Mol Gen Genet (1991) 227:106-112 0026892591001510 © Springer-Verlag 1991

The isolated N-terminal D N A binding domain of the c repressor of bacteriophage 16-3 is functional in D N A binding in vivo and in vitro G~za Dallmann 1' 2, Ferenc Marines a, P~ter Papp 1, Mikl6s Gaszner l and Lfiszl60rosz 1' 2

1 Attila J6zsef University, Department of Genetics, K6z~pfasor 52, 6726 Szeged, Hungary 2 Agricultural BiotechnologyCenter, Institute of Molecular Genetics, 2101 G6d6116,PO Box 170, Hungary

Summary. The 197 amino acid c repressor of the temperate Rhizobium meliloti phage 16-3 still regulates the OR operator of the phage after removal of its carboxyl terminal region. When cloned in the low-copy-number plasmid pGA46, a severely truncated variant (R1-77), which retains only the first 77 amino acids of the intact protein, repressed in vivo transcription from the phage promoter PR. When the R1-77 repressor was fused to E. coli ~galactosidase, the hybrid protein bound OR operator DNA in vitro. The behavior of fusion proteins derived from a point mutant is consistent with the assignment of DNA binding specificity to the amino-terminal region. Furthermore two repressor alleles bearing ts mutations that mapped in the R1-77 region (near a helix-turnhelix motif) were also temperature sensitive for regulation of the OR site, while an 18 bp "in frame" deletion mutant, which mapped in the carboxyl terminal segment, regulated the OR operator in wild-type fashion. The carboxyl terminal region of the repressor is however necessary for the control of lysogenic development of 16-3. Key words: Repressor - Operator binding domain - Repressor-/~gal fusion - Temperate phage - Rhizobium

Introduction

There are many examples known among proteins of the separation of various functions into autonomous structural domains. For example, the immunity repressor of lambdoid phages folds into two domains; one is utilized in the recognition of the operator DNA, the other in the formation of repressor oligomers from dimers. The structural and functional autonomy of these domains is well established: in vitro the physically separated domains retain their characteristic biological activities (Ogata and Gilbert 1978; Pabo et al. 1979; Sauer et al. 1979; Aiba and Krakow 1981; Wilcken-Bergman and Offprint requests to: L. Orosz (G6d616)

Mtiller-Hill 1982; Mata-Gilsinger and Ritzenthaler 1983; Anderson et al. 1984; Ptashne 1986). The results of mutational analyses in vivo (clustering of certain mutations in fine structure maps, and interallelic complementation patterns) and data from gene fusion experiments (between repressor genes and the lacZ gene) are also compatible with a domain structure of repressors in vivo (Platt et al. 1972; Pfahl et al. 1974; Mfiller-Hill and Kania 1974; Case and Giles 1975; Kania and Brown /976; Lieb 1976; Mata-Gilsinger and Ritzenthaler 1983). The functional autonomy of a repressor domain has been demonstrated directly by protein fusion between the DNA binding headpiece of the E. coli L e x A repressor and the yeast regulatory protein GAL4. The hybrid protein recognized the L e x A operators in both E. coli and in yeast (Brent and Ptashe 1985). Another example of domain autonomy is seen in the repressor of the temperate phage 16-3 of Rhizobium meliIoti, a nitrogen-fixing soil bacterium. Lysogeny of phage 16-3 is regulated by two repressors, c and X (Orosz et al. 1973; Dorgai et al. 1986). There are three operator regions in 16-3 (OL, OR and OT) which ale regulated by the c repressor (Dallmann et al. 1980, 1987). The recognition specificity of the 16-3 c repressor for DNA is related to that of the otherwise unrelated E. coli phage 434 repressor. The two repressors are partially crossfunctional on the cognate operators (Dallmann et al. 1987). In this report we focus on the structure of the 197 amino acid 16-3 c repressor and address the question of whether the c cistron can be truncated and still retain activity. Materials and methods Bacterial strains, phages and plasmids. We used E. coli

strains HB101 (Boyer and Roulland-Dussoix 1969) for plasmid propagation; HB101 RifR (gift from Z. Bfinfalvi) for conjugation; RR1 (Bolivar et al. 1977) in cases where a rec + background was required and MC1061

107 Table 1. Plasmids carrying I6-3 regulatory elements Name

Source of 16-3 Insert

pGA46 derivatives (PR-tet fusions) pDPll c+ pDPl 5 c+ pDP62 c+

insert"

Specific comments for regulatory elements

Wild type Wild type In frame deletion of R (RA 156-161) Truncated R (R1-155) ts R (Rl-197[met~5]) ts R (Rl-197[gluZ3]) Truncated R (R1-77) Truncated R (Rl_77[met15]) Double mutant R (Ri-197[gluZ3-U6]) Mutant OR (O~-1) Truncated R (R1-77) and tR deleted Truncated ts R (R1-77 [met15]) and tR deleted Truncated ts R (R1-40 [metlS]) and tR deleted

pDP61

e+

pDP5 pDP25 pDPI6 pDP6

cti3 cti4 c+ eti3

H55-H60 P57-H60 P57-H60 ASa/I-XhoI H55-H60 ASalI-HindII P57-H60 P57-H60 B59-H60 B59-H60

pDP55

cti4-U6

P57-H60

pDP65 pDP71

avirT1-6 c+

P57-H60 B59-Be/I

pDP72

cti3

B59-Bc/I

pDP9

cti3

$280

pNM480 derivatives (Pc-R-fl-gal fusions) pAV3 c+ pFM1001

cti3

B59-E61 B59-E61

Truncated R fused to fl-gal (R1-77-fl-gal) Truncated ts R fused to fl-gal (R1-77[met15]-fi-gal)

Reference for origin and/or sequence b An and Friesen (1979) This work Dallmann et al. (1987) This work This work Dallmann et al. (1987) Dallmann et al. (1987) This work This work Orosz et al. (1980) Dallmann et al. (1987) This work This work This work Minton (1984) This work This work

a Inserts are designated according to the map coordinates of restriction sites mapped on the 16-3 genome (Dorgai et al. 1983). R1-197 indicates the intact 197 amino acid repressor b Sequence of putative transcription terminator tR: 5'-TAACGCTGCCCGCCTAGATCGCGGGTGGCGTTATTTT-3'(reading toward H60 site, Fig. 1)

(Casadaban et al. 1983) for identifying p N M plasmid derivatives. R. meliloti 41, which serves as host for 16-3, the wide host range conjugative plasmid p G Y 2 , carrying the 16-3 prophage (Kiss et al. 1980; Dorgai et al. 1981), and 16-3 phage mutants (Orosz 1980) were from our own collection. The following characteristics of 16-3 strains are relevant to this study: wild-type phage, c +, forms turbid plaques at 25 ° C-38 ° C; cti3 and cti4 are temperature-sensitive repressor mutants that f o r m turbid plaques below 32 ° C (ti4) and 34 ° C (ti3), and clear plaques at higher temperatures; cti4- U6 is a double point m u t a n t that forms only clear plaques at 25 ° C-38° C (U6 is the most proximal mutation available in the c cistron; Orosz et al. 1980); avirT1-6, a product of a back-cross between wild-type 16-3 and the immunityinsensitive m u t a n t vir-1, carries the c + allele and the O~-1 m u t a n t operator (Dallmann et al. 1980, 1987). Mutations ti3, ti4 and O~-1 have all been sequenced (Dallm a n n et al. 1987). In ti3, leu 15 is replaced by met, and in ti4 gly 23 is mutated to glu. Plasmids used for investigating repressor action are listed in Table 1.

Plasmid constructions. A restriction m a p of the relevant region of the 16-3 genome (Dorgai et al. 1983) is given in Fig. 1. Table 1 lists structural and functional features of the plasmids used in this study. Plasmids that code for full-length R (i.e. R1-197) (pDP5, pDP15, pDP25, pDP55, pDP65) were constructed by inserting the P57H60 fragment from the appropriate 16-3 m u t a n t into p G A 4 6 after cleavage of the vector with PstI and HindIII. Plasmids p D P 6 and pDP16, which code for the repressor fragment R1-77, contain the B59-H60 fragment inserted in p G A 4 6 after cleavage of the vector with HindIII and BgIII. Plasmids pDP71 and pDP72 are derivatives of pDP16 and pDP6, respectively. In these two plasmids, the tR site was removed by BclI and HindIII digestion, then the ends were filled in and religated. In this plasmid family R1-77 is fused at the carboxyl end to the amino acid sequence L-P-Y-L-P-V-L-R-L-Q-QW - Q - Q - R - C - A - N - Y derived from vector D N A sequences. Plasmid pDP9, coding for R1-40, was synthesized by ligating the $280 fragment into p G A 4 6 after linearization of the vector with BglII. The correct orien-

108

,

A B

,

X

I

H55

[

56

Xhol\/SalI

7

Pc ORP R

fR

U6

ti4[gtu23]~ j,ti3[met ;51 J

--

H-T-H

• I

RI-197 R I-~0

R 1-77 Rt-lSS RAts6-161

Fig. 1A and B. Genetic and physical maps of the regulatory elements in 16-3. A C and X repressors, OL, OR and OT operator regions (see Introduction). B Anatomy of the C region and the R alleles used in this work. Restriction sites: B, BamHI; E, EcoRI, H, HindIII; P, PstI; numbers after restriction sites refer to the standard map coordinates of the 16-3 physical map (Dorgai et al. 1983). $280 is a Sau3A fragment that contains OR, PR, Pc and R1-40 (Dallmann et al. 1987). Clear plaque mutations: ti3, ti4 and U6 are point mutations in the e cistron, O~-1 is one of the operator mutations in the immunity insensitive (virulent) mutant vir-1 (see details in Materials and methods). PR and Pc stand for the rightward and e gene promoter, respectively (see Table 1 for references).

Repressor alleles are designated with reference to the portion of the intact repressor sequence that they encode. R1-197 indicates the full length repressor; the filled box indicates the helix-turn-helix motif (H-T-H, spans from the residue 26 to 45); the open box indicates the position of the 156-161 deletion; the thin line designates the repressor sequence; the zig-zag line (joined to the left end of truncated repressors) indicates the plasmid-derived sequences fused to the carboxyl terminal of the truncated repressor (see details in Materials and methods). The DNA sequence of the H55-E61 region is available from the EMBL under accession number X06420

tation of the insert (PR fused to tet) was verified by BglI digestion. In pDP9, R1-40 is fused at the carboxyl end to a polypeptide of the following structure: F-H-TY-Q-F-C-A-C-S-N-G-N-N-V-A-E-T-I-N-W-R-T-T-YS-S-S-P-A-T-I-N-R-L-D-G-G-G, as deduced from the D N A sequence of the plasmid. For plasmid pDP61, coding for R1-155, the HindII-SalI fragment was removed from the H55-H60 insert in p D P l l by partial HindII digestion and religation. The resulting plasmid structure was verified by HindIII and HindII digestions. The carboxyl end of R1-155 in pDP61 carries the tripeptide extension D-Y-N. Plasmid pDP62 carries RA156-161, an in frame deletion, and was constructed by removing 6 codons from the open reading frame (ORF) of R1-197 (in pDP15) by XhoI-SalI digestion and religation. The plasmid structure was confirmed by testing for the loss of the XhoI and SaII sites in the insert and for the presence of the SalI site in the vector. RA156-161 lacks the sequence D156-C-P-P-S-L 161. To synthesize plasmids pAV3 and pFMI001, the B59-E61 fragment from 16-3 was ligated to pNM480 after digestion of the vector with BamHI and EcoRI. The structure of the plasmids was further checked by detailed restriction analyses; pDP61 and pDP62 structures were confirmed by D N A sequencing. We used the M13mp cloning system (Messing and Vieira 1982; Norrander et al. 1983) and the chain termination method (Sanger et al. 1977) for sequencing.

a series of LB agar plates containing Cm (30 gg/ml) and tetracycline (Tc) at various concentrations. Plates were incubated for 24 h at 37 ° C or 36 h at 30 ° C, and the numbers of surviving colonies were plotted against Tc concentration. Resistance is expressed as the concentration resulting in a decrease of 50% in plating efficiency (mic, minimum inhibitory concentration). The observed decrease in colony number was accompanied by a concomitant decrease in colony size. The actual mic values were normalized by comparing the growth of the various strains plated in parallel on the same petri dish.

Tetracycline resistance assay. Strains carrying pGA46 and p D P plasmids were propagated in LB medium containing 30 gg/ml chloramphenicol (Cm). Samples of log phase cultures were plated in appropriate dilutions on

Gel retardation assay. Protein extracts for D N A binding assays were prepared as described by Singh et al. (1988). Binding reactions (30 gl) contained 10 m M TRIS-HC1 (pH 7.5), 1 m M MgC12, 1 m M CaC12, 0.1 m M EDTA, 200 m M KC1, 1 m M D T T , 250 gg/ml bovine serum albumin; 3 gg sonicated chicken DNA, 0.5 ng 32p endlabelled D N A fragment (see Fig. 1 and Dot blot assay below) and the indicated amount (see Fig. 2) of protein extracts. Incubation was for 30 r e i n a t 28 ° C and 36 ° C (in vivo, these temperatures are used routinely to identify the c + and cti alleles in phage). Mixtures were loaded immediately onto a 5% polyacrylamide gel containing 25 m M TRIS, 190 m M glycine pH 8.3 and 1 m M E D T A (Singh et al. 1988). Gels were run in the same buffer at 100 V for 1.5-2 h, dried and exposed for autoradiography. Purification of RI-77-fl-gal fusion protein. The method of Miller (1972) was followed except that Sephacryl S300 chromatography (1.5 x 30 cm column, 0.2 M NaC1/

109 NTM buffer) was used for the last step. The purity of the protein was checked by SDS polyacrylamide gel electrophoresis, and Western blotting using 3ap-labelled DNA fragments carrying 16-3 OR sequences (Bowen et al. 1980). The specific activity of the fusion protein was ~ 10% of that of intact fl-galactosidase. Dot blot assay for DNA binding. Quantitative dot blot binding assays were carried out under the conditions described by Bowen et al. (1980) but using 1 mM MgC12 and 1 mM CaC12. Nitrocellulose membrane strips (12 cm x 1 cm) carrying dilution series of pure Ri-77-flgal (ranging from 10 - 16 mol to 10-1 ~ tool) protein were prepared. The membrane strips were then incubated with 10 pmol 32p-labelled O~ and O~-1 D N A probes. Radioactivity retained by each protein sample was measured by liquid scintillation counting. DNA probes were endlabelled at the EcoRI site using the Klenow fragment of D N A polymerase I according to the procedure described by Maniatis et al. (1982); the specific activity for O + D N A was 1.76 x 105 cpm per pmol EcoRI end, 2.32 x 105 for O~-1 and 1.83 x 105 for control. The DNA probes were isolated from M13mp8 phages carrying fragment $280 (Fig. 1) (inserted at the BamHI site and recovered by digestion with EcoRI and SalI); a 207 bp EcoRI fragment of plasmid rcVX was used as a control. Transfer of a mutagenized repressor allele to the 16-3 phage chromosome. The non-conjugative pGA46 derivative, pDP62 (Cm R, see Table 1), carrying the RA156-161 allele as a P57-H60 fragment, was transformed into the E. coli rec + strain RR1, which bears the 16-3 prophage inserted in the conjugative plasmid pGY2 (KmR), in order to promote homologous recombination between the c + prophage and the 16-3 sequence cloned in pDP62. This strain was then mated with the E. coli rec- strain HBI01 (RifR) in order to rescue the recombinant plasraids in rec- recipients, making use of the conjugation ability inherited from pGY2 by the recombinant plasmid. Individual HB101 RifR transconjugants carrying the cointegrate of the two plasmids (resulting from homologous recombination between pDP62 and pGY2) were selected (they are Km R Cm R Ri~) and mated again with R. meIiloti 41 (rec+). Free phages appeared by spontaneous induction in Rhizobium recipients in each cross and the majority (28/30) produced a mixed population of phages that formed clear (30-70%) and turbid plaques. Two isolates out of 30 liberated only turbid plaque-forming phages (in these cases the cointegrated plasmids were probably produced by crossing over between the tet genes of pDP62 and pGY2). Randomly chosen single plaques (a total of 61) from the mixed populations were screened by restriction analysis for the presence of SalI and J(hoI sites in the H55-H60 region (Fig. 1). All (21/21) isolates that had lost both restriction sites (i.e. those that carried the RA 156-161 allele) formed clear plaques, while all of the turbid plaque formers (40 tested) retained the SalI and XhoI sites in their H55H60 fragment. One clear mutant was chosen for further work and its c region was checked by more detailed restriction and genetic analyses and sequencing.

Enzymes. Restriction endonucleases, T4 DNA ligase and the Klenow fragment of D N A polymerase I were obtained from Reanal (Budapest). Chemicals, isotopes. Antibiotics were obtained from Chinoin (Budapest), Serva (Heidelberg), Boehringer (Mannheim). Other chemicals were from Reanal (Budapest) or from VEPEX-Biotechnika RT (Szeged); [e_32p]_ ATP (110 TBq/mM) was purchased from Izinta (Budapest).

Results

Repressors truncated at their carboxyl ends regulate the OR operator The intact c repressor gene (including its promoter Po) and one of the cognate operator-promoter regions ORPR can be excised from the 16-3 chromosome by treatment with PstI and HindIII restriction endonucleases (Fig. 1). We fused the PR promoter to the tet gene of the promoter probe vector pGA46 (Table 1). In the resulting plasmids, the expression of the tet gene depended on the status of the controlling elements (i.e. repressor and OR) contained in the same D N A insert: if PR activity was blocked by the repressor, no Tc g colonies were obtained; if a mutation in repressor or in OR interfered with repression of PR, the plasmid-carrying cells acquired a Tc g phenotype. The mutations used are described in detail in Table 1 and in Materials and methods and the results of assays for Tc g cells are summarized in Table 2. As expected, when both the c gene and OR were intact no tet activity was noted (pDP15). On the other hand the operator mutation O~-1 resulted in constitutive expression of tet, while temperature-sensitive alleles (ti; R1-197 [met 15] and RI-197 [glu23]) of the c gene rendered the development of Tc R colonies temperature-dependent. The double mutant cti4-U6 (R1-197 [gluZa-U6]) and the very severely truncated allele R1-40 [met 15] lost their repressor activities completely and hence Tc R colonies were obtained at both 30 ° C and 37 ° C. In plasmids RA156-161, R1-77 and R1-155 (Fig. 1), which code for truncated repressors, rightward transcription from PR was still repressed (Table 2). However, the activity of the shortest N-terminal fragment, R1-77, produced by plasmids pDP6, pDP16 and pDP72 proved to be temperature-sensitive. Mutation ti3 (which changes leu 15 to met 15 in the repressor) accounts for this phenotype in pDP6 and pDP72. Further experiments are needed, however, to understand why the repression in pDP16, but not in pDP71, is ts.

R1-77-fl-gal fusion protein binds OR We cloned the B59-E61 fragment (Fig. 1) containing OR, PR, and part of the c gene in the vector pNM480 (Table 1). A fusion gene was produced, which in E. coli directed the synthesis of a protein having the R1-77 repressor at its amino terminal region fused to fl-galactosi-

110 Table 2. Effects of mutations in Plasmid in HB101

O R or

c on PR-directed tet gene expression"

e Repressor allele

Operator allele

Degree of Tc R

Repression of tet activity

30 ° C

37 ° C

pGA46

none

none

1-2

1-2

t, present pDP15 pDP62 pDP61 pDP5 pDP25 pDP16 pDP6 pDP55 pDP65

R1-197 RA156-161 R1-155 Rl.197[met ~5] Rli197[glu 23] R1- 77 Rl_77[met ~5] Rl-197[gluZ3-U6] R1-197

O~ O~ O~ O+ 0+ O~ O~ O+ O~_ i

1-3 1-3 4-6 5-7 5-7 4-6 7-9 11-13 11-13

1-3 1-3 7-9 11-13 11-13 9-11 12-14 12-14 12-14

tR deleted pDP71 pDP72 pDP9

R1-77 Rl_77[met ~s] R1-40[met ~s]

O+ O+ 0+

14-16 21-23 30-32

14-16 27-29 30 32

30 ° C

37 ° C

at 12 gg/ml Tc + + + + + + + + +

+ + + + + -

at 28 gg/ml Tc + + +

+ + -

" The degree of tetracycline resistance was assayed as described in Materials and methods and is expressed as the concentration range required to reduce the efficiency of plating (EOP) by 50% (mic). Strong repression ( + +) is defined as EOP<10 -4 where no resistant colonies were detected. In cases of intermediate repression (EOP < 10-4) only microcolonies (diam. < 0.05 mm) were observed. The degree of derepression was assessed based on EOP and colony size (see Materials and methods). At EOP_
repressor-p-gal

-

op,rator

Oi~ O1~ OR Oi~

Oi~ 01~_l

temp.CC)

28

36

c+

c*

ctl3

cli3

c*

1.3O

28 36

28

28

+

OR

1.2-

~ I.I-o

o

-~ t00

o

o

"

."

"'~ 0.0-

B

~0.8"

00-0.7 . -0

o0.6-

00l 5 l F-'-~

g

0 "..D,

0.4-

0

0

0

0.3-

Fig. 2. Specific binding of R1-77-#-gal hybrid protein to OR operator. Gel retardation assay of 32p-labelled $280 fragments carrying wild type (O+) and mutant operators (O~-1) was performed at the indicated temperatures as described in Materials and methods. e + and cti3 refer to the R1-77-#-gal and Ri-77[met15]-#-gal hybrid proteins, respectively; the positions of protein-bound $280 DNA and free $280 are indicated by the arrows B and F respectively

d a s e residues 8-1021. Two alleles o f this f u s i o n gene were c o n s t r u c t e d : one w i t h the wild t y p e R 1 - 7 7 r e g i o n (c +) a n d one with the s a m e r e g i o n d e r i v e d f r o m a ts R allele (ti3; R 1 - 7 7 [met15]). T h e D N A b i n d i n g abilities o f the c o r r e s p o n d i n g h y b r i d p r o t e i n s R1-77-#-gal a n d R1-77 [met 15]-#-gal were i n v e s t i g a t e d b y gel r e t a r d a t i o n assays, using c r u d e cell extracts ( M a t e r i a l s a n d m e t h o d s ) . It was d e m o n s t r a t e d (Fig. 2) t h a t D N A b i n d i n g c o r r e l a t e d una m b i g u o u s l y with the p h e n o t y p e o f the p h a g e strain w h i c h d o n a t e d the R 1 - 7 7 r e g i o n o f the h y b r i d p r o t e i n :

u_ 0.20.1"

01

•

•

•

•

•

•

-_ ~

=

=

~

--

=

,

-

~

=

=

0 Cm

-Control

6-17 lb-1, lb-15 lO- , lb-,3 lO-12 1b-1 1 -1o R1-77-~-gal

added (mot)

Fig. 3. Specific binding of RI-77-#-gal to Ok: Dot blot assays using defined quantities of purified protein (see Materials and methods for details)

R1-77-#-gal b o u n d O~ b o t h at 28 ° C a n d 36 ° C l e a d i n g to a c h a r a c t e r i s t i c decrease in the e l e c t r o p h o r e t i c m o b i lity o f the o p e r a t o r D N A f r a g m e n t , while R 1 - 7 7 [met 15]fl-gal shows t e m p e r a t u r e d e p e n d e n c e , b i n d i n g O + o n l y at 28 ° C. F u r t h e r m o r e R1-77-#-gal d i d n o t b i n d the m u t a n t o p e r a t o r O~-1.

111

6 amino acids lost in the carboxyl proximal region of the c repressor are essential for lysogenization by 16-3.

KmR

KmR

Discussion

×

intraspecific conjugation with

CA

C+

E. coti Rec A

CmR

cmR

in E.coti RecA I interspecific conJugation

in E. coil Rec÷

with

Rhizobium meU[oti

Rec÷ KmR C+

(9

u

C~A

~

J phage induction

CmR Fig. 4. Transfer of deletion A156-161 to the 16-3 chromosome (see details in Materials and methods). The heavy line designates 16-3 DNA sequences; Tra + indicates the genes for conjugative transfer

Our results indicate that the operator-binding domain of the Rhizobium phage 16-3 repressor can autonomously regulate one of the 16-3 operators, OR. The results of three experiments support this conclusion. First, repressors whose carboxyl-terminal regions had been deleted repressed OR in vivo. (It is worth mentioning that these repressors were probably not overexpressed, since the truncated repressor genes were cloned on low-copynumber plasmids.) Second, the fusion protein R1-77-~gal binds to the OR operator in vitro. Third, mutations in the amino-terminal domain affected the ability of repressor to block tet gene expression from PR, while a mutation mapping outside this region, in the carboxyl terminal half of the protein, did not. Although the loss of the carboxyl terminal region of the repressor still allows binding at OR, it inhibits lysogeny (as shown by RA156-161). Hence, the carboxyl terminal region plays an indispensable role in other significant processes required for lysogenic development. Our current investigations of the function of the leftward operator OL are aimed at elucidating this role. Acknowledgements. We thank L/~szl6 Dorgai and Ferenc Olasz for helpful daily discussions; Sankar Adhya and Paul Venetianer for encouragement and support; Nigel Minton for pNM vectors; Erzs~bet Bukva-HorvAt and Erzs~bet Zs6t& for technical assistance and Tom Bickle for critically reading the manuscript. This work was supported by grants to L.O. from the Hungarian Academy of Sciences, from the Ministry of Culture, and from the State Office of Technical Development (OTKA 691/86, Tt 177/1986, OMFB13/182/86).

Pure R1-77-~-gal The R1-77-B-gal protein was purified to homogeneity and its D N A binding activity in vitro was investigated by the dot blot binding assay (Materials and methods). In good agreement with the in vivo observations and with the results of the gel retardation experiments, the R1-77-~-gal protein specifically bound the D N A fragment carrying the intact OR operator but could not bind the mutant operator O~-1 (Fig. 3).

The carboxyl terminal region of the repressor is essential for lysogeny The RA156-161 repressor controlled tet gene expression from PR as efficiently the wild type repressor (Table 2, pDP62). The " i n f r a m e " deletion 156-161 lies in the carboxyl terminal region (Fig. 1). We constructed the RA156-16I allele in vitro by removing a XhoI-SalI fragment corresponding to 6 codons (Materials and methods). The mutant allele was then transferred to the phage chromosome (Fig. 4, Materials and methods) in order to test its phenotypic effect. Phages carrying the RA i56i61 allele formed clear plaques, indicating that the

References Aiba H, Krakow JS (1981) Isolation and characterization of the amino and carboxyl proximal fragments of adenosine cyclic 3',5'-phosphate receptor protein of Escherichia coli. Biochemistry 20:4774-4780 Anderson J, Ptashne M, Harrison SC (1984) Cocrystals of the DNA binding domain of phage 434 repressor and a synthetic phage 434 operator. Proc Natl Acad Sci USA 81:1307-1311 An G, Friesen JD (1979) Plasmid vehicles for direct cloning of Escherichia coli promoters. J Bacteriol 140:400-407 Bolivar F, Rodriguez RL, Greene PJ, Betlach HL, Heynecker HL, Boyer HW, Crosa JH, Falkow S (1977) Construction and characterization of new cloning vehicles. II. A multipurpose cloning system. Gene 2: 95-113 Boyer HW, Roulland-Dussoix D (1969) A complementation analysis of the restriction and modification of DNA in Eseherichia coli. J Mol Biol 41:459-472 Bowen B, Steinberg J, Laemmli UK, Weintraub H (1980) The detection of DNA binding protein by protein blotting. Nucleic Acids Res 8 : 1--20 Brent R, Ptashne M (1985) An eukaryotic transcriptional activator bearing the DNA specificity of a prokaryotic repressor. Cell 43 : 729-736 Casabadan MJ, Martinez-Arias A, Sharpira SK, Chou J (1983) /7-galactosidase gene fusion for analyzing gene expression in Escherichia coli and yeast. Methods Enzymol 100:293-308

112

Case ME, Giles NH (1975) Genetic evidence on the organization and action of the ga-1 gene product: A protein regulating the induction of three enzymes in quinate catabolism in Neurospora crassa. Proc Natl Acad Sci USA 72:553-557 Dallmann G, Olasz F, Orosz L (1980) Virulent mutants of temperate Rhizobium meliloti phage 16-3: Loci avirC and avirT, and increased recombination. Mol Gen Genet 178:443-446 Dallmann G, Papp P, Orosz L (1987) Related repressor specificity of unrelated phages. Nature 330: 398-401 Dorgai L, Olasz F, Ber6nyi M, Dallmann G, Pay A, Orosz L (1981) Orientation of the genetic and physical map of Rhizobium meliloti temperate phage 16-3. Mol Gen Genet 182:321325 Dorgai L, Polner G, J6nfis E, Garamszegi N, Ascher Z, P/ty A, Dallmann G, Orosz L (1983) The detailed physical map of the temperate phage 16-3 of Rhizobium meliloti 41. Mol Gen Genet 191:430-433 Dorgai L, Olasz F, N6meth K (1986) Lysogenic control of temperate phage 16-3 of Rhizobium meliloti 41 is governed by two distinct regions. Mol Gen Genet 205 : 568-571 Kania J, Brown DT (1976) The functional repressor parts of a tetrameric lac repressor-/~-galactosidase chimera are organized as dimers. Proc Natl Acad Sci USA 73:3529-3533 Kiss Gy, Dob6 K, Dusha I, Breznovits A, Orosz L, Vincze 1~, Kondorosi A (1980) Isolation and characterization of an Rprime factor in Rhizobium meliloti. J Bacteriol 141:121-128 Lieb M (1976) 2cI mutants: Intragenic complementation and complementation with a cI promoter mutant. Mol Gen Genet 146:291-297 Maniatis T, Fritsch EF, Sambrook J (1982) Molecular cloning. A laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY Mata-Gilsinger M, Ritzenthaler P (1983) Isolation of a functional ExuR-repressor-~-galactosidase hybrid protein by use of in vitro gene fusions. Gene 25:9-20 Messing J, Vieira J (1982) A new pair of M13 vectors for selecting either DNA strand of double-digest restriction fragments. Gene 19: 269-276 Miller JH (1972) Experiments in molecular genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, pp 3984O4 Minton NP (1984) Improved plasmid vectors for the isolation of translational lae gene fusions. Gene 31:269-273

Mi.iller-Hill B, Kania J (1974) Lac repressor can be fused to /~galactosidase. Nature 249:561-562 Norrander J, Kempe T, Messing J (1983) Construction of improved M13 vectors using oligodeoxynucleotide-directed mutagenesis. Gene 26:101-106 Ogata RT, Gilbert W (1978) An amino-terminal fragment of lac repressor binds specifically to lae operator. Proc Natl Acad Sci USA 75 : 5851-5854 Orosz L, Sv~b Z, Kondorosi A, Sik T (1973) Genetic studies in Rhizobiophage 16-3. I. Genes and functions on the chromosome. Mol Gen Genet 125:341-350 Orosz L (1980) Methods for analysis of the C cistron of temperate phage 16-3 of Rhizobium meliloti. Genetics 94:265--276 Orosz L, Rostfis K, Hotchkiss RD (1980) A comparison of twopoint, three-point and deletion mapping in the C cistron of Rhizobiophage I6-3, and an explanation for recombination pattern. Genetics 94: 249-263 Pabo CO, Sauer RT, Sturtevant JM, Ptashne M (1979) The repressor contains two domains. Proc Natl Acad Sci USA 76:t6081612 Pfahl M, Stockter C, Gronenborn B (1974) Genetic analysis of the active sites of lac repressor. Genetics 76 : 669-679 Platt T, Weber K, Ganem D, Miller J (1972) Translational restarts: AUG reinitiation of a lae repressor fragment. Proc Nat1 Acad Sci USA 69 : 89%901 Ptashne M (1986) A genetic switch; Gene control and phage 2. Cell Press/Blackwell Scientific Publications, Cambridge, Massachussetts Sanger F, Nicklen S, Coulson AR (1977) DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA 74: 5463-5467 Sauer RT, Pabo CD, Meyer BJ, Ptashne M, Backman KC (1979) Regulatory functions of the 2 repressor reside in the amino terminal domain. Nature 279:396-400 Singh H, LeBowitz JH, Baldwin AS, Sharp PA (1988) Molecular cloning of an enhancer binding protein: Isolation by screening of an expression library with a recognition site DNA. Cell 52: 415-423 Wilcken-Bergman B, Mfiller-Hill B (1982) Sequence of galR gene indicates a common evolutionary origin of lac and gal repressor in Escherichia eoli. Proc Natl Acad Sci USA 79:2427-2431 Communicated by W. Arber