Table 4.

Normalized serum Lyso-PL levels that were significantly altered in the dyslipidemic groups.

Compound name		C-4d	HFD-4d	P407-4d	C-30d	HFD-30d	P407-30d	2-way ANOVA (I)
								p	p (Corr)
Lyso-PE	(20:4)	−0.33 ± 0.05a	0.04 ± 0.06bc	0.12 ± 0.11bc	−0.18 ± 0.08ac	0.18 ± 0.08b	0.32 ± 0.10b	<0.001	<0.001
	(18:2)	−0.17 ± 0.08a	−0.06 ± 0.03ac	0.28 ± 0.12bc	−0.19 ± 0.10a	−0.02 ± 0.09ac	0.43 ± 0.14b	<0.001	<0.001
	(18:3)	−0.10 ± 0.18a	−0.22 ± 0.09a	0.36 ± 0.17ab	−0.23 ± 0.19a	−0.30 ± 0.14a	0.72 ± 0.19b	<0.001	<0.001
	(22:6)	−0.43 ± 0.10a	−0.08 ± 0.10ab	0.05 ± 0.17ab	0.10 ± 0.11b	−0.01 ± 0.08ab	0.81 ± 0.11c	<0.001	<0.001
	(16:1)	−0.26 ± 0.26a	−0.25 ± 0.13a	0.13 ± 0.24ab	0.23 ± 0.17ab	−0.20 ± 0.14a	0.95 ± 0.21b	0.001	0.02
	(16:0)	−0.14 ± 0.11a	−0.13 ± 0.09a	0.00 ± 0.21ab	0.06 ± 0.17ab	−0.34 ± 0.20a	0.67 ± 0.18b	0.005	0.09
Lyso-PC	(20:2)	−0.60 ± 0.17a	0.24 ± 0.10bc	−0.58 ± 0.22a	−0.09 ± 0.19ac	0.75 0.19b	−0.16 ± 0.21ac	<0.001	<0.001
	(20:1)	−0.08 ± 0.16	0.35 ± 0.10	−0.37 ± 0.22	−0.30 ± 0.12	0.27 ± 0.24	−0.29 ± 0.21	0.003	0.08

The levels of each metabolite were transformed to a base-2 logarithm, normalized to the internal standard and to the mean levels of the C-4d samples. The data are presented as the means of the normalized abundances ± SEM (n = 9–10). The two-way ANOVA cut-off point was calculated using the Benjamini-Hochberg correction to avoid false positives. The p-values for the main factor ‘intervention’ without (p) and with correction (p(Corr)) are shown. I: the effect of the pro-dyslipidemic intervention (HFD and P407 administration). When one-way ANOVA was also significant (p < 0.05), the post hoc analysis was used. In each row of the table, different superscript lowercase letters (^a,b,c) indicate significant different mean values (Tukey’s test, p < 0.05).