BUSCADOR RECOLECTA

LC50 tests: Naturally infected and uninfected adults of A. parthenogenetica were collected with a plankton net (0.5 mm) within the Odiel saltpan complex. Sampling was carried out on two dates (14th of April and 15th of May 2014). Toxicity experiments were conducted after 24 h of acclimation of the Artemia at 100 g/l salinity. Two experiments were carried out. The first (with Artemia collected on 14th of April) was conducted at 25ºC. The second (Artemia collected on 15th of May) was conducted at both 25 and 29ºC. Median lethal concentration (LC50) was used to quantify Aresenic toxicity in infected and uninfected adult Artemia. Ten concentrations of As between 5 and 140 mg/l were used for the experiments at 25ºC (0, 5, 20, 35, 50, 65, 85, 95, 110, 125, 140 mg/l), and ten between 4 and 67 mg/l were used for the experiment at 29°C (0, 4, 11, 18, 25, 32, 39, 46, 53, 60, 67 mg/l) in order to estimate the LC50. Three replicates were used per concentration, with each replicate made up of a group of 10 individuals. Prevalence (P) and mean abundance (MA) were calculated separately for the “infected group” on both dates. Oxidative stress analysis and lipid droplet quantification: Artemia were sampled in May 2015 from Odiel salt ponds. A subsample was used to characterize the exact parasite composition (n = 60 infected individuals) and quantify the number of lipid droplets (n = 20 infected and 20 uninfected Artemia). The numbers of cysticercoids, prevalence, mean abundance and mean intensity were calculated for each cestode species. The number of lipid droplets was estimated according to Wurtsbaugh & Gliwicz 2001 [86]. We quantified lipid levels by inspecting individuals at 30x magnification and counting the number of lipid droplets along the right side of the 5th and 6th segments of the body. The rest of the specimens were acclimated during 24h to the experimental salinity with continuous aeration and fed ad libitum with lyophilized Tetraselmis chuii algae. The toxic concentrations of 4.69 mg/l As was selected on the basis of preliminary LC50 tests. Infected and uninfected A. parthenogenetica of the same size range were allocated to 1L glass vials (100 individuals per vial) with 600 ml of experimental solution (either control (no toxic solution) or 4.69 mg/L As) during 24h at 25 ºC (12:12 photoperiod) without food. After 24h exposure, individuals were stored at ‒80℃ until biochemical analysis. We performed the biochemical analysis on pools of 20 individuals per treatment. Number of replicates varied from 1 to 12 depending on Artemia availability. The different biomarkers were determined in the whole soft tissues after homogenization and centrifugation. We quantified five parameters as a proxy for oxidative status of Artemia: activity of four enzymes (catalase CAT, superoxide dismutase SOD, glutathione peroxidase GPx and glutathione reductase GR) and lipid peroxidation levels (thiobarbituric acid reactive substances TBARS). Total protein content in the supernatant fluid was determined following a standard Bradford’s procedure. Enzyme activity was determined colorimetrically., Peer reviewed

Artemia franciscana vs A. parthenogenetica. Artemia sampling was conducted during November 2012 in two different salt pan complexes located along the Atlantic coast of South West Spain. Native A. parthenogenetica was collected from Odiel saltpans (Huelva 37°15'29"N, 6°58'25"W). Invasive A. franciscana were collected from La Tapa saltpans (Cádiz Bay 36º35’52”N, 06º13’07”W). In each locality, Artemia were collected from two separate evaporation ponds with distinct salinities (90 and 145 g/l). Artemia was sampled using a 0.5 mm mesh net, then they were immediately transported to the laboratory and transferred to plastic tanks containing aerated water from the same pond, and subjected to a natural photoperiod. Feeding rates were quantified at two salinities (90 and 145 g/l) and two temperatures (15 and 24 ºC). Prior to measuring feeding, individuals were acclimated for 12 hours in climatic chambers under experimental conditions (15ºC-90g/l, 15ºC-145g/l, 24ºC-90g/l or 24ºC-145g/l). Individuals were assigned to the experimental salinity that matched what they were exposed to in the field. We then selected 48 adult brine shrimps of similar size for each temperature-salinity treatment. We transferred them to Petri plates containing filtered (0.45 μm) and autoclaved water from the pond at the same salinity and temperature during 1h without food, in order to increase feeding motivation. To measure feeding rates, brine shrimps were placed individually into multi-well plates filled with 2.5 ml of freeze-dried green algae Tetraselmis chuii (EasyAlgae®, Spain) solution (algal concentration 0.2 mg/ml) and placed in climatic chambers at 24ºC and 15ºC. We prepared control and blank samples in triplicate for each treatment. Artemia individuals grazed during 4 hours under continuous light conditions. During this period, we gently agitated the water every 30 min with a plastic Pasteur pipette, to avoid food particle sedimentation at the bottom of the plates. At the end of the experiment we collected 1ml from each well and counted remaining algal particles using an EasyCyte Plus System flow cytometer (Guava Express Plus software). The number of consumed cells was calculated by subtracting the final number of cells from the initial number. For triplicate controls and blanks, samples of 1 ml were taken and counted before and after the experiment. Brine shrimps were anaesthetized with carbonated water before being mounted (sacrificed) in a temporary glycerol mount and examined under the microscope to confirm that no cestode parasites were present. We measured the length of each individual from the end of the abdomen (furca) to the top of the head using a stereomicroscope coupled with a videocamera (Axiovision software). Infected vs uninfected A. parthenogenetica. Artemia sampling was conducted at Odiel saltpans in spring 2013, from ponds of intermediate salinity where the prevalence of the cestodes Flamingolepis liguloides (hereafter FL), Anomotaenia tringae (AT, a shorebird parasite) and Confluaria podicipina (CP) was high (FL: pond E15 at 130g/l on 02/05/2013; AT and CP: pond E18 at 170g/l on 04/06/2013). We used the above experimental setup but with fixed temperature and salinity conditions. We conducted two independent experiments on the above dates, one for FL and another for CP-AT under similar conditions. Experiments were carried out at 130g/l salinity with a 0.2mg/l concentration of Tetraselmis chuii, calculating feeding rates as described above. Individuals collected from pond E18 were first acclimated to the experimental salinity for 12h. After 4 hours, all individuals were measured and their parasitic status confirmed as described above. Parasite identification followed Georgiev et al., 2005. For the first experiment we used 43 non-infected (hereafter NI) individuals and 55 infected with FL; for the second experiment we used 40 NI Artemia, 14 infected with CP and 27 with AT., Field study Temporal variation of chlorophyll-a concentration in relation to Artemia density and species. Samples of A. parthenogenetica (from Odiel) and A. franciscana (from La Tapa) were collected monthly (from April to December 2011) from three to four ponds of different salinity, by sweeping water at each point during 15 seconds from the entire water column (15-30 cm depth) using a net of 0.1 mm mesh. Given the highly patchy distribution of Artemia in the field, 10 to 20 points were selected at random from different parts of the pond including the center and shoreline. In some ponds there were no Artemia, for reasons that are unclear but are likely to include the abundance of predators such as fish at low salinities. At each pond, we measured temperature and salinity (with a refractometer) and collected unfiltered water samples for analysis of concentrations of chlorophyll-a (as a measure of phytoplankton abundance) and nutrients. Total nitrogen concentration (Total N) was measured by digestion with potassium persulfate (Sims, Ellsworth & Mulvaney, 1995). Total phosphorus concentration (Total P) was measured by the phosphomolybdate method (APHA, 1995). Chlorophyll-a analysis was performed by spectrophotometry using the trichromatic method (Strickland and Parsons 1968). Total (including all developmental stages: metanauplii, juveniles and adults) and adult Artemia density were determined in the laboratory. Spatial variation in chlorophyll-a concentration and turbidity in relation to A. parthenogenetica density and parasite prevalence. On 23/04/2013 we sampled A. parthenogenetica (at Odiel) by filtering 20 l through a 0.5 mm mesh net, at nine different ponds covering a wide range of salinities (75-235 g/l, S1 Table). Samples of water (1 l) were taken for chlorophyll-a analysis, following the above procedure. Salinity was measured with a refractometer and turbidity with a Snell tube (a modified Secchi disc suitable for shallow waters). In the laboratory the density of Artemia (adult density, plus total density including metanauplii and juveniles) as well as the total biomass (dry mass after 24h at 50ºC) was determined. Adult individuals (n = 100) from each pond were then randomly selected for calculation of parasite prevalence (using the above methods) so as to explore the effect of parasite infection on chlorophyll-a concentration and turbidity., Peer reviewed

All the ponds within the Veta la Palma system were surveyed monthly in 2015 by experienced observers as part of a monitoring program carried out by the Doñana Biological Station (EBD-CSIC, http:// www-rbd.ebd.csic.es/Seguimiento/seguimiento.htm). Owing to staff vacations, surveys did not take place in August. All aquatic birds present were identified to species and counted. We excluded raptors, corvids and other non-aquatic species from analyses. Every pond was visited on foot in a single day and surveyed from one or more positions along the shoreline with binoculars (8 x 50) and a telescope (20 x 60). Birds flying overhead were excluded. For each survey, the observers also identified the water level. Greater flamingos Phoenicopterus roseus are particularly abundant in the ponds (Rendón et al. 2008), and their behaviour also helped in the estimation of the water level as follows: 0 totally dry; 1 almost dry, with some dry regions in the pond and some shallow water; 2 low water level, roughly half way to the tibio-tarsal articulation of flamingoes; 3 medium level, approximately to the tibio-tarsal articulation; 4 high level, above the articulation but not exceeding total leg length, Peer reviewed