Scanning Electron Microscopy and First Molecular Data of Two Species of Lamproglena (Copepoda: Lernaeidae) from Labeo victorianus (Cyprinidae) and Clarias gariepinus (Clariidae) in Kenya

A parasitological study carried out in May 2022 and March 2023 in the Nyando River of Lake Victoria Basin, Kenya, disclosed two parasitic lernaeid copepods: Lamproglena cleopatra Humes, 1957, from the gills of a cyprinid, the Ningu Labeo victorianus Boulenger, 1901, endemic to the Lake Victoria drainage system, and Lamproglena clariae Fryer, 1957, from a clariid, the North African catfish Clarias gariepinus (Burchell, 1822). The copepods were studied and supplementary taxonomic information was presented using scanning electron micrographs and genetic data. Scanning electron microscopy (SEM) provided information on the morphology of L. cleopatra’s antennae, oral region, thoracic legs (2–5), and furcal rami not previously reported. Analyses of the partial fragments of 18S and 28S rDNA and cytochrome c oxidase subunit 1 (cox1) of the two parasites showed them to be distinct from all other Lamproglena taxa retrieved from GenBank. This study presents new taxonomic information on morphology using SEM and provides the first ribosomal (18S and 28S rDNA) and mitochondrial (mtDNA) data for these two parasite species. The cox1 data provided are the first for all 38 nominal species of Lamproglena. Notably, the study also provides a new host record for L. cleopatra and extends the geographical information of this species to Kenya.


Introduction
Lernaeidae Cobbold, 1879 comprises among other things the cosmopolitan parasitic freshwater copepods Lamproglena von Nordmann, 1832. This genus, comprising 38 nominal species, is regarded as the oldest and second-largest member of this family [1][2][3]. Out of the 38 valid species only 12 (31.59%) have been reported from Africa (Lamproglena hemprichii  Table 1. Measurements in millimetres with mean followed by standard deviation and range in parentheses of various taxonomic features of Lamproglena cleopatra Humes, 1957 for the present study and comparisons with previous studies. Humes [8] Kunutu et al. [ For SEM, four specimens fixed in 70% ethanol were prepared by dehydrating through graded ascending ethanol concentrations. The dehydration process consisted of 20 min sequential exchanges in increasing ethanol concentrations of 80%, 90%, 96%, 96%, 99.98%, and 99.98%. The samples were then dried for a 20 min sequential exchange using graded ascending series of Bis(trimethylsilyl)amine 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100%, 100%, and 100% based on the procedures outlined by Nation [20] and Dos Santos et al. [21] with adjustments on the concentrations of ethanol and Bis(trimethylsilyl)amine and timing. Following this, the copepods were transferred into a glass desiccator for 24 h at room temperature and gold coated using a Quorum TM Q150T Emscope sputter coater (Quorum Technologies Ltd., Newhaven, U.K.). The copepods were then examined using a Zeiss Sigma 500VP scanning electron microscope (Jena, Germany) at 4 kV acceleration voltages at the University of Limpopo. Photomicrographs from LM and SEM aided in the morphometric redescription of the copepods.

Phylogenetic Analyses
The novel sequence data obtained were assembled and inspected using the built-in De Novo Assembly tool in Geneious Prime v2022.2. (https://www.geneious.com). The resulting consensus sequences, 18S, 28S rDNA, and cox1, were subjected to a Basic Local Alignment Search Tool (BLAST, https://blast.ncbi.nlm.nih.gov/Blast.cgi, accessed on 10 July 2023) [24] to identify the closest congeners. Alignments for each gene/region fragment were constructed under the default parameters of MAFFT in Geneious and trimming of the 28S alignment was performed in trimAL v.1.2. using the "gappyout" parameter selection under default settings to remove gaps in the alignment [25]. There were no comparable sequences in GenBank for Lamproglena for the cox1 sequences generated in this study. The species used in the phylogenetic trees are outlined in Table 2. For all the alignments the parasitic copepod Lernea cyprinacea Linnaeus, 1758 was selected as the outgroup. The best fitting model selected for 18S and 28S rDNA alignments according to the Akaike Information Criterion (AIC) from jModelTest v2.1.4. [26] was the GTR + I + G (general time-reversible model with invariant sites and gamma distribution) model. Maximum Likelihood (ML) analyses were computed in phyML using ATGC Montpellier Bioinformatics Platform specifying AIC criterion, model selection, and a bootstrap value of 100 (http://www.atgc-montpellier.fr/, accessed on 10 July 2023) [27]. Bayesian Inference (BI) analyses were performed in MrBayes using the CIPRES [28] computational resource. The BI analyses were generated by implementing a data block criterion running two independent Markov Chain Monte Carlo (MCMC) chains of four chains for 1 million generations. A sampling of the MCMC chain was set at every 1000th generation and a burn-in was set to the first 25% of the sample generations. Phylogenetic trees generated were visualised in FigTree v1.4.4. [29]. The uncorrected pairwise distances (p-distances) were estimated in MEGA 7.0 [30] and the number of base pair differences was calculated in Geneious.

Results
A total of 20 female L. cleopatra occurred on the gills of 34 Labeo victorianus.

Taxonomic Summary
Lamproglena cleopatra Humes, 1957.  Figure 1A,B). Furcal rami ( Figure 1A,B,M,N) minute, 0.028 (0.02-0.03) wide, 0.037 (0.03-0.04) long. Each ramus with one long seta, one pore on inner and outer margins, and terminally with four setae, one blunt process, and two pores ( Figure 1N). Antennules uniramous, indistinctly two-jointed with long swollen basal podomere bearing 11-14 naked setae and small distal podomere with 5 naked setae, 1 lateral and 4 terminal. Dorsal side of antennule with circular pores ( Figure 1C-E). Antenna uniramous, indistinctly four-jointed, distal segment with five small terminal setae ( Figure 1C-E). Oral region consisting of distinct projecting sucker-like with two lateral lobes from which arises two long setae and two finger-like posterior lobes ( Figure 1C-E). Mandible not observed. Maxilla uniramous, rigid, covered with a thin layer through which distinct terminal spine projects, basal region finely granulated ( Figure 1A-E). Maxilliped equipped with three roughly equal, curved claws, with a minute spine-like protrusion on the proximal part ( Figure 1F). Legs 1-4 biramous, rami of legs indistinctly two-jointed. Endopodites of legs 1-4 all similar, terminating in a minute, rather blunt seta. Protopodite of legs 1-4 with one lateral long seta at the base before exopodite ( Figure 1G-J). Exopodite of first leg first podomere with one smaller seta and four long terminal setae on the second podomere ( Figure 1G). Second leg first exopodite podomere with one basal seta, second exopodite podomere with two small setae and a minute knob, an opening between setae and knob ( Figure 1H). Second exopodite podomere of third and fourth legs with four setae: two long, one medium, one min ( Figure 1I,J). Fifth leg made of small lobe with two long distal and one lateral seta ( Figure 1K). Spermatophore observed ( Figure 1I,L). Egg sac 0.98 × 0.24, containing about 20 eggs (19-22) ( Figure 1A). with four setae: two long, one medium, one min ( Figure 1I,J). Fifth leg made of small lobe with two long distal and one lateral seta ( Figure 1K). Spermatophore observed ( Figure  1I,L). Egg sac 0.98 × 0.24, containing about 20 eggs (19-22) ( Figure 1A).  Remarks: The parasitic copepods studied here were indistinguishable from L. cleopatra as per the available morphological information published by Humes [8] and Kunutu et al. [2] and clearly distinct from other species of this genus. The indistinguishable features were as follows: body elongated, cylindrical and divided into a cephalothorax, thorax, and abdomen; cephalothorax broader than neck; first thoracic legs fused with the head; thoracic segments marked by lateral constrictions; indistinctly segmented abdomen; three clawed maxilliped; genital somite laterally protruding and distinctly demarcated from the rest of the thorax by a deep indent; antennule larger than antenna; biramous legs; and furcal rami with long lateral processes and terminal setae. Slight variations were noted between the present material and previous records of Humes [8] and Kunutu et al. [2], but the additional taxonomic features observed in the present material were as follows: two long setae on lateral lobes of the oral region ( Figure 1C-E) and four circular pores on the furcal rami ( Figure 1N). Remarks: Based on the morphological data available from the reports of Fryer [5] and Marx and Avenant-Oldewage [14], the present material was identical to L. clariae. Following a detailed redescription of this parasite using LM and SEM by Marx and Avenant-Oldewage [14], the present study only provided the SEM images to confirm the identity of our specimen and most importantly provided genetic sequences using 18S, 28S, and cox1 markers.

Molecular Identification
This study generated a total of 11 novel sequences of the three genetic markers: 5 sequences for L. cleopatra and 6 sequences for L. clariae. The Bayesian Inference and Maximum Likelihood analyses of the 18S alignment yielded similar hypotheses (nt = 1325) (Figure 3). The newly generated sequences for L. clariae and L. cleopatra fell into the clade of Lamproglena species previously reported from Africa with strong support. The sequences for L. clariae clustered together with high nodal support and formed a separate branch to the L. monodi clade with no nodal support. The novel sequences for L. cleopatra clustered together and formed a separate clade with L. hemprichii (OP277526) at the basal position of the African clade with no nodal support. The BI and ML analyses for the 28SrDNA dataset showed similar topologies (nt = 696) (Figure 4). A clear distinction between Lamproglena species from Africa and Asia clades were observed. The sequences for L. clariae fell at the basal position of the African clade with strong nodal support. The L. cleopatra sequences clustered with the L. hemprichii (OP277527) previously reported from Zimbabwe with strong nodal support.
The results from the analysis of the 18S and 28S rDNA haplotypes showed a distinct match with all sequences of the four Lamproglena species present in GenBank. There were no cox1 mtDNA sequences available in GenBank for this genus for species comparisons. The pairwise distances (p-distances) and number of base pair differences of L. cleopatra and L. clariae for small (18S) and large (28S) subunit rDNA and all sequences belonging to the Lernaeidae used in this analysis are presented in Tables 3 and 4, respectively.

Molecular Identification
This study generated a total of 11 novel sequences of the three genetic markers: 5 sequences for L. cleopatra and 6 sequences for L. clariae. The Bayesian Inference and Maximum Likelihood analyses of the 18S alignment yielded similar hypotheses (nt = 1325) (Figure 3). The newly generated sequences for L. clariae and L. cleopatra fell into the clade of Lamproglena species previously reported from Africa with strong support. The sequences for L. clariae clustered together with high nodal support and formed a separate branch to the L. monodi clade with no nodal support. The novel sequences for L. cleopatra clustered together and formed a separate clade with L. hemprichii (OP277526) at the basal position of the African clade with no nodal support. The BI and ML analyses for the 28SrDNA dataset showed similar topologies (nt = 696) (Figure 4). A clear distinction between Lamproglena species from Africa and Asia clades were observed. The sequences for L. clariae fell at the basal position of the African clade with strong nodal support. The L. cleopatra sequences clustered with the L. hemprichii (OP277527) previously reported from Zimbabwe with strong nodal support. The two copepods in the present study, L. clariae and L. cleopatra, were distinct from other Lamproglena species by p-distances of 0.9-2.1% (13-29 bp) and 0.1-2.0% (1-30 bp) based on 18S rDNA (Table 3). For the 28S rDNA, the results showed p-distances of 16.8-23.7% (120-167 bp) and 7.1-23.3% (46-156 bp), respectively ( Table 4). The two ribosomal DNA (18S and 28S) markers produced nearly similar topologies with insignificant intraspecific branching. The unavailability of mitochondrial (cox1) marker sequences in GenBank made it impossible to construct any phylogeny tree; therefore, the p-distance and number of base pair differences are provided for cox1 sequences (nt = 683) generated from the present study (Table 5).