Novel and Divergent Viruses Associated with Australian Orchid-Fungus Symbioses

RESEARCH REPOSITORY

This is the author’s final version of the work, as accepted for publication following peer review but without the publisher’s layout or pagination. The definitive version is available at:

https://doi.org/10.1016/j.virusres.2017.11.026

Ong, J.W.L., Li, H., Sivasithamparam, K., Dixon, K.W., Jones, M.G.K. and Wylie, S.J. (2018) Novel and divergent viruses associated with Australian orchid-fungus symbioses. Virus Research, 244 . pp. 276-283.

https://researchrepository.murdoch.edu.au/id/eprint/39951

Accepted Manuscript

Title: Novel and divergent viruses associated with Australian orchid-fungus symbioses

Authors: Jamie W.L. Ong, Hua Li, Krishnapillai Sivasithamparam, Kingsley W. Dixon, Michael G.K. Jones, Stephen J. Wylie

PII: S0168-1702(17)30663-9 DOI: https://doi.org/10.1016/j.virusres.2017.11.026 Reference: VIRUS 97301

To appear in: Virus Research

Received date: 7-9-2017 Revised date: 21-11-2017 Accepted date: 24-11-2017

Please cite this article as: Ong, Jamie W.L., Li, Hua, Sivasithamparam, Krishnapillai, Dixon, Kingsley W., Jones, Michael G.K., Wylie, Stephen J., Novel and divergent viruses associated with Australian orchid-fungus symbioses.Virus Research https://doi.org/10.1016/j.virusres.2017.11.026

This is a PDF ﬁle of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its ﬁnal form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Title:

Novel and divergent viruses associated with Australian orchid-fungus symbioses

Authors and affiliations:

Jamie W.L. Onga, Hua Lia, Krishnapillai Sivasithamparama, Kingsley W. Dixonb,

Michael G.K. Jonesa, Stephen J. Wyliea*

aPlant Biotechnology Group – Plant Virology, Western Australian State Agricultural

Biotechnology Centre, School of Veterinary and Life Sciences, Murdoch University,

Perth, Western Australia 6150, Australia

bDepartment of Environment and Agriculture, School of Science, Curtin University,

Bentley, Western Australia 6102, Australia

Corresponding author details:

Dr. Stephen J. Wylie

Address: Murdoch University, 90 South St, Murdoch, Western Australia 6150

Email: [email protected]; Tel: 61 8 9360 6600

ACCEPTED MANUSCRIPT

1 Highlights

 Identified 22 novel viruses from wild terrestrial orchids and their fungal

symbionts

 First reported virome of plants and their mycorrhizal partners in natural

environment

 Two leaf-derived viruses were more related to mycoviruses than to plant

viruses

 Extends known geographical range of Betapartitivirus, Hypovirus,

Megabirnavirus, Mitovirus and Totivirus

 Inconsistent viral component associated with the same population over

consecutive years

Abstract

Terrestrial orchids represent a symbiotic union between plants and

mycorrhizal fungi. This study describes the occurrence and nature of viruses

associated with one population of wild Pterostylis sanguinea orchids and their fungal

symbionts over two consecutive years. A generic sequencing approach, which

combined dsRNA-enrichment from plant and mycelial tissues, random amplification

and high throughput sequencing, was used to identify novel viruses. The majority of

the virus-like sequences represent partial genomes, and their identification is based ACCEPTEDsolely on de novo assembly of sequencing MANUSCRIPTdata. In orchid leaf tissues we found three isolates of a novel totivirus and an unclassified virus; both resemble fungus-infecting

viruses. Two isolates of Ceratobasidium sp were isolated from orchid underground

stems contained at least 20 viruses, 16 of which were previously described as

2 alphapartitiviruses and betapartitiviruses. A hypovirus and a mitovirus were

genetically distant from existing members of the genera and did not readily fit into

recognised subgroups.

Keywords:

Mycorrhizal fungus; mycovirus; plant virus; Pterostylis orchid; virus taxonomy;

virome

ACCEPTED MANUSCRIPT

3 1. Introduction

Pterostylis is a genus comprising of over 200 species terrestrial orchids

indigenous to Australia, Indonesia, New Caledonia, New Zealand and Papua New

Guinea (Janes & Duretto, 2010; Janes et al., 2010; Brundrett, 2014). Pterostylis

orchids have short roots ranging from 5-10 cm in length and they form obligate

associations with compatible fungi that provide water and nutrients from beyond the

rhizosphere (Ramsay et al., 1986; Ramsay et al., 1987). Australian terrestrial orchids

differ from other composite organisms such as lichens in that the relationship is

broken annually when the plant partner enters its dormant below-ground phase, and it

becomes re-established when the shoot re-emerges, which may occur up to several

years later, depending on environmental conditions (Brundrett, 2014). Pterostylis

plants always establish mycorrhizal relationships with species of Ceratobasidium

fungi (Warcup, 1973; Bonnardeaux et al., 2007), but it is unclear if the same species

or strain of fungus re-establishes the relationship each year.

The viruses of cultivated orchids are widely studied. The most common

viruses are Cymbidium mosaic virus (CymMV) and Odontoglossum ringspot virus

(ORSV) from families Alphaflexiviridae and Virgaviridae, respectively (Zettler et al.,

1990). Transmission vectors for these two viruses remain unknown but they can be

spread through vegetative propagation and through trade of infected plants (Jensen,

1952; Blanchfield et al., 2001). In contrast, viruses infecting wild orchids are far less ACCEPTEDwell known. Exotic and indigenous viruses MANUSCRIPT from five genera (Dichorhavirus, Orthotospovirus, Potexvirus, Potyvirus and Tobamovirus) were described from a

mixture of captive and wild orchids in eastern Australia (Gibbs et al., 2000). In

Western Australia, wild orchids and their fungal symbionts were infected with exotic

4 and indigenous viruses of Alphapartitivirus, Betapartitivirus, Divavirus,

Endornavirus, Goravirus, Platypuvirus, Polerovirus and Potyvirus (Wylie et al.,

2012; Wylie et al., 2013a; Wylie et al., 2013b; Ong et al., 2016a; Ong et al., 2016b;

Ong et al., 2017). In Japan, wild Calanthe izu-insularis orchids were infected with

cucumber mosaic virus (genus Cucumovirus) (Kawakami et al., 2007). In India,

isolates of ORSV (genus Tobamovirus), CymMV (genus Potexvirus) and a novel

potyvirus infected wild epiphytic orchids (Sherpa et al., 2006; Singh et al., 2007).

In the current study, a generic approach based on high throughput sequencing

was used to identify RNA viruses infecting plants of Pterostylis sanguinea (dark

banded greenhood orchid) and mycorrhizal fungi associated with them. The samples

were collected from a small natural population over two consecutive years. These

orchids generate one or more new underground tubers each year, and the old tuber

dies. Tubers typically germinate unevenly, with some remaining dormant from one to

several years (Brundrett, 2014). Sixteen partitiviruses (of Alphapartitivirus and

Betapartitivirus) were previously detected and characterised from the same orchid

mycorrhizae (Ong et al., 2017). In this report, we describe the novel non-partitiviral

viruses identified from them and discuss them in ecological and evolutionary

contexts.

2. Materials and methods ACCEPTED Leaves and underground stems were MANUSCRIPT collected from P. Sanguinea orchids from their natural habitat in Western Australia. The subsequent experiments were carried

out as described in Ong et al. (2017). Briefly, nucleic acids were extracted from the

leaves and associated mycorrhizal fungi using a cellulose-based method (Morris &

5 Dodds, 1979). DNA libraries were made from randomly primed cDNA. Amplicons

were barcoded and pooled before high-throughput sequencing on an Illumina Hi-Seq

platform. De novo assembly of contigs was done using CLC Genomic Workbench

v6.5.1 (Qiagen), and subsequent analysis was done using Geneious v7.0.6

(Biomatters) and MEGA v6.06 software.

3. Results

3.1. De novo assembly

Three datasets of 153,581,198 and 92,046,118 and 35,630,376 reads, each of

101 nt, were generated from three independent Illumina sequencing runs. The first

dataset included 4,875,054 reads from Ceratobasidium sp. F-2012. Samples F-2012

and P-2012 were sequenced as part of the second dataset and generated 2,854,974 and

5,725,698 reads respectively. F-2012 was re-sequenced to improve the coverage of

the viral sequences. The third sequencing run generated 409,030 reads from P-2012

and 604,012 reads from F-2013. The remaining unaccounted reads were from other

barcoded samples that are unrelated to this project. De novo assembly of the datasets

generated a total of 119,854 contigs (12,896; 72,206 and 34,752 from each respective

dataset) ranging from 200 nt to 22,685 nt, with a N50 length of 403, 359 and 305

respectively. Of these, 90 contigs were identified as virus-like; 14 were derived from

P. sanguinea plants (474-10,716 nt) and 76 were from mycorrhizal fungi (407-8227

nt). ACCEPTED MANUSCRIPT 3.2. Identity of fungi

ITS sequences of both fungal isolates shared 99.7% nt identity, indicative they

were of the same taxon of Ceratobasidium (Genbank accessions KU239992 (F-2012)

6 and KU239993 (F-2013)). The mycorrhizal fungi isolates were not identified to the

level of species. The sequences were closest to a Ceratobasidium isolate from

Pterostylis orchids from eastern Australia (GQ405561; e-value: 0.0, 91% coverage

and 99% nucleotide identity) and to a Ceratobasidium-anamorph, Rhizoctonia sp.

(JQ859901; e-value: 0.0; 100% coverage and 97% nucleotide identity) isolated from a

diseased strawberry root in Western Australia.

3.3. Viruses from orchid-associated mycorrhizal fungi

3.3.1. Ceratobasidium mitovirus 1 (CbMV1): a proposed new mitovirus

A virus-like contig of 2850 nt was identified from Ceratobasidium in 2012

(Table 1). 82,998 sequence reads were mapped to it with pairwise identity of 84.8%,

and the mean coverage of the proposed virus genome was 3573.9-fold. There was a

single ORF (nt 235-2688) whose encoded protein product has a predicted mass of 92

kDa. An RdRp-like domain was identified at aa 228-543 (nt 916-1863) (Fig 1A),

indicative of a replicase function. Further support that the single ORF encoded a

replicase was the existence of six core RdRp motifs between aa 297-508 (nt 1123-

1758) (Hong et al., 1999). The deduced protein sequence shared closest pairwise

identities with replicases of mitoviruses (family Narnaviridae, genus Mitovirus),

which infect the mitochondria of fungi (Hillman & Esteban, 2012). Mitovirus

genomes typically comprise a single non-encapsidated positive-strand RNA of 2.3–

2.9 kb, which encodes a single protein of 80–104 kDa believed to function as a ACCEPTEDreplicase (Hillman & Esteban, 2012). MANUSCRIPT

The genome sequence was most closely related to the mycorrhizal fungus-

infecting mitoviruses Rhizophagus clarus mitovirus 1 (RcMV1-RF1) from Japan

7 (27% aa, 45% nt), Rhizophagus sp. HR1 mitovirus (RMV-HR1) from Japan (27% aa,

45% nt) and Tuber excavatum mitovirus (TeMV) isolated from Germany (20% aa,

43% nt), which together formed groups distinct from currently proposed mitovirus

clades I and II (Fig 2A) (Doherty et al., 2006; Hillman & Cai, 2013). The deduced

protein sequence shared 18-25% aa identity with other mitoviruses, figures below the

accepted species demarcation limit of <40% (Hillman & Esteban, 2012). Thus, we

propose that the sequence represents the complete genome of a previously-

undescribed member of genus Mitovirus that we designate Ceratobasidium mitovirus

1 isolate Murdoch-1 (CbMV1; GenBank accession KU291923), named after the virus

host genus and the location of its discovery.

3.3.2. Ceratobasidium virus 1 (CbV1): a proposed new mycovirus

Three partial monopartite virus sequences of 8227 nt, 7161 nt and 4051 nt

respectively were identified from the Ceratobasidium isolate collected in 2012 (Fig

1B; Table 1). These sequences were designated Ceratobasidium virus 1 (CbV1)

isolates Murdoch-2, Murdoch-3 and Murdoch-4 (GenBank accessions KU291947,

KU291948 and KU291949 respectively). The sequence of isolate Murdoch-2 was

constructed from 9593 raw sequence reads, with mean pairwise identity of 82.2% and

a 119.6-fold mean coverage per base across the genome. 6792 reads were mapped to

isolate Murdoch-3 with mean pairwise identity of 82.0% and mean coverage of 101.2-

fold per base. Isolate Murdoch-4 was generated from 2558 reads of mean pairwise ACCEPTEDidentity of 80.5% and mean coverage of 71.6MANUSCRIPT-fold per base. CbV1 isolates Murdoch-2 and Murdoch-3 had two non-overlapping ORFs – a complete ORF1 (177 kDa) and

partial ORF2 (RdRp; >74kDa and >115 kDa) while the sequence of Murdoch-4,

which represented about 45% of its genome, encoded an incomplete RdRp (>130

8 kDa). Neither a ‘slippery sequence’ nor pseudoknot, typical of ribosomal frameshift

sites (Brierley et al., 1992), was detected upstream of proposed ORF2. Comparison of

the isolates showed 43-80% nt identity between genomes, 91% aa identity (80% nt)

between the ORF1s, and 53-95% aa identity (59-81% nt) between the respective

RdRps.

Blastp analysis showed that about 10% of the translated product of ORF1s of

isolates Murdoch-2 and Murdoch-3 matched the transposase of Desulfovibrio

oxyclinae (Table 1). The encoded CbV1 RdRps shared highest identity with RdRp of

Rosellinia necatrix mycovirus 1-W1032/S5 (also named Yado-nushi virus W1032a;

Zhang et al., 2016) at a pairwise identity of 31-33% aa and 47-50% nt. CbV1 and

Rosellinia necatrix mycovirus 1-W1032/S5 were grouped together and distantly

related to other unclassified mycoviruses (Fig 2B).

3.3.3. Ceratobasidium virus 2 (CbV2): a proposed new mycovirus

A contig of 7089 nt representing Ceratobasidium virus 2 (CbV2) isolate

Murdoch-5 (GenBank accession KU291938) was constructed from 8508 reads, with

mean coverage of 127.3-fold per base across the genome and mean pairwise identity

of 46.1%. The partial genome of CbV2 encoded two non-overlapping ORFs

representing a hypothetical protein (nt 1-3340; >117 kDa) and an RdRp (nt 3604-

7089; >31 kDa) (Fig 1C). There was no evidence of ribosomal frameshift ‘slippery ACCEPTEDsequence’ sites upstream of the junction ofMANUSCRIPT the two putative ORFs. In ORF2, an RdRp_4 domain (pfam02123) was identified at aa 597-866 (nt 5392-6201) (Fig 1C)

and the core RdRp motifs V and IV of T/SGx3 Tx3 NS/Tx22 GDD (where x is any

9 residue) (Koonin, 1991) were represented at aa 758-800 (nt 5875-6003) as SGx3 Tx3

NTx29 GDD.

Blast and phylogenetic analyses showed that CbV2 grouped most closely with

an unclassified mycovirus Rhizoctonia solani RNA virus HN008 (RsRV-HN008;

Zhong et al., 2015) (Fig 2B, Table 1). The two viruses shared 45% nt identity

between genomes, 16% aa (43% nt) identity between ORF1s and 31% aa (48% nt)

identity between ORF2s.

3.3.4. Ceratobasidium hypovirus 1 (CbHV1): a proposed new hypovirus

A contig of 7143 nt was generated from 5051 sequence reads with mean

coverage of 74.7-fold and pairwise identity of 85.5%. The partial genome sequence

has a 5' untranslated region (UTR) of 89 nt and two predicted ORFs (Fig 1D). ORF1

was 3636 nt in length and is predicted to encode a protein of 133 kDa. The 3' part of

the genome was not obtained, and so ORF2 is incomplete (nt 3888-7143; >121 kDa).

Based on its length compared to related hypoviruses and the position of RdRp

domain, the partial ORF2 sequence represents about 60% of its complete ORF.

Elements resembling slippery sequences and pseudoknots (Brierley et al., 1992)

upstream of ORF2 were absent. The RdRp domain was located at aa 582-677 (nt

5631-5918) (Fig 1D) and the core RdRp motifs V and IV were represented at aa

residues 590-634 (nt 5655-5789) as TGx3 Tx3 NTx31 GDD. ACCEPTED MANUSCRIPT The virus represented by this sequence was designated Ceratobasidium

hypovirus 1 (CbHV1) isolate Murdoch-6 (GenBank accession KU291924). We

propose CbHV1 as a new hypovirus; the first hypovirus to be detected from a

10 basidiomycete host. CbHV1 shared highest aa identity (15-16%) with two definitive

members of Hypovirus (Cryphonectria hypovirus 1-2, CHV1-2; family Hypoviridae)

that infect Cryphonectria parasitica, the Chestnut blight fungus (Shapira et al., 1991;

Hillman et al., 1994) and two other proposed members infecting Fusarium species

(Fusarium graminearum hypovirus 1 (FgHV1) from China (Wang et al., 2013)) and

Sclerotinia species (Sclerotinia sclerotiorum hypovirus 2 (SsHV2) from New Zealand

(Khalifa and Pearson, 2014)). Hypoviruses are proposed to be categorised into

subgroups Alphahypovirus and Betahypovirus, distinguished by having either two or

one ORF, respectively (Nuss & Hillman, 2012; Yaegashi et al., 2012). A third

subgroup, Gammahypovirus, has been tentatively proposed by Khalifa and Pearson

(2014) to categorise SsHV2. Having two ORFs, CbHV1 was expected to share greater

sequence identity with alphahypoviruses than betahypoviruses, but phylogenetic

analysis positions CbHV1 distantly from members of each group (Fig 2C).

Comparison of CbHV1 with the hypoviruses showed similar nt identity at 41-43%

and low aa identity; 12-16% aa identity with alphahypoviruses and 11-13% aa identity

with betahypoviruses. This is consistent with the species demarcation limit of less

than 60% aa identity between CHV-1 and CHV-2 and 50% aa identity between CHV-

3 and CHV-4 (Nuss & Hillman, 2012).

3.4. Virus-like sequences identified from leaf samples

Four distantly related virus-like sequences were identified from P. sanguinea ACCEPTEDleaf tissue samples, one in 2012 and three MANUSCRIPTin 2013. Because these viruses resembled mycoviruses, but not known plant viruses, attempts to amplify fungus sequences from

the leaf samples using primers ITS1 and ITS4 were carried out. However this test did

11 not detect fungi from the leaf. This evidence suggested that these two mycovirus-like

viruses may indeed be plant viruses.

3.4.1. Pterostylis sanguinea virus A (PsVA), a mycovirus-like virus from leaf tissue

A partial monopartite virus genome sequence of 10,716 nt was identified from

leaf tissue in 2012 (Fig 1E; Table 1). 56,636 101 nt reads were mapped to the

sequence, with a pairwise identity of 87.4%, and a mean coverage of 1444.8-fold

across its partial genome. This sequence was named Pterostylis sanguinea virus A

(PsVA) isolate Murdoch-7 (GenBank accession KU291925).

The PsVA genome has two consecutive non-overlapping ORFs of 5739 nt and

3758 nt, respectively (Fig 1E). Ribosomal frameshift was not detected in this

sequence. PsVA is predicted to have an unusually long 5' UTR of 1154 nt. The first

putative translational start codon is positioned at nt 1155 corresponding to the start of

the hypothetical CP, estimated to have a mass of 208 kDa (Fig 1E). A region

encoding a Nudix hydrolase-like domain, responsible for hydrolysis of nucleoside

diphosphate derivatives, was detected at aa residues 177-297 (nt 1683-2045) (Fig 1E).

In ORF2, an RdRp domain, which corresponds to similar domains in viruses

belonging of Chrysovirus, Luteovirus, Rotavirus and Totivirus, was detected at aa

residues 934-1209 (nt 9758-10,585) (Marchler-Bauer & Bryant, 2004; Marchler-

Bauer et al., 2013). The highly conserved core RdRp motifs V and VI (S/TGx3 Tx3 ACCEPTEDNS/Tx22 GDD) (Koonin, 1991) were present MANUSCRIPT at aa 1109-1151 (nt 10,283-10,411) as SGx3 Tx4 NTx28 GDD.

12 ORF1 shared highest identity with the CP-encoding ORF1 of the monopartite

mycovirus Lentinula edodes spherical virus (LeSV; an unclassified virus), identified

from Shiitake mushroom (Lentinula edodes) in South Korea (Won et al., 2013). The

PsVA RdRp showed highest identity to RdRps of Lentinula edodes mycovirus

isolates HKA and HKB (LeV; unclassified), also from Shiitake mushroom but from

Japan (Ohta et al., 2008; Magae, 2012), the saprophytic fungus virus Phlebiopsis

gigantea large virus-1 (PgLV-1; unclassified; Kozlakidis et al., 2009) and mycorrhizal

fungus infecting Thelephora terrestris virus 1 (TtV1; unclassified; Petrzik et al., 2016)

(Fig 2B; Table 1). These unclassified viruses shared similar genome organisation of

unipartite RNA genomes encoding two ORFs and are phylogenetically grouped

within a new proposed genus, Phlegivirus (Fig 2B; Petrzik et al., 2016). With the

exception of PsVA, all Phlgiviruses are detected from basidiomycetous hosts.

ClustalW comparison of PsVA, LeV, PgLV-1 and TtV1 isolates showed 41-43% nt

identity across the genomes. Identity between the homologous proteins of PsVA,

LeV, PgLV-1 and TtV1 was 19-20% aa identity (41-43% nt) between CPs and 27-

30% aa identity (41-46% nt) between RdRps.

3.4.2. Pterostylis sanguinea totivirus A (PsTVA): three isolates of a proposed new

totivirus from orchid plants

Three related contigs of 4631 nt, 3631 nt and 3613 nt were identified from leaf

tissue of P. sanguinea plants collected in 2013 (Table 1). The sequences resembled ACCEPTEDthose of members of genus Totivirus. Totiviruses MANUSCRIPT have dsRNA genomes consisting of a single molecule 4.6-7.0 kbp in length that encodes two, usually overlapping, ORFs

(Wickner et al., 2012). The three putative totivirus isolates encoded by these three

partial genome sequences were designated Pterostylis sanguinea totivirus A (PsTVA)

13 isolates Murdoch-8, Murdoch-9 and Murdoch-10 (GenBank accessions KU291927,

KU291926 and KU291928 respectively). 2970 raw sequence reads were mapped to

PsTVA-Murdoch-8 with pairwise identity of 81.7% and 61.9-fold mean coverage per

base across its genome. Isolate Murdoch-9 was assembled from 94,091 reads at

pairwise identity of 81.9% and mean coverage of 3142.8-fold per base. 68,653 reads

were mapped to isolate Murdoch-10 with pairwise identity of 81.9% and mean

coverage per base of 3153.0-fold. Each of the three sequences encoded two non-

overlapping partial ORFs representing a CP and RdRp (Fig 1F), with no evidence of a

slippery sequence indicating ribosomal frameshifting. The CPs had a L-A CP domain,

which is typical of the yeast-infecting Totivirus type species Saccharomyces

cerevisiae L-A virus (ScV-L-A). The L-A CP domain was located at nt residues 40-

1212 (Murdoch-8), 1-245 (Murdoch-9) and 9-1142 (Murdoch-10). RdRp-4-like

domains were detected at nt 3000-4088 (Murdoch-8), nt 2009-3097 (Murdoch-9) and

nt 2948-3589 (Murdoch-10). Conserved RdRp core motifs V and VI (Koonin, 1991)

were located on the genomes of isolate Murdoch-8 at aa residues 527-564 (nt 3774-

3887) and isolate Murdoch-9 at aa 403-440 (nt 2783-2892) as SGx3 Tx3 NTx24

GDD. Comparison of the isolates showed 42-60% nt identity between genomes, 36-

44% aa identity (48-56% nt) between CPs and 62-65% aa identity (61-64% nt)

between RdRps. These figures are slightly below or above the suggested species

demarcation limit for totivirus species of <50% aa identity (Wickner et al., 2012),

indicating they may be categorised as the same species. ACCEPTED MANUSCRIPT Blastp analysis of the two proteins revealed that the closest matches to

described species were to the CP (32-37% aa identity) and RdRp (56-58% aa identity)

of black raspberry virus F (BRVF; GenBank accession NC_009890), a proposed

14 totivirus described from leaf tissue or fungus-infected leaf tissue of Rubus

occidentalis in the USA. PsTVA also grouped with totiviruses that infect fungi and

yeast (Fig 2D).

3.5. Other virus-like sequences

In addition to the viruses described above, there is evidence from 30 short

sequence fragments that were 454-3118 nt in length that other viruses were also

present (Table S1). These are not described in detail here because they were estimated

to represent less than 50% of genomes, thereby making assignment to taxonomic

groups speculative. Analysis using Blastp revealed they most closely matched viruses

from five virus families and five genera, and some were unclassified (Table S1). From

leaf samples, a megabirnavirus-like sequence (P-2012), four other totivirus-like

sequences (P-2013) and two related to viruses of the family Amalgaviridae (P-2013)

were identified (Table S1). The totivirus-like contigs were closely related to the three

PsTVA isolates and may represent more isolates of PsTVA, or belong to related

species. They shared 41-95% nt identity between genomes, 44-95% nt (12-93% aa)

identity between RdRp sequences and 48-57% nt (11-49% aa) identity between CP

sequences.

From the two fungal isolates, a further 23 virus-like contigs were identified

(454-3118 nt) that most closely resembled species within the genera Endornavirus, ACCEPTEDHypovirus, Megabirnavirus and unclassified MANUSCRIPT mycoviruses (Table S1). The four endornavirus-like contigs (454-1245 nt) detected in Ceratobasidium sp. (F-2013)

were matched to endornaviruses recently identified from mycorrhizal fungi of other

terrestrial orchids in the region (Ong et al., 2016b). RT-PCR was done using primers

15 specific to Ceratobasidium endornaviruses A-H, and they confirmed the presence of

Ceratobasidium endornaviruses G and H (data not shown) (Ong et al., 2016b).

4. Discussion

A small wild population of three and four P. sanguinea plants collected in

2012 and 2013 and mycorrhizal fungi associated with two of the plants were found to

be colonised by numerous persistent viruses, none of which had been described

previously. In addition to the six viruses described above, there were

alphapartitiviruses, betapartitiviruses and partitivirus-like sequences (family

Partitiviridae) associated with the same mycorrhizal fungi (Ong et al., 2017). There

were at least 10 partitiviruses – seven alphapartitiviruses and three betapartitiviruses –

found in fungal isolate F-2012. From isolate F-2013, five alphapartitiviruses and one

betapartitiviruses initially identified by Illumina sequence analysis. The majority of

these partitiviruses were subsequently detected in both mycorrhizal strains using

virus-specific primers (Ong et al., 2017). In total, at least 22 definitive viruses,

proposed as belonging to the genera Alphapartitivirus, Betapartitivirus, Hypovirus,

Mitovirus, Totivirus and unclassified mycoviruses were identified (Ong et al., 2017).

All but two of the new viruses were identified from pure cultures of Ceratobasidium

derived from pelotons isolated from two P. sanguinea plants. The findings extend the

geographical range of probable members of Betapartitivirus (Ceratobasidium

partitiviruses; Ong et al., 2017), Hypovirus (CbHV1; Ceratobasidium hypovirus-like ACCEPTEDcontig 1-5), Megabirnavirus (Ceratobasidium MANUSCRIPT megabirnavirus-like contigs 1-2; Pterostylis megabirnavirus-like contig 1), Mitovirus (CbMV1) and Totivirus (PsTV1;

Ceratobasidium totivirus-like contigs 1-4), which had not previously been identified

from Australia. Although tentatively assigned classification by phylogenetic analysis

16 with members of known groups, the proposed classifications are by no means certain

because many sequences represented partial genomes, and most were genetically

distant from described species.

4.1. Classification of new viruses

Most of the new viruses were tentatively classified with existing higher order

taxa, but assigning them to existing lower order taxa was often problematical. For

example, CbMV1 was proposed as a member of Mitovirus, but it does not fit easily

into the two proposed subgroups within the genus (Doherty et al., 2006). Instead it

groups with other unclassified mycorrhizae-derived mitoviruses (Fig 2A) that usually

encode tryptophan with UGG rather than UGA, which confers on them the capability

of replicating in the host cytoplasm as well as its mitochondria (Kitahara et al., 2014).

Similarly, the proposed hypovirus CbHV1 is phylogenetically closest to the

hypoviruses, but features of its genome organisation and host type (basidiomycete-

infecting) place it outside the existing two hypovirus subgroups (Yaegashi et al.

2012). Mycovirus-like PsVA, detected in leaf tissue, was grouped with unclassified

mycoviruses in a new proposed genus, Phlegivirus. The two mycoviruses, CbV1 and

CbV2 share sequence identity and genome organisation with other mycoviruses from

different continents, but none are currently assigned taxa. Together, these findings

indicate that the evolutionary history of mycoviruses is more complex than currently

recognised. ACCEPTED MANUSCRIPT PsTVA is the only new virus that shared a relatively close evolutionary history

with a previously described virus – the proposed totivirus black raspberry virus F.

Sequence identities of these two viruses are marginally above the species demarcation

17 threshold of 50% aa identity for totiviruses set by the ICTV (Wickner et al., 2012),

but given that the host species are distinct and their locations are widely separated, we

propose that they belong to different species.

4.2. Host identification

Viruses identified from plant materials are usually assumed to be plant viruses,

with no distinction being made between viruses capable of replicating in plant cells or

fungal cells that co-occur with plant tissues. It can be difficult to ascertain the true

host. The current experiment was designed to detect RNA viruses that have dsRNA

genomes, and ssRNA viruses that have a dsRNA replicative form. An alternative

approach would be to sequence total leaf RNA (but depleted of ribosomal RNA) to

eliminate the bias towards detection of dsRNA viral genomes and provide a clearer

representation of the entire virome, including the presence of fungal transcripts. This

metatranscriptomic approach was used recently in a study by Marzano et al. (2015),

which identified 22 putative mycoviruses (dsRNA, ssRNA and ssDNA) from soybean

leaf samples.

4.3. Viruses, fungi and orchids

In orchid biology, the plant-fungus symbiotic partnership is critical, but the

roles viruses may play in this relationship remain largely unknown. Most mycoviruses

appear to have little influence on fungal pathogenicity (Seo et al., 2004; Vainio et al., ACCEPTED2012), but some are demonstrated to influ enceMANUSCRIPT their hosts. The most well-known example of a mycovirus that reduces fungal pathogenicity is CHV1 of Cryphonectria

parasitica, the fungal pathogen that causes chestnut blight (Anagnostakis & Day,

1979). The term ‘Rhizoctonia decline’ was used to describe the decreased growth rate

18 and lack of sclerotia production of an isolate of Rhizoctonia solani infected with a

dsRNA virus (Castanho & Butler, 1978a; Castanho and Butler, 1978b). In contrast,

the plant pathogenic fungus Nectria radicicola became more virulent in the presence

of a 6.0 kbp dsRNA virus (Ahn & Lee, 2001).

Ecological roles of mycoviruses may be elucidated when fungal cultures are

cured of persistent viruses so that isogenic lines with and without mycoviruses can be

compared. It has been reported that in vitro culturing of some fungi can cause them to

lose mycoviruses (Bao & Roossinck, 2013; Roossinck, 2015). Because the

Ceratobasidium isolates analysed here were cultured in vitro, this study may provide

an underestimate of the number of mycoviruses that exist under natural conditions.

Ecological factors, such as drought, fire, grazing, low seed set, salinity of

habitats and weed invasion, are known to impact negatively on orchid populations,

and this is the case with many of the threatened orchid species in Western Australia

(Coates & Atkins, 2001; Swarts & Dixon, 2009; Brundrett, 2016). Whether viruses

play positive or negative roles in orchid biology remains unclear. The identification of

viruses is a step in on-going studies of the interplay between wild plants, fungi and

viruses. Such studies may shed light on understanding why populations of many

orchids in south-west Australia (and globally) are shrinking alarmingly, while other

orchid species are thriving to the point of becoming weeds, e.g. Microtis media and ACCEPTEDDisa bracteata (Bonnardeaux et al., 2007; MANUSCRIPT Swarts & Dixon, 2009; De Long et al., 2013). More broadly, it may also provide clues to understanding the roles, positive or

negative, that mycoviruses play in fungal interactions in other natural systems, and

even in artificial systems such as agriculture.

Acknowledgement

This work was supported by Australian Research Council Linkage [Grant

LP110200180] in collaboration with The Botanic Gardens and Parks Authority and

The Australian Orchid Foundation.

ACCEPTED MANUSCRIPT

20 References

Ahn, I-P and Y-H Lee. 2001. A viral double-stranded RNA up regulates the fungal

virulence of Nectria radicicola. Mol. Plant Microbe Interact. 14(4): 496-507.

Anagnostakis, SL and PR Day. 1979. Hypovirulence conversion in Endothia

parasitica. Phytopathology 69(12): 1226-1229.

Bao, X and MJ Roossinck. 2013. Multiplexed interactions: viruses of endophytic

fungi. Adv. Virus Res. 86: 37-58.

Blanchfield, AL, AM Mackenzie, A Gibbs, H Kondo, T Tamada and CR Wilson.

2001. Identification of Orchid fleck virus by reverse transcriptase‐polymerase chain

reaction and analysis of isolate relationships. J. Phytopathol. 149(11‐12): 713-718.

Bonnardeaux, Y, M Brundrett, A Batty, K Dixon, J Koch and K Sivasithamparam.

2007. Diversity of mycorrhizal fungi of terrestrial orchids: compatibility webs, brief

encounters, lasting relationships and alien invasions. Mycol. Res. 111(1): 51-61.

Brierley, I, AJ Jenner and SC Inglis. 1992. Mutational analysis of the “slippery-

sequence” component of a coronavirus ribosomal frameshifting signal. J. Mol. Biol. ACCEPTED227(2): 463-479. MANUSCRIPT

Brundrett, MC 2014. Identification and Ecology of Southwest Australian Orchids.

Perth: Western Australian Naturalists’ Club.

21 Brundrett, MC. 2016. Using vital statistics and core-habitat maps to manage critically

endangered orchids in the Western Australian wheatbelt. Aust. J. Bot. 64(1): 51-64.

Castanho, B and EE Butler. 1978a. Rhizoctonia decline: a degenerative disease of

Rhizoctonia solani. Phytopathology 68(10): 1505-1510.

Castanho, B and EE Butler. 1978b. Rhizoctonia decline: studies on hypovirulence and

potential use in biological control. Phytopathology 68(10): 1511-1514.

Coates, DJ and KA Atkins. 2001. Priority setting and the conservation of Western

Australia's diverse and highly endemic flora. Biol. Conserv. 97(2): 251-263.

De Long, JR, ND Swarts, KW Dixon and LM Egerton-Warburton. 2013. Mycorrhizal

preference promotes habitat invasion by a native Australian orchid: Microtis media.

Ann. Bot. 111(3): 409-418.

Doherty, M, RHA Coutts, CM Brasier and KW Buck. 2006. Sequence of RNA-

dependent RNA polymerase genes provides evidence for three more distinct

mitoviruses in Ophiostoma novo-ulmi isolate Ld. Virus Genes 33(1): 41-44.

Gibbs, MJ, R Koga, H Moriyama, P Pfeiffer and T Fukuhara. 2000. Phylogenetic ACCEPTEDanalysis of some large double-stranded RNA MANUSCRIPT replicons from plants suggests they evolved from a defective single-stranded RNA virus. J. Gen. Virol. 81(1): 227-233.

Hillman, BI and G Cai. 2013. The family Narnaviridae: simplest of RNA viruses.

22 Adv. Virus. Res. 86: 149-176.

Hillman, BI and R Esteban. 2012. Narnaviridae. Virus Taxonomy: Classification and

Nomenclature of Viruses: Ninth Report of the International Committee on Taxonomy

of Viruses. AMQ King, MJ Adams, EB Carstens and EJ Lefkowitz, 1055-1060. San

Diego: Elsevier Academic Press.

Hillman, BI, BT Halpern and MP Brown. 1994. A viral dsRNA element of the

chestnut blight fungus with a distinct genetic organization. Virology 201(2): 241-250.

Hong, Y, SL Dover, TE Cole, CM Brasier and KW Buck. 1999. Multiple

mitochondrial viruses in an isolate of the Dutch Elm disease fungus Ophiostoma

novo-ulmi. Virology 258(1): 118-127.

Janes, JK and MF Duretto. 2010. A new classification for subtribe Pterostylidinae

(Orchidaceae), reaffirming Pterostylis in the broad sense. Aust. Syst. Bot. 23(4): 260-

269.

Janes, JK, DA Steane, RE Vaillancourt and MF Duretto. 2010. A molecular

phylogeny of the subtribe Pterostylidinae (Orchidaceae): resolving the taxonomic

confusion. Aust. Syst. Bot. 23(4): 248-259. ACCEPTED MANUSCRIPT Jensen, DD. 1952. Virus diseases of orchids: symptoms, spread, host range,

distribution, and control practices determined by experimental transmissions. Calif.

Agric. 6(1): 3-15.

23 Kawakami, K, S Fuji and K Miyoshi. 2007. Endangered wild populations of endemic

Calanthe orchids on an isolated Japanese island tested for viruses. Aust. J. Bot. 55(8):

831-836.

Khalifa

Kitahara, R, Y Ikeda, H Shimura, C Masuta and T Ezawa. 2014. A unique mitovirus

from Glomeromycota, the phylum of arbuscular mycorrhizal fungi. Arch. Virol.

159(8): 2157-2160.

Koonin, EV. 1991. The phylogeny of RNA-dependent RNA polymerases of positive-

strand RNA viruses. J. Gen. Virol. 72: 2197-2206.

Kozlakidis, Z, CV Hacker, D Bradley, A Jamal, X Phoon, J Webber, CM Brasier, KW

Buck and RHA Coutts. 2009. Molecular characterisation of two novel double-

stranded RNA elements from Phlebiopsis gigantea. Virus Genes 39(1): 132-136.

Magae, Y. 2012. Molecular characterization of a novel mycovirus in the cultivated

mushroom, Lentinula edodes. Virol. J. 9: 60.

Marchler-Bauer, A and SH Bryant. 2004. CD-Search: protein domain annotations on ACCEPTEDthe fly. Nucleic Acids Res. 32(Web Server MANUSCRIPT issue): W327-331.

Marchler-Bauer, A, C Zheng, F Chitsaz, MK Derbyshire, LY Geer, RC Geer, NR

Gonzales, M Gwadz, DI Hurwitz, CJ Lanczycki, F Lu, S Lu, GH Marchler, JS Song,

24 N Thanki, RA Yamashita, D Zhang and SH Bryant. 2013. CDD: conserved domains

and protein three-dimensional structure. Nucleic Acids Res. 41(D1): D348-352.

Marzano, S-YL and LL Domier. 2016. Novel mycoviruses discovered from

metatranscriptomics survey of soybean phyllosphere phytobiomes. Virus Res. 213:

332-342.

Morris, TJ and JA Dodds. 1979. Isolation and analysis of double-stranded RNA from

virus-infected plant and fungal tissue. Phytopathology 69: 854-858.

Nuss, DL and BI Hillman. 2012. Hypoviridae. Virus Taxonomy: Classification and

Nomenclature of Viruses: Ninth Report of the International Committee on Taxonomy

of Viruses. AMQ King, MJ Adams, EB Carstens and EJ Lefkowitz, 1029-1033. San

Diego: Elsevier Academic Press.

Ohta, C, T Taguchi, S Takahashi, K Ohtsuka, K Eda, S Ayusawa and Y Magae. 2008.

Detection of double stranded RNA elements in cultivated Lentinula edodes.

Mushroom Science and Biotechnology 16: 155-158.

Ong, JWL, H Li, K Sivasithamparam, KW Dixon, MGK Jones and SJ Wylie. 2016b.

Novel Endorna-like viruses, including three with two open reading frames, challenge ACCEPTEDthe membership criteria and taxonomy of theMANUSCRIPT Endornaviridae. Virology 499: 203-211.

Ong, JWL, H Li, K Sivasithamparam, KW Dixon, MGK Jones and SJ Wylie. 2017.

The challenges of using high-throughput sequencing to track multiple new bipartite

25 viruses of wild orchid-fungus partnerships over consecutive years. Virology 510: 297-

304.

Ong, JWL, RD Phillips, KW Dixon, MGK Jones and SJ Wylie. 2016a.

Characterization of the first two viruses described from wild populations of hammer

orchids (Drakaea spp.) in Australia. Plant Pathology 65(1): 163-172.

Petrzik, K, T Sarkisova, J Starý, I Koloniuk, L Hrabáková and O Kubešova. 2016.

Molecular characterization of a new monopartite dsRNA mycovirus from

mycorrhizal Thelephora terrestris (Ehrh.) and its detection in soil oribatid mites

(Acari: Oribatida). Virology 489: 12-19.

Ramsay, RR, KW Dixon and K Sivasithamparam. 1986. Patterns of infection and

endophytes associated with Western Australian orchids. Lindleyana 1(3): 203-214.

Ramsay, RR, K Sivasithamparam and KW Dixon. 1987. Anastomosis groups among

Rhizoctonia-like endophytic fungi in southwestern Australian Pterostylis species

(Orchidaceae). Lindleyana 2: 161-166.

Roossinck, MJ. 2015. Metagenomics of plant and fungal viruses reveals an abundance

of persistent lifestyles. Front Microbiol. 5: 767. ACCEPTED MANUSCRIPT Seo, J-K, W-S Lim, J-H Jeong, Y-B Yoo, S-W Yie and K-H Kim. 2004.

Characterization and RT-PCR detection of dsRNA mycoviruses from the oyster

mushroom, Pleurotus ostreatus. Plant Pathol. J. 20(3): 200-205.

26 Shapira, R, GH Choi and DL Nuss. 1991. Virus-like genetic organization, expression

strategy for a double-stranded RNA genetic element associated with biological control

of chestnut blight. EMBO J 10: 731-739.

Sherpa, A, V Hallan, P Pathak and A Zaidi. 2006. Characterization of the coat protein

gene of Cymbidium mosaic virus isolates from India. J. Phytopathol. 154(5): 275-280.

Singh, MK, AR Sherpa, V Hallan and AA Zaidi. 2007. A potyvirus in Cymbidium

spp. in northern India. Australas. Plant Dis. Notes 2(1): 11-13.

Swarts, ND and KW Dixon. 2009. Terrestrial orchid conservation in the age of

extinction. Ann. Bot. 104(3): 543-556.

Vainio, EJ, R Hyder, G Aday, E Hansen, T Piri, T Doğmuş-Lehtijärvi, A Lehtijärvi, K

Korhonen and J Hantula. 2012. Population structure of a novel putative mycovirus

infecting the conifer root-rot fungus Heterobasidion annosum sensu lato. Virology

422(2): 366-376.

Wang, S, H Kondo, L Liu, L Guo and D Qiu. 2013. A novel virus in the family

Hypoviridae from the plant pathogenic fungus Fusarium graminearum. Virus Res.

174(1): 69-77. ACCEPTED MANUSCRIPT Warcup, J. 1973. Symbiotic germination of some Australian terrestrial orchids. New

Phytol. 72(2): 387-392.

27 Wickner, RB, SA Ghabrial, ML Nibert, JL Patterson and CC Wang. 2012.

Totiviridae. Virus taxonomy: classification and nomenclature of viruses: Ninth Report

of the International Committee on Taxonomy of Viruses. AMQ King, MJ Adams, EB

Carstens and EJ Lefkowitz, 639-650. San Diego: Elsevier Academic Press.

Won, H-K, S-J Park, D-K Kim, MJ Shin, N Kim, S-H Lee, Y-C Kwon, HK Ko, H-S

Ro and H-S Lee. 2013. Isolation and characterization of a mycovirus in Lentinula

edodes. J. Microbiol. 51(1): 118-122.

Wylie, SJ, H Li, KW Dixon, H Richards and MGK Jones. 2013a. Exotic and

indigenous viruses infect wild populations and captive collections of temperate

terrestrial orchids (Diuris species) in Australia. Virus Res. 171(1): 22-32.

Wylie, SJ, H Li and MGK Jones. 2013b. Donkey orchid symptomless virus: a viral

‘platypus’ from Australian terrestrial orchids. PLoS One 8(11): e79587.

Wylie, SJ, AJY Tan, H Li, KW Dixon and MGK Jones. 2012. Caladenia virus A, an

unusual new member of the family Potyviridae from terrestrial orchids in Western

Australia. Arch. Virol. 157(12): 2447-2452.

Yaegashi, H, S Kanematsu and T Ito. 2012. Molecular characterization of a new ACCEPTEDhypovirus infecting a phytopathogenic fungus, MANUSCRIPT Valsa ceratosperma. Virus Res. 165(2): 143-150.

Zettler, F, N Ko, G Wisler, M Elliott and S Wong. 1990. Viruses of orchids and their

28 control. Plant Dis. 74(9): 621-626.

Zhang, R, S Hisano, A Tani, H Kondo, S Kanematsu and N Suzuki. 2016. A

capsidless ssRNA virus hosted by an unrelated dsRNA virus. Nat. Microbiol. 1:

15001.

Zhong, J, C-Y Chen and B-D Gao. 2015. Genome sequence of a novel mycovirus of

Rhizoctonia solani, a plant pathogenic fungus. Virus Genes 51(1): 167-170.

ACCEPTED MANUSCRIPT

29 Figures legends

Figure 1. Proposed genome organisations of (A) Ceratobasidium mitovirus 1

(Mitovirus) (B) Ceratobasidium virus 1 (unclassified mycovirus) (C) Ceratobasidium

virus 2 (unclassified mycovirus) (D) Ceratobasidium hypovirus 1 (Hypovirus) (E)

Pterostylis sanguinea leaf-associated virus A (unclassified mycovirus-like) and (F)

Pterostylis sanguinea leaf-associated totivirus A (Totivirus). Asterisks indicate

incomplete 5' and/or 3' ends. Shaded boxes represent Nudix hydrolase (N) and RNA

dependent RNA polymerase (RdRp) domains.

Figure 2. Maximum likelihood phylogenetic trees constructed from polyprotein or

RdRp deduced amino acid sequences of (A) proposed mitovirus Ceratobasidium

mitovirus 1 (indicated by a dot), (B) proposed Ceratobasidium virus 1 (indicated by

dots), Ceratobasidium virus 2 (indicated by a square) and Pterostylis sanguinea leaf-

associated virus A (indicated by a triangle), (C) proposed hypovirus Ceratobasidium

hypovirus 1 (indicated by a dot), and (D) proposed totivirus Pterostylis sanguinea

leaf-associated totivirus A (indicated by dots) with those of most closely related

described viruses. 1000 bootstrap replications were carried out and branch confidence

values below 60% were omitted. Branch lengths represent calculated evolutionary

distance in units of amino acid substitutions per site.

ACCEPTED MANUSCRIPT

Table legend

Table 1. Viruses (other than partitiviruses) identified from Pterostylis sanguinea

orchids and associated Ceratobasidium sp. mycorrhizal fungi. Viruses identified from

pooled leaf tissue of three and four P. sanguinea plants (P-2012, P-2013, respectively)

are shown, and from Ceratobasidium isolates F-2012 and F-2013, each from the

underground stem of one P. sanguinea plant.

ACCEPTED MANUSCRIPT

33 Table 1. Viruses (other than partitiviruses) identified from Pterostylis sanguinea orchids and associated Ceratobasidium sp. mycorrhizal fungi. Viruses identified from pooled leaf tissue of three and four P. sanguinea plants (P-2012, P-2013, respectively) are shown, and from Ceratobasidium isolates F-2012 and F-2013, each from the underground stem of one P. sanguinea plant.

Sequence Proposed GenBank accession Isolate length (nt) Virus host Estimated Estimated GenBank Proposed virus name classification Closest Blastp match (e-value, % aa identity) of name [protein(s) (Sample no.) % ORF # % genome* accession Family, Genus closest match length (aa)] Ceratobasidium Murdoch-1 Narnaviridae, 2850 Ceratobasidium sp. Rhizophagus sp. HR1 BAN85985 (9e-83, 31%) 100% 100% KU291923 mitovirus 1 (CbMV1) Mitovirus [817] (F-2012) mitovirus like ssRNA

Ceratobasidium virus 1 Murdoch-2 Unclassified 8227 Ceratobasidium sp. Desulfovibrio oxyclinae WP_026167673 (2.2, 26%) >60% 92% KU291947 (CbV1) [988, >1593] (F-2012) transposase (ORF1); BAT50987 (1e-172, 35%) >90% Rosellinia necatrix mycovirus 1-W1032/S5 (ORF2) Ceratobasidium virus 1 Murdoch-3 Unclassified 7161 Ceratobasidium sp. Desulfovibrio oxyclinae WP_026167673 (0.75, 26%) 100% 80% KU291948 (CbV1) [1593, >632] (F-2012) transposase (ORF1); BAT50987 (3e-121, 37%) >50% Rosellinia necatrix mycovirus 1-W1032/S5 (ORF2) Ceratobasidium virus 1 Murdoch-4 Unclassified 4051 Ceratobasidium sp. Rosellinia necatrix mycovirus BAT50987 (7e-176, 34%) >70% 45% KU291949 (CbV1) [>1123] (F-2012) 1-W1032/S5

Ceratobasidium virus 2 Murdoch-5 Unclassified 7089 Ceratobasidium sp. Rhizoctonia solani RNA virus YP_009158859 (2e-04, 25%) >90% 93% KU291938 (CbV2) [>1112, >1162] (F-2012) HN008 YP_009158860 (1e-166, 31%) >90%

CeratobasidiumACCEPTED Murdoch-6 Hypoviridae, 71MANUSCRIPT43 Ceratobasidium sp. Cryphonectria hypovirus 1 NP_041091 (1e-12, 30%) 100% 56% KU291924 hypovirus 1 (CbHV1) Hypovirus [1211, >1085] (F-2012) ABI64296 (8e-30, 26%) >30%

Pterostylis sanguinea Murdoch-7 Unclassified 10,716 P. sanguinea Lentinula edodes mycovirus BAM34027 (1e-105, 26%) 100% 95-100% KU291925 leaf-associated virus A [1912, >1252] (P-2012) HKA BAM34028 (5e-157, 35%) >90% (PsVA)

Pterostylis sanguinea Murdoch-8 Totiviridae, 4631 P. sanguinea Black raspberry virus F YP_001497150 (6e-139, 36%) >90% 92% KU291927 leaf-associated totivirus Totivirus [>723, >812] (P-2013) YP_001497151 (0.0, 56%) >90% A (PsTVA)

Pterostylis sanguinea Murdoch-9 Totiviridae, 3631 P. sanguinea Black raspberry virus F YP_001497150 (2e-52, 32%) >50% 72% KU291926 leaf-associated totivirus Totivirus [>382, >685] (P-2013) YP_001497151 (0.0, 56%) >80% A (PsTVA)

Pterostylis sanguinea Murdoch-10 Totiviridae, 3613 P. sanguinea Black raspberry virus F YP_001497150 (2e-152, 37%) >90% 72% KU291928 leaf-associated totivirus Totivirus [>696, >404] (P-2013) YP_001497151 (1e-169, 56%) >50% A (PsTVA)

*Calculation of genome and protein percentage was based on sequence length of closest blastp match and/or related virus isolates. # Estimated percentage of protein was limited by lack of complete ORF

ACCEPTED MANUSCRIPT