Caloric restriction extends yeast chronological lifespan via a mechanism linking cellular aging to cell cycle regulation, maintenance of a quiescent state, entry into a non-quiescent state and survival in the non-quiescent state

A yeast culture grown in a nutrient-rich medium initially containing 2% glucose is not limited in calorie supply. When yeast cells cultured in this medium consume glucose, they undergo cell cycle arrest at a checkpoint in late G1 and differentiate into quiescent and non-quiescent cell populations. Studies of such differentiation have provided insights into mechanisms of yeast chronological aging under conditions of excessive calorie intake. Caloric restriction is an aging-delaying dietary intervention. Here, we assessed how caloric restriction influences the differentiation of chronologically aging yeast cultures into quiescent and non-quiescent cells, and how it affects their properties. We found that caloric restriction extends yeast chronological lifespan via a mechanism linking cellular aging to cell cycle regulation, maintenance of quiescence, entry into a non-quiescent state and survival in this state. Our findings suggest that caloric restriction delays yeast chronological aging by causing specific changes in the following: 1) a checkpoint in G1 for cell cycle arrest and entry into a quiescent state; 2) a growth phase in which high-density quiescent cells are committed to become low-density quiescent cells; 3) the differentiation of low-density quiescent cells into low-density non-quiescent cells; and 4) the conversion of high-density quiescent cells into high-density non-quiescent cells.


INTRODUCTION
A body of knowledge about mechanisms underlying chronological aging of the yeast Saccharomyces cerevisiae has been provided by studies in which yeast cells were cultured in a nutrient-rich liquid medium initially containing 2% glucose [1,2]. Under these so-called noncaloric restriction (non-CR) conditions yeast cells are not limited in the supply of calories [1,3,4]. When glucose is exhausted at the diauxic shift, cells in a non-CR yeast culture undergo arrest at the G 1 phase of the cell cycle. The non-CR yeast culture then differentiates into several cell populations [5][6][7][8].
One of these cell populations is a population of quiescent (Q) cells; these cells exist in a distinct nonproliferative state called G 0 [5][6][7][8][9][10][11]. Q cells are mainly daughter cells [5][6][7]. They are unbudded and uniformly sized, are refractive by phase-contrast microscopy and enclosed by a rigid cell wall, have high buoyant density, store glycogen and trehalose in bulk quantities, are highly metabolically active, exhibit high rates of mitochondrial respiration and low concentrations of reactive oxygen species (ROS), are able to form colonies when plated on fresh solid medium, can re-enter mitosis when nutrients become available following transfer to fresh liquid medium, are resistant to long-term thermal and oxidative stresses, exhibit low rates of mutations that impair
The differentiation of a non-CR yeast culture following glucose exhaustion at the diauxic shift also yields at least three subpopulations of non-quiescent (NQ) cells, most or all of which are first-and higher-generation mother cells [5-8, 10, 11]. One subpopulation of NQ cells consists of metabolically active cells that exhibit high reproductive (colony-forming) capacities, high ROS concentrations, impaired mitochondrial respiration and elevated frequencies of mutations impairing mitochondrial functionality [5-8, 10, 11]. Another subpopulation of NQ cells includes metabolically active cells that are impaired in reproductive (clonogenic) ability and are likely to be descended from NQ cells of the first subpopulation [5-8, 10, 11]. The third subpopulation of NQ cells is composed of cells that exhibit hallmarks of the apoptotic and/or necrotic modes of PCD and may derive from NQ cells of the second subpopulation [5-8, 10, 11].
In response to a depletion of glucose (as well as nitrogen, phosphate or sulfur), a signaling network of certain proteins and protein complexes orchestrates cell cycle arrest at the G 1 phase of the cell cycle, the differentiation of a chronologically aging non-CR yeast culture into populations of Q and NQ cells, and quiescence maintenance. Proteins and protein complexes integrated into this signaling network operate as network nodes, many of which are connected by physical links known to be predominantly phosphorylations and dephosphorylations that activate or inhibit specific target proteins [9,[12][13][14][15][16][17]. The core hubs of this signaling network of a quiescence program are four nutrient-sensing protein complexes, each of which exhibits a protein kinase activity and modulates many downstream effector proteins integrated into the network. These core hubs of the network are: 1) TORC1 (target of rapamycin complex 1), a key regulator of cell metabolism, growth, division and stress resistance in response to changes in the availabilities of nitrogen and carbon sources; 2) PKA (protein kinase A), an essential controller of cell metabolism, proliferation and stress resistance in response to changes in carbon source availability; 3) Snf1 (sucrose non-fermenting, protein 1), a heterotrimeric protein complex required for cell growth support and energy homeostasis maintenance after glucose exhaustion; and 4) Pho85 (phosphate metabolism, protein 85), a protein kinase associated with various cyclins to promote phosphate metabolism, glycogen and trehalose synthesis, oxidative stress response and cellular proteostasis in response to changes in the accessibility of a phosphate source or following glucose exhaustion [9,12,14,18,19].
CR is a dietary intervention that delays aging in yeast and other evolutionarily distant eukaryotic organisms [1,2,[102][103][104][105][106]. Although mechanisms linking yeast chronological aging to the quiescence program under non-CR conditions are well established [5-9, 11, 43, 44, 97, 107], it is unknown if the aging-delaying effect of CR in chronologically aging yeast is due in part to the ability of this low-calorie diet to control the quiescence program. In this study, we provide evidence that CR slows down yeast chronological aging through a mechanism which affects several key aspects of the quiescence program.

CR accelerates an age-related accumulation of low-density cells in chronologically aging yeast cultures
Chronologically aging yeast cultured under CR conditions in the nutrient-rich YP medium initially containing 0.2% glucose lived significantly longer than yeast undergoing chronological aging under non-CR conditions in YP medium initially supplemented with 2% glucose ( Figure 1A). Even before glucose is depleted from the medium at the diauxic shift, a yeast culture undergoing chronological aging under non-CR conditions is known to differentiate into cell populations with different buoyant densities [6,8]. Each of these cell populations can be purified to homogeneity by centrifugation in Percoll density gradient [5].
Using centrifugation in Percoll density gradient for separating cell populations exhibiting different densities, we investigated how CR influences the differentiation of a chronologically aging yeast culture into these distinct cell populations. Samples of cells cultured in YP medium initially containing 0.2% glucose (CR conditions) or 2% glucose (non-CR conditions) were recovered at different time-points following glucose exhaustion from the medium, which occurs 16 or 22 h after cell inoculation, respectively ( Figure 1B). We found that chronologically aging cultures of CR and non-CR yeast 1) contain both low-density (LD) and high-density (HD) cell populations during logarithmic (L), diauxic (D), post-diauxic (PD) and stationary (ST) growth phases, i.e. through the entire chronological lifespan; and 2) exhibit an age-related increase in the percentage of LD cells (Figures 1C and  1D). We also noticed that the percentage of LD cells in CR yeast cultures significantly exceeds that in non-CR yeast cultures during any of the four growth phases, i.e. at any stage of the chronological aging process ( Figure  1C and 1D). Moreover, we showed that the percentage of LD cells in CR yeast cultures reaches a plateau on day 5 of culturing (in PD growth phase), whereas the percentage of LD cells in non-CR yeast cultures attains a steady-state level only on day 10 of culturing (in ST growth phase) ( Figure 1C and 1D).
Collectively, these findings indicate that CR accelerates an age-related accumulation of LD cell population in chronologically aging yeast cultures.

CR alters cell size and the abundance of budded cells in LD and HD populations
We then assessed how CR influences the morphology of cells present in LD and HD populations. These cell populations were first recovered from yeast cultures of different chronological ages and then separated from each other by centrifugation in Percoll density gradient. Our differential interference contrast (DIC) microscopical examination and subsequent morphometric analysis of these LD and HD cell populations have revealed that through the entire chronological lifespan: 1) LD cells in CR cultures have sizes similar to those of LD cells in non-CR cultures ( Figure 1E Figure 1). We also found that, with the exception of HD cells recovered from non-CR cultures in L growth phase, most of HD cells recovered from CR and non-CR cultures were unbudded ( Figure 1E and 1I; Supplementary Figure 1). Moreover, we showed that the percentage of budded HD cells in CR cultures is significantly lower than that in non-CR cultures ( Figure 1E and 1I; Supplementary Figure 1).
In sum, these findings indicate that CR decreases the size of HD cells and lowers the percentage of budded Oncotarget 69331 www.impactjournals.com/oncotarget cells in both LD and HD populations through the entire process of chronological aging.

CR delays an age-related decline in the reproductive proficiencies of LD and HD cell populations and in their abilities to synchronously re-enter mitosis
Our observation that nearly all HD cells in CR and non-CR cultures were unbudded (except of HD cells purified from non-CR cultures in L growth phase, see Figure 1I) suggests a hypothesis that HD population contains predominantly Q cells. Q cells are known to be mainly unbudded [5]. Moreover, because LD population recovered from CR and non-CR cultures contained both budded and unbudded cells ( Figure 1H), we hypothesized that LD population consists mainly of NQ cells. NQ cells are known to represent a mixture of budded and unbudded cells [5].
To test this hypothesis, we compared the following two features of LD and HD populations: 1) the reproductive competence, i.e. the ability of a yeast cell to form a colony when plated on fresh solid medium; and 2) the ability of a yeast cell population to synchronously re-enter the mitotic cell cycle if returned containing 0.2% glucose (CR conditions) or 2% glucose (non-CR conditions). The logrank test for comparing each pair of survival curves was performed as described in Materials and Methods. Two survival curves were considered statistically different if the P value was less than 0.05. B. Kinetics of glucose consumption for WT yeast cultured in YP medium initially containing 0.2% glucose (CR conditions) or 2% glucose (non-CR conditions). C. to I. Samples of WT yeast cultured in YP medium initially containing 0.2% glucose (CR conditions) or 2% glucose (non-CR conditions) were recovered from logarithmic (L), diauxic (D), post-diauxic (PD) or stationary (ST) growth phase and subjected to centrifugation in Percoll density gradient as described in Materials and Methods. Percoll density gradients (C), the percentage of LD cells (D), differential interference contrast micrographs of LD and HD cells recovered from ST growth phase on day 7 (E), mean diameters of LD (F) or HD (G) cells, and the percentages of budded cells present in LD (H) or HD (I) populations are shown. Data in A, B, D, F -I are presented as means ± SEM (n = 3; *P < 0.05; **P < 0.01; ***P < 0.001; ns, not significant).
Oncotarget 69332 www.impactjournals.com/oncotarget to growth-promoting conditions. Both these abilities are known to be characteristic of Q cells; these fundamental characteristics of Q cells distinguish them from NQ cells [7]. We recovered LD and HD cell populations from CR or non-CR yeast cultures of different chronological ages and then separated them from each other by centrifugation in Percoll density gradient.
Our comparative analyses of the reproductive (colony-forming) capacities of LD and HD cells purified from differently aged CR and non-CR yeast cultures have revealed the following: 1) LD cells in CR cultures maintain reproductive (colony-forming) capacity for a long period of time in the course of chronological aging, whereas LD cells in non-CR cultures exhibit a very rapid age-related deterioration in reproductive competence (Figure 2A), and 2) HD cells in CR cultures sustain reproductive (colony-forming) ability through the entire process of chronological aging, whereas HD cells in non-CR cultures display an age-related gradual decline in reproductive capability ( Figure 2B).
Our comparison of the abilities of LD and HD cells purified from differently aged CR and non-CR yeast cultures to synchronously re-enter the mitotic cell cycle after cell transfer into fresh medium and incubation for 1 to 4 h has revealed the following: 1) LD cells in CR cultures remain Q cells (i.e. maintain the ability to synchronously re-enter mitosis following cell transfer) for a long period of time in the course of chronological aging, whereas LD cells in non-CR cultures become NQ cells (i.e. senescent) soon after being formed in the process of differentiation ( Figure 2C); and 2) HD cells in CR cultures continue to be Q cells through most of the chronological aging process, whereas HD cells in non-CR cultures progressively become NQ cells ( Figure 2D).
Taken together, these findings provide evidence that 1) HD population contains predominantly Q cells, whereas Oncotarget 69333 www.impactjournals.com/oncotarget LD population consists mainly of NQ cells; and 2) CR slows an age-related deterioration in the reproductive competences of Q and NQ cells and in their capabilities to synchronously re-enter the mitotic cell cycle when nutrients become available.

CR increases the abundance of daughter cells in Q and NQ populations
A culture of the budding yeast S. cerevisiae is known to contain first-and higher-generation daughter cells without bud scars and first-and higher-generation mother cells with one or more bud scars on the cell surface [108][109][110]. Each bud scar marks a division site on the surface of a mother cell and, thus, the number of bud scars accumulating on the mother cell surface is a measure of its replicative age [111][112][113].
We assessed the relative abundance of first-and higher-generation daughters and replicatively aged mothers present in Q and NQ cell populations. These cell populations were first recovered from CR or non-CR yeast cultures of different chronological ages and then separated from each other by centrifugation in Percoll density gradient. Bud scars were microscopically visualized by staining with Calcofluor White M2R. Our comparative analysis of Q and NQ cell populations purified from differently aged CR and non-CR cultures has revealed the following: 1) CR increases the abundance of first-and higher-generation daughters and decreases the percentage of first-and higher-generation mothers present in NQ cell populations through the entire chronological lifespan ( Figure   Collectively, these findings indicate that 1) through the entire chronological lifespan, CR rises the fraction of daughter cells in Q and NQ populations; and 2) late in chronological lifespan, CR prevents budding of daughter cells in NQ population and, to a lesser extent, in Q population.

CR increases the concentrations of glycogen and trehalose in Q and NQ cell populations
One of the metabolic hallmarks of Q cells that distinguish them from NQ cells is an accumulation of glycogen and trehalose, the two major glucose stores in yeast [5,114]. In addition to being reserve carbohydrate, trehalose has also been implicated in protecting yeast cells and cellular proteins from oxidative and other stresses [115][116][117][118], modulating cellular proteostasis [119][120][121], and allowing Q cells to re-enter the mitotic cell cycle when nutrients become available [114].
We investigated how CR influences the intracellular concentrations of glycogen and trehalose in Q and NQ cell populations. These cell populations were first recovered from CR or non-CR yeast cultures of different chronological ages and then separated from each other by centrifugation in Percoll density gradient. Our comparative analysis of Q and NQ cell populations purified from differently aged CR and non-CR cultures has revealed the following: 1) through most of the chronological lifespan (i.e. since day 3 of culturing), the concentrations of glycogen in NQ and Q cells from CR cultures significantly exceed those in NQ and Q cells from age-matched non-CR cultures ( Figure 4A and 4B, Figure 4: CR alters the abundance of glycogen, trehalose, triacylglycerols (TAG) and cardiolipins (CL) in Q and NQ cell populations through most of the chronological lifespan. Samples of WT yeast cultured in YP medium initially containing 0.2% glucose (CR conditions) or 2% glucose (non-CR conditions) were recovered from L, D, PD or ST growth phase and subjected to centrifugation in Percoll density gradient to purify Q and NQ cell populations, as described in Materials and Methods. The concentrations of glycogen A. and B., trehalose C. and D., TAG E. and F. and CL G. and H. were measured as described in Materials and Methods. Data are presented as means ± SEM (n = 3; *P < 0.05; **P < 0.01; ***P < 0.001).
Oncotarget 69335 www.impactjournals.com/oncotarget respectively); 2) through the entire chronological lifespan, the concentrations of glycogen in Q cells from CR cultures were higher than those in NQ cells from CR cultures of the same chronological age (compare Figure 4A and 4B); 3) for the most part (i.e. since day 5 of culturing), the concentrations of trehalose in NQ and Q cells from CR cultures considerably surpass those in NQ and Q cells from age-matched non-CR cultures ( Figure 4C and 4D, respectively); and 4) through the entire chronological lifespan, the concentrations of trehalose in Q cells from CR cultures exceed those in NQ cells from CR cultures of the same chronological age (compare Figure 4C and 4D).
In sum, these findings indicate that through most of the chronological aging process CR elicits significant rises in the concentrations of glycogen and trehalose in both Q and NQ cell populations.
We examined the effect of CR on the abundance of TAG and CL in Q and NQ cell populations that were first recovered from CR or non-CR yeast cultures of different chronological ages and then purified by centrifugation in Percoll density gradient. Our comparison of these Q and NQ cell populations has revealed the following: 1) through the entire chronological lifespan, the concentrations of TAG in NQ and Q cells from CR cultures are significantly lower than those in NQ and Q cells from age-matched non-CR cultures ( Figure 4E and 4F, respectively); 2) for the most part (i.e. since day 3 of culturing), the concentrations of CL in NQ and Q cells from CR cultures significantly exceed those in NQ and Q cells from non-CR cultures of the same chronological age ( Figure 4G and 4H, respectively); 3) through the entire chronological lifespan, the concentrations of TAG in Q and NQ cells from agematched CR and non-CR cultures are similar (compare Figure 4E and 4F); and 4) for the most part (i.e. since day 3 of culturing), the concentrations of CL in Q and NQ cells from CR and non-CR cultures of the same chronological age are also similar (compare Figure 4G and 4H).
We therefore concluded that through the entire chronological aging process CR induces a significant decline in the concentrations of TAG in Q and NQ cell were recovered from L, D, PD or ST growth phase and subjected to centrifugation in Percoll density gradient to purify Q and NQ cell populations, as described in Materials and Methods. The rate of mitochondrial respiration A. and B., ΔΨ m C. and D. and ROS E. and F. were measured as described in Materials and Methods. Data are presented as means ± SEM (n = 3; *P < 0.05; **P < 0.01; ***P < 0.001).
Oncotarget 69336 www.impactjournals.com/oncotarget populations. In contrast, through most of the chronological lifespan CR elicits a substantial rise in the abundance of CL in both these cell populations.

CR alters the dynamics of age-related changes in mitochondrial functionality in Q and NQ cells
Mitochondrial electron transport chain, electrochemical potential across the inner mitochondrial membrane (ΔΨ m ) and mitochondrial ROS play essential roles in defining longevity of chronologically aging yeast [1,2,4,125,135,[144][145][146][147][148][149][150][151][152][153]. Q cells purified from non-CR yeast cultures have been shown to exhibit high rates of mitochondrial respiration and low ROS, whereas NQ cells present in these cultures are known to have low rates of mitochondrial respiration and high ROS [5][6][7]154].
We therefore investigated the effect of CR on some longevity-defining traits of mitochondrial functionality in Q and NQ cell populations recovered from CR or non-CR yeast cultures of different chronological ages and purified by centrifugation in Percoll density gradient. These traits included the rate of mitochondrial respiration, ΔΨ m and cellular ROS, which in yeast are known to be produced primarily as by-products of mitochondrial respiration [155,156].
Our comparative analyses of these traits of mitochondrial functionality in Q and NQ cells purified from differently aged CR and non-CR yeast cultures have revealed the following: 1) through the entire chronological lifespan, the rate of mitochondrial respiration in NQ cells from CR cultures is significantly higher than that in NQ cells from age-matched non-CR cultures ( Figure 5A); 2) for the most part (i.e. since day 3 of culturing), the rate of mitochondrial respiration in Q cells from CR cultures substantially exceeds that in Q cells from age-matched non-CR cultures ( Figure 5B); 3) through most of the chronological lifespan (i.e. since day 3 of culturing), the rates of mitochondrial respiration in NQ and Q cells from CR cultures are considerably higher than those in NQ and Q cells from non-CR cultures of the same chronological age ( Figure 5C and 5D, respectively); 4) early in chronological lifespan (i.e. until day 7 for NQ cells or day 5 for Q cells) the concentrations of ROS in NQ and Q cells from CR cultures are notably lower than those in NQ and Q cells from age-matched non-CR cultures ( Figure  5E and 5F, respectively); and 5) late in chronological lifespan (i.e. after day 10 for NQ cells or day 7 for Q cells) the concentrations of ROS in NQ and Q cells from CR cultures considerably exceed those in NQ and Q cells from non-CR cultures of the same chronological age ( Figure 5E and 5F, respectively). Taken together, these findings show that CR creates were recovered from L, D, PD or ST growth phase and subjected to centrifugation in Percoll density gradient to purify Q and NQ cell populations, as described in Materials and Methods. Carbonylated cellular proteins A. and B., oxidatively damaged membrane lipids C. and D., the frequencies of spontaneous point mutations in the CAN1 gene of nDNA E. and F, and the frequencies of spontaneous point mutations in the RIB2 and RIB3 genes of mtDNA (G and H) were measured as described in Materials and Methods. Data are presented as means ± SEM (n = 3; *P < 0.05; **P < 0.01; ***P < 0.001).
Oncotarget 69337 www.impactjournals.com/oncotarget distinct patterns of mitochondrial functionality in Q and NQ cells by altering the chronology of age-related changes in mitochondrial respiration, ΔΨ m and ROS.

CR decreases the extent of age-related oxidative damage to proteins, lipids and DNA in Q and NQ cells
An age-related buildup of ROS-inflicted oxidative damage to various cellular macromolecules impairs their stability and is one of the major causes of aging in yeast and other eukaryotes [125, [157][158][159][160][161][162][163][164]. Q cells purified from non-CR yeast cultures are known to be genomically stable, whereas nuclear and mitochondrial DNA (nDNA and mtDNA, respectively) in NQ cells are likely instable because these cells exhibit an age-related excessive production of petite colonies [5,7,10]. Moreover, Q cells purified from non-CR yeast cultures do not accumulate SDS-insoluble protein aggregates, whereas NQ cells present in these cultures have been shown to amass such aggregates of irreversibly denatured/damaged proteins [106].
We therefore analyzed how CR influences the extent of age-related oxidative damage to cellular proteins, membrane lipids, nDNA and mtDNA in Q and NQ cell populations that were recovered from differently aged CR or non-CR yeast cultures and purified by centrifugation in Percoll density gradient. Our analyses have revealed the following: 1) through most of the chronological lifespan (i.e. since day 3 of culturing), the extent of oxidative damage to proteins in NQ cells from CR cultures is considerably lower than that in NQ cells from age-matched non-CR cultures ( Figure 6A); 2) late in chronological lifespan (i.e. since day 10) proteins in Q cells from CR cultures are oxidatively damaged to a lesser degree than proteins in Q cells from non-CR cultures of the same chronological age ( Figure 6B); 3) late in chronological lifespan (i.e. after day 7 for NQ cells or after day 10 for Q cells), the extent of oxidative damage to membrane lipids in Q and NQ cells from CR cultures is significantly decreased as compared to that in NQ cells from age-matched non-CR cultures ( Figure 6C and 6D, respectfully); 4) late in chronological lifespan (i.e. after day 10 for NQ cells or on day 18 for Q cells), the frequencies of spontaneous point mutations in the CAN1 gene of nDNA in Q and NQ cells from CR cultures are substantially lower than those in Q and NQ cells from agematched non-CR cultures -probably due to a decreased degree of oxidative damage to nDNA in Q and NQ cells from CR cultures ( Figure 6E and 6F); and 5) late in chronological lifespan (i.e. after day 10 for NQ cells or on day 18 for Q cells), the frequencies of spontaneous point mutations in the RIB2 and RIB3 genes of mtDNA in Q and NQ cells from CR cultures are markedly decreased as compared to those in Q and NQ cells from age-matched non-CR cultures -perhaps due to a decline in the extent of oxidative damage to mtDNA in Q and NQ cells from CR cultures ( Figure 6G and 6H).
In sum, these data imply that CR lessens the degree

Figure 7: CR increases the resistance of Q and NQ cells to long-term thermal and oxidative stresses. Samples of WT
yeast cultured in YP medium initially containing 0.2% glucose (CR conditions) or 2% glucose (non-CR conditions) were recovered from L, D, PD or ST growth phase and subjected to centrifugation in Percoll density gradient to purify Q and NQ cell populations, as described in Materials and Methods. Spot assays for monitoring thermal or oxidative stress resistance were performed as described in Materials and Methods. For chronic thermal stress, serial 10-fold dilutions of pure NQ A. and Q B. cells were spotted on plates with solid YEPD medium containing 2% glucose; these plates were initially exposed to 55 o C for 60 min, and then transferred to 30 o C and incubated for 3 days. For chronic oxidative stress, serial 10-fold dilutions of pure NQ (A) and Q (B) cells were spotted on plates with solid YEPD medium containing 2% glucose and 5 mM hydroxide peroxide; these plates were then incubated at 30 o C for 3 days.
Oncotarget 69338 www.impactjournals.com/oncotarget of age-related oxidative damage to proteins, lipids, nDNA and mtDNA in Q and NQ cells.
We assessed the effects of CR on the abilities of Q and NQ cell populations to resist chronic oxidative and thermal stresses. Q and NQ cell populations were recovered from differently aged CR or non-CR yeast cultures and purified by centrifugation in Percoll density gradient. Chronic thermal stress was administered by spotting pure Q and NQ cells on plates with solid YEPD medium containing 2% glucose, incubating these plates at 55 o C for 60 min, and then transferring them to 30 o C and incubating for 3 days. Chronic oxidative stress was applied by spotting pure Q and NQ cells on plates with solid YEPD medium containing 2% glucose and 5 mM hydroxide peroxide, and incubating them at 30 o C for 3 days. Our analyses have revealed the following: 1) early in chronological lifespan (i.e. on day 5 for thermal stress or on day 3 for oxidative stress), NQ cells from CR cultures become more resistant to both these kinds of chronic stresses as compared to NQ cells from age-matched non-CR cultures ( Figure 7A); and 2) late in chronological lifespan (i.e. since day 10), Q cells from CR cultures exhibit higher resistance to both oxidative and thermal stresses then Q cells from non-CR cultures of the same chronological age ( Figure 7B).
Collectively, these findings indicate that CR elicits a significant rise in the resistance of Q and NQ cells to long-term thermal and oxidative stresses. Q cells display the stimulatory effect of CR on the tolerance to both kinds of stresses only late in chronological lifespan, whereas in NQ cells such effect can be seen already early in life.

CR delays the onsets of the age-related apoptotic and necrotic modes of PCD in Q and NQ populations
In NQ cells from non-CR yeast cultures, the onsets of the age-related apoptotic and necrotic modes of PCD are known to occur much earlier in chronological lifespan glucose (non-CR conditions) were recovered from L, D, PD or ST growth phase and subjected to centrifugation in Percoll density gradient to purify Q and NQ cell populations, as described in Materials and Methods. Clonogenic assays for monitoring the susceptibilities of NQ (E and G) and Q (F and H) cells to the apoptotic (E and F) or liponecrotic (G and H) mode of PCD induced in response to a short-term (for 2 h) exposure to exogenous 2.5 mM hydrogen peroxide (E and F) or 0.15 mM POA (G and H), respectively, were performed as described in Materials and Methods. Data are presented as means ± SEM (n = 3; *P < 0.05; **P < 0.01; ***P < 0.001).
We therefore investigated how CR influences the onsets of age-related the apoptotic and necrotic modes of PCD in Q and NQ cell populations. Q and NQ cell populations were recovered from differently aged CR or non-CR yeast cultures and purified by centrifugation in Percoll density gradient. Apoptotic PCD was microscopically visualized by Annexin V staining for monitoring phosphatidylserine (PS) translocation from the inner to the outer leaflet of the plasma membrane, a characteristic marker of this PCD subroutine. Propidium iodide (PI) staining for measuring the extent of plasma membrane permeability for small molecules, a hallmark of necrotic PCD, was used to microscopically visualize programmed necrosis. Our analyses have revealed the following: 1) through most of the chronological lifespan (i.e. since day 3 of culturing), the percentage of cells undergoing an apoptotic or necrotic mode of PCD in NQ cells from CR cultures is significantly lower than that in NQ cells from age-matched non-CR cultures ( Figure  8A and 8C, respectively); and 2) late in chronological lifespan (i.e. since day 14 for apoptotic PCD or day 10 for necrotic PCD), the percentage of cells committed to any of these two PCD modes in Q cells from CR cultures is considerably decreased as compared to that in Q cells from non-CR cultures of the same chronological age ( Figures  8B and 8D, respectively).
We therefore concluded that CR decelerates the onsets of age-related apoptotic and necrotic PCD in Q and NQ cell populations. In Q cells these effects of CR take place only late in chronological lifespan, whereas NQ cells exhibit them already early in life. CR delays yeast chronological aging by causing specific changes at the G 1 checkpoint for cell cycle arrest and entry into the G 0 state, timing of G Q HD (high-density, quiescent) cells commitment to conversion into G Q LD (low-density, quiescent) cells, pace of progression of G Q LD (low-density, quiescent) cells through a differentiation program yielding G NQ LD (low-density, non-quiescent) cells, and rate of advancement of GQHD (high-density, quiescent) cells via a maintenance program leading to their conversion into GNQHD (high-density, non-quiescent) cells. Please see text for additional details.

CR decreases the susceptibilities of Q and NQ cells to the exogenously induced apoptotic and liponecrotic modes of PCD
Because CR delays the onsets of age-related the apoptotic and necrotic modes of PCD in Q and NQ populations of chronologically aging yeast, we examined how CR influences the susceptibilities of Q and NQ cells to each of these PCD subroutines that were elicited in response to certain exogenous stimuli. A brief exposure of yeast to exogenous hydrogen peroxide is known to decrease clonogenic survival of cells by causing mitochondria-controlled apoptotic PCD [173,176,181,184,185], whereas a short-term treatment of yeast with exogenous palmitoleic acid (POA) has been shown to reduce clonogenic survival of cells by eliciting a liponecrotic mode of PCD [180][181][182][183].
Our comparative analyses have revealed the following: 1) through most of the chronological lifespan (i.e. since day 3 of culturing), clonogenic survival of NQ cells from CR cultures briefly exposed to exogenous hydrogen peroxide (to elicit apoptotic PCD) or POA (to trigger liponecrotic PCD) significantly exceeds that of NQ cells from age-matched non-CR cultures ( Figure 8E and 8G, respectively); and 2) late in chronological lifespan (i.e. since day 10), clonogenic survival of Q cells from CR cultures subjected to a short-term treatment with exogenous hydrogen peroxide (to initiate apoptotic PCD) or POA (to induce liponecrotic PCD) is significantly higher than that of Q cells from non-CR cultures of the same chronological age ( Figure 8F and 8H, respectively).
In sum, these data indicate that CR causes a decline in the susceptibilities of Q and NQ cell populations to the apoptotic and liponecrotic modes of PCD induced in response to a short-term exposure to exogenous hydrogen peroxide or POA, respectively. In Q cells, the abilities of CR to cause these effects can be seen only late in chronological lifespan, whereas NQ cells display both effects of CR already early in life.

DISCUSSION
This study revealed that CR extends yeast chronological lifespan via a mechanism that links cellular aging to cell cycle regulation, maintenance of the Q state, entry into the NQ state and survival in the NQ state. Our comparative analyses of physical, morphological, reproductive, biochemical and physiological properties of Q and NQ cells from differently aged CR or non-CR yeast cultures suggest a hypothetical model for this mechanism. This model is depicted schematically in Figure 9.
The model posits that Q and NQ cell populations exist as several subpopulations related to each other in a chronological age-dependent manner. One of these subpopulations is a subpopulation of non-differentiated Q cells that represents a ″stem″ cell niche (Figure 9). We call these stem cells G Q HD because they are high-density cells that are arrested at the G 1 phase of the cell cycle and exist in a specialized non-proliferative state called G 0 . G Q HD cells constituting the stem cell niche are viable unbudded cells that 1) exhibit high reproductive (colony-forming) capacity; 2) are able to synchronously re-enter the mitotic cell cycle after cell transfer into fresh medium; 3) display high concentrations of glycogen and trehalose, the two major glucose stores; 4) possess low concentrations of TAG and high concentrations of CL; 5) have fully functional mitochondria; 6) exhibit low concentrations of ROS; 7) display low degree of oxidative damage to proteins, lipids and DNA; 8) are resistant to longterm thermal and oxidative stresses; and 9) display low susceptibilities to the mitochondria-controlled apoptotic and fatty acid-induced liponecrotic forms of PCD.
In L (for CR yeast cultures) or SP (for non-CR yeast cultures) phase, G Q HD cells become committed to entry into a program that we call the differentiation program ( Figure  9). Such commitment of G Q HD cells to differentiation is manifested in a significant decrease of their buoyant density, so that they are converted to a subpopulation of Q cells that we call G Q LD cells because of their low density. The chronological age-related progression of G Q LD cells through the differentiation program causes a gradual decline in the following features: viability, reproductive (colony-forming) capacity, the ability to synchronously re-enter the mitotic cell cycle after cell transfer into fresh medium, glycogen and trehalose concentrations, TAG concentration, mitochondrial functionality, the ability to maintain low concentration of ROS, the capacity to limit oxidative molecular damage, the tolerance to long-term thermal and oxidative stresses, and the resistance to the apoptotic and liponecrotic modes of PCD. The progression of G Q LD cells through the differentiation program ultimately leads to their conversion into a subpopulation of NQ cells that we call G NQ LD cells; these cells are committed to the programmed, age-related modes of apoptotic and/or liponecrotic cell death (Figure 9).
When a yeast culture enters ST phase, most (for CR yeast cultures) or many (for non-CR yeast cultures) of non-differentiated G Q HD cells from the stem cell niche have been committed to entry into and progression through the differentiation program and, thus, exist in the G Q LD and G NQ LD forms. However, a portion of cells within this culture still remains in the G Q HD form (Figure 9). Through the entire chronological lifespan, these non-differentiated stem cells progress through a program which we call the maintenance program. A progression of G Q LD stem cells through this maintenance program is manifested in a much slower deterioration in the same features as the ones whose relatively fast decline occurs during the differentiation program (Figure 9). The advancement of G Q HD cells through the maintenance program eventually results in their conversion into a subpopulation of NQ cells Oncotarget 69341 www.impactjournals.com/oncotarget that we call G NQ HD cells; these cells are susceptible to the chronological age-related subroutines of apoptotic and/or liponecrotic PCD (Figure 9).
Our findings suggest that CR delays yeast chronological aging by causing specific changes in a G 1 checkpoint for cell cycle arrest and entry into the G 0 state, a growth phase in which G Q HD cells undergo the commitment to become G Q LD cells, the differentiation of G Q LD cells into G NQ LD cells, and the conversion of G Q HD cells into G NQ HD cells. These changes are described below. Judging from the very small size of G Q HD stem cells seen in CR cultures ( Figure 1G), CR may arrest the cell cycle at a previously unknown checkpoint in early G 1 . This is unlike the large size of G Q HD stem cells observed in non-CR cultures ( Figure 1G), whose cell cycle is known to be arrested at the checkpoint START A in late G 1 [14,186].
Furthermore, CR is likely to accelerate the onset of commitment of G Q HD stem cells to entry into the differentiation program ( Figure 9). Indeed, G Q HD cells in CR cultures become committed to entry into this program already in L growth phase, whereas in CR cultures such commitment occurs only in SP phase ( Figure 1D).
Moreover, our data indicate that CR can also decrease both the pace of progression of G Q LD cells through the differentiation program and the rate of advancement of G Q HD cells via the maintenance program ( Figure 9). These effects of CR slow down yeast chronological aging because they delay the conversion of these two subpopulations of Q cells into G NQ LD and G NQ HD subpopulations of NQ cells (respectively), both of which are committed to the chronological age-related modes of apoptotic and liponecrotic PCD (Figure 9). Our findings suggest that the observed abilities of CR to decelerate the differentiation and maintenance programs are due to specific effects of this low-calorie diet on a distinct set of morphological, reproductive, biochemical and physiological processes in G Q LD and G Q HD cells (as outlined above, all these processes have been implicated in defining longevity of chronologically aging yeast). These specific effects of CR in G Q LD and G Q HD cells include the following: 1) CR slows a chronological age-related deterioration in the reproductive competences of these cells and in their capabilities to synchronously re-enter the mitotic cell cycle when nutrients become available; 2) CR increases glycogen and trehalose concentrations in these cells; 3) CR decreases the concentrations of TAG and raises CL concentrations in these cells; 4) CR improves mitochondrial functionality in these cells; 5) CR reduces ROS concentrations and lessens the degree of oxidative molecular damage in these cells; 6) CR increases the resistance of these cells to long-term thermal and oxidative stresses; and 7) CR decreases the susceptibilities of these cells to the apoptotic and liponecrotic forms of PCD ( Figure 9).
The challenge for the future is to test our hypothesis that CR decelerates the differentiation and maintenance programs of G Q LD and G Q HD cells (respectively) because this low-calorie diet modulates the above-mentioned set of longevity-defining processes. To address this challenge, it would be important to assess how genetic interventions impairing each of these processes may influence the chronological age-related relative abundance of G Q HD , G Q LD , G NQ HD and G NQ LD subpopulations in CR and non-CR yeast cultures. It would be also interesting to examine how these genetic interventions may affect longevity of chronologically aging yeast under CR and non-CR conditions.
Another challenge for the future is to explore how genetic and pharmacological interventions known to delay yeast chronological aging by modulating a network of certain signaling pathways and protein kinases may impinge on the described here mechanism that links cellular aging to cell cycle regulation, maintenance of the Q state, entry into the NQ state and survival in the NQ state. This network integrates the TORC1, PKA, Snf1 and Pho85 core hubs of the signaling network of quiescence, as well as their downstream effector proteins Sch9, Rim15 and others [1,12,14,21,28,144,175,[187][188][189][190][191][192].

Separation of quiescent and non-quiescent cells by centrifugation in Percoll density gradient
1 ml of 1.5 M NaCl (#S7653; Sigma) was placed into a 50-ml conical polypropylene centrifuge tube (#055398; Fisher Scientific), and 8 ml of the Percoll solution (#P1644; Sigma) was added to this tube. The NaCl and Percoll solutions were then mixed by pipetting. To form two Percoll density gradients, 4 ml of the NaCl/Percoll mixture was put into each of the two polyallomer tubes for an MLS-50 rotor for an Optima MAX ultracentrifuge (all from Beckman Coulter, Inc.). The tubes were centrifuged at 25,000 × g (16,000 rpm) for 15 min at 4 o C in an Optima Oncotarget 69342 www.impactjournals.com/oncotarget MAX ultracentrifuge. A sample of yeast cells was taken from a culture at a certain time-point. A fraction of the sample was diluted in order to determine the total number of cells per ml of culture using a hemacytometer (#0267110; Fisher Scientific). For each Percoll density gradient, 1 × 10 9 yeast cells were placed into a 15-ml conical polypropylene centrifuge tube (#0553912; Fisher Scientific) and then pelleted by centrifugation at 5,000 rpm for 7 min at room temperature in an IEC Centra CL2 clinical centrifuge (Thermo Electron Corporation). Pelleted cells were resuspended in 500 μl of 50 mM Tris/ HCl buffer (pH 7.5), overlaid onto the preformed Percoll gradient and centrifuged at 2,300 × g (5,000 rpm) for 30 min at 25 o C in an Optima MAX ultracentrifuge. The upper and lower fractions of cells were collected with a pipette, Percoll was removed by washing cells twice with 50 mM Tris/HCl buffer (pH 7.5) and cells were resuspended in 50 mM Tris/HCl buffer (pH 7.5) for subsequent assays.

Reproductive (colony forming) capability assay for quiescent and non-quiescent cells separated by centrifugation in Percoll density gradient
An aliquot of the upper or lower fraction of cells recovered from the Percoll gradient and washed twice with 50 mM Tris/HCl buffer (pH 7.5) was diluted in order to determine the total number of cells per fraction using a hemacytometer (#0267110; Fisher Scientific). Serial dilutions (1:10 2 to 1:10 5 ) of cells were also plated onto YEPD (1% yeast extract, 2% peptone, 2% glucose) plates in duplicate in order to count the number of viable cells per ml of each cell fraction. 100 ml of diluted culture was plated onto each plate. After 48-h incubation at 30 o C, the number of colonies per plate was counted. The number of colony forming units (CFU) equals to the number of reproductively capable cells in a sample. Therefore, the number of reproductively capable cells was calculated as follows: number of CFU × dilution factor × 10 = number of reproductively capable cells per ml. For each cell fraction assayed, % reproductive capability of the cells was calculated as follows: number of CFU per ml/total number of cells per ml × 100%.

Synchronous reentry into mitosis assay for quiescent and non-quiescent cells separated by centrifugation in Percoll density gradient
5 × 10 6 cells recovered in the upper or lower fraction of the Percoll gradient and washed twice with 50 mM Tris/ HCl buffer (pH 7.5) were harvested by centrifugation for 1 min at 21,000 × g at room temperature. Pelleted cells were washed twice with water and then inoculated into 50 ml of YP medium (1% yeast extract, 2% peptone; both from Fisher Scientific; #BP1422-2 and #BP1420-2, respectively) initially containing 0.2% or 2% glucose (#D16-10; Fisher Scientific) as carbon source. Cells were cultured for 4 h at 30 o C with rotational shaking at 200 rpm in Erlenmeyer flasks at a "flask volume/medium volume" ratio of 5:1. A sample of cells was taken from a culture at a certain time-point and examined microscopically for the percentage of cells with new buds. At least 500 cells were examined per time point, and the budding percentage was calculated as follows: (number of cells with new buds per ml/total number of cells per ml) × 100%.

Miscellaneous procedures
Chronological lifespan analysis [4], a microanalytic biochemical assay for measuring glucose concentration [193], cell size and number measurements [4], a Calcofluor White M2R (#L7009, component B; Molecular Probes/ Thermo Fisher Scientific) staining for microscopical visualization of bud scars [8], a measurement of the number of bud scars per cell [8], glycogen and trehalose measurements [194], a mass spectrometric quantitative assessment of the yeast lipidome [194], cellular respiration measurement [4], fluorescence microscopy for measuring ROS and ΔΨ m [4], a measurement of the frequency of spontaneous point mutations in the CAN1 gene of nDNA by monitoring the frequency of mutations that caused resistance to the antibiotic canavanine (#C1625; Sigma) [4], a measurement of the frequency of spontaneous point mutations in the rib2 and rib3 loci of mtDNA by monitoring the frequency of mtDNA mutations that caused resistance to the antibiotic erythromycin (#227330050; Acros Organics) [4], oxidatively damaged proteins and lipid measurements with a Protein Carbonyl Assay Kit assay kit (#10005020; Cayman Chemical) and a PeroXOquant Quantitative Peroxide Assay Kit assay kit (#23285; Thermo Scientific Pierce), respectively [181], a plating assay for analyzing the resistance to thermal and oxidative stresses [195], an Annexin V (#630109; Clontech Laboratories, Inc.) staining for microscopical visualization of externalized PS and a PI (#P3566; Life Technologies/ Molecular Probes) staining for microscopical visualization of the extent of plasma membrane permeability for small molecules [4], a cell viability assay for monitoring the susceptibility of yeast to an apoptotic mode of cell death induced by hydrogen peroxide [181], and a cell viability assay for monitoring the susceptibility of yeast to a liponecrotic mode of cell death induced by POA [181] were performed as previously described.

Statistical analysis
Statistical analysis was performed using Microsoft Excel's (2010) Analysis ToolPak -VBA. All data on cell survival are presented as mean ± SEM. The P values for comparing the means of two groups using an unpaired two-tailed t test were calculated with the help of the www.impactjournals.com/oncotarget of yeast Yak1 kinase by PKA and autophosphorylationdependent 14-3-3 binding.