Immunity, Vol. 23, 203–212, August, 2005, Copyright ©2005 by Elsevier Inc. DOI 10.1016/j.immuni.2005.07.004
A Direct Interaction between the RAG2 C Terminus
and the Core Histones Is Required
for Efficient V(D)J Recombination
Kelly L. West,1Netai C. Singha,1,3
Pablo De Ioannes,1Lynne Lacomis,2
Hediye Erdjument-Bromage,2Paul Tempst,2
and Patricia Cortes1,*
Mount Sinai School of Medicine
1425 Madison Avenue
New York, New York 10029
2Molecular Biology Program
Memorial-Sloan-Kettering Cancer Center
1275 York Avenue
New York, New York 10021
V(D)J recombination is a tightly controlled process of
somatic recombination whose regulation is mediated
in part by chromatin structure. Here, we report that
RAG2 binds directly to the core histone proteins. The
interaction with histones is observed in developing
lymphocytes and within the RAG1/RAG2 recombi-
nase complex in a manner that is dependent on the
RAG2 C terminus. Amino acids within the plant ho-
meo domain (PHD)-like domain as well as a con-
served acidic stretch of the RAG2 C terminus that is
considered to be a linker region are important for this
interaction. Point mutations that disrupt the RAG2-
histone association inhibit the efficiency of the V(D)J
recombination reaction at the endogenous immuno-
globulin locus, with the most dramatic effect in the V
V(D)J recombination is a site-specific somatic recombi-
nation reaction essential for the generation of a diverse
immune repertoire. During recombination, one each of
a number of different variable (V), diversity (D), and join-
ing (J) gene segments are brought together to form a
single continuous segment that codes for the variable
portion of immunoglobulin or T cell-receptor molecules.
During B and T cell development, seven different loci
undergo rearrangement: the immunoglobulin (Ig) heavy
chain, the κ light chain and the λ light chain loci, as well
as the T cell receptors (TCR) TCRα, β, γ, and δ loci.
All seven loci are recombined by a common lymphoid
recombinase, yet in a highly regulated cell-type- and
developmental-stage-specific manner. The recombi-
nase consists of RAG1 and RAG2 (protein products of
recombinase activating genes 1 and 2). The RAG1/2
complex is lymphoid specific and carries out DNA
cleavage to form a double-strand break, which is then
repaired by ubiquitous nonhomologous end joining
(NHEJ) proteins. These include Ku70/80, DNA-PKcs,
3Present address: Signal Transduction Program, The Burnham In-
stitute, La Jolla, California 92037.
XRCC4, LigaseIV, and Artemis. Deficiencies in RAG1,
RAG2, LigaseIV, or Artemis proteins lead to a severely
weakened immune system with agenesis of B and/or T
cells, a phenomenon known as severe combined immu-
nodeficiency (SCID) (reviewed in de Villartay, 2002;
Schwarz et al., 2003).
An active DNA recombinase can be reconstituted
in vitro with core RAG1 (amino acids 384–1008 of 1040)
and core RAG2 (amino acids 1–382 of 527). The “core”
proteins are the minimal regions necessary for recombi-
nation of exogenous plasmid substrates in nonlym-
phoid cells (Cuomo and Oettinger, 1994; Sadofsky et
al., 1994, 1993; Silver et al., 1993) and have been more
widely studied than their full-length counterparts due
to technical difficulty in purification of full-length RAG1
and RAG2. However, several studies suggest important
functions for the noncore regions of RAG1 and RAG2
including ubiquitin ligase activity (Jones and Gellert,
2003; Yurchenko et al., 2003), nuclear import (Ross et
al., 2003), cell cycle regulated protein degradation (Lee
and Desiderio, 1999), and suppression of transposition
(Elkin et al., 2003; Swanson et al., 2004; Tsai and
Schatz, 2003), though transposition remains a topic of
debate (Jiang et al., 2004). It was shown in B cell lines
that the C terminus of RAG2 was important for IgH V
to DJ rearrangement (Kirch et al., 1998). This finding is
supported by mouse models (Akamatsu et al., 2003; Li-
ang et al., 2002). RAG2 core knockin mice exhibit re-
duced overall numbers of T and B cells and partial de-
velopmental blocks at the progenitor cell stage. When
the sequences of the rearrangement products in cells
derived from these mice were analyzed, it was found
that both signal and coding joints are affected and are
highly imprecise (Talukder et al., 2004). A core RAG1
knockin mouse (Dudley et al., 2003) exhibited partial
developmental blocks at the progenitor cell stage, dra-
matic reduction of both IgH D-J and V-DJ rearrange-
ments, and imprecise coding joints (Talukder et al.,
As evidence is accumulating that the core RAGs are
less efficient and less regulated than their full-length
counterparts at recombination of endogenous loci, it
has become increasingly important to study the full-
length RAG proteins. Here, we present copurification of
endogenous RAG2 and interacting proteins from thy-
mocytes and a detailed biochemical study of full-length
RAG2. We report the novel and significant finding that
full-length RAG2, via its C-terminal domain, binds di-
rectly to the core histones H2A, H2B, H3, and H4.
Histones compose the fundamental unit of chroma-
tin, the nucleosome, in which two H2A/H2B dimers and
one H3/H4 tetramer bind to form an octamer and
around which approximately 150 base pairs of DNA are
then wrapped. The “accessibility hypothesis” (Yanco-
poulos and Alt, 1985) of V(D)J recombination proposes
that chromatin structure may regulate the recombina-
tion reaction by shielding particular recombination sig-
nal sequences (RSS) in cell types and/or in develop-
mental stages for which their recombination would be
inappropriate. In support of the accessibility hypothe-
sis, many studies have demonstrated correlations be-
tween transcription, histone modifications, and recom-
bination (Roth and Roth, 2000). Nonlymphoid immuno-
globulin and TCR loci can be induced to rearrange by
RAGs in concert with lymphoid-specific transcription
factors (Choi et al., 1996; Ghosh et al., 2001; Goebel et
al., 2001; Langerak et al., 2001; Romanow et al., 2000),
and targeted deletion of promoters or enhancers alters
recombination frequency (McMurry et al., 1997; Sikes
et al., 2002, 1999; Whitehurst et al., 2000). In addition,
distinct patterns of histone modifications have been
observed around actively recombining loci (Johnson et
al., 2003, 2004; McMurry and Krangel, 2000; Morshead
et al., 2003; Muegge, 2003). Finally, the nucleosome
blocks V(D)J recombination by core RAG1/2 in vitro
(Golding et al., 1999; Kwon et al., 1998, 2000), which
may be slightly alleviated by the chromatin-remodeling
activities of SWI/SNF and ISWI (Kwon et al., 2000; Pat-
enge et al., 2004).
Here, we present evidence that the RAG1/RAG2 re-
combinase, via the noncore region of RAG2, binds di-
rectly to the fundamental unit of chromatin: histones.
We explore an acidic domain in the C-terminal region
of RAG2 and show that it is critical both for interaction
with histone and for complete recombination of the IgH
locus in B cells. In summary, our data describe a novel
protein-protein interaction involving full-length RAG2
that is direct, occurs between endogenous proteins in
developing lymphoid cells, and is required for efficient
RAG2 Binds Core Histones via Its C Terminus
Initial studies to identify novel RAG2 C-terminal (R2CT)
binding proteins were carried out in 293T cells. DNA con-
structs encoding a Flag-thioredoxin tag (FT) or FT-tagged
R2CT were transfected, and proteins immunoprecipitated
(IP) with anti-Flag beads. Low-molecular-weight bands
that specifically co-IP with the R2CT protein were visual-
ized by silver stain (Figure 1A) and identified by mass
spectrometry analysis to be the core histones H2A,
H2B, H3, and H4. To investigate the specificity of the
histone interaction, Flag IP was performed with the
Flag-GST tag (FG) or FG-tagged R2CT (R2CT), RAG2
core (R2CR), and RAG2 full-length (R2FL). Silver stain
revealed the same pattern of low-molecular-weight pro-
teins as those previously identified to be histones,
which coimmunoprecipitate with R2CT and R2FL, but
not with the R2CR or the FG tag alone (Figure 1B).
We confirmed and expanded these results by West-
ern blotting analysis. We tested a wide panel of anti-
bodies to histone variants. Of the antibodies tested,
anti-acetylated histone H3 was the most sensitive and
was used to confirm the histone interactions observed
by silver stain (Figure 1C). It was also used as a marker
of histones throughout the study. However, in an initial
characterization of RAG2 bound histones, we carried
out a large-scale IP of tagged R2CT as well as the
negative control Flag-NLS-thioredoxin tag and per-
formed Western blotting analyses for several histone
variants (Figure 1D). Acetylated versions of histones
H2B, H3, and H4 were detected in the R2CT IPs. Meth-
ylated versions of histone H3 (trimethyl K27 and dimethyl
K4) were also found to co-IP with R2CT (Figure 1D).
The data presented in Figure 1 show that full-length
RAG2, through its C terminus, interacts with the core
histones and that several histone variants coimmuno-
precipitate with R2CT.
The C Terminus of RAG2 Binds Directly to Each
of the Four Core Histones H2A, H2B, H3, and H4
To investigate whether the RAG2-histone interaction is
direct, GST or GST-tagged R2CT (R2CT) were purified
from bacteria and immobilized on glutathione beads.
Beads were incubated with core histones purified in
bulk from HeLa cell nuclei. Coomassie staining of an
18% polyacrylamide gel revealed that the R2CT, but not
the GST tag, binds directly to the core histones (Figure
2A). The migration pattern of the four pure histones is
shown on the left for reference, with the amount loaded
equal to the amount used in the binding reaction (in-
put). The results clearly demonstrate that the C termi-
nus of RAG2 mediates a strong, direct interaction with
the core histones.
The core histones do not exist as monomers in solu-
tion but form H3/4 tetramers and H2A/B dimers (Eick-
bush and Moudrianakis, 1978) and potentially octamers
as well (Luger et al., 1997). As such, we wished to ask
which of the four could be responsible for the observed
direct binding to RAG2. Histones H2A, H2B, H3, and H4
purified individually from calf thymus were used for
in vitro binding. We observed that each of the four indi-
vidual histones was capable of efficiently binding to the
R2CT (Figure 2B). The binding was, however, influenced
by salt concentration; efficient binding of H3 was stable
in conditions tested up to 1.25 M KCl, whereas H4 bind-
ing was efficient up to 750 mM KCl, and neither H2A
nor H2B binding was observed at concentrations
greater than 100 mM KCl (data not shown).
Core histones undergo extensive posttranslational
modifications, which have an important role in modulat-
ing protein-protein interactions. To address this issue,
recombinant histones produced in bacteria were as-
sayed as above. We observed that, as was the case
with calf histones, each of the four recombinant core
histones was capable of binding to the R2CT (Figure
2C), although the efficiency of binding was reduced.
Although the in vitro data demonstrates that RAG2 is
capable of binding any of the four histones and can do
so independently of eukaryotic posttranslational modi-
fication, a consistently stronger interaction with H3 and
H4 was observed. The requirements for binding in vivo
may be quite different, however, and may have func-
tional consequences. Although defining the molecular
basis of the RAG2-histone interaction is an important
area of future study, we wished to first address the sig-
nificance of this interaction.
RAG2 Interacts with Histones in Thymocytes
To investigate the physiological significance of the
RAG2-histone interaction, thymocytes were collected
from 2- to 3-week-old mice whose lymphocytes were
actively undergoing V(D)J recombination. RAG2 IP was
performed by using both immune serum and purified
antibodies. Western blotting analysis shows successful
RAG2 Interacts with Histones
Figure 1. RAG2 Interacts with Core Histones
(A) FT or R2CT proteins were immunoprecipitated with anti-Flag beads from transfected 293T cells, and protein was visualized by silver stain.
Arrowheads indicate FT and R2CT proteins. Bands representative of those identified by mass spectrometry to be histones H2A, H2B, H3,
and H4 are indicated by the bracket. Molecular weight in kDa is at the left.
(B) As in (A), with arrowheads indicating FG or FG-tagged R2CT (R2CT), R2CR, and R2FL.
(C) FG or FG-tagged R2CT (R2CT), R2CR, and R2FL IP samples were probed with antibodies to Flag (top) and histone H3. 300 ng of purified
H3 was used as a positive control for Western blotting.
(D) FNT or FNT-tagged RAG2CT (R2CT) protein immunoprecipitates were separated on an 18% polyacrylamide gel, and several membranes
were prepared for Western blotting with antibodies to acetyl-Histone H2B (AcH2B), acetyl-Histone H4 (AcH4), acetyl-Histone H3 (AcH3),
trimethylated-Histone H3 (TriMeH3), and dimethylated-Histone H3 (DiMeH3), as well as to the Flag control. As a positive control for Western
blotting, 1–2 ?g purified calf thymus histone was run (+ve).
IP of endogenous full-length RAG2 from thymocytes
and that endogenous histone protein does co-IP with
RAG2 (Figure 3). This data supports and expands our
Figure 2. Direct Interaction between RAG2CT and the Core His-
(A) GST or GST-tagged R2CT (R2CT) was immobilized on GST beads
and incubated with buffer alone (GST, R2CT) or with core histones
purified from HeLa cell nuclei (GST+Histones, R2CT+Histones). Sam-
ples were run on SDS-PAGE and visualized by Coomassie blue
stain. The pure core histones were run on the left, with the amount
equal to the input of the reaction (Input).
(B) Independent RAG2 binding to each of the four core histones
H2A, H2B, H3, and H4 was assayed as above, with 1 ?g of each
histone individually purified from calf thymus.
(C) Binding assays were carried out as above, with 1 ?g recombi-
nant human histones purified from a bacterial source.
293T cell and in vitro observations and together de-
monstrate a strong, direct, and physiologic interaction
between the noncore region of RAG2 and histone.
RAG1/2 Recombinase Interacts
with Histones via R2CT
To characterize further the observed RAG2-histone in-
teraction in respect to its potential function in V(D)J re-
combination, we investigated whether RAG2 interacts
with histones within the context of the RAG1/RAG2
complex, which constitutes the active recombinase.
293T cells were cotransfected with Flag-NLS-thiore-
doxin-tagged RAG1 full-length (FR1) and either HA-thi-
oredoxin-tagged RAG2 core (FR1+R2CR) or HA-thiore-
doxin-tagged RAG2 full-length (FR1+R2FL). IPs were
done with anti-Flag antibody to ensure that any RAG2
Figure 3. RAG2 Interacts with Histone in Thymocytes
Thymocytes were harvested from 2- to 3-week-old mice, and cell
lysates were prepared. RAG2 was immunoprecipitated by using
rabbit immune serum (R2 Serum) or antibody purified from the
same serum (R2 Purified). Preimmune serum and a purified rabbit
IgG isotype control were included as negative controls. Proteins
were separated by SDS-PAGE and visualized by Western blotting
for RAG2 (top) and histone (bottom). 293T whole-cell extract (WCE)
was used as a control for Western blotting.
Figure 4. RAG1/2 Recombinase Interacts with Histone, Dependent
on the R2CT Domain
Flag IPs were preformed from cells expressing FR1, or coexpress-
ing FR1 and HA-thioredoxin-tagged RAG2 core (R2CR) or R2FL.
FR2 immunoprecipitate was included as a positive control. Proteins
were detected by Western blotting for RAG2 (top), Flag (middle),
and histone (bottom). WCE was used as a control for Western
immunoprecipitated in the assay was bound to RAG1.
As a negative control, FR1 was transfected alone (FR1),
whereas as a positive control Flag-NLS-thioredoxin-
tagged RAG2 full-length was transfected and immuno-
precipitated through Flag (FR2). Samples were immu-
noblotted for Flag, RAG2, and histone (Figure 4). We
observed that when the full-length RAG2 is brought
down through RAG1, the histone copurifies, whereas
RAG1 bound to R2CR is not sufficient to do so and
neither is RAG1 alone. These results indicate that the
R2CT domain links the RAG1/2 recombinase to histone,
further suggesting an important mechanistic role for
this interaction in the process of V(D)J recombination.
Amino Acids within an Acidic Patch of the R2CT
Are Important for Histone Interaction
We next developed a system to address what would be
the effect on V(D)J recombination if the histone interac-
tion was disrupted. Although RNA interference and tar-
geted protein knock down would be the most direct
approach to answer such a question, this technique is
problematic when studying proteins like the histones,
which are ubiquitous and critical for cell survival. Our
alternative approach was to first undertake a mapping
study to identify the region of RAG2 required for the
histone interaction, with the ultimate goal of generating
full-length RAG2 harboring minimally disruptive single
amino-acid point mutations that would abolish the his-
tone interaction. Mutant RAG2 would then be studied
in a B cell system.
Mapping analysis included truncations from either
the N- or C-terminal ends of the noncore region of
RAG2 as shown in Figure 5A. The full-length protein is
shown at the top and is schematically divided into its
putative domains by vertical lines. Sequence alignment
suggests that the RAG2 core (amino acids 1–387) forms
a six-bladed β propeller whereas the C-terminal domain
folds to a plant homeo domain (PHD)-like tertiary struc-
ture due to the contributions of a set of conserved cys-
teines and histidines between amino acids 420 and 480
Figure 5. Mapping of the R2CT for RAG2-Histone Interaction
(A) Schematic of constructs used in the mapping study. All con-
structs maintain an N-terminal FNT tag. Full-length RAG2, ending
at amino acid 527, is shown at the top, and vertical lines demarcate
the end of the core domain (387) and the PHD-like region (w420–
480). Numbers to the right of the protein constructs indicate the
specific amino acids that they span. Histone binding as determined
by Western blot in (B) is summarized as present (+), reduced (±), or
(B) Constructs shown in (A) were transfected into 293T cells, and
Flag IP was followed by Western blotting analysis for Flag (top) or
histone (bottom). WCE was used as a positive control for Western
(Callebaut and Mornon, 1998). This is supported by the
recent solution of the structure of a part of the RAG2 C
terminus, which demonstrates the presence of a non-
cannonical PHD finger from amino acids 419 to 481
(Elkin et al., 2005). The stretch of amino acids from 387
to 420, falling between the core and the PHD-like region
of the C terminus, has been defined as a “hinge” or
linker holding the two structural domains together
(Callebaut and Mornon, 1998). In Figure 5A, the end of
the core region is indicated by a vertical line at amino
acid 387, whereas the PHD-like region is demarcated
by vertical lines at amino acids 420 and 480. All con-
structs were tagged with Flag, a nuclear localization
signal, and thioredoxin (FNT). Anti-Flag IP was followed
by SDS-PAGE and Western blotting for Flag and his-
tone (Figure 5B).
We found that truncations from either the N- or the
C-terminal ends of the R2CT could interfere with his-
tone binding. The construct RAG2(1–485) maintained
the histone interaction, yet the interaction was lost on
truncation to amino acid 477. In truncating the RAG2
protein from amino acids 485 to 477, we delete a con-
served cysteine (478) and a histidine (481) that are among
the last conserved essential residues of the PHD-like
motif. It is possible that this domain may be important
for the observed histone interaction, and in fact PHD
domains of many different proteins have been shown
RAG2 Interacts with Histones
Figure 6. Point Mutation of Amino Acids
within a Conserved Acidic Region of RAG2
Reduces Histone Binding
(A) Amino acid sequence alignment of RAG2
protein from five different species. Residues
397–408 are boxed.
(B) IP was preformed on 293T cells express-
ing FNT or FNT-tagged RAG2 core (R2CR),
R2CT, wild-type RAG2 full-length (R2FL), and
point mutant RAG2 full-length (R2D397A–
R2D408A) proteins. Western blotting analysis
is shown for Flag (top) and histone (bottom).
(C) Flag IP was carried out from 293T cells
coexpressing various FNT-RAG2 constructs
and the Strep-RAG1 core. Western blotting
analysis is shown for Flag (top) and RAG1
(D) Cleavage reactions were performed with
protein preparations shown in (C) and a ra-
diolabeled 12-RSS substrate. Nicked (14 nu-
cleotide) and hairpin (24 nucleotide) prod-
ucts were resolved on a 16% denaturing
polyacrylamide gel and visualized by autora-
to play important roles in protein-protein interactions
involving gene regulation. Here, however, we will focus
our attention on a second observation: that amino
acids within the hinge region of RAG2 may be important
for the interaction with histones.
With the C terminus held constant (at 493 or 496 be-
cause this gives a higher expression level than con-
structs ending at 527), we made truncations from the N
terminus of the noncore region of RAG2, as shown
in the bottom half of Figure 5A. We observed that
RAG2(397–493) maintained the histone interaction well,
but the interaction was noticeably reduced upon further
truncation from 397 to 403 and lost altogether upon
truncations beyond. A construct ending at amino acid
408 consistently expressed at lower levels, yet, when
proportionately more IP sample was run to normalize
for RAG2 protein, still no interaction was observed
(data not shown).
To further investigate this region, we aligned RAG2
amino acid sequences from five different species (Fig-
ure 6A). The twelve amino acids between 397 and 408
are boxed. It is notable that there is a high degree of
evolutionary conservation among amino acids in this
region, suggesting it may serve an important function,
either directly mediated or indirectly by contributing to
the structure and/or stability of the C terminus in gen-
eral. Also, there is a high percentage of acidic amino
acids. In fact, these residues fall at the end of a larger
region that was originally noted upon cloning of RAG2
to contain 42% acidic amino acids and was suggested
to be similar to acidic activation domains of many tran-
scription factors (Oettinger et al., 1990; Silver et al.,
1993). Further, amino acids 388–450 are characterized
as being among the best conserved within RAG2 (Sa-
dofsky et al., 1994). Only one study has directly ad-
dressed the importance of amino acids in the acidic
region, and it suggests that this region may have a role
in recombination within the chromosomal context (Sil-
ver et al., 1993). The acidic content of the region was
further significant to us because it is known that many
histone chaperones rely on acidic patches of amino
acids for interaction with the basic histone proteins (re-
viewed in Akey and Luger, 2003).
We generated a panel of twelve mutants, individually
mutating each amino acid between 397 and 408 to ala-
nine within the context of full-length RAG2. We found
that mutation of four of the twelve amino acids to ala-
nine (Y402A, N403A, D406A, or E407A) significantly re-
duced the interaction of RAG2 with histone (Figure 6B).
Six of the twelve mutants were selected for further
study: two “positive control mutants” whose histone in-
teraction was not observed to be altered (R2D397A and
R2D408A) as well as the four mutants with reduced his-
tone interaction (R2Y402A, R2N403A, R2D406A, and
R2E407A). Mutants were assayed for their ability to in-
teract with RAG1 by cotransfection of FNT-RAG2 con-
structs with the Strep-RAG1 core followed by IP with
Flag beads and Western blotting for Flag and RAG1.
We observed that none of the mutations disrupted
RAG1 binding (Figure 6C). A trans cleavage assay
(Hiom and Gellert, 1998) revealed no differences be-
tween the various mutants and controls in terms of their
ability to nick the substrate or to generate a double
strand break (Figure 6D).
Diminished Histone Interaction Correlates
with Impaired Recombination
at the Endogenous Loci of B Cells
We next studied the activity of RAG2 point mutants in
a physiologically relevant setting: the endogenous lo-
cus of developing B cells. We established a B cell sys-
tem by using pro-B cells derived from RAG2−/−mice
(Kirch et al., 1998). Previously, it had been shown that
knockout of RAG2 in mice halts B and T cell develop-
ment at the progenitor cell stage with germline configu-
ration immunoglobulin and TCR loci (Shinkai et al.,
1992). However, induction of RAG2 protein expression
was sufficient to drive recombination and cell matura-
tion within one to two months (Shinkai et al., 1992;
Shockett et al., 2004). We transduced RAG2−/−cell lines
generated from knockout mice with retrovirus carrying
the FNT tag or various FNT-tagged RAG2 constructs
followed by an internal ribosomal entry site (IRES) and
a puromycin-resistance cassette. Eleven separate B
cell lines were engineered to include three negative
controls for histone IP (GFP, FNT, and R2CR) and two
positive controls (R2CT and R2FL), as well as the six
mutants characterized above in Figures 6C and 6D
(D397A, Y402A, N403A, D406A, E407A, and D408A).
The cell lines described above were selected with
puromycin and harvested approximately 4- to 8-weeks
posttransduction. We observe that as in 293T cells, the
RAG2WT, R2CT, and mutants R2D397A and R2D408A
interact with histone, whereas the mutants Y402A,
N403A, D406A, and E407A do not (Figure 7A). We also
observed that the R2D408A mutant has a slightly re-
duced histone interaction in this B cell system.
To assess recombination of the endogenous immu-
noglobulin locus in these B cell lines, we harvested
genomic DNA and performed polymerase chain reac-
tion (PCR) and Southern blotting analyses as described
(Schlissel et al., 1991). We examined IgH D to J re-
arrangement and V to DJ rearrangement of VHfamilies
Q52 and 7183, as well as Vκ to Jκ rearrangement in the
immunoglobulin light chain locus (Figure 7B). These
loci were selected because they are among the best
characterized in such rearrangement assays. The ex-
periment was repeated several times, and we observed
three distinct and reproducible trends for our histone
noninteracting RAG2 mutants Y402A, N403A, D406A,
and E407A: a pronounced defect in V to DJ rearrange-
ment at the IgH locus, a mild defect in IgH D to J re-
arrangements, and no visible impairment in Vκ to Jκ
rearrangements. The positive control mutants D397A and
D408A, which retain histone interaction, recombine all
four loci studied. Of note, the mutant D408A exhibits
slightly reduced V to DJHrearrangement in comparison
to the controls R2FL and R2D397A, which correlates
with the slightly reduced histone interaction seen in
Figure 7A. Expression of the control wild-type RAG2
induces recombination of all four loci, whereas the
R2CR control recombines D to J and Vκ to Jκ, but not
V to DJHas has been previously described (Akamatsu
et al., 2003; Kirch et al., 1998; Liang et al., 2002). Impor-
Figure 7. RAG2 Mutants Unable to Interact with Histone in B Cells
Exhibit Defects in V(D)J Recombination at the Endogenous Immu-
RAG2−/−cell lines were retrovirally transduced and selected to ex-
press constructs of interest: green fluorescent protein (GFP), FNT
or FNT-tagged RAG2 core (R2CR), R2CT, R2FL, and point mutant
RAG2 full-length (mutated amino acids 397, 402, 403, 406, 407, and
408). Retroviral transduction and selection of these cell lines were
done in duplicate to control for experimental variability. All subse-
quent experiments were performed with both sets of cell lines and
no significant differences were observed between the two.
(A) Flag immunoprecipitates were analyzed by Western blotting for
Flag (top) and histone (bottom).
(B) Genomic DNA was harvested from the above cell lines and
PCRs were carried out to amplify products of IgH, D to J; IgL, Vκ
to Jκ; IgH, VHQ52 to DJ; and IgH, VH7183 to DJ rearrangements,
as indicated to the left of each panel. 400 ng, 200 ng, or 25 ng of
DNA template was used in the reactions as indicated by triangles.
For controls FNT and R2CT, the result obtained with 400 ng of tem-
plate DNA is shown. A fragment of RAG1 was amplified as a load-
ing control (bottom).
tantly, all mutants retain the ability to interact with
RAG1 and to cleave DNA (Figures 6C and 6D). Together,
our data show a novel connection between the ability
of RAG2 to bind histones and the capacity of the re-
combinase to efficiently carry out complete V(D)J re-
Although the noncore region of RAG2 is clearly impor-
tant for efficient V(D)J recombination, the full-length
RAG2 protein has presented a significant biochemical
challenge to investigators in the field, and its analysis
has been limited. We report here purification of full-
length RAG2 from 293T cells, B cells, and thymocytes
and the identification of novel full-length RAG2 interact-
ing proteins: the core histones. We show the purifica-
tion of a RAG2 protein complex from thymocytes of
young mice and demonstrate that endogenous RAG2
RAG2 Interacts with Histones
associates with endogenous histone in this recombin-
ing cell population. We demonstrate that full-length
RAG2 simultaneously binds histone and RAG1 and that
the histone binding can be directly mediated by any of
the four core histones. Together, this suggests a poten-
tial unique role for the R2CT as a direct bridge between
chromatin and the recombinase during chromosomal
V(D)J recombination. Mapping and mutagenesis studies
revealed the importance of the PHD domain and a
highly conserved patch of acidic amino acids in the
R2CT. Focusing on the acidic region of RAG2, we show
that residues 397 through 408 are important for the his-
tone interaction and that mutation of any one of the
amino acids Y402, N403, D406, or E407 to alanine
within the context of full-length RAG2 diminishes this
observed histone interaction. Finally, these point mu-
tant RAG2 proteins, although still capable of binding
RAG1, cleaving DNA, and recombining Vκ-Jκ, are mod-
estly reduced in their ability to recombine DHto JHgene
segments and are severely impaired in their ability to
recombine VHto DJHgene segments within the endog-
enous immunoglobulin locus.
It now remains to investigate the numerous possible
roles that this important protein-protein interaction
could serve. We propose three possibilities. First, the C
terminus of RAG2 may stabilize RAG1/2 binding to the
RSS by interacting directly with nearby histones. This
stabilization may be particularly important for RSSs
that are less easy to recombine due to intrinsic se-
quence variation (Liang et al., 2002) and/or location in
the chromatin context. Alternatively or in addition, the
C terminus of RAG2 could recognize histones that bear
specific posttranslational modification patterns, bring-
ing the recombinase to or stabilizing it at very specific
RSSs that would otherwise not recombine. In this man-
ner, a particular RSS could be “marked” for recombina-
tion by modification of nearby histones, and RAG2
would play an important role in deciphering this histone
code to efficiently recruit the recombinase to the appro-
priate locus at the proper time. As a specific example,
it was shown that induction of VHto DJHrecombination,
an event that is tightly restricted to the B cell lineage,
requires Pax5 signaling to induce a specific change in
histone modification at the VHlocus (Johnson et al.,
2004). Mutant RAG2 that is not able to bind histones
may be “blind” to this code and restricted to recombi-
nation of gene segments that are not so precisely regu-
lated. Finally, a third possibility is that RAG2-histone
interaction could also serve an important role in the
postcleavage phases of the recombination reaction.
Signal and coding joints from core RAG2 knock-in mice
have large deletions and/or additions, suggesting that
the noncore region may be necessary for proper sta-
bility or protection of DNA ends during the processing
and joining phases of the reaction (Talukder et al.,
2004). Perhaps this postcleavage stability is conferred,
in part, by the RAG2-histone interaction. A second po-
tential postcleavage role for RAG2 is suggested by its
acidic region, which is reminiscent of acidic regions
found in histone chaperone proteins (Akey and Luger,
2003). Reformation of a nucleosome by histone chaper-
ones after DNA damage repair is required for cell sur-
vival (Peterson and Cote, 2004); RAG2 could play a sim-
ilar role in depositing histones back onto repaired DNA
after recombination, helping to complete recombina-
tion of specific loci and permitting protein expression,
cell survival, and expansion. In addition to the observed
direct interaction with histones, it is also important to
note that the R2CT could mediate a number of other
direct or indirect protein-protein interactions to assem-
ble a larger complex of which the histone proteins are
a part. It will be an interesting challenge for future
studies to address this issue by identifying other in-
teracting proteins to precisely define the biological
function(s) of RAG2.
A recent study has compared RAG1/RAG2 core and
RAG1/RAG2 full-length activities on naked DNA and on
a polynucleosomal substrate in vitro and concluded
that the full-length RAG2 does not confer any chroma-
tin-specific advantage in RSS cleavage (Patenge et al.,
2004). Although the authors make an important contri-
bution in developing this study of RAG activity on a
polynucleosomal substrate, the results should not be
generalized beyond the scope of the study, which is
based on an in vitro system in which purified RAGs and
the purified SWI/SNF chromatin remodeling activity are
provided to an artificial substrate. Given the estab-
lished importance of DNA sequence, including the RSS,
in precise nucleosome positioning (Baumann et al.,
2003; Simpson, 1991) and the differences that RSS se-
quences make in recombination frequency (Lee et al.,
2003; Montalbano et al., 2003; Nadel et al., 1998), it will
be important to use such a system with DNA substrates
that are equal to those encountered in vivo. Even then,
informative analysis should be carried out indepen-
dently for each RSS, as it is likely that nucleosomes will
map differently on individual gene segments, resulting
in different requirements for effective recombination.
Understanding how the RAG proteins navigate
through chromatin during V(D)J recombination will yield
insight into the mechanisms and regulation of the re-
combination reaction in vivo. It is a complicated issue
that will require a great deal of study by many groups.
Researchers have begun to tease apart the “histone
code” that governs V(D)J recombination, making use of
Chromatin IP (ChIP) assays to identify specific chroma-
tin modifications present at actively recombining or
suppressed loci (Johnson et al., 2003, 2004; Oettinger,
2004). This work suggests the intriguing possibility that
the modification of histones may provide a signal that
helps to direct recombination. ChIP analysis also re-
vealed that the SWI/SNF subunit BRG1 localizes at re-
combining loci (Morshead et al., 2003), and further work
has suggested that the activity of this remodeler may
help to provide RAG1/2 access to a nucleosomal RSS
(Kwon et al., 2000; Patenge et al., 2004). These are im-
portant initial studies, and it is likely that other yet un-
identified factors may participate in concert with RAGs,
histones, and remodeling activities to facilitate V(D)J re-
combination in the face of chromatin, and future studies
may address this possibility. Importantly, the data we
have presented here provides a direct link between the
RAG1/2 recombinase and chromatin, a finding that can
be built upon in future studies: a critical protein-protein
interaction between RAG2 and the core histones.
Antibodies were to the following: Flag (Sigma F-3165), acetyl-His-
tone H3 (Upstate Biotechnology, 06-599), acetyl-Histone H4 (Up-
state Biotechnology, 06-866), acetyl-Histone H2B (Upstate Biotech-
nology, 07-373), Histone H3 (trimethyl K27) (Abcam, ab6002), and
Histone H3 (dimethyl K9) (Abcam, ab7766). Secondary HRP-conju-
gated anti-mouse and anti-rabbit IgG were from Pierce. Flag pep-
tide and anti-Flag beads were from Sigma. Polyclonal anti-RAG2
was prepared against RAG2 (120–527). Monoclonal anti-RAG1 was
prepared against RAG1(630–1040) (Spanopoulou et al., 1995).
Expression Vectors and Mutagenesis
FNT vector backbone was generated by replacing GST with FNT in
the vector pEBG, and RAG2 constructs were subcloned into the
multiple cloning site by using BamHI and NotI. Retroviral vectors
were made from Ret-GST-IRES-Puro by replacing the GST coding
sequence with the insert of interest. Mutagenesis was done with Stra-
tagene’s QuikChange XL Site-Directed Mutagenesis Kit (200517-5).
For bacterial expression, RAG2 (387–493) was amplified from wild-
type cDNA and inserted into the MCS of pGEX6p2 (Amersham Bio-
sciences) by using BamHI and NotI. The Strep-RAG1 core (amino
acids 380–1040) was cloned with BamHI into vector pEF-Strep. De-
tails available upon request.
Cells were lysed in Buffer A-500 (25 mM Tris-Cl [pH 8.0], 500 mM
KCl, 0.5 mM EDTA, 10% glycerol, 1mM DTT, 0.05% Triton-X), soni-
cated, treated with DNase I at 80 ?g/ml for 30 minutes, and ethid-
ium bromide was added to a final concentration of 200 ?g/ml. After
centrifugation, the supernatant was incubated with anti-Flag beads
(Sigma) for 4 hr followed by washing with Buffer A, and elution in
an elution buffer as follows: 20 mM Tris-Cl (pH 7.5), 300 mM NaCl,
0.2 mM EDTA, 0.1% NP-40, 15% glycerol, 0.2 mg/ml Flag peptide.
B cells were harvested and IP performed within one month of in-
FNT-RAG2/Strep-RAG1CR complexes were purified from 293T
cells after calcium phosphate transfection. Cells were resuspended
in Buffer G (20 mM HEPES [pH 7.9], 20% glycerol, 1.5 mM Magne-
sium Chloride, 300 mM Sodium Chloride, and 0.2 mM EDTA), sub-
jected to five rounds of freeze-thaw, and attenuated for 1 hr at 4°C
followed by ultracentrifugation at 40,000 rpm for 30 min. Superna-
tant was bound to an anti-Flag affinity gel (Sigma) for 1.5 hr. Beads
were washed 4× in Buffer G, 2× in Buffer SDB (20 mM Tris [pH 8.0],
200 mM Sodium Chloride, and 20% glycerol), and eluted in Buffer
SDB with 0.2 mg/ml Flag peptide (Sigma).
Gel-resolved proteins were digested with trypsin, batch fraction-
ated on a RP micro-tip, and the peptide mixtures were analyzed
by MALDI-reTOF mass spectrometry (UltraFlex TOF/TOF; BRUKER
Daltonics; Bremen, Germany), as described (Erdjument-Bromage
et al., 1998; Winkler et al., 2002). Selected experimental masses
(m/z) were taken to search a nonredundant protein database (“NR,”
w1.7 × 106entries, National Center for Biotechnology Information,
Bethesda, MD), utilizing the PeptideSearch algorithm (Matthias
Mann, Southern Denmark University, Odense, Denmark). Mass spec-
trometric sequencing of selected peptides was done by MALDI-
TOF/TOF (MS/MS) analysis on the same prepared samples with the
UltraFlex instrument in “LIFT” mode. Fragment ion spectra were
taken to search the NR database with the MASCOT MS/MS Ion
Search program (Matrix Science Ltd., London, UK).
Western Blotting Analysis
Immunoprecipitates were separated on 6–16% SDS-PAGE gradient
gels or 18% continuous gels to separate histones. Protein was
transferred to PVDF membrane (Bio-Rad) and detected following
BL-21 Codon Plus cells (Stratagene) were transformed with
pGEX6p2 or pGEX6p2-RAG2CT and grown according to the manu-
facturer’s instructions. Induction was performed for 1 hr at 37°C
with 1 mM IPTG (Sigma). Purification was as described above but
with Glutathione Sepharose 4B beads (Amersham).
In Vitro Binding Assays
Core histones were purified from HeLa cell nuclei (Cote et al.,
1995). Calf thymus histones H2A, H2B, H3, and H4 were from
Roche. Bacterially produced recombinant human histones H2A,
H2B, H3, and H4 were from Upstate Biotechnologies. Equal
amounts of GST and GST-R2CT were incubated overnight with 7
?g of HeLa histones or 1 ?g of each individual histone in a 2 ml
reaction containing either 100 mM Buffer A (the same as Buffer
A-500 described above except with 100 mM KCl, used for calf
thymus histones H2A and H2B) or 500 mM Buffer A (calf thy-
mus histones H3 and H4). Beads were washed 5 × in Buffer A-250
or A-100, 3 × in Buffer A-100 and resuspended in 1× SDS loading
buffer for sodium dodecyl-sulfate-polyacrylamide gel electrophore-
sis (SDS-PAGE) on an 18% gel followed by Coomassie blue stain.
FNT-RAG2/Strep-RAG1CR complexes were purified as described
above, and protein levels were standardized. Reactions were as
described (Hiom and Gellert, 1998) with slight modifications (Bes-
mer et al., 1998).
Cells were maintained at 37°C and 5% CO2. 293T cells were grown
in Dulbecco’s modified Eagle’s medium (Cellgrow) with 10% bovine
calf serum (Hyclone), 2 mM glutamate, 2 mM pyruvate, and 1%
antibiotic-antimycotic (Invitrogen). RAG2−/−murine B cells were
cultured in RPMI 1640, 10% fetal bovine serum (Sigma), 2 mM glu-
tamate, 2 mM pyruvate, 1% antibiotic-antimycotic (Invitrogen), and
50 ?M β-mercaptoethanol (Fisher).
Retroviral Transduction and Preparation of Stable Cell Lines
Published protocols (He and Ting, 2002) were modified: 293T cells
were cotransfected with 10 ?g of retroviral vector carrying the gene
of interest, 5 ?g pMD.G, which encodes vesicular stomatitis G pro-
tein, and 7.5 ?g pMD.OGP, which encodes gag-pol. Supernatant
was collected 48 hr posttransfection, polybrene added to 5 ?g/ml,
and R2−/−B cells were resuspended at 0.6 million cells/ml. Trans-
duction was for 12 hr, and selection with 1.5 ?g/ml puromycin com-
menced 48 hr later.
PCR and Southern Blotting Analysis
Genomic DNA was isolated with Qiagen DNeasy Tissue Kit. PCRs
were as described (Schlissel et al., 1991). 50 ?l PCRs contained
400, 200, or 25 ng of template DNA. Primers have been described
(Dudley et al., 2003; Schlissel et al., 1991). PCR products were sep-
arated on a 1.2% agarose gel in 0.5× Tris-borate buffer and trans-
ferred for 12 hr to GeneScreen Plus nylon membrane (Pierce) in
0.4 N NaOH and 1M NaCl. To detect Vκ-Jκ rearrangement, probe
was prepared by PCR with unrearranged genomic DNA template,
primers Jκ5.3 and Ko, and α-32P dCTP. For all others, probe was
generated by using Stratagene’s Prime-It II Random Primer Label-
ing Kit. DNA template used corresponded to the fragment Muo
through JH4. For loading control, we PCR amplified RAG1 amino
We thank members of the Cortes lab for helpful discussion and
reagents; Juan Carcamo for critically reading the manuscript; An-
drea Martin, Claudia Canasta, Glaucia Furtado, and members of Dr.
Sergio Lira’s laboratory for help with mouse work; Adrian Ting for
retroviral vectors and advice; Eva Besmer and Ziva Misulovin (The
Rockefeller University, New York) for cloning vectors; and Gary A.
Rathbun and Frederick W. Alt for RAG2−/−pro-B cell lines. K.W.
was supported by the National Institutes of Health (NIH) training
grant AI 07605. P.T. was supported by the NCI Cancer Center Sup-
port grant P30 CA08748. Work in the P.C. laboratory is in part sup-
ported by grants AI45996 from NIH, RSG-04-191-01-LIB from the
American Cancer Society, Program Project Grant PO1 AI61093, a
Cancer Research Institute Investigator award, and a Leukemia and
Lymphoma Society Scholar award.
RAG2 Interacts with Histones
Received: January 24, 2005
Revised: June 1, 2005
Accepted: July 13, 2005
Published: August 23, 2005
Akamatsu, Y., Monroe, R., Dudley, D.D., Elkin, S.K., Gartner, F., Ta-
lukder, S.R., Takahama, Y., Alt, F.W., Bassing, C.H., and Oettinger,
M.A. (2003). Deletion of the RAG2 C terminus leads to impaired
lymphoid development in mice. Proc. Natl. Acad. Sci. USA 100,
Akey, C.W., and Luger, K. (2003). Histone chaperones and nucleo-
some assembly. Curr. Opin. Struct. Biol. 13, 6–14.
Baumann, M., Mamais, A., McBlane, F., Xiao, H., and Boyes, J.
(2003). Regulation of V(D)J recombination by nucleosome position-
ing at recombination signal sequences. EMBO J. 22, 5197–5207.
Besmer, E., Mansilla-Soto, J., Cassard, S., Sawchuk, D.J., Brown,
G., Sadofsky, M., Lewis, S.M., Nussenzweig, M.C., and Cortes, P.
(1998). Hairpin coding end opening is mediated by RAG1 and RAG2
proteins. Mol. Cell 2, 817–828.
Callebaut, I., and Mornon, J.P. (1998). The V(D)J recombination acti-
vating protein RAG2 consists of a six-bladed propeller and a PHD
fingerlike domain, as revealed by sequence analysis. Cell. Mol. Life
Sci. 54, 880–891.
Choi, J.K., Shen, C.P., Radomska, H.S., Eckhardt, L.A., and
Kadesch, T. (1996). E47 activates the Ig-heavy chain and TdT loci
in non-B cells. EMBO J. 15, 5014–5021.
Cote, J., Utley, R.T., and Workman, J.L. (1995). Basic analysis of
transcription factor binding to nucleosomes. Methods Molec.
Genet. 6, 108–129.
Cuomo, C.A., and Oettinger, M.A. (1994). Analysis of regions of
RAG-2 important for V(D)J recombination. Nucleic Acids Res. 22,
de Villartay, J.P. (2002). V(D)J recombination and DNA repair: les-
sons from human immune deficiencies and other animal models.
Curr. Opin. Allergy Clin. Immunol. 2, 473–479.
Dudley, D.D., Sekiguchi, J., Zhu, C., Sadofsky, M.J., Whitlow, S.,
DeVido, J., Monroe, R.J., Bassing, C.H., and Alt, F.W. (2003). Im-
paired V(D)J recombination and lymphocyte development in core
RAG1-expressing mice. J. Exp. Med. 198, 1439–1450.
Eickbush, T.H., and Moudrianakis, E.N. (1978). The histone core
complex: an octamer assembled by two sets of protein-protein in-
teractions. Biochemistry 17, 4955–4964.
Elkin, S.K., Matthews, A.G., and Oettinger, M.A. (2003). The C-ter-
minal portion of RAG2 protects against transposition in vitro.
EMBO J. 22, 1931–1938.
Elkin, S.K., Ivanov, D., Ewalt, M., Ferguson, C.G., Hyberts, S.G., Sun,
Z.Y., Prestwich, G.D., Yuan, J., Wagner, G., Oettinger, M.A., and Go-
zani, O.P. (2005). A PHD finger motif in the C terminus of RAG2
modulates recombination activity. J. Biol. Chem. 280, 28701–28710.
Erdjument-Bromage, H., Lui, M., Lacomis, L., Grewal, A., Annan,
R.S., McNulty, D.E., Carr, S.A., and Tempst, P. (1998). Examination
of micro-tip reversed-phase liquid chromatographic extraction of
peptide pools for mass spectrometric analysis. J. Chromatogr. A.
Ghosh, J.K., Romanow, W.J., and Murre, C. (2001). Induction of a
diverse T cell receptor gamma/delta repertoire by the helix-loop-
helix proteins E2A and HEB in nonlymphoid cells. J. Exp. Med. 193,
Goebel, P., Janney, N., Valenzuela, J.R., Romanow, W.J., Murre, C.,
and Feeney, A.J. (2001). Localized gene-specific induction of ac-
cessibility to V(D)J recombination induced by E2A and early B cell
factor in nonlymphoid cells. J. Exp. Med. 194, 645–656.
Golding, A., Chandler, S., Ballestar, E., Wolffe, A.P., and Schlissel,
M.S. (1999). Nucleosome structure completely inhibits in vitro
cleavage by the V(D)J recombinase. EMBO J. 18, 3712–3723.
He, K.L., and Ting, A.T. (2002). A20 inhibits tumor necrosis factor
(TNF) alpha-induced apoptosis by disrupting recruitment of TRADD
and RIP to the TNF receptor 1 complex in Jurkat T cells. Mol. Cell.
Biol. 22, 6034–6045.
Hiom, K., and Gellert, M. (1998). Assembly of a 12/23 paired signal
complex: a critical control point in V(D)J recombination. Mol. Cell
Jiang, H., Ross, A.E., and Desiderio, S. (2004). Cell cycle-depen-
dent accumulation in vivo of transposition-competent complexes
between recombination signal ends and full-length RAG proteins.
J. Biol. Chem. 279, 8478–8486.
Johnson, K., Angelin-Duclos, C., Park, S., and Calame, K.L. (2003).
Changes in histone acetylation are associated with differences in
accessibility of V(H) gene segments to V-DJ recombination during
B-cell ontogeny and development. Mol. Cell. Biol. 23, 2438–2450.
Johnson, K., Pflugh, D.L., Yu, D., Hesslein, D.G., Lin, K.I., Bothwell,
A.L., Thomas-Tikhonenko, A., Schatz, D.G., and Calame, K. (2004).
B cell-specific loss of histone 3 lysine 9 methylation in the V(H)
locus depends on Pax5. Nat. Immunol. 5, 853–861.
Jones, J.M., and Gellert, M. (2003). Autoubiquitylation of the V(D)J
recombinase protein RAG1. Proc. Natl. Acad. Sci. USA 100,
Kirch, S.A., Rathbun, G.A., and Oettinger, M.A. (1998). Dual role of
RAG2 in V(D)J recombination: catalysis and regulation of ordered
Ig gene assembly. EMBO J. 17, 4881–4886.
Kwon, J., Imbalzano, A.N., Matthews, A., and Oettinger, M.A.
(1998). Accessibility of nucleosomal DNA to V(D)J cleavage is mod-
ulated by RSS positioning and HMG1. Mol. Cell 2, 829–839.
Kwon, J., Morshead, K.B., Guyon, J.R., Kingston, R.E., and Oet-
tinger, M.A. (2000). Histone acetylation and hSWI/SNF remodeling
act in concert to stimulate V(D)J cleavage of nucleosomal DNA.
Mol. Cell 6, 1037–1048.
Langerak, A.W., Wolvers-Tettero, I.L., van Gastel-Mol, E.J., Oud,
M.E., and van Dongen, J.J. (2001). Basic helix-loop-helix proteins
E2A and HEB induce immature T-cell receptor rearrangements in
nonlymphoid cells. Blood 98, 2456–2465.
Lee, A.I., Fugmann, S.D., Cowell, L.G., Ptaszek, L.M., Kelsoe, G.,
and Schatz, D.G. (2003). A functional analysis of the spacer of V(D)J
recombination signal sequences. PLoS Biol. 1(1), E1 DOI: 10.1371/
Lee, J., and Desiderio, S. (1999). Cyclin A/CDK2 regulates V(D)J
recombination by coordinating RAG-2 accumulation and DNA re-
pair. Immunity 11, 771–781.
Liang, H.E., Hsu, L.Y., Cado, D., Cowell, L.G., Kelsoe, G., and
Schlissel, M.S. (2002). The “dispensable” portion of RAG2 is neces-
sary for efficient V-to-DJ rearrangement during B and T cell devel-
opment. Immunity 17, 639–651.
Luger, K., Rechsteiner, T.J., Flaus, A.J., Waye, M.M., and Richmond,
T.J. (1997). Characterization of nucleosome core particles contain-
ing histone proteins made in bacteria. J. Mol. Biol. 272, 301–311.
McMurry, M.T., and Krangel, M.S. (2000). A role for histone acetyla-
tion in the developmental regulation of VDJ recombination. Science
McMurry, M.T., Hernandez-Munain, C., Lauzurica, P., and Krangel,
M.S. (1997). Enhancer control of local accessibility to V(D)J recom-
binase. Mol. Cell. Biol. 17, 4553–4561.
Montalbano, A., Ogwaro, K.M., Tang, A., Matthews, A.G., Larijani,
M., Oettinger, M.A., and Feeney, A.J. (2003). V(D)J recombination
frequencies can be profoundly affected by changes in the spacer
sequence. J. Immunol. 171, 5296–5304.
Morshead, K.B., Ciccone, D.N., Taverna, S.D., Allis, C.D., and Oet-
tinger, M.A. (2003). Antigen receptor loci poised for V(D)J re-
arrangement are broadly associated with BRG1 and flanked by
peaks of histone H3 dimethylated at lysine 4. Proc. Natl. Acad. Sci.
USA 100, 11577–11582.
Muegge, K. (2003). Modifications of histone cores and tails in V(D)J
recombination. Genome Biol. 4, 211.
Nadel, B., Tang, A., Escuro, G., Lugo, G., and Feeney, A.J. (1998).
Sequence of the spacer in the recombination signal sequence af-
fects V(D)J rearrangement frequency and correlates with nonran-
dom Vkappa usage in vivo. J. Exp. Med. 187, 1495–1503.
Immunity Download full-text
Oettinger, M.A. (2004). How to keep V(D)J recombination under
control. Immunol. Rev. 200, 165–181.
Oettinger, M.A., Schatz, D.G., Gorka, C., and Baltimore, D. (1990).
RAG-1 and RAG-2, adjacent genes that synergistically activate
V(D)J recombination. Science 248, 1517–1523.
Patenge, N., Elkin, S.K., and Oettinger, M.A. (2004). ATP-dependent
remodeling by SWI/SNF and ISWI proteins stimulates V(D)J cleav-
age of 5 S arrays. J. Biol. Chem. 279, 35360–35367.
Peterson, C.L., and Cote, J. (2004). Cellular machineries for chro-
mosomal DNA repair. Genes Dev. 18, 602–616.
Romanow, W.J., Langerak, A.W., Goebel, P., Wolvers-Tettero, I.L.,
van Dongen, J.J., Feeney, A.J., and Murre, C. (2000). E2A and EBF
act in synergy with the V(D)J recombinase to generate a diverse
immunoglobulin repertoire in nonlymphoid cells. Mol. Cell 5, 343–
Ross, A.E., Vuica, M., and Desiderio, S. (2003). Overlapping signals
for protein degradation and nuclear localization define a role for
intrinsic RAG-2 nuclear uptake in dividing cells. Mol. Cell. Biol. 23,
Roth, D.B., and Roth, S.Y. (2000). Unequal access: regulating V(D)J
recombination through chromatin remodeling. Cell 103, 699–702.
Sadofsky, M.J., Hesse, J.E., McBlane, J.F., and Gellert, M. (1993).
Expression and V(D)J recombination activity of mutated RAG-1
proteins. Nucleic Acids Res. 21, 5644–5650.
Sadofsky, M.J., Hesse, J.E., and Gellert, M. (1994). Definition of a
core region of RAG-2 that is functional in V(D)J recombination. Nu-
cleic Acids Res. 22, 1805–1809.
Schlissel, M.S., Corcoran, L.M., and Baltimore, D. (1991). Virus-
transformed pre-B cells show ordered activation but not inactiva-
tion of immunoglobulin gene rearrangement and transcription. J.
Exp. Med. 173, 711–720.
Schwarz, K., Ma, Y., Pannicke, U., and Lieber, M.R. (2003). Human
severe combined immune deficiency and DNA repair. Bioessays 25,
Shinkai, Y., Rathbun, G., Lam, K.P., Oltz, E.M., Stewart, V., Mendel-
sohn, M., Charron, J., Datta, M., Young, F., Stall, A.M., et al. (1992).
RAG-2-deficient mice lack mature lymphocytes owing to inability
to initiate V(D)J rearrangement. Cell 68, 855–867.
Shockett, P.E., Zhou, S., Hong, X., and Schatz, D.G. (2004). Partial
reconstitution of V(D)J rearrangement and lymphocyte develop-
ment in RAG-deficient mice expressing inducible, tetracycline-reg-
ulated RAG transgenes. Mol. Immunol. 40, 813–829.
Sikes, M.L., Suarez, C.C., and Oltz, E.M. (1999). Regulation of V(D)J
recombination by transcriptional promoters. Mol. Cell. Biol. 19,
Sikes, M.L., Meade, A., Tripathi, R., Krangel, M.S., and Oltz, E.M.
(2002). Regulation of V(D)J recombination: a dominant role for pro-
moter positioning in gene segment accessibility. Proc. Natl. Acad.
Sci. USA 99, 12309–12314.
Silver, D.P., Spanopoulou, E., Mulligan, R.C., and Baltimore, D.
(1993). Dispensable sequence motifs in the RAG-1 and RAG-2
genes for plasmid V(D)J recombination. Proc. Natl. Acad. Sci. USA
Simpson, R.T. (1991). Nucleosome positioning: occurrence, mecha-
nisms, and functional consequences. Prog. Nucleic Acid Res. Mol.
Biol. 40, 143–184.
Spanopoulou, E., Cortes, P., Shih, C., Huang, C.M., Silver, D.P.,
Svec, P., and Baltimore, D. (1995). Localization, interaction, and
RNA binding properties of the V(D)J recombination-activating pro-
teins RAG1 and RAG2. Immunity 3, 715–726.
Swanson, P.C., Volkmer, D., and Wang, L. (2004). Full-length RAG-2,
and not full-length RAG-1, specifically suppresses RAG-mediated
transposition but not hybrid joint formation or disintegration. J.
Biol. Chem. 279, 4034–4044.
Talukder, S.R., Dudley, D.D., Alt, F.W., Takahama, Y., and Akamatsu,
Y. (2004). Increased frequency of aberrant V(D)J recombination
products in core RAG-expressing mice. Nucleic Acids Res. 32,
Tsai, C.L., and Schatz, D.G. (2003). Regulation of RAG1/RAG2-
mediated transposition by GTP and the C-terminal region of RAG2.
EMBO J. 22, 1922–1930.
Whitehurst, C.E., Schlissel, M.S., and Chen, J. (2000). Deletion of
germline promoter PD beta 1 from the TCR beta locus causes hy-
permethylation that impairs D beta 1 recombination by multiple
mechanisms. Immunity 13, 703–714.
Winkler, G.S., Lacomis, L., Philip, J., Erdjument-Bromage, H.,
Svejstrup, J.Q., and Tempst, P. (2002). Isolation and mass spec-
trometry of transcription factor complexes. Methods 26, 260–269.
Yancopoulos, G.D., and Alt, F.W. (1985). Developmentally controlled
and tissue-specific expression of unrearranged VH gene segments.
Cell 40, 271–281.
Yurchenko, V., Xue, Z., and Sadofsky, M. (2003). The RAG1 N-ter-
minal domain is an E3 ubiquitin ligase. Genes Dev. 17, 581–585.