Genetic Factors Required to Maintain Repression of a Paramutagenic Maize pl1 Allele

Corresponding author: Jay B. Hollick, Department of Plant and Microbial Biology, 111 Koshland Hall #3102, University of California, Berkeley, CA 94720-3102., hollick{at}nature.berkeley.edu (E-mail)

Communicating editor: J. A. BIRCHLER

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

A genetic screen identified two novel gene functions required to maintain mitotically and meiotically heritable gene silencing associated with paramutation of the maize purple plant 1 (pl1) locus. Paramutation at pl1 leads to heritable alterations of pl1 gene regulation; the Pl-Rhoades (Pl-Rh) allele, which typically confers strong pigmentation to juvenile and adult plant structures, changes to a lower expression state termed Pl'-mahogany (Pl'). Paramutation spontaneously occurs at low frequencies in Pl-Rh homozygotes but always occurs when Pl-Rh is heterozygous with Pl'. We identified four mutations that caused increased Pl' pigment levels. Allelism tests revealed that three mutations identified two new maize loci, required to maintain repression 1 (rmr1) and rmr2 and that the other mutation represents a new allele of the previously described mediator of paramutation 1 (mop1) locus. RNA levels from Pl' are elevated in rmr mutants and genetic tests demonstrate that Pl' can heritably change back to Pl-Rh in rmr mutant individuals at variable frequencies. Pigment levels controlled by two pl1 alleles that do not participate in paramutation are unaffected in rmr mutants. These results suggest that RMR functions are intimately involved in maintaining the repressed expression state of paramutant Pl' alleles. Despite strong effects on Pl' repression, rmr mutant plants have no gross developmental abnormalities even after several generations of inbreeding, implying that RMR1 and RMR2 functions are not generally required for developmental homeostasis.

EPIGENETIC control of gene expression plays a central role in both developmental processes and the maintenance of genome homeostasis. Much of our current understanding of these fundamental gene control mechanisms has resulted from studies of genes involved in color production (RUSSO et al. 1996 ; WU and MORRIS 1999 ). Variations in pigment expression are easily detected and often occur without directly affecting viability. These attributes have facilitated the identification and study of X chromosome inactivation (LYON 1961 ), position effect variegation (reviewed in WEILER and WAKIMOTO 1995 ), heterochromatin-induced inactivation (DORER and HENIKOFF 1994 ; HENIKOFF et al. 1995 ), transvection (JACK and JUDD 1979 ), pairing-sensitive silencing (reviewed in PIRROTTA 1997 ), paramutation (reviewed in CHANDLER et al. 2000 ), homology-dependent gene silencing (FURNER et al. 1998 ), cosuppression (reviewed in JORGENSEN 1995 ; PAL-BHADRA et al. 1997 ), transposon regulation (reviewed in FEDEROFF 1996 and in MARTIENSSEN 1996 ), quelling (COGONI and MACINO 1997 ), silencing of natural duplications (RONCHI et al. 1995 ; TODD and VODKIN 1996 ), imprinting (KERMICLE 1978 ), and DNA methylation (MEYER et al. 1993 ). Genetic screens using pigment reporters have been particularly powerful in revealing the genes and mechanisms responsible for some of these types of epigenetic control (REUTER and SPIERER 1992 ; CSINK et al. 1994 ; TALBERT et al. 1994 ; COGONI and MACINO 1997 ; FURNER et al. 1998 ). Herein we describe results of a genetic screen using a pigment reporter gene to identify mutations that affect meiotically heritable gene silencing occurring as a result of paramutation in maize.

Paramutation results in meiotically heritable changes in gene expression. Paramutable alleles are susceptible to these changes while paramutagenic alleles induce these changes. The hallmark of paramutation is the observation that the expression of a paramutable allele is invariably reduced following exposure to a paramutagenic allele in a heterozygote. Additionally, paramutable alleles themselves become paramutagenic following exposure to a paramutagenic allele. Alleles that are neither susceptible to such changes nor induce these changes are referred to as "neutral." Four distinct examples of paramutation are known in maize (reviewed in CHANDLER et al. 2000 ) and all four occur among alleles of genes encoding transcriptional activators of the anthocyanin or phlobaphene genetic pathways. Thus, paramutation occurring at these loci results in reduced pigment production, a simple and visual indicator of gene expression.

Temporal and spatial patterns of anthocyanin pigment production are specified by overlapping expression of two transcriptional activators required for the transcriptional induction of genes encoding anthocyanin biosynthetic enzymes (reviewed in DOONER et al. 1991 ). These two transcriptional activators are encoded by two sets of functionally duplicate genes: the booster 1 (b1) and red 1 (r1) genes encode basic helix-loop-helix factors and the colorless 1 (c1) and purple plant 1 (pl1) genes encode myb-like factors (reviewed in CHANDLER et al. 2000 ). In our material, seedling and anther tissues are pigmented due to the combined action of the r1 and pl1 genes while other mature plant tissues are pigmented due to overlapping b1 and pl1 functions. The Pl-Rhoades (Pl-Rh) allele of the pl1 locus is paramutable and can change to a paramutagenic state referred to as Pl'-mahogany (Pl'; HOLLICK et al. 1995 ). Pl-Rh normally conditions strong pigmentation to aerial tissues of the maize plant, but paramutation of Pl-Rh leads to reduced pl1 RNA levels and pl1 transcription that are visualized by reductions in pigment throughout the plant (HOLLICK et al. 2000 ). The paramutant state (Pl') is very stable; no changes of Pl' to a fully expressed Pl-Rh state are seen in Pl'/Pl' or Pl'/Pl-Rh plants (HOLLICK et al. 1995 ). However, Pl' can change back to Pl-Rh when Pl' is heterozygous with a neutral pl1 allele or when carried over a deletion (HOLLICK and CHANDLER 1998 ).

Several known modifier loci affect maize plant pigmentation. Recessive mutations at both anthocyanin 3 (a3) and mediator of paramutation 1 (mop1) loci lead to elevated levels of pigment in juvenile and adult plant structures (LINDSTROM 1935 ; DORWEILER et al. 2000 ). The action of A3 appears to be independent of any effects on pl1 alleles (C. C. CAREY, W. VANHEECKEREN and V. L. CHANDLER, unpublished results) but MOP1 acts specifically to maintain the repressed expression of both paramutant b1 and pl1 alleles (DORWEILER et al. 2000 ). Mutations in mop1 lead to derepression of Pl' and allow some Pl-Rh alleles to be recovered upon outcrossing (DORWEILER et al. 2000 ). Thus MOP1 plays a central role in maintaining the repression of paramutant alleles.

Herein we describe a seedling-based genetic screen that identified two novel maize loci, required to maintain repression 1 (rmr1) and rmr2, whose functions are required to maintain heritable repression of Pl'. Anthocyanin pigment and pl1 RNA levels are elevated in all plants that are homozygous for either rmr1 or rmr2 mutations and a significant fraction of the gametes produced from these plants carry nonparamutagenic Pl-Rh alleles. In contrast to mutations in mop1, mutations in rmr1 and rmr2 do not cause gross developmental abnormalities. Together with the identification of mop1 we are beginning to dissect the genetic requirements for paramutation in higher plants.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Genetic stocks:
All stocks contain dominant, functional alleles for all genes required for the production of anthocyanin pigments in seedling and anther tissues unless otherwise noted. Pistillate parents used for the mutagenesis experiment had the following pl1 and r1 genotypes: Pl'; R-r or R-r/r-g. Staminate parents for the mutagenesis experiment were originally obtained from the Maize Cooperation Stock Center (Urbana, IL; accession no. 611A: Pl-Rh sm1; R-r). The salmon silks 1 (sm1) locus is 10 cM distal to pl1. The recessive sm1 allele (sm1) confers salmon-colored silks when homozygous and thus provides a linked morphological marker to Pl-Rh. The Pl' testers used for initial crosses with putative mutants were obtained via spontaneous paramutation of Pl-Rh (HOLLICK et al. 1995 ). A632 (pl-A632; R-r), A619 (pl-A619; r-g), and W22 (pl-W22; r-g) inbred material was obtained from the USDA North Central Plant Introduction Station (Ames, IA). Additional W22 stocks (pl-W22; R-r:standard) were provided by Jerry Kermicle (University of Wisconsin, Madison). Five Pl-Rh/Pl-Rh tester stocks of different genetic backgrounds were used in crosses to determine whether or not mutant plants carried paramutagenic Pl' alleles. Material for pl1 RNA measurements was produced by crossing plants homozygous for the given EMS-derived mutation by sibling plants heterozygous for the same mutations. Additional details of the specific genetic stocks used for these experiments are available upon request.

Pollen mutagenesis:
Pollen pooled from multiple tassels was treated with EMS and applied to silks according to NEUFFER and COE 1978 . Pistillate parents were derived from two related Pl'/Pl' families. Both families together yielded 345 ears with an average of 49 kernels per ear. Germination frequency was 77% for the first 7500 M₁ seed planted. A total of 9000 M₁ seedlings were screened for dominant mutations affecting pigment production. Seedlings were grown in potting flats at an approximate density of 100 per square foot. Approximately 1000 M₁ plants were grown to maturity and self-pollinations of M₁ plants generated 495 M₂ families.

Seedling screens:
M₂ families of 30 seeds each were germinated in unheated sand benches and grown under high-intensity lighting (1660 µE/m² sec using a 1:1 mixture of sodium vapor and metal halide lamps). Visible seedling phenotypes (Table 1) were noted between 14 and 18 days post-imbibition.

Genetic crosses and stock syntheses:
Hand pollinations were used for all genetic crosses. Material for the sm1, rmr1 cosegregation test was derived by crossing a single plant heterozygous for ems136 and homozygous sm1 to a plant homozygous for ems235 but heterozygous for the recessive sm1 allele. The following syntheses and analyses were used to generate material to test the effects of the EMS-derived mutations on neutral pl1 alleles. Plants heterozygous for a given mutation (Pl' anthers) were crossed to both the A632 inbred line and a W22 line (obtained from J. Kermicle, University of Wisconsin, Madison) containing the weakly expressed pl1 alleles pl-A632 and pl-W22, respectively. The pl-A632, pl-W22, and Pl-Rh alleles are all distinct on the basis of restriction fragment length polymorphism (RFLP) analyses (J. HOLLICK, unpublished results). F₁ plants were either self-pollinated (A632 material) or backcrossed to plants homozygous for the given mutation (W22 material). F₂ (A632 material) and BC₁ (W22 material) progeny were grown to maturity. The pl1 genotypes of all plants in the F₂ and BC₁ families were determined using RFLP gel blot analysis as previously described (HOLLICK et al. 1995 ). Anther phenotypes of all homozygous pl-A632 F₂ plants were photographed for later comparison. All homozygous pl-A632 F₂ plants were crossed to plants heterozygous for the given EMS-induced mutation to determine whether the F₂ plants were also homozygous for the given EMS-induced mutation. Two F₂ plants that were homozygous pl-A632 by RFLP analysis and homozygous for rmr1-1 by testcross analysis were identified and five F₂ plants that were homozygous pl-A632 by RFLP analysis and homozygous for rmr2-1 by testcross analysis were identified. Examination of the anther photographs did not indicate any pigment differences between A632 inbred anthers or any F₂ pl-A632 plants regardless of whether or not they carried the EMS-induced mutations. The BC₁ plants that were Pl'/pl-W22 were self-pollinated and also crossed to plants heterozygous for a given EMS-induced mutation to identify BC₁ plants that were also homozygous for EMS-induced mutations. Anther pigmentation of pl-W22/pl-W22; rmr1-1/rmr1-1 and pl-W22/pl-W22; rmr2-1/rmr2-1 plants was weak and not obviously different from the anther pigmentation of heterozygous siblings or grandparental W22 plants.

RNA measurements:
RNA isolations from anther tissues and RNase-protection assays were performed as previously described (HOLLICK et al. 2000 ).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Pollen mutagenesis generated new maize mutations:
The number of mutable loci affecting paramutation-based gene silencing is unknown. To identify such loci, we performed chemical mutagenesis using ethyl methanesulfonate (EMS) to produce a high frequency of new maize mutations. M₁ seed was obtained by applying EMS-treated pollen from Pl-Rh/Pl-Rh plants to the silks of receptive Pl'/Pl' ears (MATERIALS AND METHODS). A total of 495 M₂ families were subsequently generated by self-pollination of M₁ plants to screen for recessive mutations. Small M₂ families (30 seeds each) were grown for 14–18 days (MATERIALS AND METHODS) and then visually examined for germination frequencies and unusual morphological phenotypes. Table 1 outlines the frequency of mutant phenotypes identified. Observed mutation frequencies are similar to previous EMS-pollen mutageneses (NEUFFER 1978 ), indicating that our chemical mutagenesis was highly efficient in producing new maize mutations.

Dominant mutations affecting Pl' expression were not found:
Because the primary leaf sheath of Pl'/Pl-Rh seedlings is normally weakly pigmented (HOLLICK et al. 1995 ), dominant EMS-induced mutations that either release Pl' from a repressed expression state or prohibit the establishment of paramutation might be expected to confer fully colored M₁ seedling phenotypes. Sixteen out of 9000 M₁ seedlings examined were fully colored. However, no flowering plants from these 16 fully colored seedlings had fully colored Pl-Rh-like anthers. This result indicates that our seedling screen for dominant mutations affecting anther pigmentation has an 0.2% false-positive rate. In addition, none of the 1000 M₁ plants grown to maturity had a Pl-Rh phenotype. Thus, no dominant mutations affecting either the establishment of pl1 paramutation or the maintenance of Pl' repression were identified out of 10,000 M₁ plants tested.

Genetic screens identified recessive mutations affecting seedling and anther pigmentation:
Recessive mutations that release Pl' from a repressed expression state might also be expected to confer fully colored seedling phenotypes (Fig 1). Our M₂ screen identified five families (nos. 60, 96, 136, 235, 240) that segregated fully colored seedlings to weakly colored seedlings in the following ratios (60, 4:23; 96, 1:29; 136, 6:24; 235, 3:25; 240, 8:16). In four of five families, fully colored seedlings gave rise to mature plants with fully pigmented Pl-Rh-like anthers (Fig 2A). Material from family 60 was dropped from further analyses because plants from fully colored seedlings in family 60 did not have fully pigmented anthers.

View larger version (51K): In this window In a new window Download PPT slide   Figure 1. Outline of genetic screen used to identify mutations affecting seedling pigmentation. Pl-Rh seedlings have fully colored first leaf sheaths (top right). Pl' seedlings have very weakly colored first leaf sheaths (top left). Pollen from Pl-Rh plants was treated with EMS and brushed on the silks of Pl' plants. All Pl'/Pl-Rh M1 plants had a Pl' seedling phenotype (middle). M1 plants were self-pollinated and M2 families were screened for Pl-Rh-like seedlings. (Bottom) An M2 family segregating 3:1 for Pl' and Pl-Rh-like seedlings.

View larger version (49K): In this window In a new window Download PPT slide   Figure 2. EMS-derived mutations affect anther pigment and pl1 RNA levels. (A) Anther phenotypes of plants that are heterozygous (top three variegated anthers) and homozygous (bottom three fully colored anthers) for the ems235 allele. The difference between anther pigment phenotypes was identical for both the ems96 and ems136 alleles (not shown). (B) RNase-protection assay measuring the levels of pl1 and actin RNA found in anther tissues of plants with the indicated genotypes. (C) Histogram of RNase-protection results for pl1 RNA levels measured relative to actin RNA for the indicated genotypes. Bars represent average measurements for each genotype and crossbars above each bar indicate the standard error. The number of samples measured for each genotype is as follows: ems96/+, n = 2; ems96/ems96, n = 2; ems136/+, n = 3; ems136/ems136, n = 4; ems235/+, n = 5; ems235/ems235, n = 6.

Genetic segregation tests indicated that the Pl-Rh-like phenotypes seen in three of the M₂ families (nos. 96, 136, and 235) were heritable as single-locus recessive traits. Plants with fully colored anthers derived from the three M₂ families were first crossed with Pl'/Pl' plants to complement the putative recessive mutations. All F₁ plants from these crosses had a clear Pl' phenotype (family 96, 35/35; family 136, 36/36; family 235, 36/36), indicating that the putative mutations affecting seedling and anther color are recessive. F₁ plants from families 96, 136, and 235 were self-pollinated to recover the fully colored seedling and anther trait. In all three cases, fully colored anther phenotypes were recovered at roughly a 1:3 ratio in F₂ families (Table 2), consistent with the interpretation that Pl-Rh-like phenotypes are due to single-locus recessive mutations. As expected, subsequent crosses between mutant plants and heterozygous siblings gave rise to families in which the mutant and normal phenotypes approximated a 1:1 ratio (family 96, 231:272; family 136, 100:88; family 235, 111:125).

Complementation tests define three loci:
Genetic complementation was tested in pairwise combinations among the four EMS-induced mutations and also with the recessive a3-Styles allele (STYLES and COE 1986 ) and the recessive mop1-1 allele (DORWEILER et al. 2000 ). Specific EMS-induced mutations were designated according to the family in which they were identified: ems96, ems136, ems235, and ems240. For each mutation, plants heterozygous for a given mutation (Pl' anthers or lightly colored plant in the case of a3-Styles heterozygotes) were individually crossed by plants homozygous for a different allele (Pl-Rh-like anthers or darkly colored plants in the case of a3-Styles homozygotes). Anther pigment phenotypes of progeny from these crosses were quantified (Table 3) on a 1–7 graded anther color score (ACS). ACS 7 describes the fully colored Pl-Rh phenotype, ACS 1–4 represent a stable Pl' state, and ACS 5 and 6 correspond to a typically metastable Pl' state (HOLLICK et al. 1995 ). On the basis of these results, mutations ems136 and ems235 fail to complement and thus define a single locus; ems240 is allelic to mop1; ems96 defines a third locus. Further description of the ems240 allele, designated mop1-2EMS, is provided in DORWEILER et al. 2000 .

EMS-derived mutations allow increased expression of pl1 RNA:
We used RNase-protection experiments to compare pl1 RNA levels from anthers of Pl'/Pl' plants either homozyogus or heterozygous for the ems136, ems235, or ems96 alleles. Using actin RNA as a control, pl1 RNA levels were 26-, 14-, and 9-fold greater in homozygous ems136, ems235, and ems96 mutants vs. heterozygous siblings, respectively (Fig 2B and Fig C). These results indicate that the normal functions of these genes identified by mutations are required to maintain repression of pl1 RNA accumulation.

Genes affecting repression of Pl' encode trans-acting factors:
Both mop1 and the locus defined by the ems96 allele genetically map to chromosome 2 (J. DORWEILER and V. CHANDLER, unpublished results; J. HOLLICK, unpublished results) whereas the pl1 locus is found on chromosome 6. Alleles that define the other locus (ems136 and ems235) failed to cosegregate with a genetic marker closely linked to the pl1 locus. The salmon silks 1 (sm1) locus, located 10 cM distal to pl1, normally conditions yellow maize silks but plants that are homozygous for the recessive sm1 allele (sm1) have salmon-colored silks. Genetic crosses (MATERIALS AND METHODS) were used to ask whether the fully colored anther phenotype found in homozygous mutant plants cosegregated with the recessive sm1 allele. A total of 6/15 (40%) segregant plants with fully colored anthers were sm1/sm1 and 9/21 (42%) plants with weakly colored anthers were sm1/sm1. The lack of strong cosegregation between the fully colored anther phenotype and recessive alleles of the sm1 locus indicates that the locus defined by the ems136 and ems235 alleles is distinct from the pl1 locus. Thus all three loci identified in our genetic screen define trans-acting genetic factors affecting pl1 RNA accumulation in Pl'/Pl' plants.

All three loci identified by mutation encode factors required to maintain heritable repression of Pl':
The fully colored phenotypes of seedlings and anthers, together with the increases in pl1 RNA seen when the EMS-derived mutations are homozygous, suggested that Pl' may have changed to a Pl-Rh state. Plants homozygous for mop1-1 sometimes show heritable changes of Pl' to Pl-Rh (DORWEILER et al. 2000 ). To test mutations at the other two loci, mutant plants with fully colored anthers were crossed to a series of Pl-Rh/Pl-Rh testers and the anther phenotypes of the progeny were quantified. If Pl' heritably changes to Pl-Rh in plants homozygous for the ems136, ems235, or ems96 alleles, then fully colored Pl-Rh phenotypes should be found in the resulting progeny. If Pl' does not heritably change to Pl-Rh, then only Pl' phenotypes would be found (HOLLICK et al. 1995 ). Results of these crosses (Table 4) indicate that while Pl' can be transmitted, Pl' often changes to a meiotically heritable Pl-Rh state in plants homozygous for the ems136, ems235, or ems96 alleles.

We have designated these new loci "rmr" for required to maintain repression to reflect the necessity of rmr functions for maintaining Pl' in a mitotically and meiotically heritable repressed state. The ems136 and ems235 alleles together define the rmr1 locus and are designated rmr1-1 and rmr1-2, respectively. The ems96 allele defines the rmr2 locus and is designated rmr2-1.

Pl' most frequently changed to a nonparamutagenic (Pl-Rh) state in plants homozygous for either the rmr1-1 or rmr1-2 alleles (Table 4). Approximately 70% of crosses between Pl-Rh/Pl-Rh testers and Pl'/Pl' plants homozygous for either the rmr1-1 or rmr1-2 allele produced at least one progeny plant having a Pl-Rh anther phenotype compared to only 22% of similar testcrosses with Pl'/Pl' plants homozygous for the rmr2-1 allele. This difference may partially relate to observations showing that different Pl-Rh/Pl-Rh testers themselves have different frequencies of spontaneous paramutation; Pl-Rh can spontaneously change to Pl' in the absence of Pl' (HOLLICK et al. 1995 ). To control for such differences, Pl-Rh/Pl-Rh pollen collected from a single plant was used to pollinate Pl'/Pl'; rmr1-1/rmr1-1 and Pl'/Pl'; rmr2-1/rmr2-1 plants. Both sets of crosses produced at least one progeny plant with a Pl-Rh phenotype, indicating that both rmr1-1 and rmr2-1 mutations can allow meiotically heritable derepression of Pl'. However, proportionally more Pl-Rh-like progeny were derived from crosses made with homozygous rmr1-1 plants (Table 5). In addition, progeny plants had higher anther color scores from the rmr1-1 crosses relative to the rmr2-1 crosses, suggesting that Pl' alleles are, in general, less paramutagenic when transmitted through plants homozygous for rmr1-1 as compared to rmr2-1 (Table 4 and Table 5).

Given that Pl' could sometimes change to a meiotically heritable Pl-Rh state in either homozygous rmr1-1 or rmr2-1 plants, we asked whether such newly changed Pl' alleles, formally designated Pl('), were distinguishable from a naive Pl-Rh allele. It was possible that Pl(') might retain residual paramutagenic activity relative to Pl-Rh. A cosegregation test using linked sm1 markers was used to show that a Pl(') allele transmitted from a either a rmr1-1 or rmr2-1 homozygote was indistinguishable from naive Pl-Rh in terms of its paramutagenicity (Fig 3). Thus, both RMR1 and RMR2 functions contribute to the meiotically heritable maintenance of the paramutagenic Pl' state. In the absence of RMR1 or RMR2 function, Pl' always has a somatic expression phenotype indistinguishable from Pl-Rh but this does not ensure that Pl' always changes to a meiotically heritable Pl-Rh state.

View larger version (24K): In this window In a new window Download PPT slide   Figure 3. Pl' can change to Pl-Rh in plants that are homozygous for rmr mutations. (A) Pedigree outlining two genetic crosses used to show that plants homozygous for the rmr mutations can transmit nonparamutagenic Pl-Rh alleles. Pl(') is used to identify a pl1 allele that should have been Pl' in the first generation but is not paramutagenic; the Pl(')/Pl-Rh plants in the second cross have fully colored anthers. (B) Results of crosses initiated with plants that were homozygous for the rmr1-1 allele. Histogram represents the number of plants with a given anther color score that were either sm1/sm1 (salmon-colored silks; solid bars) or Sm1/sm1 (yellow silks; open bars). (C) Results of crosses initiated with plants that were homozygous for the rmr2-1 allele. Histogram is similar to that shown in B.

rmr mutations do not affect the expression of other pl1 alleles:
We addressed the possibility that rmr functions generally affect all pl1 alleles by combining neutral pl1 alleles with either the rmr1-1 or rmr2-1 alleles (MATERIALS AND METHODS) and examining the anther phenotypes. The neutral pl-A632 and pl-W22 alleles confer weak, sunlight-dependent pigmentation to the anthers. Anthers of plants homozygous for the pl-A632 or pl-W22 alleles had visibly identical levels of pigmentation regardless of whether the plants were homozygous for a rmr mutation (Fig 4; MATERIALS AND METHODS), implying that RMR1 and RMR2 functions specifically affect the expression of paramutant Pl' alleles.

View larger version (73K): In this window In a new window Download PPT slide   Figure 4. Anther phenotypes of plants that are homozygous for the pl-A632 allele. Photographs are of tassels from an A632 plant (left), a plant homozygous for the rmr1-1 allele (center), and a plant homozygous for the rmr2-1 allele (right).

Heritable changes of Pl' to Pl-Rh in rmr1 mutants are unaffected by mode of sexual transmission:
Given that male and female gametes are differentially imprinted at a few loci in plants (KERMICLE and ALLEMAN 1990 ; VIELLE-CALZADA et al. 2000 ) at different times and locations during development, we wondered whether meiotically heritable changes of Pl' to Pl-Rh occurred with equal frequencies in the two separate somatic cell lineages or whether there were gametophyte-specific effects. This possibility was addressed using reciprocal crosses between Pl-Rh/Pl-Rh testers and plants homozygous for the rmr1-1 allele. Pollen from Pl-Rh/Pl-Rh testers was placed on receptive silks of Pl'/Pl'; rmr1-1/rmr1-1 plants and vice versa. Plants derived from seven reciprocal crosses were grown to maturity and the anther color scores determined (Table 6). When all the data are combined, the same frequency of Pl-Rh progeny was observed when Pl' was transmitted through either female or male gametes. When each reciprocal cross was examined, three sets of families had very similar frequencies. Frequencies for the remaining four sets of families varied 1.5- to 2-fold but there was no sex-specific trend. The frequencies of Pl' to Pl-Rh changes observed between individual rmr1-1/rmr1-1 plants could vary as much as 6-fold but the frequencies observed within each pair of reciprocal crosses varied no more than 2-fold. Thus, the frequency of heritable Pl' to Pl-Rh changes in rmr1-1 homozygotes appears to be intrinsic to each individual sporophyte and is not differentially affected by female vs. male gametophyte development.

rmr1 and rmr2 mutations do not appear to affect plant development:
Although mutations in mop1 can lead to a wide range of developmentally abnormal phenotypes (DORWEILER et al. 2000 ), we have not observed grossly abnormal phenotypes in homozygous rmr1 and rmr2 mutant plants. We began this series of observations by first generating F₁ individuals between A632 inbred plants and plants homozygous for either rmr1-1, rmr1-2, or rmr2-1 and then by examining the F₂ homozygous mutants derived from self-pollination. This strategy was adopted to reduce the number of other unlinked EMS-induced mutations that might potentially affect plant morphology. Of the 200 rmr F₂ plants segregating for each rmr mutation, all plants that had a fully colored anther phenotype were otherwise similar in stature, morphology, and flowering time to sibling plants. Similar uniform phenotypes were observed among F₂ plants generated from F₁ plants made by crossing the rmr mutations to either the A619 or W22 inbred lines.

Further generations of inbreeding did not produce abnormal plants. Several rmr F₂ plants (50% A632 background) with fully colored anthers that had a Pl-Rh/ Pl-Rh RFLP genotype were self-pollinated and also outcrossed to Pl-Rh/Pl-Rh testers. All F₃ plants from the self-pollinations had fully colored anthers but were otherwise similar in all other respects to testcross progeny. F₄ plants derived by self-pollination of homozygous rmr1-1 F₃ plants were also normal in appearance and similar in all respects to the previous F₃ plants. A single F₃ plant homozygous for rmr2-1 was crossed to a rmr2-1 heterozygote from a separate lineage to generate a family where rmr2-1 homozygotes and heterozygotes could be compared. Aside from differences in anther pigmentation, rmr2-1 homozygotes were indistinguishable from their heterozygous siblings. Although it remains possible that prolonged exposure of the genome to defects in RMR1 and RMR2 functions could have pleiotropic consequences, we currently have no indications that they are required for proper plant development, at least in a 50% A632 background.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Using a seedling-based genetic screen, we identified three maize loci (mop1, rmr1, and rmr2) whose trans-acting functions are required to maintain gene silencing that occurs as the result of paramutation at the pl1 locus. In all plants homozygous for mutations in mop1, rmr1, or rmr2, Pl' is expressed at a high level indistinguishable from Pl-Rh (DORWEILER et al. 2000 ; this report). While the reduced expression state (Pl') is most frequently restored upon outcrossing to nonmutant plants, an increased expression state indistinguishable from Pl-Rh can be meiotically heritable. Because pl1 paramutation is associated with changes in pl1 transcription (HOLLICK et al. 2000 ), our working model is that RMR1, RMR2, and MOP1 functions are involved with the maintenance of specific chromatin structures that prohibit high levels of pl1 transcription.

Additional rmr loci are clearly involved in Pl' repression. Our reported genetic screen represents only 10% theoretical saturation. This estimate is based on the apparent mutation rate of one detectable hit per gene per 1000 genomes screened and the roughly 95% probability that 5000 screened genomes would yield at least one detectable mutation in every given gene. These numbers suggest that 6–10 such rmr-like loci exist in maize. Indeed, 4 additional rmr loci have already been identified in ongoing genetic screens (J. HOLLICK, unpublished results).

Paramutation silencing at the pl1 locus not only leads to repression of Pl-Rh expression, it also affects the way in which the Pl-Rh allele is regulated (HOLLICK et al. 2000 ). Pl-Rh expression is light insensitive while Pl' expression is light dependent. As most pl1 alleles are regulated by light cues, the observations that rmr1 and rmr2 mutations have no effect on pl-A632 or pl-W22 suggest that RMR functions do not mediate light induction. Rather, we hypothesize that RMR functions are used in maintaining a specific regulatory state unique to Pl'. The Pl-Rh state then must actively prevent or avoid RMR action. Two general and nonexclusive modes of action are considered: (1) rmr RNA or proteins are components of a heritable chromatin structure that affects pl1 transcription, and (2) RMR functions facilitate alterations or maintenance of a heritable chromatin structure. Previous work has shown that paramutant pl1 and r1 alleles require continued allelic interactions with another paramutant or paramutagenic partner (STYLES and BRINK 1968 ; HOLLICK and CHANDLER 1998 ). It is possible that RMR functions actually mediate these allelic interactions through homology searching or specifying internuclear positions.

It is not known whether the rmr mutations described represent complete loss-of-function alleles. In fact, the observation that pl1 RNA levels are significantly lower in homozygous rmr1-2 plants vs. rmr1-1 plants hints that the two alleles have different levels of activity. Nonetheless, mutations at both the rmr1 and rmr2 loci are completely recessive, suggesting that neither the RMR1 nor RMR2 functions that affect Pl' repression are dosage sensitive as are many of the genes required to mediate examples of position effect variegation seen in Drosophila (reviewed in WEILER and WAKIMOTO 1995 ).

Despite similar effects on pl1 RNA levels, the frequency of meiotically heritable Pl' to Pl-Rh changes is clearly different in the rmr1 mutants vs. rmr2-1 mutants. Even in rmr1-2 homozygotes where the relative level of pl1 RNA is lower than in rmr2-1 homozygotes, Pl' changes to a Pl-Rh state more frequently in homozygous rmr1-2 plants. This set of results implies that derepression of Pl' in somatic tissues is, by itself, insufficient to allow a meiotically heritable change of Pl' to Pl-Rh. In addition to Pl' heritably changing to Pl-Rh more frequently in rmr1 mutant plants, those alleles that remain Pl' upon transmission appear to be less paramutagenic than those transmitted from rmr2-1 mutant plants. One interpretation is that RMR1 and RMR2 functions are distinct in terms of their role(s) in maintaining meiotically heritable repression of Pl'. Alternatively, the rmr2-1 allele may encode a partially active function with respect to maintaining the meiotically heritable Pl' state.

Although Pl' heritably changes to Pl-Rh at different frequencies between individual rmr1-1/rmr1-1 plants, similar frequencies were observed independent of female vs. male transmission in several reciprocal crosses (Table 6). In these families, the frequency of meiotically heritable changes of Pl' to Pl-Rh was established early in development prior to the point that cell lineages diverged to specify the lateral vs. apical inflorescence meristems. Frequencies of heritable Pl' to Pl-Rh changes were distinct, however, between each set of reciprocal crosses. This observation implies that different frequencies can be established and maintained through most of sporophyte development with relatively high fidelity. It remains unclear as to whether Pl' actually changes to a meiotically heritable Pl-Rh state during early development or whether the changes are only preset early.

RMR1 and RMR2 functions do not appear to be involved in general gene control mechanisms required for development. To date, plants homozygous for the rmr1-1, rmr1-2, or rmr2-1 alleles have been morphologically and developmentally indistinguishable from their heterozygous siblings. This absence of morphological defects, even after three generations of selfing, suggests that rmr1 and rmr2 are unlikely to be maize orthologues of the Arabidopsis ddm1 or met1 genes. Mutations of ddm1, a SWI2/SNF-2-like gene (JEDDELOH et al. 1999 ), and dominant inhibitors of met1, a DNA methyl transferase enzyme, have broad and cumulative effects on Arabidopsis development (FINNEGAN et al. 1996 ; KAKUTANI et al. 1996 ; RONEMUS et al. 1996 ). Because mutations in mop1 also appear to have effects on plant development (DORWEILER et al. 2000 ), it appears that the rmr1 and rmr2 loci define a class of genetic functions distinct from MOP1. Alternatively, RMR1 and RMR2 may participate in developmental pathways similar to MOP1 but have redundant developmental functions. Double mutant combinations are currently being synthesized to test this possibility.

Several other genes required to maintain transgene silencing in Arabidopsis have been recently described. Some of these genes, such as sgs1, sgs2, and sgs3 (suppressor of gene silencing) and sde1, sde2, sde3, and sde4 (silencing defective), are required for post-transcriptional silencing (ELMAYAN et al. 1998 ; DALMAY et al. 2000 ; MOURRAIN et al. 2000 ) while others, hog1 (homology-dependent gene silencing 1), sil1, sil2 (silencing), and mom1 (Morpheus' molecule 1), are required for transcriptional-based silencing (FURNER et al. 1998 ; AMEDEO et al. 2000 ). Given that paramutation at the b1 and pl1 loci affect heritable states of transcriptional control (PATTERSON et al. 1993 ; HOLLICK et al. 2000 ), the rmr genes could potentially be related to this latter class of Arabidopsis genes. It is also possible that RMR functions are specific for paramutation interactions, which to this point have not been described in Arabidopsis.

Our initial genetic dissection of the components required to maintain Pl' repression lays the foundation for further studies to understand the broad effects of RMR action and represents an initial step toward a molecular characterization of the epigenetic mechanism(s) responsible for paramutation in higher plants.

	ACKNOWLEDGMENTS

We are grateful to Jerry Kermicle (University of Wisconsin, Madison), the USDA North Central Plant Introduction Station (Ames, IA), and the Maize Genetics Cooperation Stock Center (Urbana, IL) for providing germplasm. This work was supported by a National Science Foundation Postdoctoral Fellowship in Plant Biology (BIR-9303601) and United States Department of Agriculture National Research Initiative award (97-35301-4430) to J.B.H., and by an American Cancer Society grant (N-675) and National Science Foundation grant (9603638) to V.L.C.

Manuscript received June 29, 2000; Accepted for publication September 16, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AMEDEO, P., Y. HABU, K. AFSAR, O. MITTELSTEN SCHEID, and J. PASZKOWSKI, 2000  Disruption of the plant gene MOM releases transcriptional silencing of methylated genes. Nature 405:203-206[Medline].

CHANDLER, V. L., W. B. EGGLESTON, and J. E. DORWEILER, 2000  Paramutation in Maize. Plant Mol. Biol. 43:121-145[Medline].

COGONI, C. and G. MACINO, 1997  Isolation of quelling-defective (qde) mutants impaired in posttranscriptional transgene-induced gene silencing in Neurospora crassa.. Proc. Natl. Acad. Sci. USA 94:10233-10238[Abstract/Free Full Text].

CSINK, A. K., R. LINSK, and J. A. BIRCHLER, 1994  The Lighten up (Lip) gene of Drosophila melanogaster, a modifier of retroelement expression, position effect variegation and white locus insertion alleles. Genetics 138:153-163[Abstract].

DALMAY, T., A. HAMILTON, S. RUDD, S. ANGELL, and D. C. BAULCOMBE, 2000  An RNA-dependent RNA polymerase gene in Arabidopsis is required for posttranscriptional gene silencing mediated by a transgene but not by a virus. Cell 101:543-553[Medline].

DOONER, H. K., T. P. ROBBINS, and R. A. JORGENSON, 1991  Genetic and developmental control of anthocyanin biosynthesis. Annu. Rev. Genet. 25:173-199[Medline].

DORER, D. R. and S. HENIKOFF, 1994  Expansions of transgene repeats cause heterochromatin formation and gene silencing in Drosophila. Cell 77:993-1002[Medline].

DORWEILER, J. E., C. C. CAREY, K. M. KUBO, J. HOLLICK, and J. L. KERMICLE et al., 2000  Mediator of paramutation 1 (mop1) is required for the establishment and maintenance of paramutation at multiple maize loci. Plant Cell 12:2101-2118[Abstract/Free Full Text].

ELMAYAN, T., S. BALZERGUE, F. BEON, V. BOURDON, and J. DAUBREMET et al., 1998  Arabidopsis mutants impaired in cosuppression. Plant Cell 10:1747-1758[Abstract/Free Full Text].

FEDEROFF, N., 1996 Epigenetic regulation of the maize Spm transposable element, pp. 575–592 in Epigenetic Mechanisms of Gene Regulation, edited by V. E. A. RUSSO, R. A. MARTIENSSEN and A. D. RIGGS. Cold Spring Harbor Laboratory Press, Plainview, NY.

FINNEGAN, E. J., W. J. PEACOCK, and E. S. DENNIS, 1996  Reduced DNA methylation in Arabidopsis thaliana results in abnormal plant development. Proc. Natl. Acad. Sci. USA 93:8449-8454[Abstract/Free Full Text].

FURNER, I. J., M. A. SHEIKH, and C. E. COLLETT, 1998  Gene silencing and homology dependent gene silencing in Arabidopsis: genetic modifiers and DNA methylation. Genetics 149:651-662[Abstract/Free Full Text].

HENIKOFF, S., J. M. JACKSON, and P. B. TALBERT, 1995  Distance and pairing effects on the brown^Dominant heterochromatic element in Drosophila. Genetics 140:1007-1017[Abstract].

HOLLICK, J. B. and V. L. CHANDLER, 1998  Epigenetic allelic states of a transcriptional regulatory locus exhibit overdominant gene action. Genetics 150:891-897[Abstract/Free Full Text].

HOLLICK, J. B., G. I. PATTERSON, E. H. COE, JR., K. C. CONE, and V. L. CHANDLER, 1995  Allelic interactions heritably influence the activity of a metastable maize pl allele. Genetics 141:709-719[Abstract].

HOLLICK, J. B., G. I. PATTERSON, I. M. ASMUNDSSON, and V. L. CHANDLER, 2000  Paramutation alters regulatory control of the maize pl locus. Genetics 154:1827-1838[Abstract/Free Full Text].

JACK, J. W. and B. H. JUDD, 1979  Allelic pairing and gene regulation: a model for the Zeste-white interaction in Drosophila melanogaster.. Proc. Natl. Acad. Sci. USA 76:1368-1372[Abstract/Free Full Text].

JEDDELOH, J. A., T. L. STOKES, and E. J. RICHARDS, 1999  Maintenance of genomic methylation requires a SWI2/SNF2-like protein. Nat. Genet. 22:94-97[Medline].

JORGENSEN, R. A., 1995  Cosuppression, flower color patterns, and metastable gene expression states. Science 268:686-691[Abstract/Free Full Text].

KAKUTANI, T., J. JEDDELOH, S. K. FLOWERS, K. MUNAKATA, and E. J. RICHARDS, 1996  Developmental abnormalities and epimutations associated with DNA hypomethylation mutations. Proc. Natl. Acad. Sci. USA 93:12406-12411[Abstract/Free Full Text].

KERMICLE, J. L., 1978 Imprinting of gene action in maize endosperm, pp. 357–371 in Maize Breeding and Genetics, edited by D. B. WALDEN. John Wiley and Sons, New York.

KERMICLE, J. L. and M. ALLEMAN, 1990  Gametic imprinting in maize in relation to the angiosperm life cycle. Development Suppl.: 9–14.

LINDSTROM, E. W., 1935  Some new mutants in maize. Iowa S. Coll. J. Sci. 9:451-459.

LYON, M. F., 1961  Gene action in the X-chromosome of the mouse (Mus musculus L.). Nature 190:372-373[Medline].

MARTIENSSEN, R. A., 1996 Epigenetic silencing Mu transposable elements in maize, pp. 593–608 in Epigenetic Mechanisms of Gene Regulation, edited by V. E. A. RUSSO, R. A. MARTIENSSEN and A. D. RIGGS. Cold Spring Harbor Laboratory Press, Plainview, NY.

MEYER, P., I. HEIDMANN, and I. NIEDENHOF, 1993  Differences in DNA-methylation are associated with a paramutation phenomenon in transgenic petunia. Plant J. 4:89-100[Medline].

MOURRAIN, P., C. BECLIN, T. ELMAYAN, F. FEUERBACH, and C. GODON et al., 2000  Arabidopsis SGS2 and SGS3 genes are required for posttranscriptional gene silencing and natural virus resistance. Cell 101:533-542[Medline].

NEUFFER, M. G., 1978 Induction of genetic variability, pp. 579–600, in Maize Breeding and Genetics, edited by D. B. WALDEN. John Wiley & Sons, New York.

NEUFFER, M. G. and E. H. COE, JR., 1978  Paraffin oil technique for treating mature corn pollen with chemical mutagens. Maydica 23:21-28.

PAL-BHADRA, M., U. BHADRA, and J. A. BIRCHLER, 1997  Cosuppression in Drosophila: gene silencing of Alcohol dehydrogenase by white-Adh transgenes is Polycomb dependent. Cell 90:479-490[Medline].

PATTERSON, G. I., C. J. THORPE, and V. L. CHANDLER, 1993  Paramutation, an allellic interaction, is associated with a stable and heritable reduction of transcription of the maize b regulatory gene. Genetics 135:881-894[Abstract].

PIRROTTA, V., 1997  PcG complexes and chromatin silencing. Curr. Opin. Genet. Dev. 7:249-258[Medline].

REUTER, G. and P. SPIERER, 1992  Position effect variegation and chromatin proteins. Bioessays 14:605-612[Medline].

RONCHI, A., K. PETRONI, and C. TONELLI, 1995  The reduced expression of endogenous duplications (REED) in the maize R gene family is mediated by DNA methylation. EMBO J. 14:5318-5328[Medline].

RONEMUS, M. J., M. GALBIATI, C. TICKNOR, J. CHEN, and S. L. DELLAPORTA, 1996  Demethylation-induced developmental pleiotropy in Arabidopsis.. Science 273:654-657[Abstract].

RUSSO, V. E. A., R. A. MARTIENSSEN and A. D. RIGGS (Editors), 1996 Epigenetic Mechanisms of Gene Regulation. Cold Spring Harbor Laboratory Press, Plainview, NY.

STYLES, E. D. and R. A. BRINK, 1968  The metastable nature of paramutable R alleles in maize. IV. Parallel enhancement of R action in heterozygotes with r and in hemizygotes. Genetics 61:801-811.

STYLES, E. D. and E. H. COE, JR., 1986  Unstable expression of an R allele with a3 in maize. J. Hered. 77:389-393[Abstract/Free Full Text].

TALBERT, P. B., C. D. S. LECIEL, and S. HENIKOFF, 1994  Modification of the Drosophila heterochromatic mutation brown^Dominant by linkage alterations. Genetics 136:559-571[Abstract].

TODD, J. J. and L. O. VODKIN, 1996  Duplications that suppress and deletions that restore expression from a chalcone synthase multigene family. Plant Cell 8:687-699[Abstract].

VIELLE-CALZADA, J. P., R. BASKAR, and U. GROSSNIKLAUS, 2000  Delayed activation of the paternal genome during seed development. Nature 404:91-94[Medline].

WU, C.-T. and J. R. MORRIS, 1999  Transvection and other homology effects. Curr. Opin. Genet. Dev. 9:237-246[Medline].

WEILER, K. S. and B. T. WAKIMOTO, 1995  Heterochromatin and gene expression in Drosophila. Annu. Rev. Genet. 29:577-605[Medline].

This article has been cited by other articles:

				M. Stam Paramutation: A Heritable Change in Gene Expression by Allelic Interactions In Trans Mol Plant, April 8, 2009; (2009) ssp020v1. [Abstract] [Full Text] [PDF]

				K. F. Erhard Jr., J. L. Stonaker, S. E. Parkinson, J. P. Lim, C. J. Hale, and J. B. Hollick RNA Polymerase IV Functions in Paramutation in Zea mays Science, February 27, 2009; 323(5918): 1201 - 1205. [Abstract] [Full Text] [PDF]

				L. Sidorenko and V. Chandler RNA-Dependent RNA Polymerase Is Required for Enhancer-Mediated Transcriptional Silencing Associated With Paramutation at the Maize p1 Gene Genetics, December 1, 2008; 180(4): 1983 - 1993. [Abstract] [Full Text] [PDF]

				V. Chandler and M. Alleman Paramutation: Epigenetic Instructions Passed Across Generations Genetics, April 1, 2008; 178(4): 1839 - 1844. [Full Text] [PDF]

				S. M. Gross and J. B. Hollick Multiple Trans-Sensing Interactions Affect Meiotically Heritable Epigenetic States at the Maize pl1 Locus Genetics, June 1, 2007; 176(2): 829 - 839. [Abstract] [Full Text] [PDF]

				D. M. Braun, Y. Ma, N. Inada, M. G. Muszynski, and R. F. Baker tie-dyed1 Regulates Carbohydrate Accumulation in Maize Leaves Plant Physiology, December 1, 2006; 142(4): 1511 - 1522. [Abstract] [Full Text] [PDF]

				K. M. McGinnis, C. Springer, Y. Lin, C. C. Carey, and V. Chandler Transcriptionally Silenced Transgenes in Maize Are Activated by Three Mutations Defective in Paramutation Genetics, July 1, 2006; 173(3): 1637 - 1647. [Abstract] [Full Text] [PDF]

				M. R. Woodhouse, M. Freeling, and D. Lisch The mop1 (mediator of paramutation1) Mutant Progressively Reactivates One of the Two Genes Encoded by the MuDR Transposon in Maize Genetics, January 1, 2006; 172(1): 579 - 592. [Abstract] [Full Text] [PDF]

				R. T. GRANT-DOWNTON and H. G. DICKINSON Epigenetics and its Implications for Plant Biology. 1. The Epigenetic Network in Plants Ann. Bot., December 1, 2005; 96(7): 1143 - 1164. [Abstract] [Full Text] [PDF]

				J. B. Hollick, J. L. Kermicle, and S. E. Parkinson Rmr6 Maintains Meiotic Inheritance of Paramutant States in Zea mays Genetics, October 1, 2005; 171(2): 725 - 740. [Abstract] [Full Text] [PDF]

				M. E. Blewitt, N. K. Vickaryous, S. J. Hemley, A. Ashe, T. J. Bruxner, J. I. Preis, R. Arkell, and E. Whitelaw An N-ethyl-N-nitrosourea screen for genes involved in variegation in the mouse PNAS, May 24, 2005; 102(21): 7629 - 7634. [Abstract] [Full Text] [PDF]

				V.L. CHANDLER Poetry of b1 Paramutation: cis- and trans-Chromatin Communication Cold Spring Harb Symp Quant Biol, January 1, 2004; 69(0): 355 - 362. [Abstract] [PDF]

				S. Chopra, S. M. Cocciolone, S. Bushman, V. Sangar, M. D. McMullen, and T. Peterson The Maize Unstable factor for orange1 Is a Dominant Epigenetic Modifier of a Tissue Specifically Silent Allele of pericarp color1 Genetics, March 1, 2003; 163(3): 1135 - 1146. [Abstract] [Full Text] [PDF]

				M. Stam, C. Belele, J. E. Dorweiler, and V. L. Chandler Differential chromatin structure within a tandem array 100 kb upstream of the maize b1 locus is associated with paramutation Genes & Dev., August 1, 2002; 16(15): 1906 - 1918. [Abstract] [Full Text] [PDF]

				D. Lisch, C. C. Carey, J. E. Dorweiler, and V. L. Chandler A mutation that prevents paramutation in maize also reverses Mutator transposon methylation and silencing PNAS, April 30, 2002; 99(9): 6130 - 6135. [Abstract] [Full Text] [PDF]

				D. Lisch, C. C. Carey, J. E. Dorweiler, and V. L. Chandler A mutation that prevents paramutation in maize also reverses Mutator transposon methylation and silencing PNAS, April 30, 2002; 99(9): 6130 - 6135. [Abstract] [Full Text] [PDF]