Neurons but not glial cells show reciprocal
imprinting of sense and antisense
transcripts of Ube3a
K. Yamasaki1,2, K. Joh3, T. Ohta4, H. Masuzaki2, T. Ishimaru2, T. Mukai3, N. Niikawa1,
M. Ogawa5, J. Wagstaff6and T. Kishino4,*
1Department of Human Genetics and2Department of Obstetrics and Gynecology, School of Medicine, Nagasaki
University, Nagasaki, Japan,3Department of Biochemistry, Saga Medical School, Saga, Japan,4Gene Research
Center, Nagasaki University, Nagasaki, Japan,5Laboratory for Cell Culture Development, Brain Science Institute,
RIKEN, Saitama, Japan and6Department of Pediatrics and Department of Biochemistry and Molecular Genetics,
University of Virginia School of Medicine, VA, USA
Received December 2, 2002; Revised February 7, 2003; Accepted February 14, 2003
The human UBE3A gene shows brain-specific partial imprinting, and lack of a maternally inherited allele
causes Angelman syndrome (AS), which is characterized by neurobehavioral anomalies. In several AS model
mice, imprinted Ube3a expression is detected predominantly in the hippocampus, cerebellar Purkinje cells
and the olfactory bulb. Therefore, imprinting of mouse Ube3a is thought to be region-specific with different
levels of silencing of the paternal Ube3a allele in different brain regions. To determine cell types of imprinted
Ube3a expression, we analyzed its imprinting status in embryonic brain cells by using primary cortical cell
cultures. RT–PCR and immunofluorescence were performed to determine the allelic expression of the gene.
The Ube3a gene encodes two RNA transcripts in the brain, sense and antisense. The sense transcript was
expressed maternally in neurons but biallelically in glial cells in the embryonic brain, whereas the antisense
transcript was expressed only in neurons and only from the paternal allele. Our data present evidence of
brain cell type-specific imprinting, i.e. neuron-specific imprinting of Ube3a in primary brain cell cultures.
Reciprocal imprinting of sense and antisense transcripts present only in neurons suggests that the neuron-
specific imprinting mechanism is related to the lineage determination of neural stem cells.
Genomic imprinting is a mode of gene regulation causing genetic
Most genes are expressed equally from both parental alleles,
whereas imprinted genes are expressed exclusively or preferen-
expressed is dependent upon parental inheritance because of a
differential epigenetic marking that occurs during gametogenesis.
To date, there is a growing number of genes for which imprinted
expression is either tissue-specific, specific to developmental
stage, species-specific, promoter-specific or partial. Imprinted
genes are recognized to play important roles in a considerable
The human UBE3A gene is such an imprinted gene that is
implicated in a human congenital syndrome, Angelman
syndrome (AS) (MIM 105830), characterized by severe neuro-
logic abnormalities and distinctive behavior. UBE3A was
originally identified as a cellular protein that mediates the
interaction of the human papillomavirus (HPV) E6 oncoprotein
with p53 (4). UBE3A is a member of a class of functionally
related E3 ubiquitin–protein ligases defined by a carboxy-
terminal ‘hect (homologous to the E6-AP carboxyl terminus)
the substrate specificity of ubiquitin transfer and in directly
catalyzing ubiquitin transfer to substrates (6). We and others
(7,8) found UBE3A mutations in patients with AS, a disorder
that can also be caused by maternal deletion of 15q11–q13 (9),
paternal uniparental disomy (UPD) of chromosome 15 (10) or
an ‘imprinting defect’ (11) that changes parent-specific patterns
of epigenetic modification and gene expression in 15q11–q13.
AS shows an imprinted mode of inheritance, consistent with a
*To whom correspondence should be addressed at: Nagasaki University Gene Research Center, 1-12-4 Sakamoto-machi, Nagasaki, Japan.
Tel: þ81 958497120; Fax: þ81 958497121; Email: firstname.lastname@example.org
Human Molecular Genetics, 2003, Vol. 12, No. 8
Human Molecular Genetics, Vol. 12, No. 8 # Oxford University Press 2003; all rights reserved
by guest on June 5, 2013
gene active exclusively or preferentially on the maternal
chromosome 15. The fact that the absence of a functional
maternal copy of UBE3A causes AS, whose phenotype is
restricted to neurobehavioral anomalies, indirectly suggests
brain-specific imprinting of UBE3A. Although expression of
UBE3A was initially shown not to be imprinted in cultured
human fibroblasts and lymphoblasts (12), direct evidence of
brain-specific imprinting of UBE3Awas demonstrated in human
fetal brains and AS deletion brains (13,14). RT–PCR analysis
revealed that, although UBE3A is not imprinted in most human
tissues, it is imprinted in the brain, with preferential but not
exclusive expression of the maternal allele. It remains unclear
whether imprinting is brain cell type-specific or region-specific.
In several AS model mice, imprinted Ube3a expression in the
brain has been analyzed histologically (15–17). Albrecht et al.
(15) used in situ hybridization in mice with paternal UPD of the
region of mouse chromosome 7 containing Ube3a to demon-
strate very low levels of Ube3a transcript in the hippocampus,
cerebellar Purkinje cells and the olfactory bulb compared with
wild-type mice. Ube3a maternal-deficient knockout mice
reported by us (17) and others (16) showed a number of subtle
neurologic abnormalities consistent with those of AS humans.
No histopathological abnormalities were seen in Ube3a
maternal-deficient mice; however, they revealed dramatically
reduced Ube3a expression in the hippocampus and dentate
gyrus compared with those of Ube3a paternal-deficient mice
(16,17). These data suggest that imprinting of Ube3a is
restricted to some brain regions, where its inappropriate
expression may cause neurological abnormalities. To determine
the mechanism leading to this brain-specific and region-specific
imprinting, identification of the brain cell type(s) with imprinted
expression of Ube3a is necessary, because previous data for
Ube3a imprinting in the mouse brain was based on in situ
hybridization, which did not permit sufficiently high resolution
to show imprinting status at the single-cell level.
To characterize imprinting status of Ube3a in brain cells, we
performed primary brain cell culture from E15 (embryonic
day 15) products of Ube3a knock-out mice (17) and E15–17
products from reciprocal crosses between the C57BL/6 and
PWK strains (divergent strains of Mus musculus). In vitro
primary cultures have been useful tools for study of neurons
and glial cells. Methods have been developed for primary
culture using completely defined serum-free media that
optimize survival and growth of either neurons or glial cells
(18). In the present paper, neurons and glial cells were
separately cultured in the primary culture system and
quantitated by immunofluorescent staining (IF) and RT–PCR.
In addition to the expression analysis of Ube3a-deficient mice
by IF, RNA from the cultured neurons or glial cells from
Ube3a-deficient mice and reciprocal crosses between C57BL/6
and PWK was used for imprinting analysis.
Evaluation of neurons and glial cells in primary
Cerebral cortices and skin tissues were prepared from E15–17
embryos and used for primary cultures. Neuronal cells grown
selectively in the B-27 medium were confirmed by IF using an
antibody against MAP2 (microtubule-associated protein 2) as a
neuronal cell marker. Likewise, the glial cells grown in the G-5
medium were confirmed immunohistochemically for a glial
cell marker, GFAP (glial fibrillary acidic protein). Five days
after initiation of the culture with B-27, MAP-2-positive
neurons were growing almost exclusively with a negligible
number of GFAP-positive cells (Fig. 1A), while in the G-5
medium cortical glial cells proliferated and differentiated and
cortical neurons survived initially but began to degenerate
thereafter. At 11 days, the cultures with G-5 consist almost
entirely of GFAP-positive astrocytes (Fig. 1B). Similar results
were obtained in the Map2 and Gfap expression analysis by
RT–PCR, i.e. the B-27 cultures were composed primarily of
neurons with a few glial cells and the G-5 cultures were of glial
cells (Fig. 1C).
Maternal Ube3a expression in Ube3a deficient mice
We have previously described maternal expression of Ube3a
in Ube3a-deficient mice with a lacZ-IRES transcriptional
reporter (17). LacZ staining of the adult brain demonstrated
maternal-specific expression of Ube3a in some brain regions
including hippocampus and dentate gyrus (17). In the
neonatal brain, Ube3a was highly expressed in whole
cerebrum, but we could not identify the cell type with
maternal Ube3a expression by LacZ immunofluorescence and
cell type-specific markers (data not shown). To see the
expression of Ube3a in neurons and glial cells, we used
primary cultures of embryonic brain (E15) from Ube3a-
deficient mice. Cells in primary cultures from paternal-
deficient mice with only maternally inherited Ube3a allele
(mþ/p?) and maternal-deficient mice (m?/pþ) were stained
by anti-b-galactosidase antibody and cell marker antibodies:
NESTIN for progenitor cells, MAP2 for neurons and GFAP
for glial cells. b-Galactosidase from the lacZ-IRES transcrip-
tional reporter was stained as green dots in the cytoplasm. In
mþ/p? products, b-galactosidase was detected in all of the
embryonic fibroblasts, progenitor cells and glial cells, but
wasnot detected in neurons
b-galactosidase is not strongly stained by IF, paternal
transmission of the targeted allele leads to b-galactosidase
expression in all cultured cell types except neurons. In m?/pþ
products, all of the cultured cells including neurons expressed
b-galactosidase (Fig. 2F), indicating b-galactosidase was
expressed in all of the cell types examined from the maternal
Imprinting is maintained in primary brain cell
To evaluate the imprinting stability in primary cultures, we
investigated imprinting status of the Snrpn and Gabrb3 genes
located near Ube3a on mouse chromosome 7, by RT–PCR. As
the Snrpn gene is known to be expressed most strongly in the
brain and exclusively from the paternal allele (19) and Gabrb3
is not imprinted in the brain (20), they were used as positive
838 Human Molecular Genetics, 2003, Vol. 12, No. 8
by guest on June 5, 2013
and negative controls for an imprinted gene, respectively. We
used F1 hybrids from reciprocal crosses between the C57BL/6
and PWK strains. Allelic expression of Snrpn was analyzed
using an AvaI digestion polymorphism. The C57BL/6 strain
has two AvaI sites and the PWK strain has one such a site
between exons 1 and 7 (Fig. 3A). RT–PCR products by primers
1Fand 7R, followed by AvaI digestion, revealed that Snrpn was
expressed only from paternal allele in both neurons and glial
cells in primary cultures (Fig. 3B). The Gabrb3 gene was
previously demonstrated not to be imprinted in the brain (20).
A HpyCH4IV polymorphic site was used for allelic expression
of Gabrb3 (Fig. 3C). RT–PCR revealed that the Gabrb3 gene
was expressed in both neurons and glial cells and a small
amount of Gabrb3 expression was detected even in embryonic
fibroblasts. In all of these experiments, the Gabrb3 gene was
expressed equally from the maternal and paternal alleles
Reciprocal imprinting of sense and antisense Ube3a
transcripts in Ube3a-deficient mice by RT–PCR
Expression of Ube3a in cultured neurons was assayed by
In cultured neurons, RT–PCR using oligo-dT-primed cDNA
generated two products; one is the expected product of 267bp
(Fig. 4, one asterisk) detected in mþ/pþ and mþ/p? samples
and the other detected in mþ/pþ and m?/pþ sample is the
longer product 1265bp in size (Fig. 4, two asterisks), the same
size as genomic PCR product containing intron 15. To know
whether the longer product in mþ/pþ and m?/pþ neurons
reflects premature mRNA or antisense transcript of Ube3a from
the paternal allele, strand-specific RT–PCR was performed.
The longer product was detected in cDNA primed by 15F but
not by 16iR in mþ/pþ and m?/pþ neurons, indicating that the
longer product is paternally expressed antisense transcript of
Figure 1. Evaluation of neurons or glial cells selectively grown from cultured cortical cells, by IF and RT–PCR. Brain cells derived from the E15–17 cerebrum
were plated. The cells were double-stained for mouse monoclonal anti-MAP2 and rabbit anti-GFAP, and with DAPI. (A) Cells grown 5 days after the initiation of
culture in B-27 medium. Most cells were stained for MAP2 (green) and a few glial cells for GFAP (red). (B) Cells grown for 11 days in G-5 medium. Most cells in
the figure were labeled with GFAP (red), but a very few MAP2-labeled neurons were observed in other views. Bar, 100mm. (C) RT–PCR of cultured cells (neurons)
in B-27, those (glial cells) in G-5 medium, and embryonic fibroblasts. Each cDNA concentration was adjusted for Gapdh amplification as an internal control.
A small amount of GFAP cDNA was detected in RT–PCR products from neuron cultures.
Human Molecular Genetics, 2003, Vol. 12, No. 8839
by guest on June 5, 2013
Ube3a (Fig. 4A). We also evaluated the expression of Ube3a in
cultured glial cells, and found that, in the m?/pþ sample, the
sense transcript but not antisense transcript of Ube3a was
detected (Fig. 4A). Such reciprocal imprinting of sense and
antisense Ube3a transcripts in neurons but not in glial cells
in vitro was supported by RT–PCR analysis using whole
cortical tissues at different developmental stages. In m?/pþ
whole cortices, both sense and antisense transcripts were
detected, whereas in mþ/p? only sense transcript was detected
at E16 and adult (>P42) cerebrums, when progenitor cells/
neurons and neurons/glial cells are main components, respec-
tively (Fig. 4B).
Reciprocal imprinting of sense and antisense Ube3a
transcripts is maintained in wild-type mice
We have demonstrated by IF that the maternal Ube3a promoter
was active in all cultured neurons from m?/pþ cerebrum and
that the paternal Ube3a promoter was inactive only in neurons
from mþ/p? cerebrum. The exclusive maternal expression of
Ube3a in neurons was also confirmed by RT–PCR, a more
sensitive assay than IF. To exclude the possibility that the
deleted region in Ube3a-deficient mice with lacZ substitution
may affect the imprinting status, we used F1 hybrids from
reciprocal crosses between C57BL/6 and PWK. Allele specific
expression of the sense and antisense transcripts of Ube3a was
analyzed using a polymorphic site in exon 5, where C57BL/6
has two Tsp509I sites and PWK has one Tsp509I site (Fig. 5A)
(21). RT–PCR was performed using primers 5Fex and 6Rex in
exons 5 and 6, respectively. RT–PCR products digested by
Tsp509I revealed that in cultured cortical neurons, the sense
transcript of Ube3a was exclusively expressed from the
maternal allele, whereas in cultured glial cells and fibroblasts,
it was biallelically expressed (Fig. 5B). We also examined the
imprinting status of the antisense transcript of Ube3a, which
Chamberlain et al. have already verified in unspliced form of
Ube3a in the mouse brain by RT–PCR across intron 5 using
strand-specific cDNA (21). The strand specific cDNA primed
only by 5F was amplified in RNA from total cerebral cortices
and cultured neurons but not glial cells (Fig. 5C), indicating
that the antisense transcript of Ube3a was not only exclusively
expressed from paternal allele, as expected from the previous
report by Chamberlain et al. (21), but was transcribed
specifically in neurons.
Imprinting status of Ube3a in the telencephalon/cerebral
cortices at developmental stages
reflects the partial imprinting in the brain invivo, telencephalon/
cerebral cortices were prepared from F1 hybrid mice at E10,
E16, P1, P5, P14 and P28, and used for RT–PCR assay. Before
imprinting analysis, brain cDNA from these developmental
stages was evaluated by RT–PCR using primers for Nestin as a
marker for progenitor cells, Map2 for neurons, and Gfap for
astrocytes, after normalization of cDNA concentration for
Gapdh expression. As previously reported (22), Gfap became
faintly positive around E16, and was clearly detected from birth
onwards, whereas Map2 was not positive at E10, but clearly
positive at E16. Nestin was positive for several weeks after birth
Figure 2. Imprinted Ube3a expression in Ube3a-deficient mice by IF. Primary
cultured cells from mþ/p? embryo (B, C, D, E) and from m?/pþ embryo at
E15 (F) were double-stained for anti-b-galactosidase antibody (green) and cell
marker antibodies (red), and also stained by DAPI (blue). (A) Schematic illus-
tration of Ube3a targeting construct (top), genomic locus (middle), and targeted
allele (bottom) (17). A part of exons 15 and 16 was substituted by the IRES-
cassette.(B) Fibroblasts from
b-galactosidase in all cells. (C, D) Cortical cells from E15 mþ/p? embryo,
grown 1 day (C) and 5 days (D) after the initiation of culture in B-27 medium
were stained by anti-NESTIN (red) and anti-MAP2 (red) antibodies, respec-
tively. Only NESTIN-positive progenitor cells express b-galactosidase. (E)
Cortical cells (glial cells) from E15 mþ/p? embryo, grown 11 days in G-5
medium, stained by anti-GFAP(red) antibody express b-galactosidase. (F)
Cortical cells (neurons) from E15 m?/pþ embryo, grown 5 days in B-27 med-
ium stained by anti-MAP2 (red) antibody express b-galactosidase. Arrowheads
point to b-galactosidase-positive dots (green). Bar, 10mm.
840 Human Molecular Genetics, 2003, Vol. 12, No. 8
by guest on June 5, 2013
(Fig. 6A). The imprinting status of Ube3a in the whole
telencephalon/cerebral cortices was analyzed. By quantitative
analysis of electrophoretic bands for each allele, we found that
Ube3a expression of maternal and paternal alleles was equal in
E10 telencephalon where neurogenesis had not yet commenced,
whereas partial imprinting became obvious after birth (Fig. 6B).
The antisense transcript of Ube3a was not detected by strand-
specific RT–PCR in E10 telencephalon (data not shown).
The role of human UBE3A in AS has been well established (23).
Although no pathological abnormalities are recognized in the
brain of AS patients, most AS symptoms are related to brain
dysfunction, suggesting that UBE3A expression in the central
nervous system (CNS) is affected in AS. Defining cell-types in
the brain where Ube3a is imprinted is important for our
understanding of the mechanism of imprinting in CNS and as
well as of its neuronal maintenance and function. We have
developed a cell culture model with which brain cell-type-
specific imprinting of mouse Ube3a can be characterized. To
minimize the survival of glial cells in the neuron culture, we
used serum-free B27 medium, which allows a yield of >99%
embryonic neurons 4 days after initiation of culture (24). In our
primary neuron culture, a very small number of glial cells were
detected with IF and RT–PCR, and a negligible amount of glial
cell cDNA not analyzable for gene expression was detected by
semi-quantitative RT–PCR (Fig. 1). Conversely, a negligible
number of neurons were also detected in the glial cell culture.
Another limitation of the primary neuron culture system is
maturation or aging in cultured cells. In the mouse cerebral
cortex, neurogenesis commences around stage E12, peaks
around E15, and finishes around birth (22,25); cortical
astrocytes are first detected around E16 and oligodendrocytes
around birth, but the vast majority of both glial cell types are
produced during the first postnatal month (26). In our primary
cultures, cerebral cortices were removed from E15–17 embryos
and cultured for 5 and 11 days in B-27 medium for neurons and
G-5 medium for glial cells, respectively. Therefore, cells
Figure 3. Expression analysis of Snrpn and Gabrb3. (A) Schematic representation of Snrpn exons including AvaI sites (vertical line). C57BL/6 and PWK alleles
were diagrammed showing PCR products and fragment sizes. (B) Verification of Snrpn imprinting. RT–PCR products were digested by AvaI. The same cDNAwas
used as in Figure 1C. (C) Schematic representation of Gabrb3 exons including an HpyCH4IV restriction site (vertical line). C57BL/6 and PWK alleles were dia-
grammed showing PCR products and fragment sizes. (D) Verification of biallelic expression of Gabrb3. RT–PCR products were digested by HpyCH4IV.
Human Molecular Genetics, 2003, Vol. 12, No. 8 841
by guest on June 5, 2013
harvested were adjusted at ages equivalent for P0-2 for neurons
and P6-8 for glial cells. Although the embryonic neurons
cultured are reported to retain a clear morphological and
electrophysiological phenotype (27), limitations of culture
periods and difficulties in pure neuronal cell cultures from
adult brain still prevents the analysis of matured or aged neurons
in the primary culture system. As far as cells in this culture
system are concerned, they represent neurons and glial cells in
the neonatal period. Imprinting status of the gene in whole
cortices from embryos and adult mice can be interpreted
consistent with our experiments using primary brain cell culture
system (Figs 4 and 6). However, considering the limitations of
the in vitro culture system mentioned above, expression and
imprinting status of brain cells, depending on their functions in
development, might be analyzed directly in the future.
Using this culture system, we have presented evidence for
neuron-specific imprinting of Ube3a by IF (Fig. 2) and
RT–PCR (Figs 4 and 5). We also demonstrated that the
imprinting status of a known imprinted gene, Snrpn, as well as
a non-imprinted gene, Gabrb3, were stable in our primary
culture system (Fig. 3). Our data indicated that Ube3a is
imprinted only in neurons and not imprinted in glial cells in
brain culture, in contrast with previous data by in situ
hybridization (15). Albrecht et al. reported patterns of mouse
Ube3a expression in the wild-type embryonic brain and in the
adult brains with partial paternal UPD encompassing Ube3a.
Expression of Ube3a in the neuro-epithelium was detected
since stage E8.5, and the gene was highly expressed at E15.5 in
the olfactory bulb, nasal epithelium, cerebral cortex, hippo-
campus and some other brain regions (15). On the other hand,
the expression was lower in the adult brain with partial paternal
UPD than in the wild-type embryonic brain. Lower expression
in the brain with paternal UPD may have reflected lower
expression from the paternal allele than the maternal allele.
According to the expression level of Ube3a in the paternal
UPD brain, Albrecht et al. divided the brain into three regions:
(i) an undetectable region, as in the hippocampus; (ii) a
moderately or slightly reduced region as in the cerebral cortex;
and (iii) a region indistinguishable from the normal region as in
the anterior commissure, optic chiasma and other regions (15).
They concluded that imprinting of Ube3a was region-specific,
which implies different relative activities of the Ube3a alleles
in different brain regions. Although neuron-specific imprinting
demonstrated in our study is restricted in cultured cortical
neurons and never excludes the possibility of region-specific
imprinting, plausible explanations for the discrepancy between
studies by us and by Albrecht et al. include: (i) neuron-specific
imprinting of Ube3a might lead to region-specific imprinting in
brain areas with higher density of neurons, but to less specific
imprinting in lower density areas; (ii) imprinting status in the
adult brain in vivo might be originally different from that in
the embryonic brain in vitro, depending on their functions
in the brain; (iii) RT–PCR is much more sensitive than in situ
hybridization to detect gene expression, i.e. relatively low
expression in the brain of wild-type mice with paternal UPD
might be difficult to be analyzed for imprinting status by in situ
hybridization. Lower Ube3a expression in glial cells than in
neurons may explain partial imprinting shown by RT–PCR
Figure 4. Imprinted Ube3a expression in Ube3a deficient mice by RT–PCR. (A) Cultured neurons (left) and glial cells (right) grown 5 days and 11 days after the
initiation of culture in B-27 and G-5 medium, respectively; and (B) cerebral tissues from E16 embryos (left) and adult mice (right) were subjected to RT–PCR assay.
Sense (*) and antisense (**) transcripts of Ube3a in exons 15 and 16 were detected by RT–PCR in oligo-dT primed cDNA, whereas only antisense transcript was
amplifiedbystrandspecificRT–PCRusing15FprimedcDNA(lefttop).EachcDNAconcentrationwas adjustedforGapdhamplificationas aninternal control.
842 Human Molecular Genetics, 2003, Vol. 12, No. 8
by guest on June 5, 2013
Figure 5. Imprinted Ube3a expression in F1 hybrids from reciprocal crosses between C57BL/6 and PWK. (A) Schematic representation of Ube3a exons 5 and 6
including restriction sites. C57BL/6 and PWK alleles are diagrammed showing PCR products, Tsp5091 sites (verticle line) and fragment sizes. Exon 5 in C57BL/6
has one additional Tsp509I site compared to that in PWK. Horizontal small arrows show primers used for PCR. (B) Verification of Ube3a imprinting in cultured
neurons, glial cells and fibroblasts. Each cDNA concentration was normalized by Gapdh expression in Figure 1C. (C) Verification of Ube3a antisense transcripts.
Strand-specific RT–PCR was performed using oligo-dT priming, specific 5F priming for antisense RNA and 5iR priming for unspliced sense RNA. Plus and minus
signs mean with and without Tsp509I digestion, respectively.
Human Molecular Genetics, 2003, Vol. 12, No. 8 843
by guest on June 5, 2013
assay in the whole cerebral cortex (Fig. 6B). Although cultured
embryonic neurons are reported to retain a clear morphological
and electrophysiological characters, studies by in situ hybridi-
zation of the embryonic or neonatal brain with paternal UPD
will clarify whether Ube3a imprinting in the brain is neuron-
specific or region-specific in vivo.
Recently, Herzing et al. (28) reported that UBE3A was
imprinted in fibroblasts and neural precursor cells by RNA-
FISH, whereas previously reported RT–PCR results in human
fibroblasts demonstrated equal biallelic UBE3A expression (12).
Total RNA from cycling cell population may only reflect the
accumulation of RNA from cells with unequal biallelic UBE3A
expression detected by RNA-FISH, resulting in almost equal
biallelic expression by RT–PCR assay. Interestingly they
suggested that exclusive imprinted UBE3A expression may be
related to neuronal maturation, because preferential maternal
expression is observed in undifferentiated neural cells, becom-
ing exclusively maternal as the neurons differentiate. If degree
of imprinted expression of UBE3A in the cell depends on each
cell cycle stage, RNA-FISH data may support our conclusion of
exclusive maternal expression only in neurons, which stay in G0
phase after differentiation from the progenitor cells.
We also showed neuron-specific expression of antisense
in the brain and under the control of the imprinting center at the
Prader-Willi syndrome critical region (PWS-IC) (21). In the
human brain, Rougeulle et al. (29) detected a 20-kb paternally-
expressed, intronless UBE3A antisense RNA fragment, which
was recently reported to overlap the 30UTR of a hypothetical
transcript extending from SNURF-SNRPN to Ube3a (30). In
addition to neuron-specific expression of antisense Ube3a, our
RT–PCR study has not detected any neuron-specific isoforms of
Ube3a and the bisulfite sequencing study has not found any
differences in DNA methylation at the Ube3a promoter regions
Figure 6. Evaluation of brain cells and imprinting status of Ube3a in telencephalon/cerebral cortices at developmental stages. The cDNA concentration was
adjusted according to Gapdh amplification as an internal control. (A) Telencephalon/cerebral cortices from E10, E16, P1, P5, P14 and P28 embryos of F1 hybrids
were used for RT–PCR. (B) Imprinting analysis of Ube3a in telencephalon/cerebral cortices. RT–PCR products were digested by Tsp509I. Plus and minus signs
mean with and without Tsp509I digestion, respectively.
844 Human Molecular Genetics, 2003, Vol. 12, No. 8
by guest on June 5, 2013
between parental alleles (data not shown). Although the role of
Ube3a is expressed paternally in neurons, only where the sense
Ube3a is maternally expressed, suggests that the antisense
transcript expressed in neurons is closely related to neuron-
specific imprinting of Ube3a (Fig. 7).
Our finding of neuron-specific reciprocal imprinting of sense
is notimprinted in telencephalon atE10 (Fig.6), wherethe brain
tissue consists of progenitor cells, which will differentiate to
neurons by birth and glial cells after birth (Fig. 7). Then, when
epigenetic factors that control Ube3a imprinting in neurons?
Specification of cell lineages in the developing brain is thought
to be regulated in part by epigenetic modifications of cell-type-
specific genes, besides cell-external cues including various
cytokines (31). Further investigations using the primary brain
cell culture system will help us to elucidate the underlying
epigenetic mechanisms of neuron-specific imprinting of Ube3a.
MATERIALS AND METHODS
Ube3a knockout mice
Ube3a-deficient mice with a lacZ-IRES transcriptional reporter
were generated in a C57BL/6 background (Fig. 2A) (17).
Paternal-deficient mice with only maternally inherited Ube3a
allele (mþ/p?) were produced by matings of male hetero-
zygotes with female wild-type mice (C57BL/6), and maternal-
deficient mice (m?/pþ) were vice versa. Homozygous
Ube3a-deficient mice (m?/p?) and wild-type mice (mþ/pþ)
were produced by matings of female heterozygotes with male
heterozygotes. Mice and embryos genotyped by PCR and
Southern blotting were used for further analysis (17). Cerebral
cortices were prepared from E16 embryos and P5 mice.
F1 hybrid mice of reciprocal crosses between
C57BL/6 and PWK
C57BL/6 female mice were crossed with PWK male mice
Telencephalon/cerebral cortices and embryonic fibroblasts were
prepared from E10–18 embryos, P1, P5, P14 and P28 products
of reciprocal crosses between C57BL/6 and PWK.
and viceversa (PWK?C57BL/6).
Embryos were removed from the uterus of timed pregnant mice
and placed in Petri dishes containing ice-cold HEPES. Cerebral
cortices were freed from meninges. Cerebral cortices and/or
embryonic fibroblasts were prepared from embryos and
neonatal/adult mice. Tissues were used for RNA extraction or
primary cultures. All procedures were approved by the Ethics
Review Committee for Animal Experimentation of the Animal
Center for Medical Research, Nagasaki University.
Fetal cerebral cortices without meninges were dissociated by
mechanical trituration and trypsinized with 0.25% trypsin with
EDTA at 37?C for 10min. Fetal calf serum (FCS; Bio
Whittaker) was then added to dissociate cells, followed by
filtration through sterile nylon sieve (pore size, 100mm).
Filtered cells were collected by centrifugation at 1200rpm for
Figure 7. Summary of sense and antisense transcripts of Ube3a expressed in neural lineage. In neurons, the sense and antisense transcripts are expressed exclu-
sively from the maternal (M) and paternal allele (P), respectively. In progenitor cells and glial cells, only sense transcript is expressed from both parental alleles.
Human Molecular Genetics, 2003, Vol. 12, No. 8845
by guest on June 5, 2013
10min. The cell pellet was resuspended in optimal media for
growth of neurons or glial cells.
was resuspended in NeurobasalTM(Gibco BRL) supplemented
with 1mM L-glutamine and B-27 supplement (Gibco BRL) to
ensure selective growth of cortical neurons (24,32). Cells from
the embryonic cerebral cortex were plated on polyethylenei-
1?106cells/ml, and cultured in 5% CO2at 37?C.
The cell pellet from cerebral cortices
dishes ata densityof
Glial cell cultures.
resuspended in Dulbecco’s modified Eagle’s medium MEM
(DMEM; Sigma) supplemented with 10% FCS. Cells were
plated on polyethyleneimine-coated 3.5cm plastic dishes at a
density of 1?106cells/ml, and cultured overnight in 5% CO2
at 37?C and medium was changed to NeurobasalTMwith
1mM L-glutamine and G-5 supplement (Gibco BRL). After
5–7 days in the primary culture, glial components grown were
dislodged enzymatically with 0.25% trypsin and subcultured
on new polyethyleneimine-coated plastic dishes. Cultures were
maintained in 5% CO2at 37?C for a total of 11 days.
The cell pellet from cerebral cortices was
Embryonic fibroblast culture.
E15–17 embryos and cultured in DMEM supplemented with
10% FCS. Cells were plated on plastic dishes and maintained
in 5% CO2at 37?C.
Fibroblasts were derived from
The cells cultured on plastic dishes were fixed with 4%
paraformaldehyde in PBS and subjected to immunofluorescent
staining. The following primary antibodies were used: mouse
monoclonal anti-MAP2 (microtubule-associated protein 2)
antibody, rabbit polyclonal anti-MAP2 antibody (Chemicon),
rabbit polyclonal anti-GFAP (anti-glial fibrillary acidic protein)
antibody (Dako), rabbit anti-serum to NESTIN (a gift from
M. Ogawa), and mouse monoclonal anti-b-galactosidase
antibody (Promega). Secondary antibodies were
488-conjugated goat anti-mouse IgG antibody and Alexia
568-conjugated goat anti-rabbit IgG antibody (Molecular
Probe). The cells were counterstained with DAPI to identify
nuclei. Signals were viewed under a Zeiss Axioskop fluores-
cence microscope and images were acquired with a PXL cooled
CCD camera (Photometrics).
Total RNA was isolated from cultured cells and tissues with
RNeasy (Qiagen) according to the manufacturer’s protocol. The
cDNA was generated from total RNA by SUPERSCRIPT II
RNase H-reverse transcriptase (Gibco BRL) primed with oligo
(dT)12–18 or specific forward or reverse primers. The first-
strand cDNA was synthesized at 42?C for 50min. Then,
mRNA-cDNA chains were denatured and the reverse tran-
scriptase activity was arrested by heating at 70?C for 5min. As
a control, an identical reaction was carried out without reverse
transcriptase. Primers for specific primings were as follows:
specific forward primer 15F for antisense Ube3a in the
knockout mice: 50-GGAGTTCTGGGAAATTGTTCA-30, spe-
cific reverse primer 16iR for sense Ube3a in the knockout
forward primer 5F for antisense Ube3a: 50-CACATATGATG-
AAGCTACGA-30, specific reverse primer 5iR for sense
Polymerase chain reaction (PCR)
The cDNA obtained was used to perform PCR for Nestin, Map2,
reverse, 50-TCTTCAAATCTTAGTGGCTCC-30; Map2 forward,
50-AGTCCCTCTCCCATCACCAGT-30; Map2 reverse, 50-CTC-
TACTTTACCCCCCATCTCTT-30; Gfap forward, 50-AAGC-
50-GCGATCTCGATGTCCAGGGC-30; Gapdh forward, 50-
50-TCCACCACCCTGTTGCTGTA-30. For a semi-quantitative
PCR, optimal template cDNA concentrations were determined
according to Gapdh amplification. PCR products were amplified
through 28 cycles of 30sec at 94?C, 30sec at 55?C and 30sec at
Imprinting analysis by RT–PCR
In the knockout mice, Ube3a expression was analyzed
using primers 15F and 16R in exons 15 and 16 in the knock-
out region, respectively (Fig. 2A). Primer 16R is 50-
GTTTACAGCATGCCAAATCC-30. Antisense transcript of
Ube3a in the knockout region was amplified using the same
forward/reverse primers, 15F and 16iR, as those used for
specific primings. PCR amplifications with primers 15F and
16R/16iR were performed through 30 cycles of 30sec at 94?C,
30sec at 55?C and 60sec at 72?C. In the F1 hybrid mice from
reciprocal crosses between C57BL/6 and PWK, allele specific
expression of sense and antisense transcripts of Ube3a was
analyzed using a polymorphic site in its exon 5, where the
C57BL/6 strain has two Tsp509I sites and the PWK strain has
one Tsp509I site (Fig. 2A) (21). The cDNA from cultured
cells and tissues was subjected to PCR for Snrpn and
Gabrb3 amplification using the following primers: Snrpn 1F,
50-TGTCACTGGCGTGGAAAGGA-30; Gabrb3 reverse, 50-
sense strand of Ube3a was amplified using the same forward/
reverse primers, 5F and 5iR, as those used for specific
primings. PCR amplifications with primers 5Fex and 6Rex
were performed through 30 cycles of 30sec at 94?C, 30sec at
58?C and 30sec at 72?C, by other primer pairs through 30
cycles of 30sec at 94?C, 30sec at 55?C and 30sec at 72?C.
Each PCR product was then digested with Tsp509I, AvaI and
HypCH4IV for imprinting analysis of sense and antisense
Ube3a, Snrpn and Gabrb3, respectively, and electophoresed in
2% agarose gel or 4% polyacrylamide gel.
a primer5Fex, 50-
846Human Molecular Genetics, 2003, Vol. 12, No. 8
by guest on June 5, 2013
ACKNOWLEDGEMENTS Download full-text
T.K. was supported in part by a Grant-in-Aid from the Ministry
of Health and Welfare of Japan, and by a Grant-in-Aid for
Scientific Research C (14570754) from the Ministry of
Education, Culture, Sports, Science and Technology of Japan.
1. Cattanach, B.M. and Beechey, C.V . (1990) Autosomal and X-chromosome
imprinting. Dev. Suppl., 63–72.
2. Surani, M.A., Barton, S.C. and Norris, M.L. (1984) Development of
reconstituted mouse eggs suggests imprinting of the genome during
gametogenesis. Nature, 308, 548–550.
3. Tilghman, S.M. (1999) The sins of the fathers and mothers: genomic
imprinting in mammalian development. Cell, 96, 185–193.
4. Huibregtse, J.M., Scheffner, M. and Howley, P.M. (1991) A cellular protein
mediates association of p53 with the E6 oncoprotein of human
papillomavirus types 16 or 18. EMBO J., 10, 4129–4135.
5. Huibregtse, J.M., Scheffner, M., Beaudenon, S. and Howley, P.M. (1995)
A family of proteins structurally and functionally related to the E6-AP
ubiquitin-protein ligase. Proc. Natl Acad. Sci. USA, 92, 2563–2567.
6. Scheffner, M., Nuber, U. and Huibregtse, J.M. (1995) Protein ubiquitination
involving an E1-E2-E3 enzyme ubiquitin thioester cascade. Nature, 373,
7. Kishino, T., Lalande, M. and Wagstaff, J. (1997) UBE3A/E6-AP mutations
cause Angelman syndrome. Nat. Genet., 15, 70–73.
8. Matsuura, T., Sutcliffe, J.S., Fang, P., Galjaard, R.J., Jiang, Y.H.,
Benton, C.S., Rommens, J.M. and Beaudet, A.L. (1997) De novo truncating
mutations in E6-AP ubiquitin-protein ligase gene (UBE3A) in Angelman
syndrome. Nat. Genet., 15, 74–77.
9. Knoll, J.H., Nicholls, R.D. and Lalande, M. (1989) On the parental origin of
the deletion in Angelman syndrome. Hum. Genet., 83, 205–207.
10. Malcolm, S., Clayton-Smith, J., Nichols, M., Robb, S., Webb, T.,
Armour, J.A., Jeffreys, A.J. and Pembrey, M.E. (1991) Uniparental paternal
disomy in Angelman’s syndrome. Lancet, 337, 694–697.
11. Buiting, K., Saitoh, S., Gross, S., Dittrich, B., Schwartz, S., Nicholls, R.D.
and Horsthemke, B. (1995) Inherited microdeletions in the Angelman
and Prader-Willi syndromes define an imprinting centre on human
chromosome 15. Nat. Genet., 9, 395–400.
12. Nakao, M., Sutcliffe, J.S., Durtschi, B., Mutirangura, A., Ledbetter, D.H.
and Beaudet, A.L. (1994) Imprinting analysis of three genes in the
Prader-Willi/Angelman region: SNRPN, E6-associated protein, and
PAR-2 (D15S225E). Hum. Mol. Genet., 3, 309–315.
13. Rougeulle, C., Glatt, H. and Lalande, M. (1997) The Angelman syndrome
candidate gene, UBE3A/E6-AP, is imprinted in brain. Nat. Genet., 17,
14. Vu, T.H. and Hoffman, A.R. (1997) Imprinting of the Angelman syndrome
gene, UBE3A, is restricted to brain. Nat. Genet., 17, 12–13.
15. Albrecht, U., Sutcliffe, J.S., Cattanach, B.M., Beechey, C.V., Armstrong, D.,
Eichele, G. and Beaudet, A.L. (1997) Imprinted expression of the murine
Angelman syndrome gene, Ube3a, in hippocampal and Purkinje neurons.
Nat. Genet., 17, 75–78.
16. Jiang, Y.H., Armstrong, D., Albrecht, U., Atkins, C.M., Noebels, J.L.,
Eichele, G., Sweatt, J.D. and Beaudet, A.L. (1998) Mutation of the
Angelman ubiquitin ligase in mice causes increased cytoplasmic p53 and
deficits of contextual learning and long-term potentiation. Neuron, 21,
17. Miura, K., Kishino, T., Li, E., Webber, H., Dikkes, P., Holmes, G.L. and
Wagstaff, J. (2002) Neurobehavioral and electroencephalographic
abnormalities in Ube3a maternal-deficient mice. Neurobiol. Dis., 9,
18. Bottenstein, J.E. and Sato, G.H. (1985) Cell Culture in the Neuroscience.
Plenum Press, NY.
19. Barr, J.A., Jones, J., Glenister, P.H. and Cattanach, B.M. (1995) Ubiquitous
expression and imprinting of Snrpn in the mouse. Mamm. Genome., 6,
20. Nicholls, R.D., Gottlieb, W., Russell, L.B., Davda, M., Horsthemke, B. and
Rinchik, E.M. (1993) Evaluation of potential models for imprinted and
nonimprinted components of human chromosome 15q11-q13 syndromes
by fine-structure homology mapping in the mouse. Proc. Natl Acad.
Sci. USA., 90, 2050–2054.
21. Chamberlain, S.J. and Brannan, C.I. (2001) The prader-willi syndrome
imprinting center activates the paternally expressed murine Ube3a antisense
transcript but represses paternal Ube3a. Genomics, 73, 316–322.
22. Bayer, S.A. and Altman, J. (1991) Neocortical Development.
Raven Press, NY.
23. Lossie, A.C., Whitney, M.M., Amidon, D., Dong, H.J., Chen, P.,
Theriaque, D., Hutson, A., Nicholls, R.D., Zori, R.T., Williams, C.A. et al.
(2001) Distinct phenotypes distinguish the molecular classes of Angelman
syndrome. J. Med. Genet., 38, 834–845.
24. Brewer, G.J. (1995) Serum-free B27/neurobasal medium supports
differentiated growth of neurons from the striatum, substantia nigra,
septum, cerebral cortex, cerebellum, and dentate gyrus. J. Neurosci. Res.,
25. Jacobson, M. (1991) Developmental Neurology. Plenum Press, NY.
26. Qian, X., Shen, Q., Goderie, S.K., He, W., Capela, A., Davis, A.A. and
Temple, S. (2000) Timing of CNS cell generation: a programmed sequence
of neuron and glial cell production from isolated murine cortical stem cells.
Neuron, 28, 69–80.
27. Evans, M.S., Collings, M.A. and Brewer, G.J. (1998) Electrophysiology of
embryonic, adult and aged rat hippocampal neurons in serum-free culture.
J. Neurosci. Methods, 79, 37–46.
28. Herzing, L.B., Cook, E.H.Jr. and Ledbetter, D.H. (2002) Allele-specific
expression analysis by RNA-FISH demonstrates preferential maternal
expression of UBE3A and imprint maintenance within 15q11- q13
duplications. Hum. Mol. Genet., 11, 1707–1718.
29. Rougeulle, C., Cardoso, C., Fontes, M., Colleaux, L. and Lalande, M.
(1998) An imprinted antisense RNA overlaps UBE3A and a second
maternally expressed transcript. Nat. Genet., 19, 15–16.
30. Runte, M., Huttenhofer, A., Gross, S., Kiefmann, M., Horsthemke, B. and
Buiting, K. (2001) The IC-SNURF-SNRPN transcript serves as a host for
multiple small nucleolar RNA species and as an antisense RNA for UBE3A.
Hum. Mol. Genet., 10, 2687–2700.
31. Edlund, T. and Jessell, T.M. (1999) Progression from extrinsic to intrinsic
signaling in cell fate specification: a view from the nervous system. Cell,
32. Brewer, G.J., Torricelli, J.R., Evege, E.K. and Price, P.J. (1993) Optimized
survival of hippocampal neurons in B27-supplemented Neurobasal, a new
serum-free medium combination. J. Neurosci. Res., 35, 567–576.
Human Molecular Genetics, 2003, Vol. 12, No. 8847
by guest on June 5, 2013