Table 1.
Summary of diets employed during the feeding trial.
For the live algae treatments, media used in the culture of microalgae strains and relative CCAP codes are reported.
Table 2.
Quaternary gradient used during NP-HPLC separation of spat TLE.
Mob A: Isooctane:Ethyl Acetate (99.8:0.2); Mob B: Acetone: Ethyl Acetate (2:1) + 25 mM GAA; Mob C: IPA: Water (85:15) + 15mM GAA and 7.5 mM TEA; Mob D: Isopropanol.
Table 3.
Binary gradient used during LC-MS analysis of spat TLE.
Mob A: Water + 10 mM Ammonium formate + 20 mM Formic acid; Mob B: IPA:ACN (9:1) + 10 mM ammonium formate + 20 mM formic acid.
Fig 1.
Non metric multidimensional scaling (nMDS) analisis and similarity percentages (SIMPER) analysis for relative (A-B) and absolute (C-D) fatty acid composition of diets. A: nMDS plot of the relative FA composition of diets (% total FA) provided during the feeding trial. Single FA loadings are stacked on the plot and reported in grey. B: principal variables explaining for a cumulative 75% of group differences in relative diet FA composition data evidenced by SIMPER analysis. C: nMDS plot of the absolute (μgFA mgDW-1) fatty acid composition of the diets. Single FA loadings are stacked on the plot and reported in grey. D: principal variables explaining for a cumulative 75% of group differences in absolute diet FA composition data evidenced by SIMPER analysis. Three replicates (n = 3) for each diet are here reported Charts B-D: data are reported as average (bar chart) ± 95% confidence interval; individual observations are jittered on the chart (smaller dots). Full data including all the FA observed and details on statistical significance values are found in S2 Table.
Fig 2.
Spat growth performances when subjected to the different diet treatments.
A: variation of shell length in mm between in each group between the beginning of the trial (T0) and end of the experiment (T28). B: Size increase of the spat (reported in mm) during the diet trial when subjected to different diets. C: live weight per spat (reported in mg) between the beginning of the trial (T0) and the end of the trial (T28 D: weight increase per spat throughout the trial according to the different sample groups (reported in mg). SP: spat subjected to Shellfish Paste, CYL: spat subjected to C. fusiformis 1017/2, ISO: spat subjected to I. galbana 927/1, MONO: spat subjected to M. subterranean 848/1, NANNO: spat subjected to N. oceanica 849/10, OUT: Outdoor deployed spat. A: N = 90 individual spat per group, marker in each box plot indicates the group average, band inside the box indicates the group median, box includes observation between 1st (25th percentile) and 3rd (75th percentile) quartile, whiskers represent values between ±1.5 * interquantile range (IQR), observations outsite over ±1.5*IQR are reported as outliers (black dots); violin shapes represent the variable distribution in each sample group. Figs B-C-D: N = 9, histograms represent the average ein each group and errobars the confidence interval for 95% of observations, single observation are jittered as small dots inside each histogram. Letters indicate significance levels: a: p>0.05; b: p<0.05; c: p<0.01; d p<0.001.
Fig 3.
Non metric multidimensional scaling (nMDS) analisis and similarity percentages (SIMPER) analysis for relative (A-B) and absolute (C-D) fatty acid composition of spat subjected to the diet treatments. A: nMDS plot of the relative FAME composition of spat (%FA) subjected to the feeding trial. Single FA loadings are stacked on the plot and reported in grey. B: principal variables explaining for a cumulative 75% of group differences in relative spat FA composition data evidenced by SIMPER analysis. C: nMDS plot of the absolute (μgFA mgashfreeDW-1) fatty acid composition of the spat. Single FA loadings are stacked on the plot and reported in grey. D: principal variables explaining for a cumulative 75% of group differences in absolute spat FA composition data evidenced by SIMPER analysis. T0: Spat sampled before the beginning of the trial, SP: spat fed with ShellPaste during the 4 weeks diet trial; CYL: spat fed with C. fusiformis 1017/2 during the 4 weeks diet trial; ISO: spat fed with I. galbana 927/1 during the 4 weeks diet trial; MONO: spat fed with M. subterranean 848/1 during the 4 weeks trial; NANNO: spat fed with N. oceanica 849/10 during the 4 weeks trial; OUT: spat deployed outdoor and sampled after 4 weeks. Three replicates (n = 3) for each sample group are here reported. Charts B-D: data are reported as average (histogram) ± 95% confidence interval; individual observations are jittered on the chart (smaller dots). The complete data for FA analysis of spat and statistical significance is provided in S3 Table.
Fig 4.
Non metric multidimensional scaling (nMDS) analisis and similarity percentages (SIMPER) analysis for relative (A-B) and absolute (C-D) lipid class composition of spat subjected to the diet treatments. A: nMDS plot for relative lipid class composition of the spat (%TLE) subjected to the feeding trial. Single lipid class loadings are stacked on the plot and reported in grey. B: principal variables explaining for a cumulative 75% of group differences in relative spat lipid class composition data evidenced by SIMPER analysis. C: nMDS plot of the absolute (μg mgashfreeDW-1) lipid class composition of the spat. Single lipid class loadings are stacked on the plot and reported in grey. D: principal variables explaining for a cumulative 75% of group differences in absolute spat lipid class composition data evidenced by SIMPER analysis. T0: Spat sampled before the beginning of the trial; SP: spat fed with Shellfish Paste during the 4 weeks trial; CYL: spat fed with C. fusiformis 1017/2 during the 4 weeks diet trial; ISO: spat fed with I. galbana 927/1 during the 4 weeks diet trial; MONO: spat fed with M. subterranean 848/1 during the 4 weeks trial; NAN: spat fed with N. oceanica 849/10 during the 4 weeks trial; OUT: spat deployed outdoor and sampled after 4 weeks. Three replicates (n = 3) for each sample group are here reported. Charts B-D: data are reported as average (histogram) ± 95% confidence interval; individual observations are jittered on the chart (smaller dots).The complete data for lipid class analysis and statistical significance values are found in S3 Table.
Fig 5.
Partial least squares discriminant analysis (PLS-DA) plots of untargeted lipidomics data of the spat diet groups.
A: Positive ionization mode. B: Heatmap plot reporting the group averages for Variable of important in projection (VIP) scores >1 evidenced from POS lipidomics data. C: Negative ionization mode. D: Heatmap plot reporting the group averages for Variable of important in projection (VIP) scores >1 evidenced from NEG lipidomics data. Heatmaps were plotted via ‘pheatmap’ package [72]. Main clusters evidenced by the elbow method (S8C Fig) are shown as breaks in the heatmap plot and are named as +/- (for POS/NEG) C and the relative cluster number (1 to 7). Euclidean distance was used as a distance measure, Ward as clustering algorithm. Lipids are reported as class, n° carbon and n° of double bonds (e.g. TG.58.10). Colour coding for lipid expression from Blue (Low) to red (High). For a representation of raw intensity for VIP selected lipids refer to S8D Fig.
Table 4.
Spearman rank correlation coefficients of lipids highlighted from the dataset analysed in this study and spat live weight increase (WI).
The reported p-values are adjusted for multiple comparisons [59].