University of South Florida Scholar Commons

Marine Science Faculty Publications College of Marine Science

2019

Prevalence of a vertically transmitted single-stranded DNA in spinybacked orbweavers (Gasteracantha cancriformis) from Florida, USA

Karyna Rosario University of South Florida, [email protected]

Kaitlin A. Mettel University of South Florida

Anthony M. Greco University of South Florida, [email protected]

Mya Breitbart University of South Florida, [email protected]

Follow this and additional works at: https://scholarcommons.usf.edu/msc_facpub

Part of the Life Sciences Commons

Scholar Commons Citation Rosario, Karyna; Mettel, Kaitlin A.; Greco, Anthony M.; and Breitbart, Mya, "Prevalence of a vertically transmitted single-stranded DNA virus in spinybacked orbweavers (Gasteracantha cancriformis) from Florida, USA" (2019). Marine Science Faculty Publications. 433. https://scholarcommons.usf.edu/msc_facpub/433

This Article is brought to you for free and open access by the College of Marine Science at Scholar Commons. It has been accepted for inclusion in Marine Science Faculty Publications by an authorized administrator of Scholar Commons. For more information, please contact [email protected]. 1 Prevalence of a vertically transmitted single-stranded DNA virus in spinybacked orbweavers 2 (Gasteracantha cancriformis) from Florida, USA

3 Karyna Rosario1*, Kaitlin A. Mettel1, Anthony M. Greco1, and Mya Breitbart1

4

5 1College of Marine Science, University of South Florida, 140 7th Avenue South, Saint Petersburg,

6 Florida, 33701, USA

7

8

9

10 *Corresponding author: Karyna Rosario, [email protected]

11

12 Keywords: CRESS DNA, , ssDNA, virus, vertical transmission, Araneae 13 Abstract 14 (Order Araneae, Class Arachnida) are an important group of predatory in 15 terrestrial ecosystems that have been recently identified as an untapped reservoir of single-stranded 16 (ss)DNA . Specifically, spiders harbor a diversity of ssDNA viruses encoding a replication- 17 associated protein (Rep) within a circular genome. However, little is known about the ecology of 18 novel circular Rep-encoding ssDNA (CRESS DNA) viruses. Here we investigated two CRESS 19 DNA viruses recently identified in spinybacked orbweavers (Gasteracantha cancriformis), 20 namely spinybacked orbweaver circular virus (SpOrbCV) 1 and 2. SpOrbCV-1 was detected in 21 the majority (>65%) of spider specimens from all life stages, including eggs, spiderlings, and 22 adults, demonstrating that this virus is active within spinybacked orbweavers. In contrast, 23 SpOrbCV-2 was only detected in adults at a lower (36%) prevalence. Since we also detected 24 SpOrbCV-2 in other spider species and this virus has been reported from a dragonfly, we suggest 25 that SpOrbCV-2 is accumulated in these predators through common insect prey. The prevalence 26 of SpOrbCV-1 in collected specimens allowed us to design assays to characterize this virus, which 27 represents a new group of CRESS DNA viruses, the “circularisviruses”. To our knowledge, 28 SpOrbCV-1 is the first example of a vertically transmitted virus in spiders, which may explain its 29 high prevalence in spinybacked orbweavers. Since vertically transmitted viruses infecting insects 30 (Class Insecta) can manipulate their host behavior and physiology, future studies should 31 investigate the ecological role of vertically transmitted viruses in spiders. 32 Introduction 33 Viruses encoding a replication-associated protein (Rep) within a covalently-closed, single- 34 stranded (ss)DNA genome have been increasingly identified in organisms spanning the eukaryotic 35 tree of life, from unicellular organisms to major groups of vertebrate and invertebrate metazoans 36 [1]. Current classification schemes for eukaryotic circular Rep-encoding ssDNA (CRESS DNA) 37 viruses, which include six monophyletic families (, , , 38 , , ) [40], lag behind the discovery rate of these 39 viruses. Many CRESS DNA viral species cannot be accommodated within established taxa and 40 remain unclassified, indicating that the number of CRESS DNA viral families will increase. Given 41 that many CRESS DNA viruses have only been identified through sequencing, little is known 42 about the ecology of a large number of CRESS DNA viruses, including the host species for both 43 classified (e.g., Genomoviridae and Smacoviridae) and unclassified viruses. Terrestrial arthropods 44 have been recently identified as an untapped reservoir of CRESS DNA viral diversity, highlighting 45 the need to further investigate viral prevalence and transmission in this dominant and 46 undersampled group of [2]. 47 Spiders (Order Araneae) are an important group of predatory arthropods [3] that have been 48 extensively studied due to their fascinating biology (e.g., [4-6]), potential as natural pest biocontrol 49 agents (e.g., [7, 8]), and the biotechnological potential of their silk and venom [9-11]. Previous 50 studies have investigated the influence of microbes, in particular bacteria, on spider biology [12- 51 15]. Notably, studies investigating vertically transmitted bacteria in spiders indicate that these 52 ‘reproductive parasites’ may lead to sex-ratio bias through behavior manipulation and male-killing 53 [16-18]. Although spiders harbor a diversity of viruses, there is scarce information regarding the 54 prevalence and ecological impact of viral infection in spiders. Endogenous viral sequences within 55 spider chromosomes indicate that these have been infected by both CRESS DNA viruses 56 and double-stranded DNA polyomaviruses [19, 20]. Studies within the past four years have also 57 revealed a diversity of novel extant viruses in multiple spider species, including more than 20 58 CRESS DNA [2] and 100 RNA viruses [21-24] encompassing various viral families and 59 unclassified species. Moreover, RNA viruses have been detected in multiple spider organs, 60 including brain, venom glands and silk glands, indicating that viral infection may play a role in 61 spider ecology [21]. 62 The present study investigated two CRESS DNA viruses, namely spinybacked orbweaver 63 circular virus (SpOrbCV) 1 and 2, that were reported from specimens of Gasteracantha 64 cancriformis, commonly known as the spinybacked orbweaver [2]. Spinybacked orbwearvers are 65 generally considered beneficial spiders since they trap and consume insects considered pests, such 66 as whiteflies, and are typically found in wooded areas where there is abundant foliage and branches 67 [25]. However, spinybacked orbweavers are also commonly found in man-made structures, such 68 as porches and screen pool enclosures, where they may become a nuisance. These conspicuous 69 spiders, which can be easily spotted due to their contrasting colors and spine-like projections 70 around their abdomen, are widely distributed throughout the Caribbean, Central America, and 71 Southern USA [25, 26]. To our knowledge, SpOrbCV-1 and SpOrbCV-2 are the first viruses that 72 have been identified in this tropical spider species. 73 Here we examined the prevalence of SpOrbCV-1 and -2 in different spider life stages, 74 including eggs, spiderlings, and adults, to explore the association between spinybacked 75 orbweavers and these CRESS DNA viruses. Similar to known CRESS DNA viruses [1], both 76 SpOrbCVs have a small genome (< 2 kb) encoding the Rep and a putative capsid protein. 77 SpOrbCV-1 is an unclassified CRESS DNA virus that has persisted in the Tampa Bay, Florida, 78 USA spinybacked orbweaver population for at least four years (2014 through 2018). We detected 79 SpOrbCV-1 in the majority (>65%) of specimens from all life stages, demonstrating that this virus 80 is active within spinybacked orbweavers. In contrast, the detection of SpOrbCV-2 in adult 81 specimens alone, combined with its presence in various species of spiders (this study) and a 82 dragonfly [27], indicates that this cyclovirus (genus Cyclovirus, family Circoviridae) may 83 accumulate in predatory arthropods through common insect prey. The prevalence of SpOrbCV-1 84 in collected specimens allowed us to design assays to characterize this virus, which represents a 85 new cohesive group of CRESS DNA viruses, referred to here as circularisviruses. Finally, the 86 detection of SpOrbCV-1 in eggs represents the first evidence for vertical transmission of viruses 87 in spiders, which may explain its high prevalence in spinybacked orbweavers. 88 Materials and Methods 89 Analysis of viral sequences related to SpOrbCV-1 90 91 SpOrbCV-1 (GenBank accession number MH545519) and SpOrbCV-2 (GenBank 92 accession number MH545515) were recently described as a part of a CRESS DNA viral discovery 93 effort in terrestrial arthropods [2]. Since SpOrbCV-1 cannot be accommodated within established 94 CRESS DNA viral families, we further evaluated the characteristics of this virus and other 95 unclassified CRESS DNA viruses. CRESS DNA virus genomes with top BLASTn matches, 96 including reciprocal searches, to the SpOrbCV-1 genome were obtained from GenBank in 97 December 2018. SeqBuilder from the Lasergene software package version 11.2.1 (DNASTAR, 98 Madison, WI) was used to explore genomic features in the identified genomes, including: size, 99 open reading frames (ORFs), genome organization, and putative origin of replication (ori), which 100 is marked by a conserved nonanucleotide motif at the apex of a predicted stem-loop structure [28]. 101 Genome-wide and predicted amino acid sequence pairwise identities were calculated using the 102 Sequence Demarcation Tool version 1.2 [29]. Predicted Rep amino acid sequences from each 103 genome were aligned using the MUSCLE algorithm as implemented in MEGA7 [30, 31]. 104 Alignments were then used to create a graphical representation of conserved residues found within 105 CRESS DNA viral Rep domains, including endonuclease and superfamily 3 (SF3) helicase motifs, 106 in WebLogo 3 [32]. 107 108 Collection of spinybacked orbweavers to investigate SpOrbCVs 109 110 To further investigate SpOrbCVs, spinybacked orbweaver samples were collected from 111 vegetation and man-made structures found outdoors as well as screen pool enclosures in the Tampa 112 Bay area in Florida, USA in 2017 and 2018. The majority of samples, including adults (66%, n = 113 154), eggs (89%, n = 35) and spiderlings (100%, n = 14), were collected from a single screen pool 114 enclosure that was closely monitored. Based on size and the presence of abdominal spines [26], 115 99% of sampled adults were female. All eggs were collected by carefully scraping and collecting 116 complete egg sacs from surfaces without damaging the eggs. Spiderling collections only took place 117 in 2018. To do this, egg sacs were monitored until spiderlings emerged (~2 - 5 weeks) and 118 spiderlings were then collected before they dispersed from the silk surrounding the egg sac. All 119 specimens were collected using sterile 15 or 50 ml conical tubes and kept at room temperature 120 until taken to the lab within one hour of collection. Once in the lab, samples from adults were 121 either immediately stored as whole specimens or two legs were dissected off individual specimens 122 using a sterile scalpel and stored separately. Whole and dissected specimens as well as leg 123 subsamples were stored at -80°C until further processing. 124 125 PCR assays to evaluate the prevalence of SpOrbCVs 126 127 The prevalence of SpOrbCV-1 and SpOrbCV-2 was investigated in specimens representing 128 different spinybacked orbweaver life stages through PCR. The prevalence in adults was initially 129 assessed based on positive PCR products from virions isolated from whole specimens following a 130 method previously described for discovery of CRESS DNA viruses in arthropods (Rosario et al 131 2012, Rosario et al 2018). Briefly, spiders were rinsed with sterile suspension medium (SM) buffer

132 [0.1 M NaCl, 50 mM Tris/HCl (pH 7.4), 10 mM MgSO4] to remove debris, followed by 133 homogenization in 1 ml SM buffer through bead beating in a FisherbrandTM Bead Mill 4 134 homogenizer (Fisher Scientific, Hampton, NH) for one minute using 1 mm glass beads. 135 Homogenates were then centrifuged at 6,000 x g for 6 min and the supernatants containing viral 136 particles were filtered through a 0.45-μm Sterivex filter (MilliporeSigma, Burlington, MA) to 137 remove cells. DNA was extracted from 200 μl of the filtrate containing partially purified virus 138 particles using the QIAmp MinElute Virus Spin kit (Qiagen, Hilden, Germany). However, since 139 virus particles isolated from whole specimens may include viruses from dietary content, viral 140 prevalence was further assessed in adult spiders by testing legs. For this purpose, DNA was directly 141 extracted from two legs per specimen using the Extract-N-Amp Tissue PCR Kit (MilliporeSigma, 142 St. Louis, MO) following the manufacturer’s protocol. 143 Viral prevalence in eggs and spiderlings was assessed on a per egg sac basis rather than 144 individual eggs or spiderlings. For this purpose, the silk surrounding the eggs was removed using 145 sterile pipette tips and scapels and about a third of the eggs or two spiderlings from a given sac 146 were transferred to a 1.5 ml microcentrifuge tube. After rinsing the eggs and spiderlings with SM 147 buffer, DNA was directly extracted from tissues using the Extract-N-Amp Tissue PCR Kit 148 (MilliporeSigma, St. Louis, MO). 149 To assess the prevalence of each virus, primer pairs SpOrbCV1_F2/R2 and 150 SpOrbCV2_F2/R2 (Table 1) were used for PCR assays targeting partial SpOrbCV-1 and 151 SpOrbCV-2 genomes, respectively. For this purpose, 1 µl of DNA extract from adults, eggs, and 152 spiderlings was amplified through rolling circle amplification (RCA) to enrich for small circular 153 templates [33], such as CRESS DNA genomes, using the Illustra TempliPhi Amplification kit (GE 154 Healthcare, Chicago, IL). Four microliters of DNA extract or 0.6 µl of RCA product were used as

155 template in 50 µl PCR reactions containing 1.5 mM MgCl2, 1X Apex NH4 buffer, 0.5 µM of each 156 primer, 0.2 mM of each dNTP, and 1U Apex Red Taq DNA Polymerase (Genesee Scientific, San 157 Diego, CA). Each PCR proceeded at 94ºC for 5 min, followed by 35 cycles of 94ºC for 1 min, 158 appropriate annealing temperature for 45 s (Table 1), and 72 ºC for 2.5 min, followed by a final 159 extension at 72 ºC for 8 min. Additionally, RCA products obtained from virus particles isolated 160 from a collection of 106 spiders from various families (Supplemental Table S1) were tested to 161 evaluate if SpOrbCVs could be detected in other spider species using the same PCR assay. This 162 spider collection included specimens from areas where we identified spinybacked orbweavers 163 positive for SpOrbCVs and specimens originally collected for CRESS DNA virus discovery 164 purposes [2]. 165 All PCRs included a negative process control composed of RCA product from SM buffer 166 DNA extracts processed alongside samples as well as a negative PCR control where sterile water 167 was used as template. DNA from the original spinybacked orbweavers where SpOrbCV-1 and 168 SpOrbCV-2 were detected served as positive controls. All PCR results were confirmed through 169 agarose gel electrophoresis and ethidium bromide staining. Templates resulting in strong, single 170 bands of the correct size relative to the positive control in agarose gels were considered positive, 171 whereas products producing weak bands were confirmed through Sanger sequencing with the 172 relevant PCR primers. 173 174 Egg surface decontamination to evaluate SpOrbCV-1 vertical transmission 175 176 Since eggs were positive for SpOrbCV-1, we evaluated if the virus could be detected in 177 individual eggs after surface decontamination. To test this, a total of 24 eggs from four egg sacs 178 known to be positive for SpOrbCV-1 were transferred individually to 1.5 ml microcentrifuge tubes 179 and processed through one out of four treatments. Each treatment was performed on three eggs 180 from each of two egg sacs for a total of six replicates per treatment. Individual eggs were rinsed 181 with SM buffer (untreated control) or decontaminated by submerging each egg in either 50 µl of 182 2% bleach for 5 min or 0.1% bleach for 30 mins [34]. For the remaining decontamination 183 treatment, each egg was submerged in 50 µl of 70% ethanol followed by 10% bleach for 1 min 184 each [35]. After bleach exposure, eggs were rinsed four times with 50 µl of SM buffer. The last 185 SM buffer wash was removed and DNA was immediately extracted from the rinsed egg using the 186 Extract-N-Amp Tissue PCR Kit (MilliporeSigma, St. Louis, MO). RCA products from each egg 187 were then tested using the PCR assay described above with SpOrbCV-1_F2/R2 primers. A 188 representative DNA extract from each treatment was spiked with a positive SpOrbCV-1 DNA 189 control and subjected to RCA followed by PCR to assess potential inhibition due to the 190 decontamination treatment. 191 192 Characterization of virion and tissue associated SpOrbCV-1 DNA 193 194 The single-stranded nature and polarity of the encapsidated SpOrbCV-1 DNA was 195 evaluated using approaches similar to those previously used to describe novel circular ssDNA 196 viruses [36]. All specimens selected for genome characterization assays were known to be positive 197 for SpOrbCV-1 based on leg subsamples. For virion strand or encapsidated DNA characterization 198 assays, virus particles were first purified from adult spinybacked orbweavers (n = 5) homogenized 199 in SM buffer. Virus particles were purified by filtering homogenates through a 0.22 µm Sterivex 200 filter and incubating 170 µl of filtrate at 37ºC for 2 hrs with a nuclease cocktail [37, 38] containing 201 1X Turbo DNAse Buffer (Invitrogen, Carlsbad, CA), 21U Turbo DNAse (Invitrogen, Carlsbad, 202 CA), 4.5U of Baseline-ZERO DNase (Epicentre, Madison, WI), and 112.5U Benzonase 203 (MilliporeSigma, Burlington, MA) to remove unencapsidated DNA. DNA was extracted from 204 200 µl of nuclease-treated viral particles using the QIAmp MinElute Virus Spin kit (Qiagen, 205 Hilden, Germany). The single-stranded nature of encapsidated SpOrbCV-1 DNA was then 206 assessed based on the susceptibility of the extracted DNA to S1 nuclease and restriction enzyme 207 digestion, which degrade single- and double-stranded DNA, respectively. 208 A controlled experiment using commercially available single- and double-stranded forms 209 of bacteriophage phiX174 circular DNA (New England Biolabs, Ipswich, MA) was conducted to 210 determine appropriate concentrations and incubation times for S1 nuclease treatment. Two PCR 211 assays with primer pairs, phiX174_F1/R1 and phiX174_F3/R3 (Table 1), were used to confirm 212 digestion. A total of 24U of S1 nuclease (Promega, Madison, WI) and incubation at 37ºC for 1 hr 213 was determined suitable to digest 120 ng of phiX174 ssDNA without completely degrading 120 214 ng of dsDNA. After appropriate S1 treatment conditions were determined, DNA extracted from 215 virus particles purified from spinybacked orbweavers were split into two 12 µl aliquots. One 216 aliquot was treated with 24U of S1 nuclease and incubated at 37ºC for 1 hr, while the other aliquot 217 was treated with 1 µl FastDigest EcoRI (Thermo Fisher Scientific, Waltham, MA) in 30 µl 218 reactions following manufacturer’s instructions. Products from the S1 nuclease treatment were 219 cleaned with the DNA Clean & Concentrator kit (Zymo Research, Irvine, CA) and eluted with 12 220 µl of sterile water. Cleaned digested products (2.5 µl) were used as template for a 25 µl PCR 221 reaction using SpOrbCV-1_F3/R3 primers (Table 1), which flank an EcoRI restriction site in the 222 SpOrbCV-1 genome, to confirm nuclease digestions. 223 A very similar nuclease treatment approach was used to determine if a dsDNA replicative 224 form (RF) of the SpOrbCV-1 genome could be detected in spinybacked orbweaver tissues. To do 225 this, the hard, shell-like abdomen was removed from five adult spinybacked orbweavers and DNA 226 was directly extracted from the cephalothorax. Spiderlings from six egg sacs were also investigated 227 for the presence of a RF. Total DNA was extracted from individual adults and twenty spiderlings 228 per egg sac using the ZR Tissue & Insect DNA Kit (Zymo Research, Irvine, CA) following the 229 manufacturer’s protocol. DNA extracts were treated with 24U of S1 nuclease in 30 µl reactions 230 and incubated at 37ºC for 1 hr. S1 nuclease digested products were cleaned using the DNA Clean 231 & Concentrator kit (Zymo Research, Irvine, CA) and eluted with 12 µl of sterile water. Clean S1 232 digested products (2.5 µl) were then treated with 1 µl FastDigest EcoRI following manufacturer’s 233 instructions in 10 µl reactions. Cleaned 10-fold diluted S1 digested products and products treated 234 with both S1 and EcoRI were used as template (2.5 µl) for PCR using SpOrbCV-1_F3/R3 primers 235 (Table 1) in 25 µl reactions. 236 The polarity of the encapsidated SpOrbCV-1 DNA was determined using strand-specific, 237 linear amplification with either SpOrbCV-1_F3 or SpOrbCV-1_R3 primers adapting a method 238 previously used to characterize circular ssDNA viral genomes [39]. For this purpose, 2.5 µl of 239 100-fold diluted DNA extracted from virus particles purified from adult spinybacked orbweavers

240 (n = 2) was linearly amplified for 50 cycles in 25 µl reactions containing 1.5 mM MgCl2, 1X Buffer 241 II, 0.4 µM of either primer, 0.2 mM of each dNTP, and 0.6U AmpliTaq Gold DNA Polymerase 242 (Applied Biosystems, Foster City, CA). A mock reaction containing template and all the reagents, 243 except for primer, was also included alongside strand-specific reactions to account for background. 244 Each amplification reaction proceeded at 94ºC for 5 min, followed by 50 cycles of 94ºC for 45 s, 245 52ºC for 45 s (Table 1), and 72ºC for 1 min, then a final extension at 72ºC for 8 min. One microliter 246 of linearly amplified products and mock reactions were then exponentially amplified in 25 µl PCR 247 reactions using primer pairs SpOrbCV1_F3n1/R3n1 and SpOrbCV1_F3n2/R3n2 (Table 1), which 248 are nested relative to strand-specific primers. PCR reactions contained 1.5 mM MgCl2, 1X Buffer 249 II, 0.4 µM of each primer, 0.2 mM of each dNTP, and 0.6U AmpliTaq Gold DNA Polymerase 250 (Applied Biosystems, Foster City, CA). Each reaction proceeded for 25 cycles using the same 251 thermocycling conditions listed for strand-specific linear amplification. Note that the encapsidated 252 viral DNA that is complementary to a given strand-specific primer will produce a strong signal 253 after exponential amplification. 254 255 Transmission electron microscopy analysis of SpOrbCV-1 positive specimens 256 257 Transmission electron microscopy (TEM) was used to observe virus-like particles (VLPs) 258 within spinybacked orbweaver tissues. For this purpose, legs were dissected from newly collected 259 adult spinybacked orbweavers and immediately fixed in a mixture of 4% glutaraldehyde and 2% 260 formaldehyde in 0.1M sodium cacodylate buffer, pH=7.3 using a mild vacuum. In addition, two 261 leg subsamples from each specimen were left unfixed and processed with the Extract-N-Amp 262 Tissue PCR Kit (MilliporeSigma, St. Louis, MO) to screen for SpOrbCV-1 through PCR using 263 methods outlined above. Legs from five positive specimens were chosen for TEM analysis. For 264 this purpose, fixed legs were rinsed for two hours with 0.1M sodium cacodylate buffer, pH=7.3 265 three times followed by an overnight rinse in the refrigerator and a final water rinse for 1 hour.

266 Rinsed samples were secondarily fixed in 2% aqueous OsO4 for 2 hours and rinsed in water for 30 267 minutes followed by overnight storage in water in the refrigerator. Tissue was enbloc stained in 268 0.5% aqueous uranyl acetate in the dark for 2 hours and then rinsed in water three times for 30 269 minutes each. Samples were dehydrated in an ethanol series, including 70% and 95% ethanol, for 270 15 minutes each, then two changes in 100% ethanol for 15 minutes each. A final dehydration 271 involved two 30 minute changes in 100% acetone. Leg tissues were embedded in Spurr’s resin and 272 cured in the oven. Ultrathin sections were cut and placed on copper grids for observation on a 273 Hitachi 7100 TEM at 100 kV. A Gatan Orius high-resolution digital camera was used to record 274 images. 275 276 Results and Discussion

277 SpOrbCV-1 represents a novel group of CRESS DNA viruses 278 279 SpOrbCV-1 and SpOrbCV-2 represent two distinct CRESS DNA viral groups identified 280 in spinybacked orbweavers. SpOrbCV-1 cannot be classified within established taxa, whereas 281 SpOrbCV-2 is an isolate of a species classified within the family Circoviridae (genus Cyclovirus) 282 known as dragonfly associated cyclovirus 3 (DfACyV-3; accession number JX185424) [2]. Since 283 SpOrbCV-1 shares similar genomic characteristics with unclassified viruses reported from other 284 spider species, dragonflies, and bat feces [2], we further explored the genomic sequences of 285 circularisviruses, named after the first virus reported from this group (dragonfly circularisvirus) 286 [27], and identified common features among this novel group of CRESS DNA viruses. Reciprocal 287 BLAST searches led to the identification of panicum ecklonii-associated virus (accession number 288 MH425571), a virus reported from the leaf tissue of a perennial grass [41], as an additional 289 circularisvirus. Therefore, circularisviruses include viruses identified in terrestrial arthropods, 290 feces, and plants (Table 2). 291 Circularisviruses contain small (~1.9 – 2 kb), circular genomes with two major ORFs 292 encoding Rep and capsid proteins in the same strand (Fig. 1). This genomic architecture was 293 deemed novel back in 2009 when metagenomic analyses uncovered small, circular ssDNA viral 294 genomes with gene arrangements that had only been observed in linear ssDNA viruses [42]. 295 Although similar genome organizations have been observed in CRESS DNA viruses reported from 296 various environments and organisms since then [28, 43-46], circularisviruses are the first 297 phylogenetic group whose members contain genomes with Rep and capsid protein coding 298 sequences on the same strand. Circularisviruses form a monophylogenetic clade when comparing 299 CRESS DNA viruses based on the Rep [2] and share > 57% genome-wide pairwise identity (PI) 300 among each other (Fig. 1), which is congruent with the lower limit of PIs reported for established 301 CRESS DNA viral families [47-49]. Similar to most CRESS DNA viral groups, the circularisvirus 302 Rep is more conserved than the putative capsid based on amino acid pairwise identities (Fig. 1). 303 Based on the lowest species demarcation threshold (75% genome-wide pairwise identity) for 304 established CRESS DNA viral families [50], currently there are eight distinct circularisvirus 305 species sharing <70% genome-wide PI and two genomes representing variants of SpOrbCV-1, 306 which share 91% genome-wide PI (Fig. 1). The distribution of genome-wide PIs among 307 circularisviruses should be evaluated as more representative genomes are discovered, which may 308 lead to the creation of a new CRESS DNA family. 309 To confirm the single-stranded nature of the SpOrbCV-1 genome, we took advantage of 310 S1 nuclease and EcoRI restriction endonuclease, which degrade ssDNA templates and dsDNA at 311 a specific site, respectively (Fig. 2). DNA extracts from purified virions were susceptible to S1 312 nuclease treatment but resistant to EcoRI digestion, confirming that SpOrbCV-1 virions 313 encapsidate circular ssDNA. Based on the putative ori of many CRESS DNA viruses, which 314 contains a conserved nonanucleotide motif (‘NANTATTAC’) at the apex of a stem-loop structure 315 [28], it was predicted that the protein-encoding strand of SpOrbCV-1 is the virion strand. A strand- 316 specific amplification assay on DNA extracted from virions purified from two individual 317 spinybacked orbweavers confirmed that the Rep- and capsid-encoding strand containing the 318 nonanucleotide motif ‘CAGTATTAC’ is indeed the virion strand of this circularisvirus. The 319 majority of circularisvirus genomes contain a putative ori in the protein-coding strand with the 320 conserved nonanucleotide motif ‘NANTATTAC’ (Fig. 1). However, two genomes, one reported 321 from a spider (accession number MH545522) and the other from grass (accession number 322 MH425571), have a substitution in the nonamer, namely ‘NANTGTTAC’ and ‘NANTACTAC’, 323 respectively. Similar nonanucleotide variants have been observed in other CRESS DNA viruses, 324 such as members of the family Smacoviridae and some members of the family Genomoviridae 325 [47, 48]. 326 The circularisvirus Rep contains the two-domain organization that is distinctive of 327 eukaryotic CRESS DNA viruses, including endonuclease and superfamily 3 (SF3) helicase 328 domains that are important for rolling circle replication (RCR) (Fig. 1) [51, 52]. RCR motifs found 329 within the endonuclease domain are important for RCR initiation and termination, whereas the 330 helicase domain may be involved in DNA unwinding during elongation (reviewed by [28]). The 331 circularisvirus Rep endonuclease domain contains conserved residues observed in other CRESS 332 DNA viral Reps, including an arginine finger motif found within the SF3 helicase domain of most 333 CRESS DNA viral groups [53]. However, the Walker-A motif within the SF3 helicase domain 334 contains a unique substitution in comparison to all other CRESS DNA viral groups that is 335 consistently found in all circularisvirus Reps. The Walker-A motif, which may be involved in 336 substrate binding [54], is characterized by the sequence ‘GxxxxGK(S/T)’ in most CRESS DNA 337 viral groups, with the exception of members of the families Smacoviridae and Nanoviridae which 338 contain the sequence ‘GxxGK(S/T)’ [53]. However, the circularisvirus Walker-A motif contains 339 the sequence ‘GxxxxGKD’ (Fig. 1). The biochemical implications of this acidic amino acid 340 substitution within the Walker-A motif are unknown, but it seems to be a distinguishing feature of 341 circularisvirus Reps. Interestingly, BLAST analysis revealed an endogenous circularisvirus-like 342 SF3 helicase sequence in a genome scaffold from Parasteatoda tepidariorum (accession 343 XP_015919340), known as the common house spider. Since SF3 helicases have been mainly 344 identified in small DNA and RNA viruses [54, 55] and the P. tepidariorum has significant matches 345 (BLASTp, e-value < 10-50) to the helicase domain of circularisvirus Reps, including the 346 ‘GxxxxGKD’ domain, it is possible that this spider genome acquired the helicase from a CRESS 347 DNA virus. 348 349 SpOrbCV-1 is highly prevalent in spinybacked orbweavers 350 351 SpOrbCV-1 and SpOrbCV-2 were originally identified in spinybacked orbweavers [2]; 352 however, SpOrbCV-2 shares 99% genome-wide pairwise identity with DfACyV-3, a cyclovirus 353 species originally discovered in dragonflies collected in Florida in 2010 [27]. Since spiders and 354 dragonflies are top insect predators, we hypothesized that SpOrbCV-2 is of dietary origin whereas 355 SpOrbCV-1 is specifically associated with spinybacked orbweavers. To further explore the 356 association between these CRESS DNA viruses and spinybacked orbweavers, we compared the 357 prevalence of SpOrbCV-1 and SpOrbCV-2 in specimens collected in 2017 and 2018 through PCR. 358 PCR assays indicated that SpOrbCV-1 was more prevalent in spinybacked orbweavers than 359 SpOrbCV-2 (Fig. 3). Notably, SpOrbV-1 was detected in at least 65% of specimens representing 360 various spinybacked orbweaver life stages, including adults, eggs, and spiderlings. In contrast, 361 SpOrbCV-2 was only detected in adults and the highest prevalence (36%) was detected among 362 samples composed of virus particles isolated from whole specimens compared to 2% based on leg 363 tissues alone. The vast majority (85%) of samples that were positive for SpOrbCV-2 (n = 13) were 364 also positive for SpOrbCV-1. In addition, SpOrbCV-1 was more abundant in leg samples than 365 SpOrbCV-2 given that the former was detected without RCA pre-amplification, whereas 366 SpOrbCV-2 was only detected in PCR reactions containing RCA product as template. A similar 367 trend was observed for virus particles isolated from whole bodies since 40% and 20% of 368 SpOrbCV-1 and SpOrbCV-2 positive samples, respectively, were directly identified from DNA 369 extracts, rather than RCA products. 370 Although most specimens were collected from a single screen pool enclosure, both 371 SpOrbCV-1 and SpOrbCV-2 were also detected in samples collected outside of this closely 372 monitored area. While SpOrbCV-2 was only detected in spiders from two screen pool enclosures, 373 SpOrbCV-1 was detected in samples collected from all tested sites, including six pool enclosures 374 and three outdoor areas. Therefore, SpOrbCV-1 is more widely associated with spinybacked 375 orbweavers than SpOrbCV-2. 376 Despite the high prevalence of SpOrbCV-1 in spinybacked orbweavers, including 377 SpOrbCV-2 positive specimens, this virus was not detected in any of the 106 specimens tested 378 representing various species from at least twelve families of spiders (Supplemental Table S1). 379 However, SpOrbCV-2 was detected in two specimens representing distinct spider species. One of 380 the SpOrbCV-2 positive specimens was collected within the closely monitored area where some 381 spinybacked orbweavers were also positive for this virus. Therefore, the PCR data support that 382 SpOrbCV-1 is highly prevalent and specifically associated with spinybacked orbweavers, whereas 383 SpOrbCV-2 is less prevalent and can be detected in different species of predatory arthropods 384 including insects (dragonflies) and at least three species of spiders. Although these data suggest 385 that SpOrbCV-2 is likely associated with dietary contents, at this time, we cannot rule out the 386 possibility that SpOrbCV-2 infects spinybacked orbweavers. For example, it is possible that 387 SpOrbCV-2 has different tissue tropism than SpOrbCV-1 and, thus, cannot be readily detected in 388 legs. Additionally, little is known about latent infections, a type of persistent infection where 389 virions are not produced, in arthropods [56]. PCR assays used to target viruses associated with 390 whole spiders only tested virus particles and, thus, prevalence data based on whole bodies did not 391 account for potential latent infections. 392 393 Similar VLPs are found in SpOrbCV-1 positive specimens 394 395 TEM was used to demonstrate the presence of VLPs within spider leg tissues that were 396 positive for SpOrbCV-1 by PCR. Similar icosahedral VLPs, 25 nm in diameter, were observed in 397 leg muscle tissue from each of the five specimens analyzed (Fig. 4). Although this morphology 398 and VLP size resembles that of well-characterized CRESS DNA viruses, such as circoviruses [57], 399 we cannot conclude that the observed VLPs represent SpOrbCV-1 without further research. It 400 should be noted that one of the specimens also contained a second type of VLPs with icosahedral 401 morphology, averaging 66 nm in diameter, indicating that some spinybacked orbweavers may be 402 co-infected with multiple viruses (Supplemental Fig. S1). 403 404 SpOrbCV-1 is active within spinybacked orbweavers and is vertically transmitted 405 406 Since SpOrbCV-1 was prevalent in spinybacked orbweavers we set out to determine if the 407 virus was replicating within spiders by screening tissues for a dsDNA replicative form (RF) form. 408 S1 nuclease treatment followed by EcoRI digestion of total DNA extracted from spider tissues was 409 used to detect SpOrbCV-1 dsDNA, keeping in mind that dsDNA will be resistant to S1 treatment 410 but not EcoRI digestions (Fig. 2). The five total DNA extracts from adult specimens subjected to 411 the enzyme treatments resulted in the detection of dsDNA. The dsDNA RF was also detected in 412 total DNA extracts from spiderlings, albeit with a lower detection rate (two out of six). These 413 findings confirm that SpOrbCV-1 replicates within spinybacked orbweavers, including spiderling 414 and adult life stages. 415 The prevalence of SpOrbCV-1 in eggs and spiderlings that had not emerged from the egg 416 sac indicates the vertical transmission of this virus among spinybacked orbweavers. To distinguish 417 between a transovum route of transmission, where contaminated egg surfaces lead to infected 418 offspring, and a transmission pathway where viruses are passed to offspring within eggs [34, 56], 419 we evaluated if the virus could still be detected after decontaminating egg surfaces. In at least five 420 out of six replicates, SpOrbCV-1 was detected in individual eggs after decontaminating the egg 421 surfaces using strategies from insect rearing studies, including 0.1% and 2% bleach treatments 422 (Kukan 1999). SpOrbCV-1 was also detected in three out of six eggs treated with 70% ethanol 423 followed by 10% bleach, which has been used as an effective surface decontamination strategy for 424 molecular analysis [35]. However, this may underestimate the number of positive eggs since PCR 425 inhibition was documented in the failed amplification of positive controls spiked into DNA 426 extracts from samples treated with 10% bleach. Nevertheless, the detection of SpOrbCV-1 in eggs 427 regardless of surface decontamination strategy indicates that this virus can be transmitted within 428 eggs. Based on the high prevalence of SpOrbCV-1 in female spinybacked orbweavers, we 429 speculate that this virus is maternally transmitted, which is a common form of parasite vertical 430 transmission in arthropods [58], through a transovarial route. However, the role of male spiders in 431 virus transmission cannot be excluded since the vast majority of specimens in this study were 432 female. 433 Elucidating viral transmission strategies is critical to understand how a given virus persists 434 in natural populations [59, 60]. To our knowledge, SpOrbCV-1 is the first virus found to be 435 vertically transmitted in spiders, since most studies investigating viruses in spiders have focused 436 on adults [21]. Moreover, although CRESS DNA viruses have been identified in all major lineages 437 of terrestrial arthropods [2], SpOrbCV-1 is the first CRESS DNA virus shown to be vertically 438 transmitted outside of fungal [61] and plant [62] viruses that are known to be vertically transmitted 439 within their insect vector population. Vertically transmitted viruses are expected to have low 440 virulence given that the fitness of the virus is tightly linked to that of its host [58, 60]. The high 441 prevalence of SpOrbCV-1 in seemingly healthy spinybacked orbweavers seems to follow this 442 general notion of low virulence. However, it remains to be determined if SpOrbCV-1 can also be 443 transmitted horizontally. Several DNA and RNA viruses have mixed-modes of transmission, 444 where the prevalence of either a vertical or horizontal route may provide a selective advantage 445 depending on host densities [56, 59]. Considering that vertically transmitted viruses may play a 446 major role in insect biology by manipulating their host’s physiology, behavior, and offspring sex- 447 ratios to increase transmission [59, 63-65], future work should investigate the ecological role of 448 vertically transmitted viruses in spiders. 449 525 References 526 1. Zhao L, Rosario K, Breitbart M, Duffy S. Eukaryotic circular Rep-encoding single- 527 stranded DNA (CRESS DNA) viruses: Ubiquitous viruses with small genomes and a diverse host 528 range. In: Kielian M, Mettenleiter TC, Roossinck MJ (editors). Advances in Virus Research: 529 Academic Press; 2019. pp. 71-133. 530 2. Rosario K, Mettel KA, Benner BE, Johnson R, Scott C et al. Virus discovery in all three 531 major lineages of terrestrial arthropods highlights the diversity of single-stranded DNA viruses 532 associated with invertebrates. PeerJ 2018;6:e5761. 533 3. Nyffeler M, Birkhofer K. An estimated 400–800 million tons of prey are annually killed 534 by the global spider community. The Science of Nature 2017;104(3):30. 535 4. Chen Z, Corlett RT, Jiao X, Liu S-J, Charles-Dominique T et al. Prolonged milk 536 provisioning in a . Science 2018;362(6418):1052-1055. 537 5. Girard MB, Kasumovic MM, Elias DO. Multi-modal courtship in the peacock spider, 538 Maratus volans (O.P.-Cambridge, 1874). PLOS ONE 2011;6(9):e25390. 539 6. Mammola S, Michalik P, Hebets EA, Isaia M. Record breaking achievements by spiders 540 and the scientists who study them. PeerJ 2017;5:e3972. 541 7. Michalko R, Pekár S. The biocontrol potential of Philodromus (Araneae, ) 542 spiders for the suppression of pome fruit orchard pests. Biol Control 2015;82:13-20. 543 8. Rypstra Ann L, Buddle Christopher M. Spider silk reduces insect herbivory. Biol Lett 544 2013;9(1):20120948. 545 9. Agnarsson I, Kuntner M, Blackledge TA. Bioprospecting finds the toughest biological 546 material: Extraordinary silk from a giant riverine orb spider. PLOS ONE 2010;5(9):e11234. 547 10. Saez NJ, Senff S, Jensen JE, Er SY, Herzig V et al. Spider-venom peptides as 548 therapeutics. Toxins 2010;2(12):2851. 549 11. King GF, Hardy MC. Spider-venom peptides: Structure, pharmacology, and potential for 550 control of insect pests. Annu Rev Entomol 2013;58(1):475-496. 551 12. Curry MM, Paliulis LV, Welch KD, Harwood JD, White JA. Multiple endosymbiont 552 infections and reproductive manipulations in a linyphiid spider population. Heredity 553 2015;115(2):146-152. 554 13. Zhang L, Yun Y, Hu G, Peng Y. Insights into the bacterial symbiont diversity in spiders. 555 Ecol Evol 2018;8(10):4899-4906. 556 14. Goodacre SL, Martin OY, Bonte D, Hutchings L, Woolley C et al. Microbial 557 modification of host long-distance dispersal capacity. BMC Biol 2009;7(1):32. 558 15. Goodacre SL, Martin OY. Modification of insect and behaviours by vertically 559 transmitted endosymbionts: Infections as drivers of behavioural change and evolutionary novelty. 560 Insects 2012;3(1):246-261. 561 16. Gunnarsson B, Hewitt GM, Goodacre SL. Sex ratio, mating behaviour and Wolbachia 562 infections in a sheetweb spider. Biol J Linnean Soc 2009;98(1):181-186. 563 17. Vanthournout B, Swaegers J, Hendrickx F. Spiders do not escape reproductive 564 manipulations by Wolbachia. BMC Evol Biol 2011;11(1):15. 565 18. Goodacre SL. Endosymbiont infections in spiders. In: Casas J (editor). Advances in Insect 566 Physiology: Academic Press; 2011. pp. 137-153. 567 19. Buck CB, Van Doorslaer K, Peretti A, Geoghegan EM, Tisza MJ et al. The ancient 568 evolutionary history of polyomaviruses. PLOS Pathog 2016;12(4):e1005574. 569 20. Dennis TPW, Flynn PJ, Marciel de Souza W, Singer JB, Moreau CS et al. Insights into 570 circovirus host range from the genomic fossil record. J Virol 2018;92(16): e00145-00118. 571 21. Debat HJ. An RNA virome associated to the golden orb-weaver spider Nephila clavipes. 572 Front Microbiol 2017;8:2097. 573 22. Shi M, Lin XD, Tian JH, Chen LJ, Chen X et al. Redefining the invertebrate RNA 574 virosphere. Nature 2016(540):539–543. 575 23. Li CX, Shi M, Tian JH, Lin XD, Kang YJ et al. Unprecedented genomic diversity of 576 RNA viruses in arthropods reveals the ancestry of negative-sense RNA viruses. Elife 577 2015;4:e05378. 578 24. Shean RC, Makhsous N, Crawford RL, Jerome KR, Greninger AL. Draft genome 579 sequences of six novel picorna-like viruses from Washington State spiders. Genome Announc 580 2017;5(9):e01705-01716. 581 25. Muma MH. Biological and behavioral notes on Gasteracantha cancriformis (Arachnida: 582 Araneidae). Fla Entomol 1971;54:345-352. 583 26. Levi HW. The American orb-weaver genera Colphepeira, Micrathena, and Gasteracantha 584 north of Mexico (Araneae, Araneidae). Bull Mus Comp Zool 1978;148:417-442. 585 27. Rosario K, Dayaram A, Marinov M, Ware J, Kraberger S et al. Diverse circular 586 ssDNA viruses discovered in dragonflies (Odonata: Epiprocta). J Gen Virol 2012;93(Pt 12):2668- 587 2681. 588 28. Rosario K, Duffy S, Breitbart M. A field guide to eukaryotic circular single-stranded 589 DNA viruses: Insights gained from metagenomics. Arch Virol 2012;157(10):1851-1871. 590 29. Muhire BM, Varsani A, Martin DP. SDT: A tool based on pairwise 591 sequence alignment and identity calculation. PLOS ONE 2014;9(9):e108277. 592 30. Kumar S, Stecher G, Tamura K. MEGA7: Molecular Evolutionary Genetics Analysis 593 version 7.0 for bigger datasets. Mol Biol Evol 2016;33(7):1870-1874. 594 31. Edgar RC. MUSCLE: multiple sequence alignment with high accuracy and high 595 throughput. Nucleic Acids Res 2004;32(5):1792-1797. 596 32. Crooks GE, Hon G, Chandonia JM, Brenner SE. WebLogo: a sequence logo generator. 597 Genome Res 2004;14(6):1188-1190. 598 33. Kim KH, Bae JW. Amplification methods bias metagenomic libraries of uncultured 599 single-stranded and double-stranded DNA viruses. Appl Environ Microb 2011;77(21):7663-7668. 600 34. Kukan B. Vertical transmission of nucleopolyhedrovirus in insects. J Invertebr Pathol 601 1999;74(2):103-111. 602 35. Hammer TJ, Dickerson JC, Fierer N. Evidence-based recommendations on storing and 603 handling specimens for analyses of insect microbiota. PeerJ 2015;3:e1190. 604 36. Ng TFF, Manire C, Borrowman K, Langer T, Ehrhart L et al. Discovery of a novel 605 single-stranded DNA virus from a sea turtle fibropapilloma by using viral metagenomics. J Virol 606 2009;83(6):2500-2509. 607 37. Victoria JG, Kapoor A, Li L, Blinkova O, Slikas B et al. Metagenomic analyses of 608 viruses in stool samples from children with acute flaccid paralysis. J Virol 2009;83(9):4642-4651. 609 38. Ng TF, Marine R, Wang C, Simmonds P, Kapusinszky B et al. High variety of known 610 and new RNA and DNA viruses of diverse origins in untreated sewage. J Virol 611 2012;86(22):12161-12175. 612 39. Okamoto H, Ukita M, Nishizawa T, Kishimoto J, Hoshi Y et al. Circular double- 613 stranded forms of TT virus DNA in the liver. J Virol 2000;74(11):5161-5167. 614 40. Kazlauskas D, Varsani A, Krupovic M. Pervasive chimerism in the replication- 615 associated proteins of uncultured single-stranded DNA viruses. Viruses 2018;10(4):187. 616 41. Richet C, Kraberger S, Filloux D, Bernardo P, Harkins GW et al. Novel circular DNA 617 viruses associated with Apiaceae and Poaceae from South Africa and New Zealand. Arch Virol 618 2019;164(1):237-242. 619 42. Rosario K, Duffy S, Breitbart M. Diverse circovirus-like genome architectures revealed 620 by environmental metagenomics. J Gen Virol 2009;90:2418-2424. 621 43. Pham HT, Bergoin M, Tijssen P. Acheta domesticus volvovirus, a novel single-stranded 622 circular DNA virus of the house cricket. Genome Announc 2013;1(2):e00079. 623 44. Dayaram A, Galatowitsch ML, Argüello-Astorga GR, van Bysterveldt K, Kraberger 624 S et al. Diverse circular replication-associated protein encoding viruses circulating in invertebrates 625 within a lake ecosystem. Infect Genet Evol 2016;39:304-316. 626 45. Dayaram A, Goldstien S, Argüello-Astorga GR, Zawar-Reza P, Gomez C et al. 627 Diverse small circular DNA viruses circulating amongst estuarine molluscs. Infect Genet Evol 628 2015;31:284-295. 629 46. Dayaram A, Potter KA, Pailes R, Marinov M, Rosenstein DD et al. Identification of 630 diverse circular single-stranded DNA viruses in adult dragonflies and damselflies (Insecta: 631 Odonata) of Arizona and Oklahoma, USA. Infect Genet Evol 2015;30:278-287. 632 47. Varsani A, Krupovic M. Sequence-based taxonomic framework for the classification of 633 uncultured single-stranded DNA viruses of the family Genomoviridae. Virus Evol 634 2017;3(1):vew037. 635 48. Varsani A, Krupovic M. Smacoviridae: A new family of animal-associated single- 636 stranded DNA viruses. Arch Virol 2018;163(7):2005-2015. 637 49. Rosario K, Breitbart M, Harrach B, Segalés J, Delwart E et al. Revisiting the 638 of the family Circoviridae: establishment of the genus Cyclovirus and removal of the genus 639 Gyrovirus. Arch Virol 2017;162(5):1447-1463. 640 50. Lefkowitz EJ, Dempsey DM, Hendrickson RC, Orton RJ, Siddell SG et al. Virus 641 taxonomy: the database of the International Committee on Taxonomy of Viruses (ICTV). Nucleic 642 Acids Res 2018;46(Database issue):D708-D717. 643 51. Koonin EV, Dolja VV, Krupovic M. Origins and evolution of viruses of eukaryotes: The 644 ultimate modularity. Virology 2015;479–480:2-25. 645 52. Koonin EV, Ilyina TV. Computer-assisted dissection of rolling circle DNA replication. 646 Biosystems 1993;30(1):241-268. 647 53. Kazlauskas D, Dayaram A, Kraberger S, Goldstien S, Varsani A et al. Evolutionary 648 history of ssDNA bacilladnaviruses features horizontal acquisition of the capsid gene from ssRNA 649 nodaviruses. Virology 2017;504:114-121. 650 54. Gorbalenya AE, Koonin EV. Helicases: amino acid sequence comparisons and structure- 651 function relationships. Curr Opin Struc Biol 1993;3(3):419-429. 652 55. Gorbalenya AE, Koonin EV, Wolf YI. A new superfamily of putative NTP-binding 653 domains encoded by genomes of small DNA and RNA viruses. FEBS Lett 1990;262(1):145-148. 654 56. Williams T, Virto C, Murillo R, Caballero P. Covert infection of insects by 655 baculoviruses. Front Microbiol 2017;8:1337-1337. 656 57. Breitbart M, Delwart E, Rosario K, Segalés J, Varsani A et al. ICTV virus taxonomy 657 profile: Circoviridae. J Gen Virol 2017;98(8):1997-1998. 658 58. Longdon B, Jiggins FM. Vertically transmitted viral endosymbionts of insects: Do sigma 659 viruses walk alone? Proc Biol Sci 2012;279(1744):3889-3898. 660 59. Cory JS. Insect virus transmission: Different routes to persistence. Curr Opin Insect Sci 661 2015;8:130-135. 662 60. Fine PEM. Vectors and vertical transmission: An epidemiologic perspective. Ann NY Acad 663 Sci 1975;266(1):173-194. 664 61. Liu S, Xie J, Cheng J, Li B, Chen T et al. Fungal DNA virus infects a mycophagous 665 insect and utilizes it as a transmission vector. Proc Natl Acad Sci USA 2016;113(45):12803–12808. 666 62. Wei J, He Y-Z, Guo Q, Guo T, Liu Y-Q et al. Vector development and vitellogenin 667 determine the transovarial transmission of begomoviruses. Proc Natl Acad Sci USA 668 2017;114(26):6746-6751. 669 63. Liu B, Preisser EL, Chu D, Pan H, Xie W et al. Multiple forms of vector manipulation 670 by a plant-infecting virus: Bemisia tabaci and Tomato yellow leaf curl virus. J Virol 671 2013;87(9):4929-4937. 672 64. Clem RJ, Passarelli AL. Baculoviruses: sophisticated pathogens of insects. PLOS Pathog 673 2013;9(11):e1003729. 674 65. Wang F, Fang Q, Wang B, Yan Z, Hong J et al. A novel negative-stranded RNA virus 675 mediates sex ratio in its parasitoid host. PLOS Pathog 2017;13(3):e1006201. 676 66. Male MF, Kraberger S, Stainton D, Kami V, Varsani A. Cycloviruses, 677 gemycircularviruses and other novel replication-associated protein encoding circular viruses in 678 Pacific flying fox (Pteropus tonganus) faeces. Infect Genet Evol 2016;39:279-292. 450 Author Statements

451 Authors and contributions 452 KR and MB designed study. KR, KAM, AMG, and MB collected specimens. KR and KAM 453 performed experimental work and data analysis. AMG performed transmission electron 454 microscopy analysis. KR wrote the manuscript and prepared figures. KAM and MB reviewed and 455 edited manuscript. All authors approved final draft. 456 457 Conflicts of interest 458 The authors declare that there are no conflicts of interest.

459

460 Funding information 461 This work was funded through NSF Assembling the Tree of Life Program grant DEB-1239976 to 462 KR and MB. 463 464 Acknowledgements

465 Authors would like to thank Alexandria Creasy, Catherine Scott, and William Landry for providing 466 samples, Mason Kerr for help with sample processing, and Bentley Fane for advice on nuclease 467 treatment assays. 468 Tables and Figure Legends 469 470 Table 1. Primers used for PCR assays performed during this study. 471 472 Table 2. Summary of viral isolates representing circularisviruses. 473 474 Figure 1. Genome organization (A), conserved motifs (B) and pairwise sequence comparisons (C) 475 for circularisviruses. Panel A shows a genome schematic of major open reading frames found 476 within circularisvirus genomes, including replication-associated (Rep, red arrow) and capsid (Cp, 477 grey arrow) protein coding regions, and the nonanucleotide motif marking the putative origin of 478 replication within a predicted stem-loop structure. Lower case letters within the nonanucleotide 479 motif indicate nucleotides that were observed in a single genome sequence and ‘n’ represents any 480 nucleotide. Panel B depicts conserved motifs found within the Rep through amino acid sequence 481 logos generated in WebLogo 3. The endonuclease domain sequence logo, including rolling circle 482 replication (RCR) motifs I through III, is shown above the superfamily 3 (SF3) helicase domain 483 logo, which includes Walker-A, Walker-B, Motif C, and arginine (R) finger motifs. Panel C shows 484 pairwise identities calculated using complete genomes (left) or predicted amino acid sequences 485 (right) for the Rep (dark grey) and Cp (light grey). The information for each panel was extracted 486 from the nine genome sequences representing members of the circularisviruses, namely dragonfly 487 circularisvirus (DfCir), two isolates of spinybacked orbweaver circular virus 1 (SpOrbCV-1A and 488 SpOrbCV-1B), golden silk orbweaver associated circular virus 1 (GoSOrbACV-1), longjawed 489 orbweaver circular virus 1 (LjOrbCV-1), cybaeus spider associated circular virus 1 (CySACV-1), 490 pacific flying fox feces associated circular DNA virus 6 (PfffACV-6), and panicum ecklonii- 491 associated virus (PeaV). 492 493 Figure 2. Schematic representation of methodological framework used to elucidate the single- or 494 double-stranded nature of spinybacked orbweaver circular virus 1 DNA found within virions and 495 spinybacked orbweaver tissues. Note that S1 nuclease degrades ssDNA templates (dotted line) 496 non-specifically (left panel), whereas EcoRI restriction endonuclease nicks dsDNA at a specific 497 site (shown as a gap) (right panel). PCR was used to confirm endonuclease treatment results. Grey 498 arrows represent primer positions. PCR results for DNA from virions are highlighted in blue, 499 whereas results for DNA from tissues are highlighted in orange. Note that S1 nuclease treated 500 DNA from spinybacked orbweaver tissues was then treated with EcoRI to confirm its double- 501 stranded nature (indicated by the orange spider). An example gel image is provided in 502 Supplemental Figure S2.

503 Figure 3. Images of spinybacked orbweaver life stages investigated during this study (A) and 504 percentage of specimens from each life stage that were positive for spinybacked orbweaver circular 505 virus (SpOrbCV) 1 and 2 (B). Each specimen category on the x-axis specifies if the data reflects 506 the prevalence for specimens collected during 2017, 2018 or both years and the number (N) of 507 individuals processed. Prevalence in adult specimens based on whole body (WB) and leg (L) 508 samples are distinguished.

509 Figure 4. Transmission electron microscope images of virus-like particles (VLPs) found in leg 510 tissues from spiders that tested positive for spinybacked orbweaver circular virus 1. VLPs had an 511 average diameter of 25 nm and each image originates from a different specimen.

512 513 514 Supplemental Table and Figure Legends

515 Table S1. List of spider species tested for spinybacked orbweaver circular viruses (SpOrbCV). 516 517 Figure S1. Transmission electron microscope images showing a second type of virus-like particles 518 (VLPs) observed within the leg tissues of a spinybacked orbweaver. Observed VLPs averaged ~66 519 nm in diameter. 520 521 Figure S2. Example gel image showing results for PCR assays targeting DNA extracted from 522 virions (top) or spider tissues (bottom) after treatment with S1 nuclease or EcoRI restriction 523 enzyme. Samples include an untreated positive control (U), PCR products from virion or tissue 524 samples (S), and a negative control (N). Table 1. Primers used for PCR assays performed during this study.

Targeted Primer Product Annealing Purpose Primer pair and sequences (5’ – 3’) virus positions1 length2 temp (˚C)3 Complete genome SpOrbCV1_F1: CACCACCTCCTGATTTAG 377 – 394 SpOrbCV-1 2 kb 42 recovery (inverse PCR) SpOrbCV1_R1: CATTTAAGGATGTTTTGCTT 395 – 414 Sample screening for SpOrbCV1_F2: ACGGATGAACAAGCCTGGAT 455 – 474 SpOrbCV-1 1.2 kb 50 prevalence estimates SpOrbCV1_R2: ACCGTCGAAGAAGAAGAAACG 1038 – 1058 Genome characterization SpOrbCV1_F3: ATCCAGGCTTGTTCATCCGT 455 – 474 SpOrbCV-1 600 bp 52 assays SpOrbCV1_R3: CGTTTCTTCTTCTTCGACGGT 1038 – 1058 Nested PCR for genome SpOrbCV1_F3n1: TCTGTTAAGTTTCGAGTTGAGGA 496 – 518 SpOrbCV-1 200 bp 52 characterization assays SpOrbCV1_R3n1: AGCCAAACACAATGCGAAGG 673 – 692 Nested PCR for genome SpOrbCV1_F3n2: GTGGAACCTTCGCATTGTGT 667 – 686 SpOrbCV-1 230 bp 52 characterization assays SpOrbCV1_R3n2: CTGGCGGGGAGAAAATGAAC 875 – 894 Sample screening for SpOrbCV2_F1: ATCTTATCTGGGGCCGATGG 24 – 43 SpOrbCV-2 1.1 kb 54 prevalence estimates SpOrbCV2_R1: AACATTTCTGCGGCCTCTTG 639 – 658 phiX174_F1: TCAGATATGGACCTTGCTGCT 5034 – 5054 phiX174 S1 nuclease experiment 625 bp 54* phiX174_R1: TCCTTGACGAACGTGCCAA 254 – 272 phiX174_F3: ACGCCGTTCAACCAGATATTG 5226 – 5246 phiX174 S1 nuclease experiment 530 bp 53* phiX174_R3: GGTTGAACAGCATCGGACTC 352 – 371 1Genome positions for each primer used to target spinybacked orbweavers circular viruses (SpOrbCV) and bacteriophage phiX174 (GenBank accession number NC_001422). 2Product sizes are listed in kilobases (kb) or base pairs (bp). 3Annealing temperatures highlighted with an asterisk (*) were incrementally decreased by 0.1˚C every cycle.

Table 2. Summary of viral isolates representing circularisviruses.

Accession Isolation source Virus1 Isolate Reference No.2 Source Scientific name Sample type Dragonfly circularisvirus JX185415 TO-DF3E-2010 Dragonfly Pantala flavescens Abdomen [27] Cybaeus spider ACV-1 MH545522 BC_I1644C_F12 Spider Cybaeus signifer Whole specimen [2] Golden silk orbweaver ACV-1 MH545520 PR_I0960_F8 Spider Nephila clavipes Whole specimen [2] Longjawed orbweaver CV-1 MH545521 BC_I1601_F12 Spider argyra Whole specimen [2] Spinybacked orbweaver CV-1 MH545518 FL_I0831_I69-H8 Spider Gasteracantha cancriformis Whole specimen [2] Spinybacked orbweaver CV-1 MH545519 FL_I1594-I104_E11 Spider Gasteracantha cancriformis Whole specimen [2] Pacific flying fox faeces ACV-6 KT732823 Tbat_H_77994 Bat Pteropus tonganus Feces [66] Panicum ecklonii AV MH425571 2-82-I Grass Panicum ecklonii Leaf [41] Odonata ACV-13 KM598396 US-1591LM1-12 Dragonfly Libellula quadrimaculata Abdomen [46] 1Associated circular virus (ACV); circular virus (CV); associated virus (AV) 2GenBank accession number A) Genome organization B) Rep conserved motifs

(C/T)A(G/n)T(A/g)(T/c)TAC RCR I RCR II RCR III 3

2 bits 1 C S F K T V LI L AF K YVY RY Y S 0 W F Q Y ) T H G YC KD p

C R-Finger ( Walker-A Walker-B Motif C

3

d

i 1960 - 2070 nt

s

p

a C 2 bits 1 S L D KA F F C V LAR H I R L LE VQT F KS GI I V Q MQ M I RS C 0 GPP GKD V I SD VTSNY A R C) Pairwise identities

Genome Cp PfffACV-6 SpOrbCV-1A SpOrbCV-1B OdACV-13 CySACV-1 GoSOrbCV-1

LjOrbCV-1 PeaV PfffACV-6 DfCirc PeaV PeaV KT 732823 SpOrbCV-1A DfCirc KM 598396 SpOrbCV-1A JX 185415 SpOrbCV-1B SpOrbCV-1B MH 425521 GoSOrbACV-1 OdACV-13 MH 425518 OdACV-13 CySACV-1 MH 425519 DfCirc GoSOrbCV-1 MH 425520 LjOrbCV-1 LjOrbCV-1 MH 425571

CySACV-1 PfffACV-6 MH 425522

Rep JX 185415 KT 732823 KT KM 598396 MH 425571 MH 425521 MH 425520 MH 425518 MH 425519 MH 425522 JX 185415 KT 732823 KT MH 425571 MH 425521 KM 598396 MH 425522 MH 425520 MH 425518 MH 425519 Pairwise identity (%)

20 28 36 44 52 60 68 76 84 92 100 ? ?

S1 nuclease digest EcoRI restriction digest

PCR PCR

- + + - A) Investigated spinybacked orbweaver life stages

Adult Eggs Spiderlings

B) Viral prevalence in specimens representing each life stage

100 SpOrbCV-1 SpOrbCV-2

79% 79% 80 74% 65% 60

40 36%

Positive samples (%) 20 2.5% 0% 0% Adults-WB Adults-L Eggs Spiderlings N = 28 N = 121 N = 26 N = 14 2017 2017-2018 2017-2018 2018

Table S1. List of spider species tested for spinybacked orbweaver circular viruses (SpOrbCV)

Number of SpOrbCV-1 SpOrbCV-2 Species1 Family Location2 Specimens Positives Positives Giant house spider BC 11 0 0 (Eratigena atrica) Cybaeus signifer BC 4 0 0 ND Cybaeidae BC 1 0 0 Woodlouse hunter spider BC 7 0 0 (Dysdera crocata) Sierra dome spider BC 4 0 0 (Neriere litigiosa) Running crab spider Philodromidae BC 8 0 0 (Philodromidae sp) altioculata Pimoidae BC 3 0 0 Zebra jumper spider Salticidae BC 7 0 0 (Salticus scenicus) Segestria pacifica Segestriidae BC 3 0 0 Common house spider BC 6 0 0 (Parasteatoda tepidariorum) Western black widow spider Theridiidae BC 3 0 0 (Latrodectus hesperus) False black widow spider Theridiidae BC 3 0 0 (Steatoda grossa) Candy stripe spider Theridiidae BC 2 0 0 (Enoplognatha ovata) Goldenrod crab spider BC 3 0 0 (Misumena vatia) Leucauge sp Tetragnathidae SPE-A, FL 4 0 0 Multiple (ND) ND SPE-A, FL 8 0 1 Cyrtophora sp Araneidae SPE-B, FL 8 0 0 Multiple (ND) ND SPE-B, FL 15 0 0 Leucauge sp Tetragnathidae SPE-C, FL 1 0 0 Multiple (ND) ND SPE-C, FL 2 0 0 ND ND FL 1 0 0 Leucauge sp Tetragnathidae FL 1 0 1 ND ND FL 1 0 0 1Common and scientific names are listed whenever possible. However, some specimens were not taxonomically identified (ND; not determined). Specimens representing more than one unidentified species collected from the same site were merged into the same field and labeled as ‘multiple’.

2Specimens were collected from outdoors in British Columbia (BC) and the Tampa Bay area in Florida, USA (FL). Some specimens were collected from three screen pool enclosures (SPE) distinguished by letters A through C.

Figure S1

Figure S1. Transmission electron microscope images showing a second type of virus-like particles (VLPs) observed within the leg tissues of a spinybacked orbweaver. Observed VLPs averaged ~66 nm in diameter. Figure S2

Virions

S1 nuclease EcoRI

U S1 S2 S3 S4 S5 N S1 S2 S3 S4 S5 N

Tissues

S1 nuclease S1 followed by EcoRI

U S6 S7 S8 S9 S10 N S6 S7 S8 S9 S10 N

U S6 S7 S8 S9 S10 N S6 S7 S8 S9 S10 N Figure S2. Example gel image showing results for PCR assays targeting DNA extracted from virions (top) or spider tissues (bottom) after treatment with S1 nuclease or EcoRI restriction enzyme. Samples include an untreated positive control (U), PCR products from virion or tissue samples (S), and a negative control (N).