The Drosophila roX1 RNA Gene Can Overcome Silent Chromatin by Recruiting the Male-Specific Lethal Dosage Compensation Complex

Corresponding author: Richard L. Kelley, Baylor College of Medicine, Houston, TX 77030., rkelley{at}bcm.tmc.edu (E-mail)

Communicating editor: S. HENIKOFF

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The Drosophila MSL complex consists of at least six proteins and two noncoding roX RNAs that mediate dosage compensation. It acts to remodel the male's X chromatin by covalently modifying the amino terminal tails of histones. The roX1 and roX2 genes are thought to be nucleation sites for assembly and spreading of MSL complexes into surrounding chromatin where they roughly double the rates of transcription. We generated many transgenic stocks in which the roX1 gene was moved from its normal location on the X to new autosomal sites. Approximately 10% of such lines displayed unusual sexually dimorphic expression patterns of the transgene's mini-white eye-color marker. Males often displayed striking mosaic pigmentation patterns similar to those seen in position-effect variegation and yet most inserts were in euchromatic locations. In many of these stocks, female mini-white expression was very low or absent. The male-specific activation of mini-white depended upon the MSL complex. We propose that these transgenes are inserted in several different types of repressive chromatin environments that inhibit mini-white expression. Males are able to overcome this silencing through the action of the MSL complex spreading from the roX1 gene and remodeling the local chromatin to allow transcription. The potency with which an ectopic MSL complex overcomes silent chromatin suggests that its normal action on the X must be under strict regulation.

IN Drosophila, males hypertranscribe most genes along their single X chromosome to match the output of females with two X chromosomes (CLINE and MEYER 1996 ). This hypertranscription is mediated by a large RNA-protein complex distributed at hundreds of sites along the male X chromosome (MELLER 2000 ). Mutations in the genes encoding five of the six known protein components display a distinctive male-specific lethal (MSL) phenotype, and so the products are collectively referred to as the MSL proteins. The two known RNA components of the MSL complex are roX1 and roX2 (RNA on the X; AMREIN and AXEL 1997 ; MELLER et al. 1997 ). The genes producing these two noncoding RNAs are located on the X, and their products are thought to spread in cis from the sites of synthesis when complexed with MSL proteins (KELLEY et al. 1999 ; PARK et al. 2002 ). The two roX RNAs are very dissimilar in size and sequence, and yet share a poorly understood function in dosage compensation. The MSL complex can bind the male X with either roX1 RNA or roX2 RNA alone, but only weakly if both RNAs are absent, resulting in male lethality (FRANKE and BAKER 1999 ; MELLER and RATTNER 2002 ).

When either roX gene is moved from the X to a random autosomal site, the MSL complex will bind to the roX DNA sequence anywhere in the genome (KELLEY et al. 1999 ; KAGEYAMA et al. 2001 ). The MSL complex also spreads into autosomal chromatin flanking such roX transgenes. If the MSL complex normally spreads in cis from the endogenous roX genes on the X, this could help explain how dosage compensation is targeted to the correct chromosome.

One prediction of the spreading model is that when the MSL complex is redirected to autosomal roX1 transgenes, the surrounding chromatin should be remodeled to resemble the male X, resulting in inappropriate hypertranscription. We previously reported that the affected segment of autosome becomes hyperacetylated at lysine 16 on histone H4 (KELLEY et al. 1999 ). Here we report that placing a roX1 gene next to a mini-white reporter can allow male-specific activation of eye color when the transgene is embedded in a variety of repressive autosomal chromatin environments. This reversal of silencing provides evidence that when the MSL complex is redirected to an autosome, it can remodel nearby chromatin to allow elevated levels of transcription. Thus, isolated roX transgenes inserted on autosomes can serve as a model to study MSL targeting, spreading, and chromatin remodeling.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Fly genetics:
Mosaic stocks were recovered in a screen where a second chromosome GMroX1 transposon was mobilized with P transposase (ROBERTSON et al. 1988 ). y w/Y; P{w⁺ GMroX1}57E/CyO; Ki {Js 2-3}99B/+ males were crossed to y w virgins. CyO; +/+ sons with pigmented eyes, which must have lost the original insert on the second chromosome, were recovered. Consequently, most inserts were on the third chromosome with a few recovered on CyO and the fourth chromosome. The new inserts were balanced and made homozygous if viable. The inserts were mapped on polytene chromosomes by anti-MSL1 antibody staining, which binds roX genes (KELLEY et al. 1999 ), and DNA in situ hybridization using a digoxygenin labeled roX1 probe. Due to the difficult cytology of CyO, the two mosaic lines recovered on this balancer were mapped only by sequencing the DNA flanking one end of the transposons by inverse PCR. The GMroX1-64A transgene was remobilized to recover hops with solid eye color in both sexes. The GMroX1-69C and GMroX1-64A.2 lines were recovered during this screen because of their distinctive eye patterns. The latter is at the original location, but suffered a deletion of sequences flanking the mini-white end of the transposon. To generate XXY daughters and XO sons, transgenic females were crossed to attached XY males: YS.X, In(1)EN, v ptg oc sn⁵ w y, YL oc⁸ y⁺ oc ptg.

Sequencing insertion sites:
Genomic DNA was cut with either HpaII or HhaI, ligated into circles, and then recut with HindIII. DNA flanking the transposon was amplified with primers 5'-TGAGAGGAAAGGTTGTGTGC-3' and 5'-TATCGACGGGACCACCTTAT-3', gel purified, and sequenced. The sequence was placed on Drosophila genome sequence (FlyBase release 3) using BLAST, Gadfly (http://www.fruitfly.org/annot/), and Flyenhancer (http://flyenhancer.org/Main).

Expressing MSL complex in female eyes:
P{y⁺ YEM2} (yellow eyeless msl2) carries a 3.6-kb eye-specific enhancer from the first intron of the eyeless gene (HALDER et al. 1998 ) driving transcription of the msl2 coding sequence, minus the regulatory SXL-binding sites in the 5' and 3' untranslated regions (UTRs; KELLEY et al. 1995 ). This was introduced into flies in the P{Carnegie 4-yellow} vector, a derivative of YES that lacks Su(Hw)-binding sites, so that eye pigmentation caused by GMroX1 could be assayed (PATTON et al. 1992 ; SIGRIST and PIRROTTA 1997 ).

Photography of fly heads:
Adults were aged for 3 weeks to maximize pigmentation and submerged in mineral oil. Brother and sister pairs were photographed side by side with Ektachrome 160T film using a Leica MZ12 microscope, digitized, and processed using Adobe Photoshop.

Scanning electron microscopy (SEM) of eyes:
Adults were dehydrated through an ethanol series, soaked in hexamethyldisilazane, and vacuum dried. After mounting on adhesive blocks, the flies were coated under vacuum using a Bal-Tec MED 020 high-resolution sputtering device (Technotrade International, Manchester, NH) with a platinum alloy target for 400 sec. Samples were examined in a JSM-5900 scanning electron microscope (JEOL, Peabody, MA) at an accelerating voltage of 5 kV.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Unusual male-specific mosaic eye pigmentation in roX1 transgenes:
The roX1 transgene used in this study was marked with the mini-white eye pigment gene (PIRROTTA 1988 ). In the course of experiments to insert the P{w⁺ GMroX1} transgene (hereafter referred to as GMroX1-location) in many autosomal sites, most new transgenic inserts showed the solid dark-orange (male) and solid light-orange (female) coloration typical of the CaSpeR vector. However, 10% of our roX1 transgenic stocks displayed unusual, sex-specific eye pigmentation patterns (Fig 1). In many cases, mini-white expression was very low or silent in females. In these same lines, males expressed mini-white, but frequently in mosaic sectors. All males from any single stock showed a similar mix of pigmented vs. white sectors, but there were large differences in the patterns between independent transgene insertion sites. For instance, GMroX1-39DE males had almost solid red eyes, but GMroX1-84E males had only a few small red sectors. In both lines, females had solid white eyes. Some lines, such as 64A and 80C, had a few large sectors, indicating that the decision to activate or to silence the transgene was made shortly after the eye disc formed during embryogenesis (Fig 1).

View larger version (0K): [in this window] [in a new window] [Download PPT slide]   Figure 1. Mosaic mini-white expression in GMroX1 transgenic flies. Female (left) and male (right) pairs of transgenic flies carrying hemizygous GMroX1 insertions at the indicated cytological locations are shown. For lines 35DE–99F (top) the order is XX females, XY males, XXY females, and XO males. For lines 82C, 80C, and 102C (bottom) the order is XX females, XY males, XXY females, and XO males, female and male Su(var)2-505/+, female and male Su(var)3-7/+, female and male Su(z)124/+. None of the modifiers increases mini-white expression in females. The inserts at 80C and 102C are located near heterochromatin and responded strongly to Su(var)3-7, moderately to Su(z)12, and not at all to Su(var)2-5. GMroX1-82C had the strongest response to Su(var) mutations of any euchromatic insertion. In each line, the males displayed a range of patterns with a slightly different percentage of the eye pigmented. Typical patterns are presented in each case.

The sectored eyes superficially resembled those reported in cases of position-effect variegation (PEV; WAKIMOTO 1998 ). One striking difference was that the mosaic patterns seen with roX transgenes were exclusively male specific. A second important difference was that PEV most often occurs when a normally euchromatic gene is placed near blocks of hererochromatin, but most roX1 transgenes under study were located in unremarkable euchromatic regions of polytene chromosomes (Fig 1, Table 1). The exceptions were GMroX1-80C at the heterochromatic base of 3L and GMroX1-102C in the banded portion of the heterochromatic fourth chromosome. Genes subject to PEV are often sensitive to the presence of the heterochromatic Y chromosome. We examined the mosaic roX1 transgenes in XO males and XXY females. In no case did the addition of a Y chromosome overcome silencing in females (Fig 1). In general, XO males displayed more severe transgene silencing compared to XY males. Not surprisingly, the strongest response was for the heterochromatic insertion at 80C, but some euchromatic inserts like 99C also had much less pigmentation (Fig 1). In several lines, such as GMroX1-64A and GMroX1-75C, the presence or absence of the Y had little or no effect on male pigmentation.

Numerous genes encoding proteins necessary for packaging silent chromatin have been identified as modifiers of PEV (WALLRATH 1998 ; GREWAL and ELGIN 2002 ). We tested mutations in Su(var)2-5 and Su(var)3-7 (encoding HP1; EISSENBERG et al. 1990 ) and a C2H2 Zn finger protein (CLEARD and SPIERER 2001 ), respectively, for dominant effects on GMroX1 eye pigmentation. We also tested Su(z)12, which encodes a C2H2 Zn finger protein that interacts with Polycomb group genes and was later found to affect variegation of w^m4 (BIRVE et al. 2001 ). None affected female eye color of any transgenic line (Fig 1, bottom, and data not shown). The same modifiers also had variable effects on male eye color that depended strongly on insertion site. The centric heterochromatin insert GMroX1-80C and one on the heterochromatic fourth chromosome, GMroX1-102C, responded to Su(var)3-7 and Su(z)12⁴ (Fig 1, bottom). In most cases, however, the suppressors of PEV had little effect on the male mosaic pattern (data not shown). We conclude that the roX1 inserts displaying male-specific mosaicism represent a new phenomenon distinct from PEV.

Most mosaic insertions are in gene-rich regions:
We favor a model in which the mosaic transgenes are inserted in locations unfavorable for mini-white expression. The male-specific MSL complex might assemble on nascent roX1 transcripts and then remodel the surrounding chromatin into a conformation permissive for mini-white expression. This idea is consistent with the observation that when the strongly variegated GMroX1-64A line was mobilized with P transposase, almost all new hops gave brothers and sisters with solid orange eyes (data not shown). This indicates that the silencing element was most likely in flanking DNA around 64A rather than within the transposon.

Examining the nearby genes or DNA sequence surrounding the mosaic transgenes might provide clues to the nature of the repressive chromatin. The exact locations of most transgenes were determined by sequencing the flanking DNA generated by inverse PCR (Table 1). The heterochromatic insertion at 80C landed within a degraded hoppel mobile element, and the GMroX1-75C transgene landed in a yoyo element. GMroX1-84E inserted into a low-copy repeat. The others landed in single-copy sequences, often densely packed with genes. In such a small sample size, it is surprising to recover strongly mosaic insertions at the 5' ends of both zfh1 (99F) and zfh2 (102C), the genes encoding large transcription factors containing both Zn fingers and a homeobox. The 64A insertion is near the 5' end of the scrt gene. These sites are expected to be packaged in active chromatin in at least some tissues and developmental times. It is possible that such regions are silenced in tissues where the resident genes are not needed. By contrast, GMroX1-82C landed in a region devoid of predicted genes for >25 kb on either side.

The insertion at 39DE illustrates the activating potential of the roX1 transgene. GMroX1-39DE landed in a copy of the histone H3 gene and mini-white was fully silent in females but active in males (Fig 1). The 5-kb histone gene clusters occur in tandem arrays of 100–200 copies. Others have reported that mini-white expression is repressed when embedded within tandem repeats (DORER and HENIKOFF 1994 ). The only P elements that we are aware of in this large interval were recovered using the yellow (y) body color marker, not mini-white (FlyBase, R. LEVIS, personal communication).

The endogenous genes surrounding the mosaic GMroX1 inserts might be silent in the developing eye, but active elsewhere. Histone H4 methylation at lysine 20 is a candidate for such an epigenetic silencing factor. This modification is mutually exclusive with histone H4 acetylation at lysine 16 produced by the MOF protein within the MSL complex (NISHIOKA et al. 2002 ). The pr-set7 gene is responsible for this methylation, which is found not only in centric heterochromatin, but also over the euchromatic arms. Because pr-set7 is an essential gene, we could assay only adults with lowered PR-SET7 levels. We tested seven mosaic GMroX1 lines in Df(3R) red^P93/+ males, which should produce only 50% of the amount of PR-SET7 methylase. In each case the hemizygous pr-set7 males showed mosaic eye patterns indistinguishable from their +/+ brothers (data not shown).

Pairing-sensitive repression:
When most mini-white transgenes (without roX1 sequences) are made homozygous, the eye pigmentation is darker than that in hemizygotes because of increased gene dose. Several viable mosaic inserts in this study showed significantly less pigmentation when homozygous compared to hemizygous (Fig 2). The most extreme case was GMroX1-84E in which the transgene inserted in a 4.5-kb low-copy repeat element present twice near the CG2616 gene with additional copies at 38D and 41F (FlyBase). Most GMroX1-84E homozygous males had solid white eyes with only 20% showing one or two small red sectors. In other lines, hemizygous animals had even pigmentation over the entire eye, but homozygous animals had decreased mini-white expression in both sexes. Males overcame this repression in a mosaic pattern (Fig 2). Pairing-dependent silencing of the white gene has been reported in numerous other situations and is often due to the action of the PcG proteins (KASSIS et al. 1991 ; CHAN et al. 1994 ; AMERICO et al. 2002 ). We tested whether the most dramatic pairing-sensitive insert, GMroX1-98B, was affected by reduced PcG in Scm and Pc³ mutant backgrounds, but found no effect (data not shown).

View larger version (0K): [in this window] [in a new window] [Download PPT slide]   Figure 2. Transgene pairing affects mini-white expression. Insertions are shown as hemizygous (left pair) or homozygous (right pair). For each pair, females are on the left and males are on the right. The numbers at left indicate insertion sites of transgenes. Most insertions have lower mini-white expression when the transgene is paired except the fourth chromosome insert, 102C, where homozygous males have solid red eyes. Some pairing-sensitive lines (77A, 93E, 95E, and 98B) show a slight salt-and-pepper pattern in females, which is greatly exacerbated in males. The following lines carried lethal mutations on the same chromosome and could not be tested as homozygotes: 35E, 39DE, 56D, 64A, 64A.2, 80C, 82C, and 99F.

A striking exception to this trend was the GMroX1-102C insert on the small, heterochromatic fourth chromosome. The transgene is silent in females whether hemizygous or homozygous, but pairing dramatically increases pigmentation in males so that they have solid red eyes (Fig 2). This insert lies only a few hundred nucleotides from the M371.R insertion site of P{hsp26-pt hsp70-w⁺} (SUN et al. 2000 ), and yet M371.R is a lethal insert in the 5' end of zfh2 whereas GMroX1-102C is homozygous viable. The hsp70-w⁺ marker is strongly expressed in both sexes to give dark red eyes. It is not clear how much the dramatically different eye phenotypes between the two transposons are determined by the different promoters driving white or the slightly separated insertion sites.

iroquois insertion produces dorsoventral pattern:
The GMroX1-69C insertion landed in the second intron of ara and had a distinctive pattern of eye pigmentation. The dorsal 30% of the eye was pigmented in hemizygotes of both sexes, but the ventral 70% was completely white in females and sectored in males (Fig 3A and Fig B). Others have recovered transposon inserts with dorsal red/ventral white patterns (Fig 3F) and they all map near 69D, the location of the iroquois-mirror cluster of homeotic genes, which includes ara (SUN et al. 1995 ; GOMEZ-SKARMETA et al. 1996 ; NETTER et al. 1998 ). These genes are expressed in many tissues, but in the developing eye disc they are expressed only in the dorsal half (MCNEILL et al. 1997 ; KEHL et al. 1998 ). The ventral silencing of transgenes in the iroquois complex is due to action of the Polycomb complex and the PcG proteins are found at 69CD in salivary gland polytene chromosomes (RASTELLI et al. 1993 ; NETTER et al. 1998 ). When the Pc³ mutation was introduced into GMroX1-69C flies, weak derepression of the mini-white marker was evident by light pigmentation in the ventral region of both male and female eyes, suggesting that this transgene is also silenced by the PcG proteins (Fig 3C).

View larger version (0K): [in this window] [in a new window] [Download PPT slide]   Figure 3. Insertion into the iroquois cluster. GMroX1-69C is located in the second intron of the ara gene within the 140-kb iroquois cluster of homeotic genes located at 69CD (G). The dorsal 30% of female eyes are pigmented (A), but the ventral region is completely silenced as is typical for many other mini-white marked transgenes in this region like P{w+ lacZ} mirrB1-12 (F). GMroX1-69C males have the same pattern as females except red sectors are present ventrally (B). The ventral silencing is mediated by the Pc group of proteins. y w; GMroX1-69C +/ + Pc3 females show weak derepression of mini-white in the ventral half (C). Pairing of the transgene causes ventral derepression in the ventral half of the eye in small sectors in 20% of homozygous GMroX1-69C females (D) or almost total derepression in most homozygous males (E). Such males often have a nonpigmented equator.

In contrast to pairing-dependent repression seen in most other lines, homozygous GMroX1-69C males have nearly solid red eyes with a lighter equator (Fig 3E). This is the converse pattern seen in the nearby Eq1 mini-white transgene lacking any roX1 sequences (NETTER et al. 1998 ). Pairing has little effect in females except that 10% of homozygous females have a small pigmented sector in the ventral half of the eye (Fig 3D).

Ectopic dosage compensation in females activates white expression:
Formally, several sex-specific regulatory proteins might mediate silencing in females (SXL, TRA, DSX^f) or activation in males (DSX^m, FRU). However, given the well-documented interaction between the MSL complex and the roX genes, we tested whether ectopic MSL expression in females could overcome repressive chromatin around the GMroX1. Females normally make all the MSL proteins except MSL2 (BASHAW and BAKER 1995 ; KELLEY et al. 1995 ; ZHOU et al. 1995 ). The P{y⁺ YEM2} transgene (Fig 4A) expressed MSL2 protein in the developing eye disc in both sexes under the control of the eyeless enhancer. This caused a low-penetrance female-specific eye defect (data not shown). The rough eye phenotype was most likely due to ectopic dosage compensation because it was strongly enhanced by simultaneous overexpression of MSL1 (Fig 4, C–E) but had no effect in males (Fig 4B). When P{y⁺ YEM2} was crossed into the GMroX1-102C stock, females' eyes changed from completely silent to having red sectors (Fig 4F vs. G). This demonstrates that the MSL complex is the key factor responsible for the male-specific activation of mini-white observed in these mosaic GMroX1 lines.

View larger version (0K): [in this window] [in a new window] [Download PPT slide]   Figure 4. The MSL complex overcomes mini-white silencing. P{y+YEM2} (A) was used to express the MSL complex in eyes using the eye-specific enhancer from the second intron of eyeless (dark green) to drive expression of the msl2 coding sequence (violet) from a minimal Hsp70 promoter (light green). The tra2 3' UTR and polyadenylation site (orange) was placed downstream of msl2. When P{y+YEM2} was present alone, the MSL2 protein in the eye was sufficient to cause only a very mild rough eye defect in 2–5% of females due to ectopic dosage compensation (not shown). However, this phenotype is dramatically enhanced by overexpression of MSL1 from P{w+ H83M1} (CHANG and KURODA 1998 ). (B) y w/Y; P{y+YEM2}9/P{y+YEM2}9 P{w+ H83M1} males have wild-type eyes. (C–E) Approximately 80% of similar females show sectors of cell death due to ectopic dosage compensation. (F) The mini-white gene in y w; GMroX1-102C females is silent. (G) The mini-white gene is derepressed when MSL2 is expressed in the eyes of y w; P{y+YEM2}9; GMroX1-102C females.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Expression of the white gene is exquisitely sensitive to its chromatin environment. The mini-white derivative used here is further debilitated by the loss of its eye enhancer. Several large screens have shown that expression of visible marker genes such as white and yellow variegate when placed near blocks of heterochromatin located around the centromeres (ROSEMAN et al. 1995 ; WALLRATH and ELGIN 1995 ; CRYDERMAN et al. 1998 ; YAN et al. 2002 ), the fourth chromosome (SUN et al. 2000 ), and at telomeres (CRYDERMAN et al. 1999 ). A few examples of white silencing/variegation in euchromatin regions have been reported, for example, when multicopy arrays of white transposons are generated (DORER and HENIKOFF 1994 ). In a screen using transposons carrying both white and a green fluorescent protein (GFP) reporter driven by PAX6-binding sites, up to 20% of the Drosophila euchromatic insertions expressed GFP, but not mini-white (HORN et al. 2000 ). The nature of these repressive regions is not understood, but they show that a large fraction of the euchromatin will silence mini-white expression. It is likely that all P-element screens generate many inserts at such sites that go unnoticed because the transgenic flies fail to express the marker. We postulate that we recovered inserts in a subset of these unfavorable sites because males could overcome silencing.

A model for male-specific mosaic eyes:
We are unaware of other mini-white marked transgenes with the characteristics of those described here and infer that the roX1 gene is responsible for this unusual behavior. We propose that male-specific pigmented sectors reported here are a visible manifestation of ectopic dosage compensation occurring around autosomal GMroX1 transgenes, which landed in repressive chromatin environments (Fig 5). The MSL complex is active by midembryogenesis and stays on throughout development (RASTELLI et al. 1995 ; FRANKE et al. 1996 ). The mosaic eye patterns seen here suggest that the MSL complex spreads a more "open" chromatin architecture during embryonic development when the primordial eye disc has a small number of cells. This chromatin packaging competes with uncharacterized silencing factors and can be inherited through many mitotic divisions so that large clones of cells in the adult eye share the same on/off state. These results are similar to those reported for a w⁺ transgene lacking roX1 inserted at the heterochromatic base of the X (WALLRATH et al. 1996 ). Such females suffered PEV but males had solid red eyes presumably due to the MSL complex spreading from flanking chromatin.

View larger version (0K): [in this window] [in a new window] [Download PPT slide]   Figure 5. Model for mosaic eyes. In 10% of all cases, GMroX1 inserts into a chromatin domain that is subject to silencing due to the action of heterochromatin, the Polycomb complex, repeat-induced silencing, pairing-induced silencing, or other unknown factors. In a female, these factors spread so that the transgene is packaged in the same repressive chromatin (right). The MSL complex present in males recognizes the roX1 gene and spreads into flanking chromatin as it is postulated to do on the X. The resulting histone modifications overcome silencing, allowing the mini-white reporter to be transcribed (left). The fraction of cells in which the MSL complex successfully remodels the local chromatin determines the eye pigmentation pattern.

This model is supported by the finding that ectopic expression of MSL2 in females is sufficient to overcome silencing. Thus the MSL complex is responsible for the activation, but must be targeted to the transgene. This could happen either by MSL proteins assembling on nascent roX1 transcripts (PARK et al. 2002 ) or by mature MSL complex being recruited by DNA sequences within the roX1 gene (KAGEYAMA et al. 2001 ).

An alternative interpretation of dosage compensation in Drosophila, known as the inverse model, postulates that the MSL proteins normally have two key functions in wild-type males. First, they sequester the MOF histone H4 acetyltransferase away from the autosomes by targeting it to the X chromosome. Second, the MSLs block overexpression of X-linked genes that might otherwise result from MOF-mediated nucleosome acetylation (BHADRA et al. 1999 ). In this model, histone acetylation by MOF has little effect on gene transcription in the wild-type male X chromosome, but a significant toxic effect in mle mutant males where MOF escapes from the X and hyperacetylates the entire genome. In contrast to expectations of the inverse model, we find that histone H4 acetylation caused by MOF within complete MSL complexes is a potent activator in wild-type males.

The spreading model of MSL recognition predicts that the expression of autosomal genes in the vicinity of an ectopic roX1 gene is elevated about twofold in nuclei where spreading occurs. The results reported here are consistent with the idea that the histone aectylation produced by the MSL complex near autosomal roX transgenes elevates expression of genes. In this situation we sometimes observed more than the expected twofold effect not because mini-white expression was elevated too much in males—the pigmented sectors were usually orange, suggesting modest expression—but rather the unusually low basal expression in the female eyes and silent male sectors was responsible for the large difference between "high" and "low" states. In a few lines such as 77A, 93E, 99C, and 98B, females did have detectable mini-white expression and males had sectors that appear about twofold darker than this basal expression. These results are consistent with a recent report that autosomal roX1 transgenes could modestly elevate the expression of a flanking lacZ reporter in males (HENRY et al. 2001 ).

Placing Polycomb response elements (PREs) near mini-white can cause mosaic silencing by recruiting the Polycomb complex (FAUVARQUE and DURA 1993 ; CHAN et al. 1994 ; SIGRIST and PIRROTTA 1997 ; AMERICO et al. 2002 ). This prompted us to consider whether the roX1 gene itself, rather than flanking chromatin, somehow caused repression of mini-white in females. The strongest argument against this idea is that the 69C and 80C inserts are clearly silenced by flanking sequences. Also, in many lines the males also have silent sectors, but it is difficult to imagine how roX1 would act to repress genes in males. The fact that females from the large majority of GMroX1 transgenic lines have solid eye pigmentation also suggests that the roX1 sequence does not act as a silencer. The mosaic lines presented here account for only 10% of all inserts. However, we cannot exclude the possibility that roX1 sequences mediate female silencing in combination with unknown factors in some lines.

In some circumstances PRE-mediated silencing is pairing dependent (CHAN et al. 1994 ; AMERICO et al. 2002 ). It is not clear whether roX1 sequences or flanking DNA contribute to the pairing-sensitive silencing seen in some GMroX1 lines. The only pairing-sensitive inserts located near endogenous PcG-binding sites are at 84E and 93E (RASTELLI et al. 1993 ). We did note that PREs often contain binding sites for the GAGA/Pipsqueak factors (HORARD et al. 2000 ; AMERICO et al. 2002 ) and the major MSL-binding site within the roX1 gene also contains several conserved GAGA sites (KAGEYAMA et al. 2001 and data not shown).

Another explanation for male-specific activation might be that the vigorous transcription of roX1 in males somehow prevents surrounding DNA from being packaged into silent chromatin similar to that reported for transgenes responding to GAL4 (ZINK and PARO 1995 ; AHMAD and HENIKOFF 2001 ). Although regulation of roX transcription is not well understood, let us assume that MSL proteins drive roX1 expression only in males. Transcription of roX1 RNA in cis may indeed be a prerequisite for the nearby region to escape silencing, but we postulate a narrow role. We have previously suggested that MSL proteins assemble onto nascent roX transcripts and that this favors spreading into adjacent chromatin (PARK et al. 2002 ). We propose that locally active RNA polymerase is not sufficient. For instance, strong transcription of Pax6-GFP in the eye does not guarantee expression of an adjacent white gene (HORN et al. 2000 ). We postulate that in the cases reported here roX1 transcription is significant only in that it produces MSL complexes able to spread into flanking chromatin, modifying nucleosomes along the way.

Repressive nature of flanking chromatin:
The MSL complex can overcome different mechanisms of silencing. The GMroX1-80C line is subject to severe PEV. The surprising aspect of this insert is the strength of silencing in females where neither the presence of a Y chromosome nor the presence of Su(var) mutations allowed any mini-white expression. Yet in males, the MSL complex can spread from roX1 sequences through centric heterochromatin and into the euchromatic proximal arms of 3L and 3R, activating mini-white along the way (PARK et al. 2002 ). The insertion in the iroquois cluster demonstrates that the MSL complex can overcome Polycomb-mediated silencing in the ventral half of the eye. The MSL complex can also overcome silencing due to insertion in dispersed repeats (75C and 84E).

The insertion in one of the 110 tandem copies of the histone gene cluster at 39DE is particularly interesting. Mini-white is sometimes poorly expressed within long repeats (DORER and HENIKOFF 1994 ). A second explanation rests on close proximity of the histone cluster to centric heterochromatin. A high histone gene copy number had been thought necessary to supply cells with enough histone proteins during each replication cycle. However, the discovery that Drosophila hawaiiensis carries 20 copies of the histone genes and D. hydei carries only 5–10 copies called for a new explanation (FITCH et al. 1990 ). In species with low copy number, the histone genes are located far from heterochromatin. However, the histone genes in D. melanogaster are adjacent to centric heterochromatin. Selection for increased copy number may have compensated for low expression per gene copy (FITCH et al. 1990 ).

The silencing mechanism remains a mystery for most inserts. The idea that flanking genes are kept silent because they are not needed for eye development is difficult to test because few of the surrounding genes are well characterized. Ectopic expression of the zfh1 gene, interrupted by GMroX1-99F, disrupts eye development (LAI et al. 1991 ), but zfh1 expression in wild-type eye discs has not been reported. The GMroX1-64A insert landed upstream of the scrt gene encoding a Zn finger transcription factor regulating neuron development (ROARK et al. 1995 ). Although the scrt gene is expressed in the developing neurons of the eye, the insert is located in an enhancer element driving expression in the central nervous system (EMERY and BIER 1995 ). The scrt gene appears to be silent in the nonneuronal pigment cells where mini-white should be expressed. Our failure to detect any interaction between mutations in the pr-set7 histone H4 methyltransferase gene and the mosaic transgenes cannot rule out a role for histone methylation in euchromatic gene silencing. Lowering methylation by 50% may be insufficient to change the mosaic pattern in eyes, or a different methylase may be responsible.

In summary, the MSL complex is a versatile chromatin-remodeling machine able to act on many different chromatin substrates. This might be expected for a regulator that must normally act on several thousand unrelated genes expressed in different tissues throughout development. However, this behavior raises the question of how males can keep appropriate segments of the X silent in tissues in which a gene product is not needed and might even be harmful. Presumably the MSL complex is tightly regulated on the X so that only active genes are upregulated. Others have shown that the MSL complex can radically alter the morphology of the X when certain chromatin-modifying factors, such as ISWI or NURF, are mutated (DEURING et al. 2000 ; BADENHORST et al. 2002 ; CORONA et al. 2002 ). Perhaps such proteins normally restrict the action of the MSL complex. In addition, chromosomes may be organized into loops or domains of activity in vivo so that the MSL complex can respect domain boundaries if it spreads along the chromosome in cis (GU et al. 2000 ). The roX1 transgenes studied here may subvert such regulation by placing a MSL-binding/assembly site internal to domain boundary elements.

Because of its extreme sensitivity to a chromatin environment, mini-white-based P elements are being replaced with yellow⁺ or PAX6-EGFP marked vectors for mutagenesis screens (ROSEMAN et al. 1995 ; HORN et al. 2000 ; YAN et al. 2002 ). However, the GMroX1 transposon may be useful in screens to assay for repressive chromatin environments. Simply comparing the eye color of brothers and sisters from the same stock would quickly identify euchromatic inserts subject to subtle chromatin effects.

	ACKNOWLEDGMENTS

We thank Xiaowen Chu and Hilda Kennedy for sequencing the transposon insertion sites. Lori Wallrath, Kwang Choi, Vince Pirrotta, Juerg Mueller, and the Bloomington Stock Center provided fly stocks. We thank Pam Geyer and Georg Halder for providing plasmids, Benjamin Frankfort for advice on SEM, and Milan Jamrich for the use of his photomicroscope. The SEM images were collected by Kenneth Dunner of the High Resolution Electron Microscopy Facility, UTMDACC Institutional Core grant no. CA16672. R.L.K. is supported by National Institutes of Health grant no. GM-45744. M.I.K. is an investigator with the Howard Hughes Medical Institute.

Manuscript received November 11, 2002; Accepted for publication March 3, 2003.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AHMAD, K. and S. HENIKOFF, 2001  Modulation of a transcription factor counteracts heterochromatic gene silencing in Drosophila.. Cell 104:839-847.[Medline]

AMERICO, J., M. WHITELEY, J. L. BROWN, M. FUJIOKA, and J. B. JAYNES et al., 2002  A complex array of DNA-binding proteins required for pairing-sensitive silencing by a polycomb group response element from the Drosophila engrailed gene. Genetics 160:1561-1571.[Abstract/Free Full Text]

AMREIN, H. and R. AXEL, 1997  Genes expressed in neurons of adult male Drosophila.. Cell 88:459-469.[Medline]

BADENHORST, P., M. VOAS, I. REBAY, and C. WU, 2002  Biological function of the ISWI chromatin remodeling complex NURF. Genes Dev. 16:3186-3198.[Abstract/Free Full Text]

BASHAW, G. J. and B. S. BAKER, 1995  The msl-2 dosage compensation gene of Drosophila encodes a putative DNA-binding protein whose expression is sex specifically regulated by Sex-lethal.. Development 121:3245-3258.[Abstract]

BHADRA, U., M. PAL-BHADRA, and J. A. BIRCHLER, 1999  Role of the male specific lethal (msl) genes in modifying the effects of sex chromosomal dosage in Drosophila. Genetics 152:249-268.[Abstract/Free Full Text]

BIRVE, A., A. K. SENGUPTA, D. BEUCHEL, J. LARSSON, and J. A. KENNISON et al., 2001  Su(z)12, a novel Drosophila polycomb group gene that is conserved in vertebrates and plants. Development 128:3371-3379.[Abstract/Free Full Text]

CHAN, C.-S., L. RASTELLI, and V. PIRROTTA, 1994  A Polycomb response element in the Ubx gene that determines an epigenetically inherited state of repression. EMBO J. 13:2553-2564.[Medline]

CHANG, K. A. and M. I. KURODA, 1998  Modulation of MSL1 abundance in female Drosophila contributes to the sex specificity of dosage compensation. Genetics 150:699-709.[Abstract/Free Full Text]

CLEARD, F. and P. SPIERER, 2001  Position-effect variegation in Drosophila: the modifier Su(var)3-7 is a modular DNA-binding protein. EMBO Rep. 2:1095-1100.[Medline]

CLINE, T. W. and B. J. MEYER, 1996  Vive la difference: males vs. females in flies vs. worms. Annu. Rev. Genet. 30:637-702.[Medline]

CORONA, D. F. V., C. R. CLAPIER, P. B. BECKER, and J. W. TAMKUN, 2002  Modulation of ISWI function by site-specific histone acetylation. EMBO Rep. 3:242-247.[Medline]

CRYDERMAN, D. E., M. H. CUAYCONG, S. C. R. ELGIN, and L. L. WALLRATH, 1998  Characterization of sequences associated with position-effect variegation at pericentric sites in Drosophila heterochromatin. Chromosoma 107:277-285.[Medline]

CRYDERMAN, D. E., E. J. MORRIS, H. BIESSMANN, S. C. ELGIN, and L. L. WALLRATH, 1999  Silencing at Drosophila telomeres: nuclear organization and chromatin structure play critical roles. EMBO J. 18:3724-3735.[Medline]

DEURING, R., L. FANTI, J. A. ARMSTRONG, M. SARTE, and O. PAPOULAS et al., 2000  The ISWI chromatin-remodeling protein is required for gene expression and the maintenance of higher order chromatin structure in vivo. Mol. Cell 5:355-365.[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]

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]

EMERY, J. F. and E. BIER, 1995  Specificity of CNS and PNS regulatory subelements comprising pan-neural enhancers of the deadpan and scratch genes is achieved by repression. Development 121:3549-3560.[Abstract]

FAUVARQUE, M.-O. and J.-M. DURA, 1993  polyhomeotic regulatory sequences induce developmental regulator-dependent variegation and targeted P-element insertions in Drosophila.. Genes Dev. 7:1508-1520.[Abstract/Free Full Text]

FITCH, D. H., L. D. STRAUSBAUGH, and V. BARRETT, 1990  On the origins of tandemly repeated genes: Does histone gene copy number in Drosophila reflect chromosomal location? Chromosoma 99:118-124.[Medline]

FRANKE, A. and B. S. BAKER, 1999  The roX1 and roX2 RNAs are essential components of the compensasome, which mediates dosage compensation in Drosophila.. Mol. Cell 4:117-122.[Medline]

FRANKE, A., A. DERNBURG, G. J. BASHAW, and B. S. BAKER, 1996  Evidence that MSL-mediated dosage compensation in Drosophila begins at blastoderm. Development 122:2751-2760.[Abstract]

GOMEZ-SKARMETA, J. L., R. DIEZ DEL CORRAL, E. DE LA CALLE-MUSTIENES, D. FERRES-MARCO, and J. MODOLELL, 1996  Araucan and caupolican, two members of the novel Iroquois complex, encode homeoproteins that control proneural and vein-forming genes. Cell 85:95-105.[Medline]

GREWAL, S. I. and S. C. ELGIN, 2002  Heterochromatin: new possibilities for the inheritance of structure. Curr. Opin. Genet. Dev. 12:178-187.[Medline]

GU, W., X. WEI, A. PANNUTI, and J. C. LUCCHESI, 2000  Targeting the chromatin-remodeling MSL complex of Drosophila to its sites of action on the X chromosome requires both acetyl transferase and ATPase activities. EMBO J. 19:5202-5211.[Medline]

HALDER, G., P. CALLAERTS, S. FLISTER, U. WALLDORF, and U. KLOTER et al., 1998  Eyeless initiates the expression of both sine oculis and eyes absent during Drosophila compound eye development. Development 125:2181-2191.[Abstract]

HENRY, R. A., B. TEWS, X. LI, and M. J. SCOTT, 2001  Recruitment of the male-specific lethal (MSL) dosage compensation complex to an autosomally integrated roX chromatin entry site correlates with an increased expression of an adjacent reporter gene in male Drosophila.. J. Biol. Chem. 276:31953-31958.[Abstract/Free Full Text]

HORARD, B., C. TATOUT, S. POUX, and V. PIRROTTA, 2000  Structure of a polycomb response element and in vitro binding of polycomb group complexes containing GAGA factor. Mol. Cell. Biol. 20:3187-3197.[Abstract/Free Full Text]

HORN, C., B. JAUNICH, and E. A. WIMMER, 2000  Highly sensitive, fluorescent transformation marker for Drosophila transgenesis. Dev. Genes Evol. 210:623-629.[Medline]

KAGEYAMA, Y., G. MENGUS, G. GILFILLAN, H. G. KENNEDY, and C. STUCKENHOLZ et al., 2001  Association and spreading of the Drosophila dosage compensation complex from a discrete roX1 chromatin entry site. EMBO J. 20:2236-2245.[Medline]

KASSIS, J. A., E. P. VANSICKLE, and S. M. SENSABAUGH, 1991  A fragment of engrailed regulatory DNA can mediate transvection of the white gene in Drosophila. Genetics 128:751-761.[Abstract]

KEHL, B. T., K.-O. CHO, and K.-W. CHOI, 1998  mirror, a Drosophila homeobox gene in the iroquois complex, is required for sensory organ and alula formation. Development 125:1217-1227.[Abstract]

KELLEY, R. L., I. SOLOVYEVA, L. M. LYMAN, R. RICHMAN, and V. SOLOVYEV et al., 1995  Expression of Msl-2 causes assembly of dosage compensation regulators on the X chromosomes and female lethality in Drosophila.. Cell 81:867-877.[Medline]

KELLEY, R. L., V. H. MELLER, P. R. GORDADZE, G. ROMAN, and R. L. DAVIS et al., 1999  Epigenetic spreading of the Drosophila dosage compensation complex from roX RNA genes into flanking chromatin. Cell 98:513-522.[Medline]

LAI, Z. C., M. E. FORTINI, and G. M. RUBIN, 1991  The embryonic expression patterns of zfh-1 and zfh-2, two Drosophila genes encoding novel zinc-finger homeodomain proteins. Mech. Dev. 34:123-124.[Medline]

MCNEILL, H., C. H. YANG, M. BRODSKY, J. UNGOS, and M. A. SIMON, 1997  mirror encodes a novel PBX-class homeoprotein that functions in the definition of the dorsal-ventral border in the Drosophila eye. Genes Dev. 11:1073-1082.[Abstract/Free Full Text]

MELLER, V. H., 2000  Dosage compensation: making 1X equal 2X. Trends Cell Biol. 10:54-59.[Medline]

MELLER, V. H. and B. P. RATTNER, 2002  The roX genes encode redundant male-specific lethal transcripts required for targeting of the MSL complex. EMBO J. 21:1084-1091.[Medline]

MELLER, V. H., K. H. WU, G. ROMAN, M. I. KURODA, and R. L. DAVIS, 1997  roX1 RNA paints the X chromosome of male Drosophila and is regulated by the dosage compensation system. Cell 88:445-457.[Medline]

NETTER, S., M. O. FAUVARQUE, R. DIEZ DEL CORRAL, J. M. DURA, and D. COEN, 1998  white⁺ transgene insertions presenting a dorsal/ventral pattern define a single cluster of homeobox genes that is silenced by the polycomb-group proteins in Drosophila melanogaster.. Genetics 149:257-275.[Abstract/Free Full Text]

NISHIOKA, K., J. C. RICE, K. SARMA, H. ERDJUMENT-BROMAGE, and J. WERNER et al., 2002  PR-Set7 is a nucleosome-specific methyltransferase that modifies lysine 20 of histone H4 and is associated with silent chromatin. Mol. Cell 9:1201-1213.[Medline]

PARK, Y., R. L. KELLEY, H. OH, M. I. KURODA, and V. H. MELLER, 2002  Extent of chromatin spreading determined by roX RNA recruitment of MSL proteins. Science 298:1620-1623.[Abstract/Free Full Text]

PATTON, J. S., X. V. GOMES, and P. K. GEYER, 1992  Position-independent germline transformation in Drosophila using a cuticle pigmentation gene as a selectable marker. Nucleic Acids Res. 20:5859-5860.[Free Full Text]

PIRROTTA, V., 1988 Vectors for P-mediated transformation in Drosophila, pp. 437–456 in Vectors: A Survey of Molecular Cloning Vectors and Their Uses, edited by R. L. RODRIGUEZ and D. T. DENHARDT. Butterworths, Boston.

RASTELLI, L., C. S. CHAN, and V. PIRROTTA, 1993  Related chromosome binding sites for zeste, suppressors of zeste and Polycomb group proteins in Drosophila and their dependence on Enhancer of zeste function. EMBO J. 12:1513-1522.[Medline]

RASTELLI, L., R. RICHMAN, and M. I. KURODA, 1995  The dosage compensation regulators MLE, MSL-1 and MSL-2 are interdependent since early embryogenesis in Drosophila.. Mech. Dev. 53:223-233.[Medline]

ROARK, M., M. S. STURTEVANT, J. EMERY, H. VAESSIN, and E. GRELL et al., 1995  scratch, a pan-neural gene encoding a zinc finger protein related to snail, promotes neuronal development. Genes Dev. 9:2384-2398.[Abstract/Free Full Text]

ROBERTSON, A. M., C. R. PRESTON, R. W. PHILLIS, D. JOHNSON-SCHLITZ, and W. K. BENZ et al., 1988  A stable genomic source of P element transposase in Drosophila melanogaster.. Genetics 118:461-470.[Abstract/Free Full Text]

ROSEMAN, R. R., E. A. JOHNSON, C. K. RODESCH, M. BJERKE, and R. N. NAGOSHI et al., 1995  A P element containing suppressor of Hairy-wing binding regions has novel properties for mutagenesis in Drosophila melanogaster.. Genetics 141:1061-1074.[Abstract]

SIGRIST, C. J. A. and V. PIRROTTA, 1997  Chromatin insulator elements block the silencing of a target gene by the Drosophila polycomb response element (PRE) but allow trans interactions between PREs on different chromosomes. Genetics 147:209-221.[Abstract]

SUN, F.-L., M. H. CUAYCONG, C. A. CRAIG, L. L. WALLRATH, and J. LOCKE et al., 2000  The fourth chromosome of Drosophila melanogaster: interspersed euchromatic and heterochromatic domains. Proc. Natl. Acad. Sci. USA 97:5340-5345.[Abstract/Free Full Text]

SUN, Y. H., C. J. TSAI, M. M. GREEN, J. L. CHAO, and C. T. YU et al., 1995  white as a reporter gene to detect transcriptional silencers specifying position-specific gene expression during Drosophila melanogaster eye development. Genetics 141:1075-1086.[Abstract]

WAKIMOTO, B. T., 1998  Beyond the nucleosome: epigenetic aspects of position-effect variegation in Drosophila.. Cell 93:321-324.[Medline]

WALLRATH, L. L., 1998  Unfolding the mysteries of heterochromatin. Curr. Opin. Genet. Dev. 8:147-153.[Medline]

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

WALLRATH, L. L., V. P. GUNTUR, L. E. ROSMAN, and S. C. ELGIN, 1996  DNA representation of variegating heterochromatic P-element inserts in diploid and polytene tissues of Drosophila melanogaster.. Chromosoma 104:519-527.[Medline]

YAN, C. M., K. W. DOBIE, H. D. LE, A. Y. KONEV, and G. H. KARPEN, 2002  Efficient recovery of centric heterochromatin P-element insertions in Drosophila melanogaster.. Genetics 161:217-229.[Abstract/Free Full Text]

ZHOU, S., Y. YANG, M. J. SCOTT, A. PANNUTI, and K. C. FEHR et al., 1995  Male-specific-lethal 2, a dosage compensation gene of Drosophila that undergoes sex-specific regulation and encodes a protein with a RING finger and a metallothionein-like cluster. EMBO J. 14:2884-2895.[Medline]

ZINK, D. and R. PARO, 1995  Drosophila polycomb-group regulated chromatin inhibits the accessibility of a trans-activator to its target DNA. EMBO J. 14:5660-5671.[Medline]

This article has been cited by other articles:

				A. A. Alekseyenko, E. Larschan, W. R. Lai, P. J. Park, and M. I. Kuroda High-resolution ChIP-chip analysis reveals that the Drosophila MSL complex selectively identifies active genes on the male X chromosome Genes & Dev., April 1, 2006; 20(7): 848 - 857. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kelley, R. L. Articles by Kuroda, M. I. Search for Related Content PubMed PubMed Citation Articles by Kelley, R. L. Articles by Kuroda, M. I.