The P-Ph Protein-Mediated Repression of yellow Expression Depends on Different cis- and trans-Factors in Drosophila melanogaster

Corresponding author: Pavel Georgiev, Institute of Gene Biology, Russian Academy of Sciences, 34/5 Vavilov St. Moscow 117334, Russia., georg{at}biogen.msk.su (E-mail)

Communicating editor: J. A. BIRCHLER

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The ph^P1 allele of Drosophila melanogaster encodes a chimeric P-Ph protein that contains the DNA-binding domain of the P-element transposase and the Ph protein lacking 12 amino-terminal amino acids. It has been shown that the P-Ph protein is responsible for the formation of a repressive complex on P elements inserted at the yellow locus. Here we demonstrate that an enhancer element can suppress the P-Ph-mediated inhibition of yellow transcription. However, an increase of P-element copy number at the yellow locus overcomes the enhancer effect. The mobilization of P-element transposition induced the appearance with a high frequency of Su(y) mutations that partially or completely suppressed the inhibitory effect of ph^P1 on yellow expression. The Su(y) mutations were localized at different sites on chromosomes. One strong Su(y) mutation, sn^eP1, was found to be induced by a 1.2-kb P-element insertion into the transcribed noncoding region of the singed locus. The Su(y) mutations resulted in a high level of transcription of the 1.2-kb P element that contained the sequences encoding one DNA-binding and two protein-protein interaction domains of the transposase. The effect of Su(y) mutations can be explained by the competition between the truncated transposase encoded by a 1.2-kb P element and the P-Ph protein for binding sites on P-element insertions.

THE polyhomeotic (ph) gene is a member of the group consisting of at least 20 genes with similar functions that are required for normal segmental specification in Drosophila and referred to as the Polycomb group (Pc-G) genes(JURGENS 1985 ). The Pc-G genes act in normal development as repressors of BX-C and ANT-C genes (STRUHL 1981 ; DURA et al. 1985 ; DURA and INGHAM 1988 ). Several Pc-G members were found to be associated in a multiprotein complex (FRANKE et al. 1992 ; RASTELLI et al. 1993 ; MULLER 1995 ; PLATERO et al. 1996 ; STRUTT and PARO 1997 ). Cis-regulatory elements necessary for maintaining the repressed state of homeotic genes have been identified (MULLER and BIENZ 1991 ; SIMON et al. 1993 ; CHAN et al. 1994 ; CHIANG et al. 1995 ) and designated as Pc-G response elements (PREs; SIMON et al. 1993 ).

We have discovered a new P-element-induced mutation in the polyhomeotic (ph) gene, ph^P1, which represses yellow expression in the yellow alleles containing P-element insertions (BELENKAYA et al. 1998 ). The ph^P1 mutation is induced by a 1.2-kb P-element insertion into the 5' transcribed noncoding region of the ph gene. The ph^P1 allele produces a chimeric protein, designated as P-Ph, that contains the DNA-binding domain of the P element and the Ph protein lacking 12 amino-terminal amino acids. The P-Ph chimeric protein can bind to the P-element sequences and recruit other Pc-G proteins, leading to the formation of a repressive complex.

Previously we described a family of highly unstable y mutations induced by chimeric mobile elements located 69 bp upstream to the yellow transcription start site (GEORGIEV et al. 1997 ). They appeared against the background of the y² allele caused by an insertion of gypsy at -700 bp position (GEYER et al. 1986 ; PARKHURST and CORCES 1986 ). The region of gypsy responsible for its mutagenic effect, su(Hw)-binding region, has the properties of an insulator: only yellow enhancers responsible for the body and wing pigmentation located distally from the promoter are affected, whereas all other tissues show the wild-type coloration (GEYER and CORCES 1987 , GEYER and CORCES 1992 ). Insertions inducing highly unstable y mutations consist of two identical copies of a defective P element, 1.2 kb long, and some genomic sequences located between them. One such insertion, present in the y^+s1 allele, contains a 2.9-kb genomic sequence that includes an enhancer-like element activating yellow expression in the body cuticle and wing blade (GEORGIEV et al. 1997 ). A large number of derivative alleles with deletions in the internal part of the chimeric element or with P-element rearrangements can be obtained from the y^+s1 allele.

In this article, we have characterized the interactions of the ph^P1 mutation with the y^+s1 allele and its derivatives. The inhibiting effect of ph^P1 on yellow expression can be suppressed either by the enhancer element located in the chimeric y^+s1 insertion or by the yellow wing and body enhancers. However, an increase of P-element copy number either in cis or in trans strongly elevates the ph^P1-mediated repression that overcomes the effect of the enhancers. We have also studied the effect of ph^P1 on P-element-induced mutations in the singed locus, which is expressed at the same middle pupal stage as yellow during the formation of adult bristles and hairs, and in the female germline during oogenesis (PATERSON and OAHARE 1991 ). Unexpectedly, the ph^P1 mutation does not influence the phenotype of sn mutations induced even by double P-element insertions.

During this work, we discovered a number of P-element-induced mutations designated as Su(y) that partially or completely suppress the inhibitory effect of ph^P1 on the y^+s1 allele and its derivatives. The effect of Su(y) mutations results in a high level of P-element expression. The truncated transposase competes with the chimeric P-Ph protein for the same binding sites on P elements.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Drosophila strains:
All flies were maintained at 25° in a standard yeast medium. Genetic symbols of the y alleles and the origin of the y^+s1 allele and some of its derivatives were described in other articles (GEORGIEV et al. 1992 , GEORGIEV et al. 1997 ). The w;Sb P[ry⁺2-3] (99B) e/TM6,e stock, providing a stable source of transposase (ROBERTSON et al. 1988 ), was obtained from the Bloomington Stock Center (Bloomington, IN). Hereafter P[ry⁺2-3](99B) is referred to as 2-3. The rest of the Drosophila melanogaster strains and mutations are described in LINDSLEY and ZIMM 1992 . The origin and structure of the ph^P1 strain are described in another article (BELENKAYA et al. 1998 ).

Genetic crosses:
To induce mutagenesis, y^+s1 ph^P1 females were crossed to w; Sb 2-3 e/TM6,e males to produce dysgenic males of the y^+s1ph^P1/Y; Sb 2-3 e/TM6,e/+ genotype. In each vial, 2-3 males were mated to 10–12 C(1)RM,yf females. The F₂ progeny was analyzed for mutagenesis. All males with a new y phenotype were individually mated to virgin C(1)RM,yf females and their phenotype was examined in the next generation. Determination of the yellow phenotype was performed as previously described (GEORGIEV et al. 1997 ).

Combinations of ph^P1 with different y alleles were obtained according to the following scheme:

• F₀: y¹sc^D1ph^P1w^a/y¹sc^D1ph^P1w^a x y*/Y, where y* is any y allele used;

• F₁: y¹sc^D1ph^P1w^a/y* x y¹sc^D1ph^P1w^a/Y;

• F₂: Selection of y*ph^P1w^a males (sc⁺ phenotype and orange eyes).

The introduction of ph^P1 was confirmed by Southern blot hybridization with labeled probes from the ph gene. The w^a (1-1.5) mutation was used as the closest marker for the ph gene (1-0.5).

The removal of ph^P1 from a X chromosome with a y* allele was performed according to the following scheme:

• F₀: y¹sc^D1w^a/y¹sc^D1w^a x y*ph^P1/Y;

• F₁: y¹sc^D1w^a/y*ph^P1 x y¹sc^D1w^a/Y;

• F₂: Selection of y*w^a males (sc⁺ phenotype and orange eyes).

The removal of ph^P1 was confirmed by Southern blot hybridization.

Combinations of autosome Su(y) mutations with y*ph^P1 were obtained and analyzed according to the following scheme: F₀: y*ph^P1/FM4 x y^+s1ph^P1; Su(y)/CyO or Su(y)/TM3,Sb; analysis of the y phenotype in y*ph^P1/Y; Su(y)/+ males.

To study the eye phenotype on the ph^P1 background we constructed the yph^P1w line. The su(Hw)²/su(Hw)^v mutations in the su(Hw) gene were combined with the y*ph^P1 mutations as described previously (GEORGIEV and KOZYCINA 1996 ). The su(Hw)^v mutation is deletion of the su(Hw) gene, whereas su(Hw)² is a strong mutation induced by jockey insertion in the first intron of the su(Hw) gene (HARRISON et al. 1993 ).

DNA manipulations:
DNA from adult flies was isolated using the protocol described in ASHBURNER 1989 . The genomic DNA was digested with restriction enzymes according to the supplier's instructions and separated in standard agarose gels (SAMBROOK et al. 1989 ). The DNA was transferred to Hybond N⁺ membranes and probed according to the supplier's instructions. The DNA fragments used as the probes were separated in agarose gels and purified using Gene Clean II (BIO 101, Vista, CA) according to the supplier's instructions.

The genomic DNAs were subjected to PCR to amplify sequences from the derivative alleles (SAIKI et al. 1985 ; MULLIS and FALOONA 1987 ). The primers used in DNA amplification were as follows: ATGCATTCTATGCACGAGCCTCC (y-1, 2409–2432 bp), CAGCGAAAGGTGATGTCTGACTC (y-2, 2606–2628 bp), TCTGTGGACCGTGGCGCGGTAAC (y-3, 2898–2877 bp), and ACTTCCACTTACCATCACGCCAC (y-4, 3293–3270 bp) for the yellow gene according to sequences of the yellow gene presented in GEYER et al. 1986 ; CGCTGAGGAACTCGAGAAAGGCC (c-1) and ACTGCGTGCTTCAAGCTTCTACC (c-2) for the internal genomic insertion of the y^+s1 allele; AAGCTAACAGCGCTGTTGGCTAG (sn-1, 875–898 bp) and ACACCCTTCACTTCGCCAATTCG (sn-2, 1632–1609 bp) for the singed gene according sequences of singed gene presented in PATERSON and OAHARE 1991 ; and CTCTCAACAAGCAAACGTGCACTG (p-1, 89–66 bp), CGTCCGCACACAACCTTTCCTCTC (p-2, 108–85 bp), CTCAATACGACACTCAGAATACT (p-3, 2802–2824 bp); and ATACGTTAAGTGGATGTCTCTTG (p-4, 2853–2875 bp) for the P-element according to sequences of the P element presented in OAHARE and RUBIN 1983 . The amplifications were performed with a program including 1 cycle of 3 min at 94°, 30 cycles of 1 min at 94°, 1 min at 64°–68° (the temperature depended on the primers), and 2 min at 72° and, the last, 4 min at 72°.

The PCR products were directly sequenced using a Sequenase II DNA sequencing kit for PCR product (Amersham, Arlington Heights, IL) according to manufacturer's instructions.

Northern blot hybridization:
Total RNA was isolated by homogenization in 4 M guanidine isothiocyanate, 0.2% N-lauryl sarcosine, 150 mM mercaptoethanol, 12.5 mM EDTA, and 50 mM tris-hydrochloride, pH 7.5, with subsequent phenol extraction and ethanol precipitation (PARKHURST et al. 1988 ). Poly(A)⁺RNA was selected by chromatography on oligo(dT) cellulose. Northern blot analysis was carried out as described by PARKHURST et al. 1988 .

Construction of P{w⁺, P-yellow} and P-element transformation:
The structure and expression of the yellow gene were described earlier (GEYER et al. 1986 ; GEYER and CORCES 1987 ). The 10-kb fragment including the yellow gene and its regulatory 3-kb region was provided by Dr. P. Geyer. The regulatory SalI/BamHI yellow region (3 kb) was subcloned in pGEM7 restricted by BamHI and XhoI. The XhoI-ScaI fragment of P element was eluted from the gel and blunt ligated in the pGEM7 vector. A plasmid with two ligated XhoI-ScaI fragments was selected by PCR analysis. After that, the double XhoI-ScaI fragment was cut out, blunt ligated, and cloned in the regulatory SalI/BamHI yellow region (3 kb) in the KpnI site (Figure 1). One of the obtained clones was sequenced to confirm that no unwanted changes in the yellow sequence had taken place. The 3-kb yellow fragment cut out by BamHI and XbaI was subsequently cloned in the BamHI and XbaI sites of the CaSpeR3 plasmid with a 5-kb coding region of the yellow gene (Figure 1). The new construct was denoted as P{w⁺, P-yellow}, where w⁺ indicates a miniwhite gene included into the vector and P-yellow indicates a complete yellow gene with an insertion of a small part of the P element. The control construct P{w⁺, yellow⁺} has an unchanged structure of yellow sequences. This construct was obtained by direct cloning of the 10-kb fragment including the yellow gene in the CaSpeR3 plasmid.

View larger version (33K): In this window In a new window Download PPT slide   Figure 1. Schematic presentation of the derivatives obtained from the y+s1 allele and their interaction with the phP1 and su(Hw) mutations. The directions of transcription of the yellow gene and of the 1A gene are indicated by arrows. The P-element sequences and their size are shown by boxes. The P-element orientation is indicated by an arrow. 1A-En is the enhancer located in the chimeric element. Thin lines show the deletion in the 2.9-kb genomic sequence of chimeric mobile element in the y2s10-2 allele. Sequences remaining after P-element excision are indicated below a corresponding P element. The structure of y2s2 is shown in the lower part of the diagram. Restriction enzymes are as follows: B, BamHI; H, HindIII; P, PstI; R, EcoRI; X, XhoI. The total number of black and white circles indicates the level of pigmentation in ph+ flies in 1, body; 2, wings; 3, thoracic bristles; 4, leg bristles; 5, wing bristles; and 6, abdominal bristles. The number of white circles shows the level of pigmentation in flies with the phP1 mutation. Thus the number of black circles shows the effect of the phP1 mutation. One circle represents one point as in Table 1. The effect of the phP1 mutation was also studied on the su(Hw)- and su(Hw)+ backgrounds. The su(Hw)- background corresponds to the trans-heterozygous su(Hw)v/su(Hw)2.

DNAs of the P{w⁺, P-yellow} and P{w⁺, yellow⁺} constructs and of a P element with a defective inverted repeat used as a transposase source, P25.7wc (KARES and RUBIN 1984 ), were injected into y^- ac^- w^67c preblastoderm embryos as described previously (RUBIN and SPRADLING 1982 ; SPRADLING and RUBIN 1982 ). Survivors were crossed to y^- ac^- w^67c flies and transgenic flies were separated on the basis of eye color. Chromosome assignments of the various transgene insertions were made by crossing the transformants with y^-ac^-w¹¹¹⁸ balancer stock containing dominant markers: In(2RL),CyO for the second chromosome and In(3LR)TM3,Sb for the third chromosome. Chromosomal insertions of P{w⁺, P-yellow} and P{w⁺, yellow⁺} were made homozygous and their copy number was determined by Southern blot analysis using the yellow and white sequences as probes.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Cis-elements influencing the repression effect of the ph^P1 mutation on yellow expression:
In a previous article (BELENKAYA et al. 1998 ), we found that the ph^P1 mutation completely inhibited yellow expression in y^2s alleles induced by two copies of P element inserted 69 bp upstream of the yellow promoter and mildly inhibited it in y^2s alleles induced by a single P-element copy at the same position. In addition, all described y alleles contained a gypsy insertion at -700 bp position. The su(Hw) insulator located in the 5' regulatory region of gypsy blocks the yellow wing blade and body enhancers. As a result, the y alleles display y² phenotype (yellow color of body cuticle and wings).

In this article, we studied the y^+s1 allele (GEORGIEV et al. 1997 ) that contains, in addition to a pair of 1.2-kb P elements, a duplicated 2.9-kb genomic sequence inserted between them (Figure 1). The P elements were designated as P1 and P2 depending on whether they were located distally or proximally with respect to the yellow promoter. The 2.9-kb genomic sequence originated from the 1A region of the X chromosome. An efficiently transcribed putative gene designated as the 1A gene was found in the middle of the 2.9-kb sequence (Figure 1). y^+s1 flies have wild-type level pigmentation in body cuticle and wing blade in contrast to flies with the original y² allele, in which the body and wing enhancers are blocked by su(Hw) insulators. A number of derivatives with y² phenotype were obtained from the y^+s1. All of them had deletions of various size in the 2.9-kb genomic sequence of chimeric mobile element (GEORGIEV et al. 1997 ; I. BIRUKOVA and P. GEORGIEV, unpublished results). The mapping of these deletions allowed us to localize the enhancer element between the 1A gene and P2 element (Figure 1).

The ph^P1 mutation only partially repressed yellow transcription in the y^+s1 allele: the pigmentation of the body, wings, thoracic, and leg bristles was decreased (Figure 1). Thus, the presence of the genomic insertion possessing the 1A enhancer reduces the effect of the P-Ph protein.

To further analyze the role of different cis-regulatory elements in formation of repressive complex, we studied the y^+s1 derivatives obtained after P-element mobilization in the y^+s1 ph^P1 and y^+s1 strains. The structure of insertions in some of these derivatives was described in a previous article (GEORGIEV et al. 1997 ), while for new y alleles, it was determined using Southern blot analysis and the PCR cloning and sequencing technique. The presence of nonmodified P element in the ph^P1 allele was verified by Southern blot analysis as described by BELENKAYA et al. 1998 . All derivatives may be divided for several classes because of their structure and the interaction with ph^P1. To simplify the presentation, we described here only one representative y allele for each class of derivatives, although several independently obtained y alleles with the identical or similar structure have been analyzed in each case.

The y^2s10-2 allele was obtained as a result of 1A enhancer deletion (GEORGIEV et al. 1997 ). The combination of y^2s10-2 with the ph^P1 mutation led to a complete inhibition of yellow expression (Figure 1). Thus, the 1A enhancer is responsible for the reduction of the P-Ph protein effect.

The y^+s27 allele was generated by excision of P1 element. PCR cloning and sequencing showed that only 15–17 bp from each terminus were retained (Figure 1). The ph^P1 mutation failed to influence the phenotype of y^+s27 flies, suggesting that the 1A enhancer completely overcomes ph^P1-mediated repression, if only one P element is present at the yellow locus.

The y^+s alleles associated with duplication of any P element were characterized by an almost complete inactivation of yellow expression by the ph^P1 mutation (Figure 1). The y^+sA11 allele was associated with duplication of the P1 element in the head-to-head orientation (Figure 1). Another y allele, y^+sA12, had P2-element duplication also in the head-to-head orientation. Both y^+s alleles had the same strong mutant phenotype in combination with the ph^P1 mutation. Thus, only the number of P-element copies at the yellow locus was important for the enhancement of the effect of the ph^P1 mutation, but not their localization in the chimeric element.

The y^+s22 allele was induced by an inversion of the central 2.9-kb region without P-element deletion (GEORGIEV et al. 1997 ). y^+s22 flies had the wild-type level of pigmentation indicating the 1A enhancer to act in an orientation-independent fashion. However, the ph^P1 mutation had a stronger effect on the y^+s22 alleles than on the original y^+s1 allele (Figure 1). Thus, the orientation of the 2.9-kb insertion influences the level of ph^P1-mediated repression. Possibly, the insertion of the 1A promoter between the 1A enhancer and the yellow promoter facilitates inhibition by the P-Ph protein of the 1A enhancer action on yellow transcription. Another explanation is that the inversion brings a distance-dependent negative element cooperating with ph^P1 into close vicinity to the yellow promoter. However, the y^2s2 derivative, which had, in addition to the P2 element, a 410-bp sequence just from the distal part of the 2.9-kb insertion (Figure 1), was affected by ph^P1 to the same extent as y^2s alleles with an insertion of the P2 element alone (Figure 2 and Figure 3).

View larger version (24K): In this window In a new window Download PPT slide   Figure 2. Structure of the yellow locus in wild-type and mutant y alleles and in the P{w+, P-yellow} construct. Exons of the yellow gene and the direction of transcription are indicated by arrows. The exons are separated by one intron. Transcriptional enhancers are indicated by circles. En-w, enhancer for the wing blade; En-b, enhancer for the body cuticle; En-br, enhancer for bristles. Insertions responsible for different alleles are represented by triangles. The gypsy element is inserted in the 5' region of the yellow gene in the y2 allele and its derivatives. The arrows flanking gypsy are long terminal repeats (LTRs). They indicate the direction of gypsy transcription. Deletion of DNA sequences in the y1s8 allele is represented by a horizontal line. The su(Hw) binding site is represented by an ovoid located in the 5' transcribed-untranslated region of gypsy. Restriction enzymes are as follows: H, HindIII; G, BglII; X, XhoI; K, KpnI; B, BamHI; S, ScaI. The primers used for PCR cloning are indicated by arrowheads. The P{w+, P-yellow} indicates the site of insertion (KpnI) of the double XhoI-ScaI fragments of P element. In the lower part of the diagram P{w+, P-yellow} is shown schematically with the wing (En-w) and body (En-b) enhancers, P-element sequences (P), and miniwhite and yellow fragments indicated. The arrow shows the direction of transcription of the miniwhite and yellow genes. The miniwhite gene is not shown in scale.

View larger version (36K): In this window In a new window Download PPT slide   Figure 3. Schematic presentation of the phP1 effect on P-element deletion derivatives in y alleles. Above is a structure of the 1.2-kb P element. The double line in the center of this box indicates the breakpoints of the internal deletion. IR, 31-bp inverted repeat; TPS, transposase-binding site. Below are the P-element sequences present in the alleles indicated at the left. The punctuated boxes represent P-element sequences remaining in the deletion derivatives; the black boxes represent the complete 1.2-kb copy of the P element. The numbers below the boxes are the numbers of base pairs remaining in the P-element sequences from the corresponding end to the break point. The arrows show the direction of P element in the corresponding alleles. The total number of black and white circles in each lane indicates the level of pigmentation of y*phP1 females in 1, body; 2, wings; 3, thoracic bristles; 4, leg bristles; 5, wing bristles; and 6, abdominal bristles. The number of white circles shows the level of pigmentation in the y*phP1 females (the number of black circles shows the effect of the phP1 mutation). The su(Hw)2/su(Hw)v mutations in the su(Hw) gene were combined with the y*phP1 to produce su(Hw)- background.

To further study the dependence of ph^P1 repression on the orientation of the chimeric element, we mobilized P elements in the y^+s22ph^P1 strain. As a result, we obtained a y^+s32 derivative characterized by a variegated pigmentation of bristles in combination with ph^P1 (Figure 1). The y^+s32 allele was generated by an excision of P2 element, leaving only 14 and 18 bp of the terminal repeats. Thus, the ph^P1 mutation can inhibit yellow expression in the presence of just a single distal P element located at a 3-kb distance from the yellow promoter if an inversion of the 2.9-kb sequence takes place.

The upstream region of the yellow gene contains the yellow wing and body enhancers, which are blocked due to the presence of the su(Hw) insulator (Figure 2). To study the role of yellow enhancers in the ph^P1-mediated repression, we inactivated the su(Hw) insulator, crossing the combination of y and ph^P1 mutations into su(Hw)^- background. The su(Hw)^v/su(Hw)² transheterozygote completely suppressed the mutant y phenotype of flies carrying the combination of the ph^P1 mutation with either y^+s1 or y^+s32 mutation (Figure 1). Thus, the yellow body and wing enhancers are also able to suppress a weak ph^P1-mediated repression.

The su(Hw)^v/su(Hw)² transheterozygote partially suppressed the ph^P1-mediated repression in the y^+sA11, y^+sA11, and y^+s22 alleles. This result shows that 1A and yellow enhancers cooperatively protect yellow expression from the ph^P1 repression. However, the yellow enhancers fail to influence strong repression in the y^2s10-2ph^P1; su(Hw)^v/su(Hw)² flies (Figure 1). Thus, the yellow enhancers have weaker effect on the ph^P1-mediated inhibition of transcription than the 1A enhancer in the y^+s1 allele.

The P-element sequences responsible for ph^P1-mediated repression:
In an earlier article (BELENKAYA et al. 1998 ), we demonstrated the absence of ph^P1 effect if the P-element deletion led to the loss of transposase-binding sites. Here we have further studied the role of different P-element sequences. From this, the derivatives of y^2s14 and y^+s1 alleles generated by partial deletions of the P elements were characterized. The breakpoints of the deletions in the P elements were cloned by PCR and sequenced. The effect of the ph^P1 mutation was studied in y*ph^P1/Y males and in y*/y¹ph^P1 heterozygous females.

As was shown previously, the ph^P1 mutation mildly repressed yellow expression in the y^2s14 allele containing a single P-element copy (BELENKAYA et al. 1998 ). Its derivative, y^2s34 allele, was induced by a P element with an internal deletion. The breakpoints of the deletion were mapped at distances of 767 bp from the 5' terminus and 93 bp from the 3' terminus; i.e., both termini contained transposase-binding regions, but the 11-bp inverted repeat at the 3' terminus was deleted (Figure 3). The ph^P1 mutation had no visible effect on the yellow expression in combination with y^2s34. The suppression of the ph^P1 effect may be explained by the previous observation that the 11-bp inverted repeat serves as an additional binding site for a truncated transposase (LEE et al. 1996 ).

Another derivative, y^2s42 allele, which had only 41 bp of the 3' end and 852 bp of the 5' end, was affected by ph^P1: the notum and leg bristles acquired a variegated mutant phenotype (Figure 3). The P element in the y^2s42 allele had a deletion of 3' transposase-binding sites, but it differed from the y^2s34 allele by the presence of a part of the central region of the defective P element between 767 and 852 bp. The latter included the breakpoint of the deletion and the filler sequence TAGCTACAAA, suggesting the presence of additional binding sites for the chimeric P-Ph protein in this region.

To identify the role of the central region of the P element in the ph^P1 repression, we prepared the P{w⁺, P-yellow} construct containing two 179-bp XhoI-ScaI restriction fragments of the 1.2-kb P element (the fragment between 730 and 909 bp of the latter) at the position of -340 bp from the yellow transcription start site (Figure 2). Seven independent transformants were obtained, possessing a single insertion as confirmed by Southern blot analysis. All of them displayed wild-type levels of yellow expression. The combination of the ph^P1 mutation with the P{w⁺, P-yellow} construct in five cases led to mosaic pigmentation of bristles, while the pigmentation of the body and wings was not changed (data not shown). The level of repression was the same for heterozygous and homozygous states of the P{w⁺, P-yellow} construct. As a control, we used P{w⁺, yellow⁺} transposon, which contained the unmodified yellow gene. Nine transgenic lines carrying P{w⁺, yellow⁺} transposon were obtained. The y phenotype of flies from these lines was not changed on a ph^P1 background. This result confirms the role of the central XhoI-ScaI region of the 1.2-kb P element in the ph^P1-mediated repression of the yellow expression.

We also compared eye pigmentation in either the presence or absence of the ph^P1 mutation. The miniwhite gene is very sensitive to repression by Pc-G complex. For example, a 661-bp PRE core fragment from the Ubx gene, inserted in a CaSpeR3 vector, generally causes variegated miniwhite gene expression in 50–60% of the lines (CHAN et al. 1994 ). We found that only one of six lines carrying P{w⁺, P-yellow} transposon in heterozygous state displayed a weak decrease of eye pigmentation after cross to a ph^P1w background. In the homozygous state of the P{w⁺, P-yellow} transposon, flies from four of six lines showed decrease of eye pigmentation and a weak variegation characteristic for Pc-G repression of white expression (data not shown). Thus, the ends of P elements in the presence of ph^P1 have only a weak repression potential compared to natural PREs.

As shown previously (BELENKAYA et al. 1998 ), the ph^P1 mutation mildly repressed yellow expression in the y^2s14 and y^2s16 alleles containing a single P-element copy, whereas it completely inhibited yellow expression in the y^2s11 allele induced by an insertion of two P-element copies in the head-to-head orientation (Figure 2 and Figure 3). The y^2s11 allele contains the P1 element with conserved ends and the P2 element with a deletion of 5'-terminal 82 bp. This deletion includes the 5'-terminal binding sites for the P-element transposase or the P-Ph protein. To find a minimal region of the P2 element that was sufficient for strong enhancement of the ph^P1 repressive effect, we analyzed the structure of two y^2s mutations, derivatives of y^+s1, which in combination with ph^P1 led to a complete inactivation of yellow expression. Both y alleles contained the unchanged P1 element. The y^2s25 allele contained only 39 bp of the 5' terminal sequence and 349 bp of the 3' terminus of the P2 element. The y^2s26 allele contained just 153 bp of the 3' terminal sequence of the P2 element (Figure 3). However the ph^P1 mutation still completely repressed the yellow expression in these alleles. Thus, the addition of the P-element transposase binding sites and the 11-bp inverted repeat to the single P element is sufficient for strong enhancement of the ph^P1-mediated repression.

To make the difference between strong and weak ph^P1-mediated repression more clear, we introduced the combinations of ph^P1 with the y mutations into a su(Hw)^- background (Figure 3). The su(Hw)²/su(Hw)^v transheterozygote completely suppressed the mutant yellow phenotype of flies carrying the combination of the ph^P1 mutation with either y^2s14 or y^2s16 mutation induced by single P-element insertion. Thus, the yellow body and wing enhancers are able to suppress a weak ph^P1-mediated repression. On the contrary, the su(Hw)²/su(Hw)^v transheterozygote only weakly influences the phenotype of flies carrying the ph^P1 mutation with either y^2s11, y^2s26, or y^2s25 alleles. Thus, the 3' sequences of the P element can strongly enhance the ph^P1-mediated repression.

Pairing-dependent repression of yellow expression mediated by ph^P1:
The repressive effect of the ph^P1 mutation was stronger in homozygous y^+s1 females than in y^+s1 males or heterozygous y¹/y^+s1 females (Table 1), where the y¹ allele was generated by a point mutation in the coding region of the yellow gene (GEYER et al. 1990 ). Thus, the presence of two P-element copies in each homologue resulted in the enhancement of silencing. To study the transinteraction in detail we used well-studied y alleles in combination with ph^P1. As has been shown above, the y^+s32 allele is associated with a deletion of the P2 element proximal to the yellow gene and the y^+s27 allele—with a deletion of the distal P1 element. The ph^P1 mutation influenced almost to the same extent on heterozygous y¹/y^+s32 and homozygous y^+s32/y^+s32 females and had no effect on the y^+s27 allele (Table 1).

To further study the role of P-element sequences in the formation of a ph^P1-dependent repression complex, we studied the phenotype of heterozygotes for the above mentioned y^+s1, y^+s32, or y^+s27 alleles with the y alleles possessing one, one and one-half, or two P-element copies in combination with ph^P1. For a y allele with a single P-element copy, we used the y^1s8 mutation (BELENKAYA et al. 1998 ) associated with a deletion of the yellow promoter, but still containing a single P-element copy (Figure 2). This y allele displays the same y-null phenotype as the combination of ph^P1 and y^2s alleles containing more than one P-element copy.

The combination of y^1s8 with either y^+s1, y^+s32, or y^+s27 did not significantly increase the effect of ph^P1 as compared to its effect in heterozygous females (Table 1). Thus, a single P-element copy failed to induce a strong pairing-dependent repression by the ph^P1 mutation. In contrast, the y^2s7, y^2s11, and y^2s25 alleles induced a strong pairing-dependent repression (Table 1). These y^2s alleles were induced either by two P elements or by the P2 element and the 3' terminus of the P1 element (Figure 2 and Figure 3). However, the effect of ph^P1 was weaker in heterozygous females with the y^2s26 allele. The y^2s26 allele differed from the y^2s25 allele only by a deletion of the central region of the P1 element, further confirming the role of this part of the 1.2-kb P element in the ph^P1-mediated repression of yellow transcription.

Su(y) mutations:
In the progeny obtained after P-element mobilization in y^+s1 ph^P1 flies, we could frequently observe the appearance of autosomal suppressors of ph^P1-mediated repression. These mutations, designated as Su(y), were widely spread on the X, second, and third chromosomes and displayed different levels of the suppression effect. Eleven Su(y) mutations were lethal in the homozygous state, 9 possessed normal or decreased viability, and 2 were associated with female sterility. One of the Su(y) mutations appeared simultaneously with a mutation in the ebony (e) locus, and the recombination analysis showed them to be one and the same mutation. Another strong Su(y) mutation was obtained in the singed locus (sn^eP1).

For further study, we selected three Su(y) alleles, Su(y)^P31 (third chromosome), Su(y)^P22 (second chromosome), and sn^eP1. To compare the levels of suppression, we studied the phenotype of females with the y ph^P1/y* ph^P1; Su(y)/+ genotype, where y* was one of the y alleles described above (Figure 4). All three Su(y) mutations completely suppressed the effect of ph^P1 on the y^2s14 and y^+s1 alleles. The Su(y)^P31 mutation also almost completely suppressed the effect of ph^P1 on the y alleles induced by a chimeric element containing even three P elements (y^+sA8 and y^+sA22) or an inversion of the chimeric element (y^+s12). Su(y)^P22 and sn^eP1 had only a weak suppressive effect in these combinations. However, the Su(y) mutations did not have any visible effect on the combination of ph^P1 with y^2s7, y^2s11, y^2s25, and y^2s26 alleles, in which yellow expression is completely repressed (data not shown).

View larger version (68K): In this window In a new window Download PPT slide   Figure 4. The effect of Su(y) mutations on the inhibitory effect of the phP1. (A) The effect of Su(y)P31, Su(y)P22, and sneP1 mutations on the y* phP1/y1 phP1 females. The total number of circles in each lane indicates the level of pigmentation of y* phP1/y1 phP1 females in the presence of the Su(y) mutations: 1, the body; 2, wings; 3, thoracic bristles; 4, leg bristles; 5, wing bristles; and 6, abdominal bristles. The number of black circles indicates the level of yellow expression in phP1 flies in the absence of Su(y) mutations. The dotted circles indicate variations of yellow expression. Upper lines show the level of pigmentation in ph+ flies; lower lines show the level of yellow expression in y-null mutation. (B) Northern blot analysis of transcripts hybridizing with P-element HindIII-XhoI probe in Su(y) strains. The mRNA was isolated from the y2s14 (1, 8, 12, 16, 20), y2s14; Su(y)P31/TM3,Sb (3, 6, 10, 14, 18), y2s14; Su(y)P22/CyO (4, 5, 9, 13, 17), and y2s14 sneP1 (2, 7, 11, 15, 19) strains at different stages of development: first instar larvae (1–4), second instar larvae (5–8), third instar larvae (9–12), early pupae (13–16), and late pupae (17–20). The HindIII-XhoI probe was hybridized to the 1.0-kb major transcript and some other minor transcripts of higher molecular weight.

The most plausible explanation was that Su(y) mutations were induced by P-element insertions into regulatory or transcribed regions of different genes that led to a high level of transcription of the defective P elements and to the accumulation of truncated transposase that might suppress the mutant phenotype of the y alleles competing with the P-Ph protein for binding sites. To test this hypothesis, we performed a Northern blot analysis of mRNA from the Su(y) and control strains using as a probe the P-element HindIII-XhoI fragment that included the sequences encoding the DNA-binding domain of transposase (ANDREWS and GLOOR 1995 ; LEE et al. 1996 ). A high level of mRNA hybridizing with the P-element sequences was detected only in the Su(y) strains (Figure 4). The highest level of P-element transcripts was found in the Su(y)^P31 strain at the larval, pupal, and adult stages of development. The Su(y)^P22 strain had a lower but still rather high level of P-element transcripts at all tested stages of development. In the sn^eP1 strain, a high level of P-element transcripts was detected only at the early pupal stage of development, when the singed gene was actively transcribed (PATERSON and OAHARE 1991 ).

Molecular study of the sn^eP1 allele:
To gain a better insight into the mechanism of Su(y) action, we used for molecular analysis the sn^eP1 mutation in the well-characterized singed locus (ROIHA et al. 1988 ; PATERSON and OAHARE 1991 ). The sn^eP1 mutation had a strong mutant phenotype. Southern blot hybridization revealed an ~1.0-kb insertion localized between the EcoRI and SalI restriction sites of the DNA segment (Figure 5). This insertion was within the 700-bp singed "hot spot" region (HAWLEY et al. 1988 ; ROIHA et al. 1988 ). For cloning of the P-element insertion, the oligonucleotides were designed from both sides of the insertion (Figure 5). The PCR-amplified fragment was directly sequenced. The sn^eP1 insertion was shown to be a 1180-bp internally deleted P element with a structure identical to that described for the P element responsible for superunstable y alleles and ph^P1 mutation. Based on the published sequences of a full-length P element (OAHARE and RUBIN 1983 ), this sn^eP1 P element contained an internal deletion from 830 to 2560 bp and a filler sequence TAGCTACAAA at the break point. Interestingly, as in the case of highly unstable y mutations, the P-element insertion at the singed locus was associated with a deletion: 4 bp (CGCG) at the target site were lost, and no 8 bp duplication typical of P-element insertions was observed. The P element in the sn^eP1 allele is localized in the transcribed but not translated part of the singed gene ~80 bp downstream from the signal of transcription initiation (PATERSON and OAHARE 1991 ; BURKE and KADONAGA 1996 ).

View larger version (33K): In this window In a new window Download PPT slide   Figure 5. Structure of sn alleles. Start and direction of singed transcription is indicated by an arrow. Insertions of P elements responsible for different sn alleles are represented by triangles. The P-element sequences and their size are shown by boxes. The orientation of P elements are indicated by arrows. The sequence of the region of P-element insertions is shown in the lower part of the figure. The duplicated eight nucleotides of the host DNA are underlined. The deleted cgcg nucleotides are shown by lowercase letters. The primers used for PCR cloning are shown by arrowheads.

To prove the responsibility of sn^eP1 for the Su(y) effect, we studied revertants induced by crossing with 2-3. Five sn^+P revertants were obtained and shown by Southern blot hybridization to lack the P element. As expected, the sn^+P revertants, sn^+P11 and sn^+P12, simultaneously lost the ability to suppress the mutant phenotype of the y^2s14ph^P1 combination (Figure 5).

We also studied two sn^wP mutations, sn^wP13 and sn^wP14, obtained from sn^eP1 as a result of P-element duplication in the head-to-head orientation, as in the case of the well-known sn^w mutation (ROIHA et al. 1988 ). The sn^wP mutations only slightly suppressed the y^2s14ph^P1 mutant phenotype, possibly due to a low level of P-element transcription (Figure 5).

The P elements in the previously described sn^w mutation (ENGELS 1989 ) are located in the same region as in the sn^eP1 mutation. The sn^w allele contains two P elements of 1.15 and 0.95 kb inserted at the same site in the inverse orientation. Excision of the 0.95-kb P element gives rise to sn^e, and excision of the 1.15-kb P element, to sn⁺ (ROIHA et al. 1988 ). The sn^w and sn⁺ mutations did not influence the effect of ph^P1. Only the sn^e allele had a weak suppressive effect (Figure 5). The P elements in sn^e and sn^eP1 have the same orientation and are inserted at a distance of 13 bp from each other (Figure 5). The only difference between the P elements is the deletion in the P element inserted in the sn^e allele of the second region important for the protein-protein interaction that enhances DNA binding (LEE et al. 1996 ). This result shows that a strong DNA-binding ability is important for realization of the Su(y) effect.

The ph^P1 mutation does not affect singed expression in P-element-induced sn mutations:
The singed gene of D. melanogaster plays a role in both somatic and germ cells (LINDSLEY and ZIMM 1992 ). The bristles and hairs are shortened or twisted and gnarled in sn mutants. In addition to this phenotype, many strong sn mutations lead to female sterility (LINDSLEY and ZIMM 1992 ).

The P element in the above described sn mutations was inserted at a distance of <50 bp from the core region of the downstream promoter element (DPE) of the singed gene (PATERSON and OAHARE 1991 ; BURKE and KADONAGA 1996 ). However, the ph^P1 mutation did not increase the visible mutant phenotype of the sn mutations induced by one (sn^e, sn^eP1, and sn⁺) or two (sn^w, sn^wP13, and sn^wP14) P elements (data not shown). This result could not be explained by the fact that sn mutations suppressed the ph^P1 effect, because such a weak suppressor as the sn^w mutation also was not modified by ph^P1. The ph^P1 mutation also did not influence the sn-phenotype in females, suggesting the absence of the pairing effect.

P-element-induced sn mutations did not significantly influence the singed function in gonads: females homozygous for a sn mutation were fertile. Only the sn^eP1 mutation reduced female fertility. The ph^P1 mutation did not significantly increase the sterility of females with homozygous sn mutations.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The properties of P-element-mediated repression of the yellow gene by the ph^P1 mutation:
The chimeric P-Ph protein encoded by the ph^P1 locus consists of the P-element transposase DNA-binding domain and an almost full-length Ph protein (BELENKAYA et al. 1998 ). It binds to the P-element sequences and leads to Pc-G-dependent silencing in y alleles induced by P-element insertions. Thus, against the ph^P1 background the P-element sequences function as the known response PREs. PREs were found in the regulatory regions of many homeotic genes and are responsible for transcription repression (PIRROTTA 1997 ). PREs were identified by their silencing effect on a reporter gene and by the introduction of variegated or pairing-sensitive expression of the white gene in transgenic flies (FAUVARQUE and DURA 1993 ; SIMON et al. 1993 ; CHAN et al. 1994 ; KASSIS 1994 ). DNA fragments with PRE activity are generally from several hundred to several thousand base pairs in length. They are composite in structure and can be dissected into multiple subfragments that display a progressively weakening PRE activity. PREs have a strong cooperative effect on repression. For example, in the Ubx gene, where silencing is required to be strong and stable, several weaker sites of Pc-G protein binding cooperate in establishing a strong silencing complex (MULLER and BIENZ 1991 ; POUX et al. 1996 ).

The y alleles used in this study have been generated by an insertion of the 2.9-kb genomic DNA sequence from the 1A region of X chromosome located more distally with respect to the yellow gene. The genomic duplication is flanked by two identical copies of a deleted 1.2-kb P element (GEORGIEV et al. 1997 ). The 1A enhancer located in the 2.9-kb insertion is responsible for yellow activation in the body cuticle and wing blade (GEORGIEV et al. 1997 ) compensating the yellow body and wing enhancers blocked by the su(Hw) insulator. The appearance of a strong enhancer disturbs the repressive effect of the P-Ph protein. Addition of the wing and body enhancers by inactivation of the su(Hw) function further suppresses the silencing effect of the ph^P1 mutation. It has been shown that the assembly of a silencing complex depends both on the strength of a PRE site and on the transcriptional activity of the region involved (PIRROTTA 1997 ). For example, the formation of a Pc-G complex and the binding of GAL4, which activates transcription, are mutually exclusive and the silencing state can be prevented by a strong activation of transcription (ZINK and PARO 1995 ; CAVALLI and PARO 1998 ).

The level of repression induced by ph^P1 depends directly on the number of P elements but not on their localization. The ph^P1-mediated repression may act at a distance from the yellow promoter. In the y^+s32 allele the P element is located ~3 kb from the yellow promoter and is separated from the latter by the 1A enhancer. Still in this case, the ph^P1 mutation inhibits yellow transcription. In this respect, the P elements act like PREs that are responsible for the repressive state in the adjacent genome region in a cooperative manner (ORLANDO and PARO 1993 ; PIRROTTA 1997 ).

The ph^P1 mutation induces only a weak repression in y alleles containing a single P-element copy upstream of the yellow gene. Thus, one copy of P element is not sufficient for the formation of a strong repression complex. Addition to a single copy of the 1.2-kb P element of 153-bp 3'-terminal P-element sequences strongly enhances ph^P1-mediated repression, leading to a complete inactivation of yellow expression. As expected, the deletion of transposase-binding sites or 11-bp inverted repeats reduces the ph^P1-mediated repression. Unexpectedly, the internal region of the 1.2-kb P element is also important for the ph^P1-mediated repression that suggests the presence of a potential site(s) for binding of the chimeric P-Ph protein.

PRE-containing transposons often show a dramatic enhancement of silencing when a fly is homozygous for a transposon insertion, indicating that the homologously paired PREs interact to produce a more stable and more repressive Pc-G complex (FAUVARQUE and DURA 1993 ; CHAN et al. 1994 ; KASSIS 1994 ). We did not find any pairing-dependent repression when both paired y alleles contained a single P-element copy inducing a weak or no ph^P1-mediated repression. A pairing-dependent repression was found only if one of the y alleles in heterozygous females induced a strong repression complex in the presence of ph^P1.

Truncated P-element transposase similar in structure to the KP protein suppresses the inhibitory effect of the chimeric P-Ph protein on the expression of P-element-induced y alleles:
The inhibitory effect of the ph^P1 mutation can be partially suppressed by Su(y) mutations induced by a P-element insertion into regulatory or coding regions of different genes. The inserted 1.2-kb P element is transcribed at a high level and gives rise to a truncated transposase. For example, in the sn^eP1 mutation, the 1.2-kb P element is inserted into the 5' transcribed noncoding region leading to an enhanced P-element transcription. As shown before, mRNA encoding truncated transposase was also formed as a result of the ph^P1 mutation (BELENKAYA et al. 1998 ) and this may partially decrease its effect.

The 1.2-kb P element has a deletion between 830 and 2560 bp and resembles the previously described KP element (BLACK et al. 1987 ). The DNA-binding domain is located within the region of 98 amino-terminal amino acids (RIO 1990 ; ANDREWS and GLOOR 1995 ). The KP protein also contains two protein-protein interaction regions, and dimerization of the KP protein is essential for high-affinity DNA binding (LEE et al. 1996 ). A putative leucine zipper is located between 101 and 122 amino acids of the KP protein (RIO 1990 ; ANDREWS and GLOOR 1995 ). The second protein-protein interaction region is present within a segment of 69 carboxy-terminal amino acids of the KP protein, i.e., still in the amino-terminal part of the intact P element (LEE et al. 1996 ). All these sequences are also present in the protein encoded by the 1.2-kb P element.

The KP protein binds to multiple sites on the P-element termini with a higher affinity than the full-length transposase (LEE et al. 1996 ). The binding sites include the high-affinity transposase binding sites, an 11-bp transpositional enhancer, and the terminal 31-bp inverted repeats. Both protein-protein interaction regions are important for this binding. The 1.15-kb P element in sn^e, a derivative of the sn^w mutation (ROIHA et al. 1988 ; ENGELS 1989 ), is inserted in almost the same place and has the same orientation as the P element in sn^eP1. The protein product of the 1.15-kb P element contains the DNA-binding domain and a leucine zipper, but not the second protein-protein interaction region (ROIHA et al. 1988 ; LEE et al. 1996 ). This explains why the sn^eP1 mutation strongly suppresses the ph^P1 inhibitory effect, while sn^e does not. The 2-3 construct generating a full length transposase has a very weak suppression effect on the ph^P1 mutation (T. BELENKAYA, unpublished results). This may also be explained by its lower affinity to the P-element DNA.

The defective transposase with the DNA-binding domain and two protein-protein interaction domains is most efficient for realization of the suppression effect. Thus, the suppression of ph^P1-induced inhibition is a result of the competition for binding sites on the P elements between the KP-like protein and the P-Ph protein.

The presence of a strong enhancer in the chimeric element helps the KP-like protein in blocking the assembly of a P-Ph-mediated Pc-G complex on the P element. However, the Su(y) mutations, even expressing the truncated P-element transposase at high levels at all stages of Drosophila development, fail to affect the ph^P1-mediated inhibition of y^2s alleles induced by a double P element in the absence of the 1A enhancer. The formation of a strong repression complex prevents binding of the KP protein to transposase-binding sites on the P elements. Thus, the general low level of transcription activity facilitates P-Ph protein binding to DNA and the formation of a repressive complex.

singed expression is insensitive to the ph^P1 mutation in the P-element-induced mutations:
The singed gene plays a role in somatic cells during the formation of adult bristles and hairs, and in the female germline during oogenesis (PATERSON and OAHARE 1991 ). During metamorphosis, a single 3.6-kb RNA is transcribed, but in ovaries two additional transcripts, 3.3- and 3.0-kb RNAs, appear. The mRNAs differ only in the length of the 3'-untranslated region and a single gene product is predicted for all of them. The singed gene contains a downstream promoter element (DPE) that is located only 50 bp upstream from the sites of insertions of P elements in the sn mutations (PATERSON and OAHARE 1991 ; BURKE and KADONAGA 1996 ). However, the ph^P1 mutation does not influence the bristle phenotype and female sterility of the sn mutants, even those induced by a double P element. This result may be explained either by a special organization of the singed promoter making it insensitive to the P-Ph protein or by a structure of chromatin in a vicinity of the singed locus.

As we have found here, the miniwhite expression in transposons is weakly sensitive to the ph^P1 mutation that may be explained by a large distance between the white promoter and P-element sequences in the transposon. However the effect of the ph^P1 mutation is not restricted to the yellow gene. In another article (SOLDATOV et al. 1999 ), we described the strong ph^P1-mediated repression of TAF_II40/e(y)1 expression in the e(y)^P1 allele that was generated by P-element insertion at the transcribed nontranslated region of the gene.

A relatively weak effect of the ph^P1 mutation in several cases may be explained by the low level of the P-Ph protein produced by the ph^P1 allele. To examine this possibility, we are constructing the transgenic lines producing a large amount of P-Ph protein.

	ACKNOWLEDGMENTS

The authors are greatly indebted to M. Gause for her help in some experiments and to K. O'Hare and P. Geyer for plasmids of the singed and yellow genes, respectively. This work was supported by the Russian Foundation for Basic Research, N 96-15-97032 and N 97-04-48742, by an International Research Scholar's award from the Howard Hughes Medical Institute, and by a Human Frontier Science Program Organization grant to P.G.

Manuscript received September 14, 1998; Accepted for publication April 21, 1999.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ANDREWS, J. D. and G. B. GLOOR, 1995  A role for the Kp leucine zipper in regulating P element transposition in Drosophila melanogaster.. Genetics 141:587-594[Abstract].

ASHBURNER, M., 1989 Drosophila. A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

BELENKAYA, T., A. SOLDATOV, E. NABIROCHKINA, I. BIRJUKOVA, and S. GEORGIEVA et al., 1998  P element insertion at the polyhomeotic gene leads to formation of a novel chimeric protein that negatively regulates yellow expression in P element-induced alleles of Drosophila melanogaster.. Genetics 150:687-697[Abstract/Free Full Text].

BLACK, D. M., M. S. JACKSON, M. G. KIDWELL, and G. A. DOVER, 1987  KP elements repress P-induced hybrid dysgenesis in Drosophila melanogaster.. EMBO J. 6:4125-4135[Medline].

BURKE, T. W. and J. T. KADONAGA, 1996  Drosophila TFIID binds to a conserved down stream basal promoter element that is present in many TATA-box-deficient promoters. Genes Dev. 10:711-724[Abstract/Free Full Text].

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

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].

CHIANG, A., M. B. O'CONNOR, R. PARO, J. SIMON, and B. BENDER, 1995  Discrete Polycomb binding sites in each parasegmental domain of the bithorax complex. Development 121:1681-1689[Abstract].

DURA, J.-M. and P. INGHAM, 1988  Tissue- and stage-specific control of homeotic and segmentation gene expression in Drosophila embryos by the polyhomeotic gene. Development 103:733-741[Abstract/Free Full Text].

DURA, J.-M., H. W. BROCK, and P. SANTAMARIA, 1985  polyhomeotic: a gene of Drosophila melanogaster required for correct expression of segment identity. Mol. Gen. Genet. 198:213-220[Medline].

ENGELS, W. R., 1989 P elements in Drosophila melanogaster, pp. 437–484 in Mobile DNA, edited by D. E. BERG and M. M. HOWE. American Society for Microbiology, Washington, DC.

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].

FRANKE, A., M. DECAMILLIS, D. ZINK, N. CHENG, and H. W. BROCK et al., 1992  Polycomb and polyhomeotic are constituents of a multimeric protein complex in chromatin of Drosophila melanogaster.. EMBO J. 11:2941-2950[Medline].

GEORGIEV, P. and M. KOZYCINA, 1996  Interaction between mutations in the suppressor of Hairy wing and modifier of mdg4 genes of Drosophila melanogaster affecting the phenotype of gypsy-induced mutations. Genetics 142:425-436[Abstract].

GEORGIEV, P. G., V. A. YELAGIN, E. M. BUFF, and N. P. KOLYAGIN, 1992  Properties of super unstable mutations in the Drosophila melanogaster yellow locus. Genetica 28:98-107. (in Russian)..

GEORGIEV, P., T. TIKHOMIROVA, V. YELAGIN, T. BELENKAYA, and E. GRACHEVA et al., 1997  Insertions of hybrid P elements in the yellow gene of Drosophila cause a large variety of mutant phenotype. Genetics 146:583-594[Abstract].

GEYER, P. K. and V. G. CORCES, 1987  Separate regulatory elements are responsible for the complex pattern of tissue specific and developmental transcription of the yellow locus in Drosophila melanogaster.. Genes Dev. 1:996-1004[Abstract/Free Full Text].

GEYER, P. K. and V. G. CORCES, 1992  DNA position-specific repression of transcription by a Drosophila Zn finger protein. Genes Dev. 6:1865-1873[Abstract/Free Full Text].

GEYER, P. K., C. SPANA, and V. G. CORCES, 1986  On the molecular mechanism of gypsy-induced mutations at the yellow locus of Drosophila melanogaster.. EMBO J. 5:2657-2662[Medline].

GEYER, P. K., M. M. GREEN, and V. G. CORCES, 1990  Tissue-specific transcriptional enhancers may act in trans on the gene located in the homologous chromosome: the molecular basis of transvection in Drosophila. EMBO J. 9:2247-2256[Medline].

HARRISON, D. A., D. A. GDULA, R. S. COYNE, and V. G. CORCES, 1993  A leucine zipper domain of the suppressor of Hairy-wing protein mediates its repressive effect on enhancer function. Genes Dev. 7:1966-1978[Abstract/Free Full Text].

HAWLEY, R. S., R. A. STEUBER, G. H. MARCUS, R. SOHN, and D. M. BARONAS et al., 1988  Molecular analysis of an unstable P element insertion at the singed locus of Drosophila melanogaster: evidence for intracistronic transposition of a P element. Genetics 119:85-94[Abstract/Free Full Text].

JURGENS, G., 1985  A group of genes controlling the spatial expression of the bithorax complex in Drosophila.. Nature 316:153-155.

KARES, R. E. and G. M. RUBIN, 1984  Analysis of P transposable element functions in Drosophila.. Cell 38:135-146[Medline].

KASSIS, J. A., 1994  Unusual properties of regulatory DNA from the Drosophila engrailed gene: three "pairing-sensitive" sites within a 1.6 kb region. Genetics 136:1025-1038[Abstract].

LEE, C. C., Y. M. MUL, and D. C. RIO, 1996  The Drosophila P-element KP repressor protein dimerizes and interacts with multiple sites on P-element DNA. Mol. Cell. Biol. 16:5616-5622[Abstract].

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

MULLER, J., 1995  Transcriptional silencing by the Polycomb protein in Drosophila embryos. EMBO J. 14:1209-1220[Medline].

MULLER, J. and M. BIENZ, 1991  Long range repression conferring boundaries of Ultrabithorax expression in the Drosophila embryo. EMBO J. 10:3147-3155[Medline].

MULLIS, K. B. and F. A. FALOONA, 1987  Specific synthesis of DNA in vitro via a polymerase-catalyzed chain reaction. Methods Enzymol. 155:335-350[Medline].

O'HARE, K. and G. M. RUBIN, 1983  Structure of P transposable elements and their sites of insertion and excision in the Drosophila melanogaster genome. Cell 34:25-35[Medline].

ORLANDO, V. and R. PARO, 1993  Mapping Polycomb-repressed domains in the bithorax complex using in vivo formaldehyde cross-linked chromatin. Cell 75:1187-1198[Medline].

PARKHURST, S. M. and V. G. CORCES, 1986  Interactions among the gypsy element and the yellow and suppressor of Hairy-wing loci in Drosophila melanogaster.. Mol. Cell. Biol. 6:47-53[Abstract/Free Full Text].

PARKHURST, S. M., D. A. HARRISON, M. P. REMINGTON, C. SPANA, and R. L. KELLEY et al., 1988  The Drosophila su(Hw) gene, which controls the phenotypic effect of the gypsy transposable element, encodes a putative DNA binding protein. Genes Dev. 2:1205-1215[Abstract/Free Full Text].

PATERSON, J. and K. O'HARE, 1991  Structure and transcription of the singed locus of Drosophila melanogaster.. Genetics 129:1073-1084[Abstract].

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

PLATERO, J. S., E. J. SHARP, P. N. ADLER, and J. C. EISSENBERG, 1996  In vivo assay for protein-protein interactions using Drosophila chromosomes. Chromosoma 104:393-404[Medline].

POUX, S., C. KOSTIC, and V. PIRROTTA, 1996  Hunchback-independent silencing of late Ubx enhancers by a Polycomb Group Response Element. EMBO J. 15:4713-4722[Medline].

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].

RIO, D. C., 1990  Molecular mechanisms regulating Drosophila P element transposition. Annu. Rev. Genet. 24:543-578[Medline].

ROBERTSON, H. M., C. R. PRESTSON, R. M. PHILLIPS, 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].

ROIHA, H., G. M. RUBIN, and K. O'HARE, 1988  P element insertions and rearrangements at the singed locus of Drosophila melanogaster.. Genetics 119:75-83[Abstract/Free Full Text].

RUBIN, G. M. and A. C. SPRADLING, 1982  Genetic transformation of Drosophila with transposable element vectors. Science 218:348-353[Abstract/Free Full Text].

SAIKI, R. K., S. SCHARF, F. FALOONA, K. B. MULLIS, and G. T. HORN et al., 1985  Enzymatic amplification of beta-globin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. Science 230:1350-1354[Abstract/Free Full Text].

SAMBROOK, J., E. F. FRITSCH and T. MANIATIS, 1989 Molecular Cloning: A Laboratory Manual, Ed. 2. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

SIMON, J., A. CHIANG, W. BENDER, M. J. SHIMELL, and M. O'CONNOR, 1993  Elements of the Drosophila bithorax complex that mediate repression by Polycomb group products. Dev. Biol. 158:131-144[Medline].

SOLDATOV, A., E. NABIROCHKINA, S. GEORGIEVA, T. BELENKAYA, and P. GEORGIEV, 1999  TAF_II40 protein is encoded by the e(y)1 gene: biological consequences of the mutations. Mol. Cell. Biol. 19:3769-3778[Abstract/Free Full Text].

SPRADLING, A. C. and G. M. RUBIN, 1982  Transposition of cloned P elements into germline chromosomes. Science 218:341-347[Abstract/Free Full Text].

STRUHL, G., 1981  A gene product required for correct initiation of segmental determination in Drosophila. Nature 293:36-41[Medline].

STRUTT, H. and R. PARO, 1997  The polycomb group protein complex of Drosophila melanogaster has different compositions at different target genes. Mol. Cell. Biol. 17:6773-6783[Abstract].

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].