The

Journ

al o

f Exp

erim

enta

l M

edic

ine

J. Exp. Med.

©

The Rockefeller University Press • 0022-1007/2004/11/1205/7 $8.00Volume 200, Number 9, November 1, 2004 1205–1211http://www.jem.org/cgi/doi/10.1084/jem.20041135

Brief Definitive Report

1205

Important Roles for E Protein Binding Sites within the Immunoglobulin

�

Chain Intronic Enhancer in Activating

V

�

J

�

Rearrangement

Matthew A. Inlay, Hua Tian, Tongxiang Lin, and Yang Xu

Division of Biological Sciences, University of California, San Diego, CA 92093

Abstract

The immunoglobulin

�

light chain intronic enhancer (iE

�

) activates

�

rearrangement and is requiredto maintain the earlier or more efficient rearrangement of

�

versus lambda (

�

). To understandthe mechanism of how iE

�

regulates

�

rearrangement, we employed homologous recombination tomutate individual functional motifs within iE

�

in the endogenous

�

locus, including the NF-

�

Bbinding site (

�

B), as well as

�

E1,

�

E2, and

�

E3 E boxes. Analysis of the impacts of these muta-tions revealed that

�

E2 and to a lesser extent

�

E1, but not

�

E3, were important for activating

�

rearrangement. Surprisingly, mutation of the

�

B site had no apparent effect on

�

rearrangement.Comparable to the deletion of the entire iE

�

, simultaneous mutation of

�

E1 and

�

E2 reducesthe efficiency of

�

rearrangement much more dramatically than either

�

E1 or

�

E2 mutationalone. Because E2A family proteins are the only known factors that bind to these E boxes,these findings provide unambiguous evidence that E2A is a key regulator of

�

rearrangement.

Key words: B cell development •

V(D)J

recombination • accessibility • monospecificity • transcription factor

Introduction

Each B lymphocyte generates a unique set of immunoglobulinheavy (IgH) and light (IgL) chain genes through the somaticrearrangement of

V

,

D

, and

J

gene segments. To ensure themonospecificity of each B cell,

V(D)J

recombination is reg-ulated in such a lineage- and stage-specific manner that IgHand IgL rearrangement occur at distinct stages of development(1). Because

V(D)J

rearrangement of all Ig loci involves thesame recombination machinery, cis elements within theseantigen receptor loci must play decisive roles in regulatingthe accessibility of each locus to

V(D)J

recombination machin-ery (2). Within the

�

locus, several cis-acting elementswere initially identified by their ability to activate transcrip-tion of

�

reporter constructs in B cell lines, including twoenhancers, one within the

J

�

-

C

�

intron (iE

�

) and one 3

�

of

C

�

(3

�

E

�

) (3–8). More recently, a putative third enhancer 8kb downstream of 3

�

E

�

(Ed) was also discovered (9). Thedeletion of the intronic enhancer and matrix attachmentregion (MiE

�

) led to decreased

�

rearrangement and alower

�

:

�

ratio. More importantly, MiE

�

is required for theearlier or more efficient rearrangement of

�

versus

�

loci.(10, 11). The matrix attachment region (MAR) withinMiE

�

does not contribute to these activities since the dele-

tion of the

�

MAR alone does not have an inhibitory effecton the overall level of

�

rearrangement (12). The deletionof the 3

�

enhancer (3

�

E

�

) results in a similar, though lessdramatic decrease in the ratio of

�

:

�

B cells and

�

rearrange-ment (13). In addition, 3

�

E

�

appears to play an importantrole in activating

�

transcription in mature B cells (11, 13).Although the loss of either enhancer alone does not eliminate

�

rearrangement, deletion of both enhancers from endoge-nous

�

loci results in a complete block of

�

rearrangement(11). Therefore, MiE

�

and 3

�

E

�

together are the necessaryelements for

�

rearrangement.Several functional motifs within the intronic enhancer

were identified through a battery of biochemical and cellline transfection studies. One such functional motif is theNF-

�

B binding site, denoted

�

B (14). The potential role ofthe

�

B site in

�

rearrangement was suggested by the find-ing that LPS could induce

�

germline transcription and rear-rangement through an NF-

�

B–dependent pathway (15–18). iE

�

also contains a class of protein-binding motifs re-ferred to as E boxes, which are also identified in enhancersof other antigen receptor genes (19, 20). iE

�

contains threeE boxes, labeled

�

E1,

�

E2, and

�

E3.

�

E2 was found tobind the E2A gene products E12 and E47, which are re-quired for B cell development (21–23). E2A gene productscan also bind to

�

E1 but not to

�

E3 (Murre, C., personalcommunication). E2A can induce germline transcription

Address correspondence to Yang Xu, Div. of Biological Sciences, Univer-sity of California, San Diego, 9500 Gilman Dr., La Jolla, CA 92093-0322.Phone: (858) 822-1084; Fax: (858) 534-0053; email: yangxu@ucsd.edu

Dow

nloaded from http://rupress.org/jem

/article-pdf/200/9/1205/1152365/jem20091205.pdf by guest on 03 Septem

ber 2021

Regulation of

V(D)J

Recombination by Enhancer Elements

1206

and

�

rearrangement when cotransfected with the

RAG

genes in a nonlymphoid cell line (24). However, the role ofE2A in the regulation of

�

rearrangement in the physiolog-ical context remains unclear.

To delineate the mechanism through which iE

�

activates

�

rearrangement, we employed homologous recombinationand

Cre/loxP

-mediated deletion to produce independentlytargeted deletions of four functional motifs within the en-dogenous

�

intronic enhancer. Analysis of the effects of thesemutations on

�

rearrangement revealed that the E2A-bind-ing E boxes, but not the

�

B site, were quantitatively impor-tant for the activation of

�

rearrangement. In addition, thesimultaneous mutation of both

�

E1 and

�

E2 sites impacts

�

rearrangement as severely as the deletion of the entire iE�,indicating that E2A-dependent pathways are the major me-diator of iE�’s activity in activating � rearrangement.

Materials and MethodsGeneration of m�EX Embryonic Stem Cells. The 740-bp � in-

tronic enhancer was cloned into pBluescript. The four sites weredeleted and replaced with diagnostic EcoRI sites independentlyby site-directed mutagenesis. The primers used to replace eachsite are as follows: m�B, 5�-TGGCATCTCAACGAATTCA-GAGCCATCTGG-3�; m�E1, 5�-GACTTTCCGAGAGAAT-TCAGTTGCTTAAGA-3�; m�E2, 5�-CAGTTCCTCCGAA-TTCGATTACAG-3�; and m�E3, 5�-TGGCTAAAAATTGA-ATTCAAACCATTAGAC-3�.

The m�E1/2 double mutations were generated by PCR-mediated mutagenesis. The �E1 and �E2 sequences are as fol-lows, with the underlined C nucleotides mutated to A: �E1,CATCTGGC; �E2, CAGGTGGC. The MiE� deletion (m�D)was generated as described (10). All m�EX clones were insertedinto the m�D targeting construct. 20 �g of each targeting con-struct was linearized with PvuI and transfected into embryonicstem (ES) cells. Transfectants were selected with G418 (300 �g/ml) and gancyclovir (1 �M). Homologous recombination wasscreened by Southern blot, and positive clones were then tran-siently transfected with 10–20 �g of a plasmid containing the Cregene to remove the PGK-neor gene as described (10). PGK-neor–deleted ES clones were subcloned, confirmed by Southern blot,and analyzed by RAG-2–deficient blastocyst complementation asdescribed (10). m�E1/2 ES cells were microinjected into blasto-cysts before Cre transfection and then bred to Cre-expressingmice to generate m�E1/2 knock-in mice.

Quantitative Analysis of V� J� Rearrangements. To detect V� J�

rearrangements, genomic DNA from mature B cells (CD19�/IgM�) was purified and amplified by PCR using the degenerateV� primer (V�D) (15) and a primer within the kappa intronic en-hancer (K1). PCR reactions were conducted under the followingconditions: 95�C for 5 min, followed by 26 cycles of 94�C for 1min, 60�C for 1 min, and 72�C for 3.5 min, followed by an addi-tional 10 min at 72�C. To control for the total amount of B cellDNA used in each PCR reaction, VDJH rearrangement was alsoanalyzed using a primer for the J558 family (J558) and a primerthat can bind to all four JH gene segments (JH(1�4)). The in-tensity of each PCR product was quantified with a Storm phos-phorimager. Primers used are as follows: V�D, 5�-GGCTGCAG-STTCAGTGGCAGTGGRTCWGGRAC-3�; K1, 5�-CCAT-GACTTTTGCTGGCTGTAGATTTTACCTC-3�; J558, 5�-CTTCAGTGAAGCTGTCCTGCAAGGCTT-3�; and JH(1�4),

5�-CAGCTTACCTGAGGAGACGGTGA-3�. Hybridomas weregenerated and analyzed as described (10). The University of Cali-fornia, San Diego Animal Subject Committee approved all ex-periments that involved mice.

Quantitative Analysis of � Transcription by Real-Time PCR. Bcells from the spleens of WT and homozygous m�E1/2 mutantmice were purified by magnetic-activated cell sorting using�CD43 microbeads according to the manufacturer’s protocol(Miltenyi Biotec). In m�E1/2 mutants, �� B cells were removedby costaining with 0.5 �g/million cells biotin-conjugated�-mouse Ig� antibodies (clone R26-46; BD Biosciences) and then,after washing, staining with 0.5 �g/million cells �-biotin micro-beads (Miltenyi Biotec). RNA from 1 million sorted cells waspurified using the RNAeasy kit (QIAGEN) combined with on-column DNase digestion (QIAGEN) and converted into cDNAusing the Superscript First Strand System (Invitrogen) accordingto the manufacturers’ protocols. Primers for amplifying the con-stant regions of � (C�1 and C�2) and � (C�1 and C�2) were de-signed using Primer Express software (Applied Biosystems) andused at concentrations of 200 and 400 nM, respectively. Real-time PCR reactions were performed using the SYBR GreenPCR Master Mix (Applied Biosystems) in an ABI Prism 7000 Se-quence Detection System according to the manufacturer’s pro-tocols. Relative transcription levels were calculated using ABIPrism 7000 SDS software using the standard curve method. Primersequences are as follows: C�1, 5�-ACACCTGCCGTGTGGA-TCA-3�; C�2, 5�-GAGGAAGATGTCGGCAAAGG-3�; C�1,5�-CAACTGTATCCATCTTCCCACCA-3�; and C�2, 5�-GGCACCTCCAGATGTTAACTGCT-3�.

Results and DiscussionGeneration of m�EX ES Cells. The 740-bp HindIII–

AflII region spanning the � intronic enhancer and associ-ated matrix attachment region (MiE�) was cloned intopBluescript. The NF-�B, �E1, �E2, or �E3 sites were eachindividually replaced with a diagnostic EcoRI restrictionsite by site–directed mutagenesis (Fig. 1 F). These iE� mu-tants, collectively referred to as m�EX and individually asm�B, m�E1, m�E2, and m�E3, were independently in-serted into the targeting construct used previously to deleteMiE�, so that homologous recombination between the tar-geting vector and the endogenous locus would replace theendogenous MiE� with m�EX (10) (Fig. 1 B). Homolo-gous recombination events were screened by Southernblotting with EcoRI digestion and hybridization to probeA (Fig. 1, A and C; not depicted). The PGK-neor gene wasexcised from the targeted allele by transient expression ofthe Cre gene in positive ES clones (Fig. 1 D). ES cloneswith the PGK-neor deleted were subcloned and confirmedby Southern blot with EcoRI digestion and hybridizationto probe A (Fig. 1 E, lanes 2–5).

Analysis of V� J� Recombination in m�EX B Cells. Theeffects of m�EX mutations on � rearrangement in B cellswere assayed by recombination activating gene-2–deficient(RAG-2�/�) blastocyst complementation as described (10).Since heterozygous mutant ES cells were used, this enabledus to compare the rearrangement frequency between theWT and mutant � alleles. B cells were sorted from m�EX–RAG-2�/� mice and analyzed for V� J� recombination using

Dow

nloaded from http://rupress.org/jem

/article-pdf/200/9/1205/1152365/jem20091205.pdf by guest on 03 Septem

ber 2021

Inlay et al. Brief Definitive Report1207

a quantitative PCR assay as previously described (11, 12).This assay could detect rearrangement to all four functionalJ� gene segments and distinguish PCR products derivedfrom the WT allele from those from the mutant allele due toan additional 150 bp within the mutant allele (Fig. 2 A).Therefore, the efficiency of V� J�1-5 rearrangement of themutant � allele could be compared with that of the WT al-lele in the same PCR reaction (Fig. 2 C). Our analysis indi-

cated that neither the NF-�B (m�B) nor �E3 (m�E3) sitemutations had any apparent effect on � rearrangement (Fig.2 B). However, mutation of the �E1 (m�E1) or �E2(m�E2) site reduced the rearrangement efficiency of the mu-tant � allele. The mutation of the �E2 site had a more inhib-itory effect on � rearrangement (Fig. 2 B, lane 4). In thiscontext, the ratio of the rearrangement frequency of V� toJ�1, J�2, J�4, and J�5 gene segments of the mutant allele ver-

Figure 1. Generation ofm�EX mutant ES cells. (A) Theendogenous WT � locus. Thelengths of the diagnostic restric-tion fragments and probes areshown. Markings above the dia-gram are spaced 1 kb apart. (B)The m�EX targeting construct.The mutated intronic enhancer(*) was inserted into the targetingconstruct just upstream of theloxP-flanked PGK-neor gene.Additional polylinker sequenceupstream of the mutated intronicenhancer is represented by a boxwith diagonal lines. (C) Targetedlocus with the PGK-neor geneinserted. Size of the diagnosticEcoRI fragment is shown. (D)Final configuration with thePGK-neor gene removed by Cre/loxP-mediated deletion. The sizeof the EcoRI restriction frag-ment is shown. (E) Southernblot analysis of the genomicDNA from WT ES cells and ESclones from the final configurationof each of the four heterozygousm�EX mutants. WT (13.5 kb)and neor-deleted (9.4 kb) bandsare indicated. (F) Diagram of theintronic enhancer and its func-tional motifs. The MAR, �Bsite, and the three E boxes areshown. The sequences spanningeach of the four sites that werereplaced by EcoRI sites areshown in boxes. The consensussequences of the each of the foursites are underlined.

Dow

nloaded from http://rupress.org/jem

/article-pdf/200/9/1205/1152365/jem20091205.pdf by guest on 03 Septem

ber 2021

Regulation of V(D)J Recombination by Enhancer Elements1208

sus that of the WT allele was 52.1, 31.9, 26.2, and 27.9%, re-spectively (Fig. 2 C). In m�E1 B cells, the rearrangement ef-ficiency of the mutant allele was approximately twofoldreduced for each of the four possible V� J� rearrangementscompared with that of the WT allele (Fig. 2 B, lane 3,and C).

Analysis of V� J� Recombination in m�EX �� Hybridomas.To further confirm the impact of m�EX mutations on �rearrangement efficiency and analyze the rearrangementfrequency of mutant and WT alleles in individual B cells,we generated hybridomas from the spleen cells of all fourm�EX–RAG-2�/� mice as described (10). GenomicDNA derived from �� hybridomas was analyzed bySouthern blotting to detect V� J� rearrangements at theWT and mutant alleles. Consistent with the data fromthe PCR analysis, we observed the most dramatic de-crease in the rearrangement frequency of the mutant al-lele in m�E2 hybridomas (Fig. 3 B). In this context, ofthe 67 �� hybridomas analyzed 48 had rearrangementson only the WT allele, 10 had rearrangements on onlythe mutant allele, and 9 had rearrangements on both al-leles (Fig. 3 C). The ratio of the rearrangement fre-quency of the WT allele versus that of the mutant allelewas �3:1. A less dramatic decrease was observed inm�E1 hybridomas (Fig. 3 C). Consistent with the dataderived from the PCR analysis, little difference was de-tected in the rearrangement frequency of the WT and

mutant alleles in hybridomas derived from the m�B andm�E3 B cells (Fig. 3, A and C).

To determine the contribution of the �E1 and �E2 sitesto the full activity of iE�, hybridomas were also generatedfrom B cells in which the entire intronic enhancer was de-leted from one allele (m�D). Only 3 of 43 �� m�D hy-bridomas analyzed harbored a V� J� rearrangement on themutant allele, whereas the WT allele was rearranged in all43 hybridomas (Fig. 3 C). Therefore, the reduction of therearrangement efficiency caused by the deletion of the en-tire iE� was much more dramatic than that caused by �E1or �E2 mutations alone. In addition, these data indicatethat the rearrangement of the mutant allele occurs only af-ter the rearrangement of the WT allele in m�D B cells.

Critical Roles of �E1 and �E2 in Mediating iE�’s Function inActivating � Rearrangement. Although the reduction of �rearrangement caused by the �E1 or �E2 mutations alonewas significantly less than that caused by the deletion of theentire enhancer, it is possible that these two sites have re-dundant functions since both sites can be bound by E2Afamily proteins. To test this hypothesis, we mutated both�E1 and �E2 sites simultaneously through homologous re-combination and assayed for effects on � rearrangement.The mutation introduced at both sites was a single nucle-otide (C to A) mutation that destroyed the canonical basichelix-loop-helix binding site (CANNTG) (Fig. 4 A). Todetermine the effects of the �E1/2 mutation on � rear-

Figure 2. PCR analysis ofV� J� rearrangement in m�EXheterozygous mice. (A) PCRstrategy. Examples of rearrangedV� J�1 WT and m�EX alleles areshown. Annealing locations of thedegenerate V� primer and K1primer are shown with arrow-heads, and probe B is shown with

a horizontal bar. Markers are placed at 1 kb intervals for scale. (B) Represen-tative PCR products of the rearrangement assay from B cells of WT andeach of the four m�EX heterozygous mutants are shown. Bands correspond-ing to mutant (mut) and WT V� J�1, V� J�2, V� J�4, and V� J�5 rearrange-ments are indicated. Heavy chain rearrangement was also assayed by PCR asa control for the number of B cells, using primers J558 and JH(1�4) as de-scribed (11). Band shown corresponds to rearrangement from J558 VH

family members to either JH1 or JH4 (VDJH). Only the smallest band pro-duced by this primer set is shown. (C) Statistical analysis of the ratio of therearrangement efficiency of m�EX allele versus that of the WT allele. Rear-rangement efficiency was calculated by dividing the intensity of bandsderived from the mutant allele by that of the WT allele. Means values fromat least three independent experiments are presented with error bars. Totalnumber of mice (n) analyzed for each mutation were; m�B (n 3), m�E1(n 3), m�E2 (n 5), and m�E3 (n 2).

Dow

nloaded from http://rupress.org/jem

/article-pdf/200/9/1205/1152365/jem20091205.pdf by guest on 03 Septem

ber 2021

Inlay et al. Brief Definitive Report1209

rangement, we used the same PCR assay described in Fig.2 to analyze � rearrangement of WT and mutant alleles in�� m�E1/2 B cells purified from the spleens of RAG�/�

chimeric mice or heterozygous mutant mice. The reduc-tion in � rearrangement caused by the m�E1/2 mutationwas much more severe than that caused by either �E1 or�E2 mutations alone (Fig. 4, B and C). Analysis of �� hy-bridomas derived from m�E1/2 B cells revealed a morethan 10-fold reduction in the rearrangement efficiency ofthe mutant allele, comparable to that caused by the deletionof the entire enhancer (Fig. 4 D). In addition, similar tofindings in m�D hybridomas, the m�E1/2 allele only rear-ranged after the WT allele had already rearranged.

To rule out the possibility that the greatly reduced rear-rangement efficiency of m�E1/2 alleles observed in splenicB cells is due to the impaired expression of rearranged � atm�E1/2 alleles during the transition from pre–B cells to im-mature B cells, we examined the rearrangement frequency

of WT and mutant alleles in pre–B cells (B220�/CD43�/IgM�) sorted from the BM of m�E1/2 mice (Fig. 4 E).Similar to that observed in splenic B cells, the rearrange-ment frequency of the m�E1/2 allele was significantly re-duced in pre–B cells compared with the WT allele. To fur-ther determine whether the m�E1/2 mutation affects �expression in B cells, we analyzed � expression in �� B cellspurified from the spleens of WT mice and homozygousm�E1/2 mutant mice. Similar levels of � mRNA were de-tected in WT and m�E1/2 �� B cells, indicating that them�E1/2 mutation had no apparent effect on � transcription(Fig. 4 F). Therefore, �E1 and �E2 play synergistic roles iniE�’s function in activating � rearrangement.

Sequential rearrangement of the IgH and IgL chain genesis critical to ensure B cell monospecificity. The � intronicenhancer plays an important role in this process (10, 11). Byanalyzing the functional motifs within iE� in vivo, we dem-onstrated that the �B site mutation had no apparent impact

Figure 3. Southern blot analy-sis of � rearrangement in m�EXhybridomas. Representative datafrom m�B (A) and m�E2 (B) hy-bridomas. Genomic DNA wasdigested with BamHI and hy-bridized to probe B. Lanes 1, 2,and 3 on each blot are genomicDNA from WT ES, mutant ES,and P3X63 fusion partner, respec-tively. All other lanes representindividual �� hybridoma lines.Bands representing germline WT(12.5 kb) and mutant (9.7 kb)alleles and the contribution ofthe P3X63 fusion partner are in-dicated. Unspecified bands ofvarious sizes represent rearrangedalleles. White lines indicate thatintervening lanes have been splicedout. (C) Analysis of � rearrange-ment frequency in �-expressinghybridomas. Hybridomas werescored for rearrangement of theWT allele, mutant allele, or bothalleles. Rearrangement frequen-cies are the number of WT ormutant rearranged alleles dividedby the total number of hybrid-omas. Rearrangement efficiencyis calculated by dividing the mu-tant rearrangement frequency bythe WT rearrangement frequency.

Dow

nloaded from http://rupress.org/jem

/article-pdf/200/9/1205/1152365/jem20091205.pdf by guest on 03 Septem

ber 2021

Regulation of V(D)J Recombination by Enhancer Elements1210

on � rearrangement. The �B site is bound by NF-�B familymembers and was found to play an important role in activat-ing � rearrangement of recombination substrates in cell linetransfection and transgenic studies (16, 25). However, thefindings that the impacts of �E1 and �E2 double mutationon � rearrangement are very similar to the deletion of theentire iE� rule out any important roles of other protein-binding motifs within iE�, such as the NF-�B binding site,in activating � rearrangement. Based on the findings that iE�

and 3�E� are the essential elements activating � rearrange-ment and that there is no NF-�B binding site within 3�E�, it

remains unclear how NF-�B family transcription factorsmight activate � rearrangement. Based on the findings thatNF-�B is activated during B cell activation induced bycross-linking of the antigen receptor (for review see refer-ence 26), the potential roles of the �B site in receptor editingand somatic hypermutation of � should be examined.

Our findings provide unambiguous evidence that the�E1 and �E2 sites play critical roles in activating � rear-rangement. E2A family transcription factors are the primarytranscription factors binding to the �E1/2 sites, and whencoexpressed with the RAG proteins, are able to activate �

Figure 4. Impacts of the �E1/�E2 double mutation (m�E1/2)on � rearrangement. (A) South-ern blotting analyses of m�E1/2heterozygous ES cells. The en-zymes and probe used are thesame as in Fig. 1. Genomic DNAderived from WT, PGK-neor–inserted and PGK-neor–deletedm�E1/2 ES cells are shown inlanes 1, 2, and 3, respectively.(B) PCR analysis of V� J� rear-rangement at the WT and mutantalleles. The strategy for the rear-rangement PCR assay is identicalto that described in Fig. 2. Rep-resentative data derived from ��

splenic B cells from m�E1/2–Rag2�/� chimeric mice and het-erozygous germline mice areshown in lane 3 and 4, respec-tively. As PCR controls, PCRproducts of genomic DNA de-rived from WT splenic B cellsand m�E3 splenic B cells areshown in lanes 1 and 2, respec-tively. (C) Statistical analysis ofthe ratio of the rearrangementfrequency of the mutant alleleversus that of the WT allele.Mean values derived from fourindependent experiments areshown with error bars. (D) Rep-resentative Southern blottinganalysis of hybridomas derivedfrom spleen m�E1/2 B cells. Totalnumbers of the rearranged WTand mutant allele in 31 hybrid-omas are shown. White lines in-dicate that intervening lanes havebeen spliced out. (E) PCR analysisof V� J� rearrangement of theWT and mutant allele in m�E1/2pre–B cells (lane 2). As a control,the rearrangement of WT andmutant alleles in m�E3 splenicB cells is also shown (lane 1). (F)Relative � expression in ��

B cells derived from WT andm�E1/2 homozygous mutantB cells. � transcripts were ana-lyzed by quantitative real-timePCR. mRNA levels were nor-malized to � expression. Averageof three RNA purifications areshown with error bars.

Dow

nloaded from http://rupress.org/jem

/article-pdf/200/9/1205/1152365/jem20091205.pdf by guest on 03 Septem

ber 2021

Inlay et al. Brief Definitive Report1211

rearrangement in a non–B cell line (24). Therefore, ourfindings provide a mechanism for how E2A family tran-scription factors might function in activating � rearrange-ment. Based on the findings that E2A can interact with his-tone acetyltransferases such as p300/CBP and the SAGAcomplex (27–29), �E1/2 might activate � rearrangementby increasing histone acetylation of the � locus.

The authors thank David Baltimore, Cornelis Murre, and AnnFeeney for helpful discussion and critical reading of the manuscript.

This work was supported by a National Institutes of Health grant(AI44838) to Y. Xu.

The authors have no conflicting financial interests.

Submitted: 7 June 2004Accepted: 23 August 2004

References1. Bassing, C.H., W. Swat, and F.W. Alt. 2002. The mecha-

nism and regulation of chromosomal V(D)J recombination.Cell. 109:S45–S55.

2. Yancopoulos, G.D., T.K. Blackwell, H. Suh, L. Hood, andF.W. Alt. 1986. Introduced T cell receptor variable regiongene segments recombine in pre-B cells: evidence that B andT cells use a common recombinase. Cell. 44:251–259.

3. Parslow, T.G., and D.K. Granner. 1983. Structure of a nuclease-sensitive region inside the immunoglobin kappa gene: evidencefor a role in gene regulation. Nucleic Acids Res. 11:4775–4792.

4. Emorine, L., M. Kuehl, L. Weir, P. Leder, and E.E. Max. 1983.A conserved sequence in the immunoglobulin J kappa-C kappaintron: possible enhancer element. Nature. 304:447–449.

5. Queen, C., and D. Baltimore. 1983. Immunoglobulin genetranscription is activated by downstream sequence elements.Cell. 33:741–748.

6. Picard, D., and W. Schaffner. 1984. A lymphocyte-specificenhancer in the mouse immunoglobulin kappa gene. Nature.307:80–82.

7. Queen, C., and J. Stafford. 1984. Fine mapping of an immu-noglobulin gene activator. Mol. Cell. Biol. 4:1042–1049.

8. Meyer, K.B., and M.S. Neuberger. 1989. The immunoglob-ulin kappa locus contains a second, stronger B-cell-specificenhancer which is located downstream of the constant re-gion. EMBO J. 8:1959–1964.

9. Liu, Z.-M., J.B. George-Raizen, S. Li, K.C. Meyers, M.Y.Chang, and W.T. Garrard. 2002. Chromatin structural analy-ses of the mouse Igk gene locus reveal new hypersensitivesites specifying a transcriptional silencer and enhancer. J. Biol.Chem. 277:32640–32649.

10. Xu, Y., L. Davidson, F.W. Alt, and D. Baltimore. 1996. De-letion of the Ig kappa light chain intronic enhancer/matrixattachment region impairs but does not abolish V kappa Jkappa rearrangement. Immunity. 4:377–385.

11. Inlay, M., F.W. Alt, D. Baltimore, and Y. Xu. 2002. Essen-tial roles of the kappa light chain intronic enhancer and 3�enhancer in kappa rearrangement and demethylation. Nat.Immunol. 3:463–468.

12. Yi, M., P. Wu, K.W. Trevorrow, L. Claflin, and W.T. Gar-rard. 1999. Evidence that the Igk gene MAR regulates theprobability of premature V-J joining and somatic hypermuta-tion. J. Immunol. 162:6029–6039.

13. Gorman, J.R., N. vanderStoep, R. Monroe, M. Cogne, L.Davidson, and F.W. Alt. 1996. The Ig kappa 3� enhancer in-

fluences the ratio of Ig kappa versus Ig lambda B lympho-cytes. Immunity 5:241–252.

14. Sen, R., and D. Baltimore. 1986. Multiple nuclear factors in-teract with the immunoglobulin enhancer sequences. Cell.46:705–716.

15. Schlissel, M.S., and D. Baltimore. 1989. Activation of immuno-globulin kappa gene rearrangement correlates with induction ofgermline kappa gene-transcription. Cell. 58:1001–1007.

16. Demengeot, J., E.M. Oltz, and F.W. Alt. 1995. Promotionof V(D)J recombinational accessibility by the intronic E kappaelement: role of the kappa B motif. Int. Immunol. 7:1995–2003.

17. Scherer, D.C., J.A. Brockman, H.H. Bendall, G.M. Zhang,D.W. Ballard, and E.M. Oltz. 1996. Corepression of RelA andc-rel inhibits immunoglobulin kappa gene transcription and re-arrangement in precursor B lymphocytes. Immunity. 5:563–574.

18. O’Brien, D.P., E.M. Oltz, and B.G. Van Ness. 1997. Coordi-nate transcription and V(D)J recombination of the kappa immu-noglobulin light-chain locus: NF-kappaB-dependent and -inde-pendent pathways of activation. Mol. Cell. Biol. 17:3477–3487.

19. Ephrussi, A., G.M. Church, S. Tonegawa, and W. Gilbert.1985. B lineage–specific interactions of an immunoglobulinenhancer with cellular factors in vivo. Science. 227:134–140.

20. Church, G.M., A. Ephrussi, W. Gilbert, and S. Tonegawa.1985. Cell-type-specific contacts to immunoglobulin enhanc-ers in nuclei. Nature. 313:798–801.

21. Murre, C., P.S. McCaw, and D. Baltimore. 1989. A newDNA binding and dimerization motif in immunoglobulinenhancer binding, daughterless, MyoD, and myc proteins.Cell. 56:777–783.

22. Bain, G., E.C. Maandag, D.J. Izon, D. Amsen, A.M. Kruis-beek, B.C. Weintraub, I. Krop, M.S. Schlissel, A.J. Feeney,M. van Roon, et al. 1994. E2A proteins are required forproper B cell development and initiation of immunoglobulingene rearrangements. Cell. 79:885–892.

23. Zhuang, Y., P. Soriano, and H. Weintraub. 1994. The helix-loop-helix gene E2A is required for B cell formation. Cell.79:875–884.

24. Romanow, W.J., A.W. Langerak, P. Goebel, I.L. Wolvers-Tettero, J.J. van Dongen, A.J. Feeney, and C. Murre. 2000.E2A and EBF act in synergy with the V(D)J recombinase togenerate a diverse immunoglobulin repertoire in nonlym-phoid cells. Mol. Cell. 5:343–353.

25. Coquilleau, I., P. Cavelier, F. Rougeon, and M. Goodhardt.2000. Comparison of mouse and rabbit Ei kappa enhancersindicates that different elements within the enhancer maymediate activation of transcription and recombination. J. Im-munol. 164:795–804.

26. Gugasyan, R., R. Grumont, M. Grossmann, Y. Nakamura,T. Pohl, D. Nesic, and S. Gerondakis. 2000. Rel/NF-kap-paB transcription factors: key mediators of B-cell activation.Immunol. Rev. 176:134–140.

27. Eckner, R., T.P. Yao, E. Oldread, and D.M. Livingston.1996. Interaction and functional collaboration of p300/CBPand bHLH proteins in muscle and B-cell differentiation.Genes Dev. 10:2478–2490.

28. Qiu, Y., A. Sharma, and R. Stein. 1998. p300 mediates tran-scriptional stimulation by the basic helix-loop-helix activatorsof the insulin gene. Mol. Cell. Biol. 18:2957–2964.

29. Massari, M.E., P.A. Grant, M.G. Pray-Grant, S.L. Berger, J.L.Workman, and C. Murre. 1999. A conserved motif present in aclass of helix-loop-helix proteins activates transcription by directrecruitment of the SAGA complex. Mol. Cell. 4:63–73.

Dow

nloaded from http://rupress.org/jem

/article-pdf/200/9/1205/1152365/jem20091205.pdf by guest on 03 Septem

ber 2021