Caloric Restriction And Erc Formation

Caloric restriction has been shown to extend the life span in a wide number of organisms (68-71). Growth on low glucose has been used as a model for caloric restriction in yeast and was found to extend the life span (72,73). Yeast are normally grown in medium containing 2% glucose. Growth on 0.1% glucose results in a 55% extension of the median life span (72); growth on 0.5% glucose results in a more modest 24% extension in the life span (73). In yeast, glucose activates the cAMP-dependent protein kinase pathway. Components of this pathway include the GTP/GDP-binding proteins (Ras1 and Ras2 proteins), and a GTP-GDP exchange factor (Cdc25 protein) that interacts with the Ras proteins (74,75). Mutations in CDC25 and several other genes within the protein kinase A (PKA)-signalling pathway that decreased PKA activity have been found to extend the life span (73). Growth on low glucose does not further extend the life span of the long-lived cdc25 mutant, suggesting that low glucose and low PKA activity are functioning through the same pathway. In the long-lived cdc25 mutant, rDNA recombination rates were decreased and ERC levels were decreased in the old cdc25 mutant cells. These results support that the life span extension produced by caloric restriction/PKA pathway is mediated through rates of ERC accumulation.

Mutations in other genes in the pathways governing carbon source utilization also show similar effects on ERC accumulation and the life span. Snf1p is a protein kinase that plays a central role in the response to glucose starvation (76). Sip2p binds to Snf1p (77,78), and is believed to function as an adaptor molecule between Snf1p and some of its substrates (79). Deletion of S1P2 causes accelerated aging in yeast, as evidenced by a decreased life span and early onset of the age associated phenotypes of sterility, nucleolar fragmentation, and Sir3p relocal-ization to the nucleolus (80). Markedly increased levels of ERCs have been observed in the S1P2 deletion when the cells were sorted out to 12 generations.


Among the initial observations that lead to the identification of the nucleolus as a site of genomic instability was the relocalization of the SIR proteins to the nucleolus during yeast aging (31,81). Deletions of SIR3 or SIR4 cause a modest decrease in the life span (31,81-83). This decrease in the life span in SIR3 or SIR4 deletions is caused by the simultaneous expression of both a and a mating-type information, as the deletions have no effect on the life span in diploid yeast, and the decrease in the life span caused by deletion of SIR3 or SIR4 requires the presence of the normally silenced mating-type locus. The sir3hmla or sir4hmla doubly mutant strains have normal life spans (82). SIR2 deletion has a more pronounced effect on the life span, causing a decrease in the median life span of 50% (82,83). This effect was not mediated through the silenced mating-type locus, as the decrease in the life span was still present in diploid strains and was unaffected by the deletion of the silenced mating-type locus (82). SIR2 overexpression results in a 30% extension of the median life span.

That the effects of SIR mutations on the life span were mediated through ERC accumulation is supported by the following three observations: (a) SIR4 deletion increases the rate of recombination at the rDNA. This increase in recombination requires the presence of the silenced mating-type locus. (b) Increased levels of ERCs are seen young and old SIR2 deletion strains. (c) Mutation of fob1, which markedly decreases rDNA recombination and ERC formation, suppresses the life span defect of a SIR2 deletion (82).

The analysis of the role of additional proteins involved in silencing in yeast also supports the hypothesis that increased silencing in the rDNA is associated with life span extension. Deletion in the yeast gene ZDS1, which interacts with Sir2, 3, and 4 proteins in a yeast two-hybrid assay (84), causes an increase in rDNA silencing and an increase in the life span (85). Deletion of ZDS2, which shares 34% sequence identity to ZDS1 (86,87), decreases rDNA silencing and decreases the life span (85).

Transcriptional silencing in yeast is mediated, in part, by covalent modification of core histones by acetylation. The acetylation of core histones is re-versibly catalyzed by histone acetylases and deacetylases (reviewed in ref. 87). Yeast RPD3 and HDA1 encode histone deacetylases, mutations that lead to his-tone hyperacetylation. In yeast, rpd3 or hda1sir3 deletions were found to extend the life span (83).

At first this might seem to conflict with preceding observations that an increase in silencing in the rDNA is associated with life span extension—acetyla-tion of histones is associated with transcriptionally active chromatin, and a deletion of a histone deacetylase should therefore result in an increase in transcription. An increase in rDNA silencing has been seen in a rpd3 deletion (83,88) and in a hda1sir3 deletion (83). Why deletion of a histone deacetylase should increase rDNA silencing is unclear, but it may be mediated through indirect effects on transcription at other sites.


Although increased levels of ERCs were first noticed in a yeast deleted for SGS1, and it then was proposed that the accelerated aging phenotype seen in sgsl deletions was due to increased recombination at the rDNA (60), it has been difficult to demonstrate that ERCs accumulate at a greater rate in the presence of an sgsl deletion. The state of the rDNA has been examined in nonsorted (or sorted to seven generations) of wild-type and sgs1 deletion cells, no obvious increase in the intensity of extrachromosomal rDNA bands has been seen (89,90).

Failure to detect increased levels of ERCs in sgsl mutants is in part explained by the finding that the effects of the sgs1 deletion on the life span are complex (91). The terminal arrest morphology of cells undergoing life span determination has been examined in wild-type and sgsl mutant cells. In wild-type cells, 60% of cells arrest with large, unbudded morphology (91). In an sgsl deletion, only 20% of the cells arrest with large, unbudded morphology.

The effect of age on arrest state also has been examined. Growth arrest morphology is not influenced by age in wild-type cells: 60-70% arrest with the large, unbudded morphology regardless of whether they had arrested at less than 70% of maximal life span or greater than 70% maximal life span. This was not true when the effect of age on the growth-arrest phenotype of an sgsl mutant was examined: 65% of cells that had lived less than 70% of maximal life span arrested as large cells with buds, whereas of those cells that had survived to greater than 70% of the maximal life span, only 15% arrested as large cells with buds. Old sgsl mutant cells tend to arrest as large unbudded cells; that is, they arrest with the same terminal growth-arrest phenotype of as wild-type cells.

Distribution of terminal phenotypes in the sgsl mutant is most consistent with there being two components. The first component is a G2 arrest (cells with buds) that is stochastic and age independent, whereas the second component is a G1 arrest (unbudded cells) that is age dependent and likely related to wild-type yeast aging (91). The age-associated death rate in young cells is low, so the fixed, age-independent component predominates, and most cells arrest with budded morphology. As cells age, the age-associated component increases and eventually predominates, so cells increasingly arrest with unbudded morphology.

That old sgsl mutants age similarly to wild-type cells is supported by the analysis of the effect of combining mutations that increase the life span with the sgsl deletion. SIR2 overexpression (82) or fobl deletion (67) extend the life span in wild-type SGSl strains. When the life span of sgslfobl strains or sgsl dele tion-SIR2 overexpression strains was determined, a prolongation of the maximum life span was observed with little affect on survival within the early life span (91). This is consistent with these mutations remaining capable of modulating the age-associated death rate and not affecting age-independent death rate in sgsl mutant strains.

Does sgsl mutation accelerate yeast aging? Criteria that can be analyzed on an individual cell basis during the manual determination of the life span micro-scopically—sterility, cell enlargement, nucleolar fragmentation—are accelerated by sgsl deletion (49). These analyses can be performed on sgsl mutant cells near-ing the end of their life span at a time when the age-dependent mortality rate predominates (91). Quantitation of ERC levels can only be performed on bulk populations of yeast cells, not on individual cells, and is biased towards cells earlier in their life spans (discussed in detail below). At earlier times, the age-independent (non-ERC) mortality rate predominates.


If the increased rates of rDNA recombination observed in sgsl mutants cause an increase in the rate of ERC formation, what is the expected change in the level of ERCs compared with an isogenic, wild-type SGS1 strain? This question is not simple to answer given the complex effects of the sgsl mutation discussed above. Using a computer model to describe the accumulation of ERCs in response to changing rates of production, it is possible crudely to estimate what the expected change in ERC level in an sgsl mutation.

In addition, the experimental quantitation of ERC levels is an intricate task. ERCs can be seen in Southern blots of total chromosomal DNA prepared from logarithmically growing yeast cultures (80,83,90), but extrachromosomal species represent only a small fraction of the total ribosomal DNA. This is believed to be due to the preferential accumulation of ERCs in old mother cells—old mother cells are rare in a growing culture. The fraction of cells that have survived long enough to accumulate significant numbers of ERCs is small—hence the levels of ERCs in a logarithmically growing culture are low.

It is also possible to analyze ERC accumulation as a function of age by studying age-sorted populations of yeast cells (10,60,67,80). Producing a sorted population of old cells is a four-step process, each step of which must be performed with great care (Fig. 4). This procedure takes advantage of the fact that the cell wall of the bud is synthesized de novo. The biotin that is attached to the surface of the mother cell will stay with the mother cell throughout its life span and is not transferred to any daughter cells. By sorting cells in this manner, it is possible to examine phenotypes—such as level of ERCs or gene expression—in populations of yeast cells of a crudely defined age.

10s cells

10s cells

Figure 4 Purification of old yeast cells. Logarithmically growing yeast cells are chemically labeled on their surface with biotin and allowed to grow 7-12 divisions. Cells are harvested by first incubating them with paramagnetic beads that have streptavidin covalently attached, and then by separating the beads with the "old" yeast cells attached using a magnet. The vast excess of "young" yeast cells are discarded. (Adapted from Ref. 101.)

Figure 4 Purification of old yeast cells. Logarithmically growing yeast cells are chemically labeled on their surface with biotin and allowed to grow 7-12 divisions. Cells are harvested by first incubating them with paramagnetic beads that have streptavidin covalently attached, and then by separating the beads with the "old" yeast cells attached using a magnet. The vast excess of "young" yeast cells are discarded. (Adapted from Ref. 101.)

ERC levels as a function of age are determined by performing a Southern blot, probed with an rDNA probe, on total genomic DNA prepared from the age-sorted cells. Although detecting twofold differences is within the level of quanti-tation of single bands on Southern blots, ERCs exist as multiple species of super-coiled episomes containing one or more copies of the 9.1-kb rDNA repeat (60). To quantitate ERC levels on a Southern blot, multiple bands need to be analyzed, decreasing the accuracy of the analysis. In addition, quantitation of ERCs is more sensitive than a standard genomic Southern blot to nicking or shearing of the DNA during preparation, as these will cause the supercoiled forms of ERCs to decrease and additional bands representing relaxed or linear forms to appear.

Quick Weight Loss Action

Quick Weight Loss Action

Why Indulge In Self-Pity When You Can Do Something About Your Weight Now. Say Goodbye to Your Weight Problems That Have Only Make Your Life Nothing But Miserable. Have you often felt short-changed because of your weight or physical appearance?

Get My Free Ebook

Post a comment