Frameshift mutations in the bacteriophage Mu repressor gene can confer a trans-dominant virulent phenotype to the phage

Genome Study of a Novel Virulent Phage vB_SspS_KASIA and Mu-like Prophages of Shewanella sp. M16 Provides Insights into the Genetic Diversity of the Shewanella Virome

Shewanella is a ubiquitous bacterial genus of aquatic ecosystems, and its bacteriophages are also isolated from aquatic environments (oceans, lakes, ice, and wastewater). In this study, the isolation and characterization of a novel virulent Shewanella phage vB_SspS_KASIA and the identification of three prophages of its host, Shewanella sp. M16, including a mitomycin-inducible Mu-like siphovirus, vB_SspS_MuM16-1, became the starting point for comparative analyses of phages infecting Shewanella spp. and the determination of their position among the known bacterial viruses. A similarity networking analysis revealed the high diversity of Shewanella phages in general, with vB_SspS_KASIA clustering exclusively with Colwellia phage 9A, with which it forms a single viral cluster composed of two separate viral subclusters. Furthermore, vB_SspS_MuM16-1 presented itself as being significantly different from the phages deposited in public databases, expanding the diversity of the known Mu-like phages and giving potential molecular markers for the identification of Mu-like prophages in bacterial genomes. Moreover, the functional analysis performed for vB_SspS_KASIA suggested that, despite the KASIA host, the M16 strain grows better in a rich medium and at 30 °C the phage replication cycle seems to be optimal in restrictive culture conditions mimicking their natural environment, the Zloty Stok gold and arsenic mine.

The CRISPR/Cas adaptive immune system of Pseudomonas aeruginosa mediates resistance to naturally occurring and engineered phages

Here we report the isolation of 6 temperate bacteriophages (phages) that are prevented from replicating within the laboratory strain Pseudomonas aeruginosa PA14 by the endogenous CRISPR/Cas system of this microbe. These phages are only the second identified group of naturally occurring phages demonstrated to be blocked for replication by a nonengineered CRISPR/Cas system, and our results provide the first evidence that the P. aeruginosa type I-F CRISPR/Cas system can function in phage resistance. Previous studies have highlighted the importance of the protospacer adjacent motif (PAM) and a proximal 8-nucleotide seed sequence in mediating CRISPR/Cas-based immunity. Through engineering of a protospacer region of phage DMS3 to make it a target of resistance by the CRISPR/Cas system and screening for mutants that escape CRISPR/Cas-mediated resistance, we show that nucleotides within the PAM and seed sequence and across the non-seed-sequence regions are critical for the functioning of this CRISPR/Cas system. We also demonstrate that P. aeruginosa can acquire spacer content in response to lytic phage challenge, illustrating the adaptive nature of this CRISPR/Cas system. Finally, we demonstrate that the P. aeruginosa CRISPR/Cas system mediates a gradient of resistance to a phage based on the level of complementarity between CRISPR spacer RNA and phage protospacer target. This work introduces a new in vivo system to study CRISPR/Cas-mediated resistance and an additional set of tools for the elucidation of CRISPR/Cas function.

Control of Bacteriophage Mu Lysogenic Repression

The transposable and temperate phage Mu infects Escherichia coli where it can enter the lytic life-cycle or reside as a repressed and integrated prophage. The repressor protein Rep is the key element in the lysis-lysogeny decision. We have analyzed the fate of Rep in different mutants by Western blotting under two conditions that can induce a lysogen: high temperature and stationary phase. We show that, unexpectedly, Rep accumulates under all conditions where the prophage is completely derepressed, and that this accumulation is ClpX-dependent. An analysis of the degradation kinetics shows that Rep is a target of two protease systems: inactivation of either the clpP or lon gene results in a stabilization of Rep. Such a reaction scheme explains the counterintuitive observation that derepression is correlated with high repressor concentration. We conclude that under all conditions of phage induction the repressor is sequestered in a non-active form. A quantitative simulation accounts for our experimental data. It provides a model that captures the essential features of Mu induction and explains some of the mechanisms by which the physiological signals affecting the lysis-lysogeny decision converge onto Rep.

A context-dependent ClpX recognition determinant located at the C terminus of phage Mu repressor

The bacteriophage Mu immunity repressor is a conformationally sensitive sensor that can be interconverted between forms resistant to and sensitive to degradation by ClpXP protease. Protease-sensitive repressor molecules with an altered C-terminal sequence promote rapid degradation of the wild-type repressor by inducing its C-terminal end to become exposed. Here we determined that the last 5 C-terminal residues (CTD5) of the wild-type repressor contain the motif required for recognition by the ClpX molecular chaperone, a motif that is strongly dependent upon the context in which it is presented. Although attachment of the 11-residue ssrA degradation tag to the C terminus of green fluorescent protein (GFP) promoted its rapid degradation by ClpXP, attachment of 5-27 C-terminal residues of the repressor failed to promote degradation. Disordered peptides derived from 41 and 35 C-terminal residues of CcdA (CcdA41) and thioredoxin (TrxA35), respectively, activated CTD5 when placed as linkers between GFP and repressor C-terminal sequences. However, when the entire thioredoxin sequence was included as a linker to promote an ordered configuration of the TrxA35 peptide, the resulting substrate was not degraded. In addition, a hybrid tag, in which CTD5 replaced the 3-residue recognition motif of the ssrA tag, was inactive when attached directly to GFP but active when attached through the CcdA41 peptide. Thus, CTD5 is sufficient to act as a recognition motif but has requirements for its presentation not shared by the ssrA tag. We suggest that activation of CTD5 may require presentation on a disordered or flexible domain that confers ligand flexibility.

Trans-targeting of the phage Mu repressor is promoted by conformational changes that expose its ClpX recognition determinant

Dominant negative forms of the phage Mu repressor, including the mutant Vir repressors, are not only rapidly degraded by the ClpXP protease but also promote degradation of the unmodified, wild-type repressor. This trans-targeting of the wild-type repressor depends upon a determinant within its C-terminal domain, which is needed for recognition by ClpX. An environmentally sensitive fluorescent probe (2-(4'-maleimidylanilino)naphthalene-6-sulfonic acid (MIANS)) attached to the C terminus of the full-length repressor indicated that Vir induces the movement of this domain into a more exposed configuration. Vir also promoted attachment of MIANS to the C terminus of the repressor at an accelerated rate, and it greatly increased the rate of phosphorylation of a cAMP-dependent protein kinase motif attached to the repressor C terminus. While an excess of Vir was needed to promote repressor phosphorylation at maximal rates, the presence of ClpX could increase phosphorylation rates at lower Vir levels. trans-Targeting of the Mu repressor is therefore promoted by exposing its ClpX recognition determinant, and the action of ClpX can assist Vir in exposing these determinants.

Modulation of phage Mu repressor DNA binding and degradation by distinct determinants in its C-terminal domain

Rapid degradation of the bacteriophage Mu immunity repressor can be induced in trans by mutant, protease-hypersensitive repressors (Vir) with an altered C-terminal domain (CTD). Genetic and biochemical analysis established that distinct yet overlapping determinants in the wild-type repressor CTD modulate Vir-induced degradation by Escherichia coli ClpXP protease and DNA binding by the N-terminal DNA-binding domain (DBD). Although deletions of the repressor C-terminus resulted in both resistance to ClpXP protease and suppression of a temperature-sensitive DBD mutation (cts62), some cysteine-replacement mutations in the CTD elicited only one of the two phenotypes. Some CTD mutations prevented degradation induced by Vir and resulted in the loss of intrinsic ClpXP protease sensitivity, characteristic of wild-type repressor, and at least two mutant repressors protected Vir from proteolysis. One protease-resistant mutant became susceptible to Vir-induced degradation when it also contained the cts62 mutation, which weakens DNA binding but apparently facilitates conversion to a protease-sensitive conformation. Conversely, this CTD mutation was able to suppress temperature sensitivity of DNA binding by the cts62 repressor. The results suggest that determinants in the CTD not only provide a cryptic ClpX recognition motif but also direct CTD movement that exposes the motif and modulates DNA binding.

Characterization of the cts4 repressor mutation in transposable bacteriophage Mu

Mucts4 was isolated more than 30 years ago and was the first available thermoinducible derivative of transposable phage Mu. We have characterized the cts4 mutation and the corresponding mutant protein. Contrary to previously characterized thermoinducible Mu prophages (e.g., Mucts62), Mucts4 lysogenizes at reduced frequency even at 30 degrees C. The cts4 mutation (Leu129Val) was located in this central repressor region. The cts4 protein was thermosensitive for operator DNA binding in vitro. Temperature-dependent changes in protein-protein cross-linking patterns in the absence of DNA were detected for purified wild type, cts62 and cts4 repressor proteins. The cts4 protein exhibited a subtly different electrophoretic profile, which became more marked at higher temperatures, from both the wild type and cts62. In addition the cts4 repressor generated a significantly different pattern of binding to DNA fragments carrying the early operator region. Consistent with the predicted involvement of the central leucine-rich region of the Mu repressor in the formation of multimeric forms, the cts4 mutation thus appeared to affect protein-protein interactions.

Derepression of bacteriophage mu transposition functions by truncated forms of the immunity repressor

To trigger bacteriophage Mu transposition and replication in response to physiological signals, its immunity repressor must be synchronously inactivated. Two repressor mutants (Vir), which have an altered C-terminal domain and are highly susceptible to degradation by ClpXP protease, confer a dominant negative phenotype by promoting degradation of the wild-type repressor. To search for other modified repressors that can induce Mu derepression in vivo and to determine what part of the inducing repressor molecules are needed to target the unmodified repressor population, repressor peptides with nested deletions starting at the C-terminal end were constructed. Such peptides with a C-terminal ssrA degradation tag promoted a sharp reduction in cellular levels of full-length unmodified repressor, a process largely dependent upon the clpP protease function. Only the repressor DNA-binding domain, located at the N-terminal end, was required in tagged peptides to target unmodified repressor. In addition, some repressor peptides containing the DNA-binding domain promoted derepression without the clpP function, being able to promote repressor inactivation without promoting its degradation. None of the modified repressors could promote derepression if immunity was established by a mutant repressor lacking 18 residues at its C-terminal end. The results indicate that inducing repressor peptides influence the function of the C-terminal domain of the intact repressor, a domain that regulates its degradation and DNA binding. They suggest the possibility that tagged repressor molecules, produced by stalled ribosomes, can be inducers of transposition under starvation conditions.

Bacteriophage Mu genome sequence: analysis and comparison with Mu-like prophages in Haemophilus, Neisseria and Deinococcus

We report the complete 36,717 bp genome sequence of bacteriophage Mu and provide an analysis of the sequence, both with regard to the new genes and other genetic features revealed by the sequence itself and by a comparison to eight complete or nearly complete Mu-like prophage genomes found in the genomes of a diverse group of bacteria. The comparative studies confirm that members of the Mu-related family of phage genomes are genetically mosaic with respect to each other, as seen in other groups of phages such as the phage lambda-related group of phages of enteric hosts and the phage L5-related group of mycobacteriophages. Mu also possesses segments of similarity, typically gene-sized, to genomes of otherwise non-Mu-like phages. The comparisons show that some well-known features of the Mu genome, including the invertible segment encoding tail fiber sequences, are not present in most members of the Mu genome sequence family examined here, suggesting that their presence may be relatively volatile over evolutionary time. The head and tail-encoding structural genes of Mu have only very weak similarity to the corresponding genes of other well-studied phage types. However, these weak similarities, and in some cases biochemical data, can be used to establish tentative functional assignments for 12 of the head and tail genes. These assignments are strongly supported by the fact that the order of gene functions assigned in this way conforms to the strongly conserved order of head and tail genes established in a wide variety of other phages. We show that the Mu head assembly scaffolding protein is encoded by a gene nested in-frame within the C-terminal half of another gene that encodes the putative head maturation protease. This is reminiscent of the arrangement established for phage lambda.

Conformational dynamics of a transposition repressor in modulating DNA binding

The repressor of bacteriophage Mu functions in the establishment and maintenance of lysogeny by binding to Mu operator DNA to shut down transposition. A domain at its N terminus functions in DNA binding, and temperature-sensitive mutations in this domain can be suppressed by truncations at the C terminus. To understand the role of the C-terminal tail in DNA binding, a fluorescent probe was attached to the C terminus to examine its environment and its movement with respect to the DNA binding domain. The emission spectrum of this probe indicated that the C terminus was in a relatively hydrophobic environment, comparable to the environment of the probe attached within the DNA-binding domain. Fluorescence of two tryptophan residues located within the DNA-binding domain was quenched by the probe attached to the C terminus, indicating that the C terminus is in close proximity to this domain. Addition of DNA, even when it did not contain operator DNA, reduced quenching of tryptophan fluorescence, indicating that the tail moves away from the DNA-binding domain as it interacts with DNA. The presence of the tail also produced a trypsin hypersensitive site within the DNA-binding domain; mutant repressors with an altered or truncated C terminus were relatively resistant to cleavage at this site. Interaction of the wild-type repressor with DNA greatly reduced cleavage at the site. A repressor with a temperature-sensitive mutation in the DNA-binding domain was especially sensitive to cleavage by trypsin even in the presence of DNA, and the C-terminal tail failed to move in the presence of DNA at elevated temperatures. These results indicate that the tail sterically inhibits DNA binding and that it moves during establishment of repression. Such conformational changes are likely to be involved in communication between repressor protomers for cooperative DNA binding.

The tRNA function of SsrA contributes to controlling repression of bacteriophage Mu prophage

The small regulatory RNA SsrA has both tRNA and mRNA activities. It charges alanine and interacts with stalled ribosomes, allowing for translation to resume on the SsrA mRNA moiety. Hence, unfinished peptides carry a short amino acid tag, which serves as a signal for degradation by energy-dependent proteases. In SsrA-defective Escherichia coli strains, thermoinducible mutants of the transposable bacteriophage Mu (Mucts) are no longer induced at high temperature. Here we show that truncated forms of the key regulator of Mu lysogeny, the repressor Repc, accumulate in the absence of SsrA. These forms resemble C-terminally truncated dominant Mu repressor mutants previously isolated from Mucts, which are no longer thermoinducible and bind operator DNA with a high affinity even at high temperature. Using various ssrA alleles, we demonstrate the importance of SsrA charging on the ribosome for controlling Mu prophage repression. Our results thus substantiate the previous observation that trans-translation is not the only function of the SsrA. The alternative function of SsrA appears to influence the stability of Mu lysogens by controlling the translation of the C-terminal domain of the repressor protein, which modulates the affinity of the protein for DNA and its susceptibility to proteolytic degradation.

NMR structure and functional studies of the Mu repressor DNA-binding domain

The repressor protein of bacteriophage Mu establishes and maintains lysogeny by shutting down transposition functions needed for phage DNA replication. It interacts with several repeated DNA sequences within the early operator, preventing transcription from two divergent promoters. It also directly represses transposition by competing with the MuA transposase for an internal activation sequence (IAS) that is coincident with the operator and required for efficient transposition. The transposase and repressor proteins compete for the operator/IAS region using homologous DNA-binding domains located at their amino termini. Here we present the solution structure of the amino-terminal DNA-binding domain from the repressor protein determined by heteronuclear multidimensional nuclear magnetic resonance spectroscopy. The structure of the repressor DNA-binding domain provides insights into the molecular basis of several temperature sensitive mutations and, in combination with complementary experiments using flourescence anisotropy, surface plasmon resonance, and circular dichroism, defines the structural and biochemical differences between the transposase and repressor DNA-binding modules. We find that the repressor and enhancer domains possess similar three-dimensional structures, thermostabilities, and intrinsic affinities for DNA. This latter result suggests that the higher affinity of the full-length repressor relative to that of the MuA transposase protein originates from cooperative interactions between repressor protomers and not from intrinsic differences in their DNA-binding domains. In addition, we present the results of nucleotide and amino acid mutagenesis which delimits the minimal repressor DNA-binding module and coarsely defines the nucleotide dependence of repressor binding.

Starvation-induced Mucts62-mediated coding sequence fusion: a role for ClpXP, Lon, RpoS and Crp

The formation of araB-lacZ coding sequence fusions in Escherichia coli is a particular type of chromosomal rearrangement induced by Mucts62, a thermoinducible mutant of mutator phage Mu. Fusion formation is controlled by the host physiology. It only occurs after aerobic carbon starvation and requires the phage-encoded transposase pA, suggesting that these growth conditions trigger induction of the Mucts62 prophage. Here, we show that thermal induction of the prophage accelerated araB-lacZ fusion formation, confirming that derepression is a rate-limiting step in the fusion process. Nonetheless, starvation conditions remained essential to complete fusions, suggesting additional levels of physiological regulation. Using a transcriptional fusion indicator system in which the Mu early lytic promoter is fused to the reporter E. coli lacZ gene, we confirmed that the Mucts62 prophage was derepressed in stationary phase (S derepression) at low temperature. S derepression did not apply to prophages that expressed the Mu wild-type repressor. It depended upon the host ClpXP and Lon ATP-dependent proteases and the RpoS stationary phase-specific sigma factor, but not upon Crp. None of these four functions was required for thermal induction. Crp was required for fusion formation, but only when the Mucts62 prophage encoded the transposition/replication activating protein pB. Finally, we found that thermally induced cultures did not return to the repressed state when shifted back to low temperature and, hence, remained activated for accelerated fusion formation upon starvation. The maintenance of the derepressed state required the ClpXP and Lon host proteases and the prophage Ner-regulatory protein. These observations illustrate how the cts62 mutation in Mu repressor provides the prophage with a new way to respond to growth phase-specific regulatory signals and endows the host cell with a new potential for adaptation through the controlled use of the phage transposition machinery.

Assembly of both the head and tail of bacteriophage Mu is blocked in Escherichia coli groEL and groES mutants

Like several other Escherichia coli bacteriophages, transposable phage Mu does not develop normally in groE hosts (M. Pato, M. Banerjee, L. Desmet, and A. Toussaint, J. Bacteriol. 169:5504-5509, 1987). We show here that lysates obtained upon induction of groE Mu lysogens contain free inactive tails and empty heads. GroEL and GroES are thus essential for the correct assembly of both Mu heads and Mu tails. Evidence is presented that groE mutations inhibit processing of the phage head protein gpH as well as the formation of a 25S complex suspected to be an early Mu head assembly intermediate.

PDZ-like domains mediate binding specificity in the Clp/Hsp100 family of chaperones and protease regulatory subunits

ClpX, a molecular chaperone and the regulatory subunit of the ClpXP protease, is shown to contain tandem modular domains that bind to the C-terminal sequences of target proteins in a manner that parallels functional specificity. Nuclear magnetic resonance studies show that these C-terminal sequences are displayed as disordered peptides on the surface of otherwise folded proteins. The ClpX substrate-binding domains are homologous to sequences in other Clp/Hsp100 proteins and are related more distantly to PDZ domains, which also mediate C-terminal specific protein-protein interactions. Conservation of these binding domains indicates that the mode of substrate recognition characterized here for ClpX will be a conserved feature among Clp/Hsp100 family members and a distinguishing characteristic between this chaperone family and the Hsp70 chaperones.

Communication of ClpXP protease hypersensitivity to bacteriophage Mu repressor isoforms

The immunity repressor (Rep) of bacteriophage Mu establishes and maintains lysogeny by shutting down transposition functions needed for phage DNA replication. Although Rep is stable in vivo, an altered immunity repressor (Vir) encoded by virulent, trans-dominant Mu mutants is rapidly degraded by Escherichia coli ClpXP protease. Rep and Vir are degraded at approximately the same maximal velocity (Vmax) by ClpXP, but the Km for Rep (3.6 microM) is over 20-fold higher than the Km for Vir (0.15 microM). Rep is also highly resistant to degradation in the presence of DNA whereas Vir is not. Vir increases the rate of Rep degradation by reducing its Km and imparts to Rep ClpXP sensitivity in the presence of DNA. Vir can drive at an accelerated rate the complete degradation of Rep molecules that outnumber Vir by eightfold or more. So long as Vir is present at a concentration of 0.1 microM or higher, Rep is degraded with a Km that is indistinguishable from that of Vir. These characteristics of repressor may be an important means of transducing physiological signals that induce Mu transposition in response to growth conditions or environmental stress, ClpXP hypersensitivity being disseminated among Rep molecules for the induction of Mu transposition.

In vivo mutational analysis of bacteriophage Mu operators

In bacteria lysogenic for bacteriophage Mu, the phage repressor binds to a tripartite operator region, O1,O2,O3, to repress the lytic promoter pE, located in O2, and negatively autoregulate its own synthesis at the pCM promoter located in O3. We isolated and characterized operator mutations which lead to derepression of pE. Their location in the first and third repressor-consensus-binding sequences in O2 confirms the importance of these sites for repressor/operator interactions.

C-terminal deletions can suppress temperature-sensitive mutations and change dominance in the phage Mu repressor

Mutations in an N-terminal 70-amino acid domain of bacteriophage Mu's repressor cause temperature-sensitive DNA-binding activity. Surprisingly, amber mutations can conditionally correct the heat-sensitive defect in three mutant forms of the repressor gene, cts25 (D43-G), cts62 (R47-Q) and cts71 (M28-I), and in the appropriate bacterial host produce a heat-stable Sts phenotype (for survival of temperature shifts). Sts repressor mutants are heat sensitive when in supE or supF hosts and heat resistant when in Sup degrees hosts. Mutants with an Sts phenotype have amber mutations at one of three codons, Q179, Q187, or Q190. The Sts phenotype relates to the repressor size: in Sup degrees hosts sts repressors are shorter by seven, 10, or 18 amino acids compared to repressors in supE or supF hosts. The truncated form of the sts62-1 repressor, which lacks 18 residues (Q179-V196), binds Mu operator DNA more stably at 42 degrees in vitro compared to its full-length counterpart (cts62 repressor). In addition to influencing temperature sensitivity, the C-terminus appears to control the susceptibility to in vivo Clp proteolysis by influencing the multimeric structure of repressor.

Bacteriophage Mu repressor as a target for the Escherichia coli ATP-dependent Clp Protease

Bacteriophage Mu repressor, which is stable in its wildtype form, can mutate to become sensitive to its Escherichia coli host ATP-dependent ClpXP protease. We further investigated the determinants of the mutant repressor's sensitivity to Clp. We show the crucial importance of a C-terminal, seven amino acid long sequence in which a single change is sufficient to decrease the rate of degradation of the protein. The sequence was fused at the C-terminal end of the CcdB and CcdA proteins encoded by plasmid F. CcdB, which is naturally stable, was unaffected, while CcdA, which is normally degraded by the Lon protease, became a substrate for ClpXP while remaining a substrate for Lon. In agreement with the current hypothesis on the mechanism of recognition of their substrates by energy- dependent proteases, these results support the existence, on the substrate polypeptides, of separate motifs responsible for recognition and cleavage by the protease.

A new component of bacteriophage Mu replicative transposition machinery: the Escherichia coli ClpX protein

We have shown previously that some particular mutations in bacteriophage Mu repressor, the frameshift vir mutations, made the protein very sensitive to the Escherichia coli ATP-dependent Clp protease. This enzyme is formed by the association between a protease subunit (ClpP) and an ATPase subunit. ClpA, the best characterized of these ATPases, is not required for the degradation of the mutant Mu repressors. Recently, a new potential ClpP associated ATPase, ClpX, has been described. We show here that this new subunit is required for Mu vir repressor degradation. Moreover, ClpX (but not ClpP) was found to be required for normal Mu replication. Thus ClpX has activities that do not require its association with ClpP. In the pathway of Mu replicative transposition, the block resides beyond the strand transfer reaction, i.e. after the transposition reaction per se is completed, suggesting that ClpX is required for the transition to the formation of the active replication complex at one Mu end. This is a new clear-cut case of the versatile activity of polypeptides that form multi-component ATP-dependent proteases.

Protein-DNA assemblies controlling lytic development of bacteriophage Mu

Recent analysis of the mechanism and regulation of transposition by bacteriophage Mu has emphasized the importance of controlled assembly of specific protein-DNA complexes. Both the Mu transposase and the Mu repressor engage in multiple protein-protein and protein-DNA interactions that modulate the outcome of a phage infection.

Temperature-sensitive mutations in the bacteriophage Mu c repressor locate a 63-amino-acid DNA-binding domain

Phage Mu's c gene product is a cooperative regulatory protein that binds to a large, complex, tripartite 184-bp operator. To probe the mechanism of repressor action, we isolated and characterized 13 phage mutants that cause Mu to undergo lytic development when cells are shifted from 30 to 42 degrees C. This collection contained only four mutations in the repressor gene, and all were clustered near the N terminus. The cts62 substitution of R47----Q caused weakened specific DNA recognition and altered cooperativity in vitro. A functional repressor with only 63 amino acids of Mu repressor fused to a C-terminal fragment of beta-galactosidase was constructed. This chimeric protein was an efficient repressor, as it bound specifically to Mu operator DNA in vitro and its expression conferred Mu immunity in vivo. A DNA looping model is proposed to explain regulation of the tripartite operator site and the highly cooperative nature of repressor binding.