The Smacoviridae are a family of circular single-stranded DNA viruses that have been identified in the faecal excrement of many mammals and birds, and a few arthropods [19]. The genomes of smacoviruses range from 2.3 to 3 kb in size and contain at least two bidirectionally transcribed open reading frames (ORFs), which encode a replication-associated protein (Rep) and a capsid protein (CP). These ORFs are separated by two intergenic regions, one of which contains an origin of replication. At present, more than 170 smacoviruses have been documented, with six established genera; Bovismacovirus, Cosmacovirus, Dragsmacovirus, Drosmacovirus, Huchismacovirus and Porprismacovirus. Recently, it was shown that DNA matching a smacovirus was found in the archaeon “Candidatus Methanomassiliicoccus intestinalis”, suggesting that smacoviruses may infect faeces-dwelling archaea [7]. Despite the broad range of animals with which smacoviruses have been associated, none have been documented in a member of the family Felidae. Felids, both domestic and wild, are known to harbour a multitude of viruses, including; feline immunodeficiency virus [13, 20], feline foamy virus [4, 11], feline leukaemia virus [5, 16], feline anellovirus [9, 21], feline gammaherpesvirus [2, 18], feline coronavirus [15], and several more.

In two separate studies, faecal samples were collected from bobcats (Lynx rufus) in California, USA, and African lions (Panthera leo) in the Serengeti National Park in Tanzania. Faecal samples described in this study were collected from three wild cats. One bobcat sample was collected during necropsy, the other from a live trap on the day of capture, and the African lion sample was collected shortly after defecation was observed. Viral DNA was extracted, and circular molecules were enriched from the faecal samples according to a previously described protocol [12]. Enriched DNA was sequenced on an Illumina 2500 platform, and contigs were generated through de novo assembly of paired-end reads using metaSPAdes v. 3.12.0 [1]. Three contigs with similarities to smacoviruses were identified in BLASTx searches and amplified from the three individual samples using abutting primers (Table 1) by PCR. The amplicons were then cloned and sequenced using the Sanger method [12].

Table 1 Sample information and primers used for recovery of full genomes by PCR
Full size table

We have tentatively named the three novel smacoviruses reported in this study “Lynx rufus associated smacovirus 1” (LruSmV 1, MK796234), “Lynx rufus associated smacovirus 2” (LruSmV 2, MK796235) and “Panthera leo associated smacovirus” (PlSmV, MK796236). Their genomes are 2,435 nt, 2595 nt and 2,650 nt, respectively, in size (Table 1). All contain two large open reading frames oriented in opposite directions, encoding the Rep and CP proteins. 174 complete smacovirus genome sequences available in the GenBank database were compiled and used for comparison with the three sequences reported here. Full-genome, Rep and CP amino acid sequence datasets were assembled and used to determine pairwise identity values using SDT 1.2 [14]. Genome-wide pairwise comparison showed that the three genomes share 54-57% nucleotide sequence identity. With all other smacoviruses, the closest relative for each is as follows: LruSmV 1 shares 59% sequence identity with chimpanzee associated porprismacovirus 2 (GQ351273) [3], LruSmV 2 shares 61% sequence identity with Chlorocebus cynosuros associated smacovirus (LC386199), and PlSmV shares 67% sequence identity with sheep faeces associated smacovirus 3 (KT862219) [17] (Supplementary Data 1). Rep and CP amino acid comparison with all smacoviruses showed they share 39-63% sequence identity in Rep and 35-60% sequence identity in CP (Fig. 1 and Supplementary data 1).

Fig. 1
figure 1

Maximum-likelihood phylogenetic tree based on Rep protein sequences (left), displaying the six described genera. The LruSmV 1 Rep is most similar to that of Papio cynocephalus associated smacovirus (LC386204), sharing 38% amino acid sequence identity. An expanded subtree of the Rep-based phylogeny of the porprismacoviruses is displayed on the right. Names in red letters indicate the feline-faeces-associated Reps of the genomes identified in this study. A pairwise amino acid sequence comparison of the two novel feline smacovirus Rep proteins in the genus Porprismacovirus to those of all other members of the genus is shown next to the phylogenetic tree

Full size image

Smacovirus Rep amino acid sequences were aligned using MAFFT [10], and the resulting alignment was used to construct a maximum-likelihood phylogenetic tree using PhyML 3.0 [8] with the best-fit amino acid model rtREV+G+I+F selected using ProtTest [6]. Branches with less than 0.8 aLRT branch support were collapsed. The phylogenetic tree was rooted with the Rep sequences of the nanoviruses. Based on the taxonomic guidelines put forward by the Smacoviridae subcommittee for the International Committee on Taxonomy of Viruses [19], the results of phylogenetic analysis and sequence comparisons support the assignment of each of these three smacoviruses to a distinct species.

Phylogenetic analysis showed that the Rep of LruSmV 1 is divergent, forming a singleton clade that is closest to the huchismacoviruses. The Reps of LruSmV 2 and PlSmV group in a clade consisting of the members of the genus Porprismacovirus (Fig. 1). The Rep proteins of LruSmV 2 and PlSmV share a common ancestor with porcine associated porprismacovirus 10 (KT862225) and sheep faeces associated smacovirus 3 (KT862219), respectively (Fig. 1) [17]. It is worth noting that these two LruSmVs were recovered from the faeces of individual bobcats living in the Los Angeles area (California, USA), but they are genetically distinct from each other. If smacoviruses do in fact infect archaea [7], this may indicate that these two distinct viruses are hosted by different archaeal species. Here, we describe the first smacoviruses associated with felid faeces and show that they are highly diverse, representing three distinct species. Continued research in this area is needed to determine the true hosts of these viruses and to elucidate what relationship they have to other felid viruses and those infecting other animals.