5-Lipoxygenase DNA methylation and mRNA content in the brain and heart of young and old mice

Abstract

The expression of 5-lipoxygenase (5-LOX) is affected by aging and regulated by epigenetic mechanisms including DNA methylation. We used methylation-sensitive restriction endonucleases (AciI, BstUI, HpaII, and HinP1I) to assess 5-LOX DNA methylation in brain and heart tissue samples from young (2 months) and old (22 months) mice. We also measured mRNA content for 5-LOX and the DNA methyltransferases DNMT1 and DNMT3a. In young mice, the 5-LOX mRNA content was significantly greater in the heart compared to the brain; 5-LOX DNA methylation was lower, except in the AciI assay in which it was higher in the heart. Aging decreased 5-LOX mRNA content in the heart and increased it in the brain. Aging also increased 5-LOX DNA methylation and this effect was site- (i.e., enzyme) and tissue-specific. Generally, DNMT1 and DNMT3a mRNA content was lower in the brain regions compared to the heart; the only effect of aging was observed in the mRNA content of DNMT3a, which was decreased in the heart of old mice. These results indicate a complex tissue-specific and aging-dependent interplay between the DNA methylation system and 5-LOX mRNA content. Interpretation of this data must take into account that the tissue samples contained a mixture of various cell types.

Figure 1

The sequence of the mouse 5-LOX promoter-5′ UTR and the first exon-intron region subjected to the methylation-sensitive endonuclease assay with the four endonucleases—AciI, BstUI, HinP1I, and HpaII. The yellow highlight indicates the first exon including the start codon (ATG, in red). The four endonuclease-targeted methylation-sensitive endonuclease recognition sites are highlighted by four different colors. The selected DNA sequence included 2 AciI-sensitive sites (upstream of the start codon), 6 BstUI-sensitive sites (2 upstream and 4 downstream of the start codon), 8 HinP1I-sensitive sites (2 upstream and 6 downstream of the start codon), and 3 HpaII-sensitive sites (2 upstream and 1 downstream of the start codon). The primer regions used for quantitative PCR amplification are underlined; primers were used as follows. Promoter - 5′ UTR: forward = 5′-aga gaa gga tgc gtt gga aggt-3′ and reverse = 5′-gac tcc ggg caa gtg agt gct-3′; exon-intron: forward = 5′-agt cat gcc ctc cta cac ggt ca-3′ and reverse = 5′-agt cat gcc ctc cta cac ggt ca-3′. For the input control, we used the following primers: forward = 5′-tga tgt ggc tgg cct ctt atg tga-3′, reverse = 5′-act ggg act gag tgc agg aaa tgt-3′.

Figure 2

An example of the SYBR Green RD-qPCR assay of 5-LOX DNA methylation (shown are graphs obtained from 10 cerebellar samples). The samples were digested with the methylation-sensitive endonucleases (AciI, BstUI, HinP1I, and HpaII) as described in Material and Methods. The PCR reaction for the promoter-5′UTR (10 samples each: AciI = blue, BstUI = red, HinP1I = green, and HpaII = gray) and corresponding input control regions (shown in yellow) were carried out in separate tubes. Panel A shows the dissociation curve data, which indicate the presence of only one PCR product (peak) for each specific set of primers (fluorescence (first derivative of the raw fluorescence reading multiplied by −1) on the Y-axis versus the PCR product melting temperature (°C) on the X-axis). Panel B shows examples of the amplification plots used for calculating the quantitative data (the amplification plots fluorescence (baseline-corrected raw fluorescence) on the Y-axis versus cycle number on the X-axis). In this assay, the threshold cycle is inversely proportional to the log of the initial copy number. In other words, the more template that is present initially, the fewer the number of cycles required for the fluorescence signal to be detectable above background.

Figure 3

5-LOX mRNA levels in heart and brain tissues of young and old mice. The content of 5-LOX mRNA in the tissues of young (2-month-old, open bars) and old (22-month-old, closed bars) mice was measured by real-time PCR. The results were normalized against the corresponding cyclophilin mRNA contents and are presented as units (the coefficient of variation values amplified by a factor of 1000, see text for details). Data are expressed as mean ± S.E.M (n = 5 −10); *P < .05 and **P < .01 versus the corresponding values in young (t-test).

Figure 4

The effect of aging on 5-LOX DNA methylation in the region promoter-5′UTR. The methylation-sensitive endonuclease assay with the four endonucleases—AciI, BstUI, HinP1I, and HpaII—was performed as described in the text and indicated in Figure 1. The heart and brain (C: cerebellum, FC: frontal cortex, H: hippocampus) tissue samples were obtained from 2-month-old mice (open bars) and 22-month-old mice (closed bars). The results are expressed as units (the coefficient of variation values multiplied by a factor of 1000, see text for details) and shown as mean ± S.E.M (heart, C, and H: n = 5; FC: n = 10). *P < .05, **P < .01, and ***P < .001 versus the corresponding values in samples from 2-month-old mice (t-test).

Figure 5

DNMT1 (a) and DNMT3a (b) mRNA levels in heart and brain tissues of young and old mice. The content of DNMT1/DNMT3a mRNA in the tissues of young and old mice was measured by real-time PCR. The results were normalized against the corresponding cyclophilin mRNA contents and are presented as units (the coefficient of variation values amplified by a factor of 1000, see text for details). Data are expressed as mean ± S.E.M (n = 5 −10). Note that the only aging-associated difference was observed for DNMT3a mRNA content in the heart (^# P < .01, t-test). The content of DNMT mRNA was generally greater in the heart compared to the brain; *P < .05 and *P < .001 versus the corresponding values in the heart in the same age group (Dunnett's test).