The Brucella abortus A19 vaccine strain was used as the indicator strain for phage isolation. This strain was obtained from the Yunnan Institute of Endemic Disease Control and Prevention (YIEDC). Thirty clinical isolates of Brucella were collected from Kunming, Yuxi, and Qujing cities in Yunnan Province, all bacterial strains were preserved at the YIEDC. The Brucella abortus 104M vaccine strain, the Brucella melitensis M5 vaccine strain, and the diagnostic phage Tb were obtained from the Chinese Center for Disease Control and Prevention (CCDC). For all experiments, Brucella was cultured on Brucella agar (BD, USA) and propagated in Brucella broth (BD, USA). The strains were stored at −80°C in 25% glycerol and grown at 37°C under standard aerobic incubation conditions.

The collected sheep fecal samples were soaked in SM buffer (5.8 g/L NaCl, 2.0 g/L MgSO₄, 50 ml/L of 1 M Tris, pH 7.5, and 5 ml/L of presterilized 2% gelatin) at room temperature for 24 h. For phage enrichment, 5 ml of the sample supernatant was mixed with 500 μl of the A19 vaccine strain and 10 ml of Brucella broth, followed by continuous shaking at 220 rpm at 37°C for 48 h. The mixture was then centrifuged at 6,000g for 10 min and filtered through a 0.22 μm filter to obtain the phage suspension. The A19 strain was mixed with the phage suspension and plated on a double-layer agar plate. After incubation at 37°C for 24–48 h, individual plaques were picked from the plate and propagated. The double-layer agar plate method was repeated three times to obtain a purified phage suspension.

The purified phages were placed onto 400 mesh carbon-coated grids and negatively stained with 2% phosphotungstic acid (pH 6.5). The morphology of the phages was then observed using a Hitachi HT7700 transmission electron microscope. The dimensions of three viral particles of each phage were measured, and the measurements were averaged.

The phage and host bacterial suspensions were mixed at specific MOIs (0.0001, 0.001, 0.01, 0.1, 1, and 10) and incubated at 37°C with shaking at 220 rpm for 24 h. The mixture was then passed through a 0.22 μm filter to remove bacteria. The phage titer was determined using the double-layer agar method, and the MOI that produced the highest phage titer was considered the optimal MOI.

The one-step growth curve of phage Y17 was generated according to a previously described method with minor modifications (15). The host bacteria and phage were mixed to achieve an MOI of 0.001, and the mixture was incubated at 37°C for 15 min. Afterward, the mixture was centrifuged at 10,000g for 1 min, and the pellet was resuspended in 1 ml of Brucella broth. The supernatant was discarded, and the pellet was washed 2–3 times to remove any unadsorbed phages. The pellet was then resuspended in 8 ml of Brucella broth and incubated at 37°C with shaking at 220 rpm for 2 h. At 10-min intervals, 500 μL samples were collected and immediately centrifuged at 10,000g for 1 min to pellet the cells, after which the phage titer in the supernatant was determined. The burst size was calculated as the ratio of the final phage titer to the number of initial bacterial cells infected (16).

The stability of the phage was tested under various conditions by treating a 10¹⁰ pfu/ml phage suspension for 1 h at different pH levels (0–14), UV irradiation (15 cm from the UV lamp, samples were taken every 5 min during exposure), temperatures (4, 28, 37, 42, 50, 60, 70, and 80°C), and ethanol concentrations (10%, 25%, 50%, 75%, and 95%). After treatment, the phage titer was determined via the double-layer agar method, followed by incubation at 37°C for 24–48 h.

In this study, the A19, 104M, and M5 vaccine strains, along with 30 clinical isolates of B. melitensis (as detailed in Table 1), were selected as the experimental strains. The spot test method was used to determine the host range of phages Y17 and Tb, and the lytic effect was evaluated after incubation at 28 or 37°C for 48 h.

Phage genomic DNA was extracted using the Phage Genomic DNA Extraction Kit (Abigen Corp, Beijing, China). The DNA was fragmented to the desired length, and an “A” base was added to the 3′ ends to enable ligation with adapters containing a complementary “T” base. PCR amplification of adapter-ligated DNA fragments was performed to construct the library, which was then sequenced using the Illumina NovaSeq X Plus platform. Quality control was conducted using Soapnuke (v2.0.5) to generate high-quality clean reads (17). BWA (v0.7.17) aligned reads to the host genome and filtered out host-derived sequences (18). De novo assembly was conducted with Megahit (v1.1.2 SCIME) (19). Circular genomes were identified using ccfind (v1.4.5), with overlapping terminal sequences trimmed.

The putative genes in the phage genome were identified using the RAST server (20). The whole genome sequence of the phage was aligned using NCBI BLAST+ software (v 2.16.0) to further predict the function of each putative gene. The genome map was generated with SnapGene (v6.0.2). PhageLeads was used to check the genome for temperate markers, antibiotic resistance genes, and virulence genes (21). FastANI software was used to conduct pairwise comparisons of phage genomes at the nucleotide level by calculating their average nucleotide identity (ANI) (22). Comparative genomic maps were constructed using EasyFig (v2.2.5) (23). A whole-genome phylogenetic tree was constructed with OrthoFinder (24), and a protein phylogenetic tree was constructed using MEGA 11. Both trees were constructed using the maximum likelihood method. The initial multiple sequence alignment of amino acids was generated using MEGA, and the results were subsequently visualized and analyzed with ESPript 3.0 (25).

Brucellaphage Y17, isolated from sheep fecal samples in Yunnan Province, forms clear plaques measuring 3–6 mm in diameter on the A19 vaccine strain in double-layer agar (Figure 1A). Transmission electron microscopy (TEM) revealed that phage Y17 was composed of an icosahedral head measuring 48.1 ± 2 nm and a short, noncontracted tail measuring 10.8 ± 1 nm (Figure 1B). Phage Y17 exhibited the highest infection efficiency at an MOI of 0.001 (Figure 1C). The one-step growth curve showed that the efficacy of phage Y17 significantly increased after 40 min and stabilized at 80 min, with an estimated burst size of ~187 pfu/cell (Figure 1D).

We tested the host ranges of the phages Y17 and Tb in 30 clinical isolates of B. melitensis and the M5 and 104M vaccine strains. Both Y17 and Tb demonstrated significant lytic activity against B. abortus. However, Tb exhibited little lytic activity against B. melitensis. In contrast, Y17 showed some lytic activity against most B. melitensis strains, but clear plaques were observed in only a few strains. Additionally, the lytic effect of Y17 was greater at 28°C than at 37°C. These findings indicate that these brucellaphages exhibit host specificity (Table 1).

After 1 h of UV irradiation, the titer of phage Y17 gradually decreased but retained significant activity (Figure 2A). The optimal pH for Y17 was 7. Its titer remained high and relatively stable within the pH range of 3 to 11. Phage activity was completely lost only at pH ≤ 2 or ≥12, indicating a broad pH tolerance (Figure 2B). Phage Y17 retained stable activity below 50°C, but was completely inactivated at 80°C (Figure 2C). Exposure of Y17 to various ethanol concentrations for 1 h did not cause complete inactivation, suggesting that Y17 is a non-enveloped phage (Figure 2D).

The Y17 genome comprises a 38,025 bp circular double-stranded DNA, and no tRNA genes were predicted. The GC content is 48.2%, lower than the 57% observed in its host Brucella strains (26). An online whole-genome BLASTn search revealed that phage Y17 has over 99% coverage and identity with all other lytic brucellaphages, indicating a close relationship between Y17 and other lytic brucellaphages. The RAST Server predicted a total of 56 ORFs in the Y17 genome. Among them, 31 ORFs are located on the positive strand, primarily distributed in the beginning and middle regions. Meanwhile, 25 ORFs are on the negative strand, concentrated toward the end of the genome. Each ORF of Y17 was analyzed using NCBI BLAST+ software (v 2.16.0) against all proteins of other lytic brucellaphages recorded in NCBI. The analysis predicted 26 functional proteins, while the remaining 30 were identified as hypothetical proteins. Notably, one predicted gene product of unknown function exhibited no homology to any published sequences in the database (Figure 3; Supplementary Tables S1, S2). Additionally, PhageLeads was used to screen the genome and identified no genes associated with the temperate life cycle, antibiotic resistance, or bacterial virulence. These findings highlight the safety and therapeutic potential of Y17, supporting its application in clinical and environmental settings (21).

The 26 predicted functional proteins were categorized into five functional modules: metabolism and regulatory, DNA packaging, structure, host lysis, and DNA replication and modification. The metabolism and regulation-related proteins included GcrA-like proteins (ORF4 and ORF5) and N-acetyltransferase (ORF21). Although the specific functions of these proteins remain unknown, they may play a role in genome expression and the transcriptional regulation of the host bacterium during infection (27, 28). Several proteins involved in DNA packaging work together to ensure that the DNA is efficiently and accurately packaged into the viral capsid (29). The DNA packaging proteins in Y17 include the small terminase subunit (ORF6), the large terminase subunit (ORF7), and the portal protein (ORF11). The morphogenesis-related proteins include the capsid protein (ORF16), structural proteins (ORF15, ORF17, and ORF22), tail protein (ORF19), head protein (ORF25), and tail collar protein (ORF26). The tail collar protein is likely involved in the host specificity of the phage (30, 31). Amidase (ORF20), peptidoglycan hydrolase (ORF24), and endolysin (ORF28) are likely involved in host lysis. Holin was not identified in the lysis module. The DNA replication and modification module is located at the end of the genome and include the following: DNA methyltransferase (ORF40 and ORF41), DUF6378 domain-containing protein (ORF46), type III restriction endonuclease (ORF47), DEAD/DEAH box helicase (ORF48), DNA-binding protein (ORF49), DNA-PolB associated exonuclease (ORF50), single strand DNA binding protein (ORF52), Sak4-like ssDNA annealing protein (ORF54), and bifunctional DNA primase/polymerase (ORF56).

The ANI heatmap of Y17 and all lytic brucellaphages in NCBI (a total of 21 phages) is shown in Figure 4. There is a high level of genomic similarity among all brucellaphages, with ANI values ranging from 99.32% to 100%, as calculated using FastANI (Supplementary Table S3). This indicates a high degree of genomic conservation among these phages. Y17 exhibited the lowest overall nucleotide coverage, with the majority of coverage values ranging from 0.85 to 0.92, suggesting potential divergence in genome. The highest coverage of 1 was exclusively observed with Bk2, likely due to a close evolutionary relationship (Supplementary Table S4).

We selected phage Y17 and all lytic brucellaphages for multiple collinearity comparison analyses. The comparisons were divided into two groups based on phage genome similarity: genomes smaller than 40 kb (Figure 5A) and genomes larger than 40 kb (Figure 5B).

In phages with genome lengths smaller than 40 kb, collinearity analysis revealed complete genome collinearity among Bk, Pr, R/C, S708, Wb, EF4, and Bk2. Although Y17 lacked a portion of the terminal segment, the rest of its genome also displayed complete collinearity with the others. Additionally, except for the last five ORFs of Bk2, which showed reverse homology with other phage genomes, the genomic arrangement of the remaining phages was consistent (Figure 5A).

For phages with genome lengths larger than 40 kb, collinearity analysis revealed complete genome collinearity among Tb, Fz, V_19, 02_19, 1066_19, 110_19, 11sa_19, 141_19, 177_19, 281_19, 544_19, F1, Fi, and Iz. These phages contained three major insertions relative to Y17: a partial fragment of ORF21 (amidase), ORF28 (hypothetical protein), ORF29 (carbohydrate-binding protein), and a portion of the terminal segment. Meanwhile, F1, Fi, and Iz showed reverse homology in their terminal genome segments, while the genomic arrangement of other phages remained consistent (Figure 5B).

The phylogenetic tree based on the whole genomes of lytic brucellaphages reveals two distinct clades. EF4 forms a separate clade, indicating a more distant evolutionary relationship with other phages. Y17 clusters closely with phages Iz, Bk2, S708, Wb, R/C, Pr, and Bk, indicating a close evolutionary relationship (Figure 6A).

To gain deeper insights into the evolutionary relationships of phages at different protein levels, we constructed phylogenetic trees using the amino acid sequences of the capsid protein (Figure 6B), tail collar protein (Figure 7A) and amidase to explore phage evolution. Phages with genomes smaller than 40 kb have shorter amidase sequences than those with larger genomes. Therefore, they were divided into two groups when constructing the phylogenetic tree based on amidase sequence similarity (Figures 7B, C). We excluded phage EF4 from the analysis due to its significant sequence divergence. Phage EF4 was excluded from the analysis due to its significant sequence divergence. Multiple sequence alignment of phage capsid proteins revealed high conservation, with only three variable sites. Notably, only phage Y17 exhibited amino acid variation at position 190, while all other phages remained conserved at this site (Figure 6B). The topological structures of the phylogenetic trees for the capsid protein, tail collar protein, and amidase revealed distinct variations. Among these, the evolutionary relationships of the capsid protein were the most conserved, whereas the tail collar protein displayed greater genetic diversity than the other two proteins. Notably, Y17 occupied a unique position in all three phylogenetic trees, suggesting its distinct adaptations to varying environmental pressures.