DNA methylation is not necessary for the inactivation of the Tam3 transposon at non-permissive temperature in Antirrhinum

It has been proposed that DNA methylation plays an important role in the inactivation of transposons. This view stems from a comparison of the degree of methylation of transposons in the active and inactive state. However, direct evidence for the degree of methylation required for the suppression of transposition has not been reported. Transposon Tam3 in Antirrhinum majus undergoes somatic reversal of its transposition activity, which is tightly controlled by temperature: low temperature around 15 degrees C permits transposition, high temperatures around 25 degrees C strongly inhibits it. Our previous study had shown that the methylation state of the Tam3 end regions is negatively correlated with the Tam3 transposition frequency. The results of the present study reveal that the inactive state of Tam3 copies at high temperature is unlikely to be directly coupled to the methylation state. Treatment with methylation inhibitors (5-azacytidine or 5-azacytidine+ethionine) does not affect Tam3 excision frequency in calli derived from Antirrhinum hypocotyls. The results suggest that methylation is not essential for the suppression of Tam3 transposition at high temperature, but rather that some other mechanism(s) involved in the control of Tam3 transposition may be obscured by methylation.     

The temperature-dependent change in methylation of the Antirrhinum transposon Tam3 is controlled by the activity of its transposase

The Antirrhinum majus transposon Tam3 undergoes low temperature-dependent transposition (LTDT). Growth at 15 degrees C permits transposition, whereas growth at 25 degrees C strongly suppresses it. The degree of Tam3 DNA methylation is altered somatically and positively correlated with growth temperature, an exceptional epigenetic system in plants. Using a Tam3-inactive line, we show that methylation change depends on Tam3 activity. Random binding site selection analysis and electrophoretic mobility shift assays revealed that the Tam3 transposase (TPase) binds to the major repeat in the subterminal regions of Tam3, the site showing the biggest temperature-dependent change in methylation state. Methylcytosines in the motif impair the binding ability of the TPase. Proteins in a nuclear extract from plants grown at 15 degrees C but not 25 degrees C bind to this motif in Tam3. The decrease in Tam3 DNA methylation at low temperature also requires cell division. Thus, TPase binding to Tam3 occurs only during growth at low temperature and immediately after DNA replication, resulting in a Tam3-specific decrease in methylation of transposon DNA. Consequently, the Tam3 methylation level in LTDT is regulated by Tam3 activity, which is dependent on the ability of its TPase to bind DNA and affected by growth temperature. Thus, the methylation/demethylation of Tam3 is the consequence, not the cause, of LTDT.     

Resistance to gap repair of the transposon Tam3 in Antirrhinum majus: a role of the end regions

The extremely homogeneous organization of the transposon family Tam3 in Antirrhinum majus is in sharp contrast to the heterogeneity of the copies constituting many other transposon families. To address the issue of the Tam3 structural uniformity, we examined two possibilities: (1) recent invasion of Tam3 and (2) failure of gap repair, which is involved in conversion from autonomous forms to defective forms. The phylogenetic analysis of 17 Tam3 copies suggested that the invasion of Tam3 into the Antirrhinum genome occurred at least 5 mya, which is sufficiently long ago to have produced many aberrant copies by gap repair. Thus, we investigated gap repair events at the nivea(recurrens:Tam3) (niv(rec)::Tam3) allele, where Tam3 is actively excised. We show here that the gap repair of de novo somatic Tam3 excision was arrested immediately after initiation of the process. All of the identified gap repair products were short stretches, no longer than 150 bp from the ends. The Tam3 ends have hairpin structures with low free energies. We observed that the gap repair halted within the hairpin structure regions. Such small gap repair products appear to be distributed in the Antirrhinum genome, but are unlikely to be active. Our data strongly suggest that the structural homogeneity of Tam3 was caused by immunity to gap repair at the hairpins in both of the end regions. The frequency of extensive gap repair of de novo excision products in eukaryotic transposons was found to be correlated with the free energies of the secondary structures in the end regions. This fact suggests that the fates of transposon families might depend on the structures of their ends.     

Position effect of the excision frequency of the Antirrhinum transposon Tam3: implications for the degree of position-dependent methylation in the ends of the element

We identified eight independent Tam3 copies residing in the same Antirrhinum majus genome. All the copies showed excision at 15 °C, but not at 25 °C. Under conditions promoting excision, each copy appeared to transpose in the leaves and flower lobes with a nearly constant frequency, whereas individual transposition abilities varied widely: the most active copy had an excision frequency more than 100-fold greater than that of the least active one. Despite the different transposition abilities, the structures of the eight Tam3 copies were almost identical. These results made it clear that the transpositional ability of Tam3 is regulated by chromosomal position, but they do not imply position-dependent transposase activity. The position effect of the Tam3 transposition was found to be correlated to the methylation state of the copy's end regions: DNA methylation in the Tam3 end regions tended to suppress the excision activity, and the degree of methylation was dependent on the chromosomal position. Our results also provide evidence of de novo methylation provoked by transposition of the endogenous element. We propose a mechanism of transpositional regulation of plant transposons that responds to the degree of methylation as determined by chromosomal position.     

Structural conservation of the transposon Tam3 family in Antirrhinum majus and estimation of the number of copies able to transpose

We have investigated the organization of the transposon Tam3 family in Antirrhinum majus. Genomic hybridization experiments and characterization of 40 independent Tam3 clones isolated from an A. majus plant revealed that the Tam3 family is quite conserved and the copy sizes are uniform. We did not find any copy with a deleted internal sequence, unlike what is usually observed in other transposons. This exceptionally conserved structure of the Tam3 family was confirmed by PCR and sequencing analyses. Sequencing analysis identified eight copies with sequences completely identical to that of the Tam3 transposase gene. These results suggested that a considerable number of autonomous Tam3 copies are present in the genome of A. majus. Among 24 copies which are surrounded by single copy regions of the genome, 14 copies are present as specific insertions in the line which we used, but absent in other lines. These copies are therefore predicted to be movable. If this ratio is the same for all Tam3 copies in a genome, then a maximum of 60% of the copies are estimated to be movable in the genome. The relatively high frequency of gene tagged by Tam3 might reflect the large number of movable copies in the genome.     

Genome juggling by transposons: Tam3-induced rearrangements in Antirrhinum majus

Transposable elements are well known for their ability to generate large- and small-scale rearrangements of the sequences flanking their insertion sites. These include deletions, inversions, and duplications. Tam3, a transposon from the Snapdragon (Antirrhinum majus), is highly active in the generation of such rearrangements. We have analysed a number of Tam3-induced rearrangements at the nivea (niv) locus by Southern blotting, cloning, and sequence determination. The data obtained from these analyses have led to an understanding of the mechanisms by which these complex alleles were formed. We have shown that the primary rearrangements usually occur without excision of the element and therefore result from aberrant transposition attempts. Subsequent rearrangements may occur on excision of the element. Finally, we suggest how the analysis of such rearrangements may not only provide information about Tam3 transposition but also show how transposon-induced rearrangements may influence the structure and function of the genome as a whole.     

Immobilized copies with a nearly intact structure of the transposon Tam3 in Antirrhinum majus: implications for the cis-element related to the transposition

 In this study we have focused on two copies of the transposon Tam3 isolated from an Antirrhinum majus plant which has flower variegation due to the excision of Tam3 from the nivea locus. These two copies possess a high homology, over 95%, to an active Tam3 element found in the nivea ^{recurrence:Tam3} allele. Although somatic excision of the Tam3 copy from the nivea locus can be detected at 15°C by Southern blotting, neither of the two copies showed any sign of the excision. Both of the immobilized copies were also found in five varieties from different A. majus sources, all of which contain common fragments. The results suggest that the two copies have been fixed in the genomes of many A. majus varieties. Structural differences between these immobilized copies and the known active copy were mainly observed in the subterminal regions, including the terminal inverted repeats. The immobility of the two Tam3 copies might be due to mutations within the end regions of essential cis-elements in Tam3 transposition, as reported for Ac and En/Spm. Received: 30 June 1997 / Accepted: 5 August 1997     

Non-symmetrical cytosine methylation in tobacco pollen DNA

We have detected sequence-specific non-symmetrical cytosine methylation within a 140 bp region of the promoter for the tobacco auxin-binding protein gene T85 in pollen DNA. Direct sequencing of the population of bisulphite reaction products showed that, in this region, 10 out of a possible 49 cytosine residues were methylated at a high frequency in pollen whereas the corresponding region from somatic cells (leaf DNA) did not show a detectable level of methylation. The context of these sites was 1×m⁵CpTpC, 1×m⁵CpGpT, 1×m⁵CpCpT, 2×m⁵CpTpT, 2×m⁵CpGpG, and 3×m⁵CpApT of which only m⁵CpGpG and m⁵CpGpT fitted the consensus sequence for symmetrical methylation in plants.     

Studies on the biological role of DNA methylation: V. The pattern of E.coli DNA methylation.

The distribution of the methylatable sites GATC and CCATGG was studied by analyzing the molecular average size of restriction fragments of E. coli DNA. Both sites were found to be randomly distributed, reflecting a random pattern of methylation. The methylation pattern of specific sequences such as the origin of replication and rRNA genes has been studied in wild type E. coli and a methylation deficient (dam- dcm-) mutant. These sequences were found to be methylated in wild type cells and unmethylated in the mutant indicating that there is no effect of the state of methylation of these sequences on their expression. Analysis of the state of methylation of GATC sites in newly replicating DNA using the restriction enzyme Dpn I (cleaves only when both strands are methylated) revealed no detectable hemimethylated DNA suggesting that methylation occurs at the replication fork. Taking together the results presented here and previously published data (5), we arrive at the conclusion that the most likely function of E. coli DNA methylations is probably in preventing nuclease activity.     

DNA methylation in diploid inbred lines of potatoes and its possible role in the regulation of heterosis

Self-incompatible diploid potatoes were altered to self-compatible ones by a function of S-locus inhibitor gene and continued selfing generated highly homozygous inbreds. In this study, this process was investigated for the status of DNA methylation by a simple method using genomic DNA digested by methylation-sensitive restriction enzymes prior to RAPD analysis. We detected 31 methylation-sensitive RAPD bands, of which 11 were newly appeared in the selfed progenies, and 6 of them stably inherited to subsequent generations. Aberrant segregations and paternal- or atavism-like transmission were also found. Segregating methylation-sensitive bands in initial populations became fixed in the advanced selfed progenies by 75.0–93.8%, of which 41.7% were fixed to all present and 58.3% to all absent. Because DNA methylation is generally recognized to suppress gene expression as regulatory factors, homozygosity/heterozygosity of methylated DNA may be involved in inbreeding depression/heterosis.     

A possible role of chromatin and tightly-bound chromatin proteins on enzyme-catalyzed methylation of DNA

Upon extensive digestion with DNAaseI of placenta chromatin matrix, previously "stripped" from its loosely-bound components by high-salt extraction, a fraction is obtained that contains almost no endogenous DNA methylase activity but whose DNA, if still included in this whole fraction--not if it has been purified to a protein-free condition--is a good substrate for externally added enzyme. This chromatin matrix can even cause a significant stimulation of methylation of single-stranded Micrococcus luteus DNA by placental methylase. In vivo, this phenomenon may have possible counterparts in the existence of highly-methylated regions of chromatin loops that appear to be protected by tightly-bound protein components from digestion of the "stripped loops" with DNAaseI.     

In vitro DNA methylation inhibits gene expression in transgenic tobacco.

A hemimethylated chimeric gene, containing the cauliflower mosaic virus 35S promoter, the beta-glucuronidase coding region and the polyadenylation signal of nopaline synthase, was introduced into tobacco protoplasts by polyethylene glycol mediated transfection. Hemimethylation led to complete inhibition of transient gene expression. In regenerated transgenic plants the integrated gene was constitutively hypermethylated at the sequences CpG and CpNpG and this was correlated with an inactivation of beta-glucuronidase in 12 out of 18 analyzed plant lines whereas two showed slight and four strong activity. From 10 control lines transformed with nonmethylated DNA, only two were inactive; three showed slight and five strong activity. 5-aza-cytidine treatment of plant tissue from 'hypermethylated' lines led to induction of beta-glucuronidase in most cases. Shoots regenerated from azaC treated calli revealed stable enzyme restoration and demethylation of the integrated transgene.