Harmful algal blooms (HABs) release toxic compounds in water and are increasing in frequency worldwide. The neurotoxin β-methylamino-L-alanine (BMAA) is released by HABs and has garnered much attention over the past twenty years due to its association with human neurodegenerative disorders, but its effects on wildlife are still largely unknown. This study characterized the effects of chronic exposure to environmentally relevant concentrations of BMAA on the behavior and brain size of developing zebrafish (Danio rerio). Zebrafish were continuously exposed to 0, 1, 10, or 100 μg/L waterborne BMAA between 0- and 5-days post-fertilization (dpf) before the onset of exogenous feeding. At 5 dpf, locomotion and responses to vibrational and visual stimuli were assessed. Following behavioral testing, larvae body and brain size were measured. Survival between 0 and 5 dpf did not differ between treatments. Moreover, BMAA exposure did not affect thigmotaxis, startle response magnitude, habituation to repeated presentation of vibrational startling stimuli, or relative brain size. A moderate increase in overall activity was observed in larvae exposed to 10 ug/L BMAA under light, but this effect was not seen in dark conditions, indicating that visual processing may have been affected by chronic BMAA exposure. Thus, our results show that passive continuous exposure to environmentally relevant concentrations of BMAA prior to first feeding in zebrafish does not affect overall brain development, locomotion, anxiety, and motor neuron-mediated reflexes, but suggest targeted neurotoxicity within the visual system.

Zebrafish were exposed to four BMAA treatments: 0 μg/L (control), 1 μg/L, 10 μg/L, and 100 μg/L BMAA within one hour after spawning. Mortalities were recorded for each group at 6-8 hours post fertilization (hpf), 24 hpf, and every morning thereafter until 5 days post fertilization (dpf). At 4 dpf, 6 larvae per treatment group were transferred into 24 multi-well plates for acclimation before testing for behaviour at 5 dpf using a DanioVision observation chamber (Noldus Information Technology, Wageningen, The Netherlands). Larval behavior was recorded with a Basler Ace camera with Gigabit Ethernet interface and infrared filter (1280 × 1024 resolution). The observation chamber was under constant infrared illumination, with the camera output fed into a standard PC system with Ethovision software (Noldus Information Technology). Ethovision was used to detect objects that were darker than the background, with a sensitivity setting of 113, and minimum object size of 5 pixels. The tracking rate was 25 samples/s. To minimize noise, input filters were applied. A minimum displacement of 0.2 mm was required before movement was registered between samples, and data was considered “missing” if >15 mm of movement was recorded between samples. Various larval behaviours were analyzed.

Following behavioral measurements, a subset of euthanized larvae was fixed with 4% paraformaldehyde at 4°C for 24 hrs before cryoprotection and sectioning. Prior to sectioning, larvae embedded in 4.4% agarose blocks were imaged to determine their standard length (mm) using the segmented line tool in ImageJ (Schindelin et al., 2012). The head of each larva was cut into 20 μm thick coronal sections and mounted onto slides. Micrographs of each section were taken using a Nikon 90i microscope and the area (mm²) of the brain in each section was determined by contour tracing using ImageJ. To estimate brain volume, the olfactory bulbs served as the rostral limit of the brain, while the first wrapping of developing cartilage of the vertebral column around the spinal cord demarcated the caudal endpoint of the brain. The volume (mm³) of each brain segment was determined by multiplying brain area by twice the section thickness (40 μm) to account for the use of alternate sections. The volumes of the individual segments were then summed to get the total brain volume (mm³), which was an estimate of absolute brain size. We used the residuals calculated from the linear relationship between standard length (mm) and brain volume (mm³) to represent relative brain size.