bacterial colonies was initially tested for the presence of an insert by digestion with EcoRI (GIBCO-BRL). The plasmid DNA was transferred to a nylon membrane (Amersham), UV cross-linked and hybridized to either of VHJ558-specific DNA probes, VNP.B4 (23) and L2M7-1v (nMeH1from this study), digoxygenin (DIG) labeled using the High Prime DIG system (Boehringer-Mannheim). The hybridization signal was detected with the chemiluminescent substrate CSPD (Boehringer-Mannheim). Known positive and negative controls were included with each Southern hybridization. The VHJ558+ clones were sequenced in both directions using M13 reverse and M13 forward primers (MOBIX, McMaster University, Hamilton) and numbered according to Kabat et al. (24). The DNA sequences can be accessed from GenBank under the accession numbers AF218628–AF218702.

DNA sequence analysis
The DNA sequences of unique VDJ recombinations, excluding duplicate isolates, were used to search for sequence similarity using the BLAST program (25). The DH and JH genes were assigned by visual comparison to published sequences (24–27). The hydrophobicity indices were plotted according to the method of Kyte and Doolittle (28). The pI-values for CDR3H were obtained using the web-based program (http://www.expasy.ch/tools/pi_tool.html). The VH subfamily phylogenetic trees were compiled using MegAlign (DNAStar Inc., Madison, WI, USA).

Statistical analysis
The SAS (29) software package was used to analyze statistical differences using Fischer's exact chi-squared test. The difference in the averages of AG(C/T)/TCN (hot spot) ratios and pI were calculated using the Student's t-test. The odds ratios (OR) were calculated using GraphPad InStat version 3.05 (GraphPad Software, San Diego, CA, USA). The sophisticated logistic regression analysis of mean gene expression values was performed with S-Plus 7.0.0 (Insightful Corporation, USA) and a value <1.96 was considered significant.


   Results
Top
 Abstract
 Introduction
 Methods
 Results
Discussion
 Supplementary data
 References
 
Individual VHJ558 gene expression is skewed in neonatal mev/mev as compared with mev/+ B lymphocytes
The cDNA libraries were constructed from neonatal B lymphocytes originating from a single mev/mev and mev/+ mouse each genotyped (22) for SHP-1 mutation (Figure S1; Supplementary Figure S1 is available at International Immunology Online) within 24 h after birth. Forty-seven and twenty-four unique VHJ558-D-J-Cµ rearrangements, excluding duplicate recombinations, were analyzed from an autoimmune-prone mev/mev (Table S1; Supplementary Table S1 is available at International Immunology Online) and phenotypically normal mev/+ (Table S2; Supplementary Table S2 is available at International Immunology Online) strain, respectively, for structural properties. The VH genes (codons 1–94) expressed in newborn mev/mev and mev/+ mice were grouped into subfamilies (Table 1) based on the criteria of ≥94% nucleotide similarity (30–37). Four known VHJ558 subfamilies (V186.2, V3, G4D11and VMU32) were expressed in the neonatal VDJCµ rearrangements (Table 3), in addition to six putative novel VHJ558 subfamilies [5D3 (previously unclassified), MVAR11, MVAR2/10, pM11, VH205.12 and 2.2A4]. Approximately 70% of the VDJ recombinations from both mice expressed members of the 5D3 and MVAR2/10 gene family. A predominant expression of V186.2 and V3 gene subfamilies was noted in splenic-/bone marrow-derived B cells of a 4-month-old CB.20 (30) mouse of the same IgHb haplotype as motheaten strain. This suggests differential expression of VHJ558 genes in B lymphocytes from neonatal and adult mice. The MVAR 2/10 gene expression is favored (OR: 1.92, 0.95 CI, 0.57–6.41) in the SHP-1-deficient mev/mev newborn mouse (31.9%) as compared with phenotypically normal mev/+ (16.7%) strain. Members of the pM11 subfamily were exclusively expressed in the neonatal mev/mev (14.9%) mouse, indicating a statistical trend for their preferential expression (P-value: 0.09). Such differences in VHJ558 gene expression in neonatal mev/mev as compared with mev/+ are significant based on logistic regression analyses (t-value: 1.14). These differentially expressed VHJ558 genes essentially cluster into three major subgroups (Fig. 1) with obvious differences in the germ line repertoire composition of mev/mev as compared with mev/+ due to increased negative B cell selection caused by SHP-1 deficiency. It could be argued that the skewed VHJ558 gene expression may be due to the predominant expansion of the B-1 cell population in mev/mev mice. However, it seems unlikely as the J558 subfamilies (V186.2, V3, G4D11, 205.12 and V130) known to be highly expressed in B-1 cells (30) are not used at an increased frequency in neonatal mev/mev B cells. Further, VDJ rearrangements encoded by 205.12 or 10B10S VH genes, known for increased expression in B-1 cells, were not identified. By contrast, previously unclassified VH5D3 gene subfamily is expressed at high frequency (42.55%) in neonatal mev/mev B cells, though this is not known for its higher expression in B-1 cells.


View this table:
[in this window]
[in a new window]
  		Table 1. Subfamily classification of VHJ558 genes from mev/mev and mev/+ neonatal B cells

 

View this table:
[in this window]
[in a new window]
  		Table 3. D-gene expression among neonatal J558+ VDJCµ rearrangements in neonatal mev/mev and mev/+ mice

 

Figure 1
View larger version (15K):
[in this window]
[in a new window]
  		Fig. 1. Phylogenetic tree of VHJ558+ VDJCµ recombinations in neonatal B cells from mev/mev (a) and mev/+ (b) mouse. Note predominant presence of pM11 group of VHJ558 genes in mev/mev B cells.

 
The skewed germ line VHJ558 gene expression in neonatal B cells results in statistically significant (P-value < 0.05) higher occurrence of mutation hotspots [AG(C/T)/TCN ratio] in CDR2H of IgM from autoimmune-prone mev/mev in contrast to normal mev/+ strain (Table 2). Such a germ line predisposition to mutations in CDR2H of IgM as a result of increased negative selection is likely to aid generation of somatically mutated pathogenic IgG AAbs in the periphery. Overall, these observations suggest that increased negative B cell selection during ontogeny, due to SHP-1 deficiency, results in skewed individual VHJ558 gene expression with an increased predisposition to somatic mutations in the variable region of IgM expressed as nascent primary antibody repertoire.


View this table:
[in this window]
[in a new window]
  		Table 2. Ratio of hot spots [AG(C/T)/TCN] in CDR3H of VHJ558+DJCµ recombinations of neonatal mev/mev and mev/+ B lymphocytes

 
Differential DH gene expression in VDJCµ rearrangements neonatal mev/mev and mev/+ B lymphocytes
In phenotypically normal mev/+ mouse, DSP genes were predominantly expressed (41.2%; Table 3), consistent with the D-gene expression observed in BALB/c mice (27). However, DQ gene expression in mev/+ (IgHb haplotype) is higher (35.3%) than BALB/c (IgHa haplotype) mice and reflects strain-related differences intrinsic to D–J recombination per se (38, 39). In autoimmune-prone mev/mev (42.9%) mouse, DFL D-genes were favorably expressed (OR: 1.87, 0.95 CI, 0.55–6.34) in neonatal B cells as compared with phenotypically normal mev/+ (23.5%) mouse (Table 3) and, also, indicating a statistical trend (P-value ≤ 0.1) for biased D-gene use. The most 5' member of the DFL family, DFL 16.1 gene, was predominantly expressed in both the mice, though its overall expression was higher in mev/mev (38.3%) as compared with mev/+ (16.7%) neonatal B cells. By contrast, the single DQ element closely linked to JH cluster preferred in early or primary DJ rearrangements (26, 27, 40) is expressed at relatively lower frequencies in mev/mev (21.4%) as compared with mev/+ (35.3%) B cells with odds being against its expression (OR: 0.6, 0.95 CI, 0.19–2.02). Similarly, odds are against DSP gene expression in mev/mev as compared with mev/+ strain (OR: 0.7, 0.95 CI, 0.24–2.1). Such differences in D-gene are noted to be significant between mev/mev and mev/+ by logistic regression analysis (t-value: 0.41). These differences reflect cellular selection via µ-heavy chain co-expressed with the surrogate light chain on developing B cell since DQ52 is predominantly expressed in D–J recombinations (40). This necessitates the need to determine the effect of negative selection on pro- and pre-B cell populations that arise early during ontogeny. The DST4 gene expression at expected low frequencies (4.8%) in neonatal mev/mev B cells suggests that it constitutes a functional D-gene in IgHb haplotype. No significant differences in D–D fusions or inverted D-gene segments were identified in SHP-1-deficient and phenotypically normal mice. Overall, these observations indicate a differential D-gene expression with a trend for an increased expression of most 5' DFL 16.1 gene, as a result of increased negative B cell selection during ontogeny.

JH2 gene is predominantly expressed in VDJCµ rearrangements from neonatal mev/mev and mev/+ mice
While no statistically significant differences were observed in JH gene expression (P > 0.05) in VDJCµ rearrangements in neonatal B cells from mev/mev and in mev/+ mice (Fig. 2), odds favored JH2 gene expression in mev/mev (OR: 1.4, 0.95 CI, 0.55–3.62). The JH3 gene was expressed at relatively low frequencies in mev/mev (19.15%) as compared with mev/+ (33.3%) mice with odds being against its expression (OR: 0.5, 0.95 CI, 0.17–1.53). Overall, logistic regression analysis indicates differential JH gene expression in mev/mev as compared with mev/+ strain (t-value: 0.23). A higher expression of JH3 and JH2 (mev/mev: 46.8%; mev/+: 33.3%) genes in neonatal motheaten (IgHb) as compared with fetal BALB/c (IgHa) where JH4 is preferentially (41.8%) expressed (26) suggests strain-related differences. Further, an increased expression of JH2 in neonatal mev/mev B cells (46.8%) is distinct from predominant JH1 expression in J558+ VDJ rearrangements in B-1 cells as compared with B-2 cells of adult C57BL/6 mice of identical IgHb haplotype (41). In addition, these observations differ from a higher JH1 use earlier noted in B-1a cells as compared with B-1b or B-2 cells (42).


Figure 2
View larger version (12K):
[in this window]
[in a new window]
  		Fig. 2. JH gene expression in neonatal VDJCµ recombinations from mev/mev and mev/+ mice.

 
Differential random junctional flexibility is evident in VDJCµ rearrangements from neonatal mev/mev and mev/+ mice
Analysis of nucleotide additions or deletions at D–J and V–D junction (Table 4) did not reveal any statistically significant differences in J558+ VDJ rearrangements from mev/mev and mev/+ mice (P > 0.05). However, odds for deletion of ≥1 codon at VH–D junction were favored in mev/mev strain (OR: 1.8, 0.95 CI, 0.34–9.28). The N- or P-nucleotide additions were evident in significant number of VDJ recombinations of mev/mev (42.6%) and mev/+ (33.3%) neonates. Interestingly, N-nucleotide additions were favored in VDJ recombinations from mev/mev in comparison to mev/+ strain (OR: 4.1, 0.95 CI, 0.48–34.61). In fact, N or P additions at the VDJ junctions in both neonatal homo- and heterozygous motheaten (IgHb) is higher than BALB/c (IgHa) fetal B cells (11.9%; 36), suggesting strain-related differences in CDR3H diversification. Overall, increased negative B cell selection during ontogeny contributes to differential CDR3H diversity in mev/mev as compared with mev/+ by favoring N-nucleotide additions.


View this table:
[in this window]
[in a new window]
  		Table 4. Junctional flexibility in VHJ558-D-J recombinations in neonatal mev/mev and mev/+ B lymphocytes

 
Increased negative selection does not directly lead to generation of disease-associated IgM AAbs
The CDR3H size of J558+ VDJCµ recombinations in neonatal B cells was comparable in mev/mev (7.5 ± 2.4) and mev/+ (7.0 ± 2.8) mice (Fig. 3). Variable odds either for or against selection of particular CDR size in VDJ recombinations from mev/mev and mev/+ existed with significant differences revealed by logistic regression analysis (t-value: 0.09). Nevertheless, the characteristic CDR3H size restriction (10 amino acids) noted in terminal disease state J558+ IgM AAbs (20) does not occur as a result of increased negative selection during B cell ontogeny but rather reflects the subsequent positive ligand-mediated selection in the periphery. The pI-values of CDR3H are statistically insignificant in mutant and normal mice (Tables S3 and S4; Supplementary Tables S3 and S4 are available at International Immunology Online). Interestingly, nMeV112 recombination, encoded by V186.2 subfamily gene (Table S1; Supplementary Table S1 is available at International Immunology Online), is remarkable for a CDR3H length of nine amino acids with a net positive charge (pI: 8.59) and presence of arginine residues (Table S3; Supplementary Table S3 is available at International Immunology Online). It seems that such V186.2-expressing recombinations in B-1 and conventional B cells may possibly expand in the periphery via ligand-mediated positive selection and undergo affinity maturation not via somatic hypermutation but by CDR3H size and amino acid composition based selection. An increased predisposition to hypermutations in the CDR2H region from increased negative selection in the nascent B cell repertoire is likely to facilitate generation of IgG AAbs. These observations demonstrate that increased negative selection does not directly lead to generation of disease-associated IgM AAbs during B cell ontogeny but impairs primary antibody repertoire per se.


Figure 3
View larger version (13K):
[in this window]
[in a new window]
  		Fig. 3. Distribution of CDR3H size in neonatal VDJCµ recombinations from mev/mev and mev/+ mice.

 

   Discussion
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
Supplementary data
 References
 
Susceptibility to complex systemic autoimmune diseases, such as SLE, is determined by polygenic traits (43), including those involved in regulation via antigen receptor signaling. The viable motheaten mouse model permits analysis of one such single component of polygenic trait, SHP-1 deficiency, in generation of disease-associated IgM AAbs as an outcome of increased negative B cell selection during B cell ontogeny. Earlier, we noted skewed VH (J558 and Q52) gene expression and VH + V{kappa} pairings in adult mev/mev B cells (21) followed by demonstration of unique structural properties of randomly selected J558+ IgM AAbs at terminal disease state (20), especially restricted CDR3H size (10 amino acids). While these observations suggested positive ligand-mediated selection of disease-associated IgM AAbs, it remained to be excluded if increased negative selection based on BCR signaling strength led to the emergence of these IgM AAbs during B cell ontogeny. Indeed, the architecture for CDR3H size and composition noted in mature B cells is already established at pre-B cell stage (44). The experiment outlined here unequivocally demonstrate that IgM AAbs of pathogenic potential do not directly arise as a result of increased negative selection during B cell ontogeny because none of the VDJ recombinations from mev/mev mice had structural characteristics similar to those seen in IgM AAbs at terminal disease state, for example, restricted CDR3H size (20). These observations are consistent with the fact that IgM AAbs in autoimmune mice, similar to IgG AAbs, bear structural properties similar to an induced antibody that arise via clonal selection (1, 45) with the exception of somatic hypermutations. Though, natural AAbs are also suggested to be involved in the pathogenesis of autoimmune diseases such as hemolytic anemia (46) and SLE (47) but it is uncertain as shorter CDR3H are frequently associated with polyreactivity (20, 48, 49).

Increased negative B cell selection during ontogeny influences the composition of the pre-immune repertoire as evidenced by skewed individual VHJ558 gene expression, for example, over-expression of MVAR2/10 and pM11 VH genes. It could be argued that these differences occur because of the predominance of B-1 (13) cells in mev/mev mice, but none of the VHJ558 genes known for their expression in B-1 cells are noted in the VDJ recombinations analyzed. The over-expression of DFL D-genes observed in mev/mev as compared with mev/+ mice reflects a trend that could be attributed to escape from increased negative selection. The known bias for RF1 was noted for DSP and DFL genes expressed in both the mutant and phenotypically normal mice. A similar bias was, however, not evident for DQ genes in neonatal mev/+ B cells. The possibility of counterselection for RF2 in DQ gene, expressed at lower than expected frequencies (26, 27, 40) in mev/mev as compared with normal mev/+ mouse deserves to be investigated further as it indicates negative selection at pre-B cell level co-expressing µ-heavy chain together with surrogate light chain (50). In addition, these observations appear to be in contrast to a higher proportion of D–D fusions and D-gene inversions suggested in the MRL autoimmune-prone mouse as compared with a C3H control mouse (51). But these seem to be due to DST gene expression rather than D-gene fusion and/or inversion events and, therefore, are not inconsistent with the observations from mev/mev mice. A trend toward differential expression of a previously unclassified J558 subfamily, designated pM11, as well as the DFL D-gene family, suggests that neonatal repertoire is skewed leading to an impaired primary antibody repertoire development as a result of increased negative selection. Such an impaired pre-immune repertoire per se provides the platform for the origin of AAb specificities of pathogenic potential by subsequent ligand-mediated positive selection in the periphery, influenced by the strength of BCR signaling (52, 53). These observations are supported by the observed skewed VH gene expression that underlies B cell overactivity during initial stages failing to prevent emergence of autoimmunity in SLE patients (54). This is also consistent with increased occurrence of consensus triplet (AGPy) hotspots for somatic mutation in the CDR2H region of the neonatal Ig B cell antigen receptor that provide a substrate for the production of IgG AAb of pathogenic ability. Such an enhanced mutability is reminiscent of enhanced mutation activity and abnormal selection of VH (55), V{kappa} (56) and V{lambda} (57) genes associated with emergence of autoimmunity in SLE patients. These observations are, however, in contrast to resistance of autoreactivity in DBA/2 B cells that display increased strength of BCR signaling following peptide immunization that leads to the development of anti-DNA antibodies in the haplotype matched BALB/c (H-2d) mice (52). These differences could be attributed to differential antibody repertoire dynamics, for example, predominance of B-1 cells, involved in generation of spontaneous autoimmunity. Nevertheless, these studies are consistent with our observations that BCR signaling strength that determines increased negative B cell selection influences the germ line IgM repertoire composition such that it may predispose to the development of spontaneous or induced AAbs. Interestingly, signaling through the IgM on CD5+ B cells does not induce proliferation, but instead leads to apoptosis (58). This process is probably mediated via SHP-1 constitutively associated with CD5 and CD72 or related CD5 ligand. CD5 is implicated in blocking the dissociation of SHP-1 from mIgM following BCR ligation (59). As a result of SHP-1 deficiency, it is possible that CD5 fails to negatively regulate B-1 cells and permits the escape of dangerously autoreactive B-1 cells in homozygous viable motheaten mice. Analysis of antibody specificities in the neonatal and adult mev/mev/CD5–/– mice is expected to shed light if loss of CD5-mediated negative selection leads to the generation of disease-associated IgM AAb. Indeed, recent studies with mev/mev/CD5–/– mice showed an increased life span (up to 69%), reduced pulmonary inflammation and lower serum IgM levels (60), suggesting involvement of CD5 glycoprotein in negative selection. But levels of AAbs against dsDNA and histone were not significantly different between mev/mev and mev/mev/CD5–/– mice highlighting further complexity. This is consistent with the suggestion that phosphorylated CD5 may be responsible for abnormal BCR signaling in some but not all B-1 cell (61, 62). Further studies involving experimental conditions that permit negative but not positive selection of B cells both centrally and peripherally will help understand the contribution of these factors to the generation of pathogenic IgM AAb. Thus, understanding relationships between negative and positive BCR signaling strength-based selection is essential to origin of disease-associated AAbs if these were an outcome of auto-antigen-driven positive selection or antigen-independent constitutive signaling (63). Further consideration should be given to SHP-1-mediated regulation of TCR signaling strength that influences the generation of CD4+CD25+ regulatory T cells (64) and, also, possibly Th2 cells.

Overall, the oultlined experiments suggest that the increased negative B cell selection, due to decreased threshold of responsiveness of BCR, for a ligand during ontogeny impairs the developing IgM antibody repertoire but does not directly lead to the generation of disease-associated IgM AAbs.


   Supplementary data
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Supplementary data
References
 
Supplementary Figure S1 and Tables S1–S4 are available at International Immunology Online.


   Acknowledgements
 
The authors would like to thank D. Bridle and his associates at the OVC Isolation Unit for their support with the mice breeding work. The research was supported by research grant from Natural Sciences and Engineering Research Council of Canada to A.K.K. The statistical analysis support of Arni S. R. S. Rao is gratefully acknowledged.


   Abbreviations
 
AAb, auto-antibody
DIG, digoxygenin
dsDNA, double-strand DNA
OR, odds ratio
SLE, systemic lupus erythematosus

   Notes
 
Transmitting editor: S. Izui

Received 11 July 2005, accepted 31 January 2006.


   References
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Supplementary data
 References
 

  1. Shlomchik M. J., Marshak-Rothstein A., Wolfowicz C. B., Rothstein T. L. and Weigert M. G. 1987. The role of clonal selection and somatic mutation in autoimmunity. Nature 328: 805.[CrossRef][Medline]
  2. Adib M., Ragimbeau J., Avrameas S. and Ternynck T. 1990. IgG autoantibody activity in normal mouse serum is controlled by IgM. J. Immunol. 145: 3807.[Abstract]
  3. Kiely J.-M., Braun J. and Unonaue E. R. 1985. Autoregulation of B cell growth by an immunoglobulin M autoantibody. J. Immunol. 135: 1040.[Abstract]
  4. Shultz L. D., Coman D. R., Beamer W. G. and Sidman C. L. 1984. ‘Viable motheaten’, a new allele at the motheaten locus I. Pathology. Am. J. Pathol. 116: 179.[Abstract]
  5. Matthews R. J., Bowne D. B., Flores E. and Thomas M. L. 1992. Characterization of hematopoietic intracellular protein tyrosine phosphatases: description of a phosphatase containing an SH2 domain and another enriched in proline-, glutamic acid-, serine-, and threonine-rich sequences. Mol. Cell Biol. 12: 2396.[Abstract/Free Full Text]
  6. Greiner D. L., Goldschneider E., Komsclies K. L., Medlock E. S., Bollum F. J. and Schultz L. 1986. Defective lymphopoiesis in bone marrow of motheaten (me/me) and viable motheaten (mev/mev) mutant mice I. Analysis of development of prothymocytes, early B lineage cells and terminal deoxynucleotidyl transferase-positive cells. J. Exp. Med. 164: 1129.[Abstract/Free Full Text]
  7. Shultz L. D., Schweitzer P. A., Rajan T. V. et al. 1993. Mutations at the murine motheaten locus are within the hematopoietic cell protein-tyrosine phosphatase (Hcph) gene. Cell 73: 1445.[CrossRef][Web of Science][Medline]
  8. Cyster J. G. and Goodnow C. C. 1995. Protein tyrosine phosphatase 1C negatively regulates antigen receptor signalling in B lymphocytes and determines thresholds for negative selection. Immunity 2: 13.[CrossRef][Web of Science][Medline]
  9. Pani G., Kozlowski M., Cambier J. C., Mills G. B. and Siminovitch K. A. 1995. Identification of the tyrosine phosphatase PTP1C as a B cell antigen receptor-associated protein involved in the regulation of B cell signalling. J. Exp. Med. 181: 2077.[Abstract/Free Full Text]
  10. Zhang J., Somany A. K. and Siminovitch K. A. 2000. Roles of the SHP-1 tyrosine phosphatase in the negative regulation of cell signalling. Semin. Immunol. 12: 361.[CrossRef][Web of Science][Medline]
  11. Cornall R. J., Goodnow C. C. and Cyster J. G. 1995. The regulation of self-reactive B cells. Curr. Opin. Immunol. 7: 804.[CrossRef][Web of Science][Medline]
  12. Frearson J. A. and Alexander D. R. 1997. The role of phosphotyrosine phosphatases in haematopoietic cell signal transduction. Bioessays 19: 417.[CrossRef][Web of Science][Medline]
  13. Sidman C. L., Shultz L. D., Hardy R. R., Hayakawa K. and Herzenberg L. A. 1986. Production of immunoglobulin isotypes by Ly-1+ B cells in viable motheaten and normal mice. Science 232: 1423.[Abstract/Free Full Text]
  14. Hayakawa K., Hardy R. R., Parks D. R. and Harzenberg L. A. 1983. The Ly-1B cell subpopulation in normal, immunodefective and autoimmune mice. J. Exp. Med. 157: 202.[Abstract/Free Full Text]
  15. Hayakawa K., Hardy R. R., Honda M., Harzenberg L. A. and Steinberg A. D. 1984. Ly-1 B cells: functionally distinct lymphocytes that secrete IgM autoantibodies. Proc. Natl Acad. Sci. USA 81: 2494.[Abstract/Free Full Text]
  16. Christianson S. W., Greiner D. L., Deluca D. et al. 2002. T cell developmental defects in ‘viable motheaten’ mice deficient in SHP-1 protein-tyrosine phosphatase. Developmental defects are corrected in vitro in the presence of normal hematopoietic-origin stromal cells and in vivo by exogenous IL-7. J. Autoimmun. 18: 119.[CrossRef][Web of Science][Medline]
  17. Kasturi K., Mayer R., Bona C. A., Scott V. E. and Sidman C. L. 1990. Germline V genes encode viable motheaten mouse autoantibodies against thymocytes and red blood cells. J. Immunol. 145: 2304.[Abstract]
  18. Shultz L. D. and Zurier R. B. 1978. "Motheaten": a single gene model for stem cell dysfunction and early onset autoimmunity. In Rose N. R., Bigazzi P. E. and Warner N. L., eds, Genetic Control of Autoimmune Disease. Developments in Immunology, Vol. 1, p. 229. Elsevier, North Holland, Amsterdam.
  19. Kovarik J., Kuntz L., Ryffel B. and Borel J. F. 1994. The viable motheaten (mev) mouse—a new model for arthritis. J. Autoimmun. 7: 575.[CrossRef][Web of Science][Medline]
  20. Lipsanen V., Walter B., Emara M., Siminovitch K., Lam J. and Kaushik A. 1997. Restricted CDR3 length of the heavy chain is characteristic of six randomly isolated disease-associated VHJ558+ IgM autoantibodies in lupus prone motheaten mice. Int. Immunol. 9: 655.[Medline]
  21. Saitoh Y., Kelsoe G., Bona C. and Kaushik A. 1995. Skewed VH and V{kappa} gene family expression and pairing occurs among B lymphocytes in autoimmune motheaten mice. Autoimmunity 21: 185.[Web of Science][Medline]
  22. Pani G., Siminovitch K. A. and Paige C. J. 1997. The motheaten mutation rescues B cell signalling and development in CD45-deficient mice. J. Exp. Med. 181: 581.
  23. Yancopoulos G. D. and Alt F. W. 1985. Developmentally controlled and tissue-specific expression of unrearranged VH gene segments. Cell 40: 271.[CrossRef][Web of Science][Medline]
  24. Kabat E. A., Wu T. T., Perry H. M., Gottesman K. S. and Foeller C. 1991. Sequences of proteins of immunological interest, 5th edition. National Institute of Health, USA.
  25. Altschul S. F., Madden T. L., Schaffer A. A. et al. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25: 3389.[Abstract/Free Full Text]
  26. Bangs L. A., Sanz I. E. and Teale J. M. 1991. Comparison of D, JH and junctional diversity in the fetal, adult and aged B cell repertoires. J. Immunol. 146: 1996.[Abstract]
  27. Feeney A. J. 1990. Lack of N regions in fetal and neonatal mouse immunoglobulin VDJ junctional sequences. J. Exp. Med. 172: 1377.[Abstract/Free Full Text]
  28. Kyte J. and Doolittle R. F. 1982. A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157: 105.[CrossRef][Web of Science][Medline]
  29. SAS Institute Inc. SAS/STAT software (1997): changes and enhancements through release 6.12. SAS Institute Inc., Cary, North Carolina. p. 1167.
  30. Gu H., Tarlinton D., Muller W., Rajewsky K. and Forster I. 1991. Most peripheral B cells in mice are ligand selected. J. Exp. Med. 173: 1357.[Abstract/Free Full Text]
  31. Bothwell A. L. M., Paskind M., Reth M., Imanishi-Kari T., Rajewsky K. and Baltimore D. 1981. Heavy chain variable region contribution to the b-NP family of antibodies: somatic mutation evident in a gamma-2a variable region. Cell 24: 625.[CrossRef][Web of Science][Medline]
  32. Maizels N. and Bothwell A. L. M. 1985. The T-cell-independent immune response to the hapten NP uses a large repertoire of heavy chain genes. Cell 43: 715.[CrossRef][Web of Science][Medline]
  33. Winter E., Radbruch A. and Krawinkel U. 1985. Member of novel VH gene families are found in VDJ regions of polyclonally activated B-lymphocytes. EMBO J. 4: 2861.[Web of Science][Medline]
  34. Blankenstein T., Zoebelein G. and Krawinkel U. 1984. Analysis of immunoglobulin heavy chain V-region genes belonging to the V NP-gene family. Nucleic Acids Res. 12: 6887.[Abstract/Free Full Text]
  35. Kaartinen M., Pelkonen E., Even J. and Makela O. 1988. V genes of the primary antibody response of C57BL/10 mice to the hapten phenyloxazolone. Eur. J. Immunol. 18: 1095.[Web of Science][Medline]
  36. Ono M., Yamamoto T., Kyogoku M. and Nose M. 1995. Sequence analysis of the germ-line VH gene corresponding to a nephritogenic antibody in MRL/lpr lupus mice. Clin. Exp. Immunol. 100: 284.[Web of Science][Medline]
  37. Both G. W., Taylor L., Pollard J. W. and Steele E. J. 1990. Distribution of mutations around rearranged heavy-chain antibody variable-region genes. Mol. Cell Biol. 10: 5187.[Abstract/Free Full Text]
  38. Alessandro A. and Desiderio S. V. 1991. Coordination of immunoglobulin DJH transcription and D to JH rearrangement by promoter-enhancer approximation. Mol. Cell Biol. 11: 2096.[Abstract/Free Full Text]
  39. Nitschke L., Kestler J., Tallone T., Pelkonen S. and Pelkonen J. 2001. Deletion of the DQ52 element within the Ig heavy chain locus leads to a selective reduction in VDJ recombination and altered D gene usage. J. Immunol. 166: 2540.[Abstract/Free Full Text]
  40. Tsukada S., Sugiyama H., Oka Y. and Kishimito S. 1990. Estimation of D segment usage in initial D to JH joinings in a murine immature B cell line. J. Immunol. 144: 643.
  41. Tornberg U.-C. and Holmberg D. 1995. B-1a, B-1b and B-2 cells display unique VHDJH repertoires formed at different stages of ontogeny and under different selection pressures. EMBO J. 14: 1680.[Web of Science][Medline]
  42. Kantor A. B., Merrill C. E., Herzenberg L. A. and Hillson L. L. 1997. An unbiased analysis of VH-D-JH sequences from B-1a, B-1b and conventional B cells. J. Immunol. 158: 1175.[Abstract]
  43. Santiago-Raber M.-L., Laporte C., Reninger L. and Izui S. 2004. Genetic basis of murine lupus. Autoimmun. Rev. 3: 33.[CrossRef][Web of Science][Medline]
  44. Ivanov I. I., Schelonka R. L., Zhuang Y., Gartland G. L., Zemlin M. and Schroder H. W., 2005. Development of the expressed Ig CDR-H3 repertoire is marked by focusing of constraints in length, amino acid use, and charge that are first established in early B cell progenitors. J. Immunol. 174: 7773.[Abstract/Free Full Text]
  45. Tillman D. M., Jou N. T., Hill R. J. and Marion T. N. 1992. Both IgM and IgG anti-DNA antibodies are the products of clonally selective B cell stimulation in (NZB x NZW)F1 mice. J. Exp. Med. 176: 761.[Abstract/Free Full Text]
  46. Hentati B., Payelle-Brogard B., Jouanne C., Avrameas S. and Ternynck T. 1994. Natural autoantibodies are involved in the haemolytic anaemia of NZB mice. J. Autoimmun. 7: 425.[CrossRef][Web of Science][Medline]
  47. Matsiota P., Druet P., Dosquet P., Guilbert B. and Avrameas S. 1987. Natural autoantibodies in systemic lupus erythematosus. Clin. Exp. Immunol. 69: 79.[Web of Science][Medline]
  48. Chen C., Stenzel-Poore M. P. and Rittenberg M. 1991. Natural auto- and polyreactive antibodies differing from antigen-induced antibodies in the H-chain CDR3. J. Immunol. 147: 2359.[Abstract]
  49. Dammers P. M. and Kroese G. M. 2005. Recruitment and selection of marginal zone B cells is independent of exogenous antigens. Eur. J. Immunol. 35: 2089.[CrossRef][Web of Science][Medline]
  50. Haasner D., Rolink A. and Melchers F. 1994. Influence of surrogate light chain on DHJH-reading frame 2 suppression in mouse precursor B cells. Int. Immunol. 6: 21.[Abstract/Free Full Text]
  51. Klonowski K. D., Primiano L. L. and Monestier M. 1999. Atypical VH-D-JH rearrangements in newborn autoimmune MRL mice. J. Immunol. 162: 1566.[Abstract/Free Full Text]
  52. Wang C., Khalil M., Ravetch J. and Diamond B. 2003. The naïve B cell repertoire predisposes to antigen induced systemic lupus erythematosus. J. Immunol. 170: 4826.[Abstract/Free Full Text]
  53. Cancro M. and Kearney J. F. 2004. B cell positive selection: roadmap to the primary repertoire? J. Immunol. 173: 15.[Abstract/Free Full Text]
  54. Dorner T., Farner N. L. and Lipsky P. E. 1999. Ig {lambda} and heavy chain gene usage in early untreated systemic lupus erythematosus suggests intensive B cell stimulation. J. Immunol. 163: 1027.[Abstract/Free Full Text]
  55. Jacobi A. M., Hansen A., Burmester G. R., Dorner T. and Lipsky P. E. 2000. Enhanced mutational activity and disturbed selection of mutations in VH gene rearrangements in a patient with systemic lupus erythematosus. Autoimmunity 33: 61.[Web of Science][Medline]
  56. Dorner T., Heimbacher C., Farner N. L. and Lipsky P. E. 1999. Enhanced mutational activity of V{kappa} gene rearrangements in systemic lupus erythematosus. Clin. Immunol. 92: 188.[CrossRef][Web of Science][Medline]
  57. Dorner T., Kaschner S., Hansen A., Pruss A. and Lipsky P. E. 2001. Perturbations in the impact of mutational activity on V{lambda} genes in systemic lupus erythematosus. Arthritis Res. 3: 368.[CrossRef][Web of Science][Medline]
  58. Bikah G., Carey J., Ciallella J. R., Tarakhovsky A. and Bondada S. 1996. CD5-mediated negative regulation of antigen receptor-induced growth signals in B-1 B cells. Science 274: 1906.[Abstract/Free Full Text]
  59. Sen G., Bikah G., Venkataraman C. and Bondada S. 1999. Negative regulation of antigen receptor-mediated signalling by constitutive association of CD5 with the SHP-1 protein tyrosine phosphatase in B-1 B cells. Eur. J. Immunol. 29: 3319.[CrossRef][Web of Science][Medline]
  60. Joliat M. J., Lang P. A., Lyons B. L. et al. 2002. Absence of CD5 dramatically reduces progression of pulmonary inflammatory lesions in SHP-1 protein-tyrosine phosphatase-deficient viable motheeaten mice. J. Autoimmun. 18: 105.[CrossRef][Web of Science][Medline]
  61. Sen G. H., Wu H. J., Bikah C. et al. 2002. Defective CD19-dependant signalling in B-1a and B-1b lymphocyte subpopulations. Mol. Immunol. 39: 57.[CrossRef][Web of Science][Medline]
  62. Frances R., Tumang J. R. and Rothstein L. 2005. B-1 cells are deficient in Lck: defective B cell receptor signal transduction in B-1 cells occurs in the absence of elevated Lck expression. J. Immunol. 175: 27.[Abstract/Free Full Text]
  63. Pillai S., Cariappa A. and Moran S. T. 2005. Marginal zone B cells. Annu. Rev. Immunol. 23: 161.[CrossRef][Web of Science][Medline]
  64. Carter J. D., Calabrese G. M., Naganuma M. and Lorenz U. 2005. Deficiency of the Src homology region 2 domain-containing phosphatase 1 (SHP-1) causes enrichment of CD4+CD25+ regulatory T cells. J. Immunol. 174: 6627.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Mol. Cell. Biol.Home page
B. Gozalbo-Lopez, P. Andrade, G. Terrados, B. de Andres, N. Serrano, I. Cortegano, B. Palacios, A. Bernad, L. Blanco, M. A. R. Marcos, et al.
A Role for DNA Polymerase {micro} in the Emerging DJH Rearrangements of the Postgastrulation Mouse Embryo
Mol. Cell. Biol., March 1, 2009; 29(5): 1266 - 1275.
[Abstract] [Full Text] [PDF]

Home page
 BloodHome page
Y. Kanamaru, H. Tamouza, S. Pfirsch, D. El Mehdi, C. Guerin-Marchand, M. Pretolani, U. Blank, and R. C. Monteiro
IgA Fc receptor I signals apoptosis through the FcR{gamma} ITAM and affects tumor growth
Blood, January 1, 2007; 109(1): 203 - 211.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
18/5/661    most recent
dxl003v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (3)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Cassady-Cain, R. L
Right arrow Articles by Kaushik, A. K
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Cassady-Cain, R. L
Right arrow Articles by Kaushik, A. K
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?