Copyright 0 1993 by the Genetics Society of America

Suppression of a Lethal Trisomic Phenotype in Drosophila melanogaster by Increased Dosage of an Unlinked Locus

Douglas R. Dorer,' Marilyn A. Cadden, Beth Gordesky-Gold, Garth Harries' and Alan C. Christensen

Department of Biochemistry and Molecular Biology and Jefferson Institute of Molecular Medicine, Thomas Jefferson University, Philadelphia, Pennsylvania 19107-5541

Manuscript received May 4, 1992 Accepted for publication January 7, 1993

ABSTRACT One of the most extreme examples of gene dosage sensitivity is the Triplo-lethal locus (Tpl) on the

third chromosome of Drosophila melanogaster, which is lethal when present in either one or three copies. Increased dosage of an unlinked locus, Zsis, suppresses the triplo-lethal phenotype of Tpl , but not the haplo-lethal phenotype. We have mapped Zsis to the X chromosome region 7E3-8A5, and shown that the suppression is a gene dosage effect. Altered dosage of Zsis in the presence of two copies of Tpl has no obvious effects. By examining the interactions between Zsis dosage and Tpl we suggest that Zsis does not directly repress Tpl expression, but acts downstream on the triplo-lethal phenotype . - " of Tpl.

A NEUPLOIDY is often accompanied by develop- mental defects or lethality, which is generally

assumed to be due to the simultaneous imbalance of many genes, some with major effects and others with minor effects. In humans, mice and Drosophila the severity of the phenotype is a function of the amount of the genome that is unbalanced, although some specific regions appear to play a disproportionately large role (EPSTEIN 1988; LINDSLEY et al. 1972). The best example of such a region is the Triplo-lethal locus (Tpl) of Drosophila melanogaster, which causes lethality when it is present in either three copies or one copy (DENELL 1976; LINDSLEY et al. 1972). Attempts by several groups to mutagenize Tpl have revealed that it is possible to eliminate Tpl function by deleting the region, but it may be impossible to completely elimi- nate Tpl function by point mutations or transposon insertions (DORER and CHRISTENSEN 1990; KEPPY and DENELL 1979; ROEHRDANZ and LUCCHESI 1980). One possible explanation for these results is that Tpl con- sists of a cluster of functionally redundant transcrip- tion units, perhaps like the previously described Achaete-Scute Complex (ALONSO and CABRERA 1988) or the Enhancer of split locus (DELIDAKIS et al. 1992). It has been suggested that Tpl is involved in the sex determination pathway as an autosomal element in the X:autosome ratio (BAKER and BELOTE 1983; DE- NELL 1976; LUCCHESI and MANNING 1987), but ex- periments reported here and elsewhere have shown

' Present address: Division of Basic Sciences"684, Fred Hutchinson Cancer Research Center, 1124 Columbia Street, Seattle, Washington 98104.

* Present address: Yale University, P.O. Box 3069, Yale Station, New Haven, Connecticut 06520.

Genetics 1 3 4 243-249 (May, 1993)

that this is unlikely (CHRISTENSEN and LUCCHESI 1988).

Previous genetic studies of Tpl have exploited the existence of stocks carrying a tandem duplication of Tpl on one homolog and a deficiency for Tpl on the other (DORER and CHRISTENSEN 1990; KEPPY and DENELL 1979; ROEHRDANZ and LUCCHESI 1980). When these are crossed to normal flies, all progeny die because they have either three copies of Tpl or one. Although deficiencies for Tpl are easily re- covered from this cross because they block the lethal effect of a duplication of the locus, extensive efforts to obtain point mutations that have the same pheno- type as deficiencies (null mutations) have been unsuc- cessful. A second class of mutation was recovered in two of the studies (DORER and CHRISTENSEN 1990; ROEHRDANZ and LUCCHESI 1980). These mutations are tightly linked to the known location of Tpl (poly- tene bands 83E1,2 on the third chromosome), block the lethal effect of a duplication for Tpl, and are recessive lethals, but they are not haplo-insufficient. Although these were previously thought to be hypo- morphic mutations of Tpl, we now believe them to be dominant mutations in a tightly linked, but function- ally distinct gene called Suppressor of Triplo-lethal (Su(Tp1)) that suppress the triplo-lethal phenotype of Tpl (our unpublished data).

In addition to these mutations which suppress the triplo-lethal phenotype of Tpl, it has been shown that the viability of Tpl hyperploids can be enhanced by increased dosage of the X chromosome. Normally flies with three copies of Tpl die as late embryos or early first-instar larvae, but if these individuals also have

244 D. R. Dorer et al.

three X chromosomes they may survive into the third larval instar, or possibly longer (ROEHRDANZ and Luc- CHESI 198 1). This corresponds to a lengthening of the life span from 18-24 hr to 4-5 days. Roehrdanz and Lucchesi also used X;Y translocations to show that trisomy for the entire X chromosome was not neces- sary for the effect. When they studied the distal end of the X chromosome they found that the trisomy must extend at least as far proximal as cytological region 8A in order to prolong the survival of Tpl hyperploids, and when they used proximal transloca- tions they found that the trisomic region must extend at least as far distal as region 6D in order to have the effect. Since the minimal proximal and distal translo- cations included a common region, they concluded that there is a locus between 6D and 8A that interacts with Tpl in a dosage-dependent manner. An alterna- tive interpretation of these data is that the increased viability of Tpl hyperploids is a function of the amount of the X chromosome that is duplicated; as larger and larger terminal translocations are used, an overlap near the middle of the X is an inevitable consequence of the experimental design.

We have further characterized this phenomenon, and have shown that a small region of the X chromo- some (7E-8A) is sufficient to rescue Tpl hyperploids. We call the genetic material responsible for suppress- ing the triplo-lethal phenotype Zsis, after the Egyptian goddess who resurrected her brother Osiris after he was killed. Using small duplications of the Zsis region we have recovered fertile adults of both sexes who bear three copies of Tpl. This demonstrates that suppression of the triplo-lethal phenotype by in- creased dosage of Zsis can result in virtually normal adults, and also provides further evidence that Tpl is not involved in the sex determination pathway. Al- tered dosage of Zsis itself has no obvious phenotype, nor does it have any effect on the haplo-lethal phe- notype of Tpl. We have also attempted to map Zsis using translocations, duplications, deficiencies, and a deletion constructed by the method of COOLEY, THOMPSON and SPRADLING (1 990).

MATERIALS AND METHODS

Drosophila stocks: Drosophila stocks were maintained on Formula 4-24 Instant Drosophila medium obtained from the Carolina Biological Supply Company, supplemented with live yeast. The translocation stocks T(I;Y)V16 and T(1;Y)BI 70 are described in LINDSLEY and ZIMM (1 992) and were obtained from John Merriam (University of California, Los Angeles). T p ( l ; Z ) ~ n + ~ ~ ~ f car; b d has been described (CRAYMER and ROY 1980; LEFEVRE 1981; LINDSLEY and ZIMM 1992) and consists of a transposition of the region 7A8-8A5 from the X chromosome into the second chro- mosome; the two elements of this transposition will also be referred to separately as D f l l ) ~ n + ~ ~ ~ f car and D p ( I ; 2 ) ~ n + ~ ~ ~ . This stock as well as DJI)C128, DJTI)GE202and Dp(l;3)sn'"' were provided by TERRY JOHNSON and ROB DENELL (Kansas

State University). DJI)RAZ was supplied by the Indiana University Drosophila Stock Center. Descriptions of the above markers and chromosome rearrangements and their origins can be found in (LINDSLEY and ZIMM 1992). The a2 and Oregon-R stocks used for hybrid dysgenesis, and Dp(3;3)TplT4' are described in (DORER and CHRISTENSEN 1990). The Tpl duplication/deficiency stock used predomi- nantly in this paper has been previously described (DORER and CHRISTENSEN 1990; KEPPY and DENELL 1979). Its full genotype is pX.p, In(1)EN y/pX.r", Zn(1)EN y ;; Dp(3;3)Tp121 pP/DJ3R)I8i77 p P , but for simplicity we will refer to it as Dp(Tpl)/DJTpl). A shortened derivative of D p ( l ; Z ) ~ n + ~ ~ ~ , called Dp', and the deficiency DJ1)hl" were originally created by F. FORQUICNON and M. GANS, are described by LINERUTH (1 987), and were kindly provided by STEPHAN ANDERSON of the University of Umei, Sweden. DP' contains a deletion of the distal end of D p ( l ; Z ) ~ n + ~ ' ~ , and is both ct- and sn-. The cytological limits of DJ1)hl" are described by ANDERSON and LAMBERTSSON ( 1 990) as being approximately 7C1,2-7D19-22.

Hybridization in situ to polytene chromosomes: Salivary gland squashes for in situ hybridization were prepared es- sentially as described in ENGELS et al. (1986), except that slides were not pretreated with Denhardt's solution. Plas- mids ps25.7wc (KARESS and RUBIN 1984) and Carnegie2O (SPRADLING and RUBIN 1983) were used as probes. DNA was biotinylated by nick translation with bio-16-dUTP (Be- thesda Research Laboratories) substituted for dTTP. Hy- bridizations were carried out at 39" in a buffer containing 4 X SSPE, 40% formamide and 1 rg/ml salmon sperm DNA (Sigma). The Enzo Biochemicals Detek-l-HRP kit was used for peroxidase staining as described in (ASHBURNER 1989).

Designer deletion: A deficiency for X chromosome bands 7D1-7D14-17 was constructed using the method of COOLEY, THOMPSON and SPRADLINC (1990). For target P elements, we chose the pair of P elements tandemly inserted in the singed locus at 7D1, snw (ROIHA, RUBIN and O'HARE 1988), and an insertion of a P(ry') element at 7D14-17. A sn";;ry/ry stock was obtained from SCOTT HAWLEY (Univer- sity of California, Davis) and Is(I)Prt007D-AS334;;ry/ry (for- merly called R403.1 (SPRADLING and RUBIN 1983)), here- after called P(ry+);;ry/ry, was obtained from the Indiana University Drosophila Stock Center. In order to place these P elements in cis, the two stocks were crossed and hetero- zygous snw/P(ry+);;ry/ry females were mated to sn"/Y;;ry/ry males. The progeny were screened for recombinant snw P(ry+);;ry/ry, and several lines were established and checked for the presence of both P elements by in situ hybridization. The deficiency between these two P elements was isolated by crossing snw P(ry+)/Y;;ry/ry males to +/+;;ry Sb A2-31 T M 6 4 Ubx females (A2-3 is a nonmobilizable source of P transposase (ROBERTSON et al. 1988)), collecting the snw P(ry+)/+;;ry/ry Sb A2-3 daughters and mating them to FM7al Y males. We then screened for lethal X chromosomes by selecting Sb+ female progeny and mating them individually to FM6IY males. The progeny of each female was then checked for the presence of non-FM7a males. If non-FM7a males were found, the line was discarded. Several lines with lethal X chromosomes were recovered and screened for the deficiency by examining polytene chromosome prepara- tions. One line with a deficiency for 7D1-7D14-17 was chosen for further study and was called Dfll)DesiS3. It was also further tested by standard genetic crosses to other deficiencies and recessive lethal mutations in the region to ensure that no other lethal mutations were present on the X chromosome and that it failed to complement other deficiencies that include the same region (see also below).

Suppression of Triplo-lethal 245

T( i ; Y) Bi 70 males

BS

T(1;Y)Vib famalea - FM7,yw ‘ B

Y+

x 8A

FM7,yw * B - X

BS

Y \8A

RESULTS

Suppression of Triplo-lethal by overlapping (X;Y) translocations: Since the suppression of the triplo- lethal phenotype of Tpl observed by ROEHRDANZ and LUCCHESI (1 98 1) was mediated by very large portions of the X chromosome, we set out to test this effect using hyperploidy for a much more limited region of the X chromosome. If the suppression of Tpl were merely a nonspecific effect of X hyperploidy, we would expect that reducing the duplicated region would reduce the suppression, while if a specific locus were involved, then reducing the hyperploidy for irrelevant regions of the X should improve the viability of the rescued flies. A synthetic duplication of the cytological regions 6E-8A was created by using two different (X;Y) translocation stocks: T(I;Y)BI70, broken in 8A and p, and T(l;Y)Vl6, broken in 6E1-3 and p. For T(1;Y)BI 70, the distal portion of the X chromosome is marked with BS, and the proximal with y+. For T(I;Y)V16, the distal fragment of X is marked with y+ and the proximal with (see Figure 1) . By crossing these two stocks together and choosing progeny with the appropriate markers, we obtained females carry- ing one complete X , the distal portion of T(I;Y)BI70, and the proximal portion of T(I;Y)V16. These females are essentially disomic for the X chromosome, except that they have three copies of the region between the translocation breakpoints. They were then crossed to Dp(Tpl)/Df(Tpl) males. A small number of female progeny were observed, all carrying the appropriate markers to conclude that they were of the genotype indicated in Figure 1, having three copies of Tpl and three of the 6E-8A region. Some of these females

FIGURE 1 .-Suppression of Triplo- lethal by T(X;Y) translocations. Crossing the two translocations T(I;Y)B170 and T(J;Y)V16 gave females carrying the dis- tal part of B170 and the proximal part of V16. These could be distinguished from their siblings because they were yellow and B/B/B (triple Bar eyes). They were trisomic for the region 6E-8A, that is between the two translocation break- points. When these females were mated to pX.p, In(1)EN y;; Dp(3;3)Tp121 pP/ Dfl3R)18i77 p P males (Dp(Tpl)/Dj(Tpl)), a small number of surviving progeny were recovered. These survivors were females with B/B eyes and yellow body color which carried three copies of Tpl. The hollow bars represent the euchro- matic regions of chromosomes, gray rep- resents the Y chromosome, and the black dots represent centromeres.

eclosed, but were feeble and died within a few hours after eclosion. A number of other females were seen as dead pharate adults. We conclude from this exper- iment that flies with three doses of Tpl are rescued to a limited extent by the presence of three copies of the region of the X chromosome from 6E to 8A, and this effect does not require hyperploidy for any other region of the X chromosome. This observation sup- ports the hypothesis that the duplication of a small region of the X chromosome, possibly even a single X - linked locus, suppresses the triplo-lethality of Tpl. This locus was not named by ROEHRDANZ and LUCCHESI, but we call it Zsis (see Introduction).

Suppression of Triplo-lethal by Dp(l;2)~n+~“: We confirmed ROEHRDANZ and LUCCHESI’S ( 1 98 1 ) results with Dp(l;3)sn13aI (data not shown) which does not suppress the triplo-lethal phenotype of Tpl. We then used another available rearrangement involving this region, Tp(1;2)~n+”~. This rearrangement is a trans- position of the genetic material between 7A8 and 8A5 from the X to the second chromosome (see Figure 2). When Tp( l ;2)~n+~’~ males were mated to Dp(Tpl)/ Df(Tp1) females, a number of progeny survived to adulthood. These adult flies were males with yellow body color and wild-type eyes. Female progeny were not observed in this cross, since they carried the X - linked deficiency for 7A8-8A5 and therefore could not have had a net duplication for that region. While many of the survivors lived for only a few days, some of them lived long enough to mate and proved to be fertile. Test crosses verified that they carried Dp(1;2)~n+~’~ and Dp(3;3)Tp121. The simultaneous oc- currence of the Dp(l;2)~n”’~ and Dp(3;3)Tp121 chro-

246 D. R. Dorer et al.

(1;Y) Trmslocotlono:

Dupllcotlono:

6C 1 Dp(l;3)8nr3'' 17C9

7A8[ Dp(1;a.n + I Z d I 6A5

7D5-6[ DP2 ,

8A5 7 A 8 l 7 I DP(l;2)C~*oo'

8A5

6A I Dp(1;l)GHSS E;] 8A

Ddlclencles:

Df(l)Cl28 7 D 1 0 7D5,6

Df(l)hlC 7C1-2-1 7D19-22

Df(l)GE202 7D12-16 1 7 E 3 - 4

Df(l)RA2 7018-22 r ~~ I 8A4-5

Df(1)Ded s3 7 D 1 , 2 0 7D14-17

FIGURE 2.--Summary of the cytogenetic region surroundinglsis. The extents of some of the rearrangements tested for the presence of Isis are shown aligned with a portion of BRIDGES'S (1 935) map of the X chromosome. DJI)RA2 is shown according to our revised mapping (see text). The locations of a number of nearby l o c i , including the P(ry+) element described in the text, are indicated above the map. The gray regions at the ends of rearrangements indicate the limits of certainty for the breakpoint position. The duplications are tested for Isis by asking for suppression of the triplo-lethal phenotype of Tpl. The deficiencies are tested by asking if they block the suppression of Tpl by DP(l;2)Sn"'" (see RESULTS).

mosomes in a single fly was also observed by examin- ing the salivary gland chromosomes of a surviving third instar larva (data not shown).

New Zsis duplications induced by hybrid dysge- nesis: Two new tandem duplications which include the cytological region 7C were recovered by the anti- Minute technique of GRELL ( 1 969). Dysgenic female progeny from a x2 male X Oregon-R female cross were mated to DfTl)~n+~", f curly; D p ( l ; 2 ) ~ n + ~ ' ~ / b d males. The D f ( l ) ~ n + ~ ~ ~ , f car/+; + / b d FP females usually die, since they have only one copy of the severe Minute locus, M(1)7C (DORER, ANANE-FIREMPONG and CHRISTENSEN 1991; LEFEVRE and JOHNSON 1973). From an estimated 17,000 D f T l ) ~ n + ~ ~ ~ , f car/+; b d / + zygotes (the three other possible genotypes from this cross are viable and totaled 51,253 adults), 87 adult survivors were recovered. Of these, 85 were infertile and Minute, and thus were probably rare D f T l ) ~ n + ~ ~ ~ / + survivors. The other two survivors were not Minute. Both of these have duplications which were observed

TABLE 1

Generality of rescue of Df(Tpl)/TpI* by X s i s duplications

X chromosomes Tpl duplication

of survivors Dp(3;3)Tpl21 Dp(3;3)TplT"

DP(l;I)GH65 cross: DP(l;I)GH65 64 20 FM7c 0 0 Dp(l;I)GH65fYsX.F 8 78

DP(I;I)GH95 0 1 FM7c 0 0

Dp(l;l)GH95fYsX.F 34 237

Dp(3;3)Tp12IfDf(3R)lBi77 and YsX.YL;; Dp(3;3)TplT"fDf Females of the genotype Dp(I;I)fFM7c were mated to YsX.YL;;

(3R)18i77 males. Small numbers of exceptional progeny resulting from nondisjunction were also recovered but are not shown.

genetically and cytologically to be X-linked. Dp(l;l)GH65 is a tandem duplication of 6A to 8A, and is semiviable in males. Dp(l;l)GH95 covers a larger region, and therefore has breakpoints which are less informative about the location of Isis. Dp(l;l)GH95 is very poorly viable in males.

Dp(l ; l )GH65 and Dp( l ; l )GH95 were both tested for the presence of Isis. Dp(l;l)GH65/FM7c or Dp(1;l)- GH95/FM7c females were mated to Dp(3;3)Tp121/ DfT3R)l8i77 or Dp(3;3)TplT4'/DfT3R)18i77 males. DP(3;3)Tpl21 is a duplication of 83E-84B (DENELL 1976) while Dp(3;3)TplT4' (DORER and CHRISTENSEN 1990) is a duplication of 83A-83E. As shown in Table 1 , a significant number of adult survivors are re- covered with all four combinations of Isis and Tp1 duplications, demonstrating that the suppression of Tpl is a gene-dosage-dependent phenomenon and not due to a position effect of a particular rearrangement.

Suppression of Triplo-lethal by shortened deriva- tives of Dp(l ;2)~n+~'~: The experiments described above show that duplication of the region 7A8-8A5 suppresses the triplo-lethal phenotype of Tpl, but that the region from 7A8-7C9 is probably not necessary for the effect (this region is included in Dp(l ;3)snl3al , see Figure 2). To further localize the essential genetic material necessary for this effect we tested for the presence of Isis in shortened duplications derived from D p ( 1 ; 2 ) ~ n + ~ ~ ~ . The most significant duplication that we tested is known as Dp' (see MATERIALS AND METHODS). The region of the X chromosome which is duplicated in Dp' has the same proximal breakpoint (8A5) as Dp(1;2)~n'~'~ and the distal breakpoint is probably located between 7D1 and 7D5,6 (LINERUTH 1987; our unpublished data). This duplication sup- presses Tpl and must therefore include Isis. Similar results were obtained with Dp(1;2)~t+~", which was created in our laboratory by y-irradiation of D p ( 1 ; 2 ) ~ n + ~ ~ ~ (DORER, ANANE-FIREMPONG and CHRIS- TENSEN 199 1). These experiments have ruled out any

Dp(I;I)GH95 cross:

Suppression of Triplo-lethal 247

TABLE 2

Blocking Zsis suppression with X-linked deficiencies

X chromosomes Deficiency used

of survivors DxZ)C128 DflI)GE202 Dfll)RA2 Dfl1)W DflI)Dcsis'

D W ) 2 57 0 5 0 FM 7c 16 13 37 90 1

DJl)/YsX.YL 0 7 0 0 1

FM7c/YsX.YL 8 10 8 9 1 YSX. YL 7 2 4 21 13

The number of progeny resulting from crosses involving five different deficiencies of the X chromosome are tabulated. Df(l)/ FM7c; D p ( l ; Z ) ~ n + ' ~ ~ / + females were mated to YsX.YL, In(1)EN y;; Dp(Tpl)/D Tp1) males. All progeny were D p ( l ; Z ) ~ n + ~ ~ ~ / + ; Dp(Tpl)/ +. The f X.YL male progeny result from nondisjunction of the X chromosomes in the Dfll)/FM7c female parent.

region distal to 7D as the possible location for Isis (see Figure 2).

Mappinglsis using X-linked deficiencies: Since the suppression of triplo-lethality seems to be caused by the presence of an extra copy of a locus which maps within Dp(1;2 )~n+~'~ , it follows that the deletion of this locus from the X chromosome in a D p ( 1 ; 2 ) ~ n + ~ ' ~ / + ; Dp(Tpl)/+ male should prevent the suppression. We tested five deficiencies in the 7D-8A region (see Fig- ure 2). Each test cross for these deficiencies included an internal control for the efficiency of suppression using the FM7c balancer X chromosome. These crosses were complicated by the variable, but gener- ally poor fertility of males carrying Dp(l;2)~n"'~, and the incomplete penetrance of the suppression of Tpl by Zsis. Females of the genotype DJII)IFM7c; Dp(1;2 )~n+~'~ /+ were crossed to pX.p , In(1)EN y;; Dp(Tpl)/Df(Tpl) males. Half of the progeny from this cross inherit Df(Tp1) from the male and will die. Of the Dp(Tp1)-bearing progeny that remain, half of them inherit D p ( 1 ; 2 ) ~ n + ~ ' ~ from the female, and are poten- tial survivors, depending on their X chromosome(s). The recovery of FM7c; Dp(1;2)~n+~'~/+; Dp(Tpl) /+ male progeny is an indicator of a fertile cross, and the survival of DJI1)-bearing males indicates that the de- ficiency does not delete Isis. The results from these crosses are shown in Table 2. Males of the genotype Df(I)GE202; Dp(1;2)~n+~'~/+; Dp(Tpl)/+ were easily obtained (Table 2). This clearly rules out the chro- mosomal region from 7D12-13 to 7E3-4 as being required in an extra copy to suppress Dp(Tpl)/+. In contrast, no survivors were found with Df(I)RA2 (Table 2). Cytological examination of this deficiency suggests that the distal breakpoint of Dfl )RA2 may have been incorrectly assigned to 7D10 (LINDSLEY and ZIMM 1992). This deficiency is frequently associ- ated with asynapsis distal to the deficiency in hetero- zygous females, which makes the breakpoint difficult to determine, but it appears to be in 7D18-22, and it complements DfTI)DesiS3 which is deleted for 7D1 through 7D14-17 (see below). Figure 2 indicates our

revised estimate of its breakpoints, 7D18-22 and 8A4,5. Although we did a large number of crosses with DfTI)RA2, the number of control progeny was still small. The reasons for this are not entirely clear, but we have observed that stocks carrying both DJI1)RAB and D p ( l ; Z ) ~ n + ~ ' ~ are not as fecund as those carrying only one of these rearrangements. As further evidence for DJII)RA2 being Isis-, we also performed a cross using Df(I)RA2/DJIl)RA2; Dp(I;2)~n+~'~/ b d females and Dp(Tpl)/Df(Tpl) males. There were no survivors from this cross, in contrast to a similar experiment using Df(l)GE202/Df(l)GE202; Dp(1;2)~n+'~~/bzu" which produced many progeny. Taken together, these results suggest that Zsis is de- leted by Df(l)RA2, but not by Df(l)GE202, and is therefore located somewhere in the 7E3-8A5 region. Furthermore, we have never recovered any survivors bearing DJITpl) in these crosses. If deficiencies for Isis rescued flies carrying only one dose of Tpl, we would have seen them in these crosses as D f ( l ) R A 2 / p X . p , In(1)EN y; +/+; Df(Tpl)/+ females, and this did not occur, demonstrating that altered Isis dosage has no apparent effect on the haplo-lethal phenotype of Tpl.

A similar test cross was found to produce DfI)C128; D p ( l ; 2 ) ~ n + ~ ' ~ / + ; Dp(TPl)/+ progeny at a very low ratio compared to their FM7c-bearing siblings (see Table 2). However, in spite of the low rate at which flies bearing this deficiency are recovered, the recovery of any of these flies at all suggests that the region 7D1- 5,6 is not essential for suppression of triplo-lethality by Isis. We also tested the X-linked deficiency DJI1)hl". The results with Df1)hl" are similar to those with DJIl)C128. We have seen Df1)hE"; D p ( 1 ; 2 ) ~ n + ~ ' ~ / + ; Dp(Tpl)/+ adults, though they appear at a frequency far lower than the FM7c; Dp(1;2)~n+~'~/+; Dp(Tpl) /+ control flies which survive in the same crosses (Table 2). We also tested a new deficiency which we have synthesized called DJIl)DesiS3. DfTI)DesiS3 is a "designer deletion" (COOLEY, THOMPSON and SPRADLING 1990) induced between P elements located at 7D1 and

sive crosses of Df(l)DesiS3/FM7c; Dp(1 ;2 )~n+~'~ /+ to Dp(Tpl)/Df(Tpl) we recovered almost no survivors of any type, including the FM7c control. A total of 251 females were used in this experiment, but the overall viability of this mating appears to be very poor for reasons we do not fully understand (see DISCUSSION). Since DfT1)hl" and Df(I)DesiS3 are deficient for the same region that is missing in DJII)C128, it is perhaps not surprising that the recovery of survivors is affected in the same negative manner by all three deficiencies. However, the recovery- of even small numbers of progeny is good evidence that these deficiencies are Isis+, since the probability of recovering adults with three doses of Tpl is essentially nil (KEPPY and DENELL 1979; ROEHRDANZ and LUCCHESI 1980; DORER and

7D14-17 (See MATERIALS AND METHODS). After exten-

248 D. R. Dorer et al.

CHRISTENSEN 1990; our unpublished data). Interactions between Zsis and other mutations af-

fecting Tpl: Duplications of Zsis might directly repress the expression of Tpl, or they might act downstream to suppress the lethal effects of overexpression of Tpl. The Su(TP1) mutations described in the Introduction also permit the survival of flies carrying a duplication of Tpl (DORER and CHRISTENSEN 1990; ROEHRDANZ and LUCCHESI 1980). If either Su(Tp1) or Isis represses Tpl directly we might expect combinations of these to further reduce the effective Tpl dose, The net effect would be to mimic deficiencies of Tpl which should result in lethality. We used one representative muta- tion, Su(Tpl)'O, [previously called Tpl" (ROEHRDANZ and LUCCHESI 1980)], and were able to recover both males and females with the genotype Dp(l;2)~n+'~~/+; ri Su(Tp1)" Ki PI+, suggesting that Isis acts down- stream on the triplo-lethal effects of Tpl.

DISCUSSION

In their original description of the rescue of Dp(TPl)/Tpl+ by X chromosome hyperploidy, ROEHR- DANZ and LUCCHESI (1 98 1) proposed that duplications of 7DE would be sufficient to effect suppression of Tpl. However, all of the translocations successfully used in their experiments were hyperploid for either the distal or proximal regions of the X in addition to the 7DE area. Therefore, the possibility that suppres- sion of triplo-lethality was caused by the presence of large regions of the X chromosome was not conclu- sively ruled out prior to the experiments reported here. We have shown that the suppression of Tpl is probably mediated by a single locus in the 7E3-8A5 region of the X chromosome, which we have named Isis. This work is also relevant to the previously sug- gested model of Tpl as a component of the sex deter- mination pathway (BAKER and BELOTE 1983; DENELL 1976; LUCCHESI and MANNING 1987). The existing data do not support this model for the following reasons. (1) We have obtained sexually normal males and females with a variety of combinations of dosage of both Tpl and Zsis (above, and our unpublished data). (2) None of the known genes involved in sex deter- mination (including S x l ) are located in the Zsis region (CLINE 1988). (3) Tpl does not interact with known sex-determination genes (CHRISTENSEN and LUCCHESI 1988).

We have also used a variety of duplications of both Tpl and Zsis, which serve to narrow down the location of Zsis and also demonstrate that the effects seen here are not due to position effects of a particular dupli- cation, but are true gene dosage effects. The effect of Isis on Tpl resembles the "inverse effect" described by BIRCHLER and colleagues, wherein variation in the dosage of chromosomal segments inversely affects the expression of unlinked genes (BIRCHLER 1979; BIR-

CHLER, HIEBERT and KRIETZMAN 1989; DEVLIN, HOLM and GRICLIATTI 1988; RABINOW and BIRCHLER 1989; RABINOW, NGUYEN-HYUNH and BIRCHLER 199 1). However there are significant differences, es- pecially the failure of Zsis deficiencies to suppress Tpl haplo-lethality, and the fact that duplications and de- ficiencies of Isis are fully viable. Unfortunately, defi- ciency mapping of Zsis has been somewhat compli- cated. Although the results with Df(l)RA2 and Df(l)GE202 are fairly clear, the other three deficien- cies, Df(l)hlc, Df(l)DesiS' and Df(l)C128 somewhat weaken the conclusion that Zsis is a single locus in 7E3-8A4,5. These deficiencies clearly reduce the vi- ability of flies carrying three doses of Tpl and three doses of Zsis, but that could be a specific effect on Zsis or a nonspecific effect on viability. One possibility is that Zsis consists of two or more separate genes acting synergistically to suppress Tpl, and the various defi- ciencies in this region delete some but not all of the required sequences. As an alternative, the viability of the rescued flies is poor, and as a result they may be exquisitely sensitive to further genetic imbalances. Since we observe variability in the efficiency of the rescue by the various duplications of Zsis, as well as from experiment to experiment, we prefer the latter possibility, but we can not rule out the hypothesis that Zsis consists of redundant genetic information. On this point, an example of a repeated sequence in the lsis region was described by WARING and POLLACK (1 987). This sequence, XDm, is 372 nucleotides long, occurs in dispersed tandem arrays, and is present a total of 300-400 times in the genome. It appears to be local- ized almost exclusively to the middle of the euchro- matic portion of the X chromosome, primarily be- tween cytological divisions 4 and l l . Interestingly, one of the largest clusters appears to be in the 7E-8A region (WARING and POLLACK 1987). Whether this sequence has any relationship to Zsis (or if it has any function at all) is currently unknown, but could be tested as the molecular analysis of Tpl proceeds.

Although the Zsis-Tpl interaction is a gene dosage effect, it is asymmetric. Tpl is extremely sensitive to dosage alterations while the Isis region is apparently not. Zsis duplications suppress the triplo-lethal phe- notype of Tpl, but have no effect on the viability of flies with either one or two doses of Tpl. In addition, the lack of interaction of Zsis with the Su(Tp1) muta- tions suggests that Zsis does not act directly to repress Tpl expression. The Su(Tpl)'o/+ genotype contains two copies of Tpl and a suppressor of Tpl. The suc- cessful addition of a duplication of Zsis to this genotype suggests that if Zsis and Su(TP1)" act directly on Tpl, they do not act additively or synergistically to reduce Tpl expression, since that should mimic a deficiency of Tpl. I t seems more likely that Zsis acts downstream on the triplo-lethal effects of Tpl. This in turn suggests

Suppression of Traplo-lethal 249

that the triplo-lethal and haplo-lethal phenotypes of Tpl may be different from each other. We hope that our continuing analysis of Tpl will help to resolve these questions.

We are grateful to TERRY JOHNSON, ROB DENELL, STEPHAN ANDERSSON and KATHY MATTHEWS (of the Indiana University Drosophila Stock Center) for providing stocks, to BRIAN MARIANI for many helpful comments on the manuscript, and to MARY K. BARRY for keeping her comments to herself. This work was sup- ported in part by a grant from the National Institutes of Health to A.C.C., R29-GM38483. G.H. was supported in part by the Pilot Studies Program of the Jefferson Medical College Biomedical Re- search Support Grant. D.R.D. was supported in part by a predoc- toral fellowship from the Percival E. and Ethel Brown Foerderer Foundation.

LITERATURE CITED

ALONSO, M. C., and C. V. CABRERA, 1988 The achaete-scute gene complex of Drosophila melanogaster comprises four homologous genes. EMBO J. 7: 2585-2591.

ANDERSON, S., and A. LAMBERTSSON, 1990 Characterization of a novel Minute-locus in Drosophila melanogaster: a putative ribosomal protein gene. Heredity 65: 5 1-57.

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

BAKER, B. S., and J. M. BELOTE, 1983 Sex determination and dosage compensation in Drosophila melanogaster. Annu. Rev. Genet. 17: 345-393.

BIRCHLER, J. A,, 1979 A study of enzyme activities in a dosage series of the long arm of chromosome one in maize. Genetics 92: 1211-1229.

BIRCHLER, J. A., J. C. HIEBERT and M. KRIETZMAN, 1989 Gene expression in adult metafemales of Drosophila melanogaster. Genetics 122: 869-879.

BRIDGES, C. B., 1935 Salivary chromosome maps with a key to the banding of the chromosomes of Drosophila melanogaster. J. Hered. 26: 60-64.

CHRISTENSEN, A. C., and J. C. LUCCHESI, 1988 Failure of the Triplo-lethal locus of Drosophila melanogaster to interact with sex-determining genes. Drosophila Inform. Serv. 67: 15.

CLINE, T. W., 1988 Evidence that sisterless-a and sisterless-b are two of several discrete “numerator elements” of the X/A sex determination signal in Drosophila that switch Sxl between two alternative stable expression states. Genetics 1 1 9 829-862.

COOLEY, L., D. THOMPSON and A. C. SPRADLING, 1990 Constructing deletions with defined endpoints in Dro- sophila. Proc. Natl. Acad. Sci. USA 87: 3170-3173.

CRAYMER, L., and E. ROY, 1980 New mutants-D. melanogaster. Drosophila Inform. Serv. 55: 200-204.

DELIDAKIS, C., A. PREISS, D. A. HARTLEY and S. ARTAVANIS- TSAKONAS, 1992 The Enhancer of split [E(spl)] locus of Dro- sophila encodes seven independent helix-loop-helix proteins. Proc. Natl. Acad. Sci. USA 89: 8731-8735.

DENELL, R. E., 1976 The genetic analysis of a uniquely dose- sensitive chromosomal region of Drosophila melanogaster. Ge- netics 84: 193-2 IO.

DEVLIN, R. H., G. HOLM and T. A. GRIGLIATTI, 1988 The influ- ence of whole-arm trisomy on gene expression in Drosophila. Genetics 118: 87-10],

DORER, D. R., A. ANANE-FIREMPONG and A. C. CHRISTENSEN, 199 1 Ribosomal protein SI 4 is not responsible for the Minute phenotype associated with the M(1)7C locus in Drosophila mel- anogaster. Mol. Gen. Genet. 2 3 0 8-1 1 .

DORER, D. R., and A. C. CHRISTENSEN, 1990 The unusual spec- trum of mutations induced by hybrid dysgenesis at the Triplo- lethal locus of Drosophila melanogaster. Genetics 125: 795-801.

ENCELS, W. R., C. R. PRESTON, P. THOMPSON and W. G . EGGLES- TON, 1986 In situ hybridization to Drosophila salivary chro- mosomes with biotinylated DNA probes and alkaline phospha- tase. Focus 8: 6-8.

EPSTEIN, C. J., 1988 Specificity versus nonspecificity in the path- ogenesis of aneuploid phenotypes. Am. J. Med. Genet. 29:

GRELL, E. H., 1969 Induction of duplications of genes which specify enzymes in Drosophila melanogaster. Genetics 61 (Suppl.): s23.

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

KEPPY, D. O., and R. E. DENELL, 1979 A mutational analysis of the triplo-lethal region of Drosophila melanogaster. Genetics 91:

LEFEVRE, G., 1981 The distribution of randomly recovered X- ray-induced sex-linked genetic effects in Drosophila melano- gaster. Genetics 99: 461-480.

LEFEVRE, G., and T. K. JOHNSON, 1973 Evidence for a sex-linked haplo-inviable locus in the cut-singed region of Drosophila mel- anogaster. Genetics 74: 633-645.

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

LINDSLEY, D. L., L. SANDLER, B. S. BAKER, A. T. C. CARPENTER, R. E. DENELL, J. C. HALL, P. A. JACOBS, G. L. G. MIKLOS, B. K. DAVIS, R. C. GETHMANN, R. W. HARDY, A. HESSLER, S. M. MILLER, H. NOZAWA, D. M. PARRY and M. COULD-SOMERO, 1972 Segmental aneuploidy and the genetic gross structure of the Drosophila genome. Genetics 71: 157-1 84.

LINERUTH, K., 1987 An Ovarian Protein Locus and Its Deuelop- mentally Regulated Expression in Drosophila. Ph.D. Thesis, Uni- versity of Umei, Sweden.

LUCCHESI, J. C., and J. E. MANNING, 1987 Gene Dosage Compen- sation in Drosophila melanogaster. Adv. Genet. 24: 371-429.

RABINOW, L., and J. A. BIRCHLER, 1989 A dosage-sensitive mod- ifier of retrotransposon-induced alleles of the Drosophila white

RABINOW, L., A. T . NGUYEN-HYUNH and J. A. BIRCHLER, 1991 A trans-acting regulatory gene that inversely affects the expres- sion of the white, brown, and scarlet loci in Drosophila. Genetics

ROBERTSON, H. M., C. R. PRESTON, R. W. PHILLIS, D. M. JOHNSON- SCHLITZ, W. K. BENZ and W. R. ENGELS, 1988 A stable genomic source of P element transposase in Drosophila mela- nogaster. Genetics 118: 461-470.

ROEHRDANZ, R. L., and J. C. LUCCHESI, 1980 Mutational events in the triplo- and haplo-lethal region (83DE) of the Drosophila melanogaster genome. Genetics 95: 355-366.

ROEHRDANZ, R. L., and J. C. LUCCHESI, 1981 An X chromosome locus in Drosophila melanogaster that enhances survival of the triplo-lethal genotype, Dp-(Tpl). Dev. Genet. 2: 147-1 58.

ROIHA, H., G. M. RUBIN and K. O’HARE, 1988 P element inser- tions and rearrangements at the singed locus of Drosophila melanogaster. Genetics 1 1 9 75-83.

SPRADLING, A. C., and G. M. RUBIN, 1983 The effect of chro- mosomal position on the expression of the Drosophila xanthine dehydrogenase gene. Cell 34: 47-57.

WARING, G. L., and J. C. POLLACK, 1987 Cloning and character- ization of a dispersed, multicopy, X chromosome sequence in Drosophila melanogaster. Proc. Natl. Acad. Sci. USA 84: 2843- 2847.

161-165.

421-441.

IOCUS. EMBO J. 8: 879-889.

1 2 9 463-480.

Communicating editor: R. E. DENELL