Myxobolus cerebralis, the myxozoan parasite responsible for whirling disease in salmonid fishes, has a complex life-cycle involving an invertebrate host and 2 spore stages. Water flow rate is an environmental variable thought to affect the establishment and propagation of M. cerebralis; however, experimental data that separates flow effects from those of other variables are scarce. To compare how this parameter affected parasite infection dynamics and the invertebrate and vertebrate hosts, dead, infected fish were introduced into a naïve habitat with susceptible hosts under 2 experimental flow regimes: slow (0·02 cm/s) and fast (2·0 cm/s). Throughout the 1-year study, uninfected fry were held in both systems, the outflows were screened weekly for spores and the annelid populations were monitored. We found clear differences in prevalence of infection in the worms, prevalence and severity of infection in the fish, and host survival. Both flows provided environments in which M. cerebralis could complete its life-cycle; however, both the parasite and its invertebrate host proliferated to a greater extent in the slow flow environment over the 1-year study period. This finding is of significance for aquatic systems where the flow rate can be manipulated, and should be incorporated into risk analysis assessments.

Various mechanisms that enable and improve transmission success of myxozoan actinospore stages towards fish hosts are described, based upon a combination of experimental data and functional analysis of morphological characters. For this purpose, laboratory-reared actinospores of Myxobolus cerebralis, Myxobolus parviformis, Henneguya nuesslini and Myxobolus pseudodispar were employed to exemplarily investigate aspects of host attachment and invasion. The process of polar filament discharge of M. cerebralis actinospores was analysed, showing that full discharge occurs in less than 10 msec. Additionally, a mechanism that rapidly contracts the discharged filament after discharge is described for the first time. Its purpose is most likely to bring the actinospore apex rapidly into intimate contact with the surface of the host. Unlike M. cerebralis, M. parviformis actinospores did not discharge polar filaments after mechanical and chemical stimulation, suggesting a different mode of triggering. For H. nuesslini actinospores, experimental results indicated that polar filament discharge is independent of external calcium-ion concentration but is influenced by osmolality. After attachment of an actinospore and prior to penetration into the host, an ensheathed unit (‘endospore’), containing the sporoplasm, was emitted from the valves in a manner which varied from species to species. Experimentally induced sporoplasm emission was time-dependent and was found to be independent of polar filament discharge in H. nuesslini. Remarkably, it could be concluded that the sporoplasm is able to recognize host-stimuli while still within the intact spore. An updated summary of the sequential course of events during host recognition and invasion by actinospores is given.

The ZAP Express cDNA library was constructed using mRNA extracted from the triactinomyxon spores of Myxobolus cerebralis. First-strand cDNA was synthesized using Moloney Murine leukaemia virus reverse transcriptase. Following second-strand cDNA synthesis, the double-stranded cDNA was digested with Xho I restriction enzyme, cDNA fragments less than 400 bp were removed and the remaining cDNA was ligated with the lambda ZAP Express vector. The recombinants were packaged in vitro using Gigapack III gold packaging extract. The primary cDNA library titre contained 0·5×106 clones, with 97% recombinant and only 3% non-recombinant clones. The cDNA library was then screened using the anti-triactinomyxon antibodies. Positive clones were selected and re-screened twice more to give a final selection of 526 clones. One clone (46-5) was selected randomly and subjected to in vivo excision of the pBK-CMV phagemid from the ZAP express vector. The sequence of the entire clone was obtained using rapid amplification of the cDNA ends. A search of the clone sequence against GenBank revealed that it related to ribosomal protein L23 and it had a high percentage similarity to this protein from different species. A conserved domain for ribosomal protein L23 was also identified in the clone sequence.

This study presents initial evidence for the requirement of both chemical and mechanical stimuli to discharge polar capsules of Myxobolus cerebralis actinospores, the causative agent of salmonid whirling disease. The obligate need for combined discharge triggers was concluded from data obtained in a before/after experimental set-up carried out with individual locally immobilized actinospores. Homogenized rainbow trout mucus as chemostimulus and tangency of the apical region of the spores to achieve mechanical stimulation were applied subsequently. The actinospores showed discharged polar filaments exclusively when mucus substrate application was followed by touching the polar capsule-bearing region, but not when either stimulus was offered solely to the same individuals. We measured filament discharge rates to mucus preparations in a microscopic assay using supplementary vibration stimuli to ensure mechanical excitation. The actinospores responded similarly to different frequencies, which suggested a touch-sensitive recognition mechanism. Discharge specificity for salmonid mucus could not be confirmed, as mucus of common carp and bream could trigger similar filament expulsion rates. To a lesser extent homogenized frog epidermis and bovine submaxillary mucin could also stimulate the attachment reaction. In contrast, mucus of a pulmonate freshwater snail elicited no response.