Considering South Africa’s richness in aquatic species, very little knowledge exists regarding copepods that are symbiotic on hosts ranging from invertebrates to marine mammals. In order to have any indication of the existing biodiversity of this group of organisms in South African waters, a thorough investigation of all possible hosts needs to be conducted, which in turn will most likely increase the number of recorded symbiotic copepods considerably. The current descriptive study was done in an effort to contribute to a larger study of metazoan parasites of elasmobranch hosts along the KwaZulu-Natal coast of South Africa. A total of 40 (31 Mobula kuhlii; two Mobula eregoodootenkee and seven Manta alfredi) mobulids were examined for infection by symbiotic copepods at the KwaZulu-Natal Sharks Board (KZNSB). More than 90% of all examined hosts were infected with different types of symbiotic siphonostomatoids. Collected copepod specimens were fixed and preserved in 70% ethanol and studied with both the stereo- and light microscopes using the wooden slide technique. Some selected specimens were further studied using Scanning Electron Microscopy (SEM) to elaborate on ill-defined features. A total of 13 different species of the order Siphonostomatoida distributed over five families were identified. The five families include Eudactylinidae (Eudactylina oliveri, Eudactylina diabolophila and Nemesis sp.); Caligidae (Caligus crysophrysi, Pupulina sp. 1, Pupulina sp. 2; Pupulina sp. 3, Unidentified sp. 1, Unidentified sp. 2 and Unidentified sp. 3); Kroyeriidae (Kroeyerina mobulae); Dichelesthiidae (Anthosoma crassum) and Cecropidae (Entepherus laminipes). Two of the 13 species (E. laminipes and A. crassum) are monotypic and were therefore easily identified. Eudactylina oliveri exhibited a prevalence of 75% and 100%; mean intensity of 42 and 130 parasites per host and a mean abundance of 32 and 130 individuals per host while Pupulina sp. 1 exhibited a prevalence of 61.29% and 100%; mean intensity of 41 and 5 individuals per host and a mean abundance of 2 and 5 individuals per host on M. kuhlii and M. eregoodootenkee respectively. Component populations of E. oliveri and Pupulina sp. 1 exhibited an aggregated distribution pattern on their examined hosts. The phylogenetic relationship between nine caligid species (three known Pupulina species, three collected Pupulina species and three Unidentified sp. species as ingroup) with Caligus glandifer as out-group was determined and analysed using a morphological dataset (40 characters) from previous and current descriptions. The ii exhaustive search with PAUP* retained a single most parsimonious tree with a tree length (TL) = 85; consistency index (CI) = 0.7; retention index (RI) = 0.7; homoplasy index (HI) = 0.3 and a rescaled consistency index (RCI) = 0.5. Bootstrap support for the estimated clades was mostly low with values less than 95%. The phylogenetic hypothesis of the 10 caligid species presented in the current study was derived from the phylogenetic analysis of the information for adult females and is therefore not intended to be a definitive theory but should be treated as a testable hypothesis that can be further analysed using more data. The current study provides the first record of C. chrysophrysi, Pupulina sp. 1, Pupulina sp. 2, K. mobulae and E. laminipes on M. kuhlii; E. oliveri, Pupulina sp. 1, Pupulina sp. 2 and Pupulina sp. 3 on M. eregoodootenkee; and E. diabolophila, Nemesis sp., C. chrysophrysi, E laminipes, A. crassum and the three Unidentified species on M. alfredi frequenting the east coast of South Africa and thus contributes to the knowledge of our marine biodiversity. Mobulid hosts were not carefully studied for copepod infection previously and the copepods that were reported from the mobulids were probably found by chance. Therefore future investigation into the symbiotic siphonostomatoids of more mobulid hosts and other host species may result in more reports of symbiotic Copepoda from South African waters. / Thesis (MSc. (Zoology)) -- University of Limpopo, 2013

Thesis (MSc. (Zoology)) -- University of Limpopo, 2014 / One of the 11 families of the siphonostomatoids found parasitic on elasmobranchs is the Kroyeriidae which has three accepted genera namely Kroyeria, Kroeyerina and Prokroyeria. Parasites from this family are found living on the gills (Kroyeria spp. and Prokroyeria sp.) or in the nasal fossae (Kroeyerina spp.) of Chondrichthyes. There are currently 21 nominal species in the genus Kroyeria. Kroyeria specimens were collected from the gill filaments of their elasmobranch hosts which were caught: (1) in the nets of the KwaZulu-Natal Sharks Board (KZNSB) installed along the east coast of South Africa, (2) by commercial fishermen off the west coast at Gansbaai as well as (3) during the demersal trawls of Department of Agriculture, Forestry and Fishery (DAFF) off the south and west coasts. Collected specimens were fixed and preserved in 70% ethanol. Morphological features were drawn where necessary to illustrate differences from previously described features. Host-parasite relationships of the different species were determined by calculating prevalence, mean abundance and mean intensity on their hosts as well as estimating the pattern of dispersion by calculating the coefficient of dispersion. DNA was extracted from selected identified samples. A partial fragment of the COI gene was amplified via PCR using the forward and reverse universal primers LCO 1490 and HCO 2198, or those with additional M13 tails, LCO 1490_t1 and HCO 2198_t1. Additionally, the complete 18S rDNA gene of some species was amplified using the forward and reverse primers as follows: 18Sf and 1282r for the first fragment, 554f and 614r for the second fragment and 1150f and 18sr for the third fragment. Phylogenetic relationships among different Kroyeria species were estimated by employing neighbor joining (NJ), parsimony (MP) and maximum likelihood (ML) in PAUP*. The use of real-time PCR and melt curve analysis to distinguish among different Kroyeria species based on their different melt temperatures of a part of the COI gene was also attempted. Eleven Kroyeria species were found on the gill filaments of elasmobranchs belonging to the families Carcharhinidae, Sphyrnidae and Triakidae off the coasts of South Africa. These include K. carchariaeglauci from C. leucas; K. decepta from C. obscurus; K. deetsi from C. brevipinna; K. dispar from G. cuvier; K. elongata from R. acutus; K. lineata from M. palumbes; K. longicauda from C. limbatus; K. papillipes viii from G. cuvier; K. procerobscena from both C. leucas and C. amboinensis; K. sphyrnae from both Sphyna lewini and S. zygaena and a new Kroyeria sp. from G. galeus. This is the first record of K. lineata from the south coast of South Africa and is also as a new host record for Mustelus palumbes. Three Kroyeria species have previously been reported from G. galeus, namely K. brasiliense, K. lineata and K. rhophemophaga. The new Kroyeria sp. is most similar to K. rhophemophaga which in turn shares morphological features with K. triakos. However, the Kroyeria sp. can be distinguished from both K. rhophemophaga and K. triakos in the armature of the legs. Most Kroyeria species are relatively host specific, infecting a single host or related group of host species. During this study two species, K. dispar and K. papillipes were collected from G. cuvier, while K. procerobscena and K. sphyrnae were each collected from two host species. Kroyeria sp. and K. dispar displayed very high prevalence values, 95.7% and 94.1% respectively, in contrast to the other Kroyeria species which have lower values (6.3–68.6%). Additionally, when compared to other siphonostomatoid species such as Nemesis lamna, Kroyeria species have relatively low prevalence values. Kroyeria species generally have low parasite loads (between 4 and 33 copepods per infected host), except for K. dispar which has a high mean intensity of 74 copepods per infected host. The mean abundance of Kroyeria species is also generally low (between 0 and 23 per examined host), with K. dispar (69 individuals per examined host) being an exception. Furthermore Kroyeria species generally display an aggregative pattern of distribution which is common in most copepod species indicating that individuals have social interactions. A preliminary estimation of the phylogenetic relationships among seven Kroyeria species revealed topologies with unresolved polytomies. The 18S rDNA gene did not make any significant changes on the topology, except that it produced very minimal resolution in one of the groupings. Therefore, COI is found to be a gene of choice that can be used in estimating molecular phylogenetics and population genetics of siphonostomatoids as it provides useful sequence divergence within individuals of the same species as well as among congeneric species due to its fast evolving rate. However, in this study, single species did not form monophyletic groupings. ix The 18S rDNA gene is found to be very conservative, providing no sequence divergence within individuals of the same species and very little divergence among conspecifics due to its low mutation rate and is therefore more useful at genus and family levels. With polytomies in the estimated phylogenetic relationships, haplotype networks were used to compare the distribution of different haplotypes among the different species. Haplotype sharing did occur between species e.g. for COI, H1 is shared by K. lineata, Kroyeria sp. and K. sphyrnae. This haplotype sharing by different species is unexpected and could be due to specimen misidentification before DNA extraction. Specimen misidentification is common for Kroyeria species because some of them are not easy to identify. The haplotype network results confirmed the relationships shown by the phylogenetic trees, dividing Kroyeria species into three different groupings. Real-time PCR and melt curve analysis have the potential to distinguish among Kroyeria species. However, the quality of the extracted DNA is an important factor in producing successful amplifications and determining the Tm. Therefore it is necessary to ensure that the extracted DNA has the ideal concentration of 50 ng/μl and is free of Taq polymerase inhibitors such as phenol, RNA and guanine residuals from the extraction process.