Isolation and Characterization of Micromonospora Phage ΦHAU8 and Development into a Phasmid

Abstract

ΦHAU8, a temperate Micromonospora phage, which is capable of infecting Micromonospora sp. strains 40027 and A-M-01, was isolated. The ΦHAU8 virion has a polyhedral head and a flexible tail and has a small genome (ca. 42.5 kb) with double-stranded DNA and cohesive ends. ΦHAU8 was most stable at 4°C in Difco nutrient broth within a pH range of 6 to 12. ΦHAU8 plaque formation on Micromonospora sp. strain 40027 was optimal with 32 mM Ca²⁺ and 30 mM Mg²⁺. A lysogen, LXH8, was isolated from turbid plaques, and a phasmid derivative that functions as a λ cosmid vector in Escherichia coli and as a phage in Micromonospora sp. strain 40027 was constructed. Pulsed-field gel electrophoresis of AseI-digested total DNA showed that ΦHAU8 DNA integrates into the 500-kb AseI fragment of Micromonospora sp. strain 40027.

The genus Micromonospora has a complex life cycle, which includes differentiation into substrate mycelia, aerial hyphae, and spores (20). Production of many hydrolytic enzymes, such as chitinases (6) and cellulases (14), allows Micromonospora strains to play an active role in the degradation of organic matter in their natural habitats. They are also known as producers of a wide variety of antibiotics, among which aminoglycosides are the most abundant (26).

In spite of their commercial importance and ability to produce and modify a large variety of antibiotics, reports on the genetics of Micromonospora strains are scarce. This is due in part to the lack of tools, techniques, and systems for the study of this genus. The advanced Streptomyces gene cloning methods cannot be applied to Micromonospora because Streptomyces vectors do not usually transform members of this genus satisfactorily.

In addition to plasmid-based vectors, phages have proven to be useful tools for studying the biology of their hosts, as they can affect the metabolism of the hosts (22) and can be used in the transfer of genetic information by both transduction and transfection (9). A temperate phage could provide the basis of a cloning system similar to that of the Streptomyces-specific ΦC31 derivatives (4, 18, 27) or the phage R4-derived cosmid (16, 19).

A number of Micromonospora-specific actinophages have been reported, including the lytic phages ΦUW21 and ΦUW51 (10, 11), the temperate phage MPΦWR-1 (25), and phages with undetermined infection cycles and specificities (3). Several other lytic Micromonospora phages have been used to screen for the presence of restriction enzymes (15). Phage pMLP1 was found to be present in Micromonospora carbonacea var. africana ATCC 39149 as a replicative form as well as an integrative form, and plasmid derivatives containing the site-specific att/int functions of pMLP1 were found to be able to integrate genes into the chromosome (1). None of the Micromonospora phages, however, has been developed into a gene cloning vector.

Here we describe a temperate phage (ΦHAU8) that is capable of infecting and transfecting Micromonospora sp. strain 40027 (13), a producer of fortimicin A, which exhibits potent, broad-spectrum antibacterial activity against gram-positive and -negative bacteria both in vitro and in vivo. We also report the development of ΦHAU8 into a phasmid that functions as a λ cosmid vector in Escherichia coli and as a phage in Micromonospora sp. strain 40027.

MATERIALS AND METHODS

Bacteria, phages, media, and culture methods used.

Micromonospora sp. strain 40027 (13) was a gift from Zhu Baoquan, and Micromonospora purpurea, Streptomyces citrofluorescens, Nocardia erythropolis, and Nocardia rubra were gifts from Hildgund Schrempf. Micromonospora sp. strain A-M-01, Micromonospora chalcea A015, Nocardia sp. strains A016 and A-N-01, and Streptomyces lividans ZX1 (28) were obtained from stocks in our laboratory. Micromonospora rosea A2000, Streptomyces coelicolor A3(2) and J1501 (9), Streptomyces vinaceus (24), and Amycolotopsis mediterranei were obtained from stock cultures from the John Innes Center, Norwich, United Kingdom. Phage ΦHAU3 (29) is a broad-host-range temperate Streptomyces phage. The standard reagents and buffers were those described by Kieser et al. (9). Micromonospora sp. strain 40027 and its derivatives were grown at 30°C on Bennet medium (1% glucose, 0.1% yeast extract, 0.1% beef extract, 0.2% NZ amine, 1.5% agar; pH 7.3) for production of spores and mycelium medium (30) for growth of mycelium. Difco nutrient agar and Difco nutrient broth (DNB) were used for testing phage properties and for phage storage.

Phage isolation and purification.

Phage isolation and purification were performed by the enrichment method with a mixed soil sample (10 g) obtained from various locations in Hubei, People's Republic of China. Dried samples were inoculated into DNB supplemented with 0.5% glucose, 4 mM Ca(NO)₃, 10 mM MgCl₂, and ca. 10⁷ spores of Micromonospora sp. strain 40027. After growth at 30°C for 36 h with shaking, a membrane filter-sterilized sample of the supernatant was tested for the presence of viable phages by the double-layer plate method (9) with spores of Micromonospora sp. strain 40027 as the indicator. Phages obtained from plaque soak-outs with DNB were purified by three serial, single-plaque isolations before a confluent phage lysate was prepared. The titer of phage that could be obtained from a confluent phage lysate (5-cm petri plate) could be up to 10⁷ to 10⁸ PFU.

Isolation of putative lysogenic strains.

Cells were picked from the center of turbid plaques or from growth occurring after confluent lysis on solid media. The cells were streaked and purified repeatedly though the sporulation stage. Spores of putative lysogens were washed and then tested for the spontaneous release of infectious phages by the double-layer plate method with spores of Micromonospora sp. strain 40027 as the indicator.

Test for phage lysogeny.

Freshly titrated phage lysates (>10⁸ PFU/ml) were spotted onto lawns of spores from putative lysogenic strains on Difco nutrient agar plates, on which plaques were scored after 72 h of growth at 30°C. The spot tests were repeated with spores from independently made putative lysogens to confirm the initial results and then repeated again with serial dilutions of the phage lysate (and with scoring of individual plaques) to ensure that plaque formation was not caused by lysis from without.

Total DNA was isolated by standard methods by using mycelia of putative lysogenic strains. DNA samples of Micromonospora sp. strain 40027, its lysogen, and the corresponding causative phage were digested with the same enzyme, run in agarose gels, and transferred to nylon filters. The filters were probed with ³²P-labeled phage DNA by Southern hybridization. Pulsed-field gel electrophoresis (PFGE) was performed by the method of Kieser et al. (9).

Electron microscopy of phage virions.

For electron micrographs of intact phage particles, phage were applied to carbon film picked up on copper grids, and this was followed by a brief glow discharge treatment. The adhering phage were then washed with either 1 to 2% ammonium acetate (pH 7) or 10 mM Tris-10 mM MgCl₂ (pH 7.4) and negatively stained with 1 to 2% uranyl acetate.

Phage DNA preparation and analysis for the presence of cohesive ends.

The standard method used for phage DNA was the method described by Hopwood et al. (7). Two methods were used to detect cos ends. In the first method restriction enzyme digests of phage DNA were heated to 70°C for 10 min and then quickly cooled to ice-water temperatures before they were loaded on an agarose gel. In the second method phage DNA was treated with T4 DNA ligase before it was analyzed by PFGE with the Bio-Rad CHEF MAPPER system.

RESULTS AND DISCUSSION

Isolation and basic properties of ΦHAU8.

One of the five Micromonospora phages isolated was designated ΦHAU8. It produced small turbid plaques on Micromonospora sp. strain 40027. ΦHAU8 could infect the mycelium (formation of plaques after 20 h) much more quickly than it could infect Micromonospora sp. strain 40027 spores (72 h).

ΦHAU8 plaque formation on Micromonospora sp. strain 40027 was optimal with addition of 32 mM Ca²⁺ and 30 mM Mg²⁺, and it was not sensitive to high concentrations of the chelating agents sodium pyrophosphate and EDTA (0.5 M), which can be used to inactivate some phages and to select for deletion mutants (4). ΦHAU8 was stable when it was stored in DNB at pH 6 to 10 at 4°C, and it was rapidly inactivated at lower and higher pH values and at a temperature of 60°C or higher.

The morphology of ΦHAU8 was investigated by electron microscopy. Phage ΦHAU8 particles have a polyhedral head and a flexible tail, thus placing them in the B2 type of bacteriophages as classified by Bradley (2). Their dimensions are about 56 nm (head width) by 60 nm (head length), with a tail length of ca. 220 nm (Fig. 1). Thus, ΦHAU8 is similar in morphology to Streptomyces phages ΦC31 (21) and ΦHAU3 (29).

FIG. 1.

Electron micrograph of ΦHAU8 negatively stained with uranyl acetate. Dimensions are indicated in nanometers.

ΦHAU8 could infect Micromonospora sp. strains 40027 and A-M-01 among five Micromonospora strains tested (including M. chalcea A015, M. purpurea, and M. rosea A2000), but it was unable to form plaques on the Streptomyces strains tested, including S. coelicolor A3(2) and J1501 (9), S. lividans ZX1 (28), S. citrofluorescens, and S. vinaceus (24), A. mediterranei, and Nocardia strains (N. erythropolis, N. rubra, and Nocardia sp. strains A016 and A-N-01).

Lysogeny.

ΦHAU8 gave turbid plaques on Micromonospora sp. strain 40027. Growth from these plaques resulted in phage-resistant colonies, which released ΦHAU8 at frequencies of about 10⁻⁵ per spore in a lawn of the nonlysogenic parental strain. A lysogenic strain, LXH8, was confirmed by Southern hybridization analysis by using total DNA from the phage-resistant strains, which showed strong bands identical to those seen with the DNA isolated from phage particles and bands that could be interpreted as flanking bands, consistent with integration into the host genome (data not shown). PFGE of AseI-digested total DNA showed that ΦHAU8 DNA was integrated into the ca. 500-kb AseI fragment of Micromonospora sp. strain 40027 (Fig. 2). These results indicated that ΦHAU8 could lysogenize this strain by integrating at a specific site in the chromosome. Also, a ΦHAU8-derived thiostrepton-resistant (Thio^r) phasmid, pJTU120 (see below), was shown to lysogenize Micromonospora sp. strain 40027, but its integration site has not been analyzed.

FIG. 2.

Southern hybridization with ³²P-labeled ΦHAU8 (lane A) as the probe to prove that ΦHAU8 integrated into the Micromonospora sp. strain 40027 chromosome in lysogenized strain LXH8. All samples were digested with AseI and separated by PFGE. Electrophoresis was performed for 20 h at 6 V/cm with a 20- to 80-s switch time at an included angle of 120°. The lanes labeled with lowercase letters are autoradiographs of the lanes labeled with the corresponding uppercase letters. A clear size difference (500 versus 540 kb) for the AseI fragments detected between strains 40027 (lane B) and LXH8 (lane C) was further proved by a hybridization signal detected in lane c (corresponding to lane C) but not in lane b (corresponding to lane B). AseI fragments of S. coelicolor M145 (lane D) were used as standards; the sizes (in kilobases) are indicated on the right.

ΦHAU8 DNA is ca. 45 kb long and has cos ends.

Digestion of ΦHAU8 DNA with BamHI and ClaI generated 13 and 7 fragments, respectively, and digestion with SalI or PvuII generated more than 19 fragments. The sum of the bands from the digestions with BamHI and ClaI gave an average genome size of about 42.5 kb, and consistent with this, undigested ΦHAU8 DNA migrated slightly more quickly than DNA of phage λ (48 kb) and DNA of Streptomyces phage ΦHAU3 (29) (51 kb) on a pulsed-field gel (Fig. 3). Comparison of heated and unheated samples fractionated on normal agarose gels (Fig. 3) and the ladder of multimers observed on a pulsed-field gel (Fig. 3) indicated that ΦHAU8 DNA has cos ends located on fragment J (Fig. 4), similar to DNA of phage ΦHAU3 (Fig. 3) and phage λ (data not shown).

FIG. 3.

ΦHAU8 DNA has cos ends. The gel on the left shows a ladder of multimers after ligation was observed in DNA of ΦHAU8 (lane B) by PFGE, which is similar to the DNA of ΦHAU3 (lane A) used for comparison, as well as a size standard. The sizes of DNA fragments (in kilobases) are indicated on the left and right. Electrophoresis was performed for 15.5 h at 6 V/cm with a 5- to 60-s switch time at an included angle of 120°. In the gel on the right, BamHI-digested ΦHAU8 DNA samples that were heated at 70°C (lane D) and were not heated (lane C) before loading were run in parallel. Clearly, one of the two comigrating bands (which had a twofold intensity and is indicated by the solid arrow) in lane C is a sum of the 1.3-kb fragment (corresponding to fragment J1 in Fig. 4) and the 0.7-kb fragment (corresponding to fragment J2 in Fig. 4) in lane D, which are indicated by open arrows. Lambda DNA digested with HindIII (lane E) and a 1-kb ladder (lane F) were used as size standards; the sizes (in kilobases) are indicated on the right.

FIG. 4.

BamHI restriction maps of ΦHAU8 and its phasmid derivative, pJTU120. The DNA region dispensable for plaque formation and lysogeny in Micromonospora sp. strain 40027 is shaded. The circular plasmid below ΦHAU8 is the cosmid part (pHZ1358) of the ΦHAU8-derived phasmid, pJTU120. tsr, thiostrepton resistance gene, selectable in Micromonospora sp. strain 40027; bla, β-lactamase gene, selectable in E. coli. cos at the ΦHAU8 termini indicates cohesive ends at the ends of fragments J1 and J2, and cos in circular plasmid pHZ1358 was derived from E. coil phage λ. The sizes of fragments in ΦHAU8 (in kilobases) are indicated. ori/pIJ101, rep/pIJ101, and sti/pIJ101 (5, 8) derived from the multicopy Streptomyces plasmid pIJ101 are not functional in Micromonospora sp. strain 40027 and are thus not described in detail. The presence of the restriction sites indicated in pHZ1358 other than the BamHI sites was not examined in ΦHAU8.

Cosmid cloning of ΦHAU8 and a phasmid derivative, pJTU120.

Having found that ΦHAU8 DNA has cos ends and is smaller than phage λ DNA, we decided to introduce foreign DNA into random Sau3AI sites of ΦHAU8 DNA by cosmid cloning. The cosmid vector pHZ1358 (23) (10.7 kb, tsr bla) was chosen because it carries an origin of replication (ori) derived from E. coli plasmid ColE1 that enables replication in E. coli, the bla gene can be selected in E. coli, and the tsr gene (encoding Thio^r) can be selected in Micromonospora sp. strain 40027 (12). Additionally, pHZ1358 contains oriT originating from E. coli plasmid RP4, making conjugation from E. coli into Micromonospora sp. strain 40027 (12) possible. ΦHAU8 DNA was circularized and partially digested with Sau3AI, and 1.5 μg of size-fractionated DNA (30 to 50 kb, sucrose gradient centrifugation) was ligated with 0.5 μg of BamHI-digested vector pHZ1358, packaged in vitro into phage λ heads (by using λ-specific cos and terminase), and introduced by infection into E. coli LE392.

Covalently closed circular plasmid DNA was isolated from the pool of ampicillin-resistant E. coli LE392 colonies, retransformed into the E. coli ET12567 strain carrying pUZ8002 (a derivative of E. coli plasmid RK2 with deletion of the ampicillin resistance gene [bla]) (17), and introduced by conjugation into Micromonospora sp. strain 40027. The phage particles were collected from the nearly confluent plaques (ca. 10⁷ to 10⁸ PFU), and the DNA isolated from them (ca. 50 μg/ml) was reintroduced into E. coli by transformation. (Transformation of E. coli was necessary because conjugation of Micromonospora sp. strain 40027 with two nonfunctional cosmids might have resulted in wild-type ΦHAU8 genomes by homologous recombination; such in vivo recombination would have resulted in plaques but would have eliminated the E. coli vector component.) A derivative cosmid was introduced by conjugation into Micromonospora sp. strain 40027 to confirm that it was a phasmid (pJTU120) functioning as a phage in Micromonospora sp. strain 40027 and as a plasmid in E. coli. The genome size of pJTU120 was estimated to be 49.7 kb, which is ca. 7.2 kb larger than the size of ΦHAU8; thus, a minimal package limit of ca. 50 kb was defined.

Like wild-type phage ΦHAU8, phasmid pJTU120 infected Micromonospora sp. strains 40027 and A-M-01 and gave the same efficiency of plating in both hosts as ΦHAU8.

Phasmid pJTU120 contained a Thio^r determinant selectable in Micromonospora. It lacked 3.5 kb of DNA whose deletion did not seem to change its ability to infect and lysogenize Micromonospora sp. strains 40027 and A-M-01. pJTU120 seems to be a useful candidate for development into a cloning vector for Micromonospora. The pHZ1358 vector part is not required for gene cloning in Micromonospora, and thus it should be possible to replace the region corresponding to the 10.7-kb EcoRI fragment (several EcoRI sites in ΦHAU8 are unmapped and thus are not indicated in Fig. 4) containing pHZ1358 DNA with at least a compatible size of the target DNA, although a suitable site(s) for such a replacement experiment does not seem to be immediately available. The benefit of the very efficient in vitro packaging into λ cosmids would be affected by this strategy, but the potential problems with restriction occurring when DNA is transferred from E. coli to a specific Micromonospora strain could be avoided.