Cumulative Metabolic and Epigenetic Effects of Paternal and/or Maternal Supplementation with Arachidonic Acid across Three Consecutive Generations in Mice

Abstract

Apart from the known associations between arachidonic acid (AA), weight gain, and neurological and immune function, AA exposure leads to alterations in global and gene-specific DNA methylation (DNAm) and fatty acid (FA) content in human cultured cells. However, it is unknown as to whether the latter effects occur in vivo and are maintained over extended periods of time and across generations. To address this issue, we asked whether AA supplementation for three consecutive generations (prior to coitus in sires or in utero in dams) affected offspring growth phenotypes, in addition to liver DNAm and FA profiles in mice. Twelve-week-old BALB/c mice were exposed daily to AA dissolved in soybean oil (vehicle, VH), or VH only, for 10 days prior to mating or during the entire pregnancy (20 days). On average, 15 mice were supplemented per generation, followed by analysis of offspring body weight and liver traits (x average = 36 and 10 per generation, respectively). Body weight cumulatively increased in F2 and F3 offspring generations and positively correlated with milligrams of paternal or maternal offspring AA exposure. A concomitant increase in liver weight was observed. Notably, akin to AA-challenged cultured cells, global DNAm and cis-7-hexadecenoic acid (16:1n-9), an anti-inflammatory FA that is dependent on stearoyl-CoA desaturase 1 (SCD1) activity, increased with milligrams of AA exposure. In accordance, liver Scd1 promoter methylation decreased with milligrams of germline AA exposure and was negatively correlated with liver weight. Our results show that mice retain cellular memories of AA exposure across generations that could potentially be beneficial to the innate immune system.

Keywords: DNA methylation; arachidonic acid; cis-7-hexadecenoic acid; cumulative; growth; stearoyl-CoA desaturase 1; transgenerational.

Figure 1

Mouse supplementation experiments. Three-month-old mice were supplemented with AA dissolved in vehicle (VH; soybean oil), or VH only, maternally (MAA and MVH) or paternally (PAA and PVH, respectively) for three consecutive generations (F0, F1, and F2). Administration of AA and VH supplements were performed within similar time frames. The effect of either FA supplement was subsequently analyzed in one-month-old F1, F2, and F3 offspring of AA- or VH-exposed dams or sires.

Figure 2

Cumulative germline and/or somatic offspring AA exposure of paternally and maternally supplemented mice. Filled arrows indicate total mg of AA supplement (calculated by multiplying mg of daily supplement by days of exposure) administered at each generation (F0, F1, and F2); dotted arrows indicate milligrams of germline or soma AA exposure (circles and ovals, respectively) of F1, F2 and F3 offspring generations. Cumulative AA exposure somatic and/or germline offspring exposure at each generation are indicated below each mouse generation. Similar calculations were performed of paternal and maternal vehicle supplements.

Figure 3

Offspring body weight of mice supplemented paternally or maternally for three consecutive generations. Horizontal lines indicate significant differences between each experimental group; ^bl p < 0.1, * p < 0.05, *** p < 0.001; supplemental period and n of each group is indicated under each bar graph; PAA and MAA: paternal and maternal AA supplementation, respectively; PVH and MVH: paternal and maternal VH supplementation, respectively.

Figure 4

Body weight of three-month-old dams at supplemental day 1. Y-axis indicates dam body weight at supplemental day 1: day 1 of pregnancy. PAA and PVH dams: non-supplemented dams that were crossed to AA- or VH-supplemented sires. ^bl p < 0.1, *** p < 0.001. Other symbols are as in Figure 3.

Figure 5

Correlations between milligrams of parental AA exposure or liver weight and offspring liver FA content. Diagram of SFA, MUFA, and PUFA biosynthesis pathways highlighting individual FA that showed significant (p < 0.05) or borderline significant (p < 0.1) correlations with both germline and germline + somatic PAA+MAA exposure (Pearson’s r). Large white boxes indicate the analyzed FA; FA in italics were not detectable. Squares with solid and dotted filling indicate significant and borderline significant correlations with AA exposure (grey) or liver weight (grey with black dots); “+” and “−“ indicate positive and negative correlations, respectively; promoter methylation was analyzed on the genes indicated in ovals; other abbreviations: SFA, MUFA, and PUFA, saturated, mono-, and polyunsaturated FA, respectively; MA, myristic acid; PA, palmitic acid; SA, stearic acid; ARA, arachidic acid; PAOA, palmitoleic acid; OA, oleic acid; ECA, eicosanoic acid; HDA, cis-7-hexadecenoic acid; LA, linoleic acid; ALA, alpha-linolenic acid; EA, elaidic acid; AA, arachidonic acid; EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid; DGLA, dihomo-γ-linolenic acid; Fads2, fatty acid desaturase 2; Scd1, stearoyl-CoA desaturase 1.

Figure 6

Correlations between Scd1 and Ppargc1a promoter methylation and expression of F2 offspring of AA- and VH-exposed mice. Each data point represents average normalized expression values of n = 6 mice (3 male and 3 females); Pearson’s r values are indicated in the upper left corner of each graph. Other abbreviations are as in Figure 3.