Refer to the Methods section of Griffin, et al. (2019).

Culture and culturing condition:
All assays were run with the dinoflagellate Alexandrium catenella strain BF-5, a highly toxigenic strain isolated from the Bay of Fundy, Canada. Cells were cultured in F/2 medium (Guillard, 1975) without silicate. Batch cultures were kept under exponential-phase growth at a temperature of 18 °C, with a photoperiod of 12 h of light and 12 h of darkness. Fluorescent lights were used at an intensity of ~50 µM m⁻² s⁻¹.

Nitrogen sources:
These assays simulated potential bias due to N-sources (ammonium and urea) excreted by the copepod grazer Acartia hudsonica. Treatments consisted of daily additions of either 0.083 μmol L⁻¹ ammonium, 50 μmol L⁻¹ ammonium, or 50 μmol L⁻¹ urea over a 3 day period. The lower ammonium concentration (0.083 μmol L⁻¹) is an estimate of the nitrogen excreted per day from the 15 copepods in a bottle. The higher concentration of ammonium (50 μmol L⁻¹) represents the ammonium concentration in K-medium (Keller et al., 1987). The F/2 medium in nutrient-replete conditions contains nitrate as its N-source. Urea was used at a concentration of 50 μmol L⁻¹, which is far higher than could result from copepod excretion. The controls consisted of cells grown in F/2 medium (880 µmol L⁻¹ nitrate and 36.3 µmol L⁻¹ phosphate) or 0.22 μm filtered seawater (FSW) collected from Long Island Sound (~3 µmol L⁻¹ nitrate and ~1 µmol L⁻¹ phosphate).

Alarm cues:
Crushed cells of the highly toxigenic Alexandrium catenella (strain BF-5), the low toxigenic A. catenella (strain GTCN-16), Tetraselmis sp., and Thalassisosira weissflogii were added to separate experimental bottles. Added cells were crushed via sonic dismembrator on ice. Complete lysis of cells was confirmed by microscopic examination. 140 μgC of crushed cells for each species tested were added daily to bottles during the three-day period incubations. Cell-to-carbon conversion factors used were 2.7 x 10^-3 μgC per Alexandrium cell, 8.0 x 10^-5 μgC per T. weissflogii cell, and 5.1 x 10^-5 μgC per Tetraselmis sp. cell desired quantity of carbon.

Grazer exposure:
Grazer assays measured the combined effects of copepod kairomone and other feeding-related cues on cell toxin production. Treatments consisted of additions of 15 adult female copepods (Acartia hudsonica) collected from Maine, USA, which have a history of exposure to A. catenella blooms. Copepods had been kept in culture conditions similar to the experimental algae, and grown on a mixed diet of nontoxigenic phytoplankton. At the conclusion of the grazing assays, contents of the bottles were first passed through a 63 μm mesh to separate grazers, eggs and fecal pellets from the Alexandrium catenella cells. The filtrate was then passed through a 10 μm mesh, which collected the washed A. catenella cells, which were then checked to ensure the absence of cells and resuspended in 0.22 µm-FSW in 50 mL centrifuge tubes.

Cell concentration and toxin analysis:
In the nitrogen source and alarm cue assays, which did not involve use of copepods, there was no need to pass cells through the 63 μm mesh before cells were collected on the 10 μm mesh. A fraction of the contents of each centrifuge tube was preserved using Lugol’s solution, and two subsamples were counted under an inverted microscope for cell abundance. In the remaining fraction, cells were centrifuged at 4000 × g for 20 minutes. The supernatant was carefully discarded and the cell pellet was resuspended in 0.1 M acetic acid. Cells were then lysed using a sonic dismembrator on ice. The solution was centrifuged and the extract was filtered through a 0.45 μm ultracentrifuge cartridge to remove the cell particles. The supernatant was stored at -80°C to prevent chemical reactions from occurring during the period before processing. After toxin extraction, toxin analysis by High Performance Liquid Chromatography (HPLC) with fluorescent detection was used to determine the total toxin content.