Phosphorylation Directly Regulates the Intrinsic DNACytidine Deaminase Activity of Activation-inducedDeaminase and APOBEC3G Protein*□S

Received for publication, February 28, 2011, and in revised form, May 20, 2011 Published, JBC Papers in Press, June 9, 2011, DOI 10.1074/jbc.M111.235721

Zachary L. Demorest, Ming Li, and Reuben S. Harris1

From the Department of Biochemistry, Molecular Biology, and Biophysics, Institute for Molecular Virology, and Center for GenomeEngineering, University of Minnesota, Minneapolis, Minnesota 55455

The beneficial effects of DNA cytidine deamination by activa-tion-induced deaminase (AID; antibody gene diversification)and APOBEC3G (retrovirus restriction) are tempered by prob-able contributions to carcinogenesis.Multiple regulatorymech-anisms serve to minimize this detrimental outcome. Here, weshow that phosphorylation of a conserved threonine attenuatesthe intrinsic activity of activation-induced deaminase (Thr-27)and APOBEC3G (Thr-218). Phospho-null alanine mutantsmaintain intrinsic DNA deaminase activity, whereas phospho-mimetic glutamate mutants are inactive. The phospho-mimeticvariants fail to mediate isotype switching in activated mousesplenic B lymphocytes or suppress HIV-1 replication in humanT cells. Our data combine to suggest a model in which this crit-ical threonine acts as a phospho-switch that fine-tunes theadaptive and innate immune responses and helps protect mam-malian genomic DNA from procarcinogenic lesions.

Activation-induced deaminase (AID)2 and APOBEC3G(A3G) are the archetypal members of a larger family ofpolynucleotide cytidine-to-uridine deaminases with criticalfunctions in adaptive and innate immunity(1, 2). All mam-mals have AID, apolipoprotein B mRNA-editing catalyticsubunit 1 (APOBEC1), APOBEC2 and variable numbers ofAPOBEC3s (A3) ranging from one in rodents to seven inmost primates, including humans (A3A/B/C/D/F/G/H)(3).AID, A1, A3A, A3C, and A3H are single domain proteinswith one zinc-coordinating active site, whereas several A3s,including rodent A3 and human A3B, A3D, A3F, and A3G,are double domain proteins with two zinc-coordinatingmotifs (both are conserved, but typically only one is active).Atomic structures for the catalytic domain of humanA3G(4–7) and APOBEC2(8) are available, and these enablestructure-function studies and homology models.

Considerable effort has been devoted to understanding themultiple mechanisms that combine to regulate AID and A3Gactivity. First, the transcription of each of these genes is tissue-specific, withAID being expressed predominantly in B lympho-cytes and A3G inmost cell types (9–11). Second, AID and A3Gtranscription levels are up-regulated by distinct signal trans-duction pathways (STAT/NF�B for AID and NFAT/IRF forA3G) (12, 13). Third, AID expression is regulated by at least onemicroRNA, miR-155 (14, 15). Fourth, both proteins are pre-dominantly cytoplasmic, with AID having additional nuclearimport and export capabilities (16–20). Finally, both proteinsare subject to proteasome-dependent degradation, AID in thenuclear compartment (21) and A3G in the cytoplasmic com-partment (22–24).In addition, numerous proteins have been implicated in reg-

ulatingAID andA3G function (MDM2 (25), replication proteinA (26), heat shock protein 90 (HSP90) (27), germinal center-associated nuclear protein (GANP) (28), calcium and integrin-binding protein 1 (CIB1) (29), beta-catenin-like protein 1(CTNNBL1) (30)), with protein kinase A (PKA) (31–34) beingmost relevant to this work. PKA phosphorylates AID at threo-nine 27 and serine 38, with serine 38 phosphorylation promot-ing interactions with replication protein A and facilitatingclass-switch recombination (CSR) and somatic hypermutation(31–33, 35). Phosphorylation of the homologous residue inA3G, threonine 32, also has functional consequences by ren-dering the protein less susceptible to HIV-1 Vif-induced ubiq-uitylation and degradation(34). Here, we provide evidence insupport of a new role for threonine/serine phosphorylation indirectly suppressing the intrinsic DNA deaminase activity ofAID and A3G. Extensive conservation of this particular residuesuggests that phospho-regulation may extend to most otherDNA deaminase superfamily members.

EXPERIMENTAL PROCEDURES

DNA Constructs—pEGFP-N3-A3G and pEGFP-N3-AIDhave been described (36, 37). Mutants of AID and A3G weremade by QuikChange site-directed mutagenesis (Stratagene).The retroviral vector pMX-EGFP was constructed by subclon-ing EGFP fromplasmid pEGFP-N3 (Clontech) into pMX-PIE (agift from V. Barreto) using EcoRI/NotI. pMX-AID-IRES-EGFPwas generated by first amplifying untagged AID from pEGFP-N3-AID by PCR using primers 5�-GCT AGC GCC ACC ATGGAC AGC CT and 5�-CCT GCA GGT CAA AGT CCC AAAGTA. The insert was cut withNheI/SbfI and ligated into pCSII-

* This work was supported, in whole or in part, by National Institutes of HealthGrants R01-AI064046 and R01-GM080437 and National Institutes of HealthGrant T32-AI007313 (to Z. L. D.).

□S The on-line version of this article (available at http://www.jbc.org) containssupplemental Figs. 1–3.

1 To whom correspondence should be addressed: 321 Church Street SE.,6-155 Jackson Hall, Minneapolis, MN 55455. E-mail: rsh@umn.edu.

2 The abbreviations used are: AID, activation-induced deaminase; A3G, apoli-poprotein B mRNA-editing catalytic subunit 3G; APOBEC3G, apolipopro-tein B mRNA-editing catalytic subunit; PKA, protein kinase A; CSR, class-switch recombination; EGFP, enhanced green fluorescent protein; CaMKII,calcium calmodulin-dependent kinase II; ssDNA, single-stranded DNA; Vif,viral infectivity factor.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 286, NO. 30, pp. 26568 –26575, July 29, 2011© 2011 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

26568 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 286 • NUMBER 30 • JULY 29, 2011

by guest on February 3, 2020http://w

ww

.jbc.org/D

ownloaded from

IRES-EGFP (a gift fromN. Somia). The AID-IRES-EGFP insertwas then PCR-amplified using primers 5�-GAA TTC ATGGACAGCCTC TTGATGAAC and 5�-CCA CATAGCGTAAAA GGA GCA AC, cut using EcoRI/NotI, and ligated intopMX-PIE. The MLV amphotrophic envelope vector pRK5-10A1 and the HIV-1 accessory vector �NRF were generousgifts fromN. Somia. TheMLV accessory vector pMD-OGPwasprovided by F. Randow. The Vif-deficient HIV-1IIIB provirushas been used previously (38, 39). The Escherichia coli expres-sion constructs pTrc99a-AID-GST and pTrc99a-A3G-GSTwere generated by subcloning AID from pEGFP-N3-AID orA3G frompEGFP-N3-A3G into pTrc99a-GSTusingNcoI/SalI.The untagged bacterial expression plasmids pTrc99a-hAIDand pTrc99a-A3G have been described (37).In Vitro Peptide Kinase Assays—CaMKII (New England Bio-

labs) or PKA (New England Biolabs or gift from L. MastersonandG.Veglia, Ref. 40)were incubatedwith 1�g of either kemp-tide (CLRRASLG, American Peptide Co.), a peptide containingA3G-T218 (VRGRHETYLCYE, New England Peptides), or anA3G-T218A mutant peptide (VRGRHEAYLCYE, New Eng-land Peptides) in themanufacturer’s recommended buffer sup-plemented with [�-32P]-ATP. Kinase reactions were incubatedat 30 °C for 1 h before fractionating on a 16% Tricine/urea-acrylamide gel. The gel was dried and imaged by phosphor-imaging (FLA-5000, Fuji).E. coli Mutation Assays—The rifampicin resistance assay has

been published (41, 42). BW310 E. coli cells were transformedwith pTrc99a-AID or pTrc99a-A3G expression constructs andplated on ampicillin-containing media. Four individual colo-nies were picked and seeded into media containing 1 mM iso-propyl 1-thio-�-D-galactopyranoside and 100 �g/ml ampicil-lin. After shaking overnight at 37 °C, the cultureswere plated onampicillin media to obtain viable cell counts and to 100 �g/mlrifampicin-containing media for mutational frequency. AIDand A3G expression levels were determined by Western blotanalysis using antibodies against AID (EK25G9, Cell SignalingTechnology, Inc.) or A3G (#10201 rabbit anti-A3G polyclonalserum provided by J. Lingappa through the AIDS Research andReference Reagent Program).Recombinant Protein Preparations—Protein samples for

AID/A3G were prepared by growing 500-ml cultures of BL21E. coli cells transformed with pTrc99a-GST, pTrc99a-AID-GST, pTrc99a-A3G-GST, or mutants of AID/A3G. The cul-tures were centrifuged, and the cell pellet was resuspended inlysis buffer (50 mM Tris-Cl (pH 7.9), 200 mM NaCl, 50 �M

ZnCl2, complete protease inhibitors (Roche)). After centrifuga-tion (17,000 � g, 20 min), clarified lysates were incubated over-night at 4 °C with glutathione-Sepharose (GE Healthcare). Theresin was washed four times with lysis buffer, and purified pro-tein was eluted by cleavage with tobacco etch virus (TEV) pro-tease (Invitrogen).For preparation of A3G from human cells, pcDNA3.1-

A3G-myc-his or mutants thereof were transfected intoHEK293T cells cultured in DMEM (Invitrogen) and 10% FBS(Hyclone) using TransIT-LT1 reagent (Mirus). After 48 h, cellswere lysed in buffer (25 mM HEPES (pH 7.4), 150 mM NaCl,0.5% Triton X-100, 1 mM EDTA, 1 mM MgCl2 1 mM ZnCl, 10%glycerol) and bound to nickel-nitrilotriacetic acid-agarose

beads (Qiagen). The beads were washed, and purified proteinwas eluted using imidazole (250 �M), as described (43).DNA Binding Assays—Protein samples prepared from E. coli

as above were used for AID DNA binding reactions. 6 pmol ofpurified protein was diluted serially 1:2 and mixed with 0.5pmol 32P-labeled oligo (5�-ATT ATT ATT ATT CCA ATGGAT TTA TTT ATT TWR CTA TTT ATT T) in bindingbuffer (10 mM HEPES (pH 7.6), 10% glycerol, 100 mM KCl, 10mMMgCl2, 100 �M EDTA, 500 �MDTT). Reactions were incu-bated for 30min at 37 °C before separation on a 7%Tris borate-EDTA acrylamide gel. The gel was dried and imaged by phos-phoimager (Storm, Molecular Dynamics).Protein samples purified from human cells from the above

were used for A3GDNA binding reactions. 25 pmol of purifiedprotein was diluted serially 1:2, and each dilution was mixedwith 1 pmol fluorescein-labeled oligo (5�-ATT ATTATTATTCCA ATG GAT TTA TTT ATT TAT TTA TTT ATT T-fluo-rescein) in binding buffer. The reactions were incubated for 30min at 37 °C before separation on a 7%Tris borate-EDTAacryl-amide gel. The free and bound oligos were then detected byfluorescence imaging (FLA-5000, Fuji).Oligo-based Deaminase Assays—Protein samples purified

from human cells as above were used for A3G deaminase activ-ity reactions. Starting with 1.2 pmol purified protein, 2-foldserial dilutions were made and mixed with 1 pmol of substrateoligo (5�-ATTATTATTATTCCAATGGATTTATTTATTTAT TTA TTT ATT T-fluorescein), 0.1 �g/�l RNase A (Qia-gen), and 0.001 units/�l uracil DNA glycosylase (NEB). Thereactions were incubated at 37 °C for 2 h, and then NaOH wasadded to 100 �M before incubating at 90 °C for another 30 min.The reactions were separated on a 16% Tris/urea-acrylamidegel and visualized by fluorescence imaging (FLA-5000, Fuji).Fluorescence Microscopy Studies—For steady-state A3G and

AID localization, pEGFP-N3-A3G, pEGFP-N3-AID, or mu-tants thereof were transfected into HeLa cells grown in DMEM(Invitrogen) supplemented with 10% FBS (Hyclone). 48 h later,the cellular localization was determined by fluorescencemicroscopy (Deltavision). For AID import activity, leptomy-cin-B (20 ng/ml) was added 2 h prior to imaging as above.HIV Restriction Assays—CEM-SS and CEM-GFP (courtesy

of M. Malim) were maintained in RPMI 1640 (Invitrogen) sup-plemented with 10% FBS (Hyclone). Stable cell lines expressingpEGFP-N3, pEGFP-N3-A3G, ormutants were generated in thepermissive cell line CEM-SS by electroporating linearizedDNAand selecting for stable integrants with 1mg/ml G418 (Mediat-ech) as described (38). Clones were confirmed to have similarexpression levels by Western blot analysis using an antibodyagainst A3G (rabbit polyclonal raised against a C-terminal pep-tide). Virus was produced by transfecting HIV-1 provirus usingTransIT-LT1 (Mirus) HEK293T cells maintained in DMEM(Invitrogen) supplemented with 10% FBS (Hyclone). 48 h aftertransfection, virus-containing supernatants were filtered with a0.45-�m filter. Viruses were then titered using the CEM-GFPreporter cell line as described (38). Spreading infections wereinitiated by adding virus to CEM-SS stable cell lines at a multi-plicity of infection of 0.05. Supernatants from infected cultureswere collected at 2- to 4-day intervals and added to CEM-GFP.After 48-hours, the cells were fixed in 4% paraformaldehyde

AID and A3G Phosphorylation

JULY 29, 2011 • VOLUME 286 • NUMBER 30 JOURNAL OF BIOLOGICAL CHEMISTRY 26569

by guest on February 3, 2020http://w

ww

.jbc.org/D

ownloaded from

and analyzed forGFP expression by flow cytometry (Quanta SCMPL, Beckman Coulter). Procedures for the detection of A3Gin producer cells and viral particles have been described (44).Class-switch Recombination Assays—All experiments were

conducted in accordance with the University of MinnesotaAnimal Care and Use Committee guidelines. The C57BL/6AID�/�mice have been described (45).Ex vivoCSR assayswereconducted by purifying resting B-cells from spleen bymagneticsorting (130-090-862, Miltenyi Biotec). Isolated B-cells werethen cultured in RPMI supplemented with 10% FBS, 50 ng/mlIL-4, and 50 �g/ml LPS. After 48 h, the media were replacedwith transducing viral supernatant supplemented with 20 mM

HEPES and 16 �g/ml polybrene and centrifuged (600 � g, 2 h,30 °C). The cells were then resuspended into fresh media con-taining IL-4 and LPS and cultured for 4 days. Efficiency ofswitching to IgG1 was determined by staining with anti-IgG1-PE (BD Biosciences) and analyzed by flow cytometry(FACSCanto II, BD Biosciences).

RESULTS

AID-Thr-27, A3G-Thr-32, and A3G-Thr-218 Are Homolo-gous and Located within a Region of High Sequence and Struc-tural Conservation—Prior studies demonstrated phosphoryla-tion of AID-Thr-27 in vivo and in vitro by mass spectrometryand radiolabeling(31, 33, 35) and A3G-Thr-32 by immunoblot-

ting (34). We noted that these two threonines are homolo-gous to A3G-Thr-218, whose high-resolution structures haveshown to be located within a solvent-accessible loop (4–7) (Fig.1A). This threonine anchors a conserved motif that matches aconsensus PKA phosphorylation site (R-H/R-X-T) (46) (Fig.1B). Notably, nearly all AID/A3 family members have homolo-gous threonine or serine residues at this exact position (Fig.1C). Rare exceptions are only apparent in specific mammalianlineages (carnivores and rodents) or in redundant or inactivedomains (most alleles of human A3H are unstable) (47). In thecatalytic domain of A3G, the first arginine in this motif (Arg-215) is located adjacent to the catalytic glutamate, and it hasbeen implicated in binding substrate single-stranded DNA(5–7). Taken together, these observations, and particularly thehigh level of conservation and the structural positioning next tothe active site, suggest that phosphorylation and dephosphory-lationmay serve as a posttranslational switch that helps controlthe DNA deaminase activity of these mutagenic enzymes.PKA and CaMKII Phosphorylate A3G-Thr-218 in Vitro—

AID-Thr-27 and A3G-Thr-32 can be phosphorylated by PKA(31, 33–35). To determine whether these observations extendto A3G-Thr-218, we asked whether recombinant PKA couldphosphorylate a peptide representing the soluble loop in whichthis residue resides, VRGRHET218YLCYE.We found that PKA

FIGURE 1. An active-site Thr/Ser is conserved in mammalian DNA cytidine deaminases. A, Thr-218 is positioned adjacent to the catalytic glutamate Glu-259in the crystal structure of A3G (3IR2) (4). The residues comprising the kinase recognition sequence and the side chains of the two cysteines that coordinate zincare also shown. B, an alignment of the PKA consensus sequence, R-X-X-T, found in human AID, mouse AID, human A3G C-terminal domain (CTD), and humanA3G N-terminal domain (NTD). C, a phylogenetic analysis showing the conservation of active site Thr/Ser residues in the AID/A3 proteins of nearly all mammals.Domains highlighted in orange represent the presence of a threonine, and domains highlighted in cyan contain a serine. Hs, human; Mm, mouse; Rn, rat; Bt, cow,Oa, sheep; Ss, pig; Tt, peccary; Ec, horse; Cf, dog; Fc, cat. D, PKA and CaMKII can phosphorylate A3G-Thr-218 in vitro.

AID and A3G Phosphorylation

26570 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 286 • NUMBER 30 • JULY 29, 2011

by guest on February 3, 2020http://w

ww

.jbc.org/D

ownloaded from

could readily phosphorylate this peptide but not a T218Amutant derivative that is otherwise identical (Fig. 1D). Simi-larly, CaMKII, which also phosphorylates R-X-X-Tmotifs (46),was able to phosphorylate the A3G-Thr-218 peptide but notthe alanine mutant derivative. Both enzymes were also able tophosphorylate a serine in a control peptide (Kemptide). Thesedata demonstrate that A3G-Thr-218 is a suitable substrate forat least two kinases, PKA and CaMKII.Phospho-mimetic Mutations Inhibit DNA Cytidine Deami-

nase Activity—To address whether phosphorylation is capableof attenuating the DNA cytidine deaminase activity of AID andA3G, phospho-null and phospho-mimetic derivatives of theseproteins were tested in an E. coli-based activity assay. Therifampicin-resistance (RifR) mutation assay has been usedextensively to assess intrinsic DNA cytidine deaminase activity(41, 42). Consistent with prior reports, AID and A3G triggered3- and 4-fold increases in the median RifR mutation frequencycompared with catalytically inactive controls, AID-E58Q andA3G-E259Q (Fig. 2, A and B). In comparison, phospho-mi-metic AID-T27E and A3G-T218E proteins also showed greatlyreduced activity approaching background levels. Phospho-nullalanine mutants showed slightly higher levels of mutator activ-ity. Mutation of another predicted surface threonine in AID(Thr-140) or the homologous threonine in the non-catalyticN-terminal domain of A3G (Thr-32) had little effect. All pro-teins expressed similarly in E. coli, indicating that these data arenot due to variable protein expression levels (lower panels inFig. 2, A and B).To ask whether these observations extended to A3G purified

fromhuman cells, we used aDNAoligonucleotide deaminationassay optimized to measure catalytic activity. As expected,wild-type A3G catalyzes dose-dependent cytidine-to-uridinedeamination of labeled deoxy-oligonucleotide substrates,which following uracil excision and NaOH-mediated phos-phodiester backbone cleavage, is detected as a shorter DNAfragment (Fig. 3, A, C, and E). As anticipated from the E. colimutation experiments, the A3G phospho-mimetic variantT218E showed considerably lower levels of catalytic activity.Interestingly, the A3G phospho-null variant T218A showed

significantly elevated levels of catalytic activity consistent witha proportion of the wild-type protein being already phos-phorylated (and thereby inactivated) in HEK293T cells. Takentogether, the E. coli and the purified protein activity data indi-cate that phosphorylation of the conserved threonine, AID-Thr-27 or A3G-Thr-218, may serve to attenuate the intrinsicDNAdeaminase activity of these proteins (supported further byHEK293T cell extract data in supplemental Fig. S1).DNA Binding Is Unaffected by Phospho-mimetic Substi-

tutions—To askwhether the diminished catalytic activity of thephospho-mimetic substitution mutants is due to defectivessDNA binding, we tested the ssDNA binding ability of AIDand A3G in electrophoretic mobility shift assays. A3G-myc-hisused in the deaminase reactions above was used for ssDNAbinding experiments. Purified protein was diluted serially,incubated with a fluorescently labeled oligo, and fractionatedon a native polyacrylamide gel. As expected, A3G and the cat-alytic mutant E259Q bound ssDNA in a dose-dependent man-ner (Fig. 3, B, D, and E) (48). Likewise, A3G-T218A and A3G-T218E had nearly identical ssDNA binding abilities, which wereindistinguishable from the wild-type enzyme (Fig. 3, B,D, and E).

Similar EMSA experiments were done with wild-type AIDandmutant derivatives, but the sensitivity of the assay had to beincreased by using a radiolabeled ssDNAoligo substrate. Again,thewild-type and the phospho-null and phospho-mimetic vari-ants produced near identicalmobility shifts (Fig. 3, F,G, andH).As a control to demonstrate the specificity of AID for ssDNA,an AID-R24E mutant was analyzed in parallel and shown to bedefective in DNA binding. This arginine is conserved andhomologous to A3G-Arg-215, which NMR chemical shift per-turbation and mutagenesis experiments have implicatedstrongly in DNA binding (5–7). Additional EMSA data can befound in supplemental Fig. S2. Overall, these EMSA resultsclearly show that phospho-mimetic substitutions in A3G andAID do not cause visible decreases in the ability of each proteinto bind ssDNA.Mutants of AID and A3G Localize Normally within Living

Cells—The subcellular localization of AID/APOBEC familymembers has beenwell studied (16–20). A3G is predominantlycytoplasmic. AID is also mostly cytoplasmic, but it is importedinto the nuclear compartment by an importin-� pathway andexported back to the cytoplasm by a CRM1 pathway. To askwhether our phospho-null or phospho-mimetic mutants retainnormal, steady-state subcellular distributions, we performed aseries ofAID/A3G-GFP localization studies in livingHeLa cells.No detectable alteration in the steady-state cytoplasmic distri-bution of A3G-EGFP, AID-EGFP, or their mutant derivativeswas detected (Fig. 4, A and B). Moreover, experiments done inthe presence and absence of the CRM1 inhibitor leptomycin Bindicated that the nuclear import and export activitieswere alsointact for all AID-EGFP constructs. These data therefore indi-cate that A3G, AID, and their mutant derivatives are capable ofinteracting with the cellular factors responsible for localizationand, furthermore, that AID is able to enter the nucleus, where itwill have the opportunity to access the immunoglobin locus, itsphysiologic DNA deamination target.AID-T27E Is Defective for Class-switch Recombination—One

of the physiological functions of AID is catalyzing cytidine-to-

FIGURE 2. Intrinsic DNA cytidine deaminase activity of AID and A3G con-structs. A and B, results from E. coli-based RifR mutation assays, with each Xrepresenting data from an independent culture. Median mutation frequen-cies are indicated by horizontal bars and numbers. Also shown are Westernblot analyses of AID or A3G constructs from representative cultures with anonspecific band (NSB) as a loading control.

AID and A3G Phosphorylation

JULY 29, 2011 • VOLUME 286 • NUMBER 30 JOURNAL OF BIOLOGICAL CHEMISTRY 26571

by guest on February 3, 2020http://w

ww

.jbc.org/D

ownloaded from

uridine deamination events in immunoglobulin heavy chaingene switch region DNA and thereby triggering additionalDNA repair processes that ultimately manifest as antibody iso-type switch recombination (1, 45). Therefore, as a functionaltest of AID activity, we assayed the phospho-null and phospho-mimetic mutants in an ex vivo B-cell CSR system (30–33).Naïve splenic B-lymphocytes were isolated from AID-deficientmice; cultured in the presence of IL-4 and LPS to induce cell

division and isotype switching from IgM to IgG1; transducedwith retroviruses encoding AID-IRES-EGFP, mutants of AID,or EGFP alone; and 4 days later subjected to flow cytometry forIgG1-positve cells. Mock- (not shown) or EGFP-virus trans-duced cells remained AID-defective and showed no classswitching to IgG1 (Fig. 5). Also, as expected, wild-type AIDexpression complemented the endogenous defect and enabledclass switching to IgG1 in a significant proportion of cells (rep-

FIGURE 3. A3G-T218E has diminished deaminase activity but retains DNA binding ability. A, representative images from titrated A3G oligo deaminaseassays. The upper band is the intact oligo, and the lower band is the product of deamination, uracil excision, and strand cleavage. B, representative images fromA3G ssDNA binding assays. Free oligo and protein-bound complexes are labeled. C, quantification of A3G deaminase activity in A and replicas not shown. Dataare plotted as the mean � S.D. of three independent experiments. D, quantification of A3G EMSA data in B. Data are plotted as the mean � S.D. of threeindependent experiments. E, Coomassie-stained gel illustrating the purity of A3G enzymes used in these experiments. F, representative images from AIDssDNA binding assays. G, quantification of AID EMSA data in F. Data are plotted as the mean � S.D. of three independent experiments. H, Coomassie-stainedgel illustrating the purity of AID enzymes used in the ssDNA binding assays. The identity of the AID bands was confirmed by immunoblotting (not shown).

AID and A3G Phosphorylation

26572 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 286 • NUMBER 30 • JULY 29, 2011

by guest on February 3, 2020http://w

ww

.jbc.org/D

ownloaded from

resentative plots in Fig. 5A and average of four experiments inB). Conversely and surprisingly, neither T27A nor T27E wascapable of promoting the switch to IgG1 despite similar proteinexpression levels (Fig. 5, A, B, and C).

The T27E result was anticipated on the basis of the lowerlevel of catalytic activity, but not DNA binding or localizationactivities, elicited by thismutant.However, the aforementioneddata on AID-T27A showing normal deaminase, ssDNA bind-ing, and cellular localization/trafficking activities strongly sug-gested that this variant would be capable of normal or evenelevated CSR levels, in stark contrast to the defect in CSRshownhere. This resultmakes theCSRdata setmore difficult tointerpret. One possibility, noted previously (31), is that phos-phorylation of Ser-38 may depend first upon phosphorylationof Thr-27. An alternativemay be that each of these residues hasa distinct mechanistic contribution to CSR, with our studiesfavoring a role for Thr-27 in regulating catalysis.A3G-T218E Lacks HIV-1 Restriction Activity—A3G potently

inhibits HIV-1 replication by blocking reverse transcriptionand deaminating viral cDNA cytosines to uracils (2). This anti-viral activity is most evident in HIV-1 lacking viral infectivityfactor (Vif), a small basic protein that triggersA3Gdegradation.Thus, a rigorous test of the functional activity ofA3G iswhetherit suppresses the spreading infection of Vif-deficientHIV-1 (38,

39, 49). We therefore created a panel of CEM-SS T cell linesstably expressing wild-type A3G-EGFP, an E259Q catalyticmutant control, a phospho-null T218A construct, or a phos-pho-mimetic T218E protein. As anticipated from prior studies,wild-type A3G completely suppressed the replication of Vif-deficient HIV-1, and its strong antiviral effect was largelydependent upon the integrity of the catalytic glutamate Glu-259 (50, 51) (Fig. 6A). A3G-T218A showed wild-type levels ofrestriction consistent with full or elevated levels of enzymaticactivity. In contrast, A3G-T218E failed to prevent the replica-tion of Vif-deficient HIV-1. However, this mutant protein didcause reproducible delays in peak viral replication consistentwith severely attenuated but not fully defective catalytic activ-ity. As additional controls,N-terminalA3G-T32AorT32E sub-stitutions had no discernable effect, and all cell lines supportedsimilar levels of Vif-proficient HIV-1 spreading infection (Fig.6A and data not shown). It is notable that, although we wereable to confirm A3G-Thr-32 phosphorylation by mass spec-trometry, we found no differences in the subcellular localiza-tion,HIV restriction capacity, orVif susceptibility in alanine- orglutamate-substituted derivatives (Fig. 4 and data not shown).An additional possibility is that A3G-T218Emay not restrict

HIV-1 because it is not efficiently packaged into budding viri-

FIGURE 4. Mutants of A3G and AID localize normally within living cells.A, representative fluorescent images of A3G-EGFP localization in HeLa cells.B, representative images of AID-EGFP localization in HeLa cells in the pres-ence or absence of leptomycin-B (LepB).

FIGURE 5. AID-T27E is defective for class-switch recombination. A, repre-sentative flow cytometry plots of stimulated B lymphocytes transduced with theindicated human AID-IRES-EGFP constructs. Transduction is indicated by GFPexpression and CSR by IgG1 expression. B, histogram summarizing the CSR activ-ity from four independent experiments (mean and S.D. of the percentage of IgG1cells within the GFP-positive-transduced population). C, representative immuno-blot of AID expression with tubulin (TUB) as a loading control.

AID and A3G Phosphorylation

JULY 29, 2011 • VOLUME 286 • NUMBER 30 JOURNAL OF BIOLOGICAL CHEMISTRY 26573

by guest on February 3, 2020http://w

ww

.jbc.org/D

ownloaded from

ons. To test and eliminate such a possibility, we harvested virusproduced from HEK293T cells expressing A3G and mutantsthereof and blotted for the presence of A3G in these viral par-ticles.We found no significant difference in the ability of any ofthemutants to get into virions as comparedwithwild-typeA3G(Fig. 6C).

DISCUSSION

The AID/APOBEC family of cytidine deaminases is animportant facet of the adaptive and innate immune responsesin humans. However, their mutagenic activity must be tightlyregulated to prevent potentially detrimental off-target effects.Regulation of these proteins has been described at multiple lev-

els, including transcription, microRNAs, cytoplasmic localiza-tion, proteasomal degradation, and phosphorylation (see intro-duction). Here we describe a novel phosphorylation regulatorymechanism capable of attenuating the intrinsic deaminaseactivity of AID and A3G. In this study, we demonstrate thatphospho-mimetic substitution of a highly conserved threoninerenders these proteins inactive in several independent assays.We show that ssDNA binding ability and steady-state subcellu-lar localization (and for AID, also trafficking) are unaffected,indicating that these proteins are structurally intact. In func-tional assays, this modification prevents AID from facilitatingCSR and A3G from restricting HIV-1�Vif replication. It isintriguing that two neighboring phosphorylation sites can havesuch contrasting effects on the function of AID, with Ser-38phosphorylation enabling interaction with replication proteinA and allowing CSR and somatic hypermutation, and Thr-27phosphorylation rendering the protein inactive. This begs thequestion of how PKA is regulated to distinguish between theseneighboring residues. Further studies are warranted to betterunderstand these posttranslational regulatory events and inves-tigate the possible involvement of other Ser/Thr kinases thatcan also recognize PKA consensus motifs, such as CaMKIIdescribed here.The obvious utility of posttranslational regulation by phos-

phorylation is 2-fold (illustrated by the model in supplementalFig. S3). First, a threonine- or serine-phosphorylated DNAdeaminasewould possess a low level ofDNAdeaminase activityandpose less of a threat to genomicDNA.GenomicDNA integ-rity is further ensured by the fact that AID, A3G, and manyother A3 proteins are predominantly cytoplasmic. Second, sig-nal transduction pathways, which are critical for both adaptiveand innate immune responses, could readily switch on DNAdeaminase activity by triggering the removal of the phosphategroup (phosphatase or phosphotransferase activity). Thiswould ensure an expedited immune response that could be fur-ther bolstered by up-regulating AID or A3 expression at thetranscriptional and/or translational levels.We propose that the posttranslational modification of AID

and the A3 proteins by phosphorylation provides a means ofdirectly controlling the intrinsic DNA cytidine deaminaseactivity of these proteins (supplemental Fig. S3). It is likely thatthis mechanism will be conserved in vertebrates because resi-dues homologous to AID-Thr-27, A3G-Thr-32, or Thr-218 areapparent in almost all other known polynucleotide cytidinedeaminase family members (Fig. 1). It is further possible thatdefects in these signal transduction pathways may manifest asimmunodeficiency syndromes (overphosphorylated protein),autoimmune diseases (underphosphorylated protein), and/orcarcinogenesis (underphosphorylated protein), especially incombination with other regulatory defects.

Acknowledgments—We thank L. Potter and G. Veglia for helpful dis-cussions; L.Masterson andG. Veglia for recombinant PKA; J. JohnsonandN. Krogan for expert mass spectrometry;W. Brown, A. Davis, andK. Shindo for technical assistance; and V. Barreto, J. Lingappa(through the AIDS Research and Reference Reagents Program), F.Randow, and N. Somia for contributing reagents.

FIGURE 6. A3G-T218E fails to restrict Vif-deficient HIV-1. A, the kinetics ofVif-deficient HIV-1 replication in the indicated stable A3G-expressing T cell lines.A multiplicity of infection of 0.05 was used to initiate infection on day 0, and viralinfectivity was measured on subsequent days by titering cell-free supernatantson CEM-GFP indicator cells. B, representative immunoblot of A3G-EGFP (anti-GFP) expression with tubulin (TUB) as a loading control. C, Western blot analysesindicating the presence of A3G in producer cells and viral particles. TUB and p24are loading controls for cells and viral particles, respectively.

AID and A3G Phosphorylation

26574 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 286 • NUMBER 30 • JULY 29, 2011

by guest on February 3, 2020http://w

ww

.jbc.org/D

ownloaded from

REFERENCES1. Di Noia, J. M., and Neuberger, M. S. (2007) Annu. Rev. Biochem. 76, 1–222. Malim, M. H., and Emerman, M. (2008) Cell Host Microbe 3, 388–3983. Conticello, S. G. (2008) Genome Biol. 9, 2294. Shandilya, S.M., Nalam,M.N., Nalivaika, E. A., Gross, P. J., Valesano, J. C.,

Shindo, K., Li, M.,Munson,M., Royer,W. E., Harjes, E., Kono, T.,Matsuo,H., Harris, R. S., Somasundaran, M., and Schiffer, C. A. (2010) Structure18, 28–38

5. Chen, K. M., Harjes, E., Gross, P. J., Fahmy, A., Lu, Y., Shindo, K., Harris,R. S., and Matsuo, H. (2008) Nature 452, 116–119

6. Harjes, E., Gross, P. J., Chen, K.M., Lu, Y., Shindo, K., Nowarski, R., Gross,J. D., Kotler, M., Harris, R. S., and Matsuo, H. (2009) J. Mol. Biol. 389,819–832

7. Holden, L. G., Prochnow,C., Chang, Y. P., Bransteitter, R., Chelico, L., Sen,U., Stevens, R. C., Goodman, M. F., and Chen, X. S. (2008) Nature 456,121–124

8. Prochnow, C., Bransteitter, R., Klein, M. G., Goodman, M. F., and Chen,X. S. (2007) Nature 445, 447–451

9. Koning, F. A., Newman, E. N., Kim, E. Y., Kunstman, K. J.,Wolinsky, S.M.,and Malim, M. H. (2009) J. Virol. 83, 9474–9485

10. Muramatsu, M., Sankaranand, V. S., Anant, S., Sugai, M., Kinoshita, K.,Davidson, N. O., and Honjo, T. (1999) J. Biol. Chem. 274, 18470–18476

11. Refsland, E.W., Stenglein,M. D., Shindo, K., Albin, J. S., Brown,W. L., andHarris, R. S. (2010) Nucleic Acids Res. 38, 4274–4284

12. Dedeoglu, F., Horwitz, B., Chaudhuri, J., Alt, F. W., and Geha, R. S. (2004)Int. Immunol. 16, 395–404

13. Farrow,M.A., Kim, E. Y.,Wolinsky, S.M., and Sheehy, A.M. (2011) J. Biol.Chem. 286, 2567–2577

14. Dorsett, Y., McBride, K. M., Jankovic, M., Gazumyan, A., Thai, T. H.,Robbiani, D. F., Di Virgilio, M., Reina San-Martin, B., Heidkamp, G., Sch-wickert, T. A., Eisenreich, T., Rajewsky, K., andNussenzweig,M. C. (2008)Immunity 28, 630–638

15. Teng, G., Hakimpour, P., Landgraf, P., Rice, A., Tuschl, T., Casellas, R., andPapavasiliou, F. N. (2008) Immunity 28, 621–629

16. Ito, S., Nagaoka, H., Shinkura, R., Begum, N., Muramatsu, M., Nakata, M.,and Honjo, T. (2004) Proc. Natl. Acad. Sci. U.S.A. 101, 1975–1980

17. McBride, K. M., Barreto, V., Ramiro, A. R., Stavropoulos, P., and Nussen-zweig, M. C. (2004) J. Exp. Med. 199, 1235–1244

18. Patenaude, A. M., Orthwein, A., Hu, Y., Campo, V. A., Kavli, B., Buschi-azzo, A., and Di Noia, J. M. (2009) Nat. Struct. Mol. Biol. 16, 517–527

19. Stenglein, M. D., Matsuo, H., and Harris, R. S. (2008) J. Virol. 82,9591–9599

20. Bennett, R. P., Presnyak, V.,Wedekind, J. E., and Smith, H. C. (2008) J. Biol.Chem. 283, 7320–7327

21. Aoufouchi, S., Faili, A., Zober, C., D’Orlando, O., Weller, S., Weill, J. C.,and Reynaud, C. A. (2008) J. Exp. Med. 205, 1357–1368

22. Sheehy, A. M., Gaddis, N. C., and Malim, M. H. (2003) Nat. Med. 9,1404–1407

23. Stopak, K., de Noronha, C., Yonemoto,W., andGreene,W. C. (2003)Mol.Cell 12, 591–601

24. Conticello, S. G., Harris, R. S., and Neuberger, M. S. (2003) Curr. Biol. 13,2009–2013

25. MacDuff, D. A., Neuberger, M. S., and Harris, R. S. (2006)Mol. Immunol.

43, 1099–110826. Chaudhuri, J., Khuong, C., and Alt, F. W. (2004) Nature 430, 992–99827. Orthwein, A., Patenaude, A. M., Affar el, B., Lamarre, A., Young, J. C., and

Di Noia, J. M. (2010) J. Exp. Med. 207, 2751–276528. Maeda, K., Singh, S. K., Eda, K., Kitabatake,M., Pham, P., Goodman,M. F.,

and Sakaguchi, N. (2010) J. Biol. Chem. 285, 23945–2395329. Demorest, Z. L., MacDuff, D. A., Brown, W. L., Morham, S. G., Parise,

L. V., and Harris, R. S. (2010) PLoS ONE 5, e1166030. Conticello, S. G., Ganesh, K., Xue, K., Lu, M., Rada, C., and Neuberger,

M. S. (2008)Mol. Cell 31, 474–48431. Basu, U., Chaudhuri, J., Alpert, C., Dutt, S., Ranganath, S., Li, G., Schrum,

J. P., Manis, J. P., and Alt, F. W. (2005) Nature 438, 508–51132. McBride, K.M.,Gazumyan,A.,Woo, E.M., Barreto, V.M., Robbiani, D. F.,

Chait, B. T., and Nussenzweig, M. C. (2006) Proc. Natl. Acad. Sci. U.S.A.103, 8798–8803

33. Pasqualucci, L., Kitaura, Y., Gu, H., and Dalla-Favera, R. (2006) Proc. Natl.Acad. Sci. U.S.A. 103, 395–400

34. Shirakawa, K., Takaori-Kondo, A., Yokoyama, M., Izumi, T., Matsui, M.,Io, K., Sato, T., Sato, H., andUchiyama, T. (2008)Nat. Struct.Mol. Biol. 15,1184–1191

35. Pham, P., Smolka, M. B., Calabrese, P., Landolph, A., Zhang, K., Zhou, H.,and Goodman, M. F. (2008) J. Biol. Chem. 283, 17428–17439

36. Stenglein, M. D., and Harris, R. S. (2006) J. Biol. Chem. 281, 16837–1684137. MacDuff, D. A., Demorest, Z. L., andHarris, R. S. (2009)Nucleic Acids Res.

37, 1854–186738. Albin, J. S., Hache, G.,Hultquist, J. F., Brown,W. L., andHarris, R. S. (2010)

J. Virol. 84, 10209–1021939. Hache, G., Shindo, K., Albin, J. S., and Harris, R. S. (2008) Curr. Biol. 18,

819–82440. Masterson, L. R., Shi, L., Metcalfe, E., Gao, J., Taylor, S. S., and Veglia, G.

(2011) Proc. Natl. Acad. Sci. U.S.A. 108, 6969–697441. Harris, R. S., Petersen-Mahrt, S. K., and Neuberger, M. S. (2002)Mol. Cell

10, 1247–125342. Petersen-Mahrt, S. K., Harris, R. S., and Neuberger, M. S. (2002) Nature

418, 99–10343. Shlyakhtenko, L. S., Lushnikov, A. Y., Li, M., Lackey, L., Harris, R. S., and

Lyubchenko, Y. L. (2011) J. Biol. Chem. 286, 3387–339544. Larue, R. S., Lengyel, J., Jonsson, S. R., Andresdottir, V., and Harris, R. S.

(2010) J. Virol. 84, 8193–820145. Muramatsu, M., Kinoshita, K., Fagarasan, S., Yamada, S., Shinkai, Y., and

Honjo, T. (2000) Cell 102, 553–56346. Amanchy, R., Periaswamy, B., Mathivanan, S., Reddy, R., Tattikota, S. G.,

and Pandey, A. (2007) Nat. Biotechnol. 25, 285–28647. OhAinle, M., Kerns, J. A., Li, M.M.,Malik, H. S., and Emerman,M. (2008)

Cell Host Microbe 4, 249–25948. Iwatani, Y., Takeuchi, H., Strebel, K., and Levin, J. G. (2006) J. Virol. 80,

5992–600249. Gervaix, A., West, D., Leoni, L. M., Richman, D. D., Wong-Staal, F., and

Corbeil, J. (1997) Proc. Natl. Acad. Sci. U.S.A. 94, 4653–465850. Browne, E. P., Allers, C., and Landau, N. R. (2009) Virology 387, 313–32151. Schumacher, A. J., Hache, G., Macduff, D. A., Brown, W. L., and Harris,

R. S. (2008) J. Virol. 82, 2652–2660

AID and A3G Phosphorylation

JULY 29, 2011 • VOLUME 286 • NUMBER 30 JOURNAL OF BIOLOGICAL CHEMISTRY 26575

by guest on February 3, 2020http://w

ww

.jbc.org/D

ownloaded from

Zachary L. Demorest, Ming Li and Reuben S. HarrisActivity of Activation-induced Deaminase and APOBEC3G Protein

Phosphorylation Directly Regulates the Intrinsic DNA Cytidine Deaminase

doi: 10.1074/jbc.M111.235721 originally published online June 9, 20112011, 286:26568-26575.J. Biol. Chem. 

  10.1074/jbc.M111.235721Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

Supplemental material:

  http://www.jbc.org/content/suppl/2011/06/09/M111.235721.DC1

  http://www.jbc.org/content/286/30/26568.full.html#ref-list-1

This article cites 51 references, 21 of which can be accessed free at

by guest on February 3, 2020http://w

ww

.jbc.org/D

ownloaded from