|
nif:isString
|
-
This study was performed within the distribution area of Pyrenean chamois and Cantabrian chamois in the two main mountain ranges from Northern Spain, the Pyrenees and the Cantabrian Mountains, respectively. Within these ranges, seven different geographical units were considered for the study, six from the Eastern and Central Pyrenees (Catalonia, NE Spain): Freser-Setcases National Game Reserve (PyFS), Cadí National Game Reserve (PyC), Cerdanya-Alt Urgell National Game Reserve (PyCAU), Boumort National Game Reserve (PyB), Alt Pallars National Game Reserve (PyAP) and Vall Aran Game Reserve (PyVA); and one in the Eastern Cantabrian Mountains (León, N Spain), the Natural Protected Area of Picos de Europa (CmPE) (Fig 1). In each study area, ungulate populations are managed independently along with hunting plans.
Figure data removed from full text. Figure identifier and caption: 10.1371/journal.pone.0186069.g001 Maps of the study areas and wild ruminant species composition in the Pyrenees.Location of the study areas in Cantabrian Mountains (Picos de Europa—CmPE) and Eastern and Central Spanish Pyrenees. Wild ruminant species composition by study area from the Pyrenees is showed in detail in the image of the bottom: Vall Aran (PyVA), NGR Alt Pallars (PyAP), NGR Boumort (PyB), NGR Cerdanya-Alt Urgell (PyCAU), NGR Cadí (PyC) and NGR Freser-Setcases (PyFS). The asterisk means the same study area. Chamois in PyB is scarce and not representative of the area (not shown in the map).
These areas are high mountain habitats mostly composed of alpine or subalpine ecosystems with strong seasonal influence, with the exception of PyB which has a dryer climate with a higher Mediterranean and continental influence. Altitude ranges approximately from 800 in the bottom of the valleys to 3100 meters high in the Pyrenees and from 1100 meters to 2600 meters high in the Cantabrian Mountains. Chamois is the most abundant ungulate species in most of the study areas except in PyB, where red deer (Cervus elaphus) is the predominant wild ruminant (Fig 1). Chamois population size is estimated yearly with linear transects performed by the rangers and is calculated to be about 7,500 in the six study areas from the Pyrenees and 3,800 in the study area from the Cantabrian Mountains. The chamois density varies among study areas as a result of differential pestivirus die-offs in the Pyrenees [32,33]. Mean minimum chamois abundances per square Km during the study period, calculated as estimated chamois population/area, were: PyFS, 14.7; PyC, 3.3; PyCAU, 2.7; PyAP, 1.8; PyVA, 1.8; CmPE 4.0. Other wild ungulates that cohabit with chamois include roe deer (Capreolus capreolus), red deer, mouflon, fallow deer (Dama dama) and wild boar (Sus scrofa), but with a different ruminant community composition among study areas from the Pyrenees, shown in Fig 1. Wild boars are present in all study areas and fallow deer and mouflon are not present in Cantabrian Mountains (CmPE). Domestic ruminants (i.e. cattle, sheep, and goats) and domestic horses also share these habitats with the wild species during the grazing period (May-November) in all of the study areas.
A long-term cross-sectional sampling design was performed on alpine wild ungulates hunted during the regular hunting seasons from 2009 to 2015 in the Pyrenees (n = 1556) and from 2010 to 2013 in the Cantabrian Mountains (n = 132). Samples were collected from recently hunted ungulates in proportion to the total number of animals hunted in each study area (Table 1). Four sheep flocks from the Pyrenees that graze in the alpine meadows of three of the study areas (PyVA, PyAP and PyFS) were also sampled in 2014 (around 30 sheep per flock; Table 1).
Table data removed from full text. Table identifier and caption: 10.1371/journal.pone.0186069.t001 Distribution of animals sampled by study areas and species. *Two sheep flocks.
Samples were taken between the third eyelid and the palpebral conjunctiva with sterile cotton swabs without medium from each eye separately and frozen at -20°C within 24 hours from collection. Basic information of the individuals was also registered, including ocular signs, sex, age based on the annual horn segments for chamois and mouflon [34], date and location. Geographic coordinates were also recorded in PyFS from 2012 to 2015. Age was classified in four categories in chamois according to social behavior and aging process, kids (<1 year), yearling (1–2 years), juvenile (2–3 years) and adults (>3 years). This study accomplish with current guidelines for ethical use of animals in research following the European (2010/63/EU) and Spanish (R.D. 53/2013) legislations. The approval of an ethic committee was not needed since management and sacrifice of animals were not performed for research purposes. Ungulate wild species studied are not endangered, and its abundant populations are managed along hunting plans, regulated by the competent public administrations. Samples were obtained by the rangers from hunted-harvested wild animals during the regular hunting plans from National Game Reserves and Hunting Reserves that belong to public administrations. Both samplings of wild animals and domestic livestock were performed in the frame of health surveillance programs approved by the Departament d′Agricultura, Ramaderia, Pesca, Alimentació i Medi Natural—Generalitat de Catalunya (DARPAMN, the Regional authority in charge of livestock and wildlife management).
Eye swabs were thawed, cut and mixed during one minute with 0.5 mL lysis buffer (100 mM Tris–HCl, pH 8.5, 0.05% Tween 20, 0.24 mg/mL proteinase K) in sterile tubes. The lysates of the cells were obtained by incubating the tubes at 60°C for 60 minutes. Proteinase K was then inactivated at 97°C for 15 minutes [35]. The resulting lysates were directly used as test samples for the molecular detection of M. conjunctivae. The presence of M. conjunctivae DNA in the samples was assessed with a previously described TaqMan real time PCR (qPCR) using primers LPPS-TM-L, LPPS-TM-R, and probe LPPS-TM-FT [35]. Briefly, 2.5 μL of the sample lysates, 900nM of each forward and reverse primer, 300 nM of the probe, 12.5 μL TaqMan®2x Universal PCR MasterMix (Applied Biosystems, Warrington, UK) and an exogenous internal positive control (IPC; Applied Biosystems, Warrington, UK) were introduced in each reaction well and nuclease-free water up to a total volume of 25 μL. Cycling conditions were set for 40 cycles at 95°C for 15 s and 60°C for one min, with pre-cycling steps of 50°C for 2 min and 95°C for 10 min. The threshold cycle (Ct) of each sample was defined as the number of cycle at which the fluorescent signal of the reaction crossed the threshold that was set to 0.05. Samples were analyzed per duplicate and were considered valid only if difference between the replicates was less than one Ct. Samples with Ct≤38 were interpreted as qPCR-positive. All PCR reactions were run on Applied Biosystems® 7500 Fast Real-time PCR system (Applied Biosystems, Warrington, UK).
Mycoplasma conjunctivae subtyping and cluster analyses: The lppS gene of M. conjunctivae encodes for a membrane lipoprotein that is involved in adhesion [36], which variable domain can be used for M. conjunctivae subtyping and to perform molecular epidemiology analyses [28]. For cluster analyses, samples from this study with Ct values lower than 33 at the qPCR-M. conjunctivae were considered for sequencing. Available qPCR- M. conjunctivae positive samples from a previous study that analyzed 439 sheep and goats (19 flocks) from the same study areas and period were also included [22]. To obtain lppS gene sequences of these samples, a nested PCR was first performed as described with minor modifications of the primers [28,31] (S1 Table). PCR products were then purified with the High Pure PCR Product Purification Kit (Roche Diagnostics, Rotkreuz, Switzerland). The sequences were determined with the sequencing primers Ser_start2, Ser_start0 and Ser_end0 (S1 Table) using the BigDye termination cycle sequencing kit (Applied Biosystems, Forster City, CA, USA). The resulting sequences were trimmed to contain the region that comprises the nucleotide positions 3935–5035 of the lppS gene from M. conjunctivae type strain HRC/581 (GenBank acc. number AJ318939), which corresponds to the variable lppS domain and flanking regions. Alignment and editing of the sequences were performed with the BioEdit software. A phylogenetic analysis of the sequences was then performed by the generation of cluster analyses trees built by the UPGMA statistical method and performing 1000 bootstrap replications [37]. The generation of the phylogenetic tree was performed using MEGA software [38]. For the tree construction, sequences of M. conjunctivae strains described in previous works from the same study areas were included for comparison (three chamois from PyAP and one mouflon from PyFS) [16], covering a temporal period from 2006 to 2015 (S2 Table). Sequences from other areas (n = 6) and the sequence of the type strain HRC/581 were also included in the tree (S2 Table).
Data management and statistical analyses: Each individual was considered “infected” if the qPCR was positive in one or both eye swabs. When appropriate, database was organized as recommended for proportion data [39]. Mycoplasma conjunctivae apparent prevalence was analyzed to assess 1) the relation between ocular clinical signs and the presence of M. conjunctivae and 2) the trend of M. conjunctivae infection probability during the study period in each study area. For the first analyses, a two-sided Chi-squared test for independence was performed. In the second analysis, generalized additive models (GAMs) were fitted with M. conjunctivae infection as response variable with a binomial distribution and the interaction of year with study area as predictor variables [40]. GAMs can be used to model trends as a nonlinear function of time and provide a framework for testing statistical significance of changes in the response variable frequencies [41]. Known risk factors for M. conjunctivae infection, such as sex and age category [8], were previously tested with Fisher’s exact tests to be equally represented in all the years for each study area. The absence of residual patterns and other general assumptions were confirmed to validate the model once it was fitted [42]. Statistical significance was set at p<0.05 for all the tests. The interval confidence of apparent prevalences were calculated with the “EpiR” package, the graphics were performed with the “ggplot2” package and the GAMs were implemented in the “mgcv” statistical package, all from the R statistical software [43]. The spatial data representation and mapping was made with the software QGIS 2.14 Essen [44].
|
![[This page as RDF]](http://data.gesis.org/somesci/static/rdf-icon.gif){kind=icon}