Previous studies indicated that a-linolenic acid (ALA) influences perinatal brain development. However, it is not clear whether these outcomes were the result of ALA’s direct influence upon brain cells, or if ALA-related alterations were mediated by other elongation and desaturation omega-3 species. The aim of this study was to establish, for the first time to our knowledge, whether ALA administered to mouse neural stem cells (NSC) in vitro directly alters cell cycle, apoptosis, global DNA methylation, and the expression of DNA methyltransferases Dnmt1 and Dnmt3a. NSC exposed for 96 hours to 100 µM ALA exhibited alterations of cell-cycle phases G0 /G1 and S, reduced apoptosis (1.27 IOD± 0.09 SE ALA vs 1.93 IOD ± 0.22 SE control, CT), and increased Dnmt1 and Dnmt3a gene (2.17 ratio for Dnmt1and 1.29 ratio for Dnmt3a as compared to CT, respectively), and protein expression. However, ALA exposure did not alter global DNA methylation levels (% 5mC). In conclusion, ALA exerted a direct effect upon mouse NSC by altering cell cycle phase distribution and apoptosis. In addition, the exposure of NSC to ALA for 96 hours also increased transcript and protein levels of Dnmt1 and Dnmt3a, while no alteration in global DNA methylation was noted.

•    a-linolenic acid
•    Brain
•    DNA methylation
•    Apoptosis
•    Cell cycle

Niculescu MD (2014) Alpha-Linolenic Acid Alters Cell Cycle, Apoptosis, and DNA Methyltransferase Expression in Mouse Neural Stem Cells, but not Global DNA Methylation. J Hum Nutr Food Sci 2(1): 1026.

ALA: α-Linolenic Acid; NSC: Neural Stem Cells; Dnmt1: DNA methyltransferase 1; Dnmt3a: DNA methyltransferase 3a; 5mC: 5-methylcytosine

The role of omega-3 fatty acids in perinatal development has been extensively studied (reviewed in [1]). Of special interest are the roles of omega-3 fatty acids in brain development, cognition and brain aging as discussed elsewhere [2,3]. However, most of these studies have focused on the beneficial effects of three most abundant omega-3 species (eicosapentaenoic acid, EPA; docosapentaenoic acid, DPA; and docosahexaenoic acid, DHA), while only limited studies investigated the role of the omega-3 precursor α-linolenic acid (ALA) in perinatal brain development.

We have previously reported that the perinatal administration of ALA during gestation and lactation not only altered the distribution of omega-3 and omega-6 fatty acids, as well as the epigenetic status in maternal and offspring livers [4,5], but also that the interplay between the ALA intakes during gestation and lactation induced specific changes in hippocampal development in the offspring [4]. However, it is not clear whether these outcomes were due to ALA’s direct action upon brain cells, or through its metabolic products (DPA, EPA, or DHA).

The aim of the present study was to determine whether the direct administration of ALA to mouse neural stem cells (NSC) in vitro induces alterations upon their cell-cycle, apoptosis, the expression of DNA methyltransferases, and global DNA methylation.

All reagents were purchased from Sigma-Aldrich (St. Louis, MO, USA) if not otherwise specified. Mouse NSCs were obtained and cultured as previously described [6]. In brief, the cells were cultured in Neurobasal medium (Invitrogen, Carlsbad, CA, USA) with B27 supplement without vitamin A (Invitrogen), 2 mM L-glutamine (Invitrogen), 100 U/ml penicillin-streptomycin (Invitrogen), 20 ng/ml murine EGF (Invitrogen), 20 ng/ml human βFGF (Invitrogen), and 2 mg/ml heparin (Invitrogen). At time 0, cells continued to be cultured as described above (control group, CT, n=4 independent samples per assay), or as described above plus 100 µM ALA (ALA group, n=4 independent samples per assay). At 48 hr, cells were either harvested for flow-cytometry, or continued to be cultured until 96 hours, in which case a second dose of ALA was added to the medium. At 96 hr the remaining cells were harvested and processed subsequently for either flowcytometry, DNA, RNA, or protein assessment.

Flow-cytometry was performed at 48 and 96 hr exposure at the UNC Flow Cytometry Core Facility, on a BC/Cytek FACSCaliber. Cell were prepared as described elsewhere [7]. For each sample at least 10,000 events were counted.

Immunoblotting was used for the assessment of apoptosis (anti-activated Caspase-3 rabbit antibody, Abcam, Cambridge, MA, USA) and DNA methyltransferase protein levels (anti-Dnmt1 C-17 goat antibody and anti-Dnmt3a H-295 rabbit antibody, Santa Cruz Biotechnology, Dallas, TX, USA). Horseradish-conjugated secondary antibodies were used for chemiluminescent detection, according to already published protocols [6]. Beta actin was used for normalization (Santa Cruz Biotechnology). The data were measured as previously described [8], and expressed as integrated optical density (IOD) units normalized to beta actin.

RNA and DNA extraction methods were performed using DNeazy and RNeazy kits (Qiagen, Valencia, CA, USA) according to manufacturer’s instructions. The assessment of gene expression was performed by real time RT-PCR, using commercially available Dnmt1, Dnmt3a, and 18S primers (Qiagen) and using dedicated kits and reagents according to manufacturer’s protocol. Data were retrieved and processed according to the -2??Ct method, as previously described [5]. Dnmt1 and Dnmt3a Ct values were normalized to 18S.

Global DNA methylation was assessed using reversed-phase HPLC as previously published [9]. Data were expressed as percent methylated cytosine (% 5mC) from total cytosine.

Statistical assessment.Flow-cytometry (% cells in each cellcycle phase) was assessed using ANOVA followed by TukeyKramer test for multiple comparisons. Gene expression was assessed using t-test after log2 transformation. Data obtained from immunoblotting assays and DNA global methylation was assessed using t-tests. All statistical calculations were performed using the JMP 10 software (SAS, Cary, NC, USA).

After 48 and 96 hours of exposure to either control medium (CT) or ALA-supplemented medium (ALA), the assessment of cellcycle phases revealed that ALA exposure increased the percent of cells in the G0 /G1 stage, while decreasing the percent of cells in the S phase. Specifically, the cells exposed to ALA had 63.63% ± 0.45 SE cells in G0 /G1 stage as compared to 56.00% ± 0.45 SE cell in the CT group at 48 hours. At 96 hours, the ALA group had 64.97% ± 1.01 SE cells in G0 /G1 as compared to 61.28% ± 0.79 SE in the CT group. These differences were statistically significant, as well as the difference between the percent of cells in the CT group at 48 and 96 hours, respectively (Figure 1a). Differences in the percent of cells in S phase were also present at 48 hour exposure time. ALA exposure decreased the percent of cell in S phase at 48 hr (26.32% ± 0.24 SE as compared to 33.37% ± 0.92 SE for CT group, and the difference was statistically significant (Figure 1b). However, no differences were present for the distribution of cells in the G2 /M phase, between ALA and CT groups (Figure 1c).

Figure 1: ALA induces alterations in cell cycle distribution. The assessment of the distribution of NSCs at 48 hr and 96 hr was performed using flow-cytometry. Columns marked with different letters denote statistical significance (p<0.05) as assessed by Tukey-Kramer test. Error bars indicate standard error (SE). ALA, ALA-treated cells; Control, cells without ALA supplementation to the medium.

For the rest of the assays, the 96 hr time-point was used. Apoptosis was measured using activated Caspase-3 by immunoblotting. ALA exposure of neural stem cells was associated with significantly less activation as compared to control (1.27 IOD ± 0.09 SE in ALA group vs. 1.93 IOD ± 0.22 SE in CT group, Figure 2).

Figure 2: ALA alters apoptosis in neural stem cells. Apoptosis was determined using immunoblotting on protein extracts from NSCs at 96 hr exposure to either 100 µM ALA (ALA), or control medium (CT). The picture inset shows the identification of activated Caspase-3 in all samples (n=4 per group). * denotes statistical significance (p<0.05) as determined by t-test. Error bars indicate standard error (SE). IOD:Integrated Optical Density (arbitrary units).

The assessment of Dnmt1 and Dnmt3a expression (Figure 3) revealed that ALA treatment induced overexpression of both genes. Dnmt1 expression was higher in the ALA group by 2.17 fold ± 0.17 SE as compared to control, while Dnmt3a expression was 1.30 fold ± 0.12 SE higher than control, and both differences were statistically significant.

Figure 3: Dnmt1 and Dnmt3 transcript levels are altered by ALA treatment after 96 hours. Gene expression for Dnmt1 and Dnmt3a was assessed by real time RT-PCR. Data are shown as ratio relative to the control group, after 18S normalization. *and** denote statistical significance (p<0.05 and p<0.01, respectively) versus control values for each gene, as determined by a t-test applied to log2 -transformed expression values. Error bars indicate standard error (SE).

A similar pattern was present for the protein levels of Dnmt1 and Dnmt3a (Figure 4), as assessed by immunoblotting. The ALA treatment of cells induced, at 96 hr exposure, higher Dnmt1 protein expression than in control (1.50 IOD ± 0.10 SE in ALA vs. 0.96 IOD ± 0.10 in CT). Similarly, Dnmt3a protein levels were 1.29 IOD ± 0.10 SE in ALA group vs. 0.62 IOD ± 0.07 SE in CT group).

Figure 4: Dnmt1 and Dnmt3 protein levels are altered by ALA treatment after 96 hours. Protein levels for Dnmt1 and Dnmt3a were assessed by immunoblotting. The picture inset shows the identification of Dnmt1, Dnmt3a, and β-actin in all samples (n=4 per group). Data are expressed as values normalized to β-actin. * and ** denote statistical significance (p<0.05 and p<0.01, respectively) versus control values for each protein, as determined by a t-test. Error bars indicate standard error (SE). IOD:Integrated Optical Density (arbitrary units).

The assessment of global DNA methylation (Figure 5) indicated no difference between the ALA and CT groups (3.22% ± 0.13 SE 5mC in ALA group vs. 3.23% ± 0.19 SE 5mC in CT group).

Figure 5: Assessment of global DNA methylation. Global DNA methylation was assessed by measuring the percent of 5 methylcytosine (% 5mC) as compared to total cytosine content in DNA, using reversed-phase HPLC. Statistical significance was determined using t-test. Error bars indicate standard error (SE). CT: Control group; ALA: ALA group (n=4/group)

This study investigated, for the first time to our knowledge, whether the direct exposure of mouse neural stem cells to ALA could induce alterations related to cell cycle progression, apoptosis, the expression of DNA methyltransferases, and global DNA methylation. When mouse NSCs were exposed to 100 µM ALA for 48 and 96 hours, the percent of cells in G0 /G1 was increased by the ALA exposure at both time points, while the percent of cells in S phase was decreased by ALA only at 48 hours. ALA treatment for 96 hours induced decreased apoptosis, and increased gene and protein expression of Dnmt1 and Dnmt3a. However, the ALA treatment did not alter global DNA methylation.

We have previously reported that ALA maternal intakes during gestation and lactation altered the brain development in pups [4]. Specifically, pups exposed to maternal ALA supplementation during lactation had increased proliferation and neuronal differentiation. However, the beneficial effects of such supplementation were offset by gestational ALA deficiency. We also indicated that maternal perinatal ALA deficiency increased the apoptosis in the hippocampus of pups. Moreover, we have also reported that the interplay between maternal ALA availability during gestation and lactation induced epigenetics alterations for Fads2 methylation in the livers of both mothers and pups, along with changes in the expression of DNA methyltransferases [5]. However, these studies were not able to determine whether these outcomes were the direct result of ALA supplementation, or as a consequence of increased synthesis of omega-3 desaturation or elongation products.

Dnmt1 and Dnmt3a regulate different types of DNA methylation events. While Dnmt1 is a maintenance DNA methyltransferase acting mainly during the S-phase [10], Dnmt3a is a de novo DNA methyltransferase, responsible for the establishment of nxew DNA methylation patterns that could differ from the ones inherited through cell division (reviewed in [11]). The increase in the gene and protein expression for these two DNA methyltransferases suggested that ALA could induce gene-specific epigenetic modifications in mouse NSCs in vitro, but without changing the global DNA methylation status. However, as opposed to our previously published work regarding the role of ALA in epigenetic regulation in liver (decreased Dnmts expression by ALA supplementation, [5]), the alterations reported in this study are opposite, suggesting that ALA could have specific epigenetic roles that are tissue-specific. The concentration of 100 µM used in the present study is physiologically relevant, as it was lower than the ALA concentrations detected in both liver and plasma [4].

The in vitro exposure of mouse neural stem cell to ALA alters the distribution of cells throughout cell cycle phases, apoptosis, and the gene and protein expression of Dnmt1 and Dnmt3a, but without alterations in global DNA methylation.

This work was funded, in part, by a grant to MDN from the UNC Center of Excellence in Children’s Nutrition sponsored by Mead Johnson Nutrition.

1. Haggarty P. Meeting the fetal requirement for polyunsaturated fatty acids in pregnancy. CurrOpinClinNutrMetab Care. 2014;17:151-155.

2. RombaldiBernardi J, de Souza Escobar R, Ferreira CF, PelufoSilveira P. Fetal and neonatal levels of omega-3: effects on neurodevelopment, nutrition, and growth. ScientificWorldJournal. 2012;2012:202473.

3. Luchtman DW, Song C. Cognitive enhancement by omega-3 fatty acids from child-hood to old age: findings from animal and clinical studies. Neuropharmacology. 2013; 64:550-565.

4. Niculescu MD, Lupu DS, Craciunescu CN. Maternal alpha-linolenic acid availability during gestation and lactation alters the postnatal hippocampal development in the mouse offspring. Int J DevNeurosci. 2011;29:795-802.

5. Niculescu MD, Lupu DS, Craciunescu CN. Perinatal manipulation of alpha-linolenic acid intake induces epigenetic changes in maternal and offspring livers. FASEB J. 2013;27:350-358.

6. Mehedint MG, Niculescu MD, Craciunescu CN, Zeisel SH. Choline deficiency alters global histone methylation and epigenetic marking at the Re1 site of the calbindin 1 gene. FASEB J. 2010; 24:184-195.

7. Krishan A. Rapid flow cytofluorometric analysis of mammalian cell cycle by propidium iodide staining. J Cell Biol. 1975;66:188-193.

8. Niculescu MD, Yamamuro Y, Zeisel SH. Choline availability modulates human neuroblastoma cell proliferation and alters the methylation of the promoter region of the cyclin-dependent kinase inhibitor 3 gene. J Neurochem. 2004;89:1252-1259.

9. Kuo KC, McCune RA, Gehrke CW, Midgett R, Ehrlich M. Quantitative reversed-phase high performance liquid chromatographic determination of major and modified deoxyribonucleosides in DNA. Nucleic Acids Res. 1980; 8:4763-4776.

10. Schneider K, Fuchs C, Dobay A, Rottach A, Qin W, Wolf P, et al. Dissection of cell cycle-dependent dynamics of Dnmt1 by FRAP and diffusion-coupled modeling. Nucleic Acids Res. 2013; 41: 4860-4876.

11. Chen BF, Chan WY. The de novo DNA methyltransferase DNMT3A in development and cancer. Epigenetics. 2014; 9.