Cloning of small DNA fragments containing the Escherichia coli tryptophan operon promoter and operator

9

Gene, 17 (1982) 9-18 Elsevier Biomedical Press

Cloning of small DNA fragments containing and operator (Recombinant

theyEscherichia coli tryptophan

DNA; trp repressor; oligo promoter-operators;

transcription

operon promoter

control)

David R. Russell and George N. Bennett Department of Biochemistry, Rice University,Houston, TX 77001 (U.S.A.) (Received June Sth, 1981) (Accepted August 19th, 1981)

SUMMARY

A 41-bp AZuI restriction fragment from the trp promoter-operator region has been cloned into the P/u11 site of pBR322, regenerating PvuII sites. Transformants were selected on media that allowed the selection of trp-operator-bearing plasmids. The cloned 41-bp fragment can be released from the vector by PvuII digestion, and it possesses a functional promoter and operator as demonstrated by in vivo tests. The 41-bp fragment contains several restriction sites: HincII, TaqI, RsaI, and a HpaI site that is located at the center of the operator sequence. Two new operator derivatives, symmetrical about the HpaI site, were prepared from the 41-bp fragment by joining two right-side, or two left-side P~uIWpa1 pieces together at the HpaI site. These derivatives showed in vivo operator activity. Plasmids containing up to five copies of the 41-bp trp-promoter-operator fragment have been constructed. These plasmids should be useful in preparing large amounts of the 41-bp fragment.

INTRODUCTION

Transcription initiation from the trp operon promoter of Escherichia coli is controlled by a repressoroperator interaction (reviewed recently by Platt, 1978; Crawford and Stauffer, 1980). The trp aporepressor, the product of the trpR gene, forms a complex with tryptophan and binds to the trp operator, blocking access of RNA polymerase to the usual transcription initiation site. Sequencing studies of the trp operator of E. coli and operator constitutive mutants (Bennett and Yanofsky, 1978) have identi-

Abbreviations: ACH, acid casein hydrolysate; bp, base pairs; KmR, kanamycin resistance; SMT, 5-methyltryptophan; TcR, tetracycline resistance; A, deletion. 0378-l 19/82/0000-0000/$02.75

0 Elsevier Biomedical Press

fied a short, highly symmetrical sequence near the transcription initiation site as the genetically defined operator (Hiraga, 1969). The trp operator overlaps the trp promoter and certain base changes affect both promoter and operator functions. Studies of the trp regulatory region from other related bacterial species that can be controlled by the E. coli trp repressor (Manson and Yanofsky, 1976) have shown the appearance of almost identical operator sequences (Oppenheim et al., 1980). In addition, the aroH gene of E. coli (Zurawski et al., 1981) and the trpR gene itself (Gunsalus et al., 1979; Gunsalus and Yanofsky, 1980) are under trp repressor control and contain sequences very similar to the trp operator. The pro moter and operator sequences occupy different relative positions in the three systems. In vitro binding

10

studies have shown the aporepressor-tryptophan plex can protect restriction from endonuclease 1978). Protection

tern. In addition,

sites around trp operators

cleavage (Bennett of certain

in vivo studies would be simplified

if operator function could be evaluated independently of the promoter function. This objective can

and Yanofsky,

bases in the operator

region from reaction with dimethyl presence of the repressor complex reported (Oppenheim

com-

be approached

sulfate in the has also been

by using a competition

which the ability of the “operator” pete with the wild-type

et al., 1980).

intracellular

trp repressor

Detailed studies of both trp promoter and operator function would be simplified if large quantities of

would

minimal

tions of the tp promoter.

sized DNA fragments

containing

chromosomal is studied.

studies, a small fragment containing

promoter

allow easier construction

technique

in

fragment to comoperator

for

For promoter

the trp promoter of sequence

varia-

This small promoter

frag-

or operator activities could be more readily isolated.

ment could also be useful in generating

transcription

We decided to construct

of other DNA sequences by introducing

it at desired

trp promoter well defined

a plasmid that contained

the

and operator functions within a small, fragment. In the case of repressor-

locations within a plasmid. A previously described trp deletion strain, W3IZO trpALC145, characterized by Bertrand et al. (1976) was shown to have full operator activity compared to

operator studies the cloning of the trpR gene and the amplified expression of the repressor protein (Gunsalus and Yanofsky, 1981) should allow more effective preparation of the protein component. The ability to isolate \quantities of small, well defined operator fragments by a simplified procedure would assist in the further definition of in vitro parameters of this sys-

the wild-type operator. Sequencing has shown the deletion removes a portion of the trp operon and places an AZuI site at +2 (Bennett and Yanofsky, 1978). This deletion strain was used as the starting material for the constructions described here. The trp

Aid

&I

JyI -39 (A)

pDRl1 _u@O

(B)

pDR36

+mclI

“a’

-30

pDR63

(0)

&$I-540

+2

CTGTTGACAATTAATCATCGAACTA6TTAACTAGTACGCAG 6ACAACT6TTAATTA6TA6CTT6ATCAATT6ATCAT6C6TC

CTGCGTACTAGTTAACTAGTACGCA6 GACGCATGATCAAyT6ATCAT6C6TC

QRQR

Jl&

.E!YdI +

-10

-20

QR (C)

Jiy

HeA!

;

nR

CTGTTGACAATTAATCATCGMCTAGTTAACTAGTTCGATGCAG 6ACAACT6TTAATTA6TA6CTT6ATCAATT6ATCAA6CTACTAATTAACACTT6TC

&&I

(118

BP)

ml1

(422

IPI

Bs.cRt

1[949 Fig. 1. DNA sequence of 0-p promoter-operator. (A) The sequence of the 41-bp trp promoter-operator fragment released by AluI or PvuII cleavage. The 41-bp AZuI fragment has the sequence CTG at -39 to -37 ahd CAG at -1 to +2. This sequence allows the A/u1 fragment to be inserted into a PvuII site and regenerate PvuII restriction sites at both -39 and +2. Numbering is relative to +1 for site of transcription initiation. Restriction endonuclease sites located in the fragment are indicated. The dashed line at - 11 to - 12 indicates the point of symmetry within the ttp operator. OR indicates the downstream portion of the operator from - 11 to +2. 0~ indicates the upstream portion of the operator from -12 to -39. (B) The sequence of the 26bp OROR symmetrical operator contained in pDR36. (C) The sequence of the 56-bp 0~0~ symmetrical operator contained in pDR63. (D) A schematic representation of BspRI-540 fragment from pBR322 using Sutchffe (1978) nomenclature (note BspRI is an isochizomer of HaeIII).

operon from trpAX145 was transferred via X trans. duction and cloning to pMK16. An AluI digestion yields a 41.bp trp promoter-operator

fragment.

This

(c) DNA manipulations Transformation

of plasmid

DNA into

recipient

AluI fragment was inserted into the fiuI1 site of pBR322, regenerating the PvuII site (see Figs. 1 and 2). The resulting transformant was isolated by a selec-

strains was by the procedure of Kushner (1978). Polyacrylamide gel electrophoresis, ethidium

tion procedure that allows only trp operator plasmidbearing strains to grow on selective media. This selec-

described

tion is similar in design to that used to select lac operator plasmid-bearing strains (Marians et al., 1976; Heyneker et al., 1976). The 41.bp trp fragment can then be released from the pBR322 vector by PvuII

sequencing followed Maxam and Gilbert (1977).

digestion.

We used the 41.bp

fragment

to prepare

plasmids containing up to five copies of the 0-p pro. moter-operator fragment in the PvuII site of pBR322. These plasmids allow easier isolation of larger amounts of the trp promoter-operator fragment. These plasmids were also used to produce two novel trp operator sequences that are symmetrical about the HpaI site of the wild-type operator. Relative operator activity of the novel sequences has been assessed by in vivo assays. These novel symmetrical operator sequences should also be useful for more detailed in vitro trp repressor-operator studies.

bromide staining of DNA, and photography DNA

(Sumner

fragments

Blunt-end

and Bennett, from

ligations

have been

1981). Elution

polyacrylamide were performed

and

of

DNA

in 60 mM

Tris * HCl, pH 7.6, 6 mM MgCle, 10 mM dithiothreitol, 0.4 mM ATP at 12°C for 2-20 h. Plasmid

DNA

for rapid

screening

was isolated

using phenol lysis (Klein et al., 1980). Preparative amounts of plasmid DNA were isolated as previously described (Sumner and Bennett, 1981). (d) Enzyme assays fl-Lactamase activity of plasmid bearing strains was assayed using an acidimetric assay (Rubin and Smith, 1973) in order to estimate relative copy number. Protein determinations were by the procedure of Lowry et al. (1951). &Galactosidase (Reznikoff et al., 1969; Miller, 1972) and anthranilate synthetase assays (Creighton and Yanofsky, 1970) have been described.

METHODS (e) Operator selection (a) Enzymes trp-operator-bearing AluI and DNA polymerase I (Klenow fragment) were purchased from Boehringer Mannheim. fiuI1 was purchased from New England Biolabs. The preparation of BspRI, an isochizomer of HaeIII, has been described (Sumner and Bennett, 1981). HpaI and HpaII were prepared by the procedure of Hines et al. (1980). T4 DNA ligase was prepared by the method of Panet et al. (1973), using the over-producing strain of Murray et al. (1979). (b) Bacterial strains X8060 is a trp-lac fusion strain described by Reznikoff and Thornton (1972) that places /I-galactosidase expression under the control of the frp promoter and operator. The strain used for frp operator selection, W3110 0p.41417 &&am14 leu277 su3, has been described (Morse and Yanofsky, 1969).

plasmids

in the appropriate

strain can be selected on minimal plates containing 5.methyltryptophan, acid casein hydrolysate, and indole. The selection is based on the observation that trp synthetase (II(trpA)-deficient strains cannot grow on low indole unless there is a high level of trp & (trpl?) activity. The presence of high 5.methyltryptophan in this culture maintains frp repressor binding and lowers rrp /IZ synthesis. This strain will grow on the selective media only if it is either trpR_ or trp0’ (Bennett and Yanofsky, 1978). Using a scheme analogous to that developed for lac operator cloning (Marians et al., 1976; Heyneker et al., 1976), a multicopy plasmid containing the trp0 sequence is introduced into the appropriate strain. Multiple operators will titrate away the trp repressor from the chromosomal operator, thus allowing transcription and translation of trp synthetase 02. This allows strains bearing trp operator plasmids to grow while non-

12

operator plasmid-bearing strains do not. Plasmids to be screened are transformed into W3110 tv-41417 frpRaml4 plates

leu277 su3 and plated on glucose minimal supplemented

5-methyltryptophan

with (50

indole

pg/ml),

casein (0.2%) and ampicillin

(1.5

pg/ml),

acid-hydrolyzed

(20 pg/ml). After 2 days

at 37’C, trp operator plasmid containing

strains form

colonies. Colonies formed a little faster using the su3 rrpRaml4

pmol &uII-cleaved pBR322 and incubated with T4 DNA ligase for 6 h at 12°C. The DNA was transformed into W3110 trpA1417 and plated

frpRaml4

leu277 su3

on rich media to select for all transfor-

mants, and selective media to select for those with a trp operator plasmid. Plasmids containing the 41-bp trp promoter-operator fragment exhibit a trpOc phenotype and produce colonies on the selective

strain than in a wild-type trpR’ strain. WSIIO

?~ppL(;i45

I

RESULTS

cloning region

of

trpep

Into

pMKI6

(a) Construction of plasmid pGB62 bearing the tryptophan operon of trpALCl45 The plasmid was obtained by first forming a xtrp transducing phage containing rrpALC145 and cloning an EcoRI fragment of the phage into pMK16 (Kahn et al., 1979). The XtrpALC145 phage was prepared from W3110 trpALC145 (Bertrand et al., 1976) by a technique similar to that used by Miozzari and Yanofsky (1978). htrpALC145 DNA was isolated (Zalkin et al., 1974), and upon cleavage with HpaI, the same sized frp operon fragment (frp promoter trpB) was observed as in the $80 trpALC145 phage DNA previously sequenced (Bennett and Yanofsky, 1978). Digestion of the phage DNA with EcoRI, ligation and transformation were carried out as described by Selker et al. (1977). Transformants of the trpB_ recipient strain were selected as trpB*, KmR colonies on minimal plates supplemented with 0.5% ACH, 10 mg/l indole and 20 mg/l kanamycin. A typical colony was chosen and plasrnid DNA, designated pGB62, was prepared. A HpaI digest of pGB62 DNA electrophoresed on a 5% polyacrylamide Tris-borate-EDTA gel shows the same trp operon fragment as in the ArrpALC145 and $80 trpALCl45 phage DNAs. (b) Insertion of 41-bp Au1 fragments into pBR322

trp promoter-operator

The 41-bp AluI fragment of pGB62, which contains the E. coli tp promoter-operator region, was isolated from either BspRI or &a11 restriction fragments. These fragments were digested with AluI to release the 41-bp trp promoter-operator fragment. The digested mixture (0.25 pmol) was added to 0.15

Insertion

I

fragment

of

&I

Into pBR322

Fig. 2. Construction of plasmids carrying trp promoter-operator. irpA,X’l45 was transferred to h and then into pMK16. AZuI digestion allowed the 41-bp ~rp promoter-operator fragment to be released from pCB62. This fragment was then inserted into the PvuII site of pBR322. The sequences at the ends of the 41-bp Ah1 fragment are such that they regenerate &x411 sites upon insertion into a PvuII site. Therefore, the fragment can be released upon PvuII digestion to yield a functionally active trp promoter-operator. Polymerization of the 41-bp fragment followed by insertion into pBR322 produced plasmids containing two to five promoter-operator fragments. pDR31 and pDR32 were reduced in size by cleaving with HpuI and religating to produce the two new symmetrical operator sequences contained in pDR63 and pDR36, respectively.

13

media

(see METHODS}.

The strain

transform as efficiently 30000-40000 colonies

used does not

as many strains, usually per pg pBR322. Approx.

‘promoter-operator unambiguously

(Fig. determined

mid with two restriction

2).

This

orientation

was

by digestion

of the plas-

endonucleases.

The PvuII

10% of all colonies on rich media also grew on selective media, indicat~g 10% of transfo~~g plasmids contained the rrp promoter-operator insertion. The nature of the inserted DNA was characterized

site of pBR322 is located within the @RI 54@bp fragment such that it divides the 54@bp fragment

by restriction

mapping

single

from selected

colonies.

tained

the

orientation. the direction

41-bp

of the plasmid DNA isolated Two were found

fragment

The orientation transcription

and

differed

that cononly

in

into note

I 18-bp and 422-bp that BspRI 41-bp

fragments

is a isochizomer

trp-promoter-operator

(Sutcliffe,

1978;

of HueIII).

If a

fragment

is

inserted at the PVUII site, the BspRI 54@bp fragment

is designated

based on

increases to 581 bp (see Fig, 1 (D) and Fig. 3, lanes 2 and 4). The HpaI site within the 41-bp fragment

would proceed

from the

divides the fragment into 13-bp and 28-bp fragments (see Fig. 1). Therefore, a double digest with BspRI and HpaI of the insert containing plasmid will yield fragments that indicate the orientation of the trp promoter-operator insertion. If the promoter-operator fragment is inserted so that it transcribes clockwise on pBR322, the double digest yields 422 + 13 (435 bp) and 118 t 28 (146 bp) fragments (Fig. 3, lane 3). If transcription is counter-clockwise, the fragments are 422 + 28 (450 bp) and 118 + 13 (131 bp) (Fig. 3, lane 5). Using this criterion, the clockwise-oriented trp promoter-operator plasmid was designated pDRl1 and the counter-clockwise was designated pDR12. (c) Derivatives of pBR322 containing promoter-operator inserts

multiple

41-bp

The 41-bp trp promoter-operator fragment was cleaved from pDRl1 with P~MII and isolated from a 7% polyacrylamide Tris-borate-EDTA gel. The purified fragment (0.25 pmol) was incubated with T4 DNA ligase in 10 /..dligation buffer (see METHODS) for 4 h at 12°C. PYuII-cleaved pBR322 (0.15 pmol) was then added, the volume adjusted to 20 @ and ligation was continued for 20 h. The DNA was then transformed into W3110 trpA1417 frpRan-114 Zeu277 su3. trp operator-containing plasmids were selected and 24 colonies were further characterized. Three classes were found: 8 contained a single 41-bp insert, 8 contained two 41-bp inserts (both tandem and inverted orientations) and 8 contained 3 or more Fig. 3, Restriction mapping of trp-promoter-operator-bearing plasmids. Plasmid DNA was digested and analyzed on a 5%

polyacrylamide T&borate-EDTA gel. Lane 1, BspRI-cleaved pBR322. The sizes in bp are indicated for some of the fragments; lane 2, BspRI-cleaved pDR11; lane 3, BspRI + HpuIcleaved pDRl1; lane 4, BspRI-cleaved pDR12; lane 5, BspRI and ffpakleaved pDR12.

inserts (various orientations). Fig. 4 shows the H&zII digest of several of these constructions. Lane 4 contains a single 41.bp insert in pDRl1 which increases HpaII 309-bp of pBR322 in lane 1 to a 350.bp HpaII fragment in lane 4. Lanes 5,6,7,8 show the insertiop of 2, 3, 4 and 5 of the 41.bp trp promoter-operator fragments into the &uII site, respectively.

-11

to t2 to itself about the HpaI site (OnOd). The

former

operator

sequence,

OLOL, would produce

a

56-bp PvuII fragment, while the latter, OnOn, would form a 26-bp A~11 fragment. (e) Formation of symmetric operators by HpaI reduction Two plasmids, pDR31 and pDR32, were selected from the 16 multiple insert plasmids to be used in the HpaI reduction step to obtain the symmetric operator sequences. pDR32 contained two 41-bp trp pro-

Fig. 4. Restriction mapping of multiple frp promoter-operator insertions. Plasmid DNAs were cleaved with HpaII and analyzed on a 5% polyacrylamide Tris-borate-EDTA gel. AB insertions are into the PvuII site located in HpaII-309. Lane 1, pBR322 (sizes of four largest fragments are indicated in bp); lane 2, pDR36 (OROR, 26bp insert); lane 3, pDR63 (OLOL, 56-bp insert); lane 4, pDRl1 (one 41-bp fragment insert); lane 5, pDR32 (two 41.bp fragment inserts); lane 6, pDR31 (three 41-bp fragment inserts); lane 7, pDR39 (four 41-bp fragment inserts); lane 8, pDR33, (five 41-bp fragment inserts).

(d) Approach to the construction

of novel operator

sequences The construction of completely symmetrical operator sequences, with two 0~ halves or two OR halves (Fig. l), involved a two-step procedure (see Fig. 2). Oligomers containing several 41-bp trp promoter-operator fragments were inserted into pBW22 at the PwII site and screened for those that contained the outer two inserts in an inverted orientation with respect to each other. The plasmids containing these inserts with inverted operators were then digested with HpaI, removing the operator sequences between the two outer HpaI sites, The HpaI sites were then joined with T4 DNA ligase to produce the novel operator sequences fusing the -12 to -39 sequence to itself about the k&I site (0~0,) and the

Fig. 5. Restriction mapping of symmetric operators. Plasmid DNA was digested and analyzed on a 5% polyacrylamide Trisborate-EDTA gel. Lane 1, BspRI-cleaved pDR31; lane 2, BspRI-cleaved pDR63; lane 3, BspRI + HpuI-cleaved pDR63; lane 4, BspRI-cleaved pDR32; lane 5, BspRI-cleaved pDR36; lane 6, BspRl + ffpaI-cleaved pDR36; lane 7, BspRI-cleaved pBR322 (sizes of several fragments are indicated in bp). A more accurate determination of the number of inserts in pDR31 and pDR32 is shown in Fig. 4.

15

moter-operator scription moter.

fragments

oriented

such that

would be directed outward pDR31 contained

tran-

from each pro-

three 41-bp frp promoter-

fragment

has an 82-bp insertion

lane 5 shows the reduction a 26-bp insertion

(two 41-bp) while

plasmid, pDR36, contains

into &R&540.

The double digest

operator fragments oriented such that the outer two promoters direct transcription towards each other.

in lane 6 shows the predicted

bands, 435 bp and 131

bp. The symmetric

plasmids,

pDR31

pDR36, have been shown to release 56-bp and 26-bp

and pDR32

were digested

1 pmol of each was incubated

with HpaI and

with T4 DNA ligase for

operator

operator

fragments,

respectively,

pDR63

upon PvuII

and diges-

20 h at 12°C in a final volume of 10 ~1. The DNA was

tion. The released fragments can also be digested with

then used to transform W3110 trpA 1417 @#am14 leu277 su3 and frp operatorcontaining colonies

HpaI

were selected.

to yield 28-bp and 13-bp fragments,

tively. The symmetric

The HpaI reduction

of pDR31 yielded the correct

symmetric operator by fusing the -12 to -39 region of the trp promoter-operator region to itself about the HpaI site. This operator plasmid (0~0,) contains a 56-bp PluII fragment and is designated pDR63. Fig. 5, lane 1 shows a BspRI digest of pDR3 1. The @RI540 fragment contains a 123-bp insertion (three 41-bp fragments) making it longer than BspRI-587. Lane 2 is a @RI digest of the HpaI reduction plasmid, pDR63, which contains a 56-bp insert in BspRI540. Lane 3 is a double digest, BspRI + HpaI, which splits 596-bp BspRI fragment into 450-bp and 146-bp fragments, as predicted for the operator fusion. An analogous characterization of HpaI reduction of pDR32 is shown (Fig. 5). Lane 4 shows the @RI

respec-

operator

firmed by DNA sequencing. fragments containing

sequences have been con&RI-@aI1

restriction

the operator sequences were iso-

lated and 3’-end-labeled with [o-32P]dCTP and KIenow DNA polymerase. The labeled DNA was then sequenced (1977).

by the procedure

of Maxam and Gilbert

(f) Enzyme assays of operator activity The relative operator activity of the constructed operators was measured by determining the chromosomal expression of anthranilate synthetase and fl-galactosidase activity in appropriate strains. This type of assay depends on the presence of increasing numbers of operators which act to titrate trp repres-

TABLE I Relative operator activity

Plasmid

p-Lactamase units per pg protein a

Specific activity of anthranilate synthetase b

Specific activity of p-galactosidase c

Colony phenotype on SMT-ACHindole minimal media

pBR322 pDR63 pDR36 pDRl1 pDR12 pDR32 pDR31 pDR33 pDR15 e

2.5 4.2 2.5 3.2 3.6 3.5 nd 3.9 nd

nd d 0.24 0.37 0.45 0.30 nd 0.55 0.48 nd

170 175 300 310 240 390 430 530 nd

no growth small large large large large large large large

a p-Lactamase units are defined as the amount of enzyme that hydrolyzed 1 rmol of benzyl penicillin in 1 h at 25°C and pH 7.6 (Rubin and Smith, 1973)/pg bacterial extract protein (as determined by method of Lowry et al., 1951). Strain without plasmid, <0.03 units/pg protein. b Assayed in W3110 frpA 1417 trpR14am Ieu277 su3. Strain alone 0.29. trpR_ strain, 1.5. Grown in glucose minimal + 20 pg/ml tryptophan. c Assayed in X8060 FAlacx74 kr mal-ara‘@80cUac (trp-tonB-1ac)Wl. Strain alone, - 180. Activity computed as in Miller (1972), units are nmol o-nitrophenol/min/mg protein. d nd, not determined. e Derivative of pGB62 with the trpB gene on plasmid removed.

16

sor off

the chromosomal

transcription

operator,

thus

allowing

from the trp promoter.

The anthranilate

synthetase

enzyme

levels were

(Table I). /I-galactosidase activity was measured by assaying the operator plasmid bearing derivative of X8060.

is a trp-lac fusion

X8060

when

present.

measured in W3 110 trp~.I 1417 trpRaml4 leu277 su3 strains containing the constructed operator plasmids

strain

cantly

promoter

DISCUSSION

strains

containing

contains

(41-bp

/I-galactosidase

levels

while

the

and pDRl1 have strain

frag-

yielded background

strain in

pDR12

(OROn)

acid was

activity levels.

operator-bearing

pDR36

3/3-indoleacrylic

of the trp promoter-operator

ment in the opposite orientation

which the trp promoter-operator controls the expression of fl-galactosidase. As shown in Table I, X8060 promoter-operator)

an inducer,

Insertion

or

similar

containing

pDR63 (OLOL) had no detectable activity. Apparently even though pDR63 was selected on trp operator selective media, this level of operator activity appears to be below the detection limits of either of the above assays. /3-L+ctamase activity was measured in the operator plasmid-bearing strains to determine if the insertion of the trp promoter-operator DNA affects the copy number (Uhlin and Nordstrom, 1977) of the constructed pBR322 derivatives (see Table I). All plasmids described have fl-lactamase activity levels similar to pBR322 and therefore would seem to have similar copy numbers. To further dissect which sequences are required for operator activity, a 41-bp trp promoter-operator fragment cleaved at RsaI (CT&AC between positions -5 and -4, see Fig. 1) was inserted into a SmaI site (CCCJGGG) and characterized by mapping with This new operator endonucleases. restriction sequence did not have detectable operator activity in viva, as judged by the operator selection procedure. (g) rrp promoter activity The promoter activity of the 41-bp promoteroperator fragment has been measured by inserting the fragment into pKO-1, a plasmid designed to measure promoter activity by expression of E. coli galactokinase (McKenney et al., 1981). The insertion and orientation of the 41-bp fragment into the SmaI site of pKO-1 was analyzed by restriction endonuclease mapping techniques similar to those described earlier (Fig. 3). Promoter activity of the correctly oriented trp promoter-operator was found to be stronger than a similarly inserted IacLJVS promoter fragment. Activity of the promoter was also found to increase signifi-

The construction

of several novel trp-promoter-

plasmids has been described. pDRl1

a 41-bp E. coli trp promoter-operator

region

that can be released from the plasmid upon digestion with PluII. This 41-bp fragment contains a functional promoter and operator. Several plasmids containing two to five trp promoter-operator fragments have also been characterized. Sadler et al. (1978) have reported that plasmids containing multiple identical sequences are often unstable when grown in various 6 coli strains. This was particularly evident when the repeated sequences were in an inverted orientation. pDR33, which contains 5 PvuII trp promoter-operator fragments, has been isolated several times without any detectable loss of DNA inserts and is suitable for purifying large amounts of the 41-bp trp promoteroperator fragment. Two novel symmetric trp operator sequences have been constructed. Both sequences were selected by their trp operator phenotype. Strains containing pDR36 formed colonies in 2 days similar to pDR11. Strains containing pDR63 grew more slowly (2.5 to 3 days) and produced much smaller colonies than strains containing pDR36 or 41-bp trp promoter-operators. This suggests both novel sequences have some trp repressor-binding activity in vivo, although the pDR63 sequence may have reduced affinity for operator. The enzyme assays in Table I also support this conclusion. It should be noted that pDR63 contains a large portion of the trp promoter sequence (-12 to -39) which could allow competition between RNA polymerase and trp repressor. The /.I-galactosidase assays of X8060 strains bearing pBR322 and pDR63 showed little /3-galactosidase activity. pDRl1 and pDR12 (both contain one trp promoter-operator region) and pDR36 produce /3-galactosidase at similar levels. The strains containing plasrnids that have two or more inserts, pDR32, pDR36 and pDR33, have increased P-galactosidase levels. It appears that the P-galactosidase activity increases more slowly as operators are added.

17

It could be that arise from differences

differences

in operator

in the number

activity

ACKNOWLEDGMENTS

of copies of the

pBR322 derivatives per cell. Comparison of /3-lactamase levels in the plasmid bearing strains allows an estimation of the relative copy numbers of the promoter-operator plasmid variants. The strains mea-

We would anthranilate

like to thank synthetase

Dr. C. Yanofsky

for

assays, Dr. W. Reznikoff

for

the E. coli X8060 strain, and Dr. M. Kahn for the TcRKmR

plasmid,

pMK16.

The research

sured (Table I) have similar /Uactamase levels, indicating there are no major differences in copy number of

ported

the

part by Training Grant (GM07833)

pBR322

pDR63,

derivatives.

The

strain

which has the lowest operator

containing activity,

was

found to have the highest /3-lactamase levels. This suggests the low operator activity can be attributed to the new operator sequence generated and is not due to reduced copy number. It is not possible to directly compare these operators with the wild-type E. coli operator using the above data. However, the W3110 trpALC145 deletion has been shown to have essentially wild-type operator activity (Bertrand et al., 1976). A plasmid bearing the rrpALC145 operator sequences (pDR15, a trpBderivative of pGB62 that has the same sequence as the 41-bp fragment sequence from -39 to +2) has been selected in W3110 trpam14 trpA1417 leu277 su3 on the SMT-ACH-indole minimal media, and the colonies appeared the same size as the strain containing pDRl1 or pDR36 (Table I). The wild-type trp operator exhibits near perfect symmetry from -21 to -2 with the only differences at -4 and -18. Methylation studies (Oppenheim et al., 1980) suggest the operator-repressor contacts are also symmetrical. A comparison of trp0’ mutants (Bennett and Yanofsky, 1978), the novel operator sequences described here, and operator sequences of other species compared by Oppenheim et al. (1980) suggest the sequence at or near -6 to -3 may be important in operator-repressor recognition. Therefore, it seems the trp operator may require the region from -20 to -3 as a minimum operator sequence that allows proper trp repressor recognition. The 41-bp trp promoter-operator fragment can be isolated by PvuII cleavage of pDR33, which contains 5 copies per plasmid. The blunt-end promoter fragment is then suitable for insertion at restriction endonuclease sites where transcription is desired. It is also being used to study the effects of adjacent DNA regions on promoter activity and as a starting material for making promoter variants.

tutes of Health (GM26437). Institutes

was sup-

by a research grant from the National

Insti-

D.R.R was supported,in from the National

of Medical Sciences.

REFERENCES Bennett, G.N. and Yanosky, C.: Sequence analysis of operator constitutive mutants of the tryptophan operon of Escherichia coli. J. Mol. Biol. 121 (1978) 179-192. Bertrand, K., Squires, C. and Yanofsky, C.: Transcription termination in vivo in the leader region of the tryptophan operon of Escherichia coli. J. Mol. Biol. 103 (1976) 319337. Crawford, I. and Stauffer, G.V.: Reglulation of tryptophan biosynthesis. Annu. Rev. Biochem. 49 (1980) 163-195. Creighton, T. and Yanofsky, C.: Chorismate to tryptophan (Escherichia coZi)-anthranilate synthetase, P.R. transferase, PRA isomerase, InGP synthetase, tryptophan synthetase, in Grossman, L. and Moldave, K. (Eds.), Methods in Enzymology, Vol. 17. Academic Press, New York, 1970, pp. 365-380. Gun&us, R.P. and Yanofsky, C.: Nucleotide sequence and expression of Escherichia coli trpR, the structural gene for the trp aporepressor. Proc. NatL Acad. Sci. USA 77 (1980) 7117-7121. Gunsalus, R.P., Zurawski, G. and Yanofsky, C.: Structural and functional analysis of cloned DNA containing the trpR-thr region of the Escherichia coli chromosome. J. Bact. 140 (1979) 106-113. Heyneker, H., Shine, J., Goodman, H., Boyer, H., Rosenberg, .I., Dickerson, R., Narang, S., Itakura, K., Lin, S. and Riggs, A.: Synthetic krc operator DNA is functional in vivo. Nature 263 (1976) 748-752. Hines, J., Chauncey, T. and Agarwal, K.: Preparation and properties of HpaI and HpaII endonuclease, in Grossman, L. and Moldave, K. (Eds.), Methods in Enzymology, Vol. 65. Academic Press, New York, 1980, pp. 153-163. Hiraga, S.: Operator mutants of the tryptophan operon in Escherichia coli. J. MoL Biol. 39 (1969) 159-179. Kahn, M., Kolter, R., Thomas, C., Figurski, D., Meyer, R., Remant, E. and Helinski, D.: Plasmid cloning vehicles’ derived from plasmids ColEl, F, R6K, and RK2, in Wu, R. (Ed.), Methods in Enzymology, Vol. 68. Academic Press, New York, 1979, pp. 268-280. Klein, R.D., Selsing, E. and Wells, R.D.: A rapid microscale technique for isolation of recombinant plasmid DNA

18 suitable for restriction enzyme analysis. Plasmid 3 (1980) 88-91. Kushner, S.R.: An improved method for transformation of Escherichiu coli with ColEl derived plasmids, in Boyer, H.W. and Nicosia, S. (Eds.), Genetic Engineering. Elsevier/North-HolIand Biomedical Press, Amsterdam, 1978, pp. 17-23. Lowry, O., Rosenbrough, N., Farr, A. and Randall, R.: Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193 (1951) 265-275. McKenney, K., Shimatake, H., Court, D., Schmeissner, U., Brady, C. and Rosenberg, M.: A system to study promoter and terminator signals recognized by Escherikhia coli RNA polymerase, in Chirikjian, J.G. and Papas, T. (Eds.), Gene Amplification and Analysis, Vol. II. Elsevier/ North-Holland, in press, 1981. Manson, M. and Yanofsky, C.: Tryptophan operon regulation in interspecific hybrids of enteric bacteria. J. Bacterial. 126 (1976) 679-689. Marians, K.J., Wu, R., Stawinski, J., Hozumi, T. and Narang, S.: Cloned synthetic [UC operator DNA is biologically active. Nature 263 (1976) 744-748. Maxam, A. and Gilbert, W.: A new method for sequencing DNA. Proc. Natl. Acad. Sci. USA 74 (1977) 560-564. Miller, J.H.: Experiments in Molecular Genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1972. Miozzari, G.F. and Yanofsky, C.: Translation of the leader region of the Escherichia coli tryptophan operon. J. Bacterial. 133 (1978) 1457-1466. Morse, D.E. and Yanofsky, C.: Amber mutants of the trpR regulatory gene. J. Mol. Biol. 44 (1969) 185-193. Murray, N.E., Bruce, S. and Murray, K.: Molecular cloning of the DNA ligase gene from bacteriophage T4 II. Amplification and preparation of the gene product. J. Mol. Biol. 132 (1979) 493-505. Oppenheim, D.S., Bennett, G.N. and Yanofsky, C.: Escherichia coli RNA polymerase and trp repressor interaction with the promoter-operator region of the tryptophan operon of Salmonella typhimurium. J. Mol. BioL 144 (1980) 133-142. Panet, A., van de Sande, J., Loewen, P., Khorana, H., Raae, A., Lillehauge, J. and Kleppe, K.: Physical characterization and simultaneous purification of bacteriophage T4 induced polynucleotide kinase, polynucleotide ligase, and deoxyribonucleic acid Polymerase. Biochemistry 12 (1973) 5045-5050.

Platt, T.: Regulation of gene expression in the tryptophan operon of Escherichia coli, in MilIer, J.H. and Reznikoff, W.S. (Eds.), The Operon. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1978, pp. 449. Reznikoff, W.S., Miller, J.H., Scaife, J.G. and Beckwith, J.R.: A mechanism for repressor action. J. Mol. Biol. 43 (1969) 201-213. Reznikoff, W.S. and Thornton, K.: Isolating tryptophan mutants in Escherichia coli by using a trp-lac fusion strain. J. Bact. 109 (1972) 526-532. Rubin, F. and Smith, D.: Characterization of R factor p-lactamases by the acidimetric method. Antimicrob. Ag. Chemother. 3 (1973) 68-73. Sadler, J.R., Betz, J.L., Tecklenburg, M., Goeddel, D.V., Yansura, D.G. and Caruthers, M.H.: Cloning of chemically synthesized lactose operators, II. EcoRI-linkered operators. Gene 3 (1978) 211-232. Selker, E., Brown, K. and Yanofsky, C.: Mitomycin D-induced expression of trpA of Salmonella typhimurium inserted into the plasmid ColEl. J. Bacterial. 129 (1977) 388-394. Sumner, W.11 and Bennett, G.N.: Anthramycin inhibition of restriction endonuclease cleavage and its use as a reversible blocking agent in DNA constructions. Nucl. Acids Res. 9 (1981) 2105-2119. Sutcliffe, J.G.: pBR322 restriction map derived from the DNA sequence: accurate DNA size markers up to 4361 nucleotide pairs long. Nucl. Acids Res. 5 (1978) 27212728. Uhlin, B. and Nordstrom, K.: R plasmid gene dosage effects in Escherichiu coli K-12: copy mutants of R plasmid Rldid-19. Plasmid 1 (1977) l-7. Zalkin, H., Yanofsky, C. and Squires, C.L.: Regulated in vitro synthesis of Escherichia coli tryptophan operon messenger ribonucleic acid and enzymes. J. Biol. Chem. 249 (1974) 465-475. Zurawski, G., Gunsalus, R.P., Brown, K.D. and Yanofsky, C.: Structure and regulation of aroH, the structural gene for the tryptophanrepressible 3deoxy-D-arabino-heptalosonic acid-7-phosphate synthetase of Escherichia coli. J. Mol. Biol. 145 (1981) 47-73. Communicated by A.M. Campbell.