The first next-generation sequencing approach to the mitochondrial phylogeny of African monogenean parasites (Platyhelminthes: Gyrodactylidae and Dactylogyridae)

Abstract

Background: Monogenean flatworms are the main ectoparasites of fishes. Representatives of the species-rich families Gyrodactylidae and Dactylogyridae, especially those infecting cichlid fishes and clariid catfishes, are important parasites in African aquaculture, even more so due to the massive anthropogenic translocation of their hosts worldwide. Several questions on their evolution, such as the phylogenetic position of Macrogyrodactylus and the highly speciose Gyrodactylus, remain unresolved with available molecular markers. Also, diagnostics and population-level research would benefit from the development of higher-resolution genetic markers. We aim to offer genetic resources for work on African monogeneans by providing mitogenomic data of four species (two belonging to Gyrodactylidae, two to Dactylogyridae), and analysing their gene sequences and gene order from a phylogenetic perspective.

Results: Using Illumina technology, the first four mitochondrial genomes of African monogeneans were assembled and annotated for the cichlid parasites Gyrodactylus nyanzae, Cichlidogyrus halli, Cichlidogyrus mbirizei (near-complete mitogenome) and the catfish parasite Macrogyrodactylus karibae (near-complete mitogenome). Complete nuclear ribosomal operons were also retrieved, as molecular vouchers. The start codon TTG is new for Gyrodactylus and for Dactylogyridae, as is the incomplete stop codon TA for Dactylogyridae. Especially the nad2 gene is promising for primer development. Gene order was identical for protein-coding genes and differed between the African representatives of these families only in a tRNA gene transposition. A mitochondrial phylogeny based on an alignment of nearly 12,500 bp including 12 protein-coding and two ribosomal RNA genes confirms that the Neotropical oviparous Aglaiogyrodactylus forficulatus takes a sister group position with respect to the other gyrodactylids, instead of the supposedly 'primitive' African Macrogyrodactylus. Inclusion of the African Gyrodactylus nyanzae confirms the paraphyly of Gyrodactylus. The position of the African dactylogyrid Cichlidogyrus is unresolved, although gene order suggests it is closely related to marine ancyrocephalines.

Conclusions: The amount of mitogenomic data available for gyrodactylids and dactylogyrids is increased by roughly one-third. Our study underscores the potential of mitochondrial genes and gene order in flatworm phylogenetics, and of next-generation sequencing for marker development for these non-model helminths for which few primers are available.

Keywords: Cichlidae; Cichlidogyrus; Clariidae; Gene order; Gyrodactylus; Macrogyrodactylus; Mitogenome; Monogenea; Monopisthocotylea; Phylogenomics.

Fig. 1

Mitochondrial genomes of four African monogeneans, including two members of the Gyrodactylidae: (a) Gyrodactylus nyanzae, (c) Macrogyrodactylus karibae (partial genome) and two representatives of the Dactylogyridae: (b) Cichlidogyrus halli and (d) Cichlidogyrus mbirizei (partial genome). The GC content is displayed for complete mitogenomes

Fig. 2

Sliding window analyses (window size 300 bp, step size 10 bp) of the alignment of mitochondrial protein-coding and ribosomal RNA genes used for the phylogenetic analyses of the four mitochondrial genomes of African monogeneans. The lines indicate the nucleotide diversity between two dactylogyrids (Cichlidogyrus halli and C. mbirizei, in blue) and two gyrodactylids (Gyrodactylus nyanzae and Macrogyrodactylus karibae, in red). Gene boundaries are indicated above the graph

Fig. 3

Ratio of non-synonymous to synonymous substitution rates for the protein-coding genes in two pairwise comparisons, between the mitogenomes of African dactylogyrid and gyrodactylid monogeneans, respectively. For Macrogyrodactylus karibae, no nad4 sequence was available, while the nad5 gene was lacking for Cichlidogyrus mbirizei

Fig. 4

Midpoint-rooted maximum likelihood phylogram of monopisthocotylean monogeneans based on 12 protein-coding and two ribosomal RNA genes. Support values displayed from (above branch): Shimodaira-Hasegawa-like approximate likelihood ratio test/ultrafast bootstrap, both implemented in IQ-TREE, (below branch) bootstrap in RAxML/Bayesian inference (posterior probability) in MrBayes. An asterisk (*) indicates that this partition was not withheld in the Bayesian consensus tree; the clade grouping Dactylogyrus lamellatus and Tetrancistrum nebulosi as sister to a monophyletic Cichlidogyrus was supported by a posterior probability of 58%. Branch lengths indicate the expected number of substitutions per site

Fig. 5

Family diagram explaining gene order changes between (a) African Gyrodactylus nyanzae and its Palearctic congeners (a single transposition), (b) G. nyanzae and Paragyrodactylus variegatus (a single transposition), (c) G. nyanzae and the Neotropical oviparous gyrodactylid Aglaiogyrodactylus forficulatus (two transpositions and a tandem duplication random loss event (TDRL)) and (d) Dactylogyrus lamellatus and the other dactylogyrids (a single transposition). Green boxes indicate transpositions, a dark blue box a TDRL. Only protein-coding genes, tRNA genes and rRNA genes of species for which a complete mitogenome was assembled, are shown

Fig. 6

Likelihood mapping (a) before and (b) after Gblocks trimming, demonstrating the high phylogenetic content and suggesting there is no need for alignment cleaning in the case of this dataset