Echis carinatus sochureki Kemmler

Chambrier, Alain de, Alves, Philippe V., Schuster, Rolf K., Toma & Scholz, s, 2021, Ophiotaenia echidis n. sp. (Cestoda: Proteocephalidae) from the saw-scaled viper, Echis carinatus sochureki Stemmler (Ophidia: Viperidae), one of the world’s deadliest snakes, from the United Arab Emirates, International Journal for Parasitology: Parasites and Wildlife 14, pp. 341-354 : 342

Tapeworms were found in the small intestine of six saw-scaled vipers, Echis carinatus sochureki Kemmler found in different localities of the United Arab Emirates and examined from 2003 to 2020. This subspecies occurs in northern India, Bangladesh, southern Afghanistan, Pakistan, central Iran, southern Iraq and the United Arab Emirates, with an isolated population in southeastern Arabian Peninsula ( Uetz et al., 2020). The nominotypical subspecies E. carinatus carinatus (Schneider) is limited in its distribution to peninsular India ( Uetz et al., 2020). None of 33 other snakes examined was infected, namely one Echis omanensis Babocsay , five Cerastes gasperettii Leviton et Anderson (both Viperidae ), ten Eryx jayakari Boulenger ( Boidae ), one Malpolon moilensis (Reuss) , 13 Psammophis schokari (Forskål) (both Psammophiidae ), and three Platyceps rhodorachis (Jan) (Colubriidae). A total of 58 tapeworms in different states of maturity were found, but some specimens were kept in tap water until their strobilae were totally relaxed. Therefore, only a few specimens collected in a fresh snake (host code No. UAE 4) fixed in 4% hot formalin were suitable for morphological evaluation (see below).

Tapeworms were stained with Mayer’ s carmine, dehydrated in an ethanol series, clarified by eugenol (clove oil) and mounted in Canada balsam as permanent preparations. For histology, pieces of strobila were embedded in paraffin, transversely sectioned at 12–15 μm intervals, stained with Weigert’ s hematoxylin, and counterstained with 1% eosin B (acidified with five drops of pure acetic acid for 100 ml solution) (see de Chambrier, 2001). Eggs were studied in distilled water. For scanning electron microscopy (SEM) observations, one scolex was dehydrated through a graded ethanol series, dried in hexamethyldisilazane, coated with gold (thickness of 10–20 nm) and examined in a JEOL JSM-740 1F scanning electron microscope at the Institute of Parasitology, Biology Centre of the Czech Academy of Sciences. All measurements in morphological description are given in micrometres unless otherwise indicated. Abbreviations used in description (usually if the number of measurements was> 5) are: x = mean; n = number of measurements. Host and zoogeographical realm classifications follow Uetz et al. (2020) and Holt et al. (2013), respectively. Material studied is deposited in the Natural History Museum, Geneva, Switzerland (acronym MHNG-PLAT), and in the Helminthological Collection of the Institute of Parasitology, Cesk ˇe´Budˇejovice, Czech Republic (IPCAS).

A piece of another specimen (paragenophore from host UAE 03; MHNG-PLAT-120508) was used for DNA sequencing (courtesy of J. Brabec) of the large subunit nuclear ribosomal RNA (lsr DNA; D1–D3 domains) and the partial mitochondrial cytochrome c oxidase subunit I (COI) following the methodology outlined by de Chambrier et al. (2019). The sequences were assembled and inspected for errors using Geneious version R11 ( Kearse et al., 2012), and submitted to GenBank (MW703700 – lsr DNA; MW703548 – COI); COI gene assembly was trimmed to the protein-coding region using the echinoderm translational code. Two alignments were created using the newly obtained sequences and selected members of the Proteocephalidae mostly corresponding to Clade K of de Chambrier et al. (2015) (these taxa were informed by a more comprehensive analysis of currently unpublished data) using default parameters of MAFFT ( Katoh and Standley 2013) implemented in the Guidance2 web server (http://guidance.tau.ac.il/; Sela et al. 2015): (i) alignment including only the lsr DNA sequence data, and (ii) a concatenated alignment (lsr DNA + COI) including only those representatives with available sequences for both markers (see Table 1). Unreliable positions in the alignments were identified and removed using the Gblock web server (https://ngphylogeny.fr/; Dereeper et al., 2008) with less stringent settings.

Phylogenetic reconstructions were performed with the maximum likelihood (ML) criterion using the evolutionary models implemented in ModelFinder ( Kalyaanamoorthy et al., 2017) within IQ-TREE ( Trifinopoulos et al., 2016), based on the small sample size corrected Akaike Information Criterion (AICc). The models chosen were as follows: TIM3 + F + R2 for the lsr DNA dataset alone; K3P + I for 1st codon position and GTR + F + G4 for lsr DNA, 2nd and 3rd codon positions using the concatenated dataset. The ML trees were generated via IQ-TREE and clade supports were estimated with 5000 replicates of the ultrafast bootstrap (UFBoot – Minh et al., 2013) and an SH-aLRT test with 5000 replicates ( Guindon et al., 2010). To avoid overestimation of UFBoot, we used a hill-climbing nearest neighbour interchange (NNI), as recently recommended by Hoang et al. (2018). Clades with support values of both UFBoot ≥95 and SH-aLRT ≥ 80 were considered strongly supported, while clades with only one of UFBoot ≥95 or SH-aLRT ≥ 80 were weakly supported; nodes with both UFBoot <95 or SH-aLRT <80 were unsupported. All the above-mentioned analyses were run on the computational resource CIPRES ( Miller et al., 2010).