Previous Article | Next Article ![]()
Journal of Bacteriology, February 2002, p. 1089-1094, Vol. 184, No. 4
0021-9193/01/$04.00+0 DOI: 10.1128/jb.184.4.1089-1094.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.
MH2K, a Bacteriophage of the Obligate Intracellular Parasitic Bacterium Bdellovibrio bacteriovorus
Department of Veterinary Sciences and Microbiology, University of Arizona, Tucson, Arizona,1 Department of Biochemistry and Molecular Biology, University of Florida, Gainesville, Florida2
Received 11 September 2001/ Accepted 11 November 2001
|
|
|---|
MH2K, of Bdellovibrio bacteriovorus was isolated, characterized, and sequenced. This phage is a member of the Microviridae, a family typified by bacteriophage
X174. Although B. bacteriovorus and Escherichia coli are both classified as proteobacteria,
MH2K is only distantly related to
X174. Instead,
MH2K exhibits an extremely close relationship to the Microviridae of Chlamydia in both genome organization and encoded proteins. Unlike the double-stranded DNA bacteriophages, for which a wide spectrum of diversity has been observed, the single-stranded icosahedral bacteriophages appear to fall into two distinct subfamilies. These observations suggest that the mechanisms driving single-stranded DNA bacteriophage evolution are inherently different from those driving the evolution of the double-stranded bacteriophages. |
|
|---|
The icosahedral, single-stranded DNA phages, or Microviridae, appear to be less common than double-stranded DNA phages. The DNAs of nine members of the family, isolated from very diverse hosts (proteobacteria, Spiroplasma, and Chlamydia), have been sequenced (11, 16, 17, 18, 22, 24-26). The species falls into two distinct subfamilies, although the subfamilies are not officially recognized in the virus taxonomy. One is represented by
X174 and contains the phages that propagate in proteobacteria. The other subfamily contains phages of Chlamydia and Spiroplasma. Although these two hosts are not closely related, their phages, Chp1, Chp2, and SpV4, respectively, are quite similar. The principal difference between the two subfamilies is the existence of two genes and the complexity of major coat protein. The Chlamydia and Spiroplasma phages do not encode major spike and external scaffolding proteins. Their more-complex coat proteins contain an insertion loop that forms large threefold-related protrusions (6). Protein homologies between the two subfamilies are approximately 20% or less (6), a typical value when comparing the most distantly related members of either the lambda or T4-like groups (14, 28). However, unlike tailed double-stranded DNA families, no mosaic species that bridge the evolutionary chasms have been isolated.
While the available DNA sequences suggest that the evolution of Microviridae may differ from the evolution of the more-prevalent, double-stranded DNA phages, the phages have been isolated from such distantly related hosts that only the extremes of a diversity spectrum may have been uncovered. To further elucidate the evolution of the Microviridae, we have isolated, characterized, and sequenced a new family member,
MH2K, infecting Bdellovibrio bacteriovorus, a nonenterobacterial proteobacterium with an obligate intracellular parasitic life cycle similar to that of Chlamydia. The results suggest that
MH2K is closely related to the phages of Chlamydia and Spiroplasma. In some instances,
MH2K is more closely related to an individual chlamydiaphage than chlamydiaphages are related to each other. While the number of available Microviridae genomes is still limited, these results suggest that the forces driving single-stranded DNA phage evolution may differ from those driving double-stranded DNA phage evolution.
|
|
|---|
Isolation of phage
MH2K.
A 200-ml volume of raw sewage sludge was centrifuged at 5,000 x g to remove the large debris. The supernatant was then passed through 0.45- and 0.22-µm-pore-size filters. One milliliter of this solution was mixed with 10.0 ml of exponentially growing cells at 30°C and incubated for 6 h. Cultures were passed through a 0.45-µm-pore-size filter and then concentrated to 1.0 ml in a Centricon filter. One hundred microliters was layered atop of a 5 to 30% sucrose gradient, along with wild-type
X174 as an S value marker, and spun for 1.0 h at 240,000 x g in an SW50.1 rotor. Gradients were fractionated, and the titer of infectious
X174 on Escherichia coli C for each fraction was determined. Fractions containing material with S values less than 114 were pooled and reconcentrated to a final volume of 100 µl, layered atop a second sucrose gradient, and treated as described above. Finally, fractions were plated on Bdellovibrio for the isolation of phage. Single plaques were picked and used to infect 1.0 ml of exponentially growing cells. Phage DNA was extracted and analyzed using
X174 protocols as described below.
Isolation and purification of
MH2K genomic and RF DNA.
To isolate genomic DNA, 1.0 liter of exponentially growing cells was inoculated with
MH2K at a multiplicity of infection (MOI) of 0.001. After an 18-h incubation at 33°C, infected cells were stored at 4°C for 12 h. Cellular debris, containing attached phage particles, was concentrated by centrifugation. Pellets were resuspended in 8.0 ml of borate EDTA buffer (10), and virus and DNA were extracted three times with 8.0 ml of ether. Virions were purified on CsCl gradients, and DNA was extracted as previously described (10). To determine the strandedness of the genomes, genomic DNA was digested with S1 nuclease (Promega) as specified by the manufacturer. Results were compared to those of similar digestions of genomic and replicative-form (RF)
X174 DNA. To generate RF DNA, 100 ml of exponentially growing cells was inoculated with
MH2K at an MOI of 5.0 and incubated for 6.0 h. Infected cells were concentrated by centrifugation. RF DNA was extracted and characterized by standard techniques (5).
For electron microscopy, particles were purified as described above and concentrated 10-fold with Centricon filters, stained with 2.0% uranyl acetate, and examined with a JEOL EM10A electron microscope.
Cloning and sequencing of
MH2K RF DNA.
RF DNA was digested with HindIII (1395), BglII (1515), MspAI (350 and 2816), EcoRI (392, 1090, and 4107), or StuI (1185 and 3541) endonucleases (nucleotide positions of restriction enzyme digestion sites given in parentheses). Fragments from single-enzyme digestions were cloned into a modified TOPO vector (Invitrogen) digested with enzymes yielding compatible ends. Clones were sequenced by the DNA sequencing facility at the University of Arizona. T7 and M13R primers were provided by the facility. The entire genome was sequenced twice. Overlapping clones were used to determine the order of the genes.
Sequence alignment.
The amino acid sequences of the major capsid proteins of the coliphages
3, G4,
K, and
X174; the chlamydiaphages Chp1 and Chp2; a Spiroplasma phage, SpV4; and
MH2K were aligned using the program CLUSTAL W, version 1.4 (29). Default parameters were used. The colors of the resulting alignment were modified to give a better representation of the similarities and differences between the eight sequences. The secondary structure elements of
X174, based on knowledge of the atomic resolution structure (19, 20, 21), are given under the sequence alignment.
Nucleotide sequence accession number.
The nucleotide sequence of
MH2K was deposited in GenBank under accession number AF306496.
|
|
|---|
MH2K.
RF DNA was digested and cloned into a modified TOPO vector as described in Materials and Methods and sequenced. The genetic map of
MH2K is depicted in Fig. 1. The organization and size of the 4,594-base DNA genome are very similar to those of chlamydiaphages (18, 22, 26). However, the location of gene 5 differs. In
MH2K, it is located between genes 3 and 8, as opposed to after gene 4. The
MH2K genome is approximately 20% smaller than the
X174-like phages. This is due to the absence of genes encoding the major spike and external scaffolding proteins, as is seen in SpV4 and the chlamydiaphages (6, 18, 22, 24, 26). The relationship between
MH2K and the previously isolated Microviridae phage MAC-1 (2) is unclear since no MAC-1 sequence data are available. Three other phages were isolated. Their morphologies were similar to those of previously isolated bacteriophages MAC-3, MAC-6, and VL1 (2, 30).
![]() View larger version (10K): [in a new window] |
FIG. 1. Genetic maps of MH2K, Chp2 (18), and X174 (25). Reading frames that encode proteins with Chp2 homologues were designated with the same numbers used in the Chp2 genetic map (18). Reading frames that are not found in Chp2 were assigned letters. The numbering of the nucleotides commences with the start codon of gene 1.
|
MH2K genome suggests the existence of two very strong promoters (23). They reside in the two largest noncoding regions, located between genes 5 and 8 (230 bases) and between genes 3 and 5 (247 bases). Genes 1, 2, 3, 4, 5, and 8 all have upstream ribosome binding sites and homologues within the chlamydiaphages. There are four additional open reading frames (ORFs) within the
MH2K genome, ORFs N, W, X, Y, and Z. All five ORFs are contained within overlapping reading frames. The largest of these ORFs, ORF N, is the only one with a discernible ribosome binding site. The predicted protein is very hydrophobic and is similar to the
X174 lysis protein in character.
Protein homologies.
The
MH2K proteins encoded by genes 1 to 5 and 8 were compared to homologous proteins found in Chp1, Chp2, SpV4, and
X174-like phages. The
MH2K proteins most closely resemble the chlamydiaphage proteins, especially those of Chp2 (Table 1). Considering the evolutionary distance between their respective hosts, this result was unexpected (see Discussion). In some instances, comparisons of
MH2K and Chp2 proteins reveal closer relationships than that seen between the Chp1 and Chp2 proteins. In addition, polyclonal antibodies raised against the Chp2 major coat protein cross-react with
MH2K (I. Clarke, personal communication). Only weak homologies exist with the proteins of the
X174-like phages (
X174, G4,
3, S13, and
K). There does not appear to be significant homologies between the
MH2K ORFs N, W, X, Y, and Z and the small Chp2 ORFs 6 and 7.
|
View this table: [in a new window] |
TABLE 1. Putative MH2K gene products and amino acid homologies
|
MH2K virion and virion proteins.
The T=1 capsid is primarily composed of Vp1, the major coat protein, as assayed by polyacrylamide gel electrophoresis, and measures 27 ± 2 nm in diameter (Fig. 2).
Vp2, with an amino acid composition similar to that of the
X174 DNA pilot protein, protein H, is also a component of the virion. The exact nature of Vp3 remains somewhat obscure. Like the chlamydiaphage proteins (18, 22, 26),
MH2K Vp3 most closely resembles the internal scaffolding proteins of the
X174-like phages. Various amounts (depending on the purification protocol used) of Vp3 are associated with
MH2K capsids. Whether this is caused by the selective loss of this protein due to particle instability or the copurification of capsids and procapsids remains to be determined.
![]() View larger version (134K): [in a new window] |
FIG. 2. Electron micrograph of MH2K. Magnification, x31,500. Bar, 100 nm.
|
X174, G4,
K, and
3 are coliphages. However, host range variants which infect Salmonella enterica serovar Typhimurium and Shigella have been reported (3, 12). The aligned major coat protein sequences resulted in the identification of two distinct subclasses.
MH2K, Chp1, Chp2, and SpV4 represented a class with large amino acid insertions between ß-strands E and F of the core ß-barrel motif; the
X174-like phages formed the other class (Fig. 3). These additional amino acids are located between residues 210 and 280 in the
MH2K sequence. The cryoelectron microscope image reconstruction of SpV4 (6) demonstrates that they form large protrusions at the threefold icosahedral axes of symmetry. The ß-barrel core is conserved among all the phages.
![]() View larger version (66K): [in a new window] |
FIG. 3. Multiple-sequence alignment of coat proteins of Microviridae phages. Different colors were added to give a better representation of the similarities between the eight sequences. The four X174-like phages are grouped on the bottom; the smaller Microviridae are grouped at the top. Amino acids identical in all sequences (red boldface type) and conserved (green boldface type) and semiconserved (blue boldface type) substitutions in all sequences are indicated. Amino acids identical within the subgroups (red type) and conserved (green type) and semiconserved (blue type) substitutions within the subgroups are indicated. Dashes represent gaps introduced to maximize alignment of the sequences. The secondary structure (2°) of X174 is presented on the last line. The yellow arrows represent ß-strands, and the green cylinders represent -helices.
|
|
|
|---|
proteobacteria Chlamydia and Spiroplasma, are distantly related. To further study Microviridae diversity, a virus of the
proteobacterium B. bacteriovorus,
MH2K, was isolated. The data presented here indicate that
MH2K, Chp2, Chp1, and SpV4 share a common ancestor not found in the
X174-like lineage. For two of the gene products, Vp3 and Vp5, similarities between
MH2K and the chlamydiaphage Chp2 were greater than between Chp2 and Chp1 (18, 26).
If phages reflect their hosts phylogeny, this result is surprising. B. bacteriovorus is more closely related to the
proteobacteria than to the chlamydia. However, the phylogenetic classification of B. bacteriovorus, based exclusively on 16S RNA sequences, may be incorrect. Considering the predatory nature of Bdellovibrio, which hunt and replicate inside proteobacteria, it is possible that its 16S RNA sequence was derived via a horizontal transfer, as has been seen in other bacteria (31). On the other hand, several B. bacteriovorus genomic sequences were characterized during the cloning of
MH2K (GenBank accession numbers AF339026 to 339030). These sequences were most closely related to genes of proteobacteria and exhibit no discernible homologies with those of Chlamydia.
If the classification of Bdellovibrio is correct, then the evolution of single-stranded icosahedral bacteriophages may be fundamentally different from the evolution of the double-stranded phages, which is most likely driven by horizontal exchange (13, 14). There are several factors that may limit horizontal transfer in single-stranded DNA viruses. For example, Microviridae replication does not require recombination, and recombination is not prevalent. Double-stranded RF DNA is not abundant (20 to 50 copies per cell), and recombination frequencies for mutations separated by 2,500 bases range from 10-3 to -10-4 (B. A. Fane, unpublished results). Genomes are circular throughout the replication cycle, necessitating two recombination events. Lysogenic or latent life cycles have never been observed, therefore minimizing horizontal exchange with prophages. In addition, the small T=1 capsid may restrict the incorporation of exogenous DNA sequences, or morons (13). Although it is unlikely that these small viruses can acquire morons, all members of the Microviridae appear to have preserved ORFs, found mostly in overlapping genes. Mutations could accrete in these reading frames, termed cretins (for accrete in), until a gene encoding a beneficial function is produced. Examples of cretins may include the
X174 lysis gene E, the putative
MH2K lysis gene N, and the
X174 genes A* and K, both unessential and of unclear function (7, 27). The close relationship between
MH2K and the chlamydiaphages suggest that Microviridae evolution may be driven by cretins and species jumping. In light of these results, the present classification, dividing the Microviridae into four genera, based on host range (14), should be reexamined to reflect two distinct subfamilies.
Genetic maps of
MH2K, Chp2 (18), and
X174 (25) are given in Fig. 1. Neither Chp2 nor
MH2K encodes an external scaffolding or major spike protein,
X174 D or G protein, respectively. The external scaffolding protein has at least two known functions in
X174 morphogenesis. It stabilizes the procapsids at the two- and threefold axes of symmetry (8, 9) and directs the placement of the major spike protein. These functions are either not required or performed by different proteins in
MH2K-like phages. First, there is no major spike protein. The twofold stabilization function may be performed by Vp3, the internal scaffolding protein equivalent, which in
X174 self-associates across twofold axes of symmetry. Finally, as seen in the cryoimage reconstruction of SpV4, a large coat protein insertion loop forms spikes at the threefold axis of symmetry. This large insertion loop may be a relic of the ancestral external scaffolding or major spike protein. Coding Vp3 in a normal reading frame, as opposed to a cretin, may be a related phenomenon. The B proteins of the
X174-like phages are highly divergent, yet they cross-function, suggesting that interactions are primarily nonspecific and flexible (5). In addition, amino-terminal deletions are tolerated (4), probably because the external scaffolding protein performs a similar function. With the loss of the external scaffolding protein, internal scaffolding protein interactions may need to be more specific, requiring a reading frame unconstrained by other genes.
Regardless of their evolution, the similarity between
MH2K and the chlamydiaphages is rather fortuitous. Chlamydia research has been stifled by the lack of a genetic transfer system, and it is hoped that the chlamydiaphages can serve as a basis for its establishment. To develop the in vitro packaging system necessary for fruition, the precise functions of the phage proteins and cis-acting packaging elements must be identified. These questions will be much easier to address in the
MH2K system, in which host-independent Bdellovibrio mutants can be used in plaque assays to facilitate both genetic and biochemical analyses.
This work was supported in part by funds from the University of Florida College of Medicine (M.A.-M. and R.M.), the National Science Foundation (grant MCB 9982284 to B.A F.), and the Undergraduate Research Program in Biology (Howard Hughes Medical Institute, grant 71195-521304) at the University of Arizona (K.L.B.).
|
|
|---|
X174. XXII. Site-specific mutagenesis of the A* gene reveals that A* protein is not essential for
X174 DNA replication. J. Mol. Biol. 197:47-54.[CrossRef][Medline]
X174. J. Mol. Biol. 288:595-608.[CrossRef][Medline]
X174. Genetics 128:663-671.[Abstract]
X174, p. 1-71. In R. Calendar (ed.), The bacteriophages, vol. 2. Plenum Publishing Corporation, New York, N.Y.
3: interrelationship of the genome structure and the gene products with those of the phages
X174, G4 and
K. Biochim. Biophys. Acta 1130:277-288.[Medline]
X174 at a resolution of 3.0 Å. J. Mol. Biol. 237:517-543.[CrossRef][Medline]
X174 and its functional implications. Nature 355:137-143.[CrossRef][Medline]
X174. J. Mol. Biol. 125:225-246.[CrossRef][Medline]
X174 codes for a nonessential protein. J. Virol. 33:557-560.
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»