OUP user menu

Sugar-induced apoptosis in yeast cells

David Granot, Alex Levine, Edan Dor-Hefetz
DOI: http://dx.doi.org/10.1016/S1567-1356(03)00154-5 7-13 First published online: 1 October 2003


Sugars induce death of Saccharomyces cerevisiae within a few hours in the absence of additional nutrients to support growth; by contrast, cells incubated in water or in the presence of other nutrients without sugar remain viable for weeks. Here we show that this sugar-induced cell death (SICD) is characterized by rapid production of reactive oxygen species (ROS), RNA and DNA degradation, membrane damage, nucleus fragmentation and cell shrinkage. Addition of ascorbic acid to sugar-incubated cells prevents SICD, indicating that SICD is initiated by ROS. The lack of a protection mechanism against SICD suggests that sugars use to be the limiting nutrients for yeast and are probably depleted before all other nutrients. Being the limiting nutrient, sugars became the growth-stimulating agent, signaling the presence of sufficient nutrients for growth, but in the absence of the complementing nutrients they induce apoptotic death.

  • Programmed cell death
  • Apoptosis
  • Growth stimulation
  • Starvation
  • Saccharomyces cerevisiae

1 Introduction

Yeast cells grown to the stationary phase accumulate large amounts of reserve carbohydrates such as glycogen and trehalose [1] and are able to survive for long periods under starvation conditions [2,3]. Upon transfer to growth media that contain sugar, nitrogen source and additional minor nutrients, the cells resume growth. To identify the growth-stimulating nutrient we have previously hypothesized that exposure of stationary-phase cells to such a stimulating nutrient, in the absence of additional nutrients to support growth, will force the cells out of the stationary phase and cause cell death. We have found that in the absence of additional nutrients, incubation in sugar caused loss of viability within a few hours [2,3], a phenomenon we have termed sugar-induced cell death (SICD). Incubation for more than 3 weeks either in water or in media containing all nutrients except sugar did not cause death [2]. These results identified sugar as the growth-stimulating nutrient for yeast. Indeed, short exposure (minutes) of stationary-phase yeast cells either to fermentable or to non-fermentable sugars, without additional nutrients to support growth, induced early, pre-budding growth events such as vacuole dispersal, culminating in advanced budding upon transfer to rich media [2,3].

In the present study we analyzed the process of SICD. Apoptotic characteristics have previously been demonstrated in certain yeast mutants [4,5], yeast cells expressing the pro-apoptotic mammalian Bax [6], and yeast cells exposed to pheromone [7] or acetic acid [8]. It has also been shown that reactive oxygen species (ROS) induce and regulate apoptotic death of yeast cells [9,10]. Morphologically, apoptotic characteristics in yeast resemble those observed in mammalian cells including nucleus condensation and fragmentation, DNA degradation and membrane blebbing [49]. Here we characterized the SICD process with respect to RNA and DNA degradation, cellular morphology, and ROS formation.

2 Materials and methods

2.1 Yeast and media

Yeast strain: Saccharomyces cerevisiae, Y426 MATa, ura3-52, lys2-801, ade2-101, trp1-901, leu2-98.

Rich media (YEPD): 1% Yeast Extract (Difco), 2% Bacto Peptone (Difco) and either 2% glucose or 2% acetate as carbon source.

2.2 Set-up for SICD and viability tests

Yeast cells were grown in 40 ml YEPD to a stationary phase for at least 2 days after they had reached approximately 5×108 cells ml−1. The cells were pelleted at 4000×g, washed twice with 40 ml water and resuspended in a small volume of water. The cells were transferred either to water, to 2% sorbitol as an osmotic control, or to 2% glucose, at a final concentration of 2×107 cells ml−1 in 50-ml flasks with an air:culture ratio greater than 5:1. The cultures were incubated with shaking at 37°C at which the SICD rate was accelerated as previously described [2,3]. For viability assay, aliquots were taken at the times indicated, diluted and plated on YEPD plates.

2.3 RNA and DNA extraction and gel electrophoresis

RNA, DNA and chromosomal DNA were extracted from cells immediately upon transfer into water or 2% glucose (0 h) and after various times in these media. RNA was extracted by the TRI REAGENT™ (Molecular Research Center Inc., OH, USA) according to the manufacturer's protocol.

DNA was extracted according to Ausbel et al. [11]. Chromosomal DNA was extracted as described by Gerring et al. [12]. RNA was separated on a denatured 1% agarose gel, DNA was separated on 0.8% agarose gel and chromosomal DNA was separated in 1% agarose and 0.5×TBE buffer [13], on a CHEF-DR™II apparatus (Bio-Rad) at 5–35-s pulses, 200 V, for 16 h.

2.4 Oxidative burst measurement and cell staining

Free radicals were measured with 2,7-dihydrochlorofluorescein as described previously [14]. For cell staining, 50-μl yeast samples were stained with 10 μg ml−1 propidium iodide (PI) and 2.5 μg ml−1 4,6-diamido-2-phenylindole (DAPI). The yeast cells were viewed with an Olympus IX70 fluorescent microscope and photographed with a Nikon Coolpix 950 digital camera.

2.5 Electron microscopy

For transmission electron microscopy, yeast cells were fixed with 4% glutaraldehyde, spun down, resuspended in 1% osmium-tetroxide and dehydrated with increasing concentrations of ethanol. After the 100% ethanol washes, cells were washed with 100% acetone, infiltrated with 50% acetone/50% Epon for 30 min and with 100% Epon for 20 h. Cells were then transferred to fresh 100% Epon and incubated at 56°C for 48 h before cutting and staining. For scanning electron microscopy, yeast cells were fixed with formalin/ethanol/acetic acid (4:25:1.5), dried in a critical-point drier, gold sputter-coated, and observed under a scanning electron microscope.

3 Results

3.1 SICD is accompanied by degradation of DNA and RNA

To analyze the process of SICD we followed DNA and RNA degradation of stationary-phase yeast cells incubated either in water or in glucose. Whereas DNA of water-incubated cells remained intact, the DNA of glucose-incubated cells was degraded within a few hours (Fig. 1B). Separation of chromosomes on pulsed-field gel-electrophoresis clearly indicated disappearance of the chromosomes and accumulation of about 30-kb fragments. rRNA of yeast cells incubated in glucose also disappeared within a few hours while tRNA was more stable. Most of the rRNA disappeared after 14 h incubation and complete disappearance was observed after 24 h. tRNA was still observed at 24 h and intact chromosomes still remained after 30 h.

Figure 1

SICD and DNA and RNA degradation. A: Viability of stationary-phase yeast cells after incubation in water or glucose. Cell viability as a function of time was determined using plating assays as described in Section 2. B: DNA degradation. Equal samples of cells incubated in water or glucose were taken at consecutive times for DNA extraction, and equal extraction volumes were loaded on a gel for DNA separation. C: Chromosome degradation. Equal samples of cells incubated in water or glucose were taken at consecutive times for chromosome extraction, and equal extraction volumes were loaded on a gel for chromosome separation. The arrows point to fragments of approximately 30 kb. D: RNA degradation. Equal samples of cells incubated in water or glucose were taken at consecutive times for RNA extraction, and equal extraction volumes were loaded on a denatured gel for RNA separation. W – water, G – glucose, M – DNA size marker. The numbers above the gels indicate the incubation time (h) in glucose or water.

3.2 Cells undergoing SICD shrink and lose their intracellular structure

Cell shrinkage is a common characteristic of mammalian apoptotic cell death [15,16]. We have previously used light microscopy to show that incubation in glucose causes cell shrinkage [2]. To further characterize the morphological changes during SICD we incubated stationary-phase cells for 24 h either in 2% glucose, in water or in 2% sorbitol as an osmotic control, and samples were taken for scanning and transmission electron microscopy. Scanning electron micrographs showed shrinkage in cells incubated in glucose but not in those incubated in water (Fig. 2) or in 2% sorbitol (not shown). Transmission electron micrographs of cells incubated in glucose exhibited nucleus fragmentation and vesicle formation (Fig. 3A,B), membrane blebbing (Fig. 3C), and membrane fragmentation (Fig. 3D,E). Cells incubated in water or in sorbitol appeared normal (Fig. 3F).

Figure 2

SICD involves cell shrinkage. Scanning electron micrographs of cells after 24 h incubation in water (A) or in 2% glucose (B).

Figure 3

Transmission electron micrographs of yeast cells after 24 h incubation in 2% glucose or water. A–E: Cells incubated in glucose. F: Control cell with normal phenotype. N – nucleus, FN – fragmented nucleus, V – vacuole. The black bars represent 1 μm.

3.3 SICD involves membrane damage and nuclei fragmentation

Cells incubated either in glucose or in water for 24 h were stained with PI, which stains cells with damaged membrane, or with DAPI, which stains nuclei. Most cells incubated in glucose were stained with PI while those incubated in water were mostly unstained, indicating membrane damage in glucose-incubated cells. Also, nuclei of cells incubated in glucose were less compact compared to those of cells incubated in water, and in some cells the nuclei were fragmented (Fig. 4).

Figure 4

Stained cells and nuclei of yeast undergoing SICD. Yeast cells incubated for 24 h in water or in 2% glucose were stained with PI, which stains cells with damaged membranes, and with the nuclear stain DAPI. Cells were visualized with a fluorescence microscope. The arrows point to fragmented nuclei.

3.4 SICD is initiated by ROS

ROS production during SICD was followed upon transfer of stationary-phase yeast cells either to water or glucose. Yeast cells transferred to glucose produced ROS within several minutes (Fig. 5). To test the role of the produced ROS in SICD, we checked the effect of ascorbic acid, a ROS scavenger, on cell viability in glucose. As shown in Fig. 6, ascorbic acid reduced the rate of SICD, suggesting that SICD is initiated by ROS.

Figure 5

SICD involves production of ROS. Cells were incubated either in 2% glucose or water and ROS production was followed as described in Section 2.

Figure 6

Ascorbic acid prevents SICD. Stationary-phase yeast cells were incubated either in 2% glucose or 2% glucose+10 mM ascorbic acid and their viability was followed by plating on YEPD.

3.5 SICD gives rise to petite mutants

Following incubation in glucose, cell viability was determined by plating cell samples on YEPD plates. In many of the viability experiments, in addition to the regular colonies, small colonies appeared (Fig. 7A). We suspected that those cells that were petite (ρ-cells) lost their mitochondrial respiration capability. To test whether the small colonies were petite we checked their ability to grow on acetate, a non-fermentable carbon source. Unlike the regular colonies, all the small colonies failed to grow on acetate (Fig. 7B). In addition, the small colonies were more sensitive to SICD (not shown), indicating that they were not selected by SICD but rather were newly formed mutants. We concluded that the small colonies were most likely petite mutants that were formed during the process of SICD.

Figure 7

Formation of petite colonies following incubation in glucose. A: Stationary-phase cells incubated in glucose or water for 48 h were diluted as indicated and plated on YEPD plates. B: Growth test of regular and small colonies on YEPD and on YEP media containing acetate instead of glucose as a carbon source.

4 Discussion

Death of yeast cells may be caused by different agents such as various mutations [4,5], expression of the pro-apoptotic mammalian Bax [6], acetic acid [8], pheromone arrest [7] and sugar alone [3]. Yet, cell death caused by sugar (SICD) is quite an exception because sugar, especially glucose, is the most favorable nutrient for yeast growth. However, since sugar serves as the only mitogen that stimulates growth of yeast cells [2,3], it may suggest a connection between cell cycle and apoptosis. Perhaps cells out of stationary phase and within the cell cycle are more vulnerable to stress conditions such as starvation, expression of Bax, exposure to acetic acid or pheromone arrest. A connection between cell cycle and apoptosis has been suggested in mammalian cells, and although its nature is not clear yet [1719], exit from stationary phase might be a common prerequisite for apoptosis.

SICD is not limited to S. cerevisiae but occurs also in Schizosaccharomyces pombe and Escherichia coli (unpublished results), suggesting that SICD is a common phenomenon for microorganisms that utilize sugar. It is surprising that no protection mechanism has evolved against SICD. Perhaps, in nature, sugar-utilizing microorganisms, including yeast, are not exposed to sugars in the absence of other nutrients. That suggests that sugars are the limiting nutrient in nature being depleted prior to the other nutrients. This may explain why sugars became the major growth-stimulating nutrient for yeast and perhaps for many other microorganisms as well. Sensing glucose, the limiting nutrient, may suffice to indicate the presence of the complementing nutrients required for growth. Moreover, evolutionary sensing glucose alone might be advantageous inasmuch as it may override the need to sense any other nutrient and thus provide a sufficient stimulus for growth. Indeed, none of the other nutrients stimulated growth or caused SICD-like death [2], suggesting that sugars are the only nutrients that stimulate growth.

Growth stimulation by glucose in the absence of other necessary nutrients was accompanied by rapid production of ROS (Fig. 5), while incubation in complete media (YEPD) produced no ROS (data not shown). The fact that ascorbic acid, a ROS scavenger, prevented SICD, indicates that ROS could be the immediate stimulator of SICD. Oxygen radicals are important components of metazoan and yeast apoptosis [9,20]. They induce apoptosis and are involved in cdc48- and Bax-induced apoptosis in yeast, both of which were partially suppressed by free radical spin traps [9]. Therefore, it is anticipated that the process of SICD would be similar to that shown in cdc48-, ROS- and Bax-induced death. Indeed, the hallmark of apoptosis, namely DNA degradation, nuclei fragmentation, membrane blebbing and membrane damage occur also in SICD. In addition, SICD involves cell shrinkage, a frequently described feature of mammalian apoptosis [15,16]. Since most of the classical characteristics of mammalian and yeast apoptosis were observed in SICD we conclude that sugar induces apoptotic death of yeast cells.

Classical mammalian apoptosis involves cleavage of nuclear DNA first into 30–50-kb fragments and then into oligonucleosomal fragments shown as a ladder upon gel separation [21,22]. Although DNA laddering was not observed in yeast apoptosis, accumulation of 800-kb fragments was described for the yeast S. pombe undergoing apoptosis [23]. We observed accumulation of about 30-kb DNA fragments following chromosome separation by pulsed-field gel electrophoresis, suggesting specific DNA cleavage. Specific cleavage of 28S rRNA, though not of 18S rRNA, accompanied apoptosis of various mammalian cells [24,25]. Unlike mammalian cells, SICD in yeast was accompanied by rapid degradation of all rRNAs. Yet, tRNA cleavage was slower, suggesting a regulated process of RNA degradation.

Although ROS are produced immediately upon transfer to glucose, the cells are still able to survive when transferred back to water [3]. Hence, ROS production is not sufficient for SICD commitment. However, some of the surviving cells might have acquired genetic damage. One such damage observed in the course of SICD is formation of petite colonies, probably due to loss of mitochondrial respiration (Fig. 7). SICD in the presence of glucose is entirely dependent on glucose phosphorylation [26], the initial step of glycolysis followed by the Krebs cycle in the mitochondria. We speculate that the mitochondria and the mitochondrial DNA might be more vulnerable to damages caused by radicals, which could be formed through SICD in the mitochondria. Yet, we cannot rule out the possibility that damages to other essential cellular constituents might have happened as well, and perhaps were not detected or were lethal, while the facultative nature of oxidative respiration in yeast allowed survival of the petite mutants.

The recent discovery of a yeast caspase YCA1 has confirmed the presence of programmed cell death in yeast [10] with features similar to those observed in plant and mammalian cells [15,27,28]. We have used SICD to isolate Arabidopsis genes that suppress SICD. One of the isolated genes encodes a vesicle-associated membrane protein (VAMP) that suppressed also Bax-induced apoptosis in yeast, downstream of the oxidative stress [14]. Hence, SICD may be used to uncover genes and processes involved in programmed cell death not only in yeast but in other organisms too.


This research was supported by research Grant no. IS-3226-02C from BARD, The united States-Israel Binational Agricultural Research and Development Fund, by BSF Grant 97-00250, and by The Israel Science Foundation Grant no. 582/01-2. Contribution from the Agriculture Research Organization, The Volcani Center, Bet Dagan, Israel, No. 149/2002.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
View Abstract