Corrections

APPLIED BIOLOGICAL SCIENCES

Correction for An algorithm-based topographical biomaterials

library to instruct cell fate, by Hemant V. Unadkat, Marc
Hulsman, Kamiel Cornelissen, Bernke J. Papenburg, Roman K.
Truckenmller, Anne E. Carpenter, Matthias Wessling, Gerhard
F. Post, Marc Uetz, Marcel J. T. Reinders, Dimitrios Stamatialis,
Clemens A. van Blitterswijk, and Jan de Boer, which appeared
in issue 40, October 4, 2011, of Proc Natl Acad Sci USA (108:
1656516570; rst published September 26, 2011; 10.1073/pnas.
1109861108).
The authors note that the following statement should be added
to the Acknowledgments: This work was supported in part by
NIH Grant R01 GM089652 (A.E.C.).

IMMUNOLOGY

Correction for Alternative end-joining catalyzes robust IgH

locus deletions and translocations in the combined absence of
ligase 4 and Ku70, by Cristian Boboila, Mila Jankovic, Catherine
T. Yan, Jing H. Wang, Duane R. Wesemann, Tingting Zhang, Alex
Fazeli, Lauren Feldman, Andre Nussenzweig, Michel Nussenzweig,
and Frederick W. Alt, which appeared in issue 7, February 16, 2010,
of Proc Natl Acad Sci USA (107:30343039; rst published January
25, 2010; 10.1073/pnas.0915067107).
The authors note that the National Institutes of Health Grant
AI031541 should instead appear as AI077595.
www.pnas.org/cgi/doi/10.1073/pnas.1303073110

www.pnas.org/cgi/doi/10.1073/pnas.1302919110

IMMUNOLOGY

EVOLUTION

Correction for Diversication of rhacophorid frogs provides

evidence for accelerated faunal exchange between India
and Eurasia during the Oligocene, by Jia-Tang Li, Yang Li,
Sebastian Klaus, Ding-Qi Rao, David M. Hillis, and Ya-Ping
Zhang, which appeared in issue 9, February 26, 2013, of Proc
Natl Acad Sci USA (110:34413446; rst published February
11, 2013; 10.1073/pnas.1300881110).
The authors note that, within the author line, Yang Lia,c
should instead appear as Yang Lib,c. The corrected author line
appears below. The online version has been corrected.
Jia-Tang Lia,b, Yang Lib,c, Sebastian Klausd, Ding-Qi Raoa,
David M. Hillise, and Ya-Ping Zhanga,f
www.pnas.org/cgi/doi/10.1073/pnas.1304031110

IMMUNOLOGY

Correction for Integrity of the AID serine-38 phosphorylation

site is critical for class switch recombination and somatic hypermutation in mice, by Hwei-Ling Cheng, Bao Q. Vuong,
Uttiya Basu, Andrew Franklin, Bjoern Schwer, Jillian Astarita,
Ryan T. Phan, Abhishek Datta, John Manis, Frederick W. Alt,
and Jayanta Chaudhuri, which appeared in issue 8, February 24,
2009, of Proc Natl Acad Sci USA (106:27172722; rst published
February 5, 2009; 10.1073/pnas.0812304106).
The authors note that the National Institutes of Health Grant
AI31541 should instead appear as AI077595.

Correction for Downstream class switching leads to IgE antibody production by B lymphocytes lacking IgM switch regions,
by Tingting Zhang, Andrew Franklin, Cristian Boboila, Amy
McQuay, Michael P. Gallagher, John P. Manis, Ahmed Amine
Khamlichi, and Frederick W. Alt, which appeared in issue 7,
February 16, 2010, of Proc Natl Acad Sci USA (107:30403045;
rst published February 1, 2010; 10.1073/pnas.0915072107).
The authors note that the National Institutes of Health Grant
AI031541 should instead appear as AI077595.
www.pnas.org/cgi/doi/10.1073/pnas.1303075110

IMMUNOLOGY

Correction for Robust chromosomal DNA repair via alternative

end-joining in the absence of X-ray repair cross-complementing
protein 1 (XRCC1), by Cristian Boboila, Valentyn Oksenych,
Monica Gostissa, Jing H. Wang, Shan Zha, Yu Zhang, Hua Chai,
Cheng-Sheng Lee, Mila Jankovic, Liz-Marie Albertorio Saez,
Michel C. Nussenzweig, Peter J. McKinnon, Frederick W. Alt, and
Bjoern Schwer, which appeared in issue 7, February 14, 2012, of
Proc Natl Acad Sci USA (109:24732478; rst published January
30, 2012; 10.1073/pnas.1121470109).
The authors note that the National Institutes of Health Grant
AI031541 should instead appear as AI077595.
www.pnas.org/cgi/doi/10.1073/pnas.1303078110

CORRECTIONS

www.pnas.org/cgi/doi/10.1073/pnas.1303069110

www.pnas.org

PNAS | April 2, 2013 | vol. 110 | no. 14 | 5731

Downstream class switching leads to IgE antibody

production by B lymphocytes lacking IgM
switch regions
Tingting Zhanga,b,c,d, Andrew Franklina,b,c,d, Cristian Boboilaa,b,c,d, Amy McQuaya,b,c,d, Michael P. Gallaghera,b,c,d,
John P. Manisb,e, Ahmed Amine Khamlichif, and Frederick W. Alta,b,c,d,1
a

Department of Genetics, Harvard Medical School, Boston, MA 02115; bthe Childrens Hospital; cImmune Disease Institute; dHoward Hughes Medical Institute;
Department of Pathology, Harvard Medical School, Boston, MA 02115; and fCentre National de la Recherche Scientique UMR 5089-IPBS, Universit de
Toulouse, F-31077 Toulouse, France

Contributed by Frederick W. Alt, December 30, 2009 (sent for review December 22, 2009)

Ig heavy chain (IgH) class-switch recombination (CSR) replaces the IgH

C constant region exons with one of several sets of downstream IgH
constant region exons (e.g., C, C, or C), which affects switching
from IgM to another IgH class (e.g., IgG, IgE, or IgA). Activationinduced cytidine deaminase (AID) initiates CSR by promoting DNA
double-strand breaks (DSBs) within switch (S) regions anking the
donor C (S) and a downstream acceptor CH (e.g., S, S, S) that
are then joined to complete CSR. DSBs generated in S frequently are
joined within S to form internal switch region deletions (ISD). AIDinduced ISD and mutations have been considered rare in downstream
S regions, suggesting that AID targeting to these S regions requires its
prior recruitment to S. We have now assayed for CSR and ISD in B
cells lacking S (S/ B cells). In S/ B cells activated for CSR to IgG1
and IgE, CSR to IgG1 was greatly reduced; but, surprisingly, CSR to IgE
occurred at nearly normal levels. Moreover, normal B cells had substantial S1 ISD and increased mutations in and near S1, and levels of
both were greatly increased in S/ B cells. Finally, S/ B cells
underwent downstream CSR between S1 and S on alleles that
lacked S CSR to these sequences. Our ndings show that AID targets
downstream S regions independently of CSR with S and implicate an
alternative pathway for IgE class switching that involves generation
and joining of DSBs within two different downstream S regions
before S joining.

ntibodies are comprised of Ig heavy (IgH) and light (IgL) chains.

The N-terminal portion of IgH and IgL chains, termed the variable region, binds antigens, whereas the C-terminal portion of the
IgH chain, termed the constant region, determines antibody class and
effector functions. The IgH variable region is encoded in a distinct
exon from those that encode the constant region. The variable region
exon is assembled early in B-cell development from VH, D, and JH
segments through V(D)J recombination (1). The IgH V(D)J exon is
rst expressed with proximal downstream exons that encode the C
constant region, leading to expression of IgH chains and IgM
antibody. Newly generated B cells express surface IgM and migrate to
the periphery where they can be induced to express a different IgH
class (e.g., IgG, IgE, or IgA). The mouse IgH locus contains eight sets
of CH exons, referred to as CH genes, arranged as 5-V(D)J-C-CC3-C1-C2b-C2a-C-C-3 over a 200-kb region (2). IgH class
switching involves a recombination/deletion event, termed classswitch recombination (CSR), in which the C gene is replaced with a
downstream CH gene (2, 3). Both CSR and the related somatic
hypermutation (SHM) process, which introduces mutations into
variable region exons to allow production of higher afnity antibodies,
are initiated by activation-induced cytidine deaminase (AID) (4, 5).
Each CH gene that undergoes CSR is organized into a unit from
5 to 3 that includes a transcriptional promoter followed by a
noncoding exon (termed I exon), a switch (S) region, and a set of
CH exons (6). CSR involves the introduction of a DNA doublestrand break (DSB) in the donor S region anking the C gene (S)
and into an acceptor S region anking a downstream CH gene; this
is followed by the joining of the breaks by general DSB end-joining
30403045 | PNAS | February 16, 2010 | vol. 107 | no. 7

pathways (7). S regions are long (110 kb) intronic sequences

containing characteristic repeated motifs that include AID target
motifs, and the S is the most repetitive and harbors the highest
numbers of AID targets (6, 8). Although there is some homology
between S, S, and S, there is little or no homology between S
and S regions (8). Transcription through S regions targets AID
activity, which generates primary lesions that are processed into
DSBs in S and the downstream acceptor S region required to
initiate CSR (7, 9). Thereby, specic induction of transcription
from the I-region promoter anking a particular acceptor S region
targets that region for CSR. AID also introduces lesions into IgH
and IgL variable region exons through a transcription dependent
process, and they are converted into SHMs (9, 10). During CSR,
AID also generates SHMs in S regions and immediate anking
sequences (11).
S regions serve primarily as specialized DNA structures that target
the AID DSB-inducing activity. Thus, most CSR junctions occur
within or occasionally just outside of S regions (12). In addition, B
cells that lack the donor S are greatly impaired for CSR to all tested
CH genes (13, 14). Correspondingly, deletion of S1 abrogates CSR
only to C1 (15). Finally, recombinational IgH CSR from IgM to
IgG1 can be achieved in cells lacking those S regions when DSBs are
introduced into their former sites by a yeast endonuclease (16). There
are several mechanisms by which transcribed S regions may become
AID targets. First, mammalian S regions form R loops that provide
substrates for the single-stranded DNA-specic activity of AID (17
19). Second, AID seems capable of gaining access to sequences, such
as S regions rich in AID target motifs, in a phosphorylation and
Replication Protein A (RPA)-dependent fashion (20).
AID activity can generate multiple DSBs within S (21, 22). Such
DSBs within a given S region can be religated, joined to another
DSB in the same S region, or joined to a DSB in a downstream
S region to affect CSR. Religation of a resected DSB within a given
S region or ligation of two intra-S region DSBs will generate
internal S-region deletions (ISD). ISD are usually assayed by
activating B cells for CSR and then, generating IgM-producing
hybridomas from them; theses are assayed for ISD large enough to
be viewed by Southern blotting. These large ISD occur frequently
within the S region in B cells or B-cell lines activated for CSR (23
26). They are AID-initiated (24) and mostly seem to occur through
end joining (27). However, prior studies have found ISD in
downstream S regions to be quite rare (2426, 28), leading to
suggestions that S drives CSR by providing excess DSBs to ensure

Author contributions: T.Z., J.M., and F.W.A. designed research; T.Z., A.F., C.B., A.M., and
M.G. performed research; A.A.K. contributed new reagents/analytic tools; T.Z., A.F., C.B.,
A.M., J.M., and F.W.A. analyzed data; and T.Z. and F.W.A. wrote the paper.
The authors declare no conict of interest.
1

To whom correspondence should be addressed. E-mail: alt@enders.tch.harvard.edu.

This article contains supporting information online at www.pnas.org/cgi/content/full/

0915072107/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.0915072107

Results
IgH CSR to IgE Occurs at Nearly Normal Levels in B Cells Lacking S.

To further elucidate mechanisms that target AID during IgH

CSR, we have studied mice generated previously by gene-targeted
mutation that harbor an essentially complete deletion of the
entire 4.6-kb S region in their germline (13). Mice homozygous
for this S deletion mutation (S/ mice) show a great reduction
in IgH CSR to all tested antibody classes, including IgG3, IgG2b,
IgG1, IgG2a, and IgA (13). However, CSR to IgE was not previously studied in these mice. To further characterize potential
IgH CSR defects in S/ mice, we puried splenic B cells from
WT and S/ spleens and stimulated them in vitro with antiCD40 and IL-4 to stimulate CSR from IgM to IgG1 and IgE. We
assayed for CSR to IgG1 by surface staining and ow cytometry at
days 2, 3, and 4 of stimulation. Consistent with previous studies
(13), we observed greatly decreased (between 5- and 10-fold) IgH
class switching to IgG1 at all analyzed time points (Fig. 1 A and B).
We conrmed this nding by hybridoma analyses that showed
increased levels of IgM-expressing hybridomas and greatly
decreased levels of IgG1-expressing hybridomas from S/ versus WT B cells (Fig. 1C and Table S1).
IgE class switching cannot be readily analyzed by ow cytometry, because anti-CD40 treatment induces up-regulation of
CD23 (FcRII), the Fc receptor for IgE, on activated B cells (36),
which leads to binding of soluble IgE antibodies to nonIgEproducing B cells and causes false-positive staining. Therefore, we
analyzed IgE switching by assaying the frequency of IgE-producing hybridomas generated from day 4 anti-CD40 plus IL-4
stimulated B cells. Under our current stimulation conditions,
activated WT B cells switch to IgE at quite substantial levels;
Zhang et al.

WT

10 4

10

10

10

10

-/-

10 4

2.34

AID-/-

10 4

0.37

0.41

10 3

10 3

10

9.2e-3

10

Day 2
10 1

10 1

0.41
10 0

10 4

IgG1

97.2
10 1

10 2

10 3

0.8

10 0

10 4

10

98.8

10

10

10

10

10

10 3

10

10 2

10 2

10 2

10

10 1

10 1

10

26.9

1.49
10

10 4

71.5
10

10

10

10

10 0
4

1.46

0.29
10 0

98.2
10 1

10 2

10 3

10

10

10 0

10 4

10 3

10

10 2

10 2

10 2

10

10 1

10 1

10

2.36
10

49.9
10

10

10

10

10 0
4

3.59

0.96
10 0

95.5
10 1

10 2

10 3

10 4

98.9
10

10

10

10

10

0.64

99.2
10 1

10 2

10 3

10 0

10

0.16

Day 3
10 0

46.8

10

0.86

0.72

10 0
4

10

0.15

10 4

0.2

Day 4
0.4
10 0

99.4
10 1

10 2

10 3

10 4

B220

70

Wildtype

60

S -/AID-/-

50
40
30

IMMUNOLOGY

joining to less frequently occurring DSBs in downstream S regions

(6, 16, 29). AID activity during CSR also generates SHMs in S and
adjacent sequences, including the upstream I region (29, 30).
However, such mutations again were rarely found in downstream
CH genes in wild-type (WT) B cells (29). The lack of SHM or ISD in
downstream S regions also led to suggestions that AID activity on
these sequences may only occur after AID has acted on S, perhaps
because of induction of factors or formation of complexes required
for AID access to downstream S regions (25, 29). However, AIDinitiated mutations in downstream S regions were observed in the
context of a DNA-repair-decient background (30).
IgH class switching to certain downstream CH genes (e.g., C)
often occurs in humans and mice through sequential CSR
involving two downstream acceptor S regions (3134). Among
the lines of evidence for sequential CSR are ndings of remnants
of an additional intervening S region in CSR junctions between
S and a downstream S region. In particular, such ndings suggest that most switching to IgE may involve a pathway in which
S rst joins to S1 to form a S/S1 fusion S region that subsequently switches to S (3134). Although it is assumed that
such junctions occurred by the pathway outlined above, it was
not ruled out that some might occur through a pathway in which
two downstream S regions are joined before joining with S.
However, such a pathway would require simultaneous DSBs in
two downstream S regions, which might be considered unlikely
given the apparently low frequency of DSBs in downstream S
regions found in prior studies.
Deletion of sequences upstream of S, including the IgH JH
segments and the intronic enhancer element, inactivated CSR
and led to increased S1 ISD in appropriately activated B cells
(35). Although there may be several explanations for this nding,
we considered the possibility that inactivation of S might lead to
increased AID targeting of downstream S regions. To test this
possibility and also better dene potential CSR pathways, we
assayed for IgH class switching, ISD and SHM of downstream S
regions, and downstream CSR between the S1 and S in B cells
stimulated for CSR to IgG1 and IgE.

20
10
0
1.5

2.0

2.5

3.0

3.5

4.0

4.5

Day

80

WT
S -/-

70
60
50
40
30
20
10
0
IgM

IgG1

IgE

Antibody class

Fig. 1. IgG1 and IgE CSR in S/ B cells. (A) FACS analysis on days 2, 3, and 4 of
anti-CD40/IL-4stimulated splenic B cells from WT, S/, and AID/ mice by
surface staining with antiIgG1-FITC and anti-B220-PE-Cy5 antibodies. The percentage of IgG1+B220+ cells represents CSR level to IgG1 at each time point. (B)
Statistical analysis of three independent FACS analysis experiments for IgG1
switching of anti-CD40/IL-4stimulated B cells from WT, S/, and AID/ mouse
spleens at day 2, 3, and 4. (C) Hybridoma analysis by supernatant ELISA on clones
generated from day 4 anti-CD40/IL-4stimulated B-cell fusion with NS-1 fusion
partner cell line. IgM, IgG1, and IgE single-positive clones were counted and
analyzed for percentage in total single positive clones for each IgH isotype in
three independent fusions on WT and S/ B cells. SDs calculated from the three
experiments are shown. Detailed numbers are listed in Table S1.

indeed, 40% of the recovered hybridomas were IgE producers

(Fig. 1C and Table S1) (27). In this regard, our stimulation conditions also seem to give very high levels of IgG1 class switching
(Fig. 1C) (27, 37). Unexpectedly, S/ B cells also displayed very
high levels of IgE switching; we recovered IgE-expressing S/
hybrdomas at nearly 70% the frequency of WT IgE-producing
hybridomas. Indeed, the frequency of IgE-expressing S/ B
hybridomas was more than 3-fold greater than that of IgG1expressing S/ hybridomas (Fig. 1C and Table S1).
To further elucidate the nature of the apparently high level of
IgE class switching in S/ B cells and to determine if it truly
represented CSR, we employed a PCR approach, using primers
from I and just downstream of S, to isolate potential CSR
junctions from a series of WT and S/ IgE-expressing hybridPNAS | February 16, 2010 | vol. 107 | no. 7 | 3041

Frequent Rearrangement of S1 in Normal and S-Deleted B Cells in

the Absence of CSR to S. To further analyze factors that may

regulate AID targeting on downstream S regions, we assayed for

ISD in S1 of IgM-secreting hybridomas generated from day 4
anti-CD40 plus IL-4 stimulated WT and S/ B cells. To assay for
potential rearrangements of S1 that did not involve bona de
CSR, we digested genomic hybridoma DNA with EcoRI and then
used Southern blotting to assay for non-germline fragments that
hybridized to an I1 probe. The I1 probe derives from sequences
just upstream of the S1 region and hybridizes to an 17-kb EcoR1
fragment that encompasses the I1 exon and S1 region; therefore,
any rearrangement within this region (except for CSR, which
deletes the I1 exon) will alter the size of the EcoRI fragment that
hybridizes to the I1 probe (Fig. 2A Upper; Fig. S2). In marked
contrast to other studies, we found, in three independent sets of
IgM-secreting WT hybridomas, that nearly 20% of the IgH alleles
had undergone nonCSR-associated S1 rearrangements (Fig. 2 A
and B; Table S2). Thus, our ndings indicate that, under our antiCD40 plus IL-4 activation conditions, AID targets the downstream
S1 region independently of S to S1 CSR on that IgH allele.
Strikingly, we found that nearly 80% of IgM-producing hybridomas isolated from three sets of anti-CD40 plus IL-4stimulated
S/ B cells contained S1 rearrangements on alleles in which S1
had not undergone bona de CSR with S (Fig. 2 A and B; Table
S2). Thus, there is a substantial accumulation of apparent AID
targeting events at S1 in the absence of an upstream donor S.
Frequent Mutation of the I1/S1 Region in the Absence of
Rearrangement to S. To further test whether or not AID can

target downstream S regions independently of initiating a bona

de CSR event, we assayed for SHM of sequences within the
region covering the 3 portion of I1 through the 5 portion of S1
after anti-CD40 plus IL-4 stimulation of WT and S/ B cells for
7 days. This 1.6-kb region, which contains about 500 bp from the
3 end of the I1 exon and then extends 1.1 kb into the 5 portion
of S1, has been assayed similarly for AID-initiated hypermutations by other groups (29, 30). Analyses of mutations within
the I1/S1 region from three sets of activated WT and control
AID/ B cells revealed a signicantly higher mutation level in the
WT B cells (4 104/bp versus 3 105/bp) (Fig. 3A; Table S3).
Moreover, activated S/ B cells had an even higher level of
mutations within the I1/S1 region (1.4 103/bp) (Fig. 3A). In
addition, there was a higher level of mutations within this given
sequence from the S/ samples than in WT samples (Fig. 3B).
Among the scored mutations, which include point mutations,
insertions, and deletions, there seemed to be an increased frequency of deletions in the I1/S1 sequences from the S/
samples (Table S3), potentially reecting increased AID targeting.
3042 | www.pnas.org/cgi/doi/10.1073/pnas.0915072107

I 1 probe
I 1

S 1

RI

C 1

RI

RI
vJ
-1 9/S
NS 1 2

WT IgM hybridomas

* *

-/-

- -

IgM+ hybridomas

** * ** * * * * * * ** * ** * * * ** * * *

Unswitched rearranged
S 1 alleles/Total
unswitched alleles (%)

omas (Fig. S1 Upper). We were able to isolate junctions from 80%

of assayed WT and S/ B-cell hybridomas, conrming that IgE
switching had occurred through a recombination/deletion event.
The recovered WT junctions were similar to those previously
reported (38). Thus, they frequently contained microhomology
(MH) at the junctions and less frequently were direct with no
overlapping nucleotides, or they contained insertions. These WT
S to S CSR junctions, similar to CSR junctions (31, 38), linked
sequences in the more 5 portion of S with sequences in the more
3 portion of S (Fig. S1), perhaps reecting PCR biases of the
assay, AID targeting hotspots, or continued AID activity and ISD
within fused S/S regions. CSR junctions from the S/ B-cell
hybridomas occurred in a similar region of S as those recovered
from WT. As expected (13), the junctions mostly occurred in the
region just upstream of S, but one junction occurred in a plasmid
sequence inserted through gene targeting and the other occurred
in the 5 portion of C (Fig. S1). Based on these ndings, we
conclude that B cells lacking the donor S region utilize sequences
outside of S to undergo substantial levels of class switching to IgE.

GL

12

J
Sv
9/

GL

90
80
70
60
50
40
30
20
10
0

WT

-/-

Fig. 2. S1 rearrangements on unswitched alleles in WT and S/ IgM-producing

hybridomas. (A Top) Map of the C1 gene. The EcoRI sites are indicated (RI), and
the I1 probe is indicated. (Middle and Bottom) Southern blotting analysis of
EcoRI-digested genomic DNA extracted from IgM-producing hybridomas generated from day 4 anti-CD40/IL-4stimulated WT (Middle) and S/ (Bottom) B
cells for hybridization to an I1 probe (18). I1-hybridizing germline bands of
EcoRI-digested genomic DNA from WT (129/SvJ background) mouse kidneys are
indicated with an arrow head and labeled as GL. Lanes with on top show no
bands hybridizing to the I1 probe or a JH4 or C probe (used to show that assayed
hybridomas had rearranged JH alleles and to conrm their genotype as WT or
mutant, respectively) (Fig. S2). These lanes were not counted in the calculations
shown in Fig. S2B, whereas lanes with * on top contain one or two unswitched
alleles that have undergone rearrangements of the I1-hybridizing fragment. (B)
Percentage of total unswitched S1 alleles in WT and S/ IgM-producing
hybridomas that contain rearrangements of the I1-hybridzing fragments (presumed S1 rearrangements). SDs calculated from the three experiments are
shown. Detailed numbers are listed in Table S1.

However, the overall spectrum of point mutations was similar

between WT and S/ sequences, consistent with the same intact
mutation mechanism in both (Fig. S3). These results provide further evidence that AID targets downstream S regions independently of S and moreover, that AID-initiated mutations are greatly
increased in downstream S regions on IgH alleles that lack S.
Frequent Recombination Events Between S1 and S in the Absence
of Recombination with S. To investigate AID targeting efciency

at downstream S regions, we assayed for S rearrangements in the

same sets of IgM-producing hybridomas in which we assayed for
S1 rearrangements. We rst employed Southern blotting to assay
genomic DNA from these hybridomas for rearrangements within
S. For this purpose, we assayed BamHI- or EcoRI-digested DNA
for hybridization to an I probe and separately, for hybridization to
a C probe to detect potential S ISD (Fig. 4A Upper; Fig. S4). We
noted that most of these hybridomas clearly arose from activated B
cells that had not undergone CSR to S1 involving S but that had
undergone some type of S1 rearrangement, because they contained two distinct, rearranged I1-hybridizing alleles (Fig. 2A).
Despite the apparently high level of IgE switching and particularly
high frequency of S1 rearrangements that did not involve CSR
Zhang et al.

I 1/S 1 mutation

I1-Fw
I1

15.0

C1

I1
probe

RI

12.5

RI

RI

RI

I1-Fw

10.0

I1

S
I
probe

C
C
probe

RI

S.1-Rv
S1/S

7.5
RI

5.0
2.5
0.0

WT

WT
6
3 4

-/-

**

* *

C 10 kb

* *

GL/
NS-1

8 kb

6 7 8

6 kb

0
10 kb
I1 8 kb

65

RI

C
probe

4
73

I1
probe

S-/- IgM+ hybridomas

-/-

S1

* *

* *

* *

GL

6 kb

Fig. 3. Mutations accumulate in the I1/S1 region in WT and S/ B cells.

(A) Mutation frequency in the I1/S1 region of WT and S/ B cells. SDs are
calculated from three independent experiments. Detailed numbers are listed
in Table S3. (B) The total number of I1/S1 sequences from WT and S/ B
cells analyzed is indicated in the center of each pie chart. The percentage of
individual sequences containing different numbers of mutations is listed in
different segments of the pie charts and is proportional to the segment sizes
in the pie charts.

WT IgM+ hybridomas

S-/- IgM+ hybridomas (a158)

PCR of S1-S junctions

with S in S/ B cells, we detected only one potential S ISD with

the I probe in 20 S/ IgM-producing hybridoma clones (Fig.
S4). However, when DNA from the same set of S/ IgM-producing hybridomas was assayed by hybridization to a C probe, we
observed a substantial number of rearrangements (Fig. 4B Upper).
Because many of these S/ IgM-producing hybridomas contained two rearranged I1-hybridizing alleles (Fig. 2A), the rearrangements detected with the C probe cannot all be explained by
the recombination events on the S-deleted allele that lead to class
switching (Fig. S1). Additionally, as outlined above, they are not S
ISD. Therefore, we considered the possibility that they represent
downstream CSR events between S1 and S. In support of this
notion, stripping and reprobing these blots with an I1 probe
revealed that a substantial proportion of the S rearrangements
detected with the C probe are exactly the same size as the I1hybridizing rearrangements (Fig. 4 A and B; common bands are
indicated with asterisks).
To directly test for downstream CSR between S1 and S in
activated S/ cells, we assayed for S1S junctions in genomic
DNA from a set of IgM-producing S/-activated B-cell hybridomas through PCR with a forward primer derived from the 3I1
region and reverse primer derived from the region immediately
downstream of S (Fig. 4A Lower). In unrearranged genomic
DNA, the two PCR primers are derived from regions more than 60
kb apart, and therefore, they would only be amplied if they
underwent a rearrangement, such as an S1 to S recombination,
that placed them more proximal to each other. Although we did
not amplify bands by this approach from the WT IgM-secreting
B-cell hybridoma DNA, we amplied bands of distinct sizes from
15 of 42 S/ IgM-secreting B-cell hybridomas (Fig. 4C). DNA of
cloned I1/downtream S PCR fragments from six S/ IgMproducing hybridomas conrmed that each harbored a unique S1
to S junction (Fig. S5). These ndings show that S1 and S can
harbor AID-initiated breaks simultaneously and undergo a form
of downstream CSR in the absence of bona de CSR with S.
Zhang et al.

Fig. 4. S
IgM-producing hybridomas contain S1S junctions on IgH
alleles that have not undergone CSR with S. (A) Diagram of the Southern
blotting and PCR strategies to detect S1S junctions. (A) Upper diagram
shows germline C1 and C genes, which are separated by 60 kb, and Lower
diagram shows putative CSR product between the two genes. RI, EcoRI sites.
The C and I1 Southern blot probes and the PCR primers used in the assay
are also indicated. The PCR primers would only generate a band in the
rearranged allele (Lower) because of the long distance of their separation in
the germline conguration (Upper). (B Upper and Lower) Southern blot with
C and I1 probes, respectively. Rearranged bands that cohybridize with the
C and I1 probe are indicated with an asterisk. Digested NS-1 fusion-partner
genomic DNA hybridizes to the C probe at the same size as germline
genomic DNA, which is shown as GL/NS-1 (NS-1). (C) PCR analyses of genomic
DNA from a set of WT (Left) and S/ IgM+ hybridomas (Right). The samesized bands observed in the S/ lines were observed in repeat analyses. In
analyses of three separate sets of WT hybridomas no PCR bands were
observed in 24 samples analyzed, whereas in analyses of three separate sets
of S/ hybridomas, we observed bands in 15 of 42 samples analyzed, with
10 representative samples illustrated here.

Discussion
Mechanisms that coordinate action of AID on donor and acceptor
S regions are not fully understood. One hypothesis, based on prior
ndings that AID seemed to have little or no activity on downstream
S regions other than in the context of CSR, was that AID might have
to be recruited to S and/or introduce lesions into S to gain
downstream access (21, 25, 26). However, we now show that WT
IgM-producing B cells activated for CSR to IgG1 accumulate ISD
events within S1 and mutations within I1/S1, both hallmarks of
AID activity, on alleles that have not undergone CSR with S. One
difference between our studies and most earlier studies is that our
current anti-CD40 plus IL-4activation conditions allow much
higher levels of IgG1 and IgE switching (e.g., refs. 27 and 37 and this
study). Thus, these potentially increased-activation conditions may
facilitate observation of AID targeting events in downstream S
PNAS | February 16, 2010 | vol. 107 | no. 7 | 3043

IMMUNOLOGY

Mutation frequency
(x10-4bp)

17.5

S.1-Rv

regions. Moreover, we nd that S/ B cells activated for CSR to

IgG1 and IgE accumulate greatly increased levels of S1 ISD,
downstream CSR events involving S1 and S, and mutations in I1/
S1. These ndings clearly show that AID can access downstream S
regions without rst interacting with S. Finally, the downstream
CSR observed between S1 and S, in the absence of CSR with S,
implies that AID activity can lead to simultaneous DSBs in these two
downstream S regions, thereby suggesting an alternative pathway
for sequential CSR (see below). In this regard, whereas the
dependence of I-region promoter transcription and CSR to most
downstream S regions on the 3 IgH regulatory region may preclude
activating more than one of these promoters at a time (39, 40), the
I1 promoter functions relatively independently from known IgH 3
regulatory region elements and is activated by the same pathways as
the I promoter, which could allow simultaneous activation and
facilitate downstream S1 to S CSR.
We now show that downstream S regions can be targeted for
AID-induced mutations and DSBs in the absence of AID activity
on S and in the absence of CSR with S. In this context, independent recent studies from our lab have conrmed this nding
and have also conrmed previous observations that normal B cells
activated for CSR to IgG1, under our current conditions, show
more ISD within S compared with S1 (2428). Factors that lead
to large ISD detectable by Southern blotting likely include DSB
frequency within an S region and aspects of S-region sequence
that might inuence end-joining events with respect to rejoining
DSBs within an S region or to a DSB within a separate S region
potentially by using different end-joining pathways (27). Assuming that DSB frequency is a major factor, the higher level of
detectable ISD at S may reect greater AID activity at S than
other S regions because of its transcription, sequence, or other
factors (29). In this context, a high frequency of DSBs in S may
well drive CSR to ensure that DSBs in downstream S regions,
which we now show can be independently introduced, nd a
partner S DSB for CSR (7). This general model also has been
supported by our ndings that independently introduced ISceIendonuclease DSBs within the IgH locus can be joined at long
range by general cellular repair processes to promote recombinational IgH class switching (16). Finally, the nding of greatly
increased levels of AID-initiated events on downstream S regions
in activated S/ B cells is intriguing. One explanation for this
nding is that such events simply accumulate in cells that cannot
undergo efcient CSR in the absence of S. A more interesting
possibility is that deletion of S, which is such an efcient AID
target, allows AID to act more effectively on the downstream S
regions. Although only speculative, such a model could suggest
functional roles for ISD in CSR by promoting increased AID
targeting to downstream S regions.
The S deletion we have studied removes most S AID target
sites (13), removes the major region of S involved in R-loop initiation (18), and greatly decreases the size of the CSR target
region. Therefore, this deletion should vastly diminish the level of
DSBs that could serve as upstream CSR donors, and residual DSBs
presumably should occur through low-level AID targeting by Rloop (41, 42) or nonR-loop mechanisms (7). Thus, given the
apparent role for S DSBs in driving CSR, the dramatic decrease
in CSR to most IgH isotypes in S/ B cells was expected (13).
However, our nding that IgE CSR in S/ B cells is relatively
unimpaired was indeed unexpected, especially considering that
CSR to IgG1 in the same cell population was severely decreased.
One can conceive of several possible explanations for this nding,
but we propose a general model that we consider attractive (Fig.
S6). S CSR junctions in S/ cells must involve direct or indirect
joining of S to a presumably rare DSB in the remaining region
upstream of C, suggesting that increased S DSBs might compensate for loss of S DSBs in S/ B cells and thereby, drive CSR.
In this regard, S/ B cells showed increased ISD and SHM in S1
along with substantial levels of downstream CSR between S1 and
3044 | www.pnas.org/cgi/doi/10.1073/pnas.0915072107

S. Thus, in the absence of S, frequent S1 breaks may drive

downstream CSR with S, potentially then leaving the fused S1/S
region as a major driver for CSR with infrequent breaks upstream
of C. If this model is correct, we must explain why S CSR junctions isolated from S/ cells lacked intervening S1 region
sequences. Although such intermediate sequences might be
eliminated by ongoing ISD in the hybrid S1/S region, a more
intriguing possibility is that the I1 promoter is more effective than
the I promoter in driving AID access to S and that, based on its
S homology, S might serve as a better AID target than S1 (Fig.
S6). Finally, our current work raises the possibility that some
observed sequential CSR (3134) might involve an alternative
pathway in which the downstream CSR provides the rst
intermediate sequence.
Materials and Methods
Mice. S/ mice were generated previously (13) and were maintained with a
homozygous 129/SvJ IgH locus background. The genotype of each experimental mouse was conrmed by Southern blotting analysis. S/ mice were
analyzed at 816 weeks of age with age-matched controls that included WT
mice on 129/SvJ background and AID/ mice (provided by Dr. Tasuku
Honjo). All experiments with mice followed the protocols approved by
Boston Animal Care Facility of the Childrens Hospital, Boston, MA 02115.
CSR Assay and FACS Analysis. B cells were harvested from spleens of agematched mutant mice and controls and puried by selection of the CD43negative population through a mouse B-cell enrichment kit (STEMCELL
TECHNOLOGIES). Puried B cells were cultured in RPMI-based medium
containing 50 M -ME and 15% FCS (as described in ref. 37) in the presence
of anti-CD40 antibodies (eBioscience) and IL-4 (20 ng/mL; eBioscience) to
stimulate CSR to IgG1 and IgE. Cells were kept under 0.5 million/mL and
assayed by surface staining and ow cytometry on days 2, 3, and 4 with antiIgG1-FITC and anti-B220-PE-Cy5 antibodies as previously described (38).
Hybridoma Analysis for CSR. Five to ten million anti-CD40/IL-4stimulated B
cells from each mouse were fused with NS-1 fusion partner myeloma cells on
day 4 and recovered after 7 days selection with 1 Hypoxanthine Aminopterin Thymidine (HAT) medium. Single clones from each well were picked
and screened by ELISA on their supernatants with IgM, IgG1, and IgE capturing and revealing antibodies (Southern Biotech). Only clones that were
single positive for one of the three antibody classes were counted. Clones
that were negative for all IgH isotypes or positive for more than one IgH
isotype were found at very low levels (below 5% of total clones) and not
taken into the calculations.
Somatic Hypermutation Assay. Five to 10 million anti-CD40/IL-4stimulated B
cells were harvested on day 7 for genomic DNA extraction with DNeasy Blood
and Tissue Kit (QIAGEN). iProof High-Fidelity DNA polymerase (Bio-Rad Life
Sciences) was used to amplify the I1/S1 region with forward primer Sg1.2F
(TGTCAATCCTGTTCTTAGTCAATCA) and reverse primer Sg1.2R (CCATCAGCTCTAGCCATGTAGTATT). PCR products were puried and cloned into vectors as
previously described (37). Plasmids with proper inserts were sequenced, and a
1,640-bp region from the 3 end of Ig1 into 5 Sg1 was analyzed.
Southern Blotting Analysis. At least 5 g genomic DNA isolated from each
hybridoma clone was digested with EcoRI or BamHI overnight and run on a
0.7% agarose gel. DNA was transferred from the gels to membranes that
were hybridized with corresponding probes (Figs. 2 and 4 and Figs. S2 and
S4), washed, and put on XAR lm (Kodak Biomax) for exposure.
Switch-Region Junction Cloning. SS and S1S junctions were amplied by
Advantage cDNA polymerase Mix and PCR Kits (Clontech) following their protocols.
For SmSe junction cloning, we used forward primer ImF1 (ACTCAGTCAGTCAGTGGCGTGAAGGGCT) and reverse primer S-1.Rv (CATCAGGCTTTGCTCACTCA).
For S1S junction cloning, we used forward primer Sg1.2F (TGTCAATCCTGTTCTTAGTCAATCA) and reverse primer S-1.Rv (CATCAGGCTTTGCTCACTCA).
ACKNOWLEDGMENTS. We thank Jing H. Wang, Duane Wesemann, Yu Nee
Lee, Feilong Meng, and Michael G. Kharas for stimulating discussions and
Michael Lieber and Barry Sleckman for critical review of the manuscript. This
work was supported by National Institutes of Health Grants AI031541 and
CA092625 (to F.W.A.). C.B. is supported by a Cancer Research Institute training
grant. F.W.A. is an investigator of the Howard Hughes Medical Institute.

Zhang et al.

Zhang et al.

24. Dudley DD, et al. (2002) Internal IgH class switch region deletions are positionindependent and enhanced by AID expression. Proc Natl Acad Sci USA 99:99849989.
25. Reina-San-Martin B, Chen J, Nussenzweig A, Nussenzweig MC (2007) Enhanced intraswitch region recombination during immunoglobulin class switch recombination in
53BP1-/- B cells. Eur J Immunol 37:235239.
26. Reina-San-Martin B, et al. (2003) H2AX is required for recombination between
immunoglobulin switch regions but not for intra-switch region recombination or
somatic hypermutation. J Exp Med 197:17671778.
27. Boboila C, et al. (2009) Alternative End-joining Catalyzes Robust IgH Locus Deletions
and Translocations in the Combined Absence of Ligase 4 and Ku. Proc Natl Acad Sci
USA, in press.
28. Reina-San-Martin B, Chen HT, Nussenzweig A, Nussenzweig MC (2004) ATM is
required for efcient recombination between immunoglobulin switch regions. J Exp
Med 200:11031110.
29. Schrader CE, et al. (2003) Mutations occur in the Ig Smu region but rarely in Sgamma
regions prior to class switch recombination. EMBO J 22:58935903.
30. Xue K, Rada C, Neuberger MS (2006) The in vivo pattern of AID targeting to
immunoglobulin switch regions deduced from mutation spectra in msh2-/- ung-/mice. J Exp Med 203:20852094.
31. Du L, et al. (2008) Involvement of Artemis in nonhomologous end-joining during
immunoglobulin class switch recombination. J Exp Med 205:30313040.
32. Mandler R, Finkelman FD, Levine AD, Snapper CM (1993) IL-4 induction of IgE class
switching by lipopolysaccharide-activated murine B cells occurs predominantly
through sequential switching. J Immunol 150:407418.
33. Mills FC, Mitchell MP, Harindranath N, Max EE (1995) Human Ig S gamma regions and
their participation in sequential switching to IgE. J Immunol 155:30213036.
34. Snapper CM, Finkelman FD, Stefany D, Conrad DH, Paul WE (1988) IL-4 induces coexpression of intrinsic membrane IgG1 and IgE by murine B cells stimulated with
lipopolysaccharide. J Immunol 141:489498.
35. Gu H, Zou YR, Rajewsky K (1993) Independent control of immunoglobulin switch
recombination at individual switch regions evidenced through Cre-loxP-mediated
gene targeting. Cell 73:11551164.
36. Challa A, Pound JD, Armitage RJ, Gordon J (1999) Epitope-dependent synergism and
antagonism between CD40 antibodies and soluble CD40 ligand for the regulation of
CD23 expression and IgE synthesis in human B cells. Allergy 54:576583.
37. Cheng HL, et al. (2009) Integrity of the AID serine-38 phosphorylation site is critical for
class switch recombination and somatic hypermutation in mice. Proc Natl Acad Sci
USA 106:27172722.
38. Yan CT, et al. (2007) IgH class switching and translocations use a robust non-classical
end-joining pathway. Nature 449:478482.
39. Cogn M, et al. (1994) A class switch control region at the 3 end of the
immunoglobulin heavy chain locus. Cell 77:737747.
40. Seidl KJ, et al. (1999) Position-dependent inhibition of class-switch recombination by
PGK-neor cassettes inserted into the immunoglobulin heavy chain constant region
locus. Proc Natl Acad Sci USA 96:30003005.
41. Roy D, Lieber MR (2009) G clustering is important for the initiation of transcriptioninduced R-loops in vitro, whereas high G density without clustering is sufcient
thereafter. Mol Cell Biol 29:31243133.
42. Roy D, Zhang Z, Lu Z, Hsieh CL, Lieber MR (2010) Competition between the RNA
transcript and the nontemplate DNA strand during R-loop formation in vitro: A nick
can serve as a strong R-loop initiation site. Mol Cell Biol 30:146159.

PNAS | February 16, 2010 | vol. 107 | no. 7 | 3045

IMMUNOLOGY

1. Dudley DD, Chaudhuri J, Bassing CH, Alt FW (2005) Mechanism and control of V(D)J
recombination versus class switch recombination: Similarities and differences. Adv
Immunol 86:43112.
2. Honjo T, et al. (1981) Rearrangements of immunoglobulin genes during differentiation
and evolution. Immunol Rev 59:3367.
3. Honjo T (1983) Immunoglobulin genes. Annu Rev Immunol 1:499528.
4. Muramatsu M, et al. (2000) Class switch recombination and hypermutation require
activation-induced cytidine deaminase (AID), a potential RNA editing enzyme. Cell
102:553563.
5. Revy P, et al. (2000) Activation-induced cytidine deaminase (AID) deciency causes the
autosomal recessive form of the Hyper-IgM syndrome (HIGM2). Cell 102:565575.
6. Chaudhuri J, Alt FW (2004) Class-switch recombination: Interplay of transcription,
DNA deamination and DNA repair. Nat Rev Immunol 4:541552.
7. Chaudhuri J, et al. (2007) Evolution of the immunoglobulin heavy chain class switch
recombination mechanism. Adv Immunol 94:157214.
8. Stavnezer J (1996) Immunoglobulin class switching. Curr Opin Immunol 8:199205.
9. Di Noia JM, Neuberger MS (2007) Molecular mechanisms of antibody somatic
hypermutation. Annu Rev Biochem 76:122.
10. Li Z, Woo CJ, Iglesias-Ussel MD, Ronai D, Scharff MD (2004) The generation of
antibody diversity through somatic hypermutation and class switch recombination.
Genes Dev 18:111.
11. Liu M, Schatz DG (2009) Balancing AID and DNA repair during somatic
hypermutation. Trends Immunol 30:173181.
12. Dunnick W, Hertz GZ, Scappino L, Gritzmacher C (1993) DNA sequences at
immunoglobulin switch region recombination sites. Nucleic Acids Res 21:365372.
13. Khamlichi AA, et al. (2004) Immunoglobulin class-switch recombination in mice
devoid of any S mu tandem repeat. Blood 103:38283836.
14. Luby TM, Schrader CE, Stavnezer J, Selsing E (2001) The mu switch region tandem
repeats are important, but not required, for antibody class switch recombination. J
Exp Med 193:159168.
15. Shinkura R, et al. (2003) The inuence of transcriptional orientation on endogenous
switch region function. Nat Immunol 4:435441.
16. Zarrin AA, et al. (2007) Antibody class switching mediated by yeast endonucleasegenerated DNA breaks. Science 315:377381.
17. Chaudhuri J, et al. (2003) Transcription-targeted DNA deamination by the AID
antibody diversication enzyme. Nature 422:726730.
18. Chaudhuri J, Khuong C, Alt FW (2004) Replication protein A interacts with AID to
promote deamination of somatic hypermutation targets. Nature 430:992998.
19. Huang FT, et al. (2007) Sequence dependence of chromosomal R-loops at the
immunoglobulin heavy-chain Smu class switch region. Mol Cell Biol 27:59215932.
20. Yu K, Chedin F, Hsieh CL, Wilson TE, Lieber MR (2003) R-loops at immunoglobulin class
switch regions in the chromosomes of stimulated B cells. Nat Immunol 4:442451.
21. Schrader CE, Linehan EK, Mochegova SN, Woodland RT, Stavnezer J (2005) Inducible
DNA breaks in Ig S regions are dependent on AID and UNG. J Exp Med 202:561568.
22. Stavnezer J, et al. (1999) Switch recombination in a transfected plasmid occurs
preferentially in a B cell line that undergoes switch recombination of its chromosomal
Ig heavy chain genes. J Immunol 163:20282040.
23. Alt FW, Rosenberg N, Casanova RJ, Thomas E, Baltimore D (1982) Immunoglobulin
heavy-chain expression and class switching in a murine leukaemia cell line. Nature
296:325331.