Bacteriophage application to control the contaminated water with Shigella

Shigella is one of the most important waterborne and foodborne pathogens around the world. Emergence of antibiotic-resistant Shigella has made the development of alternatives to conventional antibiotics necessary. In this study, a virulent Myoviridae bacteriophage, pSs-1 was isolated from environmental water in South Korea and showed infectivity to S. flexneri as well as S. sonnei strains. One-step growth analysis showed that pSs-1 has a short latent period (25 min) and a large burst size (97 PFU/cell). According to the genomic analysis, pSs-1 contains 164,999 bp of genome with a G + C content of 35.54% and it is considered as a member of the T4-like bacteriophage group. These results showed that pSs-1 may have potential as a biocontrol agent instead of conventional antibiotics for shigellosis.

boydii 12 . pSf-1 was able to infect all of the S. flexneri and most of S. sonnei strains, forming clear plaques 11 ; pSb-1 was able to infect all of the S. boydii strains and formed clear plaques 12 . Although the virulent Shigella phages, pSf-1 and pSb-1 showed the potential usefulness against shigellosis as reported previously, the two phages were considered to present restricted effectiveness because S. sonnei ATCC ® 11060 was not infected by any of two phages 11,12 . This study aimed to isolate a virulent phage infecting S. sonnei strains in order to make the best combination of various phages for use in a phage cocktail as previously noted that the isolation of S. sonnei phage was needed 12 .
Previous research in our laboratory led to the isolation of S. sonnei phage as a first priority rather than the isolation of different phages infecting other Shigella species. A virulent Shigella phage, pSs-1 infecting S. sonnei and S. flexneri was isolated from the Hongjecheon stream in Seoul in April 2012. From the isolated phages, pSs-1 was selected for further studies depending on the clarity of plaque and the phage titer after a single propagation.
The morphology of pSs-1 places it in the family Myoviridae according to the classification system of Ackermann 13 (Fig. 1C). The tail length and width were 120 ± 7 nm (mean ± SD) (n = 10) and 18 ± 2 nm (n = 10), respectively, and the head diameter was 98 ± 4 nm (n = 10). pSs-1 inhibited all of the S. sonnei and S. flexneri strains, producing clear plaques on all of the strains except one S. flexneri strain, ATCC ® 12022 (turbid plaque). However, no S. boydii strain was infected by pSs-1. The high EOP value was obtained with S. flexneri ATCC ® 29903 ( Fig. 1B) although no strain had a higher value than the indicator host strain, S. sonnei ATCC ® 25931 (Table 1). Furthermore, pSs-1 was not able to infect Escherichia coli strains used in this study. The result of one-step growth analysis revealed that pSs-1 had a short latent period (25 min) and a large burst size (97 PFU/cell) (Fig. 2).
The bacteriolytic effect of pSs-1 was tested on early phase cultures of S. sonnei ATCC ® 25931 (Fig. 3A) and S. flexneri ATCC ® 29903 (Fig. 3B). The OD 600 values of the uninfected control culture (MOI: 0) continued to increase during the incubation period. In contrast, the growth of bacteria infected by pSs-1 was retarded at MOIs of 0.1, 1, and 10; the bacterial growth was inhibited most effectively at an MOI of 10. pSs-1 lysed S. sonnei (3 h after incubation) more readily than S. flexneri (12 h after incubation).
The effective bacteriolytic activity and high EOP value of pSs-1 against S. sonnei and S. flexneri strains indicated that pSs-1 could be used for control of both S. sonnei and S. flexneri although pSs-1 was proved to use only S. sonnei strain as a host bacterium; the propagation trials of pSs-1 using S. flexneri strains have not resulted in high titer for further study. These results conclude that the combination of pSf-1, pSb-1, and pSs-1 can inhibit all of the Shigella strains used in this study, increasing the possibility for Shigella control. The bacteriolytic activity of pSs-1 was found to be active for 1 h over the temperature range of 4-50 °C or the pH range of 5.0-9.0 toward S. sonnei and S. flexneri strains although significant reduction of its activity was observed at 50 °C (data not shown). This result emphasized that pSs-1 could be used in various natural environments, especially poor sanitation surroundings.
Comparative genomic analysis of phage. According to the genome sequencing results, the genome size of pSs-1 is 164,999 bp, with a 35.54% total G + C content. The genomic sequence of the Shigella phage pSs-1 which was described in this study has been deposited in the GenBank database under the accession number KM501444. The predominant start codon was ATG; 10 ORFs such as ORF28, ORF51, ORF66, ORF67, ORF84, ORF112, ORF150, ORF178, ORF189, and ORF238 started with an uncommon start codon (GTG). ORFs with a length of at least 34 amino acids were selected. A total of 26 promoters, 14 transcriptional terminator regions, and 266 ORFs were predicted in the genome. However, only 121 ORFs (45.49%) were determined to be functional based on gene predictions and annotation of the genome. Concrete gene information such as positions, directions, sizes, molecular weights, and putative functions of each pSs-1 ORFs are shown in Supplementary Table S1. A total of 121 ORFs were determined to be functional. The functional analysis indicates that pSs-1 has similar functional system to those of T4-like phages. pSs-1 was proved to contain similar host lysis system to Shfl2 (virulent phage against S. flexneri) although orf217 showed higher similarity to SP18 (virulent phage against S. sonnei). Also, the predicted ORFs of phage structural genes were widely scattered across the entire genome although they were mostly located between orf155 and orf204 (74.46%, 35 ORFs in total).
The comparative genome analysis of pSs-1 with T4-like phages, such as Shfl2 and SP18 revealed that pSs-1 had approximately 97% nucleotide sequence identity with Shfl2 and 186 homologous ORFs among its 266 ORFs (Fig. 4). In addition, pSs-1 showed approximately 93% nucleotide sequence identity with SP18 and 54 homologous ORFs (Fig. 4). According to the ACT comparison results, the pSs-1 genome revealed a higher degree of similarity to phage Shfl2 than phage SP18; the results showed the forward matches in the order of the genome of Shfl2 and SP18 (Fig. 5).
The high bacteriolytic activity of pSs-1 against both S. sonnei and S. flexneri may be attributed to the high nucleotide identity with Shfl2 and SP18 14 . All of the ORFs in pSs-1 exhibited homology to sequences of T4-like phages, such as Shfl2, SP18, T4, T6, AR1, RB14, RB32, RB51, and vB_EcoM_ACG-C40, reported in the GenBank database. T4-like phages, one of the best-characterized groups of phages present common characteristics: (i) morphology of the Myoviridae family; (ii) similar host range, the family Enterobacteriacae; (iii) large genome size in the range between 160 kbp and 250 kbp; (iv) similar G + C content, ranging from 35% to 43% [15][16][17][18] . Based on these results, pSs-1 is considered as a member of the T4-like phage group since it belongs to the Myoviridae family, infects Shigella species, contains relatively large genome (164,999 bp), and possesses G + C content of 35  A total of 10 tRNA genes (cove score 41.18-78.18) were identified 19 , which is more than the average number of tRNAs in T4-like phage group. The exceptionally large number of tRNAs (24 tRNAs) in Aeh1 may be attributed to its significantly large genome (233,234 bp) compared to those of the other T4-like phages 17 . Likewise, the genome of KVP40 (244,834 bp) encodes a large number of tRNAs 20 . Although the exact function of tRNA in phage is still not clear, it demonstrates that it may contribute to short latent period and large burst size of pSs-1 since tRNA in phage is known to influence its reproduction in the host, and facilitate the improvement of propagation and the reduction of latent period 21 . The genome of pSs-1 did not contain lysogeny genes and all of the ORFs had nothing in common with pathogenic factors.
In summary, it is considered as a universal problem as a lot of shigellosis cases are reported in developed countries as well as developing countries, although shigellosis had been regarded as a problem only in developing countries. In developing countries especially where poor hygiene standards occur, a safe year-round supply of drinking water remains a problem because the effective water treatment facility is often beyond their financial capacity 22 . The successful protection using phage against shigellosis was reported with animal experiments and the safety of phage administration through drinking water was reported with phage safety test in humans 8,[23][24][25] . Shigella phage has potential uses to control or eradicate epidemic shigellosis in frequently affected area as phage is known to be cost-effective 26 . In developed countries where there are increasing concerns about antibiotic resistance, phages can be alternatives to conventional antibiotics. Phage research has been conducted continuously in Eastern Europe and Former Soviet Union countries, with leadership of the G. Eliava Institute of Bacteriophages, Microbiology and Virology in Tbilisi, Georgia. For many years, the Eliava Institute had developed and produced phages for the treatment and prophylaxis of bacterial infections including intestinal infections 27 . In a series of clinical trials, the therapeutic effectiveness of phages against infectious diseases was evaluated and no harmful effects were reported 27 . As the prevalence of a particular species of Shigella is generally considered to differ in various geographical areas 5 , a large-scale future screenings using Shigella environmental isolates from different geographical areas are required. This study may serve as a momentum to enhance the international collaborative research aiming control of waterborne infections, particularly shigellosis, on different continents.

Methods
Sampling, phage isolation and purification. Altogether 82 environmental water samples were collected from 5 different sampling sites, such as river and stream in South Korea between October 2011 and April 2012. Phage isolation was performed using a standard enrichment method using an indicator bacterium (S. sonnei ATCC ® 25931) 11 . All of the bacterial strains used in this study were cultured at 37 °C (Table 1). The conventional double-layer agar method was used for the examination of phage activity 28 . A single plaque was collected from the plate with a sterile Pasteur pipette and used to inoculate TSB containing 2 ml of log-phase S. sonnei ATCC ® 25931. This single plaque isolation procedure was repeated three successive times to obtain purified phages 11 .
Phage morphology, host range and efficiency of plating (EOP). In our study, every assay was performed in triplicate except electron microscopy. For the electron microscopy analysis, the phage suspension (7.8 × 10 8 PFU/ml) was concentrated and purified using continuous CsCl density gradient ultra-centrifugation 11 . The purified phages (1.0 × 10 11 PFU/ml, 6 ml of total volume) after the concentration/ purification using Polyethylene Glycol (PEG) and CsCl gradient centrifugation were negatively stained with 2% uranyl acetate. Electron micrographs were taken using a Zeiss TEM EM902 (Zeiss, Germany) at an accelerating voltage of 80 kV. The phage size was determined from at least 10 measurements 29 .
To evaluate the host range of pSs-1, its infectivity was tested on the bacterial strains used in this study. The presence of plaque formation and the number of plaques were determined after 24 h of incubation, and the EOP Hypothetical functions of encoded proteins were determined by comparison of amino acid sequences to the non-redundant databank using BLASTP. The + and − stranded ORFs were colored as grey and white, respectively. The CoreGenes between pSs-1 and Shfl2, between pSs-1 and SP18, and between pSs-1 and T4 were colored as yellow, green, and red, respectively. The predicted tRNA was indicated with thick arrow. Putative promoters and terminators were indicated by a bent arrow and inverted triangle on a vertical line, respectively.

Figure 5. Genome comparison of pSs-1 to its related phages (Shfl2 and SP18) using Artemis Comparison
Tool (ACT). Translated BLAST (tblastx, score cutoff: 40) was used to align translated genome sequences of phages. The blue and red lines represent the reverse and forward matches, respectively, and color intensity is proportional to the sequence homology. Nucleotide base-pairs were indicated between grey lines for each phage genome.
values were quantified by calculating the ratio of the PFU obtained with each phage-susceptible strain to the PFU obtained with the indicator strain.
One-step growth and host cell lysis. Burst size and latent period of pSs-1 were determined by the one-step growth analysis as previously described 30 . Samples (100 μl) were collected at 5 min intervals and the titers were determined by the double-layer agar method. To evaluate the bacteriolytic activity of pSs-1 against S. sonnei ATCC ® 25931 and S. flexneri ATCC ® 29903, the absorbance (OD 600 nm ) was examined in order to determine the change of viable bacteria. After 1 h of the early-exponential phase bacterial incubation, the culture was divided into four 10 ml samples, which were co-cultured with phage suspensions at different multiplicities of infection (MOIs): 0, 0.1, 1, and 10. The preparations were incubated at 37 °C with shaking at 250 rpm. Bacteria not inoculated with pSs-1 (MOI: 0) were used as a control.
Phage genome sequence analysis. The phage DNA was sequenced by GenoTech (Seoul, Korea) using Sanger sequencing and a Next Generation Sequencing System (NGS: Ion PGM 314 sequencer). The full length genome sequence was obtained by sequence assembly using CLC Genomics Workbench v.6.0.5. Contig gaps were filled by additional PCR and primer walking. Potential open reading frames (ORFs) that may encode gene products were predicted using GLIMMER and GeneMarkS, respectively 31,32 . The putative functions of the ORFs were analyzed by BLASTP searches at the National Center for Biotechnology Information (NCBI). Putative promoter regions were predicted using the Neural Network Promoter Prediction tool of the Berkeley Drosophila Genome Project (minimum promoter score: 0.9) 33 . Rho-independent transcription terminators were identified using Finding Terminators program (energy threshold value: −11) 34 . Transmembrane domains and signal sequence regions were predicted with the TMHMM program, ver. 2.0, and the SignalP 3.0 program, respectively 35 . The phage's genome map was drawn using DNA Master 19 . The genome of pSs-1 was subjected to pairwise analysis using the Artemis Comparison Tool (ACT) 36 with its close homologs, phage Shfl2 [GenBank accession number: NC_015457] and SP18 [GenBank accession number: NC_014595]. The protein sequence similarities of the phages were analyzed using CoreGenes3.0 37 .