Over the past decade research into the molecular biology of ageing in model organisms revealed that "ageing" shares common denominators indicating conserved mechanisms over different taxa. Calorie or dietary restriction without malnutrition is the most powerful evidence for this observation, because it is the only non-genetic intervention which extends lifespan of organisms across taxa, i.e. yeast, worms, fruitfly, and rodents. Additionally, species- specific pathways have evolved (1).
Yeast is fortunate enough to have two lifespans. Chronological lifespan measures the time of survival in a non-dividing state. In contrast, the more widely studied replicative lifespan measures the number of replications a mother cell undergoes before it senesces, and is regarded as a model for mitotically active cells. Restricting glucose availability has been shown to extend both chronological and replicative lifespan (1), although most studies on calorie restriction in yeast have focused on replicative lifespan.
Why yeast cells age has not been unambiguously identified yet. There is evidence that accumulation of oxidatively damaged proteins in mother cells induce senescence over the replicative cycles, because the damaged proteins are selectively retained in the mother cell (2;3). In addition, a popular and long-standing theory suggests that ageing is caused by accumulation of extrachromosomal rDNA circles (ERCs), which arise during homologous recombination of rDNA-tandem repeats. Each of these recombined repeats is excised and forms a little circle, not unlike plasmids. Since they contain an origin of replication, they autonomously replicate during each cell division, and thus reach high numbers in old yeast cells; in fact their DNA content can outweigh the yeast genome. Again they are retained in mother cells, and it is thought that they titrate housekeeping factors away from the chromosomes, which leads to decreasingly strict maintenance of the genome (4).
Today, three pathways linking calorie restriction to lifespan extension have been suggested in yeast (Figure 1). The first two propose that Sir2, an NAD+-dependent histone deacetylase which is required for gene silencing, is the effector via which calorie restriction exerts its beneficial effect on lifespan. Model I proposes that Sir2 senses shifts in nutrient-sources due to changes in the ratio of intracellular NAD+, which increases when yeast switch from fermentation (the rather inefficient energy-production when nutrients are abundant) to respiration, hence increasing the NAD+ to NADH-ratio. This leads to increased Sir2 activity and hence more silencing, thus increasing genome stability and reducing the ERC-formation, which slows ageing and thus extends lifespan (5). Model II explains the increase in Sir2-activity via the upregulation of the NAD+-salvage pathway in the nucleus which leads to a decrease in nicotinamide, an inhibitor of Sir2. This again results in Sir2-upregulation and feeds in the same pathway as does model I (6).
Model I and II are widely accepted as mechanisms for lifespan extension. Model III however omits Sir2 completely, and there is evidence that the Tor-pathway, and Sch9, the yeast PKB/AKT- homologue coordinate responses to calorie restriction and control lifespan extension. However, the further downstream mechanisms are still a black box, and it is not clear whether these pathways also modulate events at the rDNA (7;8).
In order to assess changes in gene silencing in response to calorie restriction, we are using a simple, but in its read-out very sensitive silencing assay. Because Sir2 is the only histone deacetylase required for silencing in the telomeres, the mating type locus and the rDNA, we chose to first assess its activity in the telomeres. Therefore, we worked with a reporter strain which has the marker gene URA3 inserted in the subtelomeric region. URA3 codes for an enzyme which is involved in the pyrimidine biosynthetic pathway. It converts 5'-Fluororotic acid into 5'-Fluorouracil, which is highly toxic for cells. In wildtype cells, where Sir2 is expressed and active, this gene is silenced and cells will grow on FOA-supplemented media. In a Sir2-deletion strain however, silencing is abolished, therefore URA3 is expressed, and the cells die on FOA-supplemented media. This simple assay allows for a qualitative and quantitative analysis of influences which directly act on Sir2, such as calorie restriction. (Fig 2.)
Cells are then grown on normal and FOA-containing media which are supplemented with the following range of glucose-concentrations: 0.05% and 0.5% for severe and moderate calorie restriction, and 20% for high external osmolarity, which was also shown to extend lifespan via Sir2 upregulation (9), and 2% glucose, which is the standard growth condition.
Additionally to the parent strain, we constructed a set of strains with gene deletions which are all reportedly involved in modulation of lifespan. The deletion of the replication fork barrier protein Fob1 is thought to extend lifespan via the reduction of ERCs. The deletion of the gene for hexokinase isoenzyme 2 HXK2 is generally regarded as the mimic for calorie restriction and also extends lifespan. The deletion of Sir2 leads to reduction in lifespan.
Our data show that Sir2-activity does not change in response to calorie restriction. Therefore, lifespan extension by calorie restriction in yeast is not due to an increase in Sir2 activation. However, we found increased gene silencing in the fob1Δ and hxk2Δ strains. The effect of calorie restriction in the hxk2Δ strain is an important finding, because it indicates that this deletion has additional effects on an organism and it does not exactly mimic calorie restriction. The effect in fob1Δ could be due to a redistribution of Sir2 away from the rDNA, where it is normally recruited to by Fob1.
In order to exclude locus-specific effects at the telomeres, and to assess whether model III provides better evidence for lifespan extension by calorie restriction, we are currently performing a series of silencing assays using a reporter strain with the marker inserted in the rDNA locus. Preliminary data that calorie restriction does not increase silencing at rDNA either, implying that the Sir2-independent pathway of model III may be correct. Future experiments are aimed at understanding the molecular mechanisms which lead from calorie restriction to lifespan extension.
Figure 1. Three proposed models connecting calorie restriction with lifespan extension in yeast. Model I and II suggest an upregulation of the histone deacetylase Sir2 via two different pathways. Model III has not been well defined mechanistically; however, the TOR and SCH9 kinase pathways have been suggested to be involved.
Figure 2. Sir2 activity regulates silencing of the URA3 reporter gene. In the wildtype strain, Sir2 silences the reporter gene introduced in the subtelomeric region and the cells survive on FOA-containing medium. In the Sir2 deletion strain, URA3 is expressed, and cells die on FOA.
