Telomeric Associated Sequences of Drosophila Recruit Polycomb-Group Proteins in Vivo and Can Induce Pairing-Sensitive Repression

Corresponding author: Stéphane Ronsseray, Professeur D. Anxolabéhère, Institut Jacques Monod, Couloir 42-32, Etage 4, Université Paris 7, 2 Place Jussieu, 75005 Paris, France., ronsseray{at}ijm.jussieu.fr (E-mail)

Communicating editor: J. A. BIRCHLER

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

In Drosophila, relocation of a euchromatic gene near centromeric or telomeric heterochromatin often leads to its mosaic silencing. Nevertheless, modifiers of centromeric silencing do not affect telomeric silencing, suggesting that each location requires specific factors. Previous studies suggest that a subset of Polycomb-group (PcG) proteins could be responsible for telomeric silencing. Here, we present the effect on telomeric silencing of 50 mutant alleles of the PcG genes and of their counteracting trithorax-group genes. Several combinations of two mutated PcG genes impair telomeric silencing synergistically, revealing that some of these genes are required for telomeric silencing. In situ hybridization and immunostaining experiments on polytene chromosomes revealed a strict correlation between the presence of PcG proteins and that of heterochromatic telomeric associated sequences (TASs), suggesting that TASs and PcG complexes could be associated at telomeres. Furthermore, lines harboring a transgene containing an X-linked TAS subunit and the mini-white reporter gene can exhibit pairing-sensitive repression of the white gene in an orientation-dependent manner. Finally, an additional binding site for PcG proteins was detected at the insertion site of this type of transgene. Taken together, these results demonstrate that PcG proteins bind TASs in vivo and may be major players in Drosophila telomeric position effect (TPE).

HETEROCHROMATIN was originally defined as nuclear domains that remain condensed throughout the cell cycle (HEITZ 1928 ). In most eukaryotic genomes, it corresponds to telomeric and pericentric regions of chromosomes that are relatively gene poor and late replicating. Conversely, the chromosome arms are mainly euchromatic, corresponding to gene-rich, early replicating domains that decondense during interphase. When a euchromatic gene is relocated in the neighborhood of pericentric heterochromatin by a chromosomal rearrangement or by transposition, it can be randomly silenced in some cells (WEILER and WAKIMOTO 1995 ). This phenomenon has been termed position effect variegation (PEV). In Drosophila, >100 dominant mutations that alter PEV have been identified genetically: those that decrease gene silencing have been termed suppressors of variegation [Su(var)], whereas those that increase gene silencing have been termed enhancers of variegation [E(var); REUTER and SPIERER 1992]. Most of the Su(var) genes behave as haplo-suppressors and triplo-enhancers of PEV; this dosage effect suggests that their products are components of pericentric heterochromatin (LOCKE et al. 1988 ). Indeed, the molecular characterization of the Su(var) mutations has shown that most of these genes specify chromosomal proteins or modifiers of such proteins (WALLRATH 1998 ). One of them, Su(var)205 encodes the nonhistone heterochromatic protein 1 (HP1; EISSENBERG et al. 1990 ). It has been demonstrated that HP1 interacts directly with other Su(var) products, such as Su(var)3-7 and Su(var)3-9, which are also components of heterochromatin (CLEARD et al. 1997 ; DELATTRE et al. 2000 ). These proteins have been detected by immunostaining on larval polytene chromosomes at the chromocenter, at telomeres, and at some euchromatic sites (FANTI et al. 1998 ; DELATTRE et al. 2000 ). Such multimeric complexes of heterochromatin proteins are thought to silence genes by rendering the gene inaccessible to the transcriptional machinery (PARO 1990 ) and/or by sequestering the gene sequences into a subnuclear transcriptionally inactive compartment (WAKIMOTO and HEARN 1990 ). These models are supported by results from chromatin accessibility tests (WALLRATH and ELGIN 1995 ; BOIVIN and DURA 1998 ) and structural studies of the intranuclear organization of chromosomes (CSINK and HENIKOFF 1996 ; DERNBURG et al. 1996 ), respectively, suggesting that both chromatin structure and nuclear location can play a role in PEV.

A distinction must be made between gene silencing induced by centromeric heterochromatin (PEV) and gene silencing induced by telomeric heterochromatin, termed telomeric position effect (TPE), a term already used in yeast and in man for a similar phenomenon (GOTTSCHLING et al. 1990 ; BAUR et al. 2001 ). Previous studies have shown that transgenes containing a white⁺ reporter gene and presenting TPE are always inserted into or next to subtelomeric repeated sequences called telomeric associated sequences (TASs; WALLRATH and ELGIN 1995 ; GOLUBOVSKY et al. 2001 ). These heterochromatic TASs are located next to HeT-A and TART retrotransposons that constitute chromosome ends in Drosophila (MASON and BIESSMANN 1995 ). Despite strong phenotypic similarities between PEV and TPE phenomena in Drosophila, Su(var) and E(var) mutations do not modify TPE (this study; WALLRATH and ELGIN 1995 ; CLEARD et al. 1997 ). Moreover, particular alleles of two Polycomb-group genes (PcG), Posterior sex combs (Psc) and Suppressor of zeste 2 [Su(z)2], have been shown to suppress TPE (CRYDERMAN et al. 1999 ). A more dramatic effect was observed with the Su(z)2⁵ allele, which deletes both the Psc and the Su(z)2 genes. PcG genes and their counteracting trithorax-group (trxG) genes are known to act through different multimeric complexes to maintain the developmental programs in Drosophila by fixing the transcriptional state of the homeotic genes along the antero-posterior axis (BROCK and VAN LOHUIZEN 2001 ). Moreover, Paul Precjewski and Robert Levis have found that mutations in the polyhomeotic (ph) gene, another PcG gene, also suppress TPE (P. PRECJEWSKI and R. LEVIS, personal communication). Taken together, these results suggest that PEV and TPE require two different sets of proteins. The simplest hypothesis would be that Su(var) and E(var) are responsible for PEV, whereas PcG and trxG genes are responsible for TPE.

In this report, we present an extensive genetic analysis of the effect of 50 mutant alleles of PcG and trxG genes on TPE. We show that a small number of these PcG and trxG mutations modify TPE when tested individually. These modifiers encode proteins that have been shown to be components of specific complexes (PRC1 and SWI/SNF). When tested in combination, a larger proportion of PcG mutations behave as suppressors of TPE, revealing synergistic effects between some of these genes in this phenomenon. Furthermore, in situ hybridization and immunolocalization experiments reveal a strict correlation between the presence of TASs and PcG proteins at a telomere, suggesting that TASs may recruit PcG proteins. Finally, we show that transgenes containing a TAS subunit inserted at a euchromatic site present variegated expression or a pairing-sensitive repression of the mini-white reporter gene. These effects depend on the orientation of the TAS subunit. In addition, the insertion site of such a transgene creates a new binding site for PcG proteins. All these data argue strongly in favor of an important role for the PcG genes at the level of the telomeres in Drosophila.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Drosophila stocks:
Drosophila stocks were raised on standard culture medium. Reference(s) and nature (when available) of the alleles of the PcG and trxG alleles used are summarized in Table 1. Detailed information concerning these stocks is available at FlyBase (http://flybase.bio.indiana.edu/). Four telomeric P[white⁺] insertions were used for this study. Three are P[hsp26-tag hsp70-mini-white] constructs in a y w^67c23 background previously described by WALLRATH and ELGIN 1995 : 39C5, 39C27, and 39C31 are located at the 2L, 2R, and 3R tips, respectively. The other transgene (A4-4) is a P[white-rosy] construct located at the telomere of 3R arm (SHEEN and LEVIS 1994 ). All these telomeric insertions present a variegating expression of the mini-white reporter gene and are inserted into TASs (WALTER et al. 1995 ; CRYDERMAN et al. 1999 ). Sequences on both sides of the 39C5 transgene are identical to the 0.4-kb satellite sequences previously characterized in the subtelomeric region of 2L (DDBL/EMBL/GenBank accession no. U35404).

Genetic analyses:
Crosses were performed at 25° unless otherwise specified and 5-day-old flies were used for comparisons of eye color. We checked first that the translocation T(2;3)ap^Xa (abbreviated ap^Xa), has no effect on the eye phenotype of all the telomeric inserts used: flies carrying the ap^Xa chromosomes in a y w^67c23 context were crossed with flies carrying a telomeric transgene, and the eye phenotypes of the progeny carrying either the ap^Xa chromosomes or a complete y w^67c23 chromosomal context were compared: no effect was detected whatever the telomeric insert tested (data not shown).

Assay of the effect of a single PcG or trxG mutation on TPE: Males carrying an autosomal mutant allele (Table 1) were crossed with y w^67c23; CyO/ap^Xa females. G₁ males carrying both the mutation and the ap^Xa chromosomes (y w^67c23; mutant/ap^Xa) were then mated with females carrying a telomeric transgene. In the G₂ progeny, the eye pigmentation of flies hemizygous for the telomeric insert was compared between siblings carrying either the mutant allele or the control ap^Xa chromosomes. Flies were observed under a Leica binocular and photos were acquired with a color video camera (Sony DXC-107AP) and the Adobe Premiere software.

Assay of the effect on TPE of a combination of two PcG mutations: For combination analyses, stocks heterozygous for a PcG mutation (Pc¹, Sce¹, ph⁴¹⁰, and Asx^XF23) and homozygous for a telomeric insert (39C5 and 39C27) were generated. Females from these stocks were crossed with males carrying both another PcG mutation and the ap^Xa chromosomes corresponding to G₁ males described above (males y w^67c23; mutant/ap^Xa except males ph⁴¹⁰). In detail, females harboring Pc¹ were crossed with males harboring ph⁴¹⁰, batman^11', batman¹¹, E(Pc)^D4, E(Pc)^w3, Df(3L)lxd15, E(z)⁵, Pcl¹¹, Pcl¹³, Pcl¹⁵, Sce¹, Asx^XF23, Asx^XT129, Psc¹, Psc^e24, and Psc^Arp1. Females harboring Sce¹ were crossed with males harboring Pcl¹³, Asx^XF23, esc^r4, Pc¹, and sxc¹. Females harboring ph⁴¹⁰ were crossed with males harboring batman^11', batman¹¹, E(Pc)^D4, E(Pc)^w3, Df(3L)lxd15, E(z)⁵, Pcl¹¹, Pcl¹³, Pcl¹⁵, Sce¹, Asx^XF23, Asx^XT129, Psc¹, Psc^e24, Psc^Arp1, Pc¹, Pc³, Pc⁵, Pc⁶, Pc¹⁵, Scm^D1, Scm^ET50, Scm^Su(z)302, Su(z)2^Arp1, esc^r4, and sxc¹. Females harboring Asx^XF23 were crossed with males harboring ph⁴¹⁰, E(Pc)^D4, E(Pc)^w3, E(z)⁵, Pcl¹¹, Pcl¹³, Sce¹, Pc¹, Pc³, Pc¹⁵, Psc¹, Scm^D1, Scm^ET50, Scm^Su(z)302, esc²¹, esc^r4, and sxc¹. In the G₂ progeny, eye pigmentation of heterozygous flies for the telomeric insert (39C5 or 39C27) was compared between siblings carrying either only one mutation (and the ap^Xa chromosome) or the combination of the two mutant alleles. Flies were observed as previously.

Immunostaining on polytene chromosomes:
This protocol, elaborated by Giacomo Cavalli (http://www.igh.cnrs.fr/equip/cavalli/link.labgoodies.html), was adapted from ZINK and PARO 1995 . Polyclonal rabbit anti-PC, monoclonal mouse anti-HP1, polyclonal rabbit anti-PH, polyclonal rabbit anti-PSC, and polyclonal rabbit anti-SCM (BORNEMANN et al. 1998 ) were kindly sent by G. Cavalli, S. C. R. Elgin, F. Maschat, and A. Peterson, respectively. When possible, double immunostainings were performed. The dilutions used were as follows: anti-HP1 1:40, anti-PH 1:40, anti-PC 1:60, anti-SCM 1:40, and anti-PSC were not diluted. Secondary antibodies, Alexa Fluor 488 goat anti-mouse IgG (H + L) conjugate and Alexa Fluor 594 goat anti-rabbit IgG (H + L) conjugate "highly cross adsorb" (Molecular Probes, Eugene, OR), were used at a 1:200 dilution in blocking reagent plus 2% normal goat serum (Sigma, St. Louis). Slides were mounted in Mowiol 0.13 g/ml (Calbiochem, San Diego) and glycerol 30% in Tris-HCl, pH 8.5. Chromosomes were analyzed under a fluorescent microscope (Leica) and pictures were acquired with a camera (Princeton Instruments) at x100 magnification using the Metaview software. Colors were added using the Adobe Photoshop software.

In situ hybridization on polytene chromosomes:
Larvae were allowed to develop at 17° on standard culture medium until they reached the late third instar. Salivary glands were dissected in 45% acetic acid and transferred in 15% lactic acid, 50% acetic acid in water for 3 min. The coverslip was then picked up with a poly-L-lysine-treated slide. After squashing, slides were dehydrated in a series of alcohol baths at -20°. Slides were then washed 30 min in 2x SSC at 65° and dehydrated in alcohol baths at 65°. The presence of TASs in the y w^67c23 strain was checked with a probe derived from TASs originally cloned from the X chromosome (KARPEN and SPRADLING 1992 ). The clone "BS + 9901 R1-8" used corresponds to nucleotides 5041–6910 from the sequence available from GenBank (L03284). This sequence was cloned into EcoRI sites of the Bluescript KS+ vector. Biotinylated probes were made using the nick translation system (GIBCO BRL, Gaithersburg, MD) plus Biotin-16-dUTP (Roche, Indianapolis) and were purified through a G50 column (Pharmacia, Piscataway, NJ) using 1x SSC, 0.1% SDS. Hybridization procedure was as described in the Berkeley Drosophila Genome Project (BDGP; http://www.fruitfly.org/about/methods/cytogenetics.html). After staining with Giemsa 4% in phosphate buffer, pH 6.7, slides were examined under a phase-contrast microscope.

Plasmid construction:
The 1.2-kb PstI-PstI fragment was isolated from the EcoRI-EcoRI fragment of the 1.8-kb TASs originally cloned from the X chromosome (KARPEN and SPRADLING 1992 ). The pCo plasmid contains two reporter genes: mini-white under control of its own promoter and lacZ under control of the otu gene promoter. It was originally built from the pCasPeR-AUG-ßgal plasmid (THUMMEL et al. 1988 ) and is described in RONSSERAY et al. 2001 . The pCo plasmid was digested by PstI to clone the 1.2-kb PstI-PstI fragment of 1.8-kb TASs between the lacZ gene and the 3' extremity of the P construct. In this construct, the TAS is 5.2 kb away from the mini-white gene. The TASs were cloned in both orientations and plasmids were named pCoT+ (TAS centromere-proximal side next to 3'P) and pCoT- (TAS centromere-proximal side next to lacZ; see Fig 4).

View larger version (59K): In this window In a new window Download PPT slide   Figure 1. Genetic modifiers of telomeric position effect. (A) Dominant suppressors and enhancers of TPE. All flies are hemizygous for the telomeric insert (39C5 or 39C27). Photos show males in wild-type PcG and trxG context (+) or carrying a PcG mutant allele [ph410, Pc11, Su(z)25, or ScmSu(z)302] or a trxG mutant allele (brm2, mor1, osa1). (B) Dosage effect of the ph gene. A strong suppressor effect on TPE is observed in both hemizygous ph410 males and homozygous ph410 females, whereas only a slight effect is observed in heterozygous females. The suppressor effect of the ph410 allele in males is abolished by a duplication of the ph locus carried by a free chromosome [Dp(1,f)R].

View larger version (55K): In this window In a new window Download PPT slide   Figure 2. Effect of combinations of PcG mutant alleles on the telomeric transgene 39C5. Each pair of photographs in a given frame shows an eye of a female bearing either a single heterozygous mutation (on the left in the frame) or double heterozygous mutations (on the right in the frame). For each picture, the names of the two mutations tested are indicated on either side and linked with a line. The gray circle indicates the mutant allele shown as a single mutant (on the left in the frame). All mutations used here have no effect on 39C5 when tested individually except Sce1 (weak suppressor effect) and ph410 in males (indicated by a male symbol).

View larger version (34K): In this window In a new window Download PPT slide   Figure 3. Correlation between the presence of TASs and PcG proteins in polytene chromosomes. First lane shows in situ localization of TASs. The probe used for in situ hybridization corresponds to 1.8 kb of X-TAS (see MATERIALS AND METHODS; KARPEN and SPRADLING 1992 ). In the y w67c23 (yw) strain, TASs were detected at the 2R and 3R telomeres, but not at the X, 2L, and 3L telomeres. The 2L telomere of this line contains different TASs that are not homologous to the probe used (CRYDERMAN et al. 1999 ). In the Oregon strain, TASs were detected at the tip of the X chromosome. Second lane shows examples of the detection of PcG products at the telomere by immunostaining experiments. The protein immunodetected is indicated according to its color. Arrows indicate the subtelomeric detection of PcG products. Third lane shows chromosomes stained with 4',6-diamidino-2-phenylindole. The table summarizes the observations: +, positive detection; -, absence of labeling; +/- a positive detection in some cases; nt, not tested.

View larger version (29K): In this window In a new window Download PPT slide   Figure 4. Map of the pCoT- transgene. The schematic diagram is drawn to scale. Sequence of the 1.2-kb TAS is given from distal to proximal as in the original article (KARPEN and SPRADLING 1992 ). Consensus PRE (as defined in MIHALY et al. 1998 ) containing PHO consensus binding sites are given in red, and GAF consensus binding sites are in green in the right strand and in cyan in the opposite strand. The pCoT+ transgene has the same structure except that the TASs are in opposite orientation.

Generation of transgenic flies:
pCoT+ and pCoT- plasmids (0.3 mg/ml) were injected with helper plasmid pUChsP2-3 (0.15 mg/ml) into y w^67c23 embryos following procedures described in BOIVIN and DURA 1998 . Several transgenic lines were recovered with an intact single copy of the pCoT+ (three lines) or the pCoT- (one line) element, as verified by genomic Southern blot and PCR analyses. Two pCoT- additional insertions were recovered after standard mobilization procedures using P{ry + t7.2 = Delta2-3}99B, a stable P-transposase source (ROBERTSON et al. 1988 ).

Localization of the transgenes:
The localization of the inserts was done by in situ hybridization procedures using the TAS probe and by inverse PCR. The protocol for inverse PCR was adapted from WHITELEY et al. 1992 . PCR reactions were performed with three pairs of primers: for the 5' end of the P construct, 5'-AGCGAAAGAGCAACTACGAA-3' and 5'-CGGGACCACCTTATGTTATT-3' were used; for the 3' end of P-CoT- lines 5'-CTCACTCAGACTCAATACGACACT-3' and 5'-GATTGTGCCCTTGCTCTATGGTAA-3' were used; and for the 3' end of P-CoT+ lines 5'-AGACTCAATACGACACTCAGAATA-3' and 5'-TGAAGAAGGGAAATGTAAGAAGAT-3' were used. PCR products were cloned using the TOPO TA cloning kit (Invitrogen, San Diego) and sequenced. Flyblast from BDGP allowed recovery of the location of the cloned sequence, corroborating results obtained by in situ experiments (except for the P-CoT-3 line for which the in situ hybridization was not made). Transgenes were located as follows: 61C8-D2 in the P-CoT-1 line, 57C3 in the P-CoT-2 line, 55B in the P-CoT-3 line, 63D in the P-CoT+1 line, 86E in the P-CoT+2 line, and 7B in the P-CoT+3 line.

Detection of P-lacZ expression in ovaries:
Staining of ovaries of females bearing a transgene to detect lacZ expression was performed as described in LEMAITRE et al. 1993 .

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

A subset of the Polycomb- and trithorax-group genes behave as dominant modifiers of TPE:
The insertion of a transgene containing the white⁺ reporter gene into TASs in Drosophila leads to partial repression (i.e., variegation) of the expression of white in the eye (HAZELRIGG et al. 1984 ; LEVIS et al. 1985 ; KARPEN and SPRADLING 1992 ; TOWER et al. 1993 ; ROSEMAN et al. 1995 ; WALLRATH and ELGIN 1995 ; GOLUBOVSKY et al. 2001 ). This phenomenon, also known in yeast and human cells (GOTTSCHLING et al. 1990 ; BAUR et al. 2001 ), has been generally termed telomeric position effect. It has previously been shown that TPE is not affected by mutations in Su(var) genes or by an additional Y chromosome (WALLRATH and ELGIN 1995 ; CLEARD et al. 1997 ; CRYDERMAN et al. 1999 ). It must be noted, however, that DONALDSON et al. 2002 have recently recovered a mutant allele of Su(var)3-9 (due to a substitution of one nucleotide) that affects TPE. Interestingly, a small number of PcG mutations have been shown to act as dominant suppressors of TPE (CRYDERMAN et al. 1999 ; P. PRECJEWSKI and R. LEVIS, personal communication). Therefore, we analyzed to what extent PcG genes and their counteracting trxG genes are implicated in TPE. For this purpose, the effect on TPE of a number of known mutations of the PcG and trxG genes was tested. These alleles are listed in Table 1. Following the genetic crosses described in MATERIALS AND METHODS, we found that only a small number of alleles act as dominant suppressors [ph⁴¹⁰, Pc¹¹, Su(z)2⁵, and Scm^Su(z)301] or as dominant enhancers (brm², osa¹, and mor¹) of TPE on most of the telomeric inserts tested (see Table 2 and Fig 1A). All others alleles tested (see Table 1) had no or a very weak effect on TPE, in agreement with previous results (CRYDERMAN et al. 1999 ). One exception is the apparent specific effect of two independent alleles of ash1 on the two 3R telomeric insertions tested. Variations observed among telomeric transgenes could be due either to different TAS flanking sequences or to structural difference between transgenes (white reporter gene with A4-4 lines and mini-white reporter gene with 39Cn lines). In most cases, the effect is not dependent on the sex of the flies. An exception is the X-linked ph⁴¹⁰ allele that has a weak effect in heterozygous females, whereas ph⁴¹⁰ homozygous females show a strong derepression of the mini-white expression, as strong as in hemizygous males (Fig 1B). Consistent with these results, the suppressor effect of the ph⁴¹⁰ allele can be counteracted by the addition of a free chromosome carrying the ph region [Dp(1,f)R, Fig 1B]. These results suggest that TPE may depend on the dosage of some of the PcG proteins, especially PH.

From this genetic study, it appears that among PcG mutant alleles, only particular ones are dominant suppressors of TPE. Indeed, the neomorphic Pc¹¹ allele may act as a dominant negative allele and the deletion of two PcG genes [Su(z)2⁵] or the strong hypomorphic allele ph⁴¹⁰ may induce a major loss of PcG products that cannot be compensated by other PcG products. In contrast, the loss of only one dose of a PcG protein may be insufficient to destabilize PcG complexes at telomeres or may be compensated by other PcG products. According to this hypothesis, the simultaneous mutation of two PcG genes may suppress TPE more efficiently than a single mutation. Synergistic or antagonistic effects between different PcG and trxG mutations have been reported many times (JURGENS 1985 ; KENNISON and TAMKUN 1988 ; ADLER et al. 1989 ; CAMPBELL et al. 1995 ; MILNE et al. 1999 ). Therefore, we tested the effect on TPE of double heterozygous combinations of different mutated alleles among the PcG genes that have no effect alone. Several lines containing both a mutant allele (ph⁴¹⁰, Pc¹, Asx^XF23, or Sce¹) and a telomeric insert (39C5 or 39C27) were generated (see MATERIALS AND METHODS). Females from such lines were then allowed to mate with males carrying both another PcG mutation (listed in MATERIALS AND METHODS) and the ap^Xa chromosomes as a control. In the progeny, the eye phenotype of individuals bearing two mutations was compared to their siblings bearing only one mutation and the ap^Xa chromosomes. Under these conditions, many alleles devoid of effect in a single mutant assay on 39C5 [Asx^XF23, Asx^XT129, Pc¹, Pc⁶, E(z)⁵, Psc¹, Pcl¹¹, Pcl¹³, Sce¹, and Scm^ET50] show a synergic suppressor effect in combination with another PcG mutant allele as shown in Fig 2. Out of 61 combinations tested, 14 (23%) result in suppression of TPE. These results suggest that a number of the PcG proteins, acting in a synergistic manner, may be in fact implicated in the TPE phenomenon.

We also tested the effect of mutations in the Su(var)205 gene encoding HP1 (EISSENBERG et al. 1990 ). This nonhistone heterochromatin protein appeared to be a good candidate as a TPE modifier since HP1 has been detected at all telomeres by immunostaining on polytene chromosomes and mutations in the Su(var)205 gene lead to telomere fusion (FANTI et al. 1998 ). Nevertheless, we were unable to detect any significant effect on the phenotype of 39C5 or 39C27, either with the Su(var)2-5⁰⁴ mutant allele or with a duplication of the Su(var)205 gene (DpP15; data not shown). Moreover, no effect was detected with Su(var)2-1⁰¹ (which causes an increase in the amount of acetylated histone H4, DORN et al. 1986 ; data not shown). These mutations were also tested in combination with PcG mutants (ph⁴¹⁰, Sce¹, Pc¹, and Asx^XF23) but no effect was detected (data not shown). These results confirm that HP1 is not involved in TPE despite its telomeric location (see below, WALLRATH and ELGIN 1995 ; CLEARD et al. 1997 ; CRYDERMAN et al. 1999 ).

PcG proteins are located at telomeres that contain telomeric associated sequences:
We next asked whether PcG proteins act directly on TPE, i.e., by binding telomeric sequences. The first approach was to assay for the presence of PcG products at telomeres harboring TAS. In the 39C5 line, generated from the y w^67c23 background (see MATERIALS AND METHODS), TASs are present at the 2L telomere and flank the telomeric P[w⁺] insertion: 4 kb of TAS on one side and 5.3 kb on the other side (CRYDERMAN et al. 1999 ). In situ hybridization experiments were performed on two strains, y w^67c23 and Oregon-R, using a TAS probe made from a 1.8-kb fragment of an X-chromosome TAS (KARPEN and SPRADLING 1992 ) that is homologous to the TAS present at the 2R and 3R telomere but not to the TAS at the tip of the 2L (MASON and BIESSMANN 1995 ). In the y w^67c23 strain, TASs homologous to the probe are present at the 2R and the 3R telomeres but are not detected at the tips of the X, 2L, and 3L chromosomes (Fig 3, first lane). In the Oregon-R strain, this type of TAS was detected at the tip of the X chromosome (Fig 3, first lane). These results confirm that variability in the presence of TASs exists between strains (WALTER et al. 1995 ; F. SHEEN, L. TOLAR and R. LEVIS, personal communication).

In addition, polytene chromosomes of salivary glands of third instar y w^67c23 larvae were squashed and stained with antibodies raised against PcG proteins (PC, PH, PSC, and SCM) and against HP1. PcG proteins were detected at the telomeres of the 2L, 2R, and 3R chromosome arms in the y w^67c23 strain, which contain TASs, but never at the tips of the X or 3L, in which no TASs were detected (Fig 3, second lane). In the Oregon-R strain, PH protein is detected at the tip of the X chromosome, which contains TASs (Fig 3, second lane). These results strongly suggest that PcG proteins bind TASs. In contrast, HP1 was detected at the ends of all chromosomes irrespective of the presence of TASs (Fig 3, second lane), supporting the idea that HP1 acts as a telomere-capping protein whatever the nature of telomeric sequences. Indeed, HP1 staining was also found at the tips of chromosomes that have lost TASs due to a terminal deficiency (FANTI et al. 1998 ). Interestingly, double labeling for HP1 and PcG proteins shows nonoverlapping localization of these two types of proteins at chromosome ends, HP1 being present at the true extremity of chromosomes and PcG proteins proximal with respect to HP1 (see Fig 3, Fig 2L and Fig 2R tips). This result is consistent with the idea that HP1 caps the extremities of chromosomes, whereas PcG proteins are associated with TASs that are located proximal to HeT-A/TART arrays, that is, several tens of kilobases away from chromosome ends.

X-chromosome TASs (1.2 kb) induce pairing-sensitive repression of reporter genes in an orientation-dependent manner:
It was previously reported that, in a transgene, TASs from the tip of 2L are able to silence a mini-white reporter gene contained within the transgene in an orientation-dependent manner (KURENOVA et al. 1998 ). To test whether different TAS are able to silence genes in the vicinity, P-element transgenes containing both mini-white and lacZ reporter genes and 1.2 kb of an X-chromosome TAS in both orientations were constructed (see MATERIALS AND METHODS and Fig 4). Independent transgenic lines were obtained after injection and remobilization of these constructs named pCoT+ (TAS centromere-proximal side next to 3'P) and pCoT- (TAS centromere-proximal side next to lacZ). Control lines bearing the pCo transgene (same structure without a TAS; RONSSERAY et al. 2001 ) show homogenous pigmentation under either hemizygous or homozygous conditions and no pairing sensitivity, meaning that expression of the mini-white gene is higher in a homozygous state than in a hemizygous one (Fig 5A and Fig B). By contrast, flies bearing pCoT- transgenes show a light orange eye in a hemizygous state (Fig 5D, Fig G, and Fig J) and nonuniform pigmentation (Fig 5E) or strong pairing-sensitive repression (lower expression of mini-white in homozygotes than in hemizygotes; Fig 5H and Fig K). Flies bearing pCoT+ transgenes show a yellow eye in a hemizygous state (Fig 5, M, P, and S) and almost uniform orange pigmentation in a homozygous state (Fig 5N, Fig Q, and Fig T). These results suggest first that 1.2 kb of an X-chromosome TAS represses the expression of a reporter gene independently of the orientation of the TAS (compare hemizygous P-CoT- or P-CoT+ lines with P-Co1) and, second, that this repression is pairing sensitive in an orientation-dependent manner (compare homozygous P-CoT- lines with homozygous P-CoT+ lines).

View larger version (78K): In this window In a new window Download PPT slide   Figure 5. Eye phenotype and ovarian lacZ expression of transgenic flies bearing a pCo, pCoT-, or pCoT+ transgene. Males carrying the control pCo transgene without TASs show homogenous pigmentation in the eye, in either a hemizygous (A) or a homozygous (B) state. Strong expression of the lacZ gene is detected in the germline of homozygous females (C). Males hemizygous for a pCoT- transgene exhibit a light orange eye color with no variegation (D, G, and J), while homozygous males present either a strong pairing-repressive effect (H and K) or nonuniform pigmentation (E). Males hemizygous for a pCoT+ transgene exhibit a yellow eye color with no variegation (M and P), while homozygous males present an almost uniform darker orange eye color (N and Q). The same observations were made in females: one example is given for the P-CoT+3 insert on the X chromosome (S and T). Expression of lacZ in the germline is repressed in the P-CoT- lines (F, I, and L) and in the P-CoT+ lines (O, R, and U).

In the ovaries of flies homozygous for pCo control transgenes, the expression of lacZ is strong (example P-Co1 in Fig 5C) as expected due to the presence of the otu gene promoter. By contrast, ovarian lacZ expression in homozygous P-CoT- females (Fig 5F, Fig I, and Fig L) and in homozygous P-CoT+ females (Fig 5O, Fig R, and Fig U) varies from severely reduced to partially repressed. Taken together, these results show that 1.2 kb of X-chromosome TAS can induce the repression of the expression of reporter genes both in the soma and in the germline.

X-chromosome TAS (1.2 kb) recruits PcG proteins in vivo:
Fig 4 shows the position of putative GAF and polycomb-response element (PRE) consensus sequences, the latter containing a PHO-binding site as defined by MIHALY et al. 1998 , present within the 1.2 kb of a TAS subunit originally cloned from an X chromosome by KARPEN and SPRADLING 1992 . In the P-CoT+ and P-CoT- lines, transgenes were localized by in situ hybridization (except for P-CoT-3) and inverse PCR (see MATERIALS AND METHODS). We further tested whether the 1.2-kb X-TAS is able to recruit PcG proteins de novo by looking for the presence of an additional binding site for PcG products on polytene chromosomes of transgenic larvae. In the P-CoT+3 line, the transgene was shown to be inserted in a strong endogenous binding site for PcG proteins (6B). In the P-CoT-1 line, the transgene was found to be at 61C8-D2. This region has been previously described as an endogenous site of fixation for some PcG products (PH, PC, and PCL), but not for PSC and SCM (as shown in Fig 6). For all the other transgenes, the insertion site was not an endogenous binding site for the PcG proteins tested so far. Fig 6 shows that for the four lines tested, the transgene creates a new binding site for PcG proteins (SCM and PSC or PH) but not for HP1, as detected by immunostaining. These results show that the 1.2-kb X-TAS is able to recruit PcG proteins in vivo. Taken together, our results show that the 1.2-kb X-TAS mimics many properties of a PRE: this sequence is able to repress the expression of reporter genes, it can induce pairing-sensitive repression, and it can recruit PcG proteins de novo on polytene chromosomes.

View larger version (70K): In this window In a new window Download PPT slide   Figure 6. Additional PcG labels at the insertion site of pCoT- and pCoT+ transgenes. In the P-CoT-1, P-CoT-2, P-CoT-3, and P-CoT+1 lines, transgenes are located in 61CD, 57C3, 55B, and 64C, respectively. These sites are not endogenous binding sites for PcG proteins as shown in yw67c23 (yw) control. By contrast, in the transgenic lines, a de novo binding site for PcG proteins can be detected by immunostaining at the insertion site of the transgene. The protein immunodetected is designated according to its color. The probe used for in situ hybridizations was 1.8 kb of X-TAS.

The derepressed state is not transmitted through meiosis:
An intriguing property of a PRE from the Ultrabithorax gene (Fab7) has been described by CAVALLI and PARO 1998 , CAVALLI and PARO 1999 . They constructed a UAS-PRE-lacZ-mini-white transgene in which the repression of the mini-white reporter gene induced by the Fab7 PRE could be abolished by an embryonic pulse of GAL4, thus leading to flies with red eyes. When these red-eyed flies were crossed with each other, 25% of their progeny retained a red eye pigmentation without any further pulse of GAL4, showing that the active state of the PRE can be transmitted through meiosis. We have tested therefore if the derepressed state of a telomeric transgene induced in a mutant context for a PcG gene could also be transmitted through meiosis. Previously, we observed that a small percentage of flies with dark red pigmentation appear spontaneously from a variegating stock (39C5) containing a P[w+] insertion at a telomere. Moreover, when a hemizygous female 39C5/+ is crossed with a y w^67c23 male, 2.5% of the female progeny (n = 506) and 0.6% of the male progeny (n = 452) show a spontaneous red-eye phenotype. In the reciprocal cross, 0.9% of the female progeny (n = 462) and 0.2% of the male progeny (n = 520) show a spontaneous red-eye phenotype. This derepressed state is reversible since in the progeny of such red-eyed flies the relative proportions of variegating and red-eyed flies are conserved (data not shown). The activation of the white reporter gene thus appears to be due to a lack of repression rather than a dramatic change in the structure of the telomere, i.e., a terminal deletion.

We thus tested whether the presence of a PcG mutation associated with derepression of a telomeric P[w+] transgene in parents increases the percentage of red-eyed flies in the progeny lacking the mutation. Females homozygous for ph⁴¹⁰ and hemizygous for the 39C5 insert were crossed with y w^67c23 males. In the progeny, females are heterozygous for the ph⁴¹⁰ mutation, a condition that allows a weak suppression of TPE (see Fig 1B). Among these females, 2.5% present a red eye color (n = 121). This proportion is identical to the spontaneous occurrence of red-eyed flies in the control (2.5%, n = 506, see above). Reciprocally, in the progeny of males carrying both the ph⁴¹⁰ allele and the 39C5 crossed with y w^67c23 females, similar results were obtained: 1.6% (n = 184) of the males that do not carry the ph⁴¹⁰ allele and 2.2% (n = 224) of the females (heterozygous for the ph⁴¹⁰ mutation) exhibit red eyes. These results show that the strong derepressed state observed in a ph⁴¹⁰ mutant is not transmitted through the next generation in the absence of the mutation.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Which modifiers contribute to telomeric position effect?
Among the 50 mutant alleles of PcG and trxG genes tested, <10 behave as dominant modifiers of TPE. By contrast, combination analyses reveal that 10 alleles that have no effect alone have synergistic effects on TPE. Although we cannot formally exclude the possibility that another mutation linked to each mutant PcG or trxG allele is responsible for the effect observed, for the following reasons we think that these effects are very unlikely due to differences in genetic backgrounds or linked mutations. First, the specificity of the effect of the ph⁴¹⁰ allele was demonstrated by the restoration of TPE with a duplication of the ph locus carried by a free chromosome [Dp(1,f)R] (Fig 1B). Second, for the autosomal mutations tested, the genome was partially homogenized using the y w^67c23 genetic background and the ap^Xa chromosomes, which have no effect of their own on TPE. This would restrict the location of a "linked mutation" to the autosome that carries the tested PcG or trxG mutation. Third, it is hard to explain how two linked mutations that have no effect alone would have an effect in combination (most of the combination effects were observed with two PcG mutants that have no effect alone). Fourth, different alleles of a same gene coming from different laboratories and generated in different genomic backgrounds show similar effects on TPE (alone or in combination). Fifth, the suppressor effects are seen with PcG mutations, while the enhancer effects are observed with the trxG mutated alleles. In contrast, background effects, if they exist, should distribute randomly between the two types of mutations. Sixth, it makes sense, a posteriori, that strong suppressors encode members of the PRC1 complex, whereas enhancers represent members of the counteracting SWI/SNF complex (SHAO et al. 1999 ; FRANCIS et al. 2001 ).

However, none of the single dominant modifiers identified represent amorphic alleles. Even in combination tests, most of the alleles that were shown to have an effect are not classified as amorphic [except Pc¹, Asx^XF23, Pcl¹¹, and E(z)⁵]. This observation suggests that TPE is only slightly sensitive to the dosage effect of the PcG proteins, contrary to what could be expected from what is known about PcG-mediated nontelomeric repression. Therefore, the suppressor effect is likely due to a dominant negative action of "poison" proteins produced by antimorphic or neomorphic alleles. In addition, among all the Pc alleles tested, none of the alleles described as amorphic (Pc¹, Pc³, and Pc¹⁵) present an effect alone, and surprisingly, only one (Pc¹) presents an effect in combination with several other PcG mutants. This apparent discrepancy can be resolved if Pc¹ is not in fact a null allele. In favor of this hypothesis, Pc¹ results from a deletion and a frameshift that produces a truncated mutant 55-kD protein (FRANKE et al. 1995 ). Taken together, these results show that the effect on TPE of PcG proteins can be detected by poisoning the system rather than reducing the dose by half of one or two proteins. This may suggest that PcG proteins compensate for each other particularly efficiently at telomeres.

When an autosomal mutant allele was tested alone, it was always paternally inherited. By contrast, in combination experiments, each parent contributes one mutant allele. We examined whether the combinatory effect could be due to a maternal effect for some of the alleles tested by performing, when possible, reciprocal crosses (for example, for Pc¹ and Asx^XF23, for which stocks exist that also carry the 39C5 telomeric insert). No difference was observed whatever the parental origin of the mutations (data not shown). This could be explained if both genes have in fact an effect alone but only when maternally transmitted. This is not the case since for some maternally transmitted alleles an effect was observed in some combinations and not in others. It appears therefore that the effects on the eye phenotype do not depend on a maternal inheritance. This result is consistent with the homeotic adult phenotypes scored in double PcG heterozygotes for which very few differences were observed between reciprocal crosses (CAMPBELL et al. 1995 ).

Interestingly, the subgroup of dominant suppressors that act alone on TPE (Pc, ph, Psc, and Scm) are members of the PRC1 complex that has been purified from embryonic nuclear extracts (SHAO et al. 1999 ). Some other PcG mutations, such as Asx, E(z), Pcl, or Sce, act as suppressors in combination, suggesting that the products of these genes participate with a specific telomeric PcG complex. Strikingly, this subgroup of eight PcG genes was already highlighted in a genetic interaction study showing that Pc, Scm, Psc, Pcl, Sce, and Asx are lethal when heterozygous with ph², a temperature-sensitive mutation, all combinations leading to similar phenotypes in the dying embryos (MCKEON and BROCK 1991 ).

It has been shown that telomeric inserts are less accessible than euchromatic inserts to restriction enzymes and to DAM methylase (CRYDERMAN et al. 1999 ; A. BOIVIN and S. RONSSERAY, unpublished data). In addition, we observed that the accessibility of telomeric inserts to DAM methylase increases in a ph⁴¹⁰ background and that this is correlated to derepression of the white gene (A. BOIVIN, C. GALLY and S. RONSSERAY, unpublished results). This result is similar to that obtained with the ph PRE-mini-white transgenes (BOIVIN and DURA 1998 ) suggesting that PcG products adopt a similar chromatin-based mechanism to repress their euchromatic and telomeric targets.

What kind of PRE is present within telomeric associated sequences?
PREs were initially identified by their ability to prevent ectopic activation of a Hox reporter gene construct. This capacity depends on the dose of the PcG proteins (MULLER and BIENZ 1991 ; SIMON et al. 1993 ; CHAN et al. 1994 ). Placed in a transgene, PREs can also induce mosaic expression of the flanking reporter gene (FAUVARQUE and DURA 1993 ; KASSIS 1994 ), a phenotype resembling that of PEV and TPE. Moreover, PRE-mediated repression often exhibits pairing sensitivity, defined as the lower expression of the flanking reporter gene in a homozygous state than in a heterozygous one (PIRROTTA 1997 ). In this article, we show that a 1.2-kb fragment of the 1.8-kb X-chromosome TAS induces variegation or pairing-sensitive repression in an orientation-dependent manner and creates new binding sites for the PcG proteins as detected by immunostaining on polytene chromosomes. These results demonstrate that this TAS fragment mimics some properties of a PRE and thus reinforce the parallels that can be made between telomeric silencing and PcG-mediated euchromatic repression. The ability of TASs from the left tip of chromosome 2 (2L-TAS) to retain aspects of telomeric silencing in ectopic positions was shown by KURENOVA et al. 1998 . At this telomere, TASs are composed of repeats of 457 bp that present only limited homology with TASs present at the X, 2R, and 3R telomeres (KARPEN and SPRADLING 1992 ; LEVIS 1993 ; WALTER et al. 1995 ). Analysis of the sequence of one repeat (457 bp) revealed nine GAF-binding sites but no PHO-binding site. KURENOVA et al. 1998 established several transgenic lines carrying different constructs made up of 6 kb of 2L-TAS (13 repeats) adjacent to the mini-white reporter gene and flanked by Su(Hw) insulator sequences. Depending on the orientation of the TASs inside the transgene, some lines present reduced expression of the mini-white gene when compared to lines carrying a similar transgene without TASs or with TASs in the opposite orientation. Such orientation-dependent silencing has been previously described for the Fab7 PRE of the Ubx gene (ZINK and PARO 1995 ), but does not appear to be a general property of PREs since another PRE from Ubx (Mcp) has been shown to function in both orientations (BUSTURIA et al. 1997 ). From our study, the more efficient orientation for the 1.2-kb X-TAS-induced repression appears to be the same as that described for the 2L-TASs (KURENOVA et al. 1998 ): repression appears to be stronger from the centromere-proximal side.

It was reported that repression induced by the 2L-TAS when inserted within a transgene was weakly sensitive to Su(z)2⁵ (KURENOVA et al. 1998 ). Surprisingly, we cannot detect any effect of PcG mutations on the repression induced by the 1.2-kb X-TASs, except a slight suppressor effect of Su(z)2⁵ on P-CoT-1 in a homozygous state (data not shown). At the moment, we have no explanation for why the repression induced by the 1.2-kb X-TASs in a euchromatic environment is not sensitive to modification of the dose of PcG proteins that could otherwise affect TPE.

Increasing the distance between the 2L-TAS and the mini-white gene with 2.4 kb of unrelated DNA in another transgene from KURENOVA et al. 1998 did not change the silencing capacity of 2L-TAS. In our study, the 1.2 kb of X-TAS is located >5 kb away from the mini-white gene, thus showing the silencing capability of TASs over an extended distance. Similar results were obtained with transgenes containing the Fab7 PRE (ZINK and PARO 1995 ). According to chromatin-immunoprecipitation experiments, PcG products can spread as far as 10–15 kb from PREs (STRUTT et al. 1997 ) and we might expect that repression could occur over such a distance.

In fact, what we observed with the 1.2 kb of X-TAS in the pCoT- transgenes resembles what has been observed with PREs from the Bithorax complex. Using Fab7-mini-white transgenes, ZINK and PARO 1995 showed that some insertion sites present pairing sensitivity (as observed with P-CoT-2 and P-CoT-3), while others present variegation with darker spots (as observed with P-CoT-1). The Fab7 PRE has been shown to convey a derepressed state through meiosis after being activated in the embryonic stage by the UAS/GAL4 system (CAVALLI and PARO 1998 , CAVALLI and PARO 1999 ). In the case of TPE, the derepressed state observed in a PcG mutant background is not transmitted to the next generation. A fundamental difference between these studies is that the suppressor effect we observed in the case of TPE is due to the lack of one PcG partner. In the case of Cavalli and Paro's experiment, it is hyperactivation (forced activation) induced by GAL4 via the UAS sequences that abolishes the repressor capacity of the Fab7 PRE. This activation likely involves fundamental changes in chromatin conformation and/or epigenetic marks (such as hyperacetylation) that may be different from the effect of a decrease in the dosage of a repressor. To compare TPE and the Fab7 PRE it would thus be interesting to test transmission through meiosis of the derepressed state of the UAS-Fab7 transgene induced by a PcG mutation rather than upon activation by GAL4. Different PREs thus share properties but also present particularities that likely depend on their sequence. Indeed, the dissection of another PRE from the Bithorax complex, Mcp, revealed that repression in cis and pairing-sensitive repression could be separated (MULLER et al. 1999 ). This shows that PREs may combine several regulatory properties and future dissection of the different TASs will tell us which functions telomeric PREs combine.

	FOOTNOTES

¹ Present address: Laboratoire Sénescence et longévité chez le champignon Podospora anserina, Centre de Génétique Moléculaire CNRS UPR 2167, Bât. 24 Avenue de la Terrasse, 91198 Gif-sur-Yvette Cedex, France.
² Present address: Laboratoire de Biologie Cellulaire de la Synapse Normale et Pathologique, Ecole Normale Supérieure—Inserm U497, 46, rue d'Ulm, 75005 Paris, France.
³ Present address: Laboratoire de Génétique et Biologie Cellulaire Bât. Buffon, Université Versailles Saint-Quentin 45, Ave. des Etats-Unis, 78035 Versailles Cedex, France.

	ACKNOWLEDGMENTS

This work was carried out in compliance with the current laws governing genetic experimentation in France and the USA. We thank F. Sheen, L. Tolar, P. Precjewski, and R. Levis for personal communications. We are particularly indebted to Robert Levis, Lori Wallrath, Giacomo Cavalli, Sarah Elgin, Florence Maschat, and Aidan Peterson for sending flies and antibodies. We thank N. Audugé and A. Piton for their participation to this work as rotator students. We thank A. M. Pret for her valuable help in the preparation of the manuscript. This work was supported by the Centre National de la Recherche Scientifique (UMR 7592), by a grant from the Association pour la Recherche sur le Cancer, and by the Universités Paris 6- Pierre et Marie Curie and Paris 7- Denis Diderot.

Manuscript received September 23, 2002; Accepted for publication January 24, 2003.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ADLER, P. N., J. CHARLTON, and B. BRUNK, 1989  Genetic interactions of the suppressor 2 of zeste region genes. Dev. Genet. 10:249-260.[Medline]

BAUR, J. A., Y. ZOU, J. W. SHAY, and W. E. WRIGHT, 2001  Telomere position effect in human cells. Science 292:2075-2077.[Abstract/Free Full Text]

BOIVIN, A. and J. M. DURA, 1998  In vivo chromatin accessibility correlates with gene silencing in Drosophila.. Genetics 150:1539-1549.[Abstract/Free Full Text]

BORNEMANN, D., E. MILLER, and J. SIMON, 1996  The Drosophila Polycomb group gene Sex comb on midleg (Scm) encodes a zinc finger protein with similarity to polyhomeotic protein. Development 122:1621-1630.[Abstract]

BORNEMANN, D., E. MILLER, and J. SIMON, 1998  Expression and properties of wild-type and mutant forms of the Drosophila sex comb on midleg (SCM) repressor protein. Genetics 150:675-686.[Abstract/Free Full Text]

BREEN, T. R. and I. M. DUNCAN, 1986  Maternal expression of genes that regulate the bithorax complex of Drosophila melanogaster.. Dev. Biol. 118:442-456.[Medline]

BROCK, H. W. and M. VAN LOHUIZEN, 2001  The Polycomb group–no longer an exclusive club? Curr. Opin. Genet. Dev. 11:175-181.[Medline]

BUSTURIA, A., C. D. WIGHTMAN, and S. SAKONJU, 1997  A silencer is required for maintenance of transcriptional repression throughout Drosophila development. Development 124:4343-4350.[Abstract]

CAMPBELL, R. B., D. A. SINCLAIR, M. COULING, and H. W. BROCK, 1995  Genetic interactions and dosage effects of Polycomb group genes of Drosophila. Mol. Gen. Genet. 246:291-300.[Medline]

CAVALLI, G. and R. PARO, 1998  The Drosophila Fab-7 chromosomal element conveys epigenetic inheritance during mitosis and meiosis. Cell 93:505-518.[Medline]

CAVALLI, G. and R. PARO, 1999  Epigenetic inheritance of active chromatin after removal of the main transactivator. Science 286:955-958.[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]

CHENG, N. N., D. A. SINCLAIR, R. B. CAMPBELL, and H. W. BROCK, 1994  Interactions of polyhomeotic with Polycomb group genes of Drosophila melanogaster. Genetics 138:1151-1162.[Abstract]

CLÉARD, F., M. DELATTRE, and P. SPIERER, 1997  SU(VAR)3-7, a Drosophila heterochromatin-associated protein and companion of HP1 in the genomic silencing of position-effect variegation. EMBO J. 16:5280-5288.[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]

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

DELATTRE, M., A. SPIERER, C. H. TONKA, and P. SPIERER, 2000  The genomic silencing of position-effect variegation in Drosophila melanogaster: interaction between the heterochromatin-associated proteins Su(var)3-7 and HP1. J. Cell Sci. 113:4253-4261.[Abstract]

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]

DOCQUIER, F., O. SAGET, F. FORQUIGNON, N. B. RANDSHOLT, and P. SANTAMARIA, 1996  The multi sex combs gene of Drosophila melanogaster is required for proliferation of the germline. Roux's Arch. Dev. Biol. 205:203-214.

DONALDSON, K. M., A. LUI, and G. H. KARPEN, 2002  Modifiers of terminal deficiency-associated position effect variegation in Drosophila. Genetics 160:995-1009.[Abstract/Free Full Text]

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

DURA, J. M., N. B. RANDSHOLT, J. DEATRICK, I. ERK, and P. SANTAMARIA et al., 1987  A complex genetic locus, polyhomeotic, is required for segmental specification and epidermal development in D. melanogaster. Cell 51:829-839.[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]

FANTI, L., G. GIOVINAZZO, M. BERLOCO, and S. PIMPINELLI, 1998  The heterochromatin protein 1 prevents telomere fusions in Drosophila. Mol. Cell 2:527-538.[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]

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]

FAUVARQUE, M. O., V. ZUBER, and J. M. DURA, 1995  Regulation of polyhomeotic transcription may involve local changes in chromatin activity in Drosophila. Mech. Dev. 52:343-355.[Medline]

FRANCIS, N. J., A. J. SAURIN, Z. SHAO, and R. E. KINGSTON, 2001  Reconstitution of a functional core Polycomb repressive complex. Mol. Cell 8:545-556.[Medline]

FRANKE, A., S. MESSMER, and R. PARO, 1995  Mapping functional domains of the polycomb protein of Drosophila melanogaster. Chromosome Res. 3:351-360.[Medline]

GINDHART, J. G., JR. and T. C. KAUFMAN, 1995  Identification of Polycomb and trithorax group responsive elements in the regulatory region of the Drosophila homeotic gene Sex combs reduced. Genetics 139:797-814.[Abstract]

GOLUBOVSKY, M. D., A. Y. KONEV, M. F. WALTER, H. BIESSMANN, and J. M. MASON, 2001  Terminal retrotransposons activate a subtelomeric white transgene at the 2L telomere in Drosophila. Genetics 158:1111-1123.[Abstract/Free Full Text]

GOTTSCHLING, D. E., O. M. APARICIO, B. L. BILLINGTON, and V. A. ZAKIAN, 1990  Position effect at S. cerevisiae telomeres: reversible repression of Pol II transcription. Cell 63:751-762.[Medline]

HAGSTROM, K., M. MULLER, and P. SCHEDL, 1997  A Polycomb and GAGA dependent silencer adjoins the Fab-7 boundary in the Drosophila bithorax complex. Genetics 146:1365-1380.[Abstract]

HAZELRIGG, T., R. LEVIS, and G. M. RUBIN, 1984  Transformation of white locus DNA in Drosophila: dosage compensation, zeste interaction, and position effects. Cell 36:469-481.