Types of cell death and methods of their detection in yeast Saccharomyces cerevisiae



Dominika M. Wloch-Salamon, Institute of Environmental Sciences, Jagiellonian University, Gronostajowa 7, 30-387 (room 2.2.1) Cracow, Poland. E-mail: dominika.wloch-salamon@uj.edu.pl


The occurrence of programmed cell death in unicellular organisms is a subject that arouses great interest of theoreticians and experimental scientists. Already found evolutionarily conserved genes and metabolic pathways confirmed its existence in yeast, protozoa and even bacteria. In the yeast Saccharomyces cerevisiae, at least three main types of death are distinguished: apoptosis, necrosis and autophagy. Their classification suggested by the Nomenclature Committee on Cell Death initially based on the morphological characteristics has now been extended to include the measurable biochemical characteristics. Several laboratory methods previously used to detect the types of cell death of higher eucaryotes and later developed and successfully used for the analysis of yeast cells are here critically reviewed. Their advantages and limitations are described.

Cell Death

The classic definition of death defines it as a state characterized by the cessation of signs of life. It is caused by irreversible functional imbalance and collapse of the internal organization of a system. To call a cell ‘dead’, the changes that occur in it must exceed a certain point called ‘the point of no return’ (Kroemer et al. 2009). It is proposed that a cell should be regarded as ‘dead’ when it has lost the integrity of its plasma membrane and/or it has undergone complete disintegration, including its nucleus, and/or its corpse (or its fragments) has been engulfed by a neighbouring cell in vivo (Galluzzi et al. 2009). Cell death, due to the presence of its genetic regulation, can be divided into programmed (or active) ones, such as apoptosis and autophagy, and not programmed (or nonactive), such as necrosis (Leist and Jaattela 2001; Nedelcu et al. 2011). There are still attempts to work out a precise definition of each type of cell death, together with unique protocols for their detection. Division of cell death proposed by the Nomenclature Committee on Cell Death (NCCD), initially based primarily on morphological characteristics (Kroemer et al. 2008) was recently extended to include measurable biochemical and molecular characteristics (Galluzzi et al. 2012) (Table 1). According to NCCD, it is necessary to use at least two detection methods to precisely define an observed cell death type. On the basis of previously described types of cell death, such as apoptosis, autophagy and necrosis, the types of so-called cell death subroutines have been specified, that is, caspase-dependent and caspase-independent intrinsic apoptosis, extrinsic apoptosis by death receptors, and by dependence receptors, necroptosis, autophagic cell death and mitotic catastrophe (Galluzzi et al. 2012).

Table 1. The main types of cell death, morphological and molecular characteristics of cells and the methods successfully used for its detection in the Saccharomyces cerevisiae cells. Underlined references describe methods
The main types of deathFeatures of cellsMethods for detection in yeast cells successfully used in cited research
Apoptosis/morphological features

1. Loss of cell membrane integrity

2. Loss of survival

1. Microscopy/FACS – use of appropriate dyes: DHE, rhodamine 123 (Izquierdo et al. 2011; Madeo et al. 1999)

2. (a) CFU (Gao et al. 2011; Granot et al. 2003), (b) metabolic labeling f.e: FUN_1 (Teng and Hardwick 2009)

Apoptosis/biochemical features

1. Phosphatidylserine (PS) exposure

2. Nuclear fragmentation

3. Chromatin condensation

4. Mitochondrial membrane permeabilization ΔΨm (Cytochrome c release)

5. Activation of proapoptotic yeast homologues of Bcl-2 family proteins (e.g. Ybh3p)

6. Activation of caspases

1. Fluorescence microscopy/FACS – quantification of FITC-conjugated Annexin V binding (Madeo et al. 1997; Allen et al. 2006; Buttner et al. 2007; Du et al. 2008; Gao et al. 2011)

2. a) Microscopy/FACS – TUNEL assay (Madeo et al. 1997,1999, 2002b; Granot et al. 2003; Bussche and Soares 2011; Ferreira et al. 2011; Gao et al. 2011; Guaragnella et al. 2011; Izquierdo et al. 2011; Kumar et al. 2011; Liu et al. 2011)

b) DNA ladder – Electrophoresis (not used nowadays)

c) Electron microscopy

3. a) Microscopy/FACS – DAPI staining; (Madeo et al. 2002b; Buttner et al. 2007; Du et al. 2008; Aerts et al. 2008; Bussche and Soares 2011; Izquierdo et al. 2011; Ferreira et al. 2011; Gao et al. 2011)

b) Electron microscopy (Madeo et al. 1997; Gao et al. 2011)

4. Fluorescence microscopy/FACS (Ludovico et al. 2002)

5. a) IF microscopy localization studies/

b) Immunoblotting with conformation-specific antibodies (Madeo et al. 2002b; Buttner et al. 2011)

6. FACS/IF microscopy quantification with antibodies specifically recognizing the active form caspases/cleaved caspase substrates (Madeo et al. 2002b)

Autophagy/morphological features1. Accumulation of double-membraned, autophagic vacuoles with accumulated undigested autophagic bodies

1. a) Light microscopy (Takeshige et al. 1992; Tsukada and Ohsumi 1993; Baba et al. 1994)

b) Electron microscopy (Takeshige et al. 1992)

c) Fluorescent microscopy using monodansylcadaverine (MDC) as a dye (Biederbick et al. 1995)

Autophagy/biochemical features

1. Bulk degradation of long-lived proteins

2. Delivery of the cytoplasmic components to the lysosome

3. Atg8 protein tracking

4. Nucleus accumulation in the vacuole, monitor membrane continuity or integrity (pH changes)

1. a) measurement of accumulated radioactivity after cells incubation with [14C] or [3H]-valine or leucine and washing with the trichloroacetic acid (TCA) (Mizushima 2004)

b) difference between the [14C]-valine released from cells treated with and without lysosomotropic reagents (chloroquine, ammonium chloride, bafilomycin A1) (Mizushima 2004)

2. a) measurement of vacuolar alkaline phosphatase (ALP) activity (Noda et al. 1995)

b) SDS-PAGE to separate the premature and mature form of aminopeptidase I (API) (Klionsky and Ohsumi 1999)

3. Fluorescent microscopy/FACS for LC3 (Atg8 in yeast) N-terminus protein fusion with fluorescent protein (e.g. GFP, CFP or mCherry) (Ravikumar et al. 2002; Axe et al. 2008)

4. Fluorescent microscopy using n-Rosella (genetically encoded dual colour-emission biosensor, fusion protein of DsRed.T3 and fluorine) (Nowikovsky et al. 2007; Rosado et al. 2008; Mijaljica et al. 2011)

Necrosis/morphological features

1. Rupture of plasma membrane

2. Disintegration of subcellular structures

3. Loss of survival

1. a) Fluorescence microscopy/FACS with the use of vital dyes: PI (Granot et al. 2003; Buttner et al. 2007; Bussche and Soares 2011; Gao et al. 2011; Liu et al. 2011)

b) Electron microscopy (Eisenberg et al. 2009)

2. Electron microscopy

3. a) CFU (Gao et al. 2011)

b) metabolic labelling f.e: FUN_1 (Teng and Hardwick 2009)

Necrosis/biochemical features

1. HMGB-1/Nhp6Ap yeast homologue release

2. Activation of cathepsins (CatD yeast Pep4p)

3. ROS overgeneration

1. Immunoblotting of culture medium with Nhp6Ap-specific antibodies (Eisenberg et al. 2009; Carmona-Gutierrez et al. 2011a)

2. a)Colorimetric/fluorogenic substrate-based assays in cells extract

b) immunoblotting (Carmona-Gutierrez et al. 2011a)

3. Fluorescence microscopy/FACS – quantification with ROS-sensitive probes (Madeo et al. 1999; Granot et al. 2003; Vachova and Palkova 2005)

For a long time, the occurrence of apoptosis and even the possibility of its occurrence in unicellular organisms were thought to be theoretically unfounded (Sharon et al. 2009). This kind of death has been attributed only to complex organisms, in which the controlled processes of a single cell death have an impact on the proper and efficient functioning of the whole organism. Currently, there is growing evidence that programmed cell death (PCD) is present in many unicellular organisms, such as protozoa, bacteria, slime moulds and yeasts. The presence of apoptotic phenotypes in Protista seems to confirm the hypothesis that apoptosis or some kind of it could have been developed in unicellular organisms long before the evolutionary separation between fungi, plants and animals (Shemarova 2010; Semighini et al. 2011). The most frequently reported types of death in unicellular organisms are these best known and characterized apoptosis, necrosis and autophagy, so in this article, we will focus mainly on them.

Saccharomyces cerevisiae as Versatile Model Organism

Using model organisms in research provides an opportunity to understand the universal mechanisms, and their results may serve as the reference for other organisms (including humans), on which research is too complicated or unethical. This assumption is validated by the common origin of all organisms, the conservatism of metabolic processes, pathways as well as the similarities in the genetic material. Yeast Saccharomyces cerevisiae, an eucaryotic unicellular fungus is one of the most important, extensively studied model organisms with a long distinguished experimental history (Forsburg 1999). Because of the ease of its genetic manipulations, laboratory handling and storing, it found wide applications in research. The cell biological issues explored in yeast range from signal transduction to cell-cycle control, chromosome structure to secretion (Forsburg 2001). The identified genes and proteins have been used as probes to uncover further pathways, to identify metazoan homologues and the functions that they provide to cells. Its genome sequencing in 1996, construction of mutant deletion (Winzeler et al. 1999; Giaever et al. 2002) and green fluorescent protein (GFP) fusion collections (Ghaemmaghami et al. 2003; Huh et al. 2003) started new fields of research called ‘functional genomics’ and ‘systems biology’ (Goffeau et al. 1996; Botstein and Fink 2011). Out of the c. 5800 protein coding Saccharomyces genes, 85% are of known biological role. Nearly, c. 17% of yeast genes are members of orthologous gene families associated with human disease (Heinicke et al. 2007). For the majority of these genes, their mammalian homologue is functional in yeast and complements the yeast deletion mutant. As stated in the recent comprehensive review by David Botstein and Gerald Fink, virtually all cellular level technologies for assessing protein function have remained easier with yeast than with most other organisms.

The Occurrence of Apoptosis in Saccharomyces cerevisiae

Many apoptotic features had been described much earlier than 1972 when the term ‘apoptosis’ was introduced in the morphological description of a death form (Kerr et al. 1972). Since then there have been many accurate and complete studies on different aspects of apoptosis (e.g. Elmore 2007). Briefly, apoptosis is a conservative, natural process, which is very similar in many organisms and requires energy input. This form of PCD does not lead to the disturbances of homoeostasis because it operates very quickly and efficiently, without inducing any immune response (Kroemer et al. 2009). In case of multicellular organisms, apoptosis is involved in embryonic development and plays a crucial role in a smooth operation of the whole organism. Irregularities in its functioning have serious consequences that may manifest themselves, among others, as autoimmune diseases, including cancer, neurodegenerative diseases and ischaemia damages. There are two major interlinked and mutually interacting biochemical pathways of apoptosis: external (by membrane receptors) and internal (by mitochondria) (Igney and Krammer 2002). Commissioning pathways of apoptosis causes cell membrane depolarization and release of cytochrome c molecules from mitochondria, and activation of caspases, the enzymes from the group of proteases which control apoptosis. They are expressed in most of the cells and are present as/in a form of inactive enzymes. Activation of one procaspase leads to activation of another one, which in turn causes the initiation of a protease cascade. Such activation may lead to immediate cell death by enhancing apoptotic signal. Regardless of the path causing apoptosis, the executive pathway is always the same. Active endoplasmic nucleases degrade nuclear material, and proteases degrade cytoskeleton proteins throughout a cell. It is followed by protrusion of cytoplasm to the outside (‘blebs’ forming) and emergence of so-called cells. Cell shrinkage causes formation of the apoptotic bodies. They consist of cytoplasm and tightly packed cell organelles and sometimes include parts of the nuclei (Elmore 2007). Cellular organelles in apoptotic bodies retain their integrity. Apoptotic bodies are then phagocytosed and degraded in phagolysosomes (Hacker 2000). Massive activation of caspases, loss of mitochondrial potential (ΔΨm) and the exposure of phosphatidylserine residues to the outer membrane, DNA fragmentation, cytoskeleton disintegration and cell shrinkage are considered to be the typical molecular and biochemical markers of apoptosis. It should be noted that the reaction does not induce inflammation, because apoptotic cells do not secrete their contents to the outside, and are rapidly phagocytosed (Kroemer et al. 2009).

Research on yeast apoptosis began with the microscopic observations of yeast cell division cycle mutant, cdc48 (Madeo et al. 1997). Its culture in the stationary phase showed abnormally elongated and distorted cells. Furthermore, cutting pattern of nuclear chromatin resembled the image of classical apoptotic eucaryotic cells. To confirm the occurrence of apoptosis, the methods used for detection of apoptosis in higher eucaryotes were used: TUNEL (terminal deoxynucleotidyl transferase dUTP nick end labelling) and annexin V staining, followed by electron microscopy observations (Madeo et al. 1997). As described earlier, caspases are the main factors of path-dependent apoptotic receptors in Metazoa. Detecting their presence in yeast cells was crucial for confirming the existence of apoptotic signalling pathway, which is analogous to that of higher eucaryotes (Vachova and Palkova 2007; Carmona-Gutierrez et al. 2010; Shemarova 2010). However, ‘classical’ caspases have not been detected in yeast but their sequence and functional homologue metacaspase YCA1/Yor197 (yeast caspase-1) were identified instead (Madeo et al. 2002a). Knock-outs of gene encoding this metacaspase result in inhibition of cell death process. Yca1p production activated in both oxidative and physiological stresses affects yeast cell death, but the detailed mechanism of this activation is not known yet (Oettinghaus et al. 2011). One of the first known factors inducing apoptosis in yeast was oxidative stress. The yeast that had been subjected to even low concentrations of hydrogen peroxide showed apoptotic phenotype. Additionally, increased resistance to hydrogen peroxide was observed in yeast with a deletion of yeast caspase YCA1 (Madeo et al. 1997). It was also noticed that during apoptosis in yeast, even in the absence of external oxidative stress, the level of intracellular formation of oxygen free radicals (reactive oxygen species, ROS) was higher (Madeo et al. 2002a). The release of cytochrome c from the mitochondrial membrane into the cytoplasm, triggered in yeast cells by acetic acid, was another important discovery (Ludovico et al. 2002, 2005). Currently, there is a long list of already known inducers of apoptosis in Saccharomyces cerevisiae, which includes human p53 protein, human death receptor 4, human caspases CASP10 and CASP8 (Sharon et al. 2009; Shemarova 2010). Although yeast caspase-independent pathways also involve lethal factors, homologues of higher Eukaryota such as apoptosis-inducing factor (AIF1), endonuclease G (NUC1), cyclophilin D and AMID (AIF-homologous mitochondrion-associated inducer of death) were also found (Liang et al. 2008) (Table 2). To date, roughly 40% of all death conditions tested in yeast have been executed, at least partly, via yeast metacaspase YCA1 (Carmona-Gutierrez et al. 2011b). The key regulators of apoptosis in the inner paths of higher eucaryote proteins are proteins from the Bcl-2 (B-cell lymphoma 2) family. One of them, the BH3, the equivalent of Ybh3, has been found in yeast recently (Buttner et al. 2011). So far, there are still some gaps in the full description of yeast apoptosis. However, because the process is very conservative, we can expect to find more analogues of the mammalian proteins (or slightly simplified compounds) and further proofs that these mechanisms do overlap. There were several attempts to explain a possible role of apoptosis in yeast (Buttner et al. 2006). Its role in population adaptation to various stresses has not been well proven yet. This leaves a space for the hypothesis assuming that it is simply a by-product of mature cell metabolism (Sharon et al. 2009; Nedelcu et al. 2011).

Table 2. Advantages and limitations of the popular methods used for detection of the types of cell death, in the Saccharomyces cerevisiae. Adapted after Galluzzi et al. (2009)
  1. TUNEL, terminal deoxynucleotidyl transferase dUTP nick end labelling; IF, Immunofluorescence microscopy; DAPI or 4′,6-diamidino-2-phenylindole; Δψm, mitochondrial transmembrane potential; IMS, mitochondrial intermembrane space proteins; PS, phosphatidylserine.

Survival test (Colony Forming Unit; CFU)SimpleNonspecific, not accurate for instance due to the presence of the aggregates in the population.The most basic method for accessing survival (dead/live)

Light microscopy:

a) visual inspection;


a) Rapid and inexpensive

b) High sensitivity

a) Time-consuming if large-scale quantitative applications are planned, live samples; lacks specificity

b) Prone to false-positive results, for instance due to sample processing

a) Monitoring of the abnormal cell phenotype

b) Detection of free 3′-hydroxyl ends in DNA; if performed correctly should only identify cells in the last phase of apoptosis

Electron microscopyDetection of subtle changes in organelle ultrastructure that occur early in the cascade of events leading to cell death

Inappropriate for large-scale

quantitative applications

May be poorly representative of the general sample conditions; Laborious, time-consuming, requires trained personnel; expensive equipment

Analyses morphological hallmarks of apoptosis at an ultrastructural level

IF microscopy:

a) Caspase activation

b) DAPI staining

c) Δψm-sensitive fluorochromes

d) Relocalization of proapoptotic yeast homologues of Bcl-2 family proteins


a) Allows for the highly specific identification of apoptotic cells

b) Useful to clearly identify nuclei

c) Allow for the visualization of energized mitochondria;

d) Fusion proteins allow for real-time (video or time-lapse microscopy-based) studies

e) Useful in co-staining protocols, to confirm DNA fragmentation

a) Operator dependent

b) Photobleaching

c) Δψm can be partially reduced in cell death-unrelated events

d) At least two IMS proteins

should be evaluated, to exclude artefacts

e) Prone to false-positive results, for instance due to sample processing

a) Based on antibodies that recognize active caspases or cleaved substrates

b) Fragmentation of nuclei is a classical hallmark of apoptotic cells

c) Cationic Lipophilic probes accumulate in mitochondria driven by the Δψm

e) Detection of free 3′-hydroxyl ends in DNA


a) Caspase activation assays

b) Release of mitochondrial proteins (IMS) (e.g., AIF, Cyt c)

a) based on standard laboratory equipment

b) Allows for the study of subcellular fractions and purified mitochondria (as opposed to IF)

a) Semiquantitative (the analysis involves entire cell

populations); small protein fragments (such as degradation products) may be difficult to detect

b) Time-consuming; not suitable for large-scale or high-throughput applications;

a) Based on antibodies that recognize

active caspases, their cleaved substrates or

both the inactive and active forms of caspases

b) mitochondrial outer membrane permeabilization is monitored by assessing the presence of IMS proteins in nonmitochondrial subcellular fractions

Flow cytometry:

a) Annexin V assay

b) Caspase activation assays

c) DAPI d) PI

a) Specific for an early event in the executioner phase of apoptosis; Annexin V exists conjugated with different fluorescent and nonfluorescent labels

b) Quantitative, allow for the analysis of large cell populations (as opposed to IF), on a per-cell basis.

d) e) Routinely employed in several co-staining protocols to detect necrotic cells

a) Need of spheroplasts; when plasma membranes are ruptured PS exposure can take place independently from apoptosis; PS exposure may be impaired in autophagy-deficient cells

b) Caspase activation may occur in cell death-unrelated settings; Fluorogenic substrates are prone to unspecific degradation

c, d) A high number of events is required for significance

a) Annexin V binds to PS, which in apoptotic

cells is exposed to the outer leaflet of the plasma membrane before DNA fragmentation and nuclear breakdown

b) Based on antibodies that recognize active caspases or cleaved substrates, based on cell-permeant fluorogenic substrates

c) d) DNA content analysis; Cell death is monitored by the quantification of events with a sub-G1 DNA content

Necrosis in Yeast

Necrosis was firstly described by the lack of morphological features characteristic of apoptosis and autophagy (Walker et al. 1988). Nowadays necrotic cells are morphologically characterized by oncosis (organelles and cell swelling) and breaking the continuity of the cell membrane that results in outflow of the cell contents to the outside. In addition, necrotic cells are characterized by rapid decrease in energy level, random degradation of DNA and causing inflammation in the surrounding tissue by inducing the infiltration of macrophages, neutrophils and dendritic cells and postinflammatory cytokine secretion (Kroemer et al. 2009). That is why, until recently, necrosis was seen as a random and spontaneous process, which does not require energy. Today, it is considered that necrosis may also occur in the genetically controlled form called necroptosis (Eisenberg et al. 2010; Portt et al. 2011; Galluzzi et al. 2011, 2012). Necroptosis plays a very important role in ischaemia, trauma, neurodegenerative disorders and various cardiovascular diseases and can be caused by various external factors. It is common that the factors inducing necrosis overlap with those characteristic of apoptosis, such as hydrogen peroxide, acetic acid, copper and manganese compounds, and certain pheromones, but in case of necrosis they occur in far higher concentrations.

Saccharomyces cerevisiae has recently become a model organism to study both necrosis and necroptosis (Eisenberg et al. 2010; Mccall 2010; Galluzzi et al. 2011). Chronologically aged yeast cells show the morphological and biochemical features of necrosis such as rupture of membrane, total disintegration of cellular organelles and absence of features characteristic of apoptosis, such as exposure of phosphatidylserine (PS) residues on the outside. Yeast necrosis can be caused by various chemical agents, including heat shock proteins such as Hsp90 and vacuolar H + ATP-ase or expression of genes involved in human neurodegenerative diseases (Eisenberg et al. 2010). It was observed that Nhp6Ap protein, the yeast homologue of chromatin-associated high mobility group Box1 protein (HMGB1), was released from the nucleus. However, it is still not confirmed whether Nhp6Ap, like HMGB1 in mammalian cells, acts as a signal of danger. It is known that endogenous polyamines play a role in the regulation of necrosis. In yeast, their removal led to an early necrotic death and the intensive production of ROS (Eisenberg et al. 2010). The natural polyamines, such as spermidine, have an antinecrotic role. Their delivery to cells promoted survival through the induction of autophagy (Braun et al. 2010; Eisenberg et al. 2010). Many yeast homologues of Eukaryota proteins playing a significant role in mammalian cell necrosis, such as: cyclophilin D, capalin, cathepsin and Hsp90, have been identified (Eisenberg et al. 2010) (Tab.3). But some like RIP kinase and factor PARP1 are still to be found.

Clear distinction between necrosis and apoptosis is of great importance as they may occur independently, sequentially and in parallel. In many cases, the type or dose of the stress signal (heat, radiation, hypoxia, certain medications, such as those from an anticancer group) determines the type of death (Denton et al. 2012).

Autophagy and its Morphological Characteristics in Yeast

Autophagy is a tightly regulated type of death that plays a role during proper cell growth, development and homoeostasis. It helps to maintain the balance between synthesis, degradation and subsequent recycling of cellular products. This ubiquitous process - observable from yeast to man – has been very carefully studied and characterized by specialist that is why we will described it very briefly (Yang and Klionsky 2010). Autophagy involves the recycling of intracellular components of a cell. It can be done by recruitment of random portions of cytoplasm by lysosomes or vacuoles, and then is called an indiscriminate autophagy or macroautophagy, respectively, which allows the cell to survive while environmental resources are limited (Inoue and Klionsky 2010). The original definition of autophagy corresponds to the process now often referred to as macroautophagy. It is becoming increasingly useful to distinguish the bulk sequestration of normal cytoplasm from the many specialized autophagic modes that have been characterized in recent years, such as chaperone-mediated autophagy, pexophagy, xenophagy and mitophagy (Klionsky 2004). In this article, the term ‘autophagy’ will be used synonymously with ‘macroautophagy’ unless otherwise stated. Macroautophagy is one of the major degradative pathways in eucaryotic cells, and it is the only process with the capacity to degrade entire organelles. The presence of large amounts of macrophagosomes, structures with a double membrane containing cytoplasm and digested organelles, massive vacuolization of cytoplasm and lack of chromatin condensation are the hallmarks of macroautophagy. Because the differentiation processes play an important role in many unicellular organisms that are human pathogens, the use of macrophagy has been suggested as a treatment of diseases caused by these organisms. The precise molecular mechanisms regulating autophagic cell death remain unclear (Denton et al. 2012) and ‘the point of no return’ in macroautophagy has not been clearly established yet (Kroemer et al. 2009).

The budding yeast Saccharomyces cerevisiae therefore proved to be an ideal organism to gain insights into Atg genes that are essential for autophagy (Nakatogawa et al. 2009; Yang and Klionsky 2010). Target of rapamycin (TOR) protein kinase regulates cell growth depending on the availability of nutrients and cellular stress (Powers et al. 2006; Kanki and Klionsky 2011). It belongs to a family of phosphatases type 2A, which regulates other proteins by phosphorylation or Atg1 complex. While the nutrients are available, kinase blocks macroautophagy through hyperphosphorylation of Atg13 protein (Denton et al. 2012). In the absence of nutrients, Atg13 protein phosphorylation is inhibited and Atg1–Atg13 complex is formed. The creation of an autophagosome is initiated by the binding of proteins and lipids to phagophore assembly site (PAS). This process is controlled by a complex composed of serine–threonine kinase Vps15, phosphatidylinositol kinase (PI 3-K), proteins Atg6 and Atg14. The activity of this complex allows the attachment of the following proteins: Atg18, Atg21, Atg20, Atg24 and Atg27. The next step in creating autophagosome is two series of reactions that are similar to ubiquitination. The first reaction is catalysed by a protease Atg4, whereas Atg8 protein is covalently bound to C-terminus to the rest of phosphatidylserine phagophore. In the second reaction, catalysed by Atg7p and Atg10p, the protein Atg12 is C-terminal covalently bound to lysine residues of protein Atg5. Two complexes, Atg8-PE and Atg5L′Atg12/Atg16, act as the phagophore magnifiers. Shortly after the formation of the complex, a fusion between autophagosome and outer vacuole membranes occurs. This results in trapping the material from autophagosome in the vacuolar cytosol, where it is then degraded. The yeast vacuole contains a lot of hydrolases, including proteases PrA, PRB, CPS and CPY. In addition, two of the ATG proteins also degrade the material from autophagosome. It is suggested that the protein Atg22 that is an integral membrane protein acts as vacuolar permease and exports regenerated material from inside of the vacuole to the cytosol (Nakatogawa et al. 2009; Inoue and Klionsky 2010; Kanki and Klionsky 2011; Denton et al. 2012). Up till now, several homologues of mammalian autophagy factors have been found.

Methods for the Detection of Cell Death in Saccharomyces cerevisiae

Methods used to detect dead cells and types of cell death in higher eucaryotes are based on both classical methods, such as cell staining followed by microscopic observations, as well as biochemical assays, which involve the use of highly specific antibodies, unambiguously assigned to a certain type of cell death. The methods used for higher eucaryote cells, with stressing their pros and cons, are revised in detail in the recent paper written by the most experienced scientists in the field (Galluzzi et al. 2009). For this reason, we summarize only the already published methods (the source included in the table) as successfully used for yeast cells (Tables 1 and 2). For those methods, we try to sum up the advantages and limitations (Table 2).

Survival Test (CFU – Colony Forming Unit; Clonogenic Assay)

Survival test is a basic, very simple and based only on the possibility of differentiation between the viable unicellular organisms still capable of division and forming colony and the dead ones. However, it does not allow a researcher to specify a death type. A known number of cells (accessed on the basis of the optical density of culture) are plated on solid medium and then grown under optimal conditions. The method assumes that each colony grown on the plate is derived from one cell. Knowing the number of cells plated and the number of colonies formed, the fraction of living (or dead) cells can be calculated. This technique is widely used to estimate the overall viability of the population of wild types or mutant strains (including knock-outs) or any other genetic yeast construct cells providing a quantitative method hardly feasible in other eucaryotes.

Microscopy and Flow Cytometry

Bright field microscopy provides a quick and inexpensive way to detect cell death still used in many laboratories. However, its application is very limited allowing cell normal and abnormal morphology observations, for example, growth of pseudohyphae or shrunk cells, which can be visualized by Nomarsky contrast (Vachova and Palkova 2005). A far more widely used method is fluorescence microscopy with both vital and exclusion dyes. Vital dyes selectively stain either live or dead cells, while exclusion dyes can go through the permeabilized plasma membranes. It provides a simple method to estimate the number of live and dead cells. In case of most methods based on specific dyes in vivo, fungal cell wall must be previously removed. Such spheroplasting is usually done by a litycase treatment (Rose et al. 1990). The fluorescence microscopy can be used for the detection of yeast apoptosis, for example, by visualization of phosphatidylserine externalization by means of Annexin V coupled to fluorescein isothiocyanate (FITC), which binds to it with high affinity, additionally enhanced by the presence of Ca2+. Moreover, the same spheroplasts can be incubated with propidium iodide (PI), which stains dead cells by the penetration through the disrupted cell membrane. A concurrent staining makes it possible to distinguish live, necrotic, early and late-apoptotic cells. This approach is a very popular. Another often used fluorescence dye is a 4′,6-diamidino-2-fenyloindol (DAPI), which has a strong affinity for DNA and intercalates between the double-stranded DNA. It allows the assessment of the degree of condensation and fragmentation of chromatin. Similar mode of action characterizes other fluorescent dyes like Hoechst and Draq5. Staining with rhodamine 123 or dihydroethidine has been successfully used to detect the kinetics of yeast production of reactive oxygen species (ROS). Another method adopted from the eucaryotic cell biology is TUNEL (terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate biotin nick-labelling). This method is based on fluorescent (or radioisotopic) labelling of the DNA free 3′-ends (formed during the degradation of DNA) with the enzyme terminal deoxyribonucleotide transferase (TdT). Apoptotic and necrotic cells are TUNEL-positive, while live cells are TUNEL-negative. Apoptotic cells are preferentially labelled in comparison with necrotic cells due to a larger number of DNA breaks in apoptotic cells. Detection of fluorescently stained DNA breaks is possible thanks to fluorescence microscopy, microtiter plate readers or flow cytometry. Each of them has its own advantages and pitfalls (Table 2), for example, microscopic observations allow also the simultaneous assessment of cell morphology; however, it is inappropriate for high-throughput applications (Galluzzi et al. 2009). To obtain reliable microscopic results, it is advised to carry out at least 3 independent experiments or to use more than 300 cells whose fluorescence intensity is recorded (Madeo et al. 2002a; Vachova and Palkova 2005).

Most of the protocols worked out for the fluorescent microscopy can be applied for flow cytometry after some optimization. The flow cytometry is often used for staining activity of caspases in vivo. This method, using the appropriate dyes and antibodies (e.g. FITC-VAD-FMK), can be applied for analysis of yeast cells (Table 1). A caspase activity can be also measured using a protocol for human HeLa cell line (Lauber et al. 2001). The spheroplasts are incubated with Ac-DEVD-AMC (N-acetyl-Asp-Glu-Val-Asp-aminomethylocumarin), whose formation is assessed by caspase activity measured by spectrofluorometer (Madeo et al. 2002b). The flow cytometry is considered to be a very powerful technique. It allows 10 000–1 000 000 measurements per sample, which gives more statistical power to the data. Recently, the 96-well plate cytofluorometers are broadly accessible, which creates much more possible applications in high-throughput screening (HTS) procedures.

A number of fine ultrastructural modifications occurring during yeast death can be observed by electron microscopy. It allows an experienced researcher to distinguish the mono- from double-membrane structures and to characterize autophagic organelles. The level of subcellular changes, such as peripheral accumulation of chromatin at the nuclear shell at an early stage of apoptosis, can be also tracked down with electron microscopy. Although electron microscopy can provide an impressive amount of ultrastructural information, the visual inspection of electron microphotographs should always be complemented by a robust quantitative approach.

Electrophoresis of DNA

The typically apoptotic cells show specific DNA cutting pattern, characteristic and reproducible ‘ladder’. Each DNA fragment consists of 180 bp or their multiplicity, which can be visualized by standard agarose DNA electrophoresis technique (Elmore 2007). In the case of Saccharomyces cerevisiae, instead of the classic ‘ladder’, a distinct smear was observed (Madeo et al. 1997). It is suggested that this may be due to the specific structure of chromatin in S. cerevisiae, characterized by the absence of DNA between nucleosomes connections (Sharon et al. 2009). This method does not seem useful or popular for the detection of yeast apoptosis.


Immunoblotting is a very precise and specific technique based on a standard laboratory equipment, however, requiring experience and expensive consumables. Moreover, it does not focus on a single cell but on a whole cell population. It means that an outcome result represents only a major fraction of cell population and does not make it possible to detect a single yeast individual. The whole cell fraction, subcellular fractions (e.g. mitochondrial fraction) as well as extracellular fluids can be used for the Western blotting analysis. A big advantage of immunoblot techniques is the possibility to detect intermembrane space (IMS) proteins released from the purified mitochondria. To detect the changes occurring during cell death in yeast cells, proteins such as cytochrome c, COXII, COXV, ATP6, pgk1p and Ilv5p were analysed by Western blot.


The field of ‘death science’ gets more and more attention resulting in a growing number of data and publications on different aspects of cell death. Annual International Meeting on Yeast Apoptosis gathers more and more scientist every year. Freely accessible databases Saccharomyces Genome Database ‘provides comprehensive integrated biological information for the budding yeast Saccharomyces cerevisiae along with search and analysis tools to explore these data, enabling the discovery of functional relationships between sequence and gene products in fungi and higher organisms’ (http://www.deathbase.org). More specific is recently launched DeathBase (Diez et al. 2010) which is a database of proteins involved in cell death. According to it's curator ‘it compiles relevant data on the function, structure and evolution of proteins involved in apoptosis and other forms of cell death in several organisms’. Data about advances in yeast research could be found in the nonreference species section. Those advances in biochemical and molecular characterization help in better understanding of the types of cell death allowing researchers to indentify the types of so-called death subroutines, that is, caspase-dependent and caspase-independent intrinsic apoptosis, extrinsic apoptosis by death receptors, and by dependence receptors, necroptosis and autophagic cell death (Galluzzi et al. 2012). We can therefore expect the formation of a new framework for more precise definition and description of the methods used for identification of yeast cell death, which could still serve as the best model for higher eucaryotes.


We thank two anonymous reviewers for helpful comments and Dr Joanna Rutkowska for support. This work was financed by National Science Centre grant (DEC-2011/01/M/NZ8/01031) to D.M.W-S.

Competing Financial Interests

The authors declare no competing financial interests.