Abstract:α-Lipoic acid induces a memory effect of the previous diet by maintain the mortality trajectory. α-Lipoic acid is unable to extend the lifespan by itself, but switching from dietary restriction to ad libitum feeding at 12 months of age retains the extended longevity characteristic of dietary restriction. Animals supplemented with α-Lipoic acid switching from ad libitum to DR do not exhibit an extended longevity [Merry et al., 2008]. Here we investigated the common transcriptional effect of lifespan interventions by dietary restriction and its mimetic α-lipoic acid.
Introduction
Meta-analysis Classification
Candidates
Enrichment
Discussion
Antioxidants α-lipoic acid, which is a component of several (α-ketoglutarate, branched chain and pyruvate) dehydrogenase complexes and an important facilitator of oxidative metabolism, confers a fascinating memory effect, by fixing the lifespan of previous feeding regimen [Figure: Lipoic Acid Memory Effect]. When animals are switched early in life from DR to AL and supplemented with α-lipoic acid the DR typical lifespan extension is maintained, but switching early from AL supplemented with α-lipoic acid to DR blocks the life span extending effect [Merry et al. 2008]. Interestingly, α-lipoic acid was shown in silico and in vitro to be a protein deacetylase inhibitor [Dashwood et al. 2007; van de Mark et al. 2003].
Figure: Lipoic Acid Memory Effect
ad libitum (AL)
dietary restricted (DR)
AL -> DR
DR -> AL
AL + LA
DR + LA
DR + LA -> AL
AL + LA -> DR
DR, AL -> DR, DR + LA and DR + LA -> Al are all long-lived compared to AL. This suggests that there exists a "memory" effect of lipoic acid, i.e. it does not change longevity itself but when switching diet it "remembered" the previous diet.
Whole transcriptome analysis identified genes that are differentially expressed in the increased longevity groups only (i.e. Al vs. DR->AL to AL vs. DR + LA AL). These groups share 501 genes in common which are expressed in the same direction and have similar fold changes.
Those 501 differential expressed genes are enriched for: ribosome, translation elongation, protein biosynthesis, acetylation, oxidative phosphorylation, huntington, parkinsons, alzeihmers, hydrogen ion transmembrane activity, mitochondrion, mitochondrial proton-transporting ATP synthase complex coupling factor F(0).
In groups with no changes in longevity (i.e. AL vs. DR + AL, AL + LA vs. AL, AL + LA->DR) there is a high number of differential expressed genes between these groups despite there being no change in longevity. 370 genes overlap with the differential expressed genes of the long-lived groups and are expressed in the same direction.
These genes are subtracted from the differential expressed genes of the long-lived groups. The resulting genes are enriched for ribosomal subunit, translation elongation protein biosynthesis, acetylation as well as P4Hx, vitamin binding, oxoreductase, positive regulation of protein kinase activity, oxidatitve phosphorylation and lipid metabolism.
Further removing genes differentially expressed in the same direction with lipoic acid supplementation (AL + LA vs AL) as the effect of lipoic acid alone should be removed from the differentially expressed genes separately which leaves 72 differentially expressed genes of which 59 are protein coding. These protein coding genes are associated with mitochondrion, protein biosynthesis, translation elongation, oxphos. ATPase 8, ATP binding, microtubule cytoskeleton organization, cytochrome-c oxidation activity, NADH, growth arrest and DNA-damage inducible gamma interacting protein 1 and ion transport.
From genes that are commonly differential expressed in multiple derived from the meta analyses of g juvenile growth period, aging, dietary restriction and circadian transcriptoms the following could be deduced.
Galnt1, Kcnc1, Atp8a1, Gadd45gip1, Rab3b, Sphkap, Eml5, Mrpl52 are circadian differential.
Sars (Serine--tRNA ligase, cytoplasmic) is aging-induced, but found here to be upregulated under different longevity conditions.
Tmem218 (Transmembrane protein 218) is dietary-restriction induced and also found here to be upregulated under different longevity conditions.
Pigq (Phosphatidylinositol glycan anchor biosynthesis, class Q) is aging-induced and differential expressed under dietary restriction. Here it was found to be transcriptional downregulated under different longevity conditions.
Rps12 (40S ribosomal protein S12) is circadian and aging-induced and also found to be upregulated under different longevity conditions.
Col4a5, Lamc1 (Laminin, gamma 1) and Retsat (All-trans-retinol 13,14-reductase) are aging-suppressed. Retsat encodes the PPAr-alpha-regulated and starvation-induced gene protein.
Hsd11b1 (Corticosteroid 11-beta-dehydrogenase isozyme 1) is induced during juvenile growth and under dietary restriction.
Txnrd1 (Thioredoxin reductase 1, cytoplasmic) is circadian and suppressed under dietary restriction.
Rpl37a (60S ribosomal protein L37a) is induced during aging.
Rala (Ras-related protein Ral-A) is circadian and aging-induced.
Egln3 (Egl nine homolog 3) is circadian and a ortholog of a DR-essential genes.
Bbs1 (Bardet-Biedl syndrome 1 homolog (Human) (Predicted), isoform CRA_a) is associated with juvenile growth.
Fam107a (Family With Sequence Similarity 107) is induced during aging and under dietary restriction.
The mouse and human homologs were identified and interaction networks generated. The three significant interactors are listed [Table: Top Murine Significant Interactors; Table: Top Human Significant Interactors].
Table: Top Murine Significant Interactors.
Group Symbol T S R p-value
Up Rpl18a 3 54 0.056 1.25⋅10-5
Gnl3 3 159 0.019 8.87⋅10-5
Pfdn5 2 23 0.087 0.00011
Delta Rpl18a 3 54 0.056 0.00010
Rhov 3 95 0.032 0.00021
Cftr 4 200 0.02 0.00038
Down Chrna5 1 1 1.0 0.00074
Cftr 3 200 0.015 0.00079
Scaper 1 1 1.0 0.00092
Up: [http://denigma.de/admin/network/candidate/?query=Longevity-Induced+10116&o=7] Down: [http://denigma.de/admin/network/candidate/?query=Longevity-Suppressed+10116&o=7] Delta: [http://denigma.de/admin/network/candidate/?query=Longevity-Differential+10116&o=7]
Table: Top Human Significant Interactors.
Group Symbol T S R p-value
Up RPL27 5 119 0.042 1.91⋅10-5
RPL9 5 141 0.035 2.42⋅10-5
EEF1G 6 268 0.022 4.44⋅10-5
Delta ANK3 3 19 0.157 0.00018
RPL29 3 107 0.047 0.00042
RPL27 5 119 0.042 0.00061
Down ANK3 3 19 0.158 2.14⋅10-5
ATL2 1 1 1.0 0.00106
TDRD6 1 1 1.0 0.00109
The functional enriched terms of the significant interactors for the differential expressed [Table: Murine Differential Genes], upregulated [Table: Murine Induced Genes] and suppressed genes [Table: Murine Suppressed Genes] were determined. Induced and suppressed enriched interactions were also determined for the human homologs [Table: Human Induced Genes; [Table: Human Suppressed Genes]]
Table: Murine Differential Genes.
Term T S R p-value
transport 1618 91 0.06 0.0
small GTPase mediated signal transduction 190 51 0.27 1.89e-38
acetylcholine-activated cation-selective channel activity 32 25 0.78 1.53e-35
T = Total number of interactions; S = Specific number of interactions with seed genes; R = Ration of specific/total number of interactions; p-value = p-value of the binomial test
Table: Murine Induced Genes.
Term T S R p-value
nucleotide binding 1992 92 0.046 0.0
transport 1618 74 0.046 0.0
mitochondrion 1858 83 0.045 0.0
T = Total number of interactions; S = Specific number of interactions with seed genes; R = Ration of specific/total number of interactions; p-value = p-value of the binomial test
Table: Human Induced Genes.
Term T S R p-value
intracellular 1976 75 0.035 0.0
mitochondrion 1410 102 0.072 0.0
translation 355 158 0.445 9.27e-198
T = Total number of interactions; S = Specific number of interactions with seed genes; R = Ration of specific/total number of interactions; p-value = p-value of the binomial test
Table: Murine Suppressed Genes.
Term T S R p-value
extracellular ligand-gated ion channel activity 34 33 0.97 2.42e-53
acetylcholine-activated cation-selective channel activity 32 31 0.97 3.52e-48
ion transmembrane transport 379 66 0.17 2.76e-33
T = Total number of interactions; S = Specific number of interactions with seed genes; R = Ration of specific/total number of interactions; p-value = p-value of the binomial test
Table: Human Suppressed Genes.
Term T S R p-value
glycosylphosphatidylinositol-N-acetylglucosaminyltransferase (GPI-GnT) complex 8 5 0.625 5.00e-10
acid-amino acid ligase activity 75 9 0.12 9.81e-10
protein K48-linked ubiquitination 32 7 0.21 1.15e-09
T = Total number of interactions; S = Specific number of interactions with seed genes; R = Ration of specific/total number of interactions; p-value = p-value of the binomial test
An investigation of several meta-analytic signatures found that circadian clock genes were highly enriched among top differentially expressed genes in DNA-microarrays of α-lipoic acid treatment. In fact, they were the top terms of functional enrichment and several core clock genes were highly differential [Circadian Clock Controls Ageing].
An hypothetical reason why there is no obvious changes in the transcriptome associated with the longevity of α-lipoic acid memory effect might be that it works via alterations of the oscillations of gene expression rather than altering significantly the levels of gene expressions in all tissues.
Merry BJ, Kirk AJ, Goyns MH (2008) Dietary lipoic acid supplementation can mimic or block the effect of dietary restriction on life span. Mech Ageing Dev 129: 341-348.
Dashwood RH, Ho E (2007) Dietary histone deacetylase inhibitors: from cells to mice to man. Semin Cancer Biol 17: 363-369.
van de Mark K, Chen JS, Steliou K, Perrine SP, Faller DV (2003) Alpha-lipoic acid induces p27Kip-dependent cell cycle arrest in non-transformed cell lines and apoptosis in tumor cell lines. J Cell Physiol 194: 325-340.
