OUP user menu

Programmed cell death in fission yeast

Luis Rodriguez-Menocal, Gennaro D'Urso
DOI: http://dx.doi.org/10.1016/j.femsyr.2004.07.007 111-117 First published online: 1 November 2004

Abstract

Recently a metacaspase, encoded by YCA1, has been implicated in a primitive form of apoptosis or programmed cell death in yeast. Previously it had been shown that over-expression of mammalian pro-apoptotic proteins can induce cell death in yeast, but the mechanism of how cell death occurred was not clearly established. More recently, it has been shown that DNA or oxidative damage, or other cell cycle blocks, can result in cell death that mimics apoptosis in higher cells. Also, in fission yeast deletion of genes required for triacylglycerol synthesis leads to cell death and expression of apoptotic markers. A metacaspase sharing greater than 40% identity to budding yeast Yca1 has been identified in fission yeast, however, its role in programmed cell death is not yet known. Analysis of the genetic pathways that influence cell death in yeast may provide insights into the mechanisms of apoptosis in all eukaryotic organisms.

Keywords
  • Apoptosis
  • Fission yeast
  • Cell cycle

1 Introduction

Apoptosis or programmed cell death (PCD) plays a critical role in promoting homeostasis in a wide variety of organisms. The role of apoptosis is protective by eliminating cells that have either been infected by extraneous agents, or have suffered non-repairable DNA or cellular damage. In contrast, necrosis, a form of cell death that results from overwhelming cellular injury, causes cell lysis and release of cytoplasmic material that can lead to undesirable inflammatory responses. Defects in apoptosis can result in cancer, autoimmune diseases and spreading of viral infections, while neurodegenerative disorders, AIDS and ischemic diseases are caused or enhanced by apoptosis. The large number of mutations found in cancer cells that inactivate proapoptotic proteins underscores the importance of apoptosis in defending the organism from tumorgenesis [1].

In higher eukaryotic cells, apoptosis is a highly complex process that can occur through either p53-dependent or p53-independent pathways. Because most chemotherapy agents target DNA synthesis and other essential cell cycle processes that ultimately trigger apoptotic cell death, a great emphasis has been put on understanding the mechanisms of apoptosis in response to DNA damage [2,3]. Therefore, the study of apoptosis and the mechanisms that lead to cell death in response to DNA damage and other stimuli would greatly benefit from studies using genetically tractable organisms like yeast.

Several years ago it has been demonstrated that expression of pro-apoptotic proteins such as Bax or Bak in yeast could result in a terminal cell death phenotype that had many characteristics of apoptosis-mediated cell death [4,5], including formation of membrane associated vesicles (similar to membrane blebbing), nuclear condensation, and DNA fragmentation. Moreover, the human high-mobility group box-1 protein (HMGB1), a protein over-expressed in human breast carcinoma, was identified as an inhibitor of Bak-mediated cell death in yeast [6]. Despite these interesting findings, the failure to detect yeast homologs of any known mammalian pro-apoptotic or anti-apoptotic proteins led researchers to initially discard yeast as a suitable model system for metazoan apoptosis.

However, with the discovery of a yeast metacaspase [7,8] yeast has once again emerged as a potentially powerful genetic tool to explore the basic mechanisms of eukaryotic PCD.

2 Defects in initiation of DNA replication result in the production of ROS

One of the hallmark features of cells undergoing apoptosis is the production of reactive-oxygen species or ROS [9]. ROS are produced as a by-product of mitochondrial activity that typically occurs during both the early and late stages of apoptosis. ROS can also be produced where electron transport occurs, including the endoplasmic reticulum and nuclear membrane. ROS can precede other events associated with apoptotic cell death, including mitochondrial membrane depolarization, cytochrome c release, activation of caspases, and nuclear fragmentation. This suggests that ROS might have a regulatory role in the apoptotic pathway.

Similar to mammalian cells, ROS is also produced in yeast cells undergoing cell death in response to a number of external or internal stimuli. This occurs in budding yeast cdc48 mutants [10], following treatment with low concentrations of H2O2[11], acetic acid [12], or high concentrations of mating pheromone [13]. It is also observed following over-expression of mammalian pro-apoptotic proteins, suggesting that ROS is an important regulator of PCD in yeast [8].

Previous experiments in budding yeast have shown that orc2-1 mutants, when shifted to the restrictive temperature, undergo cell death and release high levels of ROS [14]. Interestingly, deletion of the gene encoding the checkpoint kinase Mec1 suppresses the lethality of the strain, suggesting that cell death is at least partially dependent on checkpoint pathways. Because Orc2 is a critical component of the DNA replication initiation complex it is likely that inactivation of Orc2 results in the accumulation of DNA damage that triggers the cell death response. Although orc2-1 cells show the highest sensitivity to DNA damaging agents or high temperatures in the G1 phase of the cell cycle, subsequent entry into S phase is required to observe the highest levels of ROS [14]. This suggests that a decrease in the number of active replication complexes (that are assembled during G1) might result in DNA damage-mediated cell death as cells enter S phase.

Temperature-sensitive fission yeast mutants defective in DNA replication initiation proteins also produce high levels of ROS upon shift to the restrictive temperature [14]. Fig. 1 shows ROS production in two mutants, orc2-186 and orc5-H19, following shift to the restrictive temperature of 36 °C [14]. Similar to orc2-1 in Saccharomyces cerevisiae, cells incubated at the high temperature produce high levels of ROS. ROS production was measured by staining cells with the ROS-specific stain, 2′,7′-dichlorodihydrofluorescein diacetate. This reagent can permeate living yeast cells, but does not fluoresce unless oxidized by ROS. In addition to Orc2 and Orc5, mutations in Dfp1 and Cdc18, two additional initiation proteins, led to high production of ROS [14], consistent with the hypothesis that reduced replication origin activity may result in DNA damage that leads to cell death.

Figure 1

ROS staining in DNA replication initiation mutants of fission yeast. Living yeast cells in rich media were incubated for 80 min with the ROS-specific stain, 2′, 7′-dichlorodihydrofluorescein diacetate. The cells were then harvested and washed with citrate buffer (50 mM citrate buffer, pH 7.0). The cell pellets were resuspended in the same buffer and photographed with an epifluorescence microscope equipped with a filter set capable of detecting green (ROS).

As was observed in S. cerevisiae, it is not yet clear whether checkpoint pathways are required for the cell death phenotype observed in the DNA replication initiation mutants. In the case of orc5-H19, no effect of deleting rad3 (homolog of ATM/MEC1) or cds1 was observed on the production of ROS at the restrictive temperature. In contrast, ROS production in dfp1-564 was suppressed by deleting either rad3+ or cds1+[14]. Therefore, at least in the case of dfp1 mutants, the checkpoint might have a role in regulating cell death.

3 Inhibition of triacylglycerol synthesis induces cell death in fission yeast

Triacylglycerols (TAGs) are important energy storage molecules found in virtually all eukaryotic cells. In fission yeast, two enzymes encoded by the plh1 and dga1 genes synthesize the bulk of TAG [15]. Despite having non-detectable levels of TAGs, cells deleted for both of these genes (DKO strain) grow normally. However, when cells are grown to stationary phase and subsequently used to inoculate a fresh culture, a significant lag phase is observed prior to the onset of exponential growth. No such lag phase is observed when cultures are started using stationary wild-type cells. Colony forming assays confirmed that the lag phase stemmed from a rapid loss of cell viability once the DKO cells entered stationary phase. Microscopic examination revealed that many of these cells undergo nuclear DNA fragmentation upon nutrient starvation. In addition to nuclear fragmentation, these cells also display additional apopotic-like markers, including expression of phosphatidylserine on the surface of the mitochondrial membrane (assayed by annexin-V staining) and increased ROS production.

Diacylglycerol and long-chain fatty acids, the two major substrates of TAG synthesis, accumulate in DKO cells following nutrient starvation, suggesting that these molecules may be involved in promoting cell death [15]. This was tested by treating cells with a membrane-permeable DAG analog and assaying for both DNA fragmentation and cell death. An increase in cell death was observed in the DKO strain following addition of DAG that was not observed in wild-type cells. In addition, adding fatty acids directly to fission yeast cells also induced cell death that was partially reversed by expression of DAG kinase (that converts DAG to phosphatidate), suggesting that the fatty-acid-induced cell death is mediated through DAG accumulation.

Interestingly, although ROS production appears to be required for nuclear fragmentation and cell death in DKO cells treated with fatty acids, cell death still occurs in the absence of the fission yeast metacaspase gene, pca1+ (see below). Thus, although ROS might have an important regulatory role in promoting cell death upon accumulation of DAG, there is no clear evidence that the metacaspase Pca1 is involved in this form of apoptosis. This raises the important question of whether alternative caspases or other factors are involved in promoting cell death, such as apoptosis-inducing factor or AIF. There are a number of potential homologs of AIF in the Sshizosaccharomyces pombe genome. Future studies will be needed to address whether any of these play a significant role in lipo-mediated apoptosis.

4 RNA splicing and programmed cell death

Recently a small novel Ring-finger protein was identified in fission yeast that is required for efficient pre-mRNA splicing [16]. Ini1 is an essential protein required for normal cell cycle progression. Because the arrest phenotype of many cell cycle mutants is reversible, the lethality associated with the loss of Ini1 suggested that an active cell death pathway might be involved. ROS production is observed in cells depleted for Ini1, providing some evidence that cells might be undergoing programmed cell death (Rodriguez-Menocal and D'Urso, unpublished data). Interestingly, the loss of cell viability in the ini1 deletion strain was partially suppressed by deletion of the wee1+ gene, suggesting that the Wee1 kinase (a negative regulator of mitosis) might have a role in promoting cell death in the absence of Ini1 [16]. However, it is also possible that wee1 cells, being considerably smaller than wild-type cells, might be less sensitive to decreasing levels of Ini1. Nevertheless, it is interesting that other mutants, defective in mRNA splicing, can trigger cell death in yeast [17].

5 Cloning of Sch. pombe pca1+

We have recently cloned a gene from the Sch. pombe genome that encodes a protein that shares 34% identity with budding yeast YCA1 (Fig. 2, see also [7]). Fission yeast Pca1 contains many of the features found in metazoan caspases, including highly conserved catalytic cysteine 297. We have shown that pca1+ is not an essential gene similar to what has been reported for budding yeast YCA1. Although it is still debated why yeast, a unicellular organism, would benefit from having an apoptotic-like cell death pathway, recent experiments have suggested at least one explanation. When budding yeast cells are grown to stationary phase and maintained in a post-mitotic state, many of the cells die and display markers of apoptotic cell death [18]. Cell death occurs within the first few days of reaching saturation and in some cases (depending on the genetic background) no cells remain viable after as little as 30 days. Interestingly, overexpression of YAP1, a transcriptional regulator of the oxygen stress response, can suppress the lethality observed in chronologically aged cultures, suggesting that accumulation of ROS might be responsible for the loss of viability [18]. Moreover, deletion of YCA1 can also decrease cell death at least initially, however, after extended incubation (>30days) there is a significant decrease in cell viability in the absence of Yca1 that is not observed in wild-type cells. Therefore, although deletion of YCA1 offers some protection to the initial events that trigger cell death in stationary cultures, over the long-term Yca1 provides an advantage to the remaining viable cells. It was shown that cells undergoing programmed cell death in these aged cultures release certain positive factors that can stimulate growth of pre-aged cultures while having no effect on exponentially growing cells. Therefore, programmed cell death may ultimately provide a selective advantage to the colony, and ensure the long-term survivability of the population.

Figure 2

Protein sequence alignment of Sch. pombe (fission yeast), S. cerevisiae (budding yeast), Aspergillus nidulans, and Neurospora crassa Yca1. Fission yeast Pca1 shares 40% identity to budding yeast Yca1. Identical residues are indicated in black, whereas homologous residues are indicated in gray. The conserved catalytic cysteine residue at position 297 is indicated. Sequence alignment was performed using ClustalW 1.8.

Although fission yeast cells also show a significant decrease in viability when grown to and maintained at stationary phase (L. Menocal and G. D'Urso, unpublished data), it is not yet clear whether cell death is an active rather than a passive process. We are currently testing whether deletion of fission yeast pca1+ has any effect on either the short-term or long-term viability of growth-arrested cultures.

6 Checkpoints and programmed cell death

Checkpoints are responsible for delaying cell cycle progression in response to DNA damage or stalled replication forks [19,20]. Over the years, many proteins have been identified that either function as sensors or direct inhibitors of the regulatory proteins that drive the cell cycle. In fission yeast, the so-called 9-1-1 complex that is comprised of Rad1, Hus1 and Rad9 is thought to associate with chromatin and to be involved in sensing DNA damage/or stalled replication forks [21]. These sensor proteins are then required to transduce signals that ultimately lead to activation of Rad3, and either Chk1 or Cds1 protein kinases [22]. Interestingly, all of these proteins are conserved in mammalian cells where they play a role in both p53-dependent and p53-independent checkpoint pathways [23]. They have also been implicated in the regulation of apoptosis in response to irrepairable DNA damage. For example a portion of the human Rad9 protein is homologous to the Bcl-2 homology 3 domain found in Bax and other pro-apoptotic proteins. HRad9 can promote apoptosis in the absence of p53 by interfering with the anti-apoptotic proteins Bcl-2 and Bcl-xL[24]. Also, down-regulation of hRad9 by RNA interference experiments can suppress apoptosis in response to DNA-damaging agents such as methylmethane sulfonate (MMS), suggesting that Rad9 can be activated in response to DNA damage [24]. Furthermore, hRad9 has been shown to be a substrate of c-abl kinase and phosphorylation of hRad9 can promote binding to Bcl-xL[25].

Interestingly, the BH3 domains in hRad9 are conserved in fission yeast Rad9, suggesting these domains might provide a similar function in yeast. Over-expression of fission yeast Rad9 in human cells can induce apoptosis, while overexpression of human anti-apoptotic Bcl-2 family members in fission yeast can suppress the lethality of DNA-damaging agents [26]. These observations suggest that at least some of the features of metazoan apoptosis may be conserved in fission yeast. However, further experiments are needed to rigorously test this hypothesis.

7 Asymmetric cell division, aging, and cell death

Budding yeast has been used extensively as a model for cell aging [27,28]. Each cell division occurs asymmetrically, producing a daughter cell that is significantly smaller than the mother cell. Mother cells typically undergo 20–25 cell divisions before undergoing replicative senescence. As they age, they become larger [29], appear wrinkled and misshapen [30], and the rates of overall protein synthesis decrease [31]. Eventually these cells fail to divide. These observations have given rise to the hypothesis that some senescence factor or factors are asymmetrically distributed to mother cells during each cell division [32]. One potential factor might be the accumulation of ribosomal circular DNA (rDNA minicircles) that could lead to the titration of key replicative enzymes or other DNA binding factors required for sustained growth and division [33]. It has also been suggested that Sir2-dependent accumulation of oxidized proteins in the mother cell greatly contributes to cellular aging [34]. Consistent with this notion, environmental or genetic manipulations that increase reactive-oxygen species (ROS) can cause premature aging in yeast [35]. Thus, how cells regulate and control the accumulation of ROS is likely to be an important determinant of lifespan. This is consistent with experiments in Caenorhabditis elegans, where mutations that increase reactive oxygen can decrease lifespan [36,37]. Interestingly, as yeast cells age they display many markers of apoptotic cell death including increased ROS production, DNA fragmentation, and exposure of phosphatidylserine on the mitochondrial cytoplasmic membrane [38]. This suggests that, at least in yeast, aging may be coupled to PCD.

In contrast to budding yeast, fission yeast divides by medial fission producing two morphologically equivalent daughter cells. The apparent absence of asymmetric cell division suggested that fission yeast might not undergo cell aging as was observed for budding yeast. However, appearances can be deceiving. Recently microscopic analysis of individual dividing fission yeast cells showed that one of the daughter cells undergoes cell senescence after an average of nine cell generations [39]. This implies that dividing fission yeast cells do divide asymmetrically and partition cellular components differentially to the two daughter cells. For example, transport of organelles such as mitochondria occurs along microtubules in fission yeast, similar to that in mammalian cells [40]. This differs from budding yeast where the mitochondria are associated with the actin cytoskeleton [41]. Perhaps specific mechanisms are in place to ensure that the older cells receive damaged or inferior mitochondria what ultimately results in the accumulation of ROS and cell senescence. Interestingly, in budding yeast changes in actin dynamics have a dramatic effect on both lifespan and chronological aging. Increased actin dynamics increase lifespan, whilst decreasing actin dynamics reduce lifespan [42]. In the latter experiment cells that prematurely age in response to decreasing actin dynamics exhibit many markers of PCD. It will be interesting to determine if defects in either microtubule or actin dynamics have any effect on cell aging or apoptosis in fission yeast.

8 Conclusion

The cloning and characterization of a metacaspase in both S. cerevisiae and Sch. pombe suggests that yeasts have retained primitive apoptotic pathways that may show some similarity to those of higher organisms. Moreover, various mutants including those defective in DNA replication initiation, triacylglycerol synthesis, or RNA splicing can lead to a cell death phenotype that is reminiscent of metazoan apoptosis. Although it may seem peculiar that unicellular organisms would display such altruistic behavior, it is becoming clear that they do harbor mechanisms to promote cell death under certain adverse conditions. An interesting set of experiments has recently shown that deletion of the YCA1 gene from S. cerevisiae results in a selective disadvantage to the cell when grown in competition with wild-type cells [18]. This suggests that programmed cell death plays an important role in maintaining the viability of a yeast colony, providing an explanation for why programmed cell death pathways have been conserved from yeast to man.

Acknowledgements

We would like to thank Bill Burhans and Frank Madeo for valuable discussions. G.D. is supported by NIH1R01CA099034-01. L.R.M is supported by a post-doctoral fellowship from the American Heart Association.

References

  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].
  29. [29].
  30. [30].
  31. [31].
  32. [32].
  33. [33].
  34. [34].
  35. [35].
  36. [36].
  37. [37].
  38. [38].
  39. [39].
  40. [40].
  41. [41].
  42. [42].
View Abstract