Molecular Characterization of Novel Mycoviruses in Seven Umbelopsis Strains

The presence of viruses is less explored in Mucoromycota as compared to other fungal groups such as Ascomycota and Basidiomycota. Recently, more and more mycoviruses are identified from the early-diverging lineages of fungi. We have determined the genome of 11 novel dsRNA viruses in seven different Umbelopsis strains with next-generation sequencing (NGS). The identified viruses were named Umbelopsis ramanniana virus 5 (UrV5), 6a (UrV6a); 6b (UrV6b); 7 (UrV7); 8a (UrV8a); 8b (UrV8b); Umbelopsis gibberispora virus 1 (UgV1); 2 (UgV2) and Umbelopsis dimorpha virus 1a (UdV1a), 1b (UdV1b) and 2 (UdV2). All the newly identified viruses contain two open reading frames (ORFs), which putatively encode the coat protein (CP) and the RNA-dependent RNA polymerase (RdRp), respectively. Based on the phylogeny inferred from the RdRp sequences, eight viruses (UrV7, UrV8a, UrV8b, UgV1, UgV2, UdV1a, UdV1b and UdV2) belong to the genus Totivirus, while UrV5, UrV6a and UrV6b are placed into a yet unclassified but well-defined Totiviridae-related group. In UrV5, UgV1, UgV2, UrV8b, UdV1a, UdV2 and UdV1b, ORF2 is predicted to be translated as a fusion protein via a rare +1 (or −2) ribosomal frameshift, which is not characteristic to most members of the Totivirus genus. Virus particles 31 to 32 nm in diameter could be detected in the examined fungal strains by transmission electron microscopy. Through the identification and characterization of new viruses of Mucoromycota fungi, we can gain insight into the diversity of mycoviruses, as well as into their phylogeny and genome organization.


Introduction
Viruses have been identified in all main groups of fungi [1,2]. However, the vast majority of them were detected in members of the Ascomycota and Basidiomycota phyla, while the mycovirus harboring of basal fungi, such as those in the phylum Mucoromycota, have remained less explored [3]. Some recent studies suggest that the rate of virus carriage can also be high in these groups [3][4][5]. Most viruses described so far in the Mucoromycota have a linear double-stranded RNA genome and belong to the Totiviridae family [4,5]. Within this family, five genera, i.e., Giardiavirus, Leishmaniavirus, Trichomonasvirus, Totivirus and Victorivirus have been discerned, among which Totivirus and Victorivirus contain the majority of mycoviruses. Recently, two other genera have been proposed, i.e., Artivirus,

Fungal Strains and Cultivation
The 30 Umbelopsis strains involved in the study are presented in Supplementary  Table S1. Strains were maintained on malt extract agar slants (MEA; 0.5% malt extract, 0.5% yeast extract, 1% glucose, and 2% agar) at 4 • C. Mycelia for dsRNA and virus particle isolation were grown in yeast extract-glucose broth (YEG; 1% glucose, 0.5% yeast extract) at 25 • C for 4 days on an orbital shaker at 150 rpm.

Purification of Nucleic Acids
To screen the strains for the presence of dsRNA elements, the total nucleic acid extraction method of Leach et al. [15] was used with modifications [5]. To purify the dsRNA from the total nucleic acid samples produced with the abovementioned method of Leach et al. [15] for NGS and "full-length amplification of cDNA" (FLAC), a previously described CF-11 cellulose chromatography method [5] was used. dsRNA molecules were separated by electrophoresis on 0.8% agarose/TAE (40 mMTris/acetic acid, 1 mM EDTA, pH 7.6) horizontal gels and visualized in UV after ethidium bromide (0.5 µg/mL) staining. Separated molecules were recovered from the gel using the Zymoclean Gel RNA Recovery Kit (Zymo Research, Irvin, CA, USA) according to the manufacturer's instructions. The nature of the detected dsRNA elements was confirmed by their resistance to DNase I (Thermo Scientific, Waltham, MA, USA) and S1 nuclease (Thermo Scientific, Waltham, MA, USA) digestions carried out according to the recommendations of the manufacturers.

cDNA Synthesis and Sequencing of the dsRNA Molecules
For synthesis and amplification of cDNAs from dsRNA fragments, the FLAC method was used as described by Maan et al. 2007 [16] and modified by Darissa et al. 2010 [17]. Briefly, the PC3-T7 loop primer [18] was ligated to the purified dsRNA molecules; then, the primer-ligated dsRNA fragments were purified with the RNA Clean and Concentrator-5 kit (Zymo Research, Irvin, CA, USA), according to the manufacturer's protocol. Denaturation and cDNA synthesis were performed as described previously [17]. Amplification of the cDNA was performed using a 1.25 µM PC2 primer [18] and 1 unit of the Phusion High-Fidelity DNA Polymerase (Thermo Scientific, Waltham, MA, USA), as described previously [4]. In some cases, the amplification of the cDNA was performed using a 0.5 µM PC2 primer [18] and 0.5 µM specific primer designated for the adequate sequence segment (Supplementary Table S2). Amplified DNA fragments were purified after agarose gel electrophoresis using the Zymoclean Large Fragment DNA Recovery Kit (Zymo Research, Irvin, CA, USA). Purified products were then cloned into the pJET1.2/Blunt vector (Clone-JET PCR Cloning Kit, Thermo Scientific, Waltham, MA, USA). Sequences of the inserts were determined by the Eurofins Genomics Germany GmbH. FASTQ sequence files were generated by GenerateFASTQ 1.1.0.64 application of Illumina BaseSpace. Adapter trimming and quality-dependent sequence trimming were done using TrimGalore with parameters: -paired -length 36 -q 20. To filter out host sequences, reads were aligned to subject sequence databases generated from corresponding Umbelopsis sp.-specific sequences available from the National Center for Biotechnology Information (NCBI) using the BLASTn algorithm.
For the phylogenetic analysis, representative RdRp sequences of Totiviridae, Chrysoviridae and Partitiviridae were obtained from viruSite (http://www.virusite.org/index.php (accessed on 12 September 2022)). The dataset was supplemented by homologous hits of U. ramanniana RdRp sequences [4]. Sequences were aligned using MAFFT v. 7.453 [21] with the E-INS-i option. Best-fitting model for the maximum likelihood inference was selected by the ModelFinder [22] software of the IQ-TREE v. 1.6.12 package [23] based on the Bayesian Information Criterion [24]. The maximum likelihood analysis was conducted with IQ-TREE v. 1.6.12 applying the VT+F+R6 model. Statistical support of the best tree was calculated with ultrafast bootstrap approximation [25] in 5000 replicates.
The Umbelopsis phylogeny was inferred using sequences of the ITS region and part of the Mini-chromosome maintenance complex component 7 (MCM7). Alignments corresponding to the two regions were concatenated and partitioned. The dataset consisted of three partitions (rDNA, ITS1-ITS2 and MCM7). Model testing was performed on the partitioned dataset using the built-in model selection software of IQ-TREE v. 1.6.12. Best-fit models for the partitions were TPM2+F+G4 for the rDNA, TIM2e+G4 for the ITS region and TPM2u+F+G4 for the MCM7 region. Statistical support of the phylogenetic tree that best fits the dataset was tested with ultrafast bootstrap sampling with 5000 replicates.

Hybridization Analysis
dsRNA molecules and control plasmids were separated by electrophoresis on 1.0% agarose/TAE (40 mM Tris/acetic acid and 1 mM EDTA, pH 7.6) horizontal gels. Separated dsRNAs were denatured by rinsing the gel slides in 0.05 M NaOH and 0.15 M NaCl buffer for 30 min and neutralized in 1 M Tris-HCl and 1.5 M NaCl buffer (pH 7.5) for 2 × 20 min, as described by Hong et al. (1998) [26]. DNA samples were denatured in 0.5 M NaOH and 1.5 M NaCl buffer and neutralized in 0.5 M Tris and 1.5 M NaCl buffer (pH 7.5) [27]. Gel slides were blotted onto a positively charged nylon membrane (Amersham Hybond-N+, GE Healthcare) using 2× SSC buffer. After allowed to dry at room temperature, samples were immobilized with UV-crosslinking. Blots were hybridized with DIG-labeled CP and RdRp oligonucleotide probes in hybridization buffer (0.9 M NaCl, 1% SDS, 10% dextran sulfate) containing 5 µg/mL salmon sperm DNA (Invitrogen) (Supplementary Table S2). Probes were obtained by PCR from the corresponding DNA templates in the presence of digoxigenin-UTP (DIG DNA Labeling Mix, Roche, Sigma-Aldrich, Dorset, UK) using the DreamTaq polymerase (Thermo Scientific, Waltham, MA, USA). Primers applied to amplify the probes are listed in Supplementary Table S2. Hybridization was followed by immunological detection using alkaline phosphatase-conjugated anti-digoxigenin antibody (Roche, Sigma-Aldrich, Dorset, UK), according to the manufacturer's instructions.

Examination of Virus Particles
Virus particles were isolated from 30 g frozen mycelium using the method of Lot et al. (1972) [28] followed by a discontinuous sucrose density gradient (10 to 40% (w/v) sucrose in PBS) centrifugation as described earlier [5]. The purified samples were negatively stained as described earlier [4] and examined with a JEM-1400 Flash transmission electron microscope (JEOL, Tokyo, Japan). Samples were systematically screened at 30,000× magnification to detect the virus particles on the grid. Then, particles were recorded at 50,000-120,000× magnification with a 16 MP Matataki Flash scientific complementary metal-oxide semiconductor (sCMOS) camera (JEOL). Quantitative analysis to determine the size of the particles was performed using the built-in measurement tools of the electron microscope, then data expressed as mean ± standard error of the mean.

dsRNA Harboring in the Genus Umbelopsis
The presence of dsRNA elements was confirmed in five strains previously [4]. In the present study, they were found in a further two U. ramanniana strains ( Figure 1). The seven dsRNA harboring strains corresponded to the 23% of the tested Umbelopsis isolates (see Supplementary Table S1). For all previously tested strains, we achieved the same dsRNA patterns in this study. An approx. 5.0-kb fragment could be observed in almost all isolates. Out of the six tested U. ramanniana isolates, three proved to be dsRNA-harboring. The pattern observed in the U. ramanniana strain CBS 478.63 was similar to that described previously for NRRL 1296 containing the same sized 5.3-and 5.0-kb fragments (Figure 1), while CBS 243.58 contained only a single 5.3-kb fragment. microscope (JEOL, Tokyo, Japan). Samples were systematically screened at 30,000× magnification to detect the virus particles on the grid. Then, particles were recorded at 50,000-120,000× magnification with a 16 MP Matataki Flash scientific complementary metal-oxide semiconductor (sCMOS) camera (JEOL). Quantitative analysis to determine the size of the particles was performed using the built-in measurement tools of the electron microscope, then data expressed as mean ± standard error of the mean.

dsRNA Harboring in the Genus Umbelopsis
The presence of dsRNA elements was confirmed in five strains previously [4]. In the present study, they were found in a further two U. ramanniana strains ( Figure 1). The seven dsRNA harboring strains corresponded to the 23% of the tested Umbelopsis isolates (see Supplementary Table S1). For all previously tested strains, we achieved the same dsRNA patterns in this study. An approx. 5.0-kb fragment could be observed in almost all isolates. Out of the six tested U. ramanniana isolates, three proved to be dsRNA-harboring. The pattern observed in the U. ramanniana strain CBS 478.63 was similar to that described previously for NRRL 1296 containing the same sized 5.3-and 5.0-kb fragments (Figure 1), while CBS 243.58 contained only a single 5.3-kb fragment.

Whole Virus Genomes Identified in the Umbelopsis Isolates
Viral genome sequences were determined by NGS and assembled using SPAdes application with the -careful parameter. The sequencing depth (mean, median, minimal and maximal values) of the viral genome assemblies was the following:

Whole Virus Genomes Identified in the Umbelopsis Isolates
Viral genome sequences were determined by NGS and assembled using SPAdes application with the -careful parameter. The sequencing depth (mean, median, minimal and maximal values) of the viral genome assemblies was the following:  Predicted genome organization of the new viruses, including the position and length of the UTR sequences and the encoded ORFs, as well as the size and predicted molecular weight of the proteins encoded by the genomes are presented in Figures 2 and 3 (predicted secondary structures of the 5 and 3 UTRs are also presented in Supplementary Figure S1). All genomes are undivided and contain two overlapping open reading frames, ORF1 and ORF2 encoding a CP and a RdRp, respectively. Relatively short UTR sequences were detected in the genomes as their length varied between 11 and 153 and 21 and 65 nt for the 5 and 3 UTRs, respectively. In each case, ORF1 and ORF2 were predicted to be translated as a fusion protein via a ribosomal frameshift. For each genome, a possible slippery heptamer and an H-type pseudoknot facilitating the programmed ribosomal frameshifting could be proposed in the overlapping region, as presented in Figures 2 and 3. In each case of the newly identified mycoviruses, BLASTp homology search with the corresponding fragment sequence in the NCBI GenBank revealed a highest degree of identity with the CP and RdRp of viruses in the Totiviridae family (Supplementary Table S3).
The virus content of the U. ramanniana NRRL 1296 isolate was previously investigated in detail, and full genomes of four viruses named UrV1-4 were determined by the FLAC method [4]. In the present study, NGS sequencing was used to reexamine the virus genome sequences in this strain. The four previously described genomes of UrV1-4 could be detected by this method too. Additionally, NGS sequencing revealed a fifth, 5069-nt-long genome. The corresponding novel virus was named Umbelopsis ramanniana virus 5 (UrV5) ( Table 1). Interestingly, ORF1 and ORF2 were predicted to be translated as a fusion protein via a rare +1 (or −2) ribosomal frameshift ( Figure 2B).
In the case of U. ramanniana CBS 243.58, a single, 4977-nt-long virus genome was found ( Figure 2B). The identified genome showed 83.2% identity with the UrV6a genome at the nucleotide level, while the CP and RdRp proteins of the two viruses proved to be identical by 95.8 and 95%, respectively. Accordingly, this novel virus found in the strain CBS 243.58 was named to Umbelopsis ramanniana virus 6b (UrV6b), indicating that UrV6a and UrV6b are closely related and can be regarded as two versions of the same virus.
Two virus genomes with 4629 and 4622 nt in length were detected in U. gibberispora CBS 109328, and the viruses were named as Umbelopsis gibberispora virus 1 (UgV1) and Umbelopsis gibberispora virus 2 (UgV2), respectively. For both viruses, ORF1 and ORF2 were predicted to be translated as a fusion protein via a +1 (or −2) ribosomal frameshift ( Figure 3B).
In U. angularis CBS 603.68, a single 4613-nt genome was determined ( Figure 3B). The identified genome is most likely a variant of the UrV8 genome, because its sequence shows 79.6% identity with that of the UrV8a at the nucleic acid level. Moreover, amino acid sequences of the CP and RdRp of the two viruses proved to be identical by 95.5% and 92.4%, respectively. Therefore, the virus identified in U. angularis CBS 603.68 was described as Umbelopsis ramanniana virus 8b (UrV8b).
In the U. versiformis CBS 473.74 strain, NGS identified a 4624-nt viral genome, which showed 91.4% (CP) and 80.6% (RdRp) identity with the corresponding UdV1a sequences at the amino acid level. Based on this, we named it Umbelopsis dimorpha virus 1b (UdV1b) and regarded as a variant of the UdV1 virus ( Figure 3B).

Phylogenetic Analysis
Based on the RdRp amino acid sequences of the newly identified genomes and representative members from the viral families Totiviridae, Chrysoviridae and Partitiviridae, a phylogeny was inferred using the ML method ( Figure 4). This analysis demonstrated that all 11 novel viruses belong to the Totiviridae family. Eight viruses, i.e., UdV1a, UdV1b, UdV2, UrV7, UrV8a, UrV8b, UgV1 and UgV2, are part of the clade representing the genus Totivirus. Among them, UdV1a, UdV1b, UdV2, UrV7, UrV8a, UrV8b and UgV2 proved to be closely related to the previously described UrV1, UrV4 [4] and Mucor hiemalis virus 4 [5]. UgV1 seated in another subclade of the Totivirus clade together with other mycoviruses-among others, Xanthophyllomyces dendrorhous virus L 1A. UrV5, UrV6a and UrV6b are in the same clade as the earlier described UrV3 [4]. These viruses form a well-defined but yet unclassified clade together with viruses of nonfungal organisms, such as Beihai barnacle virus 15 and Diatom colony associated dsRNA virus 17 A and B.    [19] and drawn by the PseudoViewer program [20]. EFE (kcal/mol) indicates the estimated free energy.

Hybridization Analysis of the dsRNA Patterns
A hybridization analysis revealed that the UrV5 genome corresponds to the 5.3-kb fragment of the dsRNA pattern of the U. ramanniana NRRL 1296 isolate (Supplementary Figure S2).
U. ramanniana CBS 478.63 had an electrophoretic pattern consisting of four fragments ( Figure 2). During the hybridization experiments, the probes designed for the CP and RdRp of the UrV6a virus hybridized to the 5.3 kb fragment ( Figure 5). It is worth mentioning that the fragment in the U. ramanniana CBS 243.58 isolate also gave a signal of a similar strength with the UrV6a probes demonstrating the high level of sequence identity between UrV6a and UrV6b ( Figure 5) The probes designed for UrV7 and UrV8a viruses hybridized to the second, 5.0-kb fragment (Supplementary Figure S3). In the dsRNA pattern of U. ramanniana CBS 478.63, two smaller fragments can also be observed, for which we could not recover the sequence information.   RdRp of the UrV6a virus hybridized to the 5.3 kb fragment ( Figure 5). It is worth mentioning that the fragment in the U. ramanniana CBS 243.58 isolate also gave a signal of a similar strength with the UrV6a probes demonstrating the high level of sequence identity between UrV6a and UrV6b ( Figure 5) The probes designed for UrV7 and UrV8a viruses hybridized to the second, 5.0-kb fragment (Supplementary Figure S2.2). In the dsRNA pattern of U. ramanniana CBS 478.63, two smaller fragments can also be observed, for which we could not recover the sequence information.  In the case of U. gibberispora CBS 109328, probes designed for the CP and RdRp genes of UgV1 and UgV2 hybridized to the larger, 5.0-kb fragment (Supplementary Figure S2.3). For this isolate, we detected a smaller dsRNA fragment of 4.0 kb in size, which, according to our present investigation, we do not have any sequence information. Similarly, a third fragment of approx. 0.7 kb was also observed in the electrophoretic pattern. Although we did not manage to identify this fragment with NGS, a 661-nt segment could be isolated and sequenced using the FLAC method (Supplementary Data S1). However, the resulting In the case of U. gibberispora CBS 109328, probes designed for the CP and RdRp genes of UgV1 and UgV2 hybridized to the larger, 5.0-kb fragment (Supplementary Figure S4). For this isolate, we detected a smaller dsRNA fragment of 4.0 kb in size, which, according to our present investigation, we do not have any sequence information. Similarly, a third fragment of approx. 0.7 kb was also observed in the electrophoretic pattern. Although we did not manage to identify this fragment with NGS, a 661-nt segment could be isolated and sequenced using the FLAC method (Supplementary Data S1). However, the resulting sequence did not match any other sequences in the NCBI GenBank database either at the nucleotide or the amino acid level.
In U. angularis CBS 603.68, two dsRNA fragments were found, but we could only identify one viral genome, which was 4613 nt in size. Probes designed for UrV8b CP and RdRp genes hybridized with the larger 5.0-kb fragment (Supplementary Figure S5). A second, approximately 4.0-kb sized dsRNA was also detected by gel electrophoresis in this isolate (Figure 1). From this 4.0-kb dsRNA fragment, cDNA was generated by the FLAC technique, and the resulted clone was sequenced. Thus, a 3847-nt sequence was determined (Supplementary Data S1). The fragment contains a single ORF (from 1195 to 3645 nt), which was predicted to encode a putative, 816-aa hypothetical protein. However, the nucleic acid and the predicted amino acid sequence of this hypothetical protein did not show any similarities to known mycoviral sequences and other sequences in the GenBank by Blast queries.
In the case of the sample from U. dimorpha CBS 110039, we found that probes designed for both viruses (i.e., UdV1a and UdV2) hybridized to the 5.0-kb fragment (Supplementary Figure S6) The probe designed for the RdRp of UdV1a also gave a signal with the 5.0 kb fragments of U. ramanniana NRRL 1296, U. ramanniana CBS 478.63, U. angularis CBS 603.68 and U. versiformis CBS 473.74 isolates. The smaller 4.0-kb fragment detected during gel electrophoresis could also not be determined.
Probes designed for the coding regions of UdV1b hybridized to the 5.0-kb fragment of the isolate U. versiformis CBS 473.74 (Supplementary Figure S7). dsRNA pattern of this isolate also contained an approx. 1-kb fragment, which could be identified with the FLAC method (Supplementary Data S1). The exact size of the fragment was 1069 nt and proved to be a partial RdRp coding sequence, which showed 99.7% of identity with a 1070-nt section of the UrV4 genome. UrV4 was identified in the U. ramanniana NRRL 1296 isolate earlier [4]. Presumably, this could be residual or cryptic viral nucleic acid in the host fungus.

Diversity of Mycoviruses in the Genus Umbelopsis
To get an idea about the viral diversity of the genus, the virus harboring feature was compared with the phylogeny of the examined Umbelopsis strains (Figure 6). In this phylogeny, the Umbelopsis genus is split to two large clades. Among them, the clade containing U. ramanniana, U. angularis and U. gibberispora has a much higher proportion of the virus-carrying strains than the clade containing U. dimorpha and U. versiformis. U. dimorpha CBS 110039 and U. versiformis CBS 473.74, which form a sister group on the tree carry variants of the same virus species (UdV1a and UdV1b). We also identified two genome variants of the same virus (UrV6a and UrV6b) in U. ramanniana CBS 478.63 and CBS 243.58 isolates, which are more closely related to each other than to the other U. ramanniana isolates. In other cases, however, we found that, despite the close phylogenetic relationship of the virus-carrying isolates, the identified virus genomes differed to a greater extent. This can be seen in the case of U. angularis CBS 603.68 and U. gibberispora CBS 109328 isolates. Moreover, the UrV8b virus was identified in the U. angularis CBS 603.68 strain, while UrV8a was detected in the more distantly related U. ramanniana CBS 478.63 isolate.

Detection of Virus Particles in the Virus-Harboring Umbelopsis Strains
The presence of virus-like particles (VLPs) in U. ramanniana NRRL 1296 was already reported [4]. In this study, transmission electron microscopy revealed the presence of isometric VLPs in the purified extracts of the other five mycovirus-harboring Umbelopsis strains (Figure 7). VLPs about 32 nm in diameter were detected in U.

Discussion
In our studies (i.e., Kartali [29]. All viruses identified in the present study proved to be the member of the Totiviridae family. Previously, four viruses were already described in U. ramanniana NRRL 1296, Figure 7. Virus-like particles detected in five mycovirus-harboring Umbelopsis strains. The virus particles were recovered by ultracentrifugation at 78,000× g for 12 h at 4 • C, which follows with sucrose gradient density centrifugation. Purified virus particles were negatively stained with 2% uranyl acetate in 50% ethanol for 5 min (3 times) and examined under a JEM-1400 Flash transmission electron microscope. VLPs about 31-32 nm in diameter were detected in the examined Umbelopsis strains.

Discussion
In our studies (i.e., Kartali [29]. All viruses identified in the present study proved to be the member of the Totiviridae family. Previously, four viruses were already described in U. ramanniana NRRL 1296, which also belonged to the Totiviridae [4]. Myers et al. (2020) also detected a Totivirus-like genome in a strain of Umbelopsis nana [3]. Among the newly described 11 viruses, eight belong to the genus Totivirus and three seated in a well-separated clade containing other non-classified Totivirus-like viruses (Figure 4). This unclassified group contains UrV3, which was earlier described from U. ramanniana [4] and viruses identified from nonfungal organisms, i.e., from invertebrates and diatoms [30,31]. Although this unclassified group contained viruses with undivided Totiviridae-type dsRNA genomes, it proved to be closely related to the Chrysoviridae, which is a family of viruses with segmented genomes typically consisting of four dsRNA segments [32]. The close phylogenetic relationship between Totiviridae and Chrysoviridae has been reported several times [4,32,33].
UgV1 is located away from the former group and forms another subclade of Totivirus with Plasmopara viticola lesion associated toti 3 [35] and Xanthophyllomyces dendrorhous virus L1A [36]. Although we identified another virus, UgV2, from the same host (i.e., U. gibberispora CBS 109328), the similarity between the two genomes was minimal, which is supported by their location on the phylogeny (Figure 4).
When virus carriage was compared with the phylogeny of the fungal isolates ( Figure 6), it was observed that isolates representing different species may carry the genome variants of the same virus, while closely related fungal isolates contain slightly related viruses. These observations raise the possibility of horizontal virus transmission among the different fungal isolates as has recently been increasingly assumed for other viruses and hosts too [34].
Among the seven tested strains, four isolates contain multiple viruses (Table 1). Moreover, presence of five virus genomes was proven in U. ramanniana NRRL 1296 [4]. The presence of multiple viruses in the same strain have been reported from several fungi, such as from Aspergillus fumigatus [37], or from the plant pathogenic fungus Exobasidium sp. [38]. In agreement with previous reports [3][4][5], our results suggest that mixed infections can commonly occur in Mucoromycota fungus.
Genome organization of the identified viruses show the typical features of the genomes in the genus Totivirus, i.e., an undivided dsRNA genome with two partially overlapping ORFs [1]. This arrangement suggests the translation of RdRp as a fusion protein [39]. Generally, it is proposed for the totiviruses that translation of the fusion protein is carried out via a programmed -1 ribosomal frameshift [39]. However, in seven novel genomes, i.e., in those of UrV5, UgV1, UgV2, UrV8b, UdV1a, Udv1b and UdV2, a +1(-2) ribosomal frameshift could be predicted. This type of frameshift is considered as a rare mechanism in Totiviridae and has been reported mainly in the genera Trichomonas and Leishmaniavirus [40,41]. It is worth mentioning that +1(−2) ribosomal frameshift was also proposed for the Mucor hiemalis virus 4 (MhV4) belonging to the genus Totivirus [5].
Approximately the same-sized VLPs, about 31 to 32 nm in diameter, could be observed in all examined Umbelopsis strains. Morphology and size of the detected VLPs are typical for the Totiviridae family [42]. In the case of U. ramanniana CBS 478.63 and U. gibberispora CBS 109328 with multiple mycoviruses, we detected one-sized VLPs, with 31 or 32 nm in diameter (Figure 7). This could be explained with the approximately same genome sizes of the corresponding viruses ( Figures 2B and 3B). In U. dimorpha CBS 110039, we did not detect any VLPs, which could be explained by the very low attendance of VLPs in this fungus.
In conclusion, we identified and described 11 novel Totiviridae-related viruses from five Umbelopsis species and could obtain an insight into the mycoviral diversity of this fungal genus. In four strains, the presence of multiple viruses was proven. Our results also suggest that a +1 ribosomal frameshift might not be so rare a process in the Totiviridae family as previously reported.