Tobacco Documents Online

A unique zinc finger protein is associated preferentially with active ecdysone-responsive loci in Drosophila Sally A. Amero,t Sarah C.R. Elgin,2 and Ann L. Beyerx IDepamnent of Microbiology, University o~ Virginia School o~ Medicine, Charlottesville, Virginia 22908 USA~ 2Department of Biology, Washington University, St. Louis, Missouri 63130 USA 40000040

A unique, zinc finger p.rotein is associated preferentially w th active ecdysone-responsive loci in Drosophila Sally A. Amero,t Sarah C.R. Elgin,2 and Ann L. Beyer1 IDepanment of Microbiology, University o~ Virginis School of Medicine, Charlottesville, Virginia 22908 USA~ 2Department of Biology, Washington University, St. Louis, Missouri 63130 USA Using an immunochemical approach, we have identified a unique antigen, PEP (protein on ecdysone pul~fs), which is associated in thizd-instar larvae with the active ecdysone-regtdated loci on polytene chromosomes; PEP is not associated with most intermolt puffs and is found on some, but not all, heat shock-induced puffs. The distribution pattern changes with changing puffing patterns in the developmental program. We have screened an expression library and recovered a eDNA done encoding PEP. PEP possesses multiple potential nucleic acid- and protein- binding regions: a glyeine- and asparagine-rich amino terminus, four zinc finger motifs, two very acidic segments, two short basic stretches, and an alanine- and proline-rich catboxyl terminus. The Pep gone maps by in situ hybridization to the cytological locus 74F, adjacent to the early ecdysone-responsive region; however, the gone is not regulated by ecdysone at the level of transcription. The pattern of P~p expression through development suggests that maternal Pep gone transcripts are supplied to the embryo, and that the abundance of Pep gone transcripts decreases to a lower, fairly constant level thereafter. This unusual protein may play a role in the process of gene activation, or possibly in RNA processing, for a defined set of developmentally regulated loci. [Key Words: Drosophila~ ecdysone puffs~ locus 74F~ polytene chromosomes~ zinc fingers] Received October 3, 1990~ revised version accepted November 19, 1990. The giant polytene chromosomes of Drosophila rnelano- gaster provide a unique opportunity to assay the protein interactions at defined genetic loci, because specific pro- reins can be localized on the chromosomes through in- direct immunochemical assays {e.g., Elgin et al. 1988}. Analyses of this type have provided a general picture of transcriptional activity in putts and interbands from the distribution patterns of KNA polymerase II {Sass 1982~ Weeks et al. 19821 and topoisomcrase I (Fleischmarm et al. 1984}. The presence of heterogenous nuclear ribenu- cleoproteins, IhnRNPs} [S.A. Amero and A.L. Beyer, un- publ.} and small nuclear ribonucleoproteins [snlLNPs) [Sass and Pederson 1984; S.A. Amero and A.L. Beyer, unpubl.}, which are involved in packaging and processing of nascent RNA molecules (Dreyhms 1986; Zieve and Sauterer 1990}, is also observed at these sites. A coordinated program of developmental gene expres- sion is visible as a reproducible, temporal pattern of puff- ing activity on the polytene chromosomes {Ashbumer 1970). Three sets of chromosomal loci are included in the program, which is initiated in response to eedyster- old hormones lHodgetts et aL 1977}. The existing inter- molt puffs {including loci 25AC, 68C, and 90BC} regress; several "early" puffs {including loci 2B, 22B, 23E~ 63I:, 74F_~, and 75B} apl~ear quickly and then regress. Subse- quently; >103 "late" puffs appear in a temporal quence {Ashbumex and Berendes 1978~ Pongs 1988}. A number ot early genes have been cloned {Burtis et al. 1990~ Segraves and Hogness 1990; Thummel et al. 1990}, and their products may participate in activation of the late genes IUrness and Thummel 1990~ for review, see Ashbumer 19901; the products o~ these late genes are thought to initiate metamorphosis. A different program of gene expression, and a diHerent puffing pattern, ensues horn heat shook or a number of other stresses IRitossa 1962~ Ashbumer and Bonnet 19791. Nine puffs, the sites d the genes encoding the heat shock proteins {Spradling et al. 1977}, are rapidly induced at 37-38°C; concomitantly, the developmental puffs regress as transcription at these loci is repressed and splicing ot pre-m_RNAs is inhibited I¥ost and Lindquist 1986}. At least one specific factor, the heat shock transcription factor, is required in addition to the basal transcriptional machinery ~or this induction pro- cess {Parker and Topol 1984~ Wu 1984]. Using monoclonal antibodies to nuclear proteins from Drosophila embryos, we have identified an unusual pro- tein, PEP (protein on ecdysone puffs), which possesses a distinctive distribution pattern on polytene chromo- somes: PEP is associated preferentially x~fth activated, hormone-responsive loci, including b~th early and late gene~ but is not ~ound on most intcrmolt puffs, which 40000041

regress in response to hormone- The pattern of PEP lo- calization reflects the progression of the developmental program. In addition, PEP is associated with some, but not all, heat shock-induced puffs. Cloning of the Pep gene product has revealed the presence of numerous pu- tative nucleic acid- and protein-binding motifs in the unique PEP sequence. In situ hybridization using the eDNA probe places the gene at locus 741:, adjacent to an early eedysone-responsive region, but Pep gene expres- sion is not regulated at the level of transcription by eedysone. Thus, PEP is likely to be an important com- ponent of selected active loci, recruited in larval stages to complement the transcriptional machinery or splicing apparatus for eedysone-mediated gene expression. Results PEP is associated preferentially with active, ecdysone-induced puffs Monoclonal antibodies to nuclear proteins trom phiIa embryos were screened for interesting distribution patterns on polytene chromosomes by indirect immune- fluorescence, and mAbs Y1D2 and Y2All were chosen for further study. These monoclonal antibodies gave an immunofluoreseenee distribution pattern covering most visible puffs but few band and interband regions, and no nueleoplasmie oz cytoplasmic debris {Fig. 1, B}. Rem- nants of the nueleolus are not stained, mAbs Y1D2 and Y2AI 1 produce identical staining patterns on the two salivary glands from the same larva {not shown}. Subse- quent experiments described below show that the two monoclonal antibodies recognize separate epitopes in the PEP antigen. Fine mapping of the distribution pattern of PEP by the immunofluorescence assay over two latval and two pre- pupal developmental stages {Table 1} reveals a tight cor- relation between visible puffing of the developmentally regulated loci and the presence of PEP. Excluding the intermolt puffs, >90% of the visible developmental puffs are stained. The few visible puffs that are not stained occur either in the very early or very late stages~ the correlation is perfect in the middle stages. Only in several eases is staining dereeted prior to visible puffing Iloci 48B, 52C, and 82F1 or following regression Ilocus 47A}. The protein is not associated with the intermolt puffs at loci 25AC, 68C, and 90BC in the early puffing stages but is associated with the intermolt puff at locus 3C [not shown}. Particularly noteworthy is the intense staining of the ecdysone-responsive puffs at loci 74F_£ and 75B at all stages examined, as these puffs are thought to undergo early activation and late regression in re- sponse to ecdysone. Locus 71DE, on the other hand, is puffed throughout these developmental stages but is as- sociated with PEP only in the middle stages. On any given daromosome, the developmental loci listed in Ta- ble 1 constitute -50% of the total staining pattern for the PEP antigen; the additional sites may represent de- velopmental loci not scored in the original list prepared b7 Ashburner and Berendes 11978]. However, some of the~c additional sites are consistendy stained, such as loci 50CD, 52F, 61~-, 70A, and 70D (not shown), and staining at these sites shows no developmental response tO ecdysone. One instructive example of the distribution pattern of PEP is presented in Figure 1, C and D, using mAb Y1D2. Most of the developmental [hormone-responsive} loci that are puffed at stage 2 are stained [loci 22B, 62E, 63F, 69A, 74EF, and 75B} while the intermolt puff at locus 68C and the puffs at loci 21C, 23E, and 71DE ate not. Some of the latter puffs do become stained at later stages, however. Two other developmental loci 163E and 66B1 that are induced later in development differ in their staining characteristics on this chromosome~ locus 66B is stained rather brightly, while locus 63E is stained at a much lower level. It is clear by comparison to the dis- ttibution pattern of RNA polymerase H on the same chromosome {Fig. 1E} that the sites associated with PEP are a small subset of the loci associated with tl.NA poly- merase and potentially engaged in transcription. In sum, the general features of the distribution pattern of PEP eorfform to the classical models for a factor involved in expression of the hormone-induced loci. However, the additional sites stained indicate that PEP most likely is involved in other activation systems in addition to that triggered by eedysone. The distribution pattern of PEP on chromosomes from heat-shocked larvae also demonstrates a selectivity for certain active loci. Of the nine major heat shock-induced sites, PEP is found at loci 87A, 87C, 93D, and 63BC but not at loci 95D, 64F, and 70A [Fig. 2B,D}. Low-level staining is observed at locus 67B (Fig. 2D), and locus 33B was not scored. The heat shock puffs that are associated with PEP do not strictly correspond to those heat shock genes that ate also induced by ecdysone, such as the small heat shock protein genes at locus 67B {Zimmer- man et al. 19831 or with heat shock genes containing introns, the hspS~ gene at locus 63BC {Yost and Lindquist 19861, and the lasr93D gene at locus 93D seek et al. 1985~ Garbe and Pardue 1986}. Thus, PEP is not required for the processes of puffing, transcription, or RNA processing per se. The protein is associated, how- ever, with a particular subset of heat shock-induced loci, as well as a group of developmentally regulated genes and, thus, overlaps two programs of gene expression. The Pep gene encodes the YID2 and Y~4.11 epitopes Western blots of proteolytie digests of chromosomal pro- teins showed that mAbs YID2 and Y2&11 recognize dif- ferent epitopes in the PEP antigen. A nuclear extract from Drosophila embryos was digested with increasing amounts of trypsin and ~raetionated by eleetrophoresis in an SDS-polyacrylamide gel, and the peptides were transferred electrophoretically to a nitrocellulose filter for Western blot analyses. Identical filters were probad ~Sth either TBS {not sho~n], mAb Y1D2 {Fig. 3A}, or mAb YZAll [Fig. 3B}. Both monoclonal antibodies rec- ognize the tl0-kD nuclear PEP antigen~ often, hut not always, mAb Y1D2, but not mAb Y~ll, recognizes sev- 40000042

Figure 1. Distribution of PEP on polytene chromosomes from control l~'vae. Poly- tent chromosomes were squashcd in an acctic acid-formaldehyde fixative solution onto a microscope slide and stained with rnAb YID2, using an indirccr immunot'lu- orescence method ()'an~ and Elgin 1986), The chromosomes wezc photographed un- der phase-contrast and fluorescence illu- minntion and mapped according to the designations of Ldcvrc [1976). (A) The phase-contrast picture of chromosomes from a puff stagc 2 larva, whcrc thc inter- molt puffs 68C and 90BC~ ( 0 ), as well as the ccdysone puffs 42A and 74EF(4), azc clearly visible. The nucleolus (N] and chromocentez (C) are also marked. (B) The fluozesccncc picture of the same chromo- somcs shown in A, stained using mAb YID2. (A) The stained devclopmcntal loci; (0] those not stained. (C) The phase-con- trast picture of chromosome arm 3L at puff stagc 2. (4) The developmental loci that ate active at this stage~ [ 0 ) the inter- molt puff 68C~ (overlapping • •) dcvclo1~- mental puffs that will become active at later devdopmental stages (66B and ID) Thc fluotcsccRcc picture of the chro- mosome arm 3L (shown in C) stained with mAb YID2. The symbol designations arc dcseribcd in the legends B and C and de- note the level of fluorescence duc to PEP- spccific staining. [/~) The fluorescence pic- ture of the chromosome arm 3L shown in C and D, restaincd with affinity-purified antibodies specific Io~ the cathox'y-temai- hal domain of the largest subunit of RNA polymerasc II (Fisher ctal. 1989). eral smaller polypeptides in the original extract, which are likcly to hc degradation products. The di~erence be- tween the two cpitopcs is evident from the rctcntion of the YIDZ cpitope on several smaller tryptic fragments, while in the same samples, the Y2A11 cpitopc is found only in the parcntal PEP antigcn. Two prominent signals (20 and 24 -153] appear in all lanes containing trypsin, even in the absence o~ nuclear cxtract~ the~a most likely arise from nonsl~ecific zecognltion d the vast excess o{ tr~-psin in these samples. The filter incubated in TBS produced no signal. The mAb Y1D2 wa~ used to isolate a 1.8- -kb Droso- phila DNA ~agmcnt from a kgtl 1 expression library rep- resenting cDNAs from 0- to 3-hr embryos [M. Phillip and D. Brutlag, pers. comm.}. The partial eDNA fragment was shown to encode both the Y1D2 and Y2.A11 epitopes from the PEP protein in the Western blot analysis shown in Figure 4. Bacterial lysates from a X lysogen strain eaz- rying the 1.8-kb eDNA fragment, both before and a~ter induction, as well as Drosophila nucleaz proteins ~nd lysate from a lysogen carrying the unrelated HP1 gene [James and Elgin 1986), were ~ctionated in an SDS- l~olyaerylamide gel {Fig. 4A), blotted to nitrozellulose, and pzohed vrith mA_b Y1D2 IFi~ 4111. The moncelonal 40000043

Table 1o BiuS/nZ ~ PP_2pro~n to ~velo.~me~tally r~gn/at~d~nffs Stage 2 Stage 6 Stage Ann Locus P B P B P 12 Stage 18 B P B 2L 9.1C ......+ 211~ - _ ± +_ _ _ _ 22A ....... 22B'* + ++ ++ ++ - - - 22C - - + + + + ± ± 22d~~* + - + + + + + ± 25AC" + - - ± + -- - ± 34A - - + + X X + + 2K 42,A. + ++ + + ++ ++ ++ ++ 46A - - + + + X X - - 46F .... -+ ± + ± 47A .... + ± - 48B -- -+ ± + - - ± ± 52A - - ± + - - X X 52C - _+ ± ± + + X X 55E X X + + + + + + + 58BC ± + + + + - - X X 60B .... X X - - 3L 62~ + + + + + + + + ± + + 63]5 ..... + ± ± 63t=* * + + + + + ± + + + + 66B + + -* +- + + + + 68C" + + ± ++ ± + - + - 69A + + X X - + + + 71DE + ± + + + X X + - 74EF'* ++ ++ ++ ++ + + ++ ++ 75B** ++ ++ ++ ++ + + ++ + 75CD - - X X X X ± - 78D - - + + + + + + + + 3K 82Y -- + + + + + ~" + + + 85D + ± ++ +/+ X X + +/+ 85F + ++ ++ ++ + + ++ ++ 90Be" ++ - - - ++ - ++ - 91D - - ± + ± ± + - 93F .... + + - - 95F - - + + X X + - 98F + - + + .... 100DE - + - - ± + + + {*) lntermolt puffs; 1"*) eady puffs; all othess are late puffs. The puffing status of the devdopmental puffs used in this analysis is presented under the eolmmas ma*ked 1'; puffs were scored as very prominent { + + ], clearly puffed { + l, slightly puffed {-+), or not puffed ( - 1. An X denotes a lmff that was not visible in the field. Two nuclei were averaged 1or stages 2 and 6. Puffing stages were detemgmed by coml~arison of the puffing patterns to the tabulations presented by Ashbumcr a~d Berendes (1978}. The binding of PEP [eolunms marked B}, or the level of fluorescence at a puff, was scored as intense { + + ), dearly above background {+b barely detectable above backgro~md I-% or n~ detectable { - }. The + / + designation in the binding cokmm for locus 85D indicates t/mr t~.o distinct fluores=maee sisals were visible at this locus. The X chromosome was not scored. antibody recognizes one major protein species of 110 -ld9, the PEP antigen, in the nueleaz extract (Fig. 4B, lane 11, and three large p ol~,Tel~tides in the PEP lysogens [Fig. 4B, lanes 3 and 5), but nothing in the HPl lysogen (Fig. 413, lane 4]. Although some low-level e.xpression of the I~-gal/PEP fiasion protein occurs in the absence of IPTG {Fig. 4B, lane 31, the induction by IPTG is apparent by comparison [Fig. 4B, lane 5}. The same nuclear and PEP lysogen extracts were probed also with mAb Y2A11 {Fig. 4C}. The YZA11 epitope is present in the 110-kD nuclear PEP antigen {Fig. 4C, lane 1) and in the largest of the induced I~-gal/PEP ~usion proteins, but not in the r~vo smaller induced polypeptides (Fig. 4C, lane 8}. Therefore, the 1.8-kb DNA sequence encodes two epitopes ~rom the PEP antigen, confirming its origin in the Pep gene. This partial eDNA ~ragme~t was used to isolate a 2.7-kb eDNA Drosophila clone ~rom a second library made to optimize synthesis of full-length copies of mRNAs {Brown and Kafatos 1988). This longer clone was used for the analyses described below. The Pep gene maps to the cytological locus 74F The chromosomal site of the Pep gene was determined by in situ hybridization to polytene chromosomes of a biotin-labeled 2.5-kb H/ndlI[ fragment from the eDNA done. Signal was detected within the proximal portion oi the puff at the cytological locus 74F {Fig. 5], adiaeent to the early eedysone-responsive region. The puffing at this locus appears to be due to the vigorous ecdysoue induetinn of the neighboring E74 early gene {Thummel ctal. 1990), rather than the Pep gene [see below). Geno- mie Southern blot analyses (not shown) suggest that there is a single copy of the Pep gene covering no more than 8 kb of the chromosome at this site. The Pep gane is not regolated by ecdysone Transcripts from the Pep geue were examined at differ- ent developmental stages by probing a Northern blot containing poly{A}+ KNA from staged organisms with the eDNA probe used ~or the in situ hybridizations above. In every developmental stage, a single signal rep- resenting a transcript of ~2.8 kb is visible {Fig. 6J~ the signal is eahaneed, however, in the lanes containing samples trorn ovaries and 0- to 3-hr embryos. The load- ing controls for this filter representing transcripts from the ribosomal protein gene rp49 (O'Connell and Rosbash 1984) indicate overloading in the lanes containing KNA fiom 4- to 12-hr embryos and third-instar larvae. From this analysis we conclude that maternal Pep gene tran- scripts are supplied to the embryo, and the level of Pep gene transcripts decreases to a relatively constant level following embryogenesis. The abundance of Pep gene transcripts does not reflect the pulses of ecdysone that occur in mid-embryogenesis {Kraminsky et al. 1980} and in each larval inst~ IHodgetts et al. 1977} but may be correlated with the presence of eedysone in ovaries and the syncTtial zTgote (Richa~ds 19811. Is an additional e~xperiment, a Northern blot filter containing KNA sam- pies ~rom isolated larval organs that had been exposed in vitro to ecdysone or to ecdysone plus cydohexbmide was 191 40000044

Figure 2. Distribution of PEP on polytene chromosomes from hcat-shocked larvae. Larvae were maintained at 38° for 40 rain to induce the heat shock response before squashing and staining of their polytene chromosomes. The chromosomes were photographed and mapped as described in the legend to Fig. L [A} The phase-contrast picture of chromosome arm 3R, where the heat shock puffs at loci 87A, 87C, 93D, and 95D are deafly visible. [BJ The fluo- rescence picture of the same chromosome shown in A, stained using mAb YID2. Puffs lacking PEP-specific staining~ symbols mark puffs displaying significant PEP-specific staining. {G} The phase-con- trast picture of chromosome arm 3L, where the beat shock puffs at loci 63BG, 64F, 67B, and 70A are deafly visible. ID} The fluorescence picture of the same chro- mosome shown in C, stained using mAb Y1D2. Symbols denote the levd of fluores- cence due to PEP-specific staining, as de- scribed in the legend to B. probed with the same Pep eDNA fragment; the treat- ments had no effect on Pep gene expression {not showa]. The PEP antigen contains numerous mot//s suggesting nucleic acid interactions For sequence analysis, the 2.7-kb eDNA was subeloned into the BlueScript phagemid {Stratagene} in two over- lapping pieces, one 2.5-kb H/ndllI 5'-terminal fi:agment and one 300-bp EcoRI 3'-terminal ~ragment. The nude- otide sequence was determined on each strand using nested deletions; aa average o~ six readings pex b~e pair position was compiled. The anaJno zcid sequmace derived ~rom the primary nucleotide sequence predicts a molec- ular mass d 77,942 da2tons for PEP {Fig. 7A), rather than the 110-t33 figure deduced ~rom its eleetrophoretie mo- bility. The discrepancy may arise as a consequence of the highly charged domains found in the sequence {see be- low} and/or PEP may be modified post-tra~slatienally. Because both ends of the done contain only closed read- ing £rames~ the done probably covers the euti~e protein- coding sequence, that is, to the am.ino terminus of the protein, but does not necessarily extend to the transcrip- tion start site {Hultmark et al. 1986}. Severn2 potential tran~latioa start sites axe found within the first 200 nu- cleotides of o~en reading f:rz_me; of these, only th~ first 40000045

A 12345 B 12345 YiD2 Y2A11 Figure 3. Epitope mapping of mAbs Y1D2 and Y2A1]. Dzoso- ph.ila nuclear extract was digested with increasing amounts of trypsin, and identical samples were fractionated by electropho- zeais m an SDS-polyacrylamide gcI~ the polypeptides were transferred to nitrocellulose and probed with either TBS {not shown], mAb Y1D2, or mAb Y2A11. (A) Western blot analysis of the polypeptidea probed with mAb YID2. (Lane I} Drosopfifla nuclear pzoteins~ {lanes 2-41 the same proteins digested for 45 rain with 10, 25, or 50 l~g of trypsin/ml; {lane 5) trypsin alone {50 ~g/ml). [') The position of the nuclear PEP antigen; (el two nonspeeific signals that may arise from the excess of trypsin in these samples. (B) A duplicate Western blot analysis using mAb YZAI i. and in some RNA-binding proteins [/~Sdller ct aL 1985; Joho et aL 1990). The PEP fingers are distinctive in possessing five amino acid residues bet~veen the his- tidine residues rather than four {Fig. 7B}, a motif found predominantly in Drosopb_ila finger proteins {Wingender 1988}; they are not of the type found in nuclear hormone receptors {Wingender 1988}. The spacing of the fingers in PEP is unusual; most repet- itive fingers in other proteins are separated by only 7-10 amino acid residues [Berg 1988; Lee et al. 1989}. The segment from residues 396-630 in the PEP pro- rein spans two exceptionally acidic regions }boxed in Fig. 7A}, which surround a fourth zinc finger of the type mentioned above {double underline in Fig. 7A}. Acidic region I 179 residues in lettgth} possesses 9.4 glutamie acid {30%1 and 11 aspartie acid {14%} resi- dues, with a net charge of -31./mmediately preced- ing acidic region 1 there is a basic sequence, Lys-Arg- Lys-Lys-Lys-Pro-Val {overlined in Fig. 7A} similar to the nuclear localization signal found in the SV40 large T antigen, Pro-Lys-Lys-Lys-Arg-Lys-Val [Ding- wall and Laskey 1986~. Throughout most of this acidic region, alanine residues appear every five to six residues. Zinc finger 4 occurs midway between the acidic regions. Acidic region 9. {90 residues in length} possesses 84 glutamie acid [38%} and 9 aspartie acid two, centered at nudeotides 218 and 297, possess flank- ing sequences that conform to the Drosophila start site consensus sequence, {AJC}AA{A/C}AUGG {Cavener 1987}. The scanning model for translation {Kozak 1989} would suggest that the first AUG in the sequence, which represents a poor, but acceptable, match to the consen- sus, is likely to be the initiator eodon for synthesis of PEP; however, the second site, in phase with the first, presents a perfect match to the consensus sequence and might be used. It is possible that multiple sites are func- tional, as has been shown for the glueocortieoid receptor {Miesfeld et al. A number of interesting motifs are discernible within the deduced sequence for PEP: 1. The amino terminus of the protein consists of a ~e- gion 166 amino acid residues in length, of which 40 residues 19-4%1 are glyeine and 33 residues 120%1 are asparagine {both underlined in Fig. 7A}. This segment is reminiscent of the earboxy-terminal glyeine-rieh regions that exist in the rat hnRN-P A1 protein {Co- bianchi et al. 1986}, the Drosophila Hrb98DE protein {Haynes et al. 1990}, and nude.olin from several spe- cies {Caizergues-l~e~rer et al. 1989], proteins all kuovm to bind to RNA. 2. A second putative nucleic-acid binding, segment, commencing with amino acid 215, contains three zinc finger motifs {double underline in ~ig. 7A} of the X~-Cy~-X~-Cys-X~o_~z-His-X~.-His ~,pe, motifs found in many eu_karyotic transcriptional activators ([ohns~n z~d MeKuight 1989; Struh11989; Berg Lg~0) Eigure 4. Western blot analysis of fusion proteins produced in Xgtll lysogens. The 1.8-kb Drosophila DNA fragment identi- fied from a hgtll ex~pzession library with mAb Y1D2 was used to produce an IPTG-inducible ~-gal/P~P fusion protein in ~. co// strain Y1089. A similar k lysogen containing the gene for a fusion protein with the unrelated Drosopb21a heterochromatin protein HPI llames and Elgin 1986} was used as a control. ~1 SDS-polyaerylamide gel electroghor~is of Drosopb.ila nuclear proteins (lane 1], molecular weight markers (lane 2), mxinduced PEP lysogen proteins {lane 8l, induced HP1 lysogen proteims (lane 4), mad induced PEP lysogen proteins {lane 5}, stained with Coomassie blue. The sizes of the molecular mass markers are giveu in kJ]odaltous (le[t), asterisks mark the position of the nuclear PEP protein, ~md dots mark the positions of induced fusion proteins. [B). Western blot analysis of the protei~ in the gel in A, probed with mAb YID2. [C} Western blot analysis of Dro~o.~tzi.la nuclear protei~ (lane 1), mole~ula~ mv~ markers {lane 2), ann pro~ei~ hem the indue_M I?~P 1)~3e~ {lane u~ing m~b ~7.A11. 40000046

Figure 5. Loealizarion o£ th~ Pep gone on l~oly- tone chromosomes. Polytene chromos0m~s from ~d-~s~ la~ae w~¢ sq~shed omo a micr~s~e s~d~ ~ ~cetic a~d ~d pro~ed • e 2.5-kb ~ ka~t kom ~e 2.7-kb eDNA done. ~e only si~l on the c~omo- somes was mapped to ~e pro~ po~on of the e~ysone-r~pansive p~ at locus 74F, by ~ference to ~e d~i~afions of Ldewe {1976}. (10%} residues, with a net charge of -~3; interest- ingly, a second stretch o£ basic residues, Arg-Lys-Lys- Arg-Lys-Val, (overlined in Fig. 7A} resides next to acidic region 2, in a position analogous to the basic sequence preceding acidic region 1. Several functions have been observed for highly acidic domains in other chromosomal proteins (garnshaw 1987}, ineluding transcriptional activation {Ptashne 1988}, recognition of histones for polyubiquitination (Sung et al. 1988}, or nueleosome assembly/disassembly (Dingwall et al. 19871. 4. The carboxy-terminal 80 amino acids of P~.P are rich in alanine {34%}, proline 124%}, and serine and thre- onine {15% eombined}. 5. The Pep eDNA done contains 377 nueleotides be- yond the first termination eodon, ending in the se- quence AAUAAA, the consensus polyadenylation signal {Proudfoot and Whitelaw 1988}. Because the eDNA library was constructed by primer extension from the poly{AI tars of mKNAs, the existence of this Pep ~ 2.8 kb rp49~ 0.6kb Figure 6. No,them blot analysis of PEP expression during de- velopment. Samples of poly[A]÷ KNA (1 ~g] from the stagez indicated {adult males and female carcasses, ovaries, 0- to 3-hr embryos, 4- to 12-hr embwos, 12- to 20-hr embryos, first-instar larvae, second-instar larvae, third-instar larvae, early pupae, and mid-stage pupae} were fracdonated electrophoretically and transfened to a nylon membrane as described by Haynes et al. (l~0}. The RNA was probed with a lal~led 2.5-kb HindllI f~ag- ment fzom Pep eDNA. The size ~£ the Pep lxan~eript wo3 de- duced b7 eomp~ason to ILNA size mar&ors. The loading con- trois for this filter represent t~anseripts from the rilmsemal pro- tein gone r/:49 tO'Connell and Rosba_~J~ 1~$4) ~d~cl indi~te ~.'edoaflin~ of t/~e -qamI~les from 4- to 12-h.r e_mh_r)-~s and third- 6o canonical sequence at the very end of the e/one, rather than 20-30 nucleotides upstream, suggests that the e/one arose by priming from an adjacent AJ T-rich sequence. Within the ~'-untranslated sequence is found a nearly perfect match to a 450-bp genomie sequence reported previously from locus 74F~ the ge- nomie sequence was reported to hybridize to a 2.8-kb eedysone-indueible transcript in KC cells {Moritz et al. 1984}. That genomie sequence contains 29 nucle- otides not found in our cDNA~ suggesting strain-spe- cific polymorphism. The 74F genomie sequence was cloned from Canton-S flies, while the eDNA library used here was prepared from the Oregon-R strain. The genomic sequence also contains an A-rich stretch downstream of the AAUAAA motif~ which may have served as the start of the primer extension reaction during construction of the eDNA library. The overlap with the 74F sequence and the identity o~ the tran- script sizes suggest that the gone fragment reported by Moritz et al. {1984} is from the gone we have iso- lated. Thus, from the prior loealizarion of the 74F sequence on the chromosomal walk from this region {Thummel et al. 1990}, we know that the Pep gone probably resides 15-20 kb in the proximal direction from the eedysone-responsive E74 gone. It is interesting to note the linear duplication of seg- ments in the carboxy-terminal half of PEP, that is, the repetition of zinc finger, nuclear localization signal, and acidic domain. A search of Genbank version 63 for proteins found that the PEP sequence is unique. Discussion The chromosomal distribution of the PEP antigen sug- gests a tight correlation between the presence of this protein and transcription at a defined subfet of active loci. The pattern changes with changes in the develop- mental program or following heat shock. The pattera is more selective than others reported at active loci [topoi- somerase I (Heisehmann et al. 1984}, RNA polymerase II (Sass 1982; Weeks et al. 1982}, ImlLtffP proteins Amero and A.L. Beyers, unpubl.}, and snlLNP proteins (Sass and Pederson, 1984; S.A. Amero et al., in prep.)]~ as staining does not appear at most intermolt puffs or in- terbands, or at certain heat shock-induced puffs. The strong correlation between PEP and active hormonalIy re~mlated loci argues for a role for the protein in the proce~ of eedysone-mediated tmnserip~ional aetivatiom G £~.'T~S & 40000047

GGTTCGGATCGGAC~:GTAATA'i-I'GC~ATAGGCGGAGTGTTTAGACA.AGG~GGCCATCTTTCT~ ~G ~A~C~A~~ 101 TrGACA,AA.TCGCC GTTC, C/~CACCTTG'I'TTAAATAATTTCCTC GCI-I'AAACACAAACCCTACA'rTACACATCCCAT~CA~CC~~ 14 V 8 V K V N G N P ¢~ N R L V g N A K V N G N N A F R G N 203 A.A.ACCCAT~TTATGGTCAC-CGTAAAGGTCAACGGAAATCC GCAGAATC G'rr~AGTGA~TA~C GCAAA G ~CG~T~ ~C 305 CAGAAC C GCAA.CC GCAATTTTGGAGG~GGCAACAACAACTATGGTGGACCCATGGGCGC CAAC CGCATGGGCG~AT~T~C ~ C~G ~TC~G .... N P G G GQF~N N HR~ GGGQN N A~A "r N LAN NL I.. N _N L F 407 AATCCC GGC a GC GGACA GI"~TGGTAACAACATGCGAC A GGGTGGC GGTCAGATGAAC C~CCAG~T~C~~TCT~T~C~TCT~C R N ~, _N P P S L L D L P R G G G G N G N R N Q R G G P H V S R ~ G G 509 AGGAA~CAGAATCCACCATCGCTTCTCGACTTGCCCCGCGGC~GCGGTGGCATGGGAAATCGCAACCAA~GT~G~CC~TG~C~G~G~G~- -- X G N R L N N R R G Q G G G F Q N R G A T _G S G P K P P P K Q G G G 611 GC CGGCAATCGTCTCAATAACC GTC GTGGCCAGG GAGGT GGCTTC CAAAATCGTGGC GCCACTG GTA~G ~CC~CCCC~C~GG~G~ G~ G Z R K (~ N A F D R A K K L L A K N A N H K K K E P T P G E K K 1" E "713 GGTATTCGCAA~TG~TTTTGA.TC GTGCCAAGAAA~GGCTAAAAATGCCAACCATAA/u~AGAAGGAACCCAC TCCTG~ ~~TC ~G S P T K ¢" S P Y A S V P ~ D H F Y C H L C K K H N ~// D A N S F E N H 815 AGCCCTACCAAG GAGTCTCCATAC GCTAGTGTGCC GAAC GACATGTTCTACTGTCATCTGT~,CAAGAAGCACATGTGGC~T~CTC~C ~C~ Z K G R T H L H H R E G "r E E S Y R L K A N 14 -r R Q E A K "r A [ q L 917 ATCAAGGGCC GCACCCATCTGATGATGCGTGAG GGCATTGAGGA GAGCTATC GCCTCAAGGCCAACAT GATCC GT~G ~G~T~ ~TC~A~TC K S Z E F D R L K R 14 G K S K Q R Q L D Y C T N C D L N F H G ~ I" S 1019 AAGTC GATC GAGTTTGACC GC'i-rGAA GCGCATGGGCAAAAGCAAGCA GCGTCAGCT GGAC TACTIC ~TGT ~CCT~C~C~TG ~CA~TCTC G T PI R K s £ G H L Q I. K K F L H P K C 1" E C N K £ F A T R 1" O Y D T 1121 ~CCCATCGCAAGT C GGAGGGACA'tTTGCAGCTGAAGAAGTTCTTGCACCCCAAGTGCA~~C~G~CTC~A~CTAC ~TACT ,,H, L L S A E H L K K A A I~ N N T K V G E R K R ¢~ T L P Z S T E E £ E 1223 CATCTGCJI'GTC~CCGAGCATCTGAAGAAGCCTGCCGAGAACAACACCAAGGTGGGTGA~ ~ ~CA~C~T~ACC~G~G T R D L R L P Q K R K K K P V K Kit G E A A D G £ A K K E G A G D G | ]325 ACCCGCGATTTC`CGCCTGCCCCAGAAGCG`CAAGAAGAAC`CCGGTCAAGAAGGAGGGCGAA~A~C~TG~G~T~G~GGGT~CG~TG~~ G A 11 E E E V A L P V D P E D C 1" L D F N D G D E 1" P S E V DIT R L P K 1529 ~GAAGAGGAAGTGGCA~TGCCCGTGGA~CGAGGACTGCATC~TTGAC-~-rCAA~GACGGCC`~TGAGATCcC~G~G~ACCC~CTACCT~C~ N ~ q R A V G P G L Z S K L E C Y E C S V C S K F F D T £ V T A E 1~31 AACT G GCAGCGTGCTGTCG GTC CC G GTCTGATCTCCAAGCTGGAGTGCTATGAGTGCTCGGTGTGCA~C~C ~CAC C~G~CC ~C ~ H S R T A T g H R N F L K F" Z N E K S S D T K Z A Q K R A A A A L E 1733 CACTCCC GTAC GGC GACTCATCAC CGCAAC TTCTTGAAGTTTATCAACGAGAA.ATCAAGCGATAC CAAGATC ~ACA~A~T~C ~CCT G ~ G V Q T P A P A E P A P P A K T P A K T P T K A A A P A A V A G P A A 2141 GTCCAAACTCCTGCCCC GC-CTGAACCTGCAC CACCAGCCAAC_,ACCCCAGCCAAGACTCC GACCAAGGCAGCT~TC~C~A~CCC~T A A T S A D A S P S P A K K A T P A R A A A G A K A T P q R q R A R 2243 GC GGCAAC GTCG ~,~.AGAC GCCTCTCCATCTC C GGC CAAGAAGGCAAC GCCTGCTC GC GCT~C ~C ~G~C ~C~C~C ~ ~ CC~ G R Y N R Y ~ 23,!5 GGTCGCTAP_.AATC GCTACTAAGTAGACGTCGAA~ C-~..A'rTCGTAA~T~T~CC~TATA~~A~C~C~G~ 40000048

.~.ezo et a]o 2447 GG~TGAAC-~AAGAACAGAT~'r CGATCTCTT CC C GCAATTTGACATTTGCATC CC AATTTGCATATA~CT~T~T~CT~C G~ 2549 C.AACT'I-I'TAGTTGAAACAGTTGAATATATTGTACGAC GAAGCTTCTTGA~ACTGAC GA GGGAGAACA~~TC ~TCGTA~CCT 2651 ACCC GATTC CCCCCTCACAGA~CGCATACCAT GCCAGC C AGCGAACCAACCATI'~C'I-~ATGTAATCATATA~,ATTCTAT~T~C BPEP FZNGER MOT~S: FYCH LCKKHMWDANS F ENH I KGRTH Figure 7. Sequence o~ the Pep eDNA and protein. (A] The sequence of the 2.7-kb eDNA from the Pep gone was compiled using the ODBUTIL program {Staden 1980) and trans- lated using the DM program (Mount and Conrad 1986). Within the deduced protein sequence are found the tollowing motifs: a glycine (G)- and asparagine {N}-rich segment (glycines and asparagines are underlined), four zinc finger motifs (double-underlined}, two acidic regions {boxed areasJ, two nuclear localization-type signals (overlinedJ, and an alanine (A}- and proline ~P)-rich earboxyl terminus. The overlap with the 74F genomic sequence [Moil= et al. 1984) is underlined in the 3'-untranslated Pep eDNA sequence. The probe used for the in situ hybridization and Northern blot analyses is indicated between the a=owheads. IBJ Alignment of zinc finger motifs in PEP with the consensus finger sequence from Drosophila proteins IHarrison and Travers 1990}. The conserved pairs of cysteine (C) and histidine (H) residues are shaded and underlined. DYCTMGDLNFHGHIST--HRKSEGH PKCIECNKEFATRIDYDTHLLSAEH YECSVCSKFFDTEVTAEIHSRTATH DROSOPHILA CONSENSUS: L F.~..~.K.F ..... F..~ .... ~ Y The additional sites in the PEP distribution pattern sug- gest a function{s} for the protein in other activation sys- tems as well. Our observations complement the regulatory model proposed by Ashbumer and his colleagues to accommo- date the salient features of the larval/prepupal ecdysone system [Ashbumer et al. 1974~ Ashbumer 1990}. The model suggests that the genes within the early puffs are directly activated by ecdysone in conjunction with the ecdysone receptor. Activation of the early puffs does not require de novo protein synthesis but is strictly depen- dent on the presence of hormone IAshbumer 1974). Sev- eral of the early genes have been cloned and shown to respond directly to ecdysone [Burtis etal. 1990~ Segraves and Hogness 1990~ Thummel et al. 19901, as expected. Our finding of PEP on the early puffs, in the early puffing stages, introduces an additional factor to this process and suggests that PEP is recruited into early gone activation or expression. Second, in the Ashbumer model, activa- tion of the late genes is proposed to involve the products of the early genes [Ashbu.mer 1974]~ this process is de- pendent on de novo protein synthesis. This aspect of the model has been confirmed ~ecently, since the E74 pro- tein, the product of the E74 early gene, has been localized on the late puffs IUmess and Thummel 1990}. Our stud- ies add a new component to this level of eedysone-me- dinted gene expression as well, because PEP is not the product of a classic early gene but is found on the late puffs. Thus~ the evidence suggests that PEP is involved in ecdysone-mediated gene expression of both the early and late genes in the Ashburner model, and that PEP binding at these sites is regulated, perhaps indirectly, by ecdysone. The "known properties of the Pep gone and its protein product--its pattern of expression, sequence motifs, and reeruiunent to active complexes-would be expected o~ a "coactivator" transcription factor, as described by Lewin [199D), a protein that adapts the transcriptional apparatus to specific induction mechanisms, in this ease, the ecdysone induction mechanism, in this scenario, PEP may communicate between the basal transcription factors and the ecdysonc receptor, or early gene products, using zinc fingers to contact specific DNA sequences and acidic domains for protein-protein contacts. The protein may perform a similar traction in other activa- t-ion systems, possibly at the minor sites in the PEP dis- tribution pattern. Our observations to date--the tight coupling with transcription, intense staining within a puff, and certain sequence motifs~are also consistent with the possibility that PEP is a component of a ribo- nueleoprotein complex, and we note the ability of the t~anseription faetur IIIA, also a zinc finger p~otein, to contact both DNA and RNA {Miller etal. 1985}. In this context, PEP may eom_munieate between the hormon- ally activated transcription complexes, containing the eedysone receptor or the early gene products~ and the RNA packaging and processing inaehinery, containing the transcripts of the early and late genes. It will dearly be exciting to explore further the ttmetions of this unique protein. Methods and materials Generation of monoclonal antibodies Drosophila melanoga.ster {Oregon-R} strains were maintained as a laboratory population in a 25"C growth chamber, as" de- scribed by Elgin and/~liller (1977). Embryos collected at staged intervals wexe frozen and stozed at -80"C. Nuclei were isolated according to the procedures described by Niayfidd etal. {1978) including centribagadon through a sucrose cushion. Nuclear proteins from ~50 grams of embryos (6-12 hr} were extracted by homogenization in a Thomas size B homogenizer in 0.2 m~i EDTA IpH 8.0), and preboiled RNase A was added to a tlnal concentration of 1 l.tg/ml. The suspension was incubated at 37°C for30 rain; nuclei were then pelleted by eentdFagation at 16,500g in a Sorvall HB-4 rotor for 10 mln at 4"C. All buffers u~ed for homogenizing embryos, i~olating and digesting nuclei, and extracting nuelear proteins contained 5 ~tg leu1~eptin/ml~ 02 Ti units .~l~ro~snin/ml, and 0.1 m~i phenylmethylsulfcnyl flu~ri~e {P?.~SI:I [all trom Sigma] to i~i~it endogenous pro- teases. The suI:zu~atant c~n~g EDTA-~oluble hue!ear pro- 40000049

reins w~s di~ed a~'~ust 3 liters o~ 0.~ n~.l ~TA ~pH S.0), 0.1 Nu~ p~otm wac ka~onated ~ Trito~accEc ~ca-;olya~ g~s, ~d pzot~s ~ gel ~ w~¢ u~ed for i~un~on o~ BALB-c ~ as d~cfiEc~ ~y ~o ct (19~8). F~ions o~ spl~n ~]ls ~d my¢lo~ cogs, production hybfidoma c~ ~, ~d scre~g for positive hybfido~ w~c ~nducted as d~hed by I~ ~d ~# [1986], ~g ~c proc~cs o~ G~6 md ~lst~ {1982}. Epitope mapping and Western blot -na/ys~s Peptide mapp~g e~enta were candu~ted by ~ba~g 120 ~l of EDTA nu~ ~acr in ~e presence of ~creasing conc~a~o~ o~ ~s~ [Boeh~ger ~e~ Bioche~- ~J, d~ 10~ 25, or 50 ~ ~ 125 ~IT~s-HC1 [pH 6.8}, 1% SDS, and 7.5% #ycerol [~ter ~e procaine of Clevdand et ~. 1977} at 37"C for 45 m~. The ~estions were stopped by ~e ad~fion o~ SDS to a final cenc~afion of ~% ~d of ~-mereaptoe~ol to 166 ~, ~e ~pl~ were bored, ~d id~fi~ a~quots w~e loaded onto 1~% SD~polyac~l~ide gds [30% : 0.4%}, as d~ibed by Laemmli 11970}. For Western blot ~yses, proteins were transte=ed from gds to ~troeeHu- lose filters ~d probed as d~eribed by 1~ and Elg~ [1986J. Cla~fied as~tes fl~d diluted I : 20 ~ bloe~ng solution was used as ~e pm~ amibody, ~d l~I-labeled ~ti-mouse see- onda~ ~tibody [F[ab']2] ~ent {~er~ Co~.l was used at a $1ufion o~ 1 : 500 ~ bloe~ b~er. The filters were ~- posed ov~i#t to pr~l~hed Kodak X-Omat film at -80"C ~ ~ ~tensi~g screen. Screening eDNA libraries The Xgtll eDNA expression library, subeloned from the origi- nal kgtl0 library at the EcoR[ sites [M. Philip and D. Bmtlag, pets. eomm.I, was screened with mAb Y1DP., using essentially the procedure of Huynh etal. 11985}, as described by lames and Elgin [19861. However, repeated screening attempts suggested an instability of the Pep clone in this vector, so the primary semen was pex4ormed at a dilution of 9 x 10s plaques on ten 150-ram Luria broth {LBJ/ampieillin plates, using Escher~chia co//strain Y1088. Twenty-five positive plaques from the pri- mary screen were resereened by Western blot analysis of the proteins produced in a high-titer Z-ml minilysate of E. colJ strain Y1088 grown at 37°C in the presence of LPTG for 6 hr {Silhavy et al. 1984}~ three isolates produced positive signals. One phage with a 1.8-kb insert was chosen for fusther analysis. A f~-galaetosidase-PEP fusion protein was expressed in E. co~~ strain ¥1089 lysogenized with the phage insert, tollowing the proeedmes of Huynh etal. [1985]. Following temperature shock and induerion with IPTG, the bacterial cells {from 100-ml eul- turesJ were collected by eentrifugation, resuspended in 10 mr.t Tris-HC1 {pHLS), and frozen in liquid ni~ogem Foz protein analysis, the samples were fraetionated in a 7% SDS-polyacryl- amide gel for Western blot analysis. The 1.8-kb Pep insert was gel-purified by GeneClean IBIO 101l, lalx:led with [ot-s2p]dATP [Kmcrshaxa Corp.I by random priming IFeinbcrg and Vogelstein 1983}, and used to screen a second eDNA library f~om 8- to 12-ht embryos {Bro,~m and Kaf- atos 1958}. One-tenth the complexity of the library [and th~c- f~re 3 × 10" colonies} w~s screened, and seven po,itive clones were isolate& The inserts in the pNB~ vector ranged in size from 1.F to 2.7 kb~ th~ clone c~nt~inin~ the 2.7-kb inse.~, des- ignated p3221, w~s chosen f~r fi~rther analy~s. Salivary glands were disszeted/ram third instar larvae in 45% acetic a~id, and polytene e]zromos~mes were squashed onto subbed microscope slides as described b7 Fardue [1986}. Tie 2.5-kb H_fndl~ fragment item p3221 was isolated b~ GeneClean {BIO 101}, labeled with biotin-UTP (Bethesda Research Labora- toriesl by random priming IFeinherg and Vogels*.ein 19831, and hybridized as described by Paxdue 11986}. The signal was visu- alized using the DNA Deteerion System from Bethesda Re- search Laboratories. Northern blot analyMs The developmental Northern Falter reported by Haynes et aL {1990} was reprobed with the 2.5-kb Hind~I fragment from the Pep eDNA. The fragment was purified from agarose gels by CeneClean (BIO 101} and labeled with [a-zZPIdATP {Amcrsham Corp.} by random priming {Feinbcrg and Vogelstein 1983}. A total of 1 x 109 epm was included in the hybridization~ performed in the solutions recommended by Maniatis etal. {19821. The hy- bridization reaerion was allowed to incubate fur 24 hr at 65"C. The filter was washed twice in Zx SSC, 0.1% SDS~ at room temperature, and twice in 0.2xSSC~ 0.1% SDS, at 60"C. The filter was Mr-dried for 15 min, covered with plastic wrap, and exposed to Kodak X-Omat AR film at - 80°C with an intensi- fying screen. Sequence analysis The insert in p3221 was subeloned in two pieces into the Blue- Script phagemid tot sequence determination. One 2.5-kb Hin- dilI fragment covers the 5' end of the gone, extending into the pNB40 vector, and one 300-bp ~eoRI fragment overlaps ~50 bp of the HindIII fragment and extends to the 2~coRI site 20 bp into the vector. These insert fragments were gel-purified by Gone- Clean {BIO 101) and cloned into the appropriate sites of Blue- Script Plus phagemid IStratagene, La lolla, CA) in E. co//BB4 strain. Each orientation of each insert was isolated. A series of nested deletions in each strand was created using exonuclease nt and SI nuelesse {both from Boehringer Man- nheim Biochemicals} for both the p3~9.1 and kgtll plasmids. Prior to ligadon with T4 DNA ligase {Boehringer Marmheim}, the ends o~ the deletions were filled in using Klenow fragment {Bethesda Research Laboratories} and nucleoside triphosphates. VCS-M13 helper phage for making single-stranded templates, dideoxynucleoddes, primers, and Sequenase enzyme for se- quencing reaerions were all obtained from Stratagene. dATP labeled with ~sS was purchased from Amersham. Several prim- ors were synthesized by the University of Virginia Protein and Nucleic Acid Research Facility. The contiguous sequences were assembled, and consensus sequences weffe computed using the ODB UTIL program {Staden 1980}. The consensus sequence was translated into protein se- quence using the DM program IMount and Conrad 1986}. The PEP protein sequence was compared to Gcnbank using the FASTP program (Pealson and Lippman 1988}. The sequences obtained ~or the Xgtll and the p~l inserts axe identical, ex- cept for the presence of 58 bp on the amino-terminal side oi the Xgtll insert, which is missing from the p32Z1 insert. lmmnnoflnorescenee assays Immun~flu~rescenee anal}~e~ of polytene daznmosomes were c~nducted b7 the meth~:ls of Silver zml Elgin [1977}, as mcdi- GE2~ES ~- DEVF.LOP.M_.E~f 197 40000050

fled by lames and El~in [19S5). PolTte~e chromosomes were squ~h=d in a 45% ~c~d= ~d-3.7% Io~al~)'d= ~t~ ~- lutio~ ~fi~y YID~ wa~ used ~ a n~t hyhddoma ~pe~- t~t~ ~-p~fied goat ~d~odies s~fic ~or the ~rbo~'- te~d dom~ d ~e l~gest sub~t o~ ~A pd~erase H (Fisher et ~. 1989) were ~uted 1 : I~ m bloe~g buffer. ~fi- mo~e F~C-~niugated an6body ~d ~d-goat I~C- eoniugated mfibody were ho~ obt~ed horn ICN ~m~obi- do~eals and ~luted l : 100O ~d I : 3~0, respecfivdy. Yox r~- t~ ~e coverslips were ~pped off the e~omosomes v~th a razor bhde, ~d the s~d~ were washed ~ree ~m~ ~ ~S at room temperature. The c~omosomes w~e photographed ~ou~ a ~i= O~- omat fluoresc~ce ~eroscope using Kodak Tri-X film. ~ctu~ were printed ns~g ~-eon~ast filters on Kod~ Polyeon~ast pap~. Devdopmen~ stages were dete~ed by comp~son o~ ~e paling pa~ems to ~ose in Table 1 of ~hb~ ~d Ber- end~ 11978}. Acknowledgments We are truly indebted to T.C. James, Paul Adler, and Bob Kadner (or guidance and assistance, to ~ae Moon Lee and Amo Green- lea~ [or the provision o( anti-pol II antibodies, to Susan Haynes, Felix Karim, and Carl Thurnmel [or assistance with the North- em blot analyses, to Valerie Dietrich and Jack Diani for tech- nical assistance, and to Keith Yamamoto for suggestions and advice. This work was supported by National Institutes o( Health INIH] grant GM39271 to A.L.B., NIH grant GM31532 to S.C.R.E., and NIH grant 5-S07-RR05431-28, grant IN 149P from the American Cancer Society, and March of Dimes grant 1-1200 to S.A.A.A.L.B. is the recipient o( an American Cancer Society Faculty Keseareh Award. Sequence data described in this paper have been submitted to EMBL/GenBank Data Libraries under accession number X56689. The publication costs of this article were defrayed in part by payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 USC section 1734 solely to indicate this (act. Rderences Amero, S. A., T.C. James, and S.C.K. ]~lgin. 1988. Raising antis- era to proteins in gel bands. Methods Mo/. Biol. 3: 355-362. Ashburner, M. 1970. Patterns of puffing activity in the salivary gland chromosomes of Drosophila. Chromosoma 31: 856- 376. ~r. 1974. Sequential gene activation by ecdysone in poly- tene chromosomes of Drosophila melanogaster. ]I. The [eets of inhibitors of protein synthesis. Dev. Biol. 39: 141- 157. • 1990. Puffs, genes, andhorrnones revisited. Cell 61: 1-3. Ashburner, M. and H.D. Berendes. 1978. Puffing o( palytene chromosomes. In Th~ genetics and biology of Drosophila. led. M. Ashbumer and T.K.F. Wrightlz Vol. 2B, pp. 453-498 Academic Press, New York. Ashburner, M. and J-l• Bonnet. 1979. The induction of gene ae- tivit7 in Drosophila by heat shock. Cell 17: 241-254. Ashbumer, M., C. Chflaara, P. Mehzer, and G. Richards. 1974. Temporal control of putting activity in polytene chromo- somes. Cold Spring Harbor S£mp. Qnant. B~ol. 38: 655-~12. Berg~ J.M. 19B$. Prolapsed structure for the zinc-bin~hng mains ~rom transcription factor IIIA and related proteins. Froc. Nat/. Acad. So/. 85: 99-102. • 1~90. Zinc fingers and ~ther metal-binding dora.tins. ]. BIol. Chem. 265: 6513-5516. Brown, NI-L an~t 1:.C. Ka~atos. 198~. Functional cDNA libraries ~rom Drosophila embryos. I. MoI. Biol. 203: 425-437. Burr/.% K.C., C.S. Thummel, C.W. ]ones, F.D. Karim, and D.S. Hogne~s. 1990. The Drosopt~Ia 74E1: esfly puf~ contains E74, a complex eedysone-indueible gene that encodes two ets- related preteins. CelI 61: 85-99. Caizergues-Fe~er, M., P. Morinttini, C. Curie, B. Lapeyre, N. Gas, F. Amaldc, and F. Amaldi. 1989. Nucleolin horn Xeno- pns laevis: eDNA cloning and expression during develop- ment. Genes ~ Dev. 13: 324-333. Cavener, D.R. 1987. Comparisoa o( the consensus sequence flanking t~anslational start sites in Drosophila and verte- brates. Nucleic Acids Res. 15: 1353-1361. Cleveland, DAV., S.G. Fischer, M.W. Kimehner, and U.K. Laem- mli. 1977. Peptide mapping by limited proteolysis in sodium dodeeyl sulfate and analysis by gal eleetrophoresis./. Biol. Chem. 252: 1102-1t06. Cobianchi, F., D.N. SenGupta, B.A. Zmudzka, and S.H. Wilson. 1986. Structure o( rodent helix-destabilizing protein re- vealed by eDNA cloning. ~. Biol. Chem. 261: 8536-3543. Dingwall, C. and R.A. Laskey. 1986. Protein import into the cell nuelens. Annu. Rev. Ce//Biol. 2: 367-390. Dingwall, C., S.M. Dilworth, S.J. Black, S.E. Kearsey, L.S. CoW and R.A. Laskey. 1987. Nueleoplasmin eDNA sequence re- veals polygintamic acid tracts and a duster of sequences homologous to putative nuclear localization signals. EMBO /. 6: 69-74. Dreyfuss, G. 1986. Stroeture and function of nuclear and cyto- plasmic ribonueleoproteins. Annu. Rev. Ce31 Biol. 2: 459- 498. Earushaw, W.C. 1987. Anionic regions in nuclear proteins. ]. C~11 Biol. 105: 1479-14~2. Elgin, S.C.R. and D.W. Miller. 1977. Mass rearing o~ flies and mass production and hatwesting of embryos. In The genetics and biology of Drosophila {ed. M. Ashbumer and T.K.F. Wrightl, Vol. 2A, pp. 112--19,1. Academic Press, New York. Elgin, S.C.R., S.A. Amero, J.C. Eissenberg, G. l~leisehmarm, D.S. Gilmour, and T.C. James. 1988. Distribution patterns of nonhistone chromosomal proteins on polytene chromo- somes: Functional correlations. In Chromosome strucmr~ andfonction [ed. J.P. Gustafson and R. Appels}, pp. 145-156. Plenum, New York. l"einberg, A.P. and B. Vogelstein. 1988. A technique for radiola- baling DNA restriction endonuelease fragments to high spe- cific activity. Anal Biochem. 132: 6-13. ]~isher~ P.A., L, Lin~ M. McCormell, A. Greenleaf, J.-M. Lee, and D.~.. Smith. 1989. Heat shock-induced appearance of RNA polymerase lI in karyoskeletal protein-enriched Inuelear "marix"] fractions correlates with transcriptional shutdown in Drosophila melanogaster. ~. B~oI. Chem. 264: 8464-3469. Heisehmann, G., G. Pflug~elder, E.K. Steiner, K. Javahetian, G.C. Howard, J.C. Wang, and S.C.R. Elgin. 1984. Drosophila DNA topoisomerase I is associated with transcriptionally active regions of the genome. Proa. Natl. Aead. Saf. 81: 6958-6962. Galfr6, G. and C. M~ilstein. 1982. Preparation o~ monoclonal antibodies: Strategies and procedures. Methods EnzymoL 73: 3-46. Garl~e, J.D. and Mt.L Pardue- 1986. Heat shock locus 98D o[ Drosophila melanogasten A spliced RNA most strongly eonscn'ed in the intron sequence. Proc. Nat. Aca~t Sci. 83: 1812-1816. Harrison, S.D. and Ad~ Travers. 1990. The tramtra/< gene en- ec~ea a Drosophila finger protein that Interacts with r_he transcriptional regulatory re~ion and sh~vs a novel G'~J..S & DEVI~.OPSW.I~-E 4000O051

Chic ex-pr~sion pattern. E~IBO [. 9: 207-9-16. I-la)m,~, S.P,,, G. Raychaud~u~, and ~L. E~yer. 1990. Dro~o~ HI~9~D~ Io~ ~co~ {o~ prot~ iso~o~ homologous to ~c AI ~rotc~ of m~ h~ero~eneo~ nu~]e~ ~onuc]eopiot~ ~mp]~. Mo/. C~. Biol. 10: ~I~. Hodge~, ~B., B. Sage, ~d J.D. O'Co~or. 1977. Ecdyson~ ~es d~g postemb~o~c davelopment ~ Dmsop~a mel- anogas~r. Day. Biol. 60: 31~17. Hul~ark, D., R. ~z, and %V.[. Geh~g. 1986. Trash- ffoml ~d u~s~dpfi~ consul demurs ~ ~e hted l~d~ of &e h~t-shoc~ gene fisp~2. C~H 44: Hu~, T.R., R~. Yo~ ~d R.W. Da~s. 1985. Co~ct~g ~d sere~ eDNA ~br~ies ~ h~i0 ~d ~II. h Cloning led. D.M. Glove), Vol. I, pp. 49-78. I~ Press, Ox- lord. James, T.C. md S.C.R. ~. 1986. Id~t~ica~on d a no~is- tone chromosomal prote~ assodated ~ hetero~oma~ ~ DzosopfiHa malanoga~t~r and its gene. Mol. Call Bid. 6: 886~87~. ~ohnson, P.F. ~d S.L. Me~t. 1989. E~otic tio~l re~ato~ prof,s. A~u. Rev. Biocfi~. $8: 799- 8S9. Joho, K.E., M.K. Dath7, E.T. Craw[oral, ~d D.D. Brown. 1990. ~g~ prot~ s~t~a~7 s~]~ ~o TF~ ~t binds clusivdy to 5S ~A ~ X~o~u~. Call 61: 29~. Koz~, M. 1989. ~e s~ingmodd for transhfion: ~ up,re. ]. C~H Biol. 108~ 229-241. ~sky, G.P., W.C. Clark, M~ ~tdle, R.D. Gic~, B.A. Sage, J.D. O'Co~or, ~d R.B. Hodgetts. 1980. ~ducfion tran~mb]e mRNA for dopa dcc~boxylase in Drosop~a: ~ esfly response to ecdyste~ne. Pmu. Nad. Acad. ~: 417~179. ~em~i, U.K. 1970. ~eavage of strutted prot~s d~g the ass~bly of the h~d d ba~te~op~ge T4. Natur~ ~7: 68~ 685. Lee, M.S., G.P. Gippe~, K.V. Som~, D.A. Case, ~d P.E. WriSt. 1989. ~e-~ension~ solu~on s~re of a sin- ~e zinc ~ger DNA-b~ng do~. Sd~nc~ 245: 6~5~37. Ldevrc~ G. ~r. 1976. A photo~aphic represenm~on ~d ~ter- pretaffon of ~e ~l~ene ~hromosomes o~ D~osop~ila ~ozaater saliva~ gl~ds. In The geae~ ~d b~olo~ of Drosop~a [od. M. Ashbumer ~d E. Novits~J, Vo]. In, pp. ~1~6. A~demic P~s, New Yor~ Lewin, B. I9~. Co~em ~d ac~va~on o~ Pol H promot- es: A t~I d prof,-protOn interacffons. C~H 61: 1161- 1164. M~atis, T., E.F. Frits~, and ]. Sambrook, eds. 1982. Ex~a~ fion, p~i~tion, ~d ~IEsis of m~A ~om cegs. ~ Mole~ar ~o~ng: A h~ora~ man~l, pp. 187- 210. ~Id Sp~g Harbor ~borato~, ~Id Sp~g H~bor, N~w York. Ma~eld, J.E., L.~ S~m, L~. Silver, ~d S.C.R. EI~n. 1978. A prote~ rel~sed by DNAs~ I ~g~fion d Dr~op~la nuclei is prd~en6~7 ~s~iat~ wi~ ~s. C~H 14: 5~. 1986. Genetic complem~m~on of a gluc~or~id ~c~tor defi~ene7 by ~ression of cloned re~ptor ~NA. C~ 46: 38~99. ~er, J., ~D. Mc~c~, ~d ~ ~u~ 1985. Repetitive ~c- ~opu~ c~. ~.~O ]. 4: 1~9-1614. Mode, T., J~. ~m, ~d O. Pon~. 19~. Clo~z of a l~ed ~ ~ssed a~ ~ ~dystemid re~ated ~ 74EF in salivary glands of DrosotvhHa larvze. B~,IBO 3: 289-295. Mount, D.W. and 13. Cor~ad. 19~6. Improved programs for DNA and lr~utein sequence analysis. Nnclgie Adds Res. 14: 443- 454. O'Connell, P. and M. Rosbash. 1984. Sequence, structure, and codon preference of the Drosophila ribosomal protein ~19 gene. Nuclei~ Adds Res. 12: 5495-5513. Pardue, M.L 1986. In s~ta hybridization to DNA uf chromo- somes and nuclei. In Drosophila: A practical approach [ed. D.B. Roberts}, pp. 111-138. IRL Press, Oxford. Parker, C.$. and J. Topoi. 1984. A Drosophila RNA polymerase II transcription factor contains a promoter regian-speeifie DNA-btndlng activity. Ceil 37: 278-283. Pearson, W.R. and D.I. Lippman. 1988. Improved tools for bio- logical sequence comparison. Proc. Natl. Acad. 85: 2444-2448. Pongs, O. 1988. I~edystexoid-regulated gene expression in Droso- phila melanogaster. Fur. ]. Biochem. 175: 199-204. Proudfoot, N.J. and tL Whitelaw. 1988. Termination and 3'-end processing of eukaryotie RNA. In Transcription and splic- ing, led. B.D. Haines, and D.M. Glover}, pp. 97-129. IRL Press, Oxford. Ptaslme, M. 1988. How eukaryotie t~ansetiptional aetivaters work. 1Vatore 335: 683-689. Kiehards~ G. 198 I. The radioimmune assay o~ eedystetoid tittes in Drosophila melanogastar. Mol. Cell. Bndocr~oL 21: 181- 197. Ritossa, F.M. 1962. A new puffing pattern induced by temper- ature shock and DNP in Drosophila. Experientia 18: 571- 573. Ryseck, R.-P., U. Walldoff, and B. Hovemarm. 1985. Two maior RNA products are transcribed fitom heat-shock locus 93D Drosophila malanogastar. Chromosoma 9~: 17-20. Sass, H. 1982. RNA polymerase B in pol~ene chromosomes: Immunofluereseent and auteradiog~aphie anal)sis during stimulated and repressed RNA synthesis. Gell 28: 269-278. Sass, H. and T. Pede~son. 1984. Transcription-dependent local- ization of U1 and U2 small nuelea~ ribanueleoproteins at maior sites of gene activity in polytene chromosomes. [. B1ol. 180-" 911-926. Segraves, W.A. and D.S. Hogness. 1990. The 1~75 eedysone-in- dueible gene responsible fort.he 75B e~ly puff in Drosoplu'la encodes two new members of the steroid receptor super~am- ily. Genes e~ Dev. 4: 204--210." Silhavy, TJ., M.L. Berman, and L.W. Enquist. 1984. Pzeparation of 2-ml high-titer k liquid lysates. In Bxperiments with gen~ fusions, pp. 89-90, Cold Spring Harbor Laboratour, Cold Spring Harbor, New York. Sflver~ L.M. and S.C.R. Elgin. 1977. Distribution patterns three subf~aetions of Drosophila nonhistoae chromosomal proteins: Possible correlations with gene activity. 11: 971-983. Spradling, A., M.L. Pardue, and S. Penman. 1977. Messenger RNA i~ heat-shocked Drosophila cells. ]. Mol. B1ol. 109: 559-587. Stadea, R. 1989. A new computer method Ior the storage and manipulation of DNA gel reading data. Nncleiv Acids Res. 8." 3673--3694. Strtthl~ IL 1989. Helix-turn-helix, zlne-finger, and leuelne-zip- r~er motifs for eukaryot/e t~anseriptional regulatozy proteins. Tzends Biuahem. Sci. 14: 137-140. Sung, P., S. Prakash, and L. Prakash. 1988. The RAD6 protein of Saccharomyces cer~-is~ae polyubiquinates histories, and its acidic domain mecliates th2s aefiviW. Gzn~ ~ De;,. 2: 1476- 1485. 199 40000052

Thmx~mel, C.S., K.C. Bu~is, and D.S. Hotness. 1~90. Sl~atial and teml:oral l~tteras ¢f EX4 ~xanscription duxing Droso- pldla development. C¢// 61:101-111. Umess, L.D. and C.S. Thummel. 1990. Molecular interactions within the ecdyaone regulatox7 he~acby: DNA binding p~operties of the Drosoptdla ecdysone-inducible E74A pro- rein. Cell 63: 47-61. ~Veeks, ~.R., D.t~. Ceulter, and A.L. Greenleaf. 1982. Immuno- logical studies of RNA polyem~ase H using antibodies to subunits of Drosop1~la and wheat germ en~lme~ 1. Bioi. Chem. 257: 588~-5891. Wingender, E. 1988. Compilation of transcription regulating proteins. Nuc3e/c Acid,~ Re& 16: 1879-1902. Wu, C. 1984. Two protein-binding sites in chromatin impli- eared in the activation of heat-shock genes. Nature 309: 2~9-234. Yost, HJ. and S. Lindquist. 1986. RNA splicing is interrupted by heat shock and is rescued by heat shock protein synthesis. C~//45: 185-193. Zieve, G.W. and R.A. Sautezcr. 1990. Cell biology of the snRNP particles. Crit. Re~,. Bioc~em. Mol. Biol. 25: 1-46. Zimmennan, J.L., W. l~etti, and M. Meselson. 1983. Accumula- tion of a sl~eci~ic subset of D. m~lanogasterheat shock mR- NAs in normal development withou~ heat shock. 32: 1161-1170. 40000053