Characterization Of An Acinetobacter Baumannii Phage ZZ1and Establishment Of The Method For Screening The Phage That Has The Antibacterial Activity Against Its Host In Vivo

The emergence of pathogenic bacteria resistant to most currently available antimicrobial agents has become a critical problem in modern medicine. Over recent decades, some microorganisms, such as Acinetobacter baumannii, klebsiella pneumoniae, and Pseudomonas aeruginosa, are resistant to all approved antibiotics and can only be treated with experimental, and potentially toxic, drugs. What’s even worse, the discovery and development of the new antibiotics lagged behind the increase of antibiotic-resistant bacterial strains. The concern that humankind is reentering the "preantibiotics" era has become very real, and the development of alternative antiinfection modalities has become one of the highest priorities of modern medicine and biotechnology. Bacteriophages are abundant in all environments. These bacterial viruses are ecologically safe, readily producible and genetically engineered. Phage activity is very specific, attacking only host bacterial cells without affecting other (normal) microflora. The growing incidence of antibiotic-resistance pathogens has led scientists and physicians to examine the possibility of developing phage therapy as an alternative, but reliable, treatment. 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. The increasing prevalence of multi-drug and pan-drug resistant A. baumannii strains found in clinics has rendered it one of the few important nosocomial pathogens. In this study, pan-drug clinical isolates of A. baumannii were collected and used as indicator hosts to screen phages from water samples. A newly discovered phage, named ZZ1, with lytic activity against A. baumannii was selected for further characterization, including complete genome sequences of the phage. Its antibacterial action against A. baumannii was investigated both in vivo and in vitro. Our study lays the theoretic foundation for the success of phage therapy in future.Materials and methods1Isolation and Characterization of the A. baumannii phage ZZ1Several sewage samples were collected. The clinical isolates of A. baumannii were collected and used as indicator hosts to screen phages from water samples. 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 isolate was then propagated on the most sensitive bacterial host using the soft layer plaque technique and broth culture to obtain a high titer stock. Then the stocks were purified by PEG8000precipitation and chloroform extraction followed by caesium chloride (CsCl) density-gradient centrifugation. Images of phage ZZ1were developed using transmission electron microscope. The phage titer was determined by the double-layered method. The host range of ZZ1was investigated. The bacterial hosts of ZZ1were further confirmed as A. baumannii using sequence information derived from their16S rRNA gene and the nucleotide Blast analysis. The one-step growth curve, the temperature and pH stability, and the antimicrobial activity of ZZ1at different temperatures were also investigated. In addition, the genetic material from phage ZZ1was extracted and purified from phage lysate using a MiniBEST Viral RNA/DNA Extraction Kit. The nucleic acid of phage ZZ1was used to determine the susceptibility to DNase I (RNase Free), RNase A (DNase Free), and HindⅢ. Then, the phage DNA was broke into the500-bp fragments randomly by the Covaris S220 system. After the blunt ended fragments with5’-phosphorylated ends by T4DNA Polymerase、Klenow Fragment, adn T4Polynucleotide Kinase,3’-dA overhang by Klenow Fragment exo-, and adapter-modified ends by T4DNA ligase, libraries were amplified by PCR, and sequenced on the Hiseq2000. The whole genome sequence was obtained by Velvet1.2.08assembly2Bioinformatics analyses of the A. baumannii phage ZZl genomeOpen reading frames (CDS) of the phage ZZ1were identified using two bioinformatic software programs, GenMarkS and fgenesV0. The BlastP program (version2.2.15) from NCBI was used to search for sequence similarity of the predicted CDS against the NCBI nonredundant protein sequence database. Protein domain searches were conducted by using Batch CD-Search. Theoretical molecular masses (MM) and isoelectric points (pⅠ) of the ZZ1proteins were obtained using the ProtParam tool available on the ExPASy Web page. The GC content of the ZZ1genes was obtained using DNAStar Lasergene7.1. TMpred was used to identify membrane-spanning regions in CDS. Genomic comparisons were made with Mauve2.2.0and CoreGenes3.5at the nucleotide level and at the proteomic level, respectively. Cumulative GC skew and the putative origin of replication were conducted with GenSkew. tRNAs were identified using the tRNAscan-SE1.21server and confirmed using the ARAGORN program as well as by nucleotide comparison using Blastn. ZZ1phage codon usage was analysed with the CUSP and CAI programs of the EMBOSS package, version6.2.03The Reason Analyses about the inactivation of the A. baumannii phage ZZl in vivo and initial evaluation of a method for screening the phage that has the antibacterial activity against its host in vivoTo determine whether or not the inactivation of the phage ZZ1in vivo was caused by the phage-neutralizing antibody, the reticuloendothelial system, and/or the complement system, we conducted the plaque reduction neutralization test, then investigated the pharmacokinetics of ZZ1in mice and compared the antibacterial activity of ZZ1against its host cells in fresh and heat-inactivated sera, respectively. Four phages, named F19, PG3, PG17, and L2, respectively, were isolated recently. The4phages specifically and efficiently killed4different multi-drug resistance bacterial strain, K. pneumoniae KP-19, E. cloacae EC3, E. cloacae EC17, and P. aeruginosa PA2, respectively. We investigated the antibacterial activity of the4phages against the4bacterial hosts, respectively, in serum. The phage that has the powerful antibacterial activity in serum was further selected to evaluate whether the phage exerts a powerful antibacterial activity against its bacterial host cells in the blood and organs of the mice, and to test whether the reduced bacterial burden in the blood and organs could have therapeutic implications in mice.Results1Characterization of the A. baumannii phage ZZ1The ability of phage ZZ1to form clear plaques of approximately1-2mm in diameter on lawns of AB09V is indicative of lytic phage. Under the transmission electron microscope, phage ZZ1has a100-nm icosahedral head and a120-nm long contractile tail, which allowed for its inclusion in the Myoviridae family of the order Caudovirales. An initial host range investigation in23A. baumannii clinical strains suggested that only3strains can be infected by phage ZZ1. No other strains tested was found susceptible to phage ZZ1. Analysis of the partial16S rRNA genes of the3hosts of ZZ1confirmed that the3strains were A. baumannii. Under the same culture conditions, the antibacterial activity of ZZ1was highest in strain AB09V, followed by AB0902and then AB0901. One-step growth curve of ZZ1on AB09V indicated that the ZZ1latent period was approximately9min, and the burst size averaged200PFU per infected cell. The phage was stable over a wide pH range (4to9) and at extreme temperatures (between50℃and60℃), and was able to exhibit the most powerful antibacterial activity at temperatures ranging from35℃to39℃. The phage nucleic acid was susceptible to degradation by DNase and HindⅢ but unaffected by RNase. Thus, ZZ1is a dsDNA phage. A total of937,400reads with an average length of250bp were assembled into a single contig of166,801bp with114-bp inverted terminal repeats for the phage ZZ1. 2Bioinformatics analyses of the A. baumannii phage ZZ1genomeThe single copy ZZ1genome is166,687bp long. The genome has an average GC content of34.4%, which is slightly lower than the GC content of A. baumannii (38.9%to39.2%). A total of256protein-coding genes and8tRNA genes were predicted. There are approximately1.5genes per kbp, and93.6%of the ZZ1genome is predicted to encode proteins. The length of these putative coding sequences (CDSs) range in length from34to1,303amino acid residues (aa) and average203aa. Of the256CDSs of ZZ1, only95(37.1%) could be assigned a functional annotation, all of which are T4-like proteins, including36virion proteins,21proteins involved in DNA replication, recombination, repair, packaging, and processing,11proteins involved in nucleotide metabolism,5proteins involved in chaperonins/assembly catalysts,7proteins involved in transcription,3proteins involved in translation,3proteins involved in lysis,5proteins involved in host alteration/shutoff,2proteins involved in host or phage interactions, and2proteins involved in homing endonucleases and homologs. Of the161proteins with unknown functional annotations,37have the membrane-spanning regions, and23have the conserved domain information. Notably, ZZ1CDS011that has four membrane-spanning regions can be predicted to belong to Multi_Drug_Res superfamily. No other ZZ1proteins have similarities to any other known virulent, toxin, or pathogen-associated protein family.Comparative genomics of phage ZZ1, E. coli phage T4and other4T4-like Acinetobacter phages (Acj9, Acj61, Ac42, and133) revealed a total of11local colinear blocks (LCB),7of which (>10kb) are held in common among the6phages. The protein-by-protein comparison of the proteomes using BlastP and CoreGenes revealed that ZZ1shares179(69.9%) protein homologues, with Acinetobacter phage Acj9,164(64.1%) with Acj61,157(61.3%) with133, and143(55.9%) with Ac42, although it shares only110protein homologues with coliphage T4(E value<10’6), which is the evidence that the phage ZZ1belong to the same genus, the T4phage superfamily.A putative origin of replication in the region of nt81009and a putative terminus of replication in the region of nt1were revealed by using GC-skew analysis. The putative origin of replication is close to the tRNA-existent region. ZZ1has different numbers and types of tRNAs than the other4Acinetobacter phages do. Some of the ZZ1-encoded tRNAs share high sequence similarity with the tRNAs of these phages. The number and type of homologous tRNA appears to confirm the relationship between ZZ1and the other phages, suggesting some elements of vertical evolution in these tRNAs. Over half of ZZ1-encoded tRNAs (5out of8) are related to the optimal codon usage of ZZ1proteins. However, this correlation was not present in any of the other4Acinetobacter phages.3The Reason Analyses about the inactivation of the A. baumannii phage ZZ1in vivo and initial evaluation of a method for screening the phage that has the antibacterial activity against its host in vivoPhage ZZ1has no efficacy in treatment of the mouse model by AB09V infection (p>0.05). Further investigation suggested that no ZZ1-neutralizing antibody was in serum of the mice; and the reticuloendothelial system of the mice also has no immediate bearing on the inactivation of ZZ1in vivo. However, the interaction between the complement systems with bacterial cells in serum is interfering with the adsorption of ZZ1, which is the main reason of the phage losing the antibacterial activity against its bacterial host in vivo. Thus, the antibacterial activity of phage against pathogenic bacteria in isolated tissue (such as fresh serum, fresh anticoagulation blood, or other fresh homogenized tissue) might be an in vitro method of evaluating the treatment efficacy of phage in vivo.Four phages, F19, PG3, PG17, and L2, are used for evaluating the efficiency of the method. Except for the phage L2, the other3phages, F19, PG3, and PG17, have the positive antibacterial activity against its bacterial host, KP-19, EC2, and EC17respectively, in fresh serum. A single administration of the3phages led to the survival of almost all of the mice that were infected by the LD100bacterial cells:F19rescued100%of the mice (8of8), PG3rescued87.5%of the mice (7of8), and PG17rescued87.5%of the mice (7of8). Further investigation suggested that the3phages significantly decreased the bacterial burden in the blood and organs of mice in vivo (p<0.001). In addition, the administration of heat-inactivated phage did not protected mice from death (survival rate is0%,0/8), and the amount of viable bacteria in the blood and various organs was similar to that observed in the control group (p>0.05). Therefore, the ability of this phage to rescue bacteremic mice was demonstrated to be due to the functional capabilities of the phage and not to a nonspecific immune effect. These results initially confirmed that the antibacterial activity of phage against pathogenic bacteria in isolated tissue might be an efficient method of evaluating the treatment efficacy of phage in vivo.Conclusions1. Morphologically, phage ZZ1belongs to the Myoviridae family of the order Caudovirales. AB09V is the most sensitive host of ZZ1. The phage formed clear plaques of approximately1-2mm in diameter on AB09V lawns. ZZ1exhibited a9-min latent period and burst size of200PFU/cell. When ZZ1alone was incubated at different pHs and different temperatures, the phage was stable over a wide pH range (4to9) and at extreme temperatures (between50℃and60℃). Moreover, ZZ1exhibited the most powerful antibacterial activity at temperatures ranging from35℃to39℃.2. ZZ1is a dsDNA phage. Its166,687-bp genome has a GC content of34.4%, carries a total of256potential open reading frames and contains8putative tRNA genes.95protein-coding genes could be assigned a functional annotation, most of which are DNA replication and virion structural proteins. Comparative genomics analysis suggested that the phage ZZ1belong to the T4phage superfamily.3. A single i.v. injection of phage ZZ1has no efficacy in treatment of the mice that were infected by i.p. injection with the LDioo of AB09V. The complement systems are the main reason of the phage ZZ1losing the antibacterial activity against AB09V cells in vivo. The antibacterial activity of phage against pathogenic bacteria in isolated tissue might be an efficient method of evaluating the treatment efficacy of phage in vivo.