https://orcid.org/0009-0003-5695-505X
Figures
Abstract
Bacteriophages are ubiquitous and highly genetically diverse biological entities. Here we describe the isolation and bioinformatic characterization of 56 phages isolated on two Rhodococcus spp. They include both lytic and temperate phages and are grouped with previously described Rhodococcus phages into six clusters and 16 singletons based on genome similarity. Their genome sizes range from 43.9 kbp to 142 kbp and they have a G + C content ranging from 41.2% to 68.4%. Some of the Rhodococcus phages are more closely related to phages isolated on non-Rhodococcus Actinobacteria hosts than they are to phages isolated from the same host genus, demonstrating complex evolutionary histories. This study further expands the growing field of Actinobacteriophage genomics.
Citation: Garza DR, Di Blasi DL, Klyczek KK, Bruns JA, DeJong RJ, Findley AM, et al. (2026) Characterization of genomic diversity in bacteriophages infecting Rhodococcus. PLoS One 21(6): e0352686. https://doi.org/10.1371/journal.pone.0352686
Editor: M. Alejandro Dinamarca, Universidad de Valparaiso, CHILE
Received: March 31, 2026; Accepted: June 12, 2026; Published: June 29, 2026
Copyright: © 2026 Garza et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All phage sequences are available in GenBank. Accession numbers are listed in Table 1.
Funding: SGC: Howard Hughes Medical Institute (HHMI) grant # GT17581. https://www.hhmi.org/. GFH: National Institutes Health Grant #GM131729. https://www.nih.gov/. GFH: Howard Hughes Medical Institute (HHMI) Grant # GT12053. https://www.hhmi.org/. Funders did not play any role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Bacteriophages infecting bacterial hosts in the phylum Actinobacteria exhibit an expansive spectrum of diversity [1]. Genomic analyses have been conducted on large numbers of phages isolated on a single host genus, such as Mycobacterium [2], Gordonia [3], Arthrobacter [4], and Microbacterium [5], facilitated by students and faculty in the in the Science Education Alliance-Phage Hunters Advancing Genomics and Evolutionary Science (SEA-PHAGES) and Phage Hunting Integrating Research and Education (PHIRE) programs [6]. These studies revealed that populations of phages infecting a single host include those that are genomically very similar and can be grouped into clusters and sub-clusters based on nucleotide sequence similarity and shared gene content, and others that are distinct enough to remain as ‘singleton’ phages [7,8]. The phage genomes are pervasively mosaic, and they share genes both within and between clusters. This mosaicism likely arises by horizontal gene transfer and illegitimate recombination as phages migrate across a landscape of bacterial hosts and acquire new genes from other phages and bacteria [9]. Analysis of phages from additional Actinobacteria strains will help to elucidate these mechanisms.
Comparative genome analyses of actinobacteriophages are facilitated by construction of databases of annotated genomes and grouping predicted protein coding genes into ‘phamilies’ (phams) according to amino sequence relationships of the predicted proteins [7,10]. Genes with no close relatives within this database are referred to as ‘orphams’ [7]. Pham assignments are made by the program ‘Phamerator’ which also constructs genome maps illustrating genome organization, pairwise nucleotide comparisons, and pham assignments [7]. Parameters for the phamily assignments, map constructions, and other bioinformatic considerations have been previously reported [8,11,12].
The role of Rhodococcus species in bioproduction, bioremediation and disease highlight the need for a better understanding of Rhodococcus phages. The Rhodococcus genus comprises Gram-positive Actinobacteria containing mycolic acids in their cell wall [13] and are commonly found in soil and aquatic environments [14]. Rhodococcus strains have played critical roles as biocatalysts in the synthesis of several organic compounds such as acrylamide, used in the production of polyacrylamide, and applications such as oil recovery and water treatment [15]. Additionally, there has been some success in the application of Rhodococcus for bioremediation efforts in contaminated environments due to their ability to break down polycyclic aromatic compounds [16]. There are two known pathogenic species of Rhodococcus; R. fascians is the causative agent of leafy gall disease in herbaceous perennials such as dahlias, petunias, and hostas, and R. equi (also known as Prescottella equi) [17] is an animal pathogen, causing pneumonia in young horses, immunocompromised horses, and immunocompromised humans [18]. Rhodococcus erythropolis was used as a bacterial host for phage discovery because it has simple biosafety precautions and is easy to grow and use for phage propagation.
Here, we describe the isolation and characterization of 57 phages isolated on R. erythropolis or R. equi and their comparative analysis with fourteen phages previously isolated on Rhodococcus species (Table 1) [19–26]. We describe the physical and genomic characteristics of the Rhodococcus phages, and the evolutionary relationships among these and other phages isolated on Actinobacteria hosts.
Results and discussion
Rhodococcus phage isolation
Twenty-eight newly-isolated Rhodococcus phages were recovered from soil by plaque purification, using either enriched or direct methods. Most of these phages were isolated on R. erythropolis (47/56 phages); a small subset was isolated on R. equi (9/56 phages). Phages were isolated by students and faculty participating in the SEA-PHAGES or PHIRE program at 13 institutions (S1 Table). Comparative genomic analyses described here include an additional 42 Rhodococcus phages reported previously but not in the context of detailed comparative genomics. Table 1 lists the characteristics of all 70 Rhodococcus phages [20–24,26] and additional features of the phages are shown in S1 Table.
Transmission electron microscopy (Fig 1) shows that with only one exception all of the newly isolated Rhodococcus phages have siphoviral morphotypes, with isometric capsids and flexible non-contractile tails of varying lengths; the exception is Finch that has a myoviral morphology. Similarly, among the previously described Rhodococcus phages, E3 is a myovirus [24], Toil is a tectivirus (14), and all of the others are siphophages.
Phages shown are: A) ChewyVIII, B) Sleepyhead, C) Trogglehumper, D) NiceHouse, (Cluster CE); E) Whack, F) Shagrat (Cluster CR6), G) MacGully (Cluster CR7), H) Weasels2 (Cluster CB), I) Reynauld, J) Finch, K) Rasputin (Cluster CA), and L) Maselop (Cluster CR6).
Rhodococcus phage genometrics
Phage DNA was extracted, sequenced, and annotated using automated gene predictions followed by manual inspection and revision. GenBank accession numbers are shown in Table 1. Comparisons of the 56 newly isolated phages and 14 previously described phages showed that genome sizes range from 14,270 bp (RRH1) to 142,586 bp (NiceHouse) (Table 1). The genome termini vary in nature with most having short (5–11 bases) single-stranded 3’ extensions, but some are circularly permuted and presumably terminally redundant, and others have direct terminal repeats varying in length from 330 bp (Trogglehumper) to 5,291 bp (NiceHouse and Trina) (S1 Table). There is also substantial variation in G + C content (Table 1) and this is discussed in further detail below.
Phage cluster assignments and relationships
The Rhodococcus phages were clustered using both gene content similarity (GCS) – with phages sharing approximately 35% or more of their genes assigned to the same cluster – and protein equivalency quotients (PEQ), both as described previously [3,10]. These analyses are in good agreement, and revealed six distinct clusters (CA, CB, CC, CE, CF, CR) and 16 singletons (Fig 2A, 2B; S2, S3 Tables). The only minor discrepancy is that the PEQ for Cluster CF phages Jflix2 and Shagrat is marginally below the 25% cutoff value (24.2), although the GCS is above 35% (S2 Table) justifying their grouping in Cluster CF. We note that the distribution of phages among the clusters is highly heterogenous, with Cluster CA being by far the largest with 38 members, Cluster CR with 6 and all other clusters with three or fewer members (Fig 2).
A. Heat map of the gene content similarity (GCS) among Rhodococcus phages. Pairwise GCS values between Rhodococcus phage genomes were computed and displayed in a heat map using Prism [27]. B. Heat map of proteomic equivalence quotient (PEQ) values of Rhodococcus phages. Pairwise PEQ values between Rhodococcus phage genomes were computed and displayed in a heat map using Prism [27].Clusters are denoted to the right; sin denotes singleton phages. Raw data are shown in S2 and S3 Tables.
In general, there is high intra-cluster pairwise shared gene content with a mean average value of 69.3%. Clusters CE (38.4% mean GCS) and CF (42.8% mean GCS) have among the lowest intra-cluster similarities (Fig 2A, 2B). Overall, the inter-cluster similarities are low (<1% GCS) (Fig 2A, 2B) indicating that these are not actively exchanging genes even though they infect the same or similar Rhodococcus hosts. These delineations are clearly reflected in a network phylogeny based on shared gene content (Fig 3; S4, S5 Tables). However, notable departures to this are the relatively close relationships between Cluster CB and CE phages (average 24.7% GCS), and between some of the singleton phages such as Whack and REQ2 (31.6% GCS), Reynauld and Trogglehumper (19.8%), Reynauld and Mbo2 (31.5% GCS), Reynauld and DocB7 (23.1% GCS), and Finch and E3 (20.7% GCS). ChewyVIII shares an average of 23.8% GCS with the Clusters CR6 and CR7. These are also reflected in their branch lengths in the network phylogeny (Fig 3).
SplitsTree6 was used to visualize phage relationships of clusters based on presence or absence of shared phams. Clusters are indicated by colored ovals with singletons labeled by phage name. Only selected clusters of Mycobacterium, Gordonia, and Arthrobacter phages are shown to emphasize the relationships. The color of the oval indicates the isolation host as specified in the legend. The scale bar indicates substitutions per site. Evolutionary relationships of bacterial host taxa inferred from 16S rRNA sequences. The 16S rRNA gene sequences from representative Mycobacterium, Gordonia, Rhodococcus and Arthrobacter hosts were aligned with MAFFT v7.525, trimmed using trimAl v1.5.0, and analyzed by maximum likelihood in IQ-TREE v3.0.1 under the TIM3 + F + I + G4 model selected by ModelFinder [28]. Node support was calculated from 1,000 ultrafast bootstrap and 1,000 SH-aLRT replicates. The resulting tree was imported into iTOL for final visualization and annotation.
The network phylogeny also reveals interesting relationships between the Rhodococcus phages and phages isolated on Mycobacteria spp, Gordonia spp, and Arthrobacteria spp. (Fig 3). The Rhodococcus Clusters CB, CE, and CF, and eight of the singletons (Toil, Jace, REQ3, Sleepyhead, REQ2, Whack, Pine5, and Mbo4) share few if any genes with these other phages (Fig 3). However, seven Rhodococcus phages cluster with 38 previously described Gordonia phages grouped in Cluster CR (Fig 3), although Cluster CR is very diverse and can be readily subdivided into seven subclusters. The Gordonia phages occupy Subclusters CR1 to CR5, whereas the Rhodococcus phages – all of which were isolated on R. equi NRRL B-16538 are in Subclusters CR6 and CR7 (Table 1, Fig 3) [3]. The only other Actinobacteriophage cluster containing phages isolated on more than one host is Cluster A, for which Subcluster A15 phages were isolated on Gordonia sp. and all others are mycobacteriophages [1]. It is notable though that Cluster CA Rhodococcus phages have substantial shared gene content with Cluster A phages (Fig 3).
Another notable set of relationships is between Rhodococcus singleton RHH1, the Cluster AN Arthrobacter phages, and Cluster CW Gordonia phages [3,4,20]. These are among the smallest Actinobacteriophage genomes (14–16 kbp) and include mostly virion structural genes with a small number of DNA-binding protein genes. We also note that Rhodococcus singletons Trogglehumper, DocB7, Reynauld, and Mbo2 share some genes with Gordonia phage (Clusters CS, DF, and CX), and singletons Finch and E3 share genes with Gordonia (Clusters DO and DX) and Mycobacterium phages (Clusters AA and C) (Fig 3). The Rhodococcus Cluster CC phages share genes with Arthrobacter (Clusters AM, AW, and AU) and Gordonia Cluster DJ phages as noted previously [4,5,26]. These phages share similar genome architectures including many small genes with transmembrane domains clustered in the center of the genome.
Phage lifestyles.
For many of the phages, plaque morphotypes do not unambiguously show if they are lytic or temperate. However, genomic characterization provides strong clues, as temperate phage often have recognizable integrase and repressor genes. Four clusters (CB, CC, CE and CR) and four newly isolated singletons (ChewVIII, Finch, Reynauld and Trogglehumper) are predicted to be lytic, and two (CA and CF) and three of the newly isolated singletons (Jace, Sleepyhead and Whack) are predicted to be temperate.
Cluster-specific phage features
Each of the clusters and the singleton phages have notable features and are discussed in turn in the sections below.
Cluster CA.
Cluster CA is the largest cluster of Rhodococcus phages, containing 38 of the 70 annotated phages. All Cluster CA phages were isolated on R. erythropolis except for the previously described phage RGL3, which was isolated on R. globerulus Rglo35 [20]. Cluster CA phages isolated on R. erythropolis share a high degree of sequence similarity; the entire cluster has a minimum pairwise GCS value of 75.8% (Fig 2) with a minor variations (S1A and S1B Figs). A detailed map of a representative Cluster CA phage (Rasputin) is shown in Fig 4.
The genome is shown with predicted genes represented as boxes above or below the genome reflecting rightwards and leftwards transcription, respectively. Genes are colored according to their phamily designations using Phamerator.org and database Actinobacteriophage_4268. Predicted gene functions are indicated above the genome. tRNA genes are represented by gray boxes with the amino acid specified above the box. Predicted gene functions are noted above the genes. Vertical arrows indicate the positions of putative stoperator sites with upwards and downwards arrows depict two different orientations of the asymmetric stoperators.
Cluster CA phages have similarity to Cluster A phages and share closely related genome architectures [29]. Cluster CA phages share a mean GCS of 36.4% with phages in Cluster A, compared to a mean GCS of 1.3% with other Rhodococcus phages. The virion structural genes (e.g., terminase, portal protein, major capsid protein, capsid maturation protease, scaffolding protein, major tail protein, and head-to-tail connectors, tape measure protein) in the left arm are transcribed in the rightwards direction, while the right arm gene are leftwards-transcribed and encode DNA metabolism functions (e.g., DNA ph, DNA polymerase I, and ribonucleotide reductase; Fig 4). In an unusual departure in the genome organizations, the putative holin is not located near the lysin A gene (gene 7; Fig 4) to the left of the terminase genes but is positioned among the minor tail protein genes (Fig 4). The identification of Rasputin 7 and its homologues as holins is supported by noting that a homolog in Gordonia Cluster CD phage Puppers is adjacent to the Lysin A gene, which is also near the minor tail protein genes.
Similarly to Cluster A, the Cluster CA phages are temperate [30], with a centrally located integrase gene of the serine-recombinase family (e.g., Rasputin 29) and an immunity repressor gene (e.g., Rasputin 58) in the genome right arm (Fig 4). It is not uncommon to find Cluster A phage that appear to be naturally lytic due to loss of the repressor gene, although this could arise from the selection of more visible plaques in the isolation process. This seems relatively rare among the Cluster CA phages, although Nancinator is one example. A notable feature of Cluster A phages are multiple copies of stoperator sequences (13–14 bp) located throughout the genome, positioned in one direction relative to transcription, and either intergenic or overlapping gene termini [31]. The Cluster CA phages share these features, and Rasputin (Fig 4) has 25 putative stoperator sites with the consensus sequence 5’-TGTCTATTGTCAAG. This differs from the consensus sequences in Cluster A phages [32,33].
Most of the Cluster CA phages code for at least two tRNA genes near the left ends of the genomes (Fig 4, S1 Fig). These tRNA-Asn and tRNA-Trp genes are separated by a region coding for a potential third tRNA whose charging potential is ambiguous; in some genomes at least one of the tRNA genes is absent or lost (S1 Fig). Transcriptomic analysis of Cluster CA phage WC1 during R. erythropolis infection [30] shows that these tRNA genes are transcribed late in phage infection. We note that in the previously described phages RER2 and RGL3 [21], the region corresponding to the left-most 2.5 kbp of the genome including the tRNA genes and the lysin A gene are located at the extreme right end of the genome. Although it is plausible that the cos site has been moved to a new location, it is also possible that the physical left ends of the viral genome are incorrectly identified.
Cluster CB.
The three Cluster CB phages (Grayson, Peregrin, and Weasels2) are lytic, and have unusually large genomes for siphophages, averaging 133 kbp (Table 1) and 277 predicted genes (Fig 5). The genome ends have long direct terminal repeats (2,900, 2,937, and 3,268 bp for Grayson, Peregrin, and Weasels2, respectively) each encoding eight putative genes (Fig 5). As of November 2025, these three phages have the lowest G + C content (41%) among the 5,460 sequenced actinobacteriophages genomes (https://phagesdb.org). Greyson and Peregrin are more similar to each other (85.0% GCS) than either phage is to Weasels2 (62.4% and 60.6% GCS, respectively; Fig 2, S2 Fig). The genome organization of Peregrin is shown in Fig 5.
See Fig 4 for details. White boxes indicate orphams (genes with no other phamily members).
The Cluster CB genomes have a different gene organization compared to other siphoviral phages (Fig 5). Most of the genes are rightwards-transcribed, but 55 putative genes at the left end of the genome and 29 at the right end are leftwards-transcribed. The virion structural genes have a non-canonical order, in which many of the minor tail protein genes (69–74) are positioned upstream of the portal-capsid genes (Fig 5). Additionally, the large terminase subunit (112) is located downstream and over 16 kbp away from the other virion structural genes. Oddly, there are two related putative major tail subunits in Peregrin (70, 81) and Grayson (72, 83) sharing 52% amino acid identity (Fig 5), and three in Weasels2 (70, 72, 83). It is not known if all of these are present in virions. These phages also code for a glycosyltransferase and a methyltransferase (Peregrin 60 and 49, respectively) suggesting that the virions may be extensively glycosylated as reported for some other actinobacteriophages [34]. Two putative lysin genes (Lysin A and Lysin B) are also in unusual genomic locations (Fig 5). The genomes also code for several DNA metabolism functions including a DNA Pol III alpha subunit (Peregrin 100).
The Cluster CB phages code for 11–15 tRNA genes (Table 1), mostly organized into several small clusters near the center of the genomes (Fig 5). Nearby are two genes coding for RNA ligase (Peregrin 153 and 155), the latter of which is related to RtcB proteins, and these together with the tRNA genes may be involved in countering host defenses involving tRNA turnover. Peregrin and Weasels2 also encode a T4 rII-like system (Peregrin genes 116, 117) [35], which is also implicated in interfering with defense systems.
Another striking feature of the Cluster CB phages is the abundance of small open reading frames with no known function; 72.8% of the Peregrin genes are shorter than 500 bp (Fig 5). This contributes to a relatively large number of genes compared to the genome size (Peregrin has 287 genes in its ~ 133 kbp genome). Weasels2, which is more distantly related to Peregrin and Grayson has an unusually large proportion (~30%) of Orphams [7] (S2 Fig), gene present only once on all of the actinobacteriophages (as of November 2025).
Cluster CE.
Cluster CE includes virulent phages NiceHouse and Trina, both of which were isolated on R. erythropolis NRRL B-1574. However, Trina and Nicehouse are not closely related (38.4% GCS) and only barely surpass the threshold for cluster inclusion (35% GCS). They are not closely related to the CB phages (Fig 2) but share many of the same features. They have large genomes for siphophages (~140–145 kbp) and a large number of genes (~290), many of which are small and of unknown function (Fig 6, S3 Fig, Table 1). They also share an unusual virion structure gene organization but have only a single major tail subunit gene. They code for many tRNA genes (33 in Trina and 31 in NiceHouse) as well as an RNA ligase and also have the T4 rII-like system and code for a DNA pol II alpha subunit, as in the Cluster CB phages. Both Trina and Nicehouse have an abundance of orphams (50.4% and 46.5% in Tina and NiceHouse, respectively).
Cluster CF.
Cluster CF includes three temperate phages, Jflix2, Shagrat, and REQ1. Jflix2 and Shagrat were isolated on R. equi NRRL B-16538, and REQ1 reported previously [22] was isolated on R. equi Requ28. Although the PEQ values warrant cluster inclusion, some pairwise GCS values, such as between Jflix2 and Shagrat are relatively low (31.2%) (Fig 2, S4 Fig); a genome map of Jflix is shown in Fig 7. The structural gene organization is near-canonical for siphophages, although a tail protein is located near the left genome end as seen in some Cluster A phages. Interestingly, the capsid and capsid maturation protease functions are within a single gene (Jflix2 gene 11; Fig 7). Cluster CF genomes code for a serine integrase (Jflix 51) and an immunity repressor (Jflix 52), consistent with them being temperate [36]. Shagrat and Jflix2 have a lysin A (Shagrat 27 and Jflix2 26). Host range analyses showed that Jflix2 and Shagrat did not form plaques on two additional Rhodococcus species (R. globerulus ATCC 15903, R. jialingiae ATCC 31636) or on two Gordonia species (G. rubripertincta ATCC 25593, G. terrae 3612) [36].
Cluster CR.
Cluster CR phages are divided into seven subclusters (CR1 – CR7). However, the cluster contains phages isolated on both Rhodococcus spp and Gordonia spp and is divided into seven subclusters; all phages in Subclusters CR1 - CR5 were isolated on Gordonia and all of the phages in Subclusters CR6 and CR7 were isolated on Rhodococcus (Table 1). Nonetheless, the Cluster CR Rhodococcus phages do not infect Gordonia strains [36]. All Cluster CR phages have siphophage morphotypes and are predicted to be lytic. A map of Subcluster CR6 phage Apiary is shown in Fig 8, and the Cluster CR relationships are shown in S5 Fig.
Most of the Cluster CR phage genes are rightwards-transcribed (e.g., Apiary genes 1–50 and 83–95), with a central segment of leftwards-transcribed genes (e.g., Apiary genes 52–82) (Fig 8). The left-most series of rightwards-transcribed genes contains the virion structure and assembly genes, but also a block of genes – mostly of unknown function – between terminase small subunit gene (1) and the structural genes. Lysin A and holin genes are located to the right of the structural genes and the end of the putative late operon, although a gene related to mycobacteriophage Lysin B proteins is to the left the structural genes, and any role in lysis is unclear. The Cluster CR genomes do not encode tRNA’s, although they do code for an RtcB-like RNA ligase, and DNA metabolism functions including DNA Polymerase III alpha subunit, a primase/polymerase, a DNA helicase, and several putative regulators (Fig 8).
Singleton phages ChewyVIII, Finch, Jace, Sleepyhead, Whack, Reynauld, Trogglehumper.
ChewyVIII – isolated on R. erythropolis RIA 643 – is most closely related to Cluster CR phages (Fig 3); it shares a similar genome organization (Fig 9) but does not meet the GCS threshold criterion for cluster inclusion (S5Fig). It also includes an operon of leftwards-transcribed genes (7–14) near its left end that are absent from other Cluster CR phages (Fig 9). Moreover, about 50% of its genes are orphams with no closely related genes in the database used for comparisons (see Methods). ChewyVIII notably codes for an Ocr-like anti-restriction protein (ChewyVIII 86) and homologues are present in Cluster BD Streptomyces phages (Fig 9).
Rhodococcus phage Finch and the previously described Rhodococcus phage E3 are the only Rhodococcus phages with myoviral morphology. Finch is predicted to be lytic and has a relatively large genome (138,896 bp), with circularly permuted and presumably terminally redundant genome termini (Table 1, S1Table, Fig 10). With the exception of gene 61 coding for a DNA helicase, all of the genes are rightward-transcribed (Fig 10). A striking feature of the genome is the large array of small genes (88–200) – mostly of unknown function – spanning ~40 kbp, most of which are orphams (Fig 10). Finch DNA is likely modified and is not sensitive to MseI digestion (Fig 10); genes 210–215 are implicated in DNA modification because of their similarity to the DNA modification system in phage Rosebush as well as its relatives [37].
See Figs 4 and 5 for details. Also shown is the gene alignment with Mycobacterium phage Rosebush’s genes responsible for a PreQ0 pathway guanine modification [37].
Jace is a temperate siphophage sharing few genes with any other Rhodococcus phages (Fig 3). It is one of only two Rhodococcus phages with circularly permuted and presumably terminally redundant genomes, and the left was annotated as the first base of the small terminase subunit gene. Most of the genes are rightwards-transcribed, with the exceptions of genes 25, 27,28, 49, 89–91 and 93 (Fig 11). It also codes for six tRNAs, with the genes located at the extreme right end of the genome. As a temperate phage it codes for a putative repressor (gene 28) with similarities to the immunity repressors of Gordonia Cluster DN phages, and a tyrosine-integrase (gene 26) (Fig 11). A putative attP site is positioned immediately to the left of the integrase gene containing a region of 49 bp nucleotide identity with the 3’ end of a tRNA-gly gene in the Rhodococcus genome that defines the chromosomal attB site. Over half of the Jace genes are orphams with no homologues in the database used for comparisons (Fig 11).
Sleepyhead is a siphophage isolated on R. erythropolis NRRL B-1574 (Fig 12) and has 67 protein-coding genes but no tRNA genes (Table 1). The most closely related phage is Rhodococcus phage Whack, with which it shares 10 genes. Sleepyhead is predicted to be temperate, with the repressor and integrase encoded by genes 40 and 38, respectively (Fig 12). The attP site is located immediately to the left of the int gene and interestingly has a 91 bp near perfect match (90/91) to the extreme 3’ end of an acyl co-A dehydrogenase family protein (gene QIE55_RS12330 in the Rhodococcus erythropolis MGMM8 genome (Accession # NZ_CP124545.1). Presumably Sleepyhead uses this as an attB site but notably reconstructs the 3’ end of the coding region to retain functionality. Sleepyhead forms turbid plaques on a lawn of R. erythropolis and forms stable lysogens that release phage particles into culture supernatant and are immune to superinfection (not shown). Sleepyhead carries an IS3 like transposon flanked by 13 bp inverted repeats (5’-GGGCCTTGACCCCG) flanked by a 3 bp target duplication. It codes for two open reading frames (27, 28) that are likely expressed as single polypeptide DDE transposase by a programmed translational frameshift (S6 Fig), as with other IS3-like elements [38]. Related transposons are present in other actinobacteriophages infecting other bacterial hosts including UncleRicky (Mycobacterium), and Blueberry, Whitney and Cucurbita (Gordonia) (Fig 12; S7 Fig).
Whack was isolated on Rhodococcus erythropolis NRRL B-1574 and is most closely related to the singleton REQ2, but sharing only 31.6% GCS, below the threshold for cluster inclusion (Fig 2, S8 Fig). It is predicted to be temperate (Fig 13). Most of the genes are transcribed rightwards (1–31, 41–77) but nine genes (32–40) are leftwards-transcribed including the putative repressor (40) and integrase (39) genes (Fig 13). The location of the attP is unclear, and it could be located in the short intergenic region to the left of the integrase gene, but there are no compelling sequence matches to Rhodococcus genomes that might indicate the potential attB site. The virion structural genes have a canonical organization for siphophages, although the lysis cassette appears to be unusually located within the minor tail protein genes (Fig 13).
Singleton phages Trogglehumper (Fig 14) and Reynauld (Fig 15) have similar virion morphologies with isometric heads and unusually long tails (Fig 1). They also have similar genome sizes and organizations and share some genes with each other (19.8% GCS), particularly among the virion structure and assembly genes. They are also genetically related to Rhodococcus singleton phages DocB7 and Mbo2 and Gordonia phages in Clusters CS, CX, and DF (Fig 4). Their genome structures are similar, with similar lengths and G + C contents (Table 1), direct terminal repeats of 330 and 362 bases (Reynauld and Trogglehumper, respectively) and a genome organization where most of the left arm is transcribed in the forward direction and the right arm is transcribed the reverse direction. It’s likely that all these phages are lytic, and immunity repressor genes were not identified. However, they all have genes coding for a tyrosine-recombinase situated amid a series of left-wards transcribed genes in the right parts of the genome (Figs 14, 15). It seems unlikely that these recombinases are used as phage integrases for prophage integration and are probably involved in unrelated recombination events.
Variations in GC content and codon usage.
The Rhodococcus phages are unusual among the actinobacteriophages in having a wide range of GC contents, from 41.1% (Cluster CB phage Grayson) to 67.5% (singleton phage E3) and the Cluster CB and CE phages have some of the lowest GC of any of the actinobacteriophages. The Rhodococcus hosts have on average 62% G + C [39–42] such that there are substantial mismatches in G + C% content illustrated by Cluster CB phage Peregrin (41.4%) and its host R. erythropolis R138 (62.5%); this is even greater mismatch than between mycobacteriophage Patience (50.3% G + C) and its M. smegmatis host (65%) analyzed previously [43]. Not surprisingly, Peregrin has a very different profile of codon usage than R. erythropolis (Fig 16) including the predominance of codons with A or U in their third position relative to those with G or C (Fig 16). Because Peregrin codes for 14 tRNAs, this raises the question as to whether the tRNA specificities could explain the demand to translate codons that are typically rare in the host but abundant in the phage. However, this does not appear to be the case, as although Peregrin has a tRNA with an anticodon matching the rarest codon in the host (UUA), it also has tRNAs matching the related codon UUG which is very common in the host. Similarly, Peregrin has tRNAs with anticodons matching both of the lysine codons AAA and AAG, and in this instance the AAG codon is more common than AAA. We also note that in the host a single tRNA recognizes both of the tyrosine codons UAU and UAC, although the UAC codon is much more abundant (Fig 16).
Display of codon usage using RSCU (Relative Synonymous Codon Usage) as calculated in DNA Master. The boxes outlined in red match the tRNAs of each phage and the host.
Comparison of the Cluster CE phage Trina (44.7% G + C) codons usage with R. erythropolis shows a similar pattern. Again, the phage codons with A or U in their third position mostly outnumber those with C or G in each codon set (Fig 16). However, the tRNA profile is very different from that of Peregrin, and of the 33 Trina tRNAs, only 13 have similar anticodons to the Peregrin tRNAs. It is also notable that Trina has anticodons for Valine (codons GUN), Alanine (codons GCN), Glutamic Acid (codons GAR) and the Glycine codon GGC, all of which are absent from Peregrin (Fig 16). We note that some Rhodococcus phages (e.g., Cluster CA phage Rasputin) have tRNA genes but have similar G + C% to their bacterial hosts and similar codon usage profiles (Fig 16). For example, Rasputin has a tRNA-Trp although there is only a single tryptophan codon, and the other two tRNAs (tRNA-Asn and tRNA-Val) code have anticodons recognizing the codons with those codon sets that are the most abundant in the phage and the host.
These observations suggest that the role of the Rhodococcus phage-encoded tRNAs is not to compensate for the strong mismatches in codon usage that arise from substantial differences in GC. An alternative explanation is phages carry these tRNAs to cover host defenses that involve inactivation of tRNAs and abortive infection [44]. It is remarkable, however, that phages such as those in Cluster CE carry more than 30 tRNA genes, suggesting that the host defenses would involve widespread and non-specific tRNA destruction. The reason for the tRNA repertoire in phage genomes thus remains elusive and warrants further exploration.
Concluding remarks
The newly isolated Rhodococcus phages described here expand our view of the diversity of these phages and their genetic repertoires. As seen with other actinobacteriophages including those isolated on Mycobacterium, the types of Rhodococcus phages are heterogenous, with the temperate Cluster CA phages being predominant and representing over half of those described here. In contrast, there are 16 singleton phages, showing that Rhodococcus phages are greatly under sampled, and more of these singletons as well as new singleton types have yet to be isolated. An additional feature shared with Mycobacterium phages is that only two virion morphotypes are seen, with only siphophages and myophages represented. The relative abundance of lytic and temperate phages among genomic types is not dissimilar to that found in Mycobacterium phages. It is notable that the most common Mycobacterium phages and Rhodococcus phage – in Clusters A and CA, respectively – are both temperate, which might suggest that they are integrated as prophages in related environmental strains.
The Rhodococcus phages have the broadest range of GC contents among any of the actinobacteriophages, resulting in those with lower GC contents having greatly different codon usage profiles than the host. We favor the explanation described previously for mycobacteriophage Patience that the lower GC phages have largely evolved in bacterial hosts with lower GC contents and only relatively recently acquired the ability to infect and replicate in their current hosts. Although it might be anticipated that acquisition of tRNA genes could help to compensate for relative paucity of host tRNAs for infrequently used codons in the host, that does not appear to occur. Rhodococcus phages with low GC do carry a large repertoire of tRNA genes, but these do not appear to exclusively compensate for coding deficiencies.
The Rhodococcus phage genomes raise many questions about their biology and replication processes, including host ranges and their determinants, the roles of the large number of genes unknown function, the action of the lysis systems, and life cycle regulation. The isolation and characterization of these newly isolated Rhodococcus phage will help to provide answers to these questions, as well as expanding our view of actinobacteriophage diversity.
Materials and methods
Bacterial strains & media
All phages were isolated on one of two different species of Rhodococcus: R. equi (NRRL strain B-16538) or R. erythropolis (NRRL strain B-1574 or RIA strain 643) (Table 1) [29,36,45]. PYCa media (containing per 1 liter volume: 1.0 g Yeast Extract, 15.0 g Peptone, 2.5 ml of 40% Dextrose, 2.5 ml of 1 M CaCl2, and 1 ml of 1000X CHX) or BD Difco nutrient broth with added agar to 1.6 or 0.45% for agar plates or top agar overlay was used for phage isolation, purification and amplification.
Phage isolation, purification, amplification, and virion analysis
Phages were isolated from soil samples collected at several SEA-PHAGES institutions (S1 Table) using either an enriched or direct isolation protocol as described previously [29,36,45]. Multiple rounds of plaque picking and plaque assays were performed to ensure a homogenous phage population. Phage titers and plaque morphology were determined during each round of purification. High-titer phage lysates were generated by flooding plaque assay plates showing webbed lysis with phage buffer (containing per 1 liter volume: 10 ml of 1 M Tris stock (pH 7.5), 10 ml of 1 M MgSO4, 4 g NaCl, 10 ml of 100 mM CaCl2), incubating for at least 24 hrs at 4°C and then filtering through a 0.22 µm filter. Phage particles were imaged using transmission electron microscopy (TEM) where phage lysates were spotted onto carbon coated copper grids, stained with 1% uranyl acetate and imaged using a HITACHI 7800 TEM.
Genome sequencing, assembly and annotation
Phage DNA was extracted using a Promega Wizard DNA Extraction kit or phenol chloroform, and phage genomes were sequenced at either the Pittsburgh Bacteriophage Institute or Western Carolina University using Illumina, Ion Torrent, or Roche 454 methods [46]. Raw reads were assembled with Newbler (GS De Novo Assembler version 2.9), and assemblies were checked for completeness, accuracy, and genomic termini using Consed [47]. Phage genomes were annotated as described previously [48] using DNA Master, GLIMMER [49], GeneMark [50], BLAST [51], Aragorn [52] and tRNAScan-SE [53]. Gene functions were determined using Blast, HHpred [54], the Conserved Domain Database (CDD) [55], TMHMM [56] and DeepTMHMM [57]. Genome comparisons were performed using Phamerator.org and database Actinobacteriophage_4268 (created December 8, 2023).
Genome analyses
Phage genes were grouped into phamilies (phams) of closely related proteins using Phamerator via the pdm_utils database management package [7,12]. Genome architecture, and nucleotide sequence similarity were compared using Phamerator.org (https://phamerator.org/) [7]. Gene Content Similarity (GCS) analysis was performed with PhamClust using the ‘-m gcs’ flag, and data were visualized using GraphPad Prism (www.graphpad.com). PEQ values were calculated using PhamClust [10] and visualized using GraphPad Prism. For cross-host analyses, two representative phages were selected from each cluster with more than three members (S1 Table). Gene content network phylogenies were created using PhamNexus and SplitsTree6 [58] to render an unrooted tree using equal angle and Neighbornet functions. Intra-cluster pham matrix comparisons were performed using a custom code notebook on the Observable platform (https://observablehq.com/@cresawn-labs/pham-matrix). Codon usage biases were estimated using all genes from phage genomes Rasputin, Peregrin, Trina, and model host Rhodococcus erythopolis R138 (CP007255). Relative synonymous codon usage (RSCU), as part of the DNA Master program, is the ratio of the codon count to the amino acid count that was then used to normalize the codon frequencies so that the sum of RSCU for codons of each amino acid equals the number of synonymous codons for each amino acid. Actual codon counts are found in supplemental data (S6 Table).
Supporting information
S1 Table. Additional Rhodococcus phage information.
https://doi.org/10.1371/journal.pone.0352686.s001
(XLSX)
S2 Table. Pairwise GCS data for all Rhodococcus phages.
https://doi.org/10.1371/journal.pone.0352686.s002
(XLSX)
S3 Table. Pairwise PEQ values for all Rhodococcus phages.
https://doi.org/10.1371/journal.pone.0352686.s003
(XLSX)
S4 Table. Pairwise GCS values for representative Rhodococcus and non-Rhodococcus actinobacteriophages.
https://doi.org/10.1371/journal.pone.0352686.s004
(XLSX)
S5 Table. Pairwise PEQ values for representative Rhodococcus and non-Rhodococcus actinobacteriophages.
https://doi.org/10.1371/journal.pone.0352686.s005
(XLSX)
S6 Table. Codon usage table counts.
Relative Synonymous Codon Usage numbers and actual codon counts are displayed in table format for each phage or host listed.
https://doi.org/10.1371/journal.pone.0352686.s006
(XLSX)
S1 Fig. Genomic diversity of Cluster CA phages.
A) Phamerator.org map. Each phage genome is shown with predicted genes represented as boxes above or below the genome reflecting rightwards and leftwards transcription, respectively. Each colored box represents a gene, colored according to pham membership as defined by Phamerator.org. White boxes are orphams (genes with no other phamily members). The shading between genomes indicates pairwise nucleotide identity in rainbow order, with purple indicating high similarity, red indicating low similarity, and white indicating no similarity. B) Pham matrix. Each row represents a phage gene map with rectangular boxes representing genes. All boxes are the same width, irrespective of the nucleotide length of the gene. The genes are color-coded by phamily as defined by Phamerator.org, with white boxes indicating orphams. Each column of boxes represents a gene phamily, arranged by mean genomic position along the X axis.
https://doi.org/10.1371/journal.pone.0352686.s007
(PDF)
S2 Fig. Genomic diversity of Cluster CB phages.
See S1 Fig for details.
https://doi.org/10.1371/journal.pone.0352686.s008
(PDF)
S3 Fig. Genomic diversity of Cluster CE phages.
See S1 Fig for details.
https://doi.org/10.1371/journal.pone.0352686.s009
(PDF)
S4 Fig. Genomic diversity of Cluster CF phages.
See S1 Fig for details.
https://doi.org/10.1371/journal.pone.0352686.s010
(PDF)
S5 Fig. Genomic diversity of Cluster CR phages.
See S1 Fig for details of A) and B). C). Display of Gene Content Similarities (GCS) amongst the Rhodococcus Cluster CR phages and singleton ChewyVIII.
https://doi.org/10.1371/journal.pone.0352686.s011
(PDF)
S6 Fig. The nucleotide sequence and three reverse oriented translational reading frames of the Sleepyhead insertion sequence.
The direct repeats (TG) are indicated in bolded nucleotides and the left and right inverted repeats (LIR and RIR) are indicated by arrows. The gray arrows represent protein coding genes with gp28 (light gray) and gp27 (dark gray) as ORF 1 and ORF2 of the insertion sequence, respectively. The −10 and −35 boxes of a putative promoter overlapping the RIR is indicated by light green arrows and underlined nucleotides. The reading frames for Sleepyhead gp27 and 28 are highlighted in light gray. The TTTC tetramer at the 3’ end of ORF1 (gp28) that potentially signals a −1 programmed ribosomal frameshift is highlighted in light blue.
https://doi.org/10.1371/journal.pone.0352686.s012
(PDF)
S7 Fig. Alignment of DDE transposase sequences.
The DDE transposase (Pham 8239; in purple) is found in the genomes of Rhodococcus phage, Sleepyhead (Singleton), Gordonia phages Blueberry (CV), Whitney (DN1), Cucurbita (CQ1) and Mycobacterium phage UncleRicky (F1). Sleepyhead 27 encodes a DDE transposase and gene 28 has strong HHpred alignments to helix-turn-helix DNA binding domains associated with transposases (PF01710) [59]. The DDE transposase pham is found in the phage genomes of UncleRicky (Subcluster F1), Cucurbita (cluster CQ), Blueberry (Cluster CV), and Whitney (Cluster DN1) (Fig 14). The insertion sequence in these phage genomes has a second ORF that belongs to a different pham than Sleepyhead gp28, however all these genes have the same HHpred alignments to transposase associated helix-turn-helix DNA binding domains. While the transposase of some insertion sequences is expressed from a single ORF, transposases of the IS3 family typically exist as two overlapping open reading frames with ORF2 being in the −1 frame relative to ORF1 [38]. These have a −1 programmed ribosomal frameshift in the overlapping region of ORF1 and ORF2 that is signaled by a X-XXZ-ZZN heptamer or a Z-ZZN tetramer [25,60]. A T-TTC tetramer exists just upstream of the Sleepyhead gp28 stop codon that would allow a −1 frameshift and translation of Sleepyhead gp28:27 fusion protein 408 amino acids long (S6 Fig) [61,62].
https://doi.org/10.1371/journal.pone.0352686.s013
(PDF)
S8 Fig. Genome comparison of phages Whack and REQ.
Phamerator.org map. Each phage genome is shown with predicted genes represented as boxes above or below the genome reflecting rightwards and leftwards transcription, respectively. Each colored box represents a gene, colored according to pham membership as defined by Phamerator.org. White boxes are orphams (genes with no other phamily members). The shading between genomes indicates pairwise nucleotide identity in rainbow order, with purple indicating high similarity, red indicating low similarity, and white indicating no similarity.
https://doi.org/10.1371/journal.pone.0352686.s014
(PDF)
Acknowledgments
We thank University of California, Los Angeles students, Cristelle Hugo, Salvador Castillo, Britney Quijada, and Michelle Zorawik for early assistance in initiating these studies and James Madison University students, Isobel Cobb, Katherine Cooper, Madison Bendele, Ria Fisher, Zachary Jones, Zelda Shifflett for their work on the pham matrix notebook. We are grateful for support from the SEA-PHAGES program and the Howard Hughes Medical Institute.
References
- 1. Hatfull GF. Actinobacteriophages: Genomics, Dynamics, and Applications. Annu Rev Virol. 2020;7(1):37–61. pmid:32991269
- 2. Pope WH, Bowman CA, Russell DA, Jacobs-Sera D, Asai DJ, Cresawn SG, et al. Whole genome comparison of a large collection of mycobacteriophages reveals a continuum of phage genetic diversity. Elife. 2015;4:e06416. pmid:25919952
- 3. Pope WH, Mavrich TN, Garlena RA, Guerrero-Bustamante CA, Jacobs-Sera D, Montgomery MT, et al. Bacteriophages of Gordonia spp. Display a Spectrum of Diversity and Genetic Relationships. mBio. 2017;8(4):e01069-17. pmid:28811342
- 4. Klyczek KK, Bonilla JA, Jacobs-Sera D, Adair TL, Afram P, Allen KG, et al. Tales of diversity: Genomic and morphological characteristics of forty-six Arthrobacter phages. PLoS One. 2017;12(7):e0180517. pmid:28715480
- 5. Jacobs-Sera D, Abad LA, Alvey RM, Anders KR, Aull HG, Bhalla SS, et al. Genomic diversity of bacteriophages infecting Microbacterium spp. PLoS One. 2020;15(6):e0234636. pmid:32555720
- 6. Hanauer DI, Graham MJ, SEA-PHAGES, Betancur L, Bobrownicki A, Cresawn SG, et al. An inclusive Research Education Community (iREC): Impact of the SEA-PHAGES program on research outcomes and student learning. Proc Natl Acad Sci U S A. 2017;114(51):13531–6. pmid:29208718
- 7. Cresawn SG, Bogel M, Day N, Jacobs-Sera D, Hendrix RW, Hatfull GF. Phamerator: a bioinformatic tool for comparative bacteriophage genomics. BMC Bioinformatics. 2011;12:395. pmid:21991981
- 8. Gauthier CH, Cresawn SG, Hatfull GF. PhaMMseqs: a new pipeline for constructing phage gene phamilies using MMseqs2. G3 (Bethesda). 2022;12(11):jkac233. pmid:36161315
- 9. Jacobs-Sera D, Marinelli LJ, Bowman C, Broussard GW, Guerrero Bustamante C, Boyle MM, et al. On the nature of mycobacteriophage diversity and host preference. Virology. 2012;434(2):187–201. pmid:23084079
- 10. Gauthier CH, Hatfull GF. PhamClust: a phage genome clustering tool using proteomic equivalence. mSystems. 2023;8(5):e0044323. pmid:37791778
- 11. Gauthier CH, Hatfull GF. A bioinformatic ecosystem for bacteriophage genomics: PhaMMSeqs, Phamerator, pdm_utils, PhagesDB, DEPhT, and PhamClust. Viruses. 2024;16(8). pmid:39205252
- 12. Mavrich TN, Gauthier C, Abad L, Bowman CA, Cresawn SG, Hatfull GF. pdm_utils: a SEA-PHAGES MySQL phage database management toolkit. Bioinformatics. 2021;37(16):2464–6. pmid:33226064
- 13. Bell KS, Philp JC, Aw DW, Christofi N. The genus Rhodococcus. J Appl Microbiol. 1998;85(2):195–210. pmid:9750292
- 14. McLeod MP, Warren RL, Hsiao WWL, Araki N, Myhre M, Fernandes C, et al. The complete genome of Rhodococcus sp. RHA1 provides insights into a catabolic powerhouse. Proc Natl Acad Sci U S A. 2006;103(42):15582–7. pmid:17030794
- 15. Jiao S, Li F, Yu H, Shen Z. Advances in acrylamide bioproduction catalyzed with Rhodococcus cells harboring nitrile hydratase. Appl Microbiol Biotechnol. 2020;104(3):1001–12. pmid:31858190
- 16. Kuyukina MS, Ivshina IB. Application of Rhodococcus in bioremediation of contaminated environments. In: Alvarez HM. Springer. 2020.
- 17. Sangal V, Goodfellow M, Jones AL, Sutcliffe IC. A stable home for an equine pathogen: valid publication of the binomial Prescottella equi gen. nov., comb. nov., and reclassification of four rhodococcal species into the genus Prescottella. Int J Syst Evol Microbiol. 2022;72(9). pmid:36107761
- 18. Stewart A, Sowden D, Caffery M, Bint M, Broom J. Rhodococcus equi infection: A diverse spectrum of disease. IDCases. 2019;15:e00487. pmid:30656137
- 19. Gill JJ, Wang B, Sestak E, Young R, Chu K-H. Characterization of a Novel Tectivirus Phage Toil and Its Potential as an Agent for Biolipid Extraction. Sci Rep. 2018;8(1):1062. pmid:29348539
- 20. Petrovski S, Dyson ZA, Seviour RJ, Tillett D. Small but sufficient: the Rhodococcus phage RRH1 has the smallest known Siphoviridae genome at 14.2 kilobases. J Virol. 2012;86(1):358–63. pmid:22013058
- 21. Petrovski S, Seviour RJ, Tillett D. Characterization and whole genome sequences of the Rhodococcus bacteriophages RGL3 and RER2. Arch Virol. 2013;158(3):601–9. pmid:23129131
- 22. Petrovski S, Seviour RJ, Tillett D. Genome sequence and characterization of a Rhodococcus equi phage REQ1. Virus Genes. 2013;46(3):588–90. pmid:23381579
- 23. Salifu SP, Campbell Casey SA, Foley S. Isolation and characterization of soilborne virulent bacteriophages infecting the pathogen Rhodococcus equi. J Appl Microbiol. 2013;114(6):1625–33. pmid:23495898
- 24. Salifu SP, Valero-Rello A, Campbell SA, Inglis NF, Scortti M, Foley S, et al. Genome and proteome analysis of phage E3 infecting the soil-borne actinomycete Rhodococcus equi. Environ Microbiol Rep. 2013;5(1):170–8. pmid:23757146
- 25. Sekine Y, Eisaki N, Ohtsubo E. Translational control in production of transposase and in transposition of insertion sequence IS3. J Mol Biol. 1994;235(5):1406–20. pmid:8107082
- 26. Summer EJ, Liu M, Gill JJ, Grant M, Chan-Cortes TN, Ferguson L, et al. Genomic and functional analyses of Rhodococcus equi phages ReqiPepy6, ReqiPoco6, ReqiPine5, and ReqiDocB7. Appl Environ Microbiol. 2011;77(2):669–83. pmid:21097585
- 27. Yu Q, Ryan EM, Allen TM, Birren BW, Henn MR, Lennon NJ. PriSM: a primer selection and matching tool for amplification and sequencing of viral genomes. Bioinformatics. 2011;27(2):266–7. pmid:21068001
- 28. Kalyaanamoorthy S, Minh BQ, Wong TKF, von Haeseler A, Jermiin LS. ModelFinder: fast model selection for accurate phylogenetic estimates. Nat Methods. 2017;14(6):587–9. pmid:28481363
- 29. Bonilla JA, Isern S, Findley AM, Klyczek KK, Michael SF, Saha MS, et al. Genome sequences of 19 Rhodococcus erythropolis cluster CA phages. Genome announcements. 2017;5(49). pmid:29217789
- 30. Willner DL, Paudel S, Halleran AD, Solini GE, Gray V, Saha MS. Transcriptional dynamics during Rhodococcus erythropolis infection with phage WC1. BMC Microbiol. 2024;24(1):107. pmid:38561651
- 31. Brown KL, Sarkis GJ, Wadsworth C, Hatfull GF. Transcriptional silencing by the mycobacteriophage L5 repressor. EMBO J. 1997;16(19):5914–21. pmid:9312049
- 32. Pope WH, Jacobs-Sera D, Russell DA, Peebles CL, Al-Atrache Z, Alcoser TA, et al. Expanding the diversity of mycobacteriophages: insights into genome architecture and evolution. PLoS One. 2011;6(1):e16329. pmid:21298013
- 33. Mavrich TN, Hatfull GF. Evolution of Superinfection Immunity in Cluster A Mycobacteriophages. mBio. 2019;10(3):e00971-19. pmid:31164468
- 34. Freeman KG, Robotham AC, Parks OB, Abad L, Jacobs-Sera D, Lauer MJ, et al. Virion glycosylation influences mycobacteriophage immune recognition. Cell Host Microbe. 2023;31(7):1216-1231.e6. pmid:37329881
- 35. Paddison P, Abedon ST, Dressman HK, Gailbreath K, Tracy J, Mosser E, et al. The roles of the bacteriophage T4 r genes in lysis inhibition and fine-structure genetics: a new perspective. Genetics. 1998;148(4):1539–50. pmid:9560373
- 36. Radersma MD, Lathrop G, Moleakunnel KC, Harlow LA, Baker AE, Chen AJ, et al. Complete genome sequences of nine Rhodococcus equi phages. Microbiol Resour Announc. 2024;13(2):e0108823. pmid:38179906
- 37. Hutinet G, Kot W, Cui L, Hillebrand R, Balamkundu S, Gnanakalai S, et al. 7-Deazaguanine modifications protect phage DNA from host restriction systems. Nat Commun. 2019;10(1):5442. pmid:31784519
- 38. Fayet O, Prère MF. Programmed ribosomal −1 frameshifting as a tradition: The bacterial transposable elements of the IS3 family. In: Gesteland JFARF. Springer. 2010.
- 39. Delegan Y, Valentovich L, Petrikov K, Vetrova A, Akhremchuk A, Akimov V. Complete Genome Sequence of Rhodococcus erythropolis X5, a Psychrotrophic Hydrocarbon-Degrading Biosurfactant-Producing Bacterium. Microbiol Resour Announc. 2019;8(48):e01234-19. pmid:31776221
- 40. Khairy H, Meinert C, Wübbeler JH, Poehlein A, Daniel R, Voigt B, et al. Genome and Proteome Analysis of Rhodococcus erythropolis MI2: Elucidation of the 4,4´-Dithiodibutyric Acid Catabolism. PLoS One. 2016;11(12):e0167539. pmid:27977722
- 41. Strnad H, Patek M, Fousek J, Szokol J, Ulbrich P, Nesvera J, et al. Genome Sequence of Rhodococcus erythropolis Strain CCM2595, a Phenol Derivative-Degrading Bacterium. Genome Announc. 2014;2(2):e00208-14. pmid:24652983
- 42. Yoshida K, Kitagawa W, Ishiya K, Mitani Y, Nakashima N, Aburatani S, et al. Genome Sequence of Rhodococcus erythropolis Type Strain JCM 3201. Microbiol Resour Announc. 2019;8(14):e01730-18. pmid:30948473
- 43. Pope WH, Jacobs-Sera D, Russell DA, Rubin DHF, Kajee A, Msibi ZNP, et al. Genomics and proteomics of mycobacteriophage patience, an accidental tourist in the Mycobacterium neighborhood. mBio. 2014;5(6):e02145. pmid:25467442
- 44. Guerrero-Bustamante CA, Hatfull GF. Bacteriophage tRNA-dependent lysogeny: requirement of phage-encoded tRNA genes for establishment of lysogeny. mBio. 2024;15(2):e0326023. pmid:38236026
- 45. Ponce Reyes S, Park PJ, Kaluka D, Washington JM. Complete Genome Sequence of Rhodococcus erythropolis Phage Shuman. Microbiol Resour Announc. 2019;8(13):e00113-19. pmid:30923242
- 46. Russell DA, Hatfull GF. PhagesDB: the actinobacteriophage database. Bioinformatics. 2017;33(5):784–6. pmid:28365761
- 47. Gordon D, Green P. Consed: a graphical editor for next-generation sequencing. Bioinformatics. 2013;29(22):2936–7. pmid:23995391
- 48. Pope WH, Jacobs-Sera D. Annotation of Bacteriophage Genome Sequences Using DNA Master: An Overview. Methods Mol Biol. 2018;1681:217–29. pmid:29134598
- 49. Delcher AL, Harmon D, Kasif S, White O, Salzberg SL. Improved microbial gene identification with GLIMMER. Nucleic Acids Res. 1999;27(23):4636–41. pmid:10556321
- 50. Besemer J, Borodovsky M. GeneMark: web software for gene finding in prokaryotes, eukaryotes and viruses. Nucleic Acids Res. 2005;33(Web Server issue):W451-4. pmid:15980510
- 51. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. J Mol Biol. 1990;215(3):403–10. pmid:2231712
- 52. Laslett D, Canback B. ARAGORN, a program to detect tRNA genes and tmRNA genes in nucleotide sequences. Nucleic Acids Res. 2004;32(1):11–6. pmid:14704338
- 53. Chan PP, Lowe TM. tRNAscan-SE: Searching for tRNA Genes in Genomic Sequences. Methods Mol Biol. 2019;1962:1–14. pmid:31020551
- 54. Söding J, Biegert A, Lupas AN. The HHpred interactive server for protein homology detection and structure prediction. Nucleic Acids Res. 2005;33(Web Server issue):W244-8. pmid:15980461
- 55. Marchler-Bauer A, Anderson JB, Cherukuri PF, DeWeese-Scott C, Geer LY, Gwadz M, et al. CDD: a Conserved Domain Database for protein classification. Nucleic Acids Res. 2005;33(Database issue):D192-6. pmid:15608175
- 56. Krogh A, Larsson B, von Heijne G, Sonnhammer EL. Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes. J Mol Biol. 2001;305(3):567–80. pmid:11152613
- 57. Hallgren J, Tsirigos KD, Pedersen MD, Armenteros JJA, Marcatili P, Nielsen H, et al. DeepTMHMM predicts alpha and beta transmembrane proteins using deep neural networks. bioRxiv. 2022.
- 58. Huson DH, Bryant D. Application of phylogenetic networks in evolutionary studies. Mol Biol Evol. 2006;23(2):254–67. pmid:16221896
- 59. Cassier-Chauvat C, Poncelet M, Chauvat F. Three insertion sequences from the cyanobacterium Synechocystis PCC6803 support the occurrence of horizontal DNA transfer among bacteria. Gene. 1997;195(2):257–66. pmid:9305771
- 60. Weiss RB, Dunn DM, Atkins JF, Gesteland RF. Ribosomal frameshifting from -2 to +50 nucleotides. Prog Nucleic Acid Res Mol Biol. 1990;39:159–83. pmid:2247607
- 61. Chandler M, Fayet O. Translational frameshifting in the control of transposition in bacteria. Mol Microbiol. 1993;7(4):497–503. pmid:8384687
- 62. Sharma V, Prère M-F, Canal I, Firth AE, Atkins JF, Baranov PV, et al. Analysis of tetra- and hepta-nucleotides motifs promoting -1 ribosomal frameshifting in Escherichia coli. Nucleic Acids Res. 2014;42(11):7210–25. pmid:24875478