The Dose of a Putative Ubiquitin-Specific Protease Affects Position ...

MOLECULAR AND CELLULAR BIOLOGY, Oct. 1996, p. 5717–5725 0270-7306/96/$04.0010 Copyright q 1996, American Society for Microbiology

Vol. 16, No. 10

The Dose of a Putative Ubiquitin-Specific Protease Affects Position-Effect Variegation in Drosophila melanogaster SANDRA HENCHOZ,† FRANCESCO DE RUBERTIS, DANIEL PAULI,

AND

PIERRE SPIERER*

Department of Zoology and Animal Biology, University of Geneva, CH-1211 Geneva 4, Switzerland Received 1 April 1996/Returned for modification 14 May 1996/Accepted 22 July 1996

A dominant insertional P-element mutation enhances position-effect variegation in Drosophila melanogaster. The mutation is homozygous, viable, and fertile and maps at 64E on the third chromosome. The corresponding gene was cloned by transposon tagging. Insertion of the transposon upstream of the open reading frame correlates with a strong reduction of transcript level. A transgene was constructed with the cDNA and found to have the effect opposite from that of the mutation, namely, to suppress variegation. Sequencing of the cDNA reveals a large open reading frame encoding a putative ubiquitin-specific protease (Ubp). Ubiquitin marks various proteins, frequently for proteasome-dependent degradation. Ubps can cleave the ubiquitin part from these proteins. We discuss the link established here between a deubiquitinating enzyme and epigenetic silencing processes. variegation (18, 36, 41), and some general activators of the trithorax group of genes show enhancer-of-variegation phenotypes (12, 17, 60). Modifiers of PEV should help in dissecting the mechanisms controlling the general activity of chromosomal domains. Other epigenetic silencing phenomena such as X inactivation and imprinting in mammals, telomeric and mating-type silencing in Saccharomyces cerevisiae, and centromeric silencing in Schizosaccharomyces pombe may use similar mechanisms (25, 30). We report here the genetic and molecular characterization of a strong dominant enhancer of PEV, D-Ubp-64EEvar1. We find it to map at 64E on the third chromosome and to encode a putative ubiquitin-specific protease (Ubp). Members of the Ubp family have been cloned in S. cerevisiae, flies, and mammals (1). The D-Ubp-64E locus is the first enhancer of PEV found to encode an enzyme and not a transcriptional factor. We discuss the possible involvement of this protein in the establishment of genomic silencing.

Position-effect variegation (PEV) is the mosaic transcriptional silencing of a euchromatic gene relocated next to heterochromatin by a chromosomal rearrangement. A model suggests that heterochromatin spreads over the breakpoint of the rearrangement and inactivates euchromatic genes located nearby (14, 26, 46, 61). The proportion of cells in which inactivation occurs is affected by genetic modifiers acting in trans. The phenomenon is best visualized in Drosophila melanogaster by using the wm4h rearrangement, a paracentric inversion of the X chromosome (62). The white gene, required for the wild-type red color of the eye, is brought into proximity to centromeric heterochromatin and is stochastically inactivated, resulting in mottling of the eyes of the adult fly. This rearrangement has been used to isolate dominant second-site mutations that increase or reduce the number of cells in which silencing occurs, namely, enhancers [E(var)] and suppressors [Su(var)] of PEV, respectively. Some suppressor mutations such as Su(var)205, encoding HP1; Su(var)3-7, encoding a zinc finger protein; and Su(var)3-9 correspond to heterochromatin constituents (7, 15, 45, 64), as expected for genes that reduce silencing when their dose is reduced. On the other hand, several enhancers encode transcription factors. The trithorax-like gene encodes the GAGA transcription factor (17), a key component for chromatin remodelling implicated in nucleosome displacement needed for full gene activation (56, 65); E(var)393D contains a tramtrack domain, common to several transcriptional activators (12); and E(var)3-93E encodes the E2F transcription regulator necessary for progression of the cell cycle at the G1/S boundary (54). It is not known whether these factors act directly on chromatin as architectural transcription factors or by indirect effects through the regulation of their target genes. Functional relationships between modifiers of PEV and regulators of clustered homeotic genes have been proposed. Some Polycomb-group genes, which are involved in the maintenance of the inactive state of homeotic genes, are also suppressors of

MATERIALS AND METHODS Genetic analysis. All the fly stocks were maintained under standard conditions. Chromosomes and mutations are described in reference 33. The allele of D-Ubp64E was generated by the laboratory of G. Reuter in a screen for dominant enhancers of PEV caused by insertion of the P{lArB} transposon (13). It was designated E1 in this publication and is renamed D-Ubp-64EEvar1 in the present work because of the correspondence with a ubiquitin-specific protease (Ubp). The putative enhancer was previously localized to the second chromosome with the help of aneuploid segregants of the apXa translocation (13, 48) and relocalized on the third chromosome in this study by in situ hybridization on polytene chromosomes (see below and Results). The enhancer effect of the mutation on wm4h variegation was quantified by optical density pigment measurement at 480 nm as described by Reuter and Wolff (48) after a cross of wm4h; 1/1 females to w1/Y; D-Ubp-64E/TM3, Sb ryRK males. The wm4h/Y; 1/D-Ubp-64EEvar1 male offspring was compared with the wm4h/Y; 1/TM3, Sb ryRK control genotype. The enhancer effect of the mutation on a yellow variegating rearrangement was tested by scoring the color phenotype of the middle bristles of the triple row at the anterior wing margin after crosses of y1/Y; D-Ubp-64EEvar1/MKRS or y1/Y; D-Ubp-64EEvar1/TM2, Ubx ry2 males to yellow females heterozygous for one minichromosome bearing a variegating yellow1 (y1) gene [Dp (1;f)1187 (29)]. The percentage of yellow bristles in the male offspring with one wild-type copy of D-Ubp-64E was compared with that in the controls bearing two wild-type copies of D-Ubp-64E. Excisions of the P{lArB} element were induced by crossing homozygotes for the enhancer mutation with a TM3, Sb ryRK e P{ry1(D2-3)} chromosome. The dysgenic progeny was then used in two different screens. In the first one, it was crossed to the balancer stock TM6B, Tb/TM3, Sb ryRK and exceptional ry2 progeny was used to establish independent lines. These lines were tested for

* Corresponding author. Mailing address: Department of Zoology and Animal Biology, University of Geneva, 30 quai Ernest-Ansermet, CH-1211 Geneva 4, Switzerland. Phone: 41 22 702 66 66. Fax: 41 22 702 64 39. Electronic mail address: [email protected]. † Present address: Swiss Institute for Experimental Cancer Research, CH-1066 Epalinges s/Lausanne, Switzerland. 5717

5718

HENCHOZ ET AL.

MOL. CELL. BIOL.

FIG. 1. Effect on PEV of an insertional mutation at 64E. (A) Head of a fly carrying the X-chromosome rearrangement wm4h. (B) Head of a fly carrying the dominant enhancer mutation in the wm4h background (wm4h/Y; 1/1; D-Ubp-64EEvar1/TM3, Sb ryRK). (C) Disorganized area of a wm4h/Y;1/1; D-Ubp-64EEvar1/TM3, Sb ryRK eye shown by scanning electron microscopy.

precise excision of the transposon by genomic Southern blots (57) and crossed to wm4h; Cy/T(2;3)apXa Su-var(2)-101/Sb to test for reversion of the enhancer phenotype. In the second screen, the dysgenic progeny was crossed to the wm4h; Cy/T(2;3)apXa Su-var(2)101/Sb stock to directly select revertants. Exceptional revertants were crossed to the balancer stock TM6B, Tb/TM3, Sb ryRK. The ry2 progeny was used to establish independent lines. These lines were tested for precise excision of the transposon by genomic Southern blots. Molecular biology. In situ hybridizations with 3H-labeled DNA probes were done according to the method of Spierer et al. (58) on polytene chromosomes of squashed salivary glands. Plasmid rescue was obtained, after SalI digestion and ligation of genomic DNA from mutant adult flies, according to the protocol of Pirrotta (43). Screening of plasmid cDNA (4) and phage genomic (35) libraries was done according to protocols provided with the library or according to standard procedures (49), respectively. RNA in situ hybridizations with a digoxigenin-labeled cDNA probe on whole mounts (embryos and organs) were prepared as published by Tautz and Pfeifle (63) and modified by Cleard et al. (8). Northern (RNA) blot analyses were performed with the protocol of Cleard et al. (8), with some modifications. DNA probes were used in the following hybridization mixture at 428C: 23 Denhardt’s solution, 0.1% sodium dodecyl sulfate, 53 SSPE (13 SSPE is 0.18 M NaCl, 10 mM NaH2PO4, and 1 mM EDTA [pH 7.7]), 10% dextran sulfate, 50% deionized formamide, and 250 mg of denatured sonicated salmon sperm DNA per ml. All sequences were performed with the Sequenase version 2 T7 DNA polymerase sequencing kit (U.S. Biochemical), with either dGTP or dITP, or with the fmol DNA sequencing system (Promega), according to the protocols provided by manufacturers. Subclones of the cDNA were generated either by restriction enzyme digestion and ligation in the appropriate pBluescript II KS1 vector (Stratagene) as described by standard protocols (49), or by exonuclease III deletion with the Erase-a-Base system (Promega) according to the protocol provided by the manufacturer. Primers were used to complete the sequence. Germ line transformation. The plasmid used for injection was constructed by inserting the EcoRI-HindIII 4.9-kb cDNA into plasmid pHSS7, derived from pHSS6 (52), containing NotI sites at each extremity of the polylinker. NotI digestion was then used to excise the insert and to clone it downstream of the hsp70 promoter in the NotI site of the pNHT4 transformation vector (51). Injections into ry506 embryos were done according to the method of Spradling (59). ry1 transformants were selected to establish stocks by crosses to the CyO, Cy/1; TM2, Ubx ry2/MKRS, ry2 Sb stock. The P{ry1} segregation was followed. As the two transformant lines had the P{ry1} inserted on the X chromosome, a remobilization by dysgenesis was performed. P{ry1}; CyO, Cy/1; TM2, Ubx ry2/ry506 transgenic females were crossed to TM3, Sb ryRK e P{ry1(D2-3)} males, and dysgenic males were selected and crossed to CyO, Cy/1; TM2, Ubx ry2/ MKRS, ry2 Sb females. The autosomal transposition events were selected (ry1, Sb1 males; six independent events were recovered) and amplified by crosses to CyO, Cy/1; TM2, Ubx ry2/MKRS, ry2 Sb females. The P{ry1} segregation was monitored. Two of these new transgenic lines were tested for their ability to rescue the original D-Ubp-64EEvar1 phenotype and for a putative opposite (suppressor) dose effect. The two homozygous transgenic lines were crossed to wm4h; D-Ubp-64EEvar1/D-Ubp-64EEvar1 or to wm4h; 1/1 females and grown at 25 or 298C.

RESULTS A dominant enhancer mutation of PEV maps at 64E. Dominant enhancers of PEV have previously been isolated by mobilization of the P{lArB} transposon (13). Autosomal transpo-

sitions of P{lArB} were selected by monitoring the ry1 marker of P{lArB} (22). The ry1 progeny was tested for an enhancer phenotype in a variegating background (wm4h) bearing a strong suppressor [Su (var)2-101] to increase the sensitivity of the screen (see Materials and Methods and references 13 and 47 for details). Fifteen P{lArB}-induced putative enhancers of PEV [E(var)] had been isolated with this procedure (13). Here we present a genetic and molecular analysis of one of these lines, E1, which we renamed D-Ubp-64EEvar1 after molecular identification, in this report. The P{lArB} insertion of the D-Ubp-64EEvar1 line was mapped on larval salivary gland polytene chromosomes by in situ hybridization with the bacterial part of the P transposon as a probe. A unique signal was found at 64E on the left arm of the third chromosome. This localization corrects the previous assignment of the mutation to the second chromosome (13). The presence of a single insertion was confirmed by genomic Southern blot hybridization. To rigorously link the enhancer phenotype to the presence of the P transposon, a reversion test was performed. The Pinduced mutant was crossed to a transposase-producing strain, and the progeny was either analyzed for the loss of the transposon as seen by the loss of the ry1 gene and then tested for reversion or analyzed directly for reversion of the E(var) phenotype. Eleven ry2 lines from the first screen and three putative revertant lines from the second were analyzed by Southern blot hybridization. Twelve of the fourteen were internal deletions in the transposon, and two were apparent precise excisions. Nine of these 14 lines were tested more carefully by red eye pigment quantification (48). With one exception, all were revertant of the E(var) phenotype. A reduction in the transposon size appears sufficient to revert the phenotype. D-Ubp-64EEvar1 is a strong enhancer of PEV but not a regulator of homeotic genes. The enhancer effect of the mutation on PEV was detected by observation of the mottled phenotype seen in the compound eye of flies bearing a variegating rearrangement of the white gene (wm4h). Inactivation of white results in clones of white ommatidia in a red ommatidial background. Figure 1A illustrates the mottled eye phenotype of wm4h, and Fig. 1B illustrates the dominant enhancement of variegation caused by the D-Ubp-64EEvar1 mutation. The eye is almost completely white, indicating a strong E(var) effect. This result was confirmed by pigment measurements. A 5- to 10fold reduction was routinely observed in heterozygous D-Ubp64EEvar1 compared with wm4h controls. The phenotype of homozygotes for the mutation is similar to that of the

VOL. 16, 1996

DEUBIQUITINATION AND POSITION-EFFECT VARIEGATION

5719

FIG. 2. The D-Ubp-64E locus. (A) Genomic map with the position of the P{lArB} insertion. The PL3 polylinker used for the plasmid rescue is indicated. PS5 (PstI-SalI restriction fragment) and BB3.2 (BamHI restriction fragment) indicate the probes used for the screening of a genomic library. The BB3.2 probe was also used for the screening of a cDNA library and on Northern blots. (B) Approximate position of the longest cDNA. Except for the first exon and intron that were mapped by sequencing, the position of other exons was determined by Southern hybridization. They are drawn in the middle of positive restriction fragments, and their length reflects the relative intensity of hybridization. E, EcoRI; H, HindIII; S, SalI; B, BamHI.

heterozygotes. In addition, both in heterozygotes and in homozygotes, regions with a disorganization of the ommatidia are sometimes observed within white areas (Fig. 1C). We interpret this phenotype as the inactivation of the roughest locus, which is located more distally than white from the euchromaticheterochromatic junction of the wm4h rearrangement (44). The D-Ubp-64EEvar1 mutation seems able to induce an extensive spreading of centromeric heterochromatin. The effect is therefore not specific to the white gene, but more general on PEV. The D-Ubp-64EEvar1 mutation enhances variegation in other tissues such as the testes (not shown). Wild-type testes expressing the white gene have a yellow color, and wm4h testes show patches of colorless tissues. Testes of D-Ubp-64EEvar1 heterozygotes are colorless, indicating a full inactivation of the white gene. This colorless phenotype can be partially suppressed. When a strong suppressor [Su (var)2-101 (47)] mutation was placed in trans to D-Ubp-64EEvar1, rare and small yellow patches were visible. To demonstrate that the mutation affects the variegation of other rearrangements, the D-Ubp-64EEvar1 mutation was tested in trans with a minichromosome bearing a variegating yellow gene [Dp (1;f)1187 (29)]. The color phenotype was scored on the middle bristles of the triple row at the anterior wing margin. The percentage of yellow bristles was significantly higher in heterozygous flies compared with the siblings bearing two wild-type copies of D-Ubp-64E. In two independent crosses, 36.4 and 20.9% (mean values of 16 to 34 wings) of the bristles were yellow in heterozygous flies compared with 15.2 and 9.2% in control flies (t test; P , 0.001). We conclude that a reduced dose of D-Ubp-64E acts as an enhancer of PEV of different genes (white, roughest, and yellow) in different rearrangements [wm4h and Dp (1;f)1187]. Functional overlap between modifiers of PEV and regulators of homeotic genes has been reported. Suppression of PEV is found for some repressors of the Polycomb group of genes (18, 36, 41). Furthermore, some trithorax-group genes, which are general activators of homeotic genes, show dominant enhancement of PEV (17, 60), and at least one E(var), E(var)393D, was shown to be a positive regulator of the bithorax complex (12). The D-Ubp-64EEvar1 mutation was tested for genetic interactions with two regulators of homeotic genes [Pc23 and Df(3R)redP52]. The phenotypes already visible in flies heterozygous for those mutations were not affected (amplified or suppressed) in the double heterozygotes. A paternal effect, that is, the transmission of the enhancer effect by the males during several generations after removal of

the modifier mutation, has been proposed for D-Ubp-64EEvar1 (13). This paternal effect was shown by crossing wm4h females to wm4h/Y; D-Ubp-64EEvar1/Balancer males. The wm4h/Y; 1/Balancer male offspring showed an enhancer phenotype, although the enhancer mutation was not present. Such an effect had not been found in the reciprocal cross, in which the enhancer mutation was present in the female parent. This has been interpreted as an imprinting of the Y chromosome by the enhancer mutation. Other interpretations, such as the effect of a second enhancer mutation located on the Y chromosome of the D-Ubp-64EEvar1 strain, could not be excluded. To discriminate between an imprinting and a second-site mutation brought by the Y chromosome, a different test was used. To try to induce the imprinting of new Y chromosomes, wm4h; D-Ubp64EEvar1 females were crossed to males bearing a marked third chromosome. The resulting males (wm4h/Y; D-Ubp64EEvar1/marked third chromosome) were then mated to wm4h females, and variegation was compared in the males bearing either the marked third chromosome or the D-Ubp-64EEvar1 mutation. No evidence of imprinting was found in eight independent experiments (not shown). We do not know whether the paternal effect previously described was due to a modification of the Y chromosome before or during the mutagenesis

FIG. 3. Transcripts in the vicinity of the P-element insertion. Different probes were used on Northern blots (total embryonic RNA): E6.5 is an EcoRI restriction fragment of 6.5 kb, B3.7 is a BamHI restriction fragment of 3.7 kb, B3.2 is the BB3.2 probe described for Fig. 2, and B5.2 is a BamHI restriction fragment of 5.2 kb. One transcription unit on each side of the P transposon was detected. The smaller and rarer transcript, detected by the E6.5 probe after 13 days of exposure, is located at least 2.3 kb away from the insertion point. The 5.0-kb transcript strongly detected with the B3.2 probe after 11 h of exposure is located at 2.0 kb of the P{lArB} insertion point. E, EcoRI; H, HindIII; B, BamHI; S, SalI.

5720

HENCHOZ ET AL.

MOL. CELL. BIOL.

FIG. 4. Pattern of expression of the D-Ubp-64E transcript in embryos detected with the 4.9-kb cDNA as a probe. Left panels, wild-type (Canton S) embryos. Right panels, D-Ubp-64EEvar1 homozygous mutant embryos. Hybridizations were realized in parallel, and staining times are the same for wild-type and mutant embryos. (A and D) Stage 5; (B and E) stage 10; (C and F) stage 14. See text for description of patterns.

procedure or was acquired over several generations in the presence of the enhancer mutation either by true imprinting or by selection of a second-site modification. Cloning of the D-Ubp-64EEvar1 locus. Genomic DNA fragments flanking the P{lArB} insertion can be isolated by plasmid rescue (43). An 18-kb SalI fragment was cloned and found to hybridize to 64E on wild-type polytene chromosomes, thus demonstrating that the genomic sequences isolated originated from the appropriate region. With genomic fragments of the plasmid rescue as probes (PS5 and BB3.2 in Fig. 2), a phage genomic library (35) was screened to isolate sequences on both sides of the P{lArB} insertion site. The genomic region encompassing the 39 end of the transcription unit was isolated in a second step by using the 39 end of a cDNA clone (described below) as a probe. The map of the genomic region and the

position of the transposon are depicted in Fig. 2. The entire genomic walk covers about 60 kb. Transcription units on both sides of P{lArB}. To identify the transcription units on both sides of the P{lArB} insertion site, different restriction fragments were used as probes on Northern blots of total embryonic RNA (Fig. 3). A rare 2.9-kb transcript and an abundant RNA of 5.0 kb were detected with genomic probes E6.5 and BB3.2, on the 59 and the 39 ends of the P{lArB} insertion, respectively (Fig. 3). Developmental Northern blot of total embryonic and adult RNA showed that the 5.0-kb transcript is present throughout development (not shown). The BB3.2 probe was used to select cDNA clones from a 4- to 8-h cDNA library (4). About 50,000 colonies were screened, and nine clones were isolated. A 4.9-kb cDNA clone was analyzed further. Primer extension showed that about 0.15

VOL. 16, 1996


FIG. 5. Silencing depends on the dose of wild-type D-Ubp-64E. Optical density pigment measurements were at 480 nm (48). The heat shock condition was constant growth at 298C. Average pigment concentration (three measurements, five heads by measurement) at 258C is indicated by lightly dotted columns, and average pigment concentration at 298C is indicated by more heavily dotted columns. In the tests, standard deviations were less than 20%. 1 1 0, males with one wild-type dose of D-Ubp-64E in a wm4h background (wm4h/Y; CyO/1; D-Ubp-64EEvar1/1). 2 1 0, males with two wild-type copies of D-Ubp-64E (wm4h/Y; CyO/1; 1/1). 1 1 1, males with one wild-type dose of D-Ubp-64E and one copy of the heat-inducible transgene (wm4h/Y; T{hsD-Ubp-64E}MB28F1-4/ CyO; 1/D-Ubp-64EEvar1). 2 1 1, males with two wild-type doses of D-Ubp-64E and one copy of a heat-inducible transgene (wm4h/Y; T{hsD-Ubp-64E}MB28F1-4/ CyO; 1/1).

kb of the 59 untranslated region of the mRNA is missing in the 4.9-kb cDNA. The approximate position of the cDNA on the genomic walk (Fig. 2B) was determined by Southern blot hybridization. The 59 end of the 4.9-kb cDNA maps at about 2.0 kb of the P{lArB} insertion site (as determined by PCR). The missing 0.15-kb 59 untranslated region is probably included in the 3.2-kb BamHI genomic fragment located on the right side of the P element insertion site, as it is the only fragment close to the transposon that detects the 5.0-kb transcript. The direction of the transcription is outward from the P{lArB} insertion. The transcript is ubiquitous during embryonic development, and the mutation affects the transcript level. The distribution of the 5.0-kb transcript was examined in whole mounts of wild-type and mutant embryos. Results are illustrated in Fig. 4 for three selected developmental stages. The transcript is ubiquitous and strongly expressed throughout development, which confirms the results obtained by developmental Northern blots. The central nervous system is strongly stained (stages 14 to 16). In mutant embryos, the pattern of expression is similar to that of wild-type embryos, except for the amount of transcript, which is severely reduced. The reduction of transcript level in the mutant was confirmed by Northern blot hybridization on embryos and by in situ hybridization on adult testes and ovaries (not shown). Here also, only the transcript level is affected in mutant organs. The dramatic reduction of transcript abundance in mutants is a strong indication that the 5.0-kb transcript corresponds to the enhancer and that the enhancement of PEV is caused by a reduced activity of the gene. An inducible transgene rescues the D-Ubp-64EEvar1 phenotype and shows a dose-dependent suppression of PEV. The demonstration that the 5.0-kb transcript encodes the E(var) function was provided by genetic transformation. The cDNA containing the complete open reading frame was cloned in a

5721

P-transposon germ line transformation vector (pNHT4) that allowed us to put the cDNA under the control of the hsp70 heat shock promoter (51). After injection in a ry2 background, transformant lines were selected on the basis of the ry1 phenotype conferred by the marker of the transformation vector. As the two transformant lines bore an insertion on the X chromosome, a remobilization was performed and autosomal transformant lines were obtained. Rescue of the E(var) phenotype was determined by crossing the transformants to the D-Ubp-64EEvar1 mutation in a wm4h background. Transformant and control flies were grown at 258C and in a mild heat shock condition at 298C. This treatment did not affect the viability. Figure 5 shows that the presence of the transgene is able to suppress the E(var) phenotype (in a background of one resident wild-type copy of D-Ubp-64E) and to produce a suppressor effect (in a background of two resident wild-type copies of D-Ubp-64E) at 298C. Flies of the same genotypes grown at 258C without heat shock do not show these effects; the transgene seems not to be induced at this temperature. Another heat shock condition, namely, one heat shock daily at 378C for 40 min, was also tested. Results were similar to the data presented in Fig. 5. These experiments show that the 5.0-kb transcript does encode the modifier of PEV, and that its overexpression has an effect opposite from that of a loss of the gene. The deduced protein sequence corresponds to a ubiquitinspecific protease. The 4.9-kb cDNA was sequenced, allowing us to deduce an open reading frame of 898 amino acids (Fig. 6). Database searches revealed that the cDNA encodes a putative member of the family of ubiquitin-specific proteases (Ubps), also called ubiquitin C-terminal hydrolases. Members of the family are known in S. cerevisiae, D. melanogaster, and mammals (1, 27, 28). These enzymes act as ATP-independent proteases cleaving the ubiquitin polypeptide from its linear or branched conjugates. They have been divided into two major groups: one acting on small-size substrates, the UCH1 subfamily (42), and the other, the UCH2 subfamily, able to act on larger substrates (40). Assays in S. cerevisiae and D. melanogaster have revealed that enzymes of the second group share two small conserved functional domains, the Cys and His domains. The first one is disposed around a cystein residue, and the second one is disposed around two histidine residues, all thought to be part of the active site of the protease (1, 28, 40). Similar domains were found in the D-Ubp-64E protein sequence. Another motif of high similarity, present between the Cys and the His domain, contains an aspartic acid (1, 16, 40, 67). The general structure of D-Ubp-64E and selected Ubps is shown in Fig. 7A, and the Cys, His, and Asp domains are aligned with some homologs in Fig. 7B. The greatest similarity outside of these domains was found with a partial genomic sequence from the Caenorhabditis elegans genome project (accession number gp/Z47812/CET05H10_1). Mutations of another Ubp and of a proteasome component do not affect PEV. Another Drosophila UCH2 protein encoded by the fat facet gene (faf) has recently been described (19, 28). The faf gene is known to be required for eye and oocyte development (19). To determine if other Ubps have the same dose effect on PEV as does D-Ubp-64E, we have tested different faf alleles (fafBX4; fafFO8; fafBX3; fafFBB12; fafBP), including a putative null (fafBX4), on the wm4h rearrangement. No significant alteration of the variegated eye phenotype was observed. Dose-dependent modification of PEV is therefore not a general effect of ubiquitin-specific proteases in flies. Evidence that faf is involved in the regulation of the degradation of one or several proteins was provided by the observation that a mutation of a proteasome component acts as a dominant suppressor of faf (28). We therefore tested double heterozy-

5722

HENCHOZ ET AL.

MOL. CELL. BIOL.

FIG. 6. Nucleotide sequence of the cDNA encoding D-Ubp-64E. Single letters are used for the deduced amino acids of the long open reading frame.

gotes of D-Ubp-64EEvar1 and the proteasome mutation (50). The effect on wm4h variegation was not different from that of D-Ubp-64EEvar1 alone. DISCUSSION The dose of the D-Ubp-64E locus acts as a modifier of PEV in D. melanogaster. When one or two gene doses are lost, heterochromatic inactivation (silencing) is enhanced. On the other hand, overexpression of the corresponding cDNA from the heat-inducible hsp70 promoter has an opposite suppressor effect. Hence, this locus is a haplo-enhancer, “triplo”-suppressor gene. D-Ubp-64EEvar1 is, with E-var(3)-93E, which encodes the dE2F transcriptional regulator (54), the second molecularly characterized enhancer of PEV with such an opposite dose dependence effect. Besides the enhancer effect on the wm4h rearrangement, which affects both the white and the roughest loci, we have also observed an enhancement of silencing on another rearrangement variegating for the yellow gene, confirming that the effect is not gene specific. The D-Ubp-64E gene encodes a new member of the ubiquitin-specific proteases (Ubps) or ubiquitin C-terminal hydrolase family. Ubps have been isolated in different organisms such as S. cerevisiae, D. melanogaster, and mammals (1, 19, 27, 28) and classified into two major groups. The UCH1 subfamily

acts mainly on substrates of small size (42, 70), and the UCH2 subfamily can work on larger proteins (40). Enzymes belonging to this last group share two conserved catalytic sites, the Cys and His domains, containing a cystein residue and two histidine residues, respectively (1, 28, 40). All three residues are required for enzymatic activity. Although we have not demonstrated in vitro a deubiquitination activity, the excellent conservation of D-Ubp-64E in the Cys and His domains, as well as other regions, leaves little doubt that it belongs to the UCH2 subfamily. Ubiquitin is a small, 76-amino-acid protein, conserved among all eucaryotes (21, 66), which can be covalently linked via its terminal carboxyl group to the lysine ε-amino group of many target proteins. Ubiquitination, which is mediated by the action of successive ubiquitin-activating (E1), ubiquitin-conjugating (E2), and optional ubiquitin-ligase (E3) enzymes (27), provides different functions depending on the substrates. For a variety of cytoplasmic proteins, such as the cyclins that regulate cell cycle progression, polyubiquitination seems to be necessary prior to their degradation by a specific protein degradation complex called the proteasome (53). For other proteins, ubiquitination is a maturation signal (39) or a transient posttranslational modification, as for histones H2A and H2B (5, 10, 32). In the latter, ubiquitination appears associated with an increase in gene transcription by RNA polymerase II and a

VOL. 16, 1996


5723

FIG. 7. Alignment of D-Ubp-64E with some Ubps. (A) General structure of D-Ubp-64E and some other Ubps. The Cys domain is indicated by a vertical line, the Asp domain is indicated by a star, and the His domain is indicated by a double vertical line. (B) Alignment of the conserved domains. The Cys domain is centered on a unique cysteine (C) residue; the Asp domain, also called QQD (67), contains an aspartic acid (D); and the His domain is disposed around two histidine residues (H). The Cys and the His domains are part of the catalytic site of the protease (1, 28, 40). No function has been attributed yet to the Asp domain. Darkly shaded amino acids indicate matches with D-Ubp-64E and all the Ubps selected. Lightly shaded amino acids indicate at least one match with D-Ubp-64E. Because of the small size of the conserved domains and the lack of homology outside them, it is not possible to place D-Ubp-64E in a subclass with any of the already-known Ubps. The Ubps here include Ubp2 (1), Ubp5 (68), Tre2 (38), Ubpx-H (accession number: sp/P40818/UBCX-HUMAN), and faf (19).

repression of genes transcribed by RNA polymerase I (5, 9). Finally, ubiquitin seems to be a component of some membrane proteins and could be a specific signal for major histocompatibility complex class I antigen presentation and for the degradation of receptor-ligand complexes after their internalization (37, 55, 69). Except for the transient modification of histones, it is not clear why some proteins that have undergone ubiquitination are later deubiquitinated by the Ubps. One possibility is that the correct level of ubiquitination must be precisely regulated and that this requires the balanced action of the two counteracting enzymatic reactions. An obvious possibility is that deubiquitination helps to retrieve proteins incorrectly tagged for proteasome-dependent degradation. D-Ubp-64E and PEV. In the widely adopted model of Locke

et al. (34), dose-dependent modifiers of PEV were considered as protein components of multiprotein structural components of chromatin, while haplo modifiers without the triplo effect could encode enzymes modifying chromatin. It is hence surprising that an enhancer locus with a triplo suppressor effect on PEV is a ubiquitin protease. In the case of PEV, and because of the enhancer phenotype of the loss-of-function allele D-Ubp-64EEvar1, the substrate should be a positive regulator of chromatin decondensation, namely, another haplo-insufficient enhancer of PEV. Reduced deubiquitination would lead to excessive degradation of this other E(var) protein and thus to increased silencing. Overexpression of D-Ubp-64E would produce the opposite effect of lesser degradation and decreased silencing. Alternatively, it is possible that ubiquitination is required for the maturation or the assembly of a constituent of

5724

HENCHOZ ET AL.

heterochromatin, a suppressor of PEV, and that the role of D-Ubp-64E is to limit the amount of heterochromatin. Reduced expression of the deubiquitinating enzyme would increase the amount of heterochromatin, again enhancing silencing, and increased expression would have the opposite effect. In each case, there must be a rate-limiting component among the target(s) of D-Ubp-64E. The enhancer phenotype is the only effect of the mutation that we have observed, and the strong loss-of-function D-Ubp64EEvar1 allele does not affect viability. D-Ubp-64E is therefore unlikely to have a general function in the ubiquitin-dependent protein degradation pathway, such as the generation of ubiquitin monomers or the clearance of degradation products from the proteasome (40). This is supported by our inability to detect interaction with a mutation in the proteasome pathway. The factor(s) regulated by D-Ubp-64E could also be redundant or specific for chromatin decondensation of only a few genes. It is known from previous studies that variegating genes are not equally sensitive to modifiers of PEV (3, 23). The D-Ubp-64EEvar1 effect was tested on different rearrangements. Three genes, yellow, white, and roughest, in two rearrangements were found to be sensitive, while two other rearrangements, bearing either yellow or brown, were not modified (unpublished results). This observation argues in favor of a certain specificity of the D-Ubp-64E targets, which may be implicated in the regulation of only a subset of the factors affecting chromatin compaction. Because of the many roles played by ubiquitin in the cell, it is highly speculative to further attempt to define a precise role of D-Ubp-64E in the silencing of PEV. Two potential targets are worth citing. First, regulation of chromatin remodelling could be achieved by posttranslational modifications and/or changes in the abundance of some of its constituents (see reference 65). Ubiquitinated histones H2A (uH2A) and H2B (uH2B) seem to be associated with an increase in gene transcription by RNA polymerase II and with a repression of genes transcribed by RNA polymerase I (5, 9). Cleavage of ubiquitin from uH2A and uH2B is a very late event in chromosome condensation into metaphase chromosomes, and ubiquitination is an early event in their decondensation. Ubiquitin conjugation probably helps in relaxing the nucleosome structure and so increases the accessibility of DNA to some transcription factors. A modifier of PEV could be a good candidate for the regulation of H2A-H2B ubiquitination of genes transcribed by RNA polymerase II. An enhancer effect would be expected to result from a too-high level of deubiquitinated H2A-H2B. This alteration could be produced by mutations in regulatory enzymes, either a loss of function of a ubiquitinating enzyme, or a gain of function of a ubiquitin-specific protease. We expected that such a mutation would be recessive lethal. Because D-Ubp-64EEvar1 is a strong loss of function, enhances silencing, and is viable, we think that histones H2A and H2B are unlikely substrates. An alternative to the hypothesis of direct control of a chromatin component is to postulate an indirect effect on the cell cycle, as ubiquitination plays a crucial role in cell cycle regulation (6, 20). For example, a disturbance of the S/G2 transition, and a prolonged late S phase, during which heterochromatic DNA is replicated (11), could give more time to form more heterochromatin, resulting in an enhancer phenotype in a sensitive variegating background. A strong link exists among DNA replication, cell cycle, and silencing in S. cerevisiae and in D. melanogaster. In S. cerevisiae, genetic or chemical disturbances of normal cell cycle progression enhance the establishment of telomeric transcriptional silencing (31). In D. melanogaster, mutations in the suppressor of PEV Su-var(3)6, which

MOL. CELL. BIOL.

encodes a type 1 serine-threonine protein phosphatase (PP1), also affect mitosis (2), and mutations in mus209, which encodes the proliferating cell nuclear antigen, an indispensable component of the DNA replication apparatus, suppress PEV (24). Discrimination between a direct and an indirect effect of D-Ubp-64E on chromatin structure will require the identification of its targets and the analysis of PEV modifications in double mutants. Interestingly, D. Moazed and A. S. Johnson (35a) have determined that a deletion of the Ubp3 gene of S. cerevisiae enhances variegation and that the UBP3 protein interacts with the SIR4 protein. ACKNOWLEDGMENTS We warmly thank G. Reuter for providing us with the D-Ubp-64EEvar1 line before publication and for many stimulating discussions and collaboration in the early part of the work. We acknowledge A. Spierer for determining the chromosome position of the transposon and for help with microinjections of embryos, J. Wu ¨est from the Museum of Natural History of Geneva for the scanning electron microscopy micrograph, and C.-H. Tonka for technical assistance. We also thank J. Fisher-Vize for giving us the faf alleles, J. Belote for the proteasome mutation, and K. Wilkinson and L. Falquet for sharing unpublished results. This work was supported by grants from the Swiss National Science Foundation to P.S. and D.P. and by the Socie´te´ Acade´mique de Gene`ve and the canton of Geneva. REFERENCES 1. Baker, R. T., J. W. Tobias, and A. Varshavsky. 1992. Ubiquitin-specific proteases of Saccharomyces cerevisiae. Cloning of UBP2 and UBP3, and functional analysis of the UBP gene family. J. Biol. Chem. 267:23364–23375. 2. Baksa, K., H. Morawietz, V. Dombradi, M. Axton, H. Taubert, G. Szabo, I. Torok, A. Udvardy, H. Gyurkovics, B. Szoor, D. Glover, G. Reuter, and J. Gausz. 1993. Mutations in the protein phosphatase 1 gene at 87B can differentially affect suppression of position-effect variegation and mitosis in Drosophila melanogaster. Genetics 135:117–125. 3. Bishop, C. P. 1992. Evidence for intrinsic differences in the formation of chromatin domains in Drosophila melanogaster. Genetics 132:1063–1069. 4. Brown, N. H., and F. C. Kafatos. 1988. Functional cDNA libraries from Drosophila embryos. J. Mol. Biol. 203:425–437. 5. Busch, H. 1984. Ubiquitination of proteins. Methods Enzymol. 106:238–262. 6. Ciechanover, A. 1994. The ubiquitin-proteasome proteolytic pathway. Cell 79:13–21. 7. Cleard, F. 1993. Ph.D. thesis. University of Geneva, Geneva, Switzerland. 8. Cleard, F., M. Matsarskaia, and P. Spierer. 1995. The modifier of positioneffect variegation Suvar(3)7 of Drosophila: there are two alternative transcripts and seven scattered zinc fingers, each preceded by a tryptophan box. Nucleic Acids Res. 23:796–802. 9. Davie, J. R., and L. C. Murphy. 1994. Inhibition of transcription selectively reduces the level of ubiquitinated histone H2B in chromatin. Biochem. Biophys. Res. Commun. 203:344–350. 10. Davies, N., and G. G. Lindsey. 1994. Histone H2B (and H2A) ubiquitination allows normal histone octamer and core particle reconstitution. Biochim. Biophys. Acta 1218:187–193. 11. Dolfini, S. 1971. Karyotype polymorphism in a cell population of Drosophila melanogaster cultured in vitro. Chromosoma 33:196–208. 12. Dorn, R., V. Krauss, G. Reuter, and H. Saumweber. 1993. The enhancer of position-effect variegation of Drosophila, E(var)3-93D, codes for a chromatin protein containing a conserved domain common to several transcriptional regulators. Proc. Natl. Acad. Sci. USA 90:11376–11380. 13. Dorn, R., J. Szidonya, G. Korge, M. Sehnert, H. Taubert, E. Archoukieh, B. Tschiersch, H. Morawietz, G. Wustmann, G. Hoffmann, and G. Reuter. 1993. P transposon-induced dominant enhancer mutations of position-effect variegation in Drosophila melanogaster. Genetics 133:279–290. 14. Eissenberg, J. C. 1989. Position effect variegation in Drosophila: towards a genetics of chromatin assembly. Bioessays 11:14–17. 15. Eissenberg, J. C., T. C. James, D. M. Foster-Hartnett, T. Hartnett, V. Ngan, and S. C. Elgin. 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. 16. Falquet, L., N. Paquet, S. Frutiger, G. J. Hughes, K. Hoang-Van, and J. C. Jaton. 1995. cDNA cloning of a human 100 kDa de-ubiquitinating enzyme: the 100 kDa human de-ubiquitinase belongs to the ubiquitin C-terminal hydrolase family 2 (UCH2). FEBS Lett. 376:233–237. 17. Farkas, G., J. Gausz, M. Galloni, G. Reuter, H. Gyurkovics, and F. Karch. 1994. The Trithorax-like gene encodes the Drosophila GAGA factor. Nature (London) 371:806–808.

VOL. 16, 1996


18. 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. 19. Fischer-Vize, J. A., G. M. Rubin, and R. Lehmann. 1992. The fat facets gene is required for Drosophila eye and embryo development. Development 116: 985–1000. 20. Glotzer, M. 1995. Cell cycle. The only way out of mitosis. Curr. Biol. 5:970– 972. 21. Goldstein, G., M. Scheid, U. Hammerling, D. H. Schlesinger, H. D. Niall, and E. A. Boyse. 1975. Isolation of a polypeptide that has lymphocytedifferentiating properties and is probably represented universally in living cells. Proc. Natl. Acad. Sci. USA 72:11–15. 22. Grossniklaus, U., H. J. Bellen, C. Wilson, and W. J. Gehring. 1989. Pelement-mediated enhancer detection applied to the study of oogenesis in Drosophila. Development 107:189–200. 23. Hayashi, S., A. Ruddell, D. Sinclair, and T. Grigliatti. 1990. Chromosomal structure is altered by mutations that suppress or enhance position effect variegation. Chromosoma 99:391–400. 24. Henderson, D. S., S. S. Banga, T. A. Grigliatti, and J. B. Boyd. 1994. Mutagen sensitivity and suppression of position-effect variegation result from mutations in mus209, the Drosophila gene encoding PCNA. EMBO J. 13:1450–1459. 25. Hendrich, B. D., and H. F. Willard. 1995. Epigenetic regulation of gene expression: the effect of altered chromatin structure from yeast to mammals. Hum. Mol. Genet. 4:1765–1777. 26. Henikoff, S. 1990. Position-effect variegation after 60 years. Trends Genet. 6:422–426. 27. Hochstrasser, M. 1995. Ubiquitin, proteasomes, and the regulation of intracellular protein degradation. Curr. Opin. Cell Biol. 7:215–223. 28. Huang, Y., R. T. Baker, and J. A. Fischer-Vize. 1995. Control of cell fate by a deubiquitinating enzyme encoded by the fat facets gene. Science 270:1828– 1831. 29. Karpen, G. H., and A. C. Spradling. 1990. Reduced DNA polytenization of a minichromosome region undergoing position-effect variegation in Drosophila. Cell 63:97–107. 30. Kornberg, R. D., and Y. Lorch. 1995. Interplay between chromatin structure and transcription. Curr. Opin. Cell Biol. 7:371–375. 31. Laman, H., D. Balderes, and D. Shore. 1995. Disturbance of normal cell cycle progression enhances the establishment of transcriptional silencing in Saccharomyces cerevisiae. Mol. Cell. Biol. 15:3608–3617. 32. Levinger, L., and A. Varshavsky. 1982. Selective arrangement of ubiquitinated and D1 protein-containing nucleosomes within the Drosophila genome. Cell 28:375–385. 33. Lindsley, D. L., and G. G. Zimm. 1992. The genome of Drosophila melanogaster. Academic Press, Inc., New York. 34. Locke, J., M. A. Kotarski, and K. D. Tartof. 1988. Dosage-dependent modifiers of position effect variegation in Drosophila and a mass action model that explains their effect. Genetics 120:181–198. 35. Maniatis, T., R. C. Hardison, E. Lacy, J. Lauer, C. O’Connell, D. Quon, G. K. Sim, and A. Efstratiadis. 1978. The isolation of structural genes from libraries of eucaryotic DNA. Cell 15:687–701. 35a.Moazed, D., and A. S. Johnson. Personal communication. 36. Moehrle, A., and R. Paro. 1994. Spreading the silence: epigenetic transcriptional regulation during Drosophila development. Dev. Genet. 15:478–484. 37. Monaco, J., and D. Nandi. 1995. The genetics of proteasomes and antigen processing. Annu. Rev. Genet. 29:729–754. 38. Nakamura, T., J. Hillova, R. Mariage-Samson, M. Onno, K. Huebner, L. A. Cannizzaro, L. Boghosian-Sell, C. M. Croce, and M. Hill. 1992. A novel transcriptional unit of the tre oncogene widely expressed in human cancer cells. Oncogene 7:733–741. 39. Palombella, V. J., O. J. Rando, A. L. Goldberg, and T. Maniatis. 1994. The ubiquitin-proteasome pathway is required for processing the NF-kappa B1 precursor protein and the activation of NF-kappa B. Cell 78:773–785. 40. Papa, F. R., and M. Hochstrasser. 1993. The yeast DOA4 gene encodes a deubiquitinating enzyme related to a product of the human tre-2 oncogene. Nature (London) 366:313–319. 41. Paro, R., and D. S. Hogness. 1991. The Polycomb protein shares a homologous domain with a heterochromatin-associated protein of Drosophila. Proc. Natl. Acad. Sci. USA 88:263–267. 42. Pickart, C. M., and I. A. Rose. 1986. Mechanism of ubiquitin carboxylterminal hydrolase. Borohydride and hydroxylamine inactivate in the presence of ubiquitin. J. Biol. Chem. 261:10210–10217. 43. Pirrotta, V. 1986. Cloning Drosophila genes, p. 83–109. In D. B. Roberts (ed.), Drosophila, a practical approach. IRL Press, Oxford. 44. Ramos, R. G., G. L. Igloi, B. Lichte, U. Baumann, D. Maier, T. Schneider, J. H. Brandstatter, A. Frohlich, and K. F. Fischbach. 1993. The irregular chiasm C-roughest locus of Drosophila, which affects axonal projections and programmed cell death, encodes a novel immunoglobulin-like protein. Genes Dev. 7:2533–2547. 45. Reuter, G., M. Giarre, J. Farah, J. Gausz, A. Spierer, and P. Spierer. 1990.

46. 47. 48. 49. 50. 51. 52. 53. 54.

55.

56. 57. 58. 59. 60.

61. 62. 63.

64.

65. 66. 67.

68. 69.

70.

5725

Dependence of position-effect variegation in Drosophila on dose of a gene encoding an unusual zinc-finger protein. Nature (London) 344:219–223. Reuter, G., and P. Spierer. 1992. Position effect variegation and chromatin proteins. Bioessays 14:605–612. Reuter, G., W. Werner, and H. J. Hoffmann. 1982. Mutants affecting position-effect heterochromatinization in Drosophila melanogaster. Chromosoma 85:539–551. Reuter, G., and I. Wolff. 1981. Isolation of dominant suppressor mutations for position-effect variegation in Drosophila melanogaster. Mol. Gen. Genet. 182:516–519. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. Saville, K. J., and J. M. Belote. 1993. Identification of an essential gene, I(3)73Ai, with a dominant temperature-sensitive lethal allele, encoding a Drosophila proteasome subunit. Proc. Natl. Acad. Sci. USA 90:8842–8846. Schneuwly, S., A. Kuriowa, P. Baumgartner, and W. J. Gehring. 1986. Structural organization and sequence of the homeotic gene Antennapedia of Drosophila melanogaster. EMBO J. 4:733–739. Seifert, H. S., E. Y. Chen, M. So, and F. Heffron. 1986. Shuttle mutagenesis: a method of transposon mutagenesis for Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 83:735–739. Seufert, W., B. Futcher, and S. Jentsch. 1995. Role of a ubiquitin-conjugating enzyme in degradation of S- and M-phase cyclins. Nature (London) 373:78–81. Seum, C., A. Spierer, D. Pauli, J. Szidonya, G. Reuter, and P. Spierer. 1996. Position-effect variegation in Drosophila depends on dose of the gene encoding the E2F transcriptional activator and cell cycle regulator. Development 122:1949–1956. Siegelman, M., M. W. Bond, W. M. Gallatin, T. St. John, H. T. Smith, V. A. Fried, and I. L. Weissman. 1986. Cell surface molecule associated with lymphocyte homing is a ubiquitinated branched-chain glycoprotein. Science 231:823–829. Soeller, W. C., C. E. Oh, and T. B. Kornberg. 1993. Isolation of cDNAs encoding the Drosophila GAGA transcription factor. Mol. Cell. Biol. 13: 7961–7970. Southern, E. M. 1975. Detection of sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98:503–517. Spierer, P., A. Spierer, W. Bender, and D. S. Hogness. 1983. Molecular mapping of genetic and chromomeric units in Drosophila melanogaster. J. Mol. Biol. 168:35–50. Spradling, A. C. 1986. P element-mediated transformation, p. 175–196. In D. B. Roberts (ed.), Drosophila, a practical approach. IRL Press, Oxford. Tamkun, J. W., R. Deuring, M. P. Scott, M. Kissinger, A. M. Pattatucci, T. C. Kaufman, and J. A. Kennison. 1992. brahma: a regulator of Drosophila homeotic genes structurally related to the yeast transcriptional activator SNF2/SW12. Cell 68:561–572. Tartof, K. D., C. Bishop, M. Jones, C. A. Hobbs, and J. Locke. 1989. Towards an understanding of position effect variegation. Dev. Genet. 10:162–176. Tartof, K. D., C. Hobbs, and M. Jones. 1984. A structural basis for variegating position effects. Cell 37:869–878. Tautz, D., and C. Pfeifle. 1989. A non-radioactive in situ hybridization method for the localization of specific RNAs in Drosophila embryos reveals translational control of the segmentation gene hunchback. Chromosoma 98:81–85. Tschiersch, B., A. Hofmann, V. Krauss, R. Dorn, G. Korge, and G. Reuter. 1994. The protein encoded by the Drosophila position-effect variegation suppressor gene Su(var)3-9 combines domains of antagonistic regulators of homeotic gene complexes. EMBO J. 13:3822–3831. Tsukiyama, T., C. Daniel, J. Tamkun, and C. Wu. 1995. ISWI, a member of the SWI2/SNF2 ATPase family, encodes the 140 kDa subunit of the nucleosome remodeling factor. Cell 83:1021–1026. Vijay-Kumar, S., C. E. Bugg, K. D. Wilkinson, R. D. Vierstra, P. M. Hatfield, and W. J. Cook. 1987. Comparison of the three-dimensional structures of human, yeast, and oat ubiquitin. J. Biol. Chem. 262:6396–6399. Wilkinson, K. D., V. L. Tashayev, L. B. O’Connor, C. N. Larsen, E. Kasperek, and C. M. Pickart. 1995. Metabolism of the polyubiquitin degradation signal: structure, mechanism, and role of isopeptidase T. Biochemistry 34: 14535–14546. Xiao, W., T. Fontanie, and M. Tang. 1994. UBP5 encodes a putative yeast ubiquitin-specific protease that is related to the human Tre-2 oncogene product. Yeast 10:1497–1502. Yarden, Y., J. A. Escobedo, W. J. Kuang, T. L. Yang-Feng, T. O. Daniel, P. M. Tremble, E. Y. Chen, M. E. Ando, R. N. Harkins, U. Francke, V. A. Fried, A. Ullrich, and L. T. Williams. 1986. Structure of the receptor for plateletderived growth factor helps define a family of closely related growth factor receptors. Nature (London) 323:226–232. Zhang, N., K. Wilkinson, and M. Bownes. 1993. Cloning and analysis of expression of a ubiquitin carboxyl terminal hydrolase expressed during oogenesis in Drosophila melanogaster. Dev. Biol. 157:214–223.