Usurpativirus massiliensis, a new giant virus related to clandestinovirus

Abstract

Giant virus discovery in 2003 revolutionized the virus paradigm. In 2015, we introduced a new host, Vermamoeba vermiformis, that led to the discovery of Faustovirus, Orpheovirus and two giant viruses (Clandestinovirus and a faustovirus) having distinct cytopathic effects despite being co-isolated from the same environmental sample and culture. This raised concerns about the possibility that the most “discreet” virus be overlooked. Here we report on Usurpativirus, a new giant virus closely related to Clandestinovirus and co-isolated with Faustovirus LCD7 in V. vermiformis. Its non-lytic replicative cycle was primarily overlooked in presence of the lytic Faustovirus. The Usurpativirus genome encode two tRNAs and 758 predicted proteins, and share 707 and 202 orthologs with Ushikuvirus and Clandestinovirus, respectively. We detected four histone-like proteins that further challenge the compaction of DNA from these large genomes into icosahedral capsids of approximately 240 nm, as well as four capsid-associated proteins, but we did not detect any predicted proteins involved in entry/fusion that could explain the special replication strategy of this virus. Scanning electron microscopy revealed two distinct morphologies for amoebae infected with Usurpativirus, which became either rounded with a smooth surface or more flattened with a wavy surface. In addition, virions were observed attached to the outside of the amoebae, which most often had a smooth surface and rarely an undulating surface after 6 days of infection. Finally, such co-infection with two distantly-related giant viruses questions on the possible interactions between each other.

Main text

Co-culture procedures for the isolation of giant viruses were considerably improved during the past 20 years, notably by using new amoeba hosts as Vermamoeba vermiformis [1]. This led to the discovery of faustoviruses and kaumoebaviruses [1–3]. Furthermore, this allowed co-isolating two giant viruses from the same sample as was the case for the first Cedratvirus virus strain and a mimivirus, which were separated by flow cytometry [4]. In 2019, co-culturing two environmental samples on V. vermiformis led to discover two new giant viruses, Clandestinovirus ST1 and Usurpativirus LCD7, that were primarily overlooked [5]. Here, we focused on the characterization of Usurpativirus massiliensis IHUMI-LCD7, which is closely related to Clandestinovirus [6], and distantly related to Medusavirus that replicate on another amoeba, Acanthamoeba castellanii [7].

Usurpativirus was recovered from sewage water collected in July 2016 in La Ciotat, Southeastern France. It harbors an icosahedral capsid (Fig. 1A) with a mean diameter of 237 ± 6 nm (n = 18). The Usurpativirus genome (GenBank accession PV862220) is a linear doubled-stranded DNA genome with a length of 669,751 base pairs (bp), exceeding that of Clandestinovirus by 87,764 bp, and a GC-content of 47.88%. A total of 758 proteins and two tRNA (His(gtg) and Val(cac)) were predicted using GeneMarkS [8] and Aragorn [9] respectively; 242 (~ 31.9%) predicted proteins were ORFans (ORF with no hits in the NCBI GenBank). A BLASTn (https://blast.ncbi.nlm.nih.gov/Blast.cgi) search revealed a close similarity (99.62%) of the IHUMI-LCD7 genome with that recently released of Ushikuvirus sp. F10 genome (NCBI BAAHMQ010000000), an unclassified Nucleocytoviricota member from a Japanese freshwater pond that was isolated on V. vermiformis. The Ushikuvirus genome has a length of 666,605 bp in two scaffolds of 652,555 and 14,050 bp. Phylogenomic analyses showed that Usurpativirus and Ushikuvirus are closely related and more remotely related to Clandestinovirus. The three giant virus genomes share 203 predicted proteins (as assessed by searching best reciprocal hits with as thresholds 10⁻⁵ for e-value and 20% for identity). Usurpativirus and Ushikuvirus share 707 orthologs whereas Usurpativirus and Clandestinovirus share 211 orthologs. Besides, a total of 28 Usurpativirus proteins were not detected in Ushikuvirus. Phylogeny based on B family DNA polymerase showed consistent results, with the clustering of Ushikuvirus, Usurpativirus and Clandestinovirus, apart from the three medusaviruses (Fig. 2). Usurpativirus shares only 46 predicted proteins with these medusaviruses, even though they are relatively close in the phylogeny. Functional annotation of Usurpativirus predicted proteins was performed using Delta-BLAST (https://www.ncbi.nlm.nih.gov/books/NBK569856/) and InterProScan (https://www.ebi.ac.uk/interpro/search/sequence/). This notably allowed detecting four capsid proteins (one major and three minors). In contrast, we did not detect homologs of A16/G9/J5 (present in poxviruses) and F9/L1, two proteins that were suspected to be associated with the envelope/capsid of viruses and to be involved in the entry/fusion complex in Nucleocytoviricota [10]. Moreover, a myristoylated membrane protein of medusaviruses involved in replication was also absent in Clandestinovirus, Ushikuvirus, and Usurpativirus. Interestingly, we did not observe orthologs for this protein in other giant viruses devoid of an internal membrane that replicate on V. vermiformis [11]. This may be related to differences in morphology and replication mechanisms between Usurpativirus, Ushikuvirus, Clandestinovirus and other giant viruses. We detected in Usurpativirus histone-like proteins (H2B/2A, H3, H4, linker H5/H1) as previously described for Marseillevirus, Medusavirus, and Clandestinovirus [12, 13]. These proteins were reported to form nucleosome-like structures and to possibly impact the level of genome compaction. In this view, Usurpativirus has a larger genome than Marseillevirus, Clandestinovirus and Medusavirus while all four virions have a similar size. It is also worthy to mention that components of putative antiviral systems were detected in the Usurpativirus genome. Indeed, we detected one protein related to the Dynamin superfamily that was recently suspected to have antiviral properties [14] and one ortholog to protein R354 that is a component of the mimivirus defense system against virophages named MIMIVIRE [15]. It is worthy to note that no virophage was detected here, nor to date in this group of viruses [16–18].

Fig. 1.

Electron microscopy of Usurpativirus massiliensis IHUMI-LCD7. (A) Negatively stained Usurpativirus particle (arrow) viewed by transmission electron microscopy (TEM) on a Tecnai G2 TEM (FEI/Thermo-Fischer) operated at 200 keV equipped with a 4096 × 4096 pixels resolution Eagle camera (FEI). (B, C) Low-magnification scanning electron microscopy (SEM) views of Usurpativirus-infected amoeba cells. Single virions can be seen on the glass slide (B, arrowheads), as well as amoeba cells (B, C; arrows). Amoeba presenting a smooth surface (B, C; white arrow) or a ruffled surface (B, C; black arrow). (D, E) Zoom-in from boxed regions in (B) and (C) depicting Usurpativirus particles (arrows) attached to the surface of smooth-surface amoeba. (F) High-magnification view of two single Usurpativirus particles (arrows). B, C, D, E and F images were captured on a SU5000 (Hitachi High-Technologies, Tokyo, Japan) SEM with Secondary-Electrons (SE) detector in high-vacuum mode at 1 kV acceleration voltage, observation mode (spot size 30) from a supernatant 6 days post-infection

Fig. 2.

Phylogenetic tree basis on DNA polymerase B predicted protein. Alignment was performed on 97 predicted proteins from Nucleocytoviricota phylum using Muscle program (standard parameter) and 3 sequences outgroup belonging to bacteria, Archaea and eukaryote domains. FastTree was used to build the tree with standard parameter and was visualized on ITol v7 online [24, 25].

Usurpativirus, as Clandestinovirus, was found to be co-isolated in a same amoebal culture with a faustovirus, a giant virus related to Asfarviridae family [5]. First, interestingly, Usurpativirus and Clandestinovirus went initially unnoticed as they replicated differently in V. vermiformis and did not lead to amoeba lysis, whereas the lytic faustoviruses caught all attention [5, 6]. The presence of Clandestinovirus and Usurpativirus were eventually revealed by genome sequencing and secondary by electron microscopy. This suggests that close relatives to these viruses and further new, unknown giant viruses could be missed, particularly in absence of next-generation sequencing of the culture supernatant.

Second, the cytopathic effect of Usurpativirus is special, differing considerably to those of the other giant viruses with the exception of Medusavirus and Clandestinovirus. Scanning electron microscopy (Fig. 1B-F) of whole-mount preparations 6 days after the infection for Usurpativirus revealed two distinct morphologies for infected amoebas, which were either (i) rounded with a smooth surface or (ii) flattened with a ruffled surface. In addition, virions were seen attached extracellularly to amoebas, which most often had a smooth surface and rarely a ruffled surface. The stage of ruffled amoeba could correspond to the appearance of the amoebas at an advanced stage of infection. Aside from Clandestinovirus, three isolates of medusavirus have been reported to cause rounding of Acanthamoeba castellanii cells and release of new virions by exocytosis [7, 19, 20]. Additional analyses of the Usurpativirus replication cycle (data not shown) using scanning microscopy revealed, 12 h after infection, the appearance of new virions on the cell surface, suggesting that the viruses are released by exocytosis from amoebas with smooth appearance, as can be observed in Fig. 2D and E.

Third, the Usurpativirus and Clandestinovirus discoveries question on their interactions with the co-infecting viruses. There is a large panel of reported interactions between viruses including mutualism, synergy, facilitation, competition, antagonism, and even parasitism in the case of mimiviruses and virophages. As a matter of fact, co-infections of same host cells by different viral species are very common [21]. Here, Usurpativirus was able to replicate alone in V. vermiformis, ruling out its strict dependence on a faustovirus. Beyond, the co-infection of V. vermiformis by Usurpativirus with a faustovirus raises questions on the possibility of sequence exchanges between co-infecting viruses within amoebas as a putative source of genomic evolution [22]. Nevertheless, we found no evidence for such transfer by comparative genomic analysis between the two viral genomes. Another hypothesis could be that, during the initial isolation of the mixture of Usurpativirus and Faustovirus, cells infected with one of these two viruses could inhibit superinfection by the other virus at the single-cell level. This hypothesis was put forward by Aquino et al. [23] in Acanthamoeba and could also be considered in the host Vermamoeba through a possible reduction in phagocytosis.

Overall, these findings add to previous evidence of the colossal diversity and ubiquity of giant viruses in the biosphere. They justify further studies to better understand the interactions between Usurpativirus and Faustovirus in the natural environment, as well as between these viruses and their hosts. These studies could include comparisons of viral loads, replication cycle times, viral-host transcriptomes and proteomes under conditions of mono-infection and co-infection and at different stages of their replication cycles. This could help to investigate possible competition or cooperation between the two giant viruses, as well as the stages of attachment and release of the virus by the amoeba.

Author contributions

All authors approved the current version of the manuscript. JA isolated the virus and performed genomic analysis of the virus. JPB and LL performed electron microscopy, LP performed quality control for viral purity with AC. AC produced and purified the virus. JA, JPB, PC and BLS wrote the manuscript.

Funding

This work was supported by the French Government under the “Investments for the Future” program managed by the National Agency for Research (ANR), Méditerranée-Infection 10-IAHU-03), by the “Contrat Plan Etat-Région” and the European funding FEDER IHUPERF.

Data availability

The genome of Usurpativirus is available from the NCBI GenBank database under the accession number PV862220.

Declarations

Ethics approval and consent to participate

Not applicable.

Competing interests

P Colson and B La Scola are scientific advisors for BioSellal, a French biotechnology company (Dardilly, France). Other authors have no conflicts of interest to declare. Funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript.