Skip to main content

Comparative genomic analysis of bacteriophages specific to the channel catfish pathogen Edwardsiella ictaluri



The bacterial pathogen Edwardsiella ictaluri is a primary cause of mortality in channel catfish raised commercially in aquaculture farms. Additional treatment and diagnostic regimes are needed for this enteric pathogen, motivating the discovery and characterization of bacteriophages specific to E. ictaluri.


The genomes of three Edwardsiella ictaluri-specific bacteriophages isolated from geographically distant aquaculture ponds, at different times, were sequenced and analyzed. The genomes for phages eiAU, eiDWF, and eiMSLS are 42.80 kbp, 42.12 kbp, and 42.69 kbp, respectively, and are greater than 95% identical to each other at the nucleotide level. Nucleotide differences were mostly observed in non-coding regions and in structural proteins, with significant variability in the sequences of putative tail fiber proteins. The genome organization of these phages exhibit a pattern shared by other Siphoviridae.


These E. ictaluri-specific phage genomes reveal considerable conservation of genomic architecture and sequence identity, even with considerable temporal and spatial divergence in their isolation. Their genomic homogeneity is similarly observed among E. ictaluri bacterial isolates. The genomic analysis of these phages supports the conclusion that these are virulent phages, lacking the capacity for lysogeny or expression of virulence genes. This study contributes to our knowledge of phage genomic diversity and facilitates studies on the diagnostic and therapeutic applications of these phages.


Here we report the complete nucleotide sequence and annotation of the genomes of three bacteriophages specific to the gram negative bacterial pathogen Edwardsiella ictaluri, the causative agent of enteric septicemia of catfish (ESC). ESC is a primary cause of mortality in catfish farms with annual direct losses in the range of $40-60 million dollars in the U.S. [1]. Economic losses coupled with limited available treatment options for controlling ESC, and concerns regarding the development of resistance to antibiotics used in aquaculture warranted efforts to identify biological control agents that are antagonistic to E. ictaluri (e.g., bacteriophage and bacteria). In addition, the multiple days necessary to obtain a diagnostic result for E. ictaluri via biochemical tests was a motivation to identify phage that could serve as specific, rapid, and inexpensive typing agents for ESC disease isolates.

The idea of using phage as antimicrobial agents to treat bacterial infections in agriculture or aquaculture is not a new proposition [2]; however, there is now a better understanding of phage biology and genetics, and with it a better understanding of their potential and their limitations as biological control agents [3]. The most serious obstacles to successful use of phage therapy include the development of phage resistance by host bacteria, the capacity of some temperate phages to transduce virulence factors (i.e., lysogenic conversion), the possible degradation or elimination of phages by gastrointestinal pH or proteolytic activity within a fish, and the possible immune system clearance of administered phage. Potentially viable solutions are available to counter each of these concerns, including the use of multiple phages at concentrations selected to reduce the development of phage-resistant bacterial populations [4], identifying phage variants adapted to minimize GI tract and/or immune clearance [5], and by selecting bacteriophages as therapeutic agents that are well characterized at a genomic level, with no potential for inducing lysogenic conversion [2, 3, 6].

Two unique E. ictaluri-specific phages ϕeiAU (eiAU) and ϕeiDWF (eiDWF) were isolated from aquaculture ponds with a history of ESC [7]. Phage eiAU was isolated in 1985 at Auburn University and phage eiDWF was recently isolated in 2006 in western Alabama. An additional E. ictaluri-specific bacteriophage ϕeiMSLS (eiMSLS) was isolated directly from culture water from a commercial catfish aquaculture pond in Washington County, MS in 2004 (Timothy Welch, USDA National Center for Cool and Cold Water Aquaculture, WV personal communication). The isolation of each of these bacteriophages was accomplished by concentrating viruses from pond water samples by ultrafiltration and enriching for E. ictaluri-specific bacteriophages via enrichment in log-phase bacterial broth cultures. These three bacteriophages were classified initially within the family Siphoviridae due to their long, non-contractile tails, but their phylogenetic affiliation could not be assessed in the absence of phage genome sequence analysis [810]. To date no other bacteriophage morphotypes have been observed to infect E. ictaluri from pond water enrichment experiments. A genomic analysis of these three phages was initiated to examine the potential of these three bacteriophages for lysogeny, to ensure they did not harbor virulence or toxin genes and to better understand the genetic basis of their host specificity [7]. This study represents the first genomic analysis of bacteriophages specific to Edwardsiella ictaluri, and will expand scientific understanding of phage biology, and genomic information [11].

Results and Discussion

Genome characteristics

Total sequence coverage for the eiMSLS assembly was 9.8X, while coverage for the eiAU and eiDWF assemblies exceeded 30X. The genomes of phages eiAU, eiDWF, and eiMSLS are 42.80 kbp, 42.12 kbp, and 42.69 kbp, respectively. The % GC content is 55.37%, 55.54%, and 55.77% for phage eiAU, eiDWF, and eiMSLS, respectively, and is similar to the 57% GC content of host E. ictaluri genome reference strain (GenBank accession NC 012779). No tRNA genes were detected in the genome of any of the three phages. This is unlike several members of the Siphoviridae family that carry tRNA genes [12].

Open Reading Frame (ORF) analysis

A total of 54 ORFs were predicted for phage eiAU (Table 1), while 52 ORFs were predicted for eiDWF and 52 ORFs for eiMSLS. Based on sequence similarity (E value < 0.001), 40 out of 54 (74%), 37 out of 52 (71%) and 36 out of 52 (69%) of the ORFs for phages eiAU, eiDWF, and eiMSLS, respectively, share significant sequence similarity to known protein sequences contained in the GenBank nr/nt database (Table 1). Of the ORFs with significant sequence similarity to sequences in GenBank, putative functions could only be assigned to 21 out of 40 (53%), 21 out of 37 (57%) and 20 out of 36 (56%) for phages eiAU, eiDWF, and eiMSLS, respectively. Positions, sizes, sequence homologies and putative functions for each predicted ORF are presented in Table 1.

Table 1 Predicted ORFs for eiAU, eiDWF, and eiMSLS, and the most similar BLAST hits for each of the phage ORFs

The genome of phage eiAU contains several overlapping predicted ORFs, which can be an indication of translational coupling or programmed translational frameshifts [13]. Twelve possible sequence frameshifts were predicted in the eiAU genome sequence. Interestingly, one of these frameshifts is conserved in tail assembly genes of dsDNA phages [14]. In dsDNA phage genomes the order of the tail genes is highly conserved, most notably the major tail protein is always encoded upstream of the gene encoding the tape measure protein [14]. Between these two genes, two overlapping ORFs are commonly found that have a translational frameshift [15]. A similar organization of tail genes is observed in phage eiAU, in which two ORFs (22 and 23) lie between the putative phage tape tail measure protein gene (ORF21) and the major tail protein (ORF24) (Table 1). Similarly, phage eiAU contains a frameshift in the two overlapping ORFs between the phage tail measure and the major tail protein. In other phages both of these proteins are required for tail assembly even though they are not part of the mature tail structure [14].

Overall Genome Organization and Comparison

A schematic representation of one of these phages (eiAU) shows that ORFs in these three phages are organized into two groups; early genes (DNA replication) that are encoded on one strand and the late genes (head, tail, and lysis) that are encoded on the complementary strand (Figure 1). Whole genome comparisons revealed that phages eiAU, eiDWF, and eiMSLS have conserved synteny (Figure 1 and Figure 2). The overall genetic organization of the eiAU, eiDWF, and eiMSLS genomes, typically consisting of "DNA packaging-head-tail-tail fiber-lysis/lysogeny-DNA replication-transcriptional regulation" modules is shared by many phage within the Siphoviridae family [16].

Figure 1

Schematic representation of the genome sequence of bacteriophage eiAU showing its overall genomic organization. The ORFs are numbered consecutively (see Table 1) and are represented by arrows based on the direction of transcription. The numbers +1, +2, +3 represent corresponding reading frames.

Figure 2

Circular representation depicting the genomic organization of eiAU (two outermost circles, dark blue, showing each predicted ORF and its direction of transcription) and a tBLASTx comparison with the genomes of eiDWF (third circle from outside, green), eiMSLS (fourth circle from outside, light blue), and Enterobacteria phage SSL-2009a (fifth circle from outside, orange). The degree of sequence similarity to eiAU is proportional to the height of the bars in each frame. The %G+C content of eiAU is also depicted (sixth circle from outside, black). This map was created using the CGView server (Grant and Stothard, 2008).

Multiple sequence alignment analysis revealed that the eiAU, eiDWF, and eiMSLS genomes are >95% identical at the nucleotide level (Figure 2). Similarly, a high degree of sequence similarity has been observed in the genomes of recently sequence bacteriophages that infect Campylobacter[17], Eschericia coli[18], and also many Mycobacterium spp. [19]. The high similarity of some phage genomes that infect a single host species suggests that certain phage lineages may be stable over time and over distant geographic areas [17]. This observation may likely be clarified once additional genome sequences of phages infecting a common host such as E. ictaluri become available.

Comparison of head morphogenesis and structural proteins

Genome sequencing of tailed phages and prophages has revealed a common genetic organization of the genes encoding head morphogenesis and head structural proteins. These gene systems are typically organized as follows: 'terminase - portal - protease - scaffold - major head shell (coat) protein - head/tail-joining proteins - tail shaft protein - tape measure protein - tail tip/base plate proteins - tail fiber' (listed in the order of transcription) [20]. Phages eiAU, eiDWF, and eiMSLS follow a similar organization of genes encoding head morphogenesis and structural proteins, although the direction is reversed in relation to their order of transcription (Figure 1 and Table 1).

The module containing head morphogenesis and tail structure proteins in phage eiAU is the largest module, and is predicted to contain 22 ORFs (ORF14-ORF35). The consecutive ORFs 14 to 32 have significant sequence similarity with phage head morphogenesis and structural proteins, with putative function in tail assembly (ORFs 14, 17, and 18), tail fiber protein (ORF 15), phage host specificity (ORF 16), minor tail proteins (ORFs 19-21), major tail proteins (ORFs 24 and 25), major capsid proteins (ORF 29), structural proteins (ORFs 27, 30 and 33), and a phage head morphogenesis protein (ORF32) (Table 1). ORFs 28, 26, 23, and 22 could not be linked to a putative function based on BLAST search or any other similarity searches. However, all of these ORFs with the exception of ORF28 have sequence similarity to proteins identified within other phage genomes (Table 1). The protein products of ORF34 and ORF35 may encode large and small terminase subunits, respectively. ORF34 is predicted to encode the terminase large subunit. The top BLAST hit for ORF35 is the protein Gp1 encoded by Sodalis phage SO-1; however, it is possible that ORF 35 encodes a small terminase subunit as there is limited sequence similarity to a putative terminase small subunit from Listonella phage phiHSIC. This indicates that these E. ictaluri phages, similarly to most dsDNA viruses, use a DNA packaging motor consisting of two nonstructural proteins (the large and small terminase subunits) encoded by adjacent genes [21]. Most known terminase enzymes have a small subunit that specifically binds the viral DNA and the large subunit with endonuclease activity for DNA cleavage and an ATPase activity that powers DNA packaging [22, 23].

No hit for a portal protein or for a protease was obtained either by BLAST or by HmmPfam searches. ORF33 is the most likely candidate for a portal protein based on the observation that the portal protein is generally located immediately downstream of the terminase gene [13].

Lytic Cassette

The lytic cassette of phage eiAU is predicted to be encoded by ORFs 36-39. ORF36 encodes a predicted endolysin, and a putative holin protein is encoded by ORF39. All dsDNA phages studied to date use two enzymes to lyse their host, an endolysin which degrades cell wall peptidoglycan and a holin which permeabilizes the cell membrane [21]. These two proteins work in conjunction to destroy the cell wall of bacteria and subsequently lyse the cell [24]. These components of a host lysis cassette are each present in the genome of phages eiAU, eiDWF, and eiMSLS including a putative Rz lysis accessory protein encoded by ORF38 (Table 1.). The RZ protein is predicted to be a type II integral membrane protein and its function, although not fully understood, may be required for host cell lysis only in a medium containing an excess of divalent cations [25]. Phage endolysins have been linked to five enzymatic activities, including an N-acetyl muramidase or "true lysosyme", the lytic transglycosylases, the N-acetylmuramoyl-L-alanine amidases, the endo- β-N-acetylglucosaminidases, and the endopeptidases [26]. Secondary structure analysis predicts that the endolysin of eiAU is a member of the N-acetylmuramoyl-L-alanine amidases class of endolysins.

DNA replication proteins

ORFs with significant sequence similarity to proteins involved in DNA replication were identified in all three E. ictaluri-specific phage genomes. ORF44 is predicted to encode a phage replicative helicase/primease. Several phages use separate primase and helicase proteins while others use a multifunctional protein (primase/helicase) possessing both activities [13]. The helicase/primase protein works in DNA replication by unwinding double stranded DNA into single stranded DNA [27]. No predicted function could be assigned to ORFs45 and 46. Also, no predicted function could be assigned to ORF47; however, a search for secondary structures within the predicted ORF47 amino acid sequence detected a helix-hairpin-helix DNA binding motif. Additionally, no putative function could be assigned to ORF48, ORF49, or ORF50. ORF51 had as one of its top BLAST hits an isoprenylcysteine carboxyl methyltransferase known to function in methylating isoprenylated amino acids [28]. ORF52 is predicted to encode a protein similar to gp41 of Sodalis phage SO-1, but no putative function could be assigned. ORF53 is predicted to encode DNA polymerase I. Secondary structure analysis suggested that the DNA polymerase encoded by ORF53 contains a domain that is responsible for the 3'-5' exonuclease proof-reading activity of E. coli DNA polymerase I and other enzymes, and catalyses the hydrolysis of unpaired or mismatched nucleotides. The protein encoded by ORF54 is predicted to have a VUR-NUC domain, which are associated with members of the PD-(D/E) XK nuclease superfamily such as type III restriction modification enzymes. ORF2 is predicted to encode a DNA repair ATPase. A search for secondary structures within the ORF2 predicted amino acid sequence revealed a HNH endonuclease. No putative function could be assigned to ORF3. ORF4 is predicted to encode a helicase protein belonging to the SNF2 family, commonly found in proteins involved in a variety of processes including transcription regulation, DNA repair, DNA recombination, and chromatin unwinding [29]. ORF6 is predicted to encode a phage methyltransferase. Secondary structure analysis revealed that the methyltransferase predicted to be encoded by ORF6 is a C-5 cytosine-specific DNA methylase which in bacteria is a component of restriction-modification systems. Also, Mg+ and ATP binding sites were detected in the predicted protein product of ORF6. ORF7 is predicted to encode a DNA N-6-adenine-methyltransferase within a family of methyltransferase found in bacteria and phage that has site specific DNA methyltransferase activity [30].

No ORF encoding an RNA polymerase was detected in any of the phages suggesting that these phages rely on the host RNA polymerase to transcribe their genes. This is further corroborated by the observation that no phage-encoded transcription factor was detected in the genome of these phages.

Comparison of ORFs among phages eiAU, eiDWF, and eiMSLS

The three phage genomes revealed extensive homology and limited variability in their gene sequence (Figure 2). The percent identity and percent similarity of each ORF within the three phage genomes (data not shown) revealed that differences exist mainly in predicted ORFs that have no significant sequence similarity to sequences in GenBank database and also to ORFs encoding structural proteins (primarily the tail fiber genes). ORF14 (117 AA) is predicted to encode a phage tail fiber assembly protein/tail assembly chaperone, and in eiAU and eiDWF it is 100% identical, yet it is not present in eiMSLS. ORF15 (335 AA) is predicted to encode a tail fiber protein and is present in all three phages, with 100% identity in eiAU and eiDWF, however, it only has 58% identity to its counterpart in eiMSLS. ORF21 (900 AA) is predicted to encode a phage tail tape measure protein and is present in all three phages at approximately 95% identity at the amino acid level. ORF23 (118 AA) is predicted to encode a protein homologous to gp15 [Sodalis phage SO-1] which is a structural protein that plays a role in cell membrane penetration. This ORF is present in all three phages with 83% identity at the amino acid level. ORF24 (200 AA) is predicted to encode a major tail protein and is present in all three phages, with 100% identity between eiDWF and eiMSLS, and with only 90% identity between those two phage and the ORF counterpart in eiAU. Sequence differences in these structural proteins may help explain the differences observed in the efficiency of these phages to form plaques on various E. ictaluri strains [7]. Most of the structural proteins described above are expected to be involved in phage infectivity such as adsorption of the phage to the bacterial cell (ORFs 14 and 15), phage tail length (ORF21), and cell membrane penetration (ORF23).

Differences were also observed in the ORFs encoding the putative methyltransferases. In phage eiAU, ORF6 and ORF7 are predicted to encode a phage methyltransferase and a DNA N-6-adenine-methyltransferase respectively, while in phage eiDWF and eiMSLS only one larger ORF encoding a phage methyltransferase was predicted. Similarly, two methyltransferases are present in the genomes of one of two highly similar Campylobacter phages [17]. The authors suggest that the two methyltransferases may enable the phage to avoid DNA restriction in some strains through DNA methylation. This may help explain the differences observed in host range for the Campylobacter phages [17] as well as differences observed in host specificity of the E. ictaluri phages [7]. Hence, these methyltransferases may likely be involved in DNA methylation as a means of avoiding the restriction endonuclease (s) of E. ictaluri.

Classification of phages eiAU, eiDWF, and eiMSLS

The majority of the top BLAST hits for these phage genomes are to proteins belonging to lytic phages, including Yersinia phage PY100, Salmonella phage c341, and Enterobacteria phage HK97 (Table 1.). All of the components of a phage lysis cassette (endolysin, holin, and a lysis accessory protein) were detected in these phages and no sequence similarity to lysogenic phages or to any component that is associated with lysogeny such as integrase/recombination associated enzymes, repressor proteins, and anti-repressor proteins [31] were detected. These data along with results documenting the lytic capabilities of these phages [7] all indicate that these phages lack mechanisms for integration into the DNA of their host and that they are virulent phages without the capacity for lysogeny. Additionally, none of the predicted proteins have similarities to known bacterial pathogenicity factors. These observations indicate that these phages lack any lysogenic or bacterial virulence-inducing capacity that would preclude their potential use as therapeutic agents.

Taxonomic classification of these E. ictaluri-specific phages must rely upon a synthesis of morphological and genomic information, considering that phage evolution has been profoundly directed by lateral gene transfer [32], and that a rational hierarchical system of phage classification should be based on the degree of DNA and protein sequence identity for multiple genetic loci [33]. Gene modules that have been proposed for using as basis of a phage taxonomy system include the DNA packaging-head gene cluster, the structural gene architecture, and phage tail genes (excluding the tail fiber genes) [16].

A comparison of phage eiAU to Enterobacteria phage SSL-2009a was conducted due to the large number of significant BLAST hits between ORFs in the E. ictaluri phage genomes and those respective ORFs within the genome of phage SSL-2009a, which are on average 34.1% identical at the nucleotide level. A comparative genomic analysis between the genome of phage eiAU and that of phage SSL-2009a revealed that genome regions encoding many putative structural and replication proteins are shared by both phages (Figure 2). The predicted gene products with sequence similarity between the eiAU and SSL-2009a phage genomes include the putative minor tail proteins/tail tape measure, major tail proteins, major capsid proteins, head morphogenesis, phage terminase small subunit, and the phage terminase large subunit. Interestingly, other structural proteins including the host specificity proteins, the tail assembly proteins, and particularly the tail fiber/baseplate protein which has been recommended for exclusion in any sequence based phage taxonomy scheme [33] are not shared between the two genomes.

Phylogeny based on multiple genetic loci

The genetic conservation observed in the structural proteins between phage eiAU and Enterobacteria phage SSL-2009a led us to further investigate the relatedness of these E. ictaluri phages and other enterobacteria phage, based on specific phage genetic loci. The amino acid sequences of one of the conserved structural proteins (large terminase subunit) as well as one of the non structural proteins (DNA polymerase I) were chosen for phylogenetic analysis. The large terminase subunit which is a structural protein is along with the portal protein considered the most universally conserved gene sequence in phages [20], hence they are good options to aid in phage classification. Phylogenetic analysis based on the large terminase subunit amino acid sequence (Figure 3) and the DNA polymerase I amino acid sequence (Figure 4) of eiAU reveal that phages eiAU, eiDWS, and eiMSLS were most similar to phage that infect other enterobacteria (Enterobacteria phage SSL-2009a) and Sodalis glossinidius (Sodalis phage SO-1). These two phages are dsDNA viruses belonging to the Caudovirales order, one being a Siphoviridae (Sodalis phage SO-1) (NCBI accession # NC_013600) and the other an unclassified member of the Caudovirales (Enterobacteria phage SSL-2009a) (NCBI accession # NC_012223). The overall genomic organization of the three new phages is shared by many members of the Siphoviridae family of phages sequenced to date [16], and is supported by the previously described morphology of these phages [7].

Figure 3

Rooted maximum parsimony tree based on the aligned amino acid sequences of the large terminase subunit gene of phage eiAU and 25 other large terminase genes from diverse phage genomes. The numbers at the nodes represent bootstrap values based on 1,000 resamplings.

Figure 4

Rooted maximum parsimony tree based on the aligned amino acid sequences of the DNA polymerase subunit gene of phage eiAU and 33 other DNA Polymerases from diverse phage genomes. The numbers at the nodes represent bootstrap values based on 1,000 resamplings.


This is the first genomic analysis of bacteriophages that infect the bacterial pathogen E. ictaluri. Phylogenetic analysis of multiple phage gene products suggests that these phages are similar to those that infect other Enterobacteria hosts. The bioinformatic analysis of the genomes of these three E. ictaluri-specific bacteriophages corroborate previously published data that indicates that these bacteriophages are lytic, and lack any mechanism for lysogenic conversion of their host. Additionally, none of the predicted proteins have similarities to known bacterial pathogenicity factors or to toxin genes. Even though these three bacteriophages were isolated in different geographic locations within the natural range of catfish over twenty years apart, they are remarkably similar to each other at a genomic level. This genomic analysis suggests that these phages are members of a lineage that is highly stable over time and geographic regions. The information obtained from the analyses of these bacteriophage genomes will facilitate their diagnostic and therapeutic applications.


Bacteriophages and bacterial strains

Phages ϕeiAU and ϕeiDWF used in the study were originally isolated and characterized at Auburn University [7]. Phage ϕMSLS was isolated from an aquaculture pond water sample on a lawn of E. ictaluri strain I49 (Thad Cocharan National Warmwater Aquaculture Center, Aquatic Diagnostic Lab), and clear plaques were doubly purified on an E. ictaluri host. Host bacterial isolate E. ictaluri strain 219 was obtained from the Southeastern Cooperative Fish Disease Laboratory at Auburn University. E. ictaluri strains were grown on Brain Heart Infusion (BHI) medium and cryopreserved in BHI containing 10% glycerol at -80°C. In each experiment bacterial strains were grown from the original glycerol stock to maintain low passage number, virulent E. ictaluri cultures.

Isolation of phage DNA

Phages eiAU, eiDWF, and eiMSLS were propagated on E. ictaluri strain 219 using a standard soft agar overlay method [34]. Phages were harvested by flooding plates with 5 mL SM buffer (100 mM NaCl, 8 mM MgSO4·7H2O, 50 mM Tris-Cl (1 M, pH 7.5), and 0.002% (w/v) of 2% Gelatin), incubating at 30°C while shaking for 6 h, and then collecting the buffer-phage solution. Collected phage suspensions were treated for 10 min with 1% (v/v) chloroform to lyse bacterial cells, subjected to centrifugation at 3,600× g for 25 min, and then filtered through a 0.22 μm filter to remove cell debris. Phage solutions were purified over a cesium chloride gradient and concentrated by precipitation with polyethylene glycol 8000. Concentrated phage particles were resuspended in 200 μl SM buffer. Free nucleic acids from lysed bacterial host cells were degraded with 250 units of benzonase endonuclease for 2 h at 37°C, after which the benzonase was inhibited by the addition of 10 mM EDTA. The phage protein coats were degraded using proteinase K (1 mg/ml) and SDS (1%). A phenol-chloroform extraction was performed, and DNA was precipitated with ethanol. The washed DNA pellet was resuspended in T10 E1 buffer (10 mM Tris-HCl (pH 8.0), 1 mM EDTA) and stored at -20°C.

Shotgun library construction and sequencing

Shotgun subclone libraries were constructed at Lucigen Corporation (Middleton, WI) as previously described [35]. Briefly, phage genomic DNA was randomly sheared using a Hydroshear instrument (Digilab Genomic Solutions, Ann Arbor, MI) and DNA fragments from 1 to 3 kb in size were extracted from an agarose gel. Phage DNA fragments were blunt-end repaired, ligated to asymmetric adapters, amplified using a proof reading polymerase and ligated into the pSMART® GC cloning vector following manufacturer recommendations. The ligation was transfected into electrocompetent E. coli cells. E. coli transformants were robotically picked into Luria-Bertani (LB) broth containing 30 ug per ml kanamycin and 10% (w/v) glycerol in a 96-well format using a QPix2 colony picking system (Genitex Limited, Hampshire, UK). Colony PCR was performed on a representative number of clones (n = 10) to assess insert size and the percentage of subclones containing an insert. Plasmid DNA was isolated using standard alkaline-SDS lysis and ethanol precipitation. Alternately, the insert was amplified from the E. coli clone glycerol stock using a pSMART vector-specific primer set, with 30 cycles of amplification (95°C denaturation, 50°C annealing, and 72°C extension). The resultant PCR products were treated with exonuclease I and Shrimp Alkaline Phosphatase to remove oligonucleotides. Sanger sequencing from both ends of the insert was obtained using ABI PRISM BigDye™ 3.1 Terminators chemistry (Applied Biosystems, Foster City, CA), and sequencing products were resolved on an ABI 3130XL capillary electrophoresis instrument.

Contig assembly and primer walking

Raw sequence data from eiMSLS was re-assembled using LaserGene software (DNASTAR Inc., Madison, WI). The eiMSLS sequence was used as a reference for alignment of eiAU and eiDWF sequences. For the latter two genomes, raw sequence data was trimmed for quality and vector sequence was removed using Sequencher™ software (Gene Codes Corporation, Ann Arbor, MI). Contigs were re-assembled using CromasPro v.1.42 (Technelysium Pty, Tewantin, Australia) using 70% sequence match, and a minimum of 30 bp overlap. Contigs were manually edited to remove nucleotide gaps and mis-called bases. Closure of each respective phage genome was completed by primer walking using either the isolate phage DNA or amplified products as the sequencing template. Each phage was determined to have a circular genome by PCR amplification using primers directed out from the ends of the single large contig comprising the respective phage genome.

Genome sequence analysis

Open reading frames were identified using a GeneMark heuristic approach for gene prediction in prokaryotes, which is specifically designed for small virus, plasmid, or phage genomes less than 50 kb in size [36]. Additionally, GLIMMER 3.02, and NCBI's ORF Finder [37] were utilized to corroborate the predicted ORFs obtained from GenMark analysis. The % GC content of phages was calculated using geecee [38]. The tRNAscan-SE v.1.21 program was used to search for tRNA genes [39]; [40]. Gene function was predicted by comparing each phage ORF sequence against the GenBank nr/nt sequence database using the BLASTp and BLASTn [41] search algorithms. Iterative PSI-BLAST analysis was used to increase sensitivity of detecting homologous genes for ORFs resulting in hits with low E-values. Searches for secondary structures (profiles, patterns, blocks, motifs, and protein families) were performed using a web server [42]. Frameshifts were detected using FrameD [43]. The amino acid identity (AAI) of predicted protein sequences was determined by pairwise BLASTp analysis of each set of phage homologs. Dotplots were generated using the DOTMATCHER tool from EMBOSS (Ian Longden, Sanger Institute, Cambridge, UK). Pairwise global alignment and graphical representation of phage genomes was performed using the CGView server using tBLASTx with an E-value cutoff of 0.001 [44]. Genome sequences were annotated using the Artemis software package (The Institute for Genomic Research), and all sequences were deposited in the GenBank database (accession # HQ824548-HQ824705) using Sequin.

Phylogenetic analysis

The predicted amino acid sequences for phage terminase large subunit and DNA polymerase were used to conduct a phylogenetic analysis of these E. ictaluri bacteriophages. The amino acid sequence for each predicted protein was aligned with a collection of homologous sequences using the program ClustalW2 [45]. ClustalW2 multiple alignments were exported to Mega4 [46] and a maximum parsimony analysis was used to construct a phylogenetic tree, with bootstrap support (n = 1000 replicates).


  1. 1.

    Shoemaker CA, Klesius PH, Evans JJ, Arias CR: Use of modified live vaccines in aquaculture. J World Aquacult Soc 2009, 40: 573-585. 10.1111/j.1749-7345.2009.00279.x

    Article  Google Scholar 

  2. 2.

    Brussow H: Phage therapy: the Escherichia coli experience. Microbiology 2005, 151: 2133. 10.1099/mic.0.27849-0

    Article  PubMed  Google Scholar 

  3. 3.

    Skurnik M, Strauch E: Phage therapy: facts and fiction. International Journal of Medical Microbiology 2006, 296: 5-14. 10.1016/j.ijmm.2005.09.002

    Article  CAS  PubMed  Google Scholar 

  4. 4.

    Cairns BJ, Payne RJH: Bacteriophage therapy and the mutant selection window. Antimicrobial agents and chemotherapy 2008, 52: 4344. 10.1128/AAC.00574-08

    PubMed Central  Article  CAS  PubMed  Google Scholar 

  5. 5.

    Merril CR, Biswas B, Carlton R, Jensen NC, Creed GJ, Zullo S, Adhya S: Long-Circulating Bacteriophage as Antibacterial Agents. Volume 93. JSTOR; 1996:3188-3192.

    Google Scholar 

  6. 6.

    Carlton RM: Phage therapy: past history and future prospects. Arch Immunol Ther Exp 1999, 47: 267-274.

    CAS  Google Scholar 

  7. 7.

    Walakira JK, Carrias AA, Hossain MJ, Jones E, Terhune JS, Liles MR: Identification and characterization of bacteriophages specific to the catfish pathogen, Edwardsiella ictaluri. J Appl Microbiol 2008, 105: 2133-2142. 10.1111/j.1365-2672.2008.03933.x

    Article  CAS  PubMed  Google Scholar 

  8. 8.

    Rohwer F, Edwards R: The Phage Proteomic Tree: a genome-based taxonomy for phage. Journal of bacteriology 2002, 184: 4529. 10.1128/JB.184.16.4529-4535.2002

    PubMed Central  Article  CAS  PubMed  Google Scholar 

  9. 9.

    Chibani-Chennoufi S, Canchaya C, Bruttin A, Brussow H: Comparative genomics of the T4-like Escherichia coli phage JS98: implications for the evolution of T4 phages. Journal of bacteriology 2004, 186: 8276. 10.1128/JB.186.24.8276-8286.2004

    PubMed Central  Article  CAS  PubMed  Google Scholar 

  10. 10.

    Lima-Mendez G, Van Helden J, Toussaint A, Leplae R: Reticulate representation of evolutionary and functional relationships between phage genomes. Molecular biology and evolution 2008, 25: 762. 10.1093/molbev/msn023

    Article  CAS  PubMed  Google Scholar 

  11. 11.

    Serwer P, Hayes SJ, Thomas JA, Hardies SC: Propagating the missing bacteriophages: a large bacteriophage in a new class. Virology Journal 2007, 4: 21. 10.1186/1743-422X-4-21

    PubMed Central  Article  PubMed  Google Scholar 

  12. 12.

    Kropinski AM, Sibbald MJ: Transfer RNA genes and their significance to codon usage in the Pseudomonas aeruginosa lamboid bacteriophage D3. Canadian journal of microbiology 1999, 45: 791-796. 10.1139/cjm-45-9-791

    Article  CAS  PubMed  Google Scholar 

  13. 13.

    Bukovska G, Klucar L, Vlcek C, Adamovic J, Turna J, Timko J: Complete nucleotide sequence and genome analysis of bacteriophage BFK20--a lytic phage of the industrial producer Brevibacterium flavum. Virology 2006, 348: 57-71. 10.1016/j.virol.2005.12.010

    Article  CAS  PubMed  Google Scholar 

  14. 14.

    Xu J, Hendrix RW, Duda RL: Conserved translational frameshift in dsDNA bacteriophage tail assembly genes. Molecular cell 2004, 16: 11-21. 10.1016/j.molcel.2004.09.006

    Article  CAS  PubMed  Google Scholar 

  15. 15.

    Levin ME, Hendrix RW, Casjens SR: A programmed translational frameshift is required for the synthesis of a bacteriophage l tail assembly protein. J Mol Biol 1993, 234: 124-139. 10.1006/jmbi.1993.1568

    Article  CAS  PubMed  Google Scholar 

  16. 16.

    Brüssow H, Desiere F: Comparative phage genomics and the evolution of Siphoviridae: insights from dairy phages. Molecular Microbiology 2001, 39: 213-223.

    Article  PubMed  Google Scholar 

  17. 17.

    Timms AR, Cambray-Young J, Scott AE, Petty NK, Connerton PL, Clarke L, Seeger K, Quail M, Cummings N, Maskell DJ: Evidence for a lineage of virulent bacteriophages that target Campylobacter. BMC genomics 11: 214. 10.1186/1471-2164-11-214

  18. 18.

    Denou E, Bruttin A, Barretto C, Ngom-Bru C, Brüssow H, Zuber S: T4 phages against Escherichia coli diarrhea: potential and problems. Virology 2009, 388: 21-30. 10.1016/j.virol.2009.03.009

    Article  CAS  PubMed  Google Scholar 

  19. 19.

    Hatfull GF, Jacobs-Sera D, Lawrence JG, Pope WH, Russell DA, Ko CC, Weber RJ, Patel MC, Germane KL, Edgar RH: Comparative genomic analysis of sixty mycobacteriophage genomes: Genome clustering, gene acquisition and gene size. Journal of molecular biology 2010, 397: 119-143. 10.1016/j.jmb.2010.01.011

    PubMed Central  Article  CAS  PubMed  Google Scholar 

  20. 20.

    Casjens S: Prophages and bacterial genomics: what have we learned so far? Molecular microbiology 2003, 49: 277-300. 10.1046/j.1365-2958.2003.03580.x

    Article  CAS  PubMed  Google Scholar 

  21. 21.

    Lehman SM, Kropinski AM, Castle AJ, Svircev AM: Complete Genome of the Broad-Host-Range Erwinia amylovora Phage {Phi} Ea21-4 and Its Relationship to Salmonella Phage Felix O1. Applied and environmental microbiology 2009, 75: 2139. 10.1128/AEM.02352-08

    PubMed Central  Article  CAS  PubMed  Google Scholar 

  22. 22.

    Duffy C, Feiss M: The large subunit of bacteriophage [lambda]'s terminase plays a role in DNA translocation and packaging termination1. Journal of molecular biology 2002, 316: 547-561. 10.1006/jmbi.2001.5368

    Article  CAS  PubMed  Google Scholar 

  23. 23.

    Maluf NK, Yang Q, Catalano CE: Self-association Properties of the Bacteriophage [lambda] Terminase Holoenzyme: Implications for the DNA Packaging Motor. Journal of molecular biology 2005, 347: 523-542. 10.1016/j.jmb.2005.01.016

    Article  CAS  PubMed  Google Scholar 

  24. 24.

    Wang IN, Smith DL, Young R: H OLINS: The Protein Clocks of Bacteriophage Infections. Annu Rev Microbiol 2000, 54: 799-825. 10.1146/annurev.micro.54.1.799

    Article  CAS  PubMed  Google Scholar 

  25. 25.

    Summer EJ, Berry J, Tran TAT, Niu L, Struck DK, Young R: Rz/Rz1 lysis gene equivalents in phages of Gram-negative hosts. Journal of molecular biology 2007, 373: 1098-1112. 10.1016/j.jmb.2007.08.045

    Article  CAS  PubMed  Google Scholar 

  26. 26.

    Borysowski J, Weber-Dabrowska B, Gorski A: Bacteriophage endolysins as a novel class of antibacterial agents. Experimental Biology and Medicine 2006, 231: 366.

    CAS  PubMed  Google Scholar 

  27. 27.

    Egelman EH, Yu X, Wild R, Hingorani MM, Patel SS: Bacteriophage T7 helicase/primase proteins form rings around single-stranded DNA that suggest a general structure for hexameric helicases. Proc Natl Acad Sci USA 1995, 92: 3869. 10.1073/pnas.92.9.3869

    PubMed Central  Article  CAS  PubMed  Google Scholar 

  28. 28.

    Clarke S: Protein methylation. Current opinion in cell biology 1993, 5: 977-983. 10.1016/0955-0674(93)90080-A

    Article  CAS  PubMed  Google Scholar 

  29. 29.

    Eisen JA, Sweder KS, Hanawalt PC: Evolution of the SNF2 family of proteins: subfamilies with distinct sequences and functions. Nucleic acids research 1995, 23: 2715. 10.1093/nar/23.14.2715

    PubMed Central  Article  CAS  PubMed  Google Scholar 

  30. 30.

    Schneider-Scherzer E, Auer B, de Groot EJ, Schweiger M: Primary structure of a DNA (N6-adenine)-methyltransferase from Escherichia coli virus T1. DNA sequence, genomic organization, and comparative analysis. Journal of Biological Chemistry 1990, 265: 6086.

    CAS  PubMed  Google Scholar 

  31. 31.

    Oakey HJ, Cullen BR, Owens L: The complete nucleotide sequence of the Vibrio harveyi bacteriophage VHML. Journal of applied microbiology 2002, 93: 1089-1098. 10.1046/j.1365-2672.2002.01776.x

    Article  CAS  PubMed  Google Scholar 

  32. 32.

    Nelson D: Phage taxonomy: we agree to disagree. Journal of bacteriology 2004, 186: 7029. 10.1128/JB.186.21.7029-7031.2004

    PubMed Central  Article  CAS  PubMed  Google Scholar 

  33. 33.

    Proux C, Van Sinderen D, Suarez J, Garcia P, Ladero V, Fitzgerald GF, Desiere F, Brussow H: The dilemma of phage taxonomy illustrated by comparative genomics of Sfi21-like Siphoviridae in lactic acid bacteria. Journal of bacteriology 2002, 184: 6026. 10.1128/JB.184.21.6026-6036.2002

    PubMed Central  Article  CAS  PubMed  Google Scholar 

  34. 34.

    Adams MH: Bacteriophages. New York: Interscience Pub- lishers, Inc; 1959.

    Google Scholar 

  35. 35.

    Schoenfeld T, Patterson M, Richardson PM, Wommack KE, Young M, Mead D: Assembly of viral metagenomes from yellowstone hot springs. Applied and environmental microbiology 2008, 74: 4164. 10.1128/AEM.02598-07

    PubMed Central  Article  CAS  PubMed  Google Scholar 

  36. 36.

    GeneMark Gene Prediction[]

  37. 37.

    ORF Finder (Open Reading Frame Finder)[]

  38. 38.

    Calculating fractional GC content[]

  39. 39.

    Lowe TM, Eddy SR: tRNAscan-SE: A program for improved detection of transfer RNA genes in genomic sequence. Nucleic acids research 1997, 25: 955. 10.1093/nar/25.5.955

    PubMed Central  Article  CAS  PubMed  Google Scholar 

  40. 40.

    tRNAscan-SE Search Server[]

  41. 41.

    Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ: Basic local alignment search tool. Journal of molecular biology 1990, 215: 403-410.

    Article  CAS  PubMed  Google Scholar 

  42. 42.

    Motif Search Server[]

  43. 43.


  44. 44.

    Stothard P, Wishart DS: Circular genome visualization and exploration using CGView. Bioinformatics 2005, 21: 537. [] 10.1093/bioinformatics/bti054

    Article  CAS  PubMed  Google Scholar 

  45. 45.

    Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, McWilliam H, Valentin F, Wallace IM, Wilm A, Lopez R: Clustal W and Clustal X version 2.0. Bioinformatics 2007, 23: 2947. [] 10.1093/bioinformatics/btm404

    Article  CAS  PubMed  Google Scholar 

  46. 46.

    Tamura K, Dudley J, Nei M, Kumar S: MEGA4: molecular evolutionary genetics analysis (MEGA) software version 4.0. Molecular biology and evolution 2007, 24: 1596. [] 10.1093/molbev/msm092

    Article  CAS  PubMed  Google Scholar 

Download references


This study is part of a Doctoral work funded in part by the Alabama Agriculture Experiment Station (ALA080-051). Thanks to members of the Liles laboratory at Auburn University for providing support needed in completing this study. Special thanks are given to Nancy Capps for ensuring that all materials and equipments were available and in good working conditions when needed.

Author information



Corresponding author

Correspondence to Mark R Liles.

Additional information

Competing interests

The authors declare that they have no competing interests.

Authors' contributions

AC prepared phage DNA, assembled and finished phage genomes, conducted the phage genome annotation and phylogenetic analyses, and drafted the manuscript. TJW isolated the eiMSLS phage, prepared phage DNA, and contributed to finishing the eiMSLS genome sequence. GCW produced the primary DNA sequence for all three phage genomes, and assembled and finished one phage genome. DAM prepared phage genome shotgun subclone libraries for sequencing. JST was the primary supervisor of AC, helped coordinate the collaborative work, and contributed to the intellectual design of the project. MRL co-supervised AC, helped in the assembly and finishing of the phage genomes, and in the manuscript design and editing. All authors read and approved the final version of the manuscript.

Authors’ original submitted files for images

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Carrias, A., Welch, T.J., Waldbieser, G.C. et al. Comparative genomic analysis of bacteriophages specific to the channel catfish pathogen Edwardsiella ictaluri. Virol J 8, 6 (2011).

Download citation


  • Phage Genome
  • Aquaculture Pond
  • Tail Fiber
  • Phage Tail
  • Enterobacterium Phage