IS3 can function as a mobile promoter in E. coli

volume 10 Number 19 1982 Nucleic Acids Research IS3 can function as a mobile promoter in E. coU Daniel Charlier, Jacques Piette and Nicolas Glansdorff Microbiologie, Vqje Universiteit Brussel, and Research Institute of the C.E.R.I.A., Avenue Emile Gryson, 1, B-1070, Belgium Received 5 July 1982; Revised 27 August 1982; Accepted 6 September 1982 INTRODUCTION In E_. coli, silent genes can be reactivated by insertion of an IS2, IS3 or IS5 element near the 5' end of their coding part (1,2,3,^,5,6). In the case of strong reactivations due to IS2 or IS3 this effect was shown to be orientation-dependent, the element being found in orientation II with respect to the polarity of transcription of the gene turned on. From the study of IS2 insertions proximal to the gal operon, it was originally concluded that IS2 carried a strong promoter able to readthrough into adjacent genes (1). IS2 was therefore visualized as the bacterial equivalent of a "controlling element" (1). However, since the gal reactivations had been obtained under selective pressure, alternative explanations were not excluded, such as the formation of a promoter by the juxtaposition of appropriate sequences at the junction between the element and the gal genes themselves. DNA sequence analysis of such a reactivated mutant and of derived gal negative segregants actually reinforced this suspicion (7,2). It also became clear that the mere insertion in orientation II of an IS2 element 5' © I R L P r w s Limited, Oxford, E r r a n d . 0305-1048/82/1019-593582.00/0 5936 ABSTRACT We had shown previously (3) that the E.coli argE gene could be turned-on by an IS3 element inserted in orientation II near the 5' end of the gene. Here we show that this effect is due to the presence of an outward promoter located on IS3The exact site of insertion of IS3 was determined by DNA sequencing. Using the S1 nuclease mapping technique with iji vivo transcribed RNA we located the promoter responsible for argE transcription on IS3 itself outside the region involved in the inverted repeats of this element. IS3 may therefore be considered as a mobile promoter. Nucleic Acids Research to a particular gene was not sufficient for strong downstream activation (8). Besides, Me Kenney and Rosenberg (personal communication) were able to show that an IS2 element inserted in orientation II with respect to a silent galK gene borne on the pKO plasmid (9) failed to activate the gal gene. Clearly, the idea that at least some IS elements would actually harbour a strong outward promoter capable of gene activation by readthrough transcription was still in need of experimental support. Here we show that one particular IS3 element answers this description and therefore that "mobile promoters" do exist. The IS3 investigated has been described previously (3) ; it activates the argE gene -the left arm of the argECBH divergent operon- in a strain where the natural promoter of that gene has been deleted. MATERIALS AND METHODS Bacterial strains, bacterlophages and plasmids used in this work are given in table 1. Conditions for growth and composition of media have been described (13). Ampicillin was used at 25 jjg/ml, tetracyclin as indicated in the text. The preparation of phage suspensions by thermoinduction of lysogens and the purification of transducing phages on CsCl gradients were performed according to ref.11. DNA and plasmid manipulations - Preparation of plasmid DNA and transformation conditions were 6936 There were basically two non-mutually exclusive hypotheses that could explain all the data : (i) turn-on would result from the formation of new promoters either by simple juxtaposition of sequences or by rearrangements which could occur at the time of insertion, such as errors during the DNA replication round involved in the transposition act (2) ; (ii) idividual IS elements of the same class could differ from each other in such a way that, in some cases, turn-on would result from the insertion of an IS element carrying an active "outward" promoter while other IS elements of the same class would be inert as such but could possibly engage in the process outlined in the previous sentence. Nucleic Acids Research Table 1. Bacteria P4XX~ P^XB2X" PUXMN42X" P'IXB2MN42X~ P4Xsup102 P4XB2sup102 C600galK" Genotype Source or reference Hfr,P4X,metB cured of X idem, argR~ Hfr,P4X,metB,AppcargECBHn042 idem, argR~ Hfr.PtX.metB, Asup102 idem, argR~ C600,galK~,lac~ 10 This laboratory 11 11 11 11 9 Phages X13AA4 Xi3,3up102: :IS3-H 3 Plasmids Ap r Tc r pBR322 pKO-1 pCDH-102 Tcr,galK gene without promoter pCD4-110 idem but with fragment in pBR322-670bp Hindlll, argE-IS3 fragment opposite orientation pCD4-152 9 This work idem pKO-1-670bp HindIII-arp;E-IS3 fragment pCD4-179 12 idem idem but with fragment in opposite orientation. idem Symbols and abbreviations as in ref. 10 ; A - deletion. described previously (15). "Mini-lysates" were performed according to Brinboim and Doly (16). - Construction of hybrid plasmids was described by Herschfield et al. (17). - DNA fragments were isolated from agarose or acrylamide gels respectively as described by Tabak and Flavell (18), and Maxam and Gilbert (19). 5937 Nucleic Acids Research Restriction enzymes, TM DNA ligaseand polynucleotide kinase were purchased from New England Biolabs, bacterial alkaline phosphatase from Worthington, and SI nuclease from Boehringer (Mannheim) DNA sequence analysis was performed according to Maxam and Gilbert (19). Cloning of the argE-IS3 junction in pBR322 and pKO-1 A 67Obp fragment starting at the Hindlll site of argE and extending approximately 200bp into IS3 (see fig.2) was purified on a 1.5 % agarose gel from a Hindlll digest of X13AA4 and ligated either with pBR322 or pKO-1 plasmid DNA predigested with the same endonuclease. pBR322 derivatives were transformed into strain P4XMN42X" by selecting for ampicillin resistant clones on complex medium supplemented with 25 yg/ml ampicillin. Only those clones which were sensitive to a high concentration of tetracyclin (60 pg/ml) were further screened on different concentrations of the same antibiotic, ranging from 2.5 to 30 yg/ml. On this basis they could be subdivided into two classes respectively resistant to 7.5 and 20 yg tetracyclin/ml. As shown by restriction enzyme analysis, this different behaviour was found to depend upon the orientation of the inserted fragment (see text and fig.3). pKO-1 derivatives were transformed into the g_alj(~ derivative of strain C600 ; here again selection was for Ap only. Colonies purified on the same medium were then screened on Mac Conkey galactose indicator plates ; cells carrying the same fragment as above in either orientation were already red after 18 hours incubation at 32°C while controls transformed with pure Hindlll cleaved and religated pKO-1 DNA remained white under these conditions. 5938 y- P ATP was from NEN Chemicals. Enzyme assays N-a-acetylornithinase (EC (...truncated)