Surface seawater (150 L) was collected from an oligotrophic ocean basin site MS-29 (12°0.0′ N, 116°0.0′ E) with a bottom depth of 4,079 m (Figure 1A). The collected seawater was filtered sequentially through 3-μm and 0.22-μm polycarbonate membranes (Isopore™, Millipore). The filtrate was concentrated by tangential flow filtration using a 50 kDa cartridge (Pellicon® XL Cassette, Biomax® 50 kDa; polyethersulfone, Millipore, United States) to increase the viral concentration by 300 times the initial concentration. The viral concentrate was stored at 4°C in the dark.

The host of the cyanophages is Synechococcus sp. strain MW02 (NCBI accession number KP113680) belonging to clade IX, which belongs to clade IX in subcluster 5.1 and was originally isolated from the Hong Kong coastal waters impacted by various water masses (Xia et al., 2015). This clade appears to have a high tolerance to different environmental conditions and is capable of adapting to diverse ecological niches (Xia et al., 2015, 2021). The algal culture was grown in conical flasks with f/2 seawater medium under a constant illumination of approximately 25 μmol m⁻² s⁻¹ at 25°C in a 12-h/12-h light–dark cycle. Liquid infection was used to isolate the cyanophage, where aliquots of virus concentrate were added to the exponentially growing host culture in a ratio of 1:9. Phage-host suspensions were incubated for about 1 week until lysed host cells were observed based on the color and turbidity of the lysate. A separate control group was set up with pure algal culture.

The cyanophage obtained in this study was purified using a serial dilution method followed by sucrose density gradient centrifugation (Jiang et al., 2020). To begin with, the lysate was diluted at a series levels (10-fold dilution over 7 orders of magnitude) using sodium chloride-magnesium sulfate (SM) buffer (100 mM NaCl, 50 mM Tris, 10 mM MgSO₄, and 0.01% gelatin, pH 7.5). The infectivity of each dilution was tested, and the most diluted sample that induced lysis was selected for next round of continuous dilution and infection assay using the same method. After three rounds of purification, a pure lysate containing a single phage strain was obtained theoretically (Wilhelm et al., 2006). The obtained cyanophage was concentrated using high recovery centrifugal filter units (Amicon® Ultra-15, MWCO 30 kDa, Millipore; Chénard and Suttle, 2008).

The concentrated phage suspensions were further purified by sucrose density gradient centrifugation. Four sucrose gradients of 20%, 30%, 40%, and 50% (w/v) were prepared, and the suspensions were centrifuged at 110,000 ×g for 2.5 h at 4°C in a Beckman-Coulter ultra-high-speed centrifuge. A visible band between the 40 and 50% gradient was carefully extracted using a syringe with a needle and diluted in TE buffer. The dilution was then centrifuged at 110,000 ×g at 4°C for 3 h to obtain the concentrated cyanophages. The purified phage precipitates were resuspended in SM buffer and stored at 4°C.

The purified phage suspension (20 μL) was placed onto a 200-mesh copper grid. A drop of 1% (w/v) phosphotungstic acid (pH 7.2) was added onto the grid, covering the phage particles, and allowed to stain for 10 min (as described by Wilson et al., 1993). After staining, the excess stain was blotted off gently with filter paper. The stained phage particles on the grid were then observed using a transmission electron microscope (JEOL JEM-1200EX, Japan) operating at 100 kV.

To understand the growth kinetics of Nanhaivirus ms29 in its host, a one-step growth curve was conducted. Briefly, exponentially growing Synechococcus sp. strain MW02 were inoculated in medium. Cyanophage suspension was added to the host culture at a multiplicity of infection (MOI) of 0.1 to increase the chance of infection for the host cells during the first infection cycle. The samples were collected at various time points, including 0, 1, 2, 5, 8, 11, 14, 17, and 20 h after infection. The abundance of phage and host was measured using flow cytometry (McSharry, 2000).

The infectivity of Nanhaivirus ms29 was tested using nine Synechococcus strains, including Synechococcus WH7803, WH8102, MW02, MW03, LTWRed, LTWGreen, PSHK05, CCMP1333, and PCC7002. The cyanophage solution was added to each host Synechococcus culture in logarithmic growth phase at a volume ratio of 1:9, in triplicates. The cyanophage solution was replaced by the medium in the control group. The mixtures were incubated under the same conditions described above. Cell lysis was monitored and compared in the control and cyanophage solution groups every day for 2 weeks to examine the infectivity.

Phage genomic DNA was extracted using the TIANamp Virus DNA Kit (TIANGEN) and sequenced using Illumina Miseq 2 × 300 paired-end sequence method. Raw reads were cleaned up by trimming adaptor sequences and removing low-quality reads. Reads containing >40% low-quality bases (mass value ≤ 20), >10% N content, and overlap with the adapter for >15 bp with less than 3 mismatches, were removed. Clean reads were then assembled into a single contig using Velvet (Version 1.2.07). The GapCloser (v1.12) was applied to fill the gap in preliminary assembly results and the fragments shorter than 500 bp were eliminated. The complete genome of Nanhaivirus ms29 has been deposited in the GenBank database under the accession number OP807803. The prediction of the hypothetical open reading frames (ORFs) was performed with Prodigal (version 2.6.3; Hyatt et al., 2010). The prediction of the transfer RNA (tRNA) genes sequences was performed using the RAST online server¹ (Brettin et al., 2015). After that, the translated amino acid sequences were scanned for homologs by BLASTp (version 2.9.0) in Non-Redundant (nr) protein database² of the GenBank. The protein domains were predicted and analyzed by Pfam database³ (Hofle and Brettar, 1996), and Kyoto Encyclopedia of Genes and Genomes (KEGG) database.⁴ The genome information of all Kyanoviridae members was downloaded from the NCBI database, and the same steps as above were used to annotate the genome, the content of AMGs in Nanhaivirus ms29 and member genomes was determined.

To classify Nanhaivirus ms29, 57 isolated reference genome sequences of the Kyanoviridae were downloaded from the NCBI Virus database, and 56 complete genome sequences were selected for analysis. A proteome clustering tree based on genome-wide sequence similarities computed by tBLASTx was generated with the genomes of Nanhaivirus ms29 and the other 56 selected cyanophages using ViPTree online server (Nishimura et al., 2017). The conserved portal protein sequences were used to construct a single gene phylogenetic tree (Zhong et al., 2002; Weinheimer and Aylward, 2022). These specific amino acid sequences were aligned with MEGA (version 7.0.18) and trimmed with TrimAl (version 1.4.15). The maximum-likelihood (ML) phylogenetic tree was constructed using IQ-tree (multicore version 2.0.3), and the bootstrap values were based on 1,000 replicates (Nguyen et al., 2015). The phylogenetic tree was visualized with iTOL.⁵

To determine the relative abundances of Nanhaivirus ms29, RPKM (Reads Per Kilobase per Million mapped reads) was calculated using CoverM (v0.6.0).⁶ The Nanhaivirus ms29 contigs were mapped to quality-controlled reads from the 154 Global Ocean Viromes (GOV 2.0) dataset using the minimap2 (v2.17) program (Li et al., 2010, 2021). To compare the abundance distribution of this novel cyanophage with other T4-like cyanophages in global oceans, 45 cyanophage sequences from different cyanophage genera in Kyanoviridae were selected as references. The relative abundance of Nanhaivirus ms29 was compared among five marine viral ecological zones (VEZs) defined by the Global Ocean Viromes (GOV 2.0) dataset: Arctic (ARC), Antarctic (ANT), temperate and tropical epipelagic (EPI), temperate and tropical mesopelagic (MES), and bathypelagic (BATHY; Gregory et al., 2019). To map the global distribution of Synechococcus phage Nanhaivirus ms29, the relative abundance was log₁₀(x + 1) transformed and visualized using a heatmap.

Transmission electron microscopy examination showed that Nanhaivirus ms29 exhibited a T4-like cyanophage with an icosahedral head (approximately 89.945 nm in diameter) and a long straight contractile tail (length 124.166 nm; Figure 1B). The one-step growth curve showed that Nanhaivirus ms29 had an adsorption period of 1–2 h, followed by a latent period of 2–5 h (Figure 1C). The first peak release time occurred approximately 17 h after inoculation. The infection cycle of this cyanophage is similar to that of the cyanophage Ormenosvirus syn9 (Doron et al., 2016), which is its closest relative on the protein taxonomic tree. Host specificity test showed that Nanhaivirus ms29 could not infect the other 9 strains of Synechococcus.

In the new virus classification system proposed by ICTV in October 2020, T4-like cyanophages were grouped under the family Kyanoviridae. So far, this family has been classified into 45 genera and 56 species. However, it appears that this proposal has not gained widespread adoption and many studies still uses the previous classification method. Here, we used genome sequences of Nanhaivirus ms29 and other 56 representative Kyanoviridae isolates to perform phylogenetic analysis. Newly identified cyanophages were usually classified using phylogenetic trees constructed from viral marker genes such as portal proteins, major capsid proteins, and large subunits of terminases (Dion et al., 2020). Here, we used the portal-protein-encoding homolog, gene 20 (g20), to construct a single-gene phylogenetic tree of 57 phages (Figure 2A). This approach has been previously used to classify T4-like phages (Zhong et al., 2002). Based on the results of the phylogenetic tree, the 57 cyanophages strains were classified into three clusters: Cluster I, Cluster II and Cluster III. The Nanhaivirus ms29 is clustered separately in Cluster III, forming a distinct branch. To further explore the phylogenetic relationship between Nanhaivirus ms29 and the other T4-like cyanophages and evaluate their taxonomic position, we constructed a genome-wide proteomic tree (Figure 2B). The result showed that Ormenosvirus syn9, Shandvirus sh35 and Shandvirus sb64 were most closely related to Nanhaivirus ms29 in the proteome tree. Shandvirus sh35 and Shandvirus sb64 infect Synechococcus sp. WH8102 and were isolated from the Yellow and Bohai estuaries in China (You et al., 2019), while Ormenosvirus syn9 was isolated from Woods Hole Harbor in the United States and also hosted by Synechococcus sp. WH8102 (Weigele et al., 2007). Nanhaivirus ms29 is the first cyanophage isolated from the open ocean to be included in this little clade and to infect Synechococcus MW02. According to the nucleotide homology heat map of 57 cyanophages, Shandvirus sh35 had the highest nucleotide identity with Nanhaivirus ms29, but the observed similarity was only 54.7% (Figure 2C). Therefore, according to the latest genus classification criteria proposed by ICTV (nucleotide homology with existing members < 70%), we proposed that the newly discovered Synechococcus phage represents a new cyanophage genus in Kyanoviridae and named it Nanhaivirus. The new genus Nanhaivirus has been submitted to ICTV.

The genome of Nanhaivirus ms29 is a linear double-stranded DNA of 178,866 bp with a GC content of 42.5%. The GC content of most Synechococcus phages genomes typically ranges from 35.4% to 51.7% (Jiang et al., 2020). It has been shown that organisms such as bacteria, phages, and plasmids, which are dependent on parasitism for survival, tend to have low G + C content in their genomes (Rocha and Danchin, 2002), probably due to differences in the cost of relevant metabolites in cells and the limited availability of G and C relative to A and T/U (Rocha and Danchin, 2002). The Nanhaivirus ms29 contains 217 ORFs and six different tRNA types that encode for 5 amino acids (Ser, Arg, Val, Pro, Gly). After comparison in the NCBI non-redundant (NR) protein database, Pfam protein database, and KEGG protein database, 67 out of 217 ORFs (30.87%) matched with homologs of other known functional proteins, while the remaining 150 ORFs (69.13%) were predicted to be hypothetical proteins due to the lack of relevant genetic information (Figure 3). The predicted functions of these 67 ORFs included structuring (18 ORFs), DNA replication and regulation (16 ORFs), auxiliary metabolic genes (AMGs, 30 ORFs), and packaging proteins (3 ORFs).

The structural proteins of Nanhaivirus ms29 include the neck protein (ORF 50, 51), prohead core protein (ORF 72), baseplate protein (ORF 40, 41, 43, 45, 176, 180, 188), tail tube protein (ORF 69, 150), tail fiber protein (ORF 201), phage tail sheath protein (ORF 52, 68), head completion protein (ORF 178), phage tail completion protein (ORF 75), and the major capsid proteins (ORF 74). The large and small terminase subunits (ORF 67 and ORF 53) plus the portal protein (ORF 70) belong to the packaging module of Nanhaivirus ms29 (Fujisawa and Morita, 1997). The DNA replication and regulation module consist of 16 ORFs, including four DNA polymerase-associated proteins (ORF 93, 94, 98, 112), DNA methyltransferase (ORF 6, 112), DNA primase (ORF 114, 131), DNA helicase (ORF 29, 170), DNA binding protein (ORF 77, 172), transcriptional regulator (ORF 99, 168) and RNA polymerase (ORF 81), nuclease (ORF 167). These genes, which are directly involved in DNA replication and transcription, are ubiquitous in most living organisms (Stubbe, 2000).

Autonomous replication of viruses inside host cells leads to the expression of viral genes, including AMGs, which are associated with a variety of physiological activities in virus-infected hosts, including photosynthesis (Mann et al., 2003), carbon metabolism (Thompson et al., 2011), and nucleotide metabolism (Dwivedi et al., 2013). Recently, Class II AMGs, which do not exist in the KEGG database or lack explicit annotations of metabolic pathways, were also suggested to be important in cyanophage function (Hurwitz and U’Ren, 2016). In the genome of Nanhaivirus ms29, 30 AMGs were identified, including 26 Class I AMGs and 4 Class II AMGs (Figure 4). AMGs involved in photosynthesis (i.e., petE, hli, psbA, psbD, cpeT) may play an important role in maintaining the photosynthetic activity of infected host cells, providing energy for phage replication and increasing phage fitness (Lindell et al., 2005). Genes related to carbon metabolism included CP12, gnd, zwf, and talA. Previous studies have demonstrated that phage-encoded CP12 and talA can redirect the energy flow of ATP and NADPH produced during photosynthesis away from the Calvin cycle toward the biosynthesis of dNTPs for phage replication during host infection (Thompson et al., 2011). AMGs related to phosphate regulation (pstS, phoH) may involve in phosphate uptake and starvation. Nucleotide metabolism (MazG, nrdA, nrdB, thyX, pyrBI, cobS) are genes related to genome replication and transcription in cyanophages. For example, the phage-encoded MazG was proposed to regulate the cellular level of ppGpp and therefore affect transcription and translation in the host and extend the period of cell survival under the stress of phage infection (Clokie and Mann, 2006). In addition, Class II AMGs (ORF 78, 115, 146, 208, 213, 214, 215, 216, 2OG-Fe(II)_Oxy, ORF 88, 160 grxA, ORF 107 hsp., ORF 169 prnA) were also identified in genome of Nanhaivirus ms29. For example, the 2OG-Fe(II) oxygenase superfamily proteins have the highest copy number among all AMGs in the genome of Nanhaivirus ms29, which may involve in performing multiple functions (van Staalduinen and Jia, 2014).

To compare the distribution of different AMGs present in the genomes of representative T4-like cyanophages, a quantitative heatmap was constructed and the cyanophages were clustered based on their proteome (Figure 4). Among the 57 phages compared, the genome of Nanhaivirus ms29 contains a higher number of AMGs, exceeds approximately 85% of the phages analyzed. It has been suggested that the AMG repertoire of virus is correlated with its isolated environment (coastal vs. open ocean; Crummett et al., 2016). The presence of relatively large number of AMGs in the genome of Nanhaivirus ms29, which was isolated from open ocean, may indicate its coping strategies and adaptive evolution toward oligotrophic environments.

In addition, we observed that that the clade composed of Nanhaivirus ms29 and three other cyanophages exhibit a similar AMG composition, with only minor differences for some low-frequency AMGs among the clades, such as gnd, zwf, pyrBI, and pstS. To acquire phosphorus, cyanobacteria typically use an ABC-type phosphate transporter, consisting of a high affinity phosphate-binding protein (pstS) and two transmembrane proteins (pstC and pstA; Kamennaya et al., 2020). Studies have shown that the pstS gene in cyanophages can be effectively expressed during host infection and improve the phosphate uptake rate of the host after infection (Zhao et al., 2022), which may contribute to adaptation of the cyanophage to the oligotrophic environment. Coincide with our observation, pstS has been suggested to be one of the genes contributing to the difference in AMG content between coastal and open ocean phage isolates (Fuchsman et al., 2023). The enzymes produced by the gnd and zwf genes in cyanophages are involved in the pentose phosphate pathway, an important metabolic pathway providing reducing capacity and the carbon skeleton required for biosynthesis (Thompson et al., 2011). The enzymes produced by these two genes can convert 6-phosphogluconate to NADPH and ketones. This process could be important for the growth and replication of oligotrophic cyanophages, as NADPH is a crucial cofactor needed for energy production and biosynthesis (Doron et al., 2016). PyrBI that encodes catalytic subunit of aspartate carbamoyltransferase catalyzes the first step in the pyrimidine biosynthetic pathway. However, the role of pyrBI in cyanophages is still not clear. In brief, variation in content of AMGs among closely related cyanophages reflect the different horizontal evolutionary paths resulting from selection under diverse environmental pressures during vertical evolution of homologous cyanophages. Therefore, further research on AMGs among cyanophages of different habitats will deepen our understanding of cyanophage-host interactions in the environment.

The biogeographical distribution of Nanhaivirus ms29 was examined in 154 viral metagenomes from five VEZs (Vertical Ecological Zones) of the Global Ocean Viromes (GOV2.0) dataset, which included the Arctic (ARC), Antarctic (ANT), temperate and tropical epipelagic (EPI), temperate and tropical mesopelagic (MES), and bathypelagic (BATHY) regions (Gregory et al., 2019). We selected 45 cyanophage sequences from different cyanophage genera under Kyanoviridae as references and compared their absolute abundance in GOV2.0 (Figure 5A). In T4-like phages, the abundance of Nanhaivirus ms29 is relatively low, with the highest abundance observed in the EPI and MES regions. Despite its low abundance, Nanhaivirus-ms29 was widely distributed across the Indian, Pacific, Atlantic, Arctic, and Antarctic Oceans, with a higher relative abundance in the Indian Ocean than other regions (Figure 5B). Notably, Nanhaivirus ms29 was highly abundant in the nearshore waters of the Arabian Sea and the Red Sea compared to other sea areas. Furthermore, Nanhaivirus ms29 was also detected in the arctic pole.