|Home | About | Journals | Submit | Contact Us | Français|
Acinetobacter baumannii, a significant nosocomial pathogen, has evolved resistance to almost all conventional antimicrobial drugs. Bacteriophage therapy is a potential alternative treatment for multidrug-resistant bacterial infections. In this study, one lytic bacteriophage, ZZ1, which infects A. baumannii and has a broad host range, was selected for characterization.
Phage ZZ1 and 3 of its natural hosts, A. baumanni clinical isolates AB09V, AB0902, and AB0901, are described in this study. The 3 strains have different sensitivities to ZZ1, but they have the same sensitivity to antibiotics. They are resistant to almost all of the antibiotics tested, except for polymyxin. Several aspects of the life cycle of ZZ1 were investigated using the sensitive strain AB09V under optimal growth conditions. ZZ1 is highly infectious with a short latent period (9min) and a large burst size (200 PFU/cell). It exhibited the most powerful antibacterial activity at temperatures ranging from 35°C to 39°C. Moreover, when ZZ1 alone was incubated at different pHs and different temperatures, the phage was stable over a wide pH range (4 to 9) and at extreme temperatures (between 50°C and 60°C). ZZ1 possesses a 100-nm icosahedral head containing double-stranded DNA with a total length of 166,682bp and a 120-nm long contractile tail. Morphologically, it could be classified as a member of the Myoviridae family and the Caudovirales order. Bioinformatic analysis of the phage whole genome sequence further suggested that ZZ1 was more likely to be a new member of the Myoviridae phages. Most of the predicted ORFs of the phage were similar to the predicted ORFs from other Acinetobacter phages.
The phage ZZ1 has a relatively broad lytic spectrum, high pH stability, strong heat resistance, and efficient antibacterial potential at body temperature. These characteristics greatly increase the utility of this phage as an antibacterial agent; thus, it should be further investigated.
Bacteriophage therapy is one of the emerging methods used to overcome bacterial infections [1-3]. Bacteriophages are viruses that infect and kill bacteria. Theoretically, phages have several advantages over antibiotics. They are highly specific and very effectively lyse targeted pathogenic bacteria. They are safe because they have no activity against animal or plant cells. Phages are ubiquitous, so isolation of new phages is a relatively rapid process that can frequently be accomplished in days or weeks. The use of phages as therapeutic agents was initiated in 1919, 3years after their discovery, for the treatment of bacillary dysentery and continued until the 1940s. Over this time period, phages were used to treat a variety of infectious diseases . However, with the advent of antibiotics, commercial production of therapeutic phages ceased in most of the Western world . Currently, there is renewed interest in phage research and the applications of bacteriophages as potentially powerful antibacterial agents due to the emergence of drug-resistant pathogens and the dearth of new antibiotics. Several studies have shown that bacteriophages can be used successfully for therapeutic purposes, both in humans and animals [6-9]. However, more research is required before clinical use can be re-initiated. Before using a phage for therapeutic purposes, the isolation of lytic phages and characterization of the phage are essential.
In this study, clinical isolates of Acinetobacter baumannii were collected and used as indicator hosts to screen phages from water samples. A. baumannii mostly infects debilitated patients in intensive care units and is associated with high mortality rates [10,11]. Since its discovery, A. baumannii has become resistant to many common antibiotics . The increasing prevalence of multidrug- and pandrug-resistant A. baumannii strains in clinics has rendered it one of the most important nosocomial pathogens [12-15]. Fortunately, lytic phages specific to A. baumannii might provide an alternative to antibiotics for the prevention and treatment of infections caused by this bacterium. However, to the best of our knowledge, very few detailed characterizations of A. baumannii phages have been published [16,17]. This paper describes the isolation and characterization of a novel virulent phage, ZZ1, that infects A. baumannii clinical isolates.
Twenty-three A. baumannii clinical isolates were screened for phage present in a sample of fishpond water. Among these, only the strain AB09V could serve as an indicator for ZZ1 in the initial screening. This phage formed clear plaques of approximately 1-2mm in diameter on AB09V lawns. AB09V was thus used to propagate, purify and characterize the phage. As shown in Figure Figure1,1, the phage ZZ1 has a 100-nm icosahedral head and a 120-nm long contractile tail. Morphologically, phage ZZ1 can be tentatively classified as a member of the Myoviridae family in the order of Caudovirales. Most of the input phages rapidly adsorbed to AB09V cells. Appearance of ghost particles 5min after mixing phages with bacteria suggested that ejection of DNA from the phage head occurred rapidly.
Two additional natural bacterial hosts, AB0901 and AB0902, were found when the other 22 of the 23 A. baumannii clinical isolates were used to investigate the host range of ZZ1 by spot test. This test used a higher concentration of phage (108 PFU/ml) than the original screen. Interestingly, some differences were observed in the ability of the phage to lyse the 3 bacterial hosts (AB09V, AB0901, and AB0902). For example, as shown Figure Figure2,2, ZZ1 was capable of forming transparent areas on lawns of the strains AB09V, AB0901, and AB0902. However, the minimum phage concentrations required to form clear spots on each lawn were different: AB09V required 105 PFU/ml, AB0902 required 106 PFU/ml, and AB0901 required 108 PFU/ml. The values suggest that under the same culture conditions, the antibacterial activity of ZZ1 was highest in strain AB09V, followed by AB0902 and then AB0901. There might be natural resistance mechanisms in AB0901 and AB0902; thus, the strain AB09V is likely the most sensitive indicator of the phage titer of the 3 strains and is the best host for phage propagation.
The 3 strains that can be infected by ZZ1 were selected for further detailed study. They were resistant to all of the antibiotics tested except polymyxin (amikacin, gentamicin, imipenem, meropenem, cefazolin, ceftazidime, cefotaxime, cefepime, aztreonam, ampicillin, piperacillin, amoxicillin/clavulanic acid, ampicillin/sulbactam, piperacillin/tazobactam, sulfanilamide, sulfamethoxazole and trimethoprim, ciprofloxacin, levofloxacin, and tetracycline). Partial 16S rRNA genes of the 3 strains were sequenced and deposited in GenBank under the accession numbers [GenBank: JF313142] (AB09V), [GenBank: JF313143] (AB0901), and [GenBank: JF313144] (AB0902). Nucleotide blast analysis further confirmed that the three strains were A. baumannii.
Temperature and pH stability are two important parameters in the storage of therapeutic phage. Thus, the stability of ZZ1 was investigated at different pHs and temperatures. As shown in Figure Figure3,3, no obvious effect on ZZ1 activity was observed after 1h of incubation at pH levels ranging from 4 to 9. However, the phage completely lost its infectivity at pH 10 or higher and pH 3 or lower. Within 1h of incubation at pH 4, the phage infectivity decreased by approximately 87%, and a titer of 6.0×108 PFU/ml of active phage ZZ1 was detected at the end of the incubation. The maximum stability of the phage was observed at a pH between 6 and 7. In addition, the results of thermal stability tests shown in Figure Figure44 suggest that ZZ1 was relatively heat stable over 1h at temperatures between 50°C and 60°C, and no significant loss in phage activity was observed. At 70°C, the phage titer quickly dropped, and no viral particles were detected after 40min. Furthermore, phage activity was completely lost at 80°C within the first 1min of incubation. The ZZ1 phage lysate retained almost 100% of its infection activity when stored at both 25°C and 4°C for several months (data not shown).
Optimal A. baumannii growth occurs over a very broad temperature range . As shown in Figure Figure5,5, the AB09V lawns grew well on nutrient agar plates at temperatures ranging from 25°C to 42°C. However, the antimicrobial activity of ZZ1 is clearly influenced by temperature variations. When the plates were incubated at different temperatures, the minimum phage concentrations required to form clear spots on AB09V lawns were different: 105 PFU/ml at 35°C, 37°C, and 39°C; 106 PFU/ml at 30°C and 40°C; 108 PFU/ml at 25°C; and 109 PFU/ml at 42°C. Thus, we concluded that ZZ1 exhibits the most efficient antibacterial activity against the AB09V strain at temperatures ranging from 35°C to 39°C, which suggested that ZZ1 could exhibit powerful antibacterial activity against AB09V at body temperature.
The growth characteristics of ZZ1 on the sensitive indicator strain AB09V were characterized under optimal growth conditions. Phage ZZ1 exhibited high infection efficiency after mixing the phages and AB09V cells. We inferred that almost all of the A. baumannii AB09V were infected prior to the burst time of the first infected cell because the number of bacteria surviving at 9min was less than 100CFU/ml. Moreover, as shown in Figure Figure6,6, the total plaque count was 6.6×108 PFU/ml at the beginning of infection (0min), and only 2.3×108 PFU/ml remained after 9min. The difference (approximately 4.3×108 PFU/ml) originated from adsorption of multiple phage particles to one susceptible bacterial cell. The decrease in the number of phages was greater than 6-fold higher than the initial number of bacterial cells (approximately 7×107CFU/ml). These results further confirmed that almost all of the bacterial cells could be infected within the latent period (9min). The number of unattached phages at the end of the latent period (or prior to the burst time of the first infected cells) can be estimated as the difference between the number of the total plaque count and the initial number of bacterial cells. The calculated number of unattached phages was 1.6×108 PFU/ml, which is negligible compared to the phage number at the end of the experiment (1.5×1010 PFU/ml). Moreover, the number of bacteria surviving at the end of the experiment is less than 50CFU/ml, which can also be considered negligible when compared to the initial number of bacterial cells (7.0×107CFU/ml). Therefore, the average burst size was approximately 200 PFU/cell, which can be calculated as the ratio of the final count of phage particles to the initial count of infected bacterial cells.
The phage nucleic acid was susceptible to degradation by DNase and some restriction endonucleases (e.g., HindIII, EcoRI, and EcoRV) but unaffected by RNase. Thus, ZZ1 is a dsDNA phage (data not shown). The ZZ1 genome has a total length of 166,682bp and a GC content of 34.3%, which is slightly lower than that described for the A. baumannii ATCC 17978 strain (38%, accession number NC_009085). An initial NCBI nucleotide blast analysis (blastn) of the complete genome sequence indicated that ZZ1 shares limited similarities with other known phage nucleotide sequences, which confirmed its status as a novel Acinetobacter phage species. The top 4 most similar sequences found were of the Acinetobacter phages Acj9 [GenBank: HM004124.1], Acj61 [GenBank: GU911519.1], Ac42 [GenBank: HM032710.1], and 133 [GenBank: HM114315.1]. The max scores were 4662 (50% of coverage, 89% of max ident), 4448 (45% of coverage, 87% of max ident), 2634 (34% of coverage, 94% of max ident), and 2210 (31% of coverage, 92% of max ident). The four Acinetobacter phages were recently deposited in GenBank and were previously annotated as T4-like phages . No other Acinetobacter phages were hit by blastn. In addition, Enterobacteria phage T4 ranked tenth, and its max score was 1972 (28% of coverage, 83% of max ident), suggesting that the ZZ1 phage might be a new member of the T4-like phage family.
A sequence search using the NCBI open reading frame (ORF) finder revealed a total of 402 putative ORFs of 50 or more codons in the ZZ1 genome that have limited similarity to other known phage proteins. Among them, 118 ORFs have the highest similarity to predicted ORFs from the Acinetobacter phage Acj9; 47 ORFs are most similar to predicted ORFs from the Acinetobacter phage Acj61; 18 ORFs most closely resemble predicted ORFs from the Acinetobacter phage 133; and only 13 ORFs have the highest score with predicted ORFs from the Acinetobacter phage Ac42. In addition, of the 402 ORFs, 105 ORFs showed homology with sequences in GenBank with annotated function; 244 ORFs had matches with uncharacterized entries; and the remaining 53 ORFs had no match to sequenced genes in the database.
Phage therapy has been the subject of several recent reviews, and the present study reinforces the view that it is worth exploring [1,2,19]. To the best of our knowledge, the characterization of lytic phages of A. baumannii has rarely been studied, although Ackermann et al. [16,20] described the classification of an A. baumannii phage, and Soothill et al. [1,21] tested the efficacy of phage therapy for experimental A. baumannii infections in mice. In this study, we focused our efforts on the isolation and characterization of A. baumannii phages with potential for prophylactic/therapeutic use.
Phages are thought to be found wherever bacteria thrive . Acinetobacter spp. are ubiquitous organisms that can be readily isolated from non-clinical sources, such as soil, drinking and surface waters, sewage, and a variety of different foods , which suggests that phages specific to A. baumannii might also be easily isolated from nature. Recently, 10 phages were obtained from wastewater using 125 clinical isolates of A. baumannii as indicator hosts [20,23]. These phages were designated AB1 to AB9 and AB11. Examination by transmission electron microscopy suggested that phages AB1-7 and AB9 belonged to the Podoviridae family, and phages AB8 and AB11 belonged to the Myoviridae family. Two of the 10 phages, AB1 and AB2, were further characterized at 35°C and 37°C, respectively. Based on morphology and genomic analysis, the two phages were classified as new members of the ϕKMV-like phages [20,23]. In this study, the phage ZZ1, which is specific to A. baumannii, was isolated from fishpond water, which further confirmed that phages specific to A. baumannii are waiting to be exploited as an abundant natural resource. The ability of phage ZZ1 to form clear plaques on lawns of AB09V is indicative of lytic phage, and a large burst size with a short latent period is further suggestive of the lytic nature of phage ZZ1. Morphologically, ZZ1 exhibits features similar to the Myoviridae family (order Caudovirales), which, broadly, are tailed phages with icosahedral head symmetry and contractile tail structures. Genome analysis of ZZ1 showed that it is fairly similar to four other Acinetobacter phages (Acj9, Acj61, Ac42, and 133). In a recent review by Petrov et al. , the four Acinetobacter phages were initially assigned to the “T4-like Viruses” genus. Each of these Acinetobacter phages has a unique set of ORFs that occupy ~35% of the genome. That is, each represents a different type of T4-related phage genome . The genome size of the phage ZZ1 (166,682bp) is also close to the genome size of T4-like phages. These genomes vary between ~160,000 and ~250,000bp . Therefore, the above features confirmed that the phage ZZ1 is most likely a new member of the T4-like virus family of Acinetobacter phages. However, according to the 2011 Virus Taxonomy List (current) from the International Committee for the Taxonomy of Viruses (http://www.ncbi.nlm.nih.gov/ICTVdb/index.htm.), only the Acinetobacter phage 133 can be searched and classified in the unassigned genus of the Myoviridae family, most likely because the phage is inadequately characterized. At the very least, the current sequence database for the Myoviridae phages should prove to be a rich source of genetic markers for bioprospecting and a mine of reagents for basic research and biotechnology. Our future research will focus on further detailed analysis of the whole ZZ1 genome to understand the genetic characteristics of this phage.
The main aim of this study was the isolation and characterization of a lytic bacteriophage with potential for prophylactic/therapeutic use. Therefore, the antibacterial activity of the phage against its different host cells was the focus of our research. Through the preliminary in vitro host range investigation, ZZ1 was found to have different antibacterial activity against 3 of its natural hosts. The antibacterial activity of ZZ1 was highest against the strain AB09V, followed by AB0902 and then AB0901, based on the minimum phage concentration required to form clear spots at 37°C. The natural resistance mechanisms of AB0901 and AB0902 against ZZ1 are worth further investigation in future studies. With respect to its life cycle in the sensitive strain AB09V, ZZ1 proliferates efficiently, with a short latent period (9min), a large burst size (200 PFU/ml), and a high adsorption rate. Remarkably, only less than 50CFU/ml of the AB09V cells remained viable 30min after AB09V cells were mixed with ZZ1 particles at a multiplicity of infection (MOI) of 10 at 37°C. Moreover, ZZ1 exhibited the most powerful antibacterial activity at temperatures ranging from 35°C to 39°C, suggesting that the phage would be highly effective when placed inside the body at normal or near normal body temperature. In addition, ZZ1 was stable over a wide pH range (4-9) and was strongly resistant to heat. All of these features have implications for the use of this phage as a stable therapeutic agent for the treatment of A. baumannii infections, especially those caused by the strain most sensitive to the phage, AB09V. The differences in the antibacterial activity of ZZ1 against the strains tested will be the focus of our future research both in vitro and in vivo.
This study provides information about a novel virulent A. baumannii phage. Our future research will examine the application of this characterized phage in treating infections by A. baumannii clinical isolates both in vivo and in vitro.
Twenty-three clinical strains of A. baumannii were used in this study for phage isolation and phage host investigation. All of these strains were isolated from the sputum of hospitalized patients at the Henan Province People’s Hospital in Zhengzhou, China. After obtaining the approval of the Life Science Ethics Committee of Zhengzhou University and written informed consent, sputum samples were collected for the purposes of this study. The automated system BD Phoenix (Becton Dickinson Diagnostic Systems, Sparks, MD, USA) was used on clinical samples for the identification of bacteria and for antibiotic susceptibility tests. Only 3 of the 23 strains could be lysed by ZZ1; these were lysed to varying degrees. Therefore, the 3 strains were designated AB09V, AB0901, and AB0902 in our nomenclature. The 3 strains selected for use in this study were further confirmed as A. baumannii using sequence information derived from their 16S rRNA gene. Briefly, bacterial DNA was isolated as previously described . The extracted DNA was used as the PCR template to amplify the 16S ribosomal RNA coding regions. The ClustalX 2.0 program and Oligo 4.0 primer analysis software were used for universal primer design based on homology profiles among the 16S rRNA genes of A. baumannii strains reported in the GenBank sequence database. The universal primers 199f (5' CTA CGG GAG AAA GCA GGG GAT 3') and 1344r (5' TTA CTA GCG ATT CCG ACT TCA 3') were used to amplify partial 16S rRNA gene sequences. To increase the specificity of amplification and to reduce the formation of spurious byproducts, a “touchdown” PCR was performed (the annealing temperature decreased from 65 to 55°C for 20 cycles) as described previously . The PCR amplicons were purified with a CONCERT Rapid PCR purification kit (Invitrogen) and were then sequenced directly with the primers.
Phage isolation was conducted using the method described by Adams . Several water samples (municipal sewage, fishpond water, and river water) collected from different places in Zhengzhou, China, were clarified by centrifugation (12,000×g for 15min at 4°C). One percent (v/v) of a bacterial broth culture (overnight growth) along with an equal volume of nutrient broth at double concentration was added to the cleared supernatant and incubated at 37°C overnight. The next day, after centrifugation (12,000×g for 20min at 4°C), the supernatant was filtered with a 0.45μm SFCA Corning syringe filter (Corning Inc., Corning, NY) to remove the residual bacterial cells. An aliquot (0.2ml) of the filtrate was mixed with 0.1ml of an overnight culture of an A. baumannii strain and 2.5ml of molten top soft nutrient agar (0.7% agar) at 47°C then overlaid on the surface of solidified base nutrient agar (1.5% agar) at 37°C. After incubation overnight at 37°C, the phage plaques were picked from the plates, and each individual plaque was re-isolated three times to ensure the purity of the phage isolate . The phage titer was determined by the double-layered method .
Phage stocks were prepared on the most sensitive bacterial host using the soft layer plaque technique. Briefly, 10ml of an overnight AB09V bacterial culture was concentrated to 1ml by centrifugation (3,000×g for 10min). One hundred microliters of the concentrated culture (1010CFU/ml) and 0.1ml of the phage ZZ1 (107PFU/ml) were added to 2.5ml of molten top soft nutrient agar (0.4% agar) then overlaid on the surface of solidified base nutrient agar (1.5% agar). The plates were incubated for 6-8h at 37°C and were used to prepare a concentrated phage suspension (1011PFU/ml) by eluting the top agar overlaid plates in 5ml SM buffer. Phage stocks were stored at 4°C after filtration through 0.45-μm filters.
The host range of the phages was examined by spot tests on 23 A. baumannii clinical strains. A 0.1ml aliquot of bacterial overnight broth culture (109CFU/ml) was mixed with melted 0.7% soft nutrient agar (47°C), and this mixture was poured onto 1.5% solid agar to make double layer ager plates. When the top agar hardened, phage stock (5μl) from a dilution series was spotted on each plate with different bacterial strains. The plates were incubated at 37°C. When clearing zones were observed, the antibacterial activity of the phages against each bacterial host was assessed based on the minimum phage concentration required to form a completely transparent zone.
The antibacterial activity of ZZ1 against A. baumannii AB09V was evaluated by serial dilution spot testing at different temperatures. Phage stock (5μl) from a dilution series was spotted onto a lawn of AB09V in top agar. The plates were examined for cell lysis after overnight incubations at 25°C, 30°C, 35°C, 37°C, 39°C, 40°C, and 42°C. The optimal antibacterial temperature was determined by comparing the minimum phage concentration required to form a completely transparent zone.
An overnight culture of strain AB09V (1ml) was inoculated into fresh medium (100ml) and incubated with shaking at 37°C for approximately 1h to yield a cell density of approximately 7.0×107CFU/ml (at an OD600 of 0.15). A 1ml sample of a nutrient broth suspension of the phage ZZ1 at an approximate MOI of 10 was added to this culture. Samples were periodically withdrawn and immediately chilled while being further diluted to measure total phage activity (including infected bacterial cells and free phages) by the double-layered-agar plate technique. Bacterial viable counts were determined before the bacteria were mixed with the phage and were assessed periodically. Burst size was estimated from triplicate experiments using the equation described by Jiang et al. . Each experiment was performed three times, and the results are reported as the mean of three observations±standard deviation (SD).
Resistance to different pH values at 37°C was determined according to the methods described by Verma et al. . The pH of the nutrient broth was adjusted with either 1M HCl or 1M NaOH to obtain a pH within the range of 2–11. A total of 100μl of bacteriophage suspension (4.7×1011 PFU/ml) was inoculated into 10ml of pH-adjusted medium. After incubation for 1h at 37°C, the surviving phages were diluted and counted immediately using the soft agar overlay method at 37°C. Moreover, according to the methods described by Capra et al. , the stability of ZZ1 at various temperatures (50°C, 60°C, 70°C, and 80°C) was checked by incubating the phage (3.2×1010 PFU/ml) at the indicated temperature for 1h at pH 7.0 in nutrient broth; the surviving phages were then counted using the soft agar overlay method at 37°C.
AB09V cells were infected with ZZ1 during the exponential growth phase (OD600=0.35) at an MOI of approximately 100 and incubated at 37°C for 5min in nutrient broth medium. The mixture was fixed with 1% glutaraldehyde at 0°C for 60min and then centrifuged (4500×g, 3min). The phage-cell complexes from the pellets were placed on 300-mesh copper grids coated with carbon film and then stained with phosphotungstic acid (2%w/v, pH 7.2) for 30s. After drying, the preparation was examined by a transmission electron microscope.
The nucleic acid of phage ZZ1 was isolated as previously described . Purified nucleic acid was used to determine susceptibility to DNase, RNase, and restriction enzymes and was then sent to Zhejiang California International NanoSystems Institute (Hangzhou, China) for commercial sequencing. The whole genome sequence, with a total length of 166,682bp, was obtained using the Illumina Solexa Sequencing platform (Illumina, San Diego, USA) and the Swift analysis tool (http://swiftng.sourceforge.net) . The genome sequence was analyzed with the NCBI BlastX bioinformatics tool (http://blast.ncbi.nlm.nih.gov/Blast.cgi) for nucleotide analysis, and the NCBI ORF finder (http://www.ncbi.nlm.nih.gov/projects/gorf/) was used to identify ORFs, which were limited to those encoding proteins of greater than or equal to 50 amino acids. Homology assignments between genes from other phages and predicted ORFs of phage ZZ1 were based on amino acid sequence alignment searches (BlastP, http://blast.ncbi.nlm.nih.gov/Blast.cgi).
The genome sequence, with a total length of 166,682bp, for phage ZZ1 described in this work was submitted to GenBank and was assigned the accession number [GenBank: HQ698922].
The authors declare that they have no competing interests.
JJ conceived of the study and designed all the experiments and drafted the manuscript; ZJL, SWW, and DHH performed all phage-related experiments; SMW, YYM, and JW analyzed the clinical bacteria strains; FL and XDC participated in the TEM investigation; YHL, GXL, and XTW analyzed the phage genome. GQZ and ZQW participated in the design of the study and coordination and helped to draft the manuscript. All authors read and approved the final manuscript.
This study was supported by a Project of Open Research Fund Program of the State Key Laboratory of Virology of China (No. 2011007) and a Project of the Education Department of Henan Province (No. 2011C310014).