Transcription Is Required to Establish Maternal
Imprinting at the Prader-Willi Syndrome and Angelman
Emily Y. Smith, Christopher R. Futtner¤a, Stormy J. Chamberlain¤b, Karen A. Johnstone¤c, James L.
Department of Molecular Genetics and Microbiology, University of Florida College of Medicine, Gainesville, Florida, United States of America
The Prader-Willi syndrome (PWS [MIM 17620]) and Angelman syndrome (AS [MIM 105830]) locus is controlled by a bipartite
imprinting center (IC) consisting of the PWS-IC and the AS-IC. The most widely accepted model of IC function proposes that
the PWS-IC activates gene expression from the paternal allele, while the AS-IC acts to epigenetically inactivate the PWS-IC
on the maternal allele, thus silencing the paternally expressed genes. Gene order and imprinting patterns at the PWS/AS
locus are well conserved from human to mouse; however, a murine AS-IC has yet to be identified. We investigated a
potential regulatory role for transcription from the Snrpn alternative upstream exons in silencing the maternal allele using a
murine transgene containing Snrpn and three upstream exons. This transgene displayed appropriate imprinted expression
and epigenetic marks, demonstrating the presence of a functional AS-IC. Transcription of the upstream exons from the
endogenous locus correlates with imprint establishment in oocytes, and this upstream exon expression pattern was
conserved on the transgene. A transgene bearing targeted deletions of each of the three upstream exons exhibited loss of
imprinting upon maternal transmission. These results support a model in which transcription from the Snrpn upstream
exons directs the maternal imprint at the PWS-IC.
Citation: Smith EY, Futtner CR, Chamberlain SJ, Johnstone KA, Resnick JL (2011) Transcription Is Required to Establish Maternal Imprinting at the Prader-Willi
Syndrome and Angelman Syndrome Locus. PLoS Genet 7(12): e1002422. doi:10.1371/journal.pgen.1002422
Editor: Anne C. Ferguson-Smith, University of Cambridge, United Kingdom
Received September 12, 2011; Accepted October 27, 2011; Published December 29, 2011
Copyright: ? 2011 Smith et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This research was funded by a NICHD grant (NIH HD037872) http://www.nichd.nih.gov/. EYS was supported by a UF Alumni Fellowship. The funders
had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: firstname.lastname@example.org
¤a Current address: Department of Surgery, Duke University Medical Center, Durham, North Carolina, United States of America
¤b Current address: Department of Genetics and Developmental Biology, University of Connecticut Health Center, Farmington, Connecticut, United States of
¤c Current address: Institute of Biomedical and Clinical Science, Peninsula Medical School, University of Exeter, Exeter, United Kingdom
Genomic imprinting is an epigenetic phenomenon that occurs
at a subset of chromosomal regions and results in parent-of-origin
specific monoallelic gene expression. Imprinted genes are
frequently found in clusters, coordinately regulated by imprinting
centers (ICs) that direct allele-specific differences in transcription,
DNA methylation, histone modifications and replication timing
[1–4]. Appropriate control of imprinted gene expression is vital to
growth and development and errors in imprinting may lead to
developmental disorders or embryonic lethality.
Prader-Willi syndrome (PWS) and Angelman syndrome (AS) are
distinct neurogenetic disorders resulting from improper gene
expression from an imprinted domain on chromosome 15q11–
q13, the PWS/AS locus. Mutations arising from the maternal
chromosome that lead to a loss of UBE3A function are sufficient to
cause AS [5,6]. PWS results from the loss of multiple paternal gene
products encoded at the PWS/AS locus. A bipartite IC, comprised
of the PWS-IC and the AS-IC, regulates both epigenetic
reprogramming and allele-specific gene expression at the PWS/
AS locus [7,8]. Prevailing models of IC function suggest that the
PWS-IC is a positive element that activates gene expression from
the paternal allele. The AS-IC acts as a negative element to direct
inhibitory epigenetic modifications at the PWS-IC during
oogenesis, thereby silencing the paternally expressed genes on
the future maternal allele [8–10]. A subset of individuals with PWS
or AS bear microdeletions that disrupt gene expression at the
PWS/AS locus. The shortest regions of overlap for these
microdeletions define the boundaries of the IC. The PWS-IC is
located in a region of 4.3 kb, just 59 to and including exon 1 of
SNRPN, while the AS-IC is contained within 0.88 kb approxi-
mately 35 kb upstream of the PWS-IC [11,12]. Notably, within
the 0.88 kb of AS-IC sequence reside two of several alternative
upstream exons of SNRPN [13,14]. These upstream exons are
postulated to play a role in silencing the PWS-IC on the maternal
The PWS/AS locus is highly conserved from human to mouse
in both gene order and allelic gene expression patterns, providing
an excellent model for studying IC-directed imprinting mecha-
nisms within this domain (Figure 1A). Hindering these studies has
been the absence of an identifiable murine AS-IC. However, as in
the human, the murine Snrpn locus contains several alternative
PLoS Genetics | www.plosgenetics.org1 December 2011 | Volume 7 | Issue 12 | e1002422
upstream promoters and exons, herein referred to as U exons,
from which multiple alternatively spliced transcripts arise [16,17].
These U exons are transcribed exclusively from the paternal allele
in the brain as well as in oocytes [15,16,18]. U exon expression in
the oocyte suggests that transcription is essential for maternal
imprinting at the PWS/AS locus. In this report, we test whether
the murine U exons have AS-IC activity.
Previous targeted deletion approaches to knockout AS-IC
imprinting defects [19,20], suggesting that the murine AS-IC
consists of multiple elements. We have therefore taken a
transgenic approach to identify the murine AS-IC. We utilized
a BAC transgene that contains Snrpn and 100 kb of upstream
sequence, including three of the alternative U exons. We show
that this transgene is properly imprinted in multiple transgenic
lines, indicating that there is a functional AS-IC contained within
or resultedin only partial
its sequence. Transcription of the U exons from the transgene
correlates with imprint establishment in the maternal germ line.
Upon targeted deletion of the three U exons, this transgene loses
both the epigenetic imprint as well as its imprinted expression
pattern, demonstrating that the AS-IC activity originates from the
AS-IC Activity Is Contained within a 100-kb Region
Upstream of Snrpn
To establish a transgenic system that would faithfully model AS-
IC activity, it was necessary to identify a murine BAC that
contained the PWS-IC and expressed Snrpn exclusively after
paternal transmission. In addition, to study the role of the U exons,
it was important to compare BAC transgenes containing U exons
to those without. We screened the C57BL/6 RPCI-23 murine
BAC library and identified BACs 380J10 and 425D18 for further
study. The 380J10 BAC contains approximately 15 kb of sequence
upstream of Snrpn exon 1, a region lacking any alternative Snrpn U
exons. The 425D18 BAC possesses approximately 100 kb of
sequence upstream of Snrpn, including three of the nine identified
U exons (Figure 1B). We generated two independent transgenic
lines for the 380J10 BAC, 380A and 380D. 380A contains multiple
copies of the transgene while 380D is a single copy line (Figure
S1A). We examined the imprinted status of the transgene by
Northern blot analysis on postnatal (P) day 1 brain RNA. As the
endogenous Snrpn transcript would complicate transgene expres-
sion analysis, we initially crossed our transgenic lines to a mutant
bearing a 35 kb deletion of the PWS-IC (PWS-ICD35kb) . This
deletion ablates endogenous Snrpn expression when inherited
paternally. The 380A and 380D lines were analyzed upon both
maternal and paternal transmission of the transgene in combina-
tion with a paternal PWS-ICD35kballele. In both lines, transgene-
encoded Snrpn was expressed after maternal and paternal
transmission (Figure 2A). This result demonstrated that AS-IC
activity is not contained within 15 kb upstream of Snrpn, a region
deficient in U exons.
Figure 1. The PWS/AS imprinted domain and corresponding regions covered by the BAC transgenes. (A) Schematic diagram of the
murine PWS/AS locus. Genes expressed from the paternal allele are shown above the locus (blue rectangles) and maternally expressed gene are
shown below (red rectangles)(not to scale). (B) Schematic diagram of the Snrpn locus and sequences included in the 380J10 and 425D18 BACs. The
nine identified Snrpn U exons (filled red boxes) are shown as well as five additional exons, termed a-e (open boxes) . For each BAC, the extent of
genomic sequence relative to Snrpn is indicated (not to scale).
Prader-Willi and Angelman syndromes are neurobehavioral
disorders arising from dysregulation of a cluster of
imprinted genes located at chromosome 15q11–q13.
PWS results from the absence of paternally expressed
genes and AS from the absence of maternally expressed
genes. Two elements, termed the PWS-IC and the AS-IC,
are responsible for allele-specific gene expression. The
PWS-IC activates expression of paternally expressed genes,
while the AS-IC is thought to silence the PWS-IC in the
female germ line, rendering it inactive on the future
maternal allele. Mouse models have been effective for
studying the IC-directed regulation of this locus; however,
the murine AS-IC has yet to be characterized. In this study,
we have determined the identity of the AS-IC and
investigated how it functions to inactivate the PWS-IC.
Our results suggest that the murine AS-IC consists of
several promoters that direct expression of transcripts
through the PWS-IC in oocytes. Thus, faulty transcription in
oocytes may lead to AS imprinting defects.
AS-IC Activity Requires Transcription
PLoS Genetics | www.plosgenetics.org2 December 2011 | Volume 7 | Issue 12 | e1002422
We then focused on the 425D18 BAC which contains three
Snrpn U exons. To simplify transgene expression analysis, we
utilized BAC recombineering techniques to delete a region
between Snrpn exons 5 and 7, sequence shown previously to have
no effect on imprinting at the locus . We established multiple
transgenic lines using this BAC, termed BAC 425D5-7. RT-PCR
with primers located in exons 4 and 8, the exons flanking the
deletion, distinguishes expression of endogenous Snrpn from
expression of the transgene based on amplicon size (Figure 3A).
Analysis of P1 brain RNA showed the 425D5-7 transgene was
expressed upon paternal but not maternal inheritance and thus
was correctly imprinted (Figure 3B). This result was demonstrated
in three independent lines, lines A, H, and I, indicating that AS-IC
activity lies within the 425D18 BAC sequence. We chose line
425D5-7A for further analysis as we found this line to be single
copy (Figure S1B).
We further analyzed imprinting of the 425D5-7 transgene by
examining the Snrpn differentially methylated region (DMR),
which is located within the PWS-IC. In somatic cells, this DMR is
methylated on the maternal chromosome and unmethylated on
the paternal chromosome [22–24]. This DNA methylation imprint
is erased in fetal germ cells and applied as a maternal-specific
epigenetic mark during oogenesis [25–27]. To investigate whether
Figure 3. Analysis of imprinting for the 425D5-7 BAC transgene. (A) Schematic diagram of the strategy for Snrpn expression analysis by RT-
PCR. The Snrpn gene contains ten exons (numbered boxes). The 425D5-7 transgene bears a deletion between exons 5 and 7 (dashed lines). PCR
primers (arrows) in exons flanking the deletion distinguish the endogenous Snrpn product (513 bp) from transgenic Snrpn expression (350 bp). (B)
RT-PCR analysis of P1 brain RNA from the 425D5-7A, 425D5-7H, and 425D5-7I (from left to right) transgenic lines. Non-transgenic littermates (-/-) were
included as negative controls. NT indicates a control PCR reaction with no cDNA template added. Hprt amplification was performed to demonstrate
cDNA integrity. The minus sign in RT-PCR analyses indicates control samples in which reverse transcriptase was omitted during cDNA synthesis. (C)
Analysis of the DNA methylation imprint at the 425D5-7A Snrpn DMR. Genomic bisulfite sequencing was performed on P1 brain samples after both
maternal and paternal transmission of the transgene. A 364 bp region of the Snrpn DMR (spanning from 2175 to +189 relative to Snrpn exon 1) that
contains 14 CpG dinucleotides  was analyzed in the 425D5-7A line. A SNP located within the sequenced region distinguishes the transgenic from
the endogenous alleles. Each row represents an individually sequenced clone, filled circles indicate methylated CpG dinucleotides, and open circles
represent unmethylated CpG dinucleotides. DNA methylation patterns at the endogenous locus are shown in Figure S2.
Figure 2. 380J10 BAC transgene expression analysis. Northern
blot analysis for Snrpn expression in the 380J10 transgenic lines. Two
independent P1 brain samples were analyzed for each line and parental
transmission. The PWS-ICD35kb(KO) samples demonstrate the absence
of endogenous Snrpn expression upon paternal transmission of this
allele. The 380 transgenic samples all bear a paternal PWS-ICD35kballele.
b-actin was assayed as a loading control.
AS-IC Activity Requires Transcription
PLoS Genetics | www.plosgenetics.org3 December 2011 | Volume 7 | Issue 12 | e1002422
the 425D5-7 BAC transgene undergoes appropriate epigenetic
reprogramming in the germ line, we used genomic bisulfite
sequence analysis to examine the methylation status of the Snrpn
DMR in transgenic P1 brain DNA. We bred the 425D5-7A
transgene onto a C57BL/6 line congenic for the PWS/AS domain
of Mus musculus castaneus chromosome 7 (B6.cast.c7)  and
examined a region within the DMR containing 14 CpG
dinucleotides. A single nucleotide polymorphism (SNP) located
within our region of interest distinguishes C57BL/6 transgenic
from endogenous cast sequences. As expected, the paternally
transmitted 425D5-7A transgene displayed hypomethylation at the
Snrpn DMR while the maternally transmitted transgene was
hypermethylated, demonstrating the presence of the appropriate
epigenetic imprinting marks (Figure 3C). The endogenous alleles
displayed the expected mixture of methylated and unmethylated
clones as both the maternal and paternal alleles are represented in
these sequences (Figure S2). Taken together, the expression
analysis and DNA methylation analysis illustrate that the
425D18 BAC harbors complete AS-IC activity.
Snrpn U Exon Expression Patterns Are Conserved on the
The human AS-IC lies within a region of 880 bp that contains
two of several alternative upstream exons of SNRPN, designated u5
and u6 [11,13,14,29,30]. These two exons are included in various
transcripts originating from the upstream exons, suggesting a
potential role in the imprinting process. A conserved sequence from
the human AS-IC has not been identified in the mouse but the
murine Snprn locus does possess a number of alternative U exons.
Nine have been identified to date, termed U1-U9, spanning over
580 kb upstream of Snrpn exon 1 . These exons generate
multipletranscripts,most frequentlysplicing into exon 2 ofSnrpn but
occasionally splicing downstream of the Snrpn gene. Significantly,
while Snrpn is widely expressed in adult tissues, U exon expression is
detected only in postnatal brain from the paternal allele as well as in
oocytes [15,16,18]. Expression of the U exons in oocytes correlates
withimprintestablishment inthefemale germline,consistent with a
role in establishing the maternal epigenotype.
As the 425D5-7A transgene displays appropriate imprinted
expression and epigenetic imprinting marks, we hypothesized that
U exon expression from this transgene would mimic the
endogenous U exon expression patterns. To detect U exon
transcripts arising exclusively from the transgene, we used a
primer recognizing a loxP site that remained in the BAC after
deletion of the region between Snrpn exons 5 and 7. We performed
RT-PCR using a forward primer that anneals to both U1 and U2
(referred to as U-F1) and a reverse primer in the loxP sequence
(Figure 4A). Replicating the endogenous locus, BAC transgene-
encoded transcripts including the U exons were detected in P1
brain after paternal transmission of the transgene but not after
maternal transmission (Figure S3). Importantly, these transcripts
were also detected in transgenic ovary RNA from three-week-old
females, corresponding with the timing of imprint establishment in
oocytes (Figure 4B). Thus, U exon expression from the imprinted
425D5-7A transgene does mimic the endogenous expression
Snrpn U Exon Expression in the Germ Line Correlates with
Maternal Imprint Establishment
We explored the role of Snrpn U exon transcription in directing
imprinting at the PWS/AS locus by analyzing U exon expression
in the developing germ line. Imprints are erased in the germ line in
post-migratory germ cells between 10.5 to 12.5 days post coitus
(dpc), yielding biallelic expression of Snrpn in 13.5 dpc primordial
germ cells (PGCs) [27,31–34]. Therefore, we examined 13.5 dpc
PGCs for the presence of Snrpn U exon-containing transcripts,
expecting a lack of expression since imprinting is not established
until after birth in the maternal germ line. We utilized the U-F1
primer and a reverse primer complementary to Snrpn exon 3 to
analyze expression in wild type, sex-segregated, purified 13.5 dpc
PGC cDNAs (Figure 5A). Snrpn U exon-containing transcripts
were not detected in either male or female 13.5 dpc PGCs
(Figure 5B). RT-PCR was performed on the same cDNAs using
the primer set spanning Snrpn exons 4 to 8 to demonstrate that
Snrpn is expressed in both male and female 13.5 dpc PGCs. These
data confirm that U exon transcription is not detectable in the
germ line before imprints are established.
We further tested whether U exon expression contributed to
imprinting by examining the postnatal ovary. Detecting expression
in this tissue is vital as the DNA methylation imprint is established
at the PWS-IC in growing oocytes. The Snrpn DMR methylation
imprint first appears in growing oocytes between 5–25 days
postpartum (dpp) [25,26]. Therefore, we examined ovaries from
wild type female mice to determine whether U exon transcription
was active throughout this stage of imprint establishment. Ovaries
obtained from females ranging in age from P1 to P20 were
subjected RT-PCR analysis for transcripts spanning U-F1 to Snrpn
exon 3. We detected U exon-containing transcripts in the ovary
during the entire period of maternal imprint establishment at the
PWS-IC (Figure 5C). This RT-PCR reaction revealed a larger
amplicon than expected that was preferentially expressed in the
developing ovaries. DNA sequencing of the PCR products
indicated that the larger amplicon was derived from a transcript
originating at U2 and included the Snrpn c exon whereas the
smaller product included only U1 and Snrpn exons 2 and 3.
Three Snrpn U Exons Constitute the AS-IC on the 425D18
We next sought to determine whether the Snrpn U exons are
necessary for maternal imprinting of the 425D5-7 transgene. To
do so, we used BAC recombineering to make targeted deletions of
Figure 4. U exon expression from the 425D5-7A transgene
during maternal imprint establishment. (A) Schematic diagram of
the RT-PCR strategy used to analyze U exon transcription from the
425D5-7A transgene. The loxP site, unique to the transgene exon 5 to
exon 7 deletion, is depicted as an open triangle. PCR primers (arrows)
were designed to anneal to U1 or U2 and the loxP site. (B) RT-PCR
analysis of U exon expression in ovary RNA. Two wild type (2/2) and
four transgenic (Tg) sets of ovaries were analyzed from females at three
weeks of age. Transgenes were maternally transmitted. Stella amplifi-
cation was performed to verify the presence of oocyte RNA.
AS-IC Activity Requires Transcription
PLoS Genetics | www.plosgenetics.org4 December 2011 | Volume 7 | Issue 12 | e1002422
U1, U2, and U3, the three U exons contained within the 425D5-7
transgene. For each U exon, a 2.5 to 2.7 kb region spanning the
exon was deleted to eliminate promoter function. This modified
BAC was termed 425DU1-U3 (Figure 6A). Transgene copy
number is an important element in imprinted transgene studies,
especially as a multi-copy Snrpn transgene was previously shown to
be imprinted while the same transgene in single copy was
biparentally expressed . We therefore limited our analysis to
Figure 5. U exon expression in the developing germ line. (A) Schematic diagram of the RT-PCR strategy used to analyze U exon transcription.
PCR primers (arrows) were designed to anneal to U1 or U2 and Snrpn exon 3 . (B) RT-PCR analysis of U exon expression in 13.5 dpc PGCs from
female (F) and male (M) embryos. Testis and adult ovary provide negative and positive controls, respectively. The RT-PCR reactions were Southern
blotted and probed with verified product to confirm identity. Snrpn expression was also analyzed using primers spanning exons 4 to 8. (C) U exon
expression was analyzed by RT-PCR in wild type ovaries spanning from P1 to adult. The identity of the products was confirmed by DNA sequencing.
Stella expression was examined to verify the presence of oocyte RNA.
Figure 6. Analysis of imprinting of the 425DU1-U3D BAC transgene. (A) Schematic representation of the 425DU1-U3 transgene. The three
upstream exons (gray boxes) were deleted (dashed lines) from the 425D5-7 BAC to create this transgene. PCR primers (arrows) in exons flanking the
deletion distinguish the endogenous Snrpn product (513 bp) from transgenic Snrpn expression (350 bp). (B) Expression analysis of the 425DU1-U3D
transgene. RT-PCR was performed on P1 brain RNA after both maternal and paternal transmission of the transgene. (C) Analysis of the DNA
methylation imprint at the 425DU1-U3D Snrpn DMR. Genomic bisulfite sequencing was performed on P1 brain samples after both maternal and
paternal transmission of the transgene. A SNP located within the sequenced region distinguishes the transgenic from the endogenous alleles, only
the transgenic alleles are shown.
AS-IC Activity Requires Transcription
PLoS Genetics | www.plosgenetics.org5 December 2011 | Volume 7 | Issue 12 | e1002422
single copy 425DU1-U3 transgenic lines. In addition to providing
the most stringent test of imprinting, single copy transgene lines
eliminate uncertainty about which copy(s) of the transgene is being
expressed and allow for accurate determination of DNA
methylation status. The single copy line 425DU1-U3D was used
for our initial analysis (Figure S1B). Unlike the 425D5-7 transgene,
the 425DU1-U3D transgene was expressed in P1 brain upon both
paternal and maternal inheritance, indicating an absence of AS-IC
activity correlating with the deletion of the upstream exons
(Figure 6B). We also examined the methylation imprint at the
transgene Snrpn DMR and found that it was hypomethylated after
both paternal and maternal transmission, demonstrating an
absence of the maternal epigenetic imprint (Figure 6C). These
experiments show that the 425D5-7 transgene loses its imprinted
expression pattern as well as its epigenetic imprint upon removal
of the three U exons contained within it, providing further
evidence that these U exons play a major role in AS-IC function.
Imprinting Is Established in the Absence of the U Exons
when Transcription through the PWS-IC Is Activated
We obtained one additional single copy line for the 425DU1-U3
transgene, 425DU1-U3H (Figure S1B). Unexpectedly, this trans-
gene displayed imprinted expression of Snrpn (Figure 7B). We
initially questioned whether critical sequences in the BAC had
been rearranged upon integration but found that sequences for all
single copy BAC transgenes were intact between U1 and the PWS-
IC (Figure S4). Since the insertion sites for the transgenes in our
study are unknown, we questioned whether transcripts initiated
from a promoter upstream of the 425DU1-U3H transgene
insertion site might proceed through the PWS-IC in the maternal
germ line. We reasoned that this adventitious transcription could
create an artificial imprint at the PWS-IC. To investigate this
hypothesis, we performed RT-PCR on ovaries harvested from P15
females, including the imprinted 425DU1-U3H and 425D5-7A
lines as well as the non-imprinted 425DU1-U3D line. As the
425DU1-U3 transgenes do not have the U exons intact, we used
sequence from the Snrpn c exon to generate our forward primer for
RT-PCR since this exon is included in U exon transcripts
generated during oocyte growth (Figure 5C). Unlike U1-U9, the
Greek-lettered exons shown in Figure 1B have both splice acceptor
signals in addition to splice donor signals . We reasoned that
the c exon could be included in transcripts initiated from upstream
of the transgene insertion site. As before, the reverse primer in the
loxP sequence was used so as to avoid any signal from the
endogenous allele. Figure 7C demonstrates that ovaries from the
imprinted 425DU1-U3H line expressed Snrpn c exon-containing
transcripts that traversed the PWS-IC. These transcripts were also
observed in the imprinted 425D5-7A line but not in the
biparentally expressed 425DU1-U3D line. Thus, transcription
through the PWS-IC in the ovary correlates with the establishment
of the maternal imprint.
Roughly 4% of individuals with AS have an imprinting defect,
characterized by a paternal epigenotype at the PWS/AS locus on
the maternally inherited chromosome. Of these individuals, 10–
15% have a microdeletion that includes the AS-IC . The AS-
IC is thought to operate in the female germ line to epigenetically
inactivate the PWS-IC and thereby silence the paternally
expressed genes on the future maternal allele . Among these
maternally silenced genes is UBE3A-ATS, a transcript antisense to
UBE3A. UBE3A-ATS is thought to silence UBE3A expression by
an unknown mechanism [37–39]. In individuals lacking AS-IC
function, UBE3A-ATS is expressed from the maternal allele in
addition to the paternal allele and the consequent lack of UBE3A
results in AS. Hypomethylation of the maternal SNRPN DMR is
diagnostic of an AS imprinting defect .
Figure 7. Expression analysis of the imprinted 425DU1-U3H transgene. (A) Schematic representation of the 425DU1-U3 transgene. PCR
primers (black arrows) in exons flanking the deletion distinguish the endogenous Snrpn product (513 bp) from transgenic Snrpn expression (350 bp).
(B) Expression analysis of the 425DU1-U3H transgene. RT-PCR analysis of P1 brain RNA after both maternal and paternal transmission of the
transgene. (C) Analysis of transcription across the PWS-IC in transgenic ovaries during imprint establishment. P15 ovaries were isolated from 425D5-
7A, 425DU1-U3D, and 425DU1-U3H females and analyzed by RT-PCR for Snrpn c exon-containing transcripts. Primers anneal to the Snrpn c exon and
the loxP site (gray arrows in (A)). The RT-PCR product was Southern blotted and probed with verified product to confirm identity. Stella expression
was examined to demonstrate the presence of oocyte RNA.
AS-IC Activity Requires Transcription
PLoS Genetics | www.plosgenetics.org6 December 2011 | Volume 7 | Issue 12 | e1002422
The AS-IC functions in the female germ line, complicating
investigations into imprinting mechanisms at the PWS/AS locus.
Deletion analysis mapped the human AS-IC to 35 kb upstream of
the PWS-IC but deletion of similarly positioned sequences in the
mouse had no detectable consequence . Here we used BAC
transgenes to identify murine AS-IC activity. We found that
15 kb of sequence upstream of Snrpn exon 1 was insufficient to
direct imprinted expression of Snrpn. Transgenes containing
100 kb of sequence 59 to Snrpn exon 1 directed the appropriate
imprinted expression of Snrpn. Furthermore, this sequence was
sufficient to direct appropriate hypermethylation of the Snrpn
DMR following maternal but not paternal transmission. We
conclude that an AS-IC activity is contained within 15 to 100 kb
of the PWS-IC. This region contains three alternative exons, at
least two of which are expressed in oocytes and splice into Snrpn
exon 2. Deletion of these U exons from the BAC transgene
resulted in biparental Snrpn expression and DNA hypomethyla-
tion at the PWS-IC, characteristic of an imprinting defect.
Surprisingly, a second single copy line lacking the U exons
displayed appropriate maternal silencing. In contrast to the
biparentally expressed transgene but similar to imprinted
transgenes with the U exons intact, this line exhibited
transcription through the PWS-IC, consistent with the notion
that transcription is an essential element of AS-IC activity. Lastly,
we demonstrated that U exon transcription is undetectable in
fetal oogonia as they are entering meiotic prophase, a time when
imprinting is absent, but becomes evident after birth and prior to
the application of the DNA methylation imprint at the PWS-IC.
We conclude that the U exons exhibit AS-IC activity as
previously proposed [8,15].
Snrpn transcription becomes biallelic in PGCs as they colonize
the gonad at midgestation . This stage of imprinting erasure in
the germ line coincides with the removal of the DNA methylation
imprint at the PWS-IC . The appearance of the DNA
methylation imprint is tied to postnatal oocyte growth [25,26].
Our data support a model in which growing oocytes direct
transcription from the U exons through the PWS-IC. This
transcription is necessary to epigenetically modify the PWS-IC
leading to the maternal imprint (Figure 8). Importantly, our results
do not distinguish whether the RNA transcript or the act of
transcription is essential for AS-IC activity. However, since
transcription presumably arising from sequences flanking the
BAC insertion site appears to be sufficient for imprinting in the
425DU1-U3H line, we suggest that the U exon sequences per se are
Our data support a model in which U exon promoter activity
supplies AS-IC function in the mouse. The human AS-IC,
however, includes two upstream exons that are internal to
transcripts encoded in brain RNA. If the human AS-IC does not
possess promoter activity in oocytes, we speculate that the AS-IC
may serve an RNA antitermination or stabilization role.
In both human and mouse, splicing of the upstream exons to
Snrpn exon 2 can encode the Snurf polypeptide [17,41]. It is
unknown whether Snurf is expressed in oocytes but the absence of
an overt phenotype in Snurf knockout mice indicates that it is not
involved in the imprinting process .
Our BAC transgene system allows simultaneous assessment of
both the transgene and endogenous locus. U exon transcription
arising from the endogenous locus does not impose imprinting
upon a maternally transmitted transgene harboring a PWS-IC but
Figure 8. A working model for the establishment of the maternal imprint at the PWS-IC. Schematic diagrams of the locus around Snrpn at
various stages of development. The U exons are indicated by red boxes and the PWS-IC is depicted as an orange oval overlying Snrpn exon 1. Imprints
are erased in fetal germ cells and the Snrpn upstream exons are not expressed. In the neonatal period, transcription is initiated from the Snrpn
upstream exons prior to the appearance of the DNA methylation imprint in growing oocytes. These transcripts proceed through the PWS-IC,
triggering repressive epigenetic modification of this element, indicated by black lollipops. The PWS-IC on the future maternal allele is thereby
inactivated, rendering the paternally expressed genes silent.
AS-IC Activity Requires Transcription
PLoS Genetics | www.plosgenetics.org7 December 2011 | Volume 7 | Issue 12 | e1002422
lacking U exons (Figure 2 and Figure 6B). This observation
suggests that if the RNA transcript is an essential element of the
imprinting machinery, then it does not function in trans.
Wu et al.  described two mutants with a maternal imprinting
defect. Maternal transmission of an 80 kb deletion spanning from
10 to 90 kb upstream of Snrpn exon 1 resulted in an extensive loss
of DNA methylation at the PWS-IC in only three of eight
offspring. This deletion removes only U1 and U3, leaving the
remaining seven U exons intact. Our model suggests that extensive
imprinting defects were not observed in the remaining offspring
due to compensatory transcription arising from the remaining U
exons. The second mutant described by these investigators
exhibited a highly penetrant imprinting defect leading to
expression of paternal genes from the maternal allele and reduced
expression of the Ube3a gene product, E6AP. This mutant carried
an insertion 10 kb upstream of Snrpn exon 1 that was comprised of
a puromycin resistance cassette in the opposite transcriptional
orientation to Snrpn, exons 3–9 of an Hprt minigene, and a
duplication of 6 kb of endogenous sequence. Our findings support
a model in which the insertion interferes with transcription from
the U exons in oocytes.
The imprinting activity of the human AS-IC element has been
tested in mice. Five independent lines of a human P1 phage
transgene that included both the human AS-IC and PWS-IC
failed to show evidence of imprinted gene expression in the brain
. This lack of imprinting may be explained by our more recent
observation that a human PWS-IC substituted for the endogenous
murine PWS-IC can acquire a DNA methylation imprint in
oocytes but this imprint is subsequently lost during development
. Shemer et al.  characterized transgenes containing the
human AS-IC linked to the murine PWS-IC and observed
preferential Snrpn expression following paternal inheritance.
Transcription through the PWS-IC during oogenesis was not
examined in these transgenes but our model suggests that the
partial imprinting seen in these transgenes resulted from
transcription of the human upstream exons.
Zogel et al.  reported that a maternally inherited
polymorphism at the AS-IC increases the risk of imprinting
defects leading to AS. Our observations strongly support
these investigators9 hypothesis that this polymorphism may affect
a transcription factor recognition site. Our model suggests that
this polymorphism affects transcription through the PWS-IC
during oogenesis, decreasing the efficiency of epigenetic modifi-
cation at the PWS-IC and thereby raising the risk of AS imprinting
Transcription is implicated in allelic silencing at several
imprinted gene clusters (See Peters and Robson  for a review).
Our results at the PWS/AS locus are consistent with the
observation that in growing oocytes, transcripts traverse maternal
germ line DMRs at eight locations, suggesting that transcription is
commonly associated with the establishment of maternal imprints
. A recent report indicates that 35% of CpG islands that are
methylated in oocytes are downstream of active promoters .
The role of transcription through a germ line DMR in establishing
a DNA methylation imprint in oocytes has been tested at the Gnas
locus . At this locus, Nesp transcription is active in oocytes and
traverses two maternal DMRs. Similar to the U transcripts at the
PWS/AS locus, appearance of the Nesp transcript at the Gnas locus
temporally precedes the establishment of the maternal methylation
imprint in growing oocytes. Furthermore, truncation of the Nesp
transcript led to a loss of the DNA methylation imprint at several
maternal DMRs within the Gnas locus and activation of normally
silent maternal transcripts. Chotalia et al.  suggested several
mechanisms by which transcription might contribute to DNA
methylation including an interaction of the transcription complex
with the methylation machinery, creating an open chromatin
environment allowing DNA methylation machinery access to the
locus, or recruiting factors necessary for epigenetic modification.
These same mechanisms could operate to set the methylation
imprint at the PWS/AS locus. Similar to the Gnas locus, truncation
of the U exon transcripts upstream of the PWS-IC may confirm
their role in establishing imprinting throughout the PWS/AS
Materials and Methods
All animal procedures in this work were reviewed and approved
by the University of Florida Institutional Animal Care and Use
BACs 380J10 and 425D18 were originally identified by
application of a Snrpn probe to the RPCI-23 library prepared
from C57BL/6 DNA (Research Genetics). BAC 380J10 is
232.3 kb and contains chromosome 7 nucleotides (nt) 67165323
to 66935240. This region includes 15.3 kb upstream of the major
Snrpn transcription start site to 192.1 kb 39 sequence containing
nearly the entire complement of Snord116 repeats. BAC 425D18 is
192.3 kb and contains sequences from nt 67250191 to 67057909.
This span includes 100.1 kb upstream of the major Snrpn
transcription start site and 69.5 kb sequence 39 to the Snrpn
Recombineering procedures were performed in bacterial
strains DY380, EL250 and EL350 as described . First, a
loxP site in the BACe3.6 vector was replaced with a blasticidin
resistance cassette. To create a reporter gene in the BAC, a
deletion spanning Snrpn exons 5 to 7 was engineered by
homologous recombination with a floxed kanamycin (kan) resis-
tance cassette followed by Cre-mediated removal of the kan
resistance cassette in EL350 cells. This deletion removes
sequences previously shown to be unimportant for imprinting
. The three U exons in this modified BAC were deleted
individually by first subcloning a 4.6 to 5.0 kb region surrounding
each U exon and subsequently inserting a kan resistance cassette
into each clone. These clones were then used to recombineer the
modified 425D18 BAC. The U1 deletion removes 984 bp 59 to
1576 bp 39 of U1. The U2 deletion removes sequences from
1032 bp 59 to 1296 bp 39 of U2. The U3 deletion removes
sequences from 834 bp 59 to 1575 bp 39 of U3. In each case, the
kan resistance cassette was removed by recombineering. Recom-
bination between lox sites at different U exons was prevented by
using kan resistance cassettes flanked by FRT sites at U1 and by
incompatible lox sequences at U2 and U3 . Transfer of each
deletion to the BAC was confirmed by Southern blot analysis and
C57BL/6 BAC DNA was purified by CsCl gradient centrifu-
gation and supercoiled DNA was injected into fertilized FVB/N
oocytes. Strains were maintained on an FVB background except
where noted for the DNA methylation studies.
Southern Blot Analysis
Genomic DNA was digested with restriction endonucleases as
indicated and analyzed by Southern blot. The 380J10 copy
number blot was probed with a 770 bp EcoRI fragment
approximately 7.8 kb upstream of Snrpn exon 1. The 425D18
AS-IC Activity Requires Transcription
PLoS Genetics | www.plosgenetics.org8 December 2011 | Volume 7 | Issue 12 | e1002422
copy number blot was probed with a 291 bp EcoRI fragment
located 59 to the deletion between Snrpn exons 5 and 7. Relative
band intensity was digitized on the Storm 860 PhosphorImager
(GE Healthcare) and analyzed with Image Quant TL software for
transgene copy number analyses. Probes for Southern blots
demonstrating the integrity of the 425D18 derived BAC
transgenes were generated from PCR products (H, E,L) (see
Dataset S1 for primer sequences) or restriction fragments (304:
800 bp PstI fragment, 210: 770 bp EcoRI fragment).
Analysis of Gene Expression
RNA was isolated from the indicated tissues by homogenization
with RNA-Bee reagent (Tel-Test, Inc.) following manufacturer’s
instructions. RNA was subject to Northern blot analysis or DNase
treated and subsequently reverse-transcribed with Superscript II
(Invitrogen). PCR reactions were performed under standard
conditions. Primer sequences listed in Dataset S1. The Snrpn
Northern blot probe was generated from a 480 bp ClaI-PstI
DNA Methylation Analysis
Bisulfite sequence analysis was performed on whole brain
genomic DNA. DNA was subject to bisulfite conversion as
previously described . PCR amplification of the Snrpn DMR
was performed on the bisulfite converted DNA using the primer
set W18-F/W19-R  (Dataset S1). An initial 15 minute
denaturation step was followed by 38 cycles of 94uC for 45 s,
54uC for 60 s and 72uC for 90 s with a 10 minute final extension
at 72uC. Amplified products were cloned into the pGEM-T Easy
Vector System (Promega). Plasmid sequencing was performed with
ABI Prism BigDye terminator (PerkinElmer) at the UF Center for
were used for RT-PCR gene expression analysis, genomic bisulfite
sequencing analysis, or Southern or Northern blot probe
PCR primer set sequences. The primers listed above
assessed by Southern blot analysis on genomic brain DNA following
digestion with the indicated restriction endonucleases. Two samples
from each transgenic line were analyzed and the ratio of the average
transgenic to endogenous allele band intensity is displayed below the
Genomic DNAs were digested with PstI and hybridized with a probe
for the 59 end of the transgene. The endogenous allele generates a
17.8 kb fragment while the transgene produces a 12.5 kb fragment.
(B) Copy number Southern blot for the 425D18 transgenic lines.
Genomic DNAs were digested with SpeI and PvuII and hybridized
with a probe just 59 to the deletion between Snrpn exons 5 and 7. The
endogenous allele generates a 4.5 kb fragment while the transgene
produces a 3.0 kb fragment.
Transgene copy number analysis. Copy number was
Genomic bisulfite sequencing was performed on P1 brain samples
after both maternal and paternal transmission of the 425D5-7A
transgene. The methylation status of the endogenous alleles is
represented here. Each row represents an individually sequenced
clone. Filled circles indicate methylated CpG dinucleotides and
white circles represent unmethylated CpG dinucleotides.
DNA methylation analysis of the Snrpn DMR.
brain. (A) Schematic diagram of the RT-PCR strategy used to
analyze U exon transcription from the 425D5-7A transgene. The
loxP site is depicted as a white triangle. PCR primers (arrows) were
designed to anneal to U1 or U2 and the loxP site. (B) RT-PCR
analysis of U exon expression was performed on P1 brain RNA
after maternal and paternal transmission of the transgene. Hprt
amplification was performed to demonstrate cDNA integrity.
U exon usage from the 425D5-7A transgene in the
intact and not rearranged. The topmost part of the figure shows
100 kb upstream of Snrpn exon 1. Black rectangles indicate the
three U exons present in the 425D18 BAC as well as Snrpn exon 1.
The locations of five probes, termed H, E, L, 304, and 210, are
shown in red (arrows). Horizontal lines indicate the lengths and
locations of expected fragments for each restriction endonuclease
and probe combination. Below each line is the expected size of
that fragment, first from the BAC containing U exons, 425D5-7,
followed by the BAC from which the three U exons have been
deleted, 425DU1-U3. Restriction endonuclease digests of either
transgenic genomic DNA (lanes 1–3) or purified BAC DNA (lanes
4 & 5) were analyzed by Southern blot. Bands differing from the
endogenous alleles are detectable only for fragments that include
recombineered deletions of the U exons, indicating the absence of
rearrangements within that region. Fragments overlap the entire
region with the exception of a 1.4 kb span between the SpeI
fragment recognized by probe E and the PstI fragment recognized
by probe L. A non-repetitive probe for this short region could not
be identified. Lane 1: 425D5-7A genomic DNA, Lane 2: 425DU1-
U3D genomic DNA, Lane 3: 425DU1-U3H genomic DNA, Lane
4: 425D5-7 BAC DNA, Lane 5: 425DU1-U3 BAC DNA.
The single copy 425D18 derived BAC transgenes are
We thank Dr. Dan Driscoll and Dr. Bernhard Horsthemke for critical
reading of the manuscript and Dr. Susan D’Costa for help with data
analysis. We also thank Ryan Hallett and Deb Morse for technical
assistance, Dr. Jingda Shi and the UF Center for Epigenetics for
sequencing, and Ryan Fiske of the UF Mouse Models Core.
Conceived and designed the experiments: KAJ JLR. Performed the
experiments: EYS CRF SJC. Analyzed the data: EYS CRF JLR.
Contributed reagents/materials/analysis tools: EYS CRF SJC KAJ JLR.
Wrote the paper: EYS JLR.
1. Lewis A, Reik W (2006) How imprinting centres work. Cytogenet Genome Res
2. Razin A, Cedar H (1994) DNA Methylation and Genomic Imprinting. Cell 77:
3. Margueron R, Trojer P, Reinberg D (2005) The key to development:
interpreting the histone code? Curr Opin Genet Dev 15: 163–176.
4. Kitsberg D, Selig S, Brandeis M, Simon I, Keshet I, et al. (1993) Allele-specific
replication timing of imprinted gene regions. Nature 364: 459–463.
5. Kishino T, Lalande M, Wagstaff J (1997) UBE3A/E6-AP mutations cause
Angelman syndrome. Nature Genetics 15: 70–73.
6. Matsuura T, Sutcliffe JS, Fang P, Galjaard R-J, Jiang Y-H, et al. (1997) De novo
truncating mutations in E6-AP ubiquitin-protein ligase gene (UBE3A) in
Angelman syndrome. Nature Genetics 15: 74–77.
7. Buiting K, Saitoh S, Gross S, Dittrich B, Schwartz S, et al. (1995) Inherited
microdeletions in the Angelman and Prader-Willi syndromes define an
imprinting centre on human chromosome 15. Nat Genet 9: 395–400.
AS-IC Activity Requires Transcription
PLoS Genetics | www.plosgenetics.org9 December 2011 | Volume 7 | Issue 12 | e1002422
8. Dittrich B, Buiting K, Korn B, Rickard S, Buxton J, et al. (1996) Imprint Download full-text
switching on human chromosome 15 may involve alternative transcripts of the
SNRPN gene. Nat Genet 14: 163–170.
9. Brannan CI, Bartolomei MS (1999) Mechanisms of genomic imprinting.
Current Opinion in Genetics and Development 9: 164–170.
10. Shemer R, Hershko AY, Perk J, Mostoslavsky R, Tsuberi B, et al. (2000) The
imprinting box of the Prader-Willi/Angelman syndrome domain. Nature
Genetics 26: 440–443.
11. Buiting K, Lich C, Cottrell S, Barnicoat A, Horsthemke B (1999) A 5-kb
imprinting center deletion in a family with Angelman syndrome reduces the
shortest region of deletion overlap to 880 bp. Human Genetics 105: 665–666.
12. Ohta T, Gray TA, Rogan PK, Buiting K, Gabriel JM, et al. (1999) Imprinting-
mutation mechanisms in Prader-Willi syndrome. American Journal of Human
Genetics 64: 397–413.
13. Farber C, Dittrich B, Buiting K, Horsthemke B (1999) The chromosome 15
imprinting centre (IC) region has undergone multiple duplication events and
contains an upstream exon of SNRPN that is deleted in all Angelman syndrome
patients with an IC microdeletion. Hum Mol Genet 8: 337–343.
14. Wawrzik M, Spiess AN, Herrmann R, Buiting K, Horsthemke B (2009)
Expression of SNURF-SNRPN upstream transcripts and epigenetic regulatory
genes during human spermatogenesis. Eur J Hum Genet 17: 1463–1470.
15. Mapendano CK, Kishino T, Miyazaki K, Kondo S, Yoshiura K, et al. (2006)
Expression of the Snurf-Snrpn IC transcript in the oocyte and its putative role in
the imprinting establishment of the mouse 7C imprinting domain. J Hum Genet
16. Bressler J, Tsai TF, Wu MY, Tsai SF, Ramirez MA, et al. (2001) The SNRPN
promoter is not required for genomic imprinting of the Prader- Willi/Angelman
domain in mice. Nature Genetics 28: 232–240.
17. Landers M, Bancescu DL, Le Meur E, Rougeulle C, Glatt-Deeley H, et al.
(2004) Regulation of the large (,1000 kb) imprinted murine Ube3a antisense
transcript by alternative exons upstream of Snurf/Snrpn. Nucleic Acids Res 32:
18. Le Meur E, Watrin F, Landers M, Sturny R, Lalande M, et al. (2005) Dynamic
developmental regulation of the large non-coding RNA associated with the
mouse 7C imprinted chromosomal region. Dev Biol 286: 587–600.
19. Peery EG, Elmore MD, Resnick JL, Brannan CI, Johnstone KA (2007) A
targeted deletion upstream of Snrpn does not result in an imprinting defect.
Mamm Genome 18: 255–262.
20. Wu MY, Chen KS, Bressler J, Hou A, Tsai TF, et al. (2006) Mouse imprinting
defect mutations that model Angelman syndrome. Genesis 44: 12–22.
21. Yang T, Adamson TE, Resnick JL, Leff S, Wevrick R, et al. (1998) A mouse
model for Prader-Willi syndrome imprinting-centre mutations. Nature Genetics
22. Sutcliffe JS, Nakao M, Christian S, Orstavik KH, Tommerup N, et al. (1994)
Deletions of a differentially methylated CpG island at the SNRPN gene define a
putative imprinting control region. Nat Genet 8: 52–58.
23. Gabriel JM, Gray TA, Stubbs L, Saitoh S, Ohta T, et al. (1998) Structure and
function correlations at the imprinted mouse Snrpn locus. Mamm Genome 9:
24. Shemer R, Birger Y, Riggs AD, Razin A (1997) Structure of the imprinted
mouse Snrpn gene and establishment of its parental-specific methylation pattern.
Proc Natl Acad Sci U S A 94: 10267–10272.
25. Hiura H, Obata Y, Komiyama J, Shirai M, Kono T (2006) Oocyte growth-
dependent progression of maternal imprinting in mice. Genes Cells 11: 353–361.
26. Lucifero D, Mann MR, Bartolomei MS, Trasler JM (2004) Gene-specific timing
and epigenetic memory in oocyte imprinting. Hum Mol Genet 13: 839–849.
27. Hajkova P, Erhardt S, Lane N, Haaf T, El-Maarri O, et al. (2002) Epigenetic
reprogramming in mouse primordial germ cells. Mech Dev 117: 15–23.
28. Wakeland EK, Morel L, Achey K, Yui M, Longmate J (1997) Speed congenics:
A classic technique in the fast lane (relatively speaking). Immunology Today 18:
29. Ohta T, Buiting K, Kokkonen H, McCandless S, Heeger S, et al. (1999)
Molecular mechanism of angelman syndrome in two large families involves an
imprinting mutation. Am J Hum Genet 64: 385–396.
30. Saitoh S, Buiting K, Rogan PK, Buxton JL, Driscoll DJ, et al. (1996) Minimal
definition of the imprinting center and fixation of chromosome 15q11–q13
epigenotype by imprinting mutations. Proc Natl Acad Sci U S A 93: 7811–7815.
31. Szabo PE, Mann JR (1995) Biallelic expression of imprinted genes in the mouse
germ line: implications for erasure, establishment, and mechanisms of genomic
imprinting. Genes Dev 9: 1857–1868.
32. Szabo PE, Hubner K, Scholer H, Mann JR (2002) Allele-specific expression of
imprinted genes in mouse migratory primordial germ cells. Mech Dev 115:
33. Obata Y, Kono T (2002) Maternal Primary Imprinting Is Established at a
Specific Time for Each Gene throughout Oocyte Growth. J Biol Chem 277:
34. Yamazaki Y, Mann MR, Lee SS, Marh J, McCarrey JR, et al. (2003)
Reprogramming of primordial germ cells begins before migration into the
genital ridge, making these cells inadequate donors for reproductive cloning.
Proc Natl Acad Sci U S A 100: 12207–12212.
35. Blaydes SM, Elmore M, Yang T, Brannan CI (1999) Analysis of murine Snrpn
and human SNRPN gene imprinting in transgenic mice. Mammalian Genome
36. Buiting K (2010) Prader-Willi syndrome and Angelman syndrome. Am J Med
Genet C Semin Med Genet 154C: 365–376.
37. Chamberlain SJ, Brannan CI (2001) The Prader-Willi syndrome imprinting
center activates the paternally expressed murine Ube3a antisense transcript but
represses paternal Ube3a. Genomics 73: 316–322.
38. Johnstone KA, DuBose AJ, Futtner CR, Elmore MD, Brannan CI, et al. (2006)
A human imprinting centre demonstrates conserved acquisition but diverged
maintenance of imprinting in a mouse model for Angelman syndrome
imprinting defects. Hum Mol Genet 15: 393–404.
39. Rougeulle C, Cardoso C, Fontes M, Colleaux L, Lalande M (1998) An
imprinted antisense RNA overlaps UBE3A and a second maternally expressed
transcript. Nat Genet 19: 15–16.
40. Glenn CC, Saitoh S, Jong MT, Filbrandt MM, Surti U, et al. (1996) Gene
structure, DNA methylation, and imprinted expression of the human SNRPN
gene. Am J Hum Genet 58: 335–346.
41. Gray TA, Saitoh S, Nicholls RD (1999) An imprinted, mammalian bicistronic
transcript encodes two independent proteins. Proc Natl Acad Sci U S A 96:
42. Tsai TF, Jiang YH, Bressler J, Armstrong D, Beaudet AL (1999) Paternal
deletion from Snrpn to Ube3a in the mouse causes hypotonia, growth retardation
and partial lethality and provides evidence for a gene contributing to Prader-
Willi syndrome. Human Molecular Genetics 8: 1357–1364.
43. Zogel C, Bohringer S, Gross S, Varon R, Buiting K, et al. (2006) Identification
of cis- and trans-acting factors possibly modifying the risk of epimutations on
chromosome 15. Eur J Hum Genet 14: 752–758.
44. Peters J, Robson JE (2008) Imprinted noncoding RNAs. Mamm Genome 19:
45. Chotalia M, Smallwood SA, Ruf N, Dawson C, Lucifero D, et al. (2009)
Transcription is required for establishment of germline methylation marks at
imprinted genes. Genes Dev 23: 105–117.
46. Smallwood SA, Tomizawa S, Krueger F, Ruf N, Carli N, et al. (2011) Dynamic
CpG island methylation landscape in oocytes and preimplantation embryos. Nat
Genet 43: 811–814.
47. Lee EC, Yu D, Martinez de Velasco J, Tessarollo L, Swing DA, et al. (2001) A
highly efficient Escherichia coli-based chromosome engineering system adapted
for recombinogenic targeting and subcloning of BAC DNA. Genomics 73:
48. Siegel RW, Jain R, Bradbury A (2001) Using an in vivo phagemid system to
identify non-compatible loxP sequences. FEBS Lett 505: 467–473.
49. Clark SJ, Harrison J, Paul CL, Frommer M (1994) High sensitivity mapping of
methylated cytosines. Nucleic Acids Res 22: 2990–2997.
AS-IC Activity Requires Transcription
PLoS Genetics | www.plosgenetics.org10 December 2011 | Volume 7 | Issue 12 | e1002422