reStructured
======================================================================================= Circadian Cycles Control Developmental Timing, Growth, Aging and its Modulation by Diet =======================================================================================
Dietary restriction without malnutrition delays aging and extends lifespan, but the molecular mechanism is still unknown. Aging gene expression changes originate from development and growth cessation, are modulated by diet and manipulate circadian rhythmicity. A significant proportion of the genes differentially expressed under dietary restriction, during aging and the juvenile growth period are under circadian control. Circadian clock genes are among the most significant interaction partners of dietary restriction differentially expressed genes in mouse and human interactome, while clock modulators interact highly significant with genes controlling nuclear lumen, ribosomal biogenesis and cell cycle. Canonical clock genes and clock modifiers itself exhibit differential expression upon DR, as well as during aging and already during the juvenile period. Notable, for instance circadian mRNA expression of NAMPT was identified to increase during juvenile, while to decrease during aging and is induced upon dietary restriction. Changing cycles in redox state and epigenetic marks could be possible mechanism to measure time over age.
Dietary restriction (DR) in defined factors, without causing malnutrition, slows down the aging process and extends healthy lifespan in multiple species, from yeast to primates. Different regimens can affect lifespan by restricting total calorie intake (calorie restriction), certain nutrients (sugars, amino acids) or time of feeding (intermittent fasting). It is totally unclear why various forms of restricted diets are working in so many diverse life forms.
Biological rhythmicity is generated by gene expression loops and metabolic cycles. The circadian clock is an ancient intrinsic timer, ubiquitous present in all species on earth. Thus, it governs timing in evolutionary far-related species, from unicellular organisms to mammals (Eelderink-Chen et al., 2010).
In D. melanogaster starvation, DR and many longevity mutants (like Rpd3, Sir2, chico, methusalem) all upregulate takeout (to). to is a secreted potential juvenile hormone binding protein and its induction by starvation is blocked by all arrhythmic central clock mutants (Bauer et al., 2010; Galikova and Flatt, 2010).
It is known that DR invokes hormonal changes which have systemic effects on the whole body physiology and decreases core body temperature. The circadian clock controls peripheral tissues also in an endocrine manner as well via the body temperature.
The light-responsive suprachiasmatic nucleus (SCN) of the hypothalamus harbours the circadian master clock. Besides this there is also a food-entrainable pacemaker, probably in the dorsomedial hypothalamic nucleus (Mieda et al., 2006) and the pineal gland appears to be ancient pacemaker too. The molecular clock is also present in all peripheral organs and tissues as well as in every single cell of the organism. Caloric restriction and maybe intermittent fasting reset the SCN clock and generate high amplitude circadian rhythms (similar to young ages) and increased lifespan (Froy and Miskin, 2010).
Many clock gene mutants (Clock-/-; Arntl-/-; Per2-/-; Dbp/Tef/Hlf triple knockout) display phenotypes of rapid aging. Although in general clock disruption correlates often with accelerated aging, disrupted circadian rhythms does not cause abnormal phenotypes per se (e.g. Cry1,2 double knockout) and specific clock manipulation are even associated with extended longevity (Csnk1e-/- under constant darkness, alpha-MUPA and to).
Juvenile growth appears to be controlled by non-cell autonomous endocrine mechanisms as well as cell-autonomous genetic program (Finkielstain et al., 2009; Lui et al., 2008; Lui et al., 2010).
We hypothesize here that the systemic and genetic program which regulates postnatal growth continues and cause aging, is manipulated by diet (delayed by restriction) and in fact timed by biological clocks (like circadian rhythmicity).
Comparing gene expression molecular signatures can generate hypothesis which gene expression programs are shared between particular biological processes (Plaisier et al., 2010). DNA-microarray meta-analysis of postnatal (juvenile) development/growth (Finkielstain et al., 2009; Lui et al., 2008; Lui et al., 2010), aging (de Magalhaes et al., 2009; Hong et al., 2010; Swindell, 2009), dietary (caloric) restriction (Hong et al., 2010; Plank et al., 2012; Swindell, 2008a, b, 2009) and circadian rhythmicity (Yan et al., 2008) lead to the identification of transcriptional signatures, hence genes which transcript expression commonly change in multiple tissues.
All living forms are programs executed by sequence of events encoded in the sequence of the genome. The complexity arises from chaos (stochastic interactions with the environment) which based on feedback loops and oscillations give rise to structured patterns of an organism.
Here, we found that the transcriptional signatures of juvenile, aging, DR and circadian rhythmicity significantly overlap and regulate a common set of a few genes which are themselves highly co-expressed with genes related to reactive oxygen generation (e.g. associated to mitochondria and peroxisome). Among the circadian genes in this common set which have the similar expression profile of being juvenile-induced, aging-suppressed and DR-triggered is for instance NAMPT. Several clock genes are themselves differential expressed during the juvenile period, aging and upon DR. On the genetic level juvenile growth appears to shorten the period, while aging seems to reduce the amplitude and lengthening the period, whereas DR restores high amplitude rhythms. Defined classes of clock modifiers have enrichment in certain functions and in general are highly enriched for nuclear lumen association. The interactome of clock modifiers in mouse and humans are enriched for genes related to ribosome biogenesis. We identified specific DR-essential, aging and longevity associated genes which are either circadian or function as clock modifiers. Moreover, imprinted genes are present in the signatures, although only to a moderate extent. Lastly, the transcriptional signature of α-lipoic acid (a DR-mimetic) is highly enriched for circadian clock-related terms.
Mammalian transcriptional signatures (commonly differential expressed genes in multiple tissues) of aging (de Magalhaes et al., 2009; Hong et al., 2010; Swindell, 2009), DR (Hong et al., 2010; Plank et al., 2012; Swindell, 2008a, b, 2009), juvenile growth (Finkielstain et al., 2009; Lui et al., 2008; Lui et al., 2010), and circadian rhythms were merged if representing the same category and intersected if belonging to different categories. Genes were mapped to unique Entrez gene identifiers and the overlap between any two signatures determined via the hypergeomtric test.
Overlaps were visualised with Venn diagrams of colored circles via the use of pythonâs SDL (Simple Directmedia Layer) bindings, with the size of a circle being proportional to the number of genes it represents. Functional enrichment of gene sets was obtained from DAVID (http://david.abcc.ncifcrf.gov/] (Huang da et al., 2009).
Coexpression were retrieved from the coexpression database [http://coxpresdb.jp/] (Obayashi and Kinoshita, 2011)].
The interactome information was integrated from various sources as described elsewhere (Wuttke et al., submitted).
Gene expression profile of alpha-lipoic acid treatment were derived from the GEO database (GSE27625; (Finlay et al., 2012).
DR as it retards aging should be expected to act in opposite on aging-gene expression alterations. The overlap of genes which are oppositely expressed during aging and DR were more significant than those genes which change in the same direction (Figure S1). It was noted that a significant amount of genes were enriched for up- and down-regulation both during aging and under DR. Those genes which are regulated by aging and DR in the same direction could represent protective mechanisms against aging (de Magalhaes et al., 2009), which are further enhanced upon DR. This is important as it argues that it could be detrimental to simple restore aged gene expression patterns to young state. It would be much more of interest to identify the gene expression signature associated to longevity, i.e. aging suppression.
Genes going commonly up in juvenile and aging are dominated by immune system related clusters/terms (Figure S2). Of note, genes related to wound healing and cell cycle start to be down-regulated already in the juvenile and continue this trend during aging which could explain why the proliferative/regenerative potential declines with age.
Juvenile and DR commonly downregulate collagen genes, actin-related genes and cell cycle genes, while juvenile and DR both upregulate Drug metabolism and positive regulators of Apoptosis (Figure S3).
Next we asked whether there is an enrichment of circadian genes in the transcriptional signatures of DR, aging and juvenile growth (Figure 2). Although more than half of the DR-differential expressed genes were circadian, the aging and juvenile signature were also significant enriched for circadian genes.
To dissect which circadian process are differential regulated upon DR, during aging and in the juvenile growth phase the circadian signature was intersected with the respective up- and down-regulated genes (Figure S4).
Finally, it was asked which genes are common to more than two signatures (Figure 3 and Table 1).
Among these genes several (Nampt, Inmt, Klf9, Pim3, Dbp, Gstm1) are directly bound by Arntl close to their transcription start site. Most of them exhibit the common expression pattern of being up-regulated during juvenile, downregulated during aging and upregulated upon DR, which could indicate that reduced Arntl transactivation accounts at least in part for this program.
In total 27 genes are common to the transcriptional signatures of juvenile, aging, DR and circadian. Seven were metabolic enzymes, five transcription factors, three collagen genes, 7 immune related genes, 5 clock genes. At least five genes of this gene set have defined role in the clockwork. Dbp is an output transcriptional regulator. Nampt mediates an enzymatic feedback loop in the clock. Ccrn4l, chaperone Hspa8 and Cirbp are all three clock system entrained factors. Ccrn4l is a clock component and output gene and its knockdown has positive metabolic effects (lean mice clock machine). Based on the expression pattern, genes in this set can be grouped. One group consist of immune-related genes, which are increased in expression during juvenile and aging, while DR suppresses their expression. Among these genes, the histocompatibility antigen (like HLA-DRB1) were identified in genome-wide association studies to correlate with human longevity in Sicilian (Listi et al., 2010), French (Soto-Vega et al., 2005) and Okinawa (Akisaka et al., 1997) populations. Another group is formed by three collagen genes, which bind growth factors, thus interacting selectively and non-covalently with growth factors, proteins or polypepides that stimulate a cell or organism to grow or proliferate. There is no gene up-regulated during juvenile, which becomes down-regulated during aging which is also differentially expressed upon DR and exhibit circadian rhythmicity.
Coexpression (http://coxpresdb.jp/) among these genes in Table 1 clusters them in at least three different groups. Collagen genes (Col1a1, Col5a1 and Col3a1), genes related to the immune system (Igh-6, Ccl6, Cd74, H2-Aa, H2-D1, H2-Eb1 and Psmb9) and others. Many of the immune related genes (H2-Aa, H2-D1, H2-Eb1 and Psmb9) are also clustered in close proximity on chromosome 17. This argues that these modules compromise distinctive regulatory programs. 9 genes (Nampt, Inmt, Glul, Dbp, Zbtb16, Klf9, Pim3 and Hspa8) were selected as they had similar expression pattern (up in juvenile, down in aging and up in DR) and their coexpressed genes were investigated. 289 genes were found to be significantly coexpressed (Table 2). The strongest enriched clusters were terms related to control of the redox balance. Interestingly there were 4 antioxidant activities among them: Gpx1, Prdx5/6, Sod2. The PRDX5, the mitochondrial peroxidoxin progressively declines with advanced age in the nervous system of humans and PRDX6 RNAi causes arrhythmicity.
Canonical core clock and modifier genes (indicated in bold) as well as non-canonical pacemakers were identified in the transcriptional signatures of juvenile, aging and DR (Table S1). Pacemakers which knockdown via RNAi in human cell causes arrythymicity (AR), long period (LP), short period (SP), high amplitude (HA) or low amplitude (LA) circadian phenotype were annotated. From this it would be expected that juvenile growth leads to shorter period, aging in contrast would cause a long period and low amplitude, whereas DR would lead to high amplitude rhythmicity.
Remarkable, Wnt signaling which is crucial in development and stem cells was enriched in the overlaps of the transcriptional signatures as well as intersections of aging and DR with clock genes/modifiers (Table S2).
Clock modifiers from the genome-wide RNAi screen in human cells (Zhang et al., 2009) were divided into groups according to their effect on cellular clocks (i.e. high amplitude and short/long period) were assessed for specific functional enrichments (Table S3). Clock modifiers in general were enriched for terms related to nuclear lumen, stress response, RNA-binding and histone modification. High amplitude genes were associated to cell cycle control, H4 acetylation, apoptosis regulation and ion transport. The enrichment for ion transport could couple imprinted genes (also enriched for ion transport) and redox state. Short period genes are enriched for nucleolus, while long period genes are associated to nucleoplasm, RNA processing and Hedgehog signalling.
Genes associated to be required for circadian rhythmicity (here called as âPacemakersâ) and differential expressed during aging and DR were identified and checked for enrichment (Table S4). Aging clock modifiers are enriched for acetylation and DR clock modifiers for ribonucleoprotein.
In order to gather how the interactome is affected by the clock modifiers identified in the genome-wide RNAi screen in human cells, a human protein-protein interaction network in mouse functional interaction network were generated. Genes/proteins which were identified to interact with a higher specificity in connectivity with clock modifiers than expected by chance were analysed for functional enrichment (Table S5 and S6). Some clusters of functional terms were commonly enriched in both networks, such as ânuclear lumenâ and âribosome biogenesisâ.
8 DR-essential gene orthologs [Ep300, Crebbp, Sirt1, Epas1 (long period), Foxa3 (long period), Mtor (long period), Tsc2 (long period) and Sfn (high amplitude)] are in fact clock modifiers (p-value = 0.005), 25 DR-essential gene orthologs (Atg7, Dlat, Ywhab, Irs2, Ghr, Elovl6, Slc25a30, Cryab, Sc5d, Eif4b, Hspb1, Prkdc, Gclc, Elovl5, Ywhaq, Ywhaz, Mtor, Rgl1, Trp53, Mat2a, Foxo3, Ywhah, Ywhag, Egln3 and Elovl1) exhibit circadian oscillation on transcript level in multiple tissues. Of these two sets only Mtor is in common. Also 13 of 27 circadian systemic entrained factors are differentially expressed under DR (not significant).
104 of 252 aging-associated genes (derived from human genes in GenAge) were circadian (p-value < 5â10-19). 12 of 40 negative aging-suppressors (Prdx1, Bub3, Foxm1, Nos3, Polg, Rae1, Xpa, Xrcc6, Zmpste24) were circadian with high significance (p-value = 0). 6 of 18 positive gerontogenes (Cdkn1a, Cebpa, Ghr, Igf1, Irs1, and Irs2) were circadian(P(X)Â > 6 = 0.027). 3 of 11 positive aging-suppressors (Cebpb, Pck1, and Igf1) were circadian (not significant). 1 of 2 negative gerontogenes was circadian [P(X)Â > 6 = 0.031].
Genetic variants linked to longevity in genome-wide association studies give clues which players might be associated to exceptional longevity. At least three clock modulators were found to be associated to human longevity (Table S7).
Of note, thyroid hormone receptor activator and coativator activity is one of the enriched molecular functions of aging hypomethylated genes (Daniel Wuttke, unpublished).
Several genes which were differentially expressed in juvenile, aging and DR are imprinted genes (Table S8). This striking correlation might point to a causal relationship.
Imprinted genes are differentially expressed from the paternal or maternal copy duo to epigenetic regulation such as silencing DNA methylation. Imprints are established in the developing gametes and erased and reset in germ cells. The memory is set by DNA-methylation, post-translational histone modifications and protein/RNA factors. Paternally expressed genes encourage fetal growth (oncogenic), while maternally expressed genes are growth inhibitor (similar to tumor suppressors).
The circadian imprinted gene Slc22a1 transports methylnicotinamide, metformin, spermidine in both directions, is inhibited by PKA, and induced by PPARA and PPARG. All these molecules were strongly implicated in aging.
α-Lipoic acid (LA) supplementation in the ad libitum phase during feeding switch regimens can mimic or block the lifespan extending effect of DR (Merry et al., 2008). Aging suppresses hepatitic Dbp and Bhlhe41 expression. LA significantly modulated the expression of circadian clock genes and led to a high enrichment of clock-related terms in the induced genes (First cluster in DAVID analysis with an enrichment score of 3.54; Table 3). In young animals LA induced the expression of Naps2, Arntl and Cdkn1a, while it suppressed Dbp, Per2 and Per3 as well as Dact1 (dapper, antagonist of β-catenin, homolog 1). In old animals LA induced Npas2, Arntl and suppressed Tef and Jdp2, Dbp, Per2 and Per3 in the liver.
There are discrete regulatory programs operating during the life history of an organism (Table S9). First of all, there are gene regulatory patterns established during development and maturation which exhibit the same trend after adulthood (âdriftâ). Secondly, the initiation of a growth subroutine ensures that the animal stops growing in height/size, but continues action of these changes in adults cause detrimental effects. Third, there are also protective responses to the negative effects of the continuation of development and growth cessation subroutine as well as to environmental influences and perhaps damage. Forth, as these programs are not sufficient to explain all observed changes there are indices for an active aging program, also it nature remains controversial. Dietary restriction appears to counteract some but not all age-related changes.
It was totally elusive how all these regulatory programs are coordinated in time and in space at the whole organismsâ physiology. Here we present compiling evidence that biological timekeeping mechanism involving circadian cycles regulate not only daily rhythms, but also participate in controlling the development, growth and aging of an organism (Figure 4).
Further, the sexual dimorphism on lifespan by genetic mutations, DR or pharmacological interventions might be duo to imprinted genes and circadian hormones.
A specific set of genes was identified to be commonly regulated by the programs mediating juvenile growth, aging, DR response and circadian rhythmicity. Besides the immune-related and collagen gene several of these are very interesting candidates and therefore will be discussed in the following in more detail and put into context. Among them were five clock genes (Nampt, Hspa8, Glul, Dbp and Ccrn4l). At least seven genes encode metabolic enzymatic activities (Nampt, Inmt, Glul, Aldh1a1, Gstm1 and Gsta3). Three genes are clock systemic entrained factors, which encode a heat and cold shock protein, respectively (Hspa8 and Cirbp), as well as a deadenylase (Ccrn4l). Two have as primary function the regulation of cell cycle in common (Ccnd2 and Pim3). Five encode transcription factors (Zbtb16, Sox4, Klf9, Dbp and eventually Peg3). A few of all of them were implicated in aging. For instance, Hspa8 and Glul are oxidatively modified in aging rat brain (Perluigi et al., 2010).
Metabolic Enzymes ~~~~~~~~~~~~~~~~~ Nampt (nicotinamid phosphoribosyltransferase) catalyses the condensation of nicotinamide with 5-phosphoribosyl-1-pyrophosphate to yield nicotinamide mononucleotide, which is crucial step in the biosynthesis of nicotinamide adenine dinucleotide (NAD). Nampt knockout is embryonic lethal. Replicative senescence is preceded by a marked decline in Nampt expression and activity. Pharmacological suppression of Nampt activity induces premature senescence, while introducing Nampt gene into aging human cells delayed senescence and substantially lengthened cell lifespan (van der Veer et al., 2007). The substrate of Nampt, nicotinamide, increases the period in Arabidopsis (Dodd et al., 2007) and mammalian cells (Asher et al., 2008), just as aging does (Table S1). Its product NAD is considered to function as a metabolic oscillator of aging (Imai, 2010). Intracellular NAD+ levels and the NAD:NADH ratio decline in multiple organs during the progression from young to middle age (Braidy et al., 2011). Aging decreases systemic NAD biosynthesis via lowering of plasma NMN (nicotinamid adenine dinucleotide) levels (5 months vs. 19 months old) in BESTO (beta-cell specific Sirt1-overexpressing) mice (Ramsey et al., 2008). Systemic NAD biosynthesis is mediated by intra- and extracellular Nampt functions as driver that keeps up the pace of metabolism in multiple tissues/organs. NAD-dependent enzymes like Sirt1 serve as universal mediators that execute metabolic effects in a tissue-dependent manner in response to changes in systemic NAD biosynthesis (Imai, 2009). SIRT1 itself regulates amplitude and duration of circadian gene expression.
Inmt (Indolethylamine N-methyltransferase; a.k.a. Temt) catalyses the N-methylation of amines such as tryptamine (derived from e.g. tryptophan) and serotonin via the usage of S-adenosyl-L-methionine (S-adenosyl-L-methionine + amine = S-adenosyl-L-homocysteine + a methylated amine) Thus, it participated in tryptophan metabolism and couples it with the one-carbon source metabolism of (S-adeno-)methionine and related structures.
Glul (Glutamine synthetase) is a ubiquitously expressed enzyme that converts glutamate (a signaling molecule) to glutamine and glutamine into 4-aminobutanoate (gamma-aminobutyric, GABA). It is involved in ammonium detoxification and interorgan nitrogen flux (Haberle et al., 2006) as well as essential for cell proliferation. It transcription increases with development (Vardimon et al., 1993).
Aldh1a1 (aldehyde dehydrogenase family 1 member A1) was implicated in stem cell function as it is highly expressed in adult stem cells but changes during age. Aldh1a1 knockout mice develop cataracts at 6-9 months if age and had decreased proteosomal activity, increased protein oxidation and increased glutathione levels (Lassen et al., 2007). Aldh1a1 participates in retinoic acid (RA) synthesis in vivo (Retinal + NAD+ + H2O -> retinoate + NADH). RA signaling is necessary for development and RA synthesis is catalysed by multiple enzymes including Aldh1a1 (Fan et al., 2003). Retinoic signaling represses Wnt signaling via induction of Pitx2 and Dkk2 (Kumar and Duester, 2010).
Glutathione S-transferases such as are multifunctional enzymes important in the metabolism of xenobiotics, but catalyse also important reaction in hormone synthesis. For instance, Gsta3 catalyses the double bond isomerization of precursors for progesterone and testosterone during the biosynthesis of steroid hormones.
Clock Systemic Factors ~~~~~~~~~~~~~~~~~~~~~~ Hspa8 and Cirbp as well as Ccna2 (only suppressed during juvenile) are cellular stress genes (Li et al., 2010). Meal time feeding induced Hspa8, Cirbp and Ccna2 transcription ten-fold. Physiological temperature fluctuations drive rhythmic DNA binding of HSF1 (heat shock factor-1), which regulates HSP transcription (Reinke et al., 2008).
Hspa8 (heat shock 70kDa protein 8; aka Hsc70 or Hsp73) is a ubiquitous expressed chaperone, which is localised in cytoplasmic mRNP granules. Melanosome related terms were also identified in the enrichments such as in the intersection of DR and circadian (Figure 2). Upon heat stress Hspa8 translocates rapidly from the cytoplasm to the nuclei (especially to the nucleoli). HSPA8 contributes to CMA (chaperone mediated autophagy) from both sides of the lysosome membrane. The cytosolic form binds substrates and targets them to lysosomes, while the form in the lysosomal lumen is required to complete translocation across the lysosomal membrane, probably by actively pulling in (Hubbard et al., 2011). Hspa8 was found to be upregulated in an in vitro model of DR (de Cabo et al., 2003).
Cirbp (Cold-inducible mRNA binding protein) is ubiquitous expression and has protective role in the genotoxic stress response by stabilizing transcripts of genes involved in cell survival. It acts as a translational activator and is essential in cold-induced suppression of cell proliferation. Various cytoplasmic stresses, such as osmotic and heat shocks cause Cirbp translocation from the nucleus to the cytoplasm into stress granules. Moderate low temperature preserves stemness of neural stem cells and prevented apoptosis via stimulation of CIRPB (Saito et al., 2010) Cirbp-knockout mice exhibit neither gross abnormality nor defect in fertility, but had a decreased number of undifferentiated spermatogonia. Cirbp accelerates cell-cycle progression from G0 to G1 as well as from G1 to S phase in cultured mouse embryonic fibroblasts.
Ccrn4l expression increases during juvenile and further during aging, while DR counteracts this trend as it suppresses it. Ccrn4l, the Carbon catabolite repressor protein 4 homolog (a.k.a. Ccr4 or nocturnin) is an ubiquitous expressed component of the circadian clock or downstream effector of clock function. Ccrn4l homozygous knockout mice are extraordinary healthy as they were lean, resistant to diet-induced obesity and fatty liver development, showed increased circulating glucose levels and increased insulin sensitivity on a standard diet as well as impaired glucose tolerance on a high fat diet (Green et al., 2007). Ccrn4l was described as a molecular factor of senescence regulated by chromatin and methylation status during aging (Puech et al., 1997).
Cell Cycle Regulators ~~~~~~~~~~~~~~~~~~~~~ Ccnd2 is a direct Arntl circadian output target and essential for G1/S transition. It also participates in Wnt signaling.
Pim3 is a serine/threonine-protein kinase which is widely expressed, able to phosphorylate CDKN1, and involved in cell cycle progression and suppression of apoptosis.
Transcription Factors ~~~~~~~~~~~~~~~~~~~~~ Zbtb16 (alias PLZF) is enriched for up-regulation during juvenile growth and DR, while it is enriched for downregulation during aging. Zbtb16 is enriched for circadian rhythmicity and coexpressed with Per1,2. Zbtb16 maintains the undifferentiated state of spermatogonial progenitor cells (SPCs). It inhibits mTORC1 pathway activity in SPCs by inducing Ddit4. Increased mTORC1 activity inhibits the response of SPCs to the niche-derived factor GDNF (Ellisen, 2005). Ddit4 was greatly induced after-short term DR in male mice liver (Estep et al., 2009).
Sox4 knockout mice die at embryonic day 14 (Schilham et al., 1996). Sox4 was found here to be, to be progressively be suppressed during juvenile and aging but being also induced upon DR. It regulates cell differentiation, proliferation and survival in multiple essential processes (Penzo-Mendez, 2010). Sox4 enhances beta-catenin/TCF activity and proliferation.
Klf9 (Krueppel like factor 9) can activate transcription and also repress gene expression by directly binding to Sin3A a scaffold co-repressor. Its expression is sensitive to thyroid hormone and serotonin signaling. Klf9 was implicated in stem cell maintenance and differentiation of T- and B-lymphocytes and was found to be necessary for adult neurogenesis (Scobie et al., 2009). Klf9 affects keratinocyte proliferation/differentiation by controlling the expression of target genes in a daytime-dependent manner. Kl9 is substantially up-regulated in a cortisol and differentiation-state-dependent manner. Kl9 exhibits strong anti-proliferative effects and it putative target genes include proliferation/differentiation markers that also exhibit circadian expression in vivo.
Dbp (Albumin D-element binding protein) is classical circadian transcription factor which controls genes that peak in expression at puberty (Yano et al., 1992).
Peg3 (Paternally-expressed gene 3 protein) belongs to the krueppel C2H2-type zinc-finger protein family and inhibits Wnt signalling. It alters growth and development as well as regulates apoptosis (Jiang et al., 2010).
Aging is accompanied by a decrease in circadian amplitude and a shift in the period. In contrast, DR slows down aging and increases the amplitude of circadian oscillations. The causality of these correlations was not yet clear; i.e. if aging impacts on clock or vice versa. Here it was shown that this trend was reflected on the genetic and transcriptional level as aging tend overall to decrease the amplitude and prolong the period, while DR counteracted these effects by increasing the amplitude (i.e. clock resetting). DR induces the expression of the first interconnected negative limb (Periods and Cryptochromes), which likely contributes to the higher amplitude. Furthermore, during juvenile the gene expression changes cause a shortening in the period. Period prolongation is therefore part of the growth cessation subroutine which becomes initiated as an organism reaches maturity (Figure 4).
How the cellular circadian clock changes during the juvenile period is not studied, but during the adolescent there are enormous changes/alterations in circadian behaviour and sleep/wake cycle (Tarokh et al., 2010).
Juvenile growth tends to be mediated by a shortening of the period, while aging prolongs the period and lowers the amplitude. DR counteracts this trend by showing a tendency to shorten the period and increase the amplitude (thus clock resetting). These trends are consistence with the observation that DR and αMUPA increase clock gene expression amplitude and reset the clock.
As the DR-essential Mtor modifies clock function and is itself circadian on transcript level, it represents a novel feedback loop coupling Mtor signaling to the molecular clock and vice versa.
Further, as the molecular interaction network of clock modifiers is strongly enriched for functional terms related to ribosome genesis, both in mouse and human interactome, this further supports the notation that cell growth and circadian time keeping are tightly coupled.
Mtor signaling is known to be decreased under DR, it could therefore provide one pathway how DR impacts on circadian rhythmicity and especially its ability to reset the SCN.
A conformation of the role of circadian genes in the control of aging and especially its modulation by diet stems from the fact that α-lipoic acid treatment (which is able to mimic or block the effect of DR by fixing the mortality tractor of the previous feeding regimen in feeding swap experiments) was found to primarily alter clock genes regardless of age.
In this work we intersected the transcriptional signatures of juvenile growth, aging, dietary restriction and circadian rhythmicity and found a higher overlap of these processes as expected by chance. Among a common set of 26 genes around 12 genes with a similar expression profile (induced during juvenile, suppressed during aging and upregulated by DR), were strongly coexpressed with processes related to reactive oxygen generation, although they all have no obvious primary function in this process. Clock genes (modifiers and pacemakers) are themselves differentially expressed during juvenile, aging and upon DR and from genetic studies their expression suggests that juvenile phase is associated with shorting in the period, but aging reduces the amplitude and prolongs the period, while DR increases the amplitude. Interestingly different type of clock modifiers have discrete functional enrichment and were in the interaction network of pacemakers was strongly enriched for the terms related to the nucleolus (in both mouse and humans), which fragmentation was proposed to be a cause of aging. Many DR-essential, aging and some longevity associated genes were circadian or even clock modulators, providing some explanation for the here found overlap.
Several of the circadian genes commonly regulated by juvenile, aging and DR can be associated with Wnt signaling (Table 1) and the intersections between clock-modulators with aging and DR-differential genes were also enriched for Wnt signaling. A common cross-point might be GSK3 which is pivot component of Wnt signaling (Wu and Pan, 2010) as well as the circadian clock (Harms et al., 2003; Li et al., 2012) and the age-related changes of stem cells (Brack et al., 2007; Xiang et al., 2011). Progeroid LaminA (leads to premature aging in humans) shuts down Wnt signaling which alters the extracellular matrix synthesis and therefore impairs stem cell function (Hernandez et al., 2010). Interestingly LaminA is circadian expressed and increases during aging, Zmpste24 (which processes LaminA) is under clock control too. Furthermore, the huntingtin (Htt) gene is also circadian and aging induced, providing an explanation for the late onset of this disease.
We here propose that the regulatory programs governing juvenile growth, aging (and its modulation by diet) as well as circadian rhythmicity are interconnected by a common timing mechanism which probably operates on the genome via long-term epigenetic changes and on the level of oscillatory metabolites such as ROS and NAD.
Thus, aging is here hypothesised to be caused by continuation of the development program and the growth cessation program both lead to changes in chromatin structure. The location and structure of chromatin is highly dynamic and changes in a rhythmic manner. Juvenile and aging gene expression changes are established for instance by DNA methylation and demethylation, histone modifications as well as reposition of whole chromosomal locus to the nuclear periphery (LMNA). Differentially methylated genes are the product of cyclic (de)methylation reactions and targeted by defined factors, such as transcription factors and non-coding RNAs among them are imprinted genes. Some specific loci become hypermethylated while other become hypomethylated. This is controlled by an instructive program. Circadian imprinted genes are related to ion homeostasis. Altered ion transport leads change in the rhythms of cellular respiration and therefore changed redox cycles with aging drive circadian transcriptional cycles.
If developmental progression as well as aging and its modulation by diet is mediated by changes in the oscillatory state of metabolites and the decoration of the genome, α lipoic acid might for instance block a switch in this transitions (e.g. to a higher amplitude).
Many thanks to O. Vasieva for the helpful advice and JP. de Magalhaes for the great opportunity to solve aging.
Akisaka, M., Suzuki, M., and Inoko, H. (1997). Molecular genetic studies on DNA polymorphism of the HLA class II genes associated with human longevity. Tissue Antigens 50, 489-493.
Bauer, J., Antoch, M., Chang, C., Schorl, C., Kolli, S., Neretti, N., and Helfand, S.L. (2010). Comparative transcriptional profiling identifies takeout as a gene that regulates life span. Aging 2, 298-310.
Brack, A.S., Conboy, M.J., Roy, S., Lee, M., Kuo, C.J., Keller, C., and Rando, T.A. (2007). Increased Wnt signaling during aging alters muscle stem cell fate and increases fibrosis. Science 317, 807-810.
Braidy, N., Guillemin, G.J., Mansour, H., Chan-Ling, T., Poljak, A., and Grant, R. (2011). Age related changes in NAD+ metabolism oxidative stress and sirt1 activity in wistar rats. PloS one 6, e19194.
de Cabo, R., Furer-Galban, S., Anson, R.M., Gilman, C., Gorospe, M., and Lane, M.A. (2003). An in vitro model of caloric restriction. Exp Gerontol 38, 631-639.
de Magalhaes, J.P., Curado, J., and Church, G.M. (2009). Meta-analysis of age-related gene expression profiles identifies common signatures of aging. Bioinformatics 25, 875-881.
Eelderink-Chen, Z., Mazzotta, G., Sturre, M., Bosman, J., Roenneberg, T., and Merrow, M. (2010). A circadian clock in Saccharomyces cerevisiae. Proceedings of the National Academy of Sciences of the United States of America 107, 2043-2047.
Ellisen, L.W. (2005). Growth control under stress: mTOR regulation through the REDD1-TSC pathway. Cell Cycle 4, 1500-1502.
Estep, P.W., 3rd, Warner, J.B., and Bulyk, M.L. (2009). Short-term calorie restriction in male mice feminizes gene expression and alters key regulators of conserved aging regulatory pathways. PloS one 4, e5242.
Fan, X., Molotkov, A., Manabe, S., Donmoyer, C.M., Deltour, L., Foglio, M.H., Cuenca, A.E., Blaner, W.S., Lipton, S.A., and Duester, G. (2003). Targeted disruption of Aldh1a1 (Raldh1) provides evidence for a complex mechanism of retinoic acid synthesis in the developing retina. Molecular and cellular biology 23, 4637-4648.
Finkielstain, G.P., Forcinito, P., Lui, J.C., Barnes, K.M., Marino, R., Makaroun, S., Nguyen, V., Lazarus, J.E., Nilsson, O., and Baron, J. (2009). An extensive genetic program occurring during postnatal growth in multiple tissues. Endocrinology 150, 1791-1800.
Finlay, L.A., Michels, A.J., Butler, J.A., Smith, E.J., Monette, J.S., Moreau, R.F., Petersen, S.K., Frei, B., and Hagen, T.M. (2012). R-alpha-lipoic acid does not reverse hepatic inflammation of aging, but lowers lipid anabolism, while accentuating circadian rhythm transcript profiles. American journal of physiology Regulatory, integrative and comparative physiology 302, R587-597.
Froy, O., and Miskin, R. (2010). Effect of feeding regimens on circadian rhythms: implications for aging and longevity. Aging 2, 7-27.
Galikova, M., and Flatt, T. (2010). Dietary restriction and other lifespan extending pathways converge at the activation of the downstream effector takeout. Aging.
Harms, E., Young, M.W., and Saez, L. (2003). CK1 and GSK3 in the Drosophila and mammalian circadian clock. Novartis Foundation symposium 253, 267-277; discussion 102-269, 277-284.
Hernandez, L., Roux, K.J., Wong, E.S., Mounkes, L.C., Mutalif, R., Navasankari, R., Rai, B., Cool, S., Jeong, J.W., Wang, H., et al. (2010). Functional coupling between the extracellular matrix and nuclear lamina by Wnt signaling in progeria. Developmental cell 19, 413-425.
Hong, S.E., Heo, H.S., Kim, D.H., Kim, M.S., Kim, C.H., Lee, J., Yoo, M.A., Yu, B.P., Leeuwenburgh, C., and Chung, H.Y. (2010). Revealing system-level correlations between aging and calorie restriction using a mouse transcriptome. Age (Dordr) 32, 15-30.
Huang da, W., Sherman, B.T., and Lempicki, R.A. (2009). Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc 4, 44-57.
Hubbard, V.M., Valdor, R., Macian, F., and Cuervo, A.M. (2011). Selective autophagy in the maintenance of cellular homeostasis in aging organisms. Biogerontology.
Imai, S. (2009). The NAD World: a new systemic regulatory network for metabolism and aging--Sirt1, systemic NAD biosynthesis, and their importance. Cell Biochem Biophys 53, 65-74.
Imai, S.I. (2010). "Clocks" in the NAD World: NAD as a metabolic oscillator for the regulation of metabolism and aging. Biochimica et biophysica acta 1804, 1584-1590.
Jiang, X., Yu, Y., Yang, H.W., Agar, N.Y., Frado, L., and Johnson, M.D. (2010). The imprinted gene PEG3 inhibits Wnt signaling and regulates glioma growth. The Journal of biological chemistry 285, 8472-8480.
Kumar, S., and Duester, G. (2010). Retinoic acid signaling in perioptic mesenchyme represses Wnt signaling via induction of Pitx2 and Dkk2. Developmental biology 340, 67-74.
Lassen, N., Bateman, J.B., Estey, T., Kuszak, J.R., Nees, D.W., Piatigorsky, J., Duester, G., Day, B.J., Huang, J., Hines, L.M., et al. (2007). Multiple and additive functions of ALDH3A1 and ALDH1A1: cataract phenotype and ocular oxidative damage in Aldh3a1(-/-)/Aldh1a1(-/-) knock-out mice. The Journal of biological chemistry 282, 25668-25676.
Li, J., Lu, W.Q., Beesley, S., Loudon, A.S., and Meng, Q.J. (2012). Lithium impacts on the amplitude and period of the molecular circadian clockwork. PloS one 7, e33292.
Listi, F., Caruso, C., Colonna-Romano, G., Lio, D., Nuzzo, D., and Candore, G. (2010). HLA and KIR frequencies in Sicilian Centenarians. Rejuvenation research 13, 314-318.
Lui, J.C., Finkielstain, G.P., Barnes, K.M., and Baron, J. (2008). An imprinted gene network that controls mammalian somatic growth is down-regulated during postnatal growth deceleration in multiple organs. American journal of physiology Regulatory, integrative and comparative physiology 295, R189-196.
Lui, J.C., Forcinito, P., Chang, M., Chen, W., Barnes, K.M., and Baron, J. (2010). Coordinated postnatal down-regulation of multiple growth-promoting genes: evidence for a genetic program limiting organ growth. The FASEB journal : official publication of the Federation of American Societies for Experimental Biology 24, 3083-3092.
Merry, B.J., Kirk, A.J., and Goyns, M.H. (2008). Dietary lipoic acid supplementation can mimic or block the effect of dietary restriction on life span. Mechanisms of ageing and development 129, 341-348. Mieda, M., Williams, S.C., Richardson, J.A., Tanaka, K., and Yanagisawa, M. (2006). The dorsomedial hypothalamic nucleus as a putative food-entrainable circadian pacemaker. Proceedings of the National Academy of Sciences of the United States of America 103, 12150-12155.
Obayashi, T., and Kinoshita, K. (2011). COXPRESdb: a database to compare gene coexpression in seven model animals. Nucleic acids research 39, D1016-1022. Penzo-Mendez, A.I. (2010). Critical roles for SoxC transcription factors in development and cancer. The international journal of biochemistry & cell biology 42, 425-428.
Perluigi, M., Di Domenico, F., Giorgi, A., Schinina, M.E., Coccia, R., Cini, C., Bellia, F., Cambria, M.T., Cornelius, C., Butterfield, D.A., et al. (2010). Redox proteomics in aging rat brain: involvement of mitochondrial reduced glutathione status and mitochondrial protein oxidation in the aging process. J Neurosci Res 88, 3498-3507.
Plaisier, S.B., Taschereau, R., Wong, J.A., and Graeber, T.G. (2010). Rank-rank hypergeometric overlap: identification of statistically significant overlap between gene-expression signatures. Nucleic acids research 38, e169.
Plank, M., Wuttke, D., van Dam, S., Clarke, S.A., and de Magalhaes, J.P. (2012). A meta-analysis of caloric restriction gene expression profiles to infer common signatures and regulatory mechanisms. Mol Biosyst 8, 1339-1349.
Puech, A., Dupressoir, A., Loireau, M.P., Mattei, M.G., and Heidmann, T. (1997). Characterization of two age-induced intracisternal A-particle-related transcripts in the mouse liver. Transcriptional read-through into an open reading frame with similarities to the yeast ccr4 transcription factor. The Journal of biological chemistry 272, 5995-6003.
Ramsey, K.M., Mills, K.F., Satoh, A., and Imai, S. (2008). Age-associated loss of Sirt1-mediated enhancement of glucose-stimulated insulin secretion in beta cell-specific Sirt1-overexpressing (BESTO) mice. Aging cell 7, 78-88.
Reinke, H., Saini, C., Fleury-Olela, F., Dibner, C., Benjamin, I.J., and Schibler, U. (2008). Differential display of DNA-binding proteins reveals heat-shock factor 1 as a circadian transcription factor. Genes & development 22, 331-345.
Saito, K., Ohta, Y., Sami, M., Kanda, T., and Kato, H. (2010). Effect of mild restriction of food intake on gene expression profile in the liver of young rats: reference data for in vivo nutrigenomics study. The British journal of nutrition 104, 941-950.
Scobie, K.N., Hall, B.J., Wilke, S.A., Klemenhagen, K.C., Fujii-Kuriyama, Y., Ghosh, A., Hen, R., and Sahay, A. (2009). Kruppel-like factor 9 is necessary for late-phase neuronal maturation in the developing dentate gyrus and during adult hippocampal neurogenesis. The Journal of neuroscience : the official journal of the Society for Neuroscience 29, 9875-9887.
Soto-Vega, E., Richaud-Patin, Y., and Llorente, L. (2005). Human leukocyte antigen class I, class II, and tumor necrosis factor-alpha polymorphisms in a healthy elder Mexican Mestizo population. Immun Ageing 2, 13.
Swindell, W.R. (2008a). Comparative analysis of microarray data identifies common responses to caloric restriction among mouse tissues. Mechanisms of ageing and development 129, 138-153.
Swindell, W.R. (2008b). Genes regulated by caloric restriction have unique roles within transcriptional networks. Mechanisms of ageing and development 129, 580-592.
Swindell, W.R. (2009). Genes and gene expression modules associated with caloric restriction and aging in the laboratory mouse. BMC genomics 10, 585. van der Veer, E., Ho, C., O'Neil, C., Barbosa, N., Scott, R., Cregan, S.P., and Pickering, J.G. (2007). Extension of human cell lifespan by nicotinamide phosphoribosyltransferase. The Journal of biological chemistry 282, 10841-10845.
Vardimon, L., Ben-Dror, I., Havazelet, N., and Fox, L.E. (1993). Molecular control of glutamine synthetase expression in the developing retina tissue. Developmental dynamics : an official publication of the American Association of Anatomists 196, 276-282.
Xiang, X.X., Chen, L., Wang, J.H., Zhang, Y.B., and Zhang, D.Y. (2011). [Role of Wnt/beta-catenin signaling in aging of mesenchymal stem cells of rats]. Zhejiang da xue xue bao Yi xue ban = Journal of Zhejiang University Medical sciences 40, 630-640.
Yan, J., Wang, H., Liu, Y., and Shao, C. (2008). Analysis of gene regulatory networks in the mammalian circadian rhythm. PLoS Comput Biol 4, e1000193.
Yano, M., Falvey, E., and Gonzalez, F.J. (1992). Role of the liver-enriched transcription factor DBP in expression of the cytochrome P450 CYP2C6 gene. Molecular and cellular biology 12, 2847-2854.
Zhang, E.E., Liu, A.C., Hirota, T., Miraglia, L.J., Welch, G., Pongsawakul, P.Y., Liu, X., Atwood, A., Huss, J.W., 3rd, Janes, J., et al. (2009). A genome-wide RNAi screen for modifiers of the circadian clock in human cells. Cell 139, 199-210.
Table: Genes exhibiting enrichment for differentially expression during juvenile growth, aging, DR and are circadian ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Table: Coexpressed genes of the juvenile up-, aging down- and DR up-regulated circadian genes ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Table : LA induced genes in liver of young rats (1.5-fold) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Figure: Bidirectional comparisons of DR, aging and juvenile signatures ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Figure: Intersections of circadian genes with DR, aging and juvenile ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Figure: Overlap of the transcriptional signatures common to juvenile growth, aging, DR and circadian clock (CC) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Figure: Changes in circadian clock rhythmicity as a mediator of development, juvenile growth, aging and its modulation by diet ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Feed
Comment on This Data Unit