Mol Gen Genet (1992) 235:131-139 © Springer-Verlag 1992

FinOP repression of the F plasmid involves extension of the half-life of FinP antisense RNA by FinO Stuart H. Lee 1, Laura S. Frost 2, and William Paranchych 2 Department of Biochemistry 1 and Microbiology2 University of Alberta, Edmonton, Alberta, Canada, T6G 2E9 Received January 13, 1992 / Accepted May 7, 1992

Summary. The transfer operon of the F plasmid is positively regulated by the traJ gene product, expression of which, in turn, is regulated by both an antisense RNA, FinP, and the FinO protein (the FinOP system). AfinPF plasmid, pSLF20, was constructed by site-directed mutagenesis and was found to produce wild-type levels of pili encoded by the transfer operon. Transcription of the traJ gene was decreased by a factor of 3-5 fold in the presence of FinOP with no accumulation of a stable R N A duplex between the FinP R N A and the portion of the traJ m R N A which is complementary to finP. Stabilization of FinP R N A by FinO occurs in the absence of traJ transcripts, suggesting that FinO may interact directly with FinP to prevent its degradation.

Key words: F Plasmid - D N A transfer - R N A stability - Fertility inhibition - Antisense R N A

Introduction

Transfer of the F plasmid is critically dependent on the expression of the F pilus, an extracellular, filamentous organelle that identifies a suitable recipient cell and has a role in the initiation of D N A transfer from oriT (the origin of transfer); reviewed in Ippen-Ihler and Minkley 1986; Willetts and Skurray 1987). The F transfer operon contains the genes for pilus assembly and function and is positively regulated by the TraJ gene product, encoded immediately upstream of the major promoter for the transfer operon, Py (Gaffney et al. 1983; Silverman et al. 1991a, b). The 5' untranslated portion of the traJ transcript is 105 bases long and is complementary to the antisense R N A FinP (80 bases in length; Fee and Dempsey 1986) which is encoded on the opposite strand (Mullineaux and Willetts 1985). The FinP R N A has been proposed to form two stem-and-loop structures with the allelic specificity of finP, first defined by Willetts and Correspondence to." L. Frost

Maule (1986) residing in sequence differences in the loops (Finlay et al. 1986a; Koraimann et al. 1991). In addition, a second locus, finO, at the distal end of the transfer operon, which encodes a 22 kDa protein, has also been shown to be essential for repression of traJ (Finnegan and Willetts 1971). The site of action of FinOP, defined by the fisO mutation, has been mapped within the finP gene (Finnegan and Willetts 1973). The F plasmid is naturally derepressed due to the presence of an IS3 element within the finO gene (Cheah and Skurray 1986; Yoshioka et al. 1987) and is complemented byfinO supplied in trans, typically from plasmids R100 or R6-5. The derepression of the F transfer operon by the loss offinO underscores the importance offinO in FinOP repression. Site-directed mutagenesis and Northern blot analysis of cloned finP andfisO mutant genes suggest that the first stem of FinP R N A is important for recognition by FinO and that FinO may act by stabilizing FinP (Frost et al. 1989). In addition, FinP is thought to act as an antisense R N A by occluding the ribosome binding site for traJ translation. The role of FinO in catalyzing duplex formation between the 5' untranslated portion of the traJ m R N A and FinP, which is superficially similar to the control of the initiation of replication of ColE1, where the Rom protein catalyzes the interaction of R N A I and R N A II (Tomizawa 1987), is unknown. In order to understand the role of FinO in FinOP repression of traJ, we have constructed a series of clones with and without the traJ andfinP promoters, containing the minimal sequences in finP and traJ needed to affect F transfer operon expression, as assayed by their ability to alter the mating efficiency of the F plasmid in the presence and absence of FinO. Also,finP and traJ RNA, expressed from foreign promoters, were assayed for activity, with the goal of designing an in vitro system for studying FinO action. We have also created and characterized afinP- F plasmid to simplify FinP R N A quatitations and examined the fate of the traJ R N A in the presence of FinOP. We also report the apparent stabilization of FinP R N A by FinO in the absence of traJ transcripts.

132 Materials and methods

Bacterial strains, plasmids, and media. Strain JC3272 containing the F plasmid, JCFLO, was named M176 (Achtman et al. 1971). The recipient strain used in mating efficiency assays was Escherichia coli ED24 (Lac Spc r T6 r PV). The m e - strain, E. coli N3433, the wild-type E. coli N3431 (Goldblum and Apirion 1981), and the rncstrain E. coli AO160, a mutant of E. coli AO159 (Schmeissner et al. 1984), were provided by E.A. Mudd (University of Geneva) and are described in detail in Mudd et al. (1988). E. coli CQ24 ( F - ara leu lacIq lacZ: : Tn5 p u r e 9al his argG rpsL xyl ilv thi (Km * Smr)) was obtained from K. Ippen-Ihler (Texas A & M University) and is described in Edlind et al. (1986). The cloning vectors employed were pUC 18 (YanischPerron et al. 1985), pTTQ18 (Stark 1987), pT7-3 (Tabor and Richardson 1985), pGEM3Z (Promega, Madison, Wis.) and pBS (Stratagene Corp., La Jolla, Calif.). A list of the constructs used in this study is given in Fig. 1. Details of construction are available from the authors upon request. The finO product was supplied from pED104, a derivative of pACYC177 carrying a 4.0 kb PstI fragment from the R6 plasmid and expressing Km ' (Frost et al. 1989). For use in E. coli CQ24 this PstI fragment was transferred to pACYC184 which expresses Cm ~, and the resulting plasmid was named pSnO104. Bacteria were grown on Luria-Bertani (LB) media (Sambrook et al. 1989), in the presence of suitable antibiotics. M9 minimal salts medium was used as described in Sambrook et al. (1989). Ampicillin was used at a concentration of 100 ~tg/ml, kanamycin (Km) at 25 gg/ml, streptomycin (Sm) at 200 gg/ml, and spectinomycin (Spe) at 100 gg/ml. All antibiotics were supplied by Sigma (St. Louis, Mo.). Recombinant DNA techniques, sequencin# and site-directed mutaoenesis. Plasmid isolations were performed according to the method of Birnboim and Doly (1979). All manipulations with D N A were carried out as described in Sambrook et al. (1989) and Ausubel et al. (1987 with updates). All enzymes were supplied by Boehringer Mannheim (Laval, Que.) except for T4 ligase and Klenow fragment, which were supplied by BRL (Bethesda, Md.) Double-stranded D N A sequencing was performed according to Wallace et al. (1981) and site-directed mutagenesis of double-stranded plasmid D N A was as described in Frost et al. (1989). RNA Isolation. R N A was isolated and electrophoresed on denaturing agarose and polyacrylamide gels as previously described (Frost et al. 1989). R N A was dissolved in a buffer containing 5 mM TRIS-HC1, 1 mM EDTA, pH 8 and ethidium bromide at a concentration of 0.5 Ixg/ml and quantitated in a Hitachi F-2000 fluorimeter set for excitation at 525 nm and emission at 595 nm. R N A was transferred from gels to Zeta Probe (BRL) nylon membrane according to the manufacturer's directions, using a Hoeffer TE 50 Transphor electrophoresis unit. The R N A was transferred for 30 min at 15 V and a further 60 rain at 25 V in 0.5 × TBE buf-

fer at 4°C (1 x TBE buffer is 9 0 m M TRIS-borate, ZmM EDTA pH 8.0). The in vitro synthesis and use of RNA probes and endlabelled oligonucleotide probes. The plasmid of interest (0.5-1 ~tg) was linearized with the appropriate enzyme, extracted with phenol and precipitated with three volumes of ethanol. The D N A was then dissolved in the reaction mixture supplied by Promega except that the 5 × in vitro transcription buffer was according to the Boehringer-Mannheim protocol, and [a32p]CTP (New England Nuclear/DuPont, Missisauga, Ont.) was the only source of CTP. The labelled nucleotide had a specific activity of 800 Ci/mmol, and was supplied at a concentration of 40 ~tCi/~tl; 4 ~tl were used in each reaction. Transcription was allowed to progress for 45-60 rain at 37 ° C, then 27 units of RNase-free DNase I were added and allowed to digest the template for 15-20 rain. T7 and T3 R N A polyrnerases were supplied by Boehringer Mannheim and RNAguard was supplied by Pharmacia (Montreal, Que.). The probe was purified through a Nuctrap column (Stratagene). Blots were prehybridized for at least 3 h at 58 ° C in 50% formamide, 2.5 x SSC, 5 × Denhardt's solution (Sambrook et al. 1989), 1.5% SDS, and 100 ~tg/ml of E. coli strain W tRNA type XX (Sigma). The blots were probed at 58°C with 106 cpm/ml hybridization solution with the addition of 200 ~tg/ml heat-denatured calf thymus D N A (Sigma) and 200 ~tg/ml tRNA. The blots were hybridized for at least 9 h before washing for 5 min in 2 × SSC at room temperature, 10 rain in 2 × SSC, 0.1% SDS, 10 rain in 0.5 × SSC, 0.1% SDS, and 10 min in 0.1 × SSC, 0.1% SDS, at 55 ° C. Autoradiography was performed at - 70 ° C in the presence of an intensifying screen, using Kodak X-AR5 film. FinP was detected on blots using a 32p-labelled oligomer which was complementary to the first 13 bases of FinP, contained a single base sequence difference from wild-type FinP (Primer A, Frost et al. 1989) and was found to hybridize exclusively to FinP transcripts with little or no background. The primer was labelled with 32p-ATP using T4 polynucleotide kinase according to the method described in Sambrook et al. (1989) and the hybridization conditions have been described in Frost et al. (1989). Random primer labelling was performed as described in Sambrook et al. (1989). 1PTG inductions. Cells were grown to a n O D 6 o o of 0.5-0.75. IPTG (isopropylthiogalactoside, Sigma) was added to a final concentration of 1 mM from a 100 mM stock solution. After 1 rain of induction with IPTG, rifampicin (Sigma) was added to a final concentration of 200 ~tg/ml, from a fresh stock solution (20 mg/ml of methanol). At appropriate intervals, 1 ml samples were withdrawn and added to ice-cold tubes containing 10 ~tl of 1 M sodium azide and 40 ~tl of 10 mg/ml chloramphenicol. The cells were kept on ice until the end of the experiment; they were then centrifuged and the cell pellets quick-frozen at - 7 0 ° C. R N A isolation was by the hot phenol method as previously described (Frost et al. 1989).

133

l,f °

1 Plasmld Name

fin.

traM

orIT

600

I 'i ° , I

I

tray

I t[lll tl,lL,

8 0 I>g00 1000 1100 1200

1500

I,I

I I I

I

"1

2000

I

I I

Clone Boundaries Vector

,4--]~=° pSQ150

Rsal-gau3A

pTTQ18

pSBP150

Rsal-Sau3A

pBS

pSQ180

Rsal-Bglll

pTTQ18

pSQ350

HpalI-Bglll

pTTQ18

pLF400

HpalI-Bglll

pUC18

pSQ351

HpalI-Bgll]

pTTQ18

pLF401

HpalI-Bglll

pUC18

pSQ1150

BOIII.Sau3A

pTTQ18

pSQ1200

BOIII-BglII

pTTQ18

pNY300

BgllI-Bglll

pUC18

pT7.300

BgllI-Bglil

pT7-3

pSBM650

AhallI-Ahalll

pBS

pSJ99

RsaI.Rsal

pUC18

pSJ88

RsaI-Rsal

pUC18

pSJ39

RsaI.Rsal

pGEM3Z

,t,.J

-%¢ 'tr. oriT

.q.t.°

P t r aMF'-~:~

orlT

PtrsM~

orlT

P t raM~.4~

oriT

P tra~

Ptr'd~ "~]PHnP