We have identified a 34 kDa DNA‐binding protein with an HU‐like activity in the Lyme disease spirochete Borrelia burgdorferi. The 34 kDa protein is translated from an abundant transcript initiated within the gene encoding the A subunit of DNA gyrase. Translation of the 34 kDa protein starts at residue 499 of GyrA and proceeds in the same reading frame as full‐length GyrA, resulting in an N‐terminal‐truncated protein. The 34 kDa GyrA C‐terminal domain, although not homologous, substitutes for HU in the formation of the Type 1 complex in Mu transposition, and complements an HU‐deficient strain of Escherichia coli. This is the first example of constitutive expression of two gene products in the same open reading frame from a single gene in a prokaryotic cellular system.
The spirochete Borrelia burgdorferi, a causative agent of Lyme disease, is one of a few prokaryotes with a genome composed of linear as well as circular DNA molecules (Barbour and Garon, 1987; Hinnebusch and Tilly, 1993; Saint Girons et al., 1994). With a linear chromosome and up to 17 circular and linear plasmids, the B.burgdorferi genome may be the most complex of those found in the prokaryotes (Fraser et al., 1997). Whereas the telomeres of linear replicons of some prokaryotes such as Streptomyces and a number of phages contain covalently attached proteins, the linear replicons of Borrelia have covalently closed hairpin ends (Hinnebusch and Tilly, 1993). The only other prokaryotic linear replicon identified to date containing hairpin telomeres is the prophage of coliphage N15 (Svarchevsky and Rybchin, 1984; Hinnebusch and Tilly, 1993). The presence of these unusual hairpin‐containing linear replicons in the B.burgdorferi genome suggests that novel proteins may be involved in maintenance, replication and compaction of these molecules.
The small DNA‐binding proteins HU and integration host factor (IHF) function in the macromolecular metabolism of DNA to assemble higher‐order nucleoprotein complexes in other bacteria. HU and IHF, first identified for their roles in bacteriophage metabolism, are abundant in Escherichia coli, with concentrations of 30 000 dimers and 8500–17 000 dimers per cell, respectively (Rouviére‐Yaniv and Gros, 1975; Nash et al., 1977; Ditto et al., 1994). HU, which exists as a dimer encoded by hupA and hupB, binds DNA non‐specifically, while IHF, a dimer encoded by himA and hip (himD), binds at specific sequences (Nash, 1996). Many bacteria encode separate HU and IHF proteins; however, some prokaryotes, such as Bacillus subtilis, synthesize a single protein that functions in both roles (Micka et al., 1991). Hbb, an HU/IHF homolog in B.burgdorferi, was identified and shown to substitute for both HU and IHF using a genetic complementation assay in E.coli based on bacteriophage λ growth (Tilly et al., 1996).
In addition to hbb, the recently completed sequence of the B.burgdorferi genome identified a limited complement of genes encoding proteins involved in DNA replication, repair and recombination, similar to the genomes of Mycobacterium genitalium and Treponema pallidum (Fraser et al., 1995, 1997, 1998). Included within this set of borrelial genes are single copies of gyrA and gyrB, encoding the A and B subunits of DNA gyrase (GyrA and GyrB), respectively (Huang, 1992; Fraser et al., 1997). DNA gyrase, an A2B2 heterotetramer, is the only topoisomerase capable of introducing negative supercoiling into closed‐circular DNA (Gellert et al., 1976; Cozzarelli, 1980; Reece and Maxwell, 1991b; Wang, 1996). This reaction is dependent on the wrapping of DNA around the enzyme with positive writhe, which introduces negative writhe elsewhere in the molecule (Reece and Maxwell, 1991b). The domain of DNA gyrase that wraps DNA lies in the C‐terminal third of GyrA (Reece and Maxwell, 1991a). Removal of this C‐terminal domain abolishes the wrapping of DNA around the enzyme and eliminates the introduction of negative supercoiling, although not other enzymatic activities (Kampranis and Maxwell, 1996).
Pharmacological studies indicate the presence of a functional DNA gyrase activity in B.burgdorferi (Samuels and Garon, 1993). We report here that, in addition to synthesizing a functional DNA gyrase consisting of full‐length GyrA and GyrB subunits, B.burgdorferi also synthesizes the C‐terminal domain of GyrA as a distinct and abundant 34 kDa DNA‐binding protein. This unique protein forms a higher‐order nucleoprotein complex, substituting for HU in Mu transposition in vitro, and functions analogously to HU in vivo to support bacteriophage Mu growth in E.coli.
Identification of the GyrA C‐terminal domain
The 34 kDa GyrA C‐terminal domain was identified in a biochemical screen for telomere‐binding proteins in B.burgdorferi. A whole cell lysate was fractionated by heparin column chromatography, and fractions were screened for telomere‐binding activity by electrophoretic mobility shift assays (EMSAs) using a sub‐telomeric DNA fragment (Figure 1A and B). A DNA‐binding activity localized to fractions containing a 34 kDa protein, which was further purified to apparent homogeneity (Figure 1A). N‐terminal sequencing identified the 34 kDa protein as the GyrA C‐terminal domain, encompassing residues 500–810 of the full‐length GyrA protein (Figure 1C).
The large amount of the 34 kDa protein isolated, and the presence of a methionine codon, which could serve as a translational start codon, directly upstream of the N‐terminal serine residue (Figure 1C) suggested that the protein was synthesized as a distinct protein independent of full‐length GyrA. To determine whether the GyrA C‐terminal domain was present in intact cells, or whether the protein arose as a result of proteolytic cleavage during purification, whole cell B.burgdorferi lysates were examined by immunoblotting analysis using a polyclonal antiserum to a recombinant B.burgdorferi GyrA C‐terminal domain (Figure 2). The antiserum detected the 34 kDa GyrA C‐terminal domain, and an ∼91 kDa protein, predicted to be full‐length GyrA. The 34 kDa protein was present in crude lysates at a 5‐ to 20‐fold higher level than the 91 kDa protein. The cellular concentration of the GyrA C‐terminal domain was estimated to be 34 000 molecules per cell based on regression analysis of a standard curve generated from immunoblots with known concentrations of the GyrA C‐terminal domain (data not shown).
Analysis of the transcripts encoding the GyrA C‐terminal domain
To determine whether the 34 kDa protein was expressed in B.burgdorferi as a novel gene product, we used Northern analysis to identify a transcript encoding the GyrA C‐terminal domain. When using a probe specific for the 3′ end of gyrA, encoding the C‐terminal 310 amino acid residues, two separate transcripts were detected: a 1200 nucleotide and a 5000 nucleotide transcript (Figure 3A and B). The 1200 nucleotide transcript is large enough to encode the GyrA C‐terminal domain (310 amino acids); however, it is not large enough to encode full‐length GyrA (810 amino acids). A probe specific for the 5′ end of gyrA, encoding the N‐terminal domain of GyrA, only hybridized with the 5000 nucleotide transcript (data not shown). We hypothesize that the larger transcript is a bicistronic message encoding both GyrB (634 amino acids) and GyrA. This is supported by the hybridization of a gyrB‐specific probe to the 5000 nucleotide transcript (Figure 3C). As shown in Figure 3A, gyrB is located directly upstream of gyrA, with only 14 bp separating the two genes (Huang, 1992; Fraser et al., 1997). The 1200 nucleotide transcript is expressed at a 4‐fold higher level than the larger bicistronic message as assayed by quantifying the hybridization signals. We have also identified the GyrA C‐terminal domain‐specific transcript in a number of other species of Borrelia including B.garinii, B.afzelii and the relapsing fever spirochete B.hermsii, all of which contain predominantly linear genomes (data not shown).
Transcriptional start site analysis of the GyrA C‐terminal domain‐specific mRNA
Primer extension analysis was used to locate three putative transcriptional start sites for the 1200 nucleotide transcript located at nucleotides 1412, 1430 and 1435 of gyrA (Figure 4A). The presence of multiple transcriptional start sites is not uncommon in either B.burgdorferi genes or small DNA‐binding protein genes from other prokaryotes (Kohno et al., 1990; Marconi et al., 1993; Tilly et al., 1996). Although promoter sequences in B.burgdorferi have not been functionally defined, only the transcriptional start site at nucleotide 1412 has upstream sequences that resemble σ70 consensus promoters (Figure 4B). The absence of a consensus promoter sequence upstream of transcriptional start sites at nucleotides 1430 and 1435 is perhaps not unexpected, considering that the regions upstream of all three transcriptional start sites not only serve as regulatory sequences for transcription, but must maintain the open reading frame for full‐length GyrA. Two direct repeats of TATTAAT are also found in the promoter region (Figure 4B).
To confirm that the three sites identified were in fact 5′ ends of transcripts, we used a series of oligonucleotides flanking the three sites as probes in Northern hybridizations (Figure 4C). Five oligonucleotide probes complementary to regions downstream of the transcriptional start site at nucleotide 1435 all hybridized with both the 5000 nucleotide and 1200 nucleotide transcripts. However, two oligonucleotide probes complementary to regions upstream of transcriptional start sites at nucleotides 1430 and 1435 hybridized only with the 5000 nucleotide transcript. Oligonucleotide probe gyrA1441R, complementary to the region directly downstream of the transcriptional start site at nucleotide 1412, also failed to hybridize with the 1200 nucleotide transcript, suggesting that nucleotide 1412 is not used as often as the other two transcriptional start sites under standard culture conditions. These data are consistent with the primer extension data, which indicates that nucleotide 1430 serves as the major transcriptional start site (Figure 4A). The identified promoter region including all three transcriptional start sites for the 1200 nucleotide transcript functions in E.coli to regulate expression of a reporter gene (unpublished data) and to produce a functional GyrA C‐terminal domain as described below (Figure 6).
Nucleoprotein complex formation
EMSAs were used to examine DNA‐binding activity of the GyrA C‐terminal domain. As the amount of protein added to the binding reaction was increased, the mobility of the DNA–protein complex was further retarded (Figure 1B). This indicates that multiple protein molecules are capable of binding to each DNA molecule. Similar results were obtained using the ospAB promoter sequence from the 50 kb linear plasmid, an ospC coding sequence from the 26 kb circular plasmid and synthetic telomere sequences from the 17 kb linear plasmid, suggesting that the DNA‐binding activity of the protein is non‐specific (data not shown).
Based on the abundance of the GyrA C‐terminal domain and its non‐specific DNA‐binding activity, we hypothesized that the protein had an analogous function to the small DNA‐binding protein HU of E.coli. An in vitro Mu transposition assay was used to test this hypothesis. The Type 1 complex is an early intermediate in the transposition reaction, and its formation is stringently dependent on the presence of HU (Craigie et al., 1985). The GyrA C‐terminal domain substituted for HU in the in vitro reaction, as assayed by the formation of the Type 1 complex in Mu transposition (Figure 5A). The protein was capable of promoting the donor‐cleavage reaction over a wide range of concentrations (Figure 5B); however, IHF was required for the GyrA C‐terminal domain to substitute for HU in the reaction (data not shown).
Complementation of an HU mutant
Genetic manipulation of B.burgdorferi is in the initial stages of development, which have been slowed by a complex undefined medium and the availability of only one selectable marker (Samuels et al., 1994). Therefore, we investigated the in vivo activity of the GyrA C‐terminal domain in a more experimentally tractable system. We examined the ability of the gene encoding the GyrA C‐terminal domain to complement a bacteriophage Mu plating deficiency in an HU‐deficient strain of E.coli (A5196). HU, which is encoded by the hupA and hupB genes, is required for bacteriophage Mu to form plaques on E.coli. The GyrA C‐terminal domain was expressed in E.coli A5196 from plasmid pTASK1 using the natural B.burgdorferi promoter within the gyrA gene. Synthesis of the GyrA C‐terminal domain in E.coli was confirmed by Western analysis (Figure 6). Escherichia coli containing pTASK1 restored Mu growth as assayed by efficiency of plating (e.o.p.), which measured the ability of bacteriophage Mu to form plaques on the complemented E.coli hupA hupB mutant (Table I). Bacteriophage Mu readily formed plaques on wild‐type E.coli, but e.o.p. was severely reduced in the E.coli hupA hupB mutant. The gene encoding the GyrA C‐terminal domain complemented the hupA hupB mutant, restoring the Mu plating deficiency to wild‐type levels.
The GyrA C‐terminal domain in B.burgdorferi is the first example of a bacterium naturally synthesizing a domain of DNA gyrase as a separate protein, distinct from the subunits found in the holoenzyme. The synthesis of the GyrA C‐terminal domain is controlled by an expression system rarely found in a prokaryotic cellular genome: a smaller protein encoded on a separate transcript that is initiated within a full‐length gene and translated in the identical reading frame as a larger protein. In B.burgdorferi, translation of the GyrA C‐terminal domain is initiated at Met499 of GyrA, and transcription of the 1200 nucleotide mRNA encoding the protein is initiated from three separate start sites located within gyrA. The three start sites may be a means to regulate expression of the gene encoding the 34 kDa protein by using different start sites under different conditions. Based on the Northern and primer extension data, we predict that nucleotides 1430 and 1435, which are separated by one turn of the helix, are the two predominant start sites.
Expression of the gene encoding the 34 kDa protein may be complex considering that the promoter for the gene is located within gyrA, which is expressed at an ∼4‐fold lower level than the gene encoding the GyrA C‐terminal domain. The GyrA C‐terminal domain gene is also located within a gene (gyrA) that is regulated, in E.coli, by the level of DNA supercoiling (Menzel and Gellert, 1987). Adding to the complexity associated with transcription being initiated within another gene is the possibility that translation of the 34 kDa protein may be initiated from both the 1200 nucleotide transcript and the larger 5000 nucleotide bicistronic message.
As unexpected as the synthesis of this unique protein is the ability of the GyrA C‐terminal domain to substitute for HU in vitro and complement an HU‐deficient mutant of E.coli in vivo. Initial stages of Mu transposition both in vitro and in vivo require HU (Craigie et al., 1985; Huisman et al., 1989). The formation of the Type 1 complex necessitates the assembly of a multi‐protein complex consisting of Mu A, HU and IHF on a supercoiled donor molecule (Chaconas et al., 1996). We demonstrate here that the GyrA C‐terminal domain from B.burgdorferi substitutes for HU in forming the Type 1 complex in Mu transposition. The eukaryotic protein HMG1 will also substitute efficiently for HU in this reaction (Lavoie and Chaconas, 1994). HU and HMG1, although from different domains of life, share similar activities both in vitro and in vivo (Bianchi, 1994). The efficient substitution of the B.burgdorferi GyrA C‐terminal domain for HU in the Mu donor‐cleavage reaction indicates a specific HU‐like activity for the protein found only in the functionally related proteins HU and HMG1.
In E.coli, a 33 kDa protein consisting of the GyrA C‐terminal domain can be generated by proteolytic cleavage of GyrA or by expression of a recombinant gene product (Reece and Maxwell, 1989, 1991a). The naturally occurring protein from B.burgdorferi and the homologous region from the E.coli GyrA are biochemically distinct, sharing only 24% identity at the amino acid level. The GyrA C‐terminal domain from E.coli is acidic with a predicted isoelectric point (pI) of 4.0; in contrast, the naturally occurring 34 kDa protein from B.burgdorferi is basic with a predicted pI of 9.1. Like the GyrA C‐terminal domain of B.burgdorferi, the 33 kDa protein from E.coli binds DNA non‐specifically, with multiple protein molecules binding each DNA molecule (Reece and Maxwell, 1991a). Despite similar DNA‐binding activities, the recombinant 33 kDa GyrA C‐terminal domain from E.coli was not capable of substituting for HU under standard reaction conditions in the in vitro Mu donor‐cleavage reaction (data not shown), indicating that the B.burgdorferi GyrA C‐terminal domain possesses a unique HU‐like function that is not merely a consequence of binding DNA non‐specifically.
The GyrA C‐terminal domain not only forms higher‐order nucleoprotein complexes in vitro, but the gene encoding the GyrA C‐terminal domain complements an E.coli hupA hupB mutant as measured by bacteriophage Mu growth. This indicates that the protein also functions in forming these assemblies in vivo. The gene previously identified as encoding the HU/IHF homolog in B.burgdorferi, hbb, is incapable of complementing the defect in Mu growth found in hupA hupB E.coli strains (Tilly et al., 1996). However, hbb complements an HU/IHF‐deficient mutant as assayed by restoring a bacteriophage λ packaging deficiency (Tilly et al., 1996). Whereas λ growth requires either HU or IHF, Mu growth has a strict requirement for HU (Huisman et al., 1989; Mendelson et al., 1991). The previously identified HU/IHF homolog, Hbb, may be synthesized at much lower levels than the GyrA C‐terminal domain, since extensive attempts to purify Hbb from B.burgdorferi have been unsuccessful (Tilly et al., 1996). In addition, immunoblot analyses with an antibody against a synthetic peptide reveal little Hbb protein in cell lysates (S.W.Knight, C.Schwanke and D.S.Samuels, unpublished data). We report here that the GyrA C‐terminal domain is an abundant non‐specific DNA‐binding protein in B.burgdorferi with an activity and a cellular concentration similar to HU. Taken together, these data suggest that Hbb, although it shares sequence homology with HU and IHF, may have a more limited role in DNA metabolism than homologs found in other prokaryotes, and that the GyrA C‐terminal domain may provide the major HU‐like activity in B.burgdorferi.
The GyrA C‐terminal domain has not been identified as a naturally synthesized DNA‐binding protein in any other prokaryote outside of the Borrelia. However, the activities of DNA gyrase and HU are linked. HU stimulates DNA gyrase decatenation activity in vitro (Marians, 1987) and HU suppressor mutations have been mapped to gyrB (Malik et al., 1996). DNA gyrase and HU alter the superhelical path of the DNA helix using different mechanisms. DNA gyrase affects DNA topology by changing the linking number, and HU affects DNA writhe by bending or wrapping the helix. The Borrelia have apparently been able to join the supercoiling and bending activities of these two proteins more intimately by synthesizing the DNA gyrase domain partially responsible for wrapping DNA as a separate protein with an HU‐like activity. We speculate that this unique protein evolved as an adaptation for the presence of linear hairpin‐containing replicons. Alternatively, the unique expression of the GyrA C‐terminal domain gene from gyrA may be a means for Borrelia to compensate for its small genome size of ∼1.2 Mb by expressing domains of larger proteins as distinct functional products.
Materials and methods
GyrA C‐terminal domain purification
A crude lysate of high passage B.burgdorferi strain B31 was prepared from an 18 l culture grown in BSK‐H medium (Sigma) at 32°C. Nucleic acid was precipitated and the lysate was clarified as previously described (Margolis and Samuels, 1995). The clarified lysate was dialysed against 20 mM Tris–HCl pH 8 (the pH of Tris solutions was measured at 25°C), 0.5 mM EDTA, 10% glycerol (H buffer) and loaded onto a 5 ml Econo‐Pac heparin cartridge (Bio‐Rad). The column was eluted with a 100 ml linear gradient from 0–1 M NaCl in H buffer at 2 ml/min. Fractions containing the 34 kDa protein were dialysed against 10 mM sodium phosphate pH 6.8, in 10% glycerol and loaded onto a 2 ml CHT2‐I hydroxyapatite column (Bio‐Rad). The column was eluted with a 20 ml linear gradient from 10 to 500 mM sodium phosphate pH 6.8, in 10% glycerol, followed by a 10 ml linear gradient from 500 to 1000 mM sodium phosphate pH 6.8, in 10% glycerol at 1 ml/min. Fractions containing the 34 kDa protein were dialysed against 10 mM sodium phosphate pH 6.8, 0.5 mM EDTA, 10% glycerol (P buffer) and loaded onto a 1 ml Mono‐S column (Pharmacia). The column was eluted with a 15 ml linear gradient from 0 to 0.5 M NaCl in P buffer, followed by a 5 ml linear gradient from 0.5 to 1 M NaCl in P buffer at 1 ml/min. Fractions containing the 34 kDa protein were pooled and mixed with an equal volume of 3.4 M (NH4)2SO4, 100 mM sodium phosphate pH 7, 10% glycerol, 1 mM EDTA, and loaded onto a 1 ml phenyl Superose column (Pharmacia). The column was eluted with 15 ml of a 1.7–0 M (NH4)2SO4 and 50–10 mM sodium phosphate pH 7, linear gradient in 10% glycerol, 1 mM EDTA at 0.5 ml/min. Fractions containing the 34 kDa protein were dialysed against 50 mM Tris–HCl pH 7.5, 0.1 M KCl, 0.1 mM EDTA, 5 mM DTT, 50% glycerol and stored at −20°C.
Electrophoretic mobility shift assay
A 255 bp PCR product amplified from the 17 kb linear plasmid beginning 23 bp from the left end was used as an EMSA substrate. The substrate was prepared as previously described (Margolis and Samuels, 1995) using the following primers kindly provided by K.Tilly: 10 pmol of TL16g (5′‐AGACTAATAAAATAATGAATA‐3′) end‐labeled with [γ‐32P]dATP and 25 pmol of TL16h (5′‐GTATTTTGACTCAAAACTTTA‐3′). Approximately 2.1 μg of labeled DNA was incubated at 25°C for 30 min with the 34 kDa GyrA C‐terminal domain in 10 mM HEPES pH 7.5, 90 mM KCl, 300 μg/ml BSA, 200 μg/ml poly(dI‐dC), 10% glycerol. DNA–protein complexes were resolved by electrophoresis through a 4% polyacrylamide gel and visualized by autoradiography.
Whole cell B.burgdorferi lysates were prepared by boiling cells in SDS–PAGE loading buffer (125 mM Tris–HCl pH 6.8, 4% SDS, 20% glycerol, 1.4 M 2‐mercaptoethanol and 0.2% bromophenol blue). Samples were fractionated by 7.5% SDS–PAGE. For Western analysis, samples were transferred to a PVDF membrane and incubated with rabbit antiserum raised to a recombinant GyrA C‐terminal domain, followed by goat antibody to rabbit immunoglobulin G conjugated with horseradish peroxidase. The size and intensity of bands were determined using a Gel Doc 1000 system with Multi Analyst software (Bio‐Rad).
Total RNA was isolated from cultures using TRIzol™ reagent as described by the manufacturer (Gibco‐BRL). Total RNA (15 μg) was fractionated on 1.2% formaldehyde–agarose gels and transferred to nylon membranes. Hybridization with oligonucleotide probes (Marconi et al., 1993) and with PCR probes (Brown, 1996) was performed as previously described. The intensity and size of bands were determined using an Image Acquisition and Analysis system (Ambis) and a Gel Doc 1000 system with Multi Analyst software (Bio‐Rad), respectively.
Primer extension was used to map the RNA 5′ ends for the 1200 nucleotide transcript encoding the GyrA C‐terminal domain. One picomole of end‐labeled primer was incubated with 7 μg of total cellular RNA in 50 mM KCl, 10 mM Tris–HCl pH 8.3, 5 mM MgCl2, 1 mM each dNTP and 20 U RNase inhibitor. Extension was performed at 42°C with 50 U of MuLV reverse transcriptase. Extension products were resolved on a 6% polyacrylamide sequencing gel, and sizes were determined by use of a DNA sequencing ladder. Two different primers were used in the primer extension assay: gyrA1510R (5′‐GCATTAAATCCGACATAC‐3′) and gyrA1547R (5′‐CCTTTCTTTGTAAGCATAACAAC‐3′).
Mu donor cleavage reaction
Mini‐Mu donor plasmid pBL07, HU, IHF and Mu A were kindly provided by G.Chaconas. Type 1 in vitro reactions were performed as previously described with all reaction components at twice the standard concentrations (Naigamwalla and Chaconas, 1997). The GyrA C‐terminal domain, when substituted for HU, was used at a concentration of 7.5 μg/ml or as indicated. After incubation at 30°C for 5 min, the reactions were divided into equal aliquots and 1% SDS was added to one aliquot. Reactions were resolved on 1% agarose gels in 1× TAE buffer, stained with ethidium bromide (0.5 μg/ml) for 30 min and destained in H2O for 30 min.
The portion of the gyrA gene encoding the GyrA C‐terminal domain, including the promoter, was cloned into the plasmid vector pCR2.1‐TOPO (Invitrogen). Primers gyrA1213F (5′‐AAAGATGCAAGGGAGAGGC‐3′) and gyrA2433R+SphI (5′‐AGCATGCTTATTTAATAAATTTTGA‐3′) were used to PCR amplify nucleotides 1213–2433 of gyrA from B.burgdorferi strain B31. The 1220 bp product was cloned into the vector resulting in plasmid pTASK1 and transformed into TOP10F′ cells according to the manufacturer's instructions (Invitrogen). Plasmid pTASK1 was isolated and transformed into competent E.coli strain A5196 (hupA hupB) (Mendelson et al., 1991). Synthesis of the 34 kDa GyrA C‐terminal domain in E.coli strain A5196 was confirmed by Western analysis as described above using 12.5% SDS–PAGE.
Induction of Mu cts62 lysogens (kindly provided by K.Drlica) and preparations of lysate were performed as previously described (Bukhari and Ljungquist, 1977). Mu titers were measured by plaque formation on N99 lawns at 37°C on 1% Luria–Bertani (LB) Ca Mg plates (LB supplemented with 1 mM CaCl2, 2.5 mM MgSO4) with 0.5% LB top agar. The e.o.p. was measured as the ratio of bacteriophage Mu titer on the mutant strain to the titer on the wild‐type strain.
We thank Kit Tilly, George Chaconas, Tony Maxwell and Tasha Knight for thoughtful and critical reading of the manuscript, George Chaconas, Karl Drlica, Kit Tilly, Tony Maxwell, Rich Marconi, Jason Carlyon, Scott Manning, Wai Mun Huang and Bill Holben for useful discussions, Joan Strange for protein sequencing, Doug Emlen and Corbin Schwanke for help with figure preparation and Betsy Kimmel for technical assistance. Work in our laboratory is supported by grants from the National Science Foundation (MCB‐9722408), the National Institutes of Health (AI41559 and AI39695), MONTS (NSF EPSCoR), the Arthritis Foundation and The University of Montana University Grant Program.
- Copyright © 1999 European Molecular Biology Organization