Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 Dec 12.
Published in final edited form as: Mol Immunol. 2012 Apr 18;51(0):273–282. doi: 10.1016/j.molimm.2012.03.028

B cell receptor light chain repertoires show signs of selection with differences between groups of healthy individuals and SLE patients

Nathan Schoettler a, Dongyao Ni a, Martin Weigert a
PMCID: PMC4264353  NIHMSID: NIHMS573307  PMID: 22516082

Abstract

We have developed a microarray to study the expression of L-chain V genes (VL genes) in healthy and SLE patient peripheral κ- and λ-sorted B cells. In all repertoires tested, one VL gene accounts for over 10% of all gene VL expression, consistent with positive selection acting on L-chains. While a few VL genes were highly expressed in all individuals, most VL genes were expressed at different levels. Some VL genes (5 out of a total of 78) were not detected. We attribute their absence from the repertoire to negative selection. Positive selection and negative selection were also found in SLE repertoires, but expression of VL genes was different; the differences point to less regulation of VL gene repertoires in SLE. Our data shows that VL gene expression is variable and supports a model where the L-chain repertoire is generated by both positive and negative selection on L-chains.

Keywords: Receptor editing, Light chain, B cell repertoire, SLE

1. Introduction

B cell receptors are formed by pairing of Ig heavy chain (H-chain) and light chain (L-chain) that together determine the specificity of the receptor. Thus, the initial repertoire diversity is due to combinatorial association of H-chains and L-chains as well as combinatorial joining of V, D and J gene segments. (Kawasaki et al., 2001; Kawasaki et al., 1997; Matsuda et al., 1998). Additional diversity arises at the junction of V, D and J gene segments by N-nucleotide additions, P-nucleotide additions and nibbling (Weigert et al., 1980; Wu and Kabat, 1970). Selection acts on the B cell repertoire through both tolerance mechanisms and antigen selection. Tolerance selects B cells during development when autoreactive receptors are formed and acts on individual B cells through receptor editing, a process which replaces the originally rearranged L-chain with a newly-recombined L-chain (Gay et al., 1993; Halverson et al., 2004; Radic et al., 1993; Tiegs et al., 1993). In the case of antigen selection, H-chain and L-chain pairs present in the pre-immune repertoire undergo clonal expansion. We are interested in the L-chain repertoire of B cells that have undergone selection.

Previous repertoire analysis has used a variety of approaches such as isoelectric focusing, Southern blotting and PCR amplification (Bentley, 1984; de Wildt et al., 1997; Klobeck et al., 1984; Mattson et al., 1982; Meffre et al., 2001; Meffre et al., 2000; Walker et al., 1983; Williams et al., 1987); (Dorner et al., 1999; Dorner et al., 2002; Dorner et al., 2001; Farner et al., 1999; Foster et al., 1997; Girschick and Lipsky, 2002; Heimbacher et al., 2001; Jacobi et al., 2002; Kaschner et al., 2001; Lee et al., 2004; Lee et al., 2000; Meffre et al., 2004; Wardemann et al., 2004; Wrammert et al., 2008). However, these studies have limitations, including culture efficiency, PCR biases and cost. To examine L-chain repertoires, we developed a microarray that surveys all human VL genes. There are several advantages to the microarray method: it is not subject to PCR primer bias; the microarray is highly specific, permitting resolution of members within multi-gene families; and it allows comparison of individual or group repertoires. Because of these features, the microarray allows for complete repertoire analysis on a large scale.

The repertoires of κ and λ B cells from healthy individuals were determined using this method. We also studied L-chain repertoires in patients with SLE, a disease in which tolerance is broken. High expression of VL genes was observed in each of the repertoires tested. A small, but substantial number of VL genes could not be detected in the repertoires, individually and in some cases collectively. Within healthy and SLE groups, expression of individual VL genes varied from one individual to the next, and comparison of healthy and SLE patient L-chain repertoires identified several differences in Vκ and Vλ gene expression.

2. Materials and methods

2.1. Healthy and SLE blood samples

Twelve mL (12 mL) of peripheral blood was collected in BD vacutainer blood collection tubes with EDTA (Becton Dickinson) from 10 healthy donors and 10 SLE patients with evidence of active disease (identified by low complement (C3 or C4) levels or clinical signs of active SLE) after informed consent. The use of human blood was approved by the University of Chicago, Institutional Review Board (protocol 14801B). The blood was immediately processed for cell isolation and subsequent cell sorting.

2.1. FACS sorting

Blood was diluted 1:1 in PBS immediately after collection, and Lymphocyte Separation Medium (Mediatech) was used for isolation of mononuclear cells according to the manufacturer's protocol. Cells were resuspended in RPMI supplemented with 10% FBS at a concentration of 108 cells/mL. Cells were stained using the following antibodies for FACS sorting at the concentrations recommendened by the manufacturer: anti-CD138 APC (Dako), anti-Igκ FITC (BD Pharmingen), anti-Igλ PE (BD Pharmingen),anti-CD20 PE (BD Pharmingen), anti-CD20 PE (BD Pharmingen), anti-CD27 APC (eBiosciences) and anti-CD38 PE-Cy7 (eBiosciences). Eight samples from each group were phenotyped for CD27 and CD38 levels. Sorting was performed using MoFlo (Dako-Cytomation) and FACSAria (BD Biosciences) machines using the single cell sort purity mode. Lymphocytes were gated according to forward- and side-scatter profiles and doublets were excluded using forward-scatter pulse width and pulse area profiles. One-thousand events in the CD20+CD138-Igκ+Igλ- and 1,000 events in the CD20+CD138-Igκ-Igλ+ gates were sorted into RNALater (Ambion) and stored at −20°C. Although rare in peripheral blood, plasma cells were specifically excluded in the sorting strategy based upon the following: forward- and side-scatter gates were set to exclude plasma cells; CD138+ cells were not gated; and plasma cells have absent or low levels of surface CD20, κ and λ.

2.3. cDNA preparation

Total RNA for each sample was isolated using TRIZOL (Invitrogen). The RNA concentration and quality was determined using a NanoDrop (Thermo Scientific), and only RNA with 260/280 greater than 1.75 were used for amplification. Twenty ng (20 ng) of total RNA was used for cDNA amplification using Ovation RNA Amplification System V2 (Nugen) and purified using DNA Clean and Concentrator – 25 columns (Zymo research) according to Nugen protocols. If necessary, samples were stored at −80°C until the day of hybridization. Purified cDNA used for the microarray was labeled by adding 4 μg of cDNA in 20 μl to 10 μl of Ulysis Alexa-Fluor 647 (Invitrogen) and heating to 80°C for 20 minutes. Labeled cDNA was purified using Bio-Spin 6 SSC columns (Bio-Rad) after washing the columns 3 times with water. A NanoDrop was used to quantify the cDNA amount and amount of Alexa-Fluor 647.

2.4. Syndecan-1 (CD138) PCR

Primers were designed based on NCBI reference sequence for syndecan (NM_002997.4). The forward primer (AAATGGCAAAGGAAGGTGGATGGC) and reverse primer (ATACACTCCAGGCAGAAAGTCGCA) were synthesized by IDT DNA, and the PCR steps were: 95°C for 5 minutes; 30 cycles of 95°C for 30 seconds, 55°C for 30 seconds and 72°C for 50 seconds; and 72°C for 10 minutes. Reagents were used at the recommended concentrations for the polymerase, Taq DNA Polymerase (Invitrogen).

2.5. Microarray probe design

Light chain V region sequences used for probe design were obtained from publications on the human IGK and IGL loci (Kawasaki et al., 2001; Kawasaki et al., 1997; Schable and Zachau, 1993). Control genes were included on the microarray; sequences for ACTA (NCBI reference sequence: NM_001100), ACTB (NM_001101), AKR1B1 (NM_001628.2), B2M (NM_004048.2), CD19 (NM_001770), GAPDH (NM_002046), IGKC (J00241.1), LDHA (NM_005566), MS4A1 (NM_152866) and NONO (NM_001145409) were downloaded from the National Center for Biotechnology Institute gene database (http://www.ncbi.nlm.nih.gov/sites/entrez?db=gene). Probe sequences were determined by filtering VL gene sequences for: probe length (65–74 bp) aligning to any region of the VL gene, uniqueness compared with all VL genes (BLAST), self-binding (Smith and Waterman, 1981), complexity, melting temperature (69.0±1.5°C) and distance from 3' end. Probes were named and are reported using the original nomenclature to be consistent with the microarray data files. Seven pairs of Vκ genes were identical or nearly-identical, and probes meeting the design criteria correspond to identical regions in the pair. For these Vκ genes, expression is reported together (e.g. O8 and O18 are detected by the probe O8/O18). For the expression levels by cluster or position, estimates of expression for each gene within the pair were assumed to be equal. Appendix Table A.1 lists sequences for the VL gene probes. Probes were manufactured by Integrated DNA Technologies.

2.6. Reference sample

A reference sample containing the reverse-complement of all VL gene probe sequences at equal molar concentrations and 0.12 ng of each reference sequences was co-hybridized with every cDNA sample (Integrated DNA Technologies). The reference was labeled using Ulysis Alexa-Fluor 555 (Invitrogen). For the series of experiments used to estimate the amount of each gene present, the reference was labeled using the Ulysis Alexa-Fluor 647 dye (Invitrogen).

2.7. Estimation of expression level

The complementary reverse sequence of two Vκ genes (B2 and O2/O12) and two Vλ genes (2–13 and 1–19) were hybridized using the same techniques and methods as the cDNA samples. The four genes were chosen from different gene families (Figure A.2). The effect of increasing DNA concentration on signal intensity was also determined using these four genes by adding Alexa Fluor 647-labeled Vκ and Vλ targets to Alexa Fluor 647-labele reference sample at known concentrations (see 2.11 and Figure A.3), which was then hybridized along with the Alexa Fluor 555-labeled reference sample. Vκ B2 was tested at 12.2%, 21.7% and 41.0%; Vκ O2/O12 was tested at 5.2% and 10.0%; Vλ 2–13 was tested at 10.6%, 19.2% and 37.3%; and Vλ 1–19 was tested at 19.2%, 8.7% and 4.6% (percent refers to the molar amount of the gene present in the sample). Each VL gene and concentration was hybridized two times. The second hybridization of V1–19 at 4.6% had a scratch across a portion of the array and was not included in the analysis. Normalized expression values for each of these hybridizations (reference subtracted—see section 2.11 for data analysis and normalization) were compared with the concentration of the genes in the hybridized sample. The Curve Fitting Tool in Matlab was used to identify the best-equation for this data. This equation was then used to estimate expression levels for all of the cDNA samples.

2.8. Microarray spotting

Microarrays were spotted using a GeneMachines OmniGrid 100 (Genomic Solutions) onto SuperAmine 2 slides (ArrayIt). Each oligonucleotide probe was spotted twelve times per array, and the print layout was such that these twelve replicates were spotted by four different pins. In addition to the human oligonucelotides, the microarrays also had mouse oligonucleotides spotted on the array. After printing, microarrays were dehydrated following the manufacture's recommendations and stored in the dark at room temperature.

2.9. Microarray hybridization and scanning

Prior to hybridization, microarrays were re-hydrated, washed (Wash Buffers A, B and C, ArrayIt), prehybridized for two hours (BlockIt Blocking Buffer, ArrayIt) and washed again. Hybridizations were performed with 10 ng of the labeled reference sample added to 1.5 μg labeled cDNA and 2 μg human Cot-1 DNA (Invitrogen). DNA was mixed with HybIt 2 hybridization buffer (ArrayIt) at 1x concentration to a final volume of 20 μL, denatured at 80°C for 3 minutes and hybridized overnight at 37.0°C for 14–16 hours. Following hybridization, microarrays were washed with Wash Buffers A, B and C for 2 minutes each, dried and scanned using an Axon GenePix 4000B scanner (Molecular Devices). All amplified and labeled cDNA samples were hybridized two times as technical replicates. cDNA labeling, prehybridization, hybridization, washing and scanning were all performed by the same individual at approximately the same time of the day.

2.10. Data quality

Image analysis was performed using GenePix Pro 6.0 (Molecular Devices). Spot recognition and flags were checked for each spot and manually adjusted when appropriate. Because the reference sample was co-hybridized with all samples, hybridization quality could be readily observed and missing spots could be easily detected. Missing spots were consistent between arrays, and no more than 3 of 12 replicate spots per gene were missing. Therefore, only 9 replicate spots were used for analysis even though the majority of probes were spotted 12 times per array. For genes that had 12 replicate spots that were of adequate quality on all arrays, the 3 spots printed last were not analyzed. In addition, the quality of hybridization was assessed by ensuring that the positive control spots produced signal when appropriate (for example, strong signal being detected for the ACTA and ACTB probes). Replicate hybridizations were also plotted for comparison and correlation coefficients calculated to ensure consistent data between experiments.

2.11. Data normalization and analysis

Microarray data and processing has been submitted to the NCBI Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo), entry GSE33544. This reported data the Minimum Information About a Microarray Experiment (MIAME) standard. All analysis was performed using MAANOVA 2.0 software (Xiangqin Cui, Hao Wu, Gary A. Churchill, The Jackson Lab) in Matlab (Mathworks). The median spot log2 values were used for the analysis and prior to normalization, intensity-based Lowess adjustment of all data was made. Normalization was performed using mixed effects models which included variables for array (A), gene (G), sample (S), and health status (H). Two different models were used for normalization. The first model did not separate individuals into groups (healthy versus SLE) and was used to estimate expression levels in the repertoires; the following equation was used for this model: log2(yikg) = μ + Ai + Sk + Gg + (AG)ig + (SG)kg + ε ikg. The second model included a term to account for health status [(HG)hg] and used the following equation: log2(yhikg) = μ + Hh + Ai + Sk + Gg + (AG)ig + (SG)kg + (HG)hg + ε ikg. For determination of statistical significance, the (SG)kg terms from the first model (the null model where health status was not included) was compared with the (SG)kg term from the second model (in the second model health status was included in normalization). Differential gene expression was determined using the MAANOVA package, and results from two different F-tests are presented. κ-sorted and λ-sorted samples were analyzed independently for Vκ and Vλ expression, respectively. To evaluate inclusion of VL gene expression, κ-sorted and λ-sorted samples were analyzed together for expression of all VL genes.

To remove the probe effect and make intra-array comparisons, normalized values for the reference sample (containing equal concentrations of all VL gene targets) were subtracted from each sample tested. Thus, probes which hybridized well to the reference sample were assumed to also hybridize well to the experimental samples. These reference-adjusted values were then fit to a standard curve generated by spiking in known concentrations of VL gene targets. Estimates for expression levels were restricted to take on values between 0 and 30%. Expression level estimates were not further manipulated, and the sums of expression level estimates for a sample were allowed to be more or less than 100%. Comparison of biological replicates was performed to evaluate reproducibility of the method (see Appendix A.5).

Unless otherwise noted, two-tailed t-tests were performed to generate p-values, and 0.05 was used as the p-value cut-off.

2.12. NCBI database search for VL genes identified as absent

The germline VL nucleotide sequences for those VL genes identified as not being expressed in either SLE or healthy groups was searched against all entries in the NCBI database (http://www.ncbi.nlm.nih.gov/igblast/, search performed on 11/10/2010). Default parameters were used for alignment and scoring. All sequences returned in the search were manually evaluated and excluded if the entry was more closely related to another germline VL gene or if the entry was from a non-RNA source (i.e. genomic library). The number of sequences meeting the criteria is reported in the tables.

2.13. Spectratyping

Minor modifications were made to the recommended protocol for H-chain spectratype analysis (IGH Gene Clonality Assay, Invivoscribe Technologies). Instead of using genomic DNA for each PCR reaction, 40ng of amplified cDNA was used as the template. Three PCR reactions (Tube A, Tube B and Tube C) were performed on five healthy individuals (both κ- and λ-sorted B cells). Four SLE κ-sorted samples were tested, and five SLE λ-sorted samples were tested. PCR products were run on a 3130XL DNA Analyzer (Applied Biosystems). For quantitative comparison, Tube C data were used. Peak heights for all lengths within the appropriate H-chain size range were measured and each length was normalized according to the highest peak in the sample (thus, the peak had a value of 1.0). For comparison of restriction between the healthy and SLE groups, the normalized polyclonal control peak heights were subtracted from the samples at the corresponding lengths. The standard deviation in these values was determined for each sample and compared using an unpaired t-test.

3. Results

3.1. Peripheral blood B cell phenotypes in healthy individuals and SLE patients

B cells from 10 SLE patients and 10 healthy donors were phenotyped using flow cytometry. Samples were co-stained using reagents to CD27 (memory B cell marker), CD38 (germinal center/transitional cell marker) and the pan-B cell marker, CD20 (Figure 1A). The frequency of CD20+CD27+ cells was the same in healthy and SLE groups (p-value > 0.05), while the percent of CD20+CD38+ B cells was higher in samples from SLE patients (p-value < 0.05). Using a separate stain, surface CD20, light chain (κ and λ) and CD138 (plasma cell marker) levels were determined (Figure 1B). The ratio, κ:λ, of B cells was higher in SLE with a median of 1.9 compared with healthy B cells with a median of 1.4 (p-value < 0.05). Plasma cells were rare in the lymphocyte gated population, comprising less than 0.001% of cells in any of the blood samples collected. Because plasma cells contain large quantities of H- and L-chain RNA that would overshadow the true frequency of a VL gene in the repertoire, this population was excluded from the sorting gates (Figure 1A). Cells were then sorted into κ samples (CD20+CD138−κ+λ−) and λ samples (CD20+CD138−κ−λ+) for microarray analysis with κ and λ B cells tested separately. Prior to microarray hybridization, cDNA from each sample was also tested for plasma cell contamination using a syndecan-1 (CD138) PCR; contamination was not identified in any samples.

Figure 1.

Figure 1

Open in a new tab

FACS plot and phenotype of healthy and SLE B cells. (A) Peripheral blood cells were sorted into κ samples (CD20+CD138−κ+λ−) and λ samples (CD20+CD138−κ−λ+) based on the gating strategy shown at the top from a representative sample (Healthy01). A CD27 versus CD38 plot is also shown. (B) Quantification of FACS data for κ to λ ratios, CD38+ and CD27+ B cells. Both the ratio of κ-to-λ and the percentage of CD38+ were elevated in SLE (p-value <0.05), while there was no difference in the percentage of CD27+ B cells.

One thousand B cells were sorted for each sample. This number was chosen because it is large enough to include rarely expressed VL genes. According to the binomial distribution, VL genes comprising more than 0.45% of the repertoire have a 95% chance of being included in the sample. The sorted samples from healthy and SLE-affected individuals were used to generate cDNA for hybridization on identical light chain microarrays. A reference sample that included equal concentrations of the VL gene targets was co-hybridized with the experimental samples. The reference sample was also used in a series of experiments to evaluate for cross-reactivity and quantify expression levels. These experiments did not show signs of cross-reactivity, demonstrated high sensitivity and accurately quantified expression levels (see Appendix sections A.2, A.3 and A.4).

3.2. Biased VL gene expression in healthy repertoires

We evaluated B cell repertoires with the aim of addressing whether VL genes are expressed at different levels. If all VL genes were equally expressed, each Vκ gene would comprise 2.4% (1/42) of the total κ signal and each Vλ would comprise 2.8% (1/36) of the total λ signal. Most VL genes in the healthy samples were expressed at levels higher or lower than these values For example, 5 Vκ genes in the Healthy01 κ sample were expressed at more than 10% the repertoire (Figure 2, left, bars > 0.1). Other Vκ genes in different samples were similarly expressed at over 10% of the total repertoire. Vλ genes were also highly expressed (Figure 2, right; Appendix Table A.2 and Table A.3 contains complete expression level estimates). The probability of observing a single VL gene at 10% of the repertoire by chance assuming equal expression of VL genes is extremely small (less than 2×10−27). Some VL genes were highly expressed in all healthy individuals (Vκ genes A3/A19, A20 and O8/O18 and the Vλ genes 1–4, 1–7, 1–2 and 1–3) as indicated by low rank-order numbers (Appendix Figure A.5). Other VL genes were expressed at different levels between individuals with variable rank-order values.

Figure 2.

Figure 2

Open in a new tab

Distribution of VL gene expression levels in healthy individuals. The range of expression level estimates was split into 1% intervals and the number of VL genes with expression within each 1% interval was counted. The x-axis indicates the interval (i.e. 0.10 indicates expression at levels between 10% and 11%) and the y-axis indicates that number of VL genes in a given interval. The κ repertoires are shown on the left and the λ repertoires are shown on the right.

3.3. Absence of light chain V genes from the repertoires of healthy individuals

Some VL genes were not detected in any of the repertoires from healthy individuals although they appear to be functional. These included three Vκ genes (A10, L14 and O11a) and seven Vλ genes (1–9, 3–2, 3–3, 4–2, 4–3, 5–1 and 5–2). The finding that certain VL genes were absent from repertoires was consistent with data from the NCBI database. A search of all 93,626 sequences in the NCBI database for evidence of expression of these genes revealed that three had no entries, six had between one and sixteen entries and one had more than sixteen entries (Table 1).

Table 1.

List of Vl genes not expressed in any healthy repertoires. The number of entries in the NCBI database is also indicated (over 93,000 total entries).

Vκ genes Vλ genes
Name Evidence of expression in NCBI database Name Evidence of expression in NCBI database
A10 1 1–9 0
L14 3 3–2 >100
O11a 0 3–3 2
4–2 6
4–3 0
5–1 11
5–2 16
Open in a new tab

The VL genes absent from repertoires varied between individuals in both identity and number. For Vκ genes this ranged in number from 6 to 18, and the range of Vλ genes not detected was 9 to 17. As described above, VL genes present at greater than 0.45% of the repertoire are likely to have been included in our sample (95% confidence level). Therefore, these VL genes are either not expressed or expressed at levels so low they are undetectable in the repertoire.

3.4. Repertoires from SLE patients

Light chain repertoires from SLE patients shared many of the same features as repertoires from healthy individuals but also had some differences. SLE patient repertoires had high expression of VL genes in both κ and λ samples. Some of the same VL genes that were highly expressed in all healthy individuals were also highly expressed in all SLE patients, including A3/A19, A20 and 1–4 (Appendix Figure A.6 and Figure A.7). A few other genes that had inconsistent expression in healthy repertoires were highly expressed in all SLE patient repertoires (e.g. 1–19). The expression of L6, L12, 1–20 and 2–8 was variable in SLE patient repertoires but consistently high in repertoires from healthy individuals (Appendix Figure A.6 and Figure A.7).

While 10 VL genes were not expressed in any repertoires from healthy individuals, 7 VL genes were absent from all repertoires of SLE patients (Table 2). Four of the VL genes (A14, 1–9, 4–3 and 5–2) missing from all SLE samples were also not detected in any of the healthy samples. None of these showed evidence of expression in the NCBI database. Other VL genes were expressed exclusively in healthy individuals or SLE patients. Expression levels indicated that when present, these VL genes made significant contributions to the repertoire. For example, when O11a, 3–3, 4–2 and 5–1 were expressed in SLE repertoires, their expression levels were 4.3%, 4.3%, 4.8% and 6.7%, respectively. These expression levels were much higher than could be attributed to sampling error (Appendix Figure A.5). In summary, some genes that were not expressed in the healthy samples were expressed in some repertoires from SLE patients, and some VL genes that were not expressed in repertoires from SLE patients were expressed some healthy samples.

Table 2.

L-chain V genes not detected in any of the SLE repertoires.

Vκ genes Vλ genes
Name Number of sequences in NCBI database Name Number of sequences in NCBI database
A14 0 1–9 0
L1 8 4–3 0
4–6 0
5–2 16
Open in a new tab

Differences in VL gene expression between groups were detected using two different F-tests, both of which are valid in determining statistical significance alone but do so with different methods. F1 values were calculated using gene-specific standard deviations while F3 values were calculated using the standard deviation of all VL genes (Woo et al., 2005). A total of 11 genes (5 Vκs and 6 Vλs) were identified as differentially expressed, eight according to the F1 test and seven according to the F3 test (p-value < 0.05, Table 3). Four VL genes were identified as differentially expressed by both F- tests. Genes expressed at higher levels in repertoires from healthy individuals than SLE patient repertoires include the Vκs B3 and L1 and Vλs 1–2, 1–3 and 4–1. Genes expressed at higher levels in SLE include the Vκs A3/A19, A27 and O11a and the Vλs 2–8, 3–2 and 5–1. As described above, O11a, 3–2 and 5–1 were not expressed in any healthy repertoires and had rare occurrences in the NCBI database. However, in some SLE repertoires these VL genes were expressed. There was intra-group variability in expression levels with overlap in the expression levels of healthy and SLE groups for most of these VL genes.

Table 3.

Differences in Vl gene expression between healthy and SLE groups.

Gene name κ or λ Average expression in healthy Average expression in SLE F1 p-value F3 p-value
Higher expression in healthy repertoires
B3 κ 4.0% 1.9% 0.067 0.002
L1 κ 0.7% Not detected 0.047 0.113
1–2 λ 14.5% 9.6% 0.015 0.089
1–3 λ 14.3% 9.2% 0.041 0.082
4–1 λ 5.9% 2.4% 0.049 0.025
Higher expression in SLE repertoires
A3/A19 κ 23.4% 26.9% 0.067 0.002
A27 κ 8.4% 12.8% 0.083 0.029
O11a κ Not detected 0.5% 0.022 0.041
2–8 λ 2.0% 8.3% 0.025 0.001
3–2 λ Not detected <0.3% 0.024 0.094
5–1 λ Not detected 1.1% 0.008 0.001
Open in a new tab

3.5. Heavy chain restriction in the B cell repertoire

Clonal selection, both during normal development and resulting from antigen encounter, can play a role in shaping the B cell repertoire and could contribute to the biased VL gene expression observed in the microarray data. Because clonal selection restricts both a H-chain and a L-chain repertoire by increasing the frequency of a single clone, H-chain CDR3-length restrictions can be used as a marker for clonal selection. Spectratype analysis of H-chain CDR3 lengths was performed using the same cDNA that was used for microarray hybridization. Several spectratypes, such as the κ-sorted Healthy01, showed minimal signs of H-chain CDR3 length restriction as indicated by normally-distributed peak heights (Figure 3A). However, the majority of the 16 spectratype patterns tested showed evidence of restricted CDR3 lengths. Restriction between samples was quantified and compared with a polyclonal control sample. SLE-κ samples had greater H-chain restriction than healthy-κ samples, while there was no significant difference between the SLE-λ and the healthy-λ (Figure 3B). Thus, there is evidence for clonal selection in repertoires from both healthy individuals and SLE patients with a difference in the amount of selection in κ B cells.

Figure 3.

Figure 3

Open in a new tab

Heavy chain spectratype analysis. (A) Two different H-chain spectratype patterns, Healtlhy01-κ (left) and Healthy02-κ (right). (B) The polyclonal control peak heights were subtracted from the sample peak heights at the corresponding lengths. The standard deviation of these differences within a sample was then determined as a measure of restriction (standard deviation is plotted). The only significant difference was between the Healthy-κ and SLE-κ groups (p < 0.05).

3.6. Transcription of Vκ genes in λ B cells

A feature of our method is that it allows examination of repertoires at the mRNA level. Hence, we are able to analyze expression of VL genes that are not necessarily functional at the protein level. Such transcripts include aberrantly rearranged and unrearranged VL genes (Delpy et al., 2004; Staudt and Lenardo, 1991). B cells were sorted on the basis of κ and λ isotypes (CD19+κ+λ-CD138- and CD19+κ-λ+CD138-), and the differences between these two populations are of interest because rearrangement of IGL genes and expression of the λ L-chain repertoire occurs after IGK rearrangement (Belessi et al., 2005; Brauninger et al., 2001; Hieter et al., 1981; Klein et al., 2005; Korsmeyer et al., 1981; Korsmeyer et al., 1982; van der Burg et al., 2001). This temporal difference could result in different selective forces acting on B cells bearing different isotypes. Vλ gene expression was not detected above background in κ-sorted samples (Figure 4 shows one example). However, Vκ expression was detected in the λ-sorted samples, albeit significantly lower than Vλ expression (p-value less than 0.05). Thus, there is little if any expression of Vλ genes in κ+λ- B cells, but there is appreciable expression of Vκ genes in κ-λ+ B cells.

Figure 4.

Figure 4

Open in a new tab

Transcription of Vκ genes in λ-sorted B cells. B cells sorted on the basis of surface κ or λ protein were compared for expression levels of VL genes. The VL genes are grouped according to κ or λ from each sample. The two samples are from Healthy01. One Vκ (A3/A19) in the λ B cell population is expressed at a level that is higher than any Vλ.

4.0. Discussion

4.1. Selection on L-chains shapes the B cell repertoire

One-third of VL genes were not expressed in a given repertoire. Lack of VL gene expression can be due to several factors. Cis-elements can limit recombination, an example of this is the octamer motif bound by OcaB which decreases recombination of some Vκ genes when altered or deleted (Casellas et al., 2002; Nadel et al., 1998). VL genes consistently expressed at low levels or not expressed may be influenced by such cis-elements that limit recombination. Differences in the recombination signal sequences also affect VL gene recombination (Hesse et al., 1989). Although rare, nonsense codons at the end of a V region limit functional rearrangements for these VL genes. The mouse VλX and two human Vλ genes, 5–1 and 5–2, are the only VL genes with nonsense codons at the end of the V region (Kawasaki et al., 1997; Sanchez et al., 1990). Since the microarray was designed to detect only functional VL genes, the possibility that the VL genes not expressed are pseudo-genes is excluded. Negative selection on L-chains can also limit VL gene expression. While all of these mechanisms may decrease expression levels, we think selection on L-chains plays the largest role in shaping the repertoire.

B cells with autoreactive receptors undergo negative selection by receptor editing, and L-chains with intrinsic self-reactivity are negatively selected by this mechanism (Gay et al., 1993; Halverson et al., 2004; Radic et al., 1993; Tiegs et al., 1993). The mouse λX L-chain has been shown to bind the self-antigen myelin basic protein alone or in combination with VHs (Doyle et al., 2006; Galin et al., 1996). This may account for the low expression of λX observed in wild-type mice (Li et al., 2004). Other L-chains with intrinsic autoreactivity are also expected to have low expression in the repertoire. The human VL genes 5–2 and 5–1 are similar in protein sequence to the mouse VλX, and neither 5–2 nor 5–1 was detected in any of the healthy repertoires. We think their absence is not a coincidence, but rather due to negative selection on 5–1 and 5–2 because of intrinsic autoreactivity. While some VL genes were not expressed in any repertoire, there was variability in which VL genes were not expressed between individuals. This variability may reflect inter-individual differences in negative selection.

We also have evidence for positive selection on L-chains based on high expression of VL genes. A set of VL genes were uniformly highly expressed without H-chain restriction, and this is consistent with selection of particular L-chains rather than clonal selection. It is not obvious what selective forces might govern expression of these genes, but they may be L-chains that have intrinsic specificity for self-antigens. Such specificities are found at high frequencies in the repertoire and are thought to play an important housekeeping role by clearing immune complexes (Radoux et al., 1986; Shlomchik et al., 1987; Shlomchik et al., 1986). These consistently highly expressed VL genes may perform a similar function. Among the VL genes with high expression in all repertoires was Vκ A27. This is in agreement with other studies that have shown a high frequency of A27 in the repertoire (de Wildt et al., 1999; Foster et al., 1997; Jacobi et al., 2002). However, while A27 was the most frequent Vκ in these studies, we found other VL genes had consistently higher expression than A27. This discrepancy is likely due to differences in approaches and methods. While these other studies used a limited number of B cells from few individuals, our method sampled a large number of B cells from many individuals. We also avoided potential bias introduced by culture and PCR efficiency. For these reasons, our analysis provides a more comprehensive understanding of positive selection on L-chains.

Positive selection in the form of clonal expansion also contributes to the repertoire. Our repertoire analysis was performed on populations of B cells expressing κ or λ L-chains and included sub-populations of memory and plasmablast B cells. We detected concomitant H-chain restriction by spectratyping and high expression of VL genes. This pattern is consistent with clonal expansion.

Repertoire selection operates differently in κ and λ B cells, and our data show that the λ repertoire was more restricted than the κ repertoire. This restriction can be attributed to the order and timing of recombination. Initial recombination occurs at IGK, and only if recombination of IGK fails will IGL undergo recombination (Belessi et al., 2005; van der Burg et al., 2001). While B cells have the potential to undergo many recombination events, the actual number of attempts at recombination is limited and exhaustive recombination does not occur (Louzoun et al., 2002; Mehr et al., 1999). Because of these features of B cell development, receptor editing occurs predominantly at IGK and may be responsible for the less restricted repertoire observed in κ B cells. Differences in the organization of the IGK and IGL loci may also contribute to increased restriction in λ B cells. The IGK locus can be inactivated by recombination to IGKDE (the deletion element) while the IGL locus lacks this feature (Delpy et al., 2004; Klobeck and Zachau, 1986; Nemazee and Weigert, 2000; Siminovitch et al., 1985). Thus, recombination at IGL may result in a receptor that cannot be deleted. Receptor editing is consequentially different between κ and λ B cells as a result of these features, and the repertoires we analyzed here reflect this difference.

4.2. Transcription of Vk genes in lambda B cells

We detected Vκ gene expression in λ B cells but did not detect Vλ expression in κ B cells. Three types of VL gene transcripts have been reported: rearranged VL-JL mRNA which is in-frame and produces a functional L-chain; rearranged VL-JL mRNA which has a premature nonsense codon – either because recombination introduces the nonsense codon or results in the JL-CL being out-of-frame; and sterile transcription of unrearranged VL genes (Delpy et al., 2004; Staudt and Lenardo, 1991). The absence of a surface κ L-chain excludes the possibility that the observed Vκ expression in Vλ-sorted cells is a functional L-chain protein. The nonsense-mediated decay pathway is efficient in degrading transcripts with a nonsense codon at the V-J junction. Over 90% of out-of-frame κ transcripts are degraded by this mechanism (Chemin et al.). The impact of these non-functional messages on the expression level is likely to be negligible for several reasons: 1) in both κ and λ samples, a large number of VL genes were not expressed, implying that aberrant rearrangement of VL genes is relatively infrequent; 2) as compared with λ message levels, κ message levels in λ B cells was low; and 3) nonsense-mediated decay is efficient.

The Vκ gene A3/A19 is unique as it is highly expressed in both κ and λ B cells. This pattern of expression is consistent with sterile transcription of A3/A19. Sterile transcription is a phenomenon associated with κ locus activation and is a surrogate for recombination. This observation suggests that the κ locus continues to prepare for recombination despite a B cell having a functional λ L-chain. If the κ-locus is deleted before recombination begins at λ (as is typical of λ-expressing cells) (Belessi et al., 2005; Brauninger et al., 2001; Hieter et al., 1981; Klein et al., 2005; Korsmeyer et al., 1981; Korsmeyer et al., 1982; van der Burg et al., 2001), then expression of any Vκ would be unexpected. However, deletion at the κ locus is primarily Cκ deletion. Thus, some Vκ genes are likely to remain in λ B cells. It should also be remembered that this analysis was performed on a population of B cells, and it would be expected that at least some λ B cells retain the A3/A19 Vκ gene. Rearranged and functional L-chains using A3/A19 have been reported (Bensimon et al., 1994; Dorner et al., 1998; Girschick and Lipsky, 2002; Yurasov et al., 2006; Yurasov et al., 2005). We expect that some of the measured A3/A19 is functional, although the proportion of functional transcripts is unknown.

4.3. Proximity of VL to JL

We asked whether VL gene position affects expression. Vκ genes are organized in two clusters, and Vλ are organized in three clusters in humans (Frippiat et al., 1995; Kawasaki et al., 2001; Kawasaki et al., 1995). We analyzed VL gene expression according to position and found VL genes from each cluster were highly expressed. Each cluster also had VL genes that were not expressed (Figure 5). The median cluster expression of VL genes was highest for the J-proximal clusters (Figure 5). There was no difference between healthy and SLE groups in expression levels according to clusters. J-proximal VL genes have been reported to be expressed at higher levels than distal VL genes (Girschick and Lipsky, 2002; Lee et al., 2004; Richl et al., 2008). However, expression was not dictated by position as high and low expression of VL genes from all clusters was observed. Therefore positive and negative selection on L-chains appears to be more important in shaping the repertoire than VL gene proximity to J-genes.

Figure 5.

Figure 5

Open in a new tab

Expression levels by cluster and position. Average expression of VL genes per cluster, (A) Vκ clusters and (B) Vλ clusters, was calculated for each individual and is shown for both healthy and SLE groups. Red line indicates the median for each group. For the κ clusters average expression was different between proximal and distal clusters for both healthy and SLE groups (p-value less than 0.01). There was no difference between groups. The average expression was different for the λ clusters for both healthy and SLE groups (p-value <0.01, tested using one-way ANOVA), but there was no difference between groups. Expression levels for each VL gene are shown according to position for Healthy01 κ genes (C) and λ genes (D).

4.4. Repertoire selection in SLE

Previous work in our lab focused on regulation of auto-antibodies in mice and identified several VL genes important for tolerance in the 56R H-chain transgenic model (Doyle et al., 2006; Li et al., 2001; Li et al., 2004; Sekiguchi et al., 2006). These studies demonstrated that editor VL genes are expressed differently in lupus-susceptible and lupus-resistance mice. While our present analysis evaluated all VL genes, we were particularly interested in the expression of VL genes that are homologous to those identified as editors in the mouse B cell repertoire. Vλ 5–1, the human counterpart to mouse editor VλX, was identified in our screen of the SLE group as over-expressed. These two Vλs are more similar to each other than to any other human or mouse VL genes (Solomon et al., 1997). The consequence of VλX expression is context-dependent and somewhat paradoxical because while it vetoes 56R H-chain's DNA reactivity, it also forms poly-reactive auto-antibodies with this and other H-chains (Doyle et al., 2006). Assuming that 5–1 has similar potential to generate poly-reactive antibodies as does VλX, this feature of 5–1 expression is likely to be disadvantageous in establishing tolerance. Expression of 5–1 in SLE repertoires may give rise to incompletely edited or poly-reactive antibodies that are common in SLE (Yurasov et al., 2005).

The VL gene repertoires in SLE patients were similar to repertoires from healthy individuals by some measures, but subtle differences exist. Repertoires in SLE patients show signs of positive and negative selection on L-chains as indicated by high VL gene expression and absent expression of other VL genes, respectively. The number of VL genes undergoing positive and negative selection was similar in SLE patients and healthy individuals; no global defects in positive or negative selection were seen in the SLE group. We observed considerable inter-individual diversity in VL gene expression in the SLE group that was also present in the healthy group. However, there are notable differences in the VL gene repertoires between the healthy and SLE groups. There was greater variation in expression levels of Vκ genes in SLE patients than healthy individuals. VL genes that were not expressed in any healthy individual were expressed in some SLE patient repertoires. This is likely due to differences in negative selection between healthy individuals and SLE patients. A set of VL genes were expressed at different levels between the healthy and SLE groups. We acknowledge that the differences were modest but conclude that VL genes are selected differently in those without autoimmune disease than in individuals with autoimmune disease. Small differences in selection may not lead to an overwhelming repertoire shift but may lead to the appearance of disease-causing B cells and antibodies.

We have been able to study VL gene expression in a large number of individuals and different populations of B cells because of a novel approach to repertoire analysis. The results provide insight into selection on L-chains and identified differences in diseased repertoires. Using the VL gene microarray to study subsets of SLE patients and additional B cell populations will continue to improve our understanding of selection and the role of L-chains in autoimmunity.

Supplementary Material

01

Highlights

  • We developed a microarray to study B cell receptor light chain V gene repertoires

  • Negative and positive selection acts on light chains to shape the B cell repertoire

  • Κ and λ B cells are selected differently

  • Certain light chain V genes are expressed at different levels in SLE patients

Acknowledgments

Dr Tammy Utset coordinated the collection of samples from SLE patients. We would like to thank Cindy Benedict-Alderfer for her assistance in preparing this manuscript. Funding for this project was provided by National Institutes of Health grant GM020964.

Abbreviations

L-chain

Light chain

H-chain

Heavy chain

VL gene

Light chain V gene

SLE

Systemic lupus erythematosus

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Belessi C, Stamatopoulos K, Hadzidimitriou A, Hatzi K, Smilevska T, Stavroyianni N, Marantidou F, Paterakis G, Fassas A, Anagnostopoulos A, Laoutaris N. Analysis of expressed and non-expressed IGK locus rearrangements in chronic lymphocytic leukemia. Mol Med. 2005;11:52–8. doi: 10.2119/2005-00044.Belessi. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bensimon C, Chastagner P, Zouali M. Human lupus anti-DNA autoantibodies undergo essentially primary V kappa gene rearrangements. EMBO J. 1994;13:2951–62. doi: 10.1002/j.1460-2075.1994.tb06593.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bentley DL. Most kappa immunoglobulin mRNA in human lymphocytes is homologous to a small family of germ-line V genes. Nature. 1984;307:77–80. doi: 10.1038/307077a0. [DOI] [PubMed] [Google Scholar]
  4. Brauninger A, Goossens T, Rajewsky K, Kuppers R. Regulation of immunoglobulin light chain gene rearrangements during early B cell development in the human. Eur J Immunol. 2001;31:3631–7. doi: 10.1002/1521-4141(200112)31:12&#60;3631::aid-immu3631&#62;3.0.co;2-l. [DOI] [PubMed] [Google Scholar]
  5. Casellas R, Jankovic M, Meyer G, Gazumyan A, Luo Y, Roeder R, Nussenzweig M. OcaB is required for normal transcription and V(D)J recombination of a subset of immunoglobulin kappa genes. Cell. 2002;110:575–85. doi: 10.1016/s0092-8674(02)00911-x. [DOI] [PubMed] [Google Scholar]
  6. Chemin G, Tinguely A, Sirac C, Lechouane F, Duchez S, Cogne M, Delpy L. Multiple RNA surveillance mechanisms cooperate to reduce the amount of nonfunctional Ig kappa transcripts. J Immunol. 184:5009–17. doi: 10.4049/jimmunol.0902949. [DOI] [PubMed] [Google Scholar]
  7. de Wildt RM, Hoet RM, van Venrooij WJ, Tomlinson IM, Winter G. Analysis of heavy and light chain pairings indicates that receptor editing shapes the human antibody repertoire. J Mol Biol. 1999;285:895–901. doi: 10.1006/jmbi.1998.2396. [DOI] [PubMed] [Google Scholar]
  8. de Wildt RM, Steenbakkers PG, Pennings AH, van den Hoogen FH, van Venrooij WJ, Hoet RM. A new method for the analysis and production of monoclonal antibody fragments originating from single human B cells. J Immunol Methods. 1997;207:61–7. doi: 10.1016/s0022-1759(97)00108-7. [DOI] [PubMed] [Google Scholar]
  9. Delpy L, Sirac C, Magnoux E, Duchez S, Cogne M. RNA surveillance down-regulates expression of nonfunctional kappa alleles and detects premature termination within the last kappa exon. Proc Natl Acad Sci U S A. 2004;101:7375–80. doi: 10.1073/pnas.0305586101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Dorner T, Farner NL, Lipsky PE. Ig lambda and heavy chain gene usage in early untreated systemic lupus erythematosus suggests intensive B cell stimulation. J Immunol. 1999;163:1027–36. [PubMed] [Google Scholar]
  11. Dorner T, Foster SJ, Farner NL, Lipsky PE. Immunoglobulin kappa chain receptor editing in systemic lupus erythematosus. J Clin Invest. 1998;102:688–94. doi: 10.1172/JCI3113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Dorner T, Hansen A, Jacobi A, Lipsky PE. Immunglobulin repertoire analysis provides new insights into the immunopathogenesis of Sjogren's syndrome. Autoimmun Rev. 2002;1:119–24. doi: 10.1016/s1568-9972(02)00029-0. [DOI] [PubMed] [Google Scholar]
  13. Dorner T, Kaschner S, Hansen A, Pruss A, Lipsky PE. Perturbations in the impact of mutational activity on Vlambda genes in systemic lupus erythematosus. Arthritis Res. 2001;3:368–74. doi: 10.1186/ar329. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Doyle CM, Han J, Weigert MG, Prak ET. Consequences of receptor editing at the lambda locus: multireactivity and light chain secretion. Proc Natl Acad Sci U S A. 2006;103:11264–9. doi: 10.1073/pnas.0604053103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Farner NL, Dorner T, Lipsky PE. Molecular mechanisms and selection influence the generation of the human V lambda J lambda repertoire. J Immunol. 1999;162:2137–45. [PubMed] [Google Scholar]
  16. Foster SJ, Brezinschek HP, Brezinschek RI, Lipsky PE. Molecular mechanisms and selective influences that shape the kappa gene repertoire of IgM+ B cells. J Clin Invest. 1997;99:1614–27. doi: 10.1172/JCI119324. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Frippiat JP, Williams SC, Tomlinson IM, Cook GP, Cherif D, Le Paslier D, Collins JE, Dunham I, Winter G, Lefranc MP. Organization of the human immunoglobulin lambda light-chain locus on chromosome 22q11.2. Hum Mol Genet. 1995;4:983–91. doi: 10.1093/hmg/4.6.983. [DOI] [PubMed] [Google Scholar]
  18. Galin FS, Maier CC, Zhou SR, Whitaker JN, Blalock JE. Murine V lambda x and V lambda x-containing antibodies bind human myelin basic protein. J Clin Invest. 1996;97:486–92. doi: 10.1172/JCI118439. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Gay D, Saunders T, Camper S, Weigert M. Receptor editing: an approach by autoreactive B cells to escape tolerance. J Exp Med. 1993;177:999–1008. doi: 10.1084/jem.177.4.999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Girschick HJ, Lipsky PE. The kappa gene repertoire of human neonatal B cells. Mol Immunol. 2002;38:1113–27. doi: 10.1016/s0161-5890(02)00039-1. [DOI] [PubMed] [Google Scholar]
  21. Halverson R, Torres RM, Pelanda R. Receptor editing is the main mechanism of B cell tolerance toward membrane antigens. Nat Immunol. 2004;5:645–50. doi: 10.1038/ni1076. [DOI] [PubMed] [Google Scholar]
  22. Heimbacher C, Hansen A, Pruss A, Jacobi A, Reiter K, Lipsky PE, Dorner T. Immunoglobulin Vkappa light chain gene analysis in patients with Sjogren's syndrome. Arthritis Rheum. 2001;44:626–37. doi: 10.1002/1529-0131(200103)44:3<626::AID-ANR111>3.0.CO;2-T. [DOI] [PubMed] [Google Scholar]
  23. Hesse JE, Lieber MR, Mizuuchi K, Gellert M. V(D)J recombination: a functional definition of the joining signals. Genes Dev. 1989;3:1053–61. doi: 10.1101/gad.3.7.1053. [DOI] [PubMed] [Google Scholar]
  24. Hieter PA, Korsmeyer SJ, Waldmann TA, Leder P. Human immunoglobulin kappa light-chain genes are deleted or rearranged in lambda-producing B cells. Nature. 1981;290:368–72. doi: 10.1038/290368a0. [DOI] [PubMed] [Google Scholar]
  25. Jacobi AM, Hansen A, Kaufmann O, Pruss A, Burmester GR, Lipsky PE, Dorner T. Analysis of immunoglobulin light chain rearrangements in the salivary gland and blood of a patient with Sjogren's syndrome. Arthritis Res. 2002;4:R4. doi: 10.1186/ar423. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kaschner S, Hansen A, Jacobi A, Reiter K, Monson NL, Odendahl M, Burmester GR, Lipsky PE, Dorner T. Immunoglobulin Vlambda light chain gene usage in patients with Sjogren's syndrome. Arthritis Rheum. 2001;44:2620–32. doi: 10.1002/1529-0131(200111)44:11<2620::aid-art442>3.0.co;2-m. [DOI] [PubMed] [Google Scholar]
  27. Kawasaki K, Minoshima S, Nakato E, Shibuya K, Shintani A, Asakawa S, Sasaki T, Klobeck HG, Combriato G, Zachau HG, Shimizu N. Evolutionary dynamics of the human immunoglobulin kappa locus and the germline repertoire of the Vkappa genes. Eur J Immunol. 2001;31:1017–28. [PubMed] [Google Scholar]
  28. Kawasaki K, Minoshima S, Nakato E, Shibuya K, Shintani A, Schmeits JL, Wang J, Shimizu N. One-megabase sequence analysis of the human immunoglobulin lambda gene locus. Genome Res. 1997;7:250–61. doi: 10.1101/gr.7.3.250. [DOI] [PubMed] [Google Scholar]
  29. Kawasaki K, Minoshima S, Schooler K, Kudoh J, Asakawa S, de Jong PJ, Shimizu N. The organization of the human immunoglobulin lambda gene locus. Genome Res. 1995;5:125–35. doi: 10.1101/gr.5.2.125. [DOI] [PubMed] [Google Scholar]
  30. Klein F, Feldhahn N, Mooster JL, Sprangers M, Hofmann WK, Wernet P, Wartenberg M, Muschen M. Tracing the pre-B to immature B cell transition in human leukemia cells reveals a coordinated sequence of primary and secondary IGK gene rearrangement, IGK deletion, and IGL gene rearrangement. J Immunol. 2005;174:367–75. doi: 10.4049/jimmunol.174.1.367. [DOI] [PubMed] [Google Scholar]
  31. Klobeck HG, Solomon A, Zachau HG. Contribution of human V kappa II germ-line genes to light-chain diversity. Nature. 1984;309:73–6. doi: 10.1038/309073a0. [DOI] [PubMed] [Google Scholar]
  32. Klobeck HG, Zachau HG. The human CK gene segment and the kappa deleting element are closely linked. Nucleic Acids Res. 1986;14:4591–603. doi: 10.1093/nar/14.11.4591. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Korsmeyer SJ, Hieter PA, Ravetch JV, Poplack DG, Waldmann TA, Leder P. Developmental hierarchy of immunoglobulin gene rearrangements in human leukemic pre-B-cells. Proc Natl Acad Sci U S A. 1981;78:7096–100. doi: 10.1073/pnas.78.11.7096. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Korsmeyer SJ, Hieter PA, Sharrow SO, Goldman CK, Leder P, Waldmann TA. Normal human B cells display ordered light chain gene rearrangements and deletions. J Exp Med. 1982;156:975–85. doi: 10.1084/jem.156.4.975. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Lee J, Cho YJ, Lipsky PE. The V(lambda)-J(lambda) repertoire of patients with systemic lupus erythematosus manifests characteristics of the natural antibody repertoire. Arthritis Rheum. 2004;50:2604–14. doi: 10.1002/art.20439. [DOI] [PubMed] [Google Scholar]
  36. Lee J, Monson NL, Lipsky PE. The V lambda J lambda repertoire in human fetal spleen: evidence for positive selection and extensive receptor editing. J Immunol. 2000;165:6322–33. doi: 10.4049/jimmunol.165.11.6322. [DOI] [PubMed] [Google Scholar]
  37. Li H, Jiang Y, Prak EL, Radic M, Weigert M. Editors and editing of anti-DNA receptors. Immunity. 2001;15:947–57. doi: 10.1016/s1074-7613(01)00251-5. [DOI] [PubMed] [Google Scholar]
  38. Li Y, Louzoun Y, Weigert M. Editing anti-DNA B cells by Vlambdax. J Exp Med. 2004;199:337–46. doi: 10.1084/jem.20031712. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Louzoun Y, Friedman T, Luning Prak E, Litwin S, Weigert M. Analysis of B cell receptor production and rearrangement. Part I. Light chain rearrangement. Semin Immunol. 2002;14:169–90. doi: 10.1016/s1044-5323(02)00041-6. discussion 221–22. [DOI] [PubMed] [Google Scholar]
  40. Matsuda F, Ishii K, Bourvagnet P, Kuma K, Hayashida H, Miyata T, Honjo T. The complete nucleotide sequence of the human immunoglobulin heavy chain variable region locus. J Exp Med. 1998;188:2151–62. doi: 10.1084/jem.188.11.2151. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Mattson DH, Roos RP, Hopper JE, Arnason BG. Light chain composition of CSF oligoclonal IgG bands in multiple sclerosis and subacute sclerosing panencephalitis. J Neuroimmunol. 1982;3:63–76. doi: 10.1016/0165-5728(82)90018-2. [DOI] [PubMed] [Google Scholar]
  42. Meffre E, Chiorazzi M, Nussenzweig MC. Circulating human B cells that express surrogate light chains display a unique antibody repertoire. J Immunol. 2001;167:2151–6. doi: 10.4049/jimmunol.167.4.2151. [DOI] [PubMed] [Google Scholar]
  43. Meffre E, Davis E, Schiff C, Cunningham-Rundles C, Ivashkiv LB, Staudt LM, Young JW, Nussenzweig MC. Circulating human B cells that express surrogate light chains and edited receptors. Nat Immunol. 2000;1:207–13. doi: 10.1038/79739. [DOI] [PubMed] [Google Scholar]
  44. Meffre E, Schaefer A, Wardemann H, Wilson P, Davis E, Nussenzweig MC. Surrogate light chain expressing human peripheral B cells produce self-reactive antibodies. J Exp Med. 2004;199:145–50. doi: 10.1084/jem.20031550. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Mehr R, Shannon M, Litwin S. Models for antigen receptor gene rearrangement. I. Biased receptor editing in B cells: implications for allelic exclusion. J Immunol. 1999;163:1793–8. [PubMed] [Google Scholar]
  46. Nadel B, Tang A, Escuro G, Lugo G, Feeney AJ. Sequence of the spacer in the recombination signal sequence affects V(D)J rearrangement frequency and correlates with nonrandom Vkappa usage in vivo. J Exp Med. 1998;187:1495–503. doi: 10.1084/jem.187.9.1495. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Nemazee D, Weigert M. Revising B cell receptors. J Exp Med. 2000;191:1813–7. doi: 10.1084/jem.191.11.1813. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Radic MZ, Erikson J, Litwin S, Weigert M. B lymphocytes may escape tolerance by revising their antigen receptors. J Exp Med. 1993;177:1165–73. doi: 10.1084/jem.177.4.1165. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Radoux V, Chen PP, Sorge JA, Carson DA. A conserved human germline V kappa gene directly encodes rheumatoid factor light chains. J Exp Med. 1986;164:2119–24. doi: 10.1084/jem.164.6.2119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Richl P, Stern U, Lipsky PE, Girschick HJ. The lambda gene immunoglobulin repertoire of human neonatal B cells. Mol Immunol. 2008;45:320–7. doi: 10.1016/j.molimm.2007.06.155. [DOI] [PubMed] [Google Scholar]
  51. Sanchez P, Marche PN, Rueff-Juy D, Cazenave PA. Mouse V lambda × gene sequence generates no junctional diversity and is conserved in mammalian species. J Immunol. 1990;144:2816–20. [PubMed] [Google Scholar]
  52. Schable KF, Zachau HG. The variable genes of the human immunoglobulin kappa locus. Biol Chem Hoppe Seyler. 1993;374:1001–22. [PubMed] [Google Scholar]
  53. Sekiguchi DR, Yunk L, Gary D, Charan D, Srivastava B, Allman D, Weigert MG, Prak ET. Development and selection of edited B cells in B6.56R mice. J Immunol. 2006;176:6879–87. doi: 10.4049/jimmunol.176.11.6879. [DOI] [PubMed] [Google Scholar]
  54. Shlomchik M, Nemazee D, van Snick J, Weigert M. Variable region sequences of murine IgM anti-IgG monoclonal autoantibodies (rheumatoid factors). II. Comparison of hybridomas derived by lipopolysaccharide stimulation and secondary protein immunization. J Exp Med. 1987;165:970–87. doi: 10.1084/jem.165.4.970. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Shlomchik MJ, Nemazee DA, Sato VL, Van Snick J, Carson DA, Weigert MG. Variable region sequences of murine IgM anti-IgG monoclonal autoantibodies (rheumatoid factors). A structural explanation for the high frequency of IgM anti-IgG B cells. J Exp Med. 1986;164:407–27. doi: 10.1084/jem.164.2.407. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Siminovitch KA, Bakhshi A, Goldman P, Korsmeyer SJ. A uniform deleting element mediates the loss of kappa genes in human B cells. Nature. 1985;316:260–2. doi: 10.1038/316260a0. [DOI] [PubMed] [Google Scholar]
  57. Smith TF, Waterman MS. Identification of common molecular subsequences. J Mol Biol. 1981;147:195–7. doi: 10.1016/0022-2836(81)90087-5. [DOI] [PubMed] [Google Scholar]
  58. Solomon A, Weiss DT, Schell M, Ringelberg C, Ch'ang LY, Klebig ML. Identification and characterization of a human Vlambda5 (T1) germline gene that encodes structurally unique lambda light chains. Mol Immunol. 1997;34:463–70. doi: 10.1016/s0161-5890(97)00038-2. [DOI] [PubMed] [Google Scholar]
  59. Staudt LM, Lenardo MJ. Immunoglobulin gene transcription. Annu Rev Immunol. 1991;9:373–98. doi: 10.1146/annurev.iy.09.040191.002105. [DOI] [PubMed] [Google Scholar]
  60. Tiegs SL, Russell DM, Nemazee D. Receptor editing in self-reactive bone marrow B cells. J Exp Med. 1993;177:1009–20. doi: 10.1084/jem.177.4.1009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. van der Burg M, Tumkaya T, Boerma M, de Bruin-Versteeg S, Langerak AW, van Dongen JJ. Ordered recombination of immunoglobulin light chain genes occurs at the IGK locus but seems less strict at the IGL locus. Blood. 2001;97:1001–8. doi: 10.1182/blood.v97.4.1001. [DOI] [PubMed] [Google Scholar]
  62. Walker RW, Keir G, Thompson EJ. Assessment of cerebrospinal fluid immunoglobulin patterns after isoelectric focusing. Use of kappa and lambda light chain immunoperoxidase staining. J Neurol Sci. 1983;58:123–34. doi: 10.1016/0022-510x(83)90115-6. [DOI] [PubMed] [Google Scholar]
  63. Wardemann H, Hammersen J, Nussenzweig MC. Human autoantibody silencing by immunoglobulin light chains. J Exp Med. 2004;200:191–9. doi: 10.1084/jem.20040818. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Weigert M, Perry R, Kelley D, Hunkapiller T, Schilling J, Hood L. The joining of V and J gene segments creates antibody diversity. Nature. 1980;283:497–9. doi: 10.1038/283497a0. [DOI] [PubMed] [Google Scholar]
  65. Williams JM, Gorevic PD, Looney RJ, Abraham GN. Isoelectric focusing characterization of IgM-VKiiib immunoglobulin light chains and their association with anti-IgG autoantibodies in essential mixed cryoglobulinaemia, Sjogren's syndrome and rheumatoid arthritis. Immunology. 1987;62:529–36. [PMC free article] [PubMed] [Google Scholar]
  66. Woo Y, Krueger W, Kaur A, Churchill G. Experimental design for three- color and four-color gene expression microarrays. Bioinformatics. 2005;21(Suppl 1):i459–67. doi: 10.1093/bioinformatics/bti1031. [DOI] [PubMed] [Google Scholar]
  67. Wrammert J, Smith K, Miller J, Langley WA, Kokko K, Larsen C, Zheng NY, Mays I, Garman L, Helms C, James J, Air GM, Capra JD, Ahmed R, Wilson PC. Rapid cloning of high-affinity human monoclonal antibodies against influenza virus. Nature. 2008;453:667–71. doi: 10.1038/nature06890. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Wu TT, Kabat EA. An analysis of the sequences of the variable regions of Bence Jones proteins and myeloma light chains and their implications for antibody complementarity. J Exp Med. 1970;132:211–50. doi: 10.1084/jem.132.2.211. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Yurasov S, Tiller T, Tsuiji M, Velinzon K, Pascual V, Wardemann H, Nussenzweig MC. Persistent expression of autoantibodies in SLE patients in remission. J Exp Med. 2006;203:2255–61. doi: 10.1084/jem.20061446. [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Yurasov S, Wardemann H, Hammersen J, Tsuiji M, Meffre E, Pascual V, Nussenzweig MC. Defective B cell tolerance checkpoints in systemic lupus erythematosus. J Exp Med. 2005;201:703–11. doi: 10.1084/jem.20042251. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01
NIHMS573307-supplement-01.doc (336KB, doc)

RESOURCES