Molecular Detection of Cryptosporidium cuniculus in Rabbits (Oryctolagus cuniculus) from Tenerife, Canary Islands, Spain

Cryptosporidium cuniculus is a zoonotic parasite responsible for cryptosporidiosis cases and outbreaks in both humans and rabbits. Since there are no molecular Cryptosporidium spp. infection data in rabbits (Oryctolagus cuniculus) from Spain, our aim was to gather information about this parasite in wild European rabbits from Tenerife, Canary Islands (Spain). A total of 100 faecal samples were collected from rabbits from eight municipalities of Tenerife. Microscopic analysis showed that 4.0% of the samples presented structures compatible with Cryptosporidium oocyst. A nested polymerase chain reaction (PCR) targeting 18S ribosomal RNA (rRNA) gene fragments was carried out, and sequencing confirmed the identity of C. cuniculus in one sample (1.0%). The sample was successfully subtyped using nested PCR analysis of the 60-kDa glycoprotein (gp60) gene as the subtype VbA26R3. This study confirms the presence of C. cuniculus in wild rabbits from Tenerife, providing new information on the occurrence of this zoonotic parasite. Further studies are required to better understand the epidemiology of Cryptosporidium spp. in wild rabbits in Spain and their possible public health repercussions.


Introduction
The Canary Islands are a Spanish archipelago composed of eight islands and five islets located in the Atlantic Ocean, near the coast of northwest Africa (13 • 23 -18 • 8 W and 27 • 37 -29 • 24 N). The European rabbit (Oryctolagus cuniculus), an "invasive introduced" species [1] native to the Iberian Peninsula, was introduced to these islands in the 15th century [1]. Nowadays, it has been established as a species of economic and cultural interest in the islands, due to hunting activity and farming. The Canary Islands have 3.2% of the rabbit farms of Spain [2] and the wild rabbit is one of the small game hunting species in Tenerife [1]. In 2017, rabbit abundance was estimated in a mean value of 2.22 individuals/ha in Tenerife, with a standard derivation of 2.25 individuals/ha, which suggests that there is high spatial variability in the abundance of the species. In general, abundance was higher in areas of low elevation and slope [3].
The European rabbit is a well-known host of several pathogens such as helminths, viruses, and protozoa [4], including Cryptosporidium spp.
Cryptosporidium is a genus of parasitic protozoa comprising of 48 valid species and more than 100 genotypes [5] that infect a wide range of hosts, including humans, mammals, birds, reptiles, and fish [6]. The transmission of the infective stage occurs through the ingestion of sporulated oocysts via contact with an infected person/animal, ingestion of contaminated water or food, or possibly through the air [7]. Infection ranges from being asymptomatic to mild-severe diarrhea. The parasite has a cosmopolitan distribution, particularly in developing countries [8].

Sample Collection
A total of 100 faecal samples from rabbits donated by hunters (n = 82) and found dead (n = 18), from eight municipalities of Tenerife: Tegueste, San Cristóbal de La Laguna, El Sauzal, La Matanza de Acentejo, Arafo, La Orotava, Güímar, and Granadilla de Abona ( Figure 1) were collected between 2015-2017 and placed into sterile plastic containers until reception in the laboratory, then were subsequently deposited in vials containing 2.5% aqueous (w/v) potassium dichromate (K 2 Cr 2 O 7 ) solution. The samples were stored at 4 • C until the posterior analysis.

Ethical Statement
The samples used in this study were donated by hunters that hunted wild rabbits during the legal hunting season period, detailed in the numbers of the Official Bulletin of Canaries (BOC) (http://www.gobiernodecanarias.org/boc/ accessed on 11 February 2022); 128 from 2015, and 125 from 2016 and 2017, respectively. Therefore, no rabbits were sacrificed for this study. No ethical approval was required.

Staining Method
Samples were stained using the Kinyoun TB Stain Kit K (Becton, Dickinson and Company, USA) following the manufacturer's instructions and microscopically screened for Cryptosporidium spp. oocysts. The samples with oocyst-compatible structures were identified as positive and preserved until DNA extraction and PCR analysis.

DNA Extraction
An aliquot of ∼200 µL of each sample identified as positive diluted in potassium dichromate was washed with PBS-EDTA at room temperature to remove the potassium dichromate. Then, they were transferred to centrifuge tubes containing 500 µL of lysis buffer, and one freeze-thaw cycle (−80 • C to +100 • C) in boiling water was made prior the extraction procedure [70]. Total DNA was isolated with the commercial FastDNA ® Spin Kit for Soil (MP Biomedicals, Solon, OH, USA) following the manufacturer's instructions, with the homogenizer FastPrep-24™ 5G (MP Biomedicals, Solon, OH, USA) as oocyst disruptors.
The nested PCR products were sequenced at Macrogen Spain, with the secondary pair of primers in both senses.

Sequencing and Phylogenetic Analysis
The fragments of the nucleotide sequences obtained were edited with the MEGA X program [73], and subsequently aligned with other Cryptosporidium species/subtypes sequences using the ClustalW program included in MEGA X. Minor corrections, to increase the aligned sequence similarity and improve the inferences on any positional homology, were then made by hand.
A Basic Local Alignment Search Tool (BLAST) search was carried out in order to elucidate any homologies or similarities with the sequences previously published in the GenBank database.
The molecular identification of the 18S rRNA and gp60 genes was achieved by phylogenetic analysis through the Neighbor-Joining distance method with the p-distance model [74] and maximum-likelihood method with a Tamura-Nei model [75], both with at least 1000 bootstrap replications in MEGA X using the sequence of Eimeria magna (HQ173833.1) as the outgroup. Nucleotide sequences obtained in this work were submitted to the GenBank database under the accession numbers OM170342 and OM249938 for 18S rRNA and gp60 genes, respectively.

Staining and Molecular Results
In four (4.0%) of the faecal samples screened by the Kinyoun method and light microscopy, Cryptosporidium oocyst-compatible structures were found. One sample was amplified by nested PCR with the expected size, and it was identified as C. cuniculus by sequencing.
Therefore, the occurrence of C. cuniculus in wild rabbits from Tenerife was 1.0% (1/100). The positive sample was located in La Orotava from a rabbit donated by hunters, with an overall occurrence of 6.25% (1/16) in this municipality.
An alignment of 921 bp was used for the phylogeny. The result of the Neighbor-Joining analysis based on gp60 is shown in Figure 3. The results of the Neighbor-Joining (Figures 2 and 3) and the maximum-likelihood (Supplementary Material, Figures S1 and S2) analyses based on the 18S rRNA and gp60 genes identified and isolated as C. cuniculus (100% bootstrap) clearly separated to other Cryptosporidium species (Figure 2), belonging to the Vb family (100% bootstrap) and differing from the Va family (Figure 3), respectively.

Discussion
The present study provides data about the occurrence of Cryptosporidium cuniculus in wild rabbits in the Canary Islands. To date, some previous studies carried out in rabbits were based on the detection of Cryptosporidium spp. without reaching the species level due to the techniques employed: faecal examination using light microscopy in pet rabbits in Japan [76], Brazil [77] and Egypt [78]; immunofluorescent microscopy in faecal samples from pet rabbits in the UK [79]; antigen detection by enzyme-linked immunosorbent assay (ELISA) in farmed rabbits in Nigeria [80]; and staining methods in rabbits from Spain [81,82], Australia [83], the UK (see [84]), China (see [14]), Iraq [85], and Ecuador [86].
Before the re-description of the C. cuniculus species, other authors cited C. parvum in rabbits, and using light microscopy and indirect immunofluorescence in the UK [87], staining and/or histological techniques in the Czech Republic, Belgium, Hungary, and the USA (see [84]), and staining and molecular methods in China (see [14]) may have led to misdiagnoses.
In the Iberian Peninsula (Spain) from where the Canary rabbit population was introduced, previous studies have already evidenced the presence of Cryptosporidium spp. in faecal samples from rabbits: in farmed rabbits from Toledo [81], farmed rabbits from León, and wild rabbits from Madrid [82], but in these cases the specific diagnosis was not possible. To our knowledge, three publications describing the detection of C. cuniculus in Spain are available, of which two cases were in humans, one autochthonous and a Spanish travel-related case, in addition to a wastewater detection [37, 58,59]. However, the molecular detection of C. cuniculus in wild rabbits from Spain had not been reported.
The occurrence of C. cuniculus in wild rabbits from Tenerife of 1.0% (1/100) is similar to other studies carried out on wild rabbits from Europe, ranging from 0-0.9% in Germany (0/232) and the UK (1/109). These data differ from those reported in wild rabbits from the UK, 7% (2/28); Spain, 100% (7/7) [82], and The Netherlands, 6.5% (2/31) (see [84]); possibly due to the small size of these studies. Chalmers et al. 2009 [21] propose the classification of the C. cuniculus subtypes into two families, Va and Vb, depending on the TCA repetition number in tandem with the microsatellite region of the gp60 gene, and Nolan et al. 2010 [22] recommend the subtype division depends on ACA triplet repetitions in the region immediately following the microsatellite region, specifying it with an R. Following the proposed nomenclature, the novel genotype sequence obtained in this study (Acc. Number: OM249938) should be named VbA26R3.
Other authors cited VbA26subtype detection in wild rabbits from Australia [23,24,26], in Eastern grey kangaroos [24,31], in South Africa river water [42], and humans from the UK [48] and New Zealand [55], thus confirming the zoonotic potential of this subtype.
There are previous studies on molecular epidemiology in wild rabbits, but this was only located in Australia, with prevalence data ranges from 2.18-14.3%, and with the Vb family being the only family detected and VbA23subtype being the most prevalent (Table 3), highlighting C. cuniculus as the most prevalent Cryptosporidium species reported in more than 11 host species in wildlife in Australia [23,24]. Outside Australia, there is only molecular epidemiology data of farmed rabbits, with prevalence ranging from 2.38-12.73% (Table 3), and the Vb family also being the most common (see Supplementary Material, Table S1: Cryptosporidium cuniculus subtypes reported worldwide).  In 2018, Spain was the fourth highest country in the European Union/European Economic Area (EU/EEA) with confirmed cases of cryptosporidiosis [88]. The national cases in 2018 were 1526 (5.22 per 100,000 population) in 11 autonomous communities and the autonomous city of Ceuta (not all of the country is notifying cases). In that year, the Canary Islands had the lowest notification rate of cryptosporidiosis in Spain (0.46 per 100,000 population) [89].
Since 2015, cryptosporidiosis has been declared as a notable disease in Spain. From that year, 54 cases have been confirmed in the Canary Islands, and in that period none of them were in Tenerife (Table 4). Considering the previous detection of Cryptosporidium species in Tenerife in humans from 2002-2004 [60], and in wildlife [64,66,67] and wastewater [68,69], possible underdiagnosis may have occurred. The detection of the zoonotic species C. cuniculus in this study highlights the potential role that wild rabbits can play in the maintenance and transmission of this species. Although the majority of rabbit cases develop asymptomatically, the infection can be associated with different pathologies, which can affect farm productivity and even animal survival, leading to economic losses [27, 81,90]. On the other hand, while the risk of zoonotic transmission is low, it should not be dismissed since C. cuniculus has been involved in human cases (see Table 1) and outbreaks [21], even becoming the third cause of cryptosporidiosis in the UK [48] and New Zealand [56].
For these reasons, further studies are required to better understand the epidemiology of Cryptosporidium spp. in wild rabbits in the Canary Islands and their possible public health repercussions.

Conclusions
The present study constitutes the first molecular detection of C. cuniculus in wild rabbits from Tenerife (Canary Islands, Spain), leading to the identification of the VbA26R3 subtype, with potential zoonotic risk. Considering the possible implications to public health of these results, more studies are required in order to evaluate the public health and veterinary risk.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/vetsci9020091/s1: Figure S1: Maximum-Likelihood tree of Cryptosporidium spp. based on the 18S rRNA gene sequence; Figure S2: Maximum-Likelihood tree of Cryptosporidium spp. genotypes based on gp60 sequences; Table S1: Cryptosporidium cuniculus subtypes reported worldwide. Institutional Review Board Statement: Ethical review and approval were waived for this study due to the samples used in this study were donated by hunters that hunted wild rabbits during the legal hunting season period, and no rabbits were sacrificed for this study.