Comparative Analysis of Position–Effect Variegation Mutations in Drosophila melanogaster Delineates the Targets of Modifiers

Corresponding author: Steven Henikoff, Fred Hutchinson Cancer Research Center, A1-162, 1100 Fairview Ave. N., P.O. Box 19024, Seattle, WA 98109-1024, steveh{at}muller.fhcrc.org (E-mail).

Communicating editor: J. A. BIRCHLER

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

In Drosophila melanogaster, heterochromatin-induced silencing or position–effect variegation (PEV) of a reporter gene has provided insights into the properties of heterochromatin. Class I modifiers suppress PEV, and class II modifiers enhance PEV when the modifier gene is present in fewer than two doses. We have examined the effects of both class I and class II modifiers on four PEV mutations. These mutations include the inversions In(1)w^m4 and In(2R)bw^VDe2, which are classical chromosomal rearrangements that typify PEV mutations. The other mutations are a derivative of brown^Dominant, in which brown⁺ reporters are inactivated by a large block of heterochromatin, and a P[white⁺] transposon insertion associated with second chromosome heterochromatin. In general, we find that class I modifiers affect both classical and nonclassical PEV mutations, whereas class II modifiers affect only classical PEV mutations. We suggest that class II modifiers affect chromatin architecture in the vicinity of reporter genes, and only class I modifiers identify proteins that are potentially involved in heterochromatin formation or maintenance. In addition, our observations support a model in which there are different constraints on the process of heterochromatin-induced silencing in classical vs. nonclassical PEV mutations.

GENE expression depends on both intrinsic regulatory mechanisms, including enhancer–promoter interactions, and chromosomal context, including chromatin structure. Whereas intrinsic regulatory mechanisms are well defined molecularly, chromosomal context is difficult to assess and is sometimes revealed only by gene-silencing phenomena. Examples of gene silencing include X-chromosome inactivation and parental imprinting in mammals, telomere and mating-type silencing in yeast, as well as heterochromatin-induced gene silencing known as position–effect variegation (PEV) in Drosophila melanogaster (for review see HENDRICH and WILLARD 1995 ). In PEV, chromosomal rearrangements that change the position of a gene so that it is placed near heterochromatin result in the variable expression of the gene. In contrast to gene-rich euchromatin, heterochromatin has comparatively few genes, remains condensed throughout the cell cycle, and is enriched in satellite and middle repetitive DNA sequences (reviewed in CSINK et al. 1997 ). Although gene inactivation is a direct consequence of relocation to heterochromatin, it is not a normal function of heterochromatin, as evidenced by the presence of expressed genes in heterochromatin (GATTI and PIMPINELLI 1992 ). Nevertheless, the ability to inactivate a gene is presumed to reflect an intrinsic difference between heterochromatin and euchromatin with respect to chromatin structure and gene expression.

Genic modifiers of PEV have been readily recovered in different screens (DORN et al. 1993B ; LOCKE et al. 1988 ; SINCLAIR et al. 1989 , SINCLAIR et al. 1992 ; WUSTMANN et al. 1989 ) and have been considered a valuable tool in the study of heterochromatin. These modifiers have been categorized into two classes (LOCKE et al. 1988 ). Class I modifiers act to suppress PEV when only one dose of the wild-type gene is present and may enhance variegation in three doses. In genetic screens, class I modifiers were recovered as Suppressors of variegation [Su(var)s], which are gene mutations or deficiencies, and as chromosomal duplications that acted as Enhancers of variegation [E(var)s]. Because suppression of a PEV phenotype indicates a reduction in the ability of heterochromatin to silence a gene, Su(var)s were predicted to be mutations in genes that code for either structural components of heterochromatin or proteins that regulate heterochromatin components (LOCKE et al. 1988 ). Class II modifiers, isolated as mutations in E(var) genes, enhance variegation in one dose and may suppress in three doses. Therefore, E(var) genes might code for proteins that antagonize the silencing potential of heterochromatin or promote the formation of euchromatin (LOCKE et al. 1988 ).

More than 100 dosage-dependent modifiers of PEV have been identified (LOCKE et al. 1988 ; REUTER and SPIERER 1992 ; WUSTMANN et al. 1989 ). Most of these were identified by phenotypic suppression or enhancement of the chromosomal rearrangement In(1)w^m4 (w^m4). This inversion positions the white⁺ (w⁺) gene within 25 kb of pericentric heterochromatin, and the resulting mosaic inactivation of w⁺ produces patches of white tissue in a normally red eye (TARTOF et al. 1984 ). The enhancement or suppression of this phenotype can be readily scored as more or less white tissue, respectively, in a genetic screen. Modifiers of PEV isolated using w^m4 have been tested with other PEV mutations, such as the chromosomal rearrangements In(2R)bw^VDe2, In(1)y^3p, and T(2:3)Sb^V which affect the brown, yellow, and Stubble genes, respectively. Modifiers isolated using w^m4 typically behave similarly with these other PEV mutations.

A different response to genic modifiers was reported for PEV associated with the brown^Dominant (bw ^D) mutation. bw ^D contains a large heterochromatic insertion into the coding region of the brown⁺ (bw⁺) gene, generating a null allele (HENIKOFF et al. 1993 ). The heterochromatin of the bw ^D insertion is able to silence a wild-type copy of the bw⁺ gene that is present on the homolog. This is referred to as trans-inactivation, and such dominant PEV is characteristic of all variegating alleles of brown. A genetic screen designed to recover dominant modifiers of bw ^D identified unlinked mutations that suppressed trans-inactivation but failed to detect comparable enhancer mutations (TALBERT et al. 1994 ). A collection of these suppressors of bw ^D were tested for their effect on w^m4, and 33 out of 37 were found to be typical class I modifiers [i.e., Su(var) mutations]. The notable absence of enhancers of bw ^D was not caused by an inability to detect enhancement of the phenotype because two dominant enhancer mutations, both involving a rearrangement of the bw ^D chromosome, were identified. In addition, the bw ^D mutation is enhanced in males that lack a Y chromosome (P. TALBERT, personal communication), indicating that bw ^D responds to this modifier in a manner that is consistent with other PEV mutations (SPOFFORD 1976 ). These observations suggest that classical PEV mutations (i.e., gross chromosomal rearrangements such as w^m4) and nonclassical PEV mutations such as bw ^D exhibit differential responses to PEV modifiers, specifically to class II modifiers.

Failure to recover class II modifiers (i.e., E(var) mutations) in the collection of modifiers of bw ^D could be explained if the linkage enhancers of bw ^D were exceptionally strong, and therefore weaker effects would have been undetected. Alternatively, the lack of enhancers may be indicative of a bw ^D-specific property. For example, the bw ^D heterochromatic insertion is composed primarily of the simple sequence satellite (AAGAG)_n (CSINK and HENIKOFF 1996 ), which is present in heterochromatin (LOHE et al. 1993 ), and bound by GAGA protein in early embryos (RAFF et al. 1994 ). This raises the possibility that the class II modifier Trithorax-like (Trl), which codes for the GAGA protein (FARKAS et al. 1994 ), would modify the phenotype of bw ^D, but other modifiers would not. To differentiate between these and other possibilities, we examined a collection of PEV modifiers for effects on a panel of both classical and nonclassical PEV mutations. Our results indicate that insensitivity to class II modifiers is not restricted to the bw ^D mutation, but extends to two nonclassical PEV mutations, a derivative of bw ^D and a variegating w⁺ transgene insertion into heterochromatin. We rationalize these findings in terms of differences in targets of action between class I and class II modifiers. Additionally, our observations point to potentially different topological causes of heterochromatin-induced silencing in classical vs. nonclassical PEV mutations.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Fly stocks:
Flies were reared on standard cornmeal-molasses medium in vials or bottles at a constant temperature of 25°. Four PEV mutations were used in these experiments (Figure 1). The isolation and characterization of the Byron mutation has been described (HENIKOFF et al. 1995 ). P[w⁺tARry⁺t7.2AR = wA^R]B133-0923 (referred to in this paper as P[wA^R]B133) is a w⁺ transgene insertion in or near chromosome 2R heterochromatin from R. LEVIS, Syracuse University, NY. The collection of genic modifiers of PEV used in this study include both molecularly characterized mutations (references listed in Table 1) and uncharacterized modifiers that have been described previously (HEARN et al. 1991 ; SINCLAIR et al. 1983 ) or isolated in previously published genetic screens (DORN et al. 1993B ). EC stocks are documented (http://flybase.bio.indiana.edu:82/stocks/labs/reuter2.csv). All other chromosomes or mutations are described in LINDSLEY and ZIMM 1992 .

View larger version (11K): In this window In a new window Download PPT slide   Figure 1. —Depictions of PEV mutations used for comparative analysis. The classical PEV mutation wm4 is a chromosomal inversion in which the w+ gene is moved proximally to within 25 kb of X chromosome heterochromatin. In the case of bwVDe2, the bw+ gene is moved close to 2R heterochromatin. In each case, the small arrows show the position of the inversion breakpoints. The nonclassical PEV mutation Byron is a duplication of the 59E region in which the proximal element carries bw D and the distal element carries a wild-type bw D gene. P[wAR]B133 is a nonclassical PEV mutation in which a w+ transgene has inserted into or very near 2R heterochromatin. For PEV mutations involving the second chromosome, only the right arm is depicted.

Experimental design:
We examined the effects of modifiers on both classical and nonclassical PEV mutations. w^m4 and bw^VDe2 are classical PEV mutations in which two different reporter genes (w⁺ and bw⁺, respectively) are variably silenced because of their position next to heterochromatin (Figure 1). We also tested two nonclassical PEV mutations that involve the same reporters but are not gross chromosomal rearrangements. In the case of Byron, a derivative of the bw ^Dominant (bw ^D) PEV mutation, the large block of heterochromatin inserted within the bw⁺ gene inactivates bw⁺ reporter genes on the homolog in trans and on a duplication. Variegating transgene insertions into heterochromatin are also nonclassical PEV mutations. The fourth PEV mutation we examined is a w⁺ transgene, P[wA^R]B133, which is variably expressed because of its location in or near second chromosome (2R) heterochromatin (R. LEVIS, personal communication).

The modifiers used in our analysis are thought to be specific gene mutations, with the known exception being a regional duplication [the class I E(var) Dp(2;2)E39A]. The determination that a molecularly uncharacterized modifier is a class I or class II modifier cannot be made with certainty because a gain-of-function mutation of one class will behave the same as a loss-of-function mutation of the other class. For this reason, we concentrated primarily on those modifiers for which the gene has been cloned and sequenced.

The modifier lines used in this study are either homozygous lethal or infertile, requiring that the stocks be kept over balancer chromosomes. Previous studies assayed modifier effects by comparing the variegated phenotype with the modifier-bearing chromosome (e.g., w^m4;Modifier) to the pheno-type with the balancer chromosome (e.g., w^m4;Balancer). We found that this was an inaccurate test of the modifier effect because balancer chromosomes are capable of affecting the degree of variegation (data not shown). During the course of our studies, it became clear that an internal control was needed in each cross of modifier to PEV mutation. This was accomplished by first outcrossing flies carrying the modifier to a line that carried a marked chromosome. The second and third chromosomes were marked with the Sco and H ² mutations, respectively. These marked chromosomes do not modify the phenotypes of the PEV mutants used in our analysis (data not shown). The resulting male progeny that were heterozygous for the modifier and the marked chromosome were then crossed to females carrying the PEV mutation. Siblings carrying the PEV mutation with either the modifier or the marked chromosome were then compared directly. For PEV mutations that involved w⁺, only male siblings were compared.

Analysis of modifier effects on PEV mutations:
Pigment assays, which are commonly used to measure the degree of bw⁺ or w⁺ reporter expression, can be inaccurate because variation in head size will affect pigment measurements even when the observed degree of variegation is similar. In addition, pigment assays lack the required sensitivity for PEV mutations that result in very weak or strong inactivation. However, side-by-side visual comparisons of flies can be reliable and sensitive to subtle differences. We established five phenotypic ranks for each PEV mutation. Because the severity of the PEV mutations used in this study varies, phenotypic ranks were specific for each mutation. Also, the same predetermined phenotypic ranks for a given PEV mutation were used when examining the effect of each modifier analyzed in this study. Vials containing flies of a given genotype that had been aged 8–10 days at 25° were coded so that assignment of ranks could be made without knowing the specific genotype. In cases where the relevant genotypes do not show phenotypic overlap in the distribution of individuals falling into each rank, i.e., there is clear suppression or enhancement despite the phenotypic variation of individuals within a genotype, the effect of a modifier can be stated unequivocally. In cases where the distributions of the sibling phenotypes overlapped, however, the significance of the overlap was determined statistically. The nonparametric equivalent of the unpaired t-test, the Mann-Whitney U test, was used to determine significance (StatView, Abacus Concepts Inc., Berkeley, CA) and P values are reported where relevant.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Analysis of class I modifiers:
We examined the effects of four molecularly characterized class I modifiers (listed in Table 1) on our collection of PEV mutations. As expected, all class I loss-of-function mutations suppressed and the class I duplication [E(var)39A] enhanced the variegation seen in w^m4 (Table 2, column 1; Figure 2, column 1). Similar results were obtained with bw^VDe2 (Table 2, column 2; Figure 2, column 2). Class I modifiers likewise affect variegation of the nonclassical PEV mutations Byron and P[wA^R]B133 (Table 2; Figure 2, columns 3 and 4, respectively). However, one class I modifier, Su(var)3-9, did not suppress Byron and P[wA^R]B133 (Table 2, columns 3 and 4). A second allele, Su(var)3-9⁰⁵, similarly failed to suppress these nonclassical PEV mutations (data not shown).

View larger version (86K): In this window In a new window Download PPT slide   Figure 2. —Phenotypes of the PEV mutants with selected modifiers. Shown are typical phenotypes of the four PEV mutants used in this study. Examples include the phenotype in the absence of a modifier, as well as in the presence of a class I Su(var) [Su(var)3-6 in the case of wm4 and Byron; Su(var3-7) in the case of bwVDe2 and P[wAR]B133)], the class I duplication Dp(2;2)E39A, and the class II E(var)3-93D.

Analysis of class II modifiers:
Three class II modifiers that have been molecularly characterized (listed in Table 1) were also tested with our collection of PEV mutations. As previously described, w^m4 was significantly enhanced by all class II modifiers (Table 2, column 1; Figure 2, column 1). Surprisingly, a revertant of the Trl^13C allele weakly enhanced w^m4. This revertant allele, Trl^R4, had been generated by the precise excision of the P transposon in Trl^13C and so should have no effect on PEV mutants (FARKAS et al. 1994 ). The Trl^R4 chromosome also enhanced the phenotype of bw^VDe2 (see below), indicating that modification is not specific to w^m4. Therefore, it appears that the genetic background of the Trl^13C chromosome and derivatives Trl^R4 and Trl^R85 contribute to the modifying effects of mutations in the Trl gene.

We also examined the classical PEV mutation bw^VDe2 (Table 2, column 2). Like w^m4, bw^VDe2 was also enhanced by the class II modifiers Trl and E(var)3-93D (Figure 2, column 2); however, the E2F mutation had no effect on bw^VDe2.

The nonclassical PEV mutation Byron showed very different responses to class II modifiers (Table 2, column 3). E(var)3-93D and E2F had no enhancing effect on the Byron phenotype (Figure 2, column 3). With respect to the Trl series, there also was no enhancement of the Byron phenotype. The presence of either a putative null allele of Trl (Trl^R85) or the revertant allele (Trl^R4) had no consequence for the Byron phenotype. The two P element insertion alleles of Trl (Trl1^3C and Trl⁶²), considered to be hypomorphic mutations, were also examined. The Trl⁶² allele did not show enhancement of the Byron phenotype. Although the distribution of individuals into phenotypic rankings was significantly different in the presence of Trl⁶², it was in a direction that indicates suppression, not enhancement. Trl^13C had an inconsistent effect in that males appeared to be enhanced using the rank distribution assay, whereas females were not affected. Enhancement in Trl^13C males was too weak to be detected using pigment assays (data not shown). Because we can detect background enhancement of w^m4 and bw^VDe2 by a revertant of Trl^13C, the weak enhancement attributable to this chromosome is suspect. Thus, we conclude that class II modifiers do not enhance the phenotype of Byron.

P[wA^R]B133 was also examined with the class II modifiers (Table 2, column 4). E(var)3-93D and E2F had no effect (Figure 2, column 4). The Trl⁶² allele had no significant effect either; however, the number of P[wA^R]B133 individuals exhibiting an enhanced phenotype was increased in the presence of the null (Trl^R85), the hypomorphic (Trl^13C ), and the revertant (Trl^R4 ) alleles. The observed enhancement of P[wA^R]B133 by the revertant allele Trl^R4 is in keeping with the background enhancement that was detected for the classical PEV mutations. The fact that the independently derived Trl⁶² allele (i.e., with a different genetic background) does not cause enhancement indicates that the nonclassical PEV mutation P[wA^R]B133 is not affected by Trl mutations. Thus, the class II modifiers used in this study have no effect on P[wA^R]B133.

Analysis of additional modifiers of unknown class:
A collection of 11 molecularly uncharacterized modifier mutations was used to further assess the effect of modifiers of PEV on the nonclassical PEV mutation Byron (Table 3). Seven of these modifiers were also tested with P[wA^R]B133. Mutations in the putative class I modifier Su(var)2-1 result in hyperacetylation of histone H4 (DORN et al. 1986 ). This mutation strongly suppresses both Byron and P[wA^R]B133. Similarly, Su(var)208 strongly suppresses both nonclassical PEV mutations. Of nine E(var) mutations tested, six were without significant effect on the Byron phenotype (Table 3, column 1). The remaining three E(var) mutations had no effect on males, but they did enhance females. Five of the E(var) mutations were also tested on P[wA^R]B133 (Table 3, column 2), and four had no effect. Our analysis of molecularly uncharacterized modifiers extends our observations made with the class I and class II modifiers: class I modifiers affect all PEV mutations, whereas class II modifiers affect classical PEV mutations but not nonclassical PEV mutations.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The comparative analysis of PEV modifiers presented here used a method that allowed detection of significant modification. Our results indicate a difference in susceptibility to modification between nonclassical and classical PEV mutations. As demonstrated in the studies of others, classical PEV mutations were sensitive to both class I and class II modifiers. In general, the nonclassical PEV mutations examined in this study show the expected response to class I modifiers, but not to class II modifiers. This was true regardless of the dosage characteristics of the modifiers tested because both haplo-dependent and haplo-, triplo-dependent modifiers of a class behaved similarly. Our results confirm and extend inferences from previous studies that class II modifiers have no effect on nonclassical PEV mutations such as bw^D (TALBERT et al. 1994 ) or transgene insertions into heterochromatin (unpublished results cited in WALLRATH and ELGIN 1995 ).

Our results demonstrate that class I modifiers have a general role in heterochromatin-induced silencing in that they affect both classical and nonclassical PEV mutations. This is as expected if the target of class I modifier effects is heterochromatin. In contrast, class II modifiers only affect classical PEV mutations. One explanation for this observation is that the insensitivity of the nonclassical PEV mutations Byron and P[wA^R]B133 reflects an unusual composition of heterochromatin responsible for silencing. Compositional differences have been speculated to underlie differential responses to modifiers among classical PEV mutations (BISHOP 1992 ; LLOYD et al. 1997 ). However, most classical PEV mutations respond to class II modifiers even though these mutations represent diverse heterochromatin breakpoints. Occasional failure of a classical PEV mutation to respond to a modifier does not extend to the class as a whole, as is the case for Byron and P[wA^R]B133. Furthermore, we do not think that the nonclassical PEV mutations used in our study are unusual because failure to respond to Trl mutations has been noted for multiple variegating transgene inserts (unpublished results cited in WALLRATH and ELGIN 1995 ).

Alternatively, the observed differences between classical and nonclassical PEV mutations in their response to class II modifiers may reflect differences in the mechanism of reporter gene inactivation. In the case of the w⁺ reporter gene in w^m4, the proximity of heterochromatin may create a competition between open and closed chromatin at w⁺. Competition models for reporter gene inactivation have been previously described (APARICIO and GOTTSCHLING 1994 ; ELGIN 1996 ). The outcome of competition can be seen as the variegated phenotype. Modifiers that alter the balance between maintaining an active transcriptional complex and forming heterochromatin will alter the phenotype. A common theme of class II modifiers appears to be the potential for involvement in chromatin architecture (DE RUBERTIS et al. 1996 ; DORN et al. 1993A ; EBERL et al. 1997 ; FARKAS et al. 1994 ; SEUM et al. 1996 ). Mutations that affect chromatin architecture would result in enhancement by decreasing the probability that an active transcriptional complex is formed at w⁺, which in turn increases the probability that w⁺ will be inactivated by heterochromatin. Thus, any modifier that affects the competition between maintaining an active transcriptional state and a heterochromatin-induced inactive state will affect the phenotype of w^m4.

The same model would apply to the bw⁺ reporter gene that is found in cis to the breakpoint of the inversion bw^VDe2. As is the case for all bw variegating mutations, bw^VDe2 is dominant and causes trans-inactivation of the bw gene on the wild-type homolog. Trans-inactivation is also sensitive to the effects of class II modifiers, as evidenced by the enhanced phenotype of bw^VDe2. Because homologous chromosomes are paired in Dipterans (METZ 1916 ), the resulting formation of an inversion loop would bring the trans copy of the bw⁺ reporter closer to a putative heterochromatic compartment (Figure 3). Proximity to a heterochromatic compartment is responsible for the susceptibility of the trans copy of the bw⁺ reporter to class II modifiers. There would be an unstable balance between the formation of an active transcription complex on the trans copy of bw and the influence of trans-inactivating heterochromatin. As in the case of w^m4, modifiers that alter the balance between maintaining an active state and a silenced state will affect the phenotype.

View larger version (19K): In this window In a new window Download PPT slide   Figure 3. —Model of PEV in classical vs. nonclassical PEV mutations. The difference in sensitivity to class II modifiers that is observed between classical and nonclassical PEV mutants in our study is caused by differences in reporter silencing. The model (see DISCUSSION for details) is illustrated using the classical PEV mutations bwVDe2 and the nonclassical PEV mutations bw D. In the case of bwVDe2, somatic pairing generates an inversion loop that brings the bw+ reporter of the homologous chromosome into the vicinity of pericentric heterochromatin in the interphase nucleus. Heterochro-matin is depicted as a wavy line. The dose of a class I or class II modifier can affect the competition between establishing an active transcriptional complex at the bw+ reporter gene or inactivation caused by heterochromatin association. A reduction in the dose of a class I modifier decreases heterochromatic association and results in phenotypic suppression. Conversely, a reduction in the dose of a class II modifier decreases the probability of an active state and therefore increases heterochromatic association leading to an enhancement of the phenotype. In the case of the nonclassical PEV mutation bw D (or its derivative, Byron), somatic pairing of the homologs, combined with association between the heterochromatic insertion of bw D and second chromosome heterochromatin, results in mislocalization of a bw+ reporter gene to a heterochromatic compartment. The shaded oval represents a heterochromatic compartment with silencing properties. The force of heterochromatic association predominates, and therefore the dosage of class II modifiers is without phenotypic consequence. As for classical PEV mutations, heterochromatic associations in nonclassical PEV mutations are altered by changes in the dose of class I modifiers. Reducing the dose of a class I modifier decreases heterochromatic association and results in phenotypic suppression.

The mechanism of reporter gene inactivation in nonclassical PEV mutants may be qualitatively different than that for classical PEV mutants (Figure 3). In the case of bw ^D (and its derivatives), the bw⁺ reporter gene present on the homolog may be inactivated because it is mislocalized to a heterochromatic compartment of the nucleus (CSINK and HENIKOFF 1996 ; DERNBURG et al. 1996 ; TALBERT et al. 1994 ). A correlation between nuclear mislocalization and the degree of bw inactivation in the eye suggests that a heterochromatic compartment is not conducive to bw transcription (CSINK and HENIKOFF 1996 ). The effect of decreasing the dose of a gene that encodes a protein involved in establishing an open chromatin state would be negligible if reporter gene expression is already compromised by virtue of its mislocalization to a heterochromatic compartment. The same insensitivity would be observed for a transgene insertion in heterochromatin. In the case of P[wA^R]B133, the w⁺ reporter is mostly inactivated (as indicated by the extreme variegation) because it is sequestered within a heterochromatic compartment, and so dosage changes in chromatin architectural proteins would not be limiting for transcription.

There were exceptions to the above-described generalizations regarding modifier mutations. Three of the 12 E(var) mutations enhanced the phenotype of Byron females (but not males), and one of these also enhanced P[wA^R]B133. In these cases, we predict that the corresponding mutation leads to a gain of function of a class I modifier gene. Alternatively, enhancement might be caused by a mutation in a gene that encodes a product directly involved in the negative regulation of heterochromatin. Another exception, the class I modifier Su(var)3-9 did not suppress either Byron or P[wA^R]B133. This is surprising given that Su(var)3-9 was shown to suppress associations between bw ^D and heterochromatin in larval brains (CSINK and HENIKOFF 1996 ). Because Su(var)3-9 mutations are in a transcribed region that is primarily expressed in embryos (TSCHIERSCH et al. 1994 ), the product might not be present during development of the eye when the bw gene is expressed. Consistent with this explanation, we found that Su(var)3-9 ⁰¹ was a weak suppressor of classical PEV mutations (data not shown). Although Su(var)3-9 was suggested to play an important role in heterochromatinization (TSCHIERSCH et al. 1994 ), our results suggest that its primary target of action is not heterochromatin.

How do our results fit with other well-studied silencing phenomena? The potential mechanistic similarities between class I modifiers of PEV and proteins involved in repression of homeotic gene expression [Polycomb group (Pc-G) gene products] have been discussed frequently (MOEHRLE and PARO 1994 ; PIRROTTA and RASTELLI 1994 ; REUTER and SPIERER 1992 ). The class II modifiers Trl and E(var)3-93D are mutations in genes of the Trithorax group, genes that encode proteins needed for the appropriate activation of homeotic genes. This observation led investigators to conclude that this overlap was noteworthy (DORN et al. 1993A ; FARKAS et al. 1994 ). Our results provide an explanation. If the class II modifier effects reflect reporter gene sensitivity, then the loss of chromatin architectural proteins (such as encoded by Trl and E(var) 3-93D) may affect susceptible homeotic genes by a similar mechanism. That is, loss of Trl or E(var)3-93D alters the balance of an open vs. closed chromatin state, resulting in the increased probability of the inappropriate formation of a Pc-G or a heterochromatin-mediated silencing complex.

The large collection of mutations that can modify PEV phenotypes has been interpreted as an indication that heterochromatin-induced silencing is an inherently complex phenomenon (REUTER and SPIERER 1992 ). In contrast, our comparative analysis of classical and nonclassical PEV mutations supports a simpler picture of PEV. We suggest that the complexity of modifiers may reflect the number of ways it is possible to affect gene expression. Class II modifiers act at the level of reporter gene expression; therefore, modification of gene expression is not informative with respect to the properties of heterochromatin. If PEV is to be used as a tool in the analysis of heterochromatin, the identification of modifiers that act on both classical and nonclassical PEV mutations would be more informative. Such an approach would distinguish proteins that affect reporter gene expression, either directly or indirectly, from those involved in packaging heterochromatin.

	ACKNOWLEDGMENTS

We thank R. LEVIS and G. REUTER for stocks, and K. AHMAD, A. CSINK, and P. TALBERT for helpful comments on this manuscript. We also acknowledge the contributions of A. CSINK and K. WEILER in discussions that led to our model describing differences between modification of classical and nonclassical PEV mutations. This work was supported by the Howard Hughes Medical Institute.

Manuscript received August 1, 1997; Accepted for publication October 10, 1997.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

APARICIO, O. M. and D. E. GOTTSCHLING, 1994  Overcoming telomeric silencing: a trans-activator competes to establish gene expression in a cell cycle-dependent way. Genes Dev. 8:1133-1146[Abstract/Free Full Text].

BAKSA, K., H. MORAWIETZ, V. DOMBRADI, M. AXTON, and H. TAUBERT et al., 1993  Mutations in the protein phosphatase 1 gene at 87B can differentially affect suppression of position-effect variegation and mitosis in Drosophila melanogaster.. Genetics 135:117-125[Abstract].

BISHOP, C. P., 1992  Evidence for intrinsic differences in the formation of chromatin domains in Drosophila melanogaster.. Genetics 132:1063-1069[Abstract].

CLEARD, F., M. MATSARSKAIA, and P. SPIERER, 1995  The modifier of position-effect variegation Suvar(3)7 of Drosophila: there are two alternative transcripts and seven scattered zinc fingers, each preceded by a tryptophan box. (erratum: Nucleic Acids Res. 23: 3804). Nucleic Acids Res. 23:796-802[Abstract/Free Full Text].

CSINK, A. K. and S. HENIKOFF, 1996  Genetic modification of heterochromatic association and nuclear organization in Drosophila.. Nature 381:529-531[Medline].

CSINK, A. K., G. L. SASS and S. HENIKOFF, 1997 Drosophila heterochromatin: retreats for repeats, pp. 223–235 in Nuclear Organization, Chromatin Structure and Gene Expression, edited by A. OTTE and R. V. DRIEL. Oxford University Press, Oxford.

DERNBURG, A. F., K. W. BROMAN, J. C. FUNG, W. F. MARSHALL, and J. PHILIPS et al., 1996  Perturbation of nuclear architecture by long-distance chromosome interactions. Cell 85:745-759[Medline].

DE RUBERTIS, F., D. KADOSH, S. HENCHOZ, D. PAULI, and G. REUTER et al., 1996  The histone deacetylase Rpd3 counteracts genomic silencing in Drosophila and yeast. Nature 384:589-591[Medline].

DORN, R., S. HEYMAN, R. LINDIGKEIT, and G. REUTER, 1986  Suppressor mutation of position-effect variegation in Drosophila melanogaster affecting chromatin properties. Chromosoma 93:398-403.

DORN, R., V. KRAUSS, G. REUTER, and H. SAUMWEBER, 1993a  The enhancer of position-effect variegation of Drosophila, E(var)3-93D, codes for a chromatin protein containing a conserved domain common to several transcriptional regulators. Proc. Natl. Acad. Sci. USA 90:11376-11380[Abstract/Free Full Text].

DORN, R., J. SZIDONYA, G. KORGE, M. SEHNERT, and H. TAUBERT et al., 1993b  P transposon-induced dominant enhancer mutations of position-effect variegation in Drosophila melanogaster.. Genetics 133:279-290[Abstract].

EBERL, D. F., L. J. LORENZ, M. B. MELNICK, V. SOOD, and P. LASKO et al., 1997  A new enhancer of position-effect variegation in Drosophila melanogaster encodes a putative RNA helicase that binds chromosomes and is regulated by the cell cycle. Genetics 146:951-963[Abstract].

EISSENBERG, J. C., T. C. JAMES, D. M. FOSTER-HARTNETT, T. HARTNETT, and V. NGAN et al., 1990  Mutation in a heterochromatin-specific chromosomal protein is associated with suppression of position-effect variegation in Drosophila melanogaster.. Proc. Natl. Acad. Sci. USA 87:9923-9927[Abstract/Free Full Text].

ELGIN, S. C., 1996  Heterochromatin and gene regulation in Drosophila.. Curr. Opin. Genet. Dev. 6:193-202[Medline].

FARKAS, G., J. GAUSZ, M. GALLONI, G. REUTER, and H. GYURKOVICS et al., 1994  The Trithorax-like gene encodes the Drosophila GAGA factor. Nature 371:806-808[Medline].

GATTI, M. and S. PIMPINELLI, 1992  Functional elements in Drosophila melanogaster heterochromatin. Annu. Rev. Genet. 26:239-275[Medline].

HEARN, M. G., A. HEDRICK, T. A. GRIGLIATTI, and B. T. WAKIMOTO, 1991  The effect of modifiers of position-effect variegation on the variegation of heterochromatic genes of Drosophila melanogaster.. Genetics 128:785-797[Abstract].

HENDRICH, B. D. and H. F. WILLARD, 1995  Epigenetic regulation of gene expression: the effect of altered chromatin structure from yeast to mammals. Hum. Mol. Genet. 4:1765-1777[Abstract].

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].

HENIKOFF, S., K. LOUGHNEY and T. D. DREESEN, 1993 The enigma of dominant position-effect variegation in Drosophila, pp. 193–207 in John Innes Review The Chromosome, edited by J. S. HESLOP-HARRISON and R. B. FLAVELL. BIOS Scientific Publishers Ltd., Oxford.

LINDSLEY, D. L., and G. G. ZIMM, 1992 The genome of Drosophila melanogaster. Academic Press, New York.

LOCKE, J., M. A. KOTARSKI, and K. D. TARTOF, 1988  Dosage-dependent modifiers of position effect variegation in Drosophila and a mass action model that explains their effect. Genetics 120:181-198[Abstract/Free Full Text].

LOHE, A. R., A. J. HILLIKER, and P. A. ROBERTS, 1993  Mapping simple repeated DNA sequences in heterochromatin of Drosophila melanogaster.. Genetics 134:1149-1174[Abstract].

LLOYD, V. K., D. A. SINCLAIR, and T. A. GRIGLIATTI, 1997  Competition between different variegating rearrangements for limited heterochromatic factors in Drosophila melanogaster.. Genetics 145:945-959[Abstract].

METZ, C. W., 1916  Chromosome studies on the Diptera II. The paired association of chromosomes in the Diptera, and its significance. J. Exp. Zool. 21:213-280.

MOEHRLE, A. and R. PARO, 1994  Spreading the silence: epigenetic transcriptional regulation during Drosophila development. Dev. Genet. 15:478-484[Medline].

PIRROTTA, V. and L. RASTELLI, 1994  white gene expression, repressive chromatin domains and homeotic gene regulation in Drosophila.. Bioessays 16:549-556[Medline].

RAFF, J. W., R. KELLUM, and B. ALBERTS, 1994  The Drosophila GAGA transcription factor is associated with specific regions of heterochromatin throughout the cell cycle. EMBO J. 13:5977-5983[Medline].

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

REUTER, G., J. GAUSZ, H. GYURKOVICS, B. FRIEDE, and R. BANG et al., 1987  Modifiers of position-effect variegation in the region from 86C to 88B of the Drosophila melanogaster third chromosome. Mol. Gen. Genet. 210:429-436[Medline].

SEUM, C., A. SPIERER, D. PAULI, J. SZIDONYA, and G. REUTER et al., 1996  Position-effect variegation in Drosophila depends on dose of the gene encoding the E2F transcriptional activator and cell cycle regulator. Development 122:1949-1956[Abstract].

SINCLAIR, D. A. R., R. C. MOTTUS, and T. A. GRIGLIATTI, 1983  Genes which suppress position-effect variegation in Drosophila melanogaster are clustered. Mol. Gen. Genet. 191:326-333.

SINCLAIR, D. A., V. K. LLOYD, and T. A. GRIGLIATTI, 1989  Characterization of mutations that enhance position-effect variegation in Drosophila melanogaster.. Mol. Gen. Genet. 216:328-333[Medline].

SINCLAIR, D. A., A. A. RUDDELL, J. K. BROCK, N. J. CLEGG, and V. K. LLOYD et al., 1992  A cytogenetic and genetic characterization of a group of closely linked second chromosome mutations that suppress position-effect variegation in Drosophila melanogaster.. Genetics 130:333-344[Abstract].

SPOFFORD, J. B., 1976 Position-effect variegation in Drosophila, pp. 955–1018 in The Genetics and Biology of Drosophila, edited by M. ASHBURNER and E. NOVITSKI. Academic Press, New York.

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

TARTOF, K. D., C. HOBBS, and M. JONES, 1984  A structural basis for variegating position effects. Cell 37:869-878[Medline].

TSCHIERSCH, B., A. HOFMANN, V. KRAUSS, R. DORN, and G. KORGE et al., 1994  The protein encoded by the Drosophila position-effect variegation suppressor gene Su(var)3-9 combines domains of antagonistic regulators of homeotic gene complexes. EMBO J. 13:3822-3831[Medline].

WALLRATH, L. L. and S. C. ELGIN, 1995  Position effect variegation in Drosophila is associated with an altered chromatin structure. Genes Dev. 9:1263-1277[Abstract/Free Full Text].

WUSTMANN, G., J. SZIDONYA, H. TAUBERT, and G. REUTER, 1989  The genetics of position-effect variegation modifying loci in Drosophila melanogaster.. Mol. Gen. Genet. 217:520-527[Medline].

ZOLLMAN, S., D. GODT, G. G. PRIVE, J. L. COUDERC, and F. A. LASKI, 1994  The BTB domain, found primarily in zinc finger proteins, defines an evolutionarily conserved family that includes several developmentally regulated genes in Drosophila.. Proc. Natl. Acad. Sci. USA 91:10717-10721[Abstract/Free Full Text].

This article has been cited by other articles:

				P. Nisha, J. L. Plank, and A. K. Csink Analysis of Chromatin Structure of Genes Silenced by Heterochromatin in Trans Genetics, May 1, 2008; 179(1): 359 - 373. [Abstract] [Full Text] [PDF]

				B. Grau, C. Popescu, L. Torroja, D. Ortuno-Sahagun, I. Boros, and A. Ferrus Transcriptional Adaptor ADA3 of Drosophila melanogaster Is Required for Histone Modification, Position Effect Variegation, and Transcription Mol. Cell. Biol., January 1, 2008; 28(1): 376 - 385. [Abstract] [Full Text] [PDF]

				N. Cobbe, E. Savvidou, and M. M. S. Heck Diverse Mitotic and Interphase Functions of Condensins in Drosophila Genetics, February 1, 2006; 172(2): 991 - 1008. [Abstract] [Full Text] [PDF]

				B. L. Newman, J. R. Lundblad, Y. Chen, and S. M. Smolik A Drosophila Homologue of Sir2 Modifies Position-Effect Variegation but Does Not Affect Life Span Genetics, December 1, 2002; 162(4): 1675 - 1685. [Abstract] [Full Text] [PDF]

				K. M. Donaldson, A. Lui, and G. H. Karpen Modifiers of Terminal Deficiency-Associated Position Effect Variegation in Drosophila Genetics, March 1, 2002; 160(3): 995 - 1009. [Abstract] [Full Text] [PDF]

				A. K. Csink, A. Bounoutas, M. L. Griffith, J. F. Sabl, and B. T. Sage Differential Gene Silencing by trans-heterochromatin in Drosophila melanogaster Genetics, January 1, 2002; 160(1): 257 - 269. [Abstract] [Full Text] [PDF]

				J. E. Dorweiler, C. C. Carey, K. M. Kubo, J. B. Hollick, J. L. Kermicle, and V. L. Chandler mediator of paramutation1 Is Required for Establishment and Maintenance of Paramutation at Multiple Maize Loci PLANT CELL, November 1, 2000; 12(11): 2101 - 2118. [Abstract] [Full Text]

				F. Vernì, R. Gandhi, M. L. Goldberg, and M. Gatti Genetic and Molecular Analysis of wings apart-like (wapl), a Gene Controlling Heterochromatin Organization in Drosophila melanogaster Genetics, April 1, 2000; 154(4): 1693 - 1710. [Abstract] [Full Text]

				G. L. Sass and S. Henikoff Pairing-Dependent Mislocalization of a Drosophila brown Gene Reporter to a Heterochromatic Environment Genetics, June 1, 1999; 152(2): 595 - 604. [Abstract] [Full Text]

				J. Suso Platero, A. K. Csink, A. Quintanilla, and S. Henikoff Changes in Chromosomal Localization of Heterochromatin-binding Proteins during the Cell Cycle in Drosophila J. Cell Biol., March 23, 1998; 140(6): 1297 - 1306. [Abstract] [Full Text] [PDF]

				K Stankunas, J Berger, C Ruse, D. Sinclair, F Randazzo, and H. Brock The enhancer of polycomb gene of Drosophila encodes a chromatin protein conserved in yeast and mammals Development, January 10, 1998; 125(20): 4055 - 4066. [Abstract] [PDF]