A Novel Freshwater Cyanophage, Mae-Yong924-1, Reveals a New Family

Cyanobacterial blooms are a worldwide ecological issue. Cyanophages are aquatic viruses specifically infecting cyanobacteria. Little is known about freshwater cyanophages. In this study, a freshwater cyanophage, Mae-Yong924-1, was isolated by the double-layer agar plate method using Microcystis aeruginosa FACHB-924 as an indicator host. Mae-Yong924-1 has several unusual characteristics: a unique shape, cross-taxonomic order infectivity and a very unique genome sequence. Mae-Yong924-1 contains a nearly spherical head of about 100 nm in diameter. The tail or tail-like structure (approximately 40 nm in length) is like the tassel of a round Chinese lantern. It could lyse six diverse cyanobacteria strains across three orders including Chroococcales, Nostocales and Oscillatoriales. The genome of the cyanophage is 40,325 bp in length, with a G + C content of 48.32%, and 59 predicted open reading frames (ORFs), only 12 (20%) of which were functionally annotated. Both BLASTn and BLASTx scanning resulted in “No significant similarity found”, i.e., the Mae-Yong924-1 genome shared extremely low homology with sequences in NCBI databases. Mae-Yong924-1 formed a root node alone and monopolized a root branch in the proteomic tree based on genome-wide sequence similarities. The results suggest that Mae-Yong924-1 may reveal a new unknown family apparently distinct from other viruses.


Introduction
Blooms of toxic cyanobacteria have become a common occurrence in water bodies worldwide. The frequency and intensity of cyanobacterial blooms are becoming increasingly serious. Cyanobacterial blooms consume a large amount of dissolved oxygen in the water and release cyanotoxins, endangering aquaculture objects, resulting in food safety risks to products due to algal toxin residues and polluting water sources [1]. Cyanotoxins are key risk factors that contribute to the development of liver cancer. The high incidence of primary liver carcinoma (PLC) was found to correlate with emerging blooms of cyanobacteria in water reservoirs supplying drinking water [2]. Microcystis spp. are the most ecologically harmful and dominant bloom-forming cyanobacteria in freshwater lakes [3]. They usually produce microcystins (MCs) at concentrations that cause chronic poisoning or acute death to animals and humans [4]. MCs have been intensively studied among cyanotoxins due to their toxicity, causing severe liver and kidney damage, tumor promotion and gastroenteritis [5,6]. Accumulation of MCs in aquatic ecosystems increases public health concerns [7,8].
Here, we isolated a novel freshwater cyanophage, Mae-Yong924-1, using M. aeruginosa FACHB-924 as an indicator cyanobacterial strain. Mae-Yong924-1 has several unusual characteristics: cross-taxonomic order infectivity, a special shape and a unique genome sequence, forming a root node alone and monopolizing along an independent root branch in the phylogenetic tree based on complete genome similarities. We propose that Mae-Yong924-1 may represent a new family distinct from the other viruses.

Isolation and Amplification of Cyanophage
Surface water samples were collected from the Yangming Lake (north latitude, 29.773222; east longitude, 121.954422) in the Meishan campus of Ningbo University. The water samples were centrifuged (10,000× g, 20 min) and filtered (through 0.45 µm and 0.22 µm nitrocellulose filter membranes (ANPEL Laboratory Technologies Inc, Shanghai, China)) as described [13]. The enrichment and separation of the cyanophage were carried out according to the process reported in [13], employing M. aeruginosa FACHB-924 as the host. In brief, 3 mL aliquots of the filtrates were mixed with FACHB-924 cultures at the exponential phase (at the volume ratio of 1:2) and cultured for 10-20 days until the cultures were etiolated. The yellowing lysate was centrifuged, filtered and applied to the next round of enrichment, as described above. The enrichment was performed for three rounds. A pure cyanophage strain was obtained by five serial single-plaque isolations using the double-layer agar method [13]. A single plaque was dug out from the plate, mixed with 3 mL of exponentially growing FACHB-924 and cultured in a light incubator for 10-15 days. The yellowed and clarified culture suspension was centrifuged (6000× g, 20 min) and filtered (0.45 µm and 0.22 µm). Amplification culture was performed via co-cultivation of M. aeruginosa FACHB-924 and the filtrate containing the Microcystis cyanophage Mae-Yong924-1 (at a volume ratio of 5:1) until the culture turned yellow (for about 10 days).

Host Range Determination
All 14 tested cyanobacteria (Table 1) were obtained from the freshwater Algae Culture Bank, from the Institute of Hydrobiology (Wuhan, China) of the Academy of Sciences. Three hundred-microliter aliquots of the Mae-Yong924-1 suspension were added to the exponential cultures (600 µL) of the tested cyanobacterial strains in triplicate in 48-well cell culture plates and incubated in a light incubator. The phage suspension was replaced with BG11 medium in the negative control groups. All cultures were monitored daily for cell lysis via visual inspection and optical microscopy (VersaMax, Molecular Devices, Sunnyvale, CA, USA) observation and checked by measuring the absorbance at OD 680 . Cyanobacterial strains that did not lyse until the 12th day were defined as insusceptible.

Genome Extraction, Sequencing and Assembly
The cyanophage genome was extracted using a High Pure Viral RNA Kit (Roche, Basel, Switzerland, product no. 11858882001). This kit allows the extraction of DNA and RNA together. The NEBNext Ultra™ II DNA Library PrepKit for Illumina was used for constructing a 2 × 300 bp paired-end DNA library. The Illumina MiSeq (San Diego, CA, USA) sequencing platform was used for sequencing. FACHB-924 culture without the cyanophage was performed in the same way as for the control group. The reads appearing in the control group were deleted from the raw sequencing data of the experimental group. Fastp and Trimmomatic-0.36 were used to filter out bad reads (Q-value < 20) and adapters. SPAdes 3.13.0 (http://cab.spbu.ru/software/spades/; 16 February 2021) was used to assemble the clean reads. Phage genome termini were predicted by the proposed method in [18].

Phylogenetic Tree
The proteomic tree approach is effective in investigating genomes of newly sequenced viruses. ViPTree online [24,25] (https://www.genome.jp/viptree/; 10 December 2021) was used to generate a proteomic tree based on genome-wide sequence similarities, computed by tBLASTx.

Results and Discussion
After five successive single-plaque isolations, uniform plaques were obtained. Mae-Yong924-1 is the fifth cyanophage isolated with the cyanobacterium M. aeruginosa. The lysis of the co-cultures of FACHB-924 with the cyanophage Mae-Yong924-1 occurred in 7-10 days.
Observation under TEM found that negatively stained Mae-Yong924-1 possesses a nearly spherical head of approximately 100 nm in diameter. The tail or tail-like structure (approximately 40 nm in length) of Mae-Yong924-1 is unique, which is like the tassel of a round Chinese lantern (Figure 1).

Phylogenetic Tree
The proteomic tree approach is effective in investigating genomes of newly se quenced viruses. ViPTree online [24,25] (https://www.genome.jp/viptree/; 10 Decembe 2021) was used to generate a proteomic tree based on genome-wide sequence similarities computed by tBLASTx.

Results and Discussion
After five successive single-plaque isolations, uniform plaques were obtained. Mae Yong924-1 is the fifth cyanophage isolated with the cyanobacterium M. aeruginosa. Th lysis of the co-cultures of FACHB-924 with the cyanophage Mae-Yong924-1 occurred i 7-10 days.
Observation under TEM found that negatively stained Mae-Yong924-1 possesses nearly spherical head of approximately 100 nm in diameter. The tail or tail-like structur (approximately 40 nm in length) of Mae-Yong924-1 is unique, which is like the tassel of round Chinese lantern (Figure 1).  (Table 1). The results demonstrate that the Microcysti cyanophage Mae-Yong924-1 had the ability to lyse cyanobacterial strains across taxo nomic orders (Chroococcales, Nostocales and Oscillatoriales). A wider host range may be ad vantageous for the application because cyanobacterial blooms are usually caused by mul tiple cyanobacteria [9]. Although most isolated and well-studied phages have a narrow host range, an increasing number of broad-host-range phages have been found recentl [26]. The Aquamicrobium phage P14 has even been found to infect various bacterial strain across taxonomic classes, Alphaproteobacteria and Betaproteobacteria [27]. Among the si susceptible cyanobacterial strains, the lysis of M. elabens FACHB-916 by Mae-Yong924appeared to be the most efficient. Mae-Yong924-1 could fully lyse M. elabens FACHB-91 in the exponential phase in 1-3 days. Microscopic observation demonstrated that the num ber of intact cyanobacterial cells in the yellowing tested groups was significantly less tha that in the control groups ( Figure 2). Subsequently, the amplification of the cyanophag Mae-Yong924-1 was performed via co-cultivation of Mae-Yong924-1 with M. elaben FACHB-916 (at a volume ratio of 5:1). A literature search indicated that Mae-Yong924-1 i the second cyanophage found to be able to rapidly lyse M. elabens FACHB-916. In the host range experiments, lytic efficiencies occurred in 6 of the 14 tested cyanobacterial strains ( Table 1) (Table 1). The results demonstrate that the Microcystis cyanophage Mae-Yong924-1 had the ability to lyse cyanobacterial strains across taxonomic orders (Chroococcales, Nostocales and Oscillatoriales). A wider host range may be advantageous for the application because cyanobacterial blooms are usually caused by multiple cyanobacteria [9]. Although most isolated and well-studied phages have a narrow host range, an increasing number of broad-host-range phages have been found recently [26]. The Aquamicrobium phage P14 has even been found to infect various bacterial strains across taxonomic classes, Alphaproteobacteria and Betaproteobacteria [27]. Among the six susceptible cyanobacterial strains, the lysis of M. elabens FACHB-916 by Mae-Yong924-1 appeared to be the most efficient. Mae-Yong924-1 could fully lyse M. elabens FACHB-916 in the exponential phase in 1-3 days. Microscopic observation demonstrated that the number of intact cyanobacterial cells in the yellowing tested groups was significantly less than that in the control groups ( Figure 2). Subsequently, the amplification of the cyanophage Mae-Yong924-1 was performed via co-cultivation of Mae-Yong924-1 with M. elabens FACHB-916 (at a volume ratio of 5:1). A literature search indicated that Mae-Yong924-1 is the second cyanophage found to be able to rapidly lyse M. elabens FACHB-916.
The complete genome sequence of Mae-Yong924-1 was sequenced with an average sequencing depth of 169-fold. The Mae-Yong924-1 genome is 40,325 bp in length with a G + C content of 48.32%. BLASTn and BLASTx analyses resulted in "No significant similarity found", i.e., the results illustrated that the Mae-Yong924-1 genome shared extremely low homology with the sequences in current databases. Phage terminus analysis showed that Mae-Yong924-1 had no fixed phage terminus. The genome was deposited in GenBank under the accession number MZ447863. The complete genome sequence of Mae-Yong924-1 was sequenced with an average sequencing depth of 169-fold. The Mae-Yong924-1 genome is 40,325 bp in length with a G + C content of 48.32%. BLASTn and BLASTx analyses resulted in "No significant similarity found", i.e., the results illustrated that the Mae-Yong924-1 genome shared extremely low homology with the sequences in current databases. Phage terminus analysis showed that Mae-Yong924-1 had no fixed phage terminus. The genome was deposited in GenBank under the accession number MZ447863.
The CRISPR/Cas immune system is a defense strategy against foreign nucleic acids such as phage genomes of bacteria and archaea [28]. CRISPR loci are normally composed of discontinuous direct repeats separated by short stretches of DNA sequences called spacers and usually clustering with cas (CRISPR-associated) genes [29]. No matching sequence between the Mae-Yong924-1 genome and viral spacer sequences was found within the CRISPRs database.
RAST, BLASTp, Hmmer and HHpred analyses proposed that Mae-Yong924-1 has 59 predicted ORFs, which encode hypothetical proteins/peptides of 37-1011 residues in length. Sequence comparison indicated that, of the 59 predicted ORFs, only 42 (71%) had homologs in current databases, and only 12 (20%) could be functionally annotated. The rest of the ORFs shared no sequence similarity to any previously identified proteins. The 59 predicted ORFs could be divided into 5 functional groups: lysis (1 ORF), packaging (2 ORFs), structure (2 ORFs), regulation and replication (7 ORFs) and uncharacterized (47 ORFs) (Figure 3). The CRISPR/Cas immune system is a defense strategy against foreign nucleic acids such as phage genomes of bacteria and archaea [28]. CRISPR loci are normally composed of discontinuous direct repeats separated by short stretches of DNA sequences called spacers and usually clustering with cas (CRISPR-associated) genes [29]. No matching sequence between the Mae-Yong924-1 genome and viral spacer sequences was found within the CRISPRs database.
Surprisingly, only two ORFs in the Mae-Yong924-1 genome were predicted to encode putative structural proteins including a phage-related minor tail protein (ORF 54) and a major capsid protein, gpN (ORF 38). No other gene sharing significant homology with viral structural proteins was found, which is very different from most known cyanophages. This character echoes the unique morphological structure feature of Mae-Yong924-1, significantly different from that of all the reported phages ( Figure 1). Correspondingly, the Mae-Yong924-1 genome had extremely low similarity to sequences in current databases.
As described in the previous section, both BLASTn and BLASTx scanning of the Mae-Yong924-1 genome resulted in "No significant similarity found", which demonstrated the extremely low similarity between Mae-Yong924-1 and other phages in current databases. To further estimate the nucleotide sequence similarity between Mae-Yong924-1 and other phages in current (10 August 2021) public databases, the Pairwise Sequence Comparison (PASC) classification tool [31] (https://www.ncbi.nlm.nih.gov/sutils/pasc/viridty.cgi; 25 Surprisingly, only two ORFs in the Mae-Yong924-1 genome were predicted to encode putative structural proteins including a phage-related minor tail protein (ORF 54) and a major capsid protein, gpN (ORF 38). No other gene sharing significant homology with viral structural proteins was found, which is very different from most known cyanophages. This character echoes the unique morphological structure feature of Mae-Yong924-1, significantly different from that of all the reported phages ( Figure 1). Correspondingly, the Mae-Yong924-1 genome had extremely low similarity to sequences in current databases.
As described in the previous section, both BLASTn and BLASTx scanning of the Mae-Yong924-1 genome resulted in "No significant similarity found", which demonstrated the extremely low similarity between Mae-Yong924-1 and other phages in current databases. To further estimate the nucleotide sequence similarity between Mae-Yong924-1 and other phages in current (10 August 2021) public databases, the Pairwise Sequence Comparison (PASC) classification tool [31] (https://www.ncbi.nlm.nih.gov/sutils/pasc/viridty.cgi; 25 August 2021) was used. Mae-Yong924-1 shared the highest nucleotide sequence similarity, as low as 11.94%, with the most closely related phage in the PASC search. This value is well below the >50% boundaries to define a genus. The PASC value indicates that Mae-Yong924-1 belongs to an unknown new genus. The complete genome sequences of 10 cyanophages including 5 Microcystis cyanophages, 43 representative bacteriophages of the 14 families and the genus Lilyvirus of Caudovirales (which are the latest classification Viruses 2022, 14, 283 7 of 9 by the ICTV) and 3 Alloherpesviridae viruses of Herpesvirales were obtained from the NCBI database. A proteomic tree was constructed based on the complete genome sequences. In the phylogenetic tree, Mae-Yong924-1 formed a root node alone and monopolized a root branch. The phylogenetic tree illustrated that there was a far evolutionary distance between Mae-Yong924-1 and the other phages of Caudovirales (Figure 4). Mae-Yong924-1 may reveal a novel clade (at least a novel family). Of the 11 cyanophages in the proteomic tree, only 2, i.e., MaMV-DC and Ma-LMM01, have a definite taxonomic status designated by the ICTV, and the other 9 strains are unclassified. MaMV-DC and Ma-LMM01 were classified as Myoviridae family members in the current ICTV classification. However, in the proteomic tree of this study, MaMV-DC and Ma-LMM01 were found to be very distinct from other Myoviridae phages. The proteomic tree reveals global genomic similarity relationships among viruses. Phage proteomic trees are advocated by the ICTV for use as the basis of a genome-based taxonomical system for phages. We propose that the creation of a new family should comprise MaMV-DC and Ma-LMM01 in the next versions of taxonomic updates. In summary, based on the morphology and sequence characteristics, we propose that the Microcystis cyanophage Mae-Yong924-1 may represent a new clade (at least a novel family) of phages. The isolation of Mae-Yong924-1 is an exciting discovery due to its uniqueness and capability of cross-taxonomic order infectivity.