Diatoms of the genus Pseudo-nitzschia produce domoic acid (DA), a toxin that is vectored in the marine food web, thus causing serious problems for marine organisms and humans. In spite of this, knowledge of interactions between grazing zooplankton and diatoms is restricted. In this study, we examined the interactions between Calanus copepodites and toxin producing Pseudo-nitzschia. The copepodites were fed with different concentrations of toxic P. seriata and a strain of P. obtusa that previously was tested to be non-toxic. The ingestion rates did not differ among the diets (P. seriata, P. obtusa, a mixture of both species), and they accumulated 6%–16% of ingested DA (up to 420 µg per dry weight copepodite). When P. seriata was exposed to the copepodites, either through physical contact with the grazers or separated by a membrane, the toxicity of P. seriata increased (up to 3300%) suggesting the response to be chemically mediated. The induced response was also triggered when copepodites grazed on another diatom, supporting the hypothesis that the cues originate from the copepodite. Neither pH nor nutrient concentrations explained the induced DA production. Unexpectedly, P. obtusa also produced DA when exposed to grazing copepodites, thus representing the second reported toxic polar diatom.

Members of the marine dinoflagellate genus Alexandrium are known to exude allelochemicals, unrelated to well-known neurotoxins (PSP-toxins, spirolides), with negative effects on other phytoplankton and marine grazers. Physico/chemical characterization of extracellular lytic compounds of A. tamarense, quantified by Rhodomonas salina bioassay, showed that the lytic activity, and hence presumably the compounds were stable over wide ranges of temperatures and pH and were refractory to bacterial degradation. Two distinct lytic fractions were collected by reversed-phase solid-phase extraction. The more hydrophilic fraction accounted for about 2% of the whole lytic activity of the A. tamarense culture supernatant, while the less hydrophilic one accounted for about 98% of activity. Although temporal stability of the compounds is high, substantial losses were evident during purification. Lytic activity was best removed from aqueous phase with chloroform-methanol (3:1). A “pseudo-loss” of lytic activity in undisturbed and low-concentrated samples and high activity of an emulsion between aqueous and n-hexane phase after liquid-liquid partition are strong evidence for the presence of amphipathic compounds. Lytic activity in the early fraction of gel permeation chromatography and lack of activity after 5 kD ultrafiltration indicate that the lytic agents form large aggregates or macromolecular complexes.