A paragenetic perspective on integration of RNA silencing into the epigenome and its role in the biology of higher plants.

We describe features of RNA silencing and associated epigenetic imprints that illustrate potential roles for RNA interference (RNAi) in maintenance and transmission of epigenetic states between cells, throughout a plant, and perhaps even across sexual generations. Three types of transgenes can trigger RNAi of homologous endogenous plant genes: (1) "sense" transgenes that overexpress translatable transcripts, (2) inverted repeat (IR) transgenes that produce double-stranded RNA (dsRNA), and (3) antisense transgenes. Each mode of RNAi produces a different characteristic developmental silencing pattern. Single-copy transgenes are sufficient for sense-RNAi and antisense-RNAi, but not inverted repeat-RNAi. A single premature termination codon dramatically attenuates sense-RNAi, but it has no effect on antisense or inverted repeat-RNAi. We report here that antisense transgenes altered by removal of nonsense codons generate silencing patterns characteristic of sense-RNAi. Duplication of a sense overexpression transgene results in two types of epigenetic events: (1) complete loss of silencing and (2) altered developmental pattern of silencing. We also report that duplicating only the transgene promoter results in complete loss of silencing, whereas duplicating only transcribed sequences produces the second class, which are vein-based patterns. We infer that the latter class is due to systemic RNA silencing signals that interact with certain epigenetic states of the transgene to imprint it with information generated at a distance elsewhere in the plant.

Manganese homeostasis in Bacillus subtilis is regulated by MntR, a bifunctional regulator related to the diphtheria toxin repressor family of proteins.

The Bacillus subtilis yqhN gene encodes a metalloregulatory protein distantly related to the Corynebacterium diphtheriae diphtheria toxin repressor (DtxR). While DtxR mediates the iron-dependent repression of iron uptake, we demonstrate that yqhN (herein renamed mntR) encodes a manganese modulated regulator of manganese transport. An mntR mutant strain is sensitive to both manganese and cadmium, suggesting that the transport of these metals is derepressed. We selected Tn10 insertions that suppress the Mn(II) sensitivity of the mntR mutant or that increase the Cd(II) tolerance of wild-type cells, and in both cases we recovered insertions in mntH (formerly ydaR). MntH is a member of the NRAMP family of proton-coupled, metal ion transporters. MntR also regulates expression of a Mn(II) ABC transporter (MntABCD). The MntH and MntABCD transporters are both selectively repressed by Mn(II) and this regulation requires MntR. In high Mn(II) conditions, MntR functions as a Mn(II)-dependent repressor of mntH transcription. In contrast, MntR acts as a positive regulator of the mntABCD operon under low Mn(II) growth conditions. Biochemical studies demonstrate that MntR binding to the mntH control region requires Mn(II), while interaction with the mntABCD control region does not depend on Mn(II).

Homology-based control of gene expression patterns in transgenic petunia flowers.

Plant transgenes may participate in two types of homology-based gene silencing. One requires transcript homology, is post-transcriptional, and is referred to as cosuppression; the other requires promoter homology, is transcriptional, and is similar to paramutation. This paper uses flower color transgenes to address the hierarchical operation of both mechanisms in plants carrying two transgene copies. It is shown that cosuppression of homologous, endogenous flower color genes by single-copy transgenes requires that the transgene be driven by a strong promoter and that the degree of cosuppression is highly sensitive to increasing transgene dosage. Together, these observations suggest that cosuppression should be a sensitive reporter of epigenetic changes in transgene transcription, such as might be caused by paramutation-like interactions between transgene loci. Intercrosses bringing together two homologous transgene loci, one a known epimutable reporter and the other a transgene inverted repeat, result in complete loss of cosuppression in some outcross progeny and a qualitative change in morphology-based patterns of cosuppression in other outcross progeny. This paramutation-like behavior suggests that the transgenes may be altered at the transcriptional level, eliminating cosuppression altogether or changing the spatial pattern of transgene transcription to produce a new pattern of cosuppression.

The Frequency and Degree of Cosuppression by Sense Chalcone Synthase Transgenes Are Dependent on Transgene Promoter Strength and Are Reduced by Premature Nonsense Codons in the Transgene Coding Sequence.

By comparing the effects of strong and weak promoters that drive sense chalcone synthase (Chs) transgenes in large populations of independently transformed plants, we show here that a strong transgene promoter is required for high-frequency cosuppression of Chs genes and for production of the full range of cosuppression phenotypes. In addition, sense Chs transgenes driven by a cauliflower mosaic virus 35S promoter possessing a single copy of the upstream activator region (UAR) were found to produce a significantly lower degree of cosuppression than they did when the transgene promoter possessed two or four copies of the UAR. It has been shown elsewhere that 35S promoter strength increases with increasing UAR copy number. Frameshift mutations producing early nonsense codons in the Chs transgene were found to reduce the frequency and the degree of cosuppression. These results suggest that promoter strength and transcript stability determine the degree of cosuppression, supporting the hypothesis that sense cosuppression is a response to the accumulation of transcripts at high concentrations. This conclusion was shown to apply to single-copy transgenes but not necessarily to inversely repeated transgenes. The results presented here also have significance for efficient engineering of cosuppression phenotypes for use in research and agriculture.

Chlorella virus SC-1A encodes at least five functional and one nonfunctional DNA methyltransferases.

Chlorella virus SC-1A encodes at least six DNA methyltransferases (MTases): four N6-methyldeoxyadenine (m6A) MTases, M x CviSI (TGCmA), M x CviSII (CmATG), M x CviSIII (TCGmA) and M x CviSIV (GmATC), one 5-methyldeoxycytosine (m5C) MTase, M x CviSV (approximately RCmCG), and one nonfunctional m5C MTase, M x CviSVI, which is homologous to the MTase M x CviJI [RGmC(T/C/G)] produced by another chlorella virus IL-3A. Genes encoding three of the SC-1A m6A MTases (M x CviSI, M x CviSII, and M x CviSIII) and the nonfunctional m5C MTase were cloned and sequenced. Neither M x CviSI nor M x CviSIII genes hybridized to genes for their respective isomethylomers, M x CviRI and M x CviBIII, from other chlorella viruses. However, the M x CviSII gene hybridized strongly to its M x CviAII isomethylomer gene from virus PBCV-1. Like the prototype chlorella virus PBCV-1, the SC-1A genome contains inverted terminal repeats, one of which is adjacent to the nonfunctional m5C MTase. The three cloned m6A MTase genes are distributed throughout the approx. 345 kb SC-1A genome.

Chalcone synthase cosuppression phenotypes in petunia flowers: comparison of sense vs. antisense constructs and single-copy vs. complex T-DNA sequences.

Flower pigmentation patterns were scored in 185 sense Chalcone synthase (Chs) transgenotes and 85 antisense Chs transgenotes; upon first flowering, 139 (75%) of sense transgenotes were found to be phenotypically altered, as were 70 (82%) of the antisense transgenotes. The observed patterns document the range of phenotypic variations that occur, as well as confirm and extend the finding that sense Chs constructs produce several types of morphology-based flower pigmentation patterns that antisense Chs constructs do not. Long-term monitoring for epigenetic variations in one population of 44 sense Chs transgenotes showed that 43 (98%) were capable of producing a cosuppression phenotype. The primary determinant of sense-specific patterns of cosuppression of Chs was found to be the repetitiveness and organization pattern of the transgene, not 'position effects' by, or 'readthrough' from, flanking plant DNA sequences. The degree of cosuppression observed in progeny of transgenotes carrying multiple, dispersed copies as compared to that observed with a single copy of the transgene suggests that sense cosuppression of Chs is subject to a transgene dosage effect.

Analysis of 94 kb of the chlorella virus PBCV-1 330-kb genome: map positions 88 to 182.

Analysis of 94 kb of DNA, located between map positions 88 and 182 kb in the 330-kb chlorella virus PBCV-1 genome, revealed 195 open reading frames (ORFs) 65 codons or longer. One hundred and five of the 195 ORFs were considered major ORFs. Twenty-six of the 105 major ORFs resembled genes in the databases including three chitinases, a chitosanase, three serine/threonine protein kinases, two additional protein kinases, a tyrosine protein phosphatase, two ankyrins, an ornithine decarboxylase, a copper/zinc-superoxide dismutase, a proliferating cell nuclear antigen, a DNA polymerase, a fibronectin-binding protein, the yeast Ski2 protein, an adenine DNA methyltransferase and its corresponding DNA site-specific endonuclease, and an amidase. The genes for the 105 major ORFs were evenly distributed along the genome and, except for one noncoding 1788-nucleotide stretch, the genes were close together. Unexpectedly, a 900-bp region in the 1788-bp noncoding sequence resembled a CpG island.

Characterization of a protein kinase gene from two Chlorella viruses.

An open reading frame (ORF) with strong homology to eukaryotic serine/threonine protein kinases was found in the two Chlorella viruses SC-1A and PBCV-1. The deduced molecular weights of each putative protein kinase were 35 kDa and the predicted amino acid sequences of the two proteins were 95% identical. The ORF encoding the SC-1A protein kinase was over-expressed as a fusion protein in Escherichia coli. The recombinant fusion protein had autophosphorylation activity and could phosphorylate certain exogenous proteins. Antiserum against the recombinant fusion protein reacted with a 35 kDa protein plus three larger proteins from virus infected cells. The 35 kDa protein was a late protein; however, the 35 kDa protein was not packaged in the virion, even though virions contain protein kinase activity.

Ectopic expression of a viral adenine methyltransferase gene in tobacco.

Plant genomes contain both methylated adenine and cytosine residues although the roles of these methylations are not well understood. A chlorella virus adenine methyltransferase gene under the control of cauliflower mosaic virus 35S promoter in a binary plant transformation vector was expressed both in transgenic tobacco plants and transformed tobacco calli. The transgenic plants as well as transformed calli produced functional adenine methyltransferase enzyme, but the level of expression was higher in tobacco calli. A transgenic tobacco cell line that expressed the methyltransferase enzyme and carried an Arabidopsis cab3 gene containing a single target site for the adenine methyltransferase enzyme showed that the adenine residue was not methylated. HPLC analysis of genomic DNA from transgenic calli also showed no detectable levels of methylated adenine residues.

Protein glycosylation and myristylation in Chlorella virus PBCV-1 and its antigenic variants.

Chlorella virus PBCV-1 particles contain three glycoproteins, the major capsid protein Vp54 and two minor proteins Vp280 and Vp260. The major capsid protein is myristylated as well as glycosylated. Both modifications are in the carboxyl-terminal portion of the protein. A gene which is modified in a PBCV-1 antiserum-resistant mutant was cloned and sequenced. This gene has an open reading frame of 3099 bases and encodes one of the two large virion glycoproteins (Vp260). Vp260 contains 13 tandem repeats of 61 to 65 amino acids. The mutation deletes the equivalent of four of the amino acid repeat sequences and duplicates one of these sequences.

Leptospira genomes are modified at 5'-GTAC.

Genomic DNAs of 14 strains from seven species of the spirochete Leptospira were resistant to cleavage by the restriction endonuclease RsaI (5'-GTAC). A modified base comigrating with m4C was detected by chromatography. Genomic DNAs from other spirochetes, Borrelia group VS461, and Serpulina strains were not resistant to RsaI digestion. Modification at 5'-GTAm4C may occur in most or all strains of all species of Leptospira but not in all genera of spirochetes. Genus-wide DNA modification has rarely been observed in bacteria.

Evidence for virus-encoded glycosylation specificity.

Four spontaneously derived serologically distinct classes of mutants of the Paramecium bursaria chlorella virus (PBCV-1) were isolated using polyclonal antiserum prepared against either intact PBCV-1 or PBCV-1-derived serotypes. The oligosaccharide(s) of the viral major capsid protein and two minor glycoproteins determined virus serological specificity. Normally, viral glycoproteins arise from host-specific glycosylation of viral proteins; the glycan portion can be altered only by growing the virus on another host or by mutations in glycosylation sites of the viral protein. Neither mechanism explains the changes in the glycan(s) of the PBCV-1 major capsid protein because all of the viruses were grown in the same host alga and the predicted amino acid sequence of the major capsid protein was identical in the PBCV-1 serotypes. PBCV-1 antiserum resistance is best explained by viral mutations that block specific steps in glycosylation, possibly by inactivating glycosyltransferases.