NA secondary 8 M urea .
32P-Labeled RNA was analyzed by electrophoresis on 10 or 12 % denaturing polyacrylamide gels in Tris/borate/EDTA at pH 8.3 containing 7 M urea ( 10 ) .
RNA samples to be subjected to limited RNase T1 digestion were dissolved in 10 mM Tris , pH 7.9 / 10 mM MgCl2/0 .1 mM EDTA , and RNase T1 was added to a final concentration of 10 units/ml .
Incubations were carried out at 200 .
Samples were withdrawn at various times , chilled , and then diluted with the same buffer .
Diethylpyrocarbonate was added to 1 % before the samples were precipitated with ethanol in the presence of carrier tRNA .
They were then dissolved in urea/dye solution and loaded on denaturing gels .
RNA fragments were eluted from polyacrylamide gels , digested to completion with RNase T1 , and fingerprinted ( 11 ) .
All procedures using recombinant DNA were performed in accordance with the National Institutes of Health Guidelines .
ABSTRACT Transcription termination at the attenuators of the trp operons of Escherichia coli and Salmonella typhi-murium was studied in-vitro using DNA restriction fragments as templates .
Readthrough transcription beyond the terminators occurred with 5 and 30 % efficiency , respectively , in E. coli and S. typhimurium .
This difference is correlated with the stability of proposed secondary structures of the respective trp leader transcripts .
Secondary structure analyses of the two leader transcripts revealed a well-conserved pattern of RNA base pairing .
This and the possibility that trp leader RNA is translated suggest a model for regulation of transcription termination that is based on ribosome movement along the RNA and a shift between alternative RNA base-pairing configurations .
The tryptophan ( trp ) operon of Escherichia coli has two transcription control sites , a promoter-operator ( 14 ) and an attenuator ( 4 ) .
At the attenuator , a site located in the transcribed leader segment of the operon , transcription is either terminated or allowed to continue into the structural genes of the operon ( 4 ) .
Transcription termination at the attenuator appears to be regulated in response to changes in the extent of charging of tRNATrP ( 5 , 6 ) .
Salmonella typhimurium has a trp attenuator that functions much like the one in E. coil ( F. Lee and C. Ya-nofsky , unpublished ) .
The sequences of the leader-attenuator regions of the trp operons of both species have been determined ( F. Lee , K. Bertrand , G. Bennett , and C. Yanofsky , unpublished .
In this paper we report the results of in-vitro-transcription studies with restriction fragments of the tip operons of both organisms .
We show that the tip leader transcripts have appreciable secondary structure .
We suggest how this structure may play a role in regulating transcription termination at the attenuator .
MATERIALS AND METHODS Restriction fragments of the E. coli and S. typhimurium trp operons were derived from plasmids pVH153 ( 7 ) and pKB5 ( ref .
8 ; G. Bennett , K. D. Brown , and C. Yanofsky , unpublished ) , respectively .
In transcription experiments , fragments added Mg/ml ) were ( final concentration 0.1-0.5 to reaction 0.1 mixtures containing in 25 Al : 20 mM Tris acetate , pH 7.9 ; mM dithiothreitol ; 0.1 mM EDTA ; 4.0 mM Mg acetate ; 0.1 M three unlabeled KCI ; nucleoside triphosphates at 0.15 mM each and a-32P-labeled one triphosphate at 0.01-0.04 mM ; and Mg of 30-50 RNA polymerase per ml ( 9 ) .
Incubations were for After phenol 30 min at 37 ' .
extraction and ethanol precipitation with Mg of samples 50 carrier tRNA , were resuspended in 20 Ml of 0.025 % xylene cyanol and bromphenol blue containing The costs of publication of this article were defrayed in part by the payment of page charges .
This article must therefore be hereby marked `` advertisement '' in accordance with 18 U. S. C. § 1734 solely to indicate this fact .
RESULTS Transcription Termination on trp Operon Restriction Fragments .
Previous studies in-vitro established that when purified RNA polymerase transcribes the trp operon of E. coli , it generally terminates transcription after synthesizing the first 140 residues of trp RNA ( 12 ) .
The 3 ' - OH termini of the in-vitro RNA transcripts are at about the same position as the termini of the in-vivo transcripts ( 13 ) .
Thus , in-vitro , RNA polymerase must recognize the region of the trp operon that signals transcription termination in-vivo .
We have quantitated the frequency of transcription termination in-vitro by using as templates restriction fragments of the E. coli and S. typhimurium trp operons on which the trp promoter is the only promoter ( Fig. 1 ) .
The E. coli restriction fragment is 570 base pairs long and has approximately 260 base pairs following the site of transcription initiation , while the S. typhimurium DNA fragment is approximately 500 base pairs long and contains approximately 280 base pairs following the transcription start site ( F. Lee , K. Bertrand , G. Bennett , and C. Yanofsky , unpubWhen lished ) .
each of these fragments is transcribed and the products are analyzed by electrophoresis on polyacrylamide two gels , major RNA bands are observed ( Fig. 2 ) .
A prominent RNA species appears at the position expected for attenuator-terminated RNA 140 residues in length .
In addition , a fainter band appears at a position corresponding to a molecule approximately twice this length .
When these two bands are eluted and analyzed by two-dimensional fingerprinting after digestion with RNase T1 , the fingerprints of the shorter bands correspond to the first 140 nucleotides of trp RNA , while the longer bands correspond to the first 260 ( E. coli ) or 280 ( S. typhimurium ) nucleotides of trp RNA ( Fig. 3 ) .
These longer bands must therefore represent readthrough message , resulting from transcription to the ends of the fragments .
When the radioac-i i 140 I terminated 140 reodthrough i i280 FIG. 1 .
Transcription of DNA restriction fragments containing trpPOL from E. coli and S. typhimurium .
Shown are the expected lengths in nucleotides of the RNA transcripts arising from either transcription termination at the trp attenuator or readthrough transcription to the end of the DNA fragment .
tivity in each of these species is determined , we calculate that on a molar basis there is and 30 % readthrough , respectively , 5 the templates from E. coli and S. typhimurium .
on The oligonucleotide map of the short S. typhimurium transcript ( Fig. 3C ) , when related to the DNA sequence of the leader region , allows the site of transcription initiation to be assigned to approximately the same base pair as in E. coli .
Fingerprints of the short S. typhimurium RNA transcript labeled with [ a-32P ] UTP ( not shown ) reveal uridylate-rich 3 ' - oligonucleotides analogous to those found the E. coli ter-on minated transcript , indicating that termination of transcription within the S. typhimurium leader region occurs at approximately the same position as in E. coli .
In S. typhimurium , termination is within the DNA segment corresponding to the T1 a b c d On the basis of the partial digestion data and the known nucleotide sequence of the transcript , a structure such as the one pictured in Fig. 5 may be proposed .
The features of this structure include significant base pairing between residues 74-85 and 108-119 ; these two regions are joined by a non-base-paired loop 22 residues long .
As shown in Fig. 5 , a region composed entirely of guanylate and cytidylate residues , from positions 115 to 119 , is capable of base pairing with two different regions of the molecule , i.e. , with nucleotides 74-78 or 129-133 .
The fact that in the partial product in band 1 all of the transcript that is quite similar to that proposed for E. colh ( Fig. 5 ) .
Base pairing between two regions of RNA , residues 74-85 and 110-122 , would explain the relative RNase resistance of the guanylate-residues between residues 71 and 132 ( band 1 ) .
The 3 ' end of the RNA is not included in any of the bands , suggesting that the G-U bond at 136 is cleaved and the potential base pairs beyond residue 132 may not be as stable in S. typhimurtum as in E. colt .
In both cases , guanylate-residues within the long unpaired loop are somewhat resistant to hydrolysis .
The possibility of dual base pairing of one region of the RNA molecule ( residues 117-122 ) also exists in the S. typhimurium structure and involves G-C base pairs exclusively .
The nucleotide sequence of the region from residue 74 to res-idue 135 leads us to predict slightly different 3 ' stem and loop structures for S. typhimurium than for E. colt ( Fig. 5 ) .
If we consider the two possible stem and loop structures within the S. typhimurium leader RNA separately ( 15 , 16 ) , we arrive at AG -- 15 kcal/mol for the region from residues 74 to 122 and AC -- 6 kcal/mol for the region from 117 to 135 .
The lower expected stability for the 3 ' stem and loop structure in S. typhimurlum compared to E. coli ( -6 compared to -20 kcal/mol ) may account for the observed differences in the 3 ' endpoints of some of the RNase-resistant bands , as well as the greater readthrough in-vitro with the S. typhimurium trp op-eron fragment .
Transcription with ITP in Place of GTP .
Substituting the analog ITP for GTP during RNA synthesis should result in the formation of RNA molecules in which G * C base pairs are replaced by weaker I-C base pairs .
If RNA strand interactions are important in the termination process , we might expect ITP incorporation to alter the efficiency of transcription termination at the attenuator .
When this possibility was tested with ITP and the Hpa II 570 fragment from E. coli , the terminated leader transcript was absent and a single longer species was produced ( Fig. 2c ) .
Although this species has a different mobility from that of the control ( GTP ) readthrough species , it gave an identical oligonucleotide fingerprint pattern .
Apparently , then , RNA in which inosinate replaces guanylate has altered electrophoretic mobility .
The striking result is that in the presence of ITP the readthrough species is made almost exclusively .
Similar results were obtained with ITP and the S. typhimurium Hinfl fragment .
DISCUSSION Secondary Structure of tip Leader RNA and Transcription Termination .
Both E. coli and S. typhimurium contain within their trp operon leader regions transcription termination sites recognized by E. coli RNA polymerase in-vitro .
By using specific restriction fragments from each operon as templates for transcription , the amount of readthrough transcription that occurs at each attenuator can be quantitated .
Under identical conditions there is 30 and 5 % readthrough , respectively , with S. typhimurium and E. coil DNA .
The nucleotide sequences of the trp attenuator regions of the two species , although similar in general structure , have a number of differences in the transcribed regions just preceding the termination sites ( F. Lee , K. Bertrand , G. Bennett , and C. Yanofsky , unpublished ) .
These differences probably account for the nonidentical efficiencies of transcription termination since the nucleotide sequences beyond the termination sites are highly conserved in the two species ( F. Lee , K. Bertrand , G. Bennett , and C. Yanofsky , unpublished ) .
The DNA sequences beyond the termination sites in phage X are also thought not to function in transcription termination ( 18 ) .