A total of eight samples were collected from the public sewers of Jeddah: four before and four after treatment. Nine samples were camel stools. Samples were prepared as described previously (La Scola et al., 2001). The following amoebae were inoculated and incubated for 2 days under the conditions previously described (La Scola et al., 2001): A. polyphaga (strain Linc AP-1), Acanthamoeba griffini (strain ATCC50702), Acanthamoeba sp. (strain E4), Vermamoeba vermiformis (strain CDC19), and Dictyostelium discoideum (Strain ATCC44841). The following mixture of antibiotics was used to avoid bacterial overgrowth: Vancomycin (10 μg/ml), Imipenem (10 μg/ml), Doxycycline (10 μg/ml), Ciprofloxacin (20 μg/ml), and Thiabendazole (30 μg/ml). Sub-cultures were then performed every 48 h using the same conditions described above on fresh amoebae without any antibiotics. These were screened daily by inverted microscope under 400X magnification for cytopathogenic effects (La Scola et al., 2001). The size and morphology of the viral particles in wells showing amoeba lysis were determined by electron microscopy using negative staining (La Scola et al., 2010). SDMV was produced within the viral factory located in the A. polyphaga cytoplasm and stained with 4,6-diamidino-2-phenylindole (DAPI) and hemacolor. In order to work on a clonal isolate, culture supernatant containing viral particles was serially diluted from 10^-1 to 10^-11. Three 24-well microplates were prepared by adding 500 μl of suspension containing 5 × 10⁴ rinsed amoebae and four wells were inoculated with each viral dilution. Until the lysis of amoebae was almost complete at 32°C, the microplates were observed daily under an inverted microscope. The well-containing the highest dilution, which caused complete lysis, was used for virus production through consecutive sub-culturing (La Scola et al., 2010). The automated EZ1 Virus MiniKit was used to extract viral DNA. Five real-time PCR systems based on hydrolysis probes for detecting the viral groups (mimivirus lineages A, B, and C; marseilleviruses; virophages) were applied and conducted as previously described (Ngounga et al., 2013).

The SDMV genome was sequenced using the Illumina MiSeq instrument (Illumina, Inc., San Diego, CA, USA) with the paired-end strategy in line with the manufacturer’s recommendations for the Nextera XT library kit (Illumina). To do so, the extracted genome of SDMV was quantified using a Qubit assay with a high sensitivity kit (Life Technologies, Carlsbad, CA, USA) to 48.8 ng/μl, and the dilution was performed, requiring 1 ng of DNA as input. The ‘tagmentation’ step fragmented the genomic DNA. Limited cycle PCR amplification then completed the tag adapters and introduced dual-index barcodes. After purification on AMPure beads (Life Technologies, Carlsbad, CA, USA), the libraries were then normalized on specific beads according to the Nextera XT protocol (Illumina). Normalized libraries were pooled into a single library for sequencing on the MiSeq. The pooled single strand library was loaded onto the reagent cartridge and then onto the instrument along with the flow cell. Automated cluster generation and paired-end sequencing with dual index reads was performed in a single 39-h run in a 2-bp × 251-bp.

The DNA polymerase gene was used for phylogenetic analysis. Amino-acid sequences were aligned using the MUSCLE software (Edgar, 2004). A phylogenetic tree was built using the MEGA 6.0 tool (Tamura et al., 2013) and the maximum likelihood method. In addition, we performed a hierarchical-clustering based on the gene presence/absence pattern of 5,443 NCVOGs, using the MeV tool (Eisen et al., 1998) with Pearson correlation as distance metric. The phylogenetic tree was visualized using the FigTree v1.4.1 tool¹.

Six hours after infection in A. polyphaga and in order to detect the transcript of putative ORFan (the ORF_850), RNA was extracted using RNeasy Mini kit (Qiagen, Hilden, Germany) after the pellet had been centrifuged at 9,000 × g for 15 min. DNA contamination was then checked by applying PCR using the HotStar TaqDNA polymerase kit (Qiagen). DNA contamination was removed from the RNA preparation using a TURBO DNA-Free kit (Ambion, Austin, TX, USA). The extracted volume was treated using RNeasy Mini Elute clean up Kit (Qiagen) and SuperScript VILO cDNA Synthesis Kit (Invitrogen, Carlsbad, CA, USA). The following primers were used to amplify the cDNA template by PCR with the Hot Star Taq DNA Polymarase (Qiagen): SV1 R 5′-TCTGAAACGTTATGTTCCGCA-3′, SV1 F 5′-TGCATTTTCCTTGGCACAAA-3′, SV2 R 5′-TCCGGATATCTTGGGCCATT-3′, SV2 F 5′-AGATGATGAATGTGTTCCTCCA-3′.

In this study, we isolated a new virus, known as SDMV by inoculating an A. polyphaga co-culture, from a sewage water sample taken before treatment (by filtration) at the King Abdulaziz University Hospital. It could also be grown in A. griffini, although not with Acanthamoeba sp. E4, V. vermiformis and D. discoideum. Presumptive morphologic observations showed a virus with a size of 500 nm, typical of the family of APMV (Figure 1C). The viral factory of SDMV was observed in the infected A. polyphaga cytoplasm with hemacolor and DAPI staining, respectively (Figures 1A,B). Some morphology features and virus factories appeared within the cytoplasm of A. polyphaga during the life cycle of the SDMV, similar to the other Mimiviridae members, and the morphology of SDMV factories (Figures 1A,B) were also similar to those previously observed for APMV, Megavirus chiliensis and A. polyphaga moumouvirus (Suzan-Monti et al., 2007; Arslan et al., 2011; Yoosuf et al., 2012). The size of the icosahedral capsid of SDMV with around 500 nm surrounded by a dense fiber layer, is comparable with members of family Mimiviridae (Figure 1C), such as APMV, moumouvirus and Megavirus chiliensis (500, 520, and 500 nm, respectively) (Klose et al., 2010; La Scola et al., 2010; Arslan et al., 2011; Yoosuf et al., 2012). In addition, the SDMV particles showed identifying, starfish-like vertex that resemble the other Mimiviridae members (Klose et al., 2010; Arslan et al., 2011).

Real-time PCR indicated that the SDMV belongs to mimiviruses lineage B.

The SDMV genome assembly yielded two large contigs, with 174.294 and 855.762 bp KY110734). Its genome showed a low C+G% content (25.83%), which was similar to other mimiviruses (La Scola et al., 2003; Raoult et al., 2007; Assis et al., 2013). The SDMV genome encoded two tRNA molecules, tRNA-Cys (gca) and tRNA-His (gtg), as predicted in genomes of other mimiviruses from lineage B (Colson et al., 2013).

The SDMV genome was predicted to encode a total of 953 ORFs, ranging in size from 34 to 2,914 amino acids (aa), with a mean size of 324 aa. Of them, 868 ORFs were functionally characterized by similarity analyses with previously reported sequences, ranging in size from 54 to 2914 aa (mean size: 349). Moreover, 40 ORFs were predicted without similarity with previously reported sequences, characterizing it as a putative ORFan. The transcription of ORFan L850 was confirmed by qPCR (Figure 2). In addition, 42 sequences had less than 100 aa without any predicted function (Pseudo-ORFs). The functional annotation of the SDMV genome revealed a great number of putative proteins related to several metabolic processes, nucleic acid binding and manipulation, as well as signal transduction processes (Figure 3). Curiously, we observed an unusual set of aminoacyl-tRNA synthetase (aaRS), which are translation-related genes not found in viruses other than mimiviruses. This new genome was predicted to encode nine aaRS molecules, including Methionyl, Isoleucyl, Asparaginyl (two sequences), Arginyl (two sequences), Cysteinyl (two sequences), and Tyrosyl-tRNA synthetase. In addition, around 40% (341/868) of sequences predicted in the SDMV genome were hypothetical proteins. Finally, the SDMV genome was predicted to encode 155 paralogous proteins (Coverage > 70%, identity > 30%, e-value < 1e-6), divided into 41 families, of which we highlight an ankyrin-like protein and an f-box/fnip (repeat containing) protein families, comprised of 32 and 20 sequences, respectively.

For the ORFs predicted in the SDMV genome, the best hits were most frequently proteins from mimivirus lineage B (97.7%). Moreover, a comparative analysis of the SDMV gene content against all mimivirus lineages revealed a higher mean identity with and bit-score distributions within mimivirus lineage B (Supplementary Figure S1).

The amplified cDNA template, using real-time PCR with the Hot Star Taq DNA Polymarase, was positive (Figure 2), with the specific primers that indicated the presence of the ORFan gene (ORF_850), which was exclusive to the SDMV genome and has not been found in any other known organism, including other mimiviruses. The search and detection of such ORFan was performed as proof of concept of the presence of previously unreported sequences.

β-DNA polymerase-based phylogenetic analysis corroborated all previous observations, clustering the SDMV isolate with members of family Mimiviridae lineage B, which includes A. polyphaga moumouvirus, Moumouvirus Monve and Moumouvirus Goulette strains (Figure 4). In addition, a hierarchical clustering tree (Figure 5), based on phyletic patterns, was constructed using a presence-absence matrix of 5,443 NCVOG (clusters of orthologous genes shared by NCLDVs). This phylogenetic tree was generated using the MeV 4.8.1 tool (Howe et al., 2010) and showed SDMV to be in the same branch of Moumouvirus (lineage B). These analyses corroborate all previous results suggesting the clustering of SMDV into mimivirus lineage B.