”3%.... I :3: . VIC chin?” ." . f. ‘ 4’ HQ; “I V. :5?“ 4._-{:\:.. “at '22 4.29?" 2 IH%::> .531 1 ‘| 1:45:13}. ‘1‘ ‘4’ '2")? 2‘; s,,.,;.:«. JP}. 22 ::”1::4‘ L'C; m‘} 2 . . .' $2M: ran)?“ ngl'da}? ’1'." Jr. ’ (“H a u cf. .2 x: J {5 v'J.r ' A,“ “C 3'31“." 1 .\’ - 5&4; 1"}? {22:}; "1123“” . 2,. 2136.? “4 £9.41; EVE“)? =55: ' , . -47 2,3121: - £1 .m , ‘L‘AV’WI’?’ "“ 'ra‘ 113“"; D:, ”fr "4"ng Eff}: ‘HI . ”’21: r" a I 7) , . '.&'4“ if“; ~» «,u fig; 3'5» 12,2.-22321 :23 1' *Jfgfiwusfl 4211-2352” - 11,? uglfi'fi- .. f: n ,"J’ ‘ 3,; . .42: . 13.2.3.» ,2»: .22: ”2'1“”? 2..., ' < .. .5 :2 .‘ .z.‘ I """"-‘-"'r-"‘& '2?) 4D: ’. WMM‘ESK' ”may! ‘ $51 + _ .rfitzrf'i. 7, if.“ .. 3%" ~22 w: W411" - .2: mi; 52%.! 732W "05. r415“; .r'mrz' "4:2 AM"- . .-:.'.’1.-':'..2" ”€33. «1-41-1113 w. o f:%&:{?§fi: ?;;-43:.”?=::'fi?£‘3 Djn”’ 16:, ,-, ‘1“ ‘r-"’;:..,' an”: :0 If; ,, ‘ -u a ». a “'1'". ,’ .- :.':M"f‘_.._‘“' = -« 2-. =2. -=-=~- ~ _ awn... "' £3“: ~ sun-7’4 a. W {V ‘1 ’L...J";§P; rr 3 «.-z~.~I.,.r,:gw 2 , ,. m-I» a! My...) - “hens.“ #0933; . 79:191-212.. 22.1.. -. . [4-21 .. . vfmflze- 2; .I 22.2% .._ pm? 2.; 12-421. .- 4 w ...‘.“... m 4. 32 ‘- ....., - . a ”339$ ”any". .fl'» 45:43 f? . farm.“ 341.35% mm awe-c, 4- ., . ’2 42.43.... ~ 4-7'2'." ff‘x‘ir.‘ “ ' 22:07," '1‘)" . .i—zg. . . 22”.. "'2‘: ' - =. ~‘9~‘\ '2 _ «x . $15.; 3.0: L‘ = ,v .- my: a? {W 423.0“? - 1 ’icéfi'é‘ dx': .nx. " wary-215%. . ,“‘.fio'4§"fi:‘133jr= 15'2" 54:4" "1'3"" ~.v. 1 ' 1 --‘”;n2~'2~‘- ” . I 1-, J 41' %‘ "4m1:2£r-'E=’J%'m‘” “'2' “:1 . , WM .r-w ”’1, ,1.-. . , .. .. .- .>n4m:r~.'-r - . , ..... n. {cry ..29;%23§%.?"2'j‘" 2224:: Wfigw my?“ w Dunn ,} M ' ' 3' I 3 .1 . 'v ‘17:" Lia's” ' 4m {twig-‘3 "* ' " “3“" a" 1: L '7‘ 1:”? . -, 2 0%}; , x =4» .- 424! 3:32.19 "V 4 r 4. w 1; .5343! - 4.1.9”: {I ”1114' _ ,, .. 4‘ ,7" ’ . .4... ,, ”a6 ”‘ .. ,, ', -22.,” ’1’: .2” ""“‘“ '7‘22322. 52.22““..2Wfi32252222'fié2g‘gw 2332‘“ =~~ ~- , y ":22: MW 1. {trim-21%.?” ma 9 t 3 l." . " 4 D «'23:A%A£’”’ ‘ “"9 W @139; .-.. _ Itcftij'vfig: ‘ .1." z f1~w o . v on»; . v- 4-31? ' I .J.,.I..l 4 v; -z~..4.;, .4» ”.1, jnnaéil‘1fljvjrtf- .Mw44gg/frm 9:5: = "C“; J 4&7??? ' web 54,. 2.2422222: r 4 a ‘1 . 25’. _- , _ 4.. 12¢ m “t D' a. . 1-. .ngo; I0. = “I“; I‘ r :‘T' ”’32.. *“7’2 4111...“.A-mr‘4 , 3'.-... v» ’7‘ . . ~1- _ 3:2; v, - g... ” v “a“. -d t 4 .__, , ' o: - . 2:“ .. 2.1:? . ~ww‘ 9.2322; “’4’ ' _~.;;‘r.---;-n- al.’ :'£?~L.",~‘»;:"# «a: 2:32.: 2: J51; '35-; = ; gun‘s/72:; '21-’34?“ —. _ é .Jvr‘ottv'y M {AW "3 i “‘f‘. “11751;. - ‘°"' {5:73}; .. ”21:71:; ' “7:32;”: ,_. 'L .. """W-kuév “£11.32 4:“; I‘- a uni, "4' /“\J. U ‘V \J ‘J J This is to certify that the dissertation entitled GENETICS AND MOLECULAR MAPPING CF E1 WITHIN THE RIBOSOMAL DNA OF LROSOPHILA MELANOGASTER presented by REBEKAH SARAH RASOOLY has been accepted towards fulfillment of the requirements for Ph. D. degree in final-fies @M Major professor Date October 5, 1989‘ MS U i: an Affirmative Action/Equal Opportunity Institution 0- 12771 MIC HlGA NSTA - W W IIIIIIIIIIQIII) III‘I’III IIIIIQIIIIIII Inga: 31'... I PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE “II-d “L ‘4 “l V n ‘ TV! ._I‘ ) ;-, w“— ?! , - [gap 2 3 “a“ ' h ‘~ I ”’13 1 a T! D MSU Is An Affirmative Action/Equal Opportunity Institution GENETICIANDIMDIECUIARLHAPPINGIOF'EEK‘WITHIN’THE WMOFWW By lbbekah Sarah msooly ALEESSERIKERJI mitted to Michigan State University in partial fulfillment of the requirements - for the degree of DOCTUR10F PHILOSOPH! Graduate Program in Genetics 1989 43947 130 gr I' I. Am mommmmorggmm WMOFWW By Rebekah Sarah Rasooly gexmamtarm-effectdanimmofmnammmidu immmzmodcsofribosmalmdaasingle chranosane. Inthisstudyseveral feaumofagxwered'iaracterized. Geneticeaqaerimerrtsnegpedgggtothepmximal heterodirmatinofthex chronoscme. Several dcminant, maternal-effect suppressors of g (M) were identified and also manned to the same region. Using duplications arxi deletions, both Egg; and one £11295). locus were shown to be neamrrhs. Both loci were then mapped to a specific position within the ribosanal [NA array using molecular size variants of the ribosomal 111A intergenic spacer. mtg], is an x-linked meiotic nutation which induces 15-! exchange amt inhibits magnification in males, perhaps because of a defect in post-mlication repair. Although & did not induce interdircnnscmal emt'lange between the x ard X alarm, the phenotypeofngifllmaleswasstmntohaveoflmersimilaritiestothe ng'tenotypeinfanales. tbstnotably, E131 inmalec irriuoesgelc- liJceexdxargas. 'Iherasults amortahypottmisthatgexmisa site-specific enhuclease activity arttixq within the rum ard that sum), loci W the endoruclease. 'Ib Avi and Irit for their mmtimim love and support iii I am grateful for all the superb technical assistance I received. 'Ihree mfiergraduats, Chris mrke, ran-mm om, ard Wing Chan, carried out various genetic experiments. Chris caustnicted the Drum-i drrarcsaneusedinmapterlt, I-‘ah-Minhtestedseveralofthedmm inciapberB forgactivity, ardWirgmarpedthesuppressorint-he "y ;et;c_." stocktothexmrancsate. Mfellowgraduatesuflents, Peter Crawley and Susan Icotens also contributed data to this dissertation. Peter did some molecular analysis of the recarbinants and Susan identified the suppressor in Quznso‘fl. Peter, Susan, Iori Wallrath ard other members of the Friedman lab also provided intangible assistance. ‘Ihey were ideal office—mates, always ready to listen and advise and share information and know-hm. My thanks also go to Nancy Veenstra, who was a confidante and‘canpanion through many long hours of fly counting. Myguidancecmmitteehasirdeedbeenasalmeofguidarne. Will Kopadrik and tarry Snyder gave nudu-reeded outside perspectives and heights. 'Itm Friedman was an invaluable source of information about virtually every aspect of experinentaticn and graduate school. Scott Williampxshednetobeginmlewlarexperinmts, anithenhelpedmeto daeignttmardlearntomnytlmart. Aboveanibeyondallthat, my cannitteeprovidedsnportardencmraganentwhenitwasmst needed. iv ‘Behlndcarselsateadxerto"respectywrstmdentasymdo ymrself..." Wadvisor, LennyRobbirs, is thevery exanple of what the rabbishadinmind. 'mrcnglnztmygraduatecareer,hehassharedhis enormous insight, always willingtodisaassamajor resultoraminor tedmicalproblem. Mytharflcsgotohimforteadrirgnegeneticsandhow toteadrarrihowtomakeflyfocdardadnenotlerlargearflsmall Urbgsumathilltalcewiutmefrmhere. Fimlly,1caxmotfindthemrdsthatwilldescribemyfamily's contribution to this work. Instead, I have dedicated this dissertation tomymxsbaxflardmydaughter,ardrereaddmyfl1ankstomyparentsaswell. 118180me IISI’OF'EBIES.......... LISI‘OFFIGJRES.......... OWLINIKIDCI'ICN...... 'me Nucleolus Organizer . . . Areas of investigation. . . . m2.mmmssm.. Measurement of g activity . Q pierbtype measurement ........... Rateofmn-disjunction... INAextractions....... Restriction digests and electrqrioresis Molecularanalysisofrma. . . . . . . Probe labelling ard hybridizatim . . . . . . ommmrsammnucmwsmmmm. . . . . mm. 0 0 O O O O O O O O O O O O O O Mamingofigwithresqaecttoflfl. mlemlar analysis of reccnbinarrts. . amlmmwn . . ‘menatureoftheMlocus. . . . . mltiplesuppressorsofgg. . . . . vi .X .1 12 15 15 15 .18 18 18 20 20 21 23 23 25 25 28 30 30 31 Wm22fl:W.Me:mml/§mmlnn. WIw O O O O O C O O O C O O O O O O I O O O O O O O O O O ommmormmmwmmmygm. . . . . . mm. 0 O C C C O O O O O O O O O O O O O O O O O O O O W “Ia O O O O O O O O O O O O O O O O O O O O O O W O O O O O O O O O O O O O O O O O O O O O O O O O O O 0 Verification of the ccntent of the Beg-bearing target Recovery of recmbinant lnflhflfiw.car diranosanes . . . . Genetic analysis of reccmbinants. . ...... Molecular analysis of recarbinants ..... . mlemlarmapofflierm. . . . DM$ Iw O O O O O O O O O O O O SIQIIFICFXNCE AND WW8 . . ........ DefiningtheMtarget..... Lookingfortheggtranscript.. SIGIIFIQNCE........... W O O O I O O O C O O O O O O O O O I O APPENDIXA. MCWOFAP-WWCISM mm C O O O C O C O O O O O O I O I O O O O O O O O O O O 0 mm B. WOmex-x mom. 0 e o e e e o o e o o o 0 vii 35 41 42 42 47 47 48 59 59 59 61 68 '70 77 81 81 82 86 86 87 88 91 91 95 98 amputembnmmrssrusm-mmvrm. . . . . . . . .103 mm C C O O O O O O O O O O O O O O O O O O O O O O O O O O 106 viii LISTOFMES Tablel. Descriptimofdmnosanesusedinexperinents. . . . . . . 16 Table 2. Use of duplicatims an! deficiencies to classify Imitation . 25 Table3.&activityofparentalardrecmbinantdmrunosanes. . . . 26 'I‘able4.mp1erntypeofparentalardrecmbinantdiruwsatee. . . . 27 Table 5. Results of dosage apex-intents . . ............. 34 Table 6. Suppression of $3 activity . . . . . . .......... 37 Table 7. Mapping suppressor in ”y egg" _)_( duanosane ......... 41 Table 8. Mapping suppressorinAnhersthmmnoscme ......... 43 Table9.0thersu;pressorsof&........ .......... 45 Table 10. Sensitivity of target chrmoscmes ............. 60 Table 11. Recanbination data for munwmilflgfl‘ and Lunwdumnscmes . . . . .............. 62 Table 12. b_t_> phenotype of parental ard recmbinant m's. . ..... 63 Table13.&xactivityofparentalardrecmbinantno's. . . . . . . 64 Table 14.§u(3_e§1activityofparentalardrecmbinantm's. . . . . 65 'I‘ab1e15.SImnaryofresultsforparentalarrirecanbinantm's. . . 66 'I'able16.ICSlelr;thvariantsinrecmbinantm's. 1. . . . . . . . .71 Table 17. Intra-dirunosareeadxangefrequercies. ..... . . . . . 96 Table18.mL-31effectmdetadments. .. . .. . . . .. . . . .105 LISTOF FIGURES Figure1.DetectimofBQ_cactivity. . . . . . . . . . . ...... Figme2.Mtypesofm-ixfiucedeadiange........ . . . . . Figure 3. Sdmtic rqaresentatim oftheXdiranosane ....... Figure4.Mapofasirgler£NAcistrm. . . . . . . . . . ..... Figure5.Mappirgofggwithrespecttoproximalmarkersarri Figure 6. IGS lengtharraysonparentalandrecmbmant chmcscmes. Figure 7. [implications and deficiencies used for functional classificatim of fix . . . . . . . .............. Figure 8. Recarbinantsusedtomapx-lirflced suppressorin'y e13". . Figure 9 . Mating scheme for recovery of E-nediated inversions A ofafi—bearingtargetdmmsate. ...... FigmelO.&x—irrhacedhairpinexd1angeinatargetdmrmcsane caatainingbothgxardflmggw ..... . ...... Figure 11. Separatim of reccnbinant 10's of hairpin Figure 12 . Molecular naming of rum array with IGS length variants. Figure 13. HimII/I-lirdIII digest of 10's generatedbym-induced hairpinexd'aange... Figure14.ICSlergthvariantsintargetd1rmnsalem's. . . . . . Figm15.lblemlarmpoftargetduumsaiem's. . . . . . . . . Figure16.Matingsdrauetorecoverx-Xemhangeprcducts. . . . . . X .2 .4 .7 .9 .28 3O 32 39 .49 50 .53 56 69 73 74 99 m1 ngisamaternal-effectdaninantloaasofflrpsgilamlm thatiniucesexdmangebetwemtworibosanalmh (rtNA) arrayscna single dimmsare (Rdzbins, 1981) . 'lhe exdiarge is mitotic and takes place in the early zygote, usually before S of the first division. For example, when a 33; female is mated to a mle with an attached-2g dlramsanethathastmrINAblocks, lto6%ofthe_)Q(_Ydaughterswill mflergoanexdiange event thatdeletestheintervenimXaxdmraratin from the attached-)5! chrcmosane (Fig. 1) (Rdbins, 1981). These "detadments" of the X dirmcsate charge the m daughters into E sons. If the Q-induced exdiange event takes place after S phase of the first cell cycle in the embryo or at the two cell stage then only half the cells will contain the detadrment product, ard the resulting animal will be a gynandrcxnorph (part male, part fanale)- SVanscn (1987) demorstrated that m causes exchanges between any two separated blocks .of rm (:1 a single dumcscme. She tested various chmmscmes with two rm}. blocks for their ability to undergo detachment events. Evmhalfablodcofrflflateadmaflofflredmrmnscmeisa target. Furthermore, Bexcan indtneboth'spiral" ardergesthat cause deleticns, and "hairpin" exchanges that invert the intervening material in the same target drrunosane (Fig. 2) (Meat, 198738:be & Svanson, Figure 1 - Detectim of g activity A: 'IherunetSquareforatypicalflmatingisshown. 'me_R_e;g- induced exd'langes take place in _)9{_Y female zygotes, yielding fl sons or gynandranorphs. These are readily distinguished from regular sons because of the yi marker. Although non-disjunctional sons are also yi, theyhave all ofthepaternal markers (3113). B: The attached-x1 target chrcmosane is schanatically shown mwdergoing a;Re_;c-irriucedexd1ange event. 'Ihisexchangeleadstoloss of} eudrrcmatin and detadment of a complete I dxrunoscme from the attached- fi dircmosane. Hetemdironatic regions are indicated by thick lines and the Q loci (or nucleolus organizers, 139) are indicated by open boxes. Hex - induced mitotic events B/«I-9 q » REGULAR X. y NON- DISJUNCTION *'_°-:TTF- Acentric "’ + Ring -x FigureZ-‘lwotypesofg-irdrcedeidmrp (adaptedfruanbirss Swansea, 1988) ‘Ihetargetshownisanltdirmosareduplicated fortheNOand surruuriing hetemdmaratin (such as In(1)sc'i-1_s_o‘l). Maternal R_e_x activityirriucesbothtypesofexdiangeinthetargetdrmmsare ina single cross (Robbins & Swanson, 1988) . The spiral exchange deletes the intervening x euchrmetin, d'Ianging m daughters into sterile yfragnent males patroclinous for only yi‘ or gynardrmnmhs with yi' male tissue. The hairpin exchange simply inverts the material between the DD NO's. x- use 0352 ’ ’ ‘ ‘ db 4L «3139.5» anmoEoEo mEomoEoEo 89¢... Sate... OZ .amemsoxw sear. an s a 8652.". ._eaw. 3 0880525 .093. _ . s . oEo»oEo.Eo Sarah X not N 6 1988). Mmmrpedtotheproximalportimofthexmrmnsare, between 53: arr! the centrrmere (Fig. 3) (Robbins, 1981) . fix is fascinating for two reasons. First, it induces a high frequemyofendaangeintherm, maimrelymdergoesneiotic emdaarqe. 'nregeneticlerqthoftheentirecentricheterodironatinis approximately 0.01 map units (Sd'nalet & Lefevre, 1976) , while the frequency of gag-induced exchange is approximately 5%. Even though the rENA amears to be a relative "hot—spot" for recombination within the heterochranatin, with most of the exdianges within the heterodmanatin actually takirq place in the run (Williams et al., 1989) , Begs-induced Vexchangeisstill at leasttwoordersofuagnitudemore frequent. Second, Beg-helped exdiange is mitotic and takes place specifically during the earliest stages of embryogenesis. Althouqu a few other loci areknowntobeactiveat this time, includinggl, @, Marleen—d, all of these involve loss of dimmsanes or lethality, and not exchange (Baker, 1975: mvis, 1969; Gelbart, 1974; Sandler, 1970). The unleolus Organizer - 'Ihe Belt target is the nucleolus organizer (NO) fiddlmjmflIeIBSaIdZBSrlNA(Ritmsaetal., 1966). mg. W,asinmanyofl1erorganisns,themisorganizedasan arrayofrepeats. 'Iherearetmrfllharrays,mid1correspord gereticallytothebeQed (Q) loci. Oneisinthecentric heteroduunatinoftheXdrrmsaneardaeisintheshortarmofthex dircnosane. Each contains approximately 225 repeats (Tartof, 19713). Mutatims at the Q loci correspond to deificiencies of rain. The flerotypeofantatiasrangesfrunlethalitytoamreorless severe ”bolted" Wype diaracterized by reduced fertility, delayed k v I 881“ Fignea-Sdmticwltatimofflaxm '1heXd1rmosaneisGGnepmitslmg,arrithemppoeitimsofnanyof theyrenotypicmarkersusedintheseexperinutteism. Althoughthe basal tetercclmtin (heavy lines) is Wm)! 40% oftherhysical lengthoftheXduuneeane,itislessthan0.1mapmitsingenetic lergth (Hillikerdtaln 1980;8d1alet81’3fdm, 1976;W1111meta1., '1989). 'niemocwpiesabaxtcm-thirdoftlebasalheterodmtinard isslmaeanopenbmt. 8 developmlt, ttflmingof thoracic bristlesandabdaninaletdling (Limisley I: 21m, 1985). ‘meindivimalrfNAgemswithinaclusteraremtidentical (Fig. 4). There is avariety of intergenic spacer (IGS) lengths, ranging fron 4kbtoaslargea520kb (Ooenetal., 1982:1rd5k8‘rartof, 1980: Terracol, 1986). There are two sources of the IGS lergth heterogeneity. 'IheS'portimoftheIGScartainsvariablemrbersofa340bpsequence bounded by Duel sites (Williams et al., 1987). The central portion of theIGSisentirelycanposedofintegralmmbersofaMObpsequence boundedbyAluI sites (Ooen et al., 1982; Simeone et al., 1985). 'meothersalrceofheterogeleitywithinanrmharrayisthe imertsequenceswhidlinterrupttheZBScodingregion. Insert sequencesarefardinapproximtely60%of_)5dlrmcsarearri15%ofx dlranosane repeats (Pelligrini et al., 1977; Wellauer et al., 1978: White & I-bgness, 1977). The insertion sequences fall into two classes designatedas'IYpeIand'IypeII (alsoreferredtoasRlnnandRZDn, respectively; Birke et al., 1987). rIype I insertshave onlybeen reportedinthezm,while'lypeIIirsertsarefo.nriinboththeXand X'm's (Wellauer et al., 1978). There is considerable length teterogereityarmx'gfllenenbersofeadlclass,wiu11ype1rangingfrm 0.5-7.5 kb (Fenigrini et al., 1977) and Type II ranging from 1.5-4 kb (Wellmleretal., 1978). Repeatscontaininginsertiaeappeartobe inactive,basedmthefactthattherearevirmallymprimary transcripts than the interrupted repeats (Franz & ninz, 1981: Glatzer, 1979: Jamridl & Miller, 1984: Kidd & Glover, 1981; Irng & David, 1979) . IthasbeensuggestedthatthemW'lypeIardTypeII insertim secpences are transposable elements (Glover, 1981; Dawid et IIIT‘lypol .IIII‘I?If-wu Pigme4-naporasingummtuut ‘IteNOcmtaimapprmdmatelyZZSerepeatsorcismaartof, 1971). A single cistrm is shown sdmatically, with the tramcribed sequencesinbooces. nemarecodingregimsfcrfalrweciesofm, mantranscribedepacenetmascpenbm(mltzetal., 1988). Sanecistrunineadtmarointernptedinthezasccdimregimby insertionmflridnareclassifiedaseiflarlypeIcrtypeII dependingmsecpence. ‘IypeIinsertimsappeartobemiqaetottex dlrunsane,mile1ypeninsertimsarefomdbothinxflrixm's (Wellaueretal.,1978). Eadminsertclasshasaspecificintegratim siteintheZBScodirgregim,arrlthetmsiteearoeeparatedbyfaier thanlootpabihaetalq 1981). 'medashedlinaswithintheimertim secpencesardwithintheIQhrmn-tramcribedspacer) irriicatethe variable lengths oftheseregicns. 'memdallergthsarearprcodmtely skbfortheICBKbenetaL,1982),5kbfor‘1YpeIimertsand3.2 for'IypeIIinsertsmellaueretaL, 197a). AllEcoRIrestrictim sitesalfirelevantsitesforodlerenzymesarem. 10 al., 1981). In fact, based at several lines of evidence, both classes ofelenentsameartobe"nm-IflRretrotransposae"or"retroposazs", elanartswhidltratsposeviaanminternediate, butlackthe characteristic 1mg terminal repeats of retroviruses (Rogers, 1985: ximg & Eickbush, 1988a). Firstly, Iype I inserticrs are flanked by duplications or deletias of the 288 coding regim (mwid & Rebbert, 1981: Roiha SI Glover, 1981: Roiha et al., 1981), a characteristic of mnytransposmirsertimevents. Secordly, 'IypeI insertionsarefotmd astarriemarrays flankedbyrmhatmltiple sites inthex teterochrunatin and the 4th dlrunoscme. 'Ihirdly, Type II elements have a stretch of polyA at the 3' end and the shorter copies of each element fomdinthegerurealwaysappeartobetrmcatedelenents containing just the 3' ends. All these features suggest transposition via a catplete or incompletely transcribed RNA intermediate. lastly, each element has an ORF with regions of close hanology to reverse tramcriptase (Eickbush, personal canmnimtion) . Beyond this, little is known about the organization of this large gere family because it is located in leterochrcrnatin. 'Ihe heteroduunatinnekesup40%ofthelergthofthe}$dlrmosme, andall of the 1 (Hilliker et al., 1980). Nevertheless, there is very little recmbinaticn (0.02% for the x heterochrtmatin) (Schalet, 1972), no omventicnal genetic loci (Hilliker et al., 1980) , and few cytological landmarks (Gatti et al., 1976). Its a result, the Q lows, like the few other heterod'lrmatic loci lawn, has been refractory to genetic analysis. The to is also difficult to dissect with mlecular techniques, because each rqaeat unit is at least 11 kb, preventing the 11 claringofnnrethantxoorthreeinanygivenfragmentminga calventicnal vector. Althoqu there is little meiotic recmbinatim within the no, there arelmiquenedlanisnethatmaintainthenmberofintactrmAcistrms in this large gene family. (be such process is magnification (Ritcssa, 1968), a heritable increase in rum copy umber which takes place both pre-meiotieally arri maiotically in the male germ-line (Hawley arri Tartof, 1985). MaleswithaQXdlrunosaxeardcreofseveral aberrant XdirmosamsproduceganeteswithQ‘thlrunosanesatahigh frequency ranging up to 35%. Hawley et a1. (1985) propose that pre—meiotic magnification involves extradircmosanal ccpies of the rum in the germline of magnifying males. Thee extradlrmcsaral copies form an anplified "onion-skin" structure, which is resolved by successive exchanges into a magnified Q locus. They suggest that meiotic magnification is the result of unequal sister-amtid exchange. The evidencethatthethnprocessesocairbydifferentnedianisnscares from the fact that various meiotic nutants can inhibit pre-meiotic magnification without affecting meiotic magnification. Furthernore, reduction (a heritable decrease in rm expression) is only seen as a meiotic event, presumably as the reciprocal result of uneven sister- chranatid exchange. WW, lownascarpexeatimalsoaffectsrfllhcopy umber. Maflyismissingcrem, thereisanincreaseinthe sanatic copy umber of ti)“ cistrons apparently by anplificatim of the single m present (Tartof, 1971) . ampereaticn appears to be under the control of a locus known asg (cmpensaticn resporseHProomier & Tartof, 1977). m the); duuosane, g; is located withinthe 12 heterochrunatindistaltothem. 'IhelccatimoftheXdirunosaneg: isnotknown. 'nlex-lirflcedglowsactsingjstotriqjer auplificatim of the adjacent Q loom, and acts in m to signal presenceofanm. Altl'nghtheXd'Ircrmsanegcanirduce carpensatiminthex, thexmdoesnotitselfccnpensate. Magnification and cmpereatim are both intradlrrncsanal events and carrot explain another interesting feature of rm: despite the low freqencyofediarqeintheheterodlrcrnatin, thereisstrikin; hmngeneityanongtheerepeatmlitsmtheXardXdlranosanes, a' Wm that has been termed "concerted evolution" (Tartof & mwid, 1980; Zimner et al., 1980; reviewed in Arnheim, 1983). Unequal reciprocal exchange does not appear to be sufficient to explain the pattern of harngeneity (Williams et al., 1989) ard so the question remains as to what forces operate in the populaticn to prevent excessive heterogeneity. Areasofinvestigatim-Ihereareantmberofquesticnsthatcanbe askedabartgg,am,moreinnortantly,amuaero£qmstiasabmtm thatmcanbeusedtoanswer. Inthesubsequentdxaptersvarims questia'sareaddressed,usingacamsetofmaterialsalflnethcds describedinclapterz: 1) Wereisfigglocatedwithrespecttoitstarget,ttem? In bodmfiaptersBardLgereticexperinentsareusedtonapBexwith respecttobothguflardthtecws (Fig. 3). 'Iheresultsarethat Mmpswithintherm. Inadditim,in(hapter4,&g—inhced recmbimtimarflamvelmlemlarapproadtareusedtomaprith respecttomoleallarvariantswithintlermAarray. l3 2) imatistlenamreofthemlows? InChapter3deleticns ardduplicatiaeoftheproximlportimoftheXdlrumsaiearetested toseewhetherMisahypa-mypo-orreanorfil. 'Iher'eslltssuggest thatmisareanorph. 3)Arethereother10ciwhid1interactwith&x? InChapter3 severalsqlpressorsofmaredescribedanimappedprmimaltogr. Modeerociaremappedtoflem,aebygereticexperhrents (Chapter3),andtheotheratthemleallarlevelusingthesame arproadltakenformappingoftheMlows(clapter4). 4) fixatistlenedlanismofthemxexdlangearflisitgelerally applicable for large scale mapping of the it)? Both these questions are addressedindeptermwheregeg-inimedexdergeeventsareanalyzed cnamolecular level. flexdengearpearstobeconventional,ardis usedtogererateamapofnoleallarvariantsinthem. 5) Whatisthenatureofthegtarget‘? InAppendixA,aspecial dlrarosarewithasingledistalribosaral repeatandanormal,proximal mistestedasatarget. Wertledmosmearpearstobe unstable, making the experinent inconclusive. 6) Doesggplayaroleinthetmngenizatim ofrmArepeatscn theXarridecsanes? InAmerriixB,therateof&gg-irduced interdlrmcsaralendlangeisneasured. Alttnightheexperinentis diffimlttointerpret,becausea1eofthetargetdlrmcsanesappearsto besanaketwstable,3g3doesmtiniwex-Xeruergeatanythim approaddngfllerateofintradlrmnsanalexdlarge,ardmaymtirducex- Xetnchargeatall. 7) Emism involvedinmagnification, an intradlrancsanal exdlangeeventspecifictotherm‘? 'Ihisquestionisnotaddressed 14 directly. In Apperriix C, however, @341, a nutation which severely inlibits magnificatim in males, is found to have Egg-like activity. 8) Wetistleproductofttefigglcals? IntheDiscussion, data frmalltheexperinentsaredrawntogethertosuggestaworking hypothesis for $3 action. I propose that Q; either encodes an enicmclease activity, or is a imitation that has activated the silent open reading frame of a retrqaoson which encodes an endonuclease. Several eaqaerimentsarealsoproposedtotestthathypothesis. 9) What iltplicaticns do the study of Q; and its effect on the N0 have ard what are possible future lines of research? These questions are addressed in the Discussion. m2 MGM StcdtsarriCrceses-Allmatingsweredaemaconmeal, molassesand brewer's yeast medium at 25°C. Five days after mating, parents were transferred to fresh food, and progeny were scored at regular intervals. Henotypicmarkersandstaniarddlrmosanesaredescribedinmndsleya Grell (1968) and Ifindsley & zim (1985, 1987). A variety of rearranged dlramsaleswereusedextersivelyinthismrk, several ofwhichwere newly constructed. 'Iheir origins and relevant properties are listed in Table 1. mt of g activity - fl activity was detected as shown in Fig. 1. Briefly, maternal presence of Be); causes mitotic exchange between the two blocks of run on a paternally-derived target chrcmosone, in thiscaseanattached-xx. 'mestandardtaterdlrcmosateusedwas xfipxbm yf Bsyi. The result is a charge of m daughters into Means, by"detadment"ofthe¥dlramsanefrcrnthe}$aldlmnatin. 'Iheeventtakesplace inthe firstorseconddivisionofthezygote, generating whole-body "detad'ment" males or gynandrrmorfils, respectively, which can be identified because they have yi male tissue without any other paternal x-linked markers. Ex activity is calculated as: detachment progeny/(regular fenales + detactment progeny). 'Ihere arescnevariatimsmthiscalcllatimincertaincasesbecause ofthe 15 16 Tablel-mazriptimorfdlrmusedinexperimms SInEEEEEEEL_______J!§EéE!BLIIEQ§£§i§§ Starriarddlrum Qg(l;1)sd!1 In(1)dl-49 In 1 m7 In(1)soiL§g§B In(1)s<>c5—1-I—-_sc>—4R WM Inflhflllihfl 111;Y)Ba4,zx§ 152(4me Novel durum We W WA Deficient for most of x basal heterochranatin, ireluding no, and proximal eudlrcmatic loci (Fig. 8) Deficient for rest of basal X heterochromatin, including no, ard proximal euchrcmatic loci (Fig. 8) mp1 icated material fun the tip, including yi’ , to the x centrcmere (Fig. 5) Middle third of )5 euchromatin inverted used to balance inversions of entire 5 eudrrcxnatin Multiply inverted balancer of the X chromosome Deficient for middle 3/4 of z basal ieterodlratetin including N0 (Fig. 8) Diplicated for middle 3/4 of z basal heterochromatin, including NO (Fig. 8) Deficient for NO (Fig. 11) Diplicated for NO One piece of 3'! translocation, includes all basal z heterochromatin (Fig. 8) xmrmcsateinsertedbetweentwoarns ofX(Fig. 1) (1) (1) (1) (l) (1) (1) (1) (1) (1) (1) (2) Non-inverted chrunoscne duplicated form NC at tip (3) free of surrounding heterochranatin. by Eds-induced hairpin exd'lange in MW Target dumescne for gag-Mod hairpin exdlange with Belt-bearing basal heterochranatin, arfi duplicated for the NC at the tip (Fig. 10) Inverteddlrmosaresgereratedl -induced U hairpin exchange in Quzlwfly— (Fig. 10) (4) (4) 17 we 1 (alt'd) We Distal em! or WJ. Mum; distal no (Fig. 11) 1111111395.: Proximal end of Mg, imluding proximal no (Fig. 11) W Distal portion of mum We. ."ith singlematthetipfreeofsurroulfilm teterodlrunatin,arrlbaseofp_§_(1)_x:1. (4) (4) (4) References: (1) Lindsley a zim, 1987 (2) Lindsley & Grell, 1966 (3) Robbins & Slvanson, 1988 (4) this smiy 18 use of different maternal genotypes. 'mese uedified mlculaticre are mtedbothinthetactandinthetableswererelevant. mWW-mhpfim. reflectirerMe-cpreesion. msneasuredusirgeitterofbmcroases. Inbothassays, theQ henotype was evaluated in Megg heterozygotes since Mg is calpletely deficient for rm. Either individual males bearing the chrcmoscme of interest were crossed to individual m(1)gJ.-49,y Q v B/nga‘y daughters or individual females who were dlrunosare/flg were crossed to Wicfiwyig males. In both cases, therhenotypewasexpressedasperetrame, thepercentageof W/nga‘y daughters showing any level of Q phermype [(phenotypicelly 33.52) (100)/total]. Rate of rut-disjmctim - 'Ihe rate of non-disjunction in all crosses is calcllated as: (L); non-disitmctionalfiprooenv) . . regular progeny + detachnent progeny + (2 x rm-disgmlctlonal progeny). Mattractia's-IbanalyzeapartiallarQlcals,X/Omales, halezygous fanalesorX/Ql'fanaleswereused. yolnalesweregenerated bycmssingx/xmalestomy/oraralesennasse. Atleastfive irriividualsonsweredeckedfor fertilityfruneadlmatingtoereure thatnofreeXdlrmnsmewassegregating. mmwmcmmmwa mdificaticn oftl'eprrocecmreofflenderetal. (1983). Flieswere stored frozen at -70°C. 50-100 flies were grourd quickly at roan taperatmeinagrcnrl-glasshanogenizerianlsofgrirdingbuffer 19 (0.114 NaCl: 0.2!! sucrose: 5am Nazism: 0.114 Tris, 1319.1; adding 0.5% SIB and 1% dinethyl pyrocarbcnate just before grinding). The extract wasirnibatedatésoc for30minutes, thenO.3mlof8MI12 CY Y ' ca’ L ‘ r o Su(Rex)"' F J’ / . .&-.nag§.¢gg g<.-w.1.:fmw‘L‘ " “‘“"”‘ "o SU(ReX) .. 4.. ~ 5'“; a_'o’] 1 [ o SU(ReX) ? 41 Table7-Hamirgamin'yemg'xamm man 1: mm my 6:10 4:0 [ 3.590 she’s .m s Distal reambimnts (Fig. 8a) y 3 1627 1444 y g 2 603 695 y s! y 4 1704 1592 2 s2 1! f 4 1169 1160 y .6! y f ear 3 677 686 0.00 Prcicinal recmbinants (Fig. 8b) 1 I per 1 792 562 y ear 3 1753 1269 y gr, Line 4 1 859 927 NUN¥DESJUNCTTGNKL llflmlllQEIPIHIXHfinl W W :u: e ... e mill nales ... . e Pendent l 1 0.13 1 0 0.06 0 2 0.31 0 0 0.00 0 2 0.17 0 0 0.00 1 3 0.34 0 0 0.00 0 1 0.15 O 0 0 0 0 29 9 4.58 0 7 0.45 81 16 5.23 l 0 0.11 8 0 0.92 Recombinant males with 1311, m (g, gbg) and various portions of the y etc. _)_( chromosane were obtained as shown in Fig. 1, and stocked by mating to C(1)DX,y f attached-x females. Progeny which had wild—type autcscmes (gt, magi) were selected tomaintainthestocks. Males fruneadlstodtwerenetedtcygyyfgeg/m females (n47 isanXdimwsanebalancercarryingthedaninant markerfi). (ET dalmters were mated to X—SX'XLLMJ y f B'yi/Q males to measure Ber activity. Beg/recanbinant 42 crossoversmreml.'nleyg£.pare71taldumosmewasmi’,ardall ofthedistalcrossoversveregi'aswell. Line4, however, which stmldalsohavebeenmlwereitasinplecrossover,bearsa35% penetrantmlocus, internediatebetweenthemoparentaldumosanes. Clearly, Line4 ismtnerelyaeud'lrunatic crossover, buthasalso m'dergoneamagnificatim event, suggestingthatthedlange inLine4's figmex) werntypewas concomitant withanalteration in its run. 31(g1inAnherstwfld—typestodc-B1esuppreescrintheAnherststock (Table 6) was mapped to the 3 chrcmosane in two experiments. Initially, several independent lines were generated with the Amherst X chromosome (re-marked with y) and an average of 1/4 Amherst autosomes. All of theseSmeresseng (pooleddata shown inTable 6). Onelinewas further cutcrossed two times to attached-x (£1395) females frcm a non- suppressirg stock, in order to further dilute the Amherst autosanes. Tenmales frmtheprogenyofthesecordwtcrosswerestocked individually with w. These stocks contained the Amherst X dwanoscme, and various combinations of Amherst and non-Amherst autosanes. All the stocks retained Su(Rex) (data not sham). The probability that all ten stocks contained a particular Amherst autoscme is (l/2)1°, so the sugaresscr appeared to be lie-linked. Areccrbinantwasthencorstmctedthatcontainedtheproximal portioncftheAnhersthirmnsarearritheaidlrmetinofarcn- ampressingdirtmosare (yufgarkmerst). 'Ihisrecanbinant still sugaressed Rex (Table 8). The converse reconbinant, with the axiarunatinofthekmerstdlramsaneardtleprmdmalportimofamn- ampressing dutmosane (y Anherst gr) , lost suppressor activity (Table 43 ms-mwmmxm y l; 1 a: mat 2213 2496 1 0 0.04 O O 0.00 ,2 Auherstm 2375 2757 . 4 6 0.39 22 4 1.00 Mnrstfamleommtedtcmalatrmeitlerofbnmlpprossimstodcs. 'Ihe proximlpcrtimoftheAmhersthirumscnewasrecoveredinaylztgr recanbinantsmfruansc—llztgzflf'y-l femalescrossed toy/ymales. PbstofthedistalportionottheAnherstxdmnsazewasreccvered inayzgardable-crossoversmfrmAnherst/yzgyfgmtlerscmssedtcy/X males. Wimntmalswerematedtoygyfw females. 'meresulting mmmmmmmmted individuallyto xLflmnmgyrn —/9 malestotesttormressimcffigx. 44 8). 'Ihissqlpressoristhereforeintheprmdmalportimofthex dlranoscne, intheheterodircnatinorinthefiofx-amrunatinthatis prcncimaltogg. mmmd-Table9listsotherstodcswhid1cmtainsmpressor activity and preliminary napping data for sane. ‘Ihe suppressor strength ismeasuredasin'rable6. Atleastomsumressorwgyyfgr) appearstcbeautosanal (13.6. Robbirs, unpublished data). Other alppressorscamntbereadilympped(mg1flg,ygyfiamm) becausetheyareininverteddlrumcnes. Discussion 'Ihese experiments lead to two conclusiors about R_ex. One is that R_exiswithinthepmxirelheterodlrmetinofthe3d1mmsatearrithe otheristhatggproducesamvelfimctim(aneamrm)orisaextreme hypenmrfil (with many-fold higher expression of wild-type function) . We canfurtherlocnlizegbyanalyzingLineZL 'Iheexdiangethat gereratedmm27wasclearlyinfl1erNbecauselire27contains diagnostic IGS variants frunbothparents. SinceLine27 alsohas partial B_e_x activity, at least mag; sequences rust lie proximal to theexchangepoint. Becauseline27hasa$gLR£m,wecamntdetermine ifBgisarepeatedelatentsplitbythisexdlange,orisentirely proocimaltotheexchange. Inanycase,&xiseitherwithintherCNAor entirelyprmdmaltoit,becauseiffigweredistaltothem,mre27 mldhavenofixactivity. Saneotherinterestingresultshavealsoanerged. (heisthat thereismorethanmedumosanewithasuppressorofgx. Altlnagh 45 m9-Otmramotu W M nil/mm?“ 0.12 m: xsaryfsarb 0.07 autosomal 1993MB: 0.24 ? I! 0.10 proximaltogronx 'WQEB 0.08 7 Severalctheerlrarosarestockshavebeenfmrritocanydaninant suppressors of Be_x. Scmeoftheloci havebeenmamed, andtheresults areshowninthelastcolunm. Otheerlrmoscmestcckshavebeen tested, anddonotsuppress 3g, including: Y In(l)dl-49,y w l§§ _i_l__Df 1 X-LY 99 $2 Df(1)R42,y l_z f LLL_n 1 sass-3.!“ y 1! eel aSwanson, 1984 bias. Robbins, unpublished data 46 mstsugpressorsarex-limted,anautosanalmmemhasalsobem fqmd. SeveraloftheX-limcedgrmexllocihavebeenmappedtothe baseofthedlrcmosare,proximaltogr. Ofmstinterest,severa1map tothebasalheteroduunatin. Baseimthemlecllardata,itislikelythatthe§p_&1in Wisacmanywitmhuem. rim31,wheretheenhangewas eitherdistaltotheronrdistaltoalldiagnosticvariantsinthe rDiA,ladcsthis§gLR;egQ,whileLim27,vteretheexdergevaswithin thermA,hasit. Havingthusmappedwtctheproxirelheterodmtin,wecan askabcutthenatureofguiggl. IfSu(_;Rg)wasahyparom1/anorfiait wculdbemimickedbyadeficiency. Butnone of thedeficienciee tested (Table 5) suppress Beg. Alternatively, if 5mg) was a hypermrph, it would be mimicked by a duplication. Neither of the duplications tested suppressedgx (Table 5). Therefore, wecanconclude that sums-2x), like fixitself, isalsoanecmorgh oranextreneseverehypermrfll. Finally, there remains the question of locating these two loci more preciselywithrespecttothem. Areoneorbothactuallywithinthe rENA? Ifsc,thissuggeststhat&¥ispartofamlti-elementsystem, located in, arriaffectedarrlappressedbydifferentbb loci. The .experinentsdescribedintherextdapteraddresswmtherbomggggard Mareactuallywithintlerm. m4 mammmmmbm mnmlcrrm Both the genetic and molecular data presented in the previous diapter suggested that 1135 and at least one $.1_(Rex) are located within the m locus. Raw can these loci be mapped more precisely? Are they withintheronratoneenioftherENAarray? Aretheseloci composed of repeated elements, divisible by recarbinaticn, like the rum itself? 'Iwonovel tedmiqueswereadoptedtoaddressthesepoints. First, torecovernoreexdlangeswithintherENA, Bgitselfwasusedtoinduce recombination in a Be_3g—bearing target chrcnosare. The rate of spontaneous exchange in the heterochr'anatin was arproximately 0.01%: 133; activity varies considerably but generally induces rENA recanbination at least 100 times more frequently. Reg-induced "hairpin" exchange also allows the recovery of both products of each W event, the mitotic equivalent of a full tetrad. Seccndly, the exd'lange events were analyzed mlecularly, using a new technique, recently developed by William et al. (sumitted for publication) . The technicpe uses the polymorphic length variants of the rtNA intergenic spacer (IGS) as genetic markers. Each distinct length variantclassisdisperseddifferentlytlmlgl'mttlerm. SaneIGS spacerlengthclassesareclustered, mileothersaredispersedover 47 48 mostofthermA. Eadladlarqebreakpointismmedwithrespectto thelergthvariantsoftheIGs,gereratingamapofthelimitsofeadl Icralelgthvariantwithintherum. 'meBexmenotypecanthenbe positicned with respect to the IGS variants. 'mermultsoftheseeimerinentsdamrstratedthatmisa mappable functicn with discrete limits within the rDIA. Wt inadvertantly,afig_(3g_)lomswasalsomappedtoadiscretepositim within the rum, and was shown to be divisible by recombination. mm IEIQI The first step in generating exchanges is to recover an appropriate target chranosane by a single crossover between a lag-bearing dmtlosane and Mfl—w‘fi (Fig. 9). The Q chrcmosare is a regular sequence chrcnosare duplicated for the rtNA, which was generated by @- induced 'mmm" excl'lange of magi—1111M (Rctlbins s. Swanson, 1988) . The resulting target, Myra; has a distal nucleolus organizer, devoid of other heterochrcmatin, and structurally normal, M—bearing, centrmeric heterodlranatin (Fig. 10). Maleswiththistargetdlrclosalearematedtoyggyfgeg/y females. In sane of the progeny the target chrunosane will undergo spiral or hairpin exchanges (Fig. 2) between the two Q loci. spiral exchanges delete the neterial between the two m's, yielding x-fragment damnosanes. Progeny containing such chrcnosanes are readily detected assterileyigar-tX/x-fragnentmles. Hairpinemhange invertsthe neterialbetweenthetwom's, inthiscase, convertinga me dlrmosale back into an M dlranosare (Fig. 10). The progeny curtaining hairpin exdlarqes are phenotypically 49 agnmfl'u-Wma X I blank" 7 Fl? / 01(13):: N'N-n "Bu-z! X 1 1 than V ”(1:1). "’“Mfiw X [ovum Y I MOST RARE .L X Dp(1;1)lr "‘"Nmu I! or WI)! "'5’“! N”, at 1 Y y(cvvlflu) ylcvvlflu) X Y “any 010ng cred-ova eon- Fu on! an 0 crouch: 00m 0 at. rear) ”8.9”. y ‘2'") Figure9-Hatirgsdmfcrremveryon-mdiatedinversiasofa M—bearirgtarwtdlrm 'me first three steps generated the appropriate Beg-bearing target dmmsarewithaduplicationoftherflflatthetip. 'Ihischrcnosane wasstodced(mtslmn)arrithentarget—bearirgmaleewerematedto&x fanales. Asshown, theregular daughters ofthismatirg carried either theminvertedtargetoratargetwhidlhasmriergoneBex-nediated inversionofthematerialbebdeenthebuom's. Becausethetwox duanosanesofthemtheringeneratim3recadainedfreely,the generation 4 daughters had various canbinatias of the markers g, y, f andfix,butinallcaseshadoneygriamcneyimxmranosane allowing detectim of single-crossovers over nearly the entire length of thedlrarosane. 'Iheywereiniividuallymatedeithertotheynales Morwrclyfaexorfimyfarmles- 50 Figure lo-E-inducedhairpineldlarqe inatargetdlrcnnsane cartainin; both E am _.CR_€M The effect of the hairpin exchange is to invert all the material between the two NO's. The effect of the exchange event on the positions of g and Su(Rex) depends on their locations with respect to each m. For exan'ple, if 31(Rex) is distal to the NO at the tip, maryi, its position will not be altered by the exchange events. Alternatively, if itiswithintheNo, saleexchangeswill wewmtheproxinel end, while others do not affect its position. Clearly, if either kg or Su(Rex) is a repeated element within the NO, it is possible that an exchange event will split them, giving activity at both recanbinant ends. 51 Dp(1;1)wm51bLRexFE car Rex of" A A A A V V V y+ Su(Rex) H.6X-_ i ....... C00 mltOIIC hange Rex? Su(F?ex) ? _Ol 7L: -='—l- Rex? Su(F?ex)? y+ In(1)w’"51wam4R, car ‘7 Al. ‘7 A; 0 8) AL ‘ 52 identical to regular females. The only difference is that these daughters are inversim heterozygotes for virtually the entire X euchrunatin. Thus, to detect hairpin products the frequency of crossovers between y ard Q; is measured in each daughter irriividually. A female bearing a hairpin exchange dirtmosale will have markedly reduced recarbination because of the absence of single crossovers. Once the inverted exchange products (designated mylflblgwm) arereccveredandstocked, thetmrecanbinantbb loci on the chranosome are separated by single crossovers with ELLEN—W1! gt 1!) j which contains virtually no rINA (Appels & Hilliker, 1982) (Fig. 11) . The dlromosomes bearing the separated NO's are referred to as "recombinant #-distal" and "recombinant #-proximal", inorldertoindicatethesouroeoftheruIA. Theendsoftheparental lasagna—fig duamsone, which itself is not inverted, are separated by single crossovers with Df(1)X-1,y ac g f 332;, which is deleted for most of the basal heterochromatin including all of the rDNA and two proximal euchromatic lethals. Thenextstepistoneasuregegactivityardbgphenotypeofeach eniwiththestardardassays (ChapterZ). ThentNAofeachenzlis analyzed for IGS length polymorphisms (HaeIII digest probed with cloned 16:5) and insert sequences (HincII/HindIII digest probed with PA56) (see Chapter 2). The HaeIII blots are scored for the presence of each 168 lengthvariantmiquetooneortheothermoftheparentaltarget dlrunoscme. To avoid antiquity, ally easily differentiated variants are used. Hapsoftherflflarethenbegeneratedbyaproceduresimilarto that described in Willians et al. (submitted) . When approaching this 53 Figurell-Sqlaratimof recmbinantm'sofhairpineadlargeprodmrts The inverted products of hairpin exdlange events have two recombinant NO's. In order to study each individually, the two ends were separated by recombination with a dlromoscme nearly devoid of m, minim (Appels s. Hilliker, 1982) . Males containing the hairpin excharge product were mated to Mgfisglém‘y _c_:_t m f/IMlldl-lim g y 5 females. The resulting Mwfikflélfl/Mflb—W daughters were then mated to y/Y males. Threeyic_t_c_a_rsonscontainingthedistalNOarrithreeymfsons containingtheproximalNOwererecoveredandusedtoestablishstocks. AllthreestocksofeachNOweretestedforRex, almexlandb_b W- S 0.59.5 oco Rescue oco .990 auntie—«CS E: 32.2% .... m x .0 too 4; ilillllv1 e l 4 I! i u " ”IL-l SEE see its a. .1. . s i e . " [IT in“ u HHIIT to 4x 235 messeoeme 55 aralysis,however,itisinportanttomtethatwhenhairpinetcharges areusedeadlreccnbinanterdhasvariantsfranbothoftheoriginal parentaltbloci. mrthermre,mlessdeletiasweregeneratedbythe edurgeevent, all oftheoriginalparentalvariants shouldbepresent inmeorbothrecmbinantends. AdiagramoftheprocedureisshownisFig.12,whidloutlinesthe procedure using a simplified array. Adetailed example of this mapping procedure, using data for one ofthe I$ length variants, is described intheResultssecticn. InFigure12,threelengthvariantclassesareshowninthedistal parentalNo. Thefclrcopiesofaarewidelydispersedthrulghoutthe rmA, whilebothbandcareclustered. First, exchange pointsinthe distalparentaerIAareorderedaccoroirgtothenmnberofparental distal-end variants remininginead'l recarbinant'sdistalNo. For example, if all thedistal-end variantsare still present ina recanbinant'sdistalrmAarray,thatexdlaxgepointisproxineltoat leastonecopyofeachofthevariants. Afterorderingtheexchange points,thedistallimits of these variantswasdetemined. That is, if the distal em of a recanbinant lacks a variant that was originally in theparentaldistalm,theexdlangemlsthavetakenplacedistaltoall copiesofthatlengthvariant. Prooeedingfrunthetelmere,thelast exdlangeeventthatremovesall ofagivenvariantdefinesthedistal limit of its distribution. The same logic amlies to ordering the mergepomtsintheproxinalparentalmardfiniingthepmximl limitofeadrvariantbyuseoftheprmcimalrfllAarr-aysofthe recclbinants. 56 FigurelZ-lblecflarmimofrflflarraywithICBlegmvariants Twohairpinedlangesinthesanetargetdrrmosaneareshown sdleneticnlly. For sinplicity mly three I6 length variants are show, all inthedistalparentalm. First,thetwoexdlangeeventsareorderedwithrespecttoead1ou1er withinthedistalparentalm. LookingatthedistalrmAarraysofthe recaibinants, #2 clearly took place distal to #1 because fewer of the variants remainatthedistalerri(onlyain#2,while#lhasaandb). Theseexchange pointsarethenusedtomapthedistallimit ofthevariants’ distributions. Notefllattledistallimitofaisbeyarithesetwoexdlangepoints,since bothretainsanecopiesofaintheirdistalm's. midlangeevent #2 moves allcopiesofbtotheotherm,butexdlange#ldoesmt. Proceedingfrom the telanere, #2 defines thedistal limit of variantbsince it is the last exchange event to move all copies of b. Both exchange events moved all copies ofctotheproximal 170. Therefore, #1 definesthedistal limit of variant c, sinceitisthelastexdlangeeventtonoveallccpiesofc. Therextstepisdetenuinimtheorderofthetmexdlangeeventswith respecttoeadlotherbylookirgatflleprudmlm'softherecmbimnts. Again,withinfl1edistalparenta1m,#lnustbenorepmxinelsincefewer variantshavebeenmovedtoitsproximalm(onlyaandcinfl,while#2has allthree). 'nreorderedexdlangepointsarethenusedtcmaptheproximl limit ofthevariants' distributions. Theproximal limits of neitheranorc aredefiredbythesebdoedlarges,sirceaardcweremvedtotheotherm bybothexcharges. W,exchange#2mvedbtotheotherm,m1ile ecdlangefldidnotmveanycqliesofb. Prooeedingfrmthecentromere, #1 definestheproximallimitofbsinceitisthelastexdlangeeventthatdoes not move any copies of b. 57 «Housman .[ 0 I . cos. 2.83 «i . ‘ ... - ... . 02 Sioux I I + N sun I ... ... p 02 :50 0 D < 02 §¢<> gulgmc Ila! ( 4 39.2.05 58 'merextstepistodeteminetheotl'lerlimitofeadlvariant's distributim. deeterminetheproocimal limitofavariantuniqueto flleparentaldu'cnosaie'sdistalm,theprmdmalrecaibimntendsare smredforvariantsthathavebeenmvedfrmthedistalparentalm. Oneagain,theeadungepointsareorderedbythenmberoforigimlly distalvariantspresent. Tl'lenbstdistalexdlangepointisthatwhidr mvesallthedistalmvariantstotheproodmalerriofarecalbimnt. Similarly, if a reccibinant's proximal err! lacks a particular distal variant, the exdlange eventnust have takenplace proximal to all copies of that variant. Thus, proceeding franthecentrarere,thelast exdlange lackingavariant frmthedistalparentalrmAdefinesthe proximal limit of that variant. arceagain, aparallel process isused tomapthedistallimitsofvariantsmidlareuniquetothetarget dlromosare'sproximalNo. Finally,thetwosetsofexd1angepointmapsarecarpared. For exanple,therearetwonepsofexd3argeswithinthedistalpare1talerd: onegeneratedbyanalysisofthedistalendvariantsremeiningin recmbirentdistalerdsardaegexeratedfrundistalerdvariantsmved totheproxinelrecanbinanterris. Eachlnaphastheexdlangepointsina particular relative order fran telanere to centranere. The two maps can becaiparedtoseeiftheexdlangepointsareinthesanerelative order. Iftheyare, that indicates thattheexdlangesweresinple single exchanges. Differences in the relative order would indicate that therehadbeenrearrargeuentsduringtheexdlangeevents. 59 mamas Mimetmwn—WWW-m mtructimoftheW—Iggggrtargetdlrm,itwas essertialtoslmthatitdidirrieedfmctimasatargetforg activity. Malesbearirgthistargetd‘lrmnsaneweremtedtom fanales. 'nleresultsareshcwnin'l‘ablelo. Wmisa itarget, though a relatively poormewith only 0.14% detadment frequency. Thisrateislowevenwhencmiparedwiththeoriginal mmlbym‘zgdmosmemmwasusedastmsmroeormedistal erri (0.40%). Itwasalso inportanttoslmthattheproximlerrlcf the target had retained normal fix activity. The ends of the target duumsarewereseparatedanifemalesbearingthepmximlmhereneted tomales bearirqanattad1ed-)_<_Ytesterd1rmnsare(Table 13,1.ine2). Clearly, the proximal NO retained Be; activity since it induced detachments at a high rate. The separated ends of the target chromosome mealsotestedfortheabilitytosurpressthepherntypeofastardard geldrramsare. Thedistalemiwasfwrritohaveaggmex) (Table 14, Linel). Mercy 9f recombm In(_1_)hfl§&I§gm—fl,gr dlromoscmes - The mating scheme (seeaboveandFig. 9) allowedthedaughterstobetested directly for y ca; recombination without a stocking step. Oarplete progeniesweremtccmted. Instead, thefanaleswerejtrigednon- reccnbinant, ani discarded, whenever there were enoth crossover products to indicate that the data were outside the 99% (cmnlative binanial) confidence limit for an inversion heterozygote. For each m cm MW43 1998 1717 1 7 0.44 4 4 0.45% MM 2828 2816 2 5 0.30 3 2 0.14t ygyyffiex/yfatalesmremtedtomalesbearirgtminiicatedtanget dircmsane.pitatiwreombinmt,sevenlm-crossweryimalsueretsedtb establishisolinestocks. 61 Fruneadiisoline, amalewasirdividually matedto yfiyfmfateles. 'memapdistanoebecweenyardgr,markers gremlingtllexwlruratimwasthenmeasmedintheresllting Wflygyffigdamters. Inadditicn, foroneof theisoliresofeadlrecaibinant,fllecrossoverfrequercyinead1nerked regionswasmeasured. Inallcasestheisolineswereidentical (data mtsham),ardalltherecarbimntshadsignificantlyreducedmap distances, reflecting the absence of single crossover products (Table 11). Intotal,ninehairpinexdlangedlrmosamswererecovered frail anong3912dlrunosanesscreened,afrequency of 0.08%. genetic analysis of recombinants-E‘ad‘l hairpin recanbinanthastwoends thatwereseparatedardthentestedfortg,&xard&l(Rex)pherntype. For each erd,threeindepexrient recombinant linesweretested. The results are shown inTables 12-14, and summarized inTable 15. Notethatwhengactivitywasneasuredinthedistalerfls,which areyi,o1eremaybesaxebackgromdbecauseexcmigeshetweenyiardw yield one crossover type (yi' w 152], males) which is phenotypically imistinguishable fran a detachment male. To estimate the frequency of sudaevents,thenapdistancebetweenyarriwwasneasuredintm crosses: Mkigegecer/yrsnlxyrsplflalfi waxywsEl/X.Therewere6/ls,215recadoinants for the first cross (0.063% crossover frequency) and 15/9455 recanbinants forthesecadcross(0.138%crossoverfrequencywhentheregularard recmbinant males are doubled to account for the hanizygols lethality of W). AGtestixriicatesthattheresultsofthebmtestsare lumgeneous, ardtogethertheyyieldamapdistance of 0.11mapunits. 62 Mon-mmtimdatafcrmeWm W mmfimm— Mimic: 35 384 0 1 2 0 0 4 2 15 8 5 17 2034 363 l 3 2 1 0 3 1 11 4 7 15 2057 577 0 1 5 0 l 3 0 17 12 5 13 2337 538 0 7 1 0 0 2 1 20 5 4 13 2569 392 0 3 1 0 0 6 0 10 4 4 12 2934 503 2 3 2 0 2 4 3 14 2 2 11 3318 507 2 4 5 0 0 5 4 17 15 3 18 3394 457 1 3 0 l l 5 3 10 8 2 13 3488 318 0 3 5 1 0 l 0 5 3 2 9 Anhflividnlmlebearirgflairdiceteddumosmebescrossedmygyxgg/flfl females. Fivedmrcmosaue/mdaughtersberemteitoyggyfgarmales. Amarent single crossovers in the inversion chutrrzsznes are double crossovers with one event outside unmarked region, butwerecomtedassirglecrossovers incalallatingmap length. Ith length is caLculated as: (scn42x000) x100/nm+scl+0co). To confirm these results, at least two other individual males bearing the indicated dummmatedmygyyffigymfaelesmddumosam/Bgdaumtersm netedtbyzgyyfmmalestoecoretmy-Qrdistarmwasmeasnedmnamt sham). 63 mn-mmdmmmmm W airman/9112 m 132-" A: {EL W: W 565 1 0* 700 y a so 1 9g; M 125 - 313 71% 547 W 35, distal 726 599 45% 1601 35, proximal 1369 0 0% 1499 2034, distal 265 679 72% 1056 2034, proximal 672 71 10% 829 2057, M 0 0 661 1231 2057, proximal 953 3 0% 961 2337, distal 727 43 6% 840 2337, prox1mal' 804 8 1% 924 2569, distal‘ 659 4 1% 717 2569, proximal 165 502 75% 873 2934, distal o 0 bbl 918 2934, proximal 627 0 0% 724 3318, distal 661 37 5% 732 3318, proximal 617 9 1% 679 3394, distal 11 352 97% 699 3394, proximal 687 2 0% 665 3488, distal 0 0 bbl 821 3488, proximal 133 2 99% 1032 Malsbearingtheirdiczteddumosarewerecrossedto Hillel-41249; y B/nggly females. The data are pooled from three indepedent reombinant lines for each NO. mu-mwfltydpmmmmu)‘. IEEDLAR lllfiin£UUNCEfllflfla UEZMllflaflfPRDGEN! W W W 7649 4132 4 9 0.2 5 0 0.1% y a: £9 1 g; m 1257 1743 3 10 0.8 105 21 9.1% W 35, distal 1877 2257 2 133 6.0 87 42 10.3% 35, proximal 2916 2130 4 203 7.6 0 0 0% 2034, distal 909 833 1 49 5.3 35 12 8.3% 2034, prmdlnal 1860 1221 3 124 7.6 0 0 0% 2057, distal 2237 1089 5 140 6.2 3 2 0.2%ll 2057, proximal 2630 1682 0 133 5.8 1 0 0.0% 2337, distal 2258 1592 0 102 5.0 0 0 0% 2337,prcnd.ma1 2723 2087 3 103 4.2 0 0 0% 2569, distal 3292 3647 2 102 2.9 1 l 0.1% 2569, prtncinal 2248 2433 1 121 4.8 88 33 5.1% 2934, distal 2713 1590 4 130 4.3 1 0 0.0%:5 2934, proximal 2789 1870 3 96 4.1 1 0 0.0% 3318, distal 3624 3615 2 137 3.7 6 2 0.4% 3318, proximal 1369 1020 0 51 4.1 0 0 0% 3394, distal 762 845 5 39 5.1 31 14 9.1% 3394, proximal 1553 1133 1 50 3.7 1 0 0.1% 3488, distal 1847 1023 3 '42 3.0 0 0 0% 3488, proximal 691 524 2 44 6.9 19 7 3.6% Proximalremnantm'samallmumm‘xnthhiledistalrmbimntm'sareall [pupil-5.19m. Halesbaarirgflnirdicrateddlrmmcrosadtoyymyyysm fenalasardthemdtirgdaughterswrematad'tothestmflard XSPXL‘llnlllmdytbty-tastermales. Siroetl'xeraombirw'ttm’sareininvertad seqnme,fl1enmberotpa&oclim15mlasmflactsbothflnsensfltimfrmmn— disjmctimardtrnserasultirg firm fair-strand dmbleeadlamainthefanale. 'me distalm'smallmrkeduithy,somlyy—gglmlesomldhesooredaswnle- bodydetadmwts, anithedetadunentrateismlallamdas: 0&me— regulart‘anales-t (ny wsplmalas) +qynarrircn3rfla Note, waver, matinthadistalmczossas,crossoversbetuaeny—ardyooalrata frequency of 0.1% and includemeclass that is irdistinguishable trundetadments. ‘Ihispointisdismssadintlntext. 'manmbersarepaoladdatatrunthme Wmimntlimsforeadlm. asmmesem'smmanmivirgy- +mlumuoordud0tadwt6. 653 mu-mwutyotpnm-amm-o m WWW 2917 1267 1 1 0.1% 2 0 0.3% y at; as 1 m m 653 799 4 8 1.6% 40 8 6.9% mammal: 35, proximal 1680 1453 0 52 3.2% 1 0 0.1% 2034, proximal 1583 1502 1 50 3.2% 0 0 0% 2057, distal 1023 656 2 28 2.4% 53 9 5.7%! 2057, proximal 1866 1524 1 54 3.1% 1 2 0.2% 2337, distal 2110 1570 0 46 2.4% 0 0 0% 2337, proximal 2404 2105 2 87 3.8% o 0 0% 2569, distal 2879 3136 2 51 1.7% 4 0 0.3% 2934, distal 423 332 0 11 2.7% 27 3 6.6%3 2934, proximal 1269 1363 4 62 4.8% 0 o 0% 3318, distal 3393 3064 3 78 2.4% 7 8 0.6% 3318, proximal 1791 1502 1 52 3.1% 0 0 0% 3394, proximal 1676 1338 1 61 3.9% 0 2 0.1% 3488, distal 1866 1010 1 32 2.2% 23 11 3.3%II withtheexoaptimoftheparentalprmcimalm, mlym'swithmagxactivitywere smredfordnmofm.Mesbearirgmim1cateddirmosamm crossedto wwglgyfm/flflfanalasamm/Qdmghtersmmatedtom standard meyxB—yi'testermales. Detadmentratasfordistalm's wrecalmlatadasin'rablefl. 'Ihereombimm'sareinirwertedsequenoe, the mmberof patroclixns males reflectsboththoserealltirg tron mr-disjmctimard tinserosultingtront dableeltdxatgasinthetanale. aSince:thetaadistalm'sare allyi'maleswresooredasdetadmnts. 66 'Iablels-Sumaryofmntsforparmtalardrecxmbimm's %_ ML. M M m % 1b activity 31M) % fl) activity aim) 0 . . 1 W 0% 0.1% yes 71% 9.1% no remnants 2569 1% 0. 1% yes 75% 5 . 1% no 35 45% 10.3% no 0% 0% yes 2034 72% 8.3% no 10% 0% yes 3394 97% 9 . 1% no 0% o . 1% yes 2057 bbl 0.2% no 0% 0% yes 2934 1:61 0% no 0% 0% yes 2337 6% 0% yes 1% 0% yes 3318 5% 0.4% yes 1% 0% yes 3488 bbl 0% no 99% 3.6% no 'me reocnbinants are grouped according to their similarities as disalssadinthetext.A11dataweretakenfranTab1es 125-14. Note thatthedistalmgggactivitieshavealwbadtgmrdduetothe omfusicn of a crossover class with aculal detadxment males. 67 'nnsecrosswersareveryinfreqmrtmfidomtsignificzntlyaffactflxe mlailatim of M activity in ends which are clearly g. In the parentaldistalm,aniintheZ934distalm,thesmallnmberofyi’y mlescanbeomsideredcrossovers. 'mereare, however, threecases (2057, 3318and2569) wherethere naybesaxepartialfiegsactivityinthedistalelfl. I-brthese,atleast oragynararomoipn patroclinous for mlyonemarker (xi) wasreoovered. Thesegynarflrmorfixsaremfimbtedlydetadmentpmductsarflmt spontaneous. Alflughanoocasionalyimalethatlookslikea detadmentisreooveredinomtrolexperinentswherefigismtpresent, mgynarflmmrmshaveeverbeenobserved. Inonesud'lgroupofoontrol crosses, among 41,086 daughterstherewere4detadmentmalesbutm gynandrcmorghs. 'Ihus,themaxinmrateofspontanemsgynardramrphs is 0.011% (99% binomial confidence interval). Unless 2569-distal carries as, even one gynandrmorm is highly unlikely. 2057-distal ard 3318- distaleadugavetwogynardrumrphsarfl,surelyevinoesane&x activity. mehailpinreoarbinantsreooveredarequitediverse,mtthereare saregeneralpatterns. 'metargetdlrmosanewasgpardmatthe proodmalend,aragitaxagu_8axlattladistalera. Onereoanbinant, 2569issimilartothetarget,a1tlnlghtheremayalsobesanem activity atthedistalend (see above). mreeothers, 35, 2034and3394 appeartobethereverseofthetarget,havirgmardmatthedistal erdardag(Rex)amm'-tattheproximalerd. 'meotherreombinantsshowmoredramaticdianges. 'mo reoadoiiants,2057am2934,aromwmlatthedistaleraam grand 3.91%). at the proximal end, with little or none}; activity at either 68 ad. Preslmbly,mostorallofthefigxactivityhasrunainedatthe prcncinalerflardissurpressed. 'Iwootherreombinants, 2337and3318, havelostthempienotypealtogetherardhave splitggjgegl. nrterestirgly, 3318amearstoluvaamakm1alagwiulggxatthe distal end, sinoethere ispartialfigxactivity (0.38%, includimgtwo diagnostic gynardramrfils) along withthe suppressor W- Finally, 3488 has loam ocnpletely, and show. a net loss of rum ewressim,wifl1am1dista1mardaseverelymand3gxproximalm. Itappearsthat3488hasade1etimofrmhasaresultoftheexd1ange. marmalysiestmts-nafirststepinmemmeauar analysiswastopreparegermictlfllfrmfi/Omalesmeretheonlyrum wasintheendbeingstudied. nasprecludedaxalysisorthethreeppl ends,however. InthecaseofthedistalNOoftheparentaltarget W,yllg(1)so§§ females were used as the some of 1104 because thedistalmdlrumscnebearstwoelxhrunaticlethalsthataremt ooveredband1runosane,hitareooveredbyLm1)sc‘—‘§_&. Itwasinportanttodetermineumetherthediangesingbrherntype seeninthereombinanterdsuereduetoiractivatimofrmhcistrons byirsertionsequenoes,ortod1angesinroncpymmber. [NAfrom ead1 end was digested with HimII/HindIII and probed with PA56 (materials & methods). 'Ihe ratio of intensity of the band at 373 bp to alltheotherbalflsreflectstheproportimofintacttointerrupted cistrms. Asisclearfrm?ig.13,fl1ereismevidentd1angeinthat ratio,eva1M1enoarparingfi1eleftparentalerdin1are1(bpt)tothe most severelymreocnbinant tested, 3488p, lane 3. 'Ihewide rangeof mWinreombinanterdsmstbeduetodurgesinthemmber 69 35 3318 parents 2057 2934 3488 2569 2337 D P D P P D P P P D P D P bb phenotype: 4/- 4 4 4 - 4 4 4 - 4 - 4 4 1635 bp - 5... 02. 2A Pigmen-aimn/Hmamaigastorm'sgaamtadbygag-mioad hairpinexcharg mmicmtmL/Omlesbearirguieirdicatedmwasdigestedwifll HircIIardrfirdIII,e3a3eptinu1ecaseoffl3eparentaldistalmvmere magi/mg fanaleswerethesairoeofm. more themtypeof eaohuo is iraieated, witha "+" damnation? (<10% m) marda"-"denotilgag(>50%_b§)m. Notethattherewasan inocnplete digestion of the INA in lane 7 (2057 proximal). 7O ofrepeatsarflmttoanalteratimoftheprcportimsofactiveto inactive cqaies. WM-mmmamlemlarmpofmetmm arraysofthetargetdumoscue,germicflflme/Omaleswithonly one specific rmA array was digested with HaeIII and probed with IGs spacertorevealthearray of res length variants (Fig. 14). 'Ihe parentalendswerecmparedtofirdinformativevariants. Eadl reoatbinantelflwasthenscoredforthepreserneorabsenceofthse variantsinatleasttwoindeperdentdigestsbytwoirrlwendent observers. 'meresultsofthatanalysisarestmnin'rable 16. ‘Ihe datainTable16wereusedtoconstructthenapofIGSlergthvariants shmninFigurelS. The mapping procedure is described in Dcperimental Design, but it isinstmctivetofollwanexanple;thedistalarrlproximal limits of variantkwillbemapped usingthedatain'rable 16. Note, however, that recombinant 2034 will be considered separately below. ‘Ihe recatbinanterdsarealreadyorderedin'rablemaooordingtothe mmbersofvariants. Icoldrgatthereombinants'distalrmharrays (partc),35isthemostdistalexd1angeswithintheparentaldistalN0, since ithasretainedthefemstvariants. Bothexchanges 35and3394 mlsthavetakenplaoedistaltoalloopiesofkbecausekisabsentfrom theirdistalm's. 'Bleatheratdlarges, 2337, 2569ard3318,nusthave takenplaoeproximaltoatleastaleoopyofk,simetheyretainkin theirdistalNO's. Prooeedirgfrunthetelarere, W394 defines thedistallimit of variant k's distribution, since itisthelast eaadwgemwvirgallofvariantk. 71 mls-mmmmmmxm's ].l_ngi3 telanere 2337 2569 3394 3318 +3 + 35 2034 centrcmere + m. 2034 + + +/- 35 3318 3394 2569 2057 + 2337 + 2934 + 3488 telonere 72 Table 16 (curt'd) Distal (aim-bearing) invariants Q)Ml'afisbcd_gjxhlimm) telomere 2034 - - 35 - - 3394 + 3318 + 2337 + 2569 + +++ ++++++ +4-4- +4-4- +++ +++++ ++++++ ++++ +++ d)Proximal Ends centromere 3488 2569 2337 3318 2057 3394 35 2934 2034 telomere +++++++++ ++++++ +++++ +++++++++ u ++++ ++++++++ ++++++++ ++++++++ +++++++ HaeIII digests of genanic [INA fran flies bearing the indicated NO were probed with IGS clone. The autoradicgraxrs were scored for the presence (+) of each parental N0 IGS length variant. An "x" indicates that the bard was not sooreble on the autoradiograns. In addition, using data from Tables 13 and 14, the last column indicates whether Be; or Mmpresent (+). Itisnotpossibletoscoreg activity iftheNObearsMexl, andthatisindicatedby the notation "So". “Although the distal NO of 3318 did have somex) (Table 14) , it also had Eel; activity (Table 13) , based on the fact that therewerengmrfirumrpl-isproduced (seetext). 73 Hex 23 Prox. Dlst. 1 . 2 . 3 . 4 _ - b 5 _ - c - d 1! " —~_.-_——-'— Figlre 14 — as images" " intarwt m‘ 163‘s r Gaunicfllhwaspreparedfrmfliesbearirgtheirflicatedmard digested with HaeIII. The blot was probed with cloned 16$ sequexne. IGS lengthvariantsmiquetooneortheothermaremarked, using nmbersfortheagmardletters forthedistalsmrzex) m. 74 FigurelS-lbleallarmpoftarwtdlrmm's 'mepresenoeorabseroecflcs lengthvariantswassooredinthe9 g—induced reoonbinants of the target chromosome sham. Dcdlange pointsamirdicatedbyamarflweredetemiredinGcasesbyusirg information from both 10's of the recmbinant dirtmosate. In three cases (2057, 2934 and 3488), only the proximal NO was assayed because thedistalNOwasbbl. Sareexdlangepointscanmtbedistinguishedon the basis of the scorable IGS length variants, and these are shown as being at the same position. In one case, 2034, there were clearly deletions of all copies of sane length variants. The site of 2034 exd'iangeeventisslwnasaregioninsteadofasapointbasedcnthe positions of the deleted variants and the variants which remain. The proximal and distal limits of each IGS variant are shown beneath the dlromosane. The locations of 33x and 511(Rex) are shown above the chromosome. Note that even though 2337 am 2569 have identical distal breakpoints based on the ICS variants, 2569 does not move Ema!) to the proximal NO, but 2337 splits Su(Rex) . Ecchange event 2569 must, therefore, have been proximal to 2337. The proximal limitofmisirrlicatedbyadashedline, becausewecannotdetermine if any Begs activity remains in the 3394, 35, 3318 and 2034 proximal NO's because they can? M- 75 Figurels DISTAL NO L Su(Rex) : 35 2034 2934 339418 2337 2560 3488 e 9 :- ; b 1 I l - d,h,l ; ‘fi C 2 2 k ‘3 ._J__. PROXIMAL NO ; Rex leeon0seelesseesseeseeeoeeeesesees 33% 2057 2569 3394 33? m4 4 4 4 4 :— 7 __ : 4,5 3 2 i 6 3 76 Theexohargepointsinthedistalparentalmcanalsobeordered basedmthenmberofdistal—endvariantsmovedtotheproodmal reoarbinant rmAar-rays (part (1). 3488 is thenostproximal event, while 35 and 2934 are the Host distal events. Ebcdlarges 3488, 2569, 2337, am 3318 must all have taken place proximal to all copies of variantk, simetheydidmtnovevariantktotheproximal NO. Exchanges 2057, 3394, 35 and 2934 must have taken place distal to at leastonecopyofk, sincetheyhavemovedsaneoopiestotheproximal to. ~ Proceeding from the centrcmere, the proximal limit of variant k's distribution is defined by 3318, the last exchange that does not move any copies of k. In the case of 2034, sane variants are missing from both ends (j,k, arr! 3). Variant h is also missing fran the distal end. Clearly this exchange event involved a deletion at the site of exchange. In Figure 15 the 2034 breakpoint is sham spanning a region. It is also important to note that there are data fran only the proximal NO's for 2057, 2934 am33485ecausetnedista1endisggl. Theseeventsmayalsohave involved deletions, but we cannot determine that using data from only one end. The genetic data above clearly indicate, havever, that 3488 irdeed has a deletion. $u_(Rex) is no larger present at either N0 of this reoanbinant. Moreover, 3488 shaded a net loss of rD'NA expression asneasuredbymrhenotype (T‘ablelz), carparedtotheparentalNO's. W04 'memajorresultofthisworkisthemapshaminI-‘igurefi. The proximalanidistalrecmbilantm'swereusedinieperdentlytoorder tree-(chargeeventswithineadlparentalrMarrays. Therewas 77 eiwellentagreenentbetmenthetmsetsofdata. TherewereSexdange points within the distal m which could be unmibigualsly named relative toeadlotherusirgdatafrunbothrecmbinantm's. Bothsetsofdata gavethesameorder. Forexample, 35wasthemostdistalexd1angein theparentaldistalmbasedmeitherthevariantsrenainimin itsdistalm (Table 16, partc) ormvariantsmovedtoitsprootimal no (Table 16, part d). Similarly, the order of these 5 exchanges in tl'ieprcntimalmwasdetermiredusirgbothreombirantm'sarflagain, there was oarplete agreement. In other words, these five recanbinants appear to have resulted from simple single exchanges. ' There was one recombinant (2034) where molecular tetrad analysis 511041821 that, in addition to a single crossover, both NO's contain deletions, and both deletions map to the exchange site. Of course, there was only one set of molecular data for the three recanbinants with a bl} distal N0, but in one case (3488), the genetic data indicate that the exchange event involved a deletion. In 3488, M is completely missing and there is a net loss of rum expression relative to the parental dumosate. The data suggest that Be_x causes single exchanges that are sanetimes accompanied by deletions at the site of exchange. “hen the molecular nap is combined with the genetic data for each recanbinantend, thetwoloci, Bexarfialmex) canbemappedto positias within the m. For example, the three exchanges (35, 2034, and 3394) which clearly moved M to the distal and must have taken placewithinorproximaltofig. It is impossibletodetermine ifgex activity remains in the proximal NO's of these reoarbinants because mmbeennovedthere, ardtheproximal limittherefore remains ambiguous. As discussed above, it appears that 3318, 2057 and, perhaps, 78 2569 have sane M activity in their distal to. Hon its map position, it is clear that 3318 did move Beg activity, since the 3318 exdiange is proximal to 3394. Assuming that 2057 and 2569 have, in fact, moved sane mactivitytomedistalm, thedistal limitormisderiiadby excharges 2934, 2337 and 3488, which did not move any M activity to thedistalm. Moreover, if2569-distaldoeshavegex, Rexisa repeated and divisible since 2569-proximal has & as well. Thedistalandproximal limitsofwaredefinedbyexdange points 2057 and 2569, respectively. No M activity remains in 2934-distal, and ro 9.1(Rex) activity has been moved to the 2569— proximal. Notice that the 2569 arr! 2337 exchange points are identical on the basis of molecular variants, but they can be differentiated with respect to 3.1.893).- It is clear from recombinants 2337 and 3318 that Su(Rex) is a repeated element, since these two exchange events each gave rise to dir'a'nscmes with two gi_(Be_x)_ NO's. To unambigualsly determine if fiexisarepeatedelement itwouldberocessarytostartwithatarget which is free of suppressors, such as 3488. Starting with a target free of suppressors waild also permit further molecular localization of @. matmightgexbe? Be_x, whichislocated in, suppressedbyand affects various NO's, appears to be part of a mlti-element system. One nodal is that M encodes a site-specific erocmclease activity causing breaksinthetargetENA, andthatsurpressorsenoodearepressor. The breaks in the target INA would be recarbirogenic. In that light, it is interesting that at least two of the m—iniuced exchanges involved deletions, one of which was mapped to the site of exchange. Under this model, Rex-induced nicks in the two m's mild yield inter-m reambination, an! additional breaks, or mleolytic trimming of 79 broken erds, would yield deletiais that map to the eidiarqe site. It also predicts that we are likely to find deletiao among ron-reoarbinant M'sthathavebeenexposadtomaternalm. 'Bfismodeloffixactimhassmesimilaritiestoasystan described intheliterature. 'merumianihastwomajor classesof insertimsequemsswhidlintemptribosaual repeatsand vmidihaveahighdegreeofsequeloehamlogytothemmemr TYpe I and Type II rum insertion sequences (Bickbush & Robins, 1985; Eickbush, personal canmnication) . Recently, it has been shown that bothofthemmgi insertionelementshaveopenreadingframes characteristic of retrotransposons, and that at least one ORF encodes an endonuclease activity which cuts 288 ram site-specifically, jg) thrg (airke at al., 1987; Xiong 8 Eickbush, 1988a and 1988b). Along these lines, we suggest that the endomcleases encoded by the rENA insertion sequences are the cause of ;Re_x activity. Clearly, not every insertionsequence inQWactilallyproducesaBg endonuclease. Not every No is Q, yet in the literature every x-linked no examined contains both types of insertion sequences (Glover, 1981: Iorr; 8 Dawid, 1980) . In particular we have dlecked the rum arrays in two lab stocks that are neither fix ror Sing) using the PA56 prcbe (Chapterz; Fig. 13), aratheyoontainboth'lyperam'lypen inserts (data not sham). It is likely that fig; is an alteration that allows theerriomlcleasetobeproduoedatadifferentdeveloptental stageorat a higher concentration or as a more active enzyme. Allithese possibilities are corsistent with the finding that 3%; is a neanorph. Ofcourse, flaggmleasemayrotbeeroodedbyalmamCRI-‘inone oftheinserts. Itmightbeeroodedbyaoarpletelymvel gene. 80 nutterexperinentstodetermimwhetheraexisarepeatadelanentaro delineatethefiexmappositimwithinthermwillberoededto elucidatetheexactnature ofmandallad ittobeclcned. sraunmammmas Bexwasdiscoveredasamatemal-effectdanirantwhidlirduced exchange between two 10's on an attached-xx dirtnosane. It exhibits two fascinatingphenotypes. (meisthatmilrhnedahighfrequencyof exdiange in a region characterized by very little exchange. The second is that Beg-induced excl-large was a mitotic event which took place at the earliest stage of embryogenesis. For these reasons, I undertook a variety of experiments all designed to increase our widerstanding of 33:5. The results of the studies answer several important questions about @arflpointtl'lewaytoamodelofgegactivityandtoother e3q3eriments needed to further characterize 3%. The studies also highlight the utility of 3:; as a tool for analyzing a multigene family, and suggest that it has additional significance because it is an 2(- linked heteroohranatic locus, one of mly 5 Imam x—linked heteroohraratic loci. W The major conclusions from these studies were: 1) The Rex locus is a nearorph located at a glecific position within the rum array. 2) There are multiple, daninant, maternal-effect suppressors of Ex 81 82 (531%)). At least the is a neanorm located at a specific position within the rum array, and it is a divisible element. 3)&x—ixrluoede3d1argeisca1ventimal, inthattherearenogross rearrangementsofthetargetrm. I-Ianever, sanefix—irrhnedecchange events involve deleticns at the site of ecc'hange. 4) When in males, M, a meiotic mutant that also suppresses magnification, induces Rex-like exchanges in a diranosane with two NO's. Me]. for Rex agion Mienalltheseresultsaretakentogethertheyalggestthatfix encodes an eidonuclease activity specific for rENA which makes recanbirogeiic breaks. This model accounts for all of the observations. Ex is an extreme hypermorph or a moment, producing a novel activity not found in other rum. 3s; induces both magnification and recombination, two events that require INA breaks. E111 males, known to be defective in post-replication repair, also irriuce Rex-like exchanges inchratosamswithtmNO's. Presumably, mammal act by creating or failing to repair recarbirogenic breaks in the rum. Furthermore, mappearstobepartofamllti-elenent system, located in, suppressed by and irducirg exchange in various rum arrays. Inthisrespect,fiexisreniniscertoftransposons. IfBixencodesan endonuclease likethatwhich ispartoftransposases, thenanalogous roles can be suggested for the suppressors (repressor of transposition) and the targets (site-specific target for trarsposition) . There are earplesoftrarsposmsystelswithbothtransposaseardrepressorsof transposase, nostnotably, theThStrarsposmofLQLi (Isbergetal., 1982: J'dmsm et al., 1982). 83 Exdiffersframnostothertransmsoroinseveralrespects: its target specificity, time of actiam and maternal effect. 3g acts specifically m the no, and it acts very early, before there is any reported transcriptim of rum in the zygote (lumight & Miller, 1976: Zalokar, 1976) . It is reascnable to hypothsize that aloe rm transcriptimbegirs, thetargetmhbeoanes inaccessibletothegx erionuclease, arasotheonlytimeoractimisinthefirst few divisions in the erbryo. The maternal effect is obviously due to same maternal contribution. Considering the enormous transcription rate of rum inthe ovary (than & Ritossa, 1970; Mermod et al., 1977) it is possible that a Re}; gene product encoded within the rum might aoamulatetoahighcoroentrationintheeggarodiminishwitheadi embryonic division. Thereisaparallel framtheliterature forthismodelofljex activity. The insertion sequences found within same 288 coding regions belong to a class of elements knam as "ron-L'I‘R retrotransposons" or "retroposons" (Rogers, 1985; xiong & Eickbush, 1988a). These elements appeartotransposeviaanmintermodiate, likeretroviruses, butlack the characteristic lamg terminal repeats (mR's) of retroviruses. They are characterized by stretches of polyA at one end, and often have duplicatiors of the flaming INA at the site of insertion. Infect, tmclassesonLWrmirsertimsequeoes (Type I and Type II) are closely related to the mll-daracterized R1311 and R280 sequences of m E5211 (Eickbush and Robins, 1985; Eickbush, personaloamtmicatim). BotthBmandRZBmhavecpenreading frames whidmoamtainseqeroessimilartoportiotooftheretroviralmlgexes encoding reverse transcriptase (mrke et al., 1987; Xiong & Eicktush, 84 1988a). Althalghtmcm-‘lacksapranoter, ithasbeensuggestedthat it is trarscribed from an rm promoter (Xian; 8 Eickbush, 1988a) based mthefilrlingsofseveralgramsthatthereareraretrarscriptsof heart-containing rm repeats in 12, W (Jammich & Killer, 1984; Kidd 8 Glover, 1981: long 8 David, 1979). Presumably, like other reversetranscriptasegenes, thesemf'seoodeanenzymewhidmcarries alt both reverse transcription as well as integraticn via e'donucleolytic cleavage (Xiong a Eickbush, 1988b) . Thereisalsogmmmnggevideroethatthemmelenent functions as a transposon with unusual site-specificity. Firstly, Xiong et a1. (1988) have found several R230 insertions outside the filth, and the flanking [NA shows substantial sequence similarity to the integration site in the 288 geme. Although this finding suggests transposition, these non-r110. insertions appear to predate the separation of several different geographical _B_g m_o;i races and do not indicate recent activity. Secondly, the races have widely differing mmbersofRZBminser‘tsinthermA, suggestingthattheelementsmay move more frequently within the rum. This might, however, be due to the recombination mechanisms which maintain the integrity of the run array and not to trarsposition. Thirdly, the fact that the insertiors iractivatetherepeatintowhidmtheyareirsertedsuggeststhatthey mmldbeselectedagairsttmlessthereisamectanismsmiias transposition to maintain them. lastly, Xicng 8 Eidtbush (1988b) have deionstrated that the (RF in R230, when ligated to an irrlucible prmoterandecpressedinmgn, doeseoodeafmctialaleoamclease that cleaves 288 [NA in a site—specific manner. Presumably, this is the integrase or endonucleolytic functim of reverse transcriptase. However, 85 theauttorsraverotdemmstratadanyinyiygeidaucleaseactivityin Billed.- Inieed,theBexsystenomplexentstheobservatiasmadeinfi,_ m,wherethereisrodirectevideoetrattheeoomcleaseswhid1 arpeartobeenoodedbleBmaroRZBmelenentsareinfacttrarslatedor activejnyiyg. Ifggxactivityistheresult ofamutatial whidl activates an insertion sequence ORF, it slould alsobe possible to search for similar mutations inn, mod. mereisalsoal'otherretmposoninmmelmrm. TheG eletentisspecifictothedlraiocelter,animostcelerontsarefamd ataconserved insertion site withintheIGS (DiNoceraeta1., 1986: DiNooera, 1988). TheGelenent target sequeoeshowsahighdegreeof sequeoesimilaritytothetargetsitesonypeIardTypeIIinsertsin the 288 gene (DiNocera et al., 1986). G elements also contain anCRF which arrears, onthebasis of sequeoe similarity, toenoodeareverse transcriptase (DiNocera, 1988). Inlightcftheseresults,itisterptingtosuggestthat£gx activity results simply fram the errlonuclease encoded by a Type I, Type II or G element insert, particularly in view of its location in the rCNA andextrem site-specificity. Hwever, notall TYpe I ardType II irsertsomrerthegexphenotype,sinoestodcstratoontaintmse elererxtscanbeBex,§1_1LBeg)orroither(thishasrotbaendirectly testedinthecaseofGeletents:however,theyarefamiinvirtually every]; W stock (DiNocera et al., 1986)). Instead, Moouldbeanactivated endaiuclease-producin; gene. Ttoactivatimcaildbearesiltofadiangeinthestnlcturalgeie, makingtheemzymemoreactive,orad1argeintheprcmoterore1haroer 86 allowing ectopic or umsually high levels of expreesim. It is possible totestthishypothesisbyfurtherdxaracterizatimofmx, including claiirg of w. W gimme; 'mralghaxt the preceding disalssim there is the implicit assumption that E; activity is entirely specific to the rum and affectsroothertargets. 'misassxmptionhasrotbeenthoroughly tested. Swanson (1984, 1987) arr! this work have sham that fix can act on any chrmosame with two Q loci. Swanson (1987) has also sham that simple duplications of heth do rot constitute a gel; target. Evenmdlflmidihastypelimertimsequeoesatbotheosofthe chrarosame, althalgh it probably lacks rm repeats at the distal end, (Hilliker & Appels, 1982) is not a _R_e_x target. The question remains as to whether an exactly duplicated block of hetercdlromatin at either emd of the chrarosame, or a duplication of a different geme family can act as a a target. Iproposetmexperimentstoanswerthisquestim. ‘mefirstisto test mmgflflzgfi as a target, since it is duplicated for a portion of theXheterochranatin. Thesecondistotestaduplicationofthess ribosanal RNA genes, W (Nix, 1973). This duplication, a tandem repeat of the 100.18 (at 56?) and airman-ding material (spanning 56¢- 59C), is unstable in lag-term stocks, but canbe readily isolated as a empressorofm. Beg-irrhicededmbemtleseparated repeats will cause reappearance of M21113. 87 'meexperimentinArpemdithsdesignedtoaskabaltthesizeof the flex target, if it is, in fact, run-specific. That experiment, minganXdirmosanewitharormalmandaP-elementatthetip‘ cmtaining a single run repeat, was inoalclusive. The element itself had a badagrourrl level of instability that made it impossible to determine if M irlduced any exchange in the dlrmosane. Nevertheless, the question remains interesting. Modified versions oftheP-elementrmhrepeathavebeenreooveredbys. McKee (personal communication) , including elements with several rtNA repeats, loss of parts of the P-elenent vector, and loss of portions of the single rum repeat. It would be mrtlmmile to test these, first for stability, and then, if possible, as gag; targets. If Ex is rum-specific, these dlraroscmes offer the best chance at defining the specific & target W- W The model of flex activity presented above proposes that Rex activity results from activation or increased production of an edemleaseeoodedbyeoofthreeclassesofruminserticns. The EX variant allows either abrormally high or ectopic expressial of the site-specific eldcnuclease, generating recanbinogenic breaks in the rum during early embryogeiesis. The model predicts that we should be able tofil'ldmRNAcorrespondingtctheORFofthe insertionelement. In general, thesetranscriptsareverylavabmoanceorarepresentas portions of longer, low abundance, rm transcripts (Jamrich & Miller, 1984: Kid 8 Glover, 1981: long £1 David, 1979). The most logical stages toecamineareovariesanimfertilizedeggs, since itisloomthatfiex acts as a maternal-effect in early eibryogeiesis. 88 Iproposeelaminirgovariesaromfertilizedeggsfronmand arrtrolmothersfortrarscriptscorrespafiingtcflreCRF'softhetmee insertimclasses. Siroeallthreeirsertia'ohavebeelsequeoed (Eidctush, perscnal carnmoatim: DiNocera, 1988), all three prdaes are readily available. Presumably, if Rex encodes an endonuclease activity, the transcript will be at a detectable level. Itis importanttopoimtmttlatgexactivitymaymtresulttmn a structural gene mutation, but, instead, may result from a regulatory mutation. Therefore, a can, with only the coding region, may not revealthenatureofthegexalteration. Itisalsonotclear ifacmA would lead to cloning of Reg. Considering the high degree of sequence similarity between all members of a given insert class, it may be impossible to identify which partiailar element of a class eooded the mRNA. Nevertheless, this result will test a central prediction of the proposed model, and will allav further experiments to daracterize 3%. For example, is the transcript present in eggs of EM mothers? WillaP-element insertionofthecmAfranthismossageconferaRex rhe'otype? CanaP-element insertimofthectNAbeusedtcmanipulate thetimirqoffigactimbyuseofanirrhcibleprototer‘? This last question is particularly interesting, since it is not clear why m acts specifically durirg early etbryogenesis. Is the Rex product mly made during oogenesis? Or, alternatively, is the target ally accessible during early erbryogeiesis? 21W Itmaybepossibletoclalemfronadlmifatrarscript is detected, as described above. I am proposing an alternative scheme for 89 clmirqintheeventthatatranscriptisnotdetected,orifthed1m has hanologytomnltiple elements, andtoprovide awaytoclcneagx even if it is a regulatory mutatim. Before omsiderirg this line of research, itisimportanttobearinmindthatspecific,asqposedto rardan,portimsofamiddlerepeatedgenefamilyraveneverbeen clcned. Infact,theresultsinduapter4representthefirsttime genes have been mapped to specfic portions of a repeated gene family. Therefore, anyattenpttocloneiggwill involve adaptingexisting tectmologytoaspecialpurpose. Asszfim,thissd1emeforclorfim3ez may became a model for cloning specific portions of the rm. 'Bmeapproadmthatlsxmestistomapfigtoassmallagroupof rum repeats as possible (lo-30 repeats), subclone that portion of the NO,arrithenscreenalltheinsertion—containingrepeatsinthe subclones for alteratioxs. If an endonuclease transcript has been detected,thescreencanberarrwedtoonlyoneclassofinsertion elements. The first step of delimiting the Beg region involves converting the geneticmap of theglg-bearh'gflo intoa physical map. It will be at thispoirrtthatuewilllcownwhethermorereoanbimntsarereeded beforewecanproceed. 'l‘hemapping data have localizedigtoa specific portion of the rtNA array, part of which includes the entire regimspamodbyIGSlengthvariantité. Bowmanyrepeatsareactually inthisspan? Itispossibletoestimatetratbymeasurirgthe internsityofbarri#6,bylcwingtheoc\mtspermimrteineadxband (usingabeta—particleoamter). Assmnirgthattheweakestband rqaresentsaxespaoer,flrerelativeamamtsinotherbaroscanbead3ed uptoguantifythetotalnmberofrepeatsinthem,ardinead1ms 90 class. IftheweakestbardisactuallyZor3 repeats itwillbeclear immediately, since a viable hp locus camot have less than approximately 150 repeats. muetotalnmberofrepeatsinagivenbploars, thenmnber of repeats in each diagrostic IGS length class, and the positiuas of various breakpoints with respect to those length classes, it is possible to estimate the minimum size of the regim where E9; is localized. For example, if a particular region bounded by two exchanges curtains 10 copies eadm of 4 diagnostic IGS length classes, and the diagrostic IGS length classes are 50% of all IGS sequences, then the minimum size of theregiuais40repeatsaroitisreasuabletoassxmethattheregion is approximately 80 repeats long. The region can be also be subdivided byfurtherexchangeswith appropriatetargets inordertomapggtoas small a region as possible. 'Ihenextstepistoseparatethisregionfruntheotherrepeats. Onepossibilityistocxtthemintoafewfragmentsusinganenzyme whichcuts infrequentlywithinthem, suchasNotI (the only sitesare in'Iype II inserts). 'Ihose fragments, whidm still cuxtainmany repeats, can be separated by pulsed-field gel electrOphoresis and eluted. Subsequently, the individual sub-NO pieces can be digested with HaeIII andprobedwith IGSclonetodeterminewhideortiuioftheNOthey curtain, based on how many and which diagnostic IGS length variants they curtain. Mother possibility is to make yeast artificial chruxosunes curtainirg portiuxs of the no, and identify the origin of each cloned region based (:1 the ICE length variants. Inanycase, thegoal istogetsubclumesofthespecificregionof themthatcurtaixsm. 'Ihenall repeats frunthat sub-regimthat 91 curtain insertscanbeeuamiredforrestrictiurvariatiuswithinthe insert or within the surrouniim regius. Finally, P-element mediated insertiur of putative E2! sequeroes might allow an in M test, currently not available for n, 2921. W 'nreremltsofthesestuiiespointinseveral importantdirections. 1) 'memarpingoffigxardwhasplacedthemwithinthem, andassud), wehaveincreasedthernmmberofmappedx-linked heterochruratic loci by 66%. Previously, any three loci had been identified in the x heterochruratin: am, 91' and m. g is a heterodiruraticelementdistal tonhich suppressespartofthe lethality induced by hutozygous 399 (a 2nd chrurosome locus) in females (Pimpinelli et al., 1985; Sandler, 1970, 1977). gr, as discussed above, is involved in run compensation, am by is the urea itself. By studyingfiexaniw) inmoredetail, wemay learnmoreabout heterochruratic loci. Forexample, aretheserepeatedelements? Ithasbeensuggested that all furotiual elements of heterochr'uratin are repeated, ard that the absence of recurbinatim has evolved to avoid a high frequency of unequal crossing-over that would lead to sudden, dramatic changes in copy number (Sandler, 1975) . 'mis hypothesis seeks to explain both the large size of the heterochrumatic regiurs and the paucity of mutable ftmctiurswithintheheterodrruratin. film), andqurelyare repeatedelements, andperhapsfixisalso. Recently itwasshcwnthat anotherheterodrrumatic lows, theBsplocus ofthemtign W system (chrulosune 2), is a repeated element (Lyttle, 1989; 92 Pimpinelli a Dimitri, 1989) . Is this true of the other heterod'arunatic loci? 2) ngirdmsfrequentmitoticexdrange inaregiumwhich exhibits little meiotic marge. mm], in males has the same effect. Matarethesemtantsdoirgtodismptthestabilityofthermh? Are there special systems that normally reduce m mitotic end-range? For example, whatisthefmrctiureroodedbygmfigg), arricanSu(R_e;) sugaress the magi-41 lesim? Alternatively, could an extra cqny of mi; gt in a male suppress the 3g maternal effect? moerstarriingthemedmanismofg-iroucedexdmangemayalsoshed lightumthemechanismsofrmAexdange. Manymodelshavebeen advamedtoerqalainthecumcertedevolutimoftherfNAarraysonthex and X chrurosomes (Dover, 1982) . The two rtNA arrays are clearly not evolving independently, yet the rate of spontaneous x—x exdxange when outpared to the rate of 3'3 exchange is not sufficient to explain their similarities (Williams et al., 1989) . ‘Ihere must either be conversion events and/or multiple exchanges between the two dumosumes. Does Be; induce conversion am/or multiple exchanges? It should be possibletoexamineunselectedtargetdrrurosumsexposedtoggfor curversion events or double exchanges that alter the mmmber or position of IGS length variants. By further daracterizing the nature of Rfl exdange, mmaylearnabuztrmhexdrangeingeneral. 3) thasproven itsworthasatool formamingwithintherm. Ihaveusedittomaplociwithintl'rermforthefirsttime. 'lhe majoradvantageofusirggx-irducedeadange isthat it isanorderof magnitniemorefrequentthanqaurtanemseadmage, whichisanon— trivial consideratim when many recumbinants have to be recovered. 93 Altlnlghtheexperimentaldesignisabitumbersura, itoulldbe streamlinedtoallungxtobeusedgenerallytomapwithrspecttothe rm. 'memostdatmrtirgpartofusingfigxtomapmindlapterllms thathairpinelmhangeshadtoberecovered. 'Ihisurtailedatime- cusmnirg screen for inversiurs. The reasur for using hairpins was that thespiral exdlangeproduct, aux-fragment, wouldmakeanymaletoo hyperploid to be fertile, even if the cross was manipulated such that all su1shada1chru1osume. Usinga different target, where the distal no is closer to the tip (such as mumscS—lsgé) walld enable the product of spiral exchanges to be recoverec as fertile males. Spiral exchanges are readily detected in the first generation of progeny from E; crosses, unlike hairpins which must be detected by screening the first generatiul progeny individually. The m(l,°1)sc§l_sg§ target dirurosume is non-inverted, with a duplication of the NC at the tip. It is a product of w exchange (Swanson, 1987). Anym, frumandelrurosumecanbeusedtoreplace theproximalNO. therefore, thesuperstructureofanyNOcanbe dissected, mud‘rastheM-beariromtasbeenanalyzedinthepreviuls experiments. This kind of analysis may prove useful in studying the distribution of various heterochrumatic elements. For example, there is an ongoing debate aboutwhether insertiur sequences are (England et al., 1988: Long a David, 1979) or are not (Hawley & Tartof, 1983b) clustered in x-linked m's. Both'IYpeIard'IypeIIirsertsarepolymorfilicinsize. By starting with odom's withdistinct arrays of irserts, Rex-induced exchange vould clearly demonstrate whetherthe inserts are clustered. 94 Another example where fix exchargewill clarify the positiur of a leterodlrmaticelarentiswithcinsertiulelenents. AlthulghG elererrtsareurlyfumriintledlrurocenter,aniareoftenfumfiwithin tl'eICS,itisrotlmomhowtheyaredistributedwithrespecttotle rm (DiNoceraetal., 1986). uoeagain, bystartingwithtwom's withdistinctGelementanays,M-irducedemhangecanmaptlec elements. 4) Begisalocus withinamultigere family. Assud‘n thecluring ofgcanbeseenasamodelfortheclulingofspecificsegmentsofa multigene family. Mam: Insmmary,ghasbeenmappedtotlerTNAandaspecificmodel for R3 action has been developed. The further study of 82;! is likely to provide information about E; itself and the M exchange mechanism, perhaps shedding light on the more general topics of teterochrumatic loci, rIZNA stability, the molecular arrargement of the rum, and, perhaps, methods of cloning specific mariners of this and other mickile repetitive gexe families. APPENDICES WA mcmwammmmm We imestigated whether a single ribosunal cistron could serve as uehalfofakgtargetbyusinganxmrulosulecontainingbotha mmalQlomsarriasinglerMcistronireerteddistaltoy-i'. The single cistron (M) had been inserted by Pelement transposition using a defective P—element vector that lacks transposase activity (Karpen et al., 1988). In order to be able to score 13;): exulanges using Lri_h_7_1, we constructed a 1rih7luo(l;1)sc‘—’l,[rih7,gi1 yi y 1 par §u_Lf)_'yi’ chrmosume. Females who were humozygcus y (with or without other markers) were then crossed to [rib7]m(1;1)scfl/X males. Regular progeny are yi' daughters and y sons. Non-disjunctional progeny are y females and xi 2 car males. Mitotic exchanges that delete the material between 111911 and the normal hp lows, would generate yi x-dlrurosume fragments. alch exdlanges occurqu before first division in the zygote would yield yi g1 yfragment sterile males. If tte exchange occurs later, gynandrurorphs would result. The initial cross with W mothers gave significant numbers of the predicted yi gri’ sterile males (Table 17), suggesting that a single cistrul is sufficient to serve as a Beg, target. however, a control experimentusirgy/ymotlers,gavetlesamefreq\emyofyi'garisuls 95 96 1751 1544 5 11 a 0.45% y e: y f m r y W - x 3460 3343 16 24 12h 0.35% y Y x e y f M W ————-— x ——-— 1493 1.395 4 9 0° 0%“ y u y f at: r aPeroentexrzhangeiscalcallatedascurd-rangemales/(regular fanales+exmargemales), simetleadergeproductcanumlybeformeiinX/Xzygotes. b5exdlargepromctStererecoveredfraaueoftheS‘Isirgle—femalematirqssmred. cuegytlaleuasreooveredthattasapatdryyimsaic. 'miselaaeptiurisrot, however, analogmstong-iniuoed svelte, sime flex-Welderlgeeventsyield gyrardrunrfilsmtmtmosaicmales. 97 (T‘able17). m,flermmosmeitselfamearstobe unstable. 'nerearethreepossiblesulrcesofthisobserved instability: the duplicated riboscmal cistron, the P-elenerrt transformatiur vector, or uremiciralmurnefllaiyfersmn-rtdumusedto culstnlctulrtargetdlruxosure. nlplicatedribosumalcistrusulttex arenotgeerallyurstable. Manysld'ldlrunosuneshavebeenusedto shriyfigx,aniefdengesareuflyfumddenmatemalgexispreselt. Inordertotostwhethertheparentmnsg‘fldrromsareisitself usablemfemaleswerecrossedwith magnso‘fl males. Noyi cari sons were observed. The data suggest, therefore, that the exchange events observed with [rug' now-.1561; are due to the P-element transformation vector. Although it has not been reported previously, a low frequency of mitotic instability may be a characteristic even of defective P—elements. We do not, however, know whether another intact P element might have been inadvertently introduced in the course of the experiment. Itisrotlomnwratedlangeeveltistakingplace, sincethey‘l' mi progeny are all sterile. Although the excharge could be a Bar-like intradlrunosumal evert, itisalsopossiblethattleexdlangeis interdlrunosumal, betmeentleX-dlrurosumemloulsardjribll. The interthrutosural exchange would, of course, have to be a male germline, possibly pre—meiotic, event. This possibility is lent sume credeoe by tiereooveryofueclusterofyi’ggriprogenyintlecmtrolcross. 98 ms mammx—xm Qeobviulsquestiulabultyggiswhetleritcanirouce interdrmrosulalaswellasintrachrurosuralexdange. BoththeXand ychrurosureshavemloci,soanexperimentwasdesigredtomeasurethe g_e_xeffectul X-Xexchange. 'meexperimenthadtotakeintoacoount thefactthathactsveryearlyinthezygotetoinduce . intrachrurosural exchange, perhaps even before pronuclear fusion. For thisreason,exdlangehadtobemeasuredbetueendlrutosuroscontributed byasinglegamete. TIeXandaXdlrmosureswerebrulghtinbyasinglegameteby usingfiflmales. InXflmalestlethdlrulosurosgeerallydisjoin frum one another (Sandler & Braver, 1954), giving rise to approximately 50%flgametes. ByusingaXchrumsurewithoremarkerattl'eendof each arm, $15—$95, x-x exchange could be assayed by looking for loss oflmcagehetweenofyiande‘i. 'nematirgsd‘rerefortleecperimerrtisslmnini‘ig. 16. The firstpartoftlesdenedevelopedastockwithahighfrequencyof Wyi’xfi suns. First, mag—W exceptiulal daughters were recover-3d from a cross with relatively high male ton—disjunction (approximately 0.8%,datanotsl'own). Thesedaughterswerematedtoygzyfgr/X males,andrterotypicallygyfgr§5daughters (whosegerotypewas 99 ' / m, g——,:' 21:, x ,__.. :'~ ac I'Yl'/ [WV'W X revues! [curler y rvfl/ \ “Erasers x Marvin") yfcvvlflex) x W 0 mice. Lunar-Irma 03) 0 1 7'" first .9 1 35.38234? y! and I" Males yl and?! females yandy‘lmalu yd’ld'nndyl males FigIEeIG-mtirgsdnletoreooverx-Xedanpproum ThefirstthreestepsgereratedastocktoprovideasouroeofK/X/yi‘XB'5 malesasdescribedinthetext. Thestockvasmaintainedasshownin step3,toersurettepreseoeofbothamaterralmarkedxulrcmosule andantmmarkedpatemalelrulosune. 'neWy-flpsmalesweremated toygyyfm/yfenales. 'n'emajorityoftleyifidaughterswill carryanintactinBSdlrurosure,butafe¢willmrrytletmnmarkers urseparatedrrumosumesasaresultofanedarge. deistirguish tlresehngerrotypicallydistimtclassemeadzyifidamteris individually mated to xsx-xLJfi/o males and the progeny were scored for segregatimoftlemarkers. Fenalescarryingedlargeummoeunes yieldedurouniqueclassesofprogemy:y1'aruybsuls. 100 xeny!eyery—+135)mselectedambadmossed- The restdtirgstodrwasumnaintaixedbyselectingfenalesardmalesof tteparertalpheetypeateadlgeeratim. Thisensmedthepresenceof tlemaricedarrimmnarkedxmrmosuresmunirgfruntleferalearrimale, respectively)araahigafrequencyofmix§sasrestutingfrom normal disjmctimintrewiyfi female. 'nerewerettmtedlnicalprcbletswiththisstock. Onewasthat may—+95 males telded to be relatively infertile, so that roughly one- third were curpletely sterile and the remainder were less fertile (27.51155 progeny/male) than their mixfi brothers (42.8:22.4 progeny/male) . The second problem was that these two classes of sums were irriistinguishable rheotypically. In order to assess the proportion of mi}? sore in the stock, individual yiBS males were ratedtoygnyr/yfexales. mealescouldthenbe distinguishedbecausetherewererulghlyequalmmmbersofyi§ardy sursamongtheir progeny. 0fthe25 males tested, 20 (80%) weretre desired min? genotype, 4 (16%) werexgyixsfi, andue (4%) was seesaw. Intresecuripartoftlematingsdlele,ggggwastestedfor irriuctiulofx-Xetdlange. mimsmalesfruntl‘estockdescribed abovewerematedtofigxfemales. Anyexdmaroeeveltsmidltookplace inuerestutirgmixfidaughtersweremdetectableinuembecause bothmarkersuiardpfi)woudrenainineveryoell,eveniftheyhad became unlinked by a mitotic eac'hange. Therefore, each genotypically 11' 35 daughter was individually mated to xfixexI-gy g/o males and the resultirgsulswerescoredforsegregatiulofyifrumfi. 101 Oflozsndaughters screened, ully3hadurdergue interdlrcmoscmalecdlarge, a0.18% frequency. Ead'loftlesixedange products (unfrumeachevelt) wastestedforthepreselceofx fertility factors and Q expressiul. The results shaved that each event was distinct: 1) #133 - misfit! it f ere-xi and W (apparently an x chrurosume fragmentwithmaterial frumfitotheceltrumere, including grit and same 1 heterochrumatin). 2)‘ #172 - 3:de f erefi and £115 (cl-bi) 3) #1425 - alarm e f exert and W (be) ‘nestructuresoftrerecurbinantsareinferred, basedultheassumption thatthere isrmAulbothamsoftlemarkedXdlrulosure (R.S. Hawley, personal cummmmicatiul) and that exchanges can take place with either rtNA array. It is not clear what event has damaged or eliminated the x§ fertility factors in 1425. Ttelwrateofexdrangesuggeststhatmdoesnotinduce interdlrurosulal exchangeatanywhererearthe frequelcywithwhich it induces intrachrurosumal exchange. nlrthermore, a different experiment indicated that in_3§ has a certain level of intrinsic irstability, regardlessofthepreserceofig. Inthisexperiment, thesamemating sdlemewasused, ecoepttlatFlprogelyweresooreddirectly forthe preselceofbothyiandps. Incrosseswithygnygg/ymotlers, 10/1204 progeny showed loss of ue or the other marker (0.8%), while in curtrol crosses with y/y mothers 3/1363 progely showed a loss (0.2%). Although the difference is significant (x2=s.ss; 0.01_<- like detachments. Hawley et a1. (1985) suggested that maid], males developrecumbinogenicbreaks intteierIAwhidlcanmtberepairedand that these breaks are what stimulate x—x exchange, allow meiotic magnification, tut inhibit pre-meiotic magnificatiul. It is possible to speculate that these same breaks induce the detachment events. That, in turn, suggests a mechanism of fig; activity. In sure way, g); may also be generating breaks specifically in the rum, leading to detadment and magnification events in tie zygote. This is oulsistelt with the model of Be); as a site-specific eldonuclease proposed in Significance & Recummendatius. 105 www.573c— M N. Win». n Helwflcfix Hflh u. N N H .4M.NWN u E any gnome means 3 sauce 89: gnome E82 an.» «0 938.3 Hooter? do n 3 o.c c o 2h 9: an E E a H 3. m ...u H 3 HA e H . Rb 8n HT§.§ mum H M ed g u m a a o.c o 3 ~.o N a 32 mane ”Tenses" g a u an a «.o o a To c a 33 men an gang a o.c o 3 e.o e s. amen came. 3.36.? S :w«. an .e, 4.4.. Ea ion—Enid E E g a; 311mm Appels, R.A. & A.J. Hilliker. 1982. The cytogenetic boundaries of the rINA regiul within heterochrumatin of the x chrurosume of We new and their relatiul to male meiotic pairing sites. Genet. Res. 32:149-156. Arri'eim, N. 1983. Concerted evolutiul of multigee families. IN: mglutig') of Genes and Proteins, M. Nei 8 R.I(. Kcehn, eds. (Sunderland, Mass., Sinauer Assoc.), mas-61. Baker, 3.8. 1975. Paternal loss (pal): a meiotic mutant in W W causing loss of paternal dlrurosumes. Geeties &:267- 296. Baker, B.S., A.T'.C. Carpenter, M.S. Esposito, R.E. Fsposito & L. Sandler. 1976. The geetic curtrol of meiosis. Ann. Rev. Genet. 19:513-134. Bender, W., P. Spierer & D.S. Hogness. 1983. alrurosural walking and jtmrping to isolate [NA from the Ag and rosy loci and the bithorax curplex in We flew. J. Mol. Biol. 1.68:17-33. Burke, W.D., C.C. Calalang & T.H. Eickbush. 1987. The site-specific ribosumal insertion element Type II of my); mori (RZBm) contains the coding sequence for a reverse transcriptase-like enzyme. Mol. Cell. Biol. 1:2221—2230. Coen, E.S., J.M. Thoday & G. Dover. 1982. Rate of turnover of structural variants in the rtNA gee family of Drosophila W. Nature _2_9§:564-568. Davis, D.G. 1969. Chrmosume behavior under the influelce of claret- norrciis:)'unctiona]= in M13 W. Geeties 6_1:577-594. DaWid, I.B., E.O. Long, P.P. Di Nocera & I‘LL. Pardue. 1981. Riboscxnal ireertion-like elements in Mil; W are interspersed with mobile sequences. Cell 25:399-408. , Dawid, I.B. & M.L. Rebbert. 1981. Nucleotide sequemces at the bumdariesbetweengeeandinsertiulregions intherCNAof my; W. Nucleic Acids Res. 2:5011-5020. Di Nooera, P.P. 1988. Close relatiuehip between non-viral retroposons in Mile; ,mel_a_n_o_gas_te;. Nucleic Acids Res. 16:4041-4052. Di Nooera, P.P., F. Graziani 8 G. Iavorgna. 1986. Gelumic and structural organization of W W G elerents. Nucleic Acids Res. 15:675-691. Dover, G. 1982. Molecllar drive: a cdlesive mode of species evolution. Nature 222:111-117. 106 107 Eickbush, T.H. 8 B. Robirs. 1985. 2. mpg 288 ribosulal genes curtain insertiulelexentssimilartotreT‘ypeIandIIelenertsof mule Helm. EMBO J. 59281-2285- Blgland, P.R., H..W Stokes 8 R. Frankham. 1988. Clustering of rINA ourtainirgTypeI ireertiul sequences intredistalnucleolus organizer of mm; W: implicatiuis for the evolutiul of X and X rmA arrays. Geet. Res. 51:209-215. Feinberg, A.P. 8 B. Vogelstein. 1983. A technique for radiolabeling [NA restrictiul eldcnuclease fragments to high specific activity. Aral. Biodlem. 122:6-13. Feinberg, A.P. 8 B. Vogelstein. 1984. Addendum to: "A technique for radiolabeling DNA restrictiul endonuclease fragments to high specific activity." Anal. Biochem. 122:266-267. Franz, G. 8 W. Kunz. 1981. Intervening sequences in ribosural RNA genes and bobbed pheotype in m1; hm 1'. Nature 292:638-640. Gatti, M., S. Pimpinelli 8 G. Santini. 1976. Characterization of Mg heterochrumatin. drrulosuma fl:351-375. Gelbart, W.M. 1974. A new mutant cultrolling mitotic chrumosure disjunction in M}; new. Geetits 16:51-63. Glatzer, K.H. 1979. Legals of transcribed rtNA repeating units in spermatocytes of mu; hyggi: only genes without an intervening sequeoe are expressed. Chrurosura 15:161-175. Glover, D.M. 1981. The rENA of Mlle; LEM- Cell 2__6:297-298. Hawley, R.S. 8 K.D. Tartof. 1983a. The effect of mei-41 on rENA redundancy in W W. Geeties 1_0_4,:63-80. Hawley, R.S. & K.D. T‘artof. 1983b. The ribosumal mm of W W is organized differently from that of mg m1. J. Dbl. Biol. 53:499-503. Hawley, 12.8. 8 KO. Tartof. 1985. A two—stage model for the control of rum magnificatiul. Geeties 192:691-700. Hawley, R.S., C.H. Mame, M.L. Cameron, R.L. Schwartz 8 A.E. Zitron. 1985. Repair-defect mutatiors inhibit rmA magnification in my; and discriminate between meiotic and premeiotic magnification. Proc. Natl. Acad. Sci. 22:8095-8099. Hilliker, A.J., R. Appels 8 A. Schalet. 1980. The geetic analysis of D_. W heterochrunatin. Cell 21: 607-619. Indik, 2.x. 8 K.D. Tartof. 1980. Long spacers amug ribosural geles of Dressing W- Nam 2.85:477-479- 108 Isberg, R.R., A.L. Lazaar 8 M. Syvanen. Regulatiur of Th5 by the right- repeat proteins: oontrol at the level of the transpositiul reactiul? Cell 20:883-892. Jamrridl, M. 8 0. L. Miller, Jr. 1984. Tie rare transcripts of interrupted rRNA gem in M1; W are processed or degraded during syntrmis. mac J. 2: 1541-1545. Johnsul, R.C., J.C.P. Yin 8 W.S. Reznikoff. 1982. Cultrol of Tris transpositiul in L col). is mediated by protein from tie right Karpen, G.H., J.B. Schaefer 8 C.D. Laird 1988. A M12 rRNA gee located in eudlrtmatin is active in trasncriptiul arri nucleolus formatiul. Gem 8 Development 2:1745-1763. Kidd, S.J. 8 D.M. Glover. 1981. W W ribosumal [NA containing Type II insertions is variably transcribed in different strains and tissues. J. Mol. Biol. 121:645-662. Lindsley, D.L. 8 £2.11. Grell. 1968. Geeties variations of my; W. mmegie Inst. of Wash., Publ. 627. Lindsley, D.L. 8 G. Zimm. 1985. The genure of W W, part 1: gem A-K. Dros. Inform. Service _6_2. Lindsley, D.L. 8 G. Zimum. 1987. The genume of Mile new, part 3: rearrangements. Dros. Inform. Service 25. long, 8.0. 8 LB. Dawid. 1979. Ebcpression of ribosutal 111A insertions in gm melanoqaster. Cell 12:1185-1196. long, 2.0. 8 LB. Dawid. 1980. Repeated gem in eukaryotes. Am. Rev. Biodlem. 4_9:727-764. Lyttle, T‘.W. 1989. The effect of rovel dlrurosume position and variable dose on the geetic behavior of the Respurder (fl) elerent of the Segregation Distorter (£9) systan of Milo. remaster- Geeties 121:751-‘763. Maniatis, T‘., E.F. Fritsd'l 8 J. Sambrook. 1982. mlecllar during, a laboratory manual (Cold Spring Harbor Laboratory) . melt, S. L. 8 0. L. Miller, Jr. 1976. Ultrastructural patterns of RNA Synthesis during early eItrvlory<=taielrl€s~it’- of mule new. cell a: 305-319. Mermod, J.-J., M. Jacobs-Lorena 8 M. Crippa. 1977. Changes in rate of Msyntlmisardribosutalgeermmbermringoogemisof Emilie rem. Dev. Biol- 22:393-402. Mohan, J. 8 F. M. Ritossa. 1970. Regulatim of ribosumal RNa synttmis and its bearingDev ur the bobbed plmeotype in Mile W. .Biol. 22: 495-512. 109 Muller, H.J. 1932. Further stuiies u: the nature and causes of gee mutatius. Proc. 6th Intl. (mg. Geetics 1:213-255. Nix, C.E. 1973. mlecular studies of the SS RNA gem of M1; Helm. lblec. gal. Genet. 129:309-318. Pellegrini, 1%., J. mug 8 N. Davidsul. 1977. Sequence arrangement of the rENA of mu; W. Cell 19:213-224. Pimpinelli, S. 8 P. Dimitri. 1989. (ytogeetic aralysis of segregation distortiul in M113 W: the cytological organization of the Respuuer (Ba) lows. Geetics 121:765-772. Pimpinelli, S., W. Sullivan, 14. Prout 8 L. Sandler. 1985. m biological flmotiors naming to the heterodlrumatin of W W. Geetim 199:701-724. Procunier, J.D. 8 X.D. T'artof. 1977. A geetic loan having m and curtigucus 9L9 functiuls that oultrol the disproportiorate replicatiul of ribosumal RNA gem in W W. Geetics 29:67-79. Ritossa, F.M. 1968. Urstable redundancy of gem for ribosuIal RNA. m. Natl. m. 861. m:509-5160 Ritossa, F.M., K.C. Atwood 8 S. Spiegelman. 1966. m the redundaroy of Moutplerentarytoamiroacidtrarsfermmitsabseoefrum the nucleolar organizer region of Ml; W. Genetics 55:663-676. Robbins, L.G. 1981. Geetically induced mitotic exchange in the heterodlrumatin of mild new. Geetim 99:443-459. Robbins, L.G. 8 E. E. Swansul.1988. R_e;(-iniuced recumbination implies bipolar organization of the ribosumal RNA gem of W W. Genetics 129: 1053-1059. Rogers, J.R. 1985. The origin and evolutiul of retrcposons. Intl. Rev. cytol. 92:187-279. Roiha, H. 8 D.M. Glover. 1981. Diplicated rmA sequencm of variable lelgthsflankingtleshorttypelireertiueintrefiNAof m1; W. Nucleic Acids Res. 9° 5521-5532. Roiha, H., J.R. Miller, L.G. Woods 8 D.M. Glover. 1981. Arrangerents and rearrargementsofseqeromflarflcirgtleuuotypmofrm insertiurs in W W. Nature 299:749-753. Sandler, L. 1970. The regulatiul of sex dlrurosure leterochrumatic activity by an autosunal gee in m1; W. Geeties fi:481-493. --. 1975. Studies am the geetic cultrol of heterochruratin in M13 W. Israel J. Medical Sci. 11:1124-1134. 110 -—-. 1977. Bvideoe for a set of closely linked autcsumal gem that interact with sex-dumcsure heterodlrumatin in M; W. Geetim 26:567-582. Sandler, L. 8 G. Braver. 1954. The meiotic loss of unpaired dlrumosurm in me new. Genetics 12:365-377- Sdlalet, A. 1972. Crossirg-over in the major heterodlruratic region of trexulrurosuieinrormalardinverted sequeem. Dros. Inform. Service 59:111-113. Sdalet, A. 8 G. Iefevre, Jr. 1976. The proximal region of the X dlrtmosure. IN: file; Genetics and giolggy of Mile, vol. 1b, M. Ashburrer 8 B. Novitski, eds. (Iondon: Academic Press), pp. 847- 902. Simeue, A., A la Volpe 8 B. Buoinelli. 1985. Nucleotide sequence of a cumplete ribosural spacer of 9,, W. Nucleic Acids Rm. 12: 1089-1101 . Swansul, E.E. 1984. Parameters of the Rex phenotype in mug W. Ph.D. thesis, Michigan State University. Swansul, B.E. 1987. The responding site of the & locus of mg; W. Geetim 119:271-276. T‘artof, K.D. 1971. Increasing the multiplicity of ribosutal RNA gem in We rem:- Science 13:294-297- --. 1973. Regulation of ribosural RNA gee multiplicity in Drosophila W. Genetim 22:57-71. T‘artof, K.D. 8 LB. David. 1976. Similaritim and differences in the structalreofggandXdlmrosurengemofmga. Nature 292:27-30. Tautz, D., 3.14. W, D.A. Welt, C. T‘autz & G.A. Dover. 1988. Curplete sequexmofflerRMgemofmgW. Mol. Biol. Biol. 5:366-376. Terracol, R. 1986. Transcriptiul of rmA irsertiuls in fiber} mutants of Mia Emmet Genet. Res- 18:167-184- Wellauer, P.I(., LB. [Bwid 8 K.D. Tartof. 1978. X art} 1 dlrurosomal ribosuial INA of M13: cuparisul of spacers and insertions. mll 15:269-278. White, 12.5. a 0.5. Hognms. 1977. a loop mapping of the 188 and 288 sequeeminttelugarrishortrepeatingtmitsofmg W m. Cell 19:177-192. 111 Williams, S.M., G.R. Furnier, E. Fuog 8 C. Strdaeck. 1987. Evolution of the ribosumal INA spacers of W W: different patterns of variaticn a: X arr! X W. Genetics 15:225-232. Williams, S.M., J.A. Kemiscn, L.G. Rdbins 8 C. Strtbeck. 1989. Reciprocal recadainatim and the evolution of the ribosumal gene family of Mg W. Gemtics 22:617-624. Williams, S.M., L.G. Rouairs, P.D. Cluster, R.W. Allard 8 C. Strobeck. submittedtoPNAS. Stiperstrucmreoftlemribosaralgene family. Xiong, Y. 8 T.H. Eickbush. 1988a. The site-specific ribosumal INA insertion element R1.Bm belongs to a class of non-long-terminal- repeat retrotransposons. Mol. C1211. Biol. 5:114-123. Xiong, Y. 8 T.H. Eickbush. 1988b. Emotional expression of a sequence- specific endonuclease encoded by the retrotransposcn R2311. Cell i5:235-246. Xicng, Y., W.D. Burke, J.L. Jakubczak 8 T.H. Eickbush. 1988. Ribosomal INAinsertionelementsRleardRZBncantrarsposeinasequence specific manner to locations outside the 288 genes. Nucleic Acids MS. 3:10561-10573. Zalokar, M. 1976. Autoradiographic study of protein and RNA formation during early development of Mg eggs. Dev. Biol. 52:425-437. Zimmer, B.A., S.L. Martin, S.M. Beverley, Y.W. Kan 8 A.C. Wilson. 1980. Rapid duplication and loss of genes coding the alpha chains of tumglobin. Proc. Natl. Acad. Sci. 12:2158-2162. "Wiilfliifiwfllflfliifliflr