key: cord-0041354-agla82gk
authors: nan
title: Sequence specificity in transcription and translation
date: 2004-02-19
journal: J Cell Biochem
DOI: 10.1002/jcb.240290605
sha: 0c005359a1a75be3f2c1ec60cbcbf3c23a42a71e
doc_id: 41354
cord_uid: agla82gk

nan

The t r a n s c r i p t i o n i n i t i a t i o n r e g i o n s of t h e B a c i l l u s s u b t i l i s genes Spovs snd ctc a r e composed of overlap nq prom e r s t h a t a r e s e p a r a t e l y u t i l i z e d by t h e minor UNA polymerase holoenzyme forms E2 ' and t r a n s c r i p t i o n is induced a t t h e o n s e t of s p o r u l a t i o n , and t h i s QXA s y n t h e s i s is dependent upon t h e products of a c l a s s of developmental r e g u l a t o r y genes known as t h e @ l o c i .

have been a b l e t o e s t a b l i s h by means of d e l e t i q q a n s l y s 9 that@-dependent r e g u l a t i o n of spoVG is e x e r t e d a t or n e a r t h e sites of Eo spoVG promoters. s t r o n g l y t h e level of Spovc t r a n s c r i p t i o n b u t is n o t r e q u i r e d f o r t h e normal p a t t e r n of Spovc expression. and whose induction t a k e s p l a c e under n u t r i t i o n a l c o n d i t i o n s d i s t i n c t l y d i f f e r e n t than t h o s e t h a t induce spore formation. W e show t h a t t h e 3 of ctc t r a n s c r i p t i o n snd its r e g u l a t i o n i s e n t i r e l y determined by DVA sequences t h a t extend no f u r t h e r upstream than u o s i t i o n -48 ( t h a t i s , without dependence upon upstream AT-rich sequences). Nucleotide Sanford, Golomb and Riddle (J. Biol. Chem., 258: 12804-12809, 1983 ) have identified a gene, ama-1 IV, which appears to code for a subunit of RNA polymerase 11. This gene was initially defined by the dominant mutation m118. positioned near &y=U on linkage group x. Strains carrying ama-l(mll8) are resistant to a-amanitin and produce an altered RNA polymerase 11 enzyme. To continue the analysis of RNA polymerase in C. elegans, we have: (1) isolated 7 lethal alleles of ama-1 including one temperature-sensitive allele and one gamma-ray-induced chromosomal rear&ment, (2) to date identified six essential genes near & (defined by 15 lethal mutations) one or more of which, by analogy with yeast (Ingles g. aJ, Proc. Natl.

Acad. Sci. USA., 81: [2157] [2158] [2159] [2160] [2161] 1984) , may encode other polymerase subunits, (3) constructed a detailed genetic map of the region surrounding ama-1 and (4) used molecular clones encoding regions of the Drosophila RNA polymerase I I G subunit gene (Greenleaf, A.L., J. Biol. Chem., 258: 13403-13406, 1983 ) to isolate recombinants from a C. eleaans gene bank.

We are currently characterizing these recombinants, particularly the structure of the promoter of this house-keeping gene. Also, we are selecting genetic revertants of certain lethal & alleles in an attempt to identify suppressor mutations defining genes encoding other RNA polymerase 11 subunits, or other genes affecting transcriptional efficiency.

ELEGANS YOLK PROTEIN GENES, Tom Blumenthal, Karen Denison, Sarah K i r t l a n d , Jerome

Cane, and John S p i e t h , I n d i a n a U n i v e r s i t y , Bloomington, IN 47405.

l h e 5. e l e g a n s yolk p r o t e i n s a r e encoded by a s i x member gene family. The t r a n s c r i p t s from t h e s e genes accumulate t o very high l e v e l s i n t h e i n t e s t i n e of the a d u l t hermaphrodite, but a r e not found i n j u v e n i l e s , males, o r o t h e r hermaphrodite t i s s u e s . 6-Crystallin, the major protein of the developing chicken lens, consists of at least two polypeptides with molecular weights of 48K and 50K (present in a ratio of 3:1, respectively). There are two extremely similar 6-crystallin genes. 4.2 Kb apart, each containing 17 introns. Only 61 (the 5 ' gene) is known to be active in the lens; no cDNA has been isolated yet for 62. The 5' flanking region of both genes are rich in GC and have TATA boxes, however, only 61 has a CAAT box and an upstream core-enhancer like sequence.

Interestingly, the 5' flanking region of 61 shows several-fold greater promoter activity & vitro in a Hela cell extract and in vivo using the pSVO-CAT expression vector in transfected lens epithelia, than the corresponding flanking region of 62. The in vitro transcription experiments were performed with about 600 bp of flanking sequence from each gene, while the in vivo experiments were conducted with 365 bp of flanking sequence. The 61 promoter region was shown by S1 mapping to initiate RNA synthesis in vitro a t the same major sites as the authentic gene in the intact lens. Thus, the 5' flanking sequences of the 61 gene appear to have inherently stronger promoter activity than those of the 62 gene. Detailed analyses can now be undertaken to understand the bases for the differential regulation of the two members of this small family of specialized, lens genes.

-1. De Lange, T. et al. Nucl. Acids Res. (1984 ), 12, 3777-3790. 2. De Lange, T. et al. Nucl. Acids Res. (1384 . 12, 4431-4443. 3. Kooter, J.M. et al. Em0 J. (1984) . 3, 2387-2392. 4 . Van der Ploeg, L.H.T. et al. Cell (19841, in press. 0795 AUTOGENOUS REGULATION OF A DNA REPLICATION GENE, dnaA, IN E.COL1, Robert E . Braun and Andrew Wright, Tufts University,HealtFi m n c e s Campus, 136 Harrison Ave.,Boston, MA. 02111 dnaA is an essential gene whose product is required for the initiation of DNAreplication in E . coli K-12.

Results of various in vivo experiments have indicated that regulation of the expression of dnaA-play a central role in regulating the frequency of initiation 0 f -A replication. Using transcriptional and translational fusions o f the dnaA gene t o the gene, we have obtained in vivo evidence which indicates that the dnaA gene product regulates its T w n i n t h e s i s at the level of transcriptional initiation.

Results from a deletion analysis of the dnaA promoter/regulatory region suggest that both dnaA promoters are regulated by the dnaA gene product and that a site b e t w c t h e two promoters is responsible for the regulation.

DNase protection experiments showed that purified dnaA protein binds to a site between the two dnaA promoters. We have also shown that one of the the two dnaA promoters isregulated by DNA methylation.

We are currently investigating the dual role of the dnaA protein in its own regulation and in the initiation of DNA replication.

To this end we have genetically defined two domains in the dnaA protein:

one involved in autogenous regulation (DNA binding) , and the other involved in the initiation of DNA replication. GenBank, the national nucleotide sequence database, is a computer-based data bank of all published DNA and RNA sequences. The database is available on-line, on tape, and in hardcopy book form (Andersen et al. (1984) Nucleotide Sequences 1984, IRL Press, Oxford). As of September, 1984, the database contains close to 3.5 million nucleotides in over 4000 entries. In addition to being a convenient reference for researchers interested in individual sequences, the database has been and is being designed to anticipate strategies that scan over many entries in search of similar features. We have also developed algorithms and software to accomplish these searches, and are especially interested in sequence comparison algorithms (used for identifying local homology, consensus sequences and hairpin structures), prediction of protein coding regions, and correlation of primary sequence data with both secondary and tertiary structure and functional roles in the cell.

POLYMERASE 111, Michael Carey, Stephen P. Gerrard and Nicholas R. Cozzarelli, University of California, Berkeley, CA 94720 We studied the assembly of RNA polymerase I11 transcription complexes on the 55 RNA gene of Xenopus and the VAI gene of adenovirus. Complete transcription complexes and subassemblies were formed on these genes in vitro using transcription factor A and RNA polymerase I11 from Xenopus and factors B and C from HeLa. To demonstrate these complexes, the DNA and f a c t o m passed through a Sepharose 4B column. bound factors was assayed for transcription after addition of polymerase. factor A bound by itself and allowed factor C to bind, which in turn allowed factor B to bind. specific subassemblies formed imply that the order of addition to a 5s gene is factors A,C and B, and to the VA gene it is factors C and B. To study the stability of these assemblies we determined the effect of high ionic strength. Factors B and C were preincubated with the VA gene in the presence of high salt o r allowed to prebind under standard conditions and then raised to high salt. purification by Sepharose 4B chromatography. remain associated with the DNA even after 2.5 min washes with 0.8 M KC1. Kinetic data indicate a half life of dissociation of 5 min in 0.45 M KC1. are able to form in 0.8 M KC1 but 50% of the normal amount of factor C binds.

The excluded material containing DNA and

For the 55 gene,

The Only factor C alone bound to the VA gene but it allowed factor B to bind.

Factor binding was analyzed after Surprisingly 50% of the prebound complexes \ Ye h a v e a l s o found t h a t t h e rDNA s t a b l e t r a n s c r i p t i o n complex can be q u a n t i t a t i v e l y and s e l e c t i v e l y sedimented from i n v i t r o t r a n s c r i p t i o n r e a c t i o n s . Sequence Specificity in Transcription a n d Control

THE CHARACTERIZATION OF PROMOTERS AND TERMINATORS OF RNA TRANSCRIP-TION CARRIED BY IS30, Brian P . Dalrymple and Werner Arber, Biozentrum, University of Basel, Klingelbergstrasse 70, CH-4056 Basel, Switzerland Using plasmids carrying various fragments from IS30 (a resident mobile genetic element of E. coli K12) inserted in front of a promoterless galK gene several promoters of transcription have been identified. In a similar system insertion of fragments of IS30 between the -5 promoter and the galK gene has allowed us to identify sequences with terminator activity. The long open reading frame of IS30 is transcribed from a weak promoter (PA), the proposed -35 region of which is in the left hand terminal inverted repeat. Transcription from this promoter is not detectable immediately outside the right hand end. A second promoter (P,) is present in the right hand end of IS30, but transcription from this promoter is generally repressed. On the reverse strand a third promoter (P ) has been identified preceding a short open reading frame. A sequence whth weak termination activity follows this open reading frame. A much stronger terminator of transcription maps in the first two hundred bases of IS30 (but on the opposite strand to the large open reading frame). So far it is not clear whether this terminator is preceded by an additional active promoter.

OVERLAPPING Tngfet PROMOTERS COMPETE FOR RNA POLYMERASE, David W. Daniels and Kevin P. Bertrand, University of California, Irvine, C A 92717.

The tetracycline resistance determinant in transpason T n g consists of two genes, the tetA resistance gene and the tetR repressor gene, that are transcribed from divergent overlapping promoters. There are two tetR promoters (PR1 and PR2) with transcription startpoints 20 bp apart. The tetA promoter (PA) overlaps both PR1 and PR?, such that the PA-PR1 startpoints are 17 bp apart and the PA-PR2 startpoints are 37 bp apart. W e lsolated nine point mutations in PA, all but one of which lie outside the -35 and -10 regions of PX1 and PR2. To study fet transcription from supercoiled DNA templates, we constructed a plasmid in which the divergent fet promoters are flanked on both sides by the efficient rrnB tl transcription terminator. Transcription of this template by purified &.

RNA polymerase yields short PA, PR1 and PR2 RNAs. Comparison of supercoiled and linear templates indicates that supercoiling stimulates PA, inhibits PR2, and has little effect on PR1. With a supercoiled PA+ template, PA is substantially more active than either PR1 or PR2. With supercoiled templates carrying PA-down mutations, PA activity is reduced as expected, however PR1 activity, and to a lesser extent PR2 activity, are substantially increased. These results suggest that PA competes with both PR1 and PR2 for RNA polymerase. Development of the chicken lens is characterized by the sequential synthesis of three major classes of crystallins ( u , 6 and 6). Each crystallin mRNA displays a characteristic temporal and spatial pattern in the developing lens. 6-Crystallin is the first crystallin to appear and accumulates during lens induction whereas the 6-crystallin mRNAs accumulate rapidly near the end of the embryogenesis. The B-crystallins consist of several polypeptides. 635 (35K polypeptide) is particularly interesting because, unlike the other 5-crystallin polypeptides, it appears only in the lens fiber cells. We have isolated the gene for 835 and are characterizing its promoter region by functional analysis in vitro and in vivo. The 835 cDNA and most of the gene have been sequenced. Each exon encodes a structural motif of the protein. A 600 bp fragment flanking the 5' region of the gene shows characteristic features of eukaryotic promoters, including a high GC-rich structure. This putative crystallin promoter is recognized in vitro in a Hela cell extract. S1 nuclease mapping of the in vitro synthesized RNA shows several transcriptional initiation sites, clustered in a narrow region. Detailed characterizations of the fiber-cell specific promoter in vitro and in vivo and its comparison with other crystallin promoters are underway.

In p a r t i c u l a r , t h e l e v e l of 6-casein gene expression in l a c t a t i n g mammary t i s s u e i s increased 260-fold over the level in t h e v i r g i n gland.

Lambda clones containing varying regions of t h e 6-casein gene have been i s o l a t e d from a rat DNA l i b r a r y .

The gene i s composed of 9 exons. 8 i n t r o n s , and spans 7.2 kb of DNA.

Nearly t h e e n t i r e 6-casein gene has been sequenced. l cycle r e g u l a t e d a t t h e t r a n s c r i pt i o n a l l e v e l . The t r a n s c r i p t i o n r a t e i s low i n G 1 , i n c r e a s e s s i x f o l d a t t h e beginning of S phase, decreases almost immediately t h e r e a f t e r and remains low throughout t h e remainder of S and i n t o G 2 . This c e l l c y c l e r e g u l a t i o n seen i n t h e G1 t o S t r a n s i t i o n i s achieved by i n c r e a s i n g t h e r a t e of t r a n s c r i p t i o n from a s i n g l e promoter region. The major s i t e of t r a n s c r i p t i o n i n i t i a t i o n i s i n t h e most 3'-ward of t h r e e contiguous 48 base p a i r r e p e a t s . A secondary s t a r t s i t e i s located i n t h e same r e l a t i v e p o s i t i o n i n t h e middle r e p e a t . These r e p e a t s do not contain t h e u s u a l CAAT or TATA boxes t h a t many e u k a r y o t i c genes u t i l i z e , b u t i n s t e a d contain t h e sequence CACAAATA, which may be a combination of t h e two s i g n a l s , and t h e sequence GGGCGG, a hexanucleotide p r e s e n t i n t h e SV40 e a r l y promoter region. The keratins are a family of 20 different polyeptides (MW40-70kd) that comprise 8nm cytoskeletal filaments in virtually all epithelial cells. These proteins can be divided into two distinct classes on the basis of t h e ability of their mRNAs to hybridize with one or the other of two different cloned keratin cDNAs. Different pairs of type I and type I1 keratin mRNAs are coordinately expressed in different epithelia and a t different stages of differentiation and development, indicating the importance of both types of subunits in filament assembly. cDNA sequence analyses have revealed that different pairs of keratins have unique sequences in localized domains and these differences influence the properties of the resulting filaments. Through c D N A sequencing, we have learned that the two types of keratins share only about 25% homology. We have now shown that despite this divergence in sequence, the predicted structures of these proteins and also the structures of their genes are remarkably similar. We are now investigating the regulatory regions of these genes to begin to elucidate t h e basis for their differential expression. In this context, we are particularly interested in determining whether vitamin A plays any direct role in regulating the expression of the keratin genes. We have already demonstrated that the vitamin specifically influences the levels of certain keratin mRNAs. It has been proposed that the vitamin and its cellular receptor act in a fashion similar to that of steroid hormones. The basic nature of the sequence features that define a promoter sequence for E. coli RNA polymerase have been established by a variety of biochemical and genetic methods. developed rigorous,analytical methods for finding unknown patterns that occur imperfectly in a set of several sequences, and have used them to examine bacterial promoters. The algorithm easily discovers the "consensus" sequences for the -10 and -35 regions, which are identical to the results of previous analyses (Hawley and McClure, 1983) , but requireno prior assumptions about the patterns. "consensus" sequences we give a rigorous definition to this concept that is widely applicable. We also provide estimates for the statistical significance of cornon patterns discovered.

In addition to providing a rigorous basis for defining the known "consensus" regions we have found additional features in these promoters that may have functional significance. These added features were located on either side of the -35 region. Recent results relating DNA sequence to helix conformation suggest that the upstream pattern may have a function in the promoter. Possible roles are discussed in this light. The gram positive bacterium B. thuringiensis, var. Kurstaki produces a protein toxin which is lethal to the larvae of lepidopteran insects. The 6-endotoxin is produced during sporulation and deposed in the cell as parasporal crystalline protein inclusion.

The gene €or the toxin has been shown to be coded by a plasmid and/or by the chromosome. The plasmid-coded gene for the B.thuringiensis. var. Kurstaki toxin has been cloned in our laboratory and its DNA sequence determined. An open reading frame for a 138 000 daltons protein has been found.

We now have constructed E. coli lac z gene fusions with the 6-endotoxin promoter in an appropriate vector system and we hope to be able to transform a plasmid-cured B. thuringiensis strain as well as a wildtype-strain and we will answer questions concerning timing, usage of the two tandem promoters and hope to get insights in the in vivo regulation of the 6-endotoxin gene in B.thurinqiensis. One important condition to garantee high levels of expression of a cloned gene is to locate it under the control of efficient promoter sequences. It has been recently shown that in E.coli the maintenance of strong promoters on plasmid vectors is dependent on the presence, on the same plasmid, of efficient terminator regions downstream from the promoter. This condition seems to prevent the transcription process from interfering with the plasmid replication and consequently allows the maintenance of the plasmid in the cells. In order to investigate on the influence of different promoters and terminators on the F-lactamase expression in both 5 and B.subtilis we have constructed a shuttle vector in which the pBR322-coded ?.-lactamase is inserted in such a way to present an EcoRI site immediately preceeding the ATG of the gene and a Hind111 site immediately following its stop codon. In this way we have been able to easily modify both the promoter and the terminator strenght and to follow the effects of these modifications on the expression of the gene. A comparison of the behavior of E.coli and B.subtilis under these different conditions is reported. W e have p a r t i a l l y p u r i f i e d an RNA polymerase a c t i v i t y from yeast which synthesizes a d i s c r e t e t r a n s c r i p t from t h e upstream region o f the yeastG&lO gene. i n i t i a t e s w i t h i n a sequence, ATATAAGTA, previously described as a consensus yeast mitochondrial promoter [Osinga, DeHaan, Christianson and Tabak (1982) Nucleic Acids Res, 1101, 799343006] .

Although t h i s enzyme preparation i s from whole yeast c e l l s i t i s probable t h a t the a c t i v i t y corresponds t o t h e yeast mitochondrial RNA polymerase [Levens, Morimoto and Rabinowitz (1981) Lbc,, Lbc, and demonstrated a p r i m r y structure w i t h three intervening sequences. Consensus CT/AG dinucleotides were found a t the borders of the intervening sequences. Putative regulat o r y sequences (CAP s i t e , TATA box, CAT box. poly A s i t e ) s i m i l a r t o other eukaryotic gene sequences were also found. Transcription o f the Lb genes was measured w i t h s p e c i f i c probes These probes d i s t i n g u i s h the nRNA precursors from the d i f f e r e n t Lb genes, and i t was demonstrated t h a t the Lbc, and Lbc, genes are transcribed before the Lbcl and Lba genes, during nodule development. The Lba gene was transcribed a t the highest l e v e l followed by the Lbc,, Lbc, and Lbc, w i t h the lowest l e v e l . Induction followed exponential k i n e t i c s and a l l f o u r genes remained t r a n s c r i p t i o n a l l y active. An Agrobacterium based transformation system has been m d i f i e d f o r legumes. Tissue s p e c i f i c t r a n s c r i p t i o n , t r a n s c r i p t processing and the Lb p r o t e i n synthesis can therefore be followed a f t e r t r a n s f e r o f complete soybean Lb genes t o other legumes. Chimeric genes t h a t allow easy detection of Lb gene a c t i v i t y has been constructed. Regulatory sequences required f o r Lb gene a c t i v i t y can therefore be defined by d e l e t i o n and s i t e directed nutagenesis. C o r n e l i s s e n , J a n N. K o o t e r , T i t i a D e Lange, Andre B e r n a r d s a n d P i e t B o r s t ,

The N e t h e r l a n d s C a n c e r I n s t i t u t e , Amsterdam, The N e t h e r l a n d s t h a t VSG s w i t c h i n g may r e s u l t i n t r a n s i e n t e x p r e s s i o n of more t h a n o n e VSG g e n r b y a s i i i q l e t r y o a n o s o n i e ( B e r n a r d s , e t a l . , NAR 12.4153-70, 1 9 0 4 ) . Gen., 2 , 285-300, 1983 ) was placed adjacent to octcpine synthase d i n g sequences w i a either a short (150tp) or k n g (700tp) stretch of octcpine synthase 3' untranslated sequence. c h i m e r i c genes were introduced into tobacco a d petunia using an Aqrobacterim s t r a i n which conferred kanamycin resistance, ad whole plants were regenerated fran the transformed tissue.

Analysis of these plants permits the follaving ccnclusicns.

(1)-Expressicn of the chimaeric gene can be detected in the leaves of most but not a l l transformants. ( 2 ) Ihe level of expressim in the highly expressing transformnts is 10%-100% of t h a t of the endogemus cab gene. (3) The 5' end of transcription of the c h i m e r i c gene is the same a s for the cab gene.

More studies cn the regulaticn of these chimeric genes and cn the influence of 5' and 3'sequences on mFNA levels w i l l be presented. t i t r a t e d amount o f the avidin-VAI RNA gene, s t a b l e t r a n s c r i p t i o n complexes were f o m d and shown t o have the same t r a n s c r i p t i o n a l a c t i v i t y as e x t r a c t s programed with an a v i d i n -f r e e V A I RNA gene. r e s i n r e s u l t e d i n t h e selective, and v i r t u a l l y q u a n t i t a t i v e , r e t e n t i o n o f the V A I RNA gene t r a n s c r i p t i o n complex. A f t e r washing the r e s i n t o remove n o n -s p e c i f i c a l l y bound proteins, a t r a n s c r i p t i o n a l l y a c t i v e f r a c t i o n was e l u t e d by the a d d Hybrid genes were constructed containing t h e human 5' h a l f of t h e e, y and P-globin genes and t h e r a b b i t 3' half of P-globin gene t o allow a d i s t i n c t i o n between the endogenous and the constructed genes.

The human sequences have r e s p e c t i v e l y 0.98, 2.2 and 1.5 k i l o b a s e s of normal 5' f l a n k i n g sequences, w h i l s t the r a b b i t has 0.7 k i l o b a s e s of normal 3' f l a n k i n g sequences.

These were introduced i n t o human K562 c e l l s by calcium phosphate p r e c i p i t a t i o n , together with t h e AGPT gene which confers G418 r e s i s t a n c e t o t h e harboring c e l l s . Clones of G418 r e s i s t a n t c e l l s were grown and harvested f o r RNA and DNA preparation. probes t h e expression of both the endogenous and exogenous genes, or of more than one exogenous globin gene, can be d e t e c t e d i n t h e same S1 assay. we have found a high l e v e l of expression of the t r a n s f e c t e d y and E genes, and have shown t h a t n e i t h e r the endogenous nor the exogenous p gene i s expressed i n K562 c e l l s . Jan M.Kooter,T.de Lange, Albert W.C.A.Cornelissen, P e t e r W.Laird and P i e t B o r s t . The Netherlands Cancer Institute,Amsterdam,The Netherlands. I n trypanosomes many,if not a l 1 , m R N A s c o n t a i n t h e same sequence of 35 n t a t t h e i r S'end. This sequence i s n o t contiguously encoded with the remainder of t h e gene but encoded by a mini-exon embedded i n a 1.35 kb repeat.The 200 c o p i e s p e r nucleus of t h e s e r e p e a t s are tandemly linked I n c l u s t e r s of up t o 15 c o p i e s . W e have obtained evidence t h a t t h e 35-nt sequence and the. remainder of t h e mRNA a r e encoded by'two s e p a r a t e t r a n s c r i p t i o n u n i t s by a n a l y s i s of s t e a d y -s t a t e RIJA and analy'sis of nascent RNA,synthesized i n i s o l a t e d n u c l e i . F i r s t , t r a n s c r i p t i o n of t h e mini-exon r e p e a t s y i e l d s a 141-nt t r a n s c r i p t with t h e 35-nt Fequence a t i t s 5'end. N o run throuah t r a n s c r i p t i o n was d e t e c t e d showing t h a t t h e 141-nt t r a n s c r i p t i s n o t a processing product. Secondly, t r a n s c r i p t i o n of mini-excn genes is s e n s i t i v e t o a moderate l e v e l of d-amanitin while t r a n s c r i p t i o n of p r o t e i n coding genes is i n h i b i t e d by a lower l e v e l suggesting t h a t t h e s e genes a r e t r a n s c r i b e d by two d i f f e r e n t RNA polymerases.We a l s o found t h a t t h e t r a n s c r i p t i o n of genes coding f o r trypanosome s u r f a c e a n t i g e n s i s not i 

Transcription initiation from certain promoters is affected by DNA supercoiling by an unclear mechanism. The effects may be mediated by changes in DNA secondary structure induced by negative supercoiling. We used mung bean nuclease, a single-strand-specific endonuclease, to determine the location and nature of nucleotide sequences involved in recognizable DNA unwinding. A variety of reaction conditions were used (all at neutral pH) since temperature and ionic environment affect enzyme site specificity by altering the conformation of supercoiled DNA (NAR 2, 7071, '84) . Three mutually exclusive sets of sites were identified. At 37°C. sites map near the end of the ampicillin-resistance (Amp-R) gene transcript. An 80bp dA+dT-rich sequence is nicked at many positions but in a non-random manner, as observed with PM2 DNA (NAR 12, 7087, '84) .

at the RNA primer promoter for DNA replication, at the promoters for tetracycline-resistance (Tet-R) and Amp-R genes, and at the terminator for RNA-I. Inverted repeat sequences are cleaved in the non-base-paired loops of potential hairpin structures. At 27OC, sites map at a promoter activated by cyclic-AMP receptor protein and near the presumed end of the Tet-R gene transcript. Examination of the known sequences around the 27OC sites shows average dA+dT-content and no strong hairpin potential. Our results demonstrate that DNA unwinding detectable by mung bean nuclease occurs at promoter and terminator regions in preference to protein-coding regions in supercoiled pBR322 DNA. might affect promoter function is by altering local DNA secondary structure. Trypanosoma brucei displays a discontinuity in RNA synthesis,in which 35 nt from the 5' end of an RNA molecule of about 140 nt (mini-exon-derived precursor RNA or medRNA) end up at the 5' end of all mRNAs analyzed to date. Whether this discontinuous transcription results from reinitiation of transcription using the medRNA as a primer or from bimolecular splicing is unknown. Recent studies with isolated nuclei indicate that mini-exon genes,VSG genes and other protein-coding genes each differ inof-amanitin sensitivity of transcription,which suggests that different polymerases are responsible for their transcription. We have investigated whether mature mRNAs in T.brucei contain a 5' cap and if so,whether the cap is TJready present on the medRN9. chemical decapping and enzymatic recapping witha-P-GTP of total and poly(A) RNA followed by either dot blot hybridization or polyacrylamide gel electrophoresis,electroblotting and mini-exon hybridization showed that both mature mRNAs and medRNA are 5' capped. We conclude that the medRNA donates its 5' cap to the mRNA molecule. These experiments also confirmed the existence of a second minor medRNA of-125 nt and revealed other prominent small capped W A S which do not hybridize to the mini-exon. Laird,P.W. et al.,in prep.; De Lange,T. et al. (1984 ),NAR 12,3777:Kooter,J. et al. (1984 P r e l i m i n a r y evidence suggests t h a t t h e upstream i n v e r t e d repeat may be e s s e n t i a l f o r t r a n s c r i p t i o n a c t i v i t y .

This Comparison o f t h e t h r e e e a r l y N4 promoter regions shows extensive V i r i o n DNA, which i s l i n e a r and double-stranded, i s n o t used f o r t r a n s c r i p t i o n i n v i t r o . _ _ _ I n v i v o experiments suggest t h a t e a r l y t r a n s c r i p t i o n u t i l i z e s a supercoiled template and i s completely dependent upon t h e presence o f r. coli single-stranded b i n d i n g p r o t e i n (ssb).

We are t e s t i n g t h e a c t i v i t y of o u r mutant promoters as supercoiled templates i n a transc r i p t i o n assay w i t h p u r i f i e d ssb.

SPECIFICITIES OF THE BACTERIOPHAGE T3 AND T 7 RNA POLYMERASES. W.T. M c A l l i s t e r , N . J . Horn, J.F. Klement and Claire E. M o r r i s , UMDNJ-Rutgers Medical School, Piscataway. New J e r s e y 08854 The RNA polymerases encoded by b a c t e r i o p h a g e s T3, T 7 , and SP6 a l t h o u g h s i m i l a r i n s t r u c t u r e , e x h i b i t n e a r l y e x c l u s i v e t e m p l a t e s p e c i f i c i t i e s .

Nucleotide sequences of t e n promoters recognized by t h e T3 RNA polymerase have now been determined. Like t h e T 7 promoters, t h e T3 promoters c o n s i s t of a h i g h l y conserved 22 b a s e p a i r sequence. S i g n i f i c a n t d i f f e r e n c e s between t h e two k i n d s of promoters are l o c a l i z e d i n a t h r e e base p a i r r e g i o n from -10 t o -12.

By use of s y n t h e t i c promoter sequences, t h e r e l a t i v e importance of each of t h e s e t h r e e po- t o t r a n s c r i b e t h e $29 l a t e genes i n v i t r o . Analysis of t h e t r a n s c r i p t i o n a l products obtained using e i t h e r crude E. c o l i c e l l e x t r a c t s c o n t a i n i n g p4 o r p a r t i a l l y p u r i f i e d p4 showed t h a t t h e v i r a l DNA l a t e r e g i o n was indeed t r a n s - 

TRAYSCRIPTIONAL REGULATORY ELEMENTS OF THE CHICK.EN LYSOZYME PROt.lOTEK, Richard J .

Miksicek and Gunther Schutz, Gernlan Cancer Research Center, Heidelberg, F.R.G. Previous experiments With a lysozymelSVl0 T antigen f u s i o n gene have shown t h a t the chicker lysozynie promoter i s expressed e f f i c i e n t l y o n l y i n primary chicken oviduct cell: and t h a t sequences responsible f o r progesterone and g l u c o c o r t i c o i d stimulated expression o f t h i s ge'ie reside w i t h i n the f i r s t 208 bp upstrsam o f the major i n v i v o t r a n s c r i p t i o n s t a r t s i t e . order t o more c l o s e l y delineate the c i s -a c t i n g t r a ! s c X p S a l r e g u l a t o r y elements resconsi h l e f o r these effects, selected r e s t r i c t i o n fragments from the lysozflie promoter region The bacterium Rhizobium meliloti invades alfalfa root hairs and induces the plant to form symbiotic nodules in which nitrogen fixation occurs. The genes in the nodABC region of the R meliloti pSym megaplasmid are required for this invasion. These genes are closely linked, translated in the same direction and may be coordinately regulated. have constructed a derivative of a broad host range plasmid containing these genes in which a nodC-lac2 gene fusion has been substituted for the nodC gene. activity in this fusion is induced in the presence of a plant product. low molecular weight and is produced Sy both host and non-host plants. We are using RNA transcript mapping to determine the start site for transcription and genetic analysis to characterize the regulation of this locus. .,~ Y The HSV-1 genes form three major groups, designated as a, 8 and Y, whose synthesis is coordinately regulated and sequentially ordered in a cascade fashion. During productive infection, the a genes are expressed first and the products of a4 gene are required for the expression of 8 genes. The B polypeptides turn off the synthesis of a genes and enable the expression of Y genes. The yl gene expression is semi-dependent on DNA replication whereas the expression of Y2 genes requires sustained viral DNA replication. To determine how infected cells differentiate between 0 and y promoters, a 72 promoter was fused with the structural sequences of thymidine kinase (TK) gene.

In the viral genome, both 0-TK and y2-m chimeras require a4 gene products, but y2-TK required in addition viral DNA synthesis. In transformed cells 0and y2-TK chimeras required for induction functional a4 genes, but were not induced by a0 gene products. In transient expression systems, the chimeric y2TK gene and BTK gene were induced by co-transfection with DNA fragments carrying the a4 and a0 genes, but not the DNA fragments carrying a genes 22 or 27. These results demonstrate that in transient assay systems both y2-and 0gene expression require only a4 or a0 gene products. The requirements for induction of 8and y2 genes in transient assay system differ therefore from those observed in trans-activation of resident genes in transformed cells and those observed in productively infected cells. Nitrogen regulated ( n t r l and nitrogen fixation (nif) gene promoters are structurally similar to each other, but bear little resemblance to canonic EL -_-coli promoters. z t r promoters are normally activated by the ntrZ (also known as slnG) product, but can also be activated by the CfrC-related K, eneumoniae . ! i f & product.

In contrast, nlf promoters of KI pneumoniae such as the nitrogenase (nifH) promoter can only be zlffi-activated.

Sequence comparisons have shown that git\-regulated promoters share the consensus sequence CTGti-bbp-'I.ItiCA between -26 and -10, whereas the consensus sequence TlTTGCA was tound centered at -14 among several ntr-activated promoters. We analyzed 2 classes of mutants isolated atter sodlum bisultite site-directed mutaqenesis oi the KL pneumonjae ____ nitH promoter, which contains the sequence CTGG-4bp-CCCrGCA. Class A mutants tailed to respond to njtfi-mediated activation while class B mutants acquired the ability to be activated by the ntrc product. With class A mutants, we tound that a transition at any of 4 diiferent bases ot the consensus sequence reduced nitA-mediated regulation (underlined where transitions were tound: CIgG-r)bp-CrcjGA). With class B mutants, a change from CCClGCA to either TCClGCA, CCI'ltiCA, or TCT.lGCA was sufficient to confer ntyc-mediated transcription. 0841 REGULATION OF I a AND 11 a GLOBIN GENES OF THE GOAT. Jacqueline W. Pierce, Anil G.

Menon and Jerry B Lingrel, Department of Microbiology and Molecular Genetics, University of Cincinnati College of Medicine, Cincinnati, OH 45267. Sequences involved in reqlated expression of a globin genes have not been defined. We are studying differential expression of two functional a globin genes of the goat; 1 a and 11 a. These genes ore highly homologous (99%) throughout their coding regions, IVS and immediate 5' and 3' flanking regions extending 122 bp 5' to the initiation codon and 139 bp 3' to the stop codon (khan et al., 1982) .

Both l a and IIa contain all signals known to be important for the transcription including CTAxT ATA box, cap site and polyA addition site; however, l a protein is about 3-fold more abundant than 11, in both fetal and adult goats. Seqences upstream of CCAAT or downstream of polyA may dictate different levels of expression. Notably, the repeat CACCCTACACCCT which is important in the transcription of ablt 8 globin genes is found upstream of l a but not 1 1 .

. In order to delineate sequences which are involved in differential expression of I a over 11 a, w e have assayed the transient expression of these genes in HeLa cells. The I a and 11 a genes were transfected separately into HeLa cells and a mRNA levels were measured by SI wclease assay. As observed for human a genes, both goat a genes are expressed in the absence of cis enhancer sequences. We found that 1 a is expressed at a higher level than IIa. These results suggest that linkage of l a and IIa is not required to achieve differential expression of these genes. In oddition, sequences dictating the different levels of expression are located within 500 bp 5' and 200 bp 3' of the a structural genes. W e ore currently examining the expression of a fusion gene I all1 a in order to further localize sequences which determine the level of a globin gene expression.

C a l i f o r n i a , L o s A n g r l e s , CA 90024; *The Wistar I n s t i t u t e of Anatomy and Biology, P h i l a d e l p h i a , PA, 19104 . W e have a p p l i e d t h e t e c h n i q u e o f DNAase f o o t p r i n t i n g t o i d e n t i f y sequences of a d e n o v i r u s t y p e 2 (Adz) promoter DNA which a r e p r o t e c t e d by f a c t o r s i n a whole-cell t r a n s c r i p t i o n e x t r a c t .

Using a s o l u b l e HeLa c e l l e x t r a c t and a r e s t r i c t i o n fragment c o n t a i n i n g t h e promoters f o r b o t h t h e major l a t e t r a n s c r i p t i o n u n i t and t h e IVa2 gene, t h r e e r e g t r a n s l a t i o n t e r m i n a t i o n module ( P r e n t k i and K r i s c h , Gene 2, 303-313, 1984) w a s i n s e r t e d i n s e v e r a l l o c a t i o n s i n t h e s e pBR322::ISL plasrnids t o l o c a l i z e t h e o r i g i n s of t r a n s c r i p t i o n , and i n v i t r o a n a l y s i s w a s c a r r i e d o u t on some. The gal operon contains overlapping promoter regions (Musso e t a l . , C e l l 12, 847 (1977) ). The P2 promoter i s used i n the absence o f the c a t a b o l i t e a c t z a E r protein(CAP); i f the CAP/cAMP complex i s a c t i v e then RNA polymerase (RP) i n i t i a t e s a t P I , f i v e base p a i r s downstream from P2. We have shown t h a t t r a n s c r i p t i o n from P 1 involves the adjacent binding o f two CAP molecules (Shanblatt and Revzin, PNAS 80, 1594 (1983) ). Our current work i s aimed a t e l u c i d a t i n g the f u n c t i o n o f the second C A P . T o t h i s end we do i n v i t r o experiments w i t h p u r i f i e d p r o t e i n s and DNA fragments. We study DNA-protein binding by means o f gel e l e c t r ophoresis as w e l l as doing a v a r i e t y o f nuclease p r o t e c t i o n and t r a n s c r i p t i o n assays. Studi e s using both wild-type and mutant promoters, and w i t h truncated DNA fragments, show t h a t the second CAP molecule has dual functions: i t s t a b i l i z e s the "one CAP/one RP" intermediate a t P1, thus excluding P2 binding; i n addition, i t i s involved i n i n t e r a c t i o n s which a i d RP i n "melting i n " a t P1. The e n t i r e process i s e x q u i s i t e l y dependent on the CAMP l e v e l , a t UM concentrations l i k e l y t o be encountered i n vivo.

The l a c c o n t r o l region a l s o has overlapping promoters. Our data reveal a marked competition between the P 1 and P2 s i t e s f o r a v a i l a b l e RP molecules. We f i n d t h a t although a s i n g l e CAP/cAMP e n t i t y binds a t the l a c promoter, the r o l e o f CAP i s s i m i l a r t o t h a t a t gal.

can stimulate t r a n s c r i p t We previously showed that induction of interferon is due to activation of transcription and that a segment of 117 nucleotides preceding the cap site suffices for inducibility of the IFN-ol gene. To identify the minimal sequence sufficient for induction, we constructed a set of hybrid promoters in which 3' truncated IFN promoters and 5 ' truncated 0-globin promoters were combined in various fashions and joined to the O-globin transcription unit. Mouse LMTK-cells were permanently transformed with the modified genes by TK-linked transformation and correctly initiated D-globin transcripts from induced and uninduced cells were determined by quantitative 51 mapping. The results show that an IFN promoter segment extending from position -675 to -64 (relative to the IFN cap site) is sufficient to mediate viral induction of transcription when placed 56, 78 and even 109 nucleotides upstream of the D-globin cap site. A 5 ' deletion analysis of the hybrid promoter with a junction at -64 of the IFN promoter and -56 of the D-globin promoter identified, as the minimal sequences required for induction, the following 46 nucleotides located between -109 and -64: 

The 8-keto short chain fatty acid (SCFA), acetoacetate, can be utilized by wild type Escherrchla ' ' GQU as a sole carbon and energy source. The structural and regulatory genes responsible for the degradation of SCFAs are encoded by the nfn system. Three structural genes and a regulatory gene have been identified on a 6 . 4 kb fragment cloned into pBR322, resulting in the recombinant plasmid pAT0. pATO contains &QA and W which encode the and subunits of acetyacetyl CoA transferase, ntpB which encodes thiolase 11, and which encodes a positive controlling element that regulates the activity of the nfn structural genes. These four genes have been mapped and are closely linked within the 47 min region of the i L & chromosome and may comprise an operon.

We have identified the proteins encoded by pAT0 via the maxicell procedure and as expected (Frerman & aL Archives Biochem and Biophy 171:14) have found that ntpB (the asubunit) codes for a 28 kd protein, (the Bsubunit) codes for a 26 kd protein, and ntpa codes for a 42 kd protein.

We have also identified the &QC gene product as a 48 kd protein.

Preliminary evidence has shown that the regulatory properties of the &Q system deviate from other well characterized positive regulatory systems (i.e. the ~L L d , u L mal systems). 

The leader sequences o f the E. c o l i M 1 RNA gene includes t h r e e p u t a t i v e promoter homologies.

The P-1 sequence i f nearest t o the 5 -end o f mature M1 RNA and most a c t i v e i n v i t r o . The promoter sequences were cloned s i n g l y and combination i n t o the Galk expression vector, pKO-100. Only the P-1 promoter directed the synthesis of galactokinase w h i l e the P-2 and t h e P-3 promoters gave the basal l e v e l s o f a c t i v i t y i n t h i s assay. DNAase I f o o t p r i n t i n g experiments showed t h a t sequences a t P-1 were protected by RNA polymerse t o a greater extent than those a t the P-2, w h i l e t h e P-3 was not observed t o be s p e c i f i c a l l y protected by the enzyme. Bal 31 nuclease d i g e s t i o n experiments were used t o examine whether upstream sequences a f f e c t e d the level o f galactokinase made from t h e M1 RNA promoter(s). galactokinase synthesis from the P-1 promoter. i n the M1 RNA s t r u c t u r a l sequence between nucleotides +22 and +168 which reduce synthesis o f galactokinase. This r e s u l t implies t h a t these s i t e s can control gene expression of M1 RNA, perhaps by t r a n s c r i p t i o n termination. I n order t o determine which regions o f cytoplasmic and s k e l e t a l muscle chicken a c t i n genes contain c i s -a c t i n g sequences important i n r e g u l a t i n g t h e i r developmentally timed expression, we have t r a n s f e r r e d i n t a c t and h y b r i d genes i n t o a myogenic c e l l l i n e and monitored t h e i r expression.

Each o f f o u r genes, i n t a c t @-actin, i n t a c t s k e l e t a l a -a c t i n , a 5'a- 3'8 Although t h e s e genes are t r a n s c r i b e d by RNA polymerase 11, they do W e have r e c e n t l y cloned and sequenced CHICKEN U2 -232-8 (Brosius, Gene 27, 151, 1984) . W e found two fragments that direct a greater level of CAT activity than the strong hybrid pranoter, tac (deBoer et al., ENAS 80, 21, 1983 , as constructed by Amaan et al., Gene 25, 167, 1983 The metallothionein gene i s transcriptionally regulated by both glucocorticoia hormones and oy metal ions such as cadmium and zinc. Karin and co-workers have identified separate regulatory sequences which control a metallothionein (MT) gene's response to metal and glucocorticoids respectively. We have isolated a rat MT-I gene and two pseudogenes and have characterized the pattern of DNase I hypersensitive sites prior to and after induction. Both pseudogenes are cONA copies of the MT-I gene. Pseudogene 21 contains homologous sequences only as far 5' as the cap site, but pseudogene 27 contains hanologous sequences 1 3 3 b.p. further 5'-This pseudogene contains all sequences required for metal regulated expression of the MT gene and probably arose from a transcript initiated at an upstream promoter. DNase I hypersensitivity studies show a single hypersensitive site upstream fran the MT-I gene. Pseudogene 21, which lacks all sequences required for expression, lacks this hypersensitivity.Pseudogene 27, which contains sequences required for regulated expression, also does not have a hypersensitive site, leading us to predict that this gene is not expressed. This prediction is substantiated by our failure to detect transcripts corresponding to this pseudogene by a variety of experimental approaches. These sequences which are required for expression of the MT-I gene, are therefore not sufficient to direct transcription of this MT-I pseudogene. Using a previously isolated cDNA clone coding for progesterone receptor B (PRB) antigen as probes, we have studied the hormonal regulation of its expression in chick oviduct as well as other tissues. We obaerved a 50-fold increase of PRB antigen upon secondary stimulation with either estrogen or progesterone for 16 hrs.

Testosterone and dexamethasone also increase PRB antigen mRNA but t o a much lesser degree. Nuclear "run-off" assay indicated that only part of this increase (2-5 fold) was due to regulation at the transcriptional level. Therefore, both transcription of the gene and stability of mRNA are affected by the presence of steroid hormones. In addition, prolonged administration of progesterone and estrogen resulted in a reduced level of mRNA sequences in the oviduct tissue while ovalbumin mRNA sequence continued to increase. Therefore, the PRB gene was apparently self regulated. When the other tissues were examined, we were surprised to detect the mRNA sequence for PRB antigen, although some of these tissues have not been reported to bind progesterone. This suggests that progesterone target tissues may require a redefinition to those which can produce the hormone binding isoform. The promoter elements necessary for the initiation of transcription of the ovalbumin gene have been localized previously by transfer of an ovalglobin fusion gene into HeLa S3 cells. A region containing the "TATA" box, a highly conserved sequence located around 25 to 30 base pairs upstream from the cap site of most mRNA-encoding genes, is required for initiation of transcription. A second region spanning between -95 to -48 upstream from the cap site is also required for efficient initiation of transcription.

Although the requirement of the "TATA" sequence for initiation of ovalbumin gene is well demonstrated & -, the importance of the upstream sequence is not yet defined. Here we show that in addition to the "TATA" box, DNA sequences located between positions -95 and -77 upstream of the cap site are also essential for efficient in vitro initiation of ovalbumin gene using various deletion mutants of ovalglobin fusion gene as template.

The enhancement of initiation of transcription by the upstream sequence can be demonstrated using circular or linear DNA as template and total cell or nuclear extracts from HeLa cells as source of transcription factors. In addition, the upstream sequence dependency is markedly influenced by the transcription factors to DNA ratio. These results are consistent with the notion that two promoter elements, the "TATA" box and the upstream sequences are necessary for efficient initiation of transcription in vitro as well as in vivo.

Rhodopseudomonas blastica and Rhodospirillum are purple non-sulphur photosynthetic bacteria that can p o w either by photosynthesis o r by respiration. Under both growth conditions their ATP synthase uses the energy of a transmembrane proton gradient to synthesize AT?. The enzyme comists of an extrinsic membrane sector (Fi) with five polypeptides ( a , p, y, 6 and E ) and an intrinsic membrane sector (F,) with at least three subunits. The genes for the five F1 subunits of the enzyme from cloned and sequenced. Analysis of transcription by Sq?uclease mapping v r extension techniques showed that the five F1 genes form an operon in both organisms.

genes is conserved in both bacteria and is the same as that in the Escherichia coli unc operon except that genes for Fo subunits are not associated with the F1 genes ( a s h e y a r e in E. coli). and further experiments showed the existence of a second promoter internal to the operon which transcribes only the two distal genes. To study the regulation of gonadotropin gene expression, we have isolated and characterized cDNA and genomic clones f o r the 6 s u b u n i t of bovine lutropin (bLH). The bLH6 gene i s expressed i n the p i t u i t a r y and i s present as a s i n g l e copy spanning l e s s than 1.1 kbp, c o n t a i n i n g t h r e e exons encoding an m R N A of 550 n u c l e o t i d e s . The m R N A f o r bLH6 contains an unusually short 5'-untranslated region of only 6-11 nucleotides; an unexpected f i n d i n g s i n c e t h e h i g h l y conserved 6 s u b u n i t genes of t h e human LHB/CGB gene f a m i l y (expressed i n p i t u i t a r y and placenta) have 5I-untranslated regions i n excess of 350 nucleotides. Further comparison between the bLH6 and hLH6/hCG6 genes reveals t h a t both have consensus TATA sequences i n identical positions. However, when the human gonadotropin 8 subunit gene family i s expressed in placenta, transcription s t a r t s a t an upstream promoter which bears no homology t o t h e consensus TATA sequence. Recently, we t r a n s f e c t e d t h e bovine LH8 gene i n t o a human placental c e l l -l i n e which normally expresses gonadotropin 8 subunits. O u r p r e l i m i n a r y d a t a suggest t h a t t r a n s c r i p t i o n o f t h e bovine LH6 gene may begin preferentially a t an upstream promoter s i t e , closely related t o t h a t normally used by the human genes. Additional experiments i n progress should reveal whether promoter recognition for t h e gonadotropin 6 subunit genes i s tissue-specific. Alternating purine-pyrimidine DNA sequences can undergo B to Z transitions when stabilized by negative supercoiling in physiological conditions. Studies with anti-2 antibody in biological systems strongly suggest a role in gene expression or DNA packaging, but there has been no direct evidence for this in vivo. The mouse metallothionein-I promoter has a potential Z-DNA sequence 86-102 base pairs from the transcription start site within the 216 bp Sst I-Bgl I1 region of the promoter thought to be involved directly with heavy metal induction. This sequence 5'-GCGCGGGTGETATGCGTG-3' is an alternating purine/pyrimidine sequence except for the central two base pair interruption. The Sst I-Bgl I1 fragment was cloned into pUC 12 and shown not to flip into Z conformation by two-dimensional chloroquine gel electrophoresis. To facilitate the B to Z transition and study its effect on transcription, we cloned this fragment into ml3-mpl0 and inverted the central AC to CA by oligonucleotide mediated mutagenesis. This creates a 17 base alternating purinejpyrimidine sequence within the potential Z-DNA region. The wild type and mutant sequence was confirmed by dideoxy sequencing. A 216 bp region containing the mutation was re-cloned into pUC 12. Two-dimensional chloroquine gel electrophoresis of the mutant metallothionein promoter demonstrated a structural transition which may involve Z-DNA. The effect this mutation has on transcription of linked genes with respect to heavy metal induction is currently being assayed by transfection and Northern Analysis. The genes encoding rRNA in the acellular slime mold Physarum polycephalum are located on linear palindromic extrachromosomal UNA molecules of 60 kb. Transcription initiates at sites 18 kb from each end and proceeds outwards. Two introns interrupt the 26s RNA coding region.

(1) We have determined the DNA sequence of the 22 kb comprising the central spacer region. ihe structure of this region is a complex array of palindromes within palindromes. Unlike in some other organisms, the promoter sequences are not repeated in the spacer.

(2) We are characterizing a strain of p. polycephalum that contains a third rUNA intron about 1 kb in size. UNA sequencing shows that this intron is located exactly at the nucleotide where an intron is found in species of Tetrahymena. L%romosomal DNA from the strains of Physarum lacking the third intron contains low or single copy sequences homologous to this intron.

(3) We are purifying and characterizing two nuclear proteins that bind specifically to sequences on the rDNA. One protein, purified over lOO-fold, recognizes a sequence about 100 nucleotides upstream of the RNA start site. A second protein fraction recognizes sequences at or near the telomers of the rDNA. 

Hospital, Memphis, TN 38101 Frog virus 3 (FV3) is a large icosahedral DNA virus whose genes are expressed in an orderly stepwise manner. Immediate-early RNAs, defined as those RNAs synthesized in the presence of cycloheximide, are not inhibited by a-amanitin in a mutant CHO cell line with an a-amantin resistant RNA polymerase 11, but are inhibited by a-amanitin in wild-type CHO cells, implicating the host polymerase in the synthesis of immediate-early viral RNA. We have cloned and sequenced the 78 base pair promoter region of a major immediate-early FV3 gene and found an A-T rich region (TATTTTA) at -30 bp upstream from the transcription start site. This presumed promoter was ligated into a plasmid 5' to the coding region of the bacterial chloramphenicol acetyl transferase (CAT) gene; the FV3 promoter-CAT construct was then introduced into a-amanitin sensitive and resistant cells by Cap04 co-precipitation. After 24 hr, one set of dishes was treated with W-inactivated FV3; extracts were prepared for CAT assay 6 hr later. CAT synthesis occurred only in UV-FV3 treated cells having a functional RNA polymerase 11. Therefore, a trans-acting component of the virion was required for recognition of the promoter by the host enzyme.

I n a currently popular model of eukaryotic tRNA gene Control, two small coding regions corresponding to conserved parts of tRNAs direct transcription of tRNA genes. At variance with this view are the observations that certain mutant tRNA genes do not have the expected phenotypes. In some cases, removal of one of the internal control elements does not abolish transcription; in others, mutations outside the critical regions have pronounced eftects on transcriptional activity. We have shown that the requirement for particular sequences downstream from the transcription initiation site is highly dependelfaon certain transcription reaction parameters. The full control region for a Bombyx tRNA gene defined by our experiments is larger than the transcription unit itself (98 bp), extending from at least -14 to +147 bp relative to initiation (+l).

The large control region is separable into two functional domains -a region including 5' flanking sequences, and a large (5140 bp) coding and 3' flanking region that binds a necessary transcription factor. What has complicated the analysis of tRNA gene control is a DNA-binding inhibitory substance present in the cell-free extracts typically used to catalyze transcription in vitra Specitically, conditions that permit the inhibitory substance to mask the contribution of certain control elements can make the control region appear smaller. When the effects of the inhibitor are minimized by the addition of non-specific DNA to transcription reactions, assays become more sensitive to mutant phenotypes, and the full size of the Bombyx tRNAAla gene control region is observed. We propose that this finding explains much of the observed variability in the sequence requirements for transcription of different tRNA genes. in tPA activity produced by different promoters, correlates well with the differences observed between the same promoters using the less sensitive chloramphenicol acetyltransferase assay. relative strengths/activities of the anti-NP Ig heavy chain gene, SV40 early and RSV LTR promoters. In addition, we have carried out experiments to examine the function of the Ig heavy chain enhancer. The boundaries o f the t r a n s c r i p t i o n a l c o n t r o l region o f Ad2 V A 1 RNA gene have been d e l i m i t e d i n t e r n a l l y a t about t9 t o +72. However i t i s n o t known whether a l l the DNA sequence i n the region i s absolutely required f o r an all-or-none t r a n s c r i p t i o n a l c o n t r o l e f f e c t . To answer t h i s question, linker-scanning mutations w i t h Kpnl l i n k e r , dCGGTACCG, replacing the DNA sequence between the two blocks were constructed. Four mutants i n which f o u r d i f f e r e n t c l u s t e r s o f DNA sequence between t h e two blocks replaced w i t h the Kpnl l i n k e r , respectively, were obtained. The t r a n s c r i p t i o n e f f i c i e n c i e s o f these mutants were about 50% t h a t o f the w i l d type gene i n d i c a t i n g t h a t the DNA sequence between t h e two blocks i s not absolutely essential f o r t r a n s c r i p t i o n a l c o n t r o l , b u t i t may be required f o r e f f i c i e n t t r a n s c r i p t i o n . Furthermore, three mutants w i t h longer DNA distance between the two blocks were a l s o constructed. The t r a n s c r i p t i o n e f f i c i e n c i e s o f these three mutants, (+-lo), (+17) and (+27), were 140, 120 and 42% t h a t o f the w i l d type gene, respectively, i n d i c a t i n g t h a t the distance between the two blocks can be extended t o about 6 1 bp and probably more, however the optimal distance i s about 44 bp. Moreover, s i x mutants w i t h s h o r t e r distance between the two blocks were constructed. The t r a n s c r i 

INTERACTION OF RNA POLYMERASE I TRANSCRIPTION FACTOR(S) WITH rDNA PROMOTER ELEMENTS, Ingrid Grumt, Detlev Buttgereit and Joachim Clos, Institut fur Biochemie der Universitat, Rontgenring 11, D-8700 Wurzburg, F.R.G.

The transcription of the ribosomal genes is very efficiently regulated according to the proliferation rate of the cells. A more or less efficient transcription is brought about by modulation of the initiation frequency of RNA polymerase I on the rDNA. The elucidation of the molecular mechanism of this transcriptional regulation requires both the identification of the promoter sequences and the protein factors which are required for the initiat ion process. A cell-free system consisting of crude nuclear extracts from cultured Ehrlich ascites cells was used which faithfully initiates transcription on a cloned 5'terminal fragment of mouse rDNA. The following results were obtained:

(1) The cell-free transcription system reflects the rRNA synthetic activity of the cells. Only extracts from rapidly proliferating cells promote transcription of cloned rDNA; extracts from growth-inhibited cells are transcriptionally inactive. (2) The transcription of ribosomal rDNA requires extracts from homologous cells, which indicates that species-specific factor(s) are involved in the initiation reaction.

(3) Fractionation of cell-extracts on several ion exchange columns showed that in addition to RNA polymerase I at least two proteins are required for accurate and efficient transcription initiation. TFIA is present or active only in rapidly proliferating cells and copurifies with RNA polymerase I . TFIIB is a species-specific DNA binding protein which is required for stable transcription complex formation. It is present both in growing or growth arrested cells.

(4) An RNA polymerase I control region essential for the initiation of pre-rRNA transcription has been identified by mutagenesis in vitro of mouse rDNA and transcription in cellfree systems derived from Ehrlich ascite= Substitution of nucleotides between -35 and -14 by foreign DNA sequences caused a loss of template activity, which indicates that an important promoter element is located within this region. At least two evolutionary highly conserved nucleotides a G at position -16 and a T at -1 play an important role in the interaction of TFIB with the rDNA promoter. Steroid hormone effects are mediated by intracellular harmone-specific receptor proteins; the hormone-receptor interaction increases the affinity of the receptor for nuclear binding sites, thereby modulating the expression of specific genes, of steroid action are not understood, the glucocorticoid receptor is perhaps the most fully characterized eukaryotic positive transcriptional regulatory factor. coid receptor binds in vitro with high affinity to defined regions of DNA near regulated promoters, and sequences essential for these interactions are functional in vivo as hormonedependent transcriptional enhancer elements. We have employed polysome immunoadsorption using glucocorticoid receptor-specific antibodies to obtain eight overlapping cDNA clones which together appear to contain the entire coding sequence of the rat glucocorticoid receptor gene. One of these clones containing a cDNA insert of 2 . 6 kb was used for a preliminary analysis of glucocorticoia receptor gene organization and expression. receptor seems to be encoded by a single copy gene which is indistinguishable in DNA isolated from rat liver or fibroblast cell lines. Wild type rat and mouse cells accumulate a 5.8 kb receptor transcript; receptor deficient (r-) mutants, which display reduced steroid binding activ.ity, accumulate reduced amounts of the 5.8 kb transcript. transfer (ntl) mutants, which produce a receptor of 40 kd rather than the 94 kd protein observed in wild type cells, contain a receptor transcript of only 4 . 8 kb. experiments of receptor coding sequences into receptor mutant cell lines are being performed to assess directly the functional potential of these clones. By fractionating various types of transcription extracts and examining different genes, we have recently identified several promoter-specific transcription factors that impart selectivity to RNA polymerases I and 11. The first of these auxiliary transcription factors identified was Spl, which binds specifically to the "GC-boxes" that are contained within the control sequences located upstream of the SV40 early promoter. More recently, this cellular transcription factor has been shown to bind and activate not only SV40 transcription, but also several other promoters, including the immediate-early promoter of the HSV ICP4 gene as well as cellular promoters such as the mouse dihydrofolate reductase gene and the promoter for a cellular monkey gene. In each of these cases, spl was found to bind multiple GC-box containing sequences located upstream from the start site of transcription and that activation of RNA synthesis required the presence of Spl. In addition to studying the interaction of specific transcription factors with viral and cellular genes, we have also begun to investigate the transcription of developmentally regulated and tissue-specific genes in Drosophila. Transcription of the alcohol dehydrogenase gene is under the control of two tandem promoters (proximal and distal) that are activated in a temporally-regulated fashion during Drosophila development. We have fractionated a Drosophila tissue culture extract system and identified a specific transcription factor, Adf-1, that binds to the upstream region of the distal Adh promoter and activates transcription. Footprint analysis also revealed the presence of additional sequence-specific binding proteins present in the extract that recognize and interact with the upstream regions of the proximal promoter of Adh.

that is required to activate transcription from the human ribosomal promoter by RNA polymerase I. Analysis of mutant templates suggests that the "core" control element of the human ribosomal promoter is required for activation of transcription by SL1. We have purified sL1 approximately 200,000 fold and have shown that addition of SL1 can reprogram the otherwise nonpemissive mouse transcription system to recognize and initiate accurate m A synthesis from the human ribosomal RNA promoters. a protein in the nucleolus of primate cells but not rodent cells. Moreover, anti-SL1 specifically inhibits in vitro transcription initiating from the human ribosomal promoter but not the mouse promoter. These findings suggest that SL1 is a nucleolar factor that imparts promoter recognition to RNA polymerase I, and that it can discriminate between promoters from different species. DNA binding studies suggest that, unlike the RNA polymerase 11 factors, SL1 is not a sequence-specific DNA binding protein.

As a third case study, we have identified a promoter-specific transcription factor, sL1, Antibodies raised against SL1 bind preferentially to We report studies: 1) defining, in part, domains of the & region and 2) analyzing two Nus functions, the products of the nut: Three regions of potential or proven importance have been identified in the & region: u, pyGCTCTT(T)A, the & stem-loop structure and m. Two mutations have proven that & is important in the gpN reaction, w l and -5.

The -1

(2) mutation results in a transversion that substitutes an A-T bp for a T-A bp. This change from CGCTCTTA to CGCTCTTT is necessary for the h N product to function with the NusA protein from Salmonella typhimurium. Two other related phages with different & products that can function with the NusA of Salmonella, phages 21 and P22, have & sequences with the three T's. & i proteins, one of which is the NusA transcription termination factor. We have used photolabeled NUSA to study the interaction of NusA with other proteins. Free NUSA is a dimeric protein.

When NusA dimers bind to multimers of the core component of RNA polymerase, both disaggregate to monomers. In standard transcription conditions NUSA binds tightly to RNA polymerase only after the initiation subunit U70 has disociated from the enzyme and RNA polymerase has paused at a NusA-sensitive pause site. However, the NusA-binding site on RNA polymerase is different from the o7O-binding site. affinity for NUSA and which is different from the conformation R stabilized by NusA. We suggest that NusA extends transcriptional pause times at NusA-SensiFive sites by stabilizing the Rp conformation of RNA polymerase. entation I1 of the insertion element IS1. The terminal octanucleotide, CUCAAAAU, in IS1 (11) is identical to the one in the Rho-and NusA-dependent terminator trpt'.

UCAA is a signal for Rho action and that the terminal As signal NusA involvement. IS1 (11) and the other cloned NusA-dependent terminator also have the perfect upstream match GCTGTITTA, and this is a sequence believed to be a signal for NUSA action. A model for regulation by NUSA and N at the nucleic acid level will be presented.

N cannot function in a strain with the nusE71 mutation in ribosomal protein S10. We have purified two proteins that restore N function in vitro when added back to a reaction containing SlOO extract from a nusE71 mutant. One is probably S10 itself functioning in transcriptional control as an extraribosomal protein. The other is a new E. coli transcription termination factor that we have called Zeta. Some of the properties of these proteins will be discussed.

U70 stabilizes a conformation of RNA polymerase Re which has low

We have cloned two NusA-dependent terminators. One is the Rho-dependent terminator in ori-We will argue that Subsequently, t h e l o n g u n t r a n s l a t e d and u n s t r u c t u r e d t r a i l e r region i s 

Gene Q of phage lambda, and corresponding genes of related phages, are positive regulators of phage late gene expression. They encode antiterminator proteins whose role is to allow transcription through terminators that precede and block expression of the phage late genes. We have purified the late gene regulators of phage lambda and its relative phage 82. They are active as antiterminators in transcription by purified RNA polymerase, and each is specific for the late gene promoter of its own phage. NusA protein, a transcription factor required for function of the lambda gene antiterminator, greatly stimulates antitermination by Q protein but i s not absolutely required. The sequences that encode specificity for lambda Q extend from within the promoter to about nucleotide 20 of DNA encoding the late transcript, although the antitermination activity is expressed at sites far distant from the promoter. Lambda Q protein can act after initiation of RNA synthesis: transcription from the lambda late promoter pauses at nucleotide 16 of the late transcript, A three-step w d e l has been proposed to account for transcription termination at rhodependent termination sites in E. coli. 'his model is based on in vitro studies of rhodependent termination of the transcript initiated at the PI( promoter of phage X ( 1 . 2 ) and on physical chemical studies of rho-polynucleotide interactions (papers i n preparation); it is supported by analyses of the sequence and secondary structure of transcripts subject to rhodependent termination that have been reported i n the literature (Morgan s e., submitted). psuses at or near potential termination sites. This pausing does not require rho, shows an -i n vitro relaxation time of at least 10 seconds, and depends only on local DNA template composition and sequence. hexamer must exist on the proximal portion of the nasceut transcript. This putative binding site i s proposed to be 70 to 90 nucleotides in length, to be relatively unencumbered by stable secondary structure, and (other than a requirement for some cytosine residues) to be non-sequence-specific. ( i i i ) Binding to this site activates the RNA-dependent ATPase of rho, leading to termination at the site of polymerase pausing Sy presently unknown processes.

In this lecture recent studies bearing on the mechanism of each of the above steps will be presented, and the present state of our overall knowledge of rho-dependent transcription termination will be summarized. 

Biochemistry, School of Medicine, Lava1 University, Quebec, Canada Control of transcription termination has been shown to be an important mechanism in procaryotic gene expression. However, much less is known whether a similar regulation exists in eucaryotic gene transcription. To define the transcriptional unit of the H5 gene, DNA fragments covering the whole genmic locus were subcloned into Kl3. Single stranded recombinants were then fixed on nitrocellulose filters and hybridized with in vitro nuclear transcripts of anemic chicken red blood cells. The results obtained s h o d that 1) the sequences imnediatly upstream of the 5' end of H5 m m are not transcribed, 2) only the coding strand serves as template for FNA polymerase I1 3) sequences as far as 470 bp downstream of the polyadenylation site are actively transcribed whereas no transcriptional activity is detectable 80 bp further downstream. To identify mre precisely the region of transcription termination, overlapping single-stranded probes of high specific activity were annealed to total anemic chicken RNA and digested with S1 nuclease. The results obtained with this method showed that transcription , i~ y& continues beyond the site of polyadenylation, but no unique termination site was found. Interestingly, a 158 bp region situated 100 bp beyond the polyadenylation site was found to be extremely sensitive to Mase I digestion in erythrocyte chromatin. Nucleotide sequence analysis of the saw region revealed a stretch of 200 nucleotides which is capable of forming stable secondary structures. We are now examinating whether these features influence the process of termination of transcription.

and Helga Boedtker. Harvard University, Cambridge, MA 02138 We have constructed a minigene of chicken proa2(1) collagen gene to study a) the formation of the 3' end of the mRNA and the use of the four different AAUAAA signal sequences, b) the correct splicing of the 1.9 kb intron 1 and 0.6 kb intron 2 resulting in the addition of the 11 bp exon 2 to the mRNA, and c) the role of GC-rich intron 1 in the regulation of gene expression.

The minigene extends from BamHI site at -1086 to the HindIII site in intron 3 , which is ligated to HindIII site in intron 51, includinq exons 1, 2, 3 and 52 and introns 1 and 2, 200 bp of intron 3 and 4 0 0 bp of intron 51 and 5 0 0 bp of DNA sequences after the last poly-A addition site. It is ligated to NdeI-BamHI fragment (from 2297 to 375) of pBR322.

Preliminary experiments including transient transfection of human 293 cells with the minigene, indicate that this gene may be accurately transcribed and results mainly in fully spliced transcripts, 680 and 1150 bp in size, with the 680 bp species predominating. Small Quantities of unspliced transcripts 3800 and 4250 bp, could also be detected. To determine sequences necessary for correct generation of each 3' end and to study the use of poly-A addition sites in 3 ' end deletion mutants, DNA seauences from the 3' end of the gene have been inserted into SP64 vector to make anti-RNA probes.

Studies on the regulation of E . coli adenylate cyclase gene revealed that CAMP-CRP acts as a transcriptional repressor for the w expression both --in vivo and in vitro. Quantitative analyzes of % mRNA by a dot blot and an S1 digestion assays indicated that crp cells produce about 5-fold more -mRNA than do wild type cells. The level of mRNA in cells was dramatically reduced by introducing a 9 plasmid and by adding CAMP exogenously. In vitro transcription of purified DNA fragments containing the % promoter region gave direct evidence that the transcription of C J = gene is specifically inhibited by CAMP-CRP. DNAase footprinting showed that CAMP-CRP interacts with a unique site, containing a consensus CRP binding sequence, which overlaps with RNA polymerase binding region. In addition, it was shown that the REIA polymerase-promoter interaction is alteredinthepresenceof CAMP-CRP. It is concluded that CAMP-CRP inhiDits the 9 transcription by preventing the functional binding of RNA polymerase to the promoter.

Massachusetts General H o s p i t a l , Boston, MA 02114 Enzymes whose r e l a t i v e r a t e o f synthesis i s a l t e r e d by i n s u l i n a r e being i d e n t i f i e d t o serve as markers t o d e f i n e how gene expression i s a l t e r e d by i n s u l i n (i.e., t r a n s l a t i o n o r t r a n sc r i p t In eukarvotes, despite several attempts, transcription-termination sites of RNA polymerase I1 transcripts have not been clearly identified. Consequently, the regulatory elements involved in the process of transcriptiontermination and antitermination could not be properly investigated, and they are almost unknown.

;ie have developed a system of isolated nuclei in which an efficient transcription-termination at the SV40 attenuation site is occuring. This system allows to start defining the regulatory elements involved in the mechanism of termine tion and antitermination.

Experiments will be described indicating the involvement of the SV40 agnoprotein in enhancing attenuation, of a heat labile nuclear factor and RNA secondary structure involved in the process of transcription-termination and of an heat stable nuclear factor involved in the process of antitermination. 

We have evaluated the usefulness of different promoters and terminators for expressing the human tissue plasminogen activator (tPA) gene in mouse C127 cells. The tPA cDNA gene was linked by transcriptional fusion to promoters such as the mouse metallothionein (MMT) promoter, the Rous sarcoma long terminal repeat (LTR), or the mouse Moloney leukaernia virus LTR. the MMT polyA site or the SV40 early polyA site. The reconstructed tPA genes were inserted into bovine papilloma virus (BPV) vectors containing the BPV genome, the MMT gene, and bacterial plasmid DNA. Cells were transfected with the DNA constructs and tPA-producing foci were easily identified using a novel application of the fibrinagarose assay method for detecting tPA activity. Stably-transformed, tPA-producing cell lines were analyzed in detail for the physical state and copy number of the tPA gene in the cells, the level and authenticity of tPA transcription, and the amount of active tPA protein being secreted. Susan speciric role of agmprotein in the sv40 lytic cycle, we have ccnstructed nmkey a -l P cell Lines in which the agnogene, lmder the control of a retrovirus LlR, is stably integrated and colstitutively expressed. Viruses with pint ad deletion nutatians in the agnogene, thich mke cell l i n~~. hthm blot analysis of the viral DNA extracted franapz-prducing cell Lirrs infected with agm mtants dmmstrates that the agmgene has not rembimi into the mhant vinrsg to r-mt wild-type virus. lhs, the agmptein f m n the cell Lines restores wild-type plaquirtg size in tlans, indicating a positive role of the agnoprotein with respect to plaque size. n-anscripticnal studig will be dkuss€d.

smll plaques in no& a-1p cells, @lKe wild-type sized plaques in the agmptein prcduzing

The anthracycline antibiotic duanomycin is widely used in cancer chemotherapy. The drug is a potent inhibitor of both transcription and DNA replication. Previous results from this laboratory have shown that duanomycin will bind preferentially to alternating purine-pyrimidine DNA sequences. Since these are the sequences able to undergo the transition from B form DNA to the left handed 2 form, the effect of daunomycin on the B to 2 transition has been examined. Daunomycin inhibits the rate B to 2 transition. Binding of the drug to poly d(G-C) under solution conditions that favor the Z form is cooperative, a finding consistent with a model in which the drug preferentially binds to the B form, and allosterically converts 2 DNA to an intercalated B form. The allosteric conversion of 2 DNA back to the B form has been directly demonstrated by circular dichroism, sedimentation, and enzymatic methods. The conversion of 2 DNA to the B form is strongly dependent on ionic strength, and under some conditions as little as one drug molecule for every 25 b.p. is sufficient to completely convert the polymer to the B form. This is a striking demonstration of how a small molecule may exert long-range allosteric effects on DNA conformation. These observations are important for understanding the molecular mechanism by which daunomycin acts, and may be of interest as an indication of long-range conformational effects of molecules on DNA that may be important in transcrip- Aminonucieoside of puromycin (AMS), a selective inhibitor of the growth of estrogenresponsive as opposed to estrogen-unresponsive mammary tumor cells, has been found to inhibit a particular class of poly(A)-containing mRNA species in estrogen-responsive cells in rltcn. Estrogen-responsive, MCF-7 cel is and estrogen-unresponsive, BT-474 cel Is were maintained for one day prior to hormonal stimulation on media supplemented with hormonestripped serum. in a parallel series of experiments, the strlpped media was replaced for b th cell iines with meaia supplemented 51th a physiological concentration of estradioi (10 Cytoplasmic RNA was extracted from magnesium precipitated polysomes by SDS-phenol-chloroform extraction procedures, and poly(A1-containing species were separated by poiy(U)-sepharose column chromatography, eluting wlth increasing concentrations of formamide. The mRNA species were fractionated by sucrose gradients, and assayed for radioactivity through I lquld scintillation counting. Inclusion of AMS (100 ug/ml) in the stimulating medium resulted in inhibition of a particular class of poly(A)(+) mRNA In estrogen-responsive cells, within 1/2 hr, while this class was resistant to AMS in estrogen-unresponsive cells. These results indicate sane alteration in the mRNA metabol Ism of estrogen-unresponsive mammary tumor cells. and point to a possible growth regulatory function for this AMS reslstant class of mRNA in these cells. The activity of mammalian ornithine decarboxylase (ODC) changes significantly and promptly in response t o multiple effectors of cell growth and differentiation. The availability of a cDNA probe for mouse ODC has facilitated study of the mechanism of induction (PNAS g:540 ( 8 4 ) ) .

Nerve growth factor (NGF) causes differentiation of cultured rat PC12 pheochromocytoma cells and induces ODC activity about 10-fold within 4 hours.

Both time-course and dose-response experiments indicate a n excellent concordance between ODC activity and the level of ODC mRNA. A substantial (but lesser) degree of induction o f mRNA by NFG is seen when protein synthesis is inhibited by cycloheximide. W e conclude that most o r all of the induction lies a t the level of mRNA and that the action of NGF is not dependent o n protein synthesis. in press). transcription when mved to another genomic site, portions of the putative termination region have been inserted into the ademvirus (type 5) chrclrrom. 'Ibe present series of viral insertions were made within the second exon of the EIA transcription unit. labeled either in isolated nuclei or in &le cells early after infection with reconstructed viruses indicated that transcription is termi~ted if the inserted DNA contains the globin ply(A) site plus an additional 1395 nucleotides downstream. must be indirection with respect to transcription as in the B globin transcriotion unit. transcriptional termination. tion unit, the insertion of the terminator region hEd a negative cis effect on the E1B transcription unit which begins 363 base pairs downstream from the globin insert. The E1B transcription unit was the only early gene affected, and caplerentation of the virus containing the functioning terminator region within a functional ElA did not restore transcription Of the ElEl gene. n * r of functional transcription units, the terminator efficiency is slightly reduced, and E1B transcription is recovered.

Telmination of the EIA pdemvirus 'Transcription Unit by Insertion of the muse 3-MajOr Globin Terminator E l a n t . C . F r a n k l i n , U n i v e r s i t y o f U t a h , S a l t Lake C i t y , Utah 84112

The "N" t r a n s c r i p t i o n -a n t i t e r m i n a t T h e 118 nucleotide band behaves like a circular RNA and. by fingerprint analysis. has been shown t o be contained within t h e intron. This R S A comprises the circular portion of the l a r i a t t h a t has been described by others. but d o e s not contain the handle.

No o t h e r circular molecules a r e d e t e c t e d .

T h e band whose mobility depends on Proinsulin mRNA was analyzed by RNA blot hybridization in three insulin-expressing tissues from the rat, adult pancreas, an insulinoma cell line and fetal pancreas. The proinsulin mRNA transcripts from the tumor cell line and fetal pancreatic tissue were estimated to be respectively 100 bases and 50 bases larger than the adult pancreatic transcript. I t has been shown that glucose is an important regulator of proinsulin mRNA fi 9. There is known to be a marked increase in the concentration of proinsulin mRNA and insulin in the developing rat neonate although plasma glucose levels are quite low. Expression of proinsulin mRNA independent of glucose levels is also found in insulinoma tissue. In addition, there is a second TATA sequence upstream of the putative start site in rat insulin gene 11. These observations, including the fact that the transcription initiation site(s) has never been mapped in these tissues, suggested that alternative promoter sites may be important in control of initiation of gene transcription. To map the 5' end of the gene, primer extension was performed using a synthetic oligonucleotide primer complementary to the first twenty bases of the coding portion of the two rat insulin genes. The extended products of the proinsulin mRNAs from the three tissues were identical indicating that at least 95% of proinsulin mRNA transcription occurs at the putative start site. The 3' ends of the proinsulin mRNA transcripts were evaluated by ribonuclease H digestion and i t was shown that the noted size differences could be accounted for by different length poly A tails. Experiments are currently in progress to define the potential role of poly A tails in stability and physiologic regulation of proinsulin mRNA. . Boston, MA The long terminal r e p e a t s (LTRs) t h a t f l a n k i n t e g r a t e d r e t r o v i r u s DNA contain t r a n s c r i p t i o na l c o n t r o l elements i n an unusual arrangement.

The polyadenylation s i g n a l i s between t h e promoter "TATAA" box and t h e s t a r t s i t e of i n i t i a t i o n , and t h e polyadenylation s i t e is a t base 21.

Normal v i r a l t r a n s c r i p t s a r e i n i t i a t e d i n t h e upstream LTR and polyadenylated i n t h e downstream LTR.

Two o t h e r t y p e s o f v i r a l t r a n s c r i p t s can be imagined: ( i ) downstream t r a n s c r i p t s , i n i t i a t e d i n t h e downstream LTR and elongated i n t o t h e adjacent c e l l u l a r DNA and ( i i ) readthrough t r a n s c r i p t s i n i t i a t e d within t h e upstream LTR, but elongated p a s t t h e normal polyadenylation s i t e i n t h e downstream LTR. Downstream t r a n s c r i p t s have been shown t o a c t i v a t e t h e c -w gene i n r e t r o v i r u s induced lymphomas of chickens.

W e used nuclease mapping techniques t o f u r t h e r c h a r a c t e r i z e v i r a l t r a n s c r i p t s t h a t a r e elongated i n t o adjacent c e l l u l a r DNA i n c l o n a l and non-clonal populations of i n f e c t e d C e l l s . W e found readthrough RNA i n tumors and f i b r o b l a s t s i n f e c t e d i n c u l t u r e a t r e l a t i v e l y high l e v e l s , approximately 5% of v i r a l t r a n s c r i p t s . Histone mRNA and small nuclear RNAs such as U1 and U2 a r e the only stable products of RNA polywrase I1 lacking a 3' terminal poly(A) tract. Work i n other laboratories has shown that the 3' end of histone mRNA i s formed by mRNA processing; the reaction requires a stem and loop structure as well as a CAAGAA signal located at a fixed distance downstream. Like histone mRNA, u1 and U2 have a 3' terminal stem and loop structure, but the stem and loop stuctures of the two genes d i f f e r i n size and i n sequence. Sequences downstream from the stem and loop are sonewhat conserved armng U1 and U2 genes from different species, but do not strongly resemble the histone CAAG4A signal. W e have assayed normal and mutagenized h m n U1 and U2 genes using a variety of heterologous and homologous systems, including microinjection of the genes or labeled SP6 runoff transcripts i n t o Xenopus oocytes, and transient expression i n HeIa c e l l s using SV4O-based vectors. W e can show that the 3' end of U2 RNA i s generated by RNA processing i n wo d i s t i n c t steps: cleavage of a longer precursor t o an intermediate known as U2+10 (about 198 n t ) and maturation of U2+10 t o U2 (188 n t ) . In contrast t o histone mRNA processing, we find that the 3' telminal stem and loop structure of U2 i s not required for processing, and that an W4 sequence located between +6 a d + 23 downstream from the mature 3' end of U2 i s independently capable of directing an RNA cleavage at the U2+10 site. Ihe processing of U1 snRNA appears t o be substantially similar t o that of U2. t h e i r f l a n k i n g reghops were determined. The & gene c o n s i s t s of a n operon i n c l u d i n g t h e genes for t R N A d , a 15 kDa p r o t e i n whose f u n c t i o n i s unknown, nusA p r o t e i n a q I n t h e DNA region between g e n e s f o r t h e tRNA$ and t h e 15 kDa p r o t e i n , t h e r e a r e two i n v e r t e d sequences followed i n each c a s e by a run of thymidines. These a r e t h e t y p i c a l p-independent t r a n s c r i p t i o n a l termination s i g n a l s . We have identified a family of messenger RNAs whose abundance in mouse kidney and liver is regulated by testosterone. Some ar all of these RNAs are present at constitutive levels i n various other tissues. Northern blots of kidney RNA show six species ranging in size from 1350 t o 2500 nucleotides. Charccterization of cDNA clones and preliminary results of genomic DNA blots indicates that some of these mRNAs may be transcribed from the same gene, w i t h the use of alternative polyadenylation sites explaining the observed differences i n size. A single polypeptide of mol. wt. 42,000 daltons is translated from these pooled RNAs. Amino acid sequence as deduced from the cDNAs does not correspond to any known sequenced protein. Hence, we have termed this family of RNAs "MAK" (mouse-androgen-kidney).

A polymorphic difference exists between some mouse strains. A B I repetitive element is present i n the 3' untronslated region of large MAK transcripts in DBA/2J mice.

This repeat is precisely missing in the corresponding regions of C57BL/6J and BALB/cJ transcripts. Translation of a leader peptide coding region containing tandem Trp codons governs formation of the alternative secondary structures. Ribosome stalling on the Trp codons promotes formation of a secondary structure that allows transcription of the operon, while ribosome movement to the stop codon causes transcription termination. For attenuation to be efficient, translation of the leader peptide coding region must be coupled to transcription of the leader region. A transcription pause site in the leader region may accomplish this synchronization by halting transcription until the translating ribosome releases the paused RNA polymerase. We have demonstrated that the translating ribosome does indeed release the transcription pause in the a leader region during coupled transcription/translation. Plasmid DNA templates with wild type and mutant sequences incapable of directing leader peptide synthesis were analyzed for the quantity of steady state pause RNA during coupled transcription/translation reactions. The mutant templates produced levels of pause RNA 10-fold greater than wild type. All templates gave nearly identical pausing kinetics in a purified transcription system lacking translational machinery. Release of the paused transcription complex by translation of the leader peptide coding region was confirmed by adding the translation inhibitor kasugamycin to coupled reactions with the wild type template. The treatment of cultured cells with interferons results in the synthesis of a number Of unique polypeptides. We have recently characterized two cDNA's (pIF-IND1 and 2) which hybridize to two W A ' s whose transcription is activated when human diploid fibroblasts or HeLa cells are treated with type 1 interferons (Larner et al. PNAS, in press). In fibroblasts, enhanced transcriptional activity of the genes corresponding to IF-IND1 and 2 occurs with less than 30 min of interferon (IFN) treatment, and the transcriptional activation of these genes declines to basal levels after 8 hrs of IFN treatment. Treatment of fibroblasts with cycloheximide prior to the addition of IFN-a both increases the rate and the period during which these genes are transcribed compared to cells incubated with only IFN-a. The action of cycloheximide is specific, since it has no effect on the transcription rates of other genes not induced by IFN such as 8-tubulin, arginine tRNA, or 28s ribosomal RNA. HeLa cells treated with IFN-a demonstrate a similar time course for the initial transcriptional activation of IFNl and 2. However, these genes are actively transcribed for 24 hrs in the continuous presence of interferon. In addition, treatment of HeLa cells with cycloheximide prior to the addition of IFN is without effect on the rate of IFN-a-induced transcription of these genes. These results suggest that a factor(s) is present in fibroblasts, but not in HeLa cells that Selectively Prevents the transcription of the IFN-induced genes. Protein synthesis appears to be required for this factor(s) to exert its action in human fibroblasts. 

We and &Z genes using an open reading frame vector. The tribrid proteins expressed from gene fusions can be used as an antigenic source for antibodies to the exogenous gene product (Weinstock, G., et al., PNAS 80:4432-4436, 1983) . The use of the antibodies to isolate the * gene product and to localize the protein in g. halobium cells is underway.

Currently studies on the effect of oxygen tension on the levels of & and The &gene product

Aspartate transcarbamyiase in E. coli is encoded by the pyre1 operon. The DNA sequence 5' to the beglnnlng of the pyrE gene contains a short open readlng frame and a G-C rich region of dyad symmetry followed by a string of eight T's. suggesting that an attenuation mechanlsm may be responsible for the 70-fold increase in expression observed upon pyrimldlne starvation. This hypothesis was tested by subcloning the promoter region of the pyrE1 operon so as to be Immediately upstream of the €. coli galK coding sequence in the plasmld pKO1. Cells containing this piasmid. pPYRB10. exhibit 70-fold gaiactokinase regulatlon characteristic of pyrEl operon expression when starved lor pyrlmidlnes. Deletlons constructed in the promoter reglon of pPYRBlO from the 3' side produced one plasmid that exhibits wild-type regulation and several plasmids that overexpress gafK even In the presence of large pyrimidine pools.

The functionally wild-type plasmid was sequenced and found to contain the entire region of dyad symmetry. including the 8 T 's The overexpression deletions lack the region of DNA with dyad symmetry or the 8 T's.

However. ail deletions of this kind stlli exhiblt residual levels of regulation. even though one deletion extends past the entire sequence coding for the putative leader peptide up to the major promoter. These results support an attenuation model. but suggest that other mechanisms may also participate in the regulatlon of the pyrEf operon.

Levln and H K. Schachman. Universlty of Calllornla. Berkeley, 94720. In the last few years, a number of viral and bacterial terminators have been cloned, sequenced, and analyzed. The majority of these studies concern transcriptional termination in gram-negative bacteria.

Escherichia &, very little information is available on transcriptional termination in -~ B. subtilis. A number of 8. subtilis genes have been sequenced and reports indicate that some of these genes have terminator-like sequences similar to those observed in gram-negative bacteria.

to gram-negative terminators, we have used well-characterized E. coli terminators, TlT2, located at 3' end of rrnB gene to investigate this problem. that the promoterless cat cartridge derived from the transposon Tn-9 lacks transcriptional terminators. We f u s e d T T 2 terminator to the 3' end of these cartridges to study the in vivo function of prokaryotic terminators in E. coli and B. subtilis. and S1 nuclease mapping of recombinant plasmids carrying this terminator in B. subtilis demonstrate that the T1T2 terminator derived from E. coli rrnB operon functions in we are examining a set of co-ordinately regulated genes expressed early in the developmental cycle of Dictyostelium. These genes are not expressed in vegetative cells and transcripts are first detectable at about 4 hours during normal development on filter pads. of these genes is stimulated by pulsing cells with low levels of CAMP, a condition that mimics the in vivo pulsing during aggregation (4 to 8 hours into development). Expression is inhibited by high, continuous levels of CAMP, a condition found later in the developmental cycle at a time when the expression of these genes decreases in vivo.

We have examined the structure of these co-ordinately regulated genes, particularly the sequences at the 5' ends, and have identified regions of homology that may be involved in their regulation. We are attempting to better determine the function of these regions using the DNA-mediated transformation system developed in our laboratory. W e a r e investigating, by a v a r i e t y of biophysical and biochemical approaches, the following three-step mechanism for rhodependent t r a n s c r i p t i o n termination i n E. c o l i . (i) Termination is i n i t i a t e d by the rho-independent pausing of RNA polymerase i n regions of the template that a r e GC-rich or contain palindromic sequences. (ii) lhis is followed by the r e l a t i v e l y non-sequence-specific binding of rho hexamers to a s i t e on the nascent RNA that is largely devoid of secondary s t r u c t u r e . (iii) This binding, i n t u r n , a c t i v a t e s the RNAdependent ATPase of rho, which leads t o termination. rRNA transcription is down regulated in response to starvation in Acanthamoeba. This regulation can be reproduced in a faithful in vitro transcription initiation system: SlOO extracts from vegetative cells actively transcribe rRNA while extracts from starved cells cannot unless they are supplemented with RNAP I purified from vegetative cells. The levels of both the transcription initiation factor(s) [TIF I] and the levels of RNAP I assayed on nonspecific templates (i.e. calf thymus DNA) are constant, but the RNAP I from the starved cell cannot specifically initiate transcription. This property is retained in RNAP I purified to near homogeneity from the starved cell even though its specific activity on calf thymus DNA matches that of the vegetative polymerase. RNAP I from starved cells has the same subunit composition as that from vegetative, but is 5X more heat labile than vegetative enzyme when assayed on nonspecific template. The small amount of enzyme active in the specific initiation assay has heat denaturation properties of vegetative polymerase. BAL-31 deletion mapping has shown that the sequence from -47 to +8 is necessary and sufficient to promote rRNA transcriptionno far upstream sequence requirement can be detected in vitro. This sequence region has been shown to contain two domains involved in TIF-I binding plus a third required motif, and to be sufficient to demonstrate regulation in vitro. Therefore, the steps in transcription involving this DNA sequence and RNAP I are impaired by the modification of the enzyme from starved cells. Supported by NIH GM26059 and GM22580. 

t a n t f the &A(Ind?-al lele &A3.

Cel Is exhibit a so-called "spl i t-phenotype", a phenomenurn in which only a subset of the SOS responses can be detected physiologically following inducing treatments. and has a second mutation downstream. RecA protein-catalyzed route in vivo, but is degraded by the Lon protease at both permissive and non-permissive temperatures.

were measured at 3 0 ' and 42OC to determine levels of expression; the differentialexpression of SOS functions oives a plausible explanation for the "split-phenotype '' associated with lexA4l. k A 4 1 has been cloned and sequenced; it retains the lexA3 mutation We show that LexA41 protein is not cleaved by the B-aalactosidase activities of fifteen S0Sop::lac fusions I

The cluster of four a interferon genes was identified in the 28kb l o n g fragment of mouse genomic DNA. The DNA sequence analysis indicated high homology (90-95%) on nucleotide levels among three of these four genes. The fourth gene (a,,) , which showed only about 80-85% homology to the others, contained an internal deletion of 15 nucleotides in the coding regioq all of these genes coded for biologically active interferons when expressed In = , however, the activity of the ah peptide was 100-fold lower than the activity of the other a interferons. Three of the four genes were expressed in virus (Newcastle disease virus -NDV)

infected, but not in the uninfected L-cells as found by S-1 nuclease analysis. The relative levels of the a s mRNA was, however, about 10-fold higher than that of the other a genes. The DNA sequences of the promoter region of these four genes were compared and it was found that the 5' flanking region of the a4 gene contains a number of point mutations and the insertion of a 17 nucleotide long G rich repetitive sequence not present in the 5' flanking region of the other genes.

In addition to a interferon genes, NDV induces in L-cells also the expression of 61 interferon gene, while in poly rI.rC induced cells, the 6 interferon gene was expressed predominantly. Surprisingly, however, it was found that in poly rI.rC induced cells both positive and negative DNA strardsof the 6 interferon gene were transcribed, while the virus induction led to the transcription of the proper strand only. This data are the first indication that the Induction of 8 interferon gene by virus and dsRNA may not be identical and suggest a novel mechanism for the regulation of B interferon synthesis in the induced cells. The T1T2 dual terminators of the E. coli, rrnB operon are able to atop anti-terminating rRNA transcripts. through the TIT2 region. phenotype, which result in expression of a downstream gal K gene on a plasmid. these mutations map in or near rpo B: the remaining two map elsewhere. None of the mutants reads t h r o u m T 2 unless an 80 bp fragment required for anti-termination is present. indep.) or the I32 terminator (rho-dep) will atop transcription rendering the cells Cal-Therefore, the defect in the mutants may specifically affect anti-terminating transcripts.

In addition, about 50% of spontaneous rifampin resistant (rpo B) mutants allow read through of TlT2. also in normal rho-indep. termination at T1 alone, yet terminate of the IS2 rhodep terminator. These data suggest that the mechanism of termination of anti-terminating transcripts may be distinquished from those yielding normal termination. super-termination, for the ability to terminate an anti-terminating transcript.

We have obtained E. coli mutants that allow RNA polymerase to read

We have isolated four independent host mutations with this Two of

In the absence of this 80bp fragment T1 alone (rho-Two of these mutants have been examined andurned out to be defective

We propose a special term,

W e have constructed a plasmid cloning and expression system for E. coli in which direct attachment of a gene to sequences regulating transcription and t F a n m i o n can be achieved rapidly and efficiently. The system includes the following main features: 1) A multi-copy cloning plasmid contains a bank of 10 unique restriction sites which are flanked by transcriptional terminators. A gene (clockwise 5' to 3 ' ) can be inserted at one of the restriction sites. It is protected by the 5' terminator from unregulated transcription which prevents its establishment. The gene can be conveniently modified in the cloning vector to contain a restriction site at the desired 5' end. Although it is well documented that actively transcribed genes are contained within a more open, DNase I sensitive chromatin conformation, little i s known of how this structure is established. One model suggests that the DNA sequence at the transition from the inactive to active chromatin, through the binding of a sequence specific protein, causes the disruption of the highly compact closed chromatin structure. By defining the DNase I sensitive domain around the ovomucoid gene we hope to identify such a sequence by comparison to the previously defined ovalbumin chromatin domain. From a cosmid library we constructed a clone containing the ovomucoid gene and 16 kb of 5' and 18 kb of 3' flanking DNA has been isolated. The DNase I sensitivity of DNA fragments around the ovomucoid gene has been determined via a novel probe-excess solution hybridization assay using SP6 RNA polymerase synthesized RNAs as probes. The DNase I sensitive region extends approximately 15 kb beyond the 5' end of the ovomucoid gene and about 10 kb from the 3' end of the gene. Further analysis should precisely define the boundaries of this domain. As was the case with the ovalbumin DNase I sensitive domain, members of the CRI family of middle repititive sequences are found near the transition from an open to a closed structure. This is in agreement with the proposal that these CRI sequences may act through the binding of a sequence specific protein these open chromatin structures during diEferentiation of the chicken oviduct cell. A fusion gene consisting of the promoter region from the mouse metallothionein-I (IQ-I) gene joined to the coding region of the HSV thymidine kinase (TK) gene i s efficiently regulated by. zinc when transfected into baby hamster kidney cells in a transient assay. Analysis of similar plasmids containing 5' deletions, internal deletions and linker scanning mutants of the MT-I promoter region confirmed the function of the previously recognized metal regulatory sl?ment-a (MRE-a), sequence CCTTTGCGCCCG, between -55 and -44 bp from the cap site, and further localized upstream sequences with regulatory function to short regions of DNA containing sequences related but not identical to MRE-a. In total we recognize five MRE's within 180 bp of the hT-I cap site. taken the annroach of trying to build a metal-responsive promoter by inserting a 17-bp synthetic DNA fragment containing the MRE-a sequence into the non-responsive promoter of the TK gene. No induction by zinc was observed with single insertions of the regulatory sequence, whereas many different construccioci with two copies of MRE-a were inducible. The precise position of t h two MKB's relative to ex?, other or to other promoter elements had relatively little effect upon t5e efficiency of induction, but the inducibility could be further increased by introduction of a third and fourth HRE-a seauence. The MRE's act cooperatively with the TK distal promoter elements, but in the presecce of the TATA-box alone they function as zinc-dependent positive promoter elements. Transgenic mice were generated by microinjection of a mouse metallothionein I/human growth hormone fusion gene into fertilized mouse eggs followed by implantation of the embryos into pseudopregnant foster mothers. Five of twenty-six mice born after one series of injections contsined from 1 to 8 copies of the fusion gene stably integrated into their genomes and expressed human growth hormone in their serum. transgenic offspring were treated with glucocorticoids, serum growth hormone levels were induced from 1.5to 6.3-fold. of the five mice was also observed after treatment with glucocorticoids. DNA sequences responsible for the observed inductions must be present in the human growth hormone portion of the fusion gene since other fusion genes containing the mouse metallothionein I promoter were not glucocorticoid inducible in previous gene transfer experiments.

In particular. a recently identified glucocorticoid receptor binding site within the first intmmof the human growth hormone gene might be able to confer inducibility on otherwise uninducible mouse metallothionein I gene constructs.

When several of these transgenic mice and Prom an a n a l y s i s of i n v i v o t r a n s c r i p t i o n p a t t e r n s , t h e f a a t o r s which i n f l u e n o e DOVemont of RNA polymerase through the fi aontrol region and generation of l.oX DRNA 3 ' end8 emerge. S1 nualease mapping data and f i n g e r p r i n t s of PNA hybridirable RNAs with endpoints I, 11. and I11 e r e present. but a r e minor a a p o n e n t s a a p a r e d t o readthrough t r a n s c r i p t s .

Of t r a n s a r i p t i o n f a a t o r t o a n t t h e l e n g t h of i t s own t r a n s a r i p t , separating t h e LaeJ. and genes i n t o d i a t i n a t t r a n s c r i p t i o n units.

Issaquah Health Research Institute, Issaquah, Washington. 98027

African trypanosomes contain numerous variant surface glycoprotein (VSG) genes. Each trypanosome expresses these genes only during restricted stages of the life cycle, expresses only one VSG gene at a time, end can switch expression among the VSG genes. We have found that the IsTat 1.1 VSG gene is located near the telomere of a stable minichromosome (-100 kb) which lacks restriction sites over most of its length and lacks a spliced leader coding sequence. We have cloned and sequenced the expressed 1 VSG gene and have found characteristic telomeric sequences downstream. The 1 VSG gene is transcribed in variants of the 1 variant antigenic type but not in other variants from the same trypanosome clonal lineage that contain this gene nor upon conversion to another life cycle stage. The Lranscriptional activation and inactivation of the 1 VSG gene does not involve its duplication nor sequence alterations detectable by restriction enzyme mapping. Using probes cloned from the 5' flank of the expressed I VSG gene we have demonstrated apparent VSG mRNA precursors and a stable processing product. Transcription appears to be initiated near the beginning of the region lacking restriction sites.

DIFFERFNTIAL TRANSCRIPTION OF MINICHROMOSOMALVSG GENES IN TRYPANOSOMA BRUCEI. Stuart, K., Scholler, J., Rothwell, V., and Aline R. Jr.

Supported from NIH and WHO.

The mutant allele G1 of the Droso hila locus Glued has a dominant effect primarily on the development of the visual system,Pi6cluding the eye and i t K u E 1 connections to the opiic lobe of the brain. Restriction site mapping of genomic DNA clones from the normal G1 and mutant G1 locus showed that G1 contains a 9 kbp insertion of a retroviral-like transposon 8104 into a transcribed region. The insertion causes formation of a truncated polyadenylated G1 transcript which terminates within the 5' terminal repeat of 8104.

We are focussing on three aspects of this findinq. One is the mechanism of the dominant effect of G1, and the role of transposon-induced mutations in producing such dominant effects. Another is the regulation and function of the normal Glued locus during Orosophila development. The third is the remarkable cis interaction between the 8104 insertion in G1 and nearby insertions of the P-element transposon, which results in the excision of 8104 and partial restoration of normal Glued function. Hinter.Laboratory of Molecular Riology, Medical Research Concil Centre, Hills Road, Cambridge CB2 2QH. ENGLAND The several E.coli amino-acyl tRNA synthetases studied appear to have different patterns of regulation. For example, alanyl-tRNA synthetase and phenylalanyl-tRNA synthetase appear to be autoregulated (Putney and Schimmel,l981; Springer et al, 1983) . but glutaminyl-tRNA synthetase is not (Cheung and Soll, 1984) . We have characterized the 5' non-coding region of the tyrosyl-tRNA synthetase gene (tyrS) of Bacillus Stearothermophilus and have experimentally identified a promoter and terminator. Presumably, the terminator is involved i n regulation of tyrosy-tRNA synthetase levels, but our data indicates that it is not regulated directly by the enzyme. Furthermore, the terminator differs from those found i n amino acid biosynthetic operons in that there is no peptide rich in tyrosine i n the putative attenuator region. References:-Putney and Schimmel ( Genes coding for H4 histones expressed in sea urchin embryos at different stages are highly diverged and organized in a very different way. The early H4 gene is part of the tandemly reiterated 5 histone gene unit whereas the late H4 gene occurs mostly in H3-H4 pairs present in far fewer copies per genome (40 VS. several hundred early H4 genes). We have demonstrated that nuclei isolated from the sea urchin, S, puruuratus, transcribe H4 histone sequences which are specific to early and late embryonic stages, as in vivo.

late H4 RNA (>50 -fold in 4 hr) is the result of transcriptional activation. The 5' ends of the nuclear H4 transcripts have been mapped by an RNase T(protection assay. but there are also run-through transcripts from upstream promoters. terminate in an area spanning the spacer sequence between the H4 and H2B genes. The late transcripts start at several closely spaced sites near the cap site. have identified 3 upstream sequences which are present in, and unique to,all genes examined thus far. both early and late nuclei, initiation may be taking place in vitro. We are testing this possibility now.

These experiments indicate that the great increase in suggest that the nut site RNA is a component of the antitermination process.

Using gene fusions, we have

These results

To further explore the role of RNA in N function, NusA, NusB, and N proteins were purified using a sensitive in vitro assay (2, 3) , and nutR RNA was purified using a complementary M13 probe.

Interactions between N, host factors, and & RNA will be presented. The specific interaction between left-handed 2 DNA sequences i n negatigely superc3iled 6x17; R F I DNA and anti-2 DNA IgG (2arling et. a[., 4'. No[. B i c ! . 1:1984) 176, 369-415) was investigated by high resolution darkfield immuno-EM. DNA-IgG complexes were tormed under optimal binding conditions, purified by column chromatographr, and visualized alter uranyl acetate staining without use of aldehyde fixation, shadowing, or second antibad>. Biialent anti-2 lqGs bound to R F I DNAs, thus forming intramolecular bridges. They csuld also oligomerize separate molecules by intermolecular linking of 2 sequences. llndcr low ionic and thermal conditions high affinity anti-2 Iqt was retained at certain loci even 3tter restriction endonuclease cleavage of the DNA. In these cleaved molecules some superhclices could be preserved in the loops generated by the bivalent IgG. This is the tirst example 3 f IgG stabilization of local superhelical strain in a cut moleculc. L seauenccs in hXli'4 k F 1 DNA were mapped. Alternating tracts of purines and pyrimidines at nucleotides 763, 1027. 1714, 2146, 2363, 3504, 4161. 4911, and 5345 occur within the 9 difrerent anti-?. IgG binding sites, which were expressed with varying frequencies (3-53X) on the molecules. Onl'i a limited number o f sites (generally < 2 ) exist on any one moleculc. The formation ot multiple 2 sites in a given molecule, at the extracted superhelix densitv, was probably non-cooperative due to Z DNA's relaxation of torsional stress. Ditterent Z sites can cccur within several different genes, including regions where transcription is attentuated and in one case was located in front of a prorotor of transcription. of free and protein bound DNA fragments, we have shown specific binding of gal repressor to each of the two operator DNA segments. The repressor binding protects the two operator DNA fragments from DNase I digestion. Each of the protected regions is about 24 bp long and covers the 16 bp homologous operator sequence. one of the two sites prevents repressor interaction with that operator.

We have studied the relationship of the two operators with the intervening DNA. We have found that the addition of a 15 bp DNA sequence, which do not affect the promoter activities, between 9 and 9 derepresses the operon. The implication of these and other results on the mechanism of gal repressor action will be discussed.

An operator constitutive mutation (gc) at 9 and 9 are separated by 100 bp and encompass the two gal promoters, 4 and 9 .

Irwin, Susan Varnum, and Robin P. Wharton, Department of Biochemistry and Molecular Biology, Harvard University, Cambridge, MA 02138 An alpha helix (the "recognition helix") can determine DNA binding specificity of a repressor. Thus, if this (presumed) helix of the 434 phage repressor is replaced (by gene manipulation) with the (presumed) analogous region of 434 cro protein or of P22 repressor, the DNA binding specificity of the hybrid protein is in both cases that of the parent that donated the alpha helix.

Using mutagenesis in vitro we have created a mutant of the CAP protein of E.coli that binds to DNA but is defective in stimulation of transcription of &, gal, and ! @ genes. When t h i s v e h i c l e is c u t b e t w e e n 3 a n d t h e p r o m o t e r w i t h =I1 a n d = I , 5 ' a n d 3 ' o v e r h a n g s a r e c r e a t e d ; we v i s u a l i z e t h i s a r r a n g e m e n t a s a n o t c h .

We In the past two years, we have examined the sequences around the start codon with regard to their role In the efficiency of protein initiation in E. coli. Using synthetic oligonucleotide primers, mutagenized in specific regions by simultaneous couplings of all nucleotides, we obtained two large collections of mutants that differ only at: (1) the three bases preceding the start codon of the a-galactosidase messenger and (2) the three bases following the start codon (i.e., the second triplet) of this messenger. In both of these groups, we found mutants that differ dramatically in the &gal levels.

In another approach to study the sequence requirements of ribosome binding sites, we developed a portable Shine-Dalgarno region (PSDR). In this system, the Shine-Dalgarno (SD) area of a given sequence was replaced by several other sequences. We have found that an increase in the length of complementarity with the 3'-end of the 165 rRNA (i.e., the anti-Shine-Dalgarno sequence) results in a decrease in translational efficiency. Using the same system, we also found that the sequence in the spacer region; i.e., the region between the SD-sequence and the start codon, affects the mRNA translatability; A's and T's favor the expression, C's lower the expression, and G's essentially abolish translatability. (Note that G residues are almost never found in natural messengers in the spacer region.) the anti-SD sequence (as found on the 16s rRNA) and the anti-SD-sequence on the 165 rRNA with the SD-sequence. The messenger in which the sequence 5' GGAGG as the SD-region was replaced by 5' CCUCC was completely inactive. We mutated (by site-directed M13-mutagenesis procedures) the anti-SD-region in a 16s rRNA gene of the plasmid pKK3535 from 5' CCUCC (its natural sequence) to 5' GGAGG. Upon reconstruction of the plasmid, transformants were obtained that contained large deletions in the plasmid borne 16s rRNA gene, showing that such a mutation is lethal. (Similar observations were made by A. Dahlberg's group at Brown University, personal comunication.) Currently, we are replacing the authentic ribosomal RNA promoters of pKK3535 by an inducible promoter in an attempt to prevent lethality.

Expression of ribosomal RNA genes by other than its own promoters may lead to premature transcriptional termination. For this purpose, we identified the precursor region of 16s rRNA as the region that gives RNA polymerase transcribing rrn-operons antitermination properties. This region contains a "so-called" box A sequence. Using M13 mutagenesis techniques, we made several mutations in and around this region and have demonstrated that the box A sequence is crucial for antitermination.

Finally, using the same PSDR system, we exchanged the SD sequence on the messenger with

Many of the peculiarities of translation in eukaryotes can be explained by postulatina that ribosomes scan the 5'-end of messenger RNA. Our working hypothesis is that a 405 ribosomal subunit binds initially at the 5'-end of m W A and then advances linearly until it reaches the first AUG triplet: if the first AUG codon occurs in an optimal sequence c~n t e x t l ,~'~ all 40s subunits stop there and that AUG serves as the unique site of initiation. Rut if the firstAUG triplet occurs in a suboptimal context, only some 405 subunits stop and initiate there; some bypass that site and initiate at another AUG codon that lies farther downstream. This ':modified scanning model" makes several predictions which I have evaluatedbyintroducing mutations around, and upstream from, the AUG initiator codon in a cloned preproinsulin gene. To monitor the effects of sequence variations near the initiator codon, SV4O-based plasmids that carry point mutations or small insertions were introduced into monkey (COS) cells and the yield of (prelproinsulin wasmeasured 48 hours later.

A number of parameters have been studied using this system: (a) Single nucleotide changes in position -3 (i.e., 3 nucleotides upstreamfromthe AUG codon) or position +4 were found to modulate the yield of proinsulin over a 15-fold range. The optimal nucleotides are A in position -3 and G in position +4; the contributions of those two positions are not simply additive. Single nucleotide changes in positions-1, -2, -4 or -5 have much smaller effects on translational efficiency. (b) An oligonucleotide has been inserted upstream from the AUG initiator codon that is complementary to a block of 13 nucleotides, includingthe AUG codon. The formation of a stable hairpin structure in which the AUG triplet is completely base paired does not preclude translation, implying that ribosomes can melt secondary structure barriers in mRNA. (c) The ability of eukaryotic ribosomes to reinitiate translation following a terminator codon4 has been confirmed. Theefficiency of reinitiation has been shown to depend on the position of the terminator codon relative to the second (downstream) initiator codon. (d) The possibility that triplets other than AUG can serve as initiator codons in higher eukaryotes was tested by introducing point mutations that changed the AUG codon to AUA,AUU, GUG or UUG. The nonstandard initiator codons that work, at least to some extent, in prokaryotes do not function to a detectable extent in higher eukaryotes. A systematic analysis of mutational alterations within the transcribed region of the CYCl gene has revealed certain essential features of the translational process in yeast. Altered DNA sequences were obtained by a variety of techniques, including the selection of mutations, revertants and recombinant i n vivo and the alteration of cloned DNA segments i n v i t r o . The following alterations have been extensively characterized: point mutations of the AUG initiator codon; small deletions and insertion in the 5' untranslated region; bp substitutions adjacent to the AUG initiator codon; relocation of the AUG initiator codon; and formation of sequences with two AUG codons. The levels of the CYCl mRNA and of the gene product, iso-1-cytochrome c , were determined in strains containing single copies of the altered CYCl genes at the normal chromosomal position. codon is required for effective initiation of translation; however, certain other codons can be used at below 1% of the normal efficiency, and these levels can be enhanced by suppressors unlinked to the CYCl locus. (ii) Deletions and AUG relocation indicate there is no requirement for specific sequences o r a ribosome binding site adjacent to the AUG initiator codon. (iii) Insertions and deletions indicate that the distance between the CAP sites and the AUG initiator codon can be varied without appreciably affecting the efficiency of translation. (iv) Hairpin and possibly other secondary structures in the mRNA can greatly diminish translation; the degree of inhibition is dependent on the strength of the hairpin structure and on the position of the AUG codon relative to the hairpin structure. (v) The translational efficiency is influenced by nucleotides at position -1 and -3; however, this context effect varies with the position of the AUG initiator codon along the mRNA. (vi) Protein synthesis initiates at the 5 ' proximal AUG codon in most sequences. Insertion of an upstream AUG codon can prevent OK diminish initiation from the normal downstream start site, and the efficiency is a function of the sequence context. This result implies that protein synthesis can initiate concomitantly at more than one AUG codon with special contexts. The e x p r e s s i o n of t h e endonuclease m i g h t t h u s be r e g ul a t e d a t t h e l e v e l of t r a n s l a t i o n i n i t i a t i o n by a t e r t i a r y s t r u c t u r e i n t e r a c t i o n between t h e r b s and t h e s i g n a l .

when t h e RNA c h a i n i s c o m p l e t e d , t h e r b s becomes masked and hence o n l y n a s c e n t mRNA's c a n be t r a n s l a t e d . Such t r a n s c r i p t i o n s , however, cannot be t r a n s l a t e d i n eukaryotic systems because they l a c k a 5'-tenninal cap s t r u c t u r e . capping o f the RNA. cloned cDNA through a combination o f prokaryotic t r a n s c r i p t i o n and eukaryotic t r a n s l a t i o n . The cDNA encoding the mouse-liver alcohol dehydrogenase (ADH-A subunit) has been cloned from a liver cDNA library, and its sequence has been determined. The cDNA contains 1 B R nucleotides of 5 ' non-translated sequence, the entire coding sequence of the ADH-A subunit, and 133 nucleotides of 3' non-translated sequence followed by a short poly(A).

This represents the first complete sequence of a cDNA for a mammalian alcohol dehydrogenase. The amino acid sequence deduced from this cDNA closely resembles that of the horse-liver ADV-E subunit: 316 of 374 residues are identical, and 29 of the differences are conservative substitutions.

The 5 ' end of this cDNA is interesting: the second AUG initiates synthesis of the ADH polypeptide.

It is preceded by a n t h a t could direct synthesis of a tri-peptide before reaching a termination codon.

The sequences surrounding both AUGs are similar. Kozak' s hypothesis that ribosomes initiate synthesis at the first AUC: and, after termination of the peptide, continue traveling along the RNA to reinitiate at the second AUG might explain the initiation of the ADH polypeptide.

Such a mechanism might influence the efficiency of translation of this mRNA. Several studies have been conducted to identify the target cell membrane determinants that binds to the effector cells. The present studies were aimed at the chemical nature of the receptor. Human Natural Killer (NK) cells and peripheral blood lymphocytes (PBL) from hepatitis B virus surface antigen vaccine (HBsAg) are cytotoxic (CTL) to human hepatocellular carcinoma (HHCC) .When incubated with NK or CTL, neuraminidase (VCN) increased the cytotoxicity of these cells by 10-14 folds. Whereas, if the target cells were treated with VCN, they became resistant to the killing effects. When HHCC are cultured in presence of non-toxic levels (0.3 to 1.0 ug/ml) of tunicamycin (Tn), a potent inhibitor of glycosylation, the cells resist the killing of target cells from HHCC and glycosylation-deficient-HHCC, and were used in an in vitro mRNA-protein synthesis system. On Oligo(dT) columns the two types of mRRA differed in elution patterns. The protein coded by mRNA from HHCC enhanced, whereas the protein coded by mRNA from glycosylationdeficient HHCC had no effect on NK or CTL cytotoxic activities. When applied on SDS-polyacrylamide gel electrophoresis, the protein coded by mRNA from IIIICC resembled. whereas that coded by mRNA from glycosylation-deficient cells differed from HBsAg. Therefore, glycosylation of the cell membrane alters mRNA translation products.

ffects of NK and CTL cells. Both HfCC and glycosylation-deficient fIIiCC were labelled with s H-N-Acetylglucosamine. Using Ficoll-Hypaque gradient centrifugation, both types bound to the effector cells, but only W-ICC were lysed. mRNA was isolated 0975 Several bacterial plasmid vectors (pBR322 derivatives) have been constructed in an effort to express human prosomatostatin in bacterial cells. These expression vectors were designed to place the human prosomatostatin cDNA sequence under the regulation of the bacterial tryptophan promoter. Probing of cellular fractions for the prosomatostatin peptide (11,500 MU) and mRNA from cells transformed with initial plasmid constructs indicated very high levels of somatostatin specific mRNA but no prosomatostatin peptide. Northern blots revealed the somatostatin specific transcripts (300-400 bases) were of sufficient length to encode the complete precursor peptide although apparently terminating in the 3' untranslated region of the eukaryotic sequence (present in all constructs).

A computer analysis revealed terminator-like sequences in this 3' untranslated region similar to those known to terminate beta-lactamase and tetracyclineresistance genes also present on pBR322. Deletion of the 3' untranslated region and reintroduction into bacterial cells of this otherwise unmodified vector resulted in significant levels of prosomatostatin peptide expression. Analysis of RNA from these cells indicated the presence of specific transcripts two to three times the size necessary to code for the 9 2 amino acid precursor peptide. The results suggest that bacterial expression of this relatively small eukaryotic peptide requires a minimum size mRNA transcript in order to be efficiently translated by the bacterial ribosome. The 8-lactamase I gene from B. cereus NilOR, a strain isolated from stream mud was found to be located on a 8 . 3 kb EcoRI fragment, as opposed to a previously reported 4.3 kb fragment in strain 569/H. Restriction analysis indicates that the flanking regions are different in these two strains. Sequence data from the 8-lactamase genes show differences both in the Shine-Delgarno region, leader sequence and structural gene. A series of secretion vectors was constructed using the leader sequence of 8-lactamase genes. We are now comparing their function according to the sequence variation in S-D and leader region.

Dennis E. Hruby, and Christine A. Franke, Center for Gene Research, Department of Microblology, Oregon State University, Corvallis, OR 97331 The use of cloning and expression vectors to study and experimentally manipulate individual genes, independent of their normal resident environment, is a central and vital theme in modern molecular genetic experimentation. h e to a number of unique biological attributes, vaccinia virus (W) would seem to offer an ideal system for such studies. A number of laboratories have demonstrated the feasibility of this approach by constructing recombinant W strains which contain and express heterologous viral antigens. Such hybrid vaccine strains may prove useful against a variety of human and animal diseases. Unfortunately, the current methodologies employed to construct recombinant W are slow, time-consuming, expensive, and do not facilitate genetic engineering of the foreign insert. These drawbacks have thus far retarded the development of W as a generalized eukaryotic expression vector. employed to construct chimeric genes and to assay their biological activities; 2) To construct insertion plasmids containing dominant selectable markers which will allow direct, one-step, selection of W recombinants containing foreign inserts; and 3) To use W as a research tool with which to study gene systems that are not readily amenable to more conventional approaches -RNA viruses (animal and plant), histocompatibility antigens, and immunoglobulin genes.

Experiments currently in progress have three objectives: 1) To streamline the methods The cleavage products accumulate with the mutant infections but are religated with wild type. Some revertants of p r & or *phage, with a second mutation in a, lack the anticodon nuclease. Other pnk revertants lack in addition an endonuclease that cleaves the host leucine tRNAl, suggesting the existence of a phage-coded factor common to both enzymes. a transduced host strain carrying the restrictive E. cot i CTr5x locus prr (Abdul Jabbar and Snyder (19841, J. Virol 51, 522), the anticodo! nuclease reaction products appear transiently during infection, accumulate with pnk and Tljmutants and are absent with fimutants. T4e genomes of the restrictive donor and transductant bacteria, but not of the permissive recipient, contain a common DNA restriction fragment that hybridizes to a CTr5x-specific tRNA fragment probe, suggesting that encodes tRNA species vulnerable to the anticodon nuclease. Regarding the restriction mechanism, it is proposed that the anticodon nuclease reaction products inhibit late phage gene expression, unless further processed by polynucleotide kinase and RNA I igase.

In E. cot i BJMnlO, a i r a w a , J. ( 1 9 8 4 ) . C e l l s, [861] [862] [863] [864] [865] [866] [867] [868] [869] [870] .

t r a n s c r i p t p a i r i n g 13 rLttp. and n e g a t i v e r e g o l a t i o n 13 1zFzp. Ihele ind other erperimenta support our working model. Cecilia M . Arraiano. Stephanie D . Y a n r e y . a n d William P . Donovan, Univ of Georgia. A t h e n s , G A 30602 E. coli m R N A s a r e rapidly d c g r a d e d to mononucleotidcs u i t h an aLeragc half life of 60-90 ? e c o X s . Although p r e v i o u s w o r k h a s s u g g e s t e d t h e involvement of a number of r i b o n ucleases s u c h a s ribonuclease I1 a n d polvnucleotide phosphorylase in m R N A t u r n o v e r , until recently it hds not been possible to demonstrate a n y clear chdnge in t h e degraddtion p a t t e r n o f mRUAs. T h e finding t h a t (polynucleotide phosphorylase) r & (RNase 11) double m ut a n t s could not be c o n s t r u c t e d b y P I transduction s u g g e s t e d a means of examining R N A metabolism in E_. e. Using in v i t r o mutdgenesis of t h e cloned r n b s t r u c t u r a l g e n e , i t hds been possible to isolate t e m p e r a t u r e sensitive r n b mutations. allele ( r n b -500) a n d a n absolute m u t a t i o n x t h c s t r u c t u r a l gene for polynucleotide phosphoryldse (=arc conditionally lethal for growth dnd accumulate partially d e g r a d e d m R N A species at t h e nonpermissive temperature.

T h e s c partially d e g r a d e d m R U A species r a n g e in S I L C fro-n 20-500 nucleotides. I n viability IS directly correlated with t h e inability to totally d e g r a d e m R N A species. An apparently autoregulatory pathway determines the level of new tubulin synthesis in virtually all animal cells. Increase in the intracellular pool of free tubulin subunits results in dramatically lowered levels of tubulin mRNAs. To understand the molecular basis for this regulation, we have transiently introduced a cloned tubulin gene into cultured niouse fibroblasts and demonstrated that the heterologous gene is transcribed and correctly processed into stable mRNA. 1,loreover. using colchicine to induce depolymerization of endogenous microtubules and a corresponding elevation of the pool o f free tubulin subunits, we have shown that the expression of the heterologous tubulin gene is suppressed concomitantly with that of the endogenous mouse tubulin genes, nowever, such downregulation is not observed following transfection of a hybrid actin gene which is transcribed under the control of a tubulin proniotor. This observation, together with our previous findings that tubulin gene transcription in nuclei from control and colchicinetreated cells is not subject to down-regulation. strongly suggests that the autoregulation of tubulin synthesis i s not modulated on a transcriptional level. Results with other recombinant gene constructs are consistent with this hypothesis and serve to define more closely the genetic region(s) which contains the regulatory signal. Using two independent approaches we demonstrate t h a t cDNAs i n t h e l a t t e r group remain attached t o t h e mRNA d u r i n g t r a n s l a t i o n .

These r e s u l t s imply t h a t t h e ribosomal complex, once f u l l y assembled a t t h e AUG i n i t i a t i o n codon can l o c a l l y d e s t a b i l i z e secondary s t r u c t u r e s as i t moves along t h e mRNA. This a c t i v i t y may be c r i t i c a l f o r t h e t r a n s l a t i o n e l o n g a t i o n r e a c t i o n .

Mycoplasmas are genome-limited organisms: their 500 or 1000 Mdalton genomes limit the genetic complexity of these cells. G+C) puts additional constraints on their information storage and codon usage. shown that mycoplasmas arose by degenerative evolution from Gram-positive eubacteria: the initial phylogenetic branch produced mycoplasmas with 1000 Mdalton genomes (hence, cell wall loss and genome reduction were probably coupled), and subsequent branchings gave rise to mycoplasmas with 500 Mdalton genomes. Recent DNA sequence studies show that mycoplasma ribosomal protein cistrons make maximal use of codons rich in A and T, and 16s-23s rRNA spacer DNA is noncoding and rich in A and T. of G+C than eubacterial rRNAs, but are still close to 50% G+C. We have identified Shine-Delgarno sequences near the 3'-terminus of mycoplasma 16s rRNA, but no data are available on promoter or Shine-Delgarno sequences upstream from DNA coding regions. structure and regulation in mycoplasmas, we are studying mycoplasma virus L2: a temperate noncytocidal phage, containing 11.8 kb circular DS DNA. We used L2 DNA as template in & vitro E. coli and --B. subtilis coupled transciption-translation systems. Both systems produce most virion proteins, plus several other different proteins. These data will be discussed and the two in vitro systems compared, both for L2 and L51 (a mycoplasma virus containing 4.5 kb SS DNA). sequences are being described by sequencing the L2 genome, and will be discussed. Large amounts of two small (160 nuoleotide) RNAs, the VA RNAs, accumulate in the cytoplasm of adonovirus-infected cells.

A mutant virns unable to produce the major species, VA RNAI, is deficient in the translation of both viral and cellular m R N A s .

The defect occurs at the level of polypeptide chain initiation and i,s corrected in sigtn by addition of the initiation factor eIF-2 or its recycling factor GEF: none of the other initiation and elongation factors tested are effective. Furthermore. a s s a y s for GEP show that the activity of this factor is severely depressed in mutant-infected Cells. Two lines of evidence suggest that the role of VA RNA is an indirect one: addition of this PNA fails to rescue initiation activity, and mixing experiments reveal the presence of a translational inhibitor in motant-infected cells.

This inhibitor appears to be a protein kinace capable of phosphorylating the a subunit of SIP-2, thereby trapping GEF and Preventing the recycling of eIF-2 in its catalytic role. Identification is underway of the VA RNAI sequences important for its function.

&. J. Biol. Chem. 259, 8648, 1984) . Both of these factors, eIF-4A and eIF-4F. seem to share a common peptide> 46,000 daltons, although identity has not been proven yet. However, the specificity of these two factors is quite different. eIF-4A is an active ATPase in the presence of all four ribohomopolymers although poly(G) works poorly. eIF-41F is quite dependent on the presence of eIF-4B for ATPase activity and then displays an m GDP sensitive preference for globin W A . Neither protein is an effective ATPase in the presence of poly(dA), poly(dT) or oligo(dT) .

Current studies are intended to extend the biochemical studies into the area of protein chemistry to identify regions of similar sequence such as the peptide site responsible for reacting with the affinity label flourosulfonylbenzoyl adenosine (FSBA) and by double label peptide mapping of the 46,000 dalton peptides of eIF-4A and eIF-4F. At present, very preliminary data indicate a high degree of similarity, but non-identity of these two peptides. Supported in part by NIH Grant GM 26796.

G. Theodorakis and Sunandita S. Ranerji, nept. of Biochemistry, Yolecular and Cell Biology "orthwestern University. Rvanston. Illinois 60201 of one heat shock protein, HSP70, and the repression of normal cellular proteins such as modified and is apparently blocked in vivo at the level of initiation. The preferential synthesis of HSP70 following heat shock is not due to an increase in HSP7cI mRNA as the level of messenger RhJA increases only two-fold following heat shock while the level of HSP70 synthesis increases over twenty-fold, We find that HSP70 mRNA is maintained in the cytoplasm of normal red cells in a translationally repressed state that can be activated following heat shock. Over 707 of the HSP7O mRNA is associated with polysomes in its translationally repressed state, and can be released from ribosomes by treatment with EnTA. Furthermore, we find that HSP7n mRNA is associated with. XW particles with the density expected of polysomes on metrizamide gradients. Our data suggests that the control of MSP70 synthesis in avian reticulocytes is regulated at the level of elongation of protein synthesis, 

In most cases, regulation of eukaryotic gene expression at the translational level probably occurs by regulation of mRNA association with ribosomes. Several protein factors have been identified as being necessary for the binding of mRNA to 40s ribosomal subunits. However, very little is known about how these proteins act during messenger binding and how their activities might be regulated. One Of the initiation factors required for mRNA binding is eIF4A, a single polypeptide of 43kd. Interestingly, it is required as a single polypeptide for mRNA binding, and is present, along with several other proteins, in a high molecular weight complex involved in cap recognition. It also has the ability to discriminate between different mRNAs, that is, it stimulates the translation of different messengers to different extents. In order to better understand the function and possible regulation of eIF4A. we have isolated mouse cDNA clones coding for this factor. We describe the isolation and characterization (including the nucleotide sequence) of these clones. We have used these clones to examine the expression of mRNA coding for eIF4A and present evidence suggesting the existence of two different messengers coding for eIF4A. 

Simonsen and A r t h u r D. Levinson. Genentech, Inc., 460 P t . San Bruno Blvd., South San Francisco, CA 94080.

We have studied t h e consequences o f i n s e r t i n g ATG t r i p l e t s i n a l l t h r e e reading frames upstream o f t h e t r a n s l a t i o n a l i n i t i a t i o n codon o f t h e H e p a t i t i s B v i r u s surface antigen (HBsAg) gene. As expected, these a d d i t i o n a l ATG codons can severely depress t h e i n i t i a t i o n o f t r a n s l a t i o n a t t h e a u t h e n t i c s t a r t codon. Such i n h i b i t i o n , however, can be t o t a l l y suppressed by t h e presence o f in-frame t r a n s l a t i o n a l stop codons f o l l o w i n g t h e upstream ATG codon. We have p o s i t i o n e d t r a n s l a t i o n a l stop codons a t various p o i n t s a f t e r t h e upstream t r a n s l a t i o n a l s t a r t codon i n o r d e r t o ask whether t h e r a t e o f t r a n s l a t i o n a l i n i t i a t i o n a t t h e a u t h e n t i c HBsAg s t a r t codon i s a f f e c t e d by t h e distance between t h e t r a n s l a t i o n a l stop codon o f t h e u p s t r HBsAg gene.

Imnunoprecipitation o f Sgg-labeled i n t r a c e l l u l a r e x t r a c t s prepared from c e l l s t r a n s f e c t e d w i t h vectors having t h e upstream AUG i n t h e same t r a n s l a t i o n a l r e a d i n g frame as t h e HBsAg gene demonstrated t h a t t h e upstream ATG codons are e f f i c i e n t l y recognized. i n t e r n a l ATG codons a f t e r having terminated t r a n s l a t i o n from an upstream reading frame. The sequence requirement at the circularization site has been studied by two different approaches. Addition of certain short oligonucleotides, containing free 3' hydroxyl groups, vill reopen the circular form of the IVS, yielding a linear IVS with the oligonucleotide covalently attached to its 5' end. The product of this reaction will recircularize with release of the added oligonucleotide, indicating two nucleotides 5' to the site of circularization are sufficient for circularization. By varying the sequence of the added oligonucleotides in the reaction, it was found that certain combinations of pyrimidine dimers and trimers work best for reopening the IVS. Relative reactivity of oligonucleotides is sequence dependent: CU works better than W, which works approximately as well as CC; UUU works better than UU. These data support the existence of an oligopyrimidine binding site within the IVS.

In a second approach, the nucleotide sequence at the site of circularization was altered using recombinant DNA techniques. RNA was transcribed in vltro from a plasmid containing a single BBlnl linker (CGGATCCG) inserted in the IVS at the primary circularization site. The IVS was found to excise and to circularize to the UCC sequence within the linker. Deletions at this site, generated by S1 nuclease treatment of BasHI-cut plasaid, shifted the preferred sites of circularization either to the secondary site or to cryptic sites. The presence of three BamHIlinkers at this site displaced the major site of circularization out of the linker sequences to the secondary site, presumably because the repeated linker sequences form a sterloop structure. The major site of circularization in each of s i x constructs was immediately preceeded by three pyrimidines, suggesting the existence of a binding site for a tripyrimidine sequence, but position and MA secondary structure also appear to be important in the selection of circularization sites. The RNase P of Bacillus subtilis consists of protein (17K mol. wt.) and RNA (395 nucleotides) elements, both required for cleavage of 5' precursor segments from tRNAs under "physiological" conditions. P RNA alone carries out the reaction with perfect fidelity. The high cation concentrations probably in part provide counterion shielding to overcome anionic repulsion between the two interacting polynucleotides, the RNase P RNA and the precursor tRNA. However, the character of the ion dependence, inhibition of the reaction by high SO4-2 concentration and potentiation of the reaction by solvents (ethanol, DMSO) suggest that RNA conformational transition is involved in the reaction. It may be that the reason for catalysis by RNA rather than protein in the RNase P reaction is a requirement for fluidity in the structure of the catalyst, so that it can accomodate many tRNA substrates, which vary in their structural details.

The RNase P and Tetrahymena rRNA self-splicing reactions both involve phosphodiester bond scissions, but they seem mechanistically distinct. Whereas the nucleolytic self-splicing reactions require a 3'-OH group, destruction of the RNase P RNA and substrate tRNA 3'-OH groups by periodate does not result in diminution of the reaction. Some understanding of the RNase P higher order structure is imperative for future progress on the problem. This best will drive from phylogenetic comparisons, seeking common foldings among homologous RNase P RNAs with different sequences. The sequences of the RNase P RNAs of B. subtilis (395 NT) and E. coli (377 NT) are known. Even though the RNAs utilize the proteins from the other species, the RNA sequence homologies are remarkably low, ca. 40 percent in the "best" alignment. The two sequences may be credibly aligned for secondary structure comparisons over only about 25 percent of their lengths, however. In those regions, common folding is evident. The two compared sequences seem to vary by addition or deletion of short helical domains. RNase P RNAs from other organisms currently are under sequence analysis.

In the presence of high cation concentrations (NH4+>1M, Mg+'>O 

The splicing of mRNA precursors in HeLa cell extracts proceeds in two steps ( 1 ) .

In the first step, a cleavage occurs at the 5' splice site and the 5' end of the intervening sequence is joined to the 2' position of an adenosine residue near the 3 ' splice site to produce a lariat structure. specific complex.

intervening sequence and the 3' exon is cleaved at the 3 ' splice site and the 5' and 3' exons are joined by a 3'-5' phosphodiester bond. The intervening sequence is released as a lariat RNA. These reactions conserve the number of phosphate bonds in the precursor RNA and are consistent with a coupled transesterification mechanism. Although the reaction requires ATP, both the newly formed 3'-5' and 2'-5' phosphodiester bonds involve phosphate groups from the precursor RNA.

In addition to the complementarity observed between the sequences at the 5' splice site and sequences at the 5' end of Ill RNA, there is a striking complementarity between the same sequences at the 5' splice site and sequences around the site of branching. This complementarity may help determine the specificity of the splicing reaction by bringing the 5' splice site into close proximity with the branch site.

The two RNAs produced by this reaction are held together in a In the second step of the reaction, the lariat RNA containing the (1) Padgett, R.A., Konarska, M.M., Grabowski, P.J., Hardy, S.F., and Sharp, P.A. (1984) Science 225, 898-903.

Christie ', Elisabeth Ljungquist', Rzkert V q i n i a The l a t e genes of temprate coliphage P2 are organized into 4 transcription units. Expression of these genes during P2 infection depends on P2 DNA replication, E. culi RNA polymerase, ard a psitive regulatory protein e n d e d by the P2 from which ~2 l a t e mmAs are transcribed share a cOnSenSus sequence a t -10 and -35 which is different frcm that normally recognized by the host RGi plymerase. ?he= pramters do not possess obvious repeated or s p x ? t r i c a l sequences that muld suggest a specific birding site for a p s i t i v e reydatory protein.

tion of P2 late transcription by the cgr gene product, we have determined the nucleotide sequence of the ogr gene and UndertakETanaLysis of ogr protein. Employing specifically engineered plasmids in which the expression of E. coli& cistron is regulated by transcription termination, we have analyzed transcription antitermination mediated by phage lambda N protein in vivo and in vitro. Antitermination requires the presence of a nut site, the N-recognition element, between the promoter and the terminator. It also requires the direct participation of three cellular proteins NusA, NusB and S10 in roughly stoichiometric amounts. S10 is a normal component of the ribosome. In addition to S10, ribosomes contain active N, NusA and NusB proteins. It seems likely that N-action in the cell involves the ribosome. However, N-action need not involve the translational coupling phenomenon since translation upstream, across and downstream of the nut locus does not play a necessary role in antitermination. We propose that a specific interaction of the N-ribosome complex with RNA polymerase-= RNA complex leads to the formation of the antitermination apparatus. to a specific DNA site in the promoters of these genes. To study how CRP acts, we have cloned, into a bacteriophage lambda vector, the gene ( 3' which are sites at which 511 protein binds to activate wild-type ERE, flank the wild-type 

We find that the Tantip thding sites at the o w of replication a n be remved vithmt loss of tranwxtivatim. Metim a€ the a;rich 21 bp repeat r e g i m tames a Sfold decreese in T antigen stirmlated gene e x p r e s 5 m in plasdsrhere the 72 bp repeat w a r e not iresat: haever. tkn the 72 bp repeat regims are induded. the deletim of the 21 bp repeats has no effect m activated 21 bp repeats m y cmtain a raek pmmter e l m t d-ich is superseded in the bp r e p t regim. 'he mjor pmmter elenarts needed for regim. We have evidcxe of at lmst tmelawts in this regim; are isvithina 33bp - The :ilanine tRNAsAf;om silkworms (Bombyx mori) accumulnte in d tissue-specif ic fashion. One species (tRNA ) riuced constitutively in all silkworm tissueb, while n second major specyes ( : ; N : ' ' ' differences in the transcriptional properties of the genes encoding these RNAs can explain the tissue distribution of alanine tRNA. To that end, we have examined the transcriptional activities ot representatives of these two classcs o f gene5 in vitro. The two genes behave very differently in homologous transcription extracts prepnred from Bombyx mori silkglands. tional differences, we have constructed tRNAC ,itRNAsG, hybrid genes. sequences upstream from the transcription initiation site determine whether the gene behaves like a constitutive type o r silkgland-specific type gene. We are currently using partially purified componen? : : of the silkworm transcription apparCirus to learn which cornponent(s) interact differentially with the two kinds of genes. Most gene transcribed by RNA polymerase I1 contain the hexanucleotide, 5'-AAUAAA-3', just upstream from the polyadenylation site. We have examined the signal requirements for processing and polyadenylation by using derivatives of the herpes simplex virus thymidine kinase gene (HSV-tk). Three lines of evidence suggest that the hexanucleotide alone is sufficient to signal processing and polyadenylation: (i) The HSV-tk gene was resected from its 3' end. All derivatives which retained the AAUAAA were able to produce poly A+ tk mRNA in transfected Cos-1 cells. The HSV-tk gene contains two copies of the AAUAAA. Resection with Bal 31 nuclease was used to generate deletion derivatives of the tk gene. Derivatives retaining at least 50 bp of information distal to these AAUAAAs produced near wildtype levels of tk mRNA. Retention of fewer than 40 nucleotides distal t o the AAUAAAs resulted in a reduced level of poly A+ tk mRNA. (ii) Poly A+ tk mRNA was produced when the HSV-tk polyadenylation signal was replaced with an 88 bp fragment of SV40 DNA, containing an AAUAAA that is never used as a polyadenylation signal during SV40 infection. (iii) Random fragments of either simian (4.0-15 kb) or prokaryotic (0.2-1.5 kb) DNA were inserted into an HSV-tk derivative lacking a polyadenylation signal. Poly A+ tk mRNA production was restored by 83% of the simian and 92% of the prokaryotic DNA fragments. Together, these results suggest that the minimum signal necessary for processing/polyadenylation is a hexanucleotide (AAUAAA o r similar) and that additional sequences in the vicinity of the AAUAAA affect the efficiency of the reaction, probably by affecting the secondary structure of the precursor RNA. The expression of YGlOeYG103 has been analyzed mst extensively. the optimal growth tenperature of 30°C, their regulation differs. The abundance of both YGlOl and YG103 transcripts decreases upon an increase in tenperature f m n 23OC to 3loC. Hacever. the abundance of YGlOO transcripts increases after a tenperature shift, while that of YG102 changes only slightly. Using the method of gene replacenent, strains containing nutations in the four genes have been constructed. Although no phenotype of any of the single nutants as detected, haploid strains containing both the YGlOO and YG102 nutations nere not able to fonn colonies at 37OC. indicate that YGlOO or YG102 protein prOduct is needed to obtain n o m l growth rates at all tenperatures, but is essential at higher tenperatures. properties at all tenperatures. The cells grow fastest at 37OC; the severity of the effect increases with decreasing tenperature, although colony foming ability is retained even at la, tenperatures. At 37OC the YGlOl YG103 double nutant gmvs 22% slarer than wild-type; at 23OC it g m s 116% slarer. suggest that the Hsp70 nultigene fanily of yeast contains ~enes which have specialized but similar functions. Sane of the gene products (YGlOO and YG102) are required for sustained growth at high tenperature and others (YG101 and YG103) at low tenperatures. is unstable in vivo, and has a physical half-life of about 5 min. Because it is rapidly turned over, changes in the rate of synthesis of a3' will rapidly lead to changes in its intracellular concentration. amount of might be increased at high temperature, we have found that in wild type strains the amount of a32 mRNA increases after a temperature upshift. the increase is sufficient to cause the rapid increase in synthesis of hsps.

Overproduceda32 migrates as 2 spots in the isoelectric dimension of 2D gels while purified a32 migrates as only one spot. a32 was purified based on its activity. Perhaps only one isoelectric form of a32 is active. of a32 could provide a direct mechanism for both induction and shut-off of the heat shock response.

The transcriptional regulator of heat shock gene expression is the rpoH (htpR) gene . This gene has been cloned and sequenced (21. Its protein product has been identified on two-D gels as cellular protein F33.4. Its predicted structure possesses 439 identity or conservative replacement of amino acid residues with the carboxy-terminal half of RNA polymerase sigma factor (3). Protein F33.4 purifies with RNA polymerase, and purified preparations direct the initiation of transcription from heat-shock promoters without requiring normal sigma factor. On this evidence F33.4 has been designated sigma-32 (the normal sigma, sigma-70) ( 4 ) .

At temperatures above 45 C part of the differential induction of heat-shock proteins is the result of suppression of expression of most other genes, but the sudden, high rates of heat-shock protein synthesis upon shifts to the 37 -45 degree range must involve a pronounced absolute activation of heat-shock transcription (1). How transcription of heat-shock promoters by sigma-32 is activated at high temperature is not known. Our recent work is directed at this process. of a large number of agents known, or claimed, to induce the heat-shock response in other organismsincluding various oxidants, DNA replication inhibitors, and damagers of DNA or membrane integrity. Also, we have examiaed the synthesis of various dinucleotides during treatment with heat and these other agents. Our results indicate that ethanol is the only agent that closely mimics high temperature in inducing the heat-shock regulon in E. &, and that there appear to be several, perhaps related, stress regulons inducible in this organism by various toxic substances. Double-stranded DNA (dsDNA) has been found to induce the transfer of phosphate from ATP to several proteins in extracts of widely divergent eukaryotic cells.

Rabbit reticulocyte lysates, extracts of HeLa cells, XenoDus (frog) egg and oocyte extracts, Arbacia (sea urchin) egg and oocyte extracts, and SDisula (surf clam) oocyte extracts all show dsDNAdependent protein phosphorylations.

The mechanism is specific for dsDNA and will not respond to either RNA or single-stranded DNA.

The minimum sized dsDNA fragment capable of inducing phosphorylation in HeLa extracts is about 14 base-pairs. One of the proteins phosphorylated in reticulocyte lysate, HeLa, and XenoDus egg extracts has a molecular weight of 90,000 and has been identified as a heat shock protein (hsp90). HeLa extracts and reticulocyte lysates possess kinase activities that bind to DNA-cellulose; DNA-cellulose depleted extracts no longer phosphorylated hsp90 in response to DNA. The pattern of dsDN&-independent phosphorylations is not significantly altered by DNA-cellulose depletion of extracts. The h y p o t h e s i s t h a t t r a n s c r i p t i o n a l r e g u l a t o r y sequences a r e l o c a t e d w i t h i n t h e coding r e g i o n s of a v i a n sarcoma v i r u s e s (ASVs), a s w e l l as i n t h e i r l o n g t e r m i n a l r e p e a t s , i s b e i n g i n v e s t i g a t e d . A GTGGTTTG sequence, matching t h e consensus c o r e sequence observed i n many e n h a n c e r s , i s p r e s e n t i n t h e gag-coding r e g i o n of many ASVs, approximately 900 nucleo- To i n v e s t i g a t e the mechanism o f t r a n s c r i p t i o n a l a c t i v a t i o n , we have developed a h i g h copy plasmid system containing the e n t i r e s t r u c t u r a l and r e g u l a t o r y sequences o f the a c i d phosphatase ( and James W. Casey, Section of Genetics, NCI. Frederick, MD 21701 Bovine leukemia v i r u s expression is highly restricted i n vivo and i n vitro. To examine the molecular basis for the control of BLV expression, we excised BLV LTRs from cloned proviruses and fused them to the bacterial chloramphenicol acetyltransferase (CAT) gene. Plasmids carrying the CAT gene controlled by the BLV LTR or LTR fragments were introduced into a variety of c e l l s . The BLV LTR was an inactive promoter i n a l l c e l l l i n e s tested except those previously established a s BLV producers. I n FLK-BLV or BLV-bat2 c e l l s , the BLV LTR directed high l e v e l s of CAT expression. Deletion mapping experiments revealed that removal of LTR sequences located between 100bp and 170bp upstream of the RNA s t a r t s i t e reduced CAT expression by 90%. A 75bp DNA fragment encompassing t h i s region was cloned into pSV Ecat ( a pSV2cat derivative lacking SV40 72bp repeats and requiring the insertion of "enhancers" for CAT expression). The resulting p l a m i d s , p E75cat. directed CAT expression only i n the BLV producer c e l l l i n e s , indicating that the 75bp fragment contains the BLV c e l l type-specific enhancer element. Surprisingly, deletion of LTR sequences on the 3'-side of the RNA s t a r t s i t e caused an 87% reduction i n CAT expression. Whether t h i s r e l f e c t s an effect on transcription or post-transcriptional stages of gene expression i s currently being investigated. T h u s , the c e l l type-specific promoter a c t i v i t y of the BLV LTR probably r e s u l t s from an interaction of unique transcriptional f a c t o r ( s ) , present i n BLV producer c e l l s , w i t h the BLV enhancer element(s1. The SV40 enhancer contains two 8 bp s t r e t c h e s of a l t e r n a t i n g purines and pyrimidines t h a t have been implicated i n t r a n s c r i p t i o n a l a c t i v a t i o n (Nordheim 6 Rich, Nature 303) 674). W e investigated the functional r o l e of these sequences within a SV40 enhancer t h a t contained only one copy of the 72 bp element; t h e p a t t e r n of a l t e r n a t i n g purines snd pyrimidines was destroyed by making two transversion point mutations i n each 8 bp segment.

This mutant enhancer i s four-t o six-fold l e s s a c t i v e than the wild type enhancer i n a t r a n s i e n t Bela c e l l expression assay when linked t o t h e human B globin gene. point mutations 8180 grows very poorly i n CV-1 c e l l s ; r e v e r t a n t s with improved growth p o t e n t i a l could be obtained. however, a f t e r passage of mutant v i r u s stocks i n CV-1 c e l l s . Nucleotide sequence a n a l y s i s of 1 8 plaque-purified r e v e r t a n t s showed t h a t each r e v e r t a n t carried 8 tandem duplication of the mutated enhancer region, but t h e mutations were always preserved even when duplicated.

The r e v e r t a n t phenotype and improved enhancer tunction were both shorn t o r e s u l t from these duplications. Although the duplicated regions range from 45 t o 135 bp i n length and the sequences contained i n each dupliCation vary, t h e r e i s a 15 bp sequence. TCTGGAAAGTCCCCA. t h a t i s c o n s i s t e n t l y duplicated i n each reactivated enhancer. This region contains the 'core' sequence, GTGGAAAG, t h a t was i n i t i a l l y suggested (Lainins +., P U S 7 9 , 6453) t o be functionally s i g n i f i c a n t because r e l a t e d sequences were i d e n t i f i e d i n a v a r i e t y of v i r a l enhancers. Our r e s u l t s suggest t h a t the 'core' and surrounding sequences a r e a &-acting element t h a t can function independently of and compensate f o r the a l t e r n a t i n g purine and pyrimidine sequences. We are investigating the regulation of hybrid globin genes during the differentiation of murine erythroleukemia (MEL) cells. We have theref ore constructed three hybrid globin genes consisting of 1.4 kb of 5' flanking sequences from the mouse B-major gene along with its first and most of its second exon, linked to 3' sequences from me of three goat B-like genes: BC (early adult), y (fetal) or En (embryonic). Following the insertion of these constructs via transformation into a tk-MEL cell line we induced the transformants with DMSO. Three cut of the eight transformants containing the hybrid BC gene showed induced mRNA levels of 30 to 200 copies per cell with no Constitutive mRNA expressicn. One showed down-regulatim upon inductim and the rest showed no signal. Fcur out of nine transformants containing the hybrid y gene showed induced mRNA levels of 500 to 3000 copies per cell. One had constitutive expressicn and the rest showed no signal. We are currently analyzing the hybrid ED transformants; preliminary experiments with 15 different transformants, all of which induce their endogenous B-glcbin genes, showed no expression of the En hybrid before or after induction. W e a r e currently determining (i) the copy number of the Zn hybrid genes and (ii) the expression of a control mcuse-human hybrid gene in these transformants. The fact that the Bc and y hybrids do induce, but the En hybrid apparently does not, provides further evidence for the control of B-glcbin expressicn by internal sequences. (Recall that all three genes have the mcuse 3 major promoter). i t is worth noting that the goat y gene is much more homologous with goat adult enes than with goat embryonic B-like enes. Thus the inductim of the mcuse-oat hybrid is "ofnecessarily surprising, even though ME& cells do no{ induce their endogenous feta5 6-lixe genes. One of the e f f e c t s of temperature e l e v a t i o n above 35OC i s the s e l e c t i v e t r a n s l a t i o n of heat shock mRNA. t r a n s l a t e d a t h i g h temperature provided i t i s synthesized during the heat shock. A f u s i o n construct between the alcohol dehydrogenase (adh) gene and the promoter o f a heat shock gene was introduced i n t o adh-f l i e s . Such transformed f l i e s s y n t h e s i z e adh mRNA during heat shock i n massive mounts but t r a n s l a t i o n of t h i s mRNA i s r e s t r i c t e d t o non heat shock temperatures. Thus, t h e information t h a t allows the t r a n s l a t i o n machinery t o d i s c r i m i n a t e between heat shock and non heat shock mRNA must r e s i d e i n t h e primary sequence o f the heat shock mRNA.

shock gene were performed i n o r d e r t o i d e n t i f y such a sequence element. leader sequence are dispensable f o r e f f i c i e n t t r a n s l a t i o n . from t h e cap r e s u l t s in a d r a s t i c drop in expression both a t high and low temperature.

W e t e s t e d whether the t r a n s c r i p t o f a non heat shock gene i s being I n t e r n a l d e l e t i o n s of d i f f e r e n t p a r t s of the non t r a n s l a t e d leader region of a heat Over 80% of the Deletion t o w i t h i n 25 bases 1049 M y c u r r e n t i n t e r e s t i s t o know how t h e expression of genes coding f o r p r o t e i n s are regulated i n e u k a r y o t i c c e l l s . There i s r e l a t i v e l y much information on t h e DNA regions which may cont r o l t h e t r a n s c r i p t i o n l e v e l s , while l i t t l e i s known about p r o t e i n f a c t o r which might i n t e ra c t t o the DNA r e g u l a t o r y region t o r e g u l a t e t h e i r expression. There a r e few works t o search such r e g u l a t o r y p r o t e i n s using i n v i t r o t r a n s c r i p t i o n system. However, t h e r e g u l a t o r y mode of expression i n v i t r o has n o t been shown i n t h e case of most of genes s t u d i e d so f a r . This kind of i n v i t r o s t u d y i s t o t a l l y based on an observation of t h e t r a n s c r i p t i o n enhanced e f f e c t by t h e e x t r a c t s from t h e c e l l s i n which a given gene i s a c t i v e l y t r a n s c r i b e d . I t i s n o t c l e a r whether t h e enhancement of t r a n s c r i p t i o n l e v e l s observed i n v i t r o , i f any, m i m i c s t r u e l y r e g u l a t o r y mechanism of gene expression occurred i n vivo. Moreover, i t may not be assured i f the i n v i t r o t r a n s c r i p t i o n of a l l genes can be s t i m u l a t e d i n same manner w i t h the homologous c e l l e x t r a c t s . Therefore, i t seems t o be e s s e n t i a l t o e s t a b l i s h an improved ~ v i t r o t r a n s c r i p t i o n system i n which any r e g u l a t o r y p r o t e i n s could be i d e n t i f i e d . There When Drosophila cells are exposed to high temperatures, the synthesis of most proteins stops, hut there is a dramatic, coordinate induction of a small group of proteins, the heat shock proteins (hsp's). This response is brought about hy hoth an activation of heat shock gene transcription and a preferential translation of heat shock mRNAs; other cellular messages are stably retained in heat shocked cells even though they are not translated. When cells are returned to normal temperatures after a heat shock, the synthesis of hsp's continues until a certain amount of the major heat shock protein, hsp70, has accumulated. At this point, heat shock transcription shuts down and heat shock messages are destabilized and degraded. their characteristic translational regulation. long (242 bp) leader sequence on the hsp70 message, which contains two sequence elements that have been conserved in several other heat shock mRNAs. A number of linker insertion mutations and small deletions in this leader have been constructed in vitro. We are generating stably transformed Drosophila cell lines which harhor these constructions in order to compare their regulation to that of the resident wild-type genes.

We have been trying to identify soecific seouences on heat shock messaqes which confer Our attention has focused on the unusually A recombinant plasmid has been constructed in which the promoter from a Drosophila 70K dalton heat shock protein gene (hsp 70) has been fused to a bacterial chloramphenicol acetyltransferase (CAT) structural gene. Introduction of the plasmid into Drosophila cells in culture results in the expression of functional CAT enzyme. At some levels of transfected DNA, CAT expression is inducible by exposure of the transfected cells to a brief heat shock. When larger amounts of plasmid DNA are transfected, CAT activity is expressed at a very high level and heat shock control of CAT expression is lost. This DNA induction of CAT activity is, in fact, brought about by plasmid DNA. The DNA induction is demonstrated by cotransfection of a small amount of the hsp 70-CAT plasmid with a large amount of plasmid pBR322. A similar plasmid in which a CAT gene was placed under the control of a Drosophila 88F actin promoter was constructed. Expression of CAT is dependent on the actin promoter when the plasmid is introduced into Drosophila cells. CAT activity is repressed in the transfected cells after a brief exposure to a temperature elevation. These r e s u l t s s u g g e s t t h a t both t h e normal p a t t e r n of mRNA s y n t h e s i s and t h e e x p r e s s i o n of h e a t shock genes a r e t i g h t l y r e g u l a t e d .

W e a r e c u r r e n t l y l o o k i n g a t changes i n e x p r e s s i o n of s p e c i f i c mRNA's which may r e s u l t i n phenocopies. 

The translational regulation observed in Drosophila cells in heat shock conditions requires that: 1 ) messenger RNAs coding for heat shock polypeptides (hsp's) have a structural element identifying them for selection for translation and 2 ) the translational apparatus in the cell changes such that mRNAs coding for hsp's are actively translated while normal cellular mRNAs are not. An early event in heat shock in Drosophila and other organisms is the dephosphorylation of the ribosomal protein, 56. To determine whether this change plays a role in translational regulation in heat shock we have set up an RNA-dependent in vitro translation system from normal and heat shocked Drosophila Kc cells which mimics the regulation of translation seen in intact cells in these conditions. We have fractionated the lysates into washed ribosanal and supernatant fractions and used the fractions to determine whether ribosome or supernatant factors control the changes in translation regulation in heat shock.

are added to intact normal or heat shock lysates, we find that supernatant factors from control lysates restore normal patterns of protein synthesis to heat shock lysates. Ribosome fractions have no effect. I n reconstitution experiments where ribosomes are mixed with supernates fran either normal or heat shock lysates, we find the regulation of translation in heat shock is supernatant dependent and independent of ribosome source. I n the lysates, the state of phosphorylation o f S6 is the same as that seen in intact cells.

At least three other polypeptides in the lysates have changed their state of phosphorylation in heat shock. One of these is ribosomal and the other two are supernatant factors. Center, Jackson, MS 3 9 2 1 6 -4 5 0 5 . When Friend erythroleukemia cells are incubated at 4 3 degrees there is a rapid and nearly complete inhibition of protein synthesis which can be reversed when cells are returned to their normal growing temperature of 37 degrees. Examination of the recovery of FEC from heat shock indicates that most cellular mRNAs behave as a cohort and return to translation at approximately the same rate. We found a noteable exception to this rule in the case of a 78 kD basic protein (named protein A) whose rate of return to a normal synthetic rate is markedly inhibited subsequent to heat shock. For example, the synthesis of protein A is inhibited by greater than 6 0 percent compared with a host of other cell proteins when cells heat shocked for 3 0 minutes at 4 3 degrees are allowed to incorporate [ 3 5 S ] methionine for one hour at 37 degrees. Longer periods of heat shock increases the disparity between the rate of protein A synthesis and other cell proteins. We found that the inhibition of protein A synthesis is not due to an increase in protein turnover after heat shock nor to a disappearance of protein A mRNA. We show that protein A corresponds to the 78 kD polypeptide commonly found to be associated with the poly(A) tails of mammalian mRNA (PABP). Further analysis of heat-shocked FEC indicates that the composition of messenger ribonucleoprotein (mRNP) particles changes after stress so that PABP comprises a significantly smaller percentage of mRNP polypeptides. able gene and have found that approximately one in 10,000 integration events leads to the expression of the provirus from the viral promoter (1). The expression is due to an unknown cis-acting mechanism (1).

viruses along with their flanking EC DNA into A phage. The flanking DNA in each case hybridizes to a single poly A+ RNA species expressed in undifferentiated, uninfected EC cells. We have isolated cDNA clones for these RNA species and are in the process o f : 1) identifying the genes, 2) identifying the gene's promoter/enhancer. and 3 ) determining the influence that the promoter/enhancer might have on viral gene expression.

We have now cloned two of these expressed pro-Sorge, J., Cutting, A.E., Erdman, V.D., and Gautsch, J.W., Proc.Natl.Acad.Sci. 3: November, 1984.

Hormone Research Institute, University of California, San Francisco, Ca 94143 Recombinant plasmids were constructed containing 5' flanking DNA of the rat insulin I gene linked to chloramphenicol acetyltransferase (CAT) coding DNA. Use of a transient transfection procedure results in selective expression of CAT activity (50-200 fold) following introduction to insulin producing tissue culture cells (HIT cells) as compared with other cells. The insulin 5' flanking DNA sequence is able to augment the activity of the thymidine kinase (tk) promoter (20-30 fold) but only in insulin producing cells. We therefore conclude that this region contains a cell specific enhancer-like element.

A fragment from -103 to -249 efficiently enhances (about 20 fold) the tk promoter in an orientation-independent manner. To test whether DNA sequences more proximal to the transcription start site also contribute to cell specificity we removed the insulin enhancer from the intact flanking DNA region, replacing it with the murine sarcoma virus (MSV) enhancer and fusing to insulin gene sequences at -114. This hybrid retains substantial cell specificity (about 50 fold) despite the absence of significant enhancer activity in the insulin sequences. insulin 5' flanking DNA sequence appears to result from the activities of two distinct elements: one located more distal to the transcription start site and exhibiting properties of an enhancer and the second more proximal to the transcription start site. i n c u l t u r e following exposure t o a v a r i e t y of a g e n t s i n c l u d i n g incubation a t e l e v a t e d temperatures and i n f e c t i o n by adenovirus. dependent on t h e adenovirus transforming gene E1A (Nevins, 1982) . Pe have cloned t h e human VSP7n gene, c h a r a c t e r i z e d t h e t r a n s c r i p t i o n a l u n i t , and determined t h e DNA sequence f o r t h e gene and f l a n k i n g r e g i o n s . t r a n s c r i p t . during two hours of h e a t shock a t 43C. Fusion of sequences f l a n k i n g t h e 5' terminus of t h e USP70 gene t o t h e b a c t e r i a l chloramphenicol a c e t y l t r a n s f e r a s e ( The second element, which I s h a l l c a l l t h e " c i r c u i t theory" of gene r e g u l a t i o n , provides r u l e s for p r e d i c t i n g t h e p a t t e r n s of r e g u l a t o r y i n t e r a c t i o n s --hat can be c a l l e d t h e " t o p o l o g i c a l coupling" of r e g u l a t o r y c i r c u i t s . Even with only two coupled u n i t s of t r a n s c r i p t i o n , t h e number of p a t t e r n s of r e g u l a t o r y i n t e r a c t i o n s i s very l a r g e and t h e s e g i v e r i s e t o a d i v e r s e behavioral r e p e r t o i r e .

The p a t t e r n s w i l l be enumerated for an i n d u c i b l e system and those p a t t e r n s t h a t a r e p h y s i c a l l y r e a l i z a b l e w i l l be d i s t i n g u i s h e d from those t h a t a r e n o t .

P r e d i c t i o n s f o r t h e use of c l a s s i c a l , autogenous and coupled c i r c u i t s w i l l be given based upon t h e behavior of t h e i n t a c t systems.

, J e f f r e y S t r a t h e r n l . Margaret K e l l y 2 , Brenda Shaferl, and Carolyn M c G i l l l .

1NCI 

References 1. Mlzuno. T., Chou, M., and Inouye, M. (1983) Proc. Japan Acad

Proc. N a t l . Acad. Sci. USA

S t . Louis University Medical Center, S t . Louis, MO 2 Place Jussieu , 75251 P a r i s , France.A model f o r t h e organization o f the E. c o l i 16s ribosomal RNA i n t h e 30s subunit i s being constructed. This model i s based on t h e 16s KRNA secondary s t r u c t u r e , MA-protein i n t e ra c t i o n s (which a r e u s e f u l because the p o s i t i o n s of the p r o t e i n s i n t h e subunit a r e known) and photochemical and chemical intramolecular RNA c r o s s l i n k s t h a t i n d i c a t e the higher order s t r u c t u r e of the RNA. These considerations allow the assignment of most of the RNA h e l i c e s t o approximate p o s i t i o n s within t h e 30s subunit. The long d i s t a n c e secondary s t r u c t u r e i n t e r a c t i o n s t h a t a r e i n the c e n t e r of t h e secondary s t r u c t u r e remain i n the c e n t e r of t h e t h r e e dimensional form; s e v e r a l of t h e h e l i x end loops t h a t a r e on the periphery of the secondary s t r u c t u r e a r e a l s o organized so t h a t they a r e a t the c e n t e r of t h e t h r e e dimensional form. The t h r e e major secondary s t r u c t u r e domains a r e preserved f o r the most p a r t a s autonomous folded regions. An enhancer was identified in human cytomegalovirus (HCMV) DNA by co-transfecting cloned DNA from the immediate early region of HCMV with an "enhancer trap" of linear simian virus 40 (SV40) DNA lacking its own enhancer. Two replication competent SV40-like viruses were isolated containing HCMV DNA inserts of 341 and 262 bp. These sequences are located upstream of the major immediate early promoter, between nucleotides -118 and -524. Transient expression assay with a cloned rabbit E-globin gene indicated that the upstream region of the major immediate early gene of HCMV contains the strongest enhancer element identified so far. Studies with deletion mutants showed that different subsets of this enhancer can substitute for the SV40 enhancer. Studies with more extended deletions and various fusion genes are in progress.

Sequence Specificity in Transcription and Control

LOCUS, C. Szent-Cyorgyi, D.B. Finkelstein, and W.T. Carrard, The U n i v e r s i t y of Texas Health Science Center at Dallas, Dallas, Tx. 75235 We have investigated at high resolution the chromatin structure of the locus containing a Saccharomyces cerevisiae gene that encodes an 81 kd heat shock protein (the homolog of orthern hybridization enables us to determine that within 10 min of k?'!%%%) t h 2 s t e a d y state level of HSP83 transcript is induced at least tenfold over the basal level (27O). DNase I and micrococcal nuclease cutting sites in nuclei isolated from either heat shock induced or uninduced cells were mapped along the HSP83 locus by indirect end-labeling. A persistent set of nuclease hypersensitive features= noted. Regions 2.150 bp in breadth that are hypersensitive to DNase I are centered at -1625 bp, -545 bp, -155 bp, +2400 bp, and +3375 bp relative to the transcription start. The DNase I sites (except +2400 bp) each display centrally located protected regions of 20-40 bp, implying the presence o f DNA binding proteins.The Pelham heat shock consensus sequence lies within a protected subdomain o f the site at -155 bp. Sequences corresponding to poly(A) addition and possibly transcription termination reside within the site at +2400 bp. A clear array of DNA sequence-positioned nucleosomes, detectable by either nuclease, resides 3' to the DNase I hypersensitive region a t +3375 bp; a similar array may exist 5' to the site at -1625 bp. These two hypersensitive reoions are also each demarcated 5 ' and 3' by exceptionally strong micrococcal nuclease cleavages % 160 bp apart, a length which coincides with the canonical nucleosomal repeat. By infecting murine embryonal carcinoma (EC) cell lines with a recombinant transducing virus that contains neomycin resistance gene ( = ) linked to Moloney murine leukemia virus (M-MuLV) LTR, we have isolated clonal lines that are resistant to the neomycin analogue, G418.These transductant lines consist of undifferentiated EC cells as judged by morphology, tumorigenicity and cell surface antigenic markers.Analysis of the integrated neo sequence by Southern blot hybridization revealed that some of the lines have single copies whereas others have multiple copies in multiple sites.Although these transductant lines contained many copies of helper M-MuLV integrated in the cellular genome, expression of their genes was not detected either by reverse transcriptase activity or by XC plaque assay. Two F9tk-transductant lines were superinfected with a second recombinant transducing virus that contains herpes simplex virus thymidine kinase gene flanked by the M-MuLV LTR. The frequency of transduction to yield clones able to grow in HAT medium was similar to that of the parental cells. These results suggest that the expression of the K O gene linked to M-MuLV LTR in the trnasductant EC cell clones is due to a cis-acting mechanism: e.g., rearrangement of the enhancer sequence in M-MuLV LTR or effect by a cellular enhancer element flanking the integrated recombinant gene.To investigate these possibilities, the inregrated recombinant gene has been cloned together with its flanking cellular sequences, 2nd their structures are being analyzed.