Co-infection of distinct papillomavirus types in a captive North American porcupine

The family Papillomaviridae contains a large collection of small, double-stranded DNA viruses that infect a diverse range of mammals, reptilians, birds and fish [1]. All papillomavirus genomes are circular, ranging from 5.7 to almost 9 kb in size, and can be divided into an early and a late region based on the temporal expression pattern of the genes they encode [1, 2]. In humans, the diversity of papillomaviruses is well studied, with hundreds of known types that can be further classified based on their pathogenic potential [3, 4]. For most animal papillomaviruses, conversely, only limited information is available, especially for those infecting non-domesticated animals. Nonetheless, the number of known animal papillomavirus types and hosts is growing continuously [5]. Most of these newly discovered viruses cluster together with other papillomaviruses found in the same host species, in line with what is known about the shaping of the papillomavirus family tree through a long history of co-evolution of virus and host [6]. However, in several cases, divergent viruses from separate clades have been identified in the same host species, hinting at the occurrence of ancient host-jumping events [7, 8]. An example of this are Erethizon dorsatum papillomavirus 1 and 2 (EdPV-1/-2), which were both discovered in North American porcupines [9‐11]. While EdPV-2 clusters directly adjacent to the rodent-borne papillomaviruses of the genus Pipapillomavirus, albeit potentially as a separate genus, EdPV-1 is the sole member of the distantly related genus Sigmapapillomavirus, clustering next to the nupapillomavirus human papillomavirus 41. Despite their divergent evolutionary origin, both EdPV-1 and -2 were reported to cause wart-like papillomatous lesions.

Papillomavirus infections occur primarily in the epithelium and often resolve autonomously after a subclinical phase, although in some cases papillomavirus infections persist and evolve into warts or benign or malignant precancerous lesions [12]. The clinical severity of papillomavirus infections is determined both by the presence of environmental co-factors and the type of papillomavirus causing the infection [13]. Lesions can sometimes also be characterized by the presence of multiple virus types, although the clinical impact of such co-infections remains poorly understood. Papillomavirus co-infections have been reported in several host species, especially in the last decade, but an incomplete understanding of papillomavirus diversity and an associated lack of appropriate diagnostic tools likely results in their true prevalence being significantly underestimated [14‐20]. Here, we report the first case of a papillomavirus co-infection in a North American porcupine. Using nanopore sequencing, two complete genomes could be retrieved from the same lesion, indicating the presence of both EdPV-1 and -2.

This report represents only the third description of a papillomavirus infection in a North American porcupine. While the two previous cases, reported in 2005 and 2018, were caused by two distinct viruses belonging to different papillomavirus genera, the case presented here appears to have been a mixed infection, caused by variants of the two previously reported Erethizon dorsatum papillomaviruses [9‐11]. However, while both EdPV-1 and EdPV-2 were detected, it is unclear to which extent both viruses contributed to the clinical presentation of the papillomatous lesions. It is possible that co-infection with two different viruses had a cumulative effect, aggravating the progression of the disease. Conversely, it might also be that only one of the two viruses played a causal role in the clinical presentation, and that the detection of the second virus was no more than an accidental observation. Morphologically, the lesions seen here resembled those previously observed for EdPV-2 [10]. For EdPV-1, a comparison is more difficult because there are no images available of the only known EdPV-1 case. However, based on the provided description of the lesions as “multiple white to light brown lobulated, raised, firm masses on the foot pads and facial skin, ranging in size from 2 to 10 mm” it is possible that this case actually resembled the early stage of the infection described here [11]. In the case described here, the lesions regressed spontaneously without any therapeutic intervention and disappeared completely after 6–7 months. Because no serial sampling was performed, it is unclear at which stage of the infection the viral presence of either strain started to reduce and if this decline occurred at similar rates for both viruses. Although unfortunately not performed for the case described here, such serial sampling through non-invasive skin or mucosal swabs might help monitor the evolution of viral persistence in future comparable cases, providing new insights into the kinetics of animal papillomavirus infections. Likewise, the swabbing of seemingly uninfected co-housed animals could be performed to screen for subclinical infections that might otherwise be overlooked.

Co-infections by different papillomavirus types have been described previously for different animal species, in some cases by more than two types at the same time [14‐20]. In 2016, Daudt and colleagues even reported the finding of seven distinct bovine papillomavirus types in the same lesion [19]. In their report, they note that co-infections are likely more prevalent than currently known, partly because PCR-based methods might fail to detect certain types. Using several commonly used PCR screening assays, we also failed to detect papillomavirus DNA, even though we had previously successfully used two of these assays for the detection of EdPV-2 [9]. However, despite belonging to the same type, there are notable differences between this EdPV-2 variant and the previously described one, including several mutations in the primer binding sites of the aforementioned assays. Using more specific primers, we were able to successfully amplify both EdPV-1 and -2 genome fragments, but, while unlikely, it cannot be ruled out that other additional papillomavirus types might have been overlooked.

The EdPV-2 genome presented here differs significantly from the previously reported sequence. Given the low evolutionary rate of papillomaviruses, this finding clearly indicates the circulation of distinct EdPV-2 lineages [6]. Further research is needed to determine whether these lineages can be found ubiquitously, or if they are confined to specific geographic regions or Erethizon dorsatum subspecies. Interestingly, both genomes are marked by the presence of an exceptionally large secondary non-coding region between the early and late regions. Even though this region appears mostly void of known regulatory motives, its conservation in distinct lineages nonetheless hints at a functional relevance. Unlike the EdPV-2 genome, the EdPV-1 genome presented here differs by only 0.1% from the only other known EdPV-1 sequence, despite having been found more than fifteen years apart and in animals from different continents [11]. However, because both EdPV-1-infected porcupines were kept in captivity, it is plausible that there was direct contact between the ancestors of these animals or those of their cage mates in the not too distant past. Further research will be necessary to elucidate if also for EdPV-1 there exist distinct lineages and whether there are other porcupine-specific papillomavirus lineages or types that remain to be discovered.