Sarcoptic mange in Felidae: does Sarcoptes scabiei var. felis exist? A first molecular study

Domestic and wild felids are considered suitable hosts for the parasitic mite Sarcoptes scabiei, and sarcoptic mange is reported in several felid species in the scientific literature. However, the historic classification of Sarcoptes mites into host-specific varieties does not include S. scabiei var. felis. It is unclear whether sarcoptic mange transmission in felids involves canids, other sympatric species, or exclusively felids. This study aimed to characterize the genetic structure of S. scabiei mites from domestic cats (Felis catus) and Eurasian lynx (Lynx lynx carpathicus), comparing them with Sarcoptes mites from sympatric domestic and wild carnivores. Ten Sarcoptes microsatellite markers were used to genotype 81 mites obtained from skin scrapings of 36 carnivores: 4 domestic cats, one dog (Canis lupus familiaris), 4 Eurasian lynx, 23 red foxes (Vulpes vulpes), and 4 grey wolves (Canis lupus lupus) from either Italy, Switzerland or France. Two genetic clusters of S. scabiei with a geographical distribution pattern were detected: mites from cats originating from Central Italy clustered with those from sympatric wolves. In contrast, all the other mites from Switzerland, France and Northern Italy clustered together. These results strengthen the previously advanced hypothesis that genetic variants of S. scabiei have a predominant geographic-related distribution with cryptic transmission patterns. These patterns may rely on the interactions between different hosts living in the same ecological niche rather than a simple infection among hosts belonging to the same taxon, reinforcing the idea that the S. scabiei historic classification into “var” might have little ongoing relevance.

Sarcoptic mange is considered a rare disease in domestic cats (Felis catus), more frequently affected by "feline mange", a similar condition caused by the related burrowing mite Notoedres cati [19].However, several confirmed sarcoptic mange cases in domestic cats have appeared in the scientific literature in the last two decades [12,14,15,20,37,38] (Table 1), including an epidemic involving 25 animals [4].The origin of these unusual episodes has been empirically attributed to contact with affected dogs living in the same household [20,38] or, more rarely, with foxes visiting neighbouring gardens [12,20].
Similarly, in wild felids, the source of S. scabiei transmission has been associated with sympatric hosts within carnivore communities (e.g., Eurasian lynx and red foxes (Vulpes vulpes) in continental Europe [5]) or in a prey-to-predator context (e.g., gazelle Eudorcas thomsonii and cheetahs in Eastern Africa [10]).Molecular evidence of the robustness of the latter association has been obtained using molecular markers in the case of lion, cheetah and the respective favourite ruminant preys in Masai Mara, Kenya [10].
Sarcoptes scabiei has been traditionally classified into host-specific varieties.Still, growing molecular evidence shows that the mere taxon-oriented approach is insufficient to embrace the issue's complexity [7].In this regard, various molecular tools have become available to deepen our understanding of the genetic differences between Sarcoptes strains affecting different host species and to track transmission pathways more efficiently and objectively.Amongst these tools, microsatellite markers have been shown to be more informative than other markers in characterizing the populations' "strains" of S. scabiei affecting wildlife in Oceania [43], Europe [5,23,31,36,41], Africa [10], Asia [21], and South and North America [33,42], sometimes revealing unexpected patterns of spread.
In line with the infrequent occurrence of sarcoptic mange in domestic cats, Sarcoptes isolates from feline hosts have never been characterized at a molecular level.This leaves open speculation on the possible transmission pathways and reservoir hosts and possible "strain"-specific diversity regarding pathogenicity.This study aims to investigate the molecular profile of Sarcoptes mites obtained from domestic and wild felids from different European countries and to compare them with Sarcoptes mites from sympatric and allopatric wild carnivores (Felidae and Canidae families), using microsatellite markers.

Material and methods
Skin scrapings from four domestic cats and a dog were collected in Italy, France and Switzerland during a dermatological examination due to severe itch and crusted skin lesions.In contrast, skin samples were also obtained from wildlife with skin lesions compatible with sarcoptic mange and submitted for post-mortem examination (Table 2).
All samples were stored at À20 °C in 70% ethanol tubes until mite isolation and later shipped to the Department of Veterinary Sciences of Turin, Italy.Morphological criteria were applied for the preliminary identification of collected mites [8].For each skin sample, one to six mites were isolated and individually stored in 70% ethanol [3].
DNA was extracted from individual mites following the HotSHOT Plus ThermalSHOCK technique [2].Then, a 10x multiplex PCR was performed using ten validated primers extracted from the previously published panel [44] to target S. scabiei mites ( Sarms 33,34,35,36,37,38,40,41,44,45) following the PCR protocol of Soglia et al. [39].Capillary electrophoresis was performed with an Applied Biosystems SeqStudio TM .The software GeneMapper 4.0 (Applied Biosystems, Foster City, CA, USA) allowed the allele calls and microsatellite visualization.After molecular analysis, mites that did not fulfil the required criteria (eight detectable loci out of the ten analyzed) were excluded from the genetic analysis.Two population genetics analyses were applied to the 81 mite microsatellite outputs: i) Bayesian clustering and ii) principal component analysis (PCA).The first one requires Hardy-Weinberg equilibrium (HWE), while no assumptions are required for the PCA.Descriptive statistics, such as observed and expected heterozygosis (Ho and He, respectively), allelic richness (R) and HWE analysis, were carried out with software R 4.0 using the packages Adegenet 2.1.3[16].P-values for the HWE test were based on Monte Carlo permutations of alleles.The Bayesian assignment test was computed with the software STRUCTURE 2.3.4.Burn-in and run lengths of Markov chains were 10,000 and 100,000, respectively, and five independent runs for each K (for K = 1-20) were run.The ancestry model selected was the admixture type.Clusters were estimated as suggested by Evanno [6], using the DK method.
The Bayesian assignment test revealed the presence of two geographically separated clusters (Fig. 1).One cluster included mites from the cats and wolves from Central Italy (green cluster), and the other from foxes, lynx, wolves, dog and cats from France, Switzerland and Northern Italy (red cluster).
The results of the PCA are displayed in Figure 2. The multivariate analysis revealed two main clusters, mainly scattered along the first axis: (i) mites collected on cats and wolves from Central Italy; and (ii) mites collected on cats, dog, wolves, foxes and lynx from France, Switzerland and Northern Italy (see also Table 2).

Discussion
The main finding of this study is the identification, using microsatellite markers, of two genetic clusters of European S. scabiei with a geographical distribution pattern.Cat-derived mites from Central Italy clustered with those obtained from sympatric wolves.By contrast, all the other mites from Switzerland, France and Northern Italy clustered together.
These results align with previous evidence, though referring to different animal models, that genetic lineages of S. scabiei have a geographic-related distribution adding to the expected host-related distribution [23].Interspecific transmission within broad taxa (e.g., carnivores) might thus rely on direct or indirect interactions between different hosts sharing the same ecological niche [21,23,31].This finding differs from the classical idea that S. scabiei exclusively embraces host-specific varieties.It instead supports the view that various mammal species sharing the same environment have opportunities for direct or indirect contact (through predation, scavenging, territorial fights, denning, mating, etc.), which may allow the sharing of pathogens [7,32].
While numerous species-related varieties of S. scabiei have been traditionally associated with various hosts, a felid-specific variety (e.g., a putative S. scabiei var."felis") has never been described.The explanation is possibly the infrequent occurrence of sarcoptic mange in domestic cats, in which notoedric mange, caused by Notoedres cati, is predominant.Nonetheless, in the last few decades, an increasing number of sarcoptic mange cases in both domestic and wild felids have been reported [10,12,15,20,35].These cases referred to different   epidemiological scenarios in which infected dogs, foxes or wild felids' natural prey were pinpointed as the likely source of infection (Table 1).Recently, nuclear molecular markers such as microsatellites have been proven valuable for investigating population genetic differences and putative transmission pathways within S. scabiei [21,23,31,33].To the authors' knowledge, this is the first study to apply such tools to domestic/wild felid-derived mites, indicating, on a molecular basis, what is historically reported, namely that a var.felis of S. scabiei may not exist.This conclusion needs confirmation by further molecular analyses based, in parallel, on microsatellite markers and innovative genomic techniques [9,18,22,46].A larger dataset of mites from diversified geographical areas and carnivore communities is similarly recommended.
Interestingly, Bornstein et al., 2004 performed a genetic characterization (although not specifying the molecular markers employed) of Sarcoptes mites from six out of 25 infected cats involved in a single outbreak in Sweden.These authors noted that "the mites had DNA sequences identical to S. scabiei from naturally infected dogs and Swedish wildlife".These results agree with our findings (Figs. 1 and 2) and confirm that S. scabiei taxonomy cannot be simplified in clear-cut host-specific varieties or subspecies, as already outlined [7,41].Empirical information in previous studies (Table 1) suggests that dogs and foxes play a key role in the transmission of S. scabiei to domestic and wild felids in various scenarios, from urban to remote natural areas.
Unlike what was observed in two large wild felids in Eastern Africa (Table 1), no prey-to-predator pattern of S. scabiei transmission was identified in the present study.For example, the lynx mites came from the Western Alps (Table 2), where sarcoptic mange has not been reported in the main lynx prey, the roe deer (Capreolus capreolus), nor in the Northern chamois (Rupicapra rupicapra).In these areas, the disease has been detected only in smaller and rarely preyed animals, such as the red fox [25,30] and mustelids [1].
Despite the limited sample size and the use of a single (although particularly informative) class of genetic markers, this study may be considered baseline data in the unexplored field of Sarcoptes epidemiology in felid hosts.Our results, far from suggesting that domestic cats and wolves from Central Italy infected each other by direct contact (which can be considered a rare, though possible event), show that the same S. scabiei strain circulates in different carnivore hosts living in the same geographical area.It seems reasonable to assume that infected dogs or red foxes represent the missing link between wolves and cats in Central Italy (see Figs. 1 and 2).Interestingly, the owners of these cats reported that they had free outdoor access and that foxes were often seen roaming near the house.
While sarcoptic mange may not always represent a threat to domestic animals, numerous cases resulted in zoonotic transmission to the owners or members of the family in contact with the infected cat [24].Moreover, infection by Sarcoptes mites circulating in European carnivore communities can put additional pressure on species or populations of conservation concern, such as the Iberian lynx and the European wild cat (not included in this study).In these wild felids and other European wild felids not included in this study, the recognition of canids (and most likely the red fox) as the expected source of infection may have practical consequences in the planning of delicate conservation and management interventions, such as restocking and reintroductions.
In conclusion, our results suggest that domestic and wild felids in Italy and neighbouring countries are affected by the same Sarcoptes strains as those involving sympatric canids.A specific geographical pattern may reveal the transmission pathways of Sarcoptes mites in the investigated carnivore hosts rather than a "host species-related only" pattern.Thus, the continued use of the term "var" in the past scientific literature referring to Sarcoptes host-specific subspecies or "strains" may be outdated or even misleading, appearing of little relevance.Accordingly, we believe that the contemporary understanding of the broad and sometimes unexpected host associations of S. scabiei should cautiously lead to the dismission of the historic nomenclature based on questionable mite morphological differences and the exclusive view of strict host association [46].
Table S1: Results of the Hardy-Weinberg equilibrium test at locus (rows) and populations (columns) showing P values from the Monte Carlo test.P values less than 0.05 (*) and 0.01 (**) are considered significant.

Figure 1 .
Figure 1.Barplot of Sarcoptes-derived genetic clusters generated with the software Structure 2.3.4 with maximum likelihood K = 2.Each mite is represented by a single bar, and the height of each coloured segment is proportional to the membership fraction in each cluster.Felids are marked with a dotted circle.

Figure 2 .
Figure 2. Principal components analysis (PCA) of microsatellite loci representing cat-, dog-, lynx-, wolf-and fox-derived mite populations in France, Switzerland, North Italy and Central Italy.Each population is labelled with the host species and the geographical origin.The eigenvalues of the two axes are displayed in the barplot on the left.Components PCA 1 and PCA 2 explained 11.3% and 7.9% of the variance, respectively (black bars of the eigenvalues, inset).

Table 1 .
Cases of sarcoptic mange previously reported in felids.
NA: information not available; *Multiple cases reported in the same article.

Table 2 .
Origin and sample size of the animals affected by sarcoptic mange included in this study.
N: number of sampled animals; n, number of mites used for microsatellite analysis.

Table 3 .
Descriptive statistics of the Sarcoptes populations arranged by Sarms locus.