INTRODUCTION

Top Abstract Introduction Procedures Results Discussion References

Following infection of a host bacterium by a temperate bacteriophage, the phage can develop in one of two ways: lytically, whereby progeny phage are produced and released by lysis of the host cell, or lysogenically whereby the phage genome is integrated into the host cell (the prophage state) and replicated along with the host DNA. Although these two life cycles are independent, some temperate phage retain the ability to switch between them. For example, in certain classes of temperate bacteriophage, exposure to UV light or other DNA-damaging agents stimulates the phage to switch from the passive prophage state to the productive lytic cycle, the process of prophage induction.

In the lambdoid family of bacteriophage, commitment to prophage induction after UV irradiation is achieved by inactivation of the prophage maintenance function, CI, through linkage to the complex global stress response system known as the SOS response (1). Upon exposure of a cell to DNA-damaging agents, host RecA protein is converted to an activated form that catalyzes by autoproteolysis the inactivation of host cell-encoded LexA protein, a transcriptional repressor of a series of unlinked genes, the SOS genes (2). Activated RecA also catalyzes the self-cleavage of lambda CI repressor protein, relieving repression of lytic genes to bring about prophage induction (3, 4).

Coliphage 186 is a close relative of the non-inducible P2 family of phage (5). The finding that 186, like lambda, is UV-inducible (6) was therefore unexpected. Induction of a 186 prophage differs from that in lambda. RecA is not directly involved in induction of a 186 prophage, but rather 186 relies upon a phage encoded function tum, whose expression is under host LexA control (7, 8). Thus, prophage induction in 186 can be considered an SOS function. The SOS operon of 186 (Fig. 1) contains two genes under control of the p₉₅ promoter, tum and orf97 (8). The tum gene contains several alternative translational start points, and the product of one of these (Tum) is essential for prophage induction. Expression of the orf97 gene product from a plasmid blocks infection of the cell by 186; however, its absence has little effect on prophage induction (8).

View larger version (27K): [in this window] [in a new window]   Fig. 1.   Organization of the major control region and SOS operon of coliphage 186. The organization of the 186 genome is shown, with the early region from the PstI site (65.5%) to the BssHII site (76.8%) (9-11) and the SOS operon from 93.8 to 100% (7, 8) enlarged to show detail. Genes are shown as boxes (rightward genes above the line and leftward genes below the line), promoters as arrowheads, their transcripts as arrows, and terminators as stem loops. cII is the gene required to establish lysogeny; apl is required for prophage excision, and during induction, int is the integrase; 69 is of unknown function and the product of the B gene activates transcription of the late genes. CI binding sites, pR, pB, FL, and FR (12) are shown as solid boxes. The LexA-binding site over p95 (7) and the CII binding site at the start of the cII gene (19) are also indicated. The SOS operon consists of two genes, tum whose product brings about prophage induction and orf97 which is of unknown function (8).

The prophage state of coliphage 186 is, like lambda, maintained by the product of a single gene, CI (13). 186 CI represses transcription from two promoters, p_R and p_B (Fig. 1). The lytic promoter p_R directs the synthesis of genes required for excision and replication (10, 11), whereas p_B is responsible for expression of the B protein, a transcriptional activator required for expression of the late genes (14). In addition, there are two more CI binding sites, FL and FR, whose functions are unknown (12) but which may be involved in DNA looping. Exposure of a 186 lysogen to a DNA-damaging agent leads to production of the activated form of RecA, which in turn catalyzes the irreversible inactivation of host LexA, resulting in a general SOS response, including relief of the LexA-mediated repression of p₉₅. The resulting Tum protein then acts in an unknown fashion to activate lytic development (7, 8).

In the present work, we investigate the mechanism of Tum action and demonstrate that Tum acts against the CI repressor, the sole phage function required for maintenance of lysogeny. By using Tum and CI purified to near homogeneity, we show that Tum specifically prevents effective CI binding to its operator sites and that this Tum-CI interaction is reversible.

	EXPERIMENTAL PROCEDURES

Top Abstract Introduction Procedures Results Discussion References

Radiolabeled nucleotides and oligonucleotide primers were purchased from Bresagen (Adelaide, Australia). Media and general techniques for 186 are as described (15). General cloning techniques were as described (16).

Oligonucleotides

Primer A is (2872-2843) TAACGATAGGTGCAGGCACTTTGATGATTG. It is used in primer extension to detect p_R activity.

Primer B is (349-320) GATAGCGGCTTGTTCGCGCATGTGCGGCAT. It is used in primer extension to detect p_B activity.

Primer C is CGTAGTGGAGGTCATATGGATAGAGAGCT. It is used to create the NdeI site (shown in bold) at the start of the tum gene.

Primer D is GTCCCCGCGGTACGAGACGCCAGCTCTCATCTTC. It is used to remove the stop codon and introduce a SacII site (shown in bold) at the 3' end of the tum gene.

Clones

pEC303-- pEC303 is a multicopy p₉₅ tum⁺ plasmid. The 2.5-kb¹ EcoRI fragment (92 to 2.3%) from 186⁺ was ligated into the EcoRI site of pBR322.

pEC306-- pEC306 is a single copy p₉₅ tum⁺ plasmid. The 2.5-kb EcoRI fragment (92 to 2.3%) from 186cI⁺97 was ligated into the EcoRI site of pKN1562 (17).

pEC307-- pEC307 is a multicopy p₉₅ tum⁺ plasmid. The 2.0-kb PstI fragment (8578-10,597) from 186tum⁺ was ligated into the PstI site of pBR322. The orientation of the insert was such that p₉₅-directed transcription was in the opposite direction to p_Amp.

pEC308-- pEC308 is same as pEC307, except the 186 DNA source was 186cI⁺tum16.

pEC309-- pEC309 is same as pEC307, except the 186 DNA source was 18695.

pEC324-- pEC324 is a derivative of the pL expression plasmid pPLc28 containing the 3.1-kb EcoRI restriction fragment from an M13 Ecol clone (M13 Ecol 95.4/5) in which Orf95.4 and Orf95.5 expression had been abolished by mutation (8). The orientation is such that p_L transcription was in the same direction as p₉₅.

pEC330-- pEC330 is a p_lac expression plasmid carrying p15A replicon and confers kanamycin resistance. It was constructed as described (8).

pEC331-- The 1.2-kb SspI (8655) to AccI (1372) fragment (end-filled) from pEC303 was inserted into the ScaI site of pEC330. The insert was oriented for expression of Tum from p_lac.

pEC342-- The 1.3-kb HaeIII (1721) to HaeIII (2998) restriction fragment from 186cItsp was ligated into the HindIII site of a single copy pKO2-derived plasmid.

pEC363-- pEC363 is a pET3a-Tum expression vector. It was created by ligating the 300-bp NdeI (8765) to BamHI (9104) restriction enzyme fragment from the mutated M13 clone, M13EcorNdeTum⁺ ligated into the NdeI to BamHI backbone of pET3a. The carboxyl-terminal portion of the tum gene was reconstructed by ligating the XhoI (9104) to XhoII (9730) restriction fragment from M13EcorNdeTum⁺ into the recreated BamHI site of the pET3a construct.

pEC364-- pEC364 is a pET3a Tum16 expression vector. It is the same as pEC363, except the 186 DNA source was the M13 clone, M13EcorNde tum16.

pEC452-- pEC452 is a CI expression plasmid. It contains the 1.3-kb HaeIII-HaeIII (71.2-75.5%) fragment of 186⁺ inserted into the EcoRV site of pACYC184 such that CI expression was under the control of its own promoter, p_L (14).

pEC458-- pEC458 is a single copy B expression plasmid. It was constructed as described (14) such that B expression was under the control of its own promoter, p_B.

pET TumHis₆-- The tum gene from pEC324 (Orf95.4,.5 (8)) was amplified by PCR using primer C to create an NdeI site at the 5' end of the gene and primer D to create a SacII site at the 3' end of the gene. The PCR product was digested with NdeI and SacII and ligated into the NdeI, SacII backbone of pET3a which had been modified (H. Healy, University of Adelaide) to fuse the coding sequence for LVPRGSHHHHHH at the carboxyl terminus of any inserted sequence.

pET Tum16His₆-- pET Tum16His₆ was the same as pET TumHis₆ except the template for the PCR reaction was pEC364.

Primer Extension Analysis

RNA isolation was based on the procedure of Reddy and Gilman (18). The concentration and purity of RNA preparations were determined spectroscopically, an A₂₆₀/A₂₈₀ ratio of greater than 1.7 being regarded as acceptable for primer extension analysis. Primer extension reactions were performed essentially as described (19). All reactions were extended from oligonucleotides that had been labeled at the 5' end using polynucleotide kinase and [-³²P]dATP. Electrophoresis was carried out on a 6% denaturing polyacrylamide gel, and the gel was fixed, dried, and autoradiographed.

Nitrosoguanidine Mutagenesis

A lysogenic culture, C600(186cItsp) or (186cItspSA), was grown overnight at 30 °C in LB, and then diluted 10² in LB. Incubation was continued at 30 °C until A₆₀₀ = 0.8, and the culture was transferred to 39.5 °C for 20 min. Nitrosoguanidine was added (8 µg ml¹ or 16 µg ml¹), and incubation was continued for a further 20 min. Following cell lysis, cellular debris was removed by centrifugation, and aliquots of lysate were diluted 10³ into LB and stored over chloroform at 4 °C.

Protein Purification

TumHis₆-- Cells (BL21 [DE3] pLysS pET TumHis₆) were grown with vigorous shaking at 37 °C in LB (500 ml) containing ampicillin (100 µg ml¹) and chloramphenicol (30 µg ml¹) until an A_600 nm of 0.6-0.8 had been reached. IPTG (0.4 mM) was added to induce Tum expression and growth continued for a further 3 h. The cells were collected by centrifugation, washed once in 50 mM Tris-Cl, pH 8.0, 0.1 mM EDTA, 150 mM NaCl, 10% glycerol (TEG150 buffer), and stored at 20 °C until use.

1

1

1

1

1

1

1

Gel Retardation Assays

Radiolabeled DNA fragments for gel retardation assays were generated via the polymerase chain reaction, by inclusion of [-³²P]dATP in the PCR reaction mix. Fragments were purified on a 6% polyacrylamide gel before use. Samples (10 µl), containing 150-300 cpm of labeled fragment, were prepared in binding buffer (50 mM Tris-Cl, pH 8.0, 0.1 mM EDTA, 150 mM NaCl, 20% glycerol, 0.01 mM dithiothreitol, 50 ng µl¹ salmon sperm DNA) and incubated on ice for 30 min prior to loading. A separate lane of tracking dye was used. Gels (6% polyacrylamide containing 20% glycerol) were run at 4 °C. Electrophoresis was carried out at 20 mA for approximately 2 h, and the gels were dried under vacuum and the distribution of labeled DNA recorded on a phosphor screen. The screen was analyzed with the Imagequant program on a Molecular Dynamics PhosphorImager.

Western Blotting

Western blots were performed essentially as described (22), using the Supersignal chemiluminescent detection kit (Pierce).

Analytical Ultracentrifugation

Sedimentation equilibrium experiments were carried out in a Beckman XL-A analytical ultracentrifuge equipped with an An60Ti rotor and absorption optics. Samples were centrifuged for 24 h at 5 °C and scans taken every 4 h thereafter until equilibrium had been reached. Data sets were collected at 280 nm with a spacing of 0.001 cm as the average of three scans. Proteins were prepared for centrifugation by exhaustive dialysis against TEG150 buffer. The partial specific volume of TumHis₆ was calculated using the amino acid partial specific volumes of Zamyatnin (23). This value was 0.721 ml g¹. Buffer density at 5 °C was measured in an Anton-Paar precision density meter to be 1.0378 g ml¹.

	RESULTS

Top Abstract Introduction Procedures Results Discussion References

Derepression of p_R following UV Induction of Prophage-- No function other than CI is required for maintenance of the prophage state (13), and CI is known to repress transcription of the excision and replication genes from p_R (10) and transcription of the late activator gene from p_B (14). The action of Tum in prophage induction could be to activate transcription that bypasses the repressed p_R and p_B, or more likely, it could be that Tum interferes with CI function or production.

View larger version (80K): [in this window] [in a new window]   Fig. 2.   Derepression of pR following UV irradiation of a 186 lysogen. Bacterial strains of a C600(186) lysogen and a C600 nonlysogenic control were grown with aeration at 37 °C in L broth to 1.5 × 108 colony-forming units/ml. After centrifugation the cells were resuspended in an equal volume of 10 mM MgSO4 and UV-irradiated (45 J per m2). RNA was extracted from 5-ml samples at various times after irradiation and used for primer extension as described under "Experimental Procedures." The extension primer (primer A) was a 30-mer complementary to the l strand of 186 (9). Extension products were sized by comparison to a 35S-labeled sequencing ladder, of which only the G and T lanes are shown. The time after UV irradiation is indicated, as is the major 125-base extension product and a second, minor product of 180 bases. In some of the gel images, the sequencing markers have been contrast-adjusted differently from the primer extension lanes.

The Direct Action of Tum Is Equivalent to CI Inactivation-- To confirm that Tum's action is to inactivate CI rather than activate an alternative promoter, we next compared the transcription response following expression of Tum specifically (from a plasmid rather than via UV induction) with that seen after temperature inactivation of a temperature-sensitive CI mutant.

View larger version (99K): [in this window] [in a new window]   Fig. 3.   Derepression of pR upon IPTG-induced expression of Tum. Cultures of C600 bacterial strains hosting pEC342 (carrying the 186 promoters pR and pL, together with the cI gene from 186cItsp) and pEC331 (carrying p95 and tum under control of the IPTG-inducible plac promoter) or its parental (tum) counterpart pEC330 were grown at 30 °C in L broth (plus the appropriate antibiotic) to 1.5 × 108 colony-forming units/ml. Expression of Tum was induced by the addition of 1 mM IPTG, and RNA was isolated from samples taken 0, 30, and 60 min after induction. A negative control, to which IPTG was not added, and a positive control, utilizing a temperature shift (ts) to 41 °C, were also used. The extension primer was the same as that used in the experiment of Fig. 2. The major 125-base extension product and the G, T sequencing tracks used for sizing the product are indicated.

Derepression of p_B during UV Induction-- A second 186 promoter repressed by CI is p_B. To confirm that Tum acts to derepress rather than activate, the start point of the B transcript after exposure of the repressed p_B to Tum was determined by primer extension analysis, using a three-plasmid system (Fig. 4). p_B B was carried as an insert on a single copy RI-derived plasmid pEC458 (14). Transcription from p_B was detected as an 82-base extension product and was repressed by CI supplied in trans from pEC452, which is a pACYC184 derivative (14). The third plasmid pEC307 is pBR322-derived and bears the 2.0-kb PstI fragment spanning p₉₅ and tum (25). When the host carrying the three plasmids was UV-irradiated, exactly this 82-base product was detectable at 20 min after irradiation. If the tum plasmid (pEC307) was replaced by either pEC308 carrying the inactive tum16 mutant allelle or pEC309 bearing a tum deletion mutant (25), then UV-mediated activation of p_B was no longer evident. The correspondence in size of the UV-induced extension product with that of the p_B transcript was further evidence for an antirepressor role for Tum and evidence against the proposal that Tum acts to bypass repressor by activation of a new transcript.

View larger version (101K): [in this window] [in a new window]   Fig. 4.   Derepression of pB upon UV irradiation. Cultures of C600 hosting pEC458 (incorporating the B gene under the control of its own promoter, pB) (14), pEC452 (incorporating pR/pL and cI+) (14) or its parental (cI) counterpart pACYC184, and the pBR322-derivative pEC307 (containing p95 tum+), pEC308 (containing p95tum16), or pEC309 (containing p95tum) were grown at 37 °C with the appropriate antibiotic selection to 2 × 108 colony-forming units/ml. Cells were centrifuged and resuspended in 10 mM MgSO4 before being UV-irradiated (45 J per m2). RNA was extracted immediately or after the addition of one-tenth volume of 10 × L broth and incubation at 37 °C for a further 20 min. The extension primer (primer B) was a 30-mer oligonucleotide complementary to the l strand of 186 (349 to 320). The 82-base pair extension product, sized by comparison to an 35S-sequencing ladder (G and T lanes shown), is indicated.

Genetic Approach to Studying CI-Tum Interaction-- To probe further the nature of Tum-mediated prophage induction, an attempt was made to isolate mutants of CI resistant to the action of Tum. Initial attempts involved screening for the presence of turbid plaques on media containing mitomycin C, a DNA-damaging agent that causes prophage induction, and indicator bacteria hosting a single copy p₉₅ tum⁺ plasmid (pEC306). Thus, in theory, 186 tum mutants would be complemented in trans by the tum⁺ plasmid in the host strain, leaving only mutants of 186 resistant to Tum as turbid plaque formers. From a sample of 650,000 plaques, generated by plating a nitrosoguanidine-mutagenized 186cItsp stock (see "Experimental Procedures") on C600(pEC306), four potential Tum-resistant mutants were isolated. However, the mutations were mapped to the tum gene itself, identifying them as dominant tum mutants. This gave the first indication that Tum may act as an oligomer, and indeed we show Tum in solution to be a dimer (see below). An improved screen for Tum resistance was then employed, using a 186 tum deletion mutant, 186cItspSA, that removes all of tum, as well as p₉₅. Use of this phage was designed to preclude the possibility of mutation to tum dominance, as the only source of Tum is that of pEC306. Nitrosoguanidine-mutated stocks of this phage were plated for single plaques on C600(pEC306) in the presence of mitomycin C. From a screen of 550,000 plaques, three further potential Tum-resistant mutants were isolated. One of these mutants (TumR ) mapped to the left of the cI gene, in a region concerned with integration and excision of the prophage. Indeed this mutant displayed a deficiency in prophage induction common to both heat- and Tum-mediated induction and was not pursued. The remaining two mutants (TumR G and TumR 3) mapped to the cI gene, each mutant bearing a double mutation resulting in two amino acid substitutions within cI (V58A,L98F for TumR G; L115F,P158S for TumR 3). The mutants formed unstable prophage that displayed a defect in prophage induction by Tum (data not shown), suggesting a simultaneous loss of sensitivity to Tum and of repressibility. However, the defect was not specific for Tum induction as the mutants were also defective in heat induction (the cItsp allele was used in the mutational screen). We therefore concluded that CI activity and resistance to Tum were not genetically independent, and development of improved genetic screens was discontinued in favor of biochemical studies to study CI-Tum interactions.

Biochemical Approach to Studying CI-Tum Interactions

Isolation of Tum-- The expression of tum-derived proteins in maxicells (8) indicated that up to four proteins (Orf95.1, .2, .4, and .5) are expressed from the tum reading frame. Plasmid-driven expression of these various proteins also demonstrated that the tum function responsible for prophage induction corresponded to the 146 amino acid Orf95.1 protein product (8). Oligonucleotide site-directed mutagenesis was used to create an NdeI site at the GTG initiation codon of Tum (Orf95.1), in a construct (pEC324 (8)) in which Orf95.4 and Orf95.5 expression had been abolished. This modified tum gene was initially cloned into the pET3a expression vector to give pEC363. That this clone produced functional protein was shown by the plaque phenotype of 186 on the bacterial strain HMS174[DE3] hosting pEC363. Even without IPTG induction there was sufficient basal expression of Tum to give a clear plaque (no lysogens) phenotype in the presence of pEC363 but not the tum16 control pEC364. However, SDS-PAGE analysis of cell extracts prepared from induced cultures of HMS174[DE3] pEC363 showed that very little of the Tum protein produced was in a soluble form. To facilitate purification of an active, soluble form of Tum, the gene was cloned into a modified expression vector such that the expressed protein contained a carboxyl-terminal six histidine tag (pET TumHis₆). Again, analysis of the plaque phenotype of 186 when plated on HMS174[DE3] pET TumHis₆ showed that this vector, but not the corresponding tum16 construct, produced functional Tum protein. The presence of the histidine tag enabled the use of a one-step affinity purification procedure under denaturing conditions, followed by a simple refolding procedure, to yield milligram quantities of soluble protein (see "Experimental Procedures").

Tum Prevents CI Binding to DNA in Vitro-- Purified TumHis₆ and CI (isolated as described (21)) were used in a series of gel retardation assays to determine whether Tum-specific retardation of CI binding activity could be demonstrated in vitro. Several conclusions could be drawn from this series of experiments as follows. (i) As shown by Dodd and Egan (12), purified CI caused retardation of DNA fragments containing the pR (Fig. 5a) and pB (Fig. 5b) CI operator sequences. The presence of Tum resulted in a loss of the ability of CI to cause retardation of these fragments, confirming the in vivo result that Tum acts to prevent effective CI binding. (ii) Tum by itself did not give rise to retardation of the pR (Fig. 5a) or pB (Fig. 5b) containing fragments, indicating that Tum did not compete with CI for binding sites on the DNA. (iii) The inhibitory action of Tum was specific for CI binding. This was demonstrated using a DNA fragment that contained both the FR CI operator and the CII binding site. CII, a transcriptional activator required to establish lysogeny in 186, is like 186 CI, a helix turn helix protein, and binds to a 7-base pair inverted repeat sequence located at the beginning of the CII gene (19) (Fig. 1). As shown in Fig. 5c, CI binding at FR was prevented by the presence of Tum, whereas at the same Tum concentration, the binding of CII was unaffected. (iv) The inactive Tum mutant, Tum16 (7), did not prevent CI binding to pR (Fig. 5a, final lane). In fact, CI binding was somewhat improved in the presence of Tum16. However, given that substitution of the same concentration of bovine serum albumin for Tum16 also improved CI binding (not shown), this effect was considered to be a nonspecific one. (v) Inhibition by Tum of the binding of CI to pR occurred over a narrow range of Tum concentrations. The fraction of labeled pR DNA bound by 150 nM CI was measured at a series of Tum concentrations (Fig. 6). The transition from no inhibition to complete inhibition of CI binding occurred over a narrow (10-fold) range of Tum concentrations. The data were fitted as shown in Equation 1.

(Eq. 1)

where _B is the fraction of DNA bound, K is the concentration of Tum required to give 50% DNA bound, and n reflects the slope of the binding curve (26). The curve in Fig. 6 represents the best fit of the data to Equation 1, where n = 4.5 (±0.8) and K = 3.3 (± 0.1) µM. Thus, the concentration of Tum required to inhibit by 50% the binding of 150 nM CI occurred at a 20-fold molar ratio of Tum to CI, expressed in terms of monomer concentration. This relatively high molar ratio may indicate that only a fraction of the isolated Tum is active, perhaps a consequence of the need to purify Tum in a non-native state, followed by refolding.

View larger version (52K): [in this window] [in a new window]   Fig. 5.   Tum prevents CI binding in vitro. Gel retardation assays were used to assess the ability of purified Tum to prevent CI binding to its operator sites. a, CI binding to pR was prevented by Tum. The DNA used was a 32P-labeled 437-bp fragment containing the pR/pL region. The Tum concentration varied between 5.4 and 0.9 µM, whereas CI, where present, was at a concentration of 150 nM. The final lane contained 15 µM Tum16, an inactive mutant of Tum (7). b, CI binding to pB was prevented by Tum. The DNA was a 32P-labeled 435-bp fragment containing pB. The protein concentrations used were the same as those in a. c, the action of Tum was specific for CI. A 32P-labeled 323-bp fragment containing both the CII-binding site (19) and the FR CI-binding site (12) was used. Tum concentrations were between 4.5 and 0 µM, whereas CI, where present, was at a concentration of 180 nM. CII was supplied by inclusion of 1 µl of a crude extract prepared from a CII overexpressing strain (19), (estimated final CII concentration ~106-107 M).

View larger version (13K): [in this window] [in a new window]   Fig. 6.   Inhibition of CI binding as a function of Tum concentration. CI (150 nM) was incubated with approximately 60 cpm of a 32P-labeled DNA fragment containing the pR CI binding site, in the presence of a range of Tum concentrations. Following a 30-min incubation on ice, the samples were subjected to gel electrophoresis on a 6% polyacrylamide gel as described under "Experimental Procedures." The gels were dried and exposed to a phosphorimaging screen overnight, and the fraction of DNA bound was quantitated using Imagequant software (Molecular Dynamics). The results are plotted as the fraction of DNA bound (B) as a function of the log of Tum concentration, whereas the line represents the best fit to the data according to Equation 1.

Tum Reversibly Inactivates CI-- Prophage induction in lambdoid phages is achieved through a specific proteolytic inactivation of the maintenance repressor via cleavage of the repressor at an Ala-Gly site (27). Although 186 CI does not contain such a sequence (9), the possibility remained that Tum may cause proteolytic degradation of CI. To test this, purified Tum (5.7 µM), CI (150 nM), and a 437-bp ³²P-labeled DNA fragment containing pR were incubated together under solution conditions identical to those used in the gel shift assay of Fig. 5a. Following a 1-h incubation, samples were subjected to SDS-PAGE, transferred to a polyvinylidene difluoride membrane, and probed with an anti-CI polyclonal antibody. As shown in Fig. 7a, the Western blot showed no evidence of CI degradation and hence no evidence for proteolytic attack by Tum.

View larger version (69K): [in this window] [in a new window]   Fig. 7.   Reversibility of the action of Tum. a, no evidence of CI proteolysis by Tum. Samples (10 µl) containing 32P-labeled pR DNA and CI only (30 ng) (lane a), CI (150 nM) and TumHis6 (5.7 µM) (lane b), or CI (150 nM) and Tum16 His6 (14 µM) (lane c) were incubated on ice for 1 h. Following incubation, an equal volume of 2 × SDS loading buffer was added, and the samples were subjected to SDS-PAGE. The separated proteins were then electroblotted onto a polyvinylidene difluoride membrane and probed with an anti-CI antibody and a secondary horseradish peroxidase-conjugated goat anti-mouse antibody, followed by chemiluminescent detection (Pierce). b, the inactivation of CI is reversible. A gel retardation assay using CI protein before and after exposure to TumHis6 was performed. TumHis6 (4.8 µM) (or Tum16 His6 (14 µM) final two lanes) and CI (150 nM) were incubated in a total volume of 200 µl for 30 min on ice, at which time 10-µl aliquots were removed. Ni2+-NTA resin (100 µl) was then added to the remaining solution in order remove TumHis6 from solution. Following centrifugation, 10-µl aliquots of the supernatant were removed. One microliter of 32P-labeled DNA (a 323-bp DNA fragment containing the FR CI binding site) was added to each of the aliquots, and these samples were subjected to gel electrophoresis as described for the retardation assays. Lanes designated () resin indicate samples removed prior to the addition of Ni2+-NTA resin, and lanes labeled (+) resin indicate samples taken following removal of TumHis6 by Ni2+-NTA resin. The result was the same when the labeled DNA was included in the incubation, rather than added later.

Characterization of Tum-- The genetic experiments aimed at finding Tum-resistant mutants of CI revealed the existence of dominant negative tum mutants, indicative of oligomerization of Tum. To quantitate the extent of any self-association, purified TumHis₆ (monomeric molecular weight 17,958) was subjected to sedimentation equilibrium (Fig. 8). Analysis of the data, obtained at two rotor speeds, gave a weight average molecular weight of 34,310 ± 700, indicating that Tum was predominantly dimeric in solution. Inclusion of additional species in the fitting procedure (tetramer or octamer) provided no evidence of higher order association, although we cannot rule out the possibility that the histidine tag may prevent association beyond a dimer.

View larger version (30K): [in this window] [in a new window]   Fig. 8.   Analytical ultracentrifugaton of TumHis6. Sedimentation experiments were performed at 16,000 rpm (solid circles) and 24,000 rpm (open circles). The data are presented as concentration distributions as a function of radial distance. For clarity, only every second data point is shown. Experiments were done at a loading concentration of 9.5 µM. The lines represent the best fit of the data to a scheme based on a single monomeric species (Mr,app = 34,310 ± 700), and the lower residuals plot presents the difference between the experimental data and the fitted values.

	DISCUSSION

Top Abstract Introduction Procedures Results Discussion References

We have confirmed that the Tum protein of 186, produced by the phage as part of the SOS response following damage to the DNA of the host cell, is the only phage function required to derepress the CI controlled promoters p_R and p_B, an event known to lead to prophage induction. In particular, we have shown that Tum directly interferes with the ability of the CI repressor to bind to its operators and so maintain the lysogenic state.

By what mechanism might Tum inactivate CI? The sequence of Tum provides no clue as to its mechanism of action, as it displays significant homology to only two other sequences (8, 25), the functions of which are unknown. UV induction of lambdoid phages involves proteolytic cleavage of the maintenance repressor by activated host cell RecA protein. We have shown, however, that Tum does not cleave or catalyze the self-cleavage of 186 CI and that in fact CI retains the ability to bind DNA following incubation with Tum. The Tum protein sequence contains no recognized DNA binding motifs and, consistent with this, displays no ability to bind DNA in gel retardation assays. Thus, Tum does not act by competing with CI for its operator sites. All the available evidence suggests that Tum directly but reversibly inactivates the CI repressor.

This mechanism of action is reminiscent of functions described for other bacteriophage systems. The Ant protein of Salmonella phage P22 for example interferes with the activity of the c2 repressor (28). The Coi antirepressor protein of phage P1 inactivates P1 C1 repressor by forming a noncovalent complex with it, at a molar ratio of about 1:1 (29, 30), whereas a second P1 antirepressor, Ant, can specifically coprecipitate the P1 C1 repressor from cellular extracts (31). In addition, it has recently been reported that the  gene product of the satellite phage P4 efficiently derepresses a P2 prophage, possibly by direct interaction with the P2 C repressor function (32).

The most likely models for Tum-mediated inactivation of CI repressor seem to be that Tum prevents CI binding (i) by blocking the DNA binding domain of CI, (ii) by preventing CI oligomerization, or (iii) by blocking the cooperative interactions between CI subunits required for effective binding of CI to its operators. The CI repressor of 186 exists in solution in a monomer-dimer-tetramer-octamer equilibrium, the formation of octamer from dimers being a concerted process (21). The stoichiometry of CI binding to its operators is currently under investigation but, given the nature of the binding sites (12), must involve at least tetramers. This rather complex set of linked equilibria provides ample opportunity for Tum to interfere with cooperative interactions.

The isolation of dominant tum mutations indicates that Tum itself exists as a multimer of at least two monomers. The ability of Tum to dimerize was confirmed by sedimentation equilibrium. The phenotypes of different tum mutants (7) suggests that the protein has two regions of differing functional importance. Those phage mutants with the most severe loss in the ability to be induced, both spontaneously and through DNA damage, all had specific base pair changes within the amino-terminal portion of the Tum protein. 186tum13, for example, involves a conservative alanine to valine change at the 34th amino acid, yet has a tum null phenotype. The Orf95.2 product differs from Tum only in the absence of the first nine amino acids, yet it possesses no discernible ability to induce a prophage (8). In contrast, less severe tum mutants have alterations further downstream toward the carboxyl half of the protein. Indeed, the 18695C mutant has the carboxyl half of Tum removed, yet it is still able to induce a prophage, albeit at 5% efficiency compared with wild type. These observations suggest that the active site of Tum involved in inactivating CI is located in the amino-terminal half of the protein. Determination of the precise mechanism of action of Tum will require further structural information on both Tum and CI.

What are the advantages to 186 of the Tum-mediated system of prophage induction? It is in the interest of the prophage that if induction is to occur, then it should do so effectively, such that the process should be essentially irreversible and, once undertaken, should involve a total commitment to bacteriophage production. 186 ensures this by having p₉₅ strongly repressed by LexA (7, 33). Tum will only be expressed following extensive DNA damage, conditions under which the survival of the host bacterium is threatened. Furthermore, once Tum synthesis begins, intermediate concentrations of Tum that derepress p_R and p_B ineffectively must be avoided. One way to address this problem would be to use cooperative interactions, either between Tum polypeptides or between Tum and CI to enhance inactivation of CI. In this way, relatively small changes in Tum concentration would be magnified to give a more efficient induction signal. This is illustrated in Fig. 9. Although CI binding to its operators is itself a cooperative process (n > 1), we have shown that inactivation of CI by Tum occurs over quite a narrow range of Tum concentrations. The data of Fig. 9 show that this transition (solid line) is sharper than the binding isotherm for CI (dashed line). This would appear to be a necessary requirement for efficient induction, such that inactivation of CI is not negated by rebinding of free CI.

View larger version (21K): [in this window] [in a new window]   Fig. 9.   Cooperativity in prophage induction. The solid line describes the Tum-mediated inactivation of CI (from Fig. 6), depicted in terms of the fraction of free DNA (F, left ordinate axis). The dotted line describes, in terms of bound DNA (B, right ordinate axis), binding of CI to pR DNA under the same solution conditions (calculated by fitting the data from Fig. 6 of Ref. 12 to Equation 1; K = 2.8 (± 0.1) × 108 M, n = 1.7 (± 0.1)), and the dashed line represents the result expected if CI binding were noncooperative (n = 1 in Equation 1), assuming the same apparent binding constant. Note also the different concentration scales for Tum and CI.

Once Tum-mediated inactivation of CI has taken place, the phage must also ensure that newly synthesized CI does not contribute to the CI pool and allow reestablishment of repression of the lytic promoters. Further synthesis of CI by the prophage is avoided through Apl-mediated repression of p_L (24). In addition, by linking Orf97 production to Tum expression (Fig. 1), superinfection of the host cell is avoided (8), and bacteriophage production can proceed without the risk of an infecting phage adding to the CI pool.