Utilization on CDR3 Length

Molecular Medicine 4: 525-553, 1998

Molecular Medicine© 1998 The Picower Institute Press

Frequent N Addition and Clonal Relatednessamong Immunoglobulin Lambda Light ChainsExpressed in Rheumatoid Arthritis Synovia andPBL, and the Influence of VA Gene SegmentUtilization on CDR3 Length

S. Louis Bridges, Jr.Division of Clinical Immunology and Rheumatology, Department ofMedicine and Department of Microbiology, University of Alabama atBirmingham, and the Birmingham Veterans Administration MedicalCenter, Birmingham, Alabama, U.S.A.Accepted June 23, 1998.

Abstract

Background: In rheumatoid arthritis (RA), B-lineagecells in the synovial membrane secrete large amounts ofimmunoglobulin that contribute to tissue destruction.The CDR3 of an immunoglobulin light chain is formedby rearrangements of VL and JL gene segments. Additionof non-gernline-encoded (N) nucleotides at V(D)J joinsby the enzyme termninal deoxynucleotidyl transferase(TdT) enhances antibody diversity. TdT was previouslythought to be active in B cells only during heavy chainrearrangement, but we and others reported unexpect-edly high levels of N addition in kappa light chains. Wealso found clonally related kappa chains bearing unusu-ally long CDR3 intervals in RA synovium, suggestingoligoclonal expansion of a set of atypical B lymphocytes.In this study, we analyzed lambda light chain expressionto determine if N addition occurs throughout immuno-globulin gene rearrangement and to compare CDR3lengths of lambda and kappa light chains in RA patientsand normal individuals.Materials and Methods: Reverse transcription-polymer-

ase chain reaction (RT-PCR) amplification of VAM tran-scripts was performed on RA synovia and peripheral bloodlymphocytes (PBL) and normal PBL for which kappa rep-ertoires were previously analyzed. Representative A+ PCRproducts were doned and sequenced.Results: Analysis of 161 cDNA clones revealed that Naddition occurs in lambda light chains of RA patientsand normal controls. The lambda light chain reper-toires in RA were enriched for long CDR3 intervals. Inboth RA and controls, CDR3 lengths were strongly in-fluenced by which VA gene segment was present in therearrangement. Five sets of donally related sequenceswere found in RA synovia and PBL; one set was found innormal PBL.Conclusions: In humans, unlike mice, N addition en-hances antibody diversity at all stages of immunoglobu-lin assembly, and the structural diversity of lambdaCDR3 intervals is greater than that of kappa light chains.Clonally related VA gene segments in RA support anantigen-driven B-cell response.

IntroductionRheumatoid arthritis (RA) is a chronic sys-temic illness characterized by inflammationand hyperplasia of the synovial membrane ofaffected joints (1-3). Infiltration of the syno-

Address correspondence and reprint requests to: Dr. S.Louis Bridges, Jr., Division of Clinical Immunology andRheumatology, 415 Lyons-Harrison Research Building,University of Alabama at Birmingham, Birmingham, AL35294-0007, U.S.A. Phone: (205) 934-7995; Fax: (205)975-5056; E-mail: [email protected]

vium by B lymphocytes and plasma cells is ahallmark of the disease and often contributesto joint destruction (4,5). Antibodies are het-erodimeric proteins composed of two immuno-globulin heavy (H) chains and two light (L)chains, each of which has a variable (V) do-main for antigen recognition. In order to rec-ognize the virtually limitless number of anti-gens in the environment, multiple mechanismsto generate antibody diversity have evolved. Inhumans, these include the use of many differ-

526 Molecular Medicine, Volume 4, Number 8, August 1998

ent gene segments to form antibodies (6), dif-ferential pairing of different heavy and lightchains, somatic hypermutation, and the inser-tion of non-germline-encoded (N) nucleotidesat the sites of gene segment rearrangement(reviewed in refs. 7-9).

Variable heavy-chain domains are encodedby variable (VH), diversity (DH), and joining(JH) gene segments, which undergo sequentialsomatic rearrangements to become juxtaposedin the genomic DNA of the B cell (8). Afterheavy chain rearrangement, the kappa locusundergoes rearrangements of V and J gene

segments to generate light chain variable do-mains. If kappa light chain rearrangement failson both alleles, lambda light chain V-J rear-

rangements occur. In humans, about 40% ofexpressed antibodies contain lambda lightchains (9). VH and VL gene segments encodethree framework regions (FRI, FR2, and FR3),which are separated by two highly variabledomains, termed complementarity determin-ing regions (CDRs). The regions of the VH_DH-JH and VL-JL joins define the heavy andlight chain CDR3 intervals, respectively, whichare at the center of the antigen-binding site(10) and are usually directly involved in anti-gen binding.

Non-germline-encoded (N) nucleotides are

added at the DH-JH and VH->DH junctions of

immunoglobulin heavy chains by the enzyme

deoxynucleotidyl transferase (TdT). Until re-

cently, expression of TdT was thought to berestricted to the pro-B-cell stage of B-cell de-velopment, when most heavy-chain rear-

rangement occurs (1 1). However, we and oth-ers have shown that N-region addition can

occur in VK-JK joins of B cells of normal indi-viduals (12-16) and patients with RA (17,18).As was the case with kappa light chains untilrecently, N-region addition is not thought tooccur in V-J joins of lambda light chains (19),but this issue has not been formally addressed.Sequencing of the human lambda locus(20,21) has made such analysis feasible by al-lowing assignment of V and J gene segmentprogenitors.

Although N-region addition can occur inkappa light chains, the length of the CDR3intervals is highly regulated in humans, with>95% of sequenced kappa chain CDR3s fromnormal peripheral B cells having either 9 or 10

amino acids (10). We have shown that in someRA patients, the kappa light repertoire is en-

riched for unusually long CDR3 intervals

(16,22). This finding may be due to differencesin regulation of gene rearrangement, for exam-ple, prolonged expression of TdT, or to anti-genic selection of B lymphocytes expressingunusual antibodies.

The goals of the present study were to de-termine the degree of N-region addition andCDR3-length heterogeneity in lambda lightchains expressed in RA patients and normalcontrols. We focused on expression of the larg-est VA family, VAIII (21), which is frequentlyused in autoantibodies from patients with RA(23-26). To avoid potential bias in the resultsby selecting for antibodies that express a par-ticular reactivity, we used a reverse transcrip-tion-polymerase chain reaction (RT-PCR) ap-proach on unsorted cells from synovial tissueand peripheral blood lymphocytes (PBL) oftwo patients with RA and PBL of three normalcontrols. VK-JK joins expressed in these samesamples have been extensively characterized,which allowed comparison of findings in thekappa and lambda repertoires.

We report analysis of a total of 194 VAsequences-161 cDNA sequences and 33 unre-arranged genomic DNA sequences (derivedfrom 6 germline VAIII gene segments). Exten-sive N-region addition was found in the VA-JAjoins of both RA patients and normal controls.Surprisingly, two of three normal individualsexhibited more N-region addition in lambdalight chains than in kappa light chains. Thelambda repertoires of the RA patients wereenriched for unusually long CDR3 intervals,similar to the pattern previously reported fortheir kappa repertoires. The CDR3 lengths oflambda light chains in both RA and normalindividuals were affected by both N-region ad-dition and by the length of the VA gene seg-ment utilized in the rearrangement. The en-richment for uncommon CDR3 structures andevidence of oligoclonal B cell expansion in RAprovides further support for the hypothesisthat changes in the repertoire reflect selectionby an unknown antigen or sets of antigens.

Materials and MethodsPatient Characteristics and Isolation of Synovial Cellsand Peripheral Blood Mononuclear CellsClinical characterization of the RA patients (BCand AS) and normal controls used in this studyand the methods used to process the synovialtissue and peripheral blood have been reported

S. L. Bridges, Jr.: Immunoglobulin VA Expression in Rheumatoid Arthritis 527

previously (16,17). Patients BC (62, WF) andAS (42, BF), as well as normal controls LB (36,WM), LK (23, WF), and IT (62, WF) have beensubjects of in-depth analysis of V gene segmentutilization, N-region addition, and CDR3-length variability of the kappa light chain rep-ertoire (16,17,27). Synovial tissue and PBLwere obtained from AS in close temporal prox-imity. In this study, PBL from patient BC wereobtained several years after isolation of syno-vial tissue because the supply of RNA from PBLobtained at the time of surgery had been ex-hausted.

Generation of cDNA and PCR Amplification of VA-containing Transcripts

Total RNA was isolated from each sample us-ing the guanidinium isothiocyanate technique(28) or using Tri-Reagent (Molecular ResearchCenter, Inc., Cincinnati, OH). Oligo d(T)-primed first-strand cDNA was generatedfrom total RNA as previously described (16).PCR amplifications were performed on 2-Alaliquots of first-strand cDNA, using Taq DNApolymerase as previously described (16). PCRconditions were as follows: 30 cycles of dena-turation at 94°C for 1 min, annealing at 50°Cfor 2 min, extension at 720C for 4 min, with afinal extension at 720C for 10 min. A low an-nealing temperature was used to decrease thelikelihood that mutated sequences would bemissed.

The VAIII family contains three subfamiliesdefined serologically and by sequence analysis(21,29-31). Thus, three separate PCR amplifica-tions were performed on each tissue sample toenrich for sequences containing VAIHa-, VAIIIb-,and VAIIIc-derived gene segments. The threesense primers used were derived from germlineleader sequences and were each used in conjunc-tion with an antisense consensus CA primer, LB-69. Leader sequences used were LB-75 (VAIIIa),LB-76 (VAIIIb), and LB-77 (VAIIIc) (Fig. 1). Tocontrol for possible contamination, mock PCRreaction mixtures lacking template or containingproducts of the first-strand cDNA reaction with-out reverse transcriptase were prepared. None ofthe controls contained amplified product as as-sessed by ethidium-stained agarose gel electro-phoresis or Southern blot analysis using a 32p_labeled internal CA oligonucleotide, LB-70(Fig. 1).

Isolation of Genomic DNA from non-B Cells andAmplification of Germline VA Gene Segments

In an attempt to allow definitive assignment ofcDNA sequences to germline progenitor genesegments, we sequenced unrearranged VAIIIgene segments from genomic DNA of T lym-phocytes from two individuals (RA patient BCand normal control LB). PBL were obtained bystandard Ficoll-Hypaque gradient centrifuga-tion (32). T lymphocytes were isolated fromPBL by sheep red blood cell (SRBC) rosetting(33). Briefly, 1 ml of 25% 2-aminoethyliso-thiouronium bromide (AET)-SRBC was addedper 108 PBL. The PBLs were resuspended in-10 ml 5% fetal calf serum (FCS)/108 PBL.The PBL-SRBC solution was distributed evenlyinto sterile round-bottom 1 5-ml tubes and cen-trifuged for 5 min. The tubes were placed onice for 30 min, then the pellets were resus-pended in ice cold 5% FCS. Aliquots of 40 X106 cells were pipetted onto Ficoll-Hypaqueand centrifuged at 1800 rpm at 40C for 30 min.The interface, which contained the B lympho-cytes, was removed. The T cells in the pelletwere then washed three times and genomicDNA was extracted, purified by phenol extrac-tion, and precipitated in ethanol using stan-dard techniques (34).

PCR amplifications were performed on2-,ul aliquots of genomic DNA. PCR conditionswere as follows: preheat to 940C for 2 min,followed by 35 cycles of denaturation at 940Cfor 30 sec, annealing at 500C for 30 sec, exten-sion at 720C for 1 min, with a final extension at720C for 7 min. Two separate PCR amplifica-tions were performed on each genomic DNA,one to amplify VAIIIa and VAIIIc gene segments(using upstream primer LB-123 and down-stream primer LB -125), and one to amplifyVAIIIb gene segments (using upstream primerLB-124 and downstream primer LB-125)(Fig. 1). Primer LB-123 was derived from theintronic sequences of the VAIIIa gene segmentsIGLV3S2 (35) and humlv318 (29), and theVAIIIc gene segment III.1 (30). Primer LB-124was derived from intron sequences fromthe VAIIIb gene segments hsigglll 50 (26)and hsiggll295 (26). Primer LB-125 was de-rived from the heptamer, 23 base pair spacer,and nonamer sequences of the recombina-tion signal sequences (RSS) of the IGLV3S2,humlv318, and III.1 gene segments (Fig. IC).PCR products were probed with an internalprobe, LB-126, derived from a portion of the


APrimer Gene Region Orientation Sequence (5'-+3')LB-75 VXIIIa Leader Sense TCCGAATTCTCCTCTCTCACTGCACAGLB-76 VXiIb Leader Sense TCCGAATTCTCTGCACAGTCTCTGAGGCCLB-77 VXHIc Leader Sense CCCTGAATTCCTCGGCGTCCTTGCTTACTGCALB-69 CX Constant Antisense GGGAATTCGCTCCCGGGTAGAAGTCACTLB-70 Cx Internal Antisense GGGAATTCTTG(GA)CTTGGAGCTCCTCAGAGGAGGGLB-123 VkIIIaIc Intron Sense TCCAGC(CG)TG(GT)CC(CT)TGA(CT)TCTGAGCTCAGGALB-124 VXIIIb Intron Sense GTGCT(GT)(CT)(CT)CCCAGGCCCTGCTCCAGGCLB-125 VXIII RSS Antisense(GT)GeTTTCT)TGTCTCACTTCC(GT)CATCTGCCTGTGT(CT)AC(CT)GTGLB-126 VXIII FR2 Sense CAGCAGAAGCCAGGCCAG

BLeader Sequences

VXIHa 1v3 18 ATGGCCTGGACCGTTCTCCTCCTCGGCCTCCTCTCTCACTGCACAGGTGATCCCCCCiglv3s2LB-75 ...GAATTC.

VXHIb hsiggll1 50 ..... ......... .T... CC... A.. T.CTC.GAGG..hsiggll295 .CC. ... ..CC.. A.CA.TT.TCTC.GAGG..LB-76 ... GAATTCT.TCTC.GAGG..

VXmIcIII.1 .....A .T.CC ... T.G. .. TG.T...... C.G ... TLB-77 .CC.GAATTC ... ...G . TG. .T.

C

Fig. 1. (A) Sequences of oligonucleotides usedas primers and probes. Parentheses indicate sitesof degeneracy. RSS, recombination signal sequence;FR2, framework 2. (B) Comparison of VAIIIa, b, andc leader sequences and primers used for RT-PCR to

the germline humlv318 sequence. Dots indicate nu-

cleotide sequence homology. Underlined sequences

indicate EcoRI restriction sites. (C) Sequences of RSSof gene segments IGLV3S2, Iv318, and III.1.

framework 2 domain that is highly conservedamong VAIII gene segments.

Cloning and Sequencing of PCR Products

Aliquots of PCR products were subcloned usingthe TA cloning kit (Invitrogen, San Diego, CA).Plasmids were transformed into INVaFF E. coli byelectroporation using a Bio-Rad Gene Pulser(Hercules, CA). For colonies that hybridized tothe CA probe, plasmid DNA was obtained andsequenced by the dideoxy chain terminationprocedure (36) either manually or using an au-

tomated sequencer (Applied Biosystems, Inc.,Model AB373).

Sequence Analysis

The FR and CDR domains of each of the se-

quences were compared individually to corre-

sponding domains of published human VA se-

quences using the computer program SAW(Sequence Analysis Workshop) (37). Se-quences were assigned to germline gene seg-ments according to highest degree of nucleo-tide sequence homology. Levels of somatic

Recombination Signal SequencesHeptamer I| Spacer -+ Nonamer

3S2 CACGGTA ACACAGGCAGATGAGGAAGTGAG ACAAAAACAlv318 ...... ...... .........

II1.1 ...A..G . C . .. .G .C

S. L. Bridges, Jr.: Immunoglobulin VA Expression in Rheumatoid Arthritis

Table IA. Germilne derivation of VA gene segments in cDNA clones from patients with rheumatoidarthritis and normal controls

Vk FamilyGene Segmen

RA AS SynoviumAS PBLBC SynoviumBC PBL

Normal LB PBLIT PBLLK PBLrTotal

5a_00010102

rota131222120182920161

Genrline gene segments in Table 1A were reported as follows: 1v1042 (42) , DPL6 (43), 3S2 (35), lv318 (29), 2-14, 2-19, 4-2,and 5-4 (21), hsigglllS0 (20, 26), 3a, 3p, 5a, Scl, and 5c2 (20), m.1 (20, 30), 8A1 (44), lv801 (45), and 6S1 (46).

hypermutation were assessed by comparingthe FR1 through the FR3 domains (codons1-88 according to ref. 10) of each completetranscript to the appropriate germline sequenceand calculating mean divergence for each sam-ple. Truncated clones were not included in thesomatic mutation analysis.

P nucleotides are nucleotides that are pal-indromic to terminal nucleotides of coding se-quences (38). They are inserted adjacent to thecorresponding coding sequences at the junc-tion of rearranged gene segments and arethought to be a consequence of the resolutionof hairpin structures that occur during V(D)Jrecombination (39). In this study, nucleotidemismatches at the 3'-end of the VA gene seg-ment or the 5'-end of the JA gene segment thatdid not appear to be palindromic to the V or Jwere assumed to represent N-region additionrather than somatic mutation. In order to al-low valid comparisons between lambda andkappa chains expressed in the same samples,sequences of previously reported VK-JK joins(16) were reanalyzed to look for the presenceof P nucleotides.

Statistical AnalysisDifferences in the amount of somatic mutation,N-region addition, and CDR3-length heterogeneitybetween patients with RA and normal individualswere analyzed using the Chi-square test, Fisherexact test (two-tailed), or Student t-test, as appro-priate. Three-way comparisons were performedusing analysis of variance (ANOVA).

ResultsGermline VA Gene Segment Derivation of cDNASequences

In humans, there are three separate clusters ofVA gene segments on chromosome 22ql 1.2(40). There are a total of -69 VA gene seg-ments, including -36 functional VA gene seg-ments and -33 pseudogenes, which aregrouped into 10 families on the basis of nucle-otide sequence homology (20,21). Two genesegments that are not part of the major locus,so-called orphon sequences, have been re-ported on chromosome 8q1 1.2 (41). Allelicpolymorphism of the human lambda locus ap-pears to be limited (20). As mentioned above,the VAIII family is the largest VA family, com-posed of 10 functional members and 13 pseu-dogenes, which are subdivided into three sub-groups (a, b, and c) (21,29-31).

The VA gene segments of 161 cDNA cloneswere assignable to known germline gene seg-ments (Table lA). Of 10 VAIII germline genesegments with open reading frames (20,21),123 of 126 (97.6%) VAIII transcripts were de-rived from six gene segments: VAIIIa gene seg-ments IGLV3S2 (35) (12 clones, 7.5% of thetotal), 2-14 (21) (11 clones, 6.9%), andhumlv318 (29) (7 clones, 4.4%); VAIIIb genesegments hsiggllS0 (3m) (20,26) (46 clones,28.8%) and 3p (20) (11 clones, 6.9%); andVAIIIc gene segment III.1 (3r) (20,30) (35clones, 21.9%). This pattern of VAIII gene uti-lization is similar to that found by other inves-tigators, who assessed the expression of VA

529


gene segments by screening cDNA librariesgenerated from normal PBL (47). In that study,the most commonly expressed VAIII gene seg-ments were 3h (which includes closely relatedgene segments IGLV3S2, humlv318, and 2-14;ref. 20), 3r (III.1), 3p, and 3m (hsigglll50),with less frequent representation of 3e, 31, 3a,and 3j. In the present study, because of cross-

reactivity of the PCR primers, members of theVAI (9 clones, 5.6%), VAIV (8 clones, 5.0%),VAV (10 clones, 6.3%), and VAVI (9 clones,5.6%) families were also represented.

Several clones were found to have deletionsor insertions in the V gene segment (Figs. 2 and3). ASSynL191 had a three-nucleotide deletionin CDR1 and LKPBLL68 contained a three-nu-cleotide insertion in CDR1. LKPBLL1O had one

nucleotide missing from FR2 and a one-base pairinsertion in the CDR3 interval. ITPBLL75 con-

tained a ten-base pair deletion at the FR3-CDR3junction. Single base pair insertions in CDR3were seen in clones LBPBLL4, ITPBLL4, and LK-PBLL53. Overall, there were 5 out-of-framecDNA sequences, all of which were isolated fromnormal PBL.

Derivation of Unrearranged VA Gene SegmentsAmplified from T Lymphocytes

Of 33 clones from two different individuals(RA patient BC and normal control LB), sixdifferent gene segments [3i, hsigglll50 (3m),3a, 3p, III.1 (3r), and 3e] were represented(Table 1B). Thirty-one of 33 (94%) of theclones were 100% identical to known germlinegene segments. Clone LBLG9 was found tohave two nucleotide differences from the re-

ported 3p sequence: GGG instead of GAG at

codon 50 in CDR2, resulting in a glycine resi-due instead of glutamine, and ATG instead ofGTG at codon 80 in FR3, resulting in a methi-onine residue rather than a valine. CloneLBLB14 had three nucleotide differences fromthe reported 111.1 sequence: ACG instead ofACA at codon 18 in FRI, which does notchange the encoded threonine residue, GCAinstead of ACA at position 70 in FR3, resultingin alanine instead of threonine, and GTT (Val)instead of GCT (Ala) at codon 80 in FR3. Noneof the clones reported here appeared to bederived from these two variant sequences. Oneor more of these five sequence changes mayrepresent PCR artifact induced by Taq DNApolymerase (see below).

Possible Allelic Variants of VA Gene Segments

Among the cDNA sequences, we found multi-ple sets of clones that had shared differences inthe VA gene segments (Table 2) but were def-initely not clonally related, as they utilizeddifferent JA gene segments or had discordantN-region addition and/or CDR3 lengths(Fig. 3). This finding suggests the presence ofnovel allelic variants of germline gene seg-ments IGLV3S2, II1.1, hsigglllS0, and iglv6Sl,or the presence of novel gene segments closelyrelated to those germline genes. In the se-

quence analysis of unrearranged gene seg-

ments (see above), we did not isolate germlinesequences containing the variant codons shownin Table 2. Possible explanations for this dis-crepancy are the small number of genomic se-

quences we analyzed or sequence mismatchesbetween our PCR primers and intronic se-

quences of allelic variants.

Fig. 2. Deduced amino acid sequences of VAcDNA clones from RA synovia, RA PBL, andnormal PBL. Dots indicate sequence homology;single-letter amino acid abbreviations are used, X,stop codon in deduced amino acid sequence. *, ter-mination of cDNA transcript. f, Clone BCSynL32contains FRI and CDR1 domains from VA gene seg-ment 6S1 and FR2 through CDR3 domains from VAgene segment DPL8. §, Clones containing insertionsor deletions: ASSynL191 has a three-nucleotide de-letion in CDR1, denoted by dash; ASPBLL54 has atwo-base pair deletion in FRI; ITPBLL4 has a one-nucleotide insertion in the CDR3 interval; ITPBLL75contains a 10-nucleotide deletion at the FR3-CDR3junction; LBPBLL4 has a one-nucleotide insertion in

CDR3; LKPBLL1O contains a one-nucleotide deletionin FR2 and a one-nucleotide insertion (or a two nu-cleotide deletion) in CDR3; LKPBLL53 has a one-nucleotide insertion (or a two nucleotide deletion)in the CDR3 interval; LKPBLL68 has a three-nucle-otide (one codon) insertion in CDR1. |1, Clone BC-SynL38 contains a stop codon in FR2. Genbank ac-cession numbers for nucleotide sequences are asfollows: AS Synovium: AF060120-AF060150; ASPBL: AF058057-AF058078; BC Synovium:AF063714-AF063734; BC PBL: AF063694-AF063713; IT PBL: AF063735-AF063762; LB PBL:AF063765-AF063782; LK PBL: AF064494-AF064513.

.)O

S. L. Bridges, Jr.: Inumunoglobulin VA Expression in Rheumatoid Arthritis 531

v X X

A > >

W CO)

A:CO * *oU, * ,qa * .,

a . .n

MW

rn * *

U) C*/ E

E)z * z -

U) *CoU)cq . .c,Q

U) * *U)

V E * *E-4 -A C :.)

lx

P:-

II

H > 0

0a z; 0 0a 0 :X>WW

*

... n ,4,

..>4

*-C4*--.,4.z....* X * n * * * * *n * a * * X * **--4:*-V*-<:*K ........~~~~ In.D

...n E- N.D* E** 13 * * * * * 3 * * *H * *

* * *H *n *Z* * * H

. . ,4

** >4 .z >4EtE *E01*q**EzZb*****V*****V*V0

* E-H4 E* * * * 4 * * *... X A.A~~~~~~~~~~~j

*--Wn*--.X.X......~~~E-* , p

*-C, *--U]U bs**.. E-4 E-4 .,.

.0 .1M * > >

* 1 D41.:4- 4 4 * *E* r >1*4H1* *H *H 4

H l U) W

cn E-4 * 0Atq) H E* E-$==

>.. - U) *E- E->4---CKOq

q;.. . .l u

Ca *4 .......~

E *....

H

,l . . . . . .

CO . . * a a

zP:* :1 :*> * * * U*la VIq *z * n *c f*tn V* *V .. ....

p~~~~~~OO. Ot. pX . ..****V

* CY * - - sy ,E-i

* qW* * q . a . . . . .

* ,q; * * * s * s~~cj*W*.D . .. ... ...

>>.. .; > . . . . . *H* l ul: - - - -W

* )'¢ CO*S .

6

E- EzE-4 E- E- E-4 E-4 E-4 E-4 E-4 E-4 E- E- P P E4 P E4 E- E- E- E- E- E- E- E- P E- PE- E-PC!4 P E- Cl E-4 E- E-4bLtMuqVLDvv v oV V VtU

P4 444i D4 00 44 r- r- r- r. r- >4 44 No 44 rZiedv.>. 8. 5 . 8. . c>x j . 2 1304 >H

cl (It0 00 NN D1004k rVN o H v in v r- rD 0 UDC1 O n

C14 r- sr- C1 C1N N N N H HN m rs r-s cs cs cs c csq ri c r- sr-i L Ln n cl ri 64 ci cq cq (SC N(n%D o I H H4- b_ Ca 64 64 0 .4 -4. .1 1. .4 .3 .1 .4 .4 .4 .4 .4 . I .1 1. .4 .4 64H 0 0 b4aaaaaa

c'. couz cwntqul uz ucn qc f qt Q- )oXU) CO UCa cq co -I co-1cnmcco coq ce cwqco co uxo z IqHUUcn Ca uz0mw Hm Im cq uzmm u

cozmu zu nC Q1Vmc qc zC NC l QU QVVVV "c zt au ?Vfiu

._

_. _. _.

co

;I-(A.:1

l , c,* ]

Cy ofs*

cE

W

* Yc **zO -


IA

10P444

IIA04>

ACl)~~~UU)4 U)

U)

AU.U UU

AL)U..0.

CD4. 4 U).

A>4 A ) E49

UH U)

DU,)

§. 0' . ...

A: A

CflU)000WIC fI00 1.4

0A4rA

I

I".0t

v

E-E.EE- E-. E- E E- E E- E4l E-1 E- E- E-. E- E-4EHE-4El a a U)ED 0 aU)a

U)U)U)U)U)0 c E 4 -4 E-U) t' CD. ,U ) ) ) ) U U.U..ED.UN4 r4r4 P4N P z ks r- r 2 0 4P

> >>4 U)>

Ou~~~~~~~~C

A L L

E-4U)~U

ID,14U

U~~~~~~~~~~l) C

P4 ..04 P P

04 4

A Cj . .~ C Cl .

1 12 4U.. .

Ca 1 CO)ZC4 -4..

C4 ~

U

10 U)4

z

A E :'n

. .

E* COE)-U

IJ

x >4

> ~E-.

*UH

.0 .U.)c.

'0..

a'.'E . Z

.E-4

.0.

U). P*0.

E-4U) U U)

E E4E.. O P

CO)0

.04

U)11

C 4 D0

MO0404 0) 4004 040 DO4040 04 40d04 404 0404 $04 404 04(> > U)U)o m )Uo U) )o 0. OOuuH c 00)imc 4 9 m 0 0 H 04m

Fig. 2 (Continued)

The presence of sequence variations that

could represent allelic variants was not unex-

pected, as differences have been found in several

germline VAIl gene segments from different in-

dividuals. These include the germline gene seg-

ments 3p/hsiggII295 (4 nucleotide differences)

and 3m/hsigglll 50 (3 nucleotide differences)

(20). An informative example is the case of the

closely related sequences humlv3 18 (29), 2-14

(21), and IGLV3S2 (35). The humlv3l8 and 2-14

gene segments differ by only one nucleotide in

the coding sequence and are 4 nucleotides and 5

r.. E-.E E

A0

U)

U) 0U)U)

YE-. >0

Uf)

014Vnt,'U)

04

U)P4 ia

U) c

icy)

toU U) U

:1.4

E4


A

UCO) -COvE .4 .

AU -U -U

CO) *CO) *CO)

CO -C CO9

4 4

ACO CO *C

QCO *CO *CO

Un COQ U

C ) CO)

A Z4 .4A: Z~..c,j

a. O.

-0-0

W~ *)1*IO,

A>4 .4 *:4-

IH Ha H4E4

UH HC*H)WOCOCO)*c

CO *CO*CLO) :

C/-)- C-)AUI *U *

Hn *HC,H,

CQ CYCC

Hco

IA

v114P4

v1IA

AU

CY)

CDCO

ACO3

CO)

ACO

H

~CDUCO

VC'C

PA

H Hp H HHH

CO~~~~~~~~~>HCC

CO~~~~~~O

A U

coC)~~~~~C

'0 -co, CO

UV.

13:3

W Q

AU~ ~ ~ ~ * C

Cl.

CO)CO)

COl0-

COl

CO)

0.-

C'-4

0 -CO)

0-

COI

CO*

U-4

C.D

HE-

E**HH

d CO4E

H *.'):. :* >.CO*

-H.

.0.-*H0..H

.') *.E-. -CO

..4.

.0-

.C:I.

.-H

.-H

-C.i

-COz

-H

-CO~

CO

CO--

-HHH

---~ -CO:

CO-,, : : : 9 : : : : ~1 !

mL% Go WC%D IH 0 0 OH Pk t3 )d O)O0 c-4C'

,. C C COCO-4 CO 1COCOCOCOCOCOCOC O1OOOOOOOOOOOOOOOOOOOO OCO

Ln M-Fi u ~ HH c .IHHHHH HHHHH HHHH

Fig. 2 (Continued)

nucleotides different from the IGLV3S2 gene

segment, respectively. The finding that the

humlv3l8 (3h) and 2-14 gene segments are lo-

cated in the same position in the lambda locus on

chromosome 2ql11.2 in mapping studies (21,40)

suggests that they represent allelic variants of the

same gene segment. The fact that none of the

five individuals in this study expressed both

humlv3 18 and 2-14 (Table lA) provides further

support for this hypothesis.

533

CO)

U

U)

C'4

DU,

H

A:

VA

E- 0.0. 0.~COCO

ED

CO

zC:)

CO

CO

0I

0.Ul

CO

E-4


a,C,4

.CO *CO)WC WC

E-C

0

CW. cl

.CO) WC

E-i%CO

CO CO)

Ca

E.13

CO0)

*04

.Ol

*E-.H

.Cl)

.O

C,.

Ca)

.Ol

E~

.CO.

E-.

HE4

z

N, 40

CO CO CO *COCOCOCOCOCOCOCOOCOCOCOCOCOCOCOA: A: A:HAA:AA:AA:AA:AA:AA:AA:AA:

A

E

A~

AC)

C,.t

AIL . . .

P4 Co...A

AU)

V Cs)

AP4

CD A

CO

A

aca

A:

C,

CO

0

CO)E CO

H COH *H

CO) ICO)

Enol CO) CO

C,CO) *4x

n

A) CO)

6 14

A

CO) C) C

CO) Co Co CO)C O) CO)

CO3 COl CO

C ,) C

'a ) *mVO1 O, of

V

CO) CO) C OA:13

0CO)0 CO ~co

C', C

COH /

AC) *CO)

C) 1

CO) O

.C, .C,

VC, .C,

O 0 A:C OLnN %D )Ln unC3

0

A0 A0 A: ,0m m m Hog

Fig. 2 (Continued)

Variability of JA Gene Segment Expression in RA,

and Expression of a Recently Described JA3 Allelic

Variant

In humans, there are four functional A isotypes

encoded by the CAI, CA2, CA3, and CA7 genes,

each with its own JA gene segment (21,48).

Although the JAl, JA2, JA3, and JA7 gene seg-

ments (49-5 1) are the only expressed JA gene

segments, JA6 is potentially functional (52).

The JA2 and JA3 gene segments have previ-

Vl >4

0Ik C:)0

ACO

* CO

C:CO

03u

~CH

VE.

V~ -

: )i

.

m

: : C-;

: :ji

: : E.

. x

.

rei W3 WI W3


Rheumatoid Arthritis SynoviaVxIIICodonIGLV3S2ASSynL220BCSynL4BCSynL21BCSynL22BCSynL24

ASSynL182ASSynL189

ASSynL232ASSynL234

huaLv3l8ASSynL186ASSynL221ASSynL236

BCSynL36

hsibal5fl1ASSynL202ASSynL210ASSynL21 5ASSynL216ASSynL21 7BCSynL17BCSynL18BCSynL47BCSynL51

ASSynL204ASSynL214ASSynL209*ASSynL21 1*ASSynL184*

VA CDR3 F89 90 91 92 93 94 95 95A 95B 95CI

CAG GTG TGG GAC AGT AGT AGT GAT CAT CC..........T . .....C.GC.

... ... ... ..T .C. ..A .C.

... ... ... ..T .C. ..A .C.

... ... ... ..T ... ..A ... ... ..

... ..A ... T..T T. C. ... ... C.

... ... ... ..G ... ... ... ... ...

... ... ... . TT C.. GA. TC . ... ..

... ... ... ..T A. A . ... ... ...

CAG GTG TGG GAT AGT AGT AGT... ... ... .... ... ... ...

........ ... ... ..... .A. AG G.

.... ... ... ... .C. ... ...

GAT CAT..C.

.C. ..

P N

GC

CCI

... ... ... ...A T. ... ... ... ...

CAA TCA GCA GAC AGC AGT GGT ACT TAT CC... ..G ..T ... A. ..C ..C... .... ..G ... T C .M.

.T..A......... ... .T. ... ..T ..A ..A

... ... ... ... .T. ... ..A

... ... ... ... .A. .A. ...

... ... ... ... .CG ..G ..C

..G ...A.. CT.C . ..G

... ... ... ...

G .. ... .AG ...

... ... ... ...

*. ..- .*--. . ..

..G ... ... ...

..G ... ...

... ... ...

... GA. ...

..A GA. ..G

..A CT....

C..

. . .

C..

GG.

.A.

.C.

.TA

T.*TT.T......T.. ... .

* .. ... .

3-p TAC TCA ACA GAC AGC AGT GGT AAT CAT AGBCSynLl6* . . G.GBCSynL52* ... ... ... ... ... ... ... G. GBCSynL53* ... ... ... ... ... ... ... G .

ASSynL136. G.. A .

BCSynL46 ................ ..A.

2-19 TAC TCT GCG GCT GAC AAC MT CTASSynL201 ... ... ... ... ... ... ... .. III.L1ASSynL226ASSynL235BCSynL65

ASSynL25ASSynL227ASSynL230BCSynL56

CAG GCG !§ GA&C QGC AGC ACT GC... ... ... ... ... ..G... ... ... ... ... ... ... ..

............ ... .. ... G.

... A. A

... .. . . AT.

CG

TAG

MCCCCC

AAAG9 ATA

AAA

TTT

A

ccC

96 97T TAT GTC

T GTG GTA... A..

T TGG GTGC..

T TAT GTC

T TGG GTG

T GTG GTA.CT

.G... .AT

... A..

A.T

T TGG GTG

T GTG GTA

T TGG GTG

il98 99 100 101 102 103 104 105 106 107TTC GGA ACT GGG ACC MG GTC ACC GTC CTA JXl

...... ... ... ... ... ... ... ..C. ...

... ... GG...G... .. GC C...G... .. GCC.

TTC GGC GG.. ... ... ... ... ... ... ...

TTC GGC GGA GGG ACC MG CTG ACC GTC CTA JX2... ... ... ... .... ... ... ... ... ...

... ... ... ... ... ... ... ... ... ...

TTC GGC GGA GGG ACC AA& CTG ACC GTC CTA JX3... ... ..C

TTC GGA-ACT GGG ACC MG GTC ACC GTC CTA JX1... ... .G. ... ...

... ... ... ... ...

... ... ... ... ...

TTC GGC GGA GGG ACC

TTC GGC GGA GGG ACC

.T....G... ........ ..

.... ... ... .C.. ...

MG CTG ACC GTC CTA JA3... ... ... ... ...

AAG CTG ACC GTC CTA JX2... ... ... ... ... .-G. .. . . .. ... ..G... ... ... ... ... ..A ... ... ... ..C... ... ... ... ... .G. ... ... ... ...

...... ... ... ... .... ..A .. ... ... ...AT

~~~~~~~~~.. .....

... ... ... ... ... ... .... ... ... ...

... ... ... ... ... ... .... .... ... ...G

..T ... ... ... ... ... T.. ... ... ..G

... ... ... ... ... ..A ... ... ... ..CTTC GGC GGA GGG ACC MG-CTG ACC GTC CTA JA3... ... ... ... ... ... ... ... ... .......G

... ... ... ... ... ... G. .. G . .

... ... ... ... ... ... T. .. . . .G

... ... ... ... ... .........T..C..C . T ..

C.C.. ..T .. TG

TTC GGC... ....

... ...

TTC GGC... ...

GGA GGG ACC MG CTG ACC GTC CTA JX2... ... .... .GA ... ... ... ..C... ... ... .GA ... ... ... ..C... ... .... GA ... ... ... ..CGGA GGG ACC MG CTG ACC GTC CTA JX3... ... ... ...- ... *... ...- ...

*.-... .. .-...... ... ... ... ....- ...

IT TGG GTG TTC GGC GGA GGG ACC AAG CTG ACC GTC CTA JX3GG .. ... ... ... ... ... ... .................. .... ................ 10

TAT GTCAGAT . ...

9 ...~~~~~~~~~...| | T ~~~~GTGGTA

JTTCACCGIGGGACMCGAGACTGCAI .C. . .Tt G .G. ..C

ITt . ... .T

TTC GGA ACT GGG ACC AAG GTC ACC GTC CTA JX1...... ... ... ... ...

... ... .... ... ... ... ...

... ... ... ... ... ... ...

TTC GGC GGA GGG ACC MG CTG... ... ... ..I ... ... ...

..

... ... ... ... ... ..A ...

* - ..-- *--... .. ... .--. . ..

* -. ...

* .-. ...

. .. ..TACC GTC... ...

... ...

.. ...*-

* .-. -..

I...

T..CTA JX2GT

* *

.. .

.. .

VXI Vk CDR3Codon 89 90 91 92 93 94 95 95A 95B 95CIhumtvlO42 CAG TCC TAT GAC AGC AGC CTG AGT GGT TC|BCSynL32 ... G .C ... ... ... ... ... ..

ASSynL188 ... ... ... ... . T. ... ... ... ... .. 19GG aa

QEL6 CAG TCC TAT GAC AGC AGC CTG AGT GGT TCASSynL191... ... .A.

Fig. 3. Nucleotide sequences of VA-JA joins ofcDNA clones amplified from RA synovia, RAPBL, and normal PBL. Dots indicate sequence ho-mology. Palindromic (P) nucleotides are shown inlower-case letters. *, related sequences (see text fordetails). Clone BCPBLL8 is truncated in the JA genesegment. §, Clones with insertions in CDR3 inter-vals: LKPBLLIO contains a one-nucleotide deletionin FR2 and a one-nucleotide insertion in CDR3; LK-

96 97T TAT GTC

T GTG GTA.G. . .T

Jx

98 99 100 101 102 103 104 105 106 107TTC GGA ACT GGG ACC MG GTC ACC GTC CTA JA1... ..C ... ... ... ... ... ... ... ...

TTC GGC GGA GGG ACC MG CTG ACC GTC CTA JX2... ... ... ... ... ... ... ... ... ..T

IT TGG GTG TTC GGC GGA GGG ACC MG CTG ACC GTC CTA JX3... ... ... ... ...A................

13

12

11

PBLL53, ITPBLL4, and LBPBLL4 contain single nu-cleotide insertions in CDR3 intervals. The sequenceat the VA-JA join in ASSynL25 may represent an in-sertion event from codons 89 through 95A of thegermline VAIII. 1 CDR3 component. Nudeotidesshared by the junctional sequence of ASSynL25 andgermline VAIII.1 are underlined (see Discussion fordetails). The number of codons in CDR3 intervalsare shown.

535

CDR3Len1th

1111111111

1111

111 1

118

11

11

121011101011121010

1010111111

111111

1111

999

16999

.. .

* *

... ... ... ...


vxIIvCodon~auBCSynL37

VX CDR389 90 91 92 93 94 95 95ACAG ACC TGG GGC ACT GGC ATT CA

TA IASSynL59 .... G.. .. .....G. .

htintv8O1 CAG ACC TGG GGC ACT GGC ATT CAASSynL19O .. .. .. .A..... .. .

vxvICodoniv6SIBCSynLlBCSynL3BCSynL38BCSynL59

VX CDR389 90 91 92 93 94 95 95A

CAG TCT TAT GAT AGC AGC MAT CA.. ....AG. ..T G.. .G.

.. . . .....C. T.. ..

.. . . . ..G ..C.

96 97T GTG GTA

.TA..T TGG GTG

98 99 100 101 102 103 104 105 106 107TTC GGC GGA GGG ACC AAG CTG ACC GTC CTA

TTC GGC GGA GGG ACC AAG CTG ACC GTC CTA.....C.. .. .C. .. .. ..T..

JX2

JX3

T TGG GTG TTC GGC GGA GGG ACC MAG CTG ACC GTC CTA JAX3

TC c

C

ASSynLI95 .............T.

T

T

96 97GTG GTA.. .G... A. .

... A. .

.CA .TTGG GTG

Jx.98 99 100 101 102 103 104 105 106 107

TTC GGC GGA GGG ACC MAG CTG ACC GTC CTA JX2..T ........G.CA ........

.. . . .....G.T.. . . .

TTC GGC GGA GGG ACC MAG CTG ACC GTC CTA JX3.. . . ......G.. . . .

RZheumatoid Arthritis PBLVXIII VX CDR3Codon 89 90 91 92 93 94 95 95A.IGLVS2 CAG GTG TGG GAC AGT AGT AGT GAT CATIBCPBLL11 ........ T ... . . .

ASPRILI ...A....T T.G .C.......

bianIvila CAG GTG TGG GAT AGT AGT AGT GAT CATIBCPBLL3 ........ A ..........

BCPBLL5 .........c . . .

ASPBLL3 G.. . . . .CC .C. .. .AASPBLL8 .. .. .. .. .. .. .. . ...

P N P

I

a AGG a

cca

ASPBLL23 .........A. G.C .......ASPBLL39 .. .. .. .. .. .. .. .. .. .

hsi-L I O5ASPBLL48*

ASPBLL51 *

ASPBLL59

ASPBLL42

ASPBLL50ASPBLL54

ASPBLL60BCPBLL27

ASPBLL53

CMA TCA GCA GAG AGC AGT GGT ACT TAT CC..G G.T.....CC. .C. ..A CGC ...

..G G.T.....CC.C.. CGC.... .G C.. ........

...G....T

A........

.A.

...

...

...

...

..A GT.

... ...

... ...

... ...

... ...

I

.CC

.T. G*

.. . .G T.. . ..A G.

32 TAC TCA ACA GAC AGC AGT GGT MAT CAT AGBCPBLL1 .. .. .. .. .. .. .. . . . .

BCPBLL28 .. .. .. .. .. .. .. ... .

BCPBLL25..............A. GCC ... cBCPBLL26 . . . . . . . . . . . . . . . . . . .

BCPBLL31 ......... . . .

aC

TG

CT

a

C

GG

CCC

88

96 97T TAT GTC

TGTG GTA.A.C..

T TAT GTC

G.C..T GTG GTA..C

T TGG GTG

T TAT GTC

T GTG GTA..C

.C. .G

T TGG GTG

T TAT GTC

.TT GTG GTA

Jx98 99 100 101 102 103TTC GGA ACT GGG ACC MAG

TTC GGC GGA GGG ACC MAG

104 105 106 107GTC ACC GTC CTA JXl... G.. . .

CTG ACC GTC CTA JX2

TTC GGA ACT GGG ACC MAG GTC ACC GTC CTA JXl

TTC GGC GGA GGG ACC MAG CTG ACC GTC CTA JX2

TTC GGC GGA GGG ACC MAG CTG ACC GTC CTA JX3

TTC GGA

TTC GGC

ACT GGG ACC MAG GTC ACC GTC CTA JXlC.. ..G CA ... T.. . .GC.. . .G.CA........GG.G. ..........C.. .

GGA GGG ACC MAG CTG ACC GTC CTA JX2.. . . .. . . . . . .T..

TTC GGC GGA GGG ACC MAG CTG ACC GTC CTA JX3.. . . .....A .. ......

TTC GGA ACT.. ...T.

.. ..GT.TTC GGC GGA

GGG ACC MAG GTC ACC GTC CTA JXl

.. ..G.. . . . .

GGG ACC MAG CTG ACC GTC CTA JX2

11

13

121 1

1 112

1212

1 11 110

1 11212

1 1

10

1 11 1

121 110

CAG GCG TGG GAC AGC AGC ACT GCA

..A A.......A.GA.

.. . . . . .G.. ..

.. . . . . A.

.. . . . .T

.T. AT. A.. .T .C.

.. . ....G.. .

G..

..C.

..C.

..C.

..C..G.

.. . . . AT .. T..

CTCT

g AGG

TTCTTGTT

T TAT GTC TTC GGA

.G ......GG ......GG ......

..C ......

T GTG GTA TTC GGC

ACT GGG ACC MAG GTC ACC GTC CTA JXl. . . . G........C

.. ....C.. .

.. . . . . .TT.

T..

T.G

'.GGA GGG ACC MAG CTG ACC GTC CTA JX2

. ....G.. . . . .

Fig. 3. (Continued)

8

10

9

10101010

9

111.1ASPBLL2ASPBLL4ASPBLL10ASPBLL13ASPBLL14BCPBLL2*BCPBLL4*BCPBLL6BCPBLL7BCPBLL8BCPBLL9

ASPBLL12

1010910910109

11119

9


V)lICodon 89 90 91hulyi1042 CAG TCC TATASPBLL31 ... ... ...

ASPBLL35 ... ... ...

ASPBLL34 ... ... ...

VX CDR392 93 94 95 95A 95B 95CIGAC AGC AGC CTG AGT GGT TC... C.

... .

...T A........ G |g1

GCTGA

CA I

96 97T TAT GTC

.G ...

T GTG GTA

T TGG GTG

Jx98 99 100 101 102 103 104 105 106 107

TTC GGA ACT GGG ACC MG GTC ACC GTC CTA JX1... ..G ... ... ... .G. ... ... ... ..GTTC GGC GGA GGG ACC MG CTG ACC GTC CTA JX2

C..GG..CG .. ... ...MG ..C ... ..GT ...

TTC GGC GGA GGG ACC AAG CTG ACC GTC CTA JA3. --.-.... -... ... ... .. *. ... .. . -... -...

VxVCodon

BCPBLL13

VX CDR389 90 91 92 93 94 95 95AGCC ATT TGG TAC AGC AGC ACT TCT

.. .. . . . . . . . . . . . . . . T

4-2 ATG ATT TGG CAC AGC AGC GCT TCTBCPBLL18. . . ... ... ... .. T ...

5c2 ATG ATT TGG CAC AGC AGC GCT TCTBCPBLL15 . .. .. C ... ... ... ..

BCPBLL14 ... ... ... ... ... ... ...

VxVICodonlv6S1BCPBLL16

VA CDR389 90 91 92 93 94 95 95ACAG TCT TAT GAT AGC AGC MT CA

..... . . . . . . . . . . . . . .

96 97T TGG GTG

... C..

Jk98 99 100 101 102 103 104 105 106 107TTC GGC GGA GGG ACC MG CTG ACC GTC CTA JX3... ... ... ... ... ... ... ... ... ..T

I T TGG GTG TTC GGC GGA GGG ACC MG CTG ACC GTC CTA JX3CA I . ... A.... ... ... ... ... ... ...

CCTG

CCTT

T TAT GTC

T TGG GTG... . ...

TTC GGA ACT GGG ACC MG GTC ACCTTC GGC GA. G.G ..ACMG CTG .ACTTC GGC GGA GGG ACC AAG CTG ACC* . . .*-- *-... -*... --... .. . . ...

GTC CTAGC..C.GTC CTA* - *. ..

Jx1

JX3

Jx96 97 98 99 100 101 102 103 104 105 106 107

T TAT GTC TTC GGA ACT GGG ACC MG GTC ACC GTC CTA JX1.

... - * -... - * - - * - -

9

9

9

9

10

Normal PBLVXIII VX CDR3Codon 89 90 91 92 93 94 95IGLV3S2 CAG GTG TGG GAC AGT AGT AGTITPBLL56 ... ... ... .. . .A .

CAG GTG TGG... ..A ...* -. ..-. ...

*. ... ...-*

*. ... ...-*

... . .. ...-

*. ... ...-*

95A 95B 95CIGAT CAT CC

GAT AGT AGT AGT GAT CAT CC... ... .C. ... ... ..

C..G... ...GM A.. ..G... G.. A. GAAA ....... G. CA . ... ....... G .

... G.A ..C.

... G.A ..C.

....... ......... ... .......

... ... ... ... ... .CG ... ... ...

CM TCA..G A..* . ...-

... A..

... ..G

* -. ...

GCA GAC AGC AGT* -. ...

* .-. ...

.. ...

* .-. ...

C.. ...

* .-. ...

..G ... ... ... .A.

... ... ... ... ..T

... ... ... ... ..T

... ... ... ... ..T

..... ... T.C .. A.

... ... A . ... ..

..... ... ATT M. A

... ... ..G ... .T.*-..*-.... -... .. . -..

GGT ACT TAT CC... G.G .

* .-. ...

.AG

.C.

* - ... .

.C. ..

.C. ..

.C. ..

... ... ... .

. .C . .

... ..C* - ..--. ..

I ...

* . .

TT .

G..* *

* *

* *

. .C

. .C..C

*..-

*. .

.C.*. .

*. - *. .. .. ...*-

..... ... TA. .

* - . .*. ... ...*-

* - ... ..--. ...

*. .. .. ...*--

* - . .*-. .. . ..*

*. - *. .. .. ...*-

... ... .T. ..G

P N P

GTAGG

A

Ta TAGGG

GGC a

t~t

TCTG

8

C c

CCGTA

a A

96 97T GTG GTA

...C

T TAT GTC

.T. . .T

.T. . .TT GTG GTA

..G

T TAT GTC

T GTG GTAA.C..T.T..T

..TA ..

T..

T TGG GTG

caa

Jx98 99 100 101 102 103 104 105 106 107

TTC GGC GGA GGG ACC MG CTG ACC GTC CTA JX2

TTC GGA ACT GGG ACC MG GTC ACC GTC CTA JX1

... ... .G.

... ... .G.TTC GGC GGA*. .....--*

*-.. *... * ..

*-.. *...*... ... ... ..-. .-..

*-.. *.. **. *.- *-. *-- *--

.. .T............. ... ... ... ... ... ...... ... ... ... ... ... ...

GGG ACC MG CTG ACC GTC CTA... ... ... ... G.. ... ...

........... ... ... ... ... ..... GA.

9

111112121111

JX21111

TTC GGA ACT GGG ACC AAG GTC ACC GTC CTA JA1...... . ... ...G............ 11... ... ... ... ... ... ... ......... 11.... ... ... ... ... .G...... ... ... ... G 9..T .C ............. 10.C .10... ... ... ... ... ... ... ... ... ... 10TTC GGC GGA GGG ACC MG CTG ACC GTC CTA JX2... ... ... ... ... ... ... ... ... ... 11... ... ... ... ... ... ... G.. ... ..........G10... ... ... ... ... ... ... G.. ... ..........G10........... ... ... ... ... ... ... G.. ... ...G 10... ... ... ... T.. ... ... ... ... .. 8.................. G. 10

.... ... A G.... ............ 10................. ... ... ... ... ... ... ... ... ... . .......11

... ... ... ... ... ... ............ .......................1

TTC GGC GGA GGG ACC AAG CTG ACC GTC CTA J)L3.............................. 10... ... ... ... ...... ... ... ... ... 13... ... ... ... ... ... ... ... ... ... 11... ... ... ... ... ... ... ... ... ... 11... ... ... ... ... ... ...:... ... ... 11.............................. 10... A . ... ... ... ... ... ... ... ... 10....................T G..G .. ..T 10

3a CTA TCA GCA GAC AGC AGT GGT ACT TAT CCCITPBLL86. T .

T TAT GTC TTC GGA ACT GGG ACC MG GTC ACC GTC CTA JA1. ... ... ... ... ... ... ... ..G . ... ... ... ..C .

Fig. 3. (Continued)

8

11

12

huntv3ldITPBLL1I TPBLL2I TPBLL28I TPBLL66LBPBLL2H*LBPBLL4D*

LBPBLL3I TPBLL96

hsiLL150LBPBLL3FITPBLL15I TPBLL45I TPBLL70LKPBLL72*LKPBLL76*

LBPBLL2LBPBLL3J*LBPBLL3K*LBPBLL3S*LBPBLL5ITPBLL1 1ITPBLL76I TPBLL80I TPBLL81

ITPBLL12ITPBLL13ITPBLL16ITPBLL79LKPBLL9LKPBLL70*LKPBLL73*LKPBLL59

12

. .A

*..A

* *

* *


3F TAC TCA ACA GAC AGC AGT GGT MT CAT AG ILKPBLL69 ... ...I............... ... .

Z212 TAC TCT GCG GCT GAC AAC MT CTLKPBLL68 ....... ... A........ ... IIII .1LBPBLL4HI TPBLL26I TPBLL97LKPBLL25LKPBLL84LKPBLL85LKPBLL88

I TPBLL22I TPBLL92LKPBLL83LKPBLL87

LBPBLL4FLBPBLL4SI TPBLL25I TPBLL95

vxiCodonhuLnLv142LBPBLL2N

CAG GCG TGG GAC AGC AGC ACT GCA... ... ... ... ... .C.

.. . . . . . . . . . . . . . .AG T. .

............. ..... G...AA...C C.... ... ... ... ... ..A.... G.

..... ... ... ... ... . T . T... ... ... ... ... ... ... .G.

... ... ... ... ...

... ... ... ... ...

... ... ... ... ..G

... ... ... ... .T.T.. ... ... ... ...

... ... ... ... ..T

.. .. ... .... .........

..T...

* .. ..

.CT ... .

.A.A . ... ..

.. ....-

T GTG GTATCTA . .....

TTC GGC GGA GGG ACC MG CTG ACC GTC CTA JX2

T GTG GTA TTC GGC GGA GGG ACC MG CTG ACC GTC CTA JX2TG . ... ................... T.. ...G.....G

TTTA

TMG

9

TGGA

Tt caa

Vk CDR389 90 91 92 93 94 95 95A 95B 95C1|CAG TCC TAT GAC AGC AGC CTG AGT GGT TC |

T .. A C I

CC I

VX CDR389 90 91 92 93 94 95 95AGAG ACC TGG GAC AGT AAC ACT CA... .... ... ... ... .. ...

§ ......................

T TAT GTC

T GTG GTA....C..

T TGG GTG

96 97T TAT GTC

T GTG GTA.. .C.

T TGG GTG

96 97T TGG GTG

.C . ..

*. ...*-

TTC GGA ACT GGG ACC MG GTC ACC GTC CTA JA1....... ...

... ... .A. ... ... ... ..B. ... ... ...

... . ...

... . ..-

... . ...

... . ...

... . ...

TTC GGC.. .- .-.

... ...

... ...

*... .-..

TTC GGC... ...

... ...

*. ...*-

*. ...*-

... . ...

... . ...

*. ...*-

... . ...

GGA GGG

... ...

*. ...*-... ...

GGA GGG... ...

... ...

... ...

... . ... .. ... ... ... ...

... . ....... .:.ACC MG CTG ACC GTC CTA J02... ... ... ... ... ...

... ... ... ... ... ...

... ... ... ... ... ...

...... ......... ... AACC AAG CTG ACC GTC CTA J03... ... ... ... ... ...

... ... ... ... ... ...

... C.A .. G....*.-* ... ... ... ... ...-*

Jx98 99 100 101 102 103 104 105 106 107TTC GGA ACT GGG ACC MG GTC ACC GTC CTA JX1... ... ... ... ... ... ... ... ... ...

12

9

9109

119109

9999

991010

12

TTC GGC GGA GGG ACC MG CTG ACC GTC CTA JX2.............. ... ... ... ... ... ... ... ........ 12

TTC GGC GGA GGG ACC MG CTG ACC GTC CTA JX3.............................. ... ... ... ... ... ... ... ... ... 11

JX98 99 100 101 102 103 104 105 106 107TTC GGC GGA GGG ACC MG CTG ACC GTC CTA JX3...... ............ ...........

... . ....... .. ... ... .. ... ... .. ... ... .. ..

99.3

8At CAG ACC TGG GGC ACT GGC ATT CALBPBLL2F ... ... ... . C . ... ..

Nv801 CAG ACC TGG GGC ACT GGC ATT CAITPBLL68 .T..

vxv

Codon

ITPBLL69

VA CDR389 90 91 92 93 94 95 95A

GCC ATT TGG TAC AGC AGC ACT TCT.. G... .. GA.

I| IT GTG GTA TTC GGC GGA GGG ACC MG CTG ACC GTC CTA JX2a AGT I ... ... ...........................

T TGG GTG TTC GGC GGA GGG ACC AAG CTG ACC GTC CTA JX3. .I. ... .... .... .... .... .... .... .... ......... ... ...

Jx

1 96 97 98 99 100 101 102 103 104 105 106 107T TAT GTC TTC GGA ACT GGG ACC AAG GTC ACC GTC CTA JX1. ... ... ... ... ... ... .A . G ... ... ...

| T TAT GTC TTC GGA ACT GGG ACC AAG GTC ACC GTC CTA JA1I .. ....... ... ... ... ... ... ... ... ... ...

Igt

j.Q ATG ATT TGG CAC AGC AGC GCT TCTLBPBLL7 C. A

5c2 ATG ATT TGG CAC AGC AGC GCT TCTLKPBLL52 ..................

LKPBLL53 § ...... A..

ITPBLL4 §.C C.

vxVI

Codontv6SlI TPBLL75

TAT GTC... ...

GTG GTA

TGG GTG* -. .-..

T

T

TT

Vk CDR389 90 91 92 93 94 95 95A

CAG TCT TAT GAT AGC AGC AAT CA............ .G.. ..G

LBPBL2S . C. ... C. C . ... ..

LBPBLL4 ....... A.... G.

96 97T TAT GTC. *... ...

T TGG GTG

. C..

9

TTC GGA ACT GGG ACC MG GTC ACC GTC CTA JX1............ ... ... ... ... ... ... ... ........ 9

TTC GGC GGA GGG ACC AAG CTG ACC GTC CTA JX2........... ... ... ... ... ... ... ... ... ... 4.3

TTC GGC GGA GGG ACC AAG CTG ACC GTC CTA JX3.................. ... ... ... ... ... ... ... ... ... 10.3

Jx

98 99 100 101 102 103 104 105 106 107TTC GGA ACT GGG ACC AAG GTC ACC GTC CTA J01... .A. ... .C. ... ... ... ... ... ...

TTC GGC GGA GGG ACC MG CTG ACC GTC CTA JX3... ... ... ... ... .GA ..C ... C. ...

... . ....... .. ... ... .. ... ... .. ... ... .. ..

10

99.3

Fig. 3. (Continued)

ously been reported to be identical in nucleo-tide sequence (30,50,53). In this study, a totalof 158 clones contained identifiable JA gene

segments (Table IC, Fig. 4). Surprisingly, a

variant JA3 gene segment (21), (Genbank Ac-cession Number D87023) was found in 41cDNA clones (26%) and was expressed in allfive individuals in this study. The definitive

LKPBLL58 ... ... ... ... ... ... ... ... ... ..

LKPBLL2 ... ... ... ... ... ... ... ... ...

VxIVCodon

LBPBL2QLKPBLL10

II

I

A


Table lB. Germline derivation of unrearranged VA gene segments amplified from genomic DNA ofrheumatoid arthritis patient BC and normal control LB

Vk Famil1y-PBL GSen

BCLBTotal.............................. ...

150lSO 3a 13,p3 0 32 1 35 1 6

All but two genoniic sequences in Table 1B were 100% identical to reported sequences of corresponding germline gene segments(Genbank Accession numbers: LBLG9-AF063764; LBLG14; AP063763). Mismatches in number of dones analyzed in Tables IAand lB are attributable to truncated sequences.

evidence of expression of what is presumed to for cDNA amplification was derived from abe an allelic variation of the JA3 gene segment conserved region of the constant region, JAsupports our contention that shared changes in gene segment utilization in different samplesVA gene segments represent novel allelic vari- could be compared without concern for biasants. because of preferential JA amplification. The

Because the downstream PCR primer used pattern of JA expression was similar in the

Table IC. Germline derivation of JA gene segments in cDNA clones

NormalIT PBL LK PBL Total

12 (42.8%). 7 (36:8%) 59 (37.3%)8 (28.6%) 6 (316%) ;58 (36.7%/6)8:(28.6%) 6 31.6%) 41(26.0%)

28 19 158

Table 2. Possible variants of known germline VA gene segments expressed in cDNA clones frompatients with rheumatoid arthritis and normal controls

~Gene Reported VariantSegment Codon Domain Sequence SequenceIGLV3S2 92 CDR3 GAC jGAT

Total1617l33 :

Gene SementJA1

TOtal


ASfyn lfi,yn ASPflL I5CPBL L&rFUL LDrDL llrBL NorM21'

Rheumatoid Arthritis Normal Controls

three normal PBL samples (JAI 33-43%, JA229-39%, JA3 28-32%, Table IC) and closelyparalleled that seen in cDNA libraries from PBLof normal individuals (47) (Fig. 4). In RA tis-sues, however, there was more variability in JAgene segment utilization. For example, in sy-novium of patient AS, there was a slightlydecreased representation of JAl and an in-creased proportion of JA3, whereas AS PBLcontained more JA 1- and JA2-containingclones and fewer JA3-containing clones.Clones from BC synovium exhibited a markedincrease in JA2 and a concomitant decrease in

II dI

* 2

P3

Fig. 4. JA gene segmentutilization in RA synovia,RA PBL, and normal PBL.*, Utilization in lambda lightchains from normal PBL(47).

PBL showed a marked in-JA3, whereas BC

crease in JA 1.

Comparison of N-region Addition in VA and VKTranscripts from the Same Tissue Samples

The percentage of clones containing P nucleotidesand N-region addition and the mean number ofnudeotides added (among dones that contain N-region addition) are shown in Table 3. Lambdalight chain data are shown in comparison to a

similar analysis of predominantly VKIII transcriptsfrom the same tissue samples [(16), Fig. 5]. P nu-

Table 3. N region and CDR3 lengths of immunoglobulin VA and Vic clones from RA synovia and PBLand normal PBL

~ VK

Clones P N Mean Mean CDR3

1YZed Addition Addition No. N's Legth_11 27.2.% 54.5% 1.7t±1.2 9.8±1.010 20.0% 60.0% 1.7± 1.2 9.9±0.9

14 35.7% 57.1% 1.8 ±1.0 9.5±1.215 33.3% 60.0% 1.9±2.3 9.3± 1.012 25.0% 16.7% .0±2.8 9.1±0.511 18.1% 45.5% 3.8±4.1 9.5±0.711 18.1% 18.1% 2.0±1.4 9.0± 0.6

VK data are derived from (16).P addition and N addition refer to the percentage of clones containing at least one palindromic nucleotide or nucleotide of N-region addition, respectively. Mean number of nucleotides and CDR3 lengths (expressed as number of amino acid codons) areshown ± one standard deviation.* The mean number of nucleotides refers to the average number of nucleotides added among the clones that contain N-regionaddition.

U)

Cd

100-

90-

80-

70-

60-

50-

40-

30-

20-

10-

0-

TissueSampleAS SynAS PBLBC SynBC PBLLB PBLLK PBLIT PBL

11 i


Fig. 5. Mean percent homology of ex-pressed VA gene segments to corre-sponding germline gene segments. Thisanalysis includes only clones that were mu-tated; clones with VA genes identical togermline were excluded. In general, clonesfrom RA synovia (black bars) and RA PBL(gray bars) were more highly mutated(lower homology) than clones from normalPBL.

cleotides were present in 10-33% of VA transcriptsand 18-36% of VK transcripts. There was no dif-ference between proportions of VA clones with Pnucleotides among RA tissues (10-33% of clones)and clones from normal PBL (16-32%). N-regionaddition in VA clones was surprisingly common innormal PBL (range 26-43%) but was more prev-alent among clones from RA synovia (43-50%)and RA PBL (40-45%) (Fig. 6).

In all RA samples, there was more N additionin VK transcripts than in VA transcripts. Surpris-ingly, in two of the three normal individualsstudied, N regions were more common amongVA transcripts than in VK transcripts. Amongclones with N-region addition the average num-

60

0:o 50

., 40

zd 30

4 40

4)

0

n4

0' 3

04) 20 f

I v

ber of nucleotides added ranged from 1.9 to 3.3(Table 3). One of the AS synovial clones (AS-SynL2 5) contained 25 nucleotides of junctionalsequence, which may represent N-region addi-tion or insertion of nucleotides through anothermechanism (see Discussion). In both RA patientsthere was more N-region addition in VA tran-scripts from synovia than in those from PBL.

CDR3 Lengths of Lambda Light Chains from RASynovia, RA PBL, and Normal PBL

CDR3 intervals were generally longer in VAtranscripts (range 8-16 codons) than in VKtranscripts from the same tissue samples (range

2K

Fig. 6. N-region addition inVA-JA joins (black bars) and VK-JKjoins (white bars) from RA syno-via, RA PBL, and normal PBL. VKclones are from ref. 16 and derive pri-marily from members of the VKIIIrC !1- _ -r Txr _ 1l, family. Percentages of VK clones con-

ITPBL taining N-region addition vary slightlyfrom ref. 16 because of reanalysis toidentify P nucleotides.

._4)

:

4)

w

r-

4)

4.)

bObJo

0

4)

0

0

0

uI _ I... .. 1

ASSyn ASPBL BCSyn BCPBL LBPBL LKPBLRheumatoid Arthritis Nonnal

541

'IT, 'Itp I"10. oto, e,0., le, *40,,4 .el% .Iel .% e tl6e


ASSyn ASPBL BCSyn

Rheumatoid Arthritis

BCPBL LKPBL LBPBLI _

P D3 U.) U.) v% t-A

P.-~~-00 oE

t'

VX gene segment

7-1 1 codons) [(16), Table 3, Fig. 7A]. As was

the case for kappa light chains (16), VA tran-scripts from RA synovium and PBL had a

higher mean CDR3 length and contained a

higher proportion of long CDR3 intervals thandid those from normal PBL (p = 0.02, Studentt-test, two-tailed). CDR3 intervals of 11 or

more codons were found in .50% of tran-scripts from each of the four RA samples. Incontrast, CDR3 lengths of .1 1 codons were

ITPBL

Normal Controls

.- 010-.4 $000 -

Q0-

Fig. 7. (A) CDR3 lengthsof VA clones from RA sy-novia, RA PBL, and nor-mal PBL. The proportions ofclones with CDR3 intervalsof 11 codons (gray seg-ments), 12 codons (blacksegments) and >12 codons(white bars) are shown. Pro-portions of clones with <10codons are not represented.(B) Mean CDR3 lengths ofclones according to thelength of the germline com-ponent of the VA gene seg-ment utilized. Germline genesegments are grouped intothose with 29 nucleotides(white bars), 24 nucleotides(gray bars), and 23 nucleo-tides (black bars). Cloneswith out-of-frame CDR3 in-tervals (see Fig. 3) were notincluded.

found in less than half of transcripts from eachof the normal PBL samples (Fig. 7A).

Evidence of Oligoclonal Expansion of VA-expressingB Cells in RA Patients and Normal Controls

Molecular characteristics of antigen-driven, oli-goclonal B lymphocyte expansion include thepresence of clones utilizing the same V and Jgene segments with shared nucleotide changes,

A

co0)

00)

u)I-,

41

0)Bo

B

12

0&O

0

04-0

0I-0)

~0

CA,cd

10

0U,

U)ao

0v

c)au

8

I


similar N-region addition, and identical CDR3lengths. Several sets of sequences that may rep-resent the products of oligoclonal B-cell expan-sion were found among clones from RA patientAS synovium and PBL. Clones ASSynL182 and189 are very mutated (92.3% and 92.3% homol-ogy to germline IGLV3S2) and have multipleshared differences (Fig. 8A). Both clones use theJA2 gene segment and have the same CDR3length.

The VA gene segments from clones AS-SynL211, 209, and 184 are derived fromhsiglllS0 (96.6%, 91.6%, and 89.7% homology,respectively) (Fig. 8B). These three clones haveidentical VA-JA joins and CDR3 lengths. Impor-tantly, clones ASSynL211 and 209 have a sharedsomatic mutation at codon 104 in the JA3 genesegment. Because all the germline JA gene seg-ments are presumably known, this shared muta-tion is strong evidence of clonal relatedness. AS-PBLL48 and 51 are also likely to be clonallyrelated (Fig. 8C). They are heavily mutated in theVA gene segment (90.8% and 91.2% homologyto hsigglll50, respectively), have identical VA-JAjoins, and use the same heavily mutated JAIgene segment. There are five nucleotide differ-ences between ASPBLL48 and ASPBLL51, threeof which lead to amino acid replacement. PatientBC also had evidence of clonal expansion in sy-novium and PBL. Clones BCSynL47 and 51 areheavily mutated and have shared differences atcodons 28, 30, 77, and 93 (Fig. 8D); BCPBLL2and 4 have shared changes at codons 50, 95A,and 96 (Fig. 8E). Among the clones from normalPBL, only individual LK demonstrated convinc-ing evidence of clonal expansion. Clones LKP-BLL76 and 72 are 96.2% and 95.0% homolo-gous to hsigglll 50, use the same JA genesegment, and have identical VA-JA joins(Fig. 8F). All but seven differences are sharedbetween the two clones.

Levels of Somatic Hypermutation in RA Synovia andPBL and Normal PBL Transcripts Suggest theSubset ofB Cells from Which They Are DerivedVA gene segments expressed in RA synovia were,on average, slightly more mutated than thosefrom RA PBL, which were in turn more mutatedthan those from normal controls (Fig. 5). Themean percent homologies to germline nucleotidesequence were RA Syn 94.4%, RA PBL 95.4%,and normal PBL 96.7% (p = 0.002, ANOVA).Nine of 67 clones (14.9%) from normal individ-uals were unmutated, compared with only 3 of

94 (3.2%) of clones from RA samples (p = 0.03,Fisher exact, two-tailed). When unmutated se-quences were omitted from the analysis, the dif-ferences in amount of somatic mutation werestill apparent (RA Syn 94.2%, RA PBL 95.2%,and normal PBL 96.1%, p = 0.02, ANOVA).

In this cDNA analysis, quantitative differ-ences in immunoglobulin mRNA expressionamong B cells at various stages of differentiationprobably explain discrepant levels of somaticmutation between RA synovia and RA PBL. Sy-novial infiltrates from patients with long-stand-ing RA typically contain many plasma cells(54,55), which have been shown to have highimmunoglobulin mRNA levels (56). Class-switched and IgM+/IgD memory B cells havealso been reported to express higher levels ofimmunoglobulin mRNA than naive B cells (57).Marked levels of somatic mutation were noted inour previous analyses of the synovial IgG VHrepertoire of RA patient BC (mean homology togermline 91%) (58) and the synovial VK reper-toires of RA patients BC (mean homology94.9%) and AS (mean homology 95.8%) (16).Thus, the increased somatic mutation in lambdalight chains from RA synovia is likely attributableto the presence of plasma cells, which expresshigh levels of mutated sequences. There werefewer unmutated sequences from RA PBL thanfrom normal PBL. Thus, it seems likely that moreRA PBL sequences originated from memory Bcells (either class-switched or IgM+/IgD mem-ory B cells) than from the essentially unmutatednaive B cell (IgM+/IgD+) population. In contrast,more of the normal PBL sequences may havebeen derived from naive B cells.

Potential PCR ArtifactsWe used sequence comparison software that iscapable of comparing individual FR and CDRdomains with corresponding domains from mul-tiple germline gene segments. Thus, we wereable to identify sequences with crossover eventsinvolving two different gene segments. Wefound only one such crossover artifact. BC-SynL32 is a clone that contains a crossover arti-fact, with the FRI and CDR1 domains from theVAVI gene segment 1v6Sl and FR2, CDR2, FR3,and CDR3 from the VAI gene segmenthumlvlO42 (DPL8) (Fig. 2). This so-called jump-ing PCR artifact occurs when an incompleteproduct of one amplification cycle serves as aprimer for a related sequence, generating a chi-meric molecule (59-61).


leU,

04

04P:

Cl4

hi

'.4

0

'.4p4he

0m40

co40

%a4e

in4f

0hen

co4W

11.4W

enin

Cl4en

'.44W

C4Wi

0'Cl4

GoCl4

C..cl

'Coll

C4C.-

'.4

1) 4 .U

4..4gV0U

6;:.

0 .46::

* IU0* 0

ii4..

4 -[) E0

A * I

0.

0..

a * @

E4 .()

cS<

cs * :

144

P3Ac

¶14

Inin

4W

Ch0'at^

tn

0111%

%a

PC

C'.in

U

14.

0 v

1401

0 E414 .

041 4.0.o

P4hi

0

P4

0

0*V144.

01

VJ E4I :S @

4W

U,

Ul,UI

'Ul0U'

0

0s

No

49

C"

'CI4W

C4W

C

0

04

41hi

4W

'.4ICl4

clj'. '. Clj'4 .4ab a ab

AI0 1 0

N4~0 1-14 0.H 14 4

* * .D* * 0

0* * V

0 .40

b.40* *U * .:* *

E4 ..

t) * *C)o * E@

.)o * *0* *

0

0:1:0 .146 : * * f

4. .E.14...s

4E..Es . . .

4 * 1 1

g::1

rl

004

E014004a

141

14* V

. E4* 0

*4

*1

*E4

crl

Inc

enrl

12c14W

C

n

P:

a

0-'4

In0

64

0In

a

Ul

00

0

Ch

Cl%co

hiz~co

04

C"PC

4 .0.V ~* *V0E.*'E0 *1.U..*4...

E

v0. .

0.*.*014**

14 .1v

00.**14* *v4..*o*-14..*..o * * o~~C

v .~ . 0

14. ..

g - i:: v

ri~

C)* * 64q

a : :e:**@~ ~~c

0E41 E41141

A ; * *

0E .14d V :

0 * Es .4..0

I *0*

0 * *4...

0 .0.

0 . 4

04 4*4...*0...<

InCl4

'.4

a a

w w o woGo oH 010~ 044

4_..ig :,:04-~

14_. _.

14-1 : :Ee *

*TM

UV0 ..

0.

H In qv'.4A'.1 la

AiPivs go2

'4

Cn

HC1'.

u4CliCC

M..E)

*

114.

U..C * 0

9 * :4-

*0v-* 0

0..

E4.

14..*1e4

C44W

4W

04

04

Cl

Cl4

,.1P:0

IfCl4

4v04

E0 .10 E4

o *~E[-I

*

F *14..aAcb : .

vdA. .

E;*

0 '

0aV . .

0 0

0..

04 '0 *4

C; *

0 .4

0s..14..<40C14..

Clqclq

Cl

en

'4

00

an4in E4V P1

Go*

C411

14-

0 .0

M14 1

VW...U

0.o *

InEI*'.41.4

0 44

ri

PC

ml

o

r- q i

a Av«

S. L. Bridges, Jr.: Inmunoglobulin VA Expression in Rheumatoid Arthritis 545

0 * 1E4

o* b

14..e *14

14~

0-4

4..

04~

40

U0. .

0) .4.4.

4.e

0 .144 *E40 .44..

U4*00s *40 E

0 *140 .0.0..

00VV4 @14 *0 '4o4..

0s.~014*

ain

,r,I14 n

,-i IaA

oo VI to

A m m

00 . a

o 4E .* V

E0* o%D Ez *

E4

0%4

14

14

6-I

6-I

Cl MQ

abr04c

0%

%00%

mUna%

4Un0%

inai

P:V40

460%

Mn0%

Cl40%

6-i0%

a0%

0%Go

coco

4

0E-4014

1-414

E4001-40E14

0

*q 0

0 0

14.

. E414.14<* 0

00

.0 .

0 .04..

B

0 .4A

1 4 .V * 0

8 * 9

It

U:0^

64If

(MIf

0

crz. E) .0

ch Ev * :

E- 1

14..

m

0

'-INv

0

inI

.-I Re U-

v- la 14tt4il-r OOVto J (

4ina%

In4m

40at)

V * -

0.'~E * 40 .44. .

V *l

q * 41; .-

.64

.4

4 01-4

00o 0

1E4

H N 0P

W HMMvv

E14

141404E140040 0

at .n 4..

6 04OIII

04.

V i144 ..

0o4 hJEHhqr E4 .

4..

0UI

r

11 I

. 4.

av I 3 a

0

inI

u-Il 04

to W 64

P6 A141

04

lb4

C..co

%00D

0

0

U,

0co

C.'

C..

%,1-

04C..

Cli

0,tC

U,SU

Ln,

U,

N

04

0

'0

t4I'

04

Ik 4

C4v

co.cf

Mn

Ike

.000Vr 0'0

I

040 *

E144

V *

14.

4; .

E4.90.

14

0 .

00

000 E.

V0.0V4 a

4 .

0 4V

94

0 :

14E-i -

14

14

4

0

14

1-'C

'C'C

'0'C

04'C

C'C

0%Un

0inIr- r- U)4u-II6.4IA

AUXCa

040

X Ln : :

0 .M

r-I .0 0 . . r-I cnE-1 - C; : :

O C) 0 19 I.MPC, 04


Although our sequence analysis softwarecan easily identify a hybrid molecule containingtwo different germline VA gene segments, iden-tification of crossovers between two transcriptsexpressing the same VA gene segment representsa more difficult problem. In all clones thought tobe genealogically related (Fig. 8), there werenonshared differences in multiple FR and CDRdomains. To explain these findings by jumpingPCR artifact, one would have to posit that mul-tiple, closely spaced crossover occurred. This isunlikely from a statistical point of view becauseof the relatively small number of PCR cycles usedfor amplification.

Because a nonproofreading DNA polymer-ase (Taq DNA polymerase) was used in PCR,we were concerned about the possibility ofPCR errors. Estimates of the fidelity of TaqDNA polymerase vary from 2.0 x 10i5 errorsper base pair (62) to 1.1 x 10-4 errors per basepair (63,64); 2.4 X 10-5 frameshift mutationsper base pair are estimated to occur in Taq DNApolymerase PCR reactions (63). The sequenceanalysis of unrearranged VA gene segmentsfrom T lymphocytes, in which somatic hyper-mutation of immunoglobulin genes does notoccur, allowed us to estimate the Taq errorrate. From a total of 9472 base pairs that weresequenced, five single base pair substitutions(and no frameshift mutations) were found. Ifone assumes that all five of these nucleotidechanges were due to Taq error, the maximumerror rate is 5.3 X 10-4 error changes per basepair. Thus, in the reported cDNA sequences,the error rate is presumably very low (approx-imately one base pair error in every third se-quence).

DiscussionN-region Addition during Immunoglobulin LambdaLight Chain Gene Rearrangement

The finding of N-region addition in VA-JAjoins from RA patients and normal individualshas important implications regarding the gen-eration of CDR3 diversity during immunoglob-ulin gene rearrangement in normal B cells. Inhuman bone marrow, stem cells progressthrough the pro-B-cell, pre-B-cell, and imma-ture B-cell stages of differentiation before be-coming mature B cells (reviewed in ref. 65). Inthe pro-B-cell stage, the immunoglobulinheavy chain locus undergoes rearrangementand the enzyme responsible for N-region addi-

tion, TdT, is present in the nucleus of the cell.In the pre-B-cell stage, kappa light chains un-dergo rearrangement. Although TdT proteinhas not been shown to be present at the time ofkappa light chain rearrangement, N-region ad-dition is clearly present in kappa light chainsfrom normal individuals and those with RA(15-18).

The current theory for immunoglobulinlight chain rearrangement holds that if success-ful kappa rearrangement fails on both alleles,then lambda light chain rearrangement occurs.Thus, lambda light chain rearrangement isthought to occur well after the expression ofTdT. Our analysis of VA-JA joins, however, pro-vides clear evidence of N-region addition, pre-sumably due to TdT activity at the time oflambda light chain rearrangement. Surpris-ingly, in two of the three normal individualsstudied, there was N-region addition in ahigher proportion of VA transcripts than VKtranscripts. Thus, our data provide support forthe hypothesis that TdT is present throughoutall stages of immunoglobulin gene rearrange-ments in B cells from normal individuals. Wealso found more N-region addition in lambdalight chains from patients with RA than inthose from normal individuals. The same istrue of the kappa repertoire of RA patients(Fig. 6) (16), which suggests that TdT may beabnormally regulated in RA.

There are several potential explanationsfor the presence of non-germline-encoded nu-cleotides at the V-J junctions of lambda lightchains. One possibility is that lambda lightchain rearrangement occurs earlier than waspreviously thought, such as before heavy chainrearrangement or after heavy chain rearrange-ment but before kappa light chain rearrange-ment. Kubagawa et al. documented pro-B-celllines that had undergone kappa rearrange-ments at one or both alleles but containedheavy chain loci without VH-DH-JH rearrange-ments (66). Some of the kappa light chainrearrangements in these kappa-only cell linesdo contain N regions (67).

An alternative explanation for N-region ad-dition in A chains is that the current theory ofrearrangement order (heavy chain, then kappalight chain, then lambda light chain) is correct,but that TdT is expressed in B cells through themature B cells stage. Support for this finding liesin the description of a small subpopulation ofpre-B cells expressing both nuclear TdT and cy-toplasmic ,u heavy chains (68).


Allelic Variations in the Human ImmunoglobulinLambda Locus

In addition to providing another mechanism forantibody diversity, the presence of allelic variantsin the human lambda locus has important impli-cations regarding the analyses of somatic hy-permutation and assessment of clonality. Theamounts of somatic hypermutation in this studywere calculated with the assumption that cloneswere derived from known VA gene segments.Each expressed VA gene segment was assigned apresumed progenitor gene segment based onhighest degree of homology. Although the hu-man VA locus has been sequenced by two groupsof investigators (20,21), there may be unre-ported variants of VA gene segments. The levelsof somatic mutation and the degree of oligo-clonality reported here may be overestimated ifthere are VA gene segments or additional unre-ported allelic variations of previously describedgermline gene segments.

Kawasaki et al. assessed the degree of allelicvariation of the human immunoglobulin lambdalocus by comparing nucleotide differences be-tween their sequence and two cosmid clonesfrom different sources (one from a Caucasianindividual and another from a Japanese individ-ual) (21). They found 100 nucleotides over aspan of 39,236 nucleotides that were thought torepresent allelic variations. Although the num-ber of differences located within coding regionswas not reported, there appear to be relativelyfrequent single nucleotide variations in the hu-man immunoglobulin lambda locus. Thus, addi-tional novel allelic variants are likely to be foundin future studies of lambda light chain expres-sion.

Nucleotide Insertions and Deletions in VA GeneSegments May Occur during Somatic Hypermutationin a Germinal Center (GC) Reaction

Single or multiple nucleotide insertions anddeletions have been observed in heavy chainvariable (VH) domains from germinal center(GC) B cells. Using a single-cell PCR approach,Goossens et al. sequenced rearranged VH do-mains from naive and GC B cells (69). Nodeletions or insertions were found in naive Bcells, but -4% of in-frame and >40% of out-of-frame (OOF) rearrangements of GC B cellshad deletions and/or insertions of variablelength. Wilson et al. also noted insertions anddeletions in somatically mutated VH gene seg-ments from CD38+/IgD GC B cells and

CD38 /IgD memory B cells from humantonsil (70). Eight of nine insertions/deletionswere present in the CDR1 or CDR2 domains.In our previous analysis of 18 immunoglobu-lin VH gene segments from RA synovium,we found one highly mutated VH4-contain-ing transcript with a two-codon insertion inCDRI (71).

In the present study, we found two lambdalight chain clones with insertions or deletionsthat may be the result of somatic hypermuta-tion in the GC reaction. ASSynL191 has a one-codon deletion in CDR1 and LKPBLL68 con-tains a one-codon insertion in CDRI (Fig. 2).Both clones are somatically mutated: AS-SynL191 is 95.2% homologous to its progeni-tor germline gene segment, while LKPBLL68shares 92.3% homology to its germline genesegment. Several clones in the present studywere found to have single base pair insertionsor deletions that may reflect somatic mutation(69) or alternatively, frameshift artifacts in-duced by Taq DNA polymerase. Our data sug-gest that insertions and deletions occur as aresult of somatic hypermutation not only inheavy chains but in lambda light chains aswell.

CDR3 Intervals of Lambda Light Chains Are Longerthan Those of Kappa Light Chains and the LengthsAre Heavily Influenced by VA Gene SegmentUtilizationIn both RA patients and normal individuals,lambda CDR3 intervals were longer, on aver-age, than CDR3 intervals from the kappa lightchains of the same tissue samples. In immuno-globulin light chain rearrangements, the num-ber of nucleotides at the VL-JL join and thus theCDR3 length are dependent upon the amountof exonuclease activity at the V and J terminiof the join and the number of N nucleotidesinserted by TdT. Germline JK and JA gene seg-ments typically contain 7 nucleotides in theregion that encodes the CDR3 (16). VK genesegments have CDR3 components that are re-markably constant in length (23 nucleotides).In contrast to VK gene segments, there is con-siderable variation in the number of nucleo-tides in the CDR3 components of germline VAgene segments. Gene segments IGLV3S2, 2-14,humlv3I8, hsigglllSO, 3p, 3a, humlvlO42, andDPL6 contain 29 nucleotides in the germlineCDR3 component. Gene segments III.1, 4-2,hsiglv5a, hsiglv5cl, and hsiglv5c2 have 24 nu-

547


cleotides in the CDR3 component, while genesegments 2-19, 8A1, lv801, 6sl, and 4a have23 nucleotides.

In the present study, the 96 clones contain-ing VA gene segments with 29 nucleotides inthe germline CDR3 component had a meanCDR3 length of 10.9 codons, the 41 clonescontaining VA gene segments with germlineCDR3 component of 24 nucleotides had amean CDR3 length of 9.5 codons, and the 16clones with VA gene segments containing 23nucleotides had a mean CDR3 length of 9.4codons (p < 0.0001, ANOVA) (Fig. 7B). Thus,in contrast to the situation in the kappa rep-ertoire, the germline VA gene segment usedin the rearrangement has a strong influenceon the length of the expressed cDNA CDR3interval.

Enrichment for Unusually Long CDR3 Intervals inBoth Lambda and Kappa Light Chains in RAPatientsCompared with normal PBL, RA synovia andPBL had higher proportions of clones with longCDR3 intervals. This difference is explained inpart by a higher proportion of VA gene seg-ments with long CDR3 components amongthe RA clones. Gene segments IGLV3S2, 2-14,humlv318, hsigglllS0, 3p, 3a, humlvlO42, andDPL6, which contain 29 nucleotides in thegermline CDR3 component were found in64.8% of RA clones, versus 57.6% of normalcontrols. Another contributing factor to en-richment of longer CDR3 intervals in RA is ahigher proportion of clones with N-region ad-dition.

The enrichment of longer CDR3 intervals inboth the kappa and lambda light chain reper-toires of RA patients suggests either positive an-tigenic selection of unusual B cells or abnormalimmunoglobulin gene rearrangement. These ex-planations are not mutually exclusive, and bothmay occur to some degree in RA patients. In thefirst scenario, lambda light chains expressing un-usually long CDR3 intervals are generated inboth normal individuals and RA patients. Subse-quently, normal individuals show no preferentialpositive selection of B cells that by chance con-tain kappa or lambda light chains with unusualCDR3 intervals. In contrast, RA patients exhibitpositive antigenic selection and clonal outgrowthof such B cells, possibly as a result of the chronicinflammatory response. Support for this hypoth-esis is the finding that alterations in the sequence

at VK-JK joins can abrogate antigen binding ofthe antibody (72).

In addition, there is substantial evidencethat RA is an antigen-driven disease(17,24,73-77). We have previously reportedoligoclonal B-cell expansion in the immuno-globulin heavy chain (58,78) and kappa lightchain (27) repertoires of synovial tissue of RApatient BC, which suggests an antigen-drivenB-cell response. Furthermore, the clonally ex-panded kappa light chain clone contained anunusually long CDR3 interval (27). In thepresent study, we found sets of genealogicallyrelated sequences in the lambda repertoire ofRA patients BC and AS. Oligoclonality of thesynovial B-cell population thus appears to be acommon feature of RA synovium.

An alternative explanation for the unusualVA-JA joins in RA is that there may be differ-ences in the regulation of expression of pro-teins involved in the rearrangement process,such as RAG-1, RAG-2, and TdT, resulting inabnormal development of the naive B-cell rep-ertoire in bone marrow. Immunoglobulinheavy and light chain gene segments undergoa highly regulated series of V(D)J gene seg-ment rearrangements to generate a functionalantibody. V(D)J recombination requires a RSS(composed of a heptamer, 12- or 23-base pairspacer and a nonamer), RAG-1, and RAG-2(79). Formation of a double-strand break at aRSS occurs during V(D)J recombination, withthe cleavage reaction divided into two distinctsteps: nicking and hairpin formation. A nick isintroduced at the 5'-end of the signal se-quence, followed by breakage of the otherDNA strand, resulting in a hairpin structure atthe coding end and a blunt signal end (79). ADNA-binding protein complex containing Ku7O,Ku8O subunits, RAG-1, RAG-2, and DNA ligaseperforms V(D)J joining (80).

Clone ASSynL25 is of particular interest withregard to possible dysregulation of immunoglob-ulin gene segment rearrangement in RA. TheVA-JA join of this clone contains 25 nucleotidesthat do not appear to be encoded by the VAIII. 1or JA2 sequence, resulting in a CDR3 interval of16 amino acid codons (Fig. 3). One possible ex-planation for this extremely long VA-JA join isN-region addition, but a more likely explanationfor the presence of this 25 base pair insertion isthe presence of an insertion that is duplicatedfrom the VAII. 1 gene segment. Sixteen of the 25nucleotides at the VA-JA join are identical to thesequence in codons 90 to 95A of the germline


III. 1 gene segment, shown as underlined nucle-otides in Figure 3A. Thus, there is evidence ofunusual immunoglobulin rearrangement in RAsynovium.

Expression of RAG-I and RAG-2 and ReceptorEditing in B Cells in Normal Secondary LymphoidOrgans: Implications for RA and Other ChronicInflammatory Diseases

Among the mechanisms of avoiding productionof autoreactive sequences are clonal deletion,clonal anergy, and receptor editing. Receptor ed-iting involves changing an autoreactive variableregion to a different (presumably nonautoreac-tive) variable region through a secondary rear-rangement (81-83). In order for this to occur,the V(D)J recombination machinery must bemaintained or reactivated. RAG-1 and RAG-2were initially thought to be expressed in B cellsonly during B-cell development in bone marrow.However, RAG- I and RAG-2 have recently beenfound to be expressed in germinal center (GC) Bcells of mice, with evidence of secondary rear-rangements and receptor editing (84-86).

There is circumstantial evidence suggestingthat RA synovia may act as a lymphoid organ.Nodular lymphocytic infiltrates that are histolog-ically similar to normal GC have been found inthe subsynovial layer of some patients with RA(87-90), Lyme disease (91), and reactive arthritis(92), and in inflamed liver tissue from patientswith chronic hepatitis B and C (93-95). We havedocumented the presence of clonally related im-munoglobulin heavy and kappa light chain se-quences in RA synovium, which suggests in situantigenic selection (27,58,78). Berek and col-leagues have reported clonally related sequencesfrom within GC-like structures in RA synovia(77). The histologic similarities between RA sy-novia and normal lymphoid tissues containingGC, and the recent findings of RAG- 1 and RAG-2expression in a subset of normal GC B cells sug-gest that RAG- I and RAG-2 may be expressed inRA synovia. If so, secondary rearrangementsmay be possible in synovium of patients withlong-standing RA.

In summary, these data demonstrate thatin humans, N-region addition enhances anti-body diversity at all stages of immunoglobu-lin gene rearrangement. The structural diver-sity of lambda light chain CDR3 intervals isgreater than that of kappa light chains, largelydue to variability in the lengths of germline VAgene segments. As in the kappa light chain

repertoire, the lambda light chains of RA pa-tients are enriched for sequences with longCDR3 intervals. Finally, the presence ofclonally related sequences in RA synovia andPBL supports an antigen-driven B-cell re-sponse in RA. We speculate that RAG-1 andRAG-2 expression may occur in a subset of RAsynovial B cells, leading to secondary immu-noglobulin gene rearrangements, as has re-cently been reported in normal peripheral lym-phoid organs. The presence of GC-likestructures capable of generating a B-cell re-sponse within RA synovia raises the possibilitythat local antigens can propagate the chronicinflammatory response.

AcknowledgmentsThis work was supported by a VA Merit Reviewgrant and NIH grant AR44243. The technicalassistance of Stephanie Byer and Jennifer Collinsis greatly appreciated. The author thanks Drs.Ralf Kuppers, Peter Burrows, Zhixin Zhang, andHarry W. Schroeder, Jr., for critical review of themanuscript.

This work was published in part in abstractform in Arthritis Rheum. 39(Suppl.): S159, 1996and Arthritis Rheum. 40(9Suppl.): S246, 1997.

References1. Harris ED Jr. (1990) Rheumatoid arthritis: patho-

physiology and implications for treatment.N. Engl. J. Med. 322: 1277-1289.

2. Zvaifler NJ. (1973) Immunopathology of joint in-flammation in rheumatoid arthritis. Adv. Immunol.16: 265-336.

3. Arend WP. (1997) The pathophysiology and treat-ment of rheumatoid arthritis. Arthritis Rheum. 40:595-597.

4. Mellors RC, Heimer R, Corcos J, Korngold L.(1959) Cellular origin of rheumatoid factor. J. Exp.Med. 110: 875-886.

5. Smiley JD, Sachs C, Ziff M. (1968) In vitro syn-thesis of immunoglobulin by rheumatoid synovialmembrane. J. Clin. Invest. 47: 624-632.

6. Dreyer WJ, Bennett JC. (1965) The molecularbasis of antibody formation: a paradox. Proc. Natl.Acad. Sci. U.S.A. 54: 864-869.

7. Sanz I. (1991) Multiple mechanisms participate inthe generation of diversity of human H chainCDR3 regions. J. Immunol. 147: 1720-1729.

8. Tonegawa S. (1983) Somatic generation of anti-body diversity. Nature 302: 575-581.

9. Max EE. (1993) Immunoglobulins: molecular ge-netics. In: Paul WE (ed). Fundamental Immunology,3rd ed. Raven Press, New York, pp. 315-382.


10. Kabat EA, Wu rT, Perry HM, Gottesman KS, Foel-ler C. (1991) Sequences of Proteins of ImmunologicalInterest, 5th ed. NIH Publ #91-3242. U.S. Depart-ment of Health and Human Services, Bethesda,MD.

11. Desiderio SV, Yancopoulos GD, Paskind M, et al.(1984) Insertion of N regions into heavy-chaingenes is correlated with expression of terminaldeoxytransferase in B cells. Nature 311: 752-755.

12. Klein U, Kuppers R, Rajewsky K. (1993) HumanIgM+ IgD+ B cells, the major B cell subset in theperipheral blood, express VK genes with no orlittle somatic mutation throughout life. Eur. J. Im-munol. 23: 3272-3277.

13. Weber J-C, Blaison G, Martin T, Knapp A-M, Pas-quali J-L. (1994) Evidence that the VKIII geneusage is nonstochastic in both adult and newbornperipheral B cells and that peripheral CD5+ adultB cells are oligoclonal. J. Clin. Invest. 93: 2093-2105.

14. Klein R, Jaenichen R, Zachau HG. (1993) Ex-pressed human immunoglobulin kappa genes andtheir hypermutation. Eur. J. Immunol. 23: 3248-3262.

15. Victor KD, Capra JD. (1994) An apparently com-mon mechanism of generating antibody diversity:length variation of the VL-JL junction. Mol. Immu-nol. 31: 39-46.

16. Bridges SL Jr, Lee SK, Johnson ML, et al. (1995)Somatic mutation and CDR3 lengths of immuno-globulin K light chains expressed in patients withrheumatoid arthritis and normal individuals.J. Clin. Invest. 96: 831-841.

17. Lee SK, Bridges SL Jr, Koopman WJ, SchroederHW Jr. (1992) The immunoglobulin kappa lightchain repertoire expressed in the synovium of apatient with rheumatoid arthritis. Arthritis Rheum.35: 905-913.

18. Martin T, Blaison G, Levallois H, Pasquali JL.(1992) Molecular analysis of the VKII-JK junc-tional diversity of polyclonal rheumatoid factorsduring rheumatoid arthritis frequently reveals Naddition. Eur. J. Immunol. 22: 1773-1779.

19. Chuchana P, Blancher A, Brockly F, Alexandre D,Lefranc G, Lefranc M-P. (1990) Definition of thehuman immunoglobulin variable lambda (IGLV)gene subgroups. Eur. J. Immunol. 20: 1317-1325.

20. Williams SC, Frippiat JP, Tomlinson IM, Ignato-vich 0, Lefranc MP, Winter G. (1996) Sequenceand evolution of the human germline V lambdarepertoire. J. Mol. Biol. 264: 220-232.

21. Kawasaki K, Minoshima S, Nakato E, et al. (1997)One-megabase sequence analysis of the humanimmunoglobulin lambda gene locus. Genome Res.7: 250-261.

22. Bridges SL Jr, Lavelle JC, Lee SK, Byer S, Schroe-der HW Jr. (1997) CDR3 fingerprinting of immu-noglobulin kappa light chains expressed in rheu-matoid arthritis: evidence of antigenic selection or

dysregulation of gene rearrangement in B cells.Ann. N.Y. Acad. Sci. 815: 423-426.

23. Ermel RW, Kenny TP, Wong A, Solomon A, ChenPP, Robbins DL. (1994) Preferential utilization of anovel V lambda 3 gene in monoclonal rheumatoidfactors derived from the synovial cells of rheuma-toid arthritis patients. Arthritis Rheum. 37: 860-868.

24. Ermel RW, Kenny TP, Chen PP, Robbins DL.(1993) Molecular analysis of rheumatoid factorsderived from rheumatoid synovium suggests anantigen-driven response in inflamed joints. Arthri-tis Rheum. 36: 380-388.

25. Harindranath N, Goldfarb IS, Ikematsu H, et al.(1991) Complete sequence of the genes encodingthe VH and VL regions of low- and high- affinitymonoclonal IgM and IgAl rheumatoid factors pro-duced by CD5+ B cells from a rheumatoid arthritispatient. Int. Immunol. 3: 865-875.

26. Fang Q, Kannapell CC, Gaskin F, Solomon A,Koopman WJ, Fu SM. (1994) Human rheumatoidfactors with restrictive specificity for rabbit immu-noglobulin G: auto- and multi-reactivity, diverseVH gene segment usage and preferential usage ofVAIIIb. J. Exp. Med. 179: 1445-1456.

27. Lee SK, Bridges SL Jr, Kirkham PM, KoopmanWJ, Schroeder HW Jr. (1994) Evidence of antigenreceptor-influenced oligoclonal B lymphocyte ex-pansion in the synovium of a patient with long-standing rheumatoid arthritis. J. Clin. Invest. 93:361-370.

28. Chirgwin JM, Przybyla AE, MacDonald RJ, RutterWJ. (1979) Isolation of biologically active ribonu-cleic acid from sources enriched in ribonuclease.Biochemistry 18: 5294.

29. Daley MD, Peng HQ, Misener V, Liu XY, Chen PP,Siminovitch KA. (1992) Molecular analysis of hu-man immunoglobulin V lambda germline genes:subgroups V lambda III and V lambda IV. Mol.Immunol. 29: 1515-1518.

30. Combriato G, Klobeck HG. (1991) V lambda and Jlambda-C lambda gene segments of the humanimmunoglobulin lambda light chain locus are sep-arated by 14 kb and rearrange by a deletion mech-anism. Eur. J. Immunol. 21: 1513-1522.

31. Blomberg BB, Glozak MA, Donohoe ME. (1995)Regulation of human lambda light chain geneexpression. Ann. N. Y. Acad. Sci. 764: 84-98.

32. English D, Andersen BR. (1974) Single-step sep-aration of red blood cells, granulocytes and mono-nuclear leukocytes on discontinuous density gra-dients of Ficoll-Hypaque. J. Immunol. Methods 5:249-252.

33. Madsen M, Johnsen HE, Hansen PW, ChristiansenSE. (1980) Isolation of human T and B lympho-cytes by E-rosette gradient centrifugation. Char-acterization of the isolated subpopulations. J. Im-munol. Methods 33: 323-336.

34. Ausubel FM, Brent R, Kingston RE, Moore DD,Seidman JG, Smith JA, et al., eds. (1997) Prepa-


ration and analysis of DNA. In: Current Protocols inMolecular Biology. John Wiley & Sons, New York,pp. 2.1.1-2.1.2.

35. Frippiat JP, Lefranc MP. (1994) Genomic organi-sation of 34 kb of the human immunoglobulinlambda locus (IGLV): restriction map and se-quences of new V lambda III genes. Mol. Immunol.31: 657-670.

36. Sanger F, Nicklen S, Coulson AR. (1977) DNAsequencing with chain-terminating inhibitors.Proc. Natl. Acad. Sci. U.S.A. 74: 5463-5467.

37. Elgavish RA, Schroeder HW Jr. (1993) SAW: Agraphical user interface for the analysis of immu-noglobulin variable domain sequences. Biotech-niques 15: 1066-1071.

38. Lafaille JJ, DeCloux A, Bonneville M, Takagaki Y,Tonegawa S. (1989) Junctional sequences of T cellreceptor gamma delta genes: implications forgamma delta T cell lineages and for a novel inter-mediate of V-(D)-J joining. Cell 59: 859-870.

39. Lewis SM. (1994) P nucleotide insertions and theresolution of hairpin DNA structures in mamma-lian cells. Proc. Natl. Acad. Sci. U.S.A. 91: 1332-1336.

40. Frippiat JP, Williams SC, Tomlinson IM, et al.(1995) Organization of the human immunoglob-ulin lambda light-chain locus on chromosome22q1.2. Hum. Mol. Genet. 4: 983-991.

41. Frippiat JP, Dard P, Marsh S, Winter G, LefrancMP. (1997) Immunoglobulin lambda light chainorphons on human chromosome 8q1 1.2. Eur.J. Immunol. 27: 1260-1265.

42. Daley MD, Olee T, Peng HQ, Soto-Gil RW, ChenPP, Siminovitch KA. (1992) Molecular character-ization of the human immunoglobulin V lambda Igermline gene repertoire. Mol. Immunol. 29: 1031-1042.

43. Williams SC, Winter G. (1993) Cloning and se-quencing of human immunoglobulin V lambdagene segments. Eur. J. Immunol. 23: 1456-1461.

44. Ch'ang LY, Schnell MG, Ringelberg CS, Weiss D,Solomon A. (1995) Molecular characterization ofa human VAVIII germline gene. Mol. Immunol. 32:49-55.

45. Deftos M, Soto-Gil R, Quan M, Olee T, Chen PP.(1994) Utilization of a potentially universal down-stream primer in the rapid identification and char-acterization of V lambda genes from two newhuman V lambda gene families. Scand. J. Immunol.39: 95-103.

46. Ch'ang LY, Yen CP, Besl L, Schell M, Solomon A.(1994) Identification and characterization of afunctional human Ig VAVI germline gene. Mol.Immunol. 31: 531-536.

47. Ignatovich 0, Tomlinson IM, Jones PT, Winter G.(1997) The creation of diversity in the humanimmunoglobulin VA repertoire. J. Mol. Biol. 268:69-77.

48. Blomberg BB, Solomon A. (1997) The murine andhuman lambda light chain immunoglobulin loci:

organization and expression. In: Herzenberg LA,Weir DM, Herzenberg LA, Blackwell C (eds).Handbook of Experimental Immunology. BlackwellScience, Inc., Boston. Fifth Edition.

49. Bauer TR Jr, Blomberg B. (1991) The humanlambda L chain Ig locus. Recharacterization of JClambda 6 and identification of a functional JClambda 7. J. Immunol. 146: 2813-2820.

50. Vasicek TJ, Leder P. (1990) Structure and expres-sion of the human immunoglobulin lambdagenes. J. Exp. Med. 172: 609-620.

51. Niewold TA, Murphy CL, Weiss DT, Solomon A.(1996) Characterization of a light chain product ofthe human JC lambda 7 gene complex. J. Immu-nol. 157: 4474-4477.

52. Dariavach P, Lefranc G, Lefranc MP. (1987) Hu-man immunoglobulin C lambda 6 gene encodesthe Kern+Oz- lambda chain and C lambda 4 andC lambda 5 are pseudogenes. Proc. Natl. Acad. Sci.U.S.A. 84: 9074-9078.

53. Udey JA, Blomberg B. (1987) Human lambdalight chain locus: organization and DNA se-quences of three genomic J regions. Immunogenet-ics 25: 63-70.

54. Ishikawa H, Ziff M. (1976) Electron microscopicobservations of immunoreactive cells in the rheu-matoid synovial membrane. Arthritis Rheum. 19:1-14.

55. Brown KA, Perry ME, Mustafa Y, et al. (1995) Thedistribution and abnormal morphology of plasmacells in rheumatoid synovium. Scand. J. Immunol.41: 509-517.

56. Kelley DE, Perry RP. (1986) Transcriptional andposttranscriptional control of immunoglobulinmRNA production during B lymphocyte develop-ment. Nucl. Acids Res. 14: 5431-5447.

57. Klein U, Kuppers R, Rajewsky K. (1997) Evidencefor a large compartment of IgM-expressing mem-ory B cells in humans. Blood 89: 1288-1298.

58. Clausen BE, Bridges SL Jr, Lavelle JC, et al. (1998)Clonally related immunoglobulin VH domains andnonrandom use of DH gene segments in rheuma-toid arthritis synovium. Mol. Med. 4: 240-2 57.

59. Campbell MJ, Zelenetz AD, Levy S, Levy R. (1992)Use of family specific leader region primers forPCR amplification of the human heavy chain vari-able region gene repertoire. Mol. Immunol. 29:193-203.

60. Paabo S, Higuchi RG, Wilson AC. (1989) AncientDNA and the polymerase chain reaction. J. Biol.Chem. 264: 9709-9712.

61. Paabo S, Irwin DM, Wilson AC. (1990) DNA dam-age promotes jumping between templates duringenzymatic amplification. J. Biol. Chem. 265: 4718-4721.

62. Lundberg KS, Shoemaker DD, Adams MW, ShortJM, Sorge JA, Mathur EJ. (1991) High-fidelityamplification using a thermostable DNA polyer-mase isolated from Pyrococcus furiosus. Gene 108:1-6.


63. Tindall KR, Kunkel TA. (1988) Fidelity of DNAsynthesis by the Thermus aquaticus DNA poly-morphism. Biochemistry 27: 6008-6013.

64. Barnes WM. (1992) The fidelity of Taq polymer-ase catalyzing PCR is improved by an N-terminaldeletion. Gene 112: 29-35.

65. Billips LG, Lassoued K, Nunez C, et al. (1995)Human B-cell development. Ann. N.Y. Acad. Sci.764: 1-8.

66. Kubagawa H, Cooper MD, Carroll AJ, BurrowsPD. (1989) Light-chain gene expression beforeheavy-chain gene rearrangement in pre-B cellstransformed by Epstein-Barr virus. Proc. Natl.Acad. Sci. U.S.A. 86: 2356-2360.

67. Schroeder J, H. W, Burrows PD, Cooper MD,Kubagawa H. (1992) N region addition and un-usual CDR3 length distributions in K-chain onlypre-B cells [abstract]. J. Cell. Biochem. S17B:235.

68. Nishimoto N, Kubagawa H, Cooper MD. (1991)Comparison of pre-B cell differentiation in normaland X-linked agammaglobulinemia (XLA) indi-viduals [abstract]. FASEB J. 5: 1346.

69. Goossens T, Klein U, Kuppers R. (1998) Frequentoccurrence of deletions and duplications duringsomatic hypermutation: implications for onco-gene translocations and heavy chain disease. Proc.Natl. Acad. Sci. U.S.A. 95: 2463-2468.

70. Wilson PC, de Bouteiller 0, Liu YJ, et al. (1998)Somatic hypermutation introduces insertions anddeletions into immunoglobulin V genes. J. Exp.Med. 187: 59-70.

71. Bridges SL Jr, Lee SK, Koopman WJ, SchroederHW Jr. (1993) Analysis of immunoglobulingamma heavy chain expression in synovial tissueof a patient with rheumatoid arthritis. ArthritisRheum. 36: 631-641.

72. Ruff-Jamison S, Glenney J. (1993) Requirementfor both H and L chain V regions, VH and VKjoining amino acids, and the unique H chain Dregion for the high affinity binding of an anti-phosphotyrosine antibody. J. Immunol. 150:3389-3396.

73. Randen I, Brown D, Thompson KM, et al. (1992)Clonally related IgM rheumatoid factors undergoaffinity maturation in the rheumatoid synovialtissue. J. Immunol. 148: 3296-3301.

74. Randen I, Thompson KM, Pascual V, et al. (1992)Rheumatoid factor V genes from patients withrheumatoid arthritis are diverse and show evi-dence of an antigen-driven response. Immunol.Rev. 128: 49-71.

75. Martin T, Crouzier R, Blaison G, Levallois H,Pasquali JL. (1993) A minor group of rheuma-toid factors isolated from a patient with rheu-matoid arthritis is derived from somatically mu-tated Vk 1 genes further evidence thatrheumatoid factors during autoimmune dis-eases undergo an antigen driven maturation.Autoimmunity 15: 163-170.

76. Olee T, Lu EW, Huang DF, et al. (1992) Geneticanalysis of self-associating immunoglobulin Grheumatoid factors from two rheumatoid synoviaimplicates an antigen-driven response. J. Exp. Med.175: 831-842.

77. Schroder ARE, Greiner A, Seyfert C, Berek C.(1996) Differentiation of B cells in the nonlym-phoid tissue of the synovial membrane of patientswith rheumatoid arthritis. Proc. Natl. Acad. Sci.U.S.A. 93: 221-225.

78. Bridges SL Jr, Clausen BE, Lavelle JC, Fowler PG,Koopman WJ, Schroeder HW Jr. (1995) Analysisof immunoglobulin gamma heavy chains fromrheumatoid arthritis synovium: evidence of anti-gen-driven selection. Ann. N.Y. Acad. Sci. 764:450-452.

79. McBlane JF, van Gent DC, Ramsden DA, et al.(1995) Cleavage at a V(D)J recombination signalrequires only RAGI and RAG2 proteins and oc-curs in two steps. Cell 83: 387-395.

80. Gellert M. (1997) Recent advances in understand-ing V(D)J recombination. Adv. Immunol. 64: 39-64.

81. Prak EL, Weigert M. (1995) Light chain replace-ment: a new model for antibody gene rearrange-ment. J. Exp. Med. 182: 541-548.

82. Chen C, Prak EL, Weigert M. (1997) Editing dis-ease-associated autoantibodies. Immunity 6: 97-105.

83. Pelanda R, Schwers S, Sonoda E, Torres RM,Nemazee D, Rajewsky K. (1997) Receptor editingin a transgenic mouse model: site, efficiency, androle in B cell tolerance and antibody diversifica-tion. Immunity 7: 765-775.

84. Han S, Dillon SR, Zheng B, Shimoda M, SchlisselMS, Kelsoe G. (1997) V(D)J recombinase activityin a subset of germinal center B lymphocytes.Science 278: 301-305.

85. Papavasiliou F, Casellas R, Suh H, et al. (1997)V(D)J recombination in mature B cells: a mecha-nism for altering antibody responses. Science 278:298-301.

86. Han S, Zheng B, Schatz DG, Spanopoulou E, Kel-soe G. (1996) Neoteny in lymphocytes: Ragl andRag2 expression in germinal center B cells. Science274: 2094-2097.

87. Ziff M. (1974) Relation of cellular infiltration ofrheumatoid synovial membrane to its immuneresponse. Arthritis Rheum. 17: 313-319.

88. Iguchi T, Ziff M. (1986) Electron microscopicstudy of rheumatoid synovial vasculature. Inti-mate relationship between tall endothelium andlymphoid aggregation. J. Clin. Invest. 77: 355-361.

89. van Dinther-Janssen AC, Pals ST, Scheper R,Breedveld F, Meijer CJ. (1990) Dendritic cells andhigh endothelial venules in the rheumatoid syno-vial membrane. J. Rheumatol. 17: 11-17.

90. Randen I, Mellbye OJ, F0rre 0, Natvig JB. (1995)


The identification of germinal centres and follicu-lar dendritic networks in rheumatoid synovial tis-sue. Scand. J. Immunol. 41: 481-486.

91. Steere AC, Duray PH, Butcher EC. (1988) Spiro-chetal antigens and lymphoid cell surface markersin Lyme synovitis. Comparison with rheumatoidsynovium and tonsillar lymphoid tissue. ArthritisRheum. 31: 487-495.

92. Schroder ARE, Sieper J, Berek C. (1997) Antigen-dependent B cell differentiation in the synovialtissue of a patient with reactive arthritis. Mol. Med.3: 260-272.

93. Kumon I. (1992) In situ characterization of mono-

nuclear cell phenotype in intrahepatic lymphoidfollicles in patients with chronic viral hepatitis.Gastroenterol. Jpn. 27: 638-645.

94. Freni MA, Artuso D, Gerken G, et al. (1995) Focallymphocytic aggregates in chronic hepatitis C: oc-currence, immunohistochemical characterization,and relation to markers of autoimmunity. Hepa-tology 22: 389-394.

95. Mosnier JF, Degott C, Marcellin P, Henin D, Er-linger S, Benhamou JP. (1993) The intraportallymphoid nodule and its environment in chronicactive hepatitis C: an immunohistochemicalstudy. Hepatology 17: 366-371.

Documents

Utilization on CDR3 Length