Proc. Natl. Acad. Sct. USA Vol. 75, No. 11, pp. 5580-5584, November 1978 Naturally occurring down mutation: sequence of the trp promoter/operator/leader region of Shigella dysenteriae 16 (evolutionary sequence alterations/regulatory mutations/nutritional deficiencies) GIUSEPPE MIOZZARI AND CHARLES YANOFSKY Department of Biological Sciences, Stanford University, Stanford, California 94305 Contributed by Charles Yanofsky, June 8, 1978

ABSTRACT The promoter/operator/leader region of the MATERIALS AND METHODS trp of Shigell dysenteriae 16 has single base pair dif- ferences from the corresponding region of Escherichia coli at All E. coli strains used are derivatives of strain W3110. Strain positions -24 and -13. The difference at -13 was shown to be SH16 of S. dysenteriae was originally obtained from S. E. Luria. responsible for the 90% reduction in promoter function char- acteristic of the of S. dysenteriae. The base pair The E. coli/S. dysenteriae hybrid strains SDEC11 and difference at position -13 also renders the operator partially SDEC1ITP have been described (9). Hybrid strains generated constitutive. This allows the organism to maintain relativel in the course of this work are described in the legend to Table high repressed levels of the trp enzymes and increases the rel 1. All hybrid strains were generated by P1-transduction. The ative importance of attenuation as a control mechanism. These findings and the earlier observation that the construction of the plaque-forming trp transducing phages trpE protein of S. dysenteriae is only slightly active explain the 080htrpSD11 and 080trpSDllTP is described in the text. low in vivo expression of the tip operon of this organism. Plasmid pGM1 1 carries the initial portion of the trp operon of Nutritional studies sugest that involved in other amino S. dysenteriae. It was constructed by inserting the fragments acid biosynthetic pathways in S. dysenteriae 16 may be simi- generated by EcoRI digestion of 080htrpSD 1-DNA into the larly partially inactivated. EcoRI site of pVH51 (miniColEl; ref. 13). pGM11 was isolated Both DNA-DNA (1) and DNA-RNA (2) hybridization data and from the ligation reaction mixture by selecting tip+ transfor- protein primary structure comparisons (3, 4) indicate a close mants of an E. coli recipient strain, W3110/trpLD102, in relationship between the enteric Shigella dysenteriae which the initial portion of the trp operon has been deleted (14). and Escherichia coli. They share susceptibility to some of the Procedures for the isolation of phage and plasmid DNA and the same bacteriophages (5), and hybrid recombinants can be conditions for ligations and transformations have been described readily obtained from crosses of the two organisms (5-10). (15, 16). Despite these similarities, some features of the nutrition and Cells for the determination of enzyme activities were grown biochemistry of E. coli and S. dysenterae are quite different. in glucose minimal medium (17) supplemented with 100 jig of In particular, many strains of S. dysenteriae have been reported L-tryptophan per ml. The conditions for the assay of the glu- to require various amino acids or vitamins, or both, for growth tamine- and ammonia-stimulated reactions of anthranilate (11, 12). As described in this report, our laboratory strain, SH16, synthetase and of phosphoribosylanthranilate transferase have will only grow well on a glucose minimal medium when it is been described (9). Protein was determined by the method of supplemented with several amino acids. Lowry et al. (18). Enzyme specific activities are given as units The structural, functional, and regulatory properties of the of S. dysenteriae trp operon were studied by Manson and Ya- per milligram protein. nofsky (9, 10). They analyzed transduction hybrids in which DNA sequence analysis was performed by the method of the entire cysB-trp region of the E. col chromosome was re- Maxam and Gilbert (19). Details of this and related procedures placed by the corresponding region from S. dysenteriae. Al- such as phosphatase treatment, polynucleotide kinase and re- though the order of the tip structural genes of S. dysenteriae striction endonuclease reactions, agarose and polyacrylamide is identical with that in E. coli, tryptophan biosynthesis is largely gel electrophoresis, restriction site mapping, and extraction of blocked in the hybrids Deletion mapping identified two sites DNA from agarose and acrylamide gels are given in Bennett within the first structural gene of the operon that are responsible et al. (20) and Brown et al. (21). The conditions for RNA for a severe decrease in the activity of the enzyme complex that polymerase and Trp protection of restriction sites in catalyzes the initial reaction in tryptophan biosynthesis. In the trp promoter/operator region against cleavage of restriction addition, a linked site that maps outside the structural gene endonucleases are given in Bennett et al. (22). region of S. dysenteriae acts to decrease the maximal rates of Restriction endonuclease EcoRI was purchased from Miles trp mRNA and enzyme synthesis to 10% of those of wild-type Research Laboratories (Elkhart, IN). Restriction endonucleases E. colh. Thus, the tryptophan auxotrophy of S. dysenteriae arises Alu I, Hha I, HincII, HindIII, Sal I, and T4 DNA ligase were from the combination of a catalytically inefficient "first" en- obtained from New England Biolabs (Beverly, MA). Hpa I and zyme and decreased expression of the operon. Hpa II were prepared by the method of Sharp et al. (23); Hinfl In this report we present the DNA nucleotide sequence of was prepared by an unpublished procedure. T4 polynucleotide the trp promoter/operator/leader region of S. dysenternae and kinase was a gift of A. Maxam. Calf intestinal alkaline phos- of a regulatory "revertant" that has compa- phatase was purchased from Boehringer Mannheim (India- rable to that of E. coli. napolis, IN). E. coli RNA polymerase and partially purified Trp The publication costs of this article were defrayed in part by page repressor were gifts of F. Lee. charge payment. This article must therefore be hereby marked "ad- These studies were performed in accordance with the Na- vertisement" in accordance with 18 U. S. C. §1734 solely to indicate tional Institutes of Health "Guidelines for Recombinant DNA this fact. Research." 5580 Downloaded by guest on September 28, 2021 Genetics: Miozzari and Yanofsky Proc. Natl. Acad. Sci. USA 75 (1978) 5581 RESULTS promoter/operator region in 080htrpEA190 and the presumed Strain SDEC11 and its spontaneous tryptophan-independent corresponding fused Hpa I fragment from 080htrpSDII. derivative, SDEC11 TP, are transductiQn hybrids in which the Comparative digestions of either whole phage DNA or isolated entire cysB-trp-tonB region of E. coli is replaced by the cor- restriction fragments with various endonucleases allowed us responding sequences from S. dysenterzae (9). The initial to derive the restriction maps shown in the upper portion of Fig. portions of the trp operons of the two hybrid strains were 2. The homology of the two E. coli Hpa I fragments with the transferred onto phage 080htrpEA190AOC1418 by genetic large fragment from S. dysenteriae is confirmed by the pres- exchange (24). In the resulting transducing phages, ence of common restriction sites. However, there are several 080htrpSD11 and 080htrpSD11TP, deletion AtrpOC1418 is differences clustered around the region corresponding to the replaced by trpPOLED and at least the initial part of trpC from initial portion of the trp operon of E. coli; in addition to the Hpa the hybrids. I site difference, S. dysenteriae DNA has a number of restric- In an attempt to identify the S. dysenteriae trp promoter/ tion sites in this region that are absent from E. coli DNA. In operator region, the DNA of 080htrpSD11 and of particular, there is a recognition sequence for EcoRI about 500 080htrpSD 1 TP was digested with Hpa I and the restriction base pairs preceding the region containing the Hpa I site in E. pattern was compared with that obtained with an Hpa I digest coli. The position of this EcoRI site preceding the trp pro- of the DNA of the parent phage 080htrpEA190. This phage moter/operator region was confirmed by digesting phage k080trpSD1 DNA with EcoRI and cloning the fragment con- carries the entire trp operon of E. coli. The trp promoter/ taining the S. dysenteriae trp genes into the EcoRI site of operator region of E. coli has been shown to contain a charac- plasmid pVH51 (miniColEl; ref. 13). The resulting plasmid, teristic recognition sequence for Hpa I (GTTWAAC) at positions pGM11, allows expression of trpE and trpD, the first two -14 to -9 preceding the transcription initiation site (22). This structural genes of the trp operon, and, therefore, probably site can be protected from Hpa I cleavage by either RNA carries the S. dysenteriae trp promoter. Restriction analysis polymerase or Trp repressor, generating a large protected band revealed that the plasmid carries a single 6.5-megadalton insert composed of the two Hpa I fragments on either side of the that is identical in size to the EcoRI fragment that starts 500 base cleavage site (22) (Fig. 1). Comparison of lanes 1 and 2 in Fig. pairs before the presumptive trp promoter/operator region in 1 indicates that '80htrpSDll lacks one of the Hpa I recognition 480htrpSD11 and spans the entire trp operon. Because of the sites present in the parent phage 480htrpEA190. The Hpa I ease with which large amounts of plasmid DNA can be obtained restriction pattern of 080htrpSD11 DNA is, however, indis- and because of the smaller size of the plasmid, all subsequent tinguishable from the restriction pattern of 080htrpEA190 experiments involving the S. dysenteriae wild-type trp operon DNA digested in the presence of RNA polymerase or Trp re- were performed with pGM11 DNA. pressor (Fig. 1, lanes 4-7). In particular, the protected band in The 1.5-megadalton EcoRI-HindIII fragment spanning the digests of 080trpEA190 DNA is identical in size to the largest presumptive S. dysenteriae trp promoter/operator region was band observed in digests of 080htrpSDI1 DNA. This indicates isolated from purified pGM11 DNA (Fig. 2). Although the that it is the Hpa I cleavage site in the trp promoter/operator restriction data suggest that it should contain the entire initial region that is not present in S. dysenteriae. portion of the trp operon of S. dysenteriae, including trpPOL, Because this Hpa I site, which had served to identify the trpE, and the beginning of trpD, this fragment contains only promoter region of the E. coli trp operon (13, 22), is missing in a weak promoter, if any, because we failed to detect tran- S. dysenteriae, we located this region by searching for common scription in a purified in vitro transcription system (26) (results restriction sites between the Hpa I fragments flanking the trp not shown). In the absence of a functional test for promoter or operator activity, we resorted to identifying the region of the EcoRI-HiudIII fragment carrying the S. dysenteriae trp promoter by comparing the DNA nucleotide sequences of S. dysenteriae and E. coli around selected restriction sites. We first determined the DNA sequence at the HindIII end of the fragment by 5'-end labeling followed by cleavage with Sal I and DNA sequence analysis as described by Maxam and Gilbert (19) (results not shown). The sequence confirmed the identity of the HindIII site of the S. dysenteriae operon with the Hin- dIII site at the beginning of trpD of E. coli (M. van Cleemput, G. Bennett, and C. Yanofsky, unpublished data). From the distance between this HindIII site and the trp promoter/op- erator region in E. coli (21), we tentatively localized the S. dysenteriae trp promoter/operator region in the vicinity of an HincII cleavage site about 400 base pairs from the EcoRI end 1 2 3 4 5 67 891011 of the EcoRI-HindIII fragment. A detailed restriction map of FIG. 1. Agarose gel electrophoresis ofHpa I digests of the DNA this region is shown in the bottom portion of Fig. 2. This map of homologous trp transducing phages carrying either the E. coli or to that of the trp promoter/operator region S. dysenteriae trp promoter/operator region. Lanes and DNA: 1, is virtually identical 080htrpEA190 (E. coli trp operon); 2, 080htrpSD11 -(S. dysenteriae of E. coli (20). trp operon); 3, 080htrpSDllTP (S. dysenteriae Trp+ "revertant"); A segment of approximately 300 base pairs surrounding the 4, 080htrpEAl90; 5, 080htrpEA190 + RNA polymerase; 6, 480h- HincII cleavage site was subjected- to sequence analysis by using trpEAl90 + Trp repressor + tryptophan; 7, 080htrpSDll; 8, a number of overlapping restriction fragments as outlined in 080htrpSD11TP; 9, 080htrpSD11TP + RNA polymerase; 10, Fig. 2. With the exception of three base pair differences, this 080htrpSDllTP + Trp repressor + tryptophan; 11, 080htrpSD11. sequence is identical to the sequence determined for the trp Lanes 5, 6, 9, and 10 show the presence of a large band formed by protection of the Hpa I site in the trp promoter/operator region by promoter/operator/leader region in E. coli (20, 27). As illus- either RNA polymerase or Trp repressor. The protected band is trated in Fig. 3, two of the basepair differences, a T-A o C*(G identical in size to the largest band generated by Hpa I digestion of change at position -13 (eliminating the Hpa I site) and a C-G 080htrpSDI I DNA (lanes 2, 7, and 11). T.A change at position -24, are in the region immediately Downloaded by guest on September 28, 2021 5582 Genetics: Miozzari and Yanofsky Proc. Natl. Acad. Sci. USA 75 (1978)

cc C V V 0 _ _ 'a Vj V V 9 MDalI 4 E. coli

I I I E I . I * trp mRNA trpPOL E D C B A

C _ C_p -_

t. _ Q. 0 - 0 Vz 9 MDal I S. dysenteriae I ..I _~~~a I Sites ° not in E. colu O 4

120 180 240 -300 I-240 \ _ =

Hpa 11560 1 _ 2

3 4

5

FIG. 2. Restriction maps of the DNA region containing the S. dysenteriae trp promoter/operator region. The upper portion shows the major restriction sites present in the largest [9 megadalton (MDal)] fragment generated by Hpa I cleavage of q,80htrpSDI 1 DNA and compares them to the restriction sites present in the two Hpa I fragments flanking the Hpa I site in the trp promoter/operator region of 080htrpEA 190 DNA. Restriction sites present in S. dysenteriae but not in E. coli DNA are indicated below the line. The horizontal arrow marks the direction of transcription and the approximate location of the trp structural genes in the E. coli trp operon. The black bar in the S. dysenteriae map indicates the location of the EcoRI-HindIII fragment that was isolated for detailed analysis. The bottom portion shows a map of the trp promoter/operator region of S. dysenteriae. The numbering is in base pairs from the initiation site of trp mRNA transcription determined in E. coli (25). The position of restriction enzyme cleavage sites is shown by vertical arrowheads. The numbered arrows at the bottom indicate the five end-labeled restriction fragments used for DNA sequence analysis.

preceding the transcription initiation site. In E. coli this region polymerase or Trp repressor with the same efficiency as the contains the trp promoter and operator (20-22, 24). The third corresponding site in E. coli (Fig. 1, lanes 8-11). alteration, a C-G A-T change at position -106 precedes the The base pair change at position -13 in the S. dysenteriae region shown to be necessary for promoter and operator func- trp operon not only impairs promoter function but also inter- tions in E. coli (21, 24) and probably has no effect on trp operon feres with repressor-operator recognition, as demonstrated by expression. No base alterations were found in the adjoining the results presented in Table 1. Constitutive and repressed segments of DNA beyond the transcription initiation site; in levels of anthranilate synthetase were determined in strains that particular, the entire transcribed leader region of the operon carry either the E. coli or the S. dysenteriae trp promoter/ including the attenuator (14, 28) was conserved (results not operator region. Although the E. coli trpR protein repressed shown). It thus appeared likely that one or both of the alterations anthranilate synthetase activity about 40-fold in the E. coli trp in the promoter region are responsible for the severe reduction operon, only a 7-fold decrease was observed with the S. dys- in maximal trp mRNA and trp enzyme production charac- enteriae promoter/operator. Similar, but slightly higher, rep- teristic of the S. dysenteriae trp operon. ression factors were observed for both the E. coli and the S. This conclusion was confirmed by restriction enzyme and dysenteriae promoter/operator in transduction hybrids that DNA nucleotide sequence analyses of the trp promoter/oper- carry the S. dysenteriae trp repressor (60-fold and 10-fold ator region of the trp independent isolate SDECllTP. Al- repression, respectively). Because wild-type S. dysenteriae is though the trp operon of this isolate retains the C-G - T-A essentially trpE-, in these analyses we used recombinant E. change at position -24, a mutational change at position -13 colilS. dysenteriae hybrid strains that carry the S. dysenteriae regenerates the Hpa I site (Fig. 3) and restores promoter activity trp promoter/operator region and E. coli trpE. In agreement to the level observed in E. coli (9). The DNA sequence of this with the reduction in promoter and operator activities associ- isolate was obtained by sequencing in both directions from the ated with the altered DNA sequence in the S. dysenteriae trp Hpa I site in the trp promoter/operator region as described by promoter/operator region, we observed that the newly gen- Bennett and Yanofsky (24). The regenerated Hpa I site in this erated Alu I cleavage site at position -15 to -12 (AGCT; Fig. mutant is protected from Hpa I cleavage by either RNA 3) is not protected by either RNA polymerase or Trp repressor Downloaded by guest on September 28, 2021 Genetics: Miozzari and Yanofsky Proc. Natl. Acad. Sci. USA 75 (1978) 5583

Shigella "revertant' T 100% T A AI Shigellau,#&Xc^zsu duyac~gvsen teriae__'g T_ C 10% A G

Alu site [AGCTI-absent in E. coli and S. 0 dysenteriae E "revertant" 0 -50 -40 -30 -20 -10 1 10 20 E. coli I m GGTTCTGGCAAATATTCTGAAATGAGCTGTTGACAATTAATCATCGAACTAGTTAACTAGTACGCAAGTTCACGTAAAAAGGGTA3 .0 100% CCAAGACCGTT TATAAGACTTTACTCGACAACTGTTAATTAGTAGCTTGATCAATTGATCATGCGTTCAAGTGCATTTTTC CCAT WLI LW 1 5 Alu Hincil Hpa trp mRNA

E. coli operator

I E. coli promoter FIG. 3. DNA nucleotide sequence of the promoter/operator region of the trp operon. The nucleotide sequence in E. coli (20) is given at the bottom. The nucleotide sequences of S. dysenteriae and of its spontaneous trp+ "revertant" are given as a solid line where they are identical to the E. coli sequence. The vertical arrows indicate the only base pair differences in this region. The relative promoter activities of the three sequences are indicated on the left. The DNA regions involved in trp promoter and operator function in E. coli (21, 24) are indicated at the bottom. (results not shown). In contrast to this, the trp operator of quire amino acids or vitamins for growth (11, 12). Strain 16 will tryptophan-independent isolate SDEC11 TP is efficiently not grow on a glucose/salts minimal medium but will grow as recognized by the E. coli trpR protein, as indicated by the well as E. coli strains when the medium is supplemented with complete protection of the Hpa I site by Trp repressor (Fig. 1). acid-hydrolyzed casein and tryptophan. We investigated the In fact, Manson and Yanofsky (10) have shown that the rep- nutritional requirements of strain 16 by testing individual ressed levels of tryptophan synthetase [#2 are identical in strains amino acids and mixtures thereof for growth-promoting ac- having E. coli trpR+ and the trp operon of either SDEC1 TP tivity. We found that this strain has an absolute requirement or wild-type E. coli. for cysteine. However, this single supplement barely supports Various strains of S. dysenteriae have been reported to re- growth. When the glucose/salts/cysteine medium was further supplemented with tryptophan, methionine, threonine, glu- Table 1. trp enzyme levels in various transduction hybrids tamic acid, arginine, lysine, isoleucine, glutamine, glycine, Specific activities proline, leucine, or histidine, growth was stimulated appre- ASase* ciably. Medium supplemented with these 13 amino acids per- mitted a growth rate approximately equivalent to that of E. coli. Strain NH3 Gln PRTaset Addition of the seven remaining amino acids, singly or in E. coli W3110 0.24 0.12 0.11 combination, had no growth-stimulating effect. It appears, E. coli W3110 trpR 8.61 4.90 4.67 therefore, that in strain 16 the biosynthesis of 12 amino acids, SDEC-hy 4 (trpOPEDCBAE-c, including tryptophan, may be partially defective. In other ex- trpR+ S.d.) 0.14 0.07 0.09 periments we found that the cysteine requirement of strain 16 SDEC-hy 1 (trpOpS-d.-trpEAE-c, was revertible. trpR+ Ec.). 0.13 0.07 0.06 SDEC-hy 2 (trpOPSdd-trpEAEc, trpR) 0.93 0.50 0.49 DISCUSSION SDEC-hy 3 (trpOPS d -trpEAE c, trpR+ Sd) 0.09 0.05 0.06 The severe reduction in the maximal rate of trp mRNA syn- thesis observed in transduction hybrids in which the entire E. trp operon enzyme levels in various S. dysenteriae/E. coli trans- coli trp operon is replaced by the corresponding region from duction hybrids. For the four hybrid strains the origin of the trp S. dysenteriae (9) is the consequence of a single base pair dif- promoter/operator region (trpOP), the trp structural genes (trpEA), ference between the and the trp repressor gene (trpR) is indicated by E.c. (for E. coli) and promoter/operator regions of the trp op- S.d. (for S. dysenteriae). The hybrid strains were constructed as erons of the two organisms. Comparison of the DNA sequence follows: SDEC-hy 1, SDEC11 X trpE9914-transductant; SDEC-hy in the promoter/operator/leader region in E. coli (20, 27) with 2, AtrpEA2-trpR X SDEC-hy 1-transductant; SDEC-hy 3, the same region in S. dysenteriae reveals only two differences: AtrpEA2-trpR+ S.d. X SDEC-hy 1-transductant; SDEC-hy 4, trpR a CG - T-A change at position -24 and a T-A C-G change X S. dysenteriae 16-transductant. Average repressions: at E. coli trp at position -13 relative to the transcription initiation site. A operator by E. coli trp repressor, 40-fold; at E. coli trp operator by spontaneous mutation that increases the rate of trp mRNA S. dysenteriae trp repressor, 60-fold; at S. dysenteriae trp operator synthesis in the hybrid to a by E. coli trp repressor, 7.7-fold; at S. dysenteriae trp operator by strain level characteristic of wild- S. dysenteriae trp repressor, 9.8-fold. type E. coli (9) restores the E. coli sequence at position -13 * ASase, anthranilate synthetase (EC 4.1.3.27). For units of specific only. Thus, it appears that the base pair difference at -13 is activity, see ref 9. responsible for the altered promoter activity of S. dysenteriae. t PRTase, phosphoribosylanthranilate transferase (EC 2.4.2.18). For The absence of additional differences in the nucleotide se- units of specific activity, see ref 9. quence of the trp regulatory region of the two organisms, Downloaded by guest on September 28, 2021 5584 Genetics: Miozzari and Yanofsky Proc. Natl. Acad. Sci. USA 75 (1978) particularly in regions that were found to be highly variable (PCM 77-24333), and the American Heart Association. G.M. is a between E. coli and Salmonella typhimurium (29), emphasizes postdoctoral fellow supported by the Swiss National Science Founda- and S. dysenteriae. tion. C.Y. is a Career Investigator of the American Heart Associa- the close relationship between E. coli tion. The T-A - C-G base pair difference at position -13 elimi- nates the Hpa I site (positions -14 to -9 in E. coli) and falls 1. Brenner, D. J., Fanning, G. R., Skerman, F. J. & Falkow, S. (1972) within a region of 2-fold symmetry thought to be essential for J. Bacteriol. 109,953-965. both promoter and operator functions (24). A number of op- 2. Denney, R. M. & Yanofsky, C. (1972) J. Mol. Biol. 64, 319- erator constitutive mutants of the E. coli trp operon have been 339. analyzed; they were shown to have single base pair changes on 3. Creighton, T. E., Helinski, D. R., Sommerville, L. & Yanofsky, either side of the Hpa I site (positions -6/-7 and -15/-16). C. (1966) J. Bacteriol. 91, 1819-1826. No operator constitutive mutants were found that altered the 4. Li, S. & Yanofsky, C. (1972) J. Biol. Chem. 247, 1031-1037. Hpa I site at -14 to -9 (24). It was suggested that alterations 5. Luria, S. E. & Burrows, J. W. (1957) J. Bacteriol. 74, 461-476. to 6. Franklin, N. & Luria, S. E. (1961) Virology 15,299-311. in this six-base pair sequence may reduce promoter activity 7. Eisenstein, R. B. & Yanofsky, C. (1962) J. Bacteriol. 83, 193- a level prohibiting recovery of operator constitutive mutations 204. under the selective conditions used. The results presented in this 8. Sarkar, S. (1966) J. Bacteriol. 91, 1477-1488. paper provide direct evidence for the dual role of this DNA 9. Manson, M. D. & Yanofsky, C. (1976) J. Bacteriol. 126, 668- region by showing that a single base pair change leads to a se- 678. vere reduction in both promoter and operator functions. 10. Manson, M. D. & Yanofsky, C. (1976) J. Bacteriol. 126, 679- Comparison of the nucleotide sequence of different pro- 689. moters led Pribnow (30, 31) to propose that a heptamer with 11. Olitzky, L. & Koch, P. K. (1943) Nature 151,334-336. the ideal sequence TATPuATG centered about 10 base pairs 12. Johnson, R. B. & Mays, C. G. (1954) J. Bacteriol. 67,542-544. essential for 13. Hershfield, V., Boyer, H. W., Chow, L. & Helinski, D. R. (1976) before the transcription initiation site may be J. Bacteriol. 126,447-453. binding of RNA polymerase to promoters. In the trp promoter 14. Jackson, E. N. & Yanofsky, C. (1973) J. Mol. Biol. 76,89-101. of E. coli, only three positions between -13 and -7 are ho- 15. Zalkin, H. C., Yanofsky, C. & Squires, C. L. (1974) J. Biol. Chem. mologous to the ideal heptamer. The alteration at position -13 249,465-475. in the S. dysenteriae trp promoter affects the first position in 16. Selker, E., Brown, K. D. & Yanofsky, C. (1977) J. Bacteriol. 129, the "Pribnow box" and further reduces its homology with the 388-394. ideal sequence. 17. Vogel, H. J. & Bonner, D. M. (1956) J. Biol. Chem. 218, 97- An abundance of amino acids in the natural growth envi- 106. ronment of S. dysenteriae may obviate the need for a fully 18. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. of (1951) J. Biol. Chem. 193,265-275. functional tryptophan pathway, and hence the occurrence 19. Maxam, A. & Gilbert, W. (1977) Proc. Natl. Acad. Sci. USA 74, the partially inactivating mutations in trpE as well as the 90% 560-564. reduction in promoter activity. The reduction in repressor- 20. Bennett, G. N., Schweingruber, M. E., Brown, K. D., Squires, C. operator recognition would seem to counteract the effect of the & Yanofsky, C. (1978) J. Mol. Biol. 121, 113-137. promoter down mutation because it allows the repressed trp 21. Brown, K. D., Bennett, G. N., Lee, F., Schweingruber, M. E. & enzyme levels of strains carrying the S. dysenteriae trp pro- Yanofsky, C. (1978) J. Mol. Biol. 121, 153-177. moter/operator region to approach those observed in wild-type 22. Bennett, G. N., Schweingruber, M. E., Brown, K. D., Squires, C. E. coli. The purpose of this may be to permit the organism to & Yanofsky, C. (1976) Proc. Nati. Acad. Sci. USA 73, 2351- retain a minimal threshold level of trp operon enzymes in the 2355. 10-fold 23. Sharp, P. A., Sugden, B. & Sambrook, J. (1973) Biochem. 12, event that tryptophan becomes growth limiting. The 3055-3063. promoter down effect coupled with partial constitutivity in- 24. Bennett, G. N. & Yanofsky, C. (1978) J. Mol. Biol. 121, 179- creases the relative importance of attenuation in the regulation 192. of expression of the trp operon of S. dysenteriae. Our nutri- 25. Squires, C., Lee, F., Bertrand, K., Squires, C. L., Bronson, M. J. tional data suggest that at least 11 additional amino acid bio- & Yanofsky, C. (1976) J. Mol. Biol. 103,351-381. synthetic pathways carry partially inactivating mutations in 26. Lee, F., Squires, C. L., Squires, C. & Yanofsky, C. (1976) J. Mol. S. dysenteriae. By analogy to the situation in the trp Qperon, Biol. 103,383-393. the partial requirement for various amino acids could be the 27. Lee, F., Bertrand, K., Bennett, G. N. & Yanofsky, C. (1978) J. minor in the regulatory and structural Mol. Biol. 121, 193-217. consequence of changes 28. Bertrand, K., Squires, C. & Yanofsky, C. (1976) J. Mol. Biol. 103, regions of the genes concerned with the respective path- 319-337. ways. 29. Bennett, G. N., Brown, K. D. & Yanofsky, C. (1978) J. Mol. Biol. We thank Virginia Horn and Miriam Bonner for their excellent 121, 139-152. technical assistance. This work was supported by grants from the U.S. 30. Pribnow, D. (1975) Proc. Natl. Acad. Sci. USA 72,784-788. Public Health Service (GM 09738), the National Science Foundation 31. Pribnow, D. (1975) J. Mol. Biol. 99,419-443. Downloaded by guest on September 28, 2021