A central question in genomic imprinting is how a specific sequence is recognized as the target for epigenetic marking. In both mammals and plants, imprinted genes are often associated with tandem repeats and transposon-related sequences, but the role of these elements in epigenetic gene silencing remains elusive. FWA is an imprinted gene in Arabidopsis thaliana expressed specifically in the female gametophyte and endosperm. Tissue-specific and imprinted expression of FWA depends on DNA methylation in the FWA promoter, which is comprised of two direct repeats containing a sequence related to a SINE retroelement. Methylation of this element causes epigenetic silencing, but it is not known whether the methylation is targeted to the SINE-related sequence itself or the direct repeat structure is also necessary. Here we show that the repeat structure in the FWA promoter is highly diverse in species within the genus Arabidopsis. Four independent tandem repeat formation events were found in three closely related species. Another related species, A. halleri, did not have a tandem repeat in the FWA promoter. Unexpectedly, even in this species, FWA expression was imprinted and the FWA promoter was methylated. In addition, our expression analysis of FWA gene in vegetative tissues revealed high frequency of intra-specific variation in the expression level. In conclusion, we show that the tandem repeat structure is dispensable for the epigenetic silencing of the FWA gene. Rather, SINE-related sequence is sufficient for imprinting, vegetative silencing, and targeting of DNA methylation. Frequent independent tandem repeat formation events in the FWA promoter led us to propose that they may be a consequence, rather than cause, of the epigenetic control. The possible significance of epigenetic variation in reproductive strategies during evolution is also discussed.
Genomic imprinting, mono-allelic gene expression depending on the parent-of-origin, is an epigenetic process known in mammals and flowering plants. A central question in genomic imprinting is how a specific sequence is recognized as the target for epigenetic marking. In both mammals and plants, imprinted genes are often associated with tandem repeats and transposon-related sequences, but the role of these elements in epigenetic gene silencing remains elusive. FWA is an imprinted gene in Arabidopsis thaliana expressed specifically in the female gametophyte and endosperm. The FWA promoter is comprised of two direct repeats containing a sequence related to a SINE retroelement. Methylation of this element causes epigenetic silencing, but it is not known whether the methylation is targeted to the SINE-related sequence itself or the direct repeat structure is necessary. Here we show that the direct repeat structure is highly diverse in species within the genus Arabidopsis. Unexpectedly, we found that the direct repeat structure is dispensable for the epigenetic silencing and methylation of the FWA promoter. Rather, the SINE-related promoter sequence is sufficient for these features. Frequent independent formation of the tandem repeats suggests that they may be a consequence of the epigenetically controlled system.
Citation: Fujimoto R, Kinoshita Y, Kawabe A, Kinoshita T, Takashima K, Nordborg M, et al. (2008) Evolution and Control of Imprinted FWA Genes in the Genus Arabidopsis. PLoS Genet 4(4): e1000048. https://doi.org/10.1371/journal.pgen.1000048
Editor: Joseph R. Ecker, The Salk Institute for Biological Studies, United States of America
Received: December 5, 2007; Accepted: March 10, 2008; Published: April 4, 2008
Copyright: © 2008 Fujimoto 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 work is supported by Grant-in-Aid for Scientific Research (19207002 and 19060014) to T. Kakutani and Research Fellowship of the Japan Society for the Promotion of Science for Young Scientists to R. Fujimoto.
Competing interests: The authors have declared that no competing interests exist.
Parental imprinting, mono-allelic gene expression depending on its parent-of-origin, is an epigenetic process known in mammals and flowering plants. In both mammals and plants, many imprinted genes are under control of DNA methylation, as their imprinting is abolished by mutations in DNA methyltransferase genes –. Interestingly, imprinted genes often contain sequences originated from parasitic sequences such as transposons and viruses –. As DNA methylation works as a defense mechanism against parasitic sequences –, the control of imprinted genes by DNA methylation may have evolved from defense mechanisms. Another feature of imprinted genes is their frequent association with tandem repeats –. Despite the strength of the association, evidence is limited concerning whether the repeat structure itself is important for imprinting ,.
The FWA gene in the flowering plant Arabidopsis thaliana is one of the most extensively studied systems linking the control of DNA methylation and imprinting. The FWA gene was originally identified through characterization of epigenetic mutants causing a heritable late-flowering phenotype. The phenotype was due to ectopic expression of the FWA gene in vegetative tissue –. In wild type plants, the FWA gene is silent in vegetative tissues and expressed specifically in the endosperm in an imprinted manner . FWA silencing depends on cytosine methylation, as it is derepressed by mutations in the maintenance methylase gene MET1 ,,. In addition, the imprinting is established in the female gametophyte by a “one-way” activation, which depends on the DNA demethylase DEMETER ,.
In addition to these trans-acting components, a cis-requirement for the epigenetic FWA silencing has also been identified. Promoter of the FWA gene has two pairs of tandem repeats that are heavily methylated . Transcription starts from this region when the methylation is lost . The nucleotide sequence of this region is similar to a SINE retrotransposon . Using artificial de novo methylation induced by double-stranded RNA, we previously showed that the critical methylated element corresponds to the SINE-related tandem repeats . However, it is still unclear whether the SINE sequence per se can direct DNA methylation, or whether the tandem repeat structure is necessary for control of methylation and imprinted expression.
Here we investigate the evolution and natural variation of FWA promoters in A. thaliana and five other Arabidopsis species. The results demonstrate that the tandem repeat structure is necessary neither for imprinted FWA expression, nor for vegetative FWA silencing. Instead, the SINE-related sequence itself appears to be the primary target of the epigenetic control, which might have subsequently induced frequent tandem repeat formation during evolution. In addition, considerable intra-specific variation was found in vegetative FWA expression, which might present sources for epigenetic variation in reproductive strategy.
Structure of the FWA Promoter in Four Arabidopsis Species
We have previously shown that the target of epigenetic FWA silencing in A. thaliana is a SINE-related direct repeat . In order to investigate the relationship between the structure of the FWA promoter and FWA expression, we identified FWA orthologues in three related Arabidopsis species, A. halleri, A. lyrata, and A. arenosa. We found that the gene structure, including coding regions as well as introns and promoter sequences, was conserved among these species and A. thaliana (Figure S1). Notably, SINE-like sequences were found in the FWA promoters of all four species, suggesting that the insertion of this sequence occurred before divergence of these species.
Unexpectedly, however, we found that duplications in the FWA promoters were highly diverse both within and among species (Figure 1 and Figure S2). The two pairs of tandem repeats found in A. thaliana were not duplicated in the other species. On the other hand, the FWA promoters of A. lyrata and A. arenosa contain other tandem repeats, which spanned different regions within the SINE-related sequences (Figure 1B, 1C, and Figure S2). Comparison of the structure of the FWA promoter of these species revealed at least four independent duplication events: two in A. thaliana, one in A. arenosa, and one in A. lyrata. In A. lyrata, the repeat number of the FWA promoter differs between subspecies (ssp.) lyrata (three copies) and ssp. petraea (four copies) (Figure 1 and Figure S2). In A. halleri, no tandem repeat was found in the FWA promoter (Figure 1 and Figure S2).
(A) Schematic view. Sequences with similarity to AtSINE2 are shown by gray boxes. Tandem repeats covering different regions are shown by arrows with different colors. Transcription start site is shown by vertical bar, and target site duplications of the SINE insertion are shown by white boxes. (B) Four independent repeat formation events found in the Arabidopsis species examined. SINE-related sequence is inserted in the orientation opposite to the transcription of FWA. Details of the repeats are shown in panel C. (C) Alignment of nucleotide sequences of the region in Figure 1B. Arrows indicate tandem repeats.
In A. thaliana, the FWA gene is silent in vegetative tissues and expressed in the endosperm in an imprinted manner (maternal-origin-specific). Both the silencing of the paternally-derived copy and silencing in vegetative tissues depend on DNA methylation in the FWA promoter.
As is the case in A. thaliana, FWA transcripts were detected in immature seeds in A. lyrata and A. halleri. To determine whether the FWA gene shows imprinted expression in these species, RNA was isolated from immature seeds after inter-strain crosses, and the origin of FWA transcripts was identified using sequence polymorphisms in the transcript. In both A. lyrata and A. halleri, the transcript detected was derived from the allele of maternal origin (Figure 2). For example, in A. lyrata, FWA transcripts of ssp. petraea and ssp. lyrata could be distinguished using polymorphisms in the transcript length (Figure 2A and 2B), and transcript of only the maternal allele could be detected in both of the reciprocal crosses (Figure 2B). The presence of maternal transcripts in seeds may reflect imprinted expression or transmission of transcripts from the female parent. In order to distinguish between these two possibilities, we used a female parent with two FWA alleles that could be distinguished by sequence (Figure 2A), and FWA transcripts were examined in 40 individual seeds. The results show segregation of the two FWA alleles expressed in these seeds; 22 seeds showed FWA RNA from one allele of the female parent and 18 seeds from the other allele (Figure 2C). This observation suggests that the FWA transcripts did not originate from maternal diploid tissues but from transcription of the maternally-derived alleles after meiosis, as is predicted in other imprinted genes . Similarly, in A. halleri, we could only detect FWA transcripts from the maternal allele in both of the reciprocal crosses between ssp. halleri and ssp. gemmifera (Figure 2D, 2E, and 2F). In 20 individual seeds from the ssp. halleri x ssp. gemmifera cross, the 11:9 segregation of the two maternal alleles completely matches the expression pattern (Figure 2E). In summary, these results demonstrate that the FWA in both A. lyrata and A. halleri shows imprinted expression in immature seeds as it does in A. thaliana. These results are striking especially in A. halleri, because the FWA promoter of this species does not contain any tandem repeats (Figure 1). We conclude that the tandem repeat structure is not essential for the imprinting of the FWA gene.
(A) Polymorphisms of the FWA gene in A. lyrata. Two alleles (allele 1 and allele 2) in ssp. petraea and one allele in ssp. lyrata are shown. The ssp. lyrata alleles have 33-bp deletion in the transcribed region compared to the petraea alleles. Differences between the allele 1 and 2 are indicated by a gray box. (B) RT-PCR product of FWA transcript in F1 from inter-strain crosses of both orientations. Mother and father are shown before and after x in each cross, respectively. (C) Segregation of two maternal alleles in individual immature seeds. RNA was prepared from individual immature seed and RT-PCR products were sequenced. (D) Polymorphisms of the FWA gene in A. halleri. (E) Structure of RT-PCR product from immature seeds after crosses between ssp. halleri (mother) and ssp. gemmifera (father). Plants expressing the halleri alleles 1 and 2 segregated. (F) Structure of RT-PCR product from immature seeds after crosses between ssp. gemmifera (mother) and ssp. halleri (father). The gemmifera allele was expressed.
We next examined vegetative FWA expression in these species. Since A. halleri does not have tandem repeats in its FWA promoter, we were able to test whether the tandem repeat is necessary for vegetative silencing. In contrast to A. thaliana, in which the FWA gene is silent in vegetative tissues , a low level of FWA transcript was often detected in the vegetative tissues of A. arenosa, A. lyrata, and A. halleri (Figure 3). Interestingly, vegetative FWA expression shows considerable variation at least at the subspecies level. We could detect the vegetative FWA transcript in two strains of A. lyrata ssp. lyrata but not in a strain of A. lyrata ssp. petraea (Figure 3A, lanes 5 and 6). Similarly, we could detect vegetative FWA transcripts in A. halleri ssp. halleri, tatrica, ovirensis, but not in nine out of eleven strains of ssp. gemmifera (Figure 3B and Figure S3). Therefore, the tandem repeat structure is not essential for epigenetic silencing of FWA in vegetative tissues.
(A) Vegetative FWA expression in Arabidopsis species. 1, A. thaliana (Col-0); 2, A. arenosa; 3, A. halleri strain Tada (ssp. gemmifera); 4, A. halleri strain RB (ssp. halleri); 5, A. lyrata pn3 (ssp. lyrata); 6, A. lyrata strain Mue-1 (ssp. petraea). Lanes 7 and 8 are from the same strains as lane 3 and 6, respectively. In A. halleri Tada (lane 3, 7) and A. lyrata Mue-1 (lane 6, 8), FWA expression was undetectable in leaves, while detectable in pistils. (B) Difference in vegetative FWA expression in six strains of A. halleri, which belong to four subspecies, ssp. gemmifera (Tada, OID06, Mino), ssp. halleri (H-RB), ssp. tatrica (T-PLDH), and ssp. ovirensis (O-AUO). Although vegetative FWA transcription was not detectable in most natural isolates of ssp. gemmifera, it was consistently detectable in other isolates (Figure S3). (C) Vegetative FWA expression in 24 natural strains of A. thaliana. An asterisk indicates natural strains without the 45-nt tandem repeat. Two to four plants were examined for each strain and gave consistent results. fwa-d is a late flowering epigenetic allele induced in a ddm1 mutant background .
The vegetative FWA silencing observed in strains of A. halleri ssp. gemmifera and A. lyrata ssp. petraea does not seems to be due to loss of promoter function, because their FWA gene was expressed in immature seeds (Figure 2F) and in pistils (Figure 3A right panel), which contain female gametophytes. Thus, the vegetative FWA silencing is likely to have an epigenetic basis. The variation in the vegetative FWA expression was heritable; the FWA gene remained silent in the self-pollinated progeny of those plants with silent FWA, while vegetative FWA expression was detected in other strains growing in parallel (not shown).
Allotetraploids and Inter-Specific Hybrids
We further examined FWA expression in two allotetraploid species, A. kamchatica (ssp. kamchatica and kawasakiana) and A. suecica. The former is an allotetraploid between A. lyrata and A. halleri , and the latter between A. arenosa and A. thaliana .
We were able to detect vegetative FWA expression in A. kamchatica (ssp. kamchatica and kawasakiana) (Figure 4A lane 1). As expected from its allotetraploid origin, A. kamchatica has two copies of FWA genes, with one copy being structurally similar to the A. lyrata FWA gene and the other similar to the A. halleri FWA gene (Supplementary Figure S4A and S4C). We examined the expression of each copy using polymorphisms between them. In both ssp. kamchatica and ssp. kawasakiana, the A. halleri-like FWA copy was expressed in leaves, while the A. lyrata-like copy was silent (Figure 4B lane 1). Interestingly, the lyrata-type copy was also silent in pistils, which contain the female gametophytes (Figure 4B lane 2), leaving open the possibility that the silencing of the lyrata-type copy was not epigenetic but genetic (for example, by a mutation in the promoter).
(A) Vegetative FWA transcript in A. kamchatica ssp. kamchatica strain FJSB1 (Ak) and A. suecica JS7 (As). (B) In the A. kamchatica, only the halleri-type copy was transcribed in both leaves and pistils (transcript of the lyrata-type copy was undetectable). The halleri-type and lyrata-type copies were distinguished by length after RsaI digestion of the PCR products. Ah and Al in lanes 3 and 4 are samples isolated from A. halleri Tada and A. lyrata pn3, respectively. (C and D) In A. suecica, only the arenosa-type copy was transcribed in leaves, but both arenosa- and thaliana-type copies were transcribed in pistils, suggesting that the thaliana-type copy was epigenetically silent in leaves. These two types of transcripts were distinguished using specific primer pairs (C) or by restriction digestion (D). See Materials and methods for details.
We were also able to detect vegetative FWA expression in A. suecica (Figure 4C lane 1), an allotetraploid between A. arenosa and A. thaliana. As expected, A. suecica has two copies of FWA genes, which are structurally similar to the FWA gene of either A. arenosa or A. thaliana (Figure S4B and S4C). Vegetative FWA expression of A. suecica was mainly from the arenosa-type copy; we could not detect vegetative expression of the thaliana-type copy (Figure 4C and 4D). On the other hand, both arenosa-type and thaliana-type copies were expressed in pistils (lane 2 of Figure 4C and 4D). Therefore, the thaliana-type copy was specifically silenced in vegetative tissues, as is the case in the parental species A. thaliana.
In the allotetraploids, vegetative FWA expression tends to be stronger than that in their parental species. In order to evaluate the direct effects of hybridization, we examined FWA expression in artificially generated inter-specific hybrids. In hybrids between A. thaliana and A. lyrata ssp. lyrata, expression level of the lyrata FWA in leaves was increased significantly compared to that in its direct parent A. lyrata ssp. lyrata (Figure 5A and 5C). In hybrids between A. thaliana and A. halleri ssp. gemmifera strain Tada, the halleri–like copy was expressed (Figure 5B and 5C). This result is striking considering that the FWA was completely silent in vegetative tissues of the parent strain Tada (Figure 5B, lane 1–3). On the other hand, in both the A. lyrata-A. thaliana and A. halleri-A. thaliana hybrids, the A. thaliana copy remained silent (Figure 5C). In addition, in the synthetic allotetraploid of A. arenosa and A. thaliana (synthetic A. suecica; reference ), the arenosa-type copy was expressed while the thaliana-type copy was silent (Figure 5D). In summary, the FWA genes tend to be transcriptionally activated in the inter-specific hybrids. This phenomenon might be related to previous finding that the imprinted genes, such as PHERES1 and MEDEA, show abnormal expression patterns in hybrids between A. thaliana and A. arenosa . In contrast to the expression of the lyrata-, halleri- or arenosa-derived FWA copies in the hybrids, thaliana-derived copies remained silent in all the hybrids. These observations are consistent with that in the situation in natural allotetraploids; the thaliana-derived copy was most stably silenced in the vegetative tissues of both the allotetraploids and hybrids.
(A) F1 progeny from a cross between A. lyrata pn3 and A. thaliana ms-1 mutant. The F1 hybrids (lanes 4–6) showed increased FWA transcripts compared to their direct parent (lane 1) and its self-pollinated progeny (lanes 2, 3). For both FWA and GAP, lower panels are from PCR reactions with three cycles less than those in the upper panels. The ms-1 mutant was used as the female parent for efficient crossing. Essentially the same results were obtained when wild type Col plant was used for the female parent, or when A. lyrata MN47 was used as the male parent. (B) F1 progeny from a cross between A. halleri Tada (ssp. gemmifera) and A. thaliana ms-1 mutant. The F1 plants showed FWA expression, while this gene is silent in their parents. (C) The FWA expressed in the F1 hybrid was from lyrata–type or from halleri-type. The FWA genes of the thaliana-type and from lyrata– or halleri-type were distinguished using specific primer pairs for the RT-PCR (see Materials and Methods for details). The FWA gene of thaliana-type remained silent in the leaves of the hybrids (lane 1, 2). (D) Expression of FWA in synthetic hybrids between A. arenosa and A. thaliana (Ler for the lane 1, 2 and Col for the lane 3, 4). The FWA gene of thaliana-type remained silent in the leaves of the hybrids.
In A. thaliana, loss of DNA methylation in the FWA promoter induces release of epigenetic silencing. The tandem repeat is methylated in all 96 natural A. thaliana strains . The methylated region of the FWA promoter precisely matched the tandem repeat regions and the methylation extends out of the SINE-related region to the end of the repeat  (Figure 6A). We were therefore interested in whether the epigenetic silencing of the FWA is also related to DNA methylation in Arabidopsis species with less prominent repeat structure.
Ten clones from bisulfite-treated templates were examined for each sample. Red, blue, and black bars represent methylation in CG, CNG, and asymmetric sites, respectively. Blue and red-arrows shows lyrata-specific and arenosa-specific tandem repeats, respectively. Black arrows show regions duplicated in A. thaliana. Gray bars show the SINE-related sequences. The circle shows the transcription start site. A. thaliana used was strain Col. A. halleri ssp. gemmifera used was strain Tada. Strains of the other subspecies of A. halleri were those used in Figure 4.
Figure 6 and Table S1 show bisulfite-mediated genomic sequencing data demonstrating that the SINE-related region of the FWA promoter was methylated in all species examined, including A. halleri, indicating that the repeat structure is not necessary for directing DNA methylation to this region. CG-sites tend to be more heavily methylated than non-CG sites, as is the case in the FWA promoter of A. thaliana.
The methylation level is high throughout the repeat region in A. thaliana, which shows the most stable silencing (Figure 6, Table S1). Interestingly, intraspecific variation in the methylation level was found in A. lyrata (Figure 6B). The methylation level tended to be negatively correlated with the expression level; it was highest in the strain Mue-1 (ssp. petraea), which shows vegetative silencing of the FWA.
Variation of the FWA Promoter within A. thaliana
In A. lyrata, the number of repeats in the FWA promoter correlated with the level of vegetative silencing. In A. lyrata ssp. petraea strain Mue-1, which has four copies of the tandem repeat, the vegetative expression was much lower than that in the two strains of A. lyrata ssp. lyrata that have three copies of the tandem repeat (Figure 3A and Figure S4A). The silencing also correlated with high level of promoter methylation (Figure 7B).
Developmental processes are shown in green, and proposed evolutionary events in blue. See text for details.
In order to test whether the correlation between copy number in the repeats and vegetative FWA silencing is found also in A. thaliana, we examined the structure of the FWA promoter and its expression in 96 natural strains of this species. All 96 strains had the long repeat, but the short repeat was not found in three strains, Fab-4, Var2-1, and Var2-6. We then examined vegetative FWA expression in these strains and, as controls, 21 strains with two pairs of tandem repeats (Figure 3C). Vegetative FWA expression was not detected in any of the 21 control strains, while two out of three strains lacking the short repeat showed a low level of vegetative FWA expression.
Tandem Repeat Structure and the SINE-Related Element
Here we reported variation in the structure and expression of the imprinted gene FWA in the genus Arabidopsis. The promoter sequence related to a SINE retroelement was found in the FWA locus of all Arabidopsis species examined. Most unexpectedly, the tandem repeat structure in this region is not essential for the epigenetic silencing of the FWA gene. The FWA promoter of A. halleri does not have the repeat structure, but it shows imprinted expression and vegetative silencing.
Similarly, tandem repeats are often found in CpG islands of mammalian imprinted genes, but they are not always conserved between mouse and human . In addition, deletion of a conserved direct repeat element upstream of H19 had no effect on imprinting . One possible interpretation of these observations is that tandem duplications may be a consequence, rather than the cause, of mono-allelic expression. Consistent with this interpretation, we found four independent duplication events in the small region of the FWA promoter in closely related species, A. thaliana, A. lyrata, and A. arenosa.
Silencing of the FWA gene tends to be stronger in A. thaliana than in other species. The majority of A. lyrata and A. halleri strains showed vegetative FWA expression. Vegetative FWA expression was also found in examined strains of A. arenosa, A. kamchatica and A. suecica. On the other hand, the FWA gene in A. thaliana was silent in all the examined 21 accessions that have two direct repeats. Vegetative FWA expression tends to be elevated after inter-species hybridization, the clearest example being the FWA gene of the A. halleri strain Tada. Although we could not detect vegetative transcripts in this strain, FWA transcript was detected after hybridization with A. thaliana. On the other hand, vegetative expression was not detectable for the A. thaliana-derived FWA gene, even after hybridization with A. halleri, A. lyrata, or A. arenosa. The thaliana-type FWA was also silent in the natural allotetraploid A. suecica.
Stable silencing of the A. thaliana FWA in vegetative tissues might depend on the presence of tandem repeats in the promoter. Consistent with this conjecture, vegetative FWA expression, although at very low levels, was found in two of three natural accessions that do not have one of the two tandem repeats (Figure 3). The tandem direct repeats may facilitate the production of small RNA and targeting of DNA methylation to stabilize silencing . Small RNA was detected in the FWA promoter of A. thaliana . In this context, it might be interesting to see if small RNA is detectable corresponding to the FWA promoter of A. halleri, which does not have the tandem repeat. Results of modified transgene with various repeat structures of the FWA promoter suggest that the tandem repeat structure is effective in inducing de novo DNA methylation at least in a transgenic system .
Based on these observations, we propose a model for the evolution of the epigenetically controlled FWA gene (Figure 7). Tandem repeat structure does not appear to be essential for the imprinted expression, vegetative silencing, or targeting of DNA methylation in the FWA. Rather, methylation is directed to the SINE-related sequence, which functions as the core of the local heterochromatin. Subsequently, tandem repeats have been recurrently generated in this region during evolution. Repeat formation might be caused after replication arrest  in the epigenetically silent region. In addition, DEMETER DNA demethylase might induce tandem repeat by nick formation  when expressed in the embryonic cell lineage . DNA methylation would have then spread to the region of the tandem repeat, which stabilizes epigenetic silencing. The methylation, especially in CG sites, function as an epigenetic mark heritable over multiple generations (Figure 7).
FWA as a Possible Epigenetic Flowering-Time Modifier
Heritable epigenetic variation is an enigmatic genetic phenomenon, which is known both in mammals and plants ,. The FWA gene in Arabidopsis has the potential to cause epigenetic variation in flowering time, and natural variation in vegetative FWA expression was found in this study (Figure 3).
Sequence analysis of the FWA promoter in 96 natural strains of A. thaliana revealed a high level of variation in this region (Table S2 and Figure S6). C/G to T/A mutations are overrepresented, possibly reflecting high mutation rate in the methylated C to T ,. Despite the high mutation rate, alleles of intermediate frequency are significantly underrepresented in this type of variation (Tajima's D: −1.7078, P<0.05), suggesting that new mutations were rapidly eliminated (Table S2). Negative selection against mutation in the C/G sites may reflect an advantage of stable silencing of the FWA gene by cytosine methylation, especially in A. thaliana, which has a rapid life cycle.
Unlike A. thaliana, other Arabidopsis species showed vegetative FWA expression in the majority of strains. The expression level was heritable but variable within species (Figure 3 and Figure S3). The rapid and potentially reversible changes in epigenetic states might represent an important source of variation in reproductive strategy.
Materials and Methods
The sources of the 96 natural accessions (cs22660) of A. thaliana are described in Nordborg et al. . The epigenetic late-flowering fwa mutant was induced in the ddm1-1 background and backcrossed to DDM1/DDM1 background as described previously . The male-sterile mutant (ms1-1) is a kind gift from Maarten Koornneef. A. lyrata ssp. lyrata pn3 was collected in Pores Knob, Wilkes County, NC. The A. lyrata ssp. lyrata MN47 was developed at Cornell University from material originally isolated by Charles Langley in Michigan, USA. The A. lyrata ssp. petraea strain Mue-1, was collected in Muehlberg, Germany. A. halleri ssp. gemmifera Tada, OID06-6, and Mino were isolated in Inagawa-cho, Osaka, Japan, Taka-cho, Hyogo, Japan, and Mino-shi, Osaka, Japan. A. halleri ssp. halleri H-RB was isolated in Bistrita, Romania, ssp. tatrica T-PLDH1 in Vysoke Tatry, Poland, ssp. ovirensis O-AUO26 in Karinthia, Austria. A. kamchatica ssp. kamchatica FJSB1 was isolated in Subashiri, Shizuoka, Japan, and ssp. kawasakiana Shirahama was isolated in Ohmi-shirahama, Shiga, Japan. A. suecica was from Sendai Arabidopsis Seed Stock Center (JS7). Synthetic suecica (CS22665 and CS22666) and A. arenosa (CS3901) are from Arabidopsis Biological Resource Center at Ohio State University.
Cloning and Sequencing of DNA Fragments of FWA
Sequences of all primers used are shown in Supplementary Text S1. Genomic DNAs were isolated from leaves by Nucleon (Amersham Biosciences, UK). The DNA fragments spanning the FWA gene in related species were amplified by PCR and cloned into pGEM-TEasy or pCR2.1-TOPO vector using the primers lyrata-7f and lyrata-10r for lyrata and arenosa alleles, lyrata-7f+lyrata-9r for halleri alleles, FWApro1+As-3 for thaliana alleles. In each strain, multiple clones were sequenced and regions of inconsistent sequences were directly sequenced from the genomic DNA again. Sequences of both alleles were determined in out-crossing species, such as A. halleri and A. lyrata. 96 ecotypes of the tandem repeat region was amplified using primer pair Ateco96F+Ateco96R, and their nucleotide sequences were determined directly from the genomic DNA. Data were analyzed using Sequencher (Gene Codes Corporation, MI, USA). Harr plot analysis was performed with GENETYX-MAC software (Software Development Co., LTD, Tokyo, Japan). Unit size to compare was 10 bp, and dot plot matching number was 9 bp . The sequence alignment was made using ClustalW (http://www.ddbj.nig.ac.jp/search/clustalw-j.html). A phylogenetic tree was constructed by the neighbor joining method with K2P distance , and bootstrap probabilities of 1,000 trials were calculated.
Gene Expression Analysis
Total RNA was isolated from leaves and pistils from open flowers using SV Total RNA Isolation system (Promega, WI, USA). From 2 µg total RNAs, first strand cDNA was synthesized using random primers by a First-strand cDNA synthesis Kit (Amersham, NJ, USA). FWA transcript was detected by RT-PCR of 40 cycles using the first strand cDNA as a template with primer pair FWA-RT-F2+FWA-RT-R1. GAPC was used as control and amplified by RT-PCR of 25 cycles using primer pair GAP3+GAP5. In A. suecica, allele-specific RT-PCR was performed using primer pair specific for FWA in A. thaliana (AtFWA-RT-F1+AtFWA-RT-R1) and specific for FWA in A. arenosa (FWA-RT-F5+AtFWA-RT-R4). In hybrid between A. lyrata/A. halleri and A. thaliana, allele specific primer pair was used for RT-PCR to amplify FWA in A. lyrata/A. halleri (FWA-RT-F5+AtFWA-RT-R5) and that of A. thaliana (AtFWA-RT-F1+AtFWA-RT-R1). The PCR condition was 95°C for 10 sec followed by 25 or 40 cycles of 95°C for 30 sec, 57°C for 30 sec, and 72°C for 30 sec.
Total RNAs from one premature fertilized seed after 15DAP in A. lyrata and A. halleri were isolated by RNAqueous-Micro (Ambion, Texas, USA). Using all of isolated total RNAs, first strand cDNA was synthesized using FWA-specific primer AlFWAcDNA-R by SuperScript III Reverse Transcriptase (Invitrogen). FWA transcripts in seed of A. lyrata and A. halleri were amplified by RT-PCR of 35 cycles using the first strand cDNA as a template with primer pairs (AhAlFWAcDNA-F2+lyrata2r) and (FWA-RT-F3+FWA-RT-R3), respectively. Products from genomic DNA and mRNA could be distinguished by size.
FWA transcript level was compared between two strains of A. kamchatica by real-time PCR (Figure S5) with the GAP gene used as the internal control. The cDNA was amplified using SYBR Premix Ex Taq (Takara Biomedicals, Japan) on a LightCycler (Roche, Rotkreuz, Switzerland). With the primer pairs FWAreal1+FWA-RT-R1 (FWA) and GAPreal2+GAPreal3 (GAP). The PCR condition was 95 for 10sec followed by 25 (GAP) or 40 (FWA) cycles of 95°C for 5 sec, 55°C for 10 sec, and 72°C for 7 sec. Data were analyzed with LightCycler Software version 3.5 (Roche).
Detection of DNA Methylation by the Bisulfite Method
Bisulfite sequencing was performed according to Paulin et al. . After chemical bisulfite reaction, PCR fragments of repeat region of FWA were amplified using primer pairs as follow. AtFWA; (AtFWA-Bis-F1+AtFWA-Bis-R1), AaFWA; (AhFWA-Bis-F1+arenosaRTbisR), AlFWA; (arenosaRTbisF+lyrataRTbisR), and AhFWA; (AhFWA-Bis-F1+AhFWA-Bis-R2). The amplified PCR fragments were gel purified by GENECLEAN III Kit (Q-Biogene, CA, USA) and cloned into pGEM-T easy vector (Promega, WI, USA), and 10 independent clones were sequenced.
Supplementary materials and methods.
(0.03 MB PDF)
FWA genes in Arabidopsis species. Harr plot analysis shows that not only exons (black boxes) but also promoter and introns are conserved. A big gap found in each panel reflects thaliana-specific big tandem duplication. Gray boxes indicate the regions related to the SINE.
(0.05 MB PDF)
Harr plot analysis of 5′ region of the FWA gene to detect tandem repeats. Black boxes and gray bar indicate exons and the SINE-related region, respectively. In the SINE-related regions, two, three, and four tandem repeats were found in A. arenosa, A. lyrata ssp. lyrata, and A. lyrata ssp. petraea, respectively. Two pairs of tandem repeats were found in A. thaliana. No tandem repeat was found A. halleri.
(0.08 MB PDF)
Variation in vegetative FWA expression in eleven plants of A. halleri ssp. gemmifera isolated around Kyoto area in Japan. (A) The vegetative FWA expression was detectable in two isolates, while it was undetectable in the other nine isolates. (B) Same expression pattern was detected in different individuals isolated from the same area. (C) Map of the locations these plants were isolated. Abbreviation of the isolates: AK (Akebe), MH (Mikohata), IK (Ikuno), OM (Omoidegawa, Taka), RK (Rokko), TD (Tada), MK (Myouken), MN (Mino), OH (Ohara), IB (Mt. Ibuki), OE (Mt. Ooe). The plants were isolated from the field and were grown in the laboratory. Three of them, Mino, Omoidegawa, and Tada, were further characterized after isolation of the seeds and growth in the laboratory (Figure 3B).
(0.12 MB PDF)
Structure of FWA genes in allotetraploids and parental species. (A) Two FWA copies in A. kamchatica. Each of them was similar to that in A. halleri (left panel) or A. lyrata (right panel). The similarity was found not only in exons (black boxes) but also promoter and introns. Gray boxes indicate the SINE-related regions. Copy number of tandem repeats in the lyrata-like copy was three as is the case in ssp. lyrata. (B) Two FWA copies in A. suecica. Each of them was similar to that in A. thaliana (left panel) or A. arenosa (right panel). (C) A phylogenetic tree of nucleotide sequences of full length cDNA of FWA in the Arabidopsis species constructed by the neighbor-joining method. Arabis glabra FWA was used as the outgroup. Bootstrap values with 1,000 repeats are indicated at the nodes of the neighbor-joining tree.
(0.08 MB PDF)
Differences in promoter structure and expression pattern of the halleri-type FWA gene in two strains of in A. kamchatica, SHR1 (ssp. kawasakiana), and FJSB1 (ssp. kamchatica). (A) Promoter alignment revealed insertion of a possible transposable element in the SHR1 allele. Blue and red letters indicate the predicted target site duplication and terminal inverted repeat, respectively. This insertion was found in majority of A. kamchatica, ssp. kawasakiana strains isolated in Japan (not shown). Gray boxes: the SINE-related sequences. Green letter shows the target site duplication of the SINE insertion. The circle shows the transcription start site. Black arrows show regions duplicated in A. thaliana. (B) Methylation status of the halleri-type FWA gene in the two strains. The predicted transposon (white box) was heavily methylated in SHR1. (C) Expression analysis of the FWA gene. Related FWA expression level measured by real-time PCR are shown below the gel. The transposon insertion and increase in the DNA methylation correlates with reduction in the FWA transcript level in SHR1.
(0.04 MB PDF)
Sliding window analyses of silent divergence and diversity in the FWA gene. The analyses are based on the 12 strains of A. thaliana. K and p in 200 bp silent sites are plotted in 1 bp intervals. Under the sliding window plot, exons (from third to last) are shown by black boxes and repeated regions by arrows.
(0.02 MB PDF)
Proportion of methylated cytosine around the FWA promoter.
(0.04 MB PDF)
We thank Akiko Terui for technical assistance. Special thanks to Eric Richards and Hiroyuki Sasaki for critical comments on the manuscript. We acknowledge Arabidopsis Biological Resource Center at Ohio State University for the seed stocks.
Conceived and designed the experiments: R. Fujimoto, M. Nordborg, T. Kakutani. Performed the experiments: R. Fujimoto, Y. Kinoshita, T. Kinoshita, K. Takashima, M. Nordborg, T. Kakutani. Analyzed the data: R. Fujimoto, Y. Kinoshita, A. Kawabe, T. Kinoshita, K. Takashima, M. Nordborg, T. Kakutani. Contributed reagents/materials/analysis tools: R. Fujimoto, Y. Kinoshita, A. Kawabe, M. Nasrallah, K. Shimizu, H. Kudoh, T. Kakutani. Wrote the paper: R. Fujimoto, T. Kakutani.
- 1. Li E, Beard C, Jaenisch R (1993) Role for DNA methylation in genomic imprinting. Nature 366: 1008–1020.E. LiC. BeardR. Jaenisch1993Role for DNA methylation in genomic imprinting.Nature36610081020
- 2. Kaneda M, Okano M, Hata K, Sado T, Tsujimoto N, et al. (2004) Essential role for de novo DNA methyltransferase Dnmt3a in paternal and maternal imprinting. Nature 429: 900–903.M. KanedaM. OkanoK. HataT. SadoN. Tsujimoto2004Essential role for de novo DNA methyltransferase Dnmt3a in paternal and maternal imprinting.Nature429900903
- 3. Kinoshita T, Miura A, Choi Y, Kinoshita Y, Cao X, et al. (2004) One-way control of FWA imprinting in Arabidopsis endosperm by DNA methylation. Science 303: 521–523.T. KinoshitaA. MiuraY. ChoiY. KinoshitaX. Cao2004One-way control of FWA imprinting in Arabidopsis endosperm by DNA methylation.Science303521523
- 4. Barlow DP (1993) Methylation and imprinting: from host defense to gene regulation? Science 260: 309–310.DP Barlow1993Methylation and imprinting: from host defense to gene regulation?Science260309310
- 5. Nabetani A, Hatada I, Morisaki H, Oshimura M, Mukai T (1997) Mouse U2af1-rs1 is a neomorphic imprinted gene. Molecular and cellular biology 17: 789–798.A. NabetaniI. HatadaH. MorisakiM. OshimuraT. Mukai1997Mouse U2af1-rs1 is a neomorphic imprinted gene.Molecular and cellular biology17789798
- 6. Lippman Z, Gendrel AV, Black M, Vaughn MW, Dedhia N, et al. (2004) Role of transposable elements in heterochromatin and epigenetic control. Nature 430: 471–476.Z. LippmanAV GendrelM. BlackMW VaughnN. Dedhia2004Role of transposable elements in heterochromatin and epigenetic control.Nature430471476
- 7. Suzuki S, Ono R, Narita T, Pask AJ, Shaw G, et al. (2007) Retrotransposon silencing by DNA methylation can drive mammalian genomic imprinting. PLoS Genetics 3: e55.S. SuzukiR. OnoT. NaritaAJ PaskG. Shaw2007Retrotransposon silencing by DNA methylation can drive mammalian genomic imprinting.PLoS Genetics3e55
- 8. Yoder JA, Walsh CP, Bestor TH (1997) Cytosine methylation and the ecology of intragenomic parasites. Trends Genet 13: 335–340.JA YoderCP WalshTH Bestor1997Cytosine methylation and the ecology of intragenomic parasites.Trends Genet13335340
- 9. Miura A, Yonebayashi S, Watanabe K, Toyama T, Shimada H, et al. (2001) Mobilization of transposons by a mutation abolishing full DNA methylation in Arabidopsis. Nature 411: 212–214.A. MiuraS. YonebayashiK. WatanabeT. ToyamaH. Shimada2001Mobilization of transposons by a mutation abolishing full DNA methylation in Arabidopsis.Nature411212214
- 10. Singer T, Yordan C, Martienssen RA (2001) Robertson's Mutator transposons in A. thaliana are regulated by the chromatin-remodeling gene Decrease in DNA Methylation (DDM1). Genes Dev 15: 591–602.T. SingerC. YordanRA Martienssen2001Robertson's Mutator transposons in A. thaliana are regulated by the chromatin-remodeling gene Decrease in DNA Methylation (DDM1).Genes Dev15591602
- 11. Kato M, Miura A, Bender J, Jacobsen SE, Kakutani T (2003) Role of CG and non-CG methylation in immobilization of transposons in Arabidopsis. Curr Biol 13: 421–426.M. KatoA. MiuraJ. BenderSE JacobsenT. Kakutani2003Role of CG and non-CG methylation in immobilization of transposons in Arabidopsis.Curr Biol13421426
- 12. Lippman Z, May B, Yordan C, Singer T, Martienssen R (2003) Distinct mechanisms determine transposon inheritance and methylation via small interfering RNA and histone modification. PLoS Biol 1: E67.Z. LippmanB. MayC. YordanT. SingerR. Martienssen2003Distinct mechanisms determine transposon inheritance and methylation via small interfering RNA and histone modification.PLoS Biol1E67
- 13. Selker EU (2004) Genome defense and DNA methylation in Neurospora. Cold Spring Harbor symposia on quantitative biology 69: 119–124.EU Selker2004Genome defense and DNA methylation in Neurospora.Cold Spring Harbor symposia on quantitative biology69119124
- 14. Gimelbrant AA, Chess A (2006) An epigenetic state associated with areas of gene duplication. Genome Research 16: 723–729.AA GimelbrantA. Chess2006An epigenetic state associated with areas of gene duplication.Genome Research16723729
- 15. Lewis A, Mitsuya K, Constancia M, Reik W (2004) Tandem repeat hypothesis in imprinting: deletion of a conserved direct repeat element upstream of H19 has no effect on imprinting in the Igf2-H19 region. Molecular and cellular biology 24(13): 5650–5656.A. LewisK. MitsuyaM. ConstanciaW. Reik2004Tandem repeat hypothesis in imprinting: deletion of a conserved direct repeat element upstream of H19 has no effect on imprinting in the Igf2-H19 region.Molecular and cellular biology24(13)56505656
- 16. Neumann B, Kubicka P, Barlow DP (1995) Characteristics of imprinted genes. Nature Genetics 9: 12–13.B. NeumannP. KubickaDP Barlow1995Characteristics of imprinted genes.Nature Genetics91213
- 17. Yoon BJ, Herman H, Sikora A, Smith LT, Plass C, et al. (2002) Regulation of DNA methylation of Rasgrf1. Nature Genetics 30: 92–96.BJ YoonH. HermanA. SikoraLT SmithC. Plass2002Regulation of DNA methylation of Rasgrf1.Nature Genetics309296
- 18. Koornneef M, Hanhart CJ, van der Veen JH (1991) A genetic and physiological analysis of late flowering mutants in Arabidopsis thaliana. Mol Gen Genet 229: 57–66.M. KoornneefCJ HanhartJH van der Veen1991A genetic and physiological analysis of late flowering mutants in Arabidopsis thaliana.Mol Gen Genet2295766
- 19. Kakutani T (1997) Genetic characterization of late-flowering traits induced by DNA hypomethylation mutation in Arabidopsis thaliana. Plant J 12: 1447–1451.T. Kakutani1997Genetic characterization of late-flowering traits induced by DNA hypomethylation mutation in Arabidopsis thaliana.Plant J1214471451
- 20. Soppe WJ, Jacobsen SE, Alonso-Blanco C, Jackson JP, Kakutani T, et al. (2000) The late flowering phenotype of fwa mutants is caused by gain-of-function epigenetic alleles of a homeodomain gene. Molecular Cell 6: 791–802.WJ SoppeSE JacobsenC. Alonso-BlancoJP JacksonT. Kakutani2000The late flowering phenotype of fwa mutants is caused by gain-of-function epigenetic alleles of a homeodomain gene.Molecular Cell6791802
- 21. Kankel MW, Ramsey DE, Stokes TL, Flowers SK, Haag JR, et al. (2003) Arabidopsis MET1 cytosine methyltransferase mutants. Genetics 163: 1109–1122.MW KankelDE RamseyTL StokesSK FlowersJR Haag2003Arabidopsis MET1 cytosine methyltransferase mutants.Genetics16311091122
- 22. Saze H, Scheid OM, Paszkowski J (2003) Maintenance of CpG methylation is essential for epigenetic inheritance during plant gametogenesis. Nat Genet 34: 65–69.H. SazeOM ScheidJ. Paszkowski2003Maintenance of CpG methylation is essential for epigenetic inheritance during plant gametogenesis.Nat Genet346569
- 23. Choi Y, Gehring M, Johnson L, Hannon M, Harada JJ, et al. (2002) DEMETER, a DNA glycosylase domain protein, is required for endosperm gene imprinting and seed viability in Arabidopsis. Cell 110: 33–42.Y. ChoiM. GehringL. JohnsonM. HannonJJ Harada2002DEMETER, a DNA glycosylase domain protein, is required for endosperm gene imprinting and seed viability in Arabidopsis.Cell1103342
- 24. Gehring M, Huh JH, Hsieh TF, Penterman J, Choi Y, et al. (2006) DEMETER DNA glycosylase establishes MEDEA polycombe gene self-imprinting by allele-specific demethylation. Cell 124: 495–506.M. GehringJH HuhTF HsiehJ. PentermanY. Choi2006DEMETER DNA glycosylase establishes MEDEA polycombe gene self-imprinting by allele-specific demethylation.Cell124495506
- 25. Kinoshita Y, Saze H, Kinoshita T, Miura A, Soppe WJJ, et al. (2007) Control of FWA gene silencing in Arabidopsis thaliana by SINE-related direct repeats. Plant J 49: 38–45.Y. KinoshitaH. SazeT. KinoshitaA. MiuraWJJ Soppe2007Control of FWA gene silencing in Arabidopsis thaliana by SINE-related direct repeats.Plant J493845
- 26. Chaudhury AM, Berger F (2001) Maternal control of seed development. Semin Cell Dev Biol 12: 381–386.AM ChaudhuryF. Berger2001Maternal control of seed development.Semin Cell Dev Biol12381386
- 27. Shimizu K, Fuji S, Marhold K, Watanabe K, Kudoh H (2005) Arabidopsis kamchatica (Fisch. ex DC.) K. Shimizu & Kudoh and A. kamchatica subsp. kawasakiana (Makino) K. Shimizu & Kudoh, Ne Combinations. Acta Phytotax Geobot 56: 163–172.K. ShimizuS. FujiK. MarholdK. WatanabeH. Kudoh2005Arabidopsis kamchatica (Fisch. ex DC.) K. Shimizu & Kudoh and A. kamchatica subsp. kawasakiana (Makino) K. Shimizu & Kudoh, Ne Combinations.Acta Phytotax Geobot56163172
- 28. Kamm A, Galasso I, Schmidt T, Heslop-Harrison JS (1995) Analysis of a repetitive DNA family from Arabidopsis arenosa and relationships between Arabidopsis species. Plant Mol Biol 27: 853–862.A. KammI. GalassoT. SchmidtJS Heslop-Harrison1995Analysis of a repetitive DNA family from Arabidopsis arenosa and relationships between Arabidopsis species.Plant Mol Biol27853862
- 29. Madlung A, Tyagi AP, Watson B, Jiang H, Kagochi T, et al. (2005) Genomic changes in synthetic Arabidopsis polyploids. Plant J 41: 221–230.A. MadlungAP TyagiB. WatsonH. JiangT. Kagochi2005Genomic changes in synthetic Arabidopsis polyploids.Plant J41221230
- 30. Josefsson C, Dikes B, Comai L (2006) Parent-dependent loss of gene silencing during interspecies hybridization. Curr Biol 16: 1322–1328.C. JosefssonB. DikesL. Comai2006Parent-dependent loss of gene silencing during interspecies hybridization.Curr Biol1613221328
- 31. Vaughn MW, Tanurd IM, Lippman Z, Jiang H, Carrasquillo R, et al. (2007) Epigenetic natural variation in Arabidopsis thaliana. PLoS Biol 5: e174.MW VaughnIM TanurdZ. LippmanH. JiangR. Carrasquillo2007Epigenetic natural variation in Arabidopsis thaliana.PLoS Biol5e174
- 32. Hutter B, Helms V, Paulsen M (2006) Tandem repeats in the CpG islands of imprinted genes. Genomics 88: 323–332.B. HutterV. HelmsM. Paulsen2006Tandem repeats in the CpG islands of imprinted genes.Genomics88323332
- 33. Martienssen RA (2003) Maintenance of heterochromatin by RNA interference of tandem repeats. Nat Genet 35: 213–214.RA Martienssen2003Maintenance of heterochromatin by RNA interference of tandem repeats.Nat Genet35213214
- 34. Chan SW, Zhang X, Bernatavichute YV, Jacobsen SE (2006) Two-step recruitment of RNA-directed DNA methylation to tandem repeats. PLoS Biol 4: e363.SW ChanX. ZhangYV BernatavichuteSE Jacobsen2006Two-step recruitment of RNA-directed DNA methylation to tandem repeats.PLoS Biol4e363
- 35. Rothstein R, Michel B, Gangloff S (2000) Replication fork pausing and recombination or “gimme a break”. Genes & Dev 14: 1–10.R. RothsteinB. MichelS. Gangloff2000Replication fork pausing and recombination or “gimme a break”.Genes & Dev14110
- 36. Mathieu O, Reinders J, Caikavski M, Smathajitt C, Paszkowski J (2007) Transgenerational stability of the Arabidopsis epigenome is coordinated by CG methylation. Cell 130: 851–862.O. MathieuJ. ReindersM. CaikavskiC. SmathajittJ. Paszkowski2007Transgenerational stability of the Arabidopsis epigenome is coordinated by CG methylation.Cell130851862
- 37. Cubas P, Vincent C, Coen E (1999) An epigenetic mutation responsible for natural variation in floral symmetry. Nature 401: 157–161.P. CubasC. VincentE. Coen1999An epigenetic mutation responsible for natural variation in floral symmetry.Nature401157161
- 38. Rakyan VK, Blewitt ME, Drunker R, Preis JI, Whitelaw E (2002) Metastable epialleles in mammals. Trends Genet 18: 348–351.VK RakyanME BlewittR. DrunkerJI PreisE. Whitelaw2002Metastable epialleles in mammals.Trends Genet18348351
- 39. Bird AP (1980) DNA methylation and the frequency of CpG in animal DNA. Nucleic Acids Res 8: 1499–1594.AP Bird1980DNA methylation and the frequency of CpG in animal DNA.Nucleic Acids Res814991594
- 40. Duncan BK, Miller JH (1980) Mutagenic deamination of cytosine residues in DNA. Nature 287: 560–561.BK DuncanJH Miller1980Mutagenic deamination of cytosine residues in DNA.Nature287560561
- 41. Nordborg M, Hu TT, Ishino Y, Jhaveri J, Toomajian C, et al. (2005) The pattern of polymorphism in Arabidopsis thaliana. PLoS Biol 3: e196.M. NordborgTT HuY. IshinoJ. JhaveriC. Toomajian2005The pattern of polymorphism in Arabidopsis thaliana.PLoS Biol3e196
- 42. Harr R, Hagblom P, Gustafsson P (1982) Two-dimensional graphic analysis of DNA sequence homologies. Nucleic Acids Research 10: 365–374.R. HarrP. HagblomP. Gustafsson1982Two-dimensional graphic analysis of DNA sequence homologies.Nucleic Acids Research10365374
- 44. Paulin R, Grigg GW, Davey MW, Piper AA (1998) Urea improves efficiency of bisulphite-mediated sequencing of 5′-methylcytosine in genomic DNA. Nucleic Acids Research 26: 5009–5010.R. PaulinGW GriggMW DaveyAA Piper1998Urea improves efficiency of bisulphite-mediated sequencing of 5′-methylcytosine in genomic DNA.Nucleic Acids Research2650095010
- 46. Saitou N, Nei M (1987) The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol 4: 406.N. SaitouM. Nei1987The neighbor-joining method: a new method for reconstructing phylogenetic trees.Mol Biol Evol4406