The RNA Exosome Targets the AIDCytidine Deaminase to Both Strandsof Transcribed Duplex DNA SubstratesUttiya Basu,1,2,7,* Fei-Long Meng,1,7 Celia Keim,2,7 Veronika Grinstein,2 Evangelos Pefanis,2 Jennifer Eccleston,1

Tingting Zhang,1 Darienne Myers,1 Caitlyn R. Wasserman,1 Duane R. Wesemann,1 Kurt Januszyk,5 Richard I. Gregory,4

Haiteng Deng,3,6 Christopher D. Lima,5 and Frederick W. Alt1,*1Howard Hughes Medical Institute, Program in Cellular and Molecular Medicine and Immune Disease Institute, Children’s Hospital Boston,

Department of Genetics, Harvard Medical School, 300 Longwood Avenue, Boston, MA 02115, USA2Department of Microbiology and Immunology, College of Physicians and Surgeons, Columbia University, New York, NY 10032, USA3Rockefeller University, Proteomic Research Center, New York, NY 10065, USA4Children’s Hospital Boston, Harvard Stem Cell Institute, Department of Biological Chemistry andMolecular Pharmacology, Harvard MedicalSchool, 300 Longwood Avenue, Boston, MA 02115, USA5Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA6School of Life Sciences, Tsinghua University, Beijing 100084, China7These authors contributed equally to this work*Correspondence: ub2121@columbia.edu (U.B.), alt@enders.tch.harvard.edu (F.W.A.)

DOI 10.1016/j.cell.2011.01.001

SUMMARY

Activation-induced cytidine deaminase (AID) initiatesimmunoglobulin (Ig) heavy-chain (IgH) class switchrecombination (CSR) and Ig variable region somatichypermutation (SHM) in B lymphocytes by deami-nating cytidines on template and nontemplatestrands of transcribed DNA substrates. However,the mechanism of AID access to the template DNAstrand, particularly when hybridized to a nascentRNA transcript, has been an enigma. We now impli-cate the RNA exosome, a cellular RNA-processing/degradation complex, in targeting AID to both DNAstrands. In B lineage cells activated for CSR, theRNA exosome associates with AID, accumulates onIgH switch regions in an AID-dependent fashion,and is required for optimal CSR. Moreover, boththe cellular RNA exosome complex and a recombi-nant RNA exosome core complex impart robustAID- and transcription-dependent DNA deaminationof both strands of transcribed SHM substratesin vitro. Our findings reveal a role for noncodingRNA surveillance machinery in generating antibodydiversity.

INTRODUCTION

Antigen-activated B lymphocytes undergo two distinct immuno-

globulin (Ig) gene diversification processes, namely somatic

hypermutation (SHM) and Ig heavy-chain (IgH) class switch

recombination (CSR). SHM diversifies IgH and Ig light-chain

(IgL) variable region exons to allow generation of B cells with

the potential to secrete higher-affinity antibodies (reviewed by

Odegard and Schatz, 2006; Di Noia and Neuberger, 2007;

Maul andGearhart, 2010). IgH class switch recombination allows

B cells to express different classes of antibodies with different

IgH constant regions (CHs) and, as a result, different antibody

effector functions (reviewed by Chaudhuri et al., 2007; Honjo

et al., 2002). CSR involves joining DNA double-strand breaks

(DSBs) in the large repetitive switch (S) region (Sm) that lies

upstream of the Cm constant region exons to DSBs within

a downstream S region (e.g., Sg1), which replaces Cm exons

with a set of downstream CH exons to complete CSR (e.g.,

switching from IgM to IgG1). Activation-induced cytidine deam-

inase (AID) initiates both SHM and CSR (Muramatsu et al., 2000;

Revy et al., 2000) by deaminating cytidines on, respectively, tran-

scribed IgH or IgL variable region exons or transcribed IgH

S regions (Petersen-Mahrt et al., 2002). The deaminated cyti-

dines become targets of co-opted DNA repair pathways that

lead to mutations associated with variable region exon SHM or

to S region DSBs that initiate CSR (reviewed by Di Noia and Neu-

berger, 2007; Neuberger et al., 2003). To initiate both SHM and

CSR, AID equally deaminates both template and nontemplate

strands of transcribed target DNA sequences (Milstein et al.,

1998; Shen et al., 2006; Xue et al., 2006).

AID is a single-stranded (ss) DNA-specific cytidine deaminase

that lacks activity on double-stranded (ds) DNA (Chaudhuri et al.,

2003; Dickerson et al., 2003; Ramiro et al., 2003; Sohail et al.,

2003). Correspondingly, SHM and CSR require transcription

through duplex substrate V(D)J exons or S regions to target

AID activity, consistent with transcription generating a ssDNA

AID substrate (reviewed by Chaudhuri et al., 2007; Yang and

Schatz, 2007; Di Noia and Neuberger, 2007; Maul and Gearhart,

2010). Transcription through mammalian S regions generates

R loops in which the template strand is hybridized to the nascent

transcript and the nontemplate strand is looped out as ssDNA

Cell 144, 353–363, February 4, 2011 ª2011 Elsevier Inc. 353

(Daniels and Lieber, 1995; Shinkura et al., 2003; Tian and Alt,

2000; Yu et al., 2003a). In biochemical studies, transcription-

generated R loops within duplex substrates allow AID deamina-

tion butmainly on the looped-out nontemplate strand (Chaudhuri

et al., 2003). AID that is phosphorylated on serine 38 by protein

kinase A can access in vitro transcribed dsDNA SHM substrates

(e.g., V(D)J exons) that lack R loop-forming ability—again,

however, mainly on the nontemplate strand (Basu et al., 2005,

2008; Chaudhuri et al., 2004; Zarrin et al., 2004; Vuong et al.,

2009). Ectopically expressed AID mutates transcribed

substrates in bacteria and yeast but, once again, predominantly

on the nontemplate strand (Gomez-Gonzalez and Aguilera,

2007; Ramiro et al., 2003). Thus, the mechanism by which AID

accesses the template DNA strand of substrates has been

a major question (e.g., Chaudhuri et al., 2007; Chelico et al.,

2009; Di Noia and Neuberger, 2007; Liu and Schatz, 2009;

Longerich et al., 2006; Maul and Gearhart, 2010; Pavri et al.,

2010; Peled et al., 2008).

Biochemical studies suggested that AID may gain access to

both DNA strands of certain types of transcribed plasmids via

formation of RNA polymerase-generated supercoils (Besmer

et al., 2006; Shen et al., 2005; Shen and Storb, 2004). However,

such a mechanism does not readily explain how AID accesses

the template strand of a transcribed substrate if it is blocked

by a nascent RNA transcript (Maul and Gearhart, 2010). Indeed,

AID has no known activity on RNA/DNA hybrids, which form over

wide regions of mammalian S regions in the form of R loops

(Huang et al., 2006, 2007; Roy et al., 2008; Yu et al., 2003a)

and which may also form in the context of transcription bubbles

(Gomez-Gonzalez and Aguilera, 2007; Li and Manley, 2005).

RNaseH has been proposed as a candidate for exposing regions

of R-looped S region template strands to AID based on biochem-

ical ability to degrade RNA in the context of R loops (Lieber,

2010; Yu and Lieber, 2003b); however, to date, there has been

no direct evidence that implicated either RNase H or other

cellular RNA degradation factors in CSR.

The RNA exosome is an evolutionarily conserved RNA-pro-

cessing/degradation complex (reviewed by Houseley et al.,

2006; Lykke-Andersen et al., 2009; Schmid and Jensen, 2008;

Shen and Kiledjian, 2006). The nine-subunit core of the eukaryotic

RNA exosome, which includes the CsL4, Rrp4, Rrp40, Rrp41,

Rrp46,Mtr3, Rrp42, Rrp43, andRrp45 subunits, hasRNA-binding

activitybut lacks ribonucleaseactivity (Andersonetal., 2006;Grei-

mann and Lima, 2008; Oddone et al., 2007; Figure S1 available

online). RNA exosome ribonuclease activity is provided by non-

core subunits, including Rrp6 and Rrp44, that have RNA 30-50

exonuclease or endonuclease activity (Greimann and Lima,

2008; Houseley et al., 2006; Lebreton et al., 2008; Schaeffer

et al., 2009; Figure S1). The mammalian RNA exosome complex

interacts with cofactors, such as the TRAMP complex, that target

it toparticularsubstratesbasedonsequenceorstructural features

(reviewed by Houseley et al., 2006; Houseley and Tollervey, 2008;

LaCava et al., 2005). In Drosophila, the RNA exosome complex

interacts with elongating RNA polymerase II (Pol II) complexes

via the Spt5/6 transcription elongation cofactors (Andrulis et al.,

2002), and in yeast, it binds to and removes nascent RNP

complexes from template DNA (El Hage et al., 2010; Houseley

et al., 2006; Houseley and Tollervey, 2008). We now describe

354 Cell 144, 353–363, February 4, 2011 ª2011 Elsevier Inc.

studies that implicate the RNA exosome as a long-speculated

cofactor that targets AID deamination activity to both template

and nontemplate strands of transcribed dsDNA substrates.

RESULTS

AID Associates with the RNA Exosome ComplexTo elucidate factors that might promote AID access to the

template strand of transcribed substrates in the context

of RNA/DNA hybrid structures, we in vitro transcribed an

RGYW-rich dsDNA SHM substrate with T7 RNA polymerase in

the presence of Ramos human B cell lymphoma line extracts.

Transcribed DNA-protein complexes were purified via

chromatographic steps that would enrich for DNA binding

(CM-sepharose, DEAE cellulose) and macromolecular complex

formation (gel-filtration chromatography), followed by heparin

sepharose chromatography and anti-AID antibody-mediated

affinity purification to enrich complexes containing AID (Figures

1A and 1B, Figure S1B, Table S1, and Extended Experimental

Procedures). At each step, fractions enriched for AID deamina-

tion stimulatory activity were identified via a 3H-release assay

(e.g., Figure S1B and Extended Experimental Procedures).

Among proteins identified by mass spectrometric analysis of

purified complexes were multiple subunits of the RNA exosome

complex, including Mtr3, Csl4, Rrp43, Rrp40, and Rrp42 (Fig-

ure 1B and Table S1). To further elucidate potential functions,

we assayed the ability of the AID-associated, transcribed DNA

complex to enhance deamination activity of purified AID in a tran-

scribed dsDNA SHM substrate assay (Chaudhuri et al., 2003)

and found that it markedly stimulated AID activity (Figure 1C).

Of note, AID association with the RNA exosome complex and

purification of an AID stimulatory activity was not observed if

the purification was performed from a reaction without T7

polymerase, indicating that complex formation is enhanced by

transcription (Figure S1B and data not shown).

To assay for AID/RNA exosome association in vivo, we immu-

noprecipitated AID from mouse primary splenic B cells stimu-

lated with anti-CD40 plus IL-4 to induce AID and CSR to IgG1

and then assayed immunoprecipitates for RNA exosome

subunits by western blotting. These analyses revealed that AID

associated with Rrp40, Rrp46, and Mtr3 RNA exosome subunits

(Figure 2A). Likewise, AID immunoprecipitated from mouse

CH12F3 B lymphoma cells stimulated with anti-CD40, IL-4,

and TGF-b to undergo CSR to IgA, as well as AID immunoprecip-

itated from the humanRamosB lymphoma cells, associatedwith

core Rrp40 andRrp46 subunits, as well as with the Rrp6 catalytic

exosome subunit (Figure 2B). By individually expressing FLAG-

tagged versions of RNA exosome subunits along with AID in

HEK293T cells, we observed that immunoprecipitation of any

of the 11 exosome core and catalytic subunits via anti-FLAG

antibodies also pulled down AID (Figure 2C). Together, our find-

ings indicate that AID either directly or indirectly associates with

the RNA exosome complex in cells.

Exosome Core Subunit Rrp40 Is Requiredfor Optimal CSRPrevious work indicates that the absence of a given core exo-

some subunit leads to a severe defect in overall RNA exosome

A B

C

Figure 2. AID Complexes with RNA Exosome Subunits In Vivo

(A) AID immunoprecipitates from extracts of CSR-activated AID-deficient and

wild-type B cells were assayed for Rrp40, Rrp46, Mtr3, and AID (indicated on

the right) via western blotting. The left two lanes show western blotting of total

extract, and the right two lanes show western blotting of immunoprecipitated

products.

(B) AID immunoprecipitates (‘‘Anti-AIDIP’’ lanes) or control reactions without

anti-AID antibody (‘‘�AbIP’’ lanes) from Ramos (‘‘Ramos’’) and CSR-activated

CH12F3 cells (‘‘CH12F3 Sti’’) were assayed for Rrp46, Rrp40, Mtr3, and AID

(indicated on right) by western blotting. Unstimulated CH12F3 cells (‘‘CH12F3

unsti’’) were used as a negative control.

(C) AID was coexpressed with individual FLAG epitope-tagged RNA exosome

subunits in HEK293T cells. Lanes from left to right represent cells transfected

with empty vector (‘‘vector’’) as a control or the individual FLAG epitope-tag-

ged subunit indicated at the top. The top two panels show western blotting of

total extract (‘‘input’’), and the bottom three panels show western blotting with

indicated antibodies (anti-AID, anti-Rrp40, and anti-Rrp6) following immuno-

precipitation with anti-Flag antibodies. Asterisks indicate bands correspond-

ing to Flag-tagged exosome subunits. A background band corresponding to

the anti-Flag Ig light chain also is indicated (‘‘Background’’).

A B

C

Figure 1. AID Forms a Transcription-Dependent Complex with RNA

Exosome

(A) Schematic outlining steps for enrichment of transcription-dependent

AID/RNA exosome/SHM substrate complex. Details are in text and Extended

Experimental Procedures.

(B) Proteins enriched by purification scheme in (A) were analyzed by

SDS-PAGE followed by staining with Coomassie blue. Identity of proteins from

selected bands was determined by mass spectrometry; bands that contain

RNA exosome subunits are indicted on the right withmolecular weight markers

on the left.

(C) AID purified following ectopic expression in HEK293 cells was assayed in

a 3H-release in vitro transcription-dependent SHM substrate assay (Chaudhuri

et al., 2003; see also Figure 5) in the presence or absence of complex enriched

by purification scheme in (A). Percent of total transcribed DNA substrate

deaminated is presented for three separate assays.

See also Figure S1 and Table S1.

function (Jensen andMoore, 2005). Therefore, to evaluate poten-

tial roles of the RNA exosome complex in CSR, we used a knock-

down approach to reduceRrp40 in theCH12F3B lymphoma line.

For this purpose, we lentivirally introduced two different shRNAs

that targeted Rrp40, respectively, into CH12F3 cells to generate

three different knockdown lines that had Rrp40 levels ranging

from about 50% to less than 10% those of controls, including

aWT line anda line harboring a nonspecific (‘‘scrambled’’) shRNA

(Figure3BandFigureS2C). Following stimulationwith anti-CD40,

TGFb, and IL4 for 48 hr to induce CSR to IgA, Rrp40 knockdown

lines consistently displayed reduced CSR, with levels ranging

from 30%–50% of those of controls (Figures 3A and 3C and Fig-

ure S2B). However, the various Rrp40 knockdown lines prolifer-

ated similarly to controls after stimulation (Figure 3E and Figures

S2A and S2E). In addition, the knockdown and control lines ex-

pressed similar levels of Im and Ia transcripts and similar levels

of AID; whereas there were variations in transcript levels from

clone to clone, there was no correlation with Im and Ia transcripts

and Rrp40 levels (Figure 3D and Figure S2F). Finally, similar

knockdowns of the Mtr3 RNA exosome core subunit also led to

decreased CSR without markedly affecting cell proliferation,

AID levels, or Im and Ia transcription (Figure S3). Together, these

findings demonstrate that physiological levels of RNA exosome

core subunits are required for efficient CSR.

Cell 144, 353–363, February 4, 2011 ª2011 Elsevier Inc. 355

A

B C

D E

Figure 3. RNA Exosome Subunit Rrp40 Is Required for Normal CSR

(A) CH12F3 cells lentivirally infected with a scrambled short-hairpin plasmid (NS) or with shRNA against Rrp40 (shRrp40) were either not stimulated (‘‘Unsti’’) or

stimulated (‘‘Sti’’) for 2 days with anti-CD40, IL4, and TGFb and analyzed for IgA CSR by flow cytometry. ShRrp40-1 and shRrp40-2 are independent shRrp40-

expressing CH12F3 isolates. Results are representative of eight experiments; additional experiments are shown in Figures S2A and S2B.

(B) NS, shRrp40-1, and shRrp40-2 expressing stimulated and unstimulated CH12F3 isolates (shown in A) were assayed for Rrp40 and AID by western blotting.

Results are representative of four experiments; additional experiment is shown in Figure S2.

(C) Average levels and standard deviation from the mean of CSR to IgA from three independent experiments (one shown in A) performed simultaneously with

unstimulated (‘‘unsti’’) NS and stimulated (‘‘Sti’’) NS, shRrp40-1 (‘‘#1’’), and ShRrp40-2 (‘‘#2’’) CH12F3 isolates. Five additional experiments gave similar results

(Figures S2A and S2B).

(D) Total cellular RNA from three independently stimulated samples of indicated CH12F3 isolates (the ones used for C) was assayed for Im transcripts (left) and Ia

transcripts (right) via quantitative RT-PCR. Average and standard deviation from the mean is shown for the three separate experiments. An additional experiment

based on northern or RT-PCR is shown in Figure S2F.

(E) Growth curves of stimulated (NS) shRrp40-1 and shRrp40-2 CH12F3 cells calculated from three independent sets of three experiments (one used for C and

others shown Figure S2A) with a fourth set of three experiments indicated in Figure S2E. Values represent average and standard deviation from the mean. See

also Figure S2 and Figure S3.

Association of Rrp40 with S Regions in B CellsActivated for CSRIf the RNA exosome functions to target AID activity, it should

be found in association with transcribed S regions in B cells

356 Cell 144, 353–363, February 4, 2011 ª2011 Elsevier Inc.

activated for CSR. To evaluate this possibility, we performed

chromatin immunoprecipitation (ChIP) assays to test whether

Rrp40 associates with transcribed Sm sequences in activated

CH12F3 cells before and after stimulation for CSR to IgA.We first

A B

C D

Figure 4. RNA Exosome Subunit Rrp40 Is Recruited to S Regions

(A) Ch12F3 cells were either stimulated with TGFb, IL4, and CD40 or kept

unstimulated for 48 hr. Subsequently, Rrp40 was immunoprecipitated from

cell extracts under chromatin immunoprecipitation (ChIP) conditions and

immunoprecipitates analyzed for Rrp40 by western blotting with anti-Rrp40.

(B) The ‘‘ChIPed’’ Rrp40-DNA complex from Ch12F3 cells was processed to

isolate bound DNA. ChIPed DNA was tested for Sm and Sa sequences via

q-PCR. The average and standard deviations from themean for three separate

ChIP experiments are shown (see also Figure S4). Numbers indicate average

fold changes comparing stimulated and unstimulated samples. Unstimulated

samples were arbitrarily normalized as 1 (Experimental Procedures).

(C) Rrp40 ChIPs were performed on extracts from primary splenic B cells

stimulated for 2 days with anti-CD40 plus IL4. Sm and Sg1 were tested via

semiquantitative PCR; results are shown for two independent ChIP samples

for each genotype. A 5-fold serial dilution of inputs is shown with the highest

input concentration corresponding to 1/20 of total input.

(D) ChIPed DNA from activated splenic B cells was tested for Sm and Sg1 via

q-PCR. Numbers indicate average fold changes comparing WT and AID�/�

samples. AID�/� samples were arbitrarily normalized as 1 (Experimental

Procedures). Values represent the average and standard deviation from the

mean for three experiments.

See Figure S4 for more details.

employed western blotting to confirm that Rrp40 is specifically

precipitated with an anti-Rrp40 antibody, but not control IgG,

under ChIP conditions (Figure 4A). After processing immunopre-

cipitates for isolation of bound DNA, we utilized quantitative PCR

(q-PCR) to determine levels of Rrp40 bound to Sm and Sa. These

analyses demonstrated enrichment of Sm and, to a lesser extent,

Sa in the anti-Rrp40 ChIPs from stimulated versus unstimulated

CH12F3 cells (Figure 4B and Figure S4). As there is some Sm

transcription in nonactivated B cells (Muramatsu et al., 2000),

the question arises as to whether transcription per se is sufficient

to recruit the RNA exosome to S regions. To explore this ques-

tion, we assayed for Rrp40 recruitment to Sm and Sg1 in WT

and AID-deficient primary B cells activated with anti-CD40 and

IL-4, which induce germline Sg1 transcription in both cell types

(Muramatsu et al., 2000). Consistent with targeting dependent

on germline transcription, Rrp40 was recruited to Sm and Sg1

in the activated WT B cells (Figures 4C and 4D and Figure S4).

Of note, however, Rrp40 was not measurably recruited to Sm

and Sg1 in AID-deficient B cells. Together, our results indicate

that the RNA exosome complex is recruited to transcribed S

regions in B cells activated for CSR in an AID-dependent fashion.

RNA Exosome Stimulates AID Activity on Templateand Nontemplate StrandsTo further evaluate the potential ability of the cellular RNA

exosome complex to act as an AID cofactor, we substantially

purified this complex from cell-free nuclear extracts prepared

from HEK293T cells that expressed a FLAG epitope-tagged

Rrp6 exosome subunit. In this purification, we maintained rela-

tively low salt concentrations to prevent disaggregation of

protein complexes (Figure S5). To test activity, we added varying

amounts of the exosome-enriched extract to a 3H-uracil-release

in vitro transcription-dependent AID deamination assay, which

measures overall SHM substrate deamination (Figure 5A). In

this assay, T7 polymerase transcription of the SHM substrate

leads to little or no AID deamination activity, and addition of

partially purified RNA exosome extract in the absence of AID

also gives no deamination activity on the T7 transcribed

substrate (Figure 5B). However, addition of both AID and partially

purified RNA exosome led to substantial deamination of the

transcribed substrate (Figure 5B), with activity appearing to be

roughly within a range similar to that observed with phosphory-

lated AID and RPA (Basu et al., 2005, 2008; Chaudhuri et al.,

2004; see below). The AID deamination stimulatory activity

observed in these extracts is likely mediated by the exosome

complex, as we found that deamination activity cofractionated

with the RNA exosome during purification (Figure S5). Likewise,

we found similar results when we purified the exosome complex

from HEK293T cells via an approach in which affinity purification

of the complex was performed with antibodies against endoge-

nous Rrp40 (Figure S5D). Finally, we found that the RNA

exosome also stimulated AID deamination of a transcribed

dsDNA core Sm substrate and a synthetic R loop-forming

substrate (Figure 5B).

Because the RNA exosome can associate with Pol II transcrip-

tion complexes and remove nascent transcripts from transcribed

DNA (El Hage et al., 2010), we considered it as a candidate AID

cofactor for template DNA strand deamination. To test this

possibility, we performed in vitro transcription-dependent AID

dsDNA SHM substrate deamination assays in which the

Southern blotting readout reveals deamination of either template

or nontemplate strands, respectively (Figure 5C) (Chaudhuri

et al., 2004). In this assay, no deamination of either strand was

observed when only AID was added in the presence or absence

of T7 polymerase (Figure 5D and Figure S6A). As observed previ-

ously, addition of PKA (to phosphorylate AID on S38) and RPA

along with T7 polymerase and AID led to deamination of the

nontemplate strand, but not the template strand (Figure 5D

and Figure S6A). Strikingly, addition of both AID and partially

purified RNA exosome (in the absence of RPA or PKA) to the

T7 transcription reaction led to deamination of both strands of

the SHM substrate (Figure 5D and Figures S6A and S6E). In

most assays, activity on both template and nontemplate strands,

respectively, was robust, as evidenced by greatly diminished

Cell 144, 353–363, February 4, 2011 ª2011 Elsevier Inc. 357

A B

C D

Figure 5. Cellular RNA Exosome Augments

Transcription-Dependent AID Deamination

Activity on Template and Nontemplate

DNA Strands

(A) Schematic representation of 3H release assay

for AID deamination of transcribed dsDNA SHM

substrate.

(B) (Top) Results of 3H release assay in which

a SHM substrate was transcribed by T7 poly-

merase (T) in the presence of purified AID (AID),

purified HEK293 RNA exosome (Exo293), or both.

(Middle) Results of 3H release assay in which

a synthetic R loop-forming substrate was tran-

scribed by T7 polymerase (T) in the presence of

purified AID (AID), recombinant RNA exosome

(rExo), or both. (Bottom) Results of 3H release

assay in core Sm substrate were transcribed by T7

polymerase (T) in the presence of purified AID

(AID), purified HEK293 RNA exosome (Exo293), or

both. In all three panels, values represent average

and standard deviation from the mean from three

independent experiments.

(C) A schematic representation of an assay for

measuring strand-specific AID deamination of

a transcribed dsDNA SHM. The location of

template (T) and nontemplate (NT) strand probes is

indicated.

(D) The strand specificity of RPA-dependent or

RNA exosome-dependent DNA deamination was

analyzed by the assay in (C) using either non-

template (left) or template (right) strand-specific

probes. Reactions contained AID, T7 polymerase

(T), RPA, and PKA or purified HEK293 exosome

(‘‘Exosome’’) as indicated.

See also Figure S5 and Figure S6.

levels of full-length substrate strands (Figure 5D and Figure S6A).

The RNA exosome also enhances AID template strand deamina-

tion activity on transcribed core Sm substrate and synthetic R

loop-forming substrates (Figure 5B and Figures S6B and S6C).

Together, our findings indicate that the endogenous RNA exo-

some complex can function as a stimulatory cofactor for AID

deamination activity on both strands of transcribed duplex

SHM substrates.

Recombinant Core RNA Exosome StimulatesAID-Dependent Deamination on Templateand Nontemplate StrandsThe partially purified endogenous RNA exosome preparation

should contain the core structural complex as well as catalytic

subunits (e.g., Rrp6, which was used to pull down the complex)

and potentially other cofactors or contaminating proteins.

To unequivocally evaluate the ability of theRNAexosome to stim-

ulate AID activity on transcribed SHM substrates and to also

358 Cell 144, 353–363, February 4, 2011 ª2011 Elsevier Inc.

obtain insight into potential mechanisms,

we assayed a nine-subunit recombinant

human RNA exosome core complex

reconstituted from subunits that were

individually purified subsequent to

expression in bacteria (Figure 6A) (Grei-

mann and Lima, 2008; Liu et al., 2006). We found that, when

titrated for optimal activity (Figure S6E), the recombinant RNA

exosome core complex robustly stimulated AID-dependent

deamination of transcribed SHM substrates, as measured by

the H3-uracil release assay (Figure 6B and Figure S6F). In fact,

maximum deamination levels stimulated by the recombinant

core complexwere roughly similar to thoseobtainedwith partially

purified RNA exosome complex (Figure 6B and Figure S6F).

Moreover, the recombinant RNA exosome core complex also

stimulated AID-dependent deamination of both template and

nontemplate strands of the SHM substrate (Figure 6C and Fig-

ure S6G) and synthetic R loop-forming substrates (Figure S6C).

Thus, the core RNA exosome complex alone, in the absence of

the RNAase catalytic subunits, promotes AID- and transcrip-

tion-dependent SHM substrate deamination.

To further characterize potential functions of the core RNA

exosome, we assayed individual core subunits (Figure 6D and

Figure S7). Individual subunits were assayed at approximately

A B

C D

Figure 6. Recombinant Core RNA Exosome

and Individual Subunits Stimulate AID

Activity

(A) Coomassie blue stained polyacrylamide gel

electrophoresis analysis of the nine-subunit

recombinant RNA exosome complex generated

from individual subunits.

(B) Comparative analysis of the ability of RNA

exosome complex from 293T cells (Exo293) and

recombinant RNA core exosome complex (rExo)

to stimulate AID deamination activity as measured

by 3H release/SHM substrate assay in Figure 5A.

See Extended Experimental Procedures and

Figure S6 for details. Optimal amounts of RNA

exosome293T and recombinant RNA exosome

core complex were used (Figure S6 and Extended

Experimental Procedures).

(C) Assay of recombinant exosome complex to

stimulate strand-specific AID deamination of

a transcribed SHM substrate via assay in Fig-

ure 5C. Nontemplate and template strand deami-

nation are shown on left and right panels,

respectively. Reactions contained either AID,

recombinant core RNA exosome (rExo9wt), or both

(AID + rExo9wt) as indicated.

(D) Individual recombinant RNA exosome subunits

were assayed for ability to promote AID deami-

nation of a T7-transcribed SHM substrate as out-

lined in Figure 5A. Added exosome components

are indicated. Control reactions with AID alone or

AID plus T7 polymerase are on the left. A positive

control with complete recombinant core RNA

exosome (rExo9) plus T7 and AID is on the right.

For (B) and (D), values represent the average

and standard deviation from the mean for three

independent experiments. See also Figure S7 and

Figure S8.

the same molarity as when assayed in the context of the core

complex. Some, but not all, recombinant core RNA exosome

subunits stimulated AID- and transcription-dependent SHM

substrate deamination activity; however, the level of stimulation

was low compared to that of the full core complex (Figure 6D and

Figure S7A). The RNA-binding subunits Rrp40, Rrp46, Rrp41,

and Mtr3 showed low levels of activity (Figure 6D and Figures

S7A and S7C). On the other hand, other core subunits with

demonstrated RNA-binding activity and/or RNA-binding

domains, along with the Rrp6 ribonuclease subunit, lacked

detectable activity, indicating that the observed activity is not

a property of all such proteins.

DISCUSSION

Enigma of AID Targeting of the TemplateSubstrate StrandRecent studies revealed a mechanism by which AID can be

brought to transcribed substrates in the context of RNA poly-

merase II pausing (Pavri et al., 2010). Once recruited to the

substrate, transcription-basedmechanismsalso provide a poten-

tial means of providing AID with a ssDNA substrate in the context

of a duplex substrate via looping out of the nontemplate strand

(Chaudhuri et al., 2003, 2004; Gomez-Gonzalez and Aguilera,

2007; Ramiro et al., 2003; Yu and Lieber, 2003). However, AID

equally deaminates both substrate DNA strands during CSR and

SHM (Milstein et al., 1998; Shen et al., 2006; Xue et al., 2006). In

the latter context, the mechanism by which AID generates DSB

intermediates that are substrates for CSR (Petersen et al., 2001;

Schrader et al., 2005; Wuerffel et al., 1997) and for chromosomal

translocations (Ramiro et al., 2004; Robbiani et al., 2008) requires

targeting of both DNA strands (e.g., Xue et al., 2006; reviewed by

DiNoia andNeuberger, 2007). Thus, themechanismbywhichAID

accesses the templateDNAstrandhasbeenamajorquestion (e.g.

Chaudhuri et al., 2007; Chelico et al., 2009; Di Noia and Neu-

berger, 2007; Liu and Schatz, 2009; Longerich et al., 2006; Maul

and Gearhart, 2010; Pavri et al., 2010; Peled et al., 2008). We

now show that the core RNA exosome complex promotes AID

deamination of both template and nontemplate strands of

in vitro transcribedSHMsubstrates.Moreover, inBcells activated

for CSR, the RNA exosome complex associates with AID and

accumulates on S regions in an AID-dependent manner. Finally,

integrity of the RNA exosome complex is required for optimal

CSR. Thus, the RNA exosome is a long-sought cofactor that can

target AID activity to both template and nontemplate strands of

transcribed SHM and CSR targets.

Implications of AID/RNA Exosome BiochemistryOur in vivo knockdown, ChIP, and physical association studies

provide strong biological evidence that the RNA exosome

Cell 144, 353–363, February 4, 2011 ª2011 Elsevier Inc. 359

functions in CSR. However, such studies often provide limited

mechanistic insight. In addition, as elimination of RNA exosome

core subunits is cell lethal in yeast, it may be difficult to achieve

complete knockdowns or knockouts to fully test the extent of

exosome function in promoting AID activity in vivo. Classical

biochemistry has been extremely useful for elucidating potential

functions of essential replication, repair, and recombination

factors (e.g., Chaudhuri et al., 2004; Dzantiev et al., 2004; Gen-

schel and Modrich, 2003; Maldonado et al., 1996; Stillman and

Gluzman, 1985). In this context, our biochemical studies reveal

that the core RNA exosome robustly targets AID to both strands

of T7 polymerase-transcribed SHM substrates in vitro. Clearly,

T7 polymerase is structurally and mechanistically distinct from

mammalian RNA PoI II, and the in vitro system that we have

employed does not operate in the chromatin context in which

SHM and CSR occur in vivo. Also, the concentration of the

RNA exosome and of other factors in our assays may be greater

than their cellular levels. But again, it is just these aspects of the

biochemical assay that allow us to definewhat the RNA exosome

has the potential of achieving. Additional mechanisms and

factors (e.g., AID) likely will work to specifically recruit the RNA

exosome to transcribed SHM and CSR targets in vivo (see

below), potentially increasing local concentration to levels suffi-

cient to achieve activities observed in vitro.

The evolutionarily conserved RNA exosome complex is

required for degradation and/or processing of various RNAs,

including noncoding RNAs generated in cells due to error-prone

transcription or from transcription initiated by cryptic promoters

(Houseley et al., 2006; Jensen and Moore, 2005; Lykke-Ander-

sen et al., 2009; Preker et al., 2008). In bacteria, the exosome

core complex contains ribonuclease activity; however, in yeast

and mammalian cells, the core subunits lack such activity and

generally have been considered structural scaffolds (Greimann

and Lima, 2008; Houseley et al., 2006; Jensen and Moore,

2005). In eukaryotes, noncore exosome subunits or cofactors

provide ribonuclease activity (Callahan and Butler, 2010; Grei-

mann and Lima, 2008; Houseley et al., 2006; Lebreton et al.,

2008; Schaeffer et al., 2009; Staals et al., 2010; Tomecki et al.,

2010). Our biochemical data demonstrate that the recombinant

nine-subunit exosome core, in the absence of catalytic subunits,

robustly targets AID to both strands of T7 transcribed substrates.

The precise molecular mechanism by which the RNA exosome

core achieves this activity remains to be determined. In theory,

it may compete with duplex DNA for binding the nascent tran-

script, thereby displacing it from the template strand, while

also forming a scaffold that stabilizes the transcriptionally

opened template to expose both DNA strands. Our finding that

certain core exosome subunits have low but significant ability

to enhance AID deamination of in vitro transcribed substrates

is intriguing. Currently, however, there is no clear correlation

between any particular known subunit activity and which

subunits promote low-level AID access.

We previously have shown that RPA, a ssDNA-binding protein

involved in replication and repair, binds AID phosphorylated on

S38 and, thereby, targets AID to transcribed SHM substrates

in vitro, albeit mostly to the nontemplate strand (Basu et al.,

2005; Chaudhuri et al., 2004). A substantial amount of evidence

supports the significance of the AID/RPA interaction for AID

360 Cell 144, 353–363, February 4, 2011 ª2011 Elsevier Inc.

function during CSRand SHM, both in augmenting AID deamina-

tion activity and also potentially in recruiting downstream repair

pathways involved in further processing AID-initiated lesions

(Basu et al., 2005, 2008; Cheng et al., 2009; McBride et al.,

2006, 2008; Pasqualucci et al., 2006). Of note, RNA exosome-

mediated AID-targeting activity in our in vitro SHM assay does

not require AID phosphorylation on S38 or RPA, demonstrating

that the RNA exosome function in AID targeting is distinct.

However, these two AID-targeting mechanisms still might be

complementary in vivo (see below).

Potential Mechanism by Which the RNA ExosomePromotes AID Activity In VivoBoth CSR and SHM in activated B cells require transcription

through target S regions or variable region exons. Such tran-

scriptional targeting was proposed to result from association

of a mutator factor with stalled Pol II within the target sequence

(Peters and Storb, 1996). Over the years, this model has been

supported by numerous findings, including the following: AID,

now known to be the hypothetical mutator, is targeted by tran-

scription (reviewed by Chaudhuri et al., 2007; Yang and Schatz,

2007; Di Noia and Neuberger, 2007); AID associates with Pol II

(Nambu et al., 2003); and Pol II indeed stalls during transcription

through S regions (Rajagopal et al., 2009; Wang et al., 2009),

potentially augmented by transcribed S regions forming

secondary structures such as R loops (reviewed by Maul and

Gearhart, 2010). Recent studies have implicated the Spt5 tran-

scriptional elongation cofactor in recruiting AID to stalled Pol II

and, thereby, bringing AID to transcribed genomic targets (Pavri

et al., 2010). However, the AID/Spt5/Pol II interaction did not

explain how AID actually accesses its target sequences once

there and, in particular, how AID accesses the template DNA

strand (Pavri et al., 2010), a process in which we now have impli-

cated the RNA exosome. The recruitment of the RNA exosome

to S regions in B cells activated for CSR, in theory, might involve

a putative Spt5/RNA exosome interaction (Andrulis et al., 2002)

or generation of the noncoding S region transcripts (Lennon

and Perry, 1985 Lutzker and Alt, 1988), the latter of which might

attract the RNA exosome complex (reviewed by Houseley et al.,

2006; Jensen and Moore, 2005; Schmid and Jensen, 2008).

However, the marked AID dependence of exosome recruitment

to targeted S regions in activated B cells rules out Spt5 interac-

tion or germline transcription alone for RNA exosome recruit-

ment. Though AID might directly recruit the RNA exosome

complex, we thus far have not found interaction with individual

purified subunits, raising the additional possibility of indirect

recruitment, for example, via an AID/transcription-related

complex.

We propose a working model for how the RNA exosome may

enhance targeting of AID to both DNA strands of its in vivo

substrates (Figure S8). This model is based on our current

biochemical and cellular findings regarding relationships

between AID and the RNA exosome, known aspects of AID

function and recruitment to transcribed target sequences, and

known RNA exosome properties. To enhance AID activity on

the template strand, the RNA exosome must in some way

remove the template RNA. As the RNA exosome is not known

to engage RNA substrates that lack a free single-stranded 30

end, cotranscriptional RNA exosome activity in which an RNA

in the RNA/DNA hybrid is still attached to the RNA polymerase

would seem unlikely. Therefore, we suggest that the RNA

exosome would have its relevant activity on stalled Pol II units

that backtrack to reveal a free 30 end (Adelman et al., 2005) (Fig-

ure S8A). In this context, stalled RNA polymerase recruits AID

via the Spt5 transcription factor (Pavri et al., 2010). Thus, Pol II

stalling would indirectly recruit the RNA exosome via AID (Fig-

ure S8B). Likewise, Pol II stalling would also provide a suitable

substrate to engage the RNA exosome and allow it to enhance

AID substrate activity (Figure S8B), potentially via mechanisms

mentioned in the preceding section of the Discussion or below.

We further suggest that RPA might bind to and stabilize ssDNA

targets opened up by the RNA exosome and facilitate S38-phos-

phorylated AID accumulation via direct interaction (Basu et al.,

2005, 2008; Vuong et al., 2009) (Figures S8C and S8D). In addi-

tion, bound RPA may facilitate engagement of downstream

repair pathways (Chaudhuri, Khuong, and Alt, 2004; Vuong

et al., 2009). We must note that it remains possible, in vivo,

that the catalytic subunit of the RNA exosome could provide

ribonuclease activity that would degrade the nascent transcript

from the DNA/RNA hybrids on transcribed AID substrates and,

thereby, further contribute to exposing the template strand to

AID activity (Figure S8E). Finally, we do not rule out the possibility

that RNA exosome, in the context of an AID complex, acquires

unique properties as compared to properties already described

for the RNA exosome and that these unique properties are

relevant to enhancing AID activity.

EXPERIMENTAL PROCEDURES

Recombinant Plasmids, Antibodies, and Proteins

RNA exosome subunits, AID, and RPA were expressed as described (Grei-

mann and Lima, 2008; Chaudhuri et al., 2003, 2004; Stillman and Gluzman,

1985). AID antibodies were generated as described (Chaudhuri et al., 2003),

and anti-exosome subunit antibodies were purchased from Genway or

AbCam. Exosome subunit and complex purifications were as described

(Greimann and Lima, 2008). Details are in Extended Experimental Procedures.

AID Activity Assays

The 3H release assay was performed essentially as described (Chaudhuri

et al., 2003; Basu et al., 2005). The strand-specific AID and transcription-

dependent dsDNA deamination assays were performed as described (Chaud-

huri et al., 2003, 2004). Details are in Extended Experimental Procedures.

AID Complex Purification from Ramos Cells

We used a combination of DNA affinity, size chromatography, and affinity

purification for isolating AID complexes. We analyzed components of the

complex via mass spectrometry. For details, see Extended Experimental

Procedures.

RNA Exosome Purification from 293T Cells

We generated protein extracts from HEK293 cells expressing FLAG epitope-

tagged Rrp6. We used a combination of size chromatography, glycerol

gradient sedimentation, and FLAG affinity purification to isolate a protein

complex enriched with RNA exosome complex. We also isolated RNA

exosome from HEK293 cells using the same approach except that the affinity

purification step employed anti-Rrp40 antibodies. For details, see Extended

Experimental Procedures.

Immunoprecipitation of AID/RNA Exosome Complex

from B Cells and HEK293 Cells

We utilized previously published protocols to immunoprecipitate the AID

complex from B cells (Chaudhuri et al., 2004; Basu et al., 2005) or used

similarly adapted protocols for immunoprecipitating FLAG-exosome subunits

from 293 cells. For details, see Extended Experimental Procedures.

CSR Analysis of Rrp40-Deficient CH12F3 Cells

We performed shRNA-mediated knockdowns in Ch12F3 according to TRC

shRNA library protocols. Knockdown efficiencies were measured by detecting

levels of Rrp40 in these cells by western blotting. We determined levels of

germline S region transcripts in Rrp40-deficient cells via previously published

protocols (Muramatsu et al., 2000). For details, see Extended Experimental

Procedures.

Chromatin Immunoprecipitation Assays

We utilized adaptations of previously published protocols (Chaudhuri et al.,

2004) for isolating chromatin immunoprecipitates of Rrp40. See Extended

Experimental Procedures for details.

SUPPLEMENTAL INFORMATION

Supplemental Information includes Extended Experimental Procedures, eight

figures, and one table and can be found with this article online at doi:10.1016/j.

cell.2011.01.001.

ACKNOWLEDGMENTS

We thank Bjoern Schwer, Jason Rodriguez, Stephen Goff, Bruce Stillman, and

Tasuku Honjo for providing reagents and/or protocols. We also thank Bjoern

Schwer and Cosmas Giallourakis for critically reading the manuscript. This

work was supported by National Institute of Health grants AI31541 (to

F.W.A.) and GM079196 (to K.J. and C.D.L.); by Trustees of Columbia Univer-

sity Faculty startup funds (to U.B.); and by funds from Regeneron Pharmaceu-

ticals (in support of E.P.). F.-L.M. is a fellow of Cancer Research Institute of

New York. C.K. was supported by a Cancer Biology training grant

CA09503-23. K.J. is a fellow of the American Cancer Society (PF-10-236-

01-RMC). U.B. is a Special Fellow of the Leukemia and Lymphoma Society

of America and the recipient of a New Investigator Award from the Leukemia

Research Foundation. F.W.A. is an Investigator of the Howard Hughes

Medical Institute.

Received: September 30, 2010

Revised: December 27, 2010

Accepted: December 31, 2010

Published online: January 20, 2011

REFERENCES

Adelman, K., Marr, M.T., Werner, J., Saunders, A., Ni, Z., Andrulis, E.D., and

Lis, J.T. (2005). Efficient release from promoter-proximal stall sites requires

transcript cleavage factor TFIIS. Mol. Cell 17, 103–112.

Anderson, J.R., Mukherjee, D., Muthukumaraswamy, K., Moraes, K.C.,Wilusz,

C.J., and Wilusz, J. (2006). Sequence-specific RNA binding mediated by the

RNase PH domain of components of the exosome. RNA 12, 1810–1816.

Andrulis, E.D., Werner, J., Nazarian, A., Erdjument-Bromage, H., Tempst, P.,

and Lis, J.T. (2002). The RNA processing exosome is linked to elongating

RNA polymerase II in Drosophila. Nature 420, 837–841.

Basu, U., Chaudhuri, J., Alpert, C., Dutt, S., Ranganath, S., Li, G., Schrum, J.P.,

Manis, J.P., and Alt, F.W. (2005). The AID antibody diversification enzyme is

regulated by protein kinase A phosphorylation. Nature 438, 508–511.

Basu, U., Wang, Y., and Alt, F.W. (2008). Evolution of phosphorylation-

dependent regulation of activation-induced cytidine deaminase. Mol. Cell

32, 285–291.

Cell 144, 353–363, February 4, 2011 ª2011 Elsevier Inc. 361

Besmer, E., Market, E., and Papavasiliou, F.N. (2006). The transcription

elongation complex directs activation-induced cytidine deaminase-mediated

DNA deamination. Mol. Cell. Biol. 26, 4378–4385.

Callahan, K.P., and Butler, J.S. (2010). TRAMP complex enhances RNA

degradation by the nuclear exosome component Rrp6. J. Biol. Chem. 285,

3540–3547.

Chaudhuri, J., Basu, U., Zarrin, A., Yan, C., Franco, S., Perlot, T., Vuong, B.,

Wang, J., Phan, R.T., Datta, A., et al. (2007). Evolution of the immunoglob-

ulin heavy chain class switch recombination mechanism. Adv. Immunol. 94,

157–214.

Chaudhuri, J., Khuong, C., and Alt, F.W. (2004). Replication protein A interacts

with AID to promote deamination of somatic hypermutation targets. Nature

430, 992–998.

Chaudhuri, J., Tian, M., Khuong, C., Chua, K., Pinaud, E., and Alt, F.W. (2003).

Transcription-targeted DNA deamination by the AID antibody diversification

enzyme. Nature 422, 726–730.

Chelico, L., Pham, P., Petruska, J., and Goodman, M.F. (2009). Biochemical

basis of immunological and retroviral responses to DNA-targeted cytosine

deamination by activation-induced cytidine deaminase and APOBEC3G. J.

Biol. Chem. 284, 27761–27765.

Cheng, H.L., Vuong, B.Q., Basu, U., Franklin, A., Schwer, B., Astarita, J., Phan,

R.T., Datta, A., Alt, F.W., andChaudhuri, J. (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, 2717–2722.

Daniels, G.A., and Lieber, M.R. (1995). RNA:DNA complex formation upon

transcription of immunoglobulin switch regions: implications for the mecha-

nism and regulation of class switch recombination. Nucleic Acids Res. 23,

5006–5011.

Di Noia, J.M., and Neuberger, M.S. (2007). Molecular mechanisms of antibody

somatic hypermutation. Annu. Rev. Biochem. 76, 1–22.

Dickerson, S.K., Market, E., Besmer, E., and Papavasiliou, F.N. (2003). AID

mediates hypermutation by deaminating single stranded DNA. J. Exp. Med.

197, 1291–1296.

Dzantiev, L., Constantin, N., Genschel, J., Iyer, R.R., Burgers, P.M., and Mod-

rich, P. (2004). A defined human system that supports bidirectional mismatch-

provoked excision. Mol. Cell 15, 31–41.

El Hage, A., French, S.L., Beyer, A.L., and Tollervey, D. (2010). Loss of Topoi-

somerase I leads to R-loop-mediated transcriptional blocks during ribosomal

RNA synthesis. Genes Dev. 24, 1546–1558.

Genschel, J., and Modrich, P. (2003). Mechanism of 50-directed excision in

human mismatch repair. Mol. Cell 12, 1077–1086.

Gomez-Gonzalez, B., and Aguilera, A. (2007). Activation-induced cytidine

deaminase action is strongly stimulated by mutations of the THO complex.

Proc. Natl. Acad. Sci. USA 104, 8409–8414.

Greimann, J.C., and Lima, C.D. (2008). Reconstitution of RNA exosomes from

human and Saccharomyces cerevisiae cloning, expression, purification, and

activity assays. Methods Enzymol. 448, 185–210.

Honjo, T., Kinoshita, K., and Muramatsu, M. (2002). Molecular mechanism of

class switch recombination: linkage with somatic hypermutation. Annu. Rev.

Immunol. 20, 165–196.

Houseley, J., LaCava, J., and Tollervey, D. (2006). RNA-quality control by the

exosome. Nat. Rev. Mol. Cell Biol. 7, 529–539.

Houseley, J., and Tollervey, D. (2008). The nuclear RNA surveillance

machinery: the link between ncRNAs and genome structure in budding yeast?

Biochim. Biophys. Acta 1779, 239–246.

Huang, F.T., Yu, K., Balter, B.B., Selsing, E., Oruc, Z., Khamlichi, A.A., Hsieh,

C.L., and Lieber, M.R. (2007). Sequence dependence of chromosomal R-loops

at the immunoglobulin heavy-chain Smu class switch region. Mol. Cell. Biol.

27, 5921–5932.

Huang, F.T., Yu, K., Hsieh, C.L., and Lieber, M.R. (2006). Downstream

boundary of chromosomal R-loops at murine switch regions: implications for

the mechanism of class switch recombination. Proc. Natl. Acad. Sci. USA

103, 5030–5035.

362 Cell 144, 353–363, February 4, 2011 ª2011 Elsevier Inc.

Jensen, T.H., and Moore, C. (2005). Reviving the exosome. Cell 121, 660–662.

LaCava, J., Houseley, J., Saveanu, C., Petfalski, E., Thompson, E., Jacquier,

A., and Tollervey, D. (2005). RNA degradation by the exosome is promoted

by a nuclear polyadenylation complex. Cell 121, 713–724.

Lebreton, A., Tomecki, R., Dziembowski, A., and Seraphin, B. (2008). Endonu-

cleolytic RNA cleavage by a eukaryotic exosome. Nature 456, 993–996.

Lennon, G.G., and Perry, R.P. (1985). C mu-containing transcripts initiate

heterogeneously within the IgH enhancer region and contain a novel 50-non-translatable exon. Nature 318, 475–478.

Li, X., and Manley, J.L. (2005). Inactivation of the SR protein splicing factor

ASF/SF2 results in genomic instability. Cell 122, 365–378.

Lieber, M.R. (2010). The mechanism of double-strand DNA break repair by the

nonhomologous DNA end-joining pathway. Annu. Rev. Biochem. 79, 181–211.

Liu, M., and Schatz, D.G. (2009). Balancing AID and DNA repair during somatic

hypermutation. Trends Immunol. 30, 173–181.

Liu, Q., Greimann, J.C., and Lima, C.D. (2006). Reconstitution, activities, and

structure of the eukaryotic RNA exosome. Cell 127, 1223–1237.

Longerich, S., Basu, U., Alt, F., and Storb, U. (2006). AID in somatic hypermu-

tation and class switch recombination. Curr. Opin. Immunol. 18, 164–174.

Lykke-Andersen, S., Brodersen, D.E., and Jensen, T.H. (2009). Origins and

activities of the eukaryotic exosome. J. Cell Sci. 122, 1487–1494.

Lutzker, S., and Alt, F.W. (1988). Structure and expression of germ line immu-

noglobulin gamma 2b transcripts. Mol. Cell. Biol. 8, 1849–1852.

Maldonado, E., Shiekhattar, R., Sheldon, M., Cho, H., Drapkin, R., Rickert, P.,

Lees, E., Anderson, C.W., Linn, S., and Reinberg, D. (1996). A human RNA

polymerase II complex associated with SRB and DNA-repair proteins. Nature

381, 86–89.

Maul, R.W., and Gearhart, P.J. (2010). AID and somatic hypermutation. Adv.

Immunol. 105, 159–191.

McBride, K.M., Gazumyan, A., Woo, E.M., Barreto, V.M., Robbiani, D.F., Chait,

B.T., and Nussenzweig, M.C. (2006). Regulation of hypermutation by activa-

tion-induced cytidine deaminase phosphorylation. Proc. Natl. Acad. Sci.

USA 103, 8798–8803.

McBride, K.M., Gazumyan, A., Woo, E.M., Schwickert, T.A., Chait, B.T., and

Nussenzweig, M.C. (2008). Regulation of class switch recombination and

somatic mutation by AID phosphorylation. J. Exp. Med. 205, 2585–2594.

Milstein, C., Neuberger, M.S., and Staden, R. (1998). Both DNA strands of

antibody genes are hypermutation targets. Proc. Natl. Acad. Sci. USA 95,

8791–8794.

Muramatsu, M., Kinoshita, K., Fagarasan, S., Yamada, S., Shinkai, Y., and

Honjo, T. (2000). Class switch recombination and hypermutation require acti-

vation-induced cytidine deaminase (AID), a potential RNA editing enzyme. Cell

102, 553–563.

Nambu, Y., Sugai, M., Gonda, H., Lee, C.G., Katakai, T., Agata, Y., Yokota, Y.,

and Shimizu, A. (2003). Transcription-coupled events associating with immu-

noglobulin switch region chromatin. Science 302, 2137–2140.

Neuberger, M.S., Harris, R.S., Di Noia, J., and Petersen-Mahrt, S.K. (2003).

Immunity through DNA deamination. Trends Biochem. Sci. 28, 305–312.

Oddone, A., Lorentzen, E., Basquin, J., Gasch, A., Rybin, V., Conti, E., and Sat-

tler, M. (2007). Structural and biochemical characterization of the yeast exo-

some component Rrp40. EMBO Rep. 8, 63–69.

Odegard, V.H., and Schatz, D.G. (2006). Targeting of somatic hypermutation.

Nat. Rev. Immunol. 6, 573–583.

Pasqualucci, L., Kitaura, Y., Gu, H., and Dalla-Favera, R. (2006). PKA-medi-

ated phosphorylation regulates the function of activation-induced deaminase

(AID) in B cells. Proc. Natl. Acad. Sci. USA 103, 395–400.

Pavri, R., Gazumyan, A., Jankovic, M., Di Virgolio, M., Kein, I., Ansarah-

Sobrinho, C., Resch, W., Yamane, A., Reina San-Martin, B., Barreto, V.,

et al. (2010). Activation-induced cytidine deaminase targets DNA at sites of

RNA polymerase II stalling by interaction with Spt5. Cell 143, 122–133.

Peled, J.U., Kuang, F.L., Iglesias-Ussel, M.D., Roa, S., Kalis, S.L., Goodman,

M.F., and Scharff, M.D. (2008). The biochemistry of somatic hypermutation.

Annu. Rev. Immunol. 26, 481–511.

Peters, A., and Storb, U. (1996). Somatic hypermutation of immunoglobulin

genes is linked to transcription initiation. Immunity 4, 57–65.

Petersen, S., Casellas, R., Reina-San-Martin, B., Chen, H.T., Difilippantonio,

M.J., Wilson, P.C., Hanitsch, L., Celeste, A., Muramatsu, M., Pilch, D.R.,

et al. (2001). AID is required to initiate Nbs1/gamma-H2AX focus formation

and mutations at sites of class switching. Nature 414, 660–665.

Petersen-Mahrt, S.K., Harris, R.S., and Neuberger, M.S. (2002). AID mutates

E. coli suggesting a DNA deamination mechanism for antibody diversification.

Nature 418, 99–103.

Preker, P., Nielsen, J., Kammler, S., Lykke-Andersen, S., Christensen, M.S.,

Mapendano, C.K., Schierup, M.H., and Jensen, T.H. (2008). RNA exosome

depletion reveals transcription upstream of active human promoters. Science

322, 1851–1854.

Rajagopal, D., Maul, R.W., Ghosh, A., Chakraborty, T., Khamlichi, A.A., Sen,

R., and Gearhart, P.J. (2009). Immunoglobulin switch mu sequence causes

RNA polymerase II accumulation and reduces dA hypermutation. J. Exp.

Med. 206, 1237–1244.

Ramiro, A.R., Jankovic, M., Eisenreich, T., Difilippantonio, S., Chen-Kiang, S.,

Muramatsu, M., Honjo, T., Nussenzweig, A., and Nussenzweig, M.C. (2004).

AID is required for c-myc/IgH chromosome translocations in vivo. Cell 118,

431–438.

Ramiro, A.R., Stavropoulos, P., Jankovic, M., and Nussenzweig, M.C. (2003).

Transcription enhances AID-mediated cytidine deamination by exposing

single-stranded DNA on the nontemplate strand. Nat. Immunol. 4, 452–456.

Revy, P., Muto, T., Levy, Y., Geissmann, F., Plebani, A., Sanal, O., Catalan, N.,

Forveille, M., Dufourcq-Labelouse, R., Gennery, A., et al. (2000). Activation-

induced cytidine deaminase (AID) deficiency causes the autosomal recessive

form of the Hyper-IgM syndrome (HIGM2). Cell 102, 565–575.

Robbiani, D.F., Bothmer, A., Callen, E., Reina-San-Martin, B., Dorsett, Y., Di-

filippantonio, S., Bolland, D.J., Chen, H.T., Corcoran, A.E., Nussenzweig, A.,

and Nussenzweig, M.C. (2008). AID is required for the chromosomal breaks

in c-myc that lead to c-myc/IgH translocations. Cell 135, 1028–1038.

Roy, D., Yu, K., and Lieber, M.R. (2008). Mechanism of R-loop formation at

immunoglobulin class switch sequences. Mol. Cell. Biol. 28, 50–60.

Schaeffer, D., Tsanova, B., Barbas, A., Reis, F.P., Dastidar, E.G., Sanchez-

Rotunno, M., Arraiano, C.M., and van Hoof, A. (2009). The exosome contains

domains with specific endoribonuclease, exoribonuclease and cytoplasmic

mRNA decay activities. Nat. Struct. Mol. Biol. 16, 56–62.

Schmid, M., and Jensen, T.H. (2008). The exosome: a multipurpose RNA-

decay machine. Trends Biochem. Sci. 33, 501–510.

Schrader, C.E., Linehan, E.K., Mochegova, S.N., Woodland, R.T., and Stav-

nezer, J. (2005). Inducible DNA breaks in Ig S regions are dependent on AID

and UNG. J. Exp. Med. 202, 561–568.

Shen, H.M., Ratnam, S., and Storb, U. (2005). Targeting of the activation-

induced cytosine deaminase is strongly influenced by the sequence and struc-

ture of the targeted DNA. Mol. Cell. Biol. 25, 10815–10821.

Shen, H.M., and Storb, U. (2004). Activation-induced cytidine deaminase (AID)

can target both DNA strands when the DNA is supercoiled. Proc. Natl. Acad.

Sci. USA 101, 12997–13002.

Shen, H.M., Tanaka, A., Bozek, G., Nicolae, D., and Storb, U. (2006). Somatic

hypermutation and class switch recombination in Msh6(-/-)Ung(-/-) double-

knockout mice. J. Immunol. 177, 5386–5392.

Shen, V., and Kiledjian, M. (2006). A view to a kill: structure of the RNA

exosome. Cell 127, 1093–1095.

Shinkura, R., Tian, M., Smith, M., Chua, K., Fujiwara, Y., and Alt, F.W. (2003).

The influence of transcriptional orientation on endogenous switch region func-

tion. Nat. Immunol. 4, 435–441.

Sohail, A., Klapacz, J., Samaranayake, M., Ullah, A., and Bhagwat, A.S. (2003).

Human activation-induced cytidine deaminase causes transcription-depen-

dent, strand-biased C to U deaminations. Nucleic Acids Res. 31, 2990–2994.

Staals, R.H., Bronkhorst, A.W., Schilders, G., Slomovic, S., Schuster, G.,

Heck, A.J., Raijmakers, R., and Pruijn, G.J. (2010). Dis3-like 1: a novel exoribo-

nuclease associated with the human exosome. EMBO J. 29, 2358–2367.

Stillman, B.W., and Gluzman, Y. (1985). Replication and supercoiling of simian

virus 40 DNA in cell extracts from human cells. Mol. Cell. Biol. 5, 2051–2060.

Tian, M., and Alt, F.W. (2000). Transcription-induced cleavage of immunoglob-

ulin switch regions by nucleotide excision repair nucleases in vitro. J. Biol.

Chem. 275, 24163–24172.

Tomecki, R., Kristiansen, M.S., Lykke-Andersen, S., Chlebowski, A., Larsen,

K.M., Szczesny, R.J., Drazkowska, K., Pastula, A., Andersen, J.S., Stepien,

P.P., et al. (2010). The human core exosome interacts with differentially local-

ized processive RNases: hDIS3 and hDIS3L. EMBO J. 29, 2342–2357.

Vuong, B.Q., Lee, M., Kabir, S., Irimia, C., Macchiarulo, S., McKnight, G.S.,

and Chaudhuri, J. (2009). Specific recruitment of protein kinase A to the immu-

noglobulin locus regulates class-switch recombination. Nat. Immunol. 10,

420–426.

Wang, L., Wuerffel, R., Feldman, S., Khamlichi, A.A., and Kenter, A.L. (2009). S

region sequence, RNA polymerase II, and histone modifications create chro-

matin accessibility during class switch recombination. J. Exp. Med. 206,

1817–1830.

Wuerffel, R.A., Du, J., Thompson, R.J., and Kenter, A.L. (1997). Ig Sgamma3

DNA-specifc double strand breaks are induced in mitogen-activated B cells

and are implicated in switch recombination. J. Immunol. 159, 4139–4144.

Xue, K., Rada, C., and Neuberger, M.S. (2006). The in vivo pattern of AID

targeting to immunoglobulin switch regions deduced from mutation spectra

in msh2-/- ung-/- mice. J. Exp. Med. 203, 2085–2094.

Yang, S.Y., and Schatz, D.G. (2007). Targeting of AID-mediated sequence

diversification by cis-acting determinants. Adv. Immunol. 94, 109–125.

Yu, K., Chedin, F., Hsieh, C.L.,Wilson, T.E., and Lieber, M.R. (2003). R-loops at

immunoglobulin class switch regions in the chromosomes of stimulated B

cells. Nat. Immunol. 4, 442–451.

Yu, K., and Lieber, M.R. (2003). Nucleic acid structures and enzymes in the

immunoglobulin class switch recombination mechanism. DNA Repair (Amst.)

2, 1163–1174.

Zarrin, A.A., Alt, F.W., Chaudhuri, J., Stokes, N., Kaushal, D., Du Pasquier, L.,

and Tian, M. (2004). An evolutionarily conserved target motif for immunoglob-

ulin class-switch recombination. Nat. Immunol. 5, 1275–1281.

Cell 144, 353–363, February 4, 2011 ª2011 Elsevier Inc. 363