Growth characteristics of mycobacteriophages D28 and D29

Mycobacteriophages: therapeutic approach for mycobacterial infections

Tuberculosis (TB) is a significant global health threat, and cases of infection with non-tuberculous mycobacteria (NTM) causing lung disease (NTM-LD) are rising. Bacteriophages and their gene products have garnered interest as potential therapeutic options for bacterial infections. Here, we have compiled information on bacteriophages and their products that can kill Mycobacterium tuberculosis or NTM. We summarize the mechanisms whereby viable phages can access macrophage-resident bacteria and not elicit immune responses, review methodologies of pharmaceutical product development containing mycobacteriophages and their gene products, mainly lysins, in the context of drug regulatory requirements and we discuss industrially relevant methods for producing pharmaceutical products comprising mycobacteriophages, emphasizing delivery of mycobacteriophages to the lungs. We conclude with an outline of some recent case studies on mycobacteriophage therapy.

Understanding the role of the lysozyme-like domain of D29 mycobacteriophage-encoded endolysin in host cell lysis and phage propagation

Mycobacteriophages are viruses that infect and kill mycobacteria. The peptidoglycan hydrolase, lysin A (LysA), coded by one of the most potent mycobacteriophages, D29, carries two catalytic domains at its N-terminus and a cell wall-binding domain at its C-terminus. Here, we have explored the importance of the centrally located lysozyme-like catalytic domain (LD) of LysA in phage physiology. We had previously identified an R198A substitution that causes inactivation of the LD when it is present alone on a polypeptide. Here, we show that upon incorporation of the same mutation (i.e. R350A) in full-length LysA, the protein demonstrates substantially reduced activity in vitro, even in the presence of the N-terminal catalytic domain, and has less efficient mycobacterial cell lysis ability when it is expressed in Mycobacterium smegmatis. These data suggest that an active LD is required for the full-length protein to function optimally. Moreover, a mutant D29 phage harbouring this substitution (D29^R350A) in its LysA protein shows significantly delayed host M. smegmatis lysis. However, the mutant phage demonstrates an increase in burst size and plaque diameter. Taken together, our data show the importance of an intact LD region in D29 LysA PG hydrolase, and indicate an evolutionary advantage over other phages that lack such a domain in their endolysins.

Real-time PCR using mycobacteriophage DNA for rapid phenotypic drug susceptibility results for Mycobacterium tuberculosis

Managing drug-resistant Mycobacterium tuberculosis requires drug susceptibility testing, yet conventional drug susceptibility testing is slow, and molecular testing does not yield results for all antituberculous drugs. We addressed these challenges by utilizing real-time PCR of mycobacteriophage D29 DNA to evaluate the drug resistance of clinical M. tuberculosis isolates. Mycobacteriophages infect and replicate in viable bacterial cells faster than bacterial cells replicate and have been used for detection and drug resistance testing for M. tuberculosis either by using reporter cells or phages with engineered reporter constructs. Our primary protocol involved culturing M. tuberculosis isolates for 48 h with and without drugs at critical concentrations, followed by incubation with 10(3) PFU/ml of D29 mycobacteriophage for 24 h and then real-time PCR. Many drugs could be incubated instantly with M. tuberculosis and phage for 24 h alone. The change in phage DNA real-time PCR cycle threshold (C(T)) between control M. tuberculosis and M. tuberculosis treated with drugs was calculated and correlated with conventional agar proportion drug susceptibility results. Specifically, 9 susceptible clinical isolates, 22 multidrug-resistant (MDR), and 1 extensively drug-resistant (XDR) M. tuberculosis strains were used and C(T) control-C(T) drug cutoffs of between +0.3 and -6.0 yielded 422/429 (98%) accurate results for isoniazid, rifampin, streptomycin, ethambutol, amikacin, kanamycin, capreomycin, ofloxacin, moxifloxacin, ethionamide, para-aminosalicylic acid, cycloserine, and linezolid. Moreover, the ΔC(T) values correlated with isolate MIC for most agents. This D29 quantitative PCR assay offers a rapid, accurate, 1- to 3-day phenotypic drug susceptibility test for first- and second-line drugs and may suggest an approximate MIC.

Optimization of a phage amplification assay to permit accurate enumeration of viable Mycobacterium avium subsp. paratuberculosis cells

A commercially available phage amplification assay, FASTPlaqueTB (Biotec Laboratories, Ipswich, United Kingdom), when used according to the manufacturer's instructions, does not permit accurate enumeration of Mycobacterium avium subsp. paratuberculosis. The aim of this study was to optimize the phage amplification assay conditions to permit accurate quantification of viable M. avium subsp. paratuberculosis cells. The burst time for M. avium subsp. paratuberculosis was initially determined to inform decisions about optimal incubation time before plating, and then other test parameters were altered to evaluate how the correlation between plaque and colony counts was affected. The D29 mycobacteriophage replicates more slowly in M. avium subsp. paratuberculosis than in Mycobacterium smegmatis (used to optimize the commercial test originally), and the mean burst time for four M. avium subsp. paratuberulosis strains was 210 +/- 36.8 min at 37 degrees C compared to 63 +/- 17.5 min for M. smegmatis mc(2) 155. To achieve 100% correlation between plaque and colony counts, the optimized phage assay includes the following: (i) resuspension of the samples to be tested in Middlebrook 7H9 broth containing 10% oleic acid-albumin-dextrose-catalase and 2 mM calcium chloride, followed by overnight incubation at 37 degrees C before performance of the phage assay; (ii) a 2-h incubation of the sample with D29 mycobacteriophage before viricide treatment; and (iii) a further 90-min incubation after viricide treatment and neutralization up to the burst time (total incubation time, 210 min) before plating with M. smegmatis mc(2) 155 in 7H9 agar. The optimized phage amplification assay was able to detect 1 to 10 CFU/ml of M. avium subsp. paratuberculosis in spiked milk or broth within 48 h, as demonstrated by the results of several blind trials.

Development of a bacteriophage phage replication assay for diagnosis of pulmonary tuberculosis

Successful infection and replication of bacteriophages is indicative of the presence of viable bacteria. We describe here the development of a bacteriophage replication assay for the detection of Mycobacterium tuberculosis by using mycobacteriophage D29. Optimization of phage inoculate and incubation times allowed highly sensitive detection of M. bovis BCG. Fewer than 10 CFU (100 CFU/ml) were detected. No false-positive results were observed in negative samples. Application of the assay to 496 sputum specimens in the National Reference Laboratory of Zambia produced sensitivity, specificity, and positive and negative predictive values of 44.1, 92.6, 82.2, and 67.5%, respectively, compared to culture on Lowenstein-Jensen medium. The equivalent corresponding results for direct fluorescent smear microscopy were 42.3, 96.8, 91.2, and 67.6%. The small increase in sensitivity over that of direct microscopy does not justify the introduction of this technique for routine diagnosis of pulmonary tuberculosis at this time.

Nontuberculous mycobacteria in the environment

It is likely that the incidence of infection by environmental opportunistic mycobacteria will continue to rise. Part of the rise will be caused by the increased awareness of these microbes as human pathogens and improvements in methods of detection and culture. Clinicians and microbiologists will continue to be challenged by the introduction of new species to the already long list of mycobacterial opportunists (see Table 3). The incidence of infection will also rise because an increasing proportion of the population is aging or subject to some type of immunosuppression. A second reason for an increase in the incidence of environmental mycobacterial infection is that these microbes are everywhere. They are present in water, biofilms, soil, and aerosols. They are natural inhabitants of the human environment, especially drinking water distribution systems. Thus, it is likely that everyone is exposed on a daily basis. It is likely that certain human activities can lead to selection of mycobacteria. Important lessons have been taught by study of cases of hypersensitivity pneumonitis associated with exposure to metalworking fluid. First, the implicated metalworking fluids contained water, the likely source of the mycobacteria. Second, the metalworking fluids contain hydrocarbons (e.g., pine oils) and biocides (e.g., morpholine) both of which are substrates for the growth of mycobacteria [53,193]. Third, outbreak of disease followed disinfection of the metalworking fluid [136,137]. Although the metalworking fluid was contaminated with microorganisms, it was only after disinfection that symptoms developed in the workers. Because mycobacteria are resistant to disinfectants, it is likely that the recovery of the mycobacteria from the metalworking fluid [137] was caused by their selection. Disinfection may also contribute, in part, to the persistence of M avium and M intracellulare in drinking water distribution systems [33,89,240]. M avium and M intracellulare are many times more resistant to chlorine, chloramine, chlorine dioxide, and ozone than are other water-borne microorganisms [141,236]. Consequently, disinfection of drinking water results in selection of mycobacteria. In the absence of competitors, even the slowly growing mycobacteria can grow in the distribution system [33]. It is likely that hypersensitivity pneumonitis in lifeguards and therapy pool attendants [139] is caused by a similar scenario.

Inactivation of mycobacteriophage D29 using ferrous ammonium sulphate as a tool for the detection of viable Mycobacterium smegmatis and M. tuberculosis

There is still an urgent requirement for more sensitive, cost-effective methods for detection and susceptibility testing of mycobacteria in clinical samples. We have been investigating a simple bacteriophage-based system which could be used for both purposes. As this depends upon the detection of phages which have successfully infected cells, a key step is the efficient removal or inactivation of phages remaining free in the culture medium. We demonstrate here the use of ferrous ammonium sulphate as an effective agent for the inactivation of mycobacteriophage D29 without impairing phage replication in previously infected host bacteria. Using this property, we report the detection of viable Mycobacterium smegmatis, M. bovis BCG and M. tuberculosis using simple low-cost technology. The method is highly sensitive, since it is able to detect 10 colony-forming units of M. smegmatis. It is also rapid, with the detection of M. tuberculosis in sputum specimens within 48 h.

Construction of D29 shuttle phasmids and luciferase reporter phages for detection of mycobacteria

Diseases caused by Mycobacterium tuberculosis, M. leprae and M. avium, cause significant morbidity and mortality worldwide. Effective treatments require that the organisms be speciated and that drug susceptibilities for the causative organisms be characterized. Reporter phage technology has been developed as a rapid and convenient method for identifying mycobacterial species and evaluating drug resistance. In this report we describe the construction of luciferase reporter phages from mycobacteriophage D29 DNA. Shuttle phasmids were first constructed with D29 in order to identify non-essential regions of the D29 genomes and to introduce unique cloning sites within that region. Using this approach, we observed that all of the D29 shuttle phasmids had the cosmid vector localized to one area of the phage genome near one cohesive end. These shuttle phasmids had been constructed with a cosmid that could be readily excised from the D29 genome with different sets of restriction enzymes. Luciferase reporter phages were made by substituting the luciferase cassette for the cosmid vector. Recombinant phages with the luciferase cassette fall into two groups. One group produced light and had the expression cassette oriented with the promoter directing transcription away from the cohesive end. In contrast, the other group had the expression cassette in the opposite orientation and failed to produce light during lytic infection, but did produce light in L5 lysogens which are known to repress D29 promoters. These results suggest that a phage promoter of the D29 phage can occlude the expression of a promoter introduced into this region. D29 luciferase reporter phages are capable of detecting low numbers of L5 lysogens like L5 luciferase phages. However, unlike L5 luciferase phages, D29 luciferase phages can readily infect M. tuberculosis and M. bovis BCG, demonstrating that these phages can be used to evaluate drug susceptibilities of many types of mycobacteria.

INFECTION OF MYCOBACTERIUM SMEGMATIS WITH D29 PHAGE DNA

DNA extracted from D29 mycobacteriophage produced plaques when plated on Mycobacterium smegmatis 607. The host bacterium did not require alternation such as conversion to protoplasts; cells susceptible to infection with intact phage were susceptible to DNA. The bases found in calf thymus DNA constituted the bases of D29 DNA, adenine being paired with thymine and guanine with cytosine. The dissymmetry ratio (A + T/G + C) was 0.56, and the buoyant density in CsCl was 1.722 with a GC content of 63.77 per cent. The efficiency of plating of the DNA is very much lower than that of intact D29, and it penetrates the host at a slower rate. As does intact phage, D29 DNA requires calcium ions for productive infection of 607. D29 DNA is significantly inactivated by incubation with RNAase, but the inactivation probably results from a complexing with the DNA rather than from enzyme hydrolysis.

Isolation and characterization of temperature sensitive mutants of the F5 deletion mutant of mycobacteriophage D29

Seven thermosensitive mutants of the F5 deletion mutant of the mycobacteriophage D29 were described. The mutants were obtained using N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) mutagenesis, and were characterized using temperature shift assays, complementation and recombination tests, electron microscopy of infected host bacteria at non-permissive temperature, and serum blocking power. Mutants deficient in tail assembly, and mutants deficient in head and tail assembly were described. Mutants deficient in head assembly but capable of assembling tails were not isolated during this study. From the data, 3 provisional linkage map of the phage F5 was proposed.

Phage typing of Mycobacterium kansasii

An improved phage typing scheme of M. kansasii is presented. Ultrasonic treatment of the bacterial suspensions was successfully used in order to obtain homogeneous bacterial growth and to improve the reproducibility of the typing. By means of 9 phages 14 lysis patterns (phage types) could be distinguished. The phage typing was applied to 450 strains of Dutch, British and Czechoslovakian origin. The epidemiological relationships between environmental and clinical isolates of M. kansasii among the strains of Dutch and Czechoslovakian origin was confirmed by the results of phage typing.

Host-phage relationships in the genus Mycobacterium and their clinical significance

Progress made during the last 15 years in the studies on the relationships between mycobacteria and their bacteriophages is reviewed. The basic biology of the phages and the applications of studies on adaptation and host range are discussed in relation to the development of phage typing systems for epidemiological purposes. The nature of lysogeny, its natural occurrence, its experimental establishment, the effect of the lysogenic state on the host bacterium and the evidence that lysogenic mycobacteria are involved in human disease, especially sarcoidosis, is reviewed.

Permeability barrier to rifampin in mycobacteria

A strain of Mycobacterium intracellulare that exhibited natural resistance to rifampin was isolated. The ribonucleic acid polymerase was found to be susceptible to rifampin. Studies with Tween 80 suggested that a permeability barrier against rifampin was present in the intact organism. A type strain of Mycobacterium smegmatis, ATCC 607, considered resistant to rifampin also demonstrated a permeability barrier. The possibility that rifampin resistance in naturally occurring strains of mycobacteria and variable levels of resistance in strains of certain species of this organism are due to permeability barriers is discussed.

Chemical and physical properties of mycobacteriophage D29

Mycobacteriophage D29 has a head of uniform size (average diameter 65 nm) and regular shape and a tail of variable length. The stability of the bacteriophage is optimal between pH 9 and 10. The virus contain double-stranded DNA and six structural polypeptides, three major and three minor. The molecular weights of these six polypeptides, as determined by polyacrylamide gel electrophoresis in the presence of dodecylsulfate, are 150 000, 138 000, 13 000, 66 000 and 24 000. The virus contains no lipids as shown by (a) the lack of structural changes after inactivation of the bacteriophage with chloroform, (b) the absence of lipids containing [32P]phosphate or [35S]sulfate in labeled virus, and (c) the absence of an electron paramagnetic resonance spectrum in bacteriophage which had been incubated with a nitroxide-containing fatty acid.

Nature of the receptor substance of Mycobacterium smegmatis for D4 bacteriophage adsorption

The acetone-soluble fraction extracted from lyophilized cells of Mycobacterium smegmatis ATCC 607 inhibited D4, a species-specific phage active against M. smegmatis. Evidence is presented indicating that the D4 inhibition was caused by the phage receptor substance(s) contained in this fraction. A fraction eluted from silicic acid with chloroform-methanol (95:5, v/v) showed the strongest inhibition of D4 phage. This fraction contained sugars and amino acids, and its infrared absorption spectrum was practically identical with those of the mycoside C isolated from the other species of mycobacteria. Further fractionation revealed that the active material was a mixture of several closely related peptidoglycolipids all of which showed, more or less, the phage inhibition. One of the compounds was purified and partially characterized; it contains three different amino acids, allo-threonine, alanine, and phenylalanine, at a molar ratio of 1:1:1, and also three different deoxyhexoses, probably 6-deoxytalose, 3,4-di-o-methylrhamnose, and 2,3,4-tri-o-methylrhamnose. A tentative name of "mycoside C(sm)" is proposed for this substance which possesses a slightly different structure from the known types of mycoside C and is probably specific for the species of M. smegmatis. A fraction extracted from the D4-resistant mutant of M. smegmatis ATCC 607 by acetone and then by chloroform-methanol (95:5, v/v) showed no phage inhibition and had no sugar component. In addition, this fraction contained lysine, serine, and a small amount of both glycine and an unidentified amino acid.

Infection of competent Mycobacterium smegmatis with deoxyribonucleic acid extracted from bacteriophage B1

A relatively competent state of Mycobacterium smegmatis for infection with deoxyribonucleic acid (DNA) extracted from phage B1 was found in the late log phase of bacterial growth. This state of the culture was used in quantitative studies on the infectivity of the DNA. The buoyant density of B1 DNA was 1.728 g/cc in CsCl, and 1 mug of the DNA produced 84 infective centers, the phage equivalent of which was 1.5 x 10(-8). The infectivity was destroyed by catalytic amounts of deoxyribonuclease but not by specific B1 antiserum. Tween 80, which prevents phage adsorption, did not prevent DNA infection. The response of plaque-forming ability to DNA concentration suggested that two or more molecules are required to initiate an infective center. The low efficiency of DNA infection in mycobacteria was considered to be caused by a limiting population of competent cells in the culture employed; in this experiment less than 10(-5) of the cells were infected with DNA. A typical cycle of infection was observed, although the latent period was prolonged and the burst size reduced after DNA infection. The transition of B1 DNA infection to deoxyribonuclease insensitivity had a lag period of about 10 min, and increased linearly with a velocity of about 0.24 infective centers per min per mug of DNA. Half of the infective titer was inactivated by heating at 92 C for 15 min. The melting temperature was about 96 C. Species barriers were not crossed by B1 DNA; however, the DNA was infectious for a B1-resistant mutant of the host.