Short ‘1.2× Genome’ Infectious Clone Initiates Kolmiovirid Replication in Boa constrictor Cells

Human hepatitis D virus (HDV) depends on hepatitis B virus co-infection and its glycoproteins for infectious particle formation. HDV was the sole known deltavirus for decades and believed to be a human-only pathogen. However, since 2018, several groups reported finding HDV-like agents from various hosts but without co-infecting hepadnaviruses. In vitro systems enabling helper virus-independent replication are key for studying the newly discovered deltaviruses. Others and we have successfully used constructs containing multimers of the deltavirus genome for the replication of various deltaviruses via transfection in cell culture. Here, we report the establishment of deltavirus infectious clones with 1.2× genome inserts bearing two copies of the genomic and antigenomic ribozymes. We used Swiss snake colony virus 1 as the model to compare the ability of the previously reported “2× genome” and the “1.2× genome” infectious clones to initiate replication in cell culture. Using immunofluorescence, qRT-PCR, immuno- and northern blotting, we found the 2× and 1.2× genome clones to similarly initiate deltavirus replication in vitro and both induced a persistent infection of snake cells. The 1.2× genome constructs enable easier introduction of modifications required for studying deltavirus replication and cellular interactions.


Introduction
Hepatitis D virus (HDV) is a unique human pathogen. Three years after its discovery in 1977 in liver specimens of chronically hepatitis B (HBV)-infected patients [1], Rizzetto and colleagues identified it as a satellite virus of HBV [2]. One can contract HDV in two different ways, either through acute co-infection with HBV or through superinfection as a chronic HBV carrier. HBV and HDV co-infection is clinically more severe than HBV mono-infection; however, the infection usually resolves, resulting in the clearance of both viruses. Superinfection of a chronic HBV carrier by HDV results in the most severe form of viral hepatitis; these patients often face hepatic cirrhosis and development of hepatocellular carcinoma [3]. HDV is a satellite virus that utilizes the envelope proteins of HBV to assemble infectious viral particles; however, the replication of HDV within the host cell proceeds independently of HBV [4]. The single-stranded RNA genome of HDV is around 1.7 kilonucleotides (knt) long, although because of high self-complementary, it forms a double-stranded rod-like structure [5,6]. Within the cell, HDV gives rise to three different RNA species: the genome, the antigenome, and the mRNA. The antigenome is the exact complement of the genome, while the mRNA mediates the expression of the delta antigen (DAg), the sole protein encoded by the HDV genome [5,7]. During the viral life cycle, the DAg is present in two different forms, small (SDAg) and large DAg (LDAg) [8]. Cellular editing mediated by adenosine deaminase acting on RNA (ADAR1) converts the amber stop codon of the SDAg on the antigenomic strand to a tryptophan codon, thus allowing monax deltavirus [19,21]. Here, we describe a construct containing 1.2× SwSCV-1 genome, which, in a manner similar to the previously described 2× SwSCV-1 infectious clone [17], initiates virus replication, produces infectious particles upon superinfection with Haartman institute snake virus 1 (HISV-1), and results in persistent infection of the cells. Because we did not manage to decipher the exact compositions of the plasmid-based 1.2× [30] and 1.1× [31] genome constructs reported earlier, we based our 1.2× infectious clone on the RNA transfection studies of Macnaughton and Lai [29] with the aim to generate shorter DNA constructs to facilitate introduction of mutations or modifications to the virus genome. Additionally, we wanted to eliminate the T7 promoter we used in the 2× SwSCV-1 infectious clone to avoid the risk of DAg expression via this promoter due to cellular polymerases, which is reported to occur in mammalian cells [32,33].
To study infectious particle formation of the SwSCV-1-infected cell lines, we conducted superinfection studies with HISV-1 [35], earlier demonstrated to be an efficient helper virus for SwSCV-1 [17]. The superinfection studies and detection followed the protocol described [17]. For the infection, we used 600 copies of HISV-1 S segment RNA per cell, which corresponds roughly to a multiplicity of infection (MOI) of 10.

Transfection
Lipofectamine 2000 (ThermoFisher Scientific) reagent served for transfection of I/1Ki cells as described [17,37]. Briefly, we mixed 500 ng of plasmid DNA in 50 µL of OptiMEM (ThermoFisher Scientific) and 3 µL of Lipofectamine 2000 in 47 µL of OptiMEM (ThermoFisher Scientific) by pipetting up and down, and allowed the complexes to form for 15-30 min at room temperature (RT). We added 1 mL of trypsinized cells (suspension containing approximately 1.8 cm 2 of cells per ml) to the mixture and allowed the suspension to stand at RT for 15-30 min before plating. At 5-6 h post plating, we replaced the transfection mixture by fully supplemented medium and incubated the cells as described above. We scaled up the above reaction volumes depending on the amount of cells needed for each experiment.

Western Blot (WB)
For WB, we washed the cells grown on plates or flasks twice with PBS, scraped them into PBS, pelleted by centrifugation (500× g, 3-5 min), lysed the cell pellets by RIPA buffer (50 mM Tris, 150 mM NaCl, 1% Tx-100, 0.1% SDS, 0.5% sodium deoxycholate, protease inhibitor cocktail), and measured the protein concentration using the Pierce TM BCA Protein Assay Kit (ThermoFisher Scientific). For comparing DAgs of different kolmiovirids, we collected from a 12-well plate nontransfected I/1Ki cells and those transfected with FWD and REV constructs at 4 days post transfection in 100 µL of Laemmli sample buffer after two washes with PBS. We separated an equal amount of protein (or volume, 30 µL/lane, for the experiments done on 12-well plate) for each sample on SDS-PAGE using 4-20% Mini-PROTEAN ® TGX gels (Bio-Rad, Hercules, CA, USA), and

Transfection
Lipofectamine 2000 (ThermoFisher Scientific, Waltham, MA, USA) reagent served for transfection of I/1Ki cells as described [17,37]. Briefly, we mixed 500 ng of plasmid DNA in 50 µL of OptiMEM (ThermoFisher Scientific, Waltham, MA, USA) and 3 µL of Lipofectamine 2000 in 47 µL of OptiMEM (ThermoFisher Scientific, Waltham, MA, USA) by pipetting up and down, and allowed the complexes to form for 15-30 min at room temperature (RT). We added 1 mL of trypsinized cells (suspension containing approximately 1.8 cm 2 of cells per ml) to the mixture and allowed the suspension to stand at RT for 15-30 min before plating. At 5-6 h post plating, we replaced the transfection mixture by fully supplemented medium and incubated the cells as described above. We scaled up the above reaction volumes depending on the amount of cells needed for each experiment.

Immunofluorescence Staining
For immunofluorescence (IF) staining, we plated the cells on collagen coated (10 µg/cm 2 type I rat tail collagen (BD Biosciences, Franklin Lakes, NJ, USA) in 25 mM acetic acid, O/N at 4 • C) CellCarrier-96 Ultra plates (PerkinElmer, Waltham, MA, USA). After the removal of culture media, cells were fixed by incubation in 4% paraformaldehyde in PBS for~15 min at RT. The IF staining followed the protocol described [17]. We used directly labeled anti-SDAg-AF488 [17]

Detection of Circular RNA Genome
To show the circularity of the SwSCV-1 genome, we performed reverse transcription (RT) with two different SwSCV-1-specific primers: RT-1 5'-GTTTCCCCACAAATTCTTTGC-3'; RT-2 5'-CCTCTATCCTACTTCAATTCTC-3'. For the cDNA synthesis, we used Super-Script™ IV Reverse Transcriptase (ThermoFisher Scientific, Waltham, MA, USA) according to the manufacturer's recommendations. The cycling conditions for the RT reaction were the following: 5 min at 50 • C, 15 min at 55 • C, 10 min at 60 • C, and 15 min at 65 • C. Subsequently, we used three different primer pairs (PP) with neighboring 5' ends, but their 3' ends facing opposite directions, similarly to the method used by Paraskevopoulou et al. for TSRV-1 [19].

Quantitative Reverse Transcription PCR (qRT-PCR)
qRT-PCR served for quantification of viral RNA in the cells. The primers and probe were the following: forward primer 5'-GAAAGACGCGACAACTGTGAGTC-3', reverse primer 5'-GTCTAGTCCCGTTCCGGTTCTATG-3', and probe 5' 6-Fam (carboxyfluorescein)-GGAGATCCGAGAGGGGAGAAGAGGAGAGGTC-BHQ (black hole quencher)-1 3', which target SwSCV-1 RNA in genomic orientation. We isolated RNA for qRT-PCR using the GeneJET RNA Purification Kit (ThermoFisher Scientific, Waltham, MA, USA) with the addition of carrier RNA when purifying RNA from cell culture supernatants. We used TaqMan ® Fast Virus 1-Step Master Mix (ThermoFisher Scientific, Waltham, MA, USA) to set up 10 µL (half volume) reactions according to the manufacturer's recommendations with the addition of 8% DMSO to prevent secondary structure formation. The AriaMX real-time PCR system (Agilent, Santa Clara, CA, USA) served for thermal cycling of the To generate a control RNA for copy-level quantification, we used the SwSCV-1 FWD plasmid described in Szirovicza et al., 2020 [17]. Briefly, FastDigest SmaI (ThermoFisher Scientific, Waltham, MA, USA) following manufacturer's protocol served for linearization of the plasmid. The GeneJet Gel Purification Kit (ThermoFisher Scientific, Waltham, MA, USA) served for purification of the linearized plasmid after agarose gel separation. We used the TranscriptAid T7 High Yield Transcription Kit (ThermoFisher Scientific, Waltham, MA, USA) according to the manufacturer's protocol to in vitro transcribe the target RNA. Subsequently, we purified the RNA by the GeneJET RNA Purification Kit (ThermoFisher Scientific, Waltham, MA, USA), diluted it into diethyl pyrocarbonate-treated water, and stored the RNA in aliquots at −80 • C until use. The NanoDrop 2000 spectrophotometer (ThermoFisher Scientific, Waltham, MA, USA) served for quantification of the control RNA, and an online copy number calculator (http://endmemo.com/bio/dnacopynum. php; accessed on 16 December 2021) for converting the concentration to RNA copies per microliter. We ran the control RNA as 10-fold dilution series in duplicates for each run to generate a standard curve for estimation of RNA copy numbers in cell and cell culture supernatant samples.
We generated a short (~850 nucleotides long) control RNA fragment using the SwSCV-1 FWD plasmid described in Szirovicza et al., 2020 [17]. We used FastDigest EcoRV and Acc65I (ThermoFisher Scientific, Waltham, MA, USA) to linearize and digest the plasmid. Then, we purified the fragment of interest-containing the T7 promoter-from agarose gel using the GeneJet Gel Purification Kit. The plasmid fragment was further purified using SPRIselect magnetic beads (Beckman Coulter, Brea, CA, USA). TranscriptAid T7 High Yield Transcription Kit (ThermoFisher Scientific, Waltham, MA, USA) then served for in vitro transcription of the target RNA according to the manufacturer's protocol. Finally, we cleaned up the RNA using the GeneJET RNA Purification Kit (ThermoFisher Scientific, Waltham, MA, USA) and stored the RNA at −80 • C until further usage. The control RNA fragment contains the target sequence for the SwSCV-1 DAg mRNA probe described above.

Transfection with 1.2× Genome Construct Initiates Replication of Kolmiovirids
In our previous study, we showed in a transfection-based assay that a plasmid bearing the SwSCV-1 genome in duplicate could initiate SwSCV-1 replication in cell culture, with highest efficacy observed in boid kidney cells [17]. We hypothesized, based on RNA transfection studies with HDV [11], that a plasmid bearing a shorter 1.2× genome-length insert would suffice to initiate translation. With the idea that duplicating the antigenomic and genomic ribozymes would better facilitate replication, we ordered synthetic genes representing the 1.2× genome of SwSCV-1, TSRV-1, DabDV-1, CITV-1, and HDV-1. Figure 1 shows the organization of the synthetic blocks that we subsequently cloned in both reverse and forward orientation into a pCAGGS/MCS expression vector under the CAG promoter. Unlike in our earlier study, in which we included an additional T7 promoter in the antigenomic orientation, upstream of the DAg to the synthetic construct [17], we did not include additional promoters that could unintentionally facilitate DAg translation. The resulting constructs we named according to the 1.2× "kolmiovirid" FWD and 1.2× "kolmiovirid" REV scheme. The FWD constructs drive transcription of the respective kolmiovirid genome, due to which the expression or translation of the DAg should only occur following virus replication, since the DAg ORF is in antigenomic orientation. On the other hand, the REV constructs would generate antigenomic transcripts that would likely also mediate the DAg translation.
To test if the shorter constructs will facilitate replication, and to estimate the crossreactivity of our rabbit anti-SwSCV-1 DAg antiserum, we transfected I/1Ki cells with each of the constructs. IF staining of the cells transfected with REV constructs for DAg at 4 days post transfection (dpt) showed that the anti-SwSCV-1 DAg antiserum clearly cross-reacted with TSRV-1 and HDV-1, but also to some extent with DabDV-1 and CITV-1 (Figure 2A). We detected DAg in cells transfected with the FWD constructs of HDV-1, TSRV-1, and SwSCV-1, suggesting that transfection resulted in replication initiation ( Figure 2B). The staining for DAg of DabDV and CITV-1 was much less prominent on the cells transfected with FWD construct; however, the results suggest that replication occurs also with these viruses. We also performed western blot (WB) on the transfected cells and we were able to detect DAgs of SwSCV-1, HDV-1, TSRV-1, and DabDV-1 4 days after transfection with the REV constructs. However, after transfection with the FWD constructs, we were only able to detect bands for DAgs of SwSCV-1, HDV-1, and TSRV-1 ( Figure 2C). The observed differences between the results of IF staining and WB could be due to the ability of the antibody to bind conformational epitopes in case of IF staining; however, in WB this would be less likely.

The 1.2× and 2× SwSCV-1 Genome Infectious Clones Induce Similar Infection as Judged by Antigen Expression and Replication
To compare the ability of the shorter 1.2× genome length constructs to initiate replication, we transfected I/1Ki cells with 1.2× SwSCV-1 FWD and 1.2× SwSCV-1 REV and compared the antigen expression and SwSCV-1 RNA production between 1 and 5 dpt. IF staining of the transfected cells at 1-4 dpt demonstrates that both 1.2× and 2× genome constructs efficiently drive the expression of DAg (Figure 3).
To better estimate the amount of DAg produced by the different constructs, we performed WB on samples collected between 1 and 4 dpt. The results showed that similarly to the 2× genome constructs, the transfection of 1.2× constructs led to an increasing amount of DAg expression during the interval studied ( Figure 4A). As the amount of DAg appeared to increase until 4 dpt and because the results of the IF staining suggested minute differences in the amount of DAg expression through the different construct, we performed WB analysis of samples collected at 5 dpt. The results indicate that the DAg expression level in the cells is similar at 5 dpt, regardless of the construct used for transfection ( Figure 4B). DAg expression in REV constructs could be driven by the CAG promoter of the plasmid, but since levels were similar at 5 dpt, we interpret the result to suggest that DAg expression is due to replication. To add another dimension to the determination of the replication efficiency, we analyzed I/1Ki cells transfected with the four different constructs for SwSCV-1 RNA levels at 3 and 6 dpt by qRT-PCR. The 2× SwSCV-1 (especially the FWD) genome construct appeared to initiate replication more rapidly as demonstrated by the higher amount of SwSCV-1 RNA at 3 dpt ( Figure 4C). However, at 6 dpt, the cells transfected with each of the constructs showed similar SwSCV-1 RNA levels when normalized against GAPDH mRNA ( Figure 4C), supporting the WB-based interpretation that all of the studied constructs can initiate replication.
To compare the ability of the shorter 1.2× genome length constructs to initiate replication, we transfected I/1Ki cells with 1.2× SwSCV-1 FWD and 1.2× SwSCV-1 REV and compared the antigen expression and SwSCV-1 RNA production between 1 and 5 dpt. IF staining of the transfected cells at 1-4 dpt demonstrates that both 1.2× and 2× genome constructs efficiently drive the expression of DAg (Figure 3). To better estimate the amount of DAg produced by the different constructs, we performed WB on samples collected between 1 and 4 dpt. The results showed that similarly to the 2× genome constructs, the transfection of 1.2× constructs led to an increasing amount of DAg expression during the interval studied ( Figure 4A). As the amount of DAg appeared to increase until 4 dpt and because the results of the IF staining suggested minute differences in the amount of DAg expression through the different construct, we performed WB analysis of samples collected at 5 dpt. The results indicate the determination of the replication efficiency, we analyzed I/1Ki cells transfected with the four different constructs for SwSCV-1 RNA levels at 3 and 6 dpt by qRT-PCR. The 2× SwSCV-1 (especially the FWD) genome construct appeared to initiate replication more rapidly as demonstrated by the higher amount of SwSCV-1 RNA at 3 dpt ( Figure 4C). However, at 6 dpt, the cells transfected with each of the constructs showed similar SwSCV-1 RNA levels when normalized against GAPDH mRNA ( Figure 4C), supporting the WB-based interpretation that all of the studied constructs can initiate replication.

Superinfection of Cells Transfected with 1.2× SwSCV-1 FWD Construct Induces Infectious Particle Formation
To show that the 1.2× SwSCV-1 construct not only initiates virus replication in cell culture, but also induces infectious particle formation in the presence of a suitable helper virus, we superinfected 1.2× and 2× SwSCV-1 FWD transfected cells with HISV-1, a hartmanivirus demonstrated to act as a helper for SwSCV-1 [17]. We titrated the supernatants collected at 3, 6, and 9 dpi with HISV-1 on clean I/1Ki cells and used supernatants collected from non-superinfected cells as the control. IF staining of cells inoculated with the supernatants at 4 dpi for the DAg served for detecting the infected cells ( Figure 5A). We determined the number of infectious units by counting the fluorescent foci at each time point, and the results showed I/1Ki cells transfected with 1.2× or 2× SwSCV-1 FWD constructs to be equally effective in producing infectious particles following superinfection ( Figure 5B). As observed for 2× SwSCV-1 FWD in our earlier study [17], the non-superinfected 1.2× SwSCV-1 FWD cells were not able to produce infectious SwSCV-1 particles ( Figure 5A).

Transfection of Cells with the 1.2× SwSCV-1 Construct Results in Persistent Infection
In our previous study, we showed that by maintaining I/1Ki cells after transfection with the 2× SwSCV-1 FWD construct, we could generate persistently SwSCV-1-infected cell lines [17]. At the time of preparing this manuscript, we have maintained the I/1Ki-2×∆ cell line for 2.5 years, and IF staining for DAg shows the cell line to be persistently SwSCV-1-infected ( Figure 6A). To compare the replication behavior of the shorter construct further, we transfected I/1Ki cells with 1.2× SwSCV-1 FWD and continued passaging the cells. Analysis of the cells by IF staining for DAg at 8 months post initial transfection indicates that also the 1.2× SwSCV-1 FWD construct can induce persistent infection in I/1Ki cells ( Figure 6A). We compared the generated cell line, I/1Ki-1.2×∆, to I/1Ki-2×∆ cells further by analyzing the amount of DAg expression using WB. The results show that DAg expression by I/1Ki-1.2×∆ cells is at least at the level observed in I/1Ki-2×∆ cells ( Figure 6B), supporting the observation of a similar replication efficiency. To further compare the cell lines, we set up a near-infrared fluorescent northern blot assay for detection of the genomic RNA, antigenomic RNA, and DAg mRNA. As an additional control, we included an in vitro transcribed RNA of approximately 850 nucleotides corresponding roughly to the size of SwSCV-1 DAg mRNA. Initially, we prepared the samples for the denaturing agarose gel run by using an "in-house" loading dye described by Mansour and Pestov [43], but ran the RNA marker with the loading dye provided by New England Biolabs (NEB). To our surprise, the northern blot of RNA isolated from I/1Ki-1.2×∆ and I/1Ki-2×∆ cells using a probe targeting the genomic RNA resulted in the detection of a doublet band migrating at around 2.8 kilonucleotides (knt) instead of the expected 1.7 knt as compared to the RNA marker ( Figure 6C left panel). To study if the use of two different loading dyes had significantly affected the migration of the RNA, we ran the RNAs extracted from I/1Ki-2×∆, I/1Ki-1.2×∆, and clean cells as well as the in vitro-transcribed control RNA and the RNA marker in parallel with both loading dyes. Indeed, the result showed the loading dye to significantly affect the migration of the RNA, and indicated that the SwSCV-1 genomic RNA is approximately 1.7 knt in size as judged by migration ( Figure 6C). Therefore, we speculate that the doublet bands observed in the initial run likely correspond to the circular and nicked forms of the genomic RNA. With the probe targeting genomic RNA, we also detected a band migrating at around 3.4 knt, which likely represents the genome dimer reported to be present in the infected cells by other researchers [45]. In order to detect antigenomic RNA from the persistently infected cells, we had to load 5 times more RNA, which corresponds roughly to the ratio of genomic and antigenomic RNA reported for HDV [46]. We were unable to detect DAg mRNA in the persistently infected cells, even though the probe detected the in vitro transcribed control RNA ( Figure 6C). The result thus suggests that the amount of DAg mRNA in the persistently infected cells is below our detection limit.   [15]) were separated on 4-20% Mini-PROTEAN TGX gels (Bio-Rad, Hercules, CA, USA), transferred onto nitrocellulose, and the membranes probed with rabbit α-SwSCV-1 DAg antiserum and mouse monoclonal anti-pan actin antibody. The results were recorded using Odyssey Infrared Imaging System (LI-COR Biosciences, Lincoln, NE, USA). (C) Indicated amounts of total RNA isolated from I/1Ki-2×∆, I/1Ki-1.2×∆, and clean I/1Ki cells and an in vitro-transcribed control RNA (~850 nucleotides long) were prepared using two different loading dyes (2X RNA loading dye [NEB] or "in-house" loading dye prepared according to Mansour and Pestov [43]), separated on agarose gel and transferred onto nylon membrane. Probes were targeting SwSCV-1 genomic RNA and SwSCV-1 DAg mRNA (left and middle panels) and antigenomic RNA and SwSCV-1 DAg mRNA (right panel); the bands of the marker served for visualizing the RNA targets. The results were recorded using Odyssey Infrared Imaging System (LI-COR Biosciences, Lincoln, NE, USA).
The HDV RNA genome is circular [6], and we wanted to study if the SwSCV-1 RNA genome in the persistently infected cells shares this characteristic feature. To that end, we designed RT primers to transcribe cDNA going over the potential cleavage sites of the genomic RNA using RNA extracted from I/1Ki-2×∆ and I/1Ki-1.2×∆ cells as the template ( Figure 7A-RT primer 1 and 2). To show that the genome is circular, we designed three primer pairs (PPs) targeting a region that is continuous with certainty, i.e., the DAg ORF. The PPs designed have their 3' ends facing opposite directions on the reverse-complementary template strands ( Figure 7A-PP1 to PP3). As a control, we performed the exact same reactions with the same set of templates and primers, but without the addition of the RT enzyme. With both RT primers and PPs 1-3, we succeeded in amplifying the near complete SwSCV-1 genome, from templates generated in the presence of RT enzyme, indicating that the SwSCV-1 genome is indeed circular ( Figure 7B). DAg staining in green, and the right panels show an overlay. The images were captured using Opera Phenix High Content Screening System (PerkinElmer) with 20× objective. (B) Samples of naïve I/1Ki cells, I/1Ki-2×Δ cells, I/1Ki-1.2×Δ cells, and the brain homogenates of SwSCV-1-infected boa constrictors (F18-4 and F-18-5, of [15]) were separated on 4-20% Mini-PROTEAN TGX gels (Bio-Rad), transferred onto nitrocellulose, and the membranes probed with rabbit α-SwSCV-1 DAg antiserum and mouse monoclonal anti-pan actin antibody. The results were recorded using Odyssey Infrared Imaging System (LI-COR Biosciences). (C) Indicated amounts of total RNA isolated from I/1Ki-2×Δ, I/1Ki-1.2×Δ, and clean I/1Ki cells and an in vitro-transcribed control RNA (~850 nucleotides long) were prepared using two different loading dyes (2X RNA loading dye [NEB] or "in-house" loading dye prepared according to Mansour and Pestov [43]), separated on agarose gel and transferred onto nylon membrane. Probes were targeting SwSCV-1 genomic RNA and SwSCV-1 DAg mRNA (left and middle panels) and antigenomic RNA and SwSCV-1 DAg mRNA (right panel); the bands of the marker served for visualizing the RNA targets. The results were recorded using Odyssey Infrared Imaging System (LI-COR Biosciences).
The HDV RNA genome is circular [6], and we wanted to study if the SwSCV-1 RNA genome in the persistently infected cells shares this characteristic feature. To that end, we designed RT primers to transcribe cDNA going over the potential cleavage sites of the genomic RNA using RNA extracted from I/1Ki-2×Δ and I/1Ki-1.2×Δ cells as the template ( Figure 7A-RT primer 1 and 2). To show that the genome is circular, we designed three primer pairs (PPs) targeting a region that is continuous with certainty, i.e., the DAg ORF. The PPs designed have their 3' ends facing opposite directions on the reversecomplementary template strands ( Figure 7A-PP1 to PP3). As a control, we performed the exact same reactions with the same set of templates and primers, but without the addition of the RT enzyme. With both RT primers and PPs 1-3, we succeeded in amplifying the near complete SwSCV-1 genome, from templates generated in the presence of RT enzyme, indicating that the SwSCV-1 genome is indeed circular ( Figure 7B).

Inoculation of Naïve I/1Ki Cells with SwSCV-1 Results in Productive
Lastly, we wanted to study if SwSCV-1 released from cells originally transfected with 1.2× and 2× SwSCV-1 FWD (experiment conducted 6 months post transfection) superinfected with HISV-1 would result in productive infection in naïve I/1Ki cells. We used supernatants collected at 3 dpi from the HISV-1 superinfected cells (earlier observed to contain adequate amount of infectious SwSCV-1, Figure 6A) at two dilutions, 1:5 and 1:100, to inoculate naïve I/1Ki cells, and collected samples from the inoculated cells at 3, 6, and 9 dpi. IF staining, qRT-PCR, and WB served to monitor, respectively, the increase in the number of infected cells, SwSCV-1 RNA, and DAg within the cells.
IF staining of the inoculated cells for DAg at 3 dpi showed prominent nuclear staining; however, at 6 and 9 dpi, after the infection had properly established and spread to new cells, DAg showed more pronounced cytoplasmic staining ( Figure 8A). To assess the spread of infection, we quantified the number of infected cells at each time point based on the IF staining for DAg. The increase in the number of infected cells and the DAg amount coincided with the increase in the amount of SwSCV-1 RNA in the cells as studied by qRT-PCR ( Figure 8C). The number of infected cells increased roughly 4-fold during the course of the infection ( Figure 8B). WB from the cell pellets collected 3, 6, and 9 dpi showed an increase in the amount of DAg over the course of the experiment ( Figure 8D).

Discussion
The identification of novel HDV-like agents, significantly divergent from HDV [14,15,[18][19][20][21], which until 2018 was the sole representative of the previously unassigned genus Deltavirus, has increased the interest in HDV and kolmiovirid research. The identification of HDVlike agents in various host species without traces of hepadnaviruses by others and us led to questioning the strict association of HDV and HBV. Co-incidentally, Perez-Vargas and colleagues showed that HDV is able to use helper viruses other than HBV to form infectious particles [16]. We demonstrated that SwSCV-1 efficiently utilized reptarenaand hartmaniviruses as its helpers, and that the co-expression of different arena-and orthohantavirus glycoproteins can drive infectious particle formation [17]. Construction of infectious clones is the first step in demonstrating that the sequences recovered through metatranscriptomic analyses are indeed complete and capable of driving replication. We reported generation of such a clone by inserting two copies of the SwSCV-1 genome in headto-tail fashion into a mammalian expression vector, pCAGGS [17]. The same approach was proven functional for TSRV-1 [19] as well as for the HDV-like agents in Taeniopygia guttata and Marmota monax [21]. The first HDV infectious clone contained a trimeric HDV genome, and the authors utilized a dimeric genome-containing plasmid with deletion in the DAg ORF to demonstrate the protein's role in replication [4]. Constructs containing multiples of the genome make synthetic inserts longer and complicate mutational studies because each modification needs to be inserted/generated multiple times. This motivated us to attempt generation of 1.2× genome infectious clones for initiation of kolmiovirid replication. The availability of tools and reagents at hand forced us to focus on comparing the replication initiation between 2× and 1.2× SwSCV-1 genome clones in depth, but we were also able to demonstrate that a similar approach might work for the recently identified kolmiovirids.
The replication of HDV occurs via rolling circle replication by cellular RNA polymerases [11], during which the genomic and antigenomic ribozymes cut the produced genome multimers into unit-length pieces [47]. The recently identified kolmiovirids presumably share the same replication strategy and possess the genomic and antigenomic ribozymes [14,15,19,21,48]. Based on the HDV literature [49][50][51], we reasoned that duplicating the genomic and antigenomic ribozyme sequences would facilitate initiation of replication and/or production of unit-length genome (and antigenome). Indeed, RNA transfection studies with HDV have shown 1.2× genome copies to be most efficient in induction of virus replication [29], and a similar approach has been applied to generate HDV infectious clones, although we were unable to decipher the exact organization of the constructs [30,31,52]. By applying the same principle, we constructed 1.2× genome infectious clones for HDV-1, SwSCV-1, TSRV-1, DabDV-1, and CITV-1 in both genomic and antigenomic sense, and tested the clones in I/1Ki cells, which efficiently support replication of SwSCV-1 following transfection with the 2× genome clone [17]. In the REV constructs, the CAG promoter of the pCAGGS vector should mediate DAg translation, which provides a source of DAg for the first rounds of replication. For HDV, the reports suggest existence of an internal promoter that could drive the production of DAg [28,53], such a promoter would presumably act in both our REV and FWD constructs and could also contribute to replication initiation. Indeed, the comparison of DAg production following transfection with 1.2× and 2× genome SwSCV-1 FWD and REV constructs demonstrated detectable DAg levels to appear earlier in cells transfected with REV constructs ( Figure 4A). We thus used the IF staining of DAg from I/1Ki cells transfected with the 1.2× genome REV constructs (HDV-1, TSRV-1, DabDV-1, and CITV-1) to estimate the cross-reactivity of the anti-SwSCV-1 DAg antiserum [15] with DAg of the different viruses. The antiserum appeared to cross-react best with TSRV-1 DAg that is the closest relative of SwSCV-1 from the kolmiovirids included [19]. The fact that HDV-1 DAg showed prominent nuclear staining, as would be expected based on the HDV literature [54], increased our confidence in the specificity of the IF staining. The antiserum appeared to cross-react moderately well with the DAg of DabDV-1, showing mainly cytoplasmic staining. While the CITV-1 DAg appeared to be barely detectable with the antiserum, the results did suggest production of CITV-1 DAg. Although the expression of DAg could come from internal promoters, we think that the DAg produced following transfection of FWD constructs is due to initiation of replication. While the inability of our SwSCV-1 DAg antiserum to cross-react with the DAgs of the other viruses tested likely explains the lower signal, it is also likely that the different kolmiovirids are not replicating optimally in the B. constrictor cells. We hypothesize that each of the kolmiovirids would be through promoter usage adapted to replicating in a specific host species.
Along with our data on the ability of 1.2× and 2× genome SwSCV-1 constructs to initiate replication and to induce a persistent infection, we demonstrated the presence of circular SwSCV-1 RNA in the infected cells. We also demonstrated that inoculation of naïve I/1Ki cells with supernatants containing infectious SwSCV-1 and a suitable helper virus results in productive infection. These observations strongly support our conclusion that plasmid transfection of I/1Ki cells indeed efficiently initiates SwSCV-1 replication. The 1.2× genome construct design described herein (analogous to constructs described for HDV) could help to reduce the complexity of introducing mutations, and facilitate synthetic gene design for molecular biology studies of the recently identified kolmiovirids and those that will be identified in the future. Our results with SwSCV-1 show that the 1.2× genome clone is at least as efficient as the 2× genome clone in initiation of replication. Furthermore, our results indicate that introduction of the insert in either genomic or antigenomic orientation functions equally well in the B. constrictor kidney cell model. Further studies with HDV-1, TSRV-1, DabDV-1, and CITV-1 in cell lines of various species could serve to demonstrate species specificity of the viruses, and to provide first evidence on the potential role of kolmiovirid promoters in mediating species-specific replication.