Phylogenetic Relationships and Evolution of the Genus Eganvirus (186-Type) Yersinia pestis Bacteriophages

Plague is an endemic infectious disease caused by Yersinia pestis. In this study, we isolated fourteen phages with similar sequence arrangements to phage 186; these phages exhibited different lytic abilities in Enterobacteriaceae strains. To illustrate the phylogenetic relationships and evolutionary relationships between previously designated 186-type phages, we analysed the complete sequences and important genes of the phages, including whole-genome average nucleotide identity (ANI) and collinearity comparison, evolutionary analysis of four conserved structural genes (V, T, R, and Q genes), and analysis of the regulatory genes (cI, apl, and cII) and integrase gene (int). Phylogenetic analysis revealed that thirteen of the newly isolated phages belong to the genus Eganvirus and one belongs to the genus Felsduovirus in the family Peduoviridae, and these Eganvirus phages can be roughly clustered into three subgroups. The topological relationships exhibited by the whole-genome and structural genes seemed similar and stable, while the regulatory genes presented different topological relationships with the structural genes, and these results indicated that there was some homologous recombination in the regulatory genes. These newly isolated 186-type phages were mostly isolated from dogs, suggesting that the resistance of Canidae to Y. pestis infection may be related to the wide distribution of phages with lytic capability.


Introduction
Plague, caused by the Gram-negative bacterial pathogen Yersinia pestis, is a notorious zoonotic disease.Plague primarily infects humans through flea bites, mainly causing bubonic plague with swollen lymph nodes.Other means of infection include contacting blood or inhaling aerosol droplets with Y. pestis, which can cause pneumonic plague [1].Currently, plague occurs in some countries, including China [2,3], the Democratic Republic of the Congo, Uganda, Madagascar, and the United States [4].
To date, public health measures and effective antibiotic treatments have significantly reduced plague infections worldwide.However, the disease remains ineradicable, and endemic natural plague foci persist across Africa, Asia, and the Americas [5].Multidrug resistance in Y. pestis strains, mediated by conjugative plasmids, has been reported in Madagascar [6,7].Furthermore, novel streptomycin resistance mechanisms, including a rpsL gene (ribosomal protein S12) mutation at amino acid 43 (K43R), have been identified in strains from China and Madagascar [8,9].Additionally, a case of plasmid-mediated Viruses 2024, 16, 748 2 of 15 doxycycline resistance was documented in a Y. pestis strain isolated from a rat in Madagascar in 1998 [10].
Bacteriophages were first explored for their ability to treat plague cases in 1925 when Dr. d'Herelle attempted to treat four bubonic plague cases using Y. pestis phage (YpsP-G) [11].In addition, phage therapy could be a new potential strategy for treating plague infection or as a biocontrol method for eliminating the endemics of animal plague in natural plague foci [12], such as the genus Gaprivervirus Y. pestis phages with broad host range characteristics, vB_YpeM_MHS112 (OP750247) and vB_YpeM_GMS130 (OP750248) [13].
In fact, phages shape the composition and evolution of bacterial communities in nature and are, therefore, important in ecosystem cycles [14].Phage communities affect the composition of infected bacteria through various mechanisms, including lysogenic conversion, transduction, and host gene disruption, when they integrate into the bacterial genome as prophages [14].
P2-like phages are a group of related temperate phages that are commonly found in γ-proteobacteria.Generally, the genome architecture of P2-like phages seems to be more conserved than that of majority phages from other families or classes [15].Structural genes are more conserved than transcriptional switches in P2-like phages, with the latter controlling the lytic versus the lysogenic growth cycle [15].Previous studies have distinguished P2-like phages into two major types, i.e., p2-type phages and 186-type phages [15][16][17].P2-type phages and 186-type phages are morphologically indistinguishable, both having an icosahedral capsid and a contractile tail.However, the genes and patterns of regulatory genes of the 186-type phages are quite different from those of p2-type phages, and the 186-type phages appear more complex and contain the additional gene cII [15].According to the latest classification of bacteriophages by the International Committee on Taxonomy of Viruses (ICTV), phage 186-type phages were found in the genus Eganvirus in subfamily Peduovirinae, while P2-type phages were found in the genus Peduovirus [18].
Our team isolated several wild-type Y.pestis phages, including previously called T4-like phages [12,19] and P2-like phages [20].Through genomic analysis, we identified 14 Y. pestis phages that are highly homologous to phage 186 in Yunnan plague natural foci.In this study, the morphological characteristics and host range of these new phages were observed, and we further conducted comparative genomic analysis, including wholegenome average nucleotide identity (ANI) and collinearity comparison.In addition, we performed a phylogenetic analysis of four capsid-related structural genes as in the previous literature [15], namely, the capsid scaffold gene (V), the major capsid precursor gene (T), the small terminase subunit gene (R), and the capsid completion gene (Q), which correspond to the structural genes O, N, M, and L in P2-type phages.We also compared the phylogenetic relationships of regulatory genes associated with 186-type phages, including the cI, apl, int, and cII genes, which are crucial for lysogeny/lysis of 186-type phages.

Isolation and Identification of Bacteriophages
Fourteen wild-type phages were isolated in 2016 and 2018 from the anal swabs of dogs (Pug, Chinese rural, and Pekingese) and the caecum of small rodents (Apodemus chevrieri) in two natural plague foci in Yunnan, China (Table 1).All samples were soaked in PBSglycerol (1:1, v/v), centrifuged at 9000× g for 10 min, and then filtered through a 0.22 µm pore size membrane filter (polyethersulfone membrane, hydrophilic, Millex ® -GP filter).The Y. pestis vaccine strain EV76 was used as an indicator host.The above filtrate was incubated with 200 µL of the EV76 strain in LB broth (OD600 ≈ 0.4) and shaken (220 rpm) at 28 • C for 24 h.Next, the phages were filtered through a 0.22 µm filter and added to 500 µL of Y. pestis EV76 culture until the liquid became clear.Single phage plaques were separated and purified by a double agar plaque assay according to the literature [13,20].The natural plague foci included F: Rattus flavipectus, which is in Yunnan-Guangdong-Fujian provinces; E: Apodemus chevrieri-Eothenomys miletus, which is in the highlands of northwestern Yunnan Province; and *: MLG42, which belongs to the genus Felsduovirus.

Transmission Electron Microscopy
The purified phage preparations were spotted onto 400 mesh carbon-coated grids, negatively stained with 2% phosphotungstic acid, and then visualised using a Hitachi HT7700 80-kV transmission electron microscope (Tokyo, Japan) with a Gatan 832.10 W CCD camera (Gatan, Pleasanton, CA, USA) operating in Gatan Digital Micrograph software [21].The diameters of the heads and lengths of the tails were determined from electron micrographs.At least three well-defined micrographs were used to measure the sizes of the phages, and the averaged measurements are listed in Table 1.

Host Range Analysis
The bacterial strains used for the host range assay are listed in Supplementary Table S1.A total of 114 Enterobacteriaceae strains (15 genera) were selected, including the genera Yersinia, Escherichia, Salmonella, Shigella, Cronobacter, Klebsiella, Enterobacter, Enterococcus, Cronobacter, Serratia, Proteus, Providencia, Morganella, Citrobacter, and Pantoea.A spot testing assay was used for host range analysis [13].Briefly, 400 µL of bacterial culture liquid was added to 50 • C molten LB semisolid agar, mixed and poured into the prepared 36 grids of nutrient solid plates to make a double-layer plate.After solidification of the top agar matrix, 4 µL of phage filtrate was dropped on top agar and incubated at 37 • C (for Enterobacteriaceae bacteria) or 28 • C (for EV76 and other Y. pestis strains) for 12-36 h to observe the plaque results.The host range was visualised through Tree Visualisation by One Table (TvBOT, https://www.chiplot.online/tvbot.html,v 2.6) (accessed on 22 December 2023), as illustrated in a heatmap [22].

Bacteriophage DNA Extraction and Amplification
Phages were propagated in 500 µL of LB medium supplemented with Y. pestis EV76 strain culture with shaking (220 rpm) overnight at 28 • C, and phage DNA was extracted using an ABigen corporation Lambda phage genomic DNA extraction kit (ABigen Corp., Beijing, China) according to previous methods [13].Soapnuke (v2.0.5) was used for quality control.

Phage Genome Sequencing, Assembly, Annotation, and Comparison
The whole genomes of fourteen 186-related phages were sequenced by a commercial sequencing company (MAGIGENE, Shenzhen, China).BWA (v0.7.17) software was used to align and clean the reads associated with the host genome to remove the host genomic sequence.Megahit software was used for de novo assembly (v1.1.2SCIME) [23].Meta-GeneMark (v3.38) software was used to predict the genes of contigs in the target genome (sequences with nucleic acid lengths less than 150 bp were filtered) [24,25].Two methods were used to annotate gene function, i.e., BLASTP (v2.9.0+) software was used to compare the gene protein sequence with the virus sequence in the UniProtKB/Swiss-Prot database (ViralZone, reviewed Proein, https://viralzone.expasy.org/)(accessed on 15 August 2023) to obtain functional information for the genes [26].Second, the putative gene function was predicted using the RAST tool (https://rast.nmpdr.org/,accessed on 22 August 2023) [27] and supplemented by analyses performed with PHASTER (https://phaster.ca/,accessed on 12 January 2024) [28,29] and the Galaxy web service (https://usegalaxy.org/,accessed on 12 January 2024) [30] to enhance the accuracy of the presumed gene function prediction.The genomic sequence comparisons among phages were conducted using Easyfig software [31].FastANI software was used to conduct pairwise comparisons of phage genomes at the nucleotide level by calculating their average nucleotide identity (ANI) [32].The values of the ANI multiplied by the coverage were used in homological analysis.

Phylogenetic Relationship Analysis
A total of 62 genomes of P2-like phages or prophages were used to analyse phylogenetic relationships in this study, including 14 newly isolated 186-related phages in this study (Table 1), P2-like phages, or prophages previously used in phylogenetic analysis in the literature [15], and all seven phage strains in the genus of Eganvirus (186, EtG, PsP3, SW9, BIS20, SI22, and SEN1) obtained from the National Center for Biotechnology Information (NCBI) database (Supplementary Table S2).Another three phages (vB_EcoM-613R3, vB_Sal_PHB48, 3625_26581) highly homologous to phage 186 were also included in this study.Our newly isolated phages were identified as 186-type phages that met the following two conditions: (1) classified as a member of the Peduovirinae subfamily; and (2) had four late structure genes of phage 186 (V, T, R, and Q genes) and complete regulatory region gene arrangements (cI, apl, cII, and int genes).In addition, to illustrate the phylogenetic relationships and population positions of these 186-type phages (prophages), the phylogenetic relationships of 15 Peduovirus phages (including phages the P2, L-413C, P2 Hy dis, Wφ, vB_YpM_22, vB_YpM_46, vB_YpM_50, etc.) were also investigated via genome and structure protein phylogenetic relationship analysis (Supplemental Table S2).
Phylogenetic trees were constructed using IQ-TREE (v2.2.2.6) software [33] and the maximum likelihood (ML) method [34] based on the inferred amino acid sequences from four structural genes as mentioned in the previous literature [15] (notes: V, T, R, and Q genes analysed together) and four regulatory genes (int, cI, apl, and cII genes).A phylogenetic tree of the full-length genome was constructed using the Bayesian inference (BI) method and OrthoFinder software [35].To elucidate phylogenetic relationships, we utilised PhyloSuite software (v1.2.2) [36], which employs the MAFFT (v7.313) [37] function for aligning protein sequences.The concatenated sequences of structural genes were then analysed phylogenetically using IQ-TREE software with default parameters (-m TEST -nt AUTO -bb 1000) [33].The constructed phylogenetic trees were then imported into tvBOT for further enhancement and visualisation [22].
The consistency of major clusters was verified through homogeneity tests, and branches with less than 50% support were collapsed.The stability of the phylogenetic tree was evaluated for robustness using 1000 bootstrap replicates, and the corresponding values were indicated on the branches.

Accession Numbers
The whole-genome sequences of the newly isolated 186-related phages in this study have been deposited in GenBank, and the detailed accession numbers can be found in Table 1.

Morphology of the Y. pestis Phages of the Genus Eganvirus
Electron microscopic observations indicate that the phages of the genus Eganvirus Y. pestis have a typical icosahedral head and long contractile tail.Thus, based on these morphologic characteristics, these phages have Caudoviricetes characteristics (Figure 1).These newly isolated Eganvirus phage genera exhibited roughly consistent head and tail sizes.The average head diameter was 71 ± 8 nm, and the average tail length was 156 ± 16 nm (Table 1).However, the size of phage MLG42 was notably greater than that of its counterpart (Figure 1).
The consistency of major clusters was verified through homogeneity tests, and branches with less than 50% support were collapsed.The stability of the phylogenetic tree was evaluated for robustness using 1000 bootstrap replicates, and the corresponding values were indicated on the branches.

Accession Numbers
The whole-genome sequences of the newly isolated 186-related phages in this study have been deposited in GenBank, and the detailed accession numbers can be found in Table 1.

Morphology of the Y. pestis Phages of the Genus Eganvirus
Electron microscopic observations indicate that the phages of the genus Eganvirus Y. pestis have a typical icosahedral head and long contractile tail.Thus, based on these morphologic characteristics, these phages have Caudoviricetes characteristics (Figure 1).These newly isolated Eganvirus phage genera exhibited roughly consistent head and tail sizes.The average head diameter was 71 ± 8 nm, and the average tail length was 156 ± 16 nm (Table 1).However, the size of phage MLG42 was notably greater than that of its counterpart (Figure 1).

Host Range Determination
Host range analysis was conducted for 13 of the 14 newly isolated Y. pestis phages (MDG94 was not recovered successfully) in this study.These phage isolates can infect twenty wild-type Y.pestis strains (biovars including Antiqua, Mediaevalis, Orientalis, and Microtus) in China.Host lysis experiments revealed that these phages presented different infecting host profiles; for example, MDG45 could not lyse any of the tested strains, whereas other phages were sensitive to at least one host strain (Figure 2, Supplemental Table S1).MLG42, MHG7, and MHG48 could completely lyse some Yersinia pseudotuberculosis, while MHG96, MHG101, MHG54, MHG7, MHG48, and GCS667 could incompletely lyse at least one Y.pseudotuberculosis; phage GCS667 incompletely lysed six of seven tested Y. pseudotuberculosis; phages MLG42, MHG7, and MHG48 could completely lyse at least one Yersinia enterocolitica, whereas MHG54 and GCS667 incompletely lysed Y. enterocolitica.In addition, phage MHG39 was the only phage that could lyse Yersinia kristensenii.

Host Range Determination
Host range analysis was conducted for 13 of the 14 newly isolated Y. pestis phages (MDG94 was not recovered successfully) in this study.These phage isolates can infect twenty wild-type Y.pestis strains (biovars including Antiqua, Mediaevalis, Orientalis, and Microtus) in China.Host lysis experiments revealed that these phages presented different infecting host profiles; for example, MDG45 could not lyse any of the tested strains, whereas other phages were sensitive to at least one host strain (Figure 2, Supplemental Table S1).MLG42, MHG7, and MHG48 could completely lyse some Yersinia pseudotuberculosis, while MHG96, MHG101, MHG54, MHG7, MHG48, and GCS667 could incompletely lyse at least one Y.pseudotuberculosis; phage GCS667 incompletely lysed six of seven tested Y. pseudotuberculosis; phages MLG42, MHG7, and MHG48 could completely lyse at least one Yersinia enterocolitica, whereas MHG54 and GCS667 incompletely lysed Y. enterocolitica.In addition, phage MHG39 was the only phage that could lyse Yersinia kristensenii.Regarding the lysing panels of Escherichia coli, phage MHG39 was particularly notable and could lyse five E. coli strains, followed by MHG48 lysing two strains, and MHG51, MHG101, MLG42, and MHG54 each lysing one strain; however, their lytic profiles were different.Both MHG39 and MHG54 could lyse Shigella flexneri, and MHG55, MHG75, and MHG39 could lyse Shigella boydii.Shigella sonnei could be lysed by more than half of the 186-related phages.Additionally, Cronobacter malonate-positive strains could be partially lysed by MHG54 and MHG48, and Enterobacter cloacae could be completely lysed by GCS667.Salmonella blegdam was lysed completely or partially by MHG39, MHG54 and MHG48.

Homology and Average Nucleotide Identity (ANI)
Except for phage MLG42, the genomic sequences of the newly isolated Y. pestis phages were homologous to those of the genus Eganvirus.The ANI of these Y. pestis phages ranged from 97.359%-99.99%.In total, the genus Eganvirus phages were considerably phylogenetically different from the genus Peduovirus phages, as phages previously called P2-like phages (value of ANI multiple coverage ranging from 25% to 36%) (Figure 3, Supplemental Table S3).Regarding the lysing panels of Escherichia coli, phage MHG39 was particularly notable and could lyse five E. coli strains, followed by MHG48 lysing two strains, and MHG51, MHG101, MLG42, and MHG54 each lysing one strain; however, their lytic profiles were different.Both MHG39 and MHG54 could lyse Shigella flexneri, and MHG55, MHG75, and MHG39 could lyse Shigella boydii.Shigella sonnei could be lysed by more than half of the 186related phages.Additionally, Cronobacter malonate-positive strains could be partially lysed by MHG54 and MHG48, and Enterobacter cloacae could be completely lysed by GCS667.Salmonella blegdam was lysed completely or partially by MHG39, MHG54 and MHG48.

Homology and Average Nucleotide Identity (ANI)
Except for phage MLG42, the genomic sequences of the newly isolated Y. pestis phages were homologous to those of the genus Eganvirus.The ANI of these Y. pestis phages ranged from 97.359%-99.99%.In total, the genus Eganvirus phages were considerably phylogenetically different from the genus Peduovirus phages, as phages previously called P2-like phages (value of ANI multiple coverage ranging from 25% to 36%) (Figure 3, Supplemental Table S3).

Genomic Collinearity Comparison
Collinearity analysis indicated that the genome sequences of phages from the newly isolated genus Eganvirus (MDG45 and GCS667 as representatives) were similar to those of E. coli phage 186 or ETG (Figure 4A,B) [38].These newly isolated Y. pestis phages presented highly similar gene arrangements; however, some detailed differences existed when comparing whole-genome sequences (Supplemental Figure S1).MHG101, MHG48, MHG54, MHG7, MHG39, and MHG51 exhibited almost the same gene arrangements, while MHG75, MHG96, MHG38, MDG45, and MDG94 exhibited similar arrangements of a second type; furthermore, GCS667 and MHG55 had more differences in their genomic sequences when compared to the other genomes (Supplemental Figure S1).

Genomic Collinearity Comparison
Collinearity analysis indicated that the genome sequences of phages from the newly isolated genus Eganvirus (MDG45 and GCS667 as representatives) were similar to those of E. coli phage 186 or ETG (Figure 4A,B) [38].These newly isolated Y. pestis phages presented highly similar gene arrangements; however, some detailed differences existed when comparing whole-genome sequences (Supplemental Figure S1).MHG101, MHG48, MHG54, MHG7, MHG39, and MHG51 exhibited almost the same gene arrangements, while MHG75, MHG96, MHG38, MDG45, and MDG94 exhibited similar arrangements of a second type; furthermore, GCS667 and MHG55 had more differences in their genomic sequences when compared to the other genomes (Supplemental Figure S1).

Phylogenetic Relationship between Genomic Proteins and Four Conserved Structural P
These 13 newly isolated Eganvirus phages clustered together with phages from er sequenced genera of Eganvirus (phage 186, EtG, PsP3, SW9, and BIS20), but it w vious that the phages of these genera could be roughly clustered into three subg based on their genomic sequences (Figure 5A).The phage MLG42 was clustered in genus Felsduovirus together with Fels-2, SopEφ, 3625_26581, and φSEN2 (Figure 5A P2-like phages possess conserved late structural genes.A total of 48 P2-like p or prophages, most of which were described in the previous literature [15,39], with 13 phages of the newly isolated genus Eganvirus Y. pestis (Table 1) and 1 pha the genus Felsduovirus Y. pestis (MLG42), were selected for the comparison of fou served structural proteins (V, T, R, and Q for 186-type phages, corresponding to O, and L for the genus Peduovirus).The phylogenetic relationships of these previously tified P2-like phages are illustrated in Figure 5B.One notable feature of these rel

Phylogenetic Relationship between Genomic Proteins and Four Conserved Structural Proteins
These 13 newly isolated Eganvirus phages clustered together with phages from other sequenced genera of Eganvirus (phage 186, EtG, PsP3, SW9, and BIS20), but it was obvious that the phages of these genera could be roughly clustered into three subgroups based on their genomic sequences (Figure 5A).The phage MLG42 was clustered into the genus Felsduovirus together with Fels-2, SopEφ, 3625_26581, and φSEN2 (Figure 5A).
P2-like phages possess conserved late structural genes.A total of 48 P2-like phages or prophages, most of which were described in the previous literature [15,39], along with 13 phages of the newly isolated genus Eganvirus Y. pestis (Table 1) and 1 phage of the genus Felsduovirus Y. pestis (MLG42), were selected for the comparison of four conserved structural proteins (V, T, R, and Q for 186-type phages, corresponding to O, N, M, and L for the genus Peduovirus).The phylogenetic relationships of these previously identified P2-like phages are illustrated in Figure 5B.One notable feature of these relationships was that the phylogenetic topology of the four structural genes (Figure 5B) was analogous to the ships was that the phylogenetic topology of the four structural genes (Figure 5B) was analogous to the topology of the whole genomic sequences (Figure 5A), indicating the exclusion of diversity and discrimination scales in their differentiation.

Regulatory Region Genes
A comparison of the topological relationships among the apl, cI, and int genes revealed distinct differences (Supplemental Figure S2).For example, one subgroup of Eganvirus phages (depicted in deep blue in Figure 6) appeared to be closer to phages in the genus Felsduovirus when phylogenetic relationships were based on the cI or int genes.Moreover, another subgroup 2 (shown in Figure 5) of Eganvirus phages (light blue in Supplemental Figure S2) seemed to have a closer relationship with the phage Felsduovirus.
The most significant feature of genus Peduovirus phages and genus Eganvirus phages was that the latter contains the additional gene cII.Excluding the diversity and discrimination scales in their differentiation, one interesting observation about the phylogenetic topology of the CII protein (Figure 6) was that it was unexpectedly analogous to the topology of the four structural proteins (Supplemental Figure S3).That is, phages belonging to the same genus also had identical sorting of these four constructed proteins (Figures 5 and 6).
Among the 13 newly isolated Eganvirus phages, the int, cI, and cII genes are highly homologous.However, surprisingly, the APL protein of GCS667 showed a different phylogenetic relationship with the other phages, as it clustered more closely with phage 186, EtG, and two prophages (φESP and ΦKPN).Corresponding clustering results based on genomic sequences also revealed a similar result (Figure 4 and Supplemental Figure S2).

Regulatory Region Genes
A comparison of the topological relationships among the apl, cI, and int genes revealed distinct differences (Supplemental Figure S2).For example, one subgroup of Eganvirus phages (depicted in deep blue in Figure 6) appeared to be closer to phages in the genus Felsduovirus when phylogenetic relationships were based on the cI or int genes.Moreover, another subgroup 2 (shown in Figure 5) of Eganvirus phages (light blue in Supplemental Figure S2) seemed to have a closer relationship with the phage Felsduovirus.
The most significant feature of genus Peduovirus phages and genus Eganvirus phages was that the latter contains the additional gene cII.Excluding the diversity and discrimination scales in their differentiation, one interesting observation about the phylogenetic topology of the CII protein (Figure 6) was that it was unexpectedly analogous to the topology of the four structural proteins (Supplemental Figure S3).That is, phages belonging to the same genus also had identical sorting of these four constructed proteins (Figures 5 and 6).
Among the 13 newly isolated Eganvirus phages, the int, cI, and cII genes are highly homologous.However, surprisingly, the APL protein of GCS667 showed a different phylogenetic relationship with the other phages, as it clustered more closely with phage 186, EtG, and two prophages (φESP and ΦKPN).Corresponding clustering results based on genomic sequences also revealed a similar result (Figure 4 and Supplemental Figure S2).

Diversity of Phages from the Genus Eganvirus
The newly isolated Y. pestis 186-type phages exhibit a high degree of genomic consistency.This consistency was not only observed in terms of genomic arrangement (99-100%), but also in high homology to conserved structural genes.One interesting phenomenon is that when the apl and cI genes were compared, the phage GCS667 clustered apart from other Eganvirus phages, and instead GCS667 clustered with the phages  Many phages can lyse Y. pestis, including lytic phages previously called Podoviridae phages (such as Pokrovskaya [40], A1122 [41], and Yep-phi [42]), as well as eight recently reported new phages with Podoviridae characteristics (YpEc56, YpEc56D, YpEc57, YpEe58, YpEc1, YpEc2, YpEc11, and YpYeO9) [43].Because several such phages have specific lytic effects on Y. pestis, some are routinely used for the diagnosis of this disease.However, some such phages also exhibit lytic effects on some Y. pseudotuberculosis strains [11].

Diversity of Phages from the Genus Eganvirus
The newly isolated Y. pestis 186-type phages exhibit a high degree of genomic consistency.This consistency was not only observed in terms of genomic arrangement (99-100%), but also in high homology to conserved structural genes.One interesting phenomenon is that when the apl and cI genes were compared, the phage GCS667 clustered apart from other Eganvirus phages, and instead GCS667 clustered with the phages EtG and φESP (prophage).In contrast GCS667 clustered with other newly isolated Eganvirus phages when the int, cII, and structural genes were compared (Figures 5 and 6, Supplemental Figure S2).
The total genus Eganvirus phages presented a certain degree of consistency (Supplemental Table S3), however, they exhibited clear divergence, even though the genomic sequence or the key regulatory genes could differentiate at least two to three subgroups (Figure 5A).This scenario indicated that there were some differences in the genus Eganvirus phages and genus Eganvirus phages that can be subdivided into more varieties in the future.

Phylogenetic Relationship between Regulatory Genes and Structural Genes
In this study, we analysed 14 newly isolated phages, 13 of which belong to the genus Eganvirus, while one (MLG42) belongs to the genus Felsduovirus.These phages were compared with other members of the genus Eganvirus and further compared with P2-like phages using previous methods [15] to illustrate their phylogenetic relationships based on four conserved structural genes or key regulatory genes.The phages in this comparison included the genera Entnonagintavirus, Felsduovirus, Longwoodvirus, Bielevirus, Hpunavirus, and Irtavirus, as well as the more common Peduovirus (only representative P2-like phages were included).
One obvious convergence among these 186-type phages could be observed when these phages were grouped together based on whole-genome sequences or four relatively conserved structural genes (V, T, R, and Q).The phylogenetic trees based on the two methods revealed a general concordance in the phylogenetic arrangement, with only subtle shifts observed.This scenario indicates that the general genomic diversity of these phages could also be represented by the variation in the four relatively structural genes.
The CII protein, a member of the helix-turn-helix family of transcriptional activators, is pivotal for establishing lysogeny by activating a promoter (pE) that initiates the expression of the CI immune repressor, which is essential for efficient lysogenic establishment [49].If we disregard the diversity discrimination scales in their differentiation, the phylogenetic topology of the CII protein was more closely analogous to that of the four conserved structural genes.
Despite these similarities, interesting phylogenetic discrepancies could be observed, such as the relationships illustrated based on the cI, apl, and int genes (Figure 6), which differ from those based on the evolution of structural genes or genomic diversity subtyping (Figure 5).Such phenomena indicate that the evolution of the cI, apl, and int genes differs from that of the structural genes, and this distinction is likely attributed to homologous recombination [15].Phage MLG42, which was isolated from a domestic dog anal swab, exhibited marked differences from the other phages.The distinct nature of the phage MLG42 can be reflected through many features, such as the phylogenetic relationships of regulatory and structural genes, attachment site attP, genome sequence, and even electron microscopy morphology (Figure 1).Therefore, we concluded that MLG42 is a new member of the genus Felsduovirus.

Phages Lysing Y. pestis Are Widely Distributed in the Environment
Yunnan is considered the origin of the Third World plague pandemic in the middle of the 19th century [50].All Eganvirus phages in this study were isolated from Yunnan Province.An overall investigation of phages in indicator animals (dogs and cats) was performed from 2015-2018 in Yunnan Province [51].The investigation covered 41 villages in 26 towns located in 10 counties or city districts and involved a total of 1003 dogs.Phages were isolated from anal swabs of these indicator animals, using Y. pestis EV76 as the host indicator.A total of 102 phages lysing Y. pestis were isolated from dogs, and these phages presented different polymorphisms based on electronical microscopic morphology; some were identified as P2-like phages (such as GMG73) or T4-like phages [13].These results indicate that phages that can lyse Y. pestis are widely distributed in the environment.In addition, most phages in this study were isolated from Menghai County.Menghai County belongs to the Rattus flavipectus plague focus, and one bubonic plague case with animal plague epizootics was confirmed in Menghai County in 2020.However, due to a lack of opportunities, the phage-carrying situation in dogs was not investigated until 2020.

The Significance of Phages Isolated from Dogs
In this study, many Eganvirus phages were isolated from dog samples.Canidae (such as wolves, dogs, coyotes, and foxes) are resistant to plague disease [52].Although coyotes and shepherd dogs are frequently infected with Y. pestis through the consumption of plague-infected prey, dogs or coyotes are thought to be resistant to Y. pestis infection, and these animals generate plague-protective F1 antibodies without developing disease symptoms [53].Indeed, dogs are believed to be indicator animals for surveillance of plague zoonosis in plague foci [54].
Recent research on LcrV (the needle cap protein of the Y. pestis type III secretion system) and the N-formylpeptide receptor (FPR1) revealed that dogs lack susceptibility to Y. pestis [55].FPR1 is a plague receptor on immune cells in both humans and mice, and Y. pestis binds to the FPR1 receptor to promote the translocation of bacterial effectors through a type III secretion system, which selectively destroys immune cells in humans.This mechanism enables Y. pestis to reproduce in the bloodstream and be transmitted to new human or mouse hosts through fleabites and mass propagation [55].Because Canidae have lost Fpr1 and the ability to respond to N-formyl peptides, it can be inferred that plague resistance in Canidae may be related to the loss of FPR1 in these species [55].
Many Yersinia phages have been isolated from dogs by our team, as mentioned above.These phages do not rely solely on Y. pestis as their host bacteria; instead, they can propagate through a broader range of Enterobacteriales species and form an ecological barrier against Y. pestis invasion.This scenario maybe could provide another explanation for the resistance of dogs or coyotes to Y. pestis infection, potentially related to the widespread distribution of phages with the ability to lyse Y. pestis.

Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/v16050748/s1, Figure S1: Phylogenetic relationships of the newly isolated phages identified in this study based on their genomic sequences.Figure S2: Phylogenetic relationships of the inferred amino acid sequences of the regulatory region genes (int, cI, apl, and cII).Figure S3: Phylogenetic relationships of the inferred amino acid sequences of the structural genes (V, T, R, and Q).Table S1: Strains used in the host range assay in this study.Table S2: 186-type phages and P2-like phages used in phylogenetic analysis in this study.Table S3: Identity and coverage multiplication values of Eganvirus vs. other P2-like Phages.

Figure 1 .
Figure 1.Transmission electron microscopy images of the morphology of the MHG48 (left) and MLG42 (right) phages.

Figure 1 .
Figure 1.Transmission electron microscopy images of the morphology of the MHG48 (left) and MLG42 (right) phages.

Figure 2 .
Figure 2. Host range of phages from the newly isolated genus Eganvirus Y. pestis phages.Notes: The bacteria used for host range tests are shown with different colours, and only the bacteria exhibiting lysis phenomena are listed on the vertical axis.The bacteria without lysis among the 114 host strains tested are not shown (detailed results are listed in Supplemental TableS1).Dark blue indicates complete lysis (clear circular plaques), light blue indicates incomplete lysis (faint and blurry circular lysis or dispersed lytic points), and colourless indicates no lysis.

Figure 2 .
Figure 2. Host range of phages from the newly isolated genus Eganvirus Y. pestis phages.Notes: The bacteria used for host range tests are shown with different colours, and only the bacteria exhibiting lysis phenomena are listed on the vertical axis.The bacteria without lysis among the 114 host strains tested are not shown (detailed results are listed in Supplemental TableS1).Dark blue indicates complete lysis (clear circular plaques), light blue indicates incomplete lysis (faint and blurry circular lysis or dispersed lytic points), and colourless indicates no lysis.

Figure 3 .
Figure 3. Homology and average nucleotide identity (ANI).Notes: The line length of classification on left do not reflect the discrimination degrees.

Figure 3 .
Figure 3. Homology and average nucleotide identity (ANI).Notes: The line length of classification on left do not reflect the discrimination degrees.
whole genomic sequences (Figure5A), indicating the exclusion of diversity and discrimination scales in their differentiation.Viruses 2024, 16, x FOR PEER REVIEW 9 of 15

Figure 5 .
Figure 5. Phylogenetic relationship of genomic proteins (A) and four conserved structural proteins (B).

Figure 5 .
Figure 5. Phylogenetic relationship of genomic proteins (A) and four conserved structural proteins (B).

Figure 6 .
Figure 6.Phylogenetic relationship of the inferred amino acid sequences of the cII gene.

Figure 6 .
Figure 6.Phylogenetic relationship of the inferred amino acid sequences of the cII gene.

Table 1 .
Newly isolated 186-type phage information in this study.