Downloaded from genesdev.cshlp.

org on October 2, 2010 - Published by Cold Spring Harbor Laboratory Press

Cytidine deaminases: AIDing DNA demethylation?

Eric L. Fritz and F. Nina Papavasiliou

Genes Dev. 2010 24: 2107-2114

Access the most recent version at doi:10.1101/gad.1963010

References This article cites 74 articles, 19 of which can be accessed free at:
http://genesdev.cshlp.org/content/24/19/2107.full.html#ref-list-1
Email alerting Receive free email alerts when new articles cite this article - sign up in the box at the
service top right corner of the article or click here

Topic collections Articles on similar topics can be found in the following collections

Chromatin and Gene Expression (120 articles)

Immunology (13 articles)

To subscribe to Genes & Development go to:

http://genesdev.cshlp.org/subscriptions

Copyright © 2010 by Cold Spring Harbor Laboratory Press

 Downloaded from genesdev.cshlp.org on October 2, 2010 - Published by Cold Spring Harbor Laboratory Press

REVIEW

Cytidine deaminases: AIDing

DNA demethylation?
Eric L. Fritz and F. Nina Papavasiliou1
Laboratory of Lymphocyte Biology, The Rockefeller University, New York, New York 10065, USA

The presence of 5-methylcytosine (5-mC) in DNA is Significant progress has been made in elucidating the
a vital epigenetic mark in vertebrates. While the en- mechanisms of establishing and maintaining DNA meth-
zymes responsible for methylating DNA in vertebrates ylation in vertebrates (Law and Jacobsen 2010). The effec-
have been identified, the means by which this mark can tors of DNA cytosine methylation are the DNA methyl-
be removed are still unclear. Recently, it has been shown transferase (DNMT) family of enzymes (Goll and Bestor
that activation-induced cytidine deaminase (AID) con- 2005). DNMT3A and DNMT3B act as de novo methyl-
tributes to the demethylation of DNA in certain systems. transferases, establishing the basic somatic methylation
This enzyme has been intensely studied in its role as pattern early in development. In contrast, DNMT1 serves
a key driver of antibody diversification in B cells, but as a ‘‘maintenance’’ methyltransferase, acting on hemi-
recent observations from early development in zebrafish methylated DNA and thereby preserving methylation
and mice as well as heterokaryons point to a role beyond state down through multiple cell divisions.
immunology. This review takes stock of the reports DNA cytosine methylation increases rapidly around the
linking AID and related deaminases to DNA demethyl- time of implantation but thereafter remains grossly stable
ation, and describes the many important questions left to (Hemberger et al. 2009). At the blastocyst stage, coinciding
be answered in this field. with lineage fixation, the inner cell mass has gross meth-
ylation levels equivalent to those in fully differentiated
cells (Geiman and Muegge 2010). Whether changes in DNA
DNA cytosine methylation is a key epigenetic mark in methylation at specific loci are a general mechanism of gene
vertebrates. It occurs predominantly in the sequence con- regulation in differentiated cells or one restricted to silenc-
text CpG, allowing for symmetrical strand modification ing of pluripotency-associated genes after early develop-
and thus heritable transmission of epigenetic information ment is a controversial issue (Illingworth and Bird 2009).
through cell divisions. The means by which cytosine The removal of DNA cytosine methylation in verte-
methylation in DNA is regulated is an issue of great brates is a far murkier subject than its addition (Ooi and
interest because of its importance in differentiation (Reik Bestor 2008). It is clear that ‘‘passive,’’ or DNA replication-
2007), cell reprogramming (Simonsson and Gurdon 2004; dependent, demethylation can occur if DNMTs are absent
Mikkelsen et al. 2008), retroelement suppression (Walsh or prevented from accessing newly synthesized DNA.
et al. 1998), parental imprinting (Edwards and Ferguson- ‘‘Active,’’ or replication-independent, demethylation has
Smith 2007), and X-chromosome inactivation (Sado et al. been reported to occur in a number of contexts in somatic
2000). cells (Zhu 2009). However, it is unclear in these cases if
The majority (70%–80%) of cytosines in the sequence decreased promoter methylation is a driver of increased
context CpG are methylated in differentiated cells in transcription or a downstream consequence. While active
mammals (Ehrlich et al. 1982). Methylation levels are demethylation in differentiated cells remains controversial
much lower in regions with high CpG density, termed (Ooi and Bestor 2008), there are several apparent examples
‘‘CpG islands’’ (Illingworth and Bird 2009). Methylation of in normal physiology: the paternal genome immediately
such islands in the promoters of genes results in silencing after fertilization (Mayer et al. 2000; Oswald et al. 2000), the
(Bird 2002) and can regulate alternative promoter usage formation of primordial germ cells (PGCs) in the embryo
when found within gene bodies (Maunakea et al. 2010). (Hajkova et al. 2002), and the differentiation of peripheral
Additionally, cytosine methylation in non-CpG contexts blood monocytes to dendritic cells (Klug et al. 2010).
has been found at significant levels in human embryonic DNA demethylation has also been described in other
stem (ES) cells (Ramsahoye et al. 2000) and at detectable important contexts. Aberrant hypomethylation of onco-
levels in differentiated cells (Lister et al. 2009). genes, as well as hypermethylation of tumor suppressor
genes, have both been implicated in oncogenesis (Ehrlich
[Keywords: Cytidine deaminase; AICDA; AID; active demethylation; 2009). Additionally, it is clear that demethylation of the
epigenetic reprogramming] promoters of certain pluripotency-associated genes, such
1
Corresponding author.
E-MAIL [email protected]; FAX (212) 327-7319. as Oct4 and Nanog, is a requisite for ex vivo cellular
Article is online at http://www.genesdev.org/cgi/doi/10.1101/gad.1963010. reprogramming (Mikkelsen et al. 2008). In the relatively

GENES & DEVELOPMENT 24:2107–2114 Ó 2010 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/10; www.genesdev.org 2107
 Downloaded from genesdev.cshlp.org on October 2, 2010 - Published by Cold Spring Harbor Laboratory Press

Fritz and Papavasiliou

slow process of reprogramming by transcription factor Revy et al. 2000). The 24-kDa protein, encoded by the
overexpression, this demethylation is thought to occur Aicda gene, is conserved among jawed vertebrates and is
passively. However, an active mechanism is likely involved closely related to the APOBEC family of zinc-coordinating
in the fast reprogramming processes of nuclear transfer and polynucleotide cytidine deaminases (Conticello 2008). It
heterokaryon formation (Yamanaka and Blau 2010). is well established that AID is able to convert cytosine
A number of potential mechanisms have been proposed to uracil in ssDNA, as demonstrated in Escherichia coli
for active demethylation in mammals (Table 1). The most (Petersen-Mahrt et al. 2002) and in vitro (Bransteitter et al.
direct posits that the methyl group could simply be 2003; Chaudhuri et al. 2003; Dickerson et al. 2003).
enzymatically removed. The Alkb family of enzymes In the decade since AID’s discovery, a broadly accepted
have been shown to demethylate 1-methyladenosine and model for its roles in antibody diversification has emerged
3-methylcytosine via oxidized intermediates, releasing (Delker et al. 2009). AID initiates CSR and SHM by
the methyl carbon as formaldehyde and regenerating the conversion of cytosine to uracil in different regions of
original base (Falnes et al. 2002; Trewick et al. 2002). the Ig loci (Fig. 1A). CSR occurs as a result of the double-
However, no comparable Alkb family enzyme that ac- stranded breaks frequently produced in the course of repair
cepts 5-methylcytosine (5-mC) as a substrate has been of such lesions in the S regions of the IgH locus. Joining of
observed in vertebrates. The methyl-CpG-binding protein breaks in different S regions results in a different constant
MBD2 has been proposed as such a direct DNA demethy- region immediately downstream from the transcribed
lase (Bhattacharya et al. 1999), but these findings could V(D)J and, consequently, to antibodies of a different isotype
not be replicated (Bird 2002). Furthermore, a number of (Stavnezer et al. 2008). SHM is initiated by AID cytosine
studies indicate that DNA is broken and subsequently deamination within the V(D)J region of Ig loci. Repair of
repaired during the course of demethylation (Kress et al. these lesions proceeds with an unusually high error rate,
2006; Barreto et al. 2007; Hajkova et al. 2010), suggesting leading to mutations that can alter the affinity of the
that the entire base is removed. A hypothesis consistent encoded antibody (Peled et al. 2008). Selection of cells
with these reports is that DNA demethylation in mam- bearing these mutated Igs accomplishes affinity matura-
mals is achieved by targeted removal of 5-mC by a glyco- tion. AID is required for both of these processes: Aicda /
sylase without previous modification of the base. This is mice exhibit a complete lack of secondary Ig isotypes and
thought to be the dominant mechanism of active DNA no somatic mutations in Ig variable regions during an
demethylation in plants, with the DME/ROS1 family of immune response (Muramatsu et al. 2000). In humans,
glycosylases serving this function in Arabidopsis thaliana mutations in the AICDA gene result in a similar condition
(Zhu 2009). MBD4 has been proposed as such a glycosylase known as hyper-IgM syndrome type 2 (Revy et al. 2000).
in mammals (Kim et al. 2009), although previous in vitro
work found this enzyme to be far more active on thymi-
AID beyond the immune system
dine than 5-mC (Zhu et al. 2000).
Yet another proposed class of mechanisms for DNA Despite the lack of an obvious nonimmune phenotype in
cytosine demethylation involves alteration of 5-mC fol- AID-deficient mice, there are signs that AID has additional
lowed by removal of the altered base and subsequent functions outside of antibody diversification. Although it
replacement with unmethylated cytosine. There is evi- is expressed at highest levels in germinal center B cells, it
dence that this may proceed by conversion of 5-mC to a is also found in many other cell types; namely, oocytes,
5-mC radical by the elongator complex member Elp3 PGCs, ES cells (Morgan et al. 2004), breast tissue (Pauklin
(Okada et al. 2010), or to thymidine by DNMT3 family et al. 2009), and prostate epithelial cells (Lin et al. 2009),
enzymes under conditions of low S-adenosylmethionine although its expression in PGCs has been challenged
(Kangaspeska et al. 2008; Metivier et al. 2008). It recently (Hajkova et al. 2010). As AID is a DNA mutator,
has also been shown that 5-mC can be converted to its expression outside of B cells cannot be dismissed as
5-hydroxymethylcytosine (5-hmC) by the enzyme TET1 inconsequential. Indeed, AID has been shown to act
(Kriaucionis and Heintz 2009; Tahiliani et al. 2009), beyond the Ig loci, causing point mutations (Liu et al.
raising the possibility that demethylation could proceed 2008), double-stranded breaks (Hasham et al. 2010), and
through this intermediate. The requirement of TET1 for translocations (Pasqualucci et al. 2008; Lin et al. 2009;
proper maintenance of methylation state at the Nanog Robbiani et al. 2009) throughout the genome. Through
promoter in ES cells lends credence to this model these processes, AID has been shown to contribute to
(Ito et al. 2010). Finally, a series of recent studies have tumorigenesis (Okazaki et al. 2007). As such, expression of
suggested that the cytidine deaminase AID (activation- AID in non-antibody-producing cells would likely be
induced cytidine deaminase) and its relatives are also selected against in evolution unless it had some function
potential candidates for agents of active DNA demethyl- in these tissues. Further suggestions of functions for AID
ation in vertebrates. beyond the immune system come from studies of lower
vertebrates. As in mice, AID is expressed during early
development in Danio rerio (Rai et al. 2008), Xenopus
The classical view of AID
laevis (Marr et al. 2007), and the newt Pleurodeles waltl
AID was initially identified as a factor required for class (Bascove and Frippiat 2010). The broad conservation of this
switch recombination (CSR) and immunoglobulin (Ig) pattern of AID expression strongly suggests a function in
somatic hypermutation (SHM) (Muramatsu et al. 2000; early development.

2108 GENES & DEVELOPMENT

 Table 1. Partial list of pathways implicated in active DNA demethylation in vertebrates
Citation System Main findings and caveats
Studies implicating DNA repair
Barreto et al. 2007 Xenopus oocytes; findings Gadd45-coordinated DNA repair promotes genome-wide demethylation.
confirmed in human and
mouse cell lines
Rai et al. 2008 Zebrafish early embryo Gadd45 family proteins and MBD4 are required for active demethylation.
Metivier et al. 2008 Hormone-responsive cell lines TDG, BER, and DNMT3 proteins are required for active demethylation
at the pS2 gene promoter.
Hajkova et al. 2010 Mouse PGCs and zygotes BER proteins Parp1, Ape1, and Xrcc1 are mediators of demethylation;
Tet1, but not AID, is expressed in PGCs.
Kim et al. 2009 Mouse cell lines; wild-type Hormone-mediated recruitment of MBD4 to the Cyp27b1 promoter is
and Mbd4 / tissue necessary for active demethylation.
Studies implicating DNA deamination
Morgan et al. 2004 In vitro assays, mouse tissues AID can deaminate 5-mC and is expressed in pluripotent tissues. These
studies first suggested the potential for AID to act as a demethylase.
Rai et al. 2008 Zebrafish early embryo AID and Apobec2 are required for demethylation of exogenous DNA
in the zebrafish embryo.
/
Popp et al. 2010 Aicda and wild-type mice The genomes of Aicda / PGCs were hypermethylated (50% increase,
genome-wide) when compared with wild-type controls; however,
Aicda / animals are healthy and fertile.
Bhutani et al. 2010 Mouse/human heterokaryons Epigenetic reprogramming toward pluripotency, in a setting where active
demethylation is required, is facilitated by AID.
Studies implicating DNA oxidation and other pathways
Tahiliani et al. 2009 In vitro assays, human cell line Human Tet1 converts 5-mC to 5-hmC; it is hypothesized that 5-hmC could be
an intermediate in demethylation.
Ito et al. 2010 Mouse PGCs and zygotes, All three mouse Tet proteins (Tet1, Tet2, and Tet3) can mediate hydroxylation
in vitro assays of 5-mC; Tet1 itself is required for ES cell maintenance.
Okada et al. 2010 Mouse zygotes The radical SAM domain of Elp3, a component of the RNA polymerase
elongator complex, is necessary for DNA demethylation at paternal pronuclei.
Downloaded from genesdev.cshlp.org on October 2, 2010 - Published by Cold Spring Harbor Laboratory Press

GENES & DEVELOPMENT

Cytidine deaminases in demethylation

2109
 Downloaded from genesdev.cshlp.org on October 2, 2010 - Published by Cold Spring Harbor Laboratory Press

Fritz and Papavasiliou

inates at most sites, it is likely the norm, while frequent

mutation as observed at the Ig loci is the special case.

AID-dependent demethylation in lower vertebrates

The AID model of demethylation received a measure of in
vivo validation from work in D. rerio early embryos (Rai
et al. 2008). Introduction of a methylated DNA fragment
into single-cell embryos induces expression of AID along
with related putative cytidine deaminases Apobec2a and
Apobec2b. Additionally, overexpression of AID along with
the T-G glycosylase MBD4 leads to efficient demethyla-
Figure 1. AID can initiate distinct processes through a single tion of a methylated DNA fragment, with observable
enzymatic activity. (A) Deamination of cytosine yields uracil. conversion of 5-mC to T. Morpholino knockdown of
Excision of uracil by UNG and repair of the subsequent abasic AID or MBD4 in D. rerio single-cell stage embryos results
site leads to error-prone repair and SHM when it occurs in the
in hypermethylation of the promoter of the neurogenesis-
V(D)J regions of the Ig loci. Although error-prone repair also gives
related transcription factor neurod2 at 80% epiboly. While
rise to mutations at nearby bases, only the outcome of repair on
C is shown. (B) Deamination of 5-mC by AID yields thymine. the large number of cell divisions between morpholino
Excision of thymine by either of the mammalian T-G-specific injection and the observed difference in methylation at
glycosylases (TDG or MBD4) and subsequent error-free replace- neurod2 seems to point toward a passive mechanism,
ment with cytosine would yield loss of methylation at that base both AID and MBD4 were also detectable at the locus by
on one strand. (C) Deamination of cytosine followed by excision chromatin immunoprecipitation (ChIP), suggesting that
of uracil and error-free long-patch BER could lead to replacement the promoter’s methylation state was dynamic through
and net demethylation of neighboring 5-mCs. development and not just the result of perturbations at the
single-cell stage. Consequently, AID morphants displayed
The first direct evidence that AID might have functions severe defects in neurogenesis consistent with decreased
beyond the standard model of antibody diversification neurod2 expression. The tissue- and locus-specific pheno-
came in 2004, when it was shown by Petersen-Mahrt and type suggests that active demethylation may serve as a
colleagues (Morgan et al. 2004) that AID, along with the mechanism for regulating specific lineage decisions, as
related cytidine deaminase APOBEC1, can convert 5-mC opposed to only resetting methylation state grossly in the
in ssDNA to thymidine in vitro. This observation led to zygote and germline.
the proposal that these enzymes could be the elusive Intriguingly, Rai et al. (2008) also found that the Gadd45
vertebrate DNA cytosine demethylase. family of proteins, previously implicated in active demeth-
AID could initiate demethylation by a damage and ylation in X. laevis (Barreto et al. 2007), were involved
repair mechanism similar to that used in SHM (Fig. 1B). in AID-dependent demethylation. Combined knockdown
The deamination of 5-mC by AID yields thymidine. This of four of the six Gadd45 family members found in zebra-
T would then be removed by a T-G mismatch-specific fish sharply reduced demethylation of a reporter and in-
glycosylase, of which two, TDG and MBD4, are known in creased methylation of the genome as a whole. It was thus
mammals (Millar et al. 2002; Hardeland et al. 2003). The suggested that Gadd45 serves as a scaffold to couple AID
resulting abasic site would then be replaced by an unme- and MBD4, supported by the facts that Gadd45a promotes
thylated cytidine via base excision repair (BER) processes, their association with a methylated reporter plasmid and
yielding the net removal of methylation without alter- that the three proteins also coimmunoprecipitate. The
ation of sequence. BER could also proceed through either apparent physical interaction of AID and MBD4, enforced
short or long patch repair (Fortini and Dogliotti 2007), by Gadd45, provides a potential mechanism for the tight
potentially yielding demethylation of multiple neighbor- coupling of deamination to repair that would be necessary
ing cytosines in the latter case (Fig. 1C). This mechanism for demethylation without attendant widespread muta-
would allow for demethylation after deamination of tion, as has been noted (Law and Jacobsen 2010).
cytosine or 5-mC, and could give rise to the appearance
of processive demethylation, despite originating from a
AID-dependent demethylation in mammalian PGCs
single deamination event.
It may seem counterintuitive to invoke the mechanism Although there is no direct evidence for deaminase-de-
of SHM to explain that of DNA demethylation, which pendent effects on cytosine methylation in mammalian
proceeds with fidelity. However, the fact that AID-induced somatic tissue, a recent study has implicated AID in
lesions are apparently resolved through multiple path- establishment of the hypomethylated state of mouse
ways makes this more plausible. At a small number of PGCs (Popp et al. 2010). In this study, genomic DNA from
loci—most dramatically, Ig—repair of AID-induced lesions sperm, total fetus, placenta, and male and female embry-
is particularly error-prone, while, at other genomic loca- onic day 13.5 (E13.5) PGCs from wild-type and Aicda /
tions, it results in almost undetectable levels of mutation mice was bisulfite-converted and sequenced using the
(Liu et al. 2008). How these different pathways are selected Illumina ultrahigh-throughput platform, a method termed
is still unclear, but, given that high-fidelity repair predom- BS-Seq or MethylC-Seq (Cokus et al. 2008; Lister et al.

2110 GENES & DEVELOPMENT

 Downloaded from genesdev.cshlp.org on October 2, 2010 - Published by Cold Spring Harbor Laboratory Press

Cytidine deaminases in demethylation

2008). The resulting data provide a genome-wide map of et al. 2010). This system uses polyethylene glycol fusion of
cytosine methylation at single-nucleotide resolution. mouse ES cells with human fibroblasts to induce repro-
While a lack of sequencing depth precluded quantitative gramming of the human nucleus to an ES-like state with
measures of methylation for every genomic cytosine, the high frequency. Because reprogramming is known to in-
coverage was sufficient to detect a significant increase in volve demethylation, heterokaryon formation allows for
methylation in Aicda / PGCs. This difference was more study of demethylation in mammalian cells, albeit in a
pronounced for female than male PGCs, and was roughly nonphysiological context. Knockdown of AID was found
homogeneous throughout the genome. The broad nature to significantly inhibit reprogramming, as measured by
of the methylation increase implies that AID functions transcript levels of the pluripotency markers Oct4 and
without regard to specific loci, and thus is a plausible Nanog. Methylation of the promoters of these genes was
component of germline methylation erasure. also significantly increased as a result of reduced levels of
While the evidence that AID has a role in demethylation AID. As heterokaryons are nondividing, AID-dependent
is suggestive, there are still considerable caveats. Most demethylation in this system is necessarily active.
significantly, a recent report detected no AID expression in AID is known to act only on ssDNA, which presumably
mouse PGCs at E11.5, although a small level of AID occurs transiently in vivo during transcription (Shen et al.
expression was observed at E12.5 (Hajkova et al. 2010). 2009). Yet AID-dependent demethylation clearly occurs at
As the epigenetic reprogramming of PGCs begins at E11.5, the promoters of silenced genes in heterokaryons in the
AID cannot be the sole agent of demethylation. This idea absence of replication. Whether these promoters share
is consistent with the occurrence of significant, albeit distinguishing structural features that would give rise to
reduced, demethylation in the PGCs of Aicda / mice transient ssDNA, such as Z/B DNA junctions (de Rosa
compared with somatic cells. Whether AID-independent et al. 2010) or other non-B structures (Raghavan et al.
demethylation in these cells is a result of 5-mC deamina- 2004), is not known. However, at least for the limited
tion by another member of the AID/APOBEC family or is number of genes studied, there seems to be no clear re-
due to any of the mechanisms mentioned above is an open lation between AID occupancy and transcription: While
question. It is also possible that the AID/APOBEC family Oct4 and Nanog are both transcribed and occupied, Tp53 is
of deaminases have differing target gene preferences for transcribed but not occupied, and the constant (C) region of
demethylation, and thus play complementary roles in the IgH is occupied but presumably not transcribed (although
genome-wide removal of cytosine methylation. this was not investigated). Additionally, AID could be
Another issue raised by these results is the functional found by ChIP at the human fibroblast Oct4 and Nanog
importance of the observed PGC hypermethylation, as loci, but not at the same genes in the mouse ES nuclei. This
Aicda / mice are fully viable. Even if AID were the sole provides evidence for AID’s (or an AID-containing com-
agent of active demethylation, it is not clear that loss of plex’s) affinity for silenced (or at least methylated) genes.
this mechanism would result in a drastic phenotype. This Whether this affinity is caused by inherent tendency of AID
view is supported by evidence from A. thaliana, in which to bind 5-mC or a more intricately targeted system of
loss of all three 5-mC-removing glycosylases in vegetative reprogramming toward pluripotency also active in PGCs
tissue leads to viable plants displaying increased methyl- remains to be determined. Whatever the case may be, the
ation only at certain loci, and no genome-wide increase behavior of AID in heterokaryons likely has implications
(Penterman et al. 2007). As DNA methylation is essential for the biology of PGCs, and likely B cells as well.
for parental imprinting, Popp et al. (2010) suggested that
the consequences of PGC AID deficiency may lie in reten-
Future directions
tion of a parental-like epigenetic state. Data supporting
this view lies in a small but significant difference that In certain lower vertebrates, AID appears to be involved
exists between wild-type and Aicda / mice in the relation- in a process other than antibody diversification. Although
ship between litter size and birth weight. In normal mice, the early embryos of X. laevis and D. rerio provide a more
pups that are part of large litters tend to have lower accessible in vivo system for studying active demethylation,
birth weights, while in Aicda / mice this compensation the discrepancies between findings in these lower ver-
is absent (Popp et al. 2010). Hypermethylated elements tebrates and mammals calls into question the existence
responsible for this phenotype have not been identi- of a single demethylation mechanism shared between
fied. New technologies for methylation-sensitive high- these groups. In particular, there is no evidence that any
throughput sequencing (Flusberg et al. 2010) and the member of the Gadd45 family is involved in mediating
decreasing cost of ‘‘traditional’’ deep sequencing should demethylation in mammals. While Gadd45a overex-
make the quantitative, single-nucleotide resolution stud- pression alone has been shown to promote demethyla-
ies of PGC methylation feasible in the near future to tion in X. laevis (Barreto et al. 2007), and is required for
address this question. AID- and MBD4-dependent demethylation in zebrafish,
it also appears to be dispensable for normal regulation of
cytosine methylation in mice (Engel et al. 2009). Further-
AID-dependent demethylation in heterokaryons
more, the gene is not expressed in the mouse oocyte (Jin
Corroboration of AID’s capacity to demethylate DNA in et al. 2008), precluding its involvement in post-fertilization
mammalian cells came in a recent study of reprogram- paternal genome demethylation. One hypothesis that
ming during interspecies heterokaryon formation (Bhutani would account for the differences between observations

GENES & DEVELOPMENT 2111

 Downloaded from genesdev.cshlp.org on October 2, 2010 - Published by Cold Spring Harbor Laboratory Press

Fritz and Papavasiliou

in mice and zebrafish is that AID is redundant with other Cokus SJ, Feng S, Zhang X, Chen Z, Merriman B, Haudenschild
APOBEC family deaminases in demethylation (Morgan CD, Pradhan S, Nelson SF, Pellegrini M, Jacobsen SE. 2008.
et al. 2004). Phylogenetics lends some support to this idea, Shotgun bisulphite sequencing of the Arabidopsis genome
as there are several members of the APOBEC gene family reveals DNA methylation patterning. Nature 452: 215–219.
Conticello SG. 2008. The AID/APOBEC family of nucleic acid
that first appear in eutherians (Conticello et al. 2005).
mutators. Genome Biol 9: 229.
Even restricting the discussion to mammalian systems, Conticello SG, Thomas CJ, Petersen-Mahrt SK, Neuberger MS.
there are significant questions regarding AID’s role in 2005. Evolution of the AID/APOBEC family of polynucleo-
demethylation. If AID is acting as a demethylase in PGCs tide (deoxy)cytidine deaminases. Mol Biol Evol 22: 367–377.
and heterokaryons, the question arises as to whether it Delker RK, Fugmann SD, Papavasiliou FN. 2009. A coming-of-
does the same in B cells, given that there is enough AID in age story: Activation-induced cytidine deaminase turns 10.
activated B cells to mutate a number of non-Ig loci (Liu Nat Immunol 10: 1147–1153.
et al. 2008) and cause double-stranded breaks throughout de Rosa M, de Sanctis D, Rosario AL, Archer M, Rich A,
the genome in the absence of proper DNA repair (Hasham Athanasiadis A, Carrondo MA. 2010. Crystal structure of a
et al. 2010). One possible explanation is that there are yet junction between two Z-DNA helices. Proc Natl Acad Sci
107: 9088–9092.
unidentified cell type-specific cofactors of AID that either
Dickerson SK, Market E, Besmer E, Papavasiliou FN. 2003. AID
promote demethylation in pluripotent cells or suppress mediates hypermutation by deaminating single stranded
this activity in B cells. It is also conceivable that demeth- DNA. J Exp Med 197: 1291–1296.
ylation might occur in B cells at some rate, with the effect Edwards CA, Ferguson-Smith AC. 2007. Mechanisms regulating
being less dramatic due to more frequent DNA methyl- imprinted genes in clusters. Curr Opin Cell Biol 19: 281–289.
ation, resulting in a dynamic equilibrium. Ehrlich M. 2009. DNA hypomethylation in cancer cells. Epi-
Whatever the case, AID’s likely role in demethylation genomics 1: 239–259.
is pertinent to those interested in epigenetics, develop- Ehrlich M, Gama-Sosa MA, Huang LH, Midgett RM, Kuo KC,
ment, and immunology. The identification of a promising McCune RA, Gehrke C. 1982. Amount and distribution of
candidate for vertebrate DNA cytidine demethylase has 5-methylcytosine in human DNA from different types of
tissues of cells. Nucleic Acids Res 10: 2709–2721.
raised more questions than it has answered; elucidation
Engel N, Tront JS, Erinle T, Nguyen N, Latham KE, Sapienza C,
of the particulars of AID’s relationship to DNA demeth- Hoffman B, Liebermann DA. 2009. Conserved DNA meth-
ylation will surely be the subject of much inquiry in the ylation in Gadd45a / mice. Epigenetics 4: 98–99.
coming years. Falnes PØ, Johansen RF, Seeberg E. 2002. AlkB-mediated oxida-
tive demethylation reverses DNA damage in Escherichia
coli. Nature 419: 178–182.
Acknowledgments Flusberg BA, Webster DR, Lee JH, Travers KJ, Olivares EC, Clark
TA, Korlach J, Turner SW. 2010. Direct detection of DNA
We thank Jayanta Chaudhuri, Claire Hamilton, and Brad Rosenberg
methylation during single-molecule, real-time sequencing.
for thoughtful comments on the manuscript. Work relating to
Nat Methods 7: 461–465.
AID and its roles in Ig mutation and cancer in F.N.P.’s laboratory
Fortini P, Dogliotti E. 2007. Base damage and single-strand break
is supported by NCI grant no. CA098495 and the Starr Cancer
repair: Mechanisms and functional significance of short- and
Consortium. E.L.F. is supported by a predoctoral training grant
long-patch repair subpathways. DNA Repair (Amst) 6: 398–
from the Cancer Research Institute.
409.
Geiman TM, Muegge K. 2010. DNA methylation in early
development. Mol Reprod Dev 77: 105–113.
References
Goll MG, Bestor TH. 2005. Eukaryotic cytosine methyltrans-
Barreto G, Schafer A, Marhold J, Stach D, Swaminathan SK, ferases. Annu Rev Biochem 74: 481–514.
Handa V, Doderlein G, Maltry N, Wu W, Lyko F, et al. 2007. Hajkova P, Erhardt S, Lane N, Haaf T, El-Maarri O, Reik W,
Gadd45a promotes epigenetic gene activation by repair- Walter J, Surani MA. 2002. Epigenetic reprogramming in
mediated DNA demethylation. Nature 445: 671–675. mouse primordial germ cells. Mech Dev 117: 15–23.
Bascove M, Frippiat JP. 2010. Molecular characterization of Hajkova P, Jeffries SJ, Lee C, Miller N, Jackson SP, Surani MA.
Pleurodeles waltl activation-induced cytidine deaminase. 2010. Genome-wide reprogramming in the mouse germ line
Mol Immunol 47: 1640–1649. entails the base excision repair pathway. Science 329:
Bhattacharya SK, Ramchandani S, Cervoni N, Szyf M. 1999. A 78–82.
mammalian protein with specific demethylase activity for Hardeland U, Bentele M, Jiricny J, Schar P. 2003. The versatile
mCpG DNA. Nature 397: 579–583. thymine DNA-glycosylase: A comparative characterization
Bhutani N, Brady JJ, Damian M, Sacco A, Corbel SY, Blau HM. of the human, Drosophila and fission yeast orthologs.
2010. Reprogramming towards pluripotency requires AID- Nucleic Acids Res 31: 2261–2271.
dependent DNA demethylation. Nature 463: 1042–1047. Hasham MG, Donghia NM, Coffey E, Maynard J, Snow KJ,
Bird A. 2002. DNA methylation patterns and epigenetic mem- Ames J, Wilpan RY, He Y, King BL, Mills KD. 2010.
ory. Genes Dev 16: 6–21. Widespread genomic breaks generated by activation-induced
Bransteitter R, Pham P, Scharff MD, Goodman MF. 2003. cytidine deaminase are prevented by homologous recombi-
Activation-induced cytidine deaminase deaminates deoxy- nation. Nat Immunol 11: 820–826.
cytidine on single-stranded DNA but requires the action of Hemberger M, Dean W, Reik W. 2009. Epigenetic dynamics of
RNase. Proc Natl Acad Sci 100: 4102–4107. stem cells and cell lineage commitment: Digging Wadding-
Chaudhuri J, Tian M, Khuong C, Chua K, Pinaud E, Alt FW. ton’s canal. Nat Rev Mol Cell Biol 10: 526–537.
2003. Transcription-targeted DNA deamination by the AID Illingworth RS, Bird AP. 2009. CpG islands–‘a rough guide.’
antibody diversification enzyme. Nature 422: 726–730. FEBS Lett 583: 1713–1720.

2112 GENES & DEVELOPMENT

 Downloaded from genesdev.cshlp.org on October 2, 2010 - Published by Cold Spring Harbor Laboratory Press

Cytidine deaminases in demethylation

Ito S, D’Alessio AC, Taranova OV, Hong K, Sowers LC, Zhang Y. Millar CB, Guy J, Sansom OJ, Selfridge J, MacDougall E,
2010. Role of Tet proteins in 5mC to 5hmC conversion, ES- Hendrich B, Keightley PD, Bishop SM, Clarke AR, Bird A.
cell self-renewal and inner cell mass specification. Nature 2002. Enhanced CpG mutability and tumorigenesis in MBD4-
466: 1129–1133. deficient mice. Science 297: 403–405.
Jin SG, Guo C, Pfeifer GP. 2008. GADD45A does not promote Morgan HD, Dean W, Coker HA, Reik W, Petersen-Mahrt SK.
DNA demethylation. PLoS Genet 4: e1000013. doi: 10.1371/ 2004. Activation-induced cytidine deaminase deaminates
journal.pgen.1000013. 5-methylcytosine in DNA and is expressed in pluripotent
Kangaspeska S, Stride B, Metivier R, Polycarpou-Schwarz M, tissues: Implications for epigenetic reprogramming. J Biol
Ibberson D, Carmouche RP, Benes V, Gannon F, Reid G. Chem 279: 52353–52360.
2008. Transient cyclical methylation of promoter DNA. Muramatsu M, Kinoshita K, Fagarasan S, Yamada S, Shinkai Y,
Nature 452: 112–115. Honjo T. 2000. Class switch recombination and hypermuta-
Kim MS, Kondo T, Takada I, Youn MY, Yamamoto Y, Takahashi tion require activation-induced cytidine deaminase (AID), a
S, Matsumoto T, Fujiyama S, Shirode Y, Yamaoka I, et al. potential RNA editing enzyme. Cell 102: 553–563.
2009. DNA demethylation in hormone-induced transcrip- Okada Y, Yamagata K, Hong K, Wakayama T, Zhang Y. 2010. A
tional derepression. Nature 461: 1007–1012. role for the elongator complex in zygotic paternal genome
Klug M, Heinz S, Gebhard C, Schwarzfischer L, Krause SW, demethylation. Nature 463: 554–558.
Andreesen R, Rehli M. 2010. Active DNA demethylation in Okazaki IM, Kotani A, Honjo T. 2007. Role of AID in tumori-
human postmitotic cells correlates with activating histone genesis. Adv Immunol 94: 245–273.
modifications, but not transcription levels. Genome Biol 11: Ooi SK, Bestor TH. 2008. The colorful history of active DNA
R63. doi: 10.1186/gb-2010-11-6-r63. demethylation. Cell 133: 1145–1148.
Kress C, Thomassin H, Grange T. 2006. Active cytosine demeth- Oswald J, Engemann S, Lane N, Mayer W, Olek A, Fundele R,
ylation triggered by a nuclear receptor involves DNA strand Dean W, Reik W, Walter J. 2000. Active demethylation of
breaks. Proc Natl Acad Sci 103: 11112–11117. the paternal genome in the mouse zygote. Curr Biol 10:
Kriaucionis S, Heintz N. 2009. The nuclear DNA base 5- 475–478.
hydroxymethylcytosine is present in Purkinje neurons and Pasqualucci L, Bhagat G, Jankovic M, Compagno M, Smith P,
the brain. Science 324: 929–930. Muramatsu M, Honjo T, Morse HC III, Nussenzweig MC,
Law JA, Jacobsen SE. 2010. Establishing, maintaining and Dalla-Favera R. 2008. AID is required for germinal center-
modifying DNA methylation patterns in plants and animals. derived lymphomagenesis. Nat Genet 40: 108–112.
Nat Rev Genet 11: 204–220. Pauklin S, Sernandez IV, Bachmann G, Ramiro AR, Petersen-
Lin C, Yang L, Tanasa B, Hutt K, Ju BG, Ohgi K, Zhang J, Rose Mahrt SK. 2009. Estrogen directly activates AID transcrip-
DW, Fu XD, Glass CK, et al. 2009. Nuclear receptor-induced tion and function. J Exp Med 206: 99–111.
chromosomal proximity and DNA breaks underlie specific Peled JU, Kuang FL, Iglesias-Ussel MD, Roa S, Kalis SL,
translocations in cancer. Cell 139: 1069–1083. Goodman MF, Scharff MD. 2008. The biochemistry of
Lister R, O’Malley RC, Tonti-Filippini J, Gregory BD, Berry CC, somatic hypermutation. Annu Rev Immunol 26: 481–511.
Millar AH, Ecker JR. 2008. Highly integrated single-base Penterman J, Zilberman D, Huh JH, Ballinger T, Henikoff S,
resolution maps of the epigenome in Arabidopsis. Cell 133: Fischer RL. 2007. DNA demethylation in the Arabidopsis
523–536. genome. Proc Natl Acad Sci 104: 6752–6757.
Lister R, Pelizzola M, Dowen RH, Hawkins RD, Hon G, Tonti- Petersen-Mahrt SK, Harris RS, Neuberger MS. 2002. AID
Filippini J, Nery JR, Lee L, Ye Z, Ngo QM, et al. 2009. Human mutates E. coli suggesting a DNA deamination mechanism
DNA methylomes at base resolution show widespread for antibody diversification. Nature 418: 99–103.
epigenomic differences. Nature 462: 315–322. Popp C, Dean W, Feng S, Cokus SJ, Andrews S, Pellegrini M,
Liu M, Duke JL, Richter DJ, Vinuesa CG, Goodnow CC, Jacobsen SE, Reik W. 2010. Genome-wide erasure of DNA
Kleinstein SH, Schatz DG. 2008. Two levels of protection methylation in mouse primordial germ cells is affected by
for the B cell genome during somatic hypermutation. Nature AID deficiency. Nature 463: 1101–1105.
451: 841–845. Raghavan SC, Swanson PC, Wu X, Hsieh CL, Lieber MR. 2004.
Marr S, Morales H, Bottaro A, Cooper M, Flajnik M, Robert J. A non-B-DNA structure at the Bcl-2 major breakpoint region
2007. Localization and differential expression of activation- is cleaved by the RAG complex. Nature 428: 88–93.
induced cytidine deaminase in the amphibian Xenopus upon Rai K, Huggins IJ, James SR, Karpf AR, Jones DA, Cairns BR. 2008.
antigen stimulation and during early development. J Immu- DNA demethylation in zebrafish involves the coupling of
nol 179: 6783–6789. a deaminase, a glycosylase, and gadd45. Cell 135: 1201–1212.
Maunakea AK, Nagarajan RP, Bilenky M, Ballinger TJ, Ramsahoye BH, Biniszkiewicz D, Lyko F, Clark V, Bird AP,
D’Souza C, Fouse SD, Johnson BE, Hong C, Nielsen C, Jaenisch R. 2000. Non-CpG methylation is prevalent in
Zhao Y, et al. 2010. Conserved role of intragenic DNA embryonic stem cells and may be mediated by DNA meth-
methylation in regulating alternative promoters. Nature 466: yltransferase 3a. Proc Natl Acad Sci 97: 5237–5242.
253–257. Reik W. 2007. Stability and flexibility of epigenetic gene regula-
Mayer W, Niveleau A, Walter J, Fundele R, Haaf T. 2000. tion in mammalian development. Nature 447: 425–432.
Demethylation of the zygotic paternal genome. Nature 403: Revy P, Muto T, Levy Y, Geissmann F, Plebani A, Sanal O,
501–502. Catalan N, Forveille M, Dufourcq-Labelouse R, Gennery A,
Metivier R, Gallais R, Tiffoche C, Le Peron C, Jurkowska RZ, et al. 2000. Activation-induced cytidine deaminase (AID)
Carmouche RP, Ibberson D, Barath P, Demay F, Reid G, et al. deficiency causes the autosomal recessive form of the hyper-
2008. Cyclical DNA methylation of a transcriptionally ac- IgM syndrome (HIGM2). Cell 102: 565–575.
tive promoter. Nature 452: 45–50. Robbiani DF, Bunting S, Feldhahn N, Bothmer A, Camps J,
Mikkelsen TS, Hanna J, Zhang X, Ku M, Wernig M, Schorderet Deroubaix S, McBride KM, Klein IA, Stone G, Eisenreich TR,
P, Bernstein BE, Jaenisch R, Lander ES, Meissner A. 2008. et al. 2009. AID produces DNA double-strand breaks in non-
Dissecting direct reprogramming through integrative geno- Ig genes and mature B cell lymphomas with reciprocal
mic analysis. Nature 454: 49–55. chromosome translocations. Mol Cell 36: 631–641.

GENES & DEVELOPMENT 2113

 Downloaded from genesdev.cshlp.org on October 2, 2010 - Published by Cold Spring Harbor Laboratory Press

Fritz and Papavasiliou

Sado T, Fenner MH, Tan SS, Tam P, Shioda T, Li E. 2000. X

inactivation in the mouse embryo deficient for Dnmt1:
Distinct effect of hypomethylation on imprinted and random
X inactivation. Dev Biol 225: 294–303.
Shen HM, Poirier MG, Allen MJ, North J, Lal R, Widom J, Storb
U. 2009. The activation-induced cytidine deaminase (AID)
efficiently targets DNA in nucleosomes but only during
transcription. J Exp Med 206: 1057–1071.
Simonsson S, Gurdon J. 2004. DNA demethylation is necessary
for the epigenetic reprogramming of somatic cell nuclei. Nat
Cell Biol 6: 984–990.
Stavnezer J, Guikema JE, Schrader CE. 2008. Mechanism and
regulation of class switch recombination. Annu Rev Immu-
nol 26: 261–292.
Tahiliani M, Koh KP, Shen Y, Pastor WA, Bandukwala H, Brudno
Y, Agarwal S, Iyer LM, Liu DR, Aravind L, et al. 2009.
Conversion of 5-methylcytosine to 5-hydroxymethylcyto-
sine in mammalian DNA by MLL partner TET1. Science
324: 930–935.
Trewick SC, Henshaw TF, Hausinger RP, Lindahl T, Sedgwick B.
2002. Oxidative demethylation by Escherichia coli AlkB
directly reverts DNA base damage. Nature 419: 174–178.
Walsh CP, Chaillet JR, Bestor TH. 1998. Transcription of IAP
endogenous retroviruses is constrained by cytosine methyl-
ation. Nat Genet 20: 116–117.
Yamanaka S, Blau HM. 2010. Nuclear reprogramming to a plu-
ripotent state by three approaches. Nature 465: 704–712.
Zhu JK. 2009. Active DNA demethylation mediated by DNA
glycosylases. Annu Rev Genet 43: 143–166.
Zhu B, Zheng Y, Angliker H, Schwarz S, Thiry S, Siegmann M,
Jost JP. 2000. 5-Methylcytosine DNA glycosylase activity is
also present in the human MBD4 (G/T mismatch glycosy-
lase) and in a related avian sequence. Nucleic Acids Res 28:
4157–4165.

2114 GENES & DEVELOPMENT