OUP user menu

Endocytosis inhibition during H2O2-induced apoptosis in yeast

Clara Pereira, Cláudia Bessa, Lucília Saraiva
DOI: http://dx.doi.org/10.1111/j.1567-1364.2012.00825.x 755-760 First published online: 1 November 2012


Yeast revealed to be a versatile organism for studying endocytosis. Here, inhibition of endocytosis by H2O2 and its correlation with apoptotic cell death were ascertained in Saccharomyces cerevisiae. We found that H2O2 causes alterations in vacuolar morphology and a concentration-dependent inhibition of endocytosis. We also found that H2O2-induced endocytosis inhibition is a reversible process that occurs in the early phase of the apoptotic cascade, preceding chromatin condensation and DNA fragmentation. Additionally, mutants affecting early steps of the endocytic pathway display sensitivity to H2O2. As endocytosis inhibition was also observed with acetic acid, it may be a broader cellular dysfunction of oxidative stress-induced toxicity in yeast.

  • yeast
  • endocytosis
  • H2O2
  • apoptosis


Endocytosis is a general mechanism by which eukaryotic cells internalize extracellular molecules through the formation of vesicles from the plasma membrane. The opposite process, expel of specific molecules and delivery of lipids and proteins to the plasma membrane by the fusion of internal membranes, is called exocytosis. It is the control of these two processes that regulates the interaction between the cell and its environment. Besides its role in cell physiology, vesicular trafficking has also been connected to the regulation of signaling pathways and complex programs like cell cycle, mitosis, and apoptosis. Consequently, it has been implicated in human diseases (Scita & Fiore, 2010).

Reflecting its importance for the eukaryotic cell, many features of the vesicular trafficking pathways are evolutionarily conserved. Many studies on endocytosis have been performed in Saccharomyces cerevisiae, which has proven to be an extremely versatile model. For instance, several components of the endocytic process have been identified through genetic studies in yeast. It is consensual that the ability to describe this process in detail in yeast can lead to considerable understanding of endocytosis in higher eukaryotes (Novick et al., 1980; Shaw et al., 2001; Burston et al., 2009).

In mammals, the vesicular trafficking system appears to be affected by oxidative stress caused by reactive oxygen species (ROS) such as hydrogen peroxide (H2O2). Like all aerobically growing organisms, yeast also suffers exposure to moderate oxidative stress and has developed multiple mechanisms for preventing and counteracting its effects, including adaptation to increased resistance and widespread changes in gene expression (Collinson & Dawes, 1992). However, for high levels of oxidative stress, cells undergo a form of apoptotic cell death, exhibiting features like chromatin condensation and DNA fragmentation (Madeo et al., 1999). In this study, the impact of H2O2 on endocytosis and its relation with apoptotic cell death were addressed in yeast.

Materials and methods

Yeast strains, plasmids and growth conditions

Saccharomyces cerevisiae strains and plasmids used in this study are listed in Table 1. W303 strain was transformed with the YX232-mtGFP plasmid by the lithium acetate method. Yeast cultures were grown in synthetic complete (SC) medium with 0.67% (w/v) yeast nitrogen base w/o amino acids (Difco), 2% (w/v) glucose (Sigma-Aldrich) and 0.2% (w/v) dropout (mix; Sigma-Aldrich), and the required amino acids. Transformed yeasts were grown in selective SC medium lacking tryptophan. For the assays with the endocytic mutants, cells were grown in rich [YEPD; 1% (w/v) yeast extract, 2% (w/v) peptone, 2% (w/v) glucose] medium. Cells were grown to exponential phase with continuous shacking (160 r.p.m.) at 30 °C.

View this table:
Table 1

Yeast strains and plasmids used in this study

S. cerevisiae
W303Mata ura3-1 leu2–3, 112 his3–11,15 trp1-1 ade2-1 can1–100Lab collection
BY4741Mat a; his3Δ 1; leu2Δ 0; met15Δ 0; ura3Δ 0EUROSCARF collection
ede1ΔBy4741; YBL047c::kanMX4EUROSCARF collection
sla1ΔBy4741; YBL007c::kanMX4EUROSCARF collection
end3ΔBy4741; YNL084c::kanMX4EUROSCARF collection
ent1ΔBy4741; YDL161w::kanMX4EUROSCARF collection
rvs161ΔBy4741; YCR009c::kanMX4EUROSCARF collection
bzz1ΔBy4741; YHR114w::kanMX4EUROSCARF collection
vps21ΔBy4741; YOR089c::kanMX4EUROSCARF collection
vps41ΔBy4741; YDR080w::kanMX4EUROSCARF collection
ypt7ΔBy4741; YML001w::kanMX4EUROSCARF collection
YX232-GFPmtGFP under control of TP1 promoterWestermann & Neupert (2000)

Viability and calcofluor white sensitivity assays

For H2O2 viability assays, exponential cultures were treated with 1–10 mM H2O2 for up to 5 h or with 80 and 140 mM acetic acid for 1 h with shaking at 30 °C. Viability was assessed by colony-forming unit (CFU) counts after 2 days incubation at 30 °C on Sabouraud Dextrose Agar (Difco) plates and expressed as percentage of time zero. Sensitivity to calcofluor white was assessed basically as described by Ram et al. (1994). Briefly, cells were grown overnight in YEPD, and serial dilutions plated on Sabouraud Dextrose Agar plates containing 0 or 50 μg mL−1 calcofluor white (Fluka; Sigma-Aldrich). Plates were incubated in the dark at 30 °C and photographed after 2 days.

Staining with FM4-64

Endocytosis was assessed using the lipophilic dye FM4-64 (N-(3-thiethylammoniumpropyl)-4-(p-diethyl-aminophe-nylhexatrienyl) pyridinium dibromide; Molecular Probes), basically as described (Vida & Emr, 1995). Briefly, cells were incubated with 3 μg mL−1 FM4-64 for 1 h at 30 °C (time required for FM4-64 to reach the vacuole in untreated wild-type (wt) cells), and thereafter observed under a microscope. The percentage of cells with endocytic inhibition (lacking vacuolar staining) was estimated by counting at least 200 cells per sample.

Cell death assays

ROS production was assessed using dihydroethidium (DE; Sigma-Aldrich). W303 cells were incubated with 10 μg mL−1 DE for 30 min at 30 °C, washed once, and visualized under a microscope. Nuclear staining was performed in fixed cells with 2 μg mL−1 4,6-diamido-2-phenyl-indole (DAPI; Sigma-Aldrich) as described (Silva et al., 2005). DNA fragmentation was assessed by TUNEL (In Situ Cell Death Detection Kit, Fluorescein; Roche Applied Science) as described (Silva et al., 2005). Mitochondria were visualized using a mitochondria-localized green fluorescent protein (mtGFP) encoded by YX232-mtGFP. For the determination of apoptotic phenotypes, at least 200 cells per sample were evaluated.

Fluorescence microscopy

Samples were observed under an Eclipse E400 fluorescence microscope (Nikon) under appropriate filter setting. Images were captured by Digital Sight camera (Nikon DS-5Mc) with software for image acquisition (Nikon ACT-2U).

Statistical analysis

Statistical analysis was performed by Paired t-test with the software GraphPad Prism. Statistical significance was accepted at P< 0.05.


Oxidative stress disturbs endocytosis in a concentration-dependent manner

The effect of two inducers of oxidative stress and apoptosis, H2O2 and acetic acid (Madeo et al., 1999; Pereira et al., 2010), on fluid-phase endocytosis was investigated in the W303 strain using the fluorescent membrane probe FM4-64. As reported by Vida & Emr (1995), we observed that the dye initially stains the plasma membrane, followed by the cytoplasmic endocytic intermediates and finally, after 1 h incubation, the vacuolar membrane (Fig. 1a, 0 mM). However, for 1 h incubation with the dye, cells exposed to low concentrations (1–3 mM) of H2O2 exhibited defects in the movement through endocytic intermediates to the vacuole, visible by the presence of dot-like staining as described in (Vida & Emr, 1995; Fig. 1a, 3 mM). Additionally, for higher concentrations of H2O2, cells only displayed plasma membrane staining, indicating a defect in the early movement of the dye from the plasma membrane to endocytic intermediates (Fig. 1a, 5 and 10 mM). When the percentage of cells with endocytic inhibition (lacking vacuolar staining) was quantified, a close correlation with the loss of viability (assessed by CFU counts) was observed (Fig. 1b). Together, the results obtained showed that H2O2 causes a concentration-dependent inhibition of yeast endocytosis (Fig. 1a and b). The same results were obtained for the BY4741 strain (data not shown).

Figure 1

H2O2 and acetic acid lead to yeast endocytosis inhibition in a concentration-dependent manner. W303 cells were treated with the indicated concentration of H2O2 and acetic acid for 1 h and compared to control yeast (untreated cells; 0 mM). (a) Representative photomicrographs of H2O2-treated cells stained with FM4-64 (left side) and respective bright-field images (right side). (b) Quantification of viability and endocytosis inhibition (lacking vacuolar staining) for H2O2-treated cells. Values are mean ± SE (n = 3). (c) Representative photomicrographs of acetic acid-treated cells stained with FM4-64. Bar, 10 μm.

Additionally, the vacuolar morphology was also affected, with a swollen and less fragmented vacuole than that observed in untreated cells (Fig. 1a, 3 mM). A similar effect on yeast endocytosis was also observed for acetic acid. Indeed, when cells were treated with 80 and 140 mM acetic acid for 1 h, an inhibition of endocytosis, more severe for the highest concentration tested, was also detected (Fig. 1c).

Interestingly, for longer incubation times (4 h), a restoration of the vacuolar staining in cells exposed to low concentrations of H2O2 was observed (Fig. 2). However, this was not observed for higher concentrations of H2O2 (5 mM). In this case, endocytosis remained impaired except if the oxidant is removed and cells allowed to recover for 3 h in fresh media. The recovery of endocytosis occurred before the recovery in clonogenicity, which required an additional 2 h (data not shown).

Figure 2

Inhibition of endocytosis is a reversible process. W303 cells were treated with 3 mM H2O2 for 1 and 4 h. Representative photomicrographs of FM4-64 stained cells (left side) and respective bright-field images (right side). Bar, 5 μm.

Endocytosis is an early event in the H2O2-induced apoptotic program

To correlate endocytosis impairment with the apoptotic program, ROS production, mitochondrial network fragmentation, chromatin condensation and DNA fragmentation were monitored upon exposure to 5 mM H2O2. For time zero, the percentage of cells displaying these apoptotic features was close to zero (Fig. 3). However, with only 30 min treatment, the majority of cells exhibited ROS production and an almost simultaneous mitochondrial network fragmentation (Fig. 3a and b). At this time point, about half of the cells (57%) displayed a blockage in endocytosis as evidenced by the absence of vacuolar staining (Fig. 3b). After 1 h treatment, although the majority of the cells exhibited endocytic inhibition, the number of cells with chromation condensation and DNA fragmentation was still low. Indeed, a significant increase in the percentage of cells exhibiting these two apoptotic markers was only achieved with longer incubation times (5 h; Fig. 3b).

Figure 3

Endocytosis is an early event in H2O2-induced yeast apoptosis. W303 cells were incubated with 5 mM H2O2 for up to 5 h. (a) Photomicrographs illustrative of apoptotic markers were obtained with untreated (control) and H2O2-treated cells for 5 h. Bar, 10 μm. (b) Quantification of endocytosis inhibition and apoptotic markers is expressed as percentage of total cells. (c) Quantification of cell viability. In (b) and (c), values are mean ± SE (n = 3).

Endocytic mutants display different responses to H2O2-induced apoptosis

To assess the impact of vesicular trafficking on the cellular response to H2O2, several null yeast mutants with decreased endocytosis were used. From the proteins recruited at early phases of endocytosis, Ede1p (early immobile phase), Ent1p, End3p, Sla1p (Mid/late immobile phase) and Rvs161p, Bzz1p (Actin/mobile phase) (Boettner et al., 2011) were studied. From the proteins recruited at late steps of apoptosis, Vps21p required for vesicle transport, Vps41p and Ypt7 involved in transport from late endosomes to the vacuole (Lachmann et al., 2011), were studied (Fig. 4a).

Figure 4

H2O2 sensitivity for endocytosis mutants. (a) Schematic illustration of the role of the proteins under study in the endocytic pathway. Ede1 is an early factor with a role in the initiation of endocytic sites. It is followed by the assembly of the clathrin coat promoted by Ent1, followed by actin recruitment. This is promoted by a complex containing End3 and Sla1 (among others). Sla1 is a negative regulator preventing premature actin assembly. Its inhibition may be lifted by the competitive binding of Bzz1. Bzz1 along with actin generated tension and proteins like Rvs161 promotes invagination and vesicle scission. After vesicle release, the endocytic coat is disassembled and the endocytic vesicle becomes associated with actin cables and moves into the cell. The endocytic vesicle then fuses with early endosomes, a process promoted by Vps21. Early endosomes matures into late endosomes, Ypt7 is recruited and finally along with Vps41 participates in the fusion with the vacuole (adapted from Boettner et al., 2011; Lachmann et al., 2011). (b) Quantification of viability in wt (BY4741) and endocytosis mutant strains (deleted in indicated proteins) treated with H2O2 for 1 h. Values are mean ± SE (n = 4).

For 1-h treatment with 3 mM H2O2, end3Δ and rvs161Δ showed a significant sensitivity, around 28% and 34%, respectively (Fig. 4b). The remaining deletion mutants exhibited a response to H2O2 not significantly different from the wt. There was no correlation between H2O2 sensitivity and mutants with obvious defects in the endomembrane system (namely ypt7Δ, vps41Δ and vps21Δ; Fig. 5a). Because several mutants with decreased endocytosis (namely rvs161Δ, bzz1Δ, ent1Δ and ede1Δ; Fig. 5a) do not exhibit accumulation of endocytic intermediates, H2O2-induced endocytosis inhibition in wt cells may be more severe.

Figure 5

Depiction of the endocytic mutants for fluid-phase endocytosis and sensitivity to calcofluor white. (a) Representative photomicrographs of wt (BY4741) and endocytosis mutant strains (deleted in indicated proteins) stained with FM4-64 (left side) and respective bright-field images (right side). Bar, 10 μm. (b) Sensitivity to calcofluor white. Wt and endocytosis mutant strains (deleted in indicated proteins) were spotted on Sabouraud plates containing 0 (data not shown) or 50 μg mL−1 of calcofluor white and monitored after 2 days of growth. For sla1Δ and end3Δ cell concentrations were adjusted to compensate the growth defects.

Because many endocytosis mutants exhibit alteration in the cell wall, which could affect the response to H2O2, sensitivity assays using calcofluor white were performed to discard this hypothesis. Mutants with defects in the cell wall are generally more sensitive to this anionic dye that interferes with the construction and stress response of the cell wall (Ram et al., 1994). The results obtained showed a strong sensitivity to calcofluor for sla1Δ and end3Δ strains, moderate for ede1Δ and bzz1Δ strains, and normal for the remaining mutants (Fig. 5b). As such, a correlation between cell wall defects and H2O2 sensitivity was not observed.


Internalization of FM4-64 is a fast and straightforward way to monitor endocytosis in yeast. The uptake of FM4-64 is time, temperature, and adenosine-5′-triphosphate (ATP)-dependent and can be divided into two stages: movement from the plasma membrane to endocytic intermediates (stage 1) and from endocytic intermediates to the vacuole (stage 2; Ram et al., 1994). Herein, it is shown for the first time that H2O2 prevents both stages, stage 2 for low concentrations and stage 1 for higher concentrations. Recovery of endocytic trafficking upon H2O2 removal indicates that endocytosis is not permanently damaged but instead inhibited/delayed. Inhibition of endocytosis can be correlated with loss of clonogenicity, and it occured after ROS production but before chromatin condensation or DNA fragmentation. Chromatin condensation and specially DNA fragmentation are typically late apoptotic events, which occur after the majority of cells have already lost clonogenicity (Pereira et al., 2007). Additionally, endocytosis inhibition occurred almost simultaneously with mitochondrial network fragmentation that is an early dysfunction. Because both of these processes are highly dependent on the actin cytoskeleton (Pereira et al., 2010; Scita & Fiore, 2010), and it is known that H2O2 causes actin depolymerization (Vilella et al., 2005), a process with a role in apoptosis progression in yeast and other organisms (Leadsham et al., 2010). This suggests that actin damage may be involved in the observed endocytic inhibition. This hypothesis is supported by the fact that the endocytic mutants with a strong sensitivity to H2O2, rvs161Δ and end3Δ lead to strong defects in actin organization (Benedetti et al., 1994; Munn et al., 1995). Although Slap1 is also involved in actin organization, it plays a distinct role from Rvs161p and End3. In fact, unlike Rvs161p and End3, Sla1p is an inhibitor of the actin polymerization process (Holtzman et al., 1993). Because endocytic inhibition was also observed in response to acetic acid and was further reported for ethanol and heat shock (Meaden et al., 1999), it may be a cellular dysfunction common to several stressors in yeast.

As H2O2-dependent changes in endocytic trafficking in yeast can be compared with those reported for mammalian cells, the knowledge provided in this study may also contribute to the understanding of the toxicity mechanisms of oxidative stress in human diseases.


We thank B. Westermann for providing the plasmid YX232-mtGFP. This work was supported by FCT through REQUIMTE (grant no. PEst-C/EQB/LA0006/2011) and C. Pereira (SFRH/BPD/44209/2008) fellowship.


  • Editor: Jens Nielsen


View Abstract