Copyright 0 1993 by the Genetics Society of America

P Transposon-Induced Dominant Enhancer Mutations of Position-Effect Variegation in Drosophila melanogaster

Rainer Dorn,* Janos Szidonya?’ Gunter Korge? Madeleine Sehnert,* Helge Taubert,* Essmail Archoukieh,* Bettina Tschiersch,* Henning Morawietz,* Gerold Wustmann,*’*

Gyula Hoffmannt and Gunter Reuter*” *Znstitutjiir Genetik, Martin-Luther-Universitat, 0-0-4020 HallelS., Germany, +Znstitute o f Genetics, Biological Research Center .f

the Hungarian Academy ofsciences, H-6701 Szeged, Hungary, and SZnstitut f u r Genetik, Freie Universitrit Berlin, 0-1000 Berlin 33, Germany

Manuscript received May 18, 1992 Accepted for publication October 7, 1992

ABSTRACT P transposon induced modifier mutations of position-effect variegation (PEV) were isolated with

the help of hybrid dysgenic crosses (7r2 strain) and after transposition of the mutator elements pUChsneory+ and P[IArB]. Enhancer mutations were found with a ten times higher frequency than suppressors. The 19 pUChsneory+- and 15 P[lArB]-induced enhancer mutations can be used for cloning of genomic sequences at the insertion sites of the mutator elements via plasmid rescue. Together with a large sample of X-ray-induced (48) and spontaneous (93) enhancer mutations a basic genetic analysis of this group of modifier genes was performed. On the basis of complementation and mapping data we estimate the number of enhancer genes at about 30 in the third chromosome and between 50 and 60 for the whole autosome complement. Therefore, enhancer of PEV loci are found in the Drosophila genome as frequently as suppressor genes. Many of the enhancer mutations display paternal effects consistent with the hypothesis that some of these mutations can induce genomic imprinting. First studies on the developmentally regulated gene expression of PEV enhancer genes were performed by &plactosidase staining in P[IArB] induced mutations.

I N recent years it has become apparent that, at a primary level, gene activity is regulated by re- gional changes in chromatin structure. In order to identify genes involved in the regulation of changes in chromatin structure several groups have employed the phenomenon of position-effect variegation (PEV) and screened for dominant modifier mutations of PEV (REUTER and WOLFF 1981 ; SINCLAIR, MOTTUS and GRIGLIATTI 1983; LOCKE, KOTARSKI and TARTOF 1988; WUSTMANN et al. 1989). Because in PEV gene inactivation is caused by a change in chromatin con- densation (heterochromatinization) the dominant sup- pressor and enhancer mutations isolated were sup- posed to identify loci that might encode structural or regulatory chromatin components. Cloning and se- quencing of several of these genes appears to confirm this hypothesis (JAMES and ELGIN 1986; REUTER et al. 1990). Therefore, further genetic studies of dominant modifier mutations of PEV were performed to dissect and analyze the complex genetic basis of chromatin assembly as well as its functional implication in regu- lation of gene activities or epigenetic developmental programs.

’ Present address: University of Horticulture and Food Industry, Depart- ment of Plant Genetics and Selection, Menesi str. 44, H-1118 Budapest, Hungary.

burg. Germany. P.O. Box 68, Germany. * Present address: Institut fur Neurobiologie und Hirnforschung, Magde- ’ Corresponding author.

Genetics 133: 279-290 (February, 1992)

Cytogenetic studies revealed the existence of both dominant suppressors and enhancers of PEV (REUTER and WOLFF 198 1 ; SINCLAIR, MOTTUS and GRICLIATTI 1983; REUTER et al. 1986; SINCLAIR, LLOYD and GRIG- LIATTI 1989). In these studies the total number of PEV modifier genes in the Drosophila genome was estimated to be about 100-1 20 (HENIKOFF 1979; WUSTMANN et al. 1989). By studying their dosage- dependent effects four classes of genes have been distinguished: in one class of genes a deficiency causes suppression (haplo-suppressors) and in another group of loci a deficiency displays an enhancer effect (haplo- enhancers). For some genes of both classes a duplica- tion was found to result in an opposite triplo-effect (haplo-suppressors with a triplo-enhancer effect and haplo-enhancers with a triplo-suppressor effect). The majority of the modifier loci of both classes do not display triplo-effects (REUTER and SZIDONYA 1983; SZIDONYA and REUTER 1988; LOCKE, KOTARSKI and TARTOF 1988; WUSTMANN et al. 1989). These dosage- dependent effects of PEV modifier genes might reflect opposing chromatin condensation and decondensa- tion processes.

The studies of the PEV modifier genes undertaken to date were concentrated almost exclusively on sup- pressors. Three of these loci have been cloned (JAMES and ELCIN 1986; REUTER et al. 1990; K. BAKSA, H.

280 R. Dorn et al.

MORAWIETZ, M. J. AXTON, V. DOMBRADI H. TAUB- ERT, G . SZABO, I. TOROK, A. UDVARDY, H. GYURKOV- ICS, D. M. GLOVER, J. GAUSZ and G. REUTER, manu- script submitted). They encode a heterochromatin- specific protein, a zinc finger nuclear protein, and a type I phosphatase, respectively. Sequencing of these genes appears to confirm the hypothesis that such genes encode regulatory or structural components of chromatin. In addition, the mutations of another sup- pressor locus, Su-var(2)1, which were found to reduce H4 deacetylation and chromatin packaging (DORN et al. 1986), are butyrate sensitive and display a lethal interaction to extra Y heterochromatin (REUTER, DORN and HOFFMANN 1982). The mutations of two other suppressor loci show identical pleiotropic ge- netic effects (REUTER et al. 1986).

Suppressor mutations of PEV strongly inhibit het- erochromatinization of euchromatic regions in varie- gating rearrangements as shown by cytological analy- sis of salivary gland giant chromosomes (REUTER, WERNER and HOFFMANN 1982) suggesting that these loci are involved in chromatin condensation (EISSEN- BERG 1989; GRIGLIATTI 199 1). Alternatively, the re- ciprocal class of genes which mutate to enhancers of PEV (haplo-enhancer loci) might encode chromatin components involved in decondensation processes and therefore might also be involved in the control of gene activation. Although the enhancers of PEV con- stitute a functionally interesting group of loci they have not been the subject of any detailed genetic and molecular studies.

In order to address this lack of knowledge we un- dertook a series of large scale experiments to isolate enhancers of PEV after X-ray mutagenesis to recover both rearrangements and single site mutations as well as P element and modified P element transposon induced mutations which would be helpful in further molecular studies. The modified P elements used were pUChsneory+ or P[lArB] to facilitate cloning of en- hancer loci. In addition the P[IArB] element, which contains the lac 2 marker, allowed the first insight into the developmental and tissue specific expression of enhancer genes.

From the results of these studies it became apparent that the enhancers of PEV represent a complex group of genes comparable in number to the suppressor loci.

MATERIALS AND METHODS

Isolation of mutants and study of revertants: Descrip- tions of the chromosomes and mutations used in this study can be found in LINDSLEY and ZIMM (1992). The translo- cations T(2;3)apxa + In(2L)Cy, apXaCy E-var(3)l'' and T(2;3)apxa, ap.xaSu-var(2)10' were used to screen for new1 induced dommant suppressors and enhancers of In(1)w" ' PEV. The dominant enhancer effect of E-var(3)1°' as well as the dominant suppressor effect of Su-var(2)1°' are very sensitive to the effect of other modifier mutations (Reuter and WOLFF 198 1; REUTER, WERNER and HOFFMANN 1982;

7

REUTER, DORN and HOFFMANN 1982; REUTER, HOFFMANN and WOLFF 1983). Therefore, E-var(3)1°' was used to test for new suppressors, and Su-var(2)1°' was used to select newly induced enhancers. Enhancer E-var(3)1°' is dominant to the suppressor effect of an additional Y chromosome, it effectively excludes the recovery of false positives from the suppressor mutation screen which arise from nondisjunction of the Y chromosome.

Mutagenized or hybrid-dys enic w'"''/Y males were crossed to wm4'; Sco/T(2;3)a$ + In(2L)Cy,apxn Cy E- var(3)lo'/TM2 females and the offspring w'"'~; +/T(2;3)apX0 Cy E-var(3)1°'/+ flies were inspected for new suppressor mutations. Enhancer mutations were selected in the w'"''; +/T(2;3)apxa Su-var(2)1°'/+ offspring of a cross of mutagen- ized w'"'" males or females with wm4'; Cy/T(2;3)apx" Su- var(2)1°'/Sb females or males. In addition, strong suppres- sors can be identified in siblings with the genotype wm4'; Cy/ +; Sb/+.

P hybrid-dysgenic males were recovered from a cross (at 18 ") of w~~~ females to r 2 males, a strain containing many P elements (ENGELS 1989) and crossed to females, as set out above, to screen for enhancers or suppressors.

The w'"''/, Icarus-neo/+ males were recovered by cross- ing a strain carrying the lethal insertion at 27A balanced over Cy0 (STELLER and PIROTTA 1986) tow'"4h; +/+ females. The F1 progeny was heat shocked (37" for 1 hr) during late embryonic and early first larval instar development.

Transpositions of the pUChsneory+ element were induced in heterozygotes with a fl Sb P[ry+ A2-31 (99B) (ROBERT- SON et al. 1988) or TM3,ry Sb e P[ry+ A2-31 chromosome (G. REUTER, G. HOFFMANN, R. DORN, J. GAUSZ and H. SAUMWEBER, manuscript submitted). In the crosses different pUChsneory+ insertions, all without effect on PEV, were used: one insertion in 67C, two independent insertions into the Cy0 chromosome and three different X chromosomal insertions. New transpositions of the pUChsneory+ element were selected as ry+ males in the Cy+Sb+ offsprin from a cross of ryJo6 females with Cy0 pUChsneory+/+; ryJo'Sb P[r A2-31 (99B) males or as ry+Sb+ males from a cross of ry 2: females with X pUChsneory+/Y; ryJo6/TM3,9Sb e P[ry+ A2- 31 males. Only a single ry+ animal was collected from each vial. Each and tested for its effect on PEV by crossing with w""; Cy/T(2;3)apxa Su-var(2)1°'/Sb females. The wm4'/, +/ T(2;3)apx"Su-var(2)l0'/ryJo6 male offspring were inspected for enhancer mutations, while their wm4'/, Cy/+; Sb/ryJo6 siblings were examined for suppressor mutations. Using a backcross of wm4'/, +/T(2;3)apxaSu-var(2)loO'/ry506 excep- tional males show an enhancer effect (mottled instead of suppressed red eye phenotype). The putative enhancer can be localized on the chromosomes with the help of aneuploid segregants of the upxa translocation (REUTER and WOLFF 198 1). Strains were constructed by crossing to CyO/Sco; ryJo6 and TM3,ryRK Sb e/ryJo6 females, which is also a possibility to control the segregation of the ry+ marker gene in pUChsneory+.

Transpositions of the P[IArB]-transposon which was in- serted into a Cy0 balancer chromosome (GROSSNIKLAUS et al. 1989) were selected as Cy+ Sb' ry+ male exceptional flies in the offspring of a cross of Cy0 P[IArB]/Sp; ryJo6/ryJo6 Sb P[ry+ A2-31 (99B) males with ryJo6/ryJo6 females. All other crosses were identical to those performed in the experiments with pUChsneory+.

In revertant analyses of the pUChsneory+ induced en- hancer mutations, ry- chromosomes were selected for sec- ond and third chromosomal mutations from the offspring of CyO/E-pUChsneory+; TM3, 9 Sb e P[ry+ A2-3]/ryJo6 or E-pUChsneory+/TM3, rym Sb e P[ry+ A2-31 males, respec- tively. The P-induced suppressor mutations were tested for

Position-Effect Variegation 28 1

their ability to revert somatically in wmrh/Y males in the presence of P[ry+ A2-31. Eyes showing white mottled sectors on a suppressed almost red eye background indicated so- matic reversion of the insertional suppressor mutation.

The effects of the modifier mutations on white variega- tion in wmrh flies was quantified by measuring the red eye pigments using previously employed methods (REUTER, HOFFMANN and WOLFF 1983). Measurements of the effects of enhancer mutations were always performed in flies which also carried the Su-~ar(Z)1~' mutation.

Complementation analysis and mapping studies: Reces- sive lethal or recessive sterile mutations were tested for allelism by complementation analysis. The newly induced modifier mutations which were homozygous viable and fer- tile or which complement all known modifiers of PEV were mapped recombinationally or with duplications by testing phenotypic rescue of the mutation in duplication/modifier mutation trans-heterozygotes. In many cases two wild-type copies of a modifier gene quench the effect of the dominant mutant allele (WUSTMANN et al. 1989) resulting in a normal mottled phenotype if the duplication covers the wild-type function of the mutated locus. For the mutation M Suz4, which is both a strong dominant suppressor and a Minute, new tandem duplications were selected after irradiating

method (Ashburner 1989). Molecular studies: The pUChsneory+ transposon was

constructed by insertion of a 7.3-kb HindIII ry+ fragment from the Carnegie 20 vector (RUBIN and SPRADLINC 1983) into the HindIII site in the polylinker of the pUChsneo plasmid (STELLER and PIROTTA 1985).

Nucleic acid purification and analysis: DNA from flies was isolated by the method described by JOWETT (1986). For plasmid rescue the DNA was purified by CsCl gradient centrifugation as described by SAMBROOK, FRITSCH and MANIATIS (1 989). Southern blot hybridization and plasmid preparation was done according to SAMBROOK, FRITSCH and MANIATIS (1989). Labeling of DNA was accomplished by random oligonucleotide priming with [a-"P]dATP (HODG- SON and FISK 1987). Plasmid rescue of the genomic DNA from the insertional mutants was performed according to the method described by STELLER and PIROTTA (1985). Transformation was done on competent DH5a-cells pre- pared according to HANAHAN (1 983).

&Galactosidase staining: Staining of embryos, larvae and adults was performed according to the protocol of BELLEN et al. (1 989).

In situ hybridization: Cytological positions of the inser- tions and genomic rescue fragments were determined by in situ hybridization to the polytene chromosomes following a modified procedure of PARDUE and GALL (1 975).

w 1 4 h . , +/+ females with 4000 R using the anti-Minute

RESULTS

Isolation of mutations: Previous studies have sug- gested that the modifier genes of PEV represent haplo dose sensitive functions and that the dominant muta- tions are amorphic or hypomorphic in nature (LOCKE, KOTARSKI and TARTOF 1988; WUSTMANN et al. 1989). Until now about 390 mutations were isolated, how- ever, only one antimorphic and one neomorphic mu- tation was identified (REUTER et al. 1987; G. REUTER and J. SZABAD, manuscript submitted). Therefore, our results strongly support the hypothesis that the dom- inant mutant effect is the result of either a loss of function or a reduced activity of the gene product.

Using EMS (2.5 mM) or X-rays (4,000 R) as muta- gens, suppressor and enhancer mutations were se- lected with a frequency between 1.3 x and 2.2 X and with lower frequencies if females were irradiated (Table 1). There were no significant differ- ences in mutation frequencies between the two types of mutations in the X-ray experiments regardless of whether females or males had been irradiated. The high mutation rates are a result of the fact that there is a large group of genes which produce the same phenotype when mutated.

Since it appears that P-induced insertional muta- tions are often weak hypomorphs (WILLIAMS, PAPPU and BELLO 1988), any mutant isolation schemes have to be very sensitive to allow the identification of new mutations. Using the mutations Su-vur(2)l0' and E- var(3)1°' (REUTER, WERNER and HOFFMANN 1982; REUTER, HOFFMANN and WOLFF 1983) which are sensitive to weak enhancer and weak suppressor mu- tations, respectively, we were able to increase the sensitivity of our screens.

In our experiments using ?r2 hybrid dysgenic males, enhancer mutations were recovered at a rate compa- rable with those of X-ray experiments (Table l), how- ever, suppressor mutations were recovered at a signif- icantly lower rate. T h e recovery of a relatively high number of enhancer mutations suggested that modi- fied P elements can also be used successfully as muta- tor elements.

In one experiment the Icarus-neo element (STELLER and PIROTTA 1,986) was mobilized. Since none of the isolated mutations proved to represent a transposition of the Icarus-neo element they have to be classified as spontaneous in origin.

Two additional modified P elements, pUChsneory+ inserted into 67C and P[IArB] inserted into the Cy0 chromosome (cf. MATERIALS AND METHODS) were mo- bilized using the P[ry+ A2-31 (99B) transposase pro- ducing element. Transpositions of the pUChsneory' element were recovered with a frequency of 1 O-'(data not shown). In our initial experiments enhancer mu- tations were recovered with a frequency of 2.1 X 1 O-4. However, molecular studies demonstrated that only two of the mutants represented insertions of intact pUChsneory+. This indicated a consistent and high spontaneous mutation rate in the strains used, and that two classes of mutations, pUChsneory+ induced and spontaneous mutations, were recovered with the spontaneous mutations which preponderate over the transposable element insertional mutations. There- fore, in order to select only those mutations caused by insertion of the modified P element it was necessary to monitor transpositions of the mutator element first and then to test their phenotypic effect on PEV. Out of 1,895 pUChsneory+ and 669 P[lArB] independ- ently isolated transpositions, 34 insertional enhancer

282 R. Dorn et al.

TABLE 1

Frequency of dominant PEV modifier mutation induced by P transposon mutagenesis compared with EMS and X-ray mutagenesis

Frequency of mutations"

Expt. Suppressore

Mutagen Individuals [insertions] [insertions] Enhancers

1 P element (a2 dysgenic cross) 227,000 fliesb 5(4.4 x 5q5.1 X

2 Icarus-neo (27A insertion and newly transformed strains) 3 17,168 fliesb 2(2.0 X 10-5) ~ ~ ( 4 . 6 X 10-4)

3 pUChsneoy+ (67C insertion) 253,879 fliesb 0(

Position-Effect Variegation 283

The two suppressors isolated from the experiment with lcarus-neo do not revert. One of these mutations [M Suz4] also expresses a Minute phenotype. The following duplications were isolated: DP(2;2)M+Suz4+-

Dp(2;2)M'Su2"-2 [24B-C; 29B-C], DP(2;2)M+Suz4+-3 [23B; 25B], Dp(2;2)M+SuZ4+-4 [23F; 25C-Dl and Dp(2;2)M+SuZ4+-5 [no rearrangements visible]. Dupli- cation mapping places the suppressor and the Minute phenotypes between 24B-C and 25A. The Dp(2;2)M+Suz4+-5 chromosome, which does not show any visible cytological rearrangement, rescues both the Minute and suppressor phenotype of M SuZ4, sug- gesting that the two functions are closely linked or even map to the same locus.

Enhancer mutations: In the X-ray experiments 48 enhancer mutations were isolated. Five represent du- plications covering the Su-var(2)l locus; these quench the suppressor effect of the Su-var(2)1°' mutation but do not show any effect on variegation in the absence of Su-var(2)lu'. Of the remaining 43 mutations, 20 are located on the second chromosome, 21 on the third chromosome and two mutations are associated with 2;3 translocations. All second chromosome mu- tations and 13 of the third chromosome mutations are recessive lethal. Several of these mutations represent deficiencies.

In the 7r2 experiment (Table 1) 58 mutations were isolated, 24 (with 4 clusters) located on the second and 34 (with 4 clusters) on the third chromosome. Of the second chromosome mutations, 7 are homozygous viable, 15 are recessive lethal and 2 are semilethal. In complementation studies the 15 recessive lethal mu- tations were found to represent 10 genes. Only one is represented by several (6) alleles. Most of these mu- tations have a relatively weak enhancer effect. How- ever, two are tandem duplications [Dp(2;2)23d and Dp(2;2)33] covering the triplo-enhancer function in 29A (WUSTMANN et al. 1989) and are strong en- hancers. A third strong enhancer is lethal over Df12L)clh3 (SZIDONYA and REUTER 1988) and quenches the triplo-suppressor effect of region 26B. Therefore, this mutation is allelic to the haplo-enhancer and triplo-suppressor locus in region 26B (REUTER and SZIDONYA 1983). The mutation does not revert in the presence of the P[ry+ A2-31 (99B) element and the results of complementation studies suggest that it rep- resents a small deletion in region 26B (data not shown).

Of the 34 mutations in the third chromosome 26 are recessive lethal or semilethal and recessive female sterile, and 8 are homozygous viable. Three of the homozygous viable mutations were mapped recombi- nationally to 56.4 (E-var(3)12), 61 .O (E-var(3)13) and 77.7 (E-var(3)14) indicating that several independent enhancer loci represent nonessential functions (Figure

1 [21E1-2; 23A1-2 + 23A1-2; 25Dl-21,

1). In complementation studies with the recessive le- thal and semilethal female sterile third chromosomal mutations, alleles were identified for four genes: 6 alleles for E-var(3)3 at 84.3, two mutations are alleles of E-var(3)4 at 57.5, four alleles for E-var(3)S at 58.0, and three are alleles of E-var(3)8 at 15.8 (Figure 1). All the remaining recessive lethal mutations (1 1) rep- resent independent loci. Five of the 11 mutations analyzed did not revert regardless whether the en- hancer effect or recessive lethal phenotype was tested. Therefore, we conclude that several of the mutations isolated in the 7r2 screen represent chromosomal re- arrangements displaying a haplo- or triplo-enhancer effect (deficiencies and duplications) or are sponta- neous in origin (mobilization of other elements).

A large sample of enhancer mutations were isolated in the Icarus-neo experiment (14 for the second and 54 for the third chromosome). None of these muta- tions proved to represent an Icarus-neo transposition. The Icarus-neo transposon at 27A was subsequently found to be rearranged (V. PIRROTTA, personal com- munication). However, analysis of the mutations re- covered has shown that 20 of the third chromosomal mutations were alleles of the E-var(3)4 locus (semile- thal and recessive female sterile) and six of the muta- tions are alleles of the E-var(3)5 locus (recessive lethal). This suggests that these loci may be hot spots for mutations in the stocks used. However, the mutations do not revert in the presence of P[q+ A2-31 and no Icarus-neo homology could be detected in genomic Southern blots of the mutant lines. Since several of the alleles were isolated in clusters (identical mutations from sibs of single dysgenic males), these mutations may have resulted from insertions of other elements mobilized by the cross.

Similar results were obtained in the experiment with the pUChsneory+ element at 67C (Table 1). Only two out of 27 mutations (E-var(2)1°' at 26A7-9 and E-var(3)3" at 93D) represent new insertions of an intact pUChsneory+ mutator element. The remaining mutations do not revert with P [ q + A2-31 and no homology to P elements or pUChsneoq+ could be found in the stocks. Three of these mutations were isolated as clusters and, again, could be due to the insertion of another mobile element.

Twenty-eight out of the 34 enhancer mutations isolated from transpositions of pUChsneory+ or P[lArB] were localized by in situ hybridization (Tables 3 and 4; Figure 1). Southern blot studies of the tested mutants indicate an intact mutator element. Twenty out of the 34 mutations are recessive lethal, semilethal or recessive female sterile. Allelism with other en- hancer mutations was found in the following cases: 195 is an allele of E-var(3)lO and the mutations 125 and 512 are alleles of E-var(3)15 defining two addi- tional recessive lethal enhancer genes.

284 R. Dorn et al.

2L

3L

0.0 13.0 16.5 a/ dp cl

41 .O 54.5 J PI

I I I I X) 30

.- * ;7 ;LJ - ; 3;, 31 I

I ' I I "... I I 1 I ' I I I 32 33 34 35 36 37 38 39 40

- 1817 €!E E &J lL 190 I 57.5 67.0 162

- - 75.5

0 C 100.5

I I PX 107.0 SP

I 1' I I I 1 55 60 70 80 90 108

2R 4: 4; T-1

44 45 -1 46 47 48 49 50 51 I 52 I 53 I" " I I I 54 55 56

E 61 E-var(3)7

- E-va 3)6

2.0 1 5% 0.0 26.0 TU S8

I 10

I I I I 61 62 63 64

100.7

82 83 84 85

195 125 - - FIGURE 1.-Mapping of enhancers of PEV. The genetic (top) and the cytogenetic maps (below) of each autosome arm are shown. On the

genetic map enhancer genes already localized by crossover analysis are shown. On the cytogenetic map regions displaying an enhancer effect if deleted (haplo-enhancers) are indicated by black bars. The positions of pUChsneory+ (solid numbers) and P[IArB] mutations (numbers in boxes) as determined by in situ hybridization are shown below the cytogenetic maps. Underlined mutations were used for isolation of flanking genomic DNA fragments via plasmid rescue in Escherichia coli. In situ hybridization with the genomic DNA fragments gave a signal at sites - - identical with the insertion sites.

The strength of the enhancer mutation was deter- mined by quantifying the amount of red eye pigments. The mutations were studied in the presence of the Su- var(2)Io' mutation (Tables 3 and 4). Eleven of the 34 pUChsneory+ and P[lArB] insertional mutations ex- hibit a strong enhancer effect. The remaining muta- tions represent medium or weak enhancers (Tables 3 and 4). Twenty-one of the 34 insertional mutations express significant paternal effects (Tables 3 and 4), i e . , if the male parent carried an enhancer mutation his progeny display an enhancer phenotype regardless of whether or not they received the enhancer muta- tion from their father (Figure 2).

Revertant analysis was performed in 4 pUChsneory+ (162,189,1817 and 129) induced enhancer mutations (Table 5). The revertant chromosomes were selected by reversion from ry+ to ry- in a rySo6 background and were tested by complementation analysis (loss of le- thality and sterility) and for loss of the enhancer effect by a cross to Su-var(2)1°' flies. The data obtained

suggest that the different phenotypic effects of the mutations are due to the insertion of the mutator element (Table 5). In several cases the revertant chro- mosomes still retained the enhancer and lethality or sterility phenotypes suggesting that excision of the mutator element was imprecise. Further characteriza- tion of these revertants has shown rearrangements of the mutator elements or deficiencies in the flanking genomic sequences (R. DORN, unpublished results). Some of these revertants probably represent amorphic mutations of the locus.

Finally as a result of complementation studies of all mutations of the third chromosome 8 enhancer loci were found to be represented by several alleles: 24 alleles for E-var(3)4 at 57.5, 9 alleles for E-var(3)S at 58.0, 3 alleles for E-var(3)6 at 57.0, 2 alleles for E- var(3)7 at 2.0, 3 alleles for E-var(3)8 at 15.8, 3 alleles for E-var(3)9 and 2 alleles for E-var(3)lO and E- var(3)Zl each. With the pUChsneory+ and P[IArB] insertional mutations 17 additional essential enhancer

Position-Effect Variegation 285

TABLE 3

pUChsneory+ induced enhancers of PEV summary of genetic analyses

Effect on w" variegation

Line Enhancer effect

Locus Viability/Fertilitya w*'* strain With $21' effects' Paternal

Second chromosome 61 53E

130 ND 162* 26A7-9 182 ND 183 56DE 189* 34D3-8 190 37EF

1817 21B6 Third chromosome

48 91F/92A1.2 62 70EF

I13* 86E(F) 125 87A 129* 93D 181 75B5-10 188 88A6-10 195 85C11-13 236 86F

1847 93E6-11 3118 85F9-1

sl/fs -

+ + -

sl/fs + + + +

Medium Weak Strong Strong Weak Weak Weak Strong

Strong Weak Weak Weak Strong Weak Medium Weak Weak Strong Weak

0.73 1 . 1 0.15 0.53 0.97 0.92 1.3 0.58

0.65 0.80 1 . 1 0.73 0.41

0.68 1.22

ND

ND

ND

ND

l.lO(-) 0.34(+) 0.56(+) 0.45(+) 0.60(+)

0.51(+) 0.62(+)

l.lO(-)

ND

0.48(+) 0.78(+) 0.59(+) 0.49(+)

0.68(+) 0.62(+)

ND

ND

ND ND

a SI, semilethal; fs, recessive female sterile. ' Enhancer effect and pigment ratio between the two sibling genotypes Su-var(2)1°' E mutation/Su-var(2)1°' E+ from crosses of w ~ ~ ~ / w ~ ~ ~ ; Su-uar(2)1°'/Cy females with the w""//Y; E/Bal (Cy or 7 " ) males. nd, no pigment measurements performed. The suppressor Su-var(2)1°' was used in order to quantify the enhancer effect of the mutations (MATERIALS AND METHODS). Su-var(2)1°' is stronger in males if maternally originated and weak enhancers are covered by the suppressor (pigment ratio is around 1 .O).

C Paternal effects of the mutations were quantified by comparing the relative content of red eye pigment of the Su-var(2)lo' offspring from crosses of W ~ ' ~ / W " ' ~ ~ ; Su-var(2)1°'/Cy females with males carrying the enhancer and wmfh males without the enhancer mutation. If the enhancer displays a paternal effect white variegation is enhanced in Su-var(2)I0' E' offspring males when the father carried the mutation compared to crosses where the father did not. (+) or (-), with or without paternal effect.

* Genomic fragments were obtained by plasmid rescue.

TABLE 4

P[lArB] induced enhancers of PEV. summary of the genetic analyses ~~

Effect on wn' variegation

Line Enhancer e,ffect

Locus ViabiIity/Fertility" W n 4 h stram With Sub effects' Paternal

Second chromosome 1 ND sl/fs Weak 1.60 0.20(+)

45* 57E1-4 + Strong 0.62 21 1

0.28(+) 31EF + Strong 0.47 0.38(+)

Third chromosome 512 64E - Strong 0.58 0.20(+) 631

27 A I 85A1-5

21 2 75c3-5 26 I 231 * 222

71 2

31 35

For details, see Table 3.

85F9-16 - Strong 0.20 0.16(+) ND +

+ + +

93D5-7 + 87B7-10 +

Weak Weak Weak Weak ND 0.18(+) Weak Medium 0.79 0.30(+)

66B1-4 + Weak 0.98 0.22(+) 0.42(+)

79E1-4 + Weak

ND

ND ND

ND

ND

ND 87B7-10

ND ND

ND - Strong 0.51 1.12

ND + Weak 1.25 0.21(+) ND

286 R. Dorn et al.

Cross of Suppressor females with I Enhancer males

-.. , Females

m m4h'w m 4 h ; +lSu w m 4 h t Y; +lSu II

FIGURE 2.-Paternal effects of enhancer mutations. For quanti- fication of the enhancer effect the dominant suppressor mutation Su-uar(2)1°' was used. (R1) Comparison of relative pigment values between the sibling Su/E and Su/+ males. If the enhancer shows a strong effect R1 is

Position-Effect Variegation 287

TABLE 6

/M"ctosidase staining of P(IArB)-induced PEV enhancer mutations

Stages and tissues of &galactosidase expression

Third larval instar Adults

Line Embryo BIG SG ID 0 T M T PV 0 T M T

Second chromosome I Zygotic +

45 Zygotic + 211 Zygotic +

27 Zygotic - 31 Zygotic + 35 Zygotic + 71 Zygotic -

212 Zygotic + 222 Zygotic + 231 Maternal + 261 Zygotic + 512 Maternal +

Third chromosome

+ + + + + + + + + + + + + - - - - - + - - + + + - + - " " - - + - + + + + - - + + - + - + + + + + + + " " " - - - + - + + - - + + - - + + + - - + + - + + + + + - + + - + + + + + + + + - + + + + + + + + -

B, Brain; G, ventral ganglion; SG, salivary glands; ID, imaginal discs; 0, ovaries; T, testes; MT, malpighian tubules; PV, proventri- culus.

the mutations interesting staining patterns were found in larval or adult gonads (Figure 3, d-f).

DISCUSSION

Genetic studies clearly demonstrated the complex genetic basis of PEV (HENIKOFF 1979; REUTER and WOLFF 198 1 ; SINCLAIR, MOTTUS and GRIGLIATTI 1983; WUSTMANN et al. 1989; SINCLAIR et al. 1992). Four different classes of genes were described all expressing a haplo-dependent phenotype but differing in their triplo-effects. Only some of the haplo-sup- pressors and haplo-enhancers display an opposite tri- plo-dependent effect. The haplo-dependent enhancer genes have been studied but not in detail (SINCLAIR, LLOYD and GRICLIATTI 1989). According to the com- plementation studies performed with recessive lethal or sterile mutations we estimate the number of en- hancer genes at about 30 in the third chromosome and between 50 and 60 for the whole autosome com- plement. For nine genes in the third chromosome more than two alleles have been identified. Therefore, enhancer of PEV loci are found in the Drosophila genome as frequently as suppressor genes.

Using P transposons as mutator elements, enhancer mutations of PEV were induced with a 10 times higher frequency than suppressor mutations. This might in- dicate target site specificity of the P element. A critical proof could consist of a sequence comparison at the insertion site of the different mutations isolated. BOWNES (1990) showed preferential insertion of P elements into genes being actively transcribed at the time of transposition and a higher transcriptional ac- tivity of enhancer genes in male germ-line cells, there-

fore, this could also explain the differences found in mutant frequencies. Different sensitivities in the mu- tant isolation tests used can be excluded on the basis of the results of EMS and X-ray mutagenesis. The enhancer mutation E-var(2)I which was used to test for newly induced suppressors is very sensitive to the effect of suppressor mutations and even weak domi- nant suppressors could be isolated in the screens.

Systematic molecular studies of enhancer loci of PEV became possible by the successful isolation of pUChsneory+ and P[lArB] insertional mutations. Ge- nomic sequences from the insertion sites were isolated by plasmid rescue for further molecular cloning of the genes. The insertional mutations also represent a useful material for genetic analyses of the correspond- ing genes. By in situ hybridization with both transpo- son specific probes (pUC18) and genomic DNA frag- ments, isolated from the insertion site after plasmid rescue, precise chromosomal mapping of the locus can be performed. The ry+ function associated with this transposon can be used as a marker for crossover studies and for phenotypic detection of revertants or the selection of deficiencies. Reversion studies for several of the mutations suggest that the enhancement of PEV as well as the recessive lethal or female sterile phenotype, result from the insertion of the mobile element.

Many of the enhancer mutations display significant paternal effects (Figure 2). Comparable paternal ef- fects have already been detected with variant lines of wm4 and were shown to be caused by an imprinting of the Y chromosome (REUTER, WOLFF and FRIEDE 1985; REUTER and SPIERER 1992). Twenty-three out of 34 mutations show significant paternal effects. The inser- tional mutations should facilitate a molecular analysis of these paternal effects.

Almost all of the P[lArB]-induced mutations showed a specific P-galactosidase-staining pattern dur- ing development. Because its selection was dependent on a dominant mutant effect (enhancement of PEV), the insertion does not only lead to a disruption of the normal gene function, but also indicates an enhancer- like cis-acting regulatory element of an E-var gene. The mutations may be hypomorphic, but they defin- itively show that the identified regulatory elements are connected with the gene function. While BELLEN et al. (1989), WILSON et al. (1989), GROSSNIKLAUS et al. (1989), and BIER et al. (1989) show a good corre- lation between lac2 expression in transposants, tran- script detection and mutant phenotype of a gene, BELLEN et al. (1 989) have also shown that the complete expression pattern of the affected gene cannot be detected by means of a single insertion. Most of our P[IArB]-induced mutations express &galactosidase in larval brain, imaginal discs, salivary glands, and in testes as well as in ovaries of both larvae and adults,

288 R. Dorn et al.

FIGURE 3.--&Galactosidase staining of some P[lArB] induced enhancer mutations. (a and b) Staining of ovaries in mutation 231. Staining of nuclei in nurse cells and the oocyte in (a) and complete staining of egg chamber at stage 13 p). (c) In ovaries of mutation 261 the nuclei of nurse cells and follicle cells are stained. (d) Complete staining of adult testis in the mutant line 512 and (e) complete testis staining with exception of the tip in mutant line 231. (0 In line 21 l a gradient of staining in larval testis anlagen with strong staining of spermatogonia is found.

Position-Effect Variegation 289

suggesting that the genes are active in somatic and germ-line cells. Staining of salivary glands, Malpighian tubules and imaginal discs might indicate genes en- coding chromosomal constituents found in both mi- totic and polytene chromosomes.

Two of the mutations (231 and 512) identify ma- ternal functions. Examination of @-galactosidase expression in ovaries demonstrates maternal gene ac- tivity in both cases. There is compelling evidence that many of the chromatin genes identified as suppressors or enhancers of PEV represent maternal functions. Maternal transcripts were identified for Su-var(2)5, Su-var(?)7 and modulo all of which encode chromo- somal proteins (EISSENBERG et al. 1990; REUTER et al. 1990; KREJCI et al. 1989). Pole cell transplantation experiments (H. TAUBERT, unpublished results) also suggest that several of the suppressor and enhancer mutations of PEV are expressed maternally. Some of these genes could represent maternally expressed chromatin functions whose products are also neces- sary for cleavage division. These genes could be abun- dantly transcribed. This is indeed the case for the enhancer genes identified by mutations 162 [E- var(2)1] in 26A and 129 [E-var(3)?] in 93D (R. DORN, V. KRAUS, A. SCHUBERT, H. SAUMWEBER and G. REU- TER, manuscript in preparation).

The pattern of @-galactosidase staining also indi- cates gene expression in mitotic active tissues like brain or imaginal discs. From the 15 P[lArB]-induced mutations 10 show intensive staining of the brain and ventral ganglion and 7 show staining of imaginal discs. The recessive lethal mutations should also be studied for a mitotic phenotype.

The mutants isolated have allowed a basic genetic and developmental analysis of E-var loci. Because the mutations enhance chromatin condensation in PEV the normal function of the genes should be causally connected with decondensation of chromatin, a main prerequisite of gene activity. Molecular cloning and a study of their gene products can be performed now for many enhancer loci of PEV.

We are very grateful to T. GRIGLIATTI, R. MOTTUS, M. HAR- RINGTON, E. KNUST and A. HOFMANN for critical reviewing the manuscript and many helpful suggestions. This work was supported by the Deutsche Forschungsgemeinschaft (KO 434/12-1 and Re 91 1/1-1).

LITERATURE CITED

ASHBURNER, M., 1989 Drosophila: A Laboratory Handbook. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.

BELLEN, H. J., C. J. O'KANE, C. WILSON, U. GROSSNIKLAUS, R. K. PEARSON and W. J. GEHRING, 1989 P-element mediated en- hancer detection: a versatile method to study development in Drosophila. Genes Dev. 3: 1288-1300.

BIER, E., H. VASSIN, S. SHEPHARD, K. LEE, K. MCCALL, S. BARBEL, L. ACKERMANN, T. UEMURAU, E. GRELL, L. Y. JAN and N. Y. JAN, 1989 Searching for pattern and mutation in the Dro-

sophila genome with a P-lacZ vector. Genes Dev. 3: 1273- 1287.

BOWNFS, M., 1990 Preferential insertion of P elements into genes expressed in the germ-line of Drosophila melanogaster. Mol. Gen. Genet. 222 457-460.

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

EISSENBERG, J. C., 1989 Position effect variegation in Drosophila: towards a genetics of chromatin assembly. BioEssays 11: 14- 17.

EISSENBERG, J. C., T. C. JAMES, D. M. FOSTER-HARTNETT, T. HATRNETT, V. NGANN and S. C. R. 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.

ENGELS, W. R., 1989 P elements in Drosophila, pp. 437-484 in Mobile DNA edited by D. E. BERG and M. M. How. American Society for Microbiology, Washington, D.C.

GRIGLIATTI, T., 1991 Position-effect variegation-an assay for nonhistone chromosomal proteins and chromatin assembly and modifying factors. Methods Cell Biol. 35: 587-627.

GROSSNIKLAUS, U., H. J. BELLEN, C. WILSON and W. J. GEHRING, 1989 P-element-mediated enhancer detection applied to the study of oogenesis in Drosophila. Development 107: 189-200.

HANAHAN, D., 1983 Studies on transformation of Escherichia coli with plasmids. J. Mol. Biol. 166: 557-580.

HENIKOFF, S., 1979 Position-effects and variegation enhancers in an autosomal region of Drosophila melanogaster. Genetics 93:

HODCSON, C. P., and R. Z. FISK, 1985 Hybridization probe size control: optimized "oligolabelling." Nucleic Acids Res. 15: 6295.

JAMES, T. C., and S. C. R. ELGIN, 1986 Identification of a nonhi- stone chromosomal protein associated with heterochromatin in Drosophila melanogaster and its gene. Mol. Cell. Biol. 6 3862- 3872.

JOWETT, T., 1986 Preparation of nucleic acids, pp. 275-286 in Drosophila: A Practical Approach, edited by D. B. ROBERTS. IRL Press, Oxford.

KREJCI, E., V. GARZINO, C. MARY, N. BENNANI and J. PRADEL, 1989 Modulo, a new maternally expressed Drosophila gene encodes a DNA-binding protein with distinct acidic and basic regions. Nucleic Acids Res. 17: 8101-8115.

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

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

PARDUE, L. M., and J. G. GALL, 1975 Nucleic acid hybridization to the DNA of cytological preparations. Methods Cell Biol. 10: 1-16.

REUTER, G., R. DORN and H. J. HOFFMANN, 1982 Butyrate sen- sitive suppressor of position-effect variegation mutations in Drosophila melanogaster. Mol. Gen. Genet. 188: 480-485.

REUTER, G., H. J. HOFFMANN and I. WOLFF, 1983 Genetic study of position-effect variegation in Drosophila melanogaster: In(1)w"' as a standard rearrangement for the isolation and characterization of suppressor and enhancer mutants. Biol. Zentralbl. 102: 281-298.

REUTER, G., and P. SPIERER, 1992 Position effect variegation and chromatin proteins. BioEssays 1 4 605-6 12.

REUTER, G., and J. SZIDONYA, 1983 Cytogenetic analysis of var- iegation suppressors and a dominant temperature-sensitive le- thal in region 23-26 of chromosome 2L in Drocophila melano- gaster. Chromosoma 88: 277-285.

106-115.

290 R. Dorn et al.

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, 198 1 Isolation of dominant suppressor mutations for position-effect variegation in Drosophila melano- gaster. Mol. Gen. Genet. 182: 516-519.

REUTER, G., I . WOLFF and B. FRIEDE, 1985 Functional properties of the heterochromatic sequences inducing wm4 position-effect variegation in Drosophila melanogaster. Chromosoma 93: 132- 139.

REUTER, G., R. DORN, G. WUSTMANN, B. FRIEDE and G . RAUH, 1986 Third chromosome suppressor of position-effect varie- gation loci in Drosophila melanogaster. Mol. Gen. Genet. 202: 481-487.

REUTER, G., J. GAUSZ, H. GYURKOVICS, B. FRIEDE, R. BANG, A. SPIERER, L. M. C. HALL and P. SPIERER, 1987 Modifiers of position-effect variegation in the region from 86C to 88B of the Drosophila melanogaster third chromosome. Mol. Gen. Ge- net. 2 1 0 429-436.

REUTER, G . , M. GIARRE, J. FARAH, J. GAUSZ, A. SPIERER and P. SPIERER, 1990 Dependence of position-effect variegation in Drosophila on dose of gene encoding an unusual zinc-finger protein. Nature 344: 219-223.

ROBERTSON, H. M., C. R. PRESTON, R. W. PHILLIS, D. JOHNSON- SCHLITZ, W. K. B ~ ~ z a n d W. R. ENGELS, 1988 A stablesource of P-element transposase in Drosophila melanogaster. Genetics 118: 461-470.

RUBIN, G. M., and A. C. SPRADLING, 1982 Genetic transformation of Drosophila melanogaster with transposable element vectors. Science 218: 348-353.

RUBIN, G. M., and A. C. SPRADLING, 1983 Vectors for P element- mediated gene transfer in Drosophila. Nucleic. Acids Res. 11:

SAMBROOK, J., E. F. FRITSCH and T . MANIATIS, 1989 Molecular 6341-6351.

Cloning: A Laboratory Manual, Ed. 2. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.

SINCLAIR, D. A. R., V. K. LLOYD and T. A. GRIGLIATTI, 1989 Characterization of mutations that enhance position- effect variegation in Drosophila melanogaster. Mol. Gen. Genet.

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

SINCLAIR, D. A. R., A. A. RUDDELL, J. K. BROCK, N. J. CLEGG, V. K. LLOYD and T. A. GRIGLIATTI, 1992 A cytogenetic and genetic characterization of a group of closely linked second chromosome mutations that suppress position-effect variega- tion in Drosophila melanogaster. Genetics 1 3 0 333-344.

STELLER, H., and V. PIROTTA, 1985 A transposable P vector that confers selectable G4 18 resistance to Drosophila larvae. EMBO J. 4: 167-171.

STELLER, H., and V. PIROTTA, 1986 P transposons controlled by the heat shock promoter. Mol. Cell. Biol. 6 1640-1649.

SZIDONYA, J., and G. REUTER, 1988 Cytogenetic analysis of the echinoid ( e d ) , dumpy ( d p ) and clot ( c l ) region in Drosophila melanogaster. Genet. Res. 51: 197-208.

WILLIAMS, J. A,, S. S. PAPPU and J. B. BELL, 1988 Molecular analysis of hybrid dysgenic-induced derivatives of a P-element allele at the vg locus. Mol. Cell. Biol. 8: 1489-1497.

WILSON, C., R. K. PEARSON, H. J. BELLEN, C. J. O’KANE, U. GROSSNIKLAUS and W. J. GEHRING, 1989 P-element-mediated enhancer detection: an efficient method for isolating and char- acterizing developmentally regulated genes in Drosophila. Genes Dev. 3: 1301-1313.

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

2 1 6 328-333.

Communicating editor: R. E. DENELL