Ahmad I. Bukhari

Analysis of bacteriophage Mu and λ-Mu hybrid DNAs by specific endonucleases☆

The DNAs of plaque-forming particles bearing various portions of bacteriophages λ and Mu were cleaved by the specific endonucleases of Hemophilus influenzae and analyzed by polyacrylamide gel electrophoresis. Our extensive mapping data on λ and Mu DNA allowed us to identify the parts of each parent in the hybrids. The technique revealed that both ends of Mu DNA are heterogeneous in size, and suggested that a portion of DNA at the immunity end of vegetative Mu is not inserted in the prophage. The known ability of Mu to invert a specific part of its own DNA (G loop inversion) was observed mainly in phage grown by induction of a lysogen, and the gene involved was deduced to lie within, or at least very close to, the G loop region itself. The mode of growth of Mu, by lytic infection or by induction of a lysogen, also affected the phage DNA in its pattern of modification. Thus a specific endonuclease, Hin II, failed to cleave the DNA of Mu induced from a lysogen, even though it cleaved at several sites the DNA of phage grown by lytic infection.

Reversal of mutator phage Mu integration

Abstract The temperate bacteriophage Mu causes mutations by inserting its DNA randomly into the genes of its host bacterium Escherichia coli . It is shown here that Mu DNA can be precisely excised from the different integration sites and that as a result wild-type function of the gene into which Mu was inserted is restored. The excision of Mu DNA is observable only if the Mu prophage carries mutations at the X locus. Thus, lac + revertants from six strains, containing heat-inducible prophage Mu cts 62 at different locations in the Z gene of the lac operon, were readily obtained by first introducing the X mutation into Mu cts 62. The lac + revertants produced wild-type β-galactosidase, and no trace of Mu DNA could be detected in them; this indicates that the junction of Mu DNA and host DNA can be specifically recognized. However, the excision of Mu DNA is generally not perfect, because in most cases it does not lead to the wild-type genotype. The function of gene A of Mu appears to be required for excision. Since the lethal functions of Mu are completely blocked in the Mu cts 62 X prophage, the X locus probably has a regulatory function. At least one X mutation is caused by an insertion of about 900 base-pairs in Mu DNA. The discovery of the X mutants opens the way for studying the reversible interaction of the host and Mu chromosomes, and for using Mu to manipulate the host genome in various ways.

Predominant end-products of prophage Mu DNA transposition during the lytic cycle are replicon fusions

Abstract We present biochemical and genetic experiments which strongly suggest that the net result of transposition of prophage Mu and internally deleted Mu derivatives (mini-Mus) during the lytic cycle is replicon fusion. When a Mu prophage located on pSC101, a low copy number plasmid, is induced, virtually all of the pSC101::Mu plasmid copies enter the host chromosome within 33 minutes. By cleaving the total host DNA with restriction endonucleases and by hybridization with 32P-labeled pSC101 DNA, we have found that the fused structures contain directly repeated copies of Mu at each junction of the plasmid and host DNA. These fused structures, called cointegrates, have been seen with Mu as low level genetic events and with almost all other transposable elements. Genetic analysis of mini-Mu transposition from pSC101 to an F′pro+lac episome has also shown that transposition invariably is associated with fusion of the whole plasmid with the episome; no simple linear insertions of a mini-Mu molecule into the F′ episome were recovered. Our results also indicate that jumping of Mu from a chromosomal location onto an F′ episome rarely results in the linear insertion of a single copy of Mu DNA; instead, the episome apparently first undergoes Mu-mediated fusion with the chromosome followed by release of Mu-containing episomes that carry extensive deletions or insertions. Integration of Mu into an F′ episome during lysogenization, however, leads to simple point insertions of the Mu genome. Our data strongly suggest that the end-products of Mu DNA transposition during prophage induction and during lysogenization are not the same. We infer, therefore, that the process of Mu DNA transposition can occur by either of two alternate pathways which differ with respect to the end-products they generate.

In vitro and in vivo manipulations of bacteriophage Mu DNA: Cloning of Mu ends and construction of mini-Mu's carrying selectable markers

Recombinant plasmids carrying one or both ends of the bacteriophage Mu genome were constructed by molecular cloning. Transposable mini-Mus with selectable markers (ampicillin resistance, kanamycin resistance or the entire lac operon of Escherichia coli) inserted between the Mu ends were also constructed. As a source of lac operon DNA, a pBR322 derivative with a 27 kb insert containing the lac operon was constructed. The plasmids with both ends of Mu (mini-Mus) conferred full Mu immunity upon the host cells. However, the same mini-Mus containing kan or lac inserts were defective in immunity. A summary of the construction and physical characterization, including restriction endonuclease cleavage maps and some of the biological properties of the plasmids, is presented.

The invertible DNA segments of coliphages Mu and P1 are identical

Electron microscopic-heteroduplex studies show that the G sequence of Mu DNA, an invertible region three kilobases in length near the S end, is identical to the invertible region found in bacteriophage P1 DNA.

Genetic mapping of prophage Mu

Abstract The gene order of several independently isolated Mu prophages has been determined by deletion mapping. The prophages in which deletions were produced were located in the lacI gene, the lacZ gene, and the trp operon of Escherichia coli . The Mu deletions were tested for marker rescue against a set of sixteen Mu amber mutants. Marker rescue tests were done by constructing strains lysogenic for both cryptic Mu prophages and Mu amber mutants and then examining the resulting heterogenotes, partially diploid for Mu, for spontaneous production of wild-type Mu particles. The gene order of all Mu prophages was identical, although prophages were found in either of the two orientations with respect to the host markers. It can be inferred therefore that the temperate bacteriophage Mu, which inserts randomly into the chromosome of E. coli , uses a specific site on its own DNA for integration.

Infecting bacteriophage Mu DNA forms a circular DNA-protein complex

Upon superinfection of immune (lysogenic) cells with bacteriophage Mu, a form of Mu DNA accumulates that sediments about twice as fast as the linear phage DNA marker in neutral sucrose gradients. This form is also detected upon infection of sensitive cells with Mu. We have purified it and examined its physical nature. Under the electron microscope it appears circular and supertwisted. Upon treatment with Pronase, phenol or sodium dodecyl sulfate, however, it is converted to a linear Mu-length form, indicating that the circle is not covalently closed. The linear DNA still has heterogeneous host sequences at its termini. The circular DNA is resistant to the action of Escherichia coli exonuclease III and T7 exonuclease, but becomes sensitive to these nucleases after treatment with Pronase showing the presence of a protein that binds non-covalently to the ends of the DNA to circularize it as well as protect it from digestion with exonucleases. The complex is resistant to high salt (up to 6 M-NaCl) but can undergo transitions between forms that are partially open, open circular, linear and circular dimers and trimers. Examination of DNA from mature phage particles reveals that a circular DNA species is present in at least 0.1 to 1% of the population. The purified complex is extremely efficient in transfection of E. coli spheroplasts. We estimate the molecular weight of the protein in this DNA-protein complex to be approximately 64,000, and suggest that this complex might represent the integrative precursor of infecting Mu DNA.

Plaque-forming λ-Mu hybrids

Abstract We have isolated plaque-forming λ particles carrying different segments of the temperate phage Mu DNA. These hybrids were obtained by heat induction of dilysogens in which Mu was inserted into the lac genes in λp lac 5 c I857 S 7. The hybrid DNA molecules, derived from three different insertions of Mu, were digested with the specific endonucleases of Hemophilus influenzae , and the resulting fragments were separated by polyacrylamide gel electrophoresis. Comparison of the fragments produced from λp lac 5 c I857 S 7, Mu, and the presumptive hybrids confirmed that a part of the λ genome had been deleted and that portions of Mu DNA had been added. The amount of Mu DNA present did not correspond to the amount of λ DNA absent. Analysis of the fragments obtained from the vegetative Mu DNA indicated that both ends of Mu DNA are variable in length.

Behavior of bacteriophage Mu DNA upon infection of Escherichia coli cells

The question whether the ends of bacteriophage Mu DNA are fused to form a ring in host cells is critical to the understanding of the mechanism of integrative recombination between Mu DNA and host DNA. We have examined the fate of 32P-labeled Mu DNA, after infection of sensitive and immune (lysogenic) cells, by sedimentation in sucrose gradients, ethidium bromide/CsCl density centrifugation and by electrophoresis of parental Mu DNA and its fragments in agarose gels. We find that the parental Mu DNA cannot be detected as covalently closed circles at any stage during the Mu life cycle. An interesting form of Mu DNA can be seen after superinfection of immune cells. This form sediments about twice as fast as the mature phage DNA marker in neutral sucrose gradients but yields linear molecules upon phenol extraction. Upon infection of sensitive cells, most of the parental DNA associates with a large complex, presumably containing the host chromosome. When Mu-sensitive cells are infected with unlabeled Mu particles and Mu DNA examined at different times after infection by fractionation in 0.3% agarose gels and hybridization with 32P-labeled Mu DNA, Mu sequences are found to appear with the bulk host DNA as the phage lytic cycle progresses. However, no distinct replicative or integrative intermediate of Mu, that behaves differently from linear Mu DNA and is separate from the host DNA, can be detected.

Heteroduplex electron microscopy of phage Mu mutants containing IS1 insertions and chloramphenicol resistance transposons.

We have examined by electron microscopy the DNA heteroduplexes of six bacteriophage Mu mutants, Mu X cam, generated by the insertion of the Tn9 transposon for chloramphenicol resistance. Tn9 was found to be 2.8 +/- 0.2 kilobases (kb) in length and to consist of a cam determinant flanked by two IS1 sequences arranged in a direct order. In two of the six Mu X cam mutants, the Tn9 insertion was at a fixed location, 3.9 kb from the left, or c, end. In the other four mutants, the position of the insertion varied, even though the lysogenic cultures induced were grown from single colonies. The insertion was located at either 3.3 kb, 3.9 kb, or, less frequently, at 4.4 kb from the left end of the DNA. Furthermore, at low frequencies, the insertions were found to be in an orientation opposite to what predominated in the preparation. Thus, Tn9 in the Mu X cam mutants examined could appear to undergo rapid rearrangements during Mu growth or over a few generations of cell growth. One of the Tn9 insertion sites was apparently the same as that for a 0.8 kb insertion found in a Mu X mutant. This latter insertion was identified as an IS1 sequence. The DNA molecules from all the Mu X cam mutant phage particles were found to be missing the bacterial DNA at the S (right) end, along with a variable amount of the adjoining Mu DNA in the beta region. This observation supports the headful packaging model for Mu DNA.