Novel papillomaviruses in free-ranging Iberian bats: no virus-host co-evolution, no strict host specificity, and hints for recombination

Abstract

Papillomaviruses (PVs) are widespread pathogens. However, the extent of PV infections in bats remains largely unknown. This work represents the first comprehensive study of PVs in Iberian bats. We identified four novel PVs in the mucosa of free-ranging Eptesicus serotinus (EserPV1, EserPV2, and EserPV3) and Rhinolophus ferrumequinum (RferPV1) individuals and analyzed their phylogenetic relationships within the viral family. We further assessed their prevalence in different populations of E. serotinus and its close relative E. isabellinus. Although it is frequent to read that PVs co-evolve with their host, that PVs are highly species-specific, and that PVs do not usually recombine, our results suggest otherwise. First, strict virus-host co-evolution is rejected by the existence of five, distantly related bat PV lineages and by the lack of congruence between bats and bat PVs phylogenies. Second, the ability of EserPV2 and EserPV3 to infect two different bat species (E. serotinus and E. isabellinus) argues against strict host specificity. Finally, the description of a second noncoding region in the RferPV1 genome reinforces the view of an increased susceptibility to recombination in the E2-L2 genomic region. These findings prompt the question of whether the prevailing paradigms regarding PVs evolution should be reconsidered.

Keywords: bats; biodiversity; evolution; papillomavirus; phylogeny; wildlife.

Fig. 1.—

Bayesian amino acid phylogenetic reconstruction for the E1–E2–L2–L1 concatenation. Branch lengths are drawn to scale, with the scale bar indicating the evolutionary distance in substitutions per site. Numbers above the branches indicate Bayesian posterior probabilities and ML bootstrap support values. Maximum support values are indicated with an asterisk (*), while values below 0.50 and 50 are indicated with a dash (-). Color code highlights the four PV crown groups: red, Alpha + OmikronPVs; green, Beta + XiPVs; blue, Delta + ZetaPVs; ochre, Lambda + MuPVs. Viruses whose detailed phylogenetic relationships could not be disentangled are labeled in black. Silhouettes represent the infected hosts. Taxonomic classification of both hosts (host order) and viruses (PV genera) are included. Gray dots highlight the five lineages encompassing bat PVs. Branches corresponding to clades or PVs that contain an E2–L2 region and may thus reflect individual recombination events are highlighted with a black star. The novel bat PVs described here are highlighted with black arrows.

Fig. 2.—

Lambda + MuPVs supernetwork. The network was constructed using the best-known maximum-likelihood trees of each individual nucleotide PV gene (E6, E7, E1, E2, L2, and L1). Color code represents the different orders of the hosts. Specific PVs tropisms and outcome of the corresponding infections are indicated in the inset. PV genera are specified in gray.

Fig. 3.—

Prevalence of EserPV1, EserPV2, EserPV3, and other PVs DNA in the screened samples recovered from seven different Iberian E. serotinus and E. isabellinus colonies.

Fig. 4.—

Tanglegram linking the phylograms of bat PVs and their hosts. Congruence between the phylogenetic relationships among host bat species after Bayesian inference (left) and bat PVs after maximum likelihood inference (right). Please note that the PV taxa depicted here are representatives of highly polyphyletic PV crown groups. Color code for the lines linking both phylograms corresponds to colors used in figure 2 for the different PV crown groups.