4. Discussion
To our knowledge, this study is the first to use the sensitive ddPCR
assay, a biotechnological refinement of conventional PCR, to detect and
quantify highly pathogenic bovine δPVs in clinically healthy sheep by
liquid biopsy. Our previous studies have shown that ddPCR has both high
specificity and sensitivity for the detection and quantification of BPV
DNA in healthy and diseased cattle (De Falco et al., 2020) as well as in
healthy goats (Cutarelli et al., submitted). The current study provides
further evidence that ddPCR is a very useful approach to detect and
quantify BPV in the blood of healthy sheep and allows us to gain
diagnostic and epidemiological insights into BPV presence in ovine
species as data on the prevalence and types of BPVs in sheep are not
currently available.
The liquid biopsy approach to the detection of circulating BPV DNA has
garnered growing interest in PV studies (Veyer et al., 2020). Indeed, PV
detection in the blood can be used as diagnostic, prognostic, and
epidemiological markers (Bishop et al., 2016).
Our study showed that BPV-2 is the most prevalent BPV genotype in
healthy sheep, similar to other ruminants, such as cattle and goats. The
highest number of copies of this genotype was found in Sar from sheep
flocks in which cross-infections by BPVs have been previously reported
(Roperto et al., 2018a), which suggests that copy numbers may correlate
with the risk of cutaneous and mucosal lesions that progress to cancer.
Furthermore, this study reports the first detection of BPV-14 in sheep.
This genotype, chronologically the last BPV type identified in cattle,
has never been described in the ovine species. Furthermore, our results
demonstrated a statistically significant prevalence of BPV-14 in Apu
compared to Sar, Cal, Bas, and Cam, which clearly showed that BPV
genotype prevalence has a territorial divergence in these regions.
We compared the sensitivity of ddPCR with that of RT-qPCR in evaluating
the same liquid biopsy, demonstrating that ddPCR has superior
sensitivity compared to RT-qPCR. Therefore, our results suggest that
ddPCR is by far the most sensitive and accurate assay for BPV detection.
It is worth noting that it has been shown that ddPCR outperforms RT-qPCR
in terms of the sensitivity, specificity, and reproducibility of
oncogenic human papillomavirus detection and quantification (Biron et
al., 2016; Jeannot et al., 2016; Carow et al., 2017).
BPV genotype detection in the blood of healthy sheep suggests that the
bloodstream can be the primary site of BPV infection. As bovine δPVs are
known to infect ovine species and result in anatomoclinical diseases, it
is conceivable that these viruses may spread through the blood, which
could be responsible for secondary tissue localization and infection.
However, as in humans (IARC, 2007), further epidemiological studies are
required to enhance the understanding of BPV transmission via the
bloodstream.
Here, we detected a higher percentage of BPVs in sheep that were in
close contact with cattle herds, about which numerous case reports of
BPV infection have been described. Indeed, the sheep flocks in our study
shared bracken fern-infested lands with pasture-residing cattle for
grazing. The evidence from epidemiological studies of cattle is
sufficiently strong to suggest that in the presence of BPV infection,
the toxic components of bracken ferns such as ptaquiloside (PT), a
water-soluble norsesquiterpenoid glycoside, are ecological co-factors in
the development of severe diseases due to BPVs, including chronic
enzootic hematuria (CEH), a clinical syndrome caused by bladder tumors
(Smith, 1990). Thus, it is conceivable that PT may also be a co-factor
of diseases in sheep. Indeed, PT has recently been detected in
biological matrices from healthy sheep (Virgilio et al., 2015). PT is
known to hamper the immune system and may play an important role in
cross-species transmission and infection of bovine δPVs. It is worth
noting that outbreaks of CEH have also been reported in sheep (Prakash
et al., 1996; Cianciolo and Mohr, 2016). Furthermore, bovine δPV
infection resulting in clinical disease are known to occur in sheep
(Mazzucchelli-de-Souza et al., 2018; Roperto et al., 2018a; Savini et
al., 2020). Therefore, the detection of bovine δPV DNA in the blood of
sheep means that sheep can be infected by these PVs, which may make δPVs
an additional, potential cause of ovine disease. Furthermore, our
results suggest that clinically healthy sheep may represent a reservoir
for bovine δPVs. Thus, it is conceivable that sheep may play a role in
intra- and interspecies bovine δPV transmission and infection. In this
context, very precise quantitation of very low viral copy numbers can
provide more precise monitoring of latent BPV DNA reservoirs.
Finally, it is well known that PV distribution varies considerably by
geography (Chen et al., 2019). Therefore, ddPCR may be an essential tool
for improving diagnostic procedures thus allowing the identification of
the genotypic distribution of BPV and a better understanding of the
possible, geographical divergence of BPV prevalence in different areas.
The ddPCR assay appears to possess high sensitivity and accuracy, which
is valuable for addressing the molecular burden of BPV infections and
useful for defining an accurate ecological epidemiology. This baseline
information improves our knowledge about the molecular mechanisms of the
disease and provides insights into necessary measures for reducing the
risk of BPV infection and/or co-infection.
In conclusion, ddPCR is presently being used to detect very low nucleic
acid concentrations and, therefore, appears to be of interest in the
diagnosis of infectious diseases, including viral diseases (Li et al.,
2018). DdPCR has proven to be a valuable new technology and with
additional improvements in prospect it is likely to become an
indispensable tool in diagnostic, prognostic and epidemiological virus
research (Rutsaert et al. 2018). Therefore, the ddPCR method may provide
a new and promising tool for evaluating the BPV viral load in clinical
samples. Future PV research warrants the use of this molecular approach
to assess PV type-specific pathogenetic pathways of disease, including
carcinogenicity. Indeed, available evidence from BPV distribution lends
strong support to the notion that the risk of an animal developing a
BPV-associated disease varies substantially according to the specific
BPV type with which the animal is infected. Finally, the ddPCR approach
may provide a better understanding of the complex interactions between
multiple BPV types during coinfections, as the possible interference
resulting from multiple PV genotypes in coinfection cases remains an
open question (Biryukov and Meyers, 2018).