Copyright(cid:211) 1998bytheGeneticsSocietyofAmerica Transvection in the Drosophila Abd-B Domain: Extensive Upstream Sequences Are Involved in Anchoring Distant cis-Regulatory Regions to the Promoter La´szlo´ Sipos,* Jo´zsef Miha´ly,*,† Franc¸ois Karch,† Paul Schedl,‡ Ja´nos Gausz* and Henrik Gyurkovics* *HungarianAcademyofSciences,BiologicalResearchCenter,InstituteofGenetics,H-6701Szeged,Hungary,†UniversityofGeneva,Department ofZoologyand AnimalBiology,CH-1211Geneva4,Switzerlandand ‡DepartmentofBiology,PrincetonUniversity,Princeton,NewJersey08544 ManuscriptreceivedJune12, 1997 Acceptedfor publicationMarch 2, 1998 ABSTRACT The Abd-B gene, one of the three homeotic genes in the Drosophila bithorax complex (BX-C), is requiredfor the properidentityof the fifth throughtheeighth abdominal segments (corresponding to parasegments 10–14) of the fruitfly. The morphological difference between these four segments is due to the differential expression of Abd-B, which is achieved by the action of the parasegment-specific cis- regulatoryregionsinfra-abdominal-5(iab-5),-6,-7and-8.Thedominantgain-of-functionmutationFrontab- dominal-7 (Fab-7)removesa boundary separating twoof thesecis-regulatory regions,iab-6and iab-7.Asa consequence of the Fab-7 deletion, the parasegment 12- (PS12-) specific iab-7 is ectopically activated in PS11. This results in the transformation of the sixth abdominal segment (A6) into the seventh (A7) in Fab-7 flies.Herewe reportthat pointmutationsof theAbd-B gene in transsuppressthe Fab-7phenotype in a pairing-dependent manner and thus represent a type of transvection. We show that the observed suppressionis theresultof trans-regulationof thedefectiveAbd-B gene bytheectopically activatediab-7. Unlikepreviously demonstratedcases of trans-regulation in theAbd-B locus, trans-suppression ofFab-7 is sensitivetoheterozygosityforchromosomalrearrangementsthatdisturbhomologouspairingatthenearby Ubx locus. However, in contrast to Ubx, thetransvection weobserved in the Abd-B locusis insensitive to theallelicstatusofzeste.AnalysisofdifferentdeletionallelesofAbd-Bthatenhancetrans-regulationsuggests that an extensive upstream region, different from the sequences required for transcription initiation, mediatesinteractionsbetweentheiabcis-regulatoryregionsandtheproximalAbd-Bpromoter.Moreover, wefind thatthe amountof DNAdeletedin theupstreamregionisroughlyproportionalto thestrength oftrans-interaction,suggesting that thisregionconsistsof numerousdiscrete elementsthatcooperatein tethering the iab regulatory domains to Abd-B. Possible implications of the tethering complex for the regulation ofAbd-B are discussed.Inaddition,wepresent evidencethat thetenacityof trans-interactions in theAbd-B gene mayvary, depending uponthe tissue andstageof development. IN eukaryotes, gene activity can be controlled by ex- such long-distance regulatory interactions is provided tensiveregulatoryregionsthatarelocatedmanykilo- by the phenomenon of transvection in Drosophila. bases away from the promoter. In the most widely ac- Transvection, first described by Lewis (1954) in the cepted model for such long-distance interactions, the Ultrabithorax (Ubx) locus, refers to a partial interallelic intervening sequences between the regulatory regions complementationthatdependsuponthepairingofho- andthepromoterarethoughtto“loopout”(Pirrotta mologouschromosomes.Inthebest-documentedcases, 1991). Although the looping model can account for such as in the yellow (Geyer et al. 1990) and the Ubx long-distance interactionsbetweenregulatoryelements loci (Martinez-Laborda et al. 1992), transvection ap- andpromoters,italsoposesaproblem.Mostregulatory pears to involve trans-regulation, that is, a regulatory elements are rather promiscuous in their interactions element on one of the homologues controls the pro- andarecapableofcontrollingtheactivityofmanydiffer- moteractivityof thecorrespondinggeneontheother, ent promoters, irrespective of their origin. Thus, the paired homologue. This unusual trans-interaction can looping model raises the question of how enhancers be exploited to learn more about the mechanisms re- are able to distinguish their target promoter from the sponsible for specifying conventional interactions be- promoters of other nearby genes. tween regulatory elements and promoters in cis. TheUbxgeneisapartofthehomeoticbithoraxcom- Onemodelsystemforstudying thefactorsgoverning plex (BX-C). Although BX-C contains only two other homeotic genes, abdominal-A (abd-A) and Abdominal-B (Abd-B) (Sa´nchez-Herrero et al. 1985; Tiong et al. Corresponding author: Henrik Gyurkovics, Hungarian Academy of 1985), these three genes assign proper identity to the Sciences, BiologicalResearchCenter,Institute ofGenetics, P.O.B. third thoracic (T3) and all of the abdominal (A1–A9) 521, H-6701Szeged,Hungary. E-mail:[email protected] segments, corresponding to parasegments (PS) 5 to Genetics149:1031–1050(June,1998) 1032 L.Siposet al. PS14. This is achieved through the control exerted by lectedfrom17-hregglaysweredechorionated,fixedanddevi- tellinized according to the procedure of Mitchison and PS-specific cis-regulatory elements on the individual Sedat(1983),modified asdescribedin Karchetal. (1990). genes of the complex. Thus, abx/bx and pbx/bxd ele- We modified this procedure further asfollows. The time of mentsregulateUbxexpressioninPS5andPS6(Beachy fixationwasreduced to12 min. Devitellinizedembryos were et al. 1985; Hogness et al. 1985; White and Wilcox washedandrehydratedin1(cid:51)PBT(1(cid:51)PBT(cid:53)1(cid:51)phosphate- 1985), and iab-2, -3 and -4 elements regulate abd-A in buffered saline (cid:49) 0.1% Triton X-100 (cid:49) 0.1 bovine serum PS7,PS8andPS9,respectively(Karchetal.1990;Mac- albumin,1(cid:51) phosphate-bufferedsaline (cid:53)137 mmNaCl,2.7 ias et al. 1990). iab-5 - iab-8 elements regulate the Abd-B mm KCl, 10.1 mm Na2HPO4, 1.8 mm KH2PO4; pH (cid:53) 7.5). Embryoswereincubatedwithprimaryantibody(monoclonal class A transcription unit that corresponds to the Abd-B mouse-anti-ABD-B1A2E9;Celnikeretal.1990)dilutedin1(cid:51) mfunction(Casanovaetal.1986)inPS10throughPS13 PBT 1:1 on a rotating wheelovernight at 4(cid:56)C.Antibody was (Celniker et al. 1990; Boulet et al. 1991; Sa´nchez- removed, and embryos were washed six times with PBT for Herrero1991).Specificregulatoryelementsthatregu- about20minonrotatingwheel.Horseradishperoxidase-con- late the longer transcription units (class B, C and (cid:103); jluugteadteidn rparbebchiti-lalnedti-(m4o(cid:56))uPseBTse1co:2n0d0a,rwyasanadtidbeoddyan(dDAinKcOub)a,teddi- Zavortnik and Sakonju 1989), corresponding to the withtheembryosonarotatingwheelfor31/2hrat4(cid:56).These Abd-B r subfunction (Casanova et al. 1986), have not conditions reduced background staining without significant yetbeenidentified,buttheyareexpectedtospecifythe loss of specificstaining.Embryos were thenwashed againas expressionpattern oftheclass B, Cand(cid:103)RNAspecies described above. After removing PBT, embryos were briefly rinsedtwiceinstainingbuffer(0.1mcitricacid, 0.05mNH- in PS14 andPS15. 4 acetate,pH(cid:53) 5.7adjustedwith NHOH)and stainedin the Recently,anunusualtypeoftrans-regulationhasbeen mixture of 980 (cid:109)l buffer, 20 (cid:109)l DA4B solution (25 mg/ml describedfortheAbd-BclassAtranscriptionunitandits stocksolution,finalcc0.5mg/ml)and2(cid:109)lHO (30%stock regulatory regions (Hendrickson and Sakonju 1995; solution,finalcc0.06%).Thereactionwasstop2ped2 bydilution Hopmannetal.1995).Incontrasttotheclassicaltrans- with PBT and embryos were rinsed six times with PBT. Em- vection of Lewis (1955), this trans-regulation is ex- bryoswere stored in PBT with 0.05% NaN3 and mounted in 9:1 glycerol:10 (cid:51) PBT. For the dissection of CNS, we used tremelyresistant to disruption of homologouspairing. tungstenneedles. Inthisarticle,weprovideevidencethattrans-interaction Chromosomes generated by recombination: McpB116iab-7Sz intheAbd-Bgenemayvaryinitsresistancetodisruption Abd-BD16:The synthesisof this chromosome wasdone in three of homologous pairing in different tissues. With the consecutive steps.First, we generated the chromosome McpB116 helpofadominantgain-of-functionmutation,Fab-7,we iab-7Szbyrecombination. Out of (cid:122)29,000male progenyof the crossbetweenMcpB116/iab-7SzvirginsandOregon-Rmales,weiden- show that transvection of a more regular type, closely tifiedtwoflieswithdarklypigmentedA4[duetothedominant resembling theclassical case in the Ubxlocus, can also mutation,Miscadastralpigmentation(Mcp)]andanadditionalru- be demonstrated in Abd-B in the adult stage. Detailed dimentaryseventhtergite(iab-7(cid:50)).Inthesecondstep,Microcepha- analysis of this transvection suggeststhe existence of a lus(Mc) (adominantmutationresultingin astrongreduction multicomponenttetheringmechanismthatmayensure of the head capsule) was recombined onto the McpB116 iab-7Sz chromosome.McpB116iab-7Sz/Fab-7McvirginswerematedwithOre- cis-autonomyoftheAbd-BdomainwithintheBX-C.This gon-Rmales.Among(cid:122)35,000F males,wefoundthreerecombi- mechanism involves an extensive region upstream of 1 nantsshowingMcp,iab-7andMc,butnoFab-7phenotype.Finally, the Abd-B gene, which appears to be distinct from the weisolatedrecombinantsbetweenthechromosomesMcpB116iab- sequences required for transcription initiation. 7SzMcandAbd-BD16.(McpandMcservedasflankingmarkersto detectrecombinationevents.)Dp(3;1)bxd111/(cid:49);Abd-BD16/McpB116 iab-7SzMcvirginswereallowedtomatewithOregon-Rmalesand theirmale progeny((cid:122)90,000) were scoredforthepresenceof MATERIALS AND METHODS Mcp and the absence of Mc phenotype. Of the 39 such males General procedures: Fly stocks were maintained on stan- found, eight were sterile and 23 did not carry Abd-BD16 [based dardyeast-cornmeal medium.Crosseswereperformedat25(cid:56) oncomplementationwithDf(3R)P9].Theremainingeightlines enmasse,unlessotherwisenoted.Unlessdescribedinthisarti- carryingtheAbd-BD16weretestedfortheirabilitytosuppressthe cle, all genetic variants used are described in the following Fab-7 phenotype. Although all of them showed suppression, references:iab-7MX2,Abd-BD14,Abd-BD16,Df(3R)C4(Karchetal. theyfell intotwodiscrete categories:fiveofthem hadanA6 1985);Abd-BRD18,Df(3R)U110 (Hopmannetal.1995);Abd-BS1, tergite larger than the thin A7(similar to the phenotype of Abd-BS4(Tiongetal.1985);Abd-BR41,Fab-7,In(3R)Fab-7iab-7R7, Fab-7/Abd-BD16),whereasintheremainingthreethesetergites Df(3R)R59(Gyurkovicsetal.1990); iab-7164(Celniker etal. wereequallylargebothin A6and inA7. Weconcluded that 1990);iab-7Sz(Gallonietal.1993);McpB116(Karchetal.1994); thefirstgroupcarriedMcpB116only,andthesecondgrouphad TM3SbP(ry(cid:49))(cid:68)2-3(Reuteretal.1993);UC21-10,1-d(McCall bothMcpB116 andiab-7Sz.Weconfirmed thepresence of both etal.1994); Cbx1,Df(3R)P9,Dp(3;3)P5,In(3LR)TM1, In(3LR)- lesionsbySouthernblot analysis.The sameprotocolwasfol- TM3 Sb Ser, In(3LR)TM6B Tb, In(3LR)TM6C Sb, Mc, Tp(3;1) lowed in the generation of recombinant chromosomes that bxd111,Ubx1, z1,za,zop,zv77h(Lindsley andZimm 1992). In(3R) containedAbd-BD14 instead of Abd-BD16. Fab-7iab-7R5was isolatedas an X-ray-induced revertantof Fab-7, Fab-7 Abd-BD16: Two apparently contradictory observations anditsbreakpointwithintheBX-Cwasclonedanddetermined promptedustostudytransvectionintheAbd-Bdomain.First, bySouthernanalysis.Cytologicalanalysisofpolytenechromo- we found that (when the homologues are paired) an Abd-B somes was performed as described by Ashburner (1989). point mutation trans to the Fab-7 mutation suppresses the AdultabdominalcuticlesweremountedasdescribedbyDun- Fab-7 gain-of-function phenotype so that the A6 tergite in can (1982), and Southern blot analyses were done as de- males is larger than the A7 tergite (trans-suppression, de- scribedbyBenderetal. (1983). scribedindetailintheresultssection).Second,wenoticed Immunohistochemical staining of embryos: Embryos col- in an earlier study (Gyurkovics et al. 1990) that when the Transvection inthe Abd-B Domain 1033 Fab-7 mutationispresentbothin cisandin transtoanAbd-B totheFab-7mutation.]Fab-7UC21-10,1-d/(cid:49)fliesshowamoder- mutation (Fab-7/Fab-7 Abd-BR41) the A6 and A7 tergites are ateFab-7phenotypethatisstronglyreducedifthesechromo- equally thin. We presumed that this phenotypic difference somes are over an Abd-B null mutation, suggesting that the between the two types of mutant combinations was not due Fab-7 phenotype originates mainly from an interaction with tosome unusualpropertiesofAbd-BR41(whichisanAbd-Bm(cid:50) the paired wild-type Abd-B gene in trans. However, the re- mutation,Gyurkovicsetal.1990),buttothepresenceofthe maining weak Fab-7 phenotype indicates that UC21-10,1-d is Fab-7mutationin cistoAbd-BR41.Totestifthis isthecase,we aleakyAbd-Bmutation.Impreciseexcisionsoftheinsertthat attempted to isolate recombinants carrying both Fab-7 and result in the deletion of the promoter region of the Abd-B Abd-BD16 (an Abd-B mr null mutation;Boulet etal. 1991)on classAtranscriptionunitshouldeliminatetheremainingFab- the samechromosome,basedontheassumptionthat Fab-7/ 7phenotypeinfliesheterozygouswiththenullmutationAbd- Fab-7Abd-BD16 fliesshouldhave a phenotypesimilarto Fab-7/ BD16.(The excision of theP element can be detectedby the Fab-7 Abd-BR41. For this purpose, ry506Fab-7/Abd-BD16 females lossof thery(cid:49)markergene carriedbythe UC21-10,1-dinser- werecrossedtory506Fab-7/ry506Fab-7males,andthemaleprog- tion.) We therefore isolated nine ry(cid:50) mutations with no re- eny were scoredfor the presence of equally thin A6and A7 mainingFab-7phenotypefromtheprogeny((cid:122)20,000)ofthe tergites. Out of (cid:122)35,000males, threesuch flies were found. crossbetweenFab-7UC21-10,1-d/TM3SbP(ry(cid:49))(cid:68)2-3malesand After establishing stocks of these lines, the presence of the Dp(3;1)bxd111/Dp(3;1)bxd111; ry506Abd-BD16/ry506 Abd-BD16 virgins. Fab-7 deletion was confirmed by Southern blot analysis and Incontrast,allnineputative5(cid:57)deletionsshowedsomedegree the presence of Abd-BD16 bygenetic complementation test in oftransformationofA6towardA7(Fab-7phenotype)intrans all three putative recombinants. Having established the use- towild-typechromosomes.Thisphenotypeallowedustosepa- fulnessofthismethod,weconstructeddoublemutantcombi- rate the chromosomes carrying the new derivatives of the nations of Fab-7 with Abd-BD14 and UC21-10,1-d in the same insertionalmutationfromthery506Abd-BD16chromosomeand way. toestablishstocks.Twoofthem,Abd-BPSz1andAbd-BPSz2,showed R5LR7R:Fortheconstructionofthischromosome,wetook significant complementation over iab-7Sz, i.e., the size of A7 advantageof thefactthat thedesiredproductof recombina- wasreducedcomparedto Abd-BD16/iab-7Sz,.Southernanalysis tion between the two inversions In(3R)iab-7R7 and In(3R)iab- detectedadeletionof10.8kbbetweenmappositions(cid:49)155.6– 7R5 should contain a deletion in the region 87C1-2;87D1-4, 156.8 and (cid:49)165.6–166.6(3.2 kbdownstream and 7.6 kbup- includingthekarmoisin(kar)locusat87C8(Gauszetal.1979). streamof the site of the original insertion) in Abd-BPSz1, and Tp(3;1)bxd111/(cid:49) ; iab-7R5/iab-7R7 females were crossed to cu a 5.5-kb deletion between map positions (cid:49)155.6–156.8 and kar/cukarmales,andtheprogenywerescoredforkarpheno- (cid:49)160.2–161.3(3.2kbdownstreamand2.3kbupstreamofthe type.Of 180flies scored,we recoveredsevenkar individuals. siteof theoriginal insertion)in Abd-BPSz2. Stocks of these putative recombinants were established and checked cytologically. All of them contained the inversion In(3R)87C1-2;89E3-4andthe deficiency Df(3R)87C1-2;87D1-4. RESULTS Isolation of rearrangements that eliminate trans-suppres- In adult wild-type males the program specifying the sion:We induced chromosomal rearrangements on the Fab-7 development of the A7 abdominal segment does not chromosome by irradiating Fab-7/Fab-7 homozygous males with Xrays(4000rads;1000rads/min,0.5-mmAl filter)and produce a visible tergite or sternite (Figure 1a). This crossing them to Abd-BD16/TM6B Tb virgins. After 6days the developmental program depends on the normal func- parents were discarded. Among the male progeny (about tioningoftheiab-7cis-regulatorydomainthatisrespon- 3000) we identified 12 flies in which trans-suppression was sibleforgeneratingtheappropriatelevelofAbd-BRNA/ weaker or no longer visible. We then tested these new re- protein expression in PS12/A7. When iab-7 is deleted arrangements over the Cbx1Ubx1 chromosomes. All but one showedsignificantreductionoftheweakdominantwingphe- on both homologues, Abd-B expression in PS12/A7 notypegeneratedbytheCbx1mutationthroughthemisexpres- comes under the control of theiab-6 cis-regulatory do- sion of Ubx(cid:49) on the homologous chromosome. Cytological main, resulting in thetransformation of PS12/A7 into examinationoftheexceptionalcaserevealedabreakpoint at a copy of PS11/A6 (Galloni et al. 1993). Because of the BX-C, 89E1,2. By complementation analysis, we showed thehaplo-insufficiencyoftheAbd-Bhomeoticgene,dis- thatthebreakpointinactivatedtheiab-4cis-regulatoryregion. Weisolatedtransvection-disruptingrearrangementsonthe ruptions in the normal developmental program of Fab-7 Abd-BD16 chromosome ina similar way.IrradiatedFab-7 PS12/A7areevidentinanimalsheterozygousforaniab- Abd-BD16/TM6C Sb males were crossed to Cbx1Ubx1/TM1 fe- 7mutation,suchasiab-7Sz,whichdeletestheiab-7region males.Outof(cid:122)1500F1fliesweisolatedfiveindividualsshow- (Gallonietal.1993).Inthiscase, A7assumes aniden- ing significantly reduced or no Cbx phenotype. Cytological tityinbetweenthatofthenormalA6andA7.Themixed examinationofthetwomutationsthatcompletelyeliminated the Cbx phenotype, TSR-11A Fab-7 Abd-BD16 and TSR-59A identity of A7 is manifested as a rudimentary tergite- Fab-7 Abd-BD16, showed the presence of the rearrangement likestructureposteriortoA6withacharacteristicshape T(3;4)89A-B;102A in the first, and Df(3R)88D;89D (cid:49) andsize(Karchetal. 1985). Thisphenotypic transfor- Tp(3;2)89E;98C;39A inthe secondmutation. mationisshowninFigure1bandsummarizeddiagram- Isolationofnew deletionsinthe Abd-Bgene by P-element maticallyinFigure2.AsimilareffectonA7development remobilization: In order to generate deletions that remove is also observed in animals heterozygous for an Abd-B sequencesaroundthepromoteroftheAbd-BclassAtranscrip- tion unit, theP-elementinsertion UC21-10,1-d,localized 253 mutation;however,inthiscasethedisruptionsindevel- bpupstreamof theproximalAbd-B promoter(McCall etal. opment are not restricted to A7, and the morphology 1994), was mobilized by introducing the transposase source oftheadjacentsegments A6andA5arealteredaswell. P(ry(cid:49))(cid:68)2-3.First,theFab-7mutationwasrecombinedontothe A phenotype exactly the opposite of that produced chromosome carrying theinsert. [Incontrast to theoriginal by iab-7 mutations is observed for the gain-of-function chromosome, these recombinants survive as adult homozy- gotes, suggesting that the reported lethality (McCall et al. mutationFab-7;PS11/A6is transformedintoPS12/A7. 1994)shouldbeduetosecond-sitemutationslocatedproximal This is due to the ectopic activation of iab-7 in PS11/ 1034 L.Siposet al. A6. Asa consequence,the number of visible segments notype of Fab-7 in A6. This partial suppression is not in Fab-7 homozygous males is reduced from six to five duetosomeunusualpropertiesoftheAbd-BD16;precisely (Gyurkovics et al. 1990). Although disruptions in A6 the same A6 phenotype was observed when other Abd-B development are also found in Fab-7/(cid:49) males, the ex- point mutations (Abd-BS1, Abd-BS4; Tiong et al. 1985) tentofthetransformationis reducedcomparedtothat were combined with Fab-7 (data not shown). observed in homozygous mutant animals. Instead of These observations could mean that the activity of the PS12/A7 identity seen in homozygotes, PS11/A6 Abd-B is “more haplo-insufficient” in A6 than in A7. If assumesamixedPS11/A6-PS12/A7 identityin animals this were true, we would expect to observe a similar heterozygousfortheFab-7mutation.Infact,themixed suppressionoftheFab-7phenotypeinA6bythedeletion identity phenotype of PS11/A6 observed in Fab-7/(cid:49) Df(3R)P9, which removes the entire BX-C, including males closely resembles the mixed identity phenotype the Abd-B gene.However,as was observedfor theFab-7/ ofPS12/A7observediniab-7(cid:50)/(cid:49)(orAbd-B(cid:50)/(cid:49))male iab-7Sz combination, segments A6 and A7 in Fab-7/ flies(see Figure 1, c and b). Df(3R)P9malesshowthesameintermediateA6-A7phe- The incomplete transformation of PS11/A6 in notype (see Figure 1f). Similarly, the combination of Fab-7/(cid:49) males could be attributed to a haplo-insuffi- Fab-7 with a deletion, Df(3R)R59, which removes both ciencyforiab-7.InFab-7/(cid:49)males,theiab-7cis-regulatory iab-6 and iab-7 but not the Abd-B gene, has the same domainontheFab-7-containinghomologue isactive in phenotypeas Fab-7/iab-7Sz(datanotshown), indicating PS11/A6, whereas the iab-7 cis-regulatory domain on thatthefunctioningofiab-6intransisirrelevantforthe thewild-typehomologueisnot.Ifregulationonlyoccurs phenotype of Fab-7 in A6. in cis, Abd-B expression in PS11/A6 should be driven Chromosomal rearrangements eliminate the effect byinteractions betweentheiab-7 cis-regulatorydomain of the Abd-B point mutations: The suppression of the andtheAbd-BgeneonlyontheFab-7homologue.Such Fab-7 gain-of-function phenotype by Abd-B point muta- interactions should not occur on the wild-type homo- tions in trans implies that our original assumption, logue in PS11/A6 because iab-7 is inactive. This inter- namely,thatregulatoryinteractionsonlyoccur incis, is pretation is supported by the phenotype of Fab-7/iab- incorrect. Instead, it suggests that some type of trans- 7Szmutantanimals.Inthisgenotype,onlyasinglecopy regulatoryinteractionor“transvection”mustalsooccur. oftheiab-7cis-regulatorydomainshouldbeavailableto Inparticular,atrans-regulatoryinteractioncouldpoten- drive Abd-B expression, not only in A6 and butalso in tially explainwhy theFab-7 gain-of-function phenotype (cid:49) A7. In males carrying this mutant combination, both ofFab-7/Abd-B animals isstronger thanthatof Fab-7/ (cid:50) PS11/A6 andPS12/A7would be expectedto assume an Abd-B animals; in the former case, the Abd-B genes identityinbetweenthatofPS11/A6andPS12/A7.This in cis and in trans would produce functional products, is the case. Moreover, as illustrated in Figure 1d and whereas in the latter case, functional products would Figure2,segmentsA6andA7intheFab-7/iab-7Szmales onlybe produced by the Abd-B in cis. closelyresembletheA7segmentfoundiniab-7(cid:50)/(cid:49)(Fig- Transvection is classically tested by generating chro- ure 1b)males. mosomalrearrangementsthatdisturbhomologouspair- Taken together,these results would seemto suggest ing.Forthispurpose,wefirstscreenedforX-ray-induced that,iftherestoftheAbd-Bdomainremainsunchanged, mutationsontheFab-7chromosome,whichreducedor thephenotype/identityofPS12/A7(orPS11/A6inthe eliminatedthesuppressionoftheFab-7gain-of-function Fab-7mutant) directlyreflectsthe number ofactiveiab-7 phenotypebyAbd-BD16,thatis,forFab-7*/Abd-BD16males cis-regulatorydomains. inwhich theA6 tergitemore closely resembledthe A7 Fab-7issuppressedbyAbd-Bpointmutationsintrans: tergite. [We used Abd-BD16 as a representative allele in We nextexamined theeffects of combining Fab-7 with thisandsubsequentexperimentsbecauseitisamolecu- anAbd-Bpointmutation,Abd-BD16.Ifourhypothesisthat larly characterized mutation that does not code for a the identity of PS11/A6 and PS12/A7 depends upon detectableprotein(Bouletetal.1991).]Outof(cid:122)3000 the number of iab-7 domains that are active in these mutagenized chromosomes,we recovered 12 indepen- twoparasegmentsis correct, then Fab-7/Abd-BD16 males dentFab-7*/Abd-BD16malesin whichtheA6tergite was wouldbeexpectedto havethesameintermediate phe- as thin or nearly as thin as the A7 tergite. For further notype in A6 as was observed in Fab-7/iab-7Sz males. analysis, we established stocks of these 12 Fab-7* chro- However, as illustrated in Figure 1e and Figure 2, this mosomes. is not the case. Although A7 has the same phenotype Chromosomalrearrangementsthatinterferewiththe inthetwomutant combinations,thephenotype ofseg- postulatedtransvectionbetweenFab-7 andAbd-B might ment A6 in Fab-7/Abd-BD16 males differs from that in also beexpected to disrupt pairing-dependent interac- Fab-7/iab-7Szmales.TheA6tergiteintheFab-7/Abd-BD16 tions elsewhere in BX-C. A well-known example of a genotypeisenlargedrelativetotheA7tergite,andmore pairing-dependentregulatoryinteractioninBX-Cisthe closelyresembles the wild-typeA6 segment.This unex- wing transformation induced by the Cbx1Ubx1 chromo- pectedfindingsuggeststhat,unlike iab-7Sz,theAbd-BD16 some when it is paired with a wild-type copy of BX-C mutationpartially suppressesthe gain-of-function phe- (Lewis 1955,1985).The12 Fab-7* chromosomes were Transvection inthe Abd-B Domain 1035 Figure 1.—Abdominal cuticles ofadult males.(a) (cid:49)/(cid:49). In wild type,A7 is represented onlyby a pairof tracheal openings (arrowhead). (b) iab-7Sz/(cid:49). The haplo-insuffiency of the iab-7 cis-regulatory region results in the addition of a rudimentary, seventh tergite. (c) Fab-7/(cid:49). Ectopic activationof iab-7 in A6 reduces the sixth tergite to the size of the seventhin iab-7Sz/(cid:49). (d) Fab-7/iab-7Sz. Because only a single iab-7 is active in A6 and A7 (that of the Fab-7 chromosome), both sixth and seventh tergites are reduced to the size of the seventhin iab-7Sz/(cid:49). (e) Fab-7/Abd-BD16. A7 is similar to that of the previous genotype, but the A6tergite isenlarged. (f)Fab-7/Df(3R)P9. If, in addition to iab-7, the transcopy of Abd-B is also deleted,both A6 and A7 tergites areagain similar to those of the genotype Fab-7/iab-7Sz.(Additionally, due to the partial transformationof A5 into A4, lack of dark pigmentationin patches of A5is alsoevident.) (g) Fab-7/McpB116iab-7SzAbd-BD16. Thesize of thesixth tergite is similar to that of thesixthtergite in Fab-7/Abd-BD16 males e,indicating that the presenceor theabsence of an inactivecopy of iab-7 in trans is indifferent for the suppression of the Fab-7 phenotype. However, A7 is enlarged to the size of A6 in Fab-7/ McpB116iab-7SzAbd-BD16 trans-heterozygotes. As shown in h, if Fab-7 is replaced with a wild-type chromosome in this genotype (McpB116iab-7SzAbd-BD16/(cid:49)), A6 becomes similar to wild type, but A7 remains about the same asin Fab-7/McpB116iab-7SzAbd-BD16, indicating that the Fab-7 deletion only ectopically activates the wild-type iab-7 in PS11, but does not alter its ability to trans- regulateAbd-B. 1036 L.Siposet al. TABLE1 Listofbreakpointsofrearrangements thatreducetrans-suppression inFab-7/AbdD16males Cytologyin the Effect on TSR-XSz, Fab-7 criticalregion trans-suppression Note TSR-1Sz, Fab-7 T(2;3)2R heterochromatin; 89B (cid:49)(cid:49)(cid:49) TSR-2Sz, Fab-7 heterochromatin(?);87B (cid:49)(cid:49) TSR-3Sz, Fab-7 T(2;3)2R heterochromatin, 87F-88A (cid:49)(cid:49) TSR-4Sz, Fab-7 In(3R)heterochromatin; 90E-F (cid:49)(cid:49)(cid:49) TSR-5Sz, Fab-7 T(2;3)26A; 3Rheterochromatin (cid:49)(cid:49) TSR-6Sz, Fab-7 T(2;3)28E;87B8,9 (cid:49)(cid:49) TSR-7Sz, Fab-7 T(2;3)59B-C;81F (cid:49)(cid:49) TSR-8Sz, Fab-7 In(2R)81F;88B (cid:49)(cid:49) TSR-9Sz, Fab-7 T(2;3)58F;3R heterochromatin (cid:49)(cid:49) homozygousviable TSR-10Sz, Fab-7 In(3LR)62B; 81F (cid:49)(cid:49) homozygousviable TSR-11Sz, Fab-7 T(Y;3R)87B (cid:49) TSR-12Sz, Fab-7 In(3LR)79E-80B; 89E (cid:49)(cid:49)(cid:49) iab-4(cid:50) Additionalrearrangements outsideofthecriticalregionare presentin TSRSz(cid:50)1,3,5,6,8and12;onlythe relevant breakpointsare listed.(cid:49)(cid:49)(cid:49),strong effect; (cid:49)(cid:49),moderateeffect;(cid:49), weakeffect. testedfortheirabilitytointeractwiththeCbx1Ubx1chro- maininPS11/A6islimiting;otherwise,thepresenceof mosome, and we found that all but one significantly a competing Abd-B point mutant in trans to the Fab-7 suppressed the wing phenotype induced by the Cbx1 chromosomewouldhavenophenotypicconsequences. mutation.Moreover,thestrengthofthephenotypicef- This supposition—that the regulatory capacity of iab-7 fectsofthese11Fab-7*chromosomescloselyparalleled islimiting—seemstobecorrect,becausethephenotype thatfoundincombinationwithAbd-BD16:mutationsthat of PS11/A6, when the Abd-B gene in trans is either re- exertedastrongeffectontheFab-7/Abd-BD16phenotype movedentirely(asinFab-7/Df(3R)P9)ornolongeravail- alsostronglysuppressedthewingphenotype(notshown). able for pairing (as in TSR,Fab-7/Abd-BD16), is indistin- AsexpectedfromthesuppressionoftheCbx1pheno- guishable from Fab-7/(cid:49). type,cytologicalexaminationofthese11Fab-7*chromo- Competition between the iab-7 cis-regulatory do- somesrevealed thattheyallhadrearrangementsaffect- mains: In the experiments described in the previous ingtherightarmofthethirdchromosome.Theposition section, nonproductive trans-interaction between the ofthebreakpoints forthe11Fab-7*chromosomes that iab-7 regulatory domain on the Fab-7 homologue and suppress the Cbx1 phenotype (TSR-1Sz,Fab-7 to TSR- theAbd-Bpointmutationontheotherhomologuesup- 11Sz,Fab-7)arelistedinTable1.All11ofthemhaveone pressedthegain-of-functionphenotypeinA6.Wespecu- breakpoint between BX-C and the centromere, and a lated that the trans-interactions between iab-7 and the secondbreakpointoutsideofthisregion.Thisarrange- mutant Abd-B gene were permitted in A6 because the ment of breakpoints follows the rules established by iab-7 regulatory domain on the homologue containing Lewis(1955)forchromosomalaberrationsthatdisrupt themutantAbd-Bgeneisnotactiveinthissegment(see transvection in BX-C.The one exception,TSR-12Sz,Fab-7, Gallonietal.1993).Incontrast,suchtrans-interactions provedtobean iab-4mutation, andit hasaninversion might not be as strong in A7 because cis-interactions breakpoint in BX-C (see Table 1). Although this re- betweentheAbd-Bgeneandtheiab-7regulatorydomain arrangementshoulddisruptpairingintheAbd-Bregion onthesamehomologuewouldreducetheabilityofthe of BX-C, it would not be expected to affect pairing in Abd-B promoter to interact in trans. This model could theUbx region of BX-C. potentially explain the phenotypic difference between TheseresultsarguethatthesuppressionofFab-7gain- A6 andA7 in Fab-7/Abd-BD16 males (see Figure 2). of-function phenotype by Abd-B point mutations is If this model is correct, cis-competition in segment likely to be due to a nonproductive trans-regulation. A7shouldbeeliminatedbydeletingtheiab-7regulatory Whena wild-type geneis present onboth homologues domain on the Abd-BD16-containing chromosome. The (Fab-7/(cid:49)), both cis- and trans-regulatory interactions lackofcompetitionfromtheiab-7domainincisshould can contribute to the level of ABD-B protein that is result in an increase in the frequency or strength of ultimatelyexpressed inPS11/A6.Although thesesame trans-interactions between the iab-7 domain on the Fab-7 cis- and trans-interactions would also occur in Fab-7/ chromosome andthe mutant Abd-B gene onthe other Abd-BD16 flies, functional Abd-B protein would only be homologue. producedbytheAbd-BgeneontheFab-7chromosome. Totestthisprediction, werecombinedan iab-7dele- It should be noted that this model requires that the tionmutant,iab-7Sz,ontotheAbd-BD16chromosome.The regulatorycapacityoftheectopicallyactivatediab-7 do- phenotypeof(cid:49)/iab-7SzAbd-BD16andFab-7/iab-7SzAbd-BD16 Transvection inthe Abd-B Domain 1037 malesisshowninFigure1handFigure1g,respectively. twoactiveiab-7 domains.Inthesecond,cis-interactions Two findings are of interest. betweentheiab-7regulatorydomainandtheAbd-Bgene First, the gain-of-function phenotype of Fab-7 in A6 onthesamehomologuewould,becauseofcompetition, is stillsuppressedinFab-7/iab-7SzAbd-BD16males(Figure tend to reduce or suppress trans-interactions between 1g),andthissegmentcloselyresemblestheA6segment thisAbd-Bgeneandtheiab-7 regulatorydomainonthe foundinFab-7/Abd-BD16males(Figure 1e).Thesimilar- other homologue. When there is no competing active itybetweenthesetwomutantcombinationswouldargue regulatory domain in cis, the Abd-B gene will be more thattheiab-7cis-regulatorydomainontheAbd-BD16chro- readilyavailabletoengageintrans-interactions.Itisnot mosome does not contributeto thesuppression of the possible to distinguish between these two models with Fab-7phenotypeinA6andwouldbeconsistentwiththe thegenetictools presently available; however,we favor idea that the iab-7 cis-regulatory domain is normally thesecond model, as ouranalysis ofthe effectsof pro- inactive inthissegment (Gallonietal. 1993).Second, moter deletions (see below) suggests that the balance segment A7 inboth(cid:49)/iab-7SzAbd-BD16 (Figure 1h)and betweencis-andtrans-interactionscanbeshiftedbycom- Fab-7/iab-7SzAbd-BD16 (Figure 1g) males is even more petition. strongly transformedtowardanA6identitythan itisin Mutations that delete sequences near the 5(cid:57) end of (cid:49)/Abd-BD16 or in Fab-7/Abd-BD16 males. In fact, A6 and the Abd-Bgene enhancetrans-regulation: Theability of A7 are indistinguishable in Fab-7/iab-7SzAbd-BD16 males the iab-7 regulatory domain to interact with the Abd-B and closely resemble the appearance of A6 in Fab-7/ genesincisandintransisnotequal.Ifthecis-andtrans- Abd-BD16 and A7 in (cid:49)/iab-7SzAbd-BD16. It should be em- interactions were symmetrical, the phenotype of Fab- phasized thatthistransformationin A7identitycannot 7(cid:49)/(cid:49)Abd-B and Fab-7Abd-B/(cid:49)(cid:49) would be identical. be attributedsolely tothe lack ofiab-7 andAbd-B func- However, thisis clearlynot the case. Incontrast to the tionontheiab-7SzAbd-BD16chromosome.Thus,thetrans- modestbutdetectabletransformation ofA6toward A7 formationofA7towardA6inFab-7/iab-7SzAbd-BD16males in Fab-7/Abd-BD16 flies, we see no trace of an A6 to A7 ismuchmoreextremethanitiswhenFab-7iscombined transformation in Fab-7Abd-BD16/(cid:49) flies. Similarly, with deficiencies e.g., Df(3R)P9, Figure 1f, or Df(3R)C4, Abd-BD16 does not complement iab-7Szto any detectable not shown, which completely remove both iab-7 and degree in A7 (Figure 6a). Abd-B (compare Figure 1g and f). As expected, if the We reasoned thatit might be possibleto strengthen Fab-7chromosomeisreplacedbyitsrearrangedversion the interaction of iab-7 with the trans copy of Abd-B by inFab-7/iab-7SzAbd-BD16,i.e.,inTSR,Fab-7/iab-7SzAbd-BD16 weakening its interactions with the Abd-B gene in cis. (not shown), the phenotype of A6 and A7 becomes Because the regulatory interactions of iab-7 with the indistinguishable from that of TSR,Fab-7/Abd-BD16 or Abd-B gene are likely to be mediated, at least in part, Fab-7/Df(3R)P9(showninFigure2andFigure1f,respec- by sequences in the vicinity ofthe Abd-B promoter,we tively). thought that it might be possible to weaken the cis- These findings indicate that trans-regulatory interac- interactionsbysubstitutingAbd-Bpointmutantswithan tionsin segmentA7 havenophenotypic consequences Abd-B 5(cid:57) deletion mutant, Abd-BD14 (m(cid:50); Karch et al. if an iab-7 domain is present on the chromosome con- 1985),whichlackssequencesaroundtheproximalpro- taining the mutant Abd-B gene. To demonstrate that moter (Boulet et al. 1991). As anticipated, we found this domain must not only be present but also active, thatAbd-BD14weaklycomplementstheiab-7loss-of-func- we recombined Fab-7 with the Abd-B point mutation tionmutant iab-7SzinA7(Figure 6b).Moreover,unlike Abd-BD16 and then examined the phenotype of Fab-7/ inFab-7Abd-BD16/(cid:49),wewereabletodetectaweaktrans- Fab-7Abd-BD16 males. In these animals, both copies of formationofA6intoA7inFab-7Abd-BD14/(cid:49)adults.This iab-7willbeectopicallyactivatedinA6.Ifwearecorrect phenotype of weak trans-interaction is best seen in the in assuming that the regulatory domain in cis must be sixth sternites of females (see Figure 3a). active in order to reduce trans-regulatory interactions Toruleoutthepossibilitythat thisweaktransforma- with themutant Abd-Bgene,thenthephenotype ofA6 tion is due to a “leakiness” of the Abd-BD14 mutation, andA7intheseanimalsshouldbeidenticalandresem- we examined thephenotype of Fab-7Abd-BD14/Df(3R)P9 ble theA7segment inFab-7/Abd-BD16 animals. Indeed, hemizygotes and Fab-7Abd-BD14/Fab-7Abd-BD14 homozy- this is the case (see Figure 2). gotes. In neither case did we detect transformation of Two different(but not mutually exclusive) hypothe- A6intoA7(notshown).Infact,itappearsthattheweak ses could explain the phenotypic difference between transformationobservedinFab-7Abd-BD14/(cid:49)animalsre- the segments in which one vs. two iab-7 regions are quiresthatthewild-typecopyoftheAbd-Bgeneintrans active, e.g., the A7 of Fab/iab-7Sz Abd-BD16 and Fab-7/ can pair. Thus, transformation is not detectable when Abd-BD16. The first postulates that the iab-7 regulatory pairingisefficientlydisrupted(Figure 3b).Conversely, domainsengagenotonlyin cis-butalsointrans-regula- trans-interactioncanbedetectedinbothFab-7Abd-BD14/ tory interactions (indicated by dotted arrows in Figure iab-7Sz and Fab-7Abd-BD14/Df(3R)R59 flies (not shown), 2).Inthiscase,theresultingphenotype dependsupon where the trans copy of the iab-7 regulatory domain is thesumoftheinteractionscontributedbyeitheroneor deleted but the Abd-B gene is intact. 1038 L.Siposet al. Figure 2.—Diagram- matic recapitulation of the most relevant results (pre- sented in part in Figure 1) and their interpretation. Continuous line represents theactiveformofiab-6and iab-7; shadedrectangle,the inactive (closed) form of iab-7; small black dots, the Fab-7 and the putative Fab-8 boundaries; vertical lines, the endpoints of Fab-7 and iab-7Sz deletions; filled cir- cle,theproximalAbd-Bpro- moter; bold arrow, the Abd-B class A transcription unit;andX,thepresenceof apointmutation.Theregu- latory interaction between Abd-B and iab-7 is repre- sented by curved arrows. Dottedarrowsandquestion marks symbolize the possi- bilityofmutualtrans-regula- tion when iab-7 cis-regula- tory regions are intact and activeonbothhomologues. Thewidthofblackoblongs in column 2roughly corre- spondstothesizeoftheter- gites of the different geno- types. OtherlinesofevidencealsosuggestthattheAbd-BD14 trans.Thesedifferencesarepairingdependent,because promoter deletion mutant is a weaker competitor for the phenotypes of Abd-BD14/TSR,Fab-7 (Figure 4c) and interactions with iab-7, either in cis or in trans, than a Abd-BD16/TSR,Fab-7fliesareindistinguishable(notshown). structurallywild-typeAbd-Bgene.First,thetrans-suppres- IsolationofnewAbd-Bmutationswhichenhancetrans- sionoftheFab-7gain-of-functionphenotypeinsegment regulation by iab-7: The observation that the small A6 of Abd-BD14/Fab-7 males is significantly weaker (but Abd-BD14deletionenhancesiab-7 trans-regulationpoints still detectable) than in the combination of Fab-7 with to sequences in the vicinity of the proximal Abd-B pro- any of the Abd-B point mutants (compare Figure 4, b moteraspotentialtargetsforcis-(andtrans-)regulatory and a). Second, the haplo-insufficient phenotype of interactions.However,thefindingthatAbd-BD14stillsup- Abd-BD14/(cid:49) in A7 is clearly weaker than that of Abd-B presses Fab-7 in trans to some extent (compare Figure point mutations (compare Figure 4, b and a), which 4,bandc) suggeststhatadditionalsequencesmayalso suggestsahyperactivationofthewild-typeAbd-Bgenein berelevant for trans-regulation. Transvection inthe Abd-B Domain 1039 Figure3.—Adultcuticlesof (a)Fab-7Abd-BD14/(cid:49),(b) TSR-10Sz,Fab-7Abd-BD14/In(3LR)TM3,SbSer females.Whenpairingof the homologues isnormal a,A6is weaklytransformed towardA7asevidencedbythe presence ofcharacteristics thatare normally foundon thewild-type A7,such ashairs pointing towardthemiddleofthe sixthsterniteandthe absenceof trichomesaround thesternite(indicatedbyarrowheads).Whenpairingisefficientlydisrupted(bytherearrangementsTSR-10SzandIn(3RLR)TM3), this mild Fab-7phenotypeisnolongerdetectableb. Inaddition,thehaplo-insufficientphenotypein A7isclearlystronger(the size oftheseventhsterniteisenlarged)thanina,indicatingthatthehyperactivationof theAbd-B geneisalsoeliminatedinthe absence of somaticpairing. To further characterize sequences from the Abd-B greatest in the largest promoter deletion, that is, Fab- genethatcontributetotheseregulatoryinteractionswe 7Abd-BPSz1/(cid:49) (cid:46) Fab-7Abd-BPSz2/(cid:49) (cid:46) Fab-7Abd-BD14/(cid:49) (cid:46) took advantage of a hypomorphic Abd-B mutation, Fab-7Abd-BD16/(cid:49). Interestingly, precisely the same rela- UC21-10,1-d,whichhasaP-transposonatposition(cid:49)159 tionship is observed for the complementation of the on the molecular map (McCall et al. 1994) just up- iab-7deletion,iab-7Sz,inA7.Thus,thestrongestcomple- stream of the transcription start site of the Abd-B class mentation is observed in Fab-7Abd-BPSz1/iab-7Sz flies, Atranscript.Byitself,thisinsertionalmutation,likethe which is followed by Fab-7Abd-BPSz2/iab-7Sz, Fab-7Abd- promoter deletion in Abd-BD14, weakly enhances iab-7 BD14/iab-7Sz, andfinally, Fab-7Abd-BD16/iab-7Sz. The simi- trans-regulation. Thus, Fab-7UC21-10,1-d in trans to a laritybetweenthesetwoassays suggestedthat wecould wild-typeAbd-BgenetransformsA6towardA7.Thatthis measure the ability of any Abd-B mutation to promote transformation of A6 to A7 is not due simply to the trans-regulation by examining its complementation of residual Abd-B activity of the UC21-10,1-d allele, but iab-7SzinA7.Usingthisassay,wetestedthetrans-regulat- rather arises from trans-activation of the wild-type ing ability of three previously isolated deficiencies, Abd-B gene, is suggested by the finding that the A6- Df(3R)C4(Karchetal.1985),Df(3R)U110andAbd-BRD18 A7transformationisstronglyreducedwhenFab-7UC21- (Hopmann et al. 1995) that remove the distal part of 10,1-d is combined with the point mutant Abd-BD16 (le- BX-C. As expected, Df(3R)C4, which removes iab-7 thality coveredby Dp(3;1)bxd111). and part of iab-6 in addition to Abd-B, does not show To identify sequences that contribute to cis- (and any complementation. However, both Df(3R)U110 and trans-) regulatory interactions, we mobilized the UC21- Abd-BRD18, which leave iab-7 intact, complement iab-7Sz 10,1-dtransposonontheFab-7UC21-10,1-dchromosome to a high degree (Figure 6, d and c, respectively). As and then selected for excision events that eliminated expected, these two deletions do not suppress the the remaining Abd-B m activity (see materials and Fab-7 phenotype in trans, and the phenotype of A6 in methods). We then tested these new Abd-B alleles for Fab-7/Df(3R)U110 and Fab-7/Abd-BRD18 (not shown) is their ability to enhance trans-regulation of a wild-type similar to Fab-7/(cid:49)(see Figure 1c). Abd-B. Outof ninenew Abd-B alleles, weidentified two Theapproximate limits of the complementing dele- mutations (Abd-BPSz1 and Abd-BPSz2) that enhanced the tions are shown in Figure 5. These deletions can be Fab-7 gain-of-function phenotype inA6 when transtoa ordered in increasing strength of iab-7Sz complemen- wild-type copy of BX-C more strongly than Abd-BD14. tationasfollows:Abd-BD14(cid:44)Abd-BPSz2(cid:44)Abd-BRD18(cid:35)Abd- Genomic Southern blotting revealed that Abd-BPSz1 has BPSz1. From thisdata it would appear that trans-regula- a 10.8-kbdeletionbetweenmapposition(cid:49)155.6–156.8 tion is strongest in the case of deletions that remove and (cid:49)165.6–166.6, andAbd-BPSz2carries a smaller dele- the largest region of Abd-B DNA on the 5(cid:57) side of the tionofabout5.5-kbfrom(cid:49)155.6–156.8to(cid:49)160.2–161.3 Abd-Bgene(compareFigure6,bandc).Itis alsoclear (Figure 5). thatdeficienciesthatstrongly enhancetrans-regulation Other 5(cid:57) deficiencies also enhance trans-regulation: extend well beyond the region thought to correspond When we compared the transformation of A6 into A7 tothe proximal Abd-B promoter(Busturia andBienz in Fab-7 Abd-B(cid:50)/(cid:49) flies, it appeared that the strength 1993). of the trans-activation of the wild-type Abd-B gene is We note that unlike Abd-B “point mutations,” the 1040 L.Siposet al. Figure 4.—Abdominal cuticlesof adult malesof the genotypes(a) Fab-7/Abd-BD16, (b) Fab-7/Abd-BD14 and (c) TSR-1Sz,Fab-7/ Abd-BD14. Suppressionof the Fab-7 phenotype bythe deletion mutant Abd-BD14 is significantly weaker(but stilldetectable) than inthecombinationwiththepointmutationAbd-BD16(comparethesizeofthesixthtergitesina,bandc).Duetothehyperactivation of thewild-type Abd-B gene, the haplo-insufficient phenotype of A7 is suppressed in b (A7tergite is missing and only a single hair at the tracheal opening, indicated by arrowhead, shows that it is not completely wild type). When pairing is efficiently disrupted c,both hyperactivation and trans-suppression are eliminated,and A6and A7tergites are equallythin. The thickness ofarrows in thedrawingsunder eachof theabdomens indicatesthe relativestrength of cis- andtrans-regulation. Abd-BPSz2, Abd-BRD18 or Abd-BPSz1 deletions do not show sometimesappears to beclonal even withina hemiter- any sign of an A7 to A6 transformation when they are gite(Figure6d).Df(3R)U110alsodiffersfromtheother trans to a wild-type BX-C. This difference presumably Abd-Bdeficienciesinthatitfailstocomplementtheloss- reflectsthe abilityofthese largepromoterdeletions to of-function iab-7 phenotype of Df(3R)R59, a deficiency enhancetrans-regulatoryinteractionswiththewild-type that extends proximally from the 3(cid:57) end of the Abd-B Abd-Bgene.Infact,inthecaseoftheselargerdeletions, geneto thebxdregion. Theonlysignof somecomple- thehyperactivationofthewild-typeAbd-B geneintrans mentation in Df(3R)U110/Df(3R)R59males is the pres- is even stronger than that observed in the case of the ence of black pigmentation in tergite A6 and A7, a smaller promoter deletion, Abd-BD14, because a weak phenotype corresponding to the result of the pairing- A7 to A6 transformation (the presence of at least a insensitive, long-range interaction described by Hop- single bristle in tergite A7; Figure 4b) is observed in mann et al. (1995) and Hendrickson and Sakonju Abd-BD14/(cid:49) males. However, these promoter deletions (1995).Allofthe otherAbd-Bdeletionmutations com- do show the expected haplo-insufficient phenotype plement the R59 deficiency in A7 to a degree roughly when they are combined with a chromosome like TSR, similar to that observed for iab-7Sz (data not shown). Fab-7thatdisruptspairing.Inthiscase,theirphenotype ThelackofthiscomplementationbetweenDf(3R)U110 is indistinguishable fromthatof TSR,Fab-7/Abd-BD14 (see and Df(3R)R59 does not seem to be due to the fact Figure 4c). thatthesetwodeficiencies overlapin theregionjust3(cid:57) Df(3R)U110 isanunusual case. Itisthelargest ofthe to the Abd-B gene. Like Df(3R)U110, the deletion Abd-Bdeficiencies thatstill retainsiab-7, andit extends in Abd-BRD18 overlaps Df(3R)R59 in this 3(cid:57) region, yet from near the3(cid:57) end of Abd-B into the 90A region. As Abd-BRD18 shows the same degree of complementation expectedfromthelargesizeofthisdeficiency,itexhibits of Df(3R) R59 as that observed for the Abd-BPSz1, which by far the strongest enhancement of trans-regulation does not overlap. whenintranstoiab-7Sz.However,unlikeanyoftheother Themostlikelyexplanationfortheunusualbehavior deficiencies,theextentofcomplementationbythetrans of Df(3R)U110 is illustrated in Figure 7. In the case copyofiab-7variesfromhemitergitetohemitergiteand of nonoverlapping deficiency combinations such as