Calcium signaling and copper toxicity in Saccharomyces cerevisiae cells

Abstract

To respond to metal surpluses, cells have developed intricate ways of defense against the excessive metallic ions. To understand the ways in which cells sense the presence of toxic concentration in the environment, the role of Ca²⁺ in mediating the cell response to high Cu²⁺ was investigated in Saccharomyces cerevisiae cells. It was found that the cell exposure to high Cu²⁺ was accompanied by elevations in cytosolic Ca²⁺ with patterns that were influenced not only by Cu²⁺ concentration but also by the oxidative state of the cell. When Ca²⁺ channel deletion mutants were used, it was revealed that the main contributor to the cytosolic Ca²⁺ pool under Cu²⁺ stress was the vacuolar Ca²⁺ channel, Yvc1, also activated by the Cch1-mediated Ca²⁺ influx. Using yeast mutants defective in the Cu²⁺ transport across the plasma membrane, it was found that the Cu²⁺-dependent Ca²⁺ elevation could correlate not only with the accumulated metal, but also with the overall oxidative status. Moreover, it was revealed that Cu²⁺ and H₂O₂ acted in synergy to induce Ca²⁺-mediated responses to external stress.

Introduction

Heavy metal stress is an important environmental burden caused by numerous anthropomorphic activities such as smelting, mining, industrial manufacturing, sewage and industrial waste, metal-containing pesticides and fertilizers, transportation, etc. (He et al. 2005; Tchounwou et al. 2012). Many heavy metals accumulate in living organisms, which may be subsequently transferred to human body via the food chain, leading to a variety of diseases (Järup 2003; Boyd 2010; Gall et al. 2015). Heavy metals are tricky pollutants because they are natural components of the earth crust, and unlike the organic pollutants, they are non-degradable (Wuana and Okieimen 2011). Of special interest are the metals essential for life (Cu, Co, Fe, Mn, Ni, Zn) which are necessary in minute amounts and for which cells have developed intricate mechanisms of uptake, intracellular traffic, buffering, and storage. Nevertheless, even the essential metals become toxic when the environmental levels get higher than the physiological threshold, due to excessive accumulation which leads to rapid and non-specific binding to biomolecules (Khan et al. 2015). For this reason, an immediate response to the sudden changes in heavy metal concentrations is essential for cell survival and adaptation.

The living organisms have developed various mechanisms to respond to environmental insults, and one of the conserved ways to transmit an external signal into the eukaryotic cell is by inducing transient elevations in Ca²⁺ concentrations within the cytosol ([Ca²⁺]_cyt). Acting as a second messenger, Ca²⁺ triggers a variety of cascade responses by temporarily activating Ca²⁺-binding components of signaling pathways which can lead either to adaptation to the environmental changes or to cell death (Bootman et al. 2012).

In a previous study, we investigated the involvement of Ca²⁺ in the response to high Mn²⁺, Co²⁺, Ni²⁺, Cu²⁺, Zn²⁺, Cd²⁺, or Hg²⁺ using Saccharomyces cerevisiae as model. This study revealed that the yeast cells responded through sharp increases in [Ca²⁺]_cyt when exposed to high Cd²⁺, and to a lesser extent to Cu²⁺, but not to Mn²⁺, Co²⁺, Ni²⁺, Zn²⁺, or Hg²⁺ (Ruta et al. 2014). In the present study, we focused on investigating the role of Ca²⁺ in mediating the cell response to high concentrations of external Cu²⁺ and we found that the cell exposure to high Cu²⁺ determined broad and prolonged [Ca²⁺]_cyt waves which showed a different pattern from the [Ca²⁺]_cyt pulses induced by high Cd²⁺.

Copper is one of the most important essential transition metals, and a variety of enzymes require copper as a cofactor for electron transfer reactions (De Freitas et al. 2003). Nevertheless, when in excess, copper is very toxic in the free form because of its ability to produce free radicals when cycling between oxidized Cu²⁺ and reduced Cu⁺¹. Copper uptake, buffering, and traffic in S. cerevisiae have been extensively studied and reviewed (Nevitt et al. 2012). Under high concentration conditions, Cu²⁺ is transported by the low-affinity plasma membrane transporters Fet4 (Hassett et al. 2000) and Smf1 (Cohen et al. 2000), which can also transport Mn²⁺, Fe²⁺, and Zn²⁺ to fulfill the cellular demand for these cations. Additionally, Pho84, a high-affinity inorganic phosphate transporter, was found to act also as a low-affinity transporter for divalent cations, including Cu²⁺ (Jensen et al. 2003). The Cu²⁺ ions are also transported by the cell surface Fet3/Ftr1 high-affinity iron uptake system which also acts as a Cu¹⁺ oxidase (Fet3) and Cu²⁺ transporter (Ftr1), being induced by high copper (Labbé et al. 1999; Gross et al. 2000).

There are numerous reports indicating that yeasts use Ca²⁺-mediated signaling to respond to a variety of environmental stimuli (Batiza et al. 1996; Kanzaki et al. 1999; Locke et al. 2000; Matsumoto et al. 2002; Pinontoan et al. 2002; Viladevall et al. 2004; Popa et al. 2010; Rao et al. 2010; Courchesne et al. 2011; Roberts et al. 2012; Ruta et al. 2014; Rigamonti et al. 2015). Ca²⁺ acts as a second messenger in response to certain stimuli, a case when its concentration in the cytosol ([Ca²⁺]_cyt) increases abruptly; in yeast, this can be a consequence of calcium influx via the Cch1/Mid1 Ca²⁺ channel on the plasma membrane (Batiza et al. 1996; Catterall 2000; Matsumoto et al. 2002), release of vacuolar Ca²⁺ into the cytosol through the vacuole-located Ca²⁺ channel Yvc1 (Palmer et al. 2001; Denis and Cyert 2002), or both (Popa et al. 2010; Khan et al. 2015). Once the message is delivered, the normal very low level of [Ca²⁺]_cyt is restored through the action of Ca²⁺ pumps and exchangers, reviewed by Cunningham (2011).

In the present study, the Ca²⁺-dependent response to surplus Cu²⁺ in the yeast environment was investigated.

Materials and methods

Yeast strains, yeast manipulation, plasmids, and growth media

The S. cerevisiae strains used in this study were isogenic to the “wild-type” (WT) parental strain BY4741 (MATa; his3Δ1; leu2Δ0; met15Δ0; ura3Δ0) (Brachmann et al 1998). The knock-out mutants used were cch1Δ, cnb1Δ, mid1Δ, yvc1Δ, fet4Δ, smf1Δ, pho84Δ, fet3Δ, ftr1Δ, and yap1Δ. All strains were purchased from EUROSCARF (www.euroscarf.de). Cell storage, growth, and manipulation were done as described by Sherman et al. (1986), using yeast extract–polypeptone–dextrose (YPD) or in synthetic complete dextrose (SD) media lacking specific amino acids when selective conditions were imposed. For solid media, 2 % agar was used. Unless otherwise specified, all chemicals were purchased from Sigma-Aldrich.

In vivo monitoring of [Ca²⁺]_cyt

Monitoring of [Ca²⁺]_cyt changes in cells exposed to external stimuli was done using an apo-aequorin complementary DNA (cDNA) expression system (Nakajima-Shimada et al. 1991). In this study, yeast strains were transformed with the multicopy plasmid pYX212-cytAEQ harboring the apo-aequorin cDNA under the control of a constitutive yeast promoter (Tisi et al. 2002). Yeast transformation was performed by a modified lithium acetate method (Schiestl and Gietz 1989). Transformed cells were maintained on SD-Ura selective medium and prepared for Ca²⁺-dependent luminescence detection as described (Tisi et al. 2015) with slight modifications. Overnight pre-cultures of cells expressing apo-aequorin were diluted to density 5 × 10⁶ cells/mL with fresh SD-Ura supplemented with 10 μM CuCl₂ and incubated with shaking (200 rpm) at 28 °C for four additional hours. At this concentration, the supplemental CuCl₂ was completely non-toxic, but it suppressed the expression of the CTR1 gene which encodes the high-affinity transporter for Cu (Dancis et al. 1994). Cells were harvested by centrifugation and resuspended in SD-Ura containing 10 μM CuCl₂ (10⁸ cells/mL). To reconstitute functional aequorin, native coelenterazine was added to the cell suspension (from a methanol stock, 50-μM final concentration) and the cells were incubated for 1 h at 28 °C in the dark. Cells were treated with coelenterazine sequentially, maintaining a constant time of incubation before addition of stressors. The excess coelenterazine was washed away by centrifugation with 0.1 M 2-(N-morpholino)ethanesulfonic acid (MES)–Tris buffer, pH 6.5, prepared with Ca²⁺-free reagents and deionized water. The cells (approximately 10⁷ cells/determination) were finally resuspended in 0.1 M MES–Tris buffer, pH 6.5, containing 1 mM CaCl₂ and transferred to the luminometer tube. To evaluate the requirement of extracellular Ca²⁺contribution, the coelenterazine-treated cells were washed three times with 1 M MES–Tris buffer, pH 6.5, containing 20 mM MgCl₂ and re-suspended in 0.1 M MES–Tris buffer, pH 6.5, containing 10 μM (1,2-bis(o-aminophenoxy) ethane-N,N,N′,N′-tetraacetic acid) (BAPTA, cell-impermeant Ca²⁺ chelator). A cellular luminescence baseline was determined for each strain by approximately 1 min of recordings at 1/s intervals. Thirty seconds after ensuring a stable signal, chemicals tested were injected from sterile stocks to give the final concentrations indicated, and the Ca²⁺-dependent light emission was monitored in a single tube luminometer (Turner Biosystems, 20ⁿ/20). The light emission was measured at 1 s intervals for at least 10 min after the stimulus and reported as relative luminescence units/second (RLUs/s). To ensure that the total reconstituted aequorin was not limiting in our assay, at the end of each experiment, aequorin expression and activity were checked by lysing cells with 1 % Triton X-100; only the cells with considerable residual luminescence were considered. Relative luminescence emission was normalized to an aequorin content giving a total light emission of 10⁶ RLUs in 10 min after lysing cells with 1 % Triton X-100. The relative maximum luminescence (RLM) was the average of the RLUs flanking the maximum value (ten values on each side) relatively to the average luminescence baseline recorded before cells were exposed to Cu²⁺.

Assessment of cell growth using spot assay

Overnight pre-cultures were washed and shifted to fresh medium (to approximately 10⁶ cells/mL) then grown for 4 h at 28 °C under strong agitation (200 rpm). The exponentially growing cells were serially diluted in sterile water and stamped on agar plates containing various chemicals using a pin replicator (approximately 4 μL/spot). Plates were photographed after 3–4 days of incubation at 28 °C. Chemicals were added to the agar media from sterile stocks, after autoclaving. To mimic calcium depletion, the medium was supplemented with increasing concentrations of the Ca²⁺ chelator ethylene glycol-bis(2-aminoethylether)-N,N,N′,N′-tetraacetic acid (EGTA).

Intracellular copper assay

Copper loading of the cells was done as follows. Overnight pre-cultures were diluted in fresh YPD to density 10⁶ cells/mL. The cells were incubated with shaking for four additional hours at 28 °C before CuCl₂ was added from sterile stocks. For metal accumulation assay, the cells were harvested at various times by centrifugation and washed three times with ice-cold solutions containing 1 mM MES–Tris buffer, pH 6.0, and 5 mM ethylenediaminetetraacetic acid (EDTA, tetrasodium salt) for a better removal of the cell wall-bound copper. All centrifugation (1 min, 5000 rpm) was done at 4 °C. Cells were finally suspended in deionized water (10⁹ cells/mL) and used for metal assay. The samples were digested for 24 h with nitric acid of ultrapure grade (Merck, Germany) and stabilized in Tris/HCl buffer (pH 8). Apart from cell digestion, HNO₃ ensures the complete oxidation of Cu¹⁺ to Cu²⁺. The copper released by cell digestion was determined colorimetrically with the chromogenic reagent bis(cyclohexanone) oxaldihydrazone (Cuprizone) as described (Marczenko and Balcerzak 2000) and normalized to cell total protein as nanomoles copper per milligram cell protein. Cellular total protein was determined by the method described by Bradford (1976) using a Shimadzu UV–vis spectrophotometer (UV mini-1240).

Reproducibility of the results and statistics

All experiments were repeated at least three times, and only those with trends that were fully consistent among the independent experiments were considered. Values in graphs are means ± standard deviation (SD). For aequorin luminescence determinations, traces represent the mean ± SD from three independent experiments performed on different days (n = 3). For copper accumulation experiments, the values were expressed as the mean ± SD of triplicate determinations made on 2 days (n = 6). For visual results (photographs), one representative example is shown.

Student’s t test was used for the statistical analysis of each strain in control conditions compared with Cu²⁺ treatment. A p < 0.01 was deemed indicative of a statistically significant difference for these tests. For copper accumulation, the analysis of mutant strains compared with the wild type was performed using a one-way analysis of variance followed by Dunnett’s test for multiple comparisons. A p < 0.05 was deemed indicative of a statistically significant difference for these tests. *p < 0.05, **p < 0.01, and ***p < 0.001.

Results and discussion

Yeast cells exposed to high Cu²⁺ respond through slow but transient increase in [Ca²⁺]_cyt

In a previous study, we showed that the presence of high Cd²⁺ in the environment induced sudden [Ca²⁺]_cyt elevations in normal S. cerevisiae cells. Other divalent metal ions tested showed no effect with the exception of Cu²⁺, whose presence in the growth media induced broad elevations of [Ca²⁺]_cyt (Ruta et al. 2014). To determine whether the cell response to high Cu²⁺ was mediated by Ca²⁺ in yeast cells, we made use of the Ca²⁺-induced luminescence of a photoprotein, aequorin, a system suitable for detecting transient modifications in the [Ca²⁺]_cyt (Nakajima-Shimada et al. 1991). For this purpose, yeast cells were transformed with a plasmid harboring the cDNA of the luminescent Ca²⁺ reporter apo-aequorin under the control of a constitutive promoter, which afforded abundant transgenic protein within the cytosol (Tisi et al. 2002). Transgenic parental (wild type (WT)) cells expressing apo-aequorin were pre-treated with the cofactor coelenterazine to reconstitute functional aequorin, and then, the cells were exposed to Cu²⁺ shocks directly in the luminometer tube. By testing various concentrations, it was noticed that the cells responded to external Cu²⁺ ions through slow, but transient luminescence peaks which indicated similar elevations in the [Ca²⁺]_cyt (Fig. 1a). It was noted that while the wild-type cells exposed to high Cd²⁺ had required no more than 100 s to restore the low level of [Ca²⁺]_cyt (Ruta et al. 2014), the cells exposed to high Cu²⁺ needed longer time to recover after the Cu²⁺ shock by restoring the normal levels of [Ca²⁺]_cyt (Fig. 1a). The maximum luminescence recorded increased with Cu²⁺ concentration (Fig. 1a), to reach a plateau at concentrations higher than 0.5 mM (Fig. 1b). Inversely, the time required to achieve the maximum luminescence decreased when increasing the concentration of external Cu²⁺ (Fig. 1c). For Cu²⁺ concentrations between 0.1 and 0.5 mM, the time to reach the maximum luminescence was higher than 200 s (Fig. 1a), approximately ten times slower than the response to Cd²⁺, for which the luminescence peak was reached after approximately 20 s following the exposure to Cd²⁺ shock (Ruta et al. 2014). The luminescence traces recorded for Cu²⁺ were broad even when the cells were exposed to concentrations as high as 10 mM (data not shown), indicating that high Cu²⁺ induced [Ca²⁺]_cyt elevations, but the cell response was different from the response to high Cd²⁺.

Fig. 1

Changes in the [Ca²⁺]_cyt upon exposure to exogenous Cu²⁺. Wild-type cells BY4147 were transformed with the plasmid pYX212-cytAEQ, and the cells expressing coelenterazine-reconstituted aequorin were exposed to various concentrations of Cu²⁺ directly in the luminometer tube as described in “Materials and methods” section. Cu²⁺-induced Ca²⁺-dependent aequorin luminescence was recorded on samples of approximately 10⁷ cells and normalized to an aequorin content giving a total light emission of 10⁶ RLUs in 10 min after lysing cells with 1 % Triton X-100. a Effect of Cu²⁺ concentration upon Ca²⁺-dependent luminescence. Traces showing Ca²⁺-dependent aequorin luminescence in response to increasing Cu²⁺ concentrations: 0.1 mM (purple), 0.2 mM (yellow), 0.5 mM (green), and 1 mM (blue). The arrow indicates the time when Cu²⁺ (in the form of CuCl₂ stock solution) was added. Dotted lines represent the luminescence traces recorded in cells transformed with the control vector (pYX212). The luminescence traces represent the mean ± SD from independent experiments performed on three different days. b Effect of Cu²⁺ concentration on the maximum intensity of the Ca²⁺-dependent aequorin luminescence. The relative maximum luminescence (RLM) was the average of the RLUs flanking the maximum value recorded (10 values on each side) divided by the average luminescence recorded before cells were exposed to Cu²⁺. c Time required by aequorin-expressing cells to attain maximum Ca²⁺-dependent aequorin luminescence following exposure to Cu²⁺. In b, c values are mean ± SD of three determinations made on two distinct experiments (n = 6)

Ca²⁺-mediated response to high Cu²⁺ depends predominantly on internal Ca²⁺ stores

To further characterize the Ca²⁺ response to exogenous Cu²⁺, we examined whether the Cu²⁺-dependent Ca²⁺ fluxes had external or internal source. In the wild-type cells expressing functional aequorin, the Cu²⁺ exposure induced broad and transient luminescence peaks caused by the increase in the cytosolic Ca²⁺ (Fig. 2a, full line). The peak intensity was attenuated when measurements were done in Ca²⁺-free resuspension buffer (Fig. 2a, dotted line), suggesting that the Cu²⁺-dependent increase in [Ca²⁺]_cyt depended on both external and internal Ca²⁺ stores. Under stress conditions, the increase in [Ca²⁺]_cyt is mainly the result of the plasma membrane Ca²⁺ channel Cch1/Mid1 activity (Batiza et al. 1996; Matsumoto et al. 2002; Rao et al. 2010) and/or of the vacuole-located Ca²⁺ channel Yvc1 (Palmer et al. 2001; Denis and Cyert 2002). To determine whether the Cu²⁺-mediated release of [Ca²⁺]_cyt occurs through these channels, we measured the Cu²⁺-induced luminescence of knock-out cch1Δ, mid1Δ, or yvc1Δ mutant cells expressing the apo-aequorin cDNA. It was noticed that the cells lacking Cch1 showed a lower increase in [Ca²⁺]_cyt when exposed to Cu²⁺ surplus (Fig. 2b, full line), indicating that Cch1 is required to achieve the [Ca²⁺]_cyt maximum peak. The luminescence trace did not change significantly in cch1Δ cells resuspended in Ca²⁺-free buffer (Fig. 2b, dotted line), suggesting that the Ca²⁺ ions that come from outside the cell enter mainly through the Cch1/Mid1 channel and that Cch1 integrity is essential for Ca²⁺ entry under high Cu²⁺. Nevertheless, as an increase in [Ca²⁺]_cyt, could still be recorded in cch1Δ cells, other ways of Ca²⁺ release are likely to exist under Cu²⁺ stress, when probably the Ca²⁺ internal stores are also mobilized. In contrast to cch1Δ cells, the mid1Δ cells responded to high Cu²⁺ similarly to the wild-type cells (Fig. 2c, full line) exhibiting an attenuation of the Ca²⁺-dependent luminescence in Ca²⁺-free buffer (Fig. 2c, dotted line). This observation indicated that if the Ca²⁺ ions entered the cell via the Cch1/Mid1 channel, Cch1 would be sufficient for inducing [Ca²⁺]_cyt elevations in response to Cu²⁺. As the luminescence trace was broader in mid1Δ than in the wild-type cells, it is highly probable that Mid1 is necessary mainly for tuning the Ca²⁺ influx through Cch1 in response to high Cu²⁺. On the other hand, in aequorin-expressing yvc1Δ cells, lacking the vacuole-located Ca²⁺ channel Yvc1p, the Cu²⁺-dependent [Ca²⁺]_cyt pulse was almost undetectable (Fig. 2d), suggesting that Yvc1p is the main contributor to the [Ca²⁺]_cyt pool under Cu²⁺ stress. It was reported that the release of vacuolar Ca²⁺ via Yvc1p is stimulated by the Ca²⁺ ions which enter the cytosol from outside or are released from the vacuole by Yvc1p itself in a positive feedback (Palmer et al. 2001; Denis and Cyert 2002); therefore, it is possible that Yvc1p massively releases Ca²⁺ into the cytosol only when the Cu²⁺-induced Ca²⁺, which enters the cell via the Cch1p channel, reaches a critical threshold. This would explain why the cells lacking the Cch1 (but having intact Yvc1) responded through less strong, but nonetheless significant [Ca²⁺]_cyt elevations. It was demonstrated that the release of Ca²⁺ from intracellular stores stimulates the extracellular Ca²⁺ influx, a process known as capacitative calcium entry (Locke et al. 2000), which is probably blocked in cch1Δ cells. Inversely, the release of vacuolar Ca²⁺ via Yvc1p can be stimulated by small amounts of Ca²⁺ from outside the cell as well as by the Ca²⁺ released from the vacuole by Yvc1p itself, in a positive feedback called Ca²⁺-induced Ca²⁺ release (CICR) (Palmer et al. 2001; Zhou et al. 2003; Su et al. 2009a, b), thus explaining the almost complete lack of [Ca²⁺]_cyt increase in cells devoid of Yvc1, but having a functional Cch1/Mid1 system.

Fig. 2

Effect of mutations affecting Ca²⁺ transport upon [Ca²⁺]_cyt elevations in response to Cu²⁺. Isogenic strains expressing coelenterazine-reconstituted cytAEQ were exposed to Cu²⁺ (1-mM final concentration) directly in the luminometer tube as described in “Materials and methods” section. Dotted lines represent the traces showing Ca²⁺-dependent aequorin luminescence of cells resuspended in Ca²⁺-free buffer. The arrows indicate the addition of Cu²⁺. The luminescence traces represent the mean ± SD of independent experiments performed on three different days. a Wild-type strain. b Null-mutant cch1Δ strain. c Null-mutant mid1Δ strain. d Null-mutant yvc1Δ strain. Inset: as main figure, but magnified. e Effect of Cu²⁺ on the growth of mutants with defects in Ca²⁺ transport and homeostasis. Exponentially growing yeast cells from the isogenic strains WT, cch1Δ, mid1Δ, yvc1Δ, and cnb1Δ were 10-fold serially diluted (from 10⁷ cells/mL, left) and were spotted (approximately 4 μL) on SD plates supplemented with CuCl₂. Cells were photographed after 3–4-day incubation at 28 °C. The experiments were repeated three times and the results were similar. One representative set of plates is shown

Although different in intensity, the Cu²⁺-induced [Ca²⁺]_cyt elevations in wild-type, cch1Δ, mid1Δ, or yvc1Δ cells followed a pattern similar to the [Ca²⁺]_cyt elevations induced in these strains by H₂O₂ (Popa et al. 2010). This observation prompted the idea that Cu²⁺ may indirectly induce [Ca²⁺]_cyt elevations by generating oxidative stress.

Mutants with defects in Ca²⁺ homeostasis exhibit various Cu²⁺ tolerance phenotypes

As Cu²⁺ stress was signaled within the wild-type cells via elevations in the [Ca²⁺]_cyt, the next step was to check the tolerance to high Cu²⁺ of the mutants defective in Ca²⁺ transport and homeostasis (Fig. 2e). It was noticed that cch1Δ sensitivity to Cu²⁺ was slightly lower than that of the wild type, while and yvc1Δ cells were Cu²⁺-tolerant (Fig. 2e). The mid1Δ cells were clearly more sensitive than the wild type; thus, the sensitivity to Cu²⁺ paralleled the sensitivity to H₂O₂ (Popa et al. 2010). In terms of Cu²⁺ uptake, the strains were not significantly different (data not shown), indicating that the differences in the tolerance to Cu²⁺ are not the result of different Cu²⁺ accumulation but are rather caused by the different ways in which the cells signal the presence of high Cu²⁺.

The sudden increase in [Ca²⁺]_cyt under environmental conditions forms the basis of Ca²⁺ action as a second messenger. Once in the cytosol, the Ca²⁺ ions bind to the universal Ca²⁺ sensor protein calmodulin, which in turn can bind and activate calcineurin. Calcineurin is the protein phosphatase required for yeast to adapt to a variety of environmental stresses (Cyert 2003) by regulating calcium homeostasis at both transcriptional (via the transcription factor Crz1p) and post-transcriptional levels (for review, Cunningham 2011). Since Cu²⁺-induced [Ca²⁺]_cyt elevations may induce calcineurin activation, the effect of high Cu²⁺ on cnb1Δ cells, defective in the regulatory subunit of calcineurin, was tested. It was found that the cnb1Δ cells exhibited Cu²⁺ tolerance similar to yvc1Δ cells (Fig. 2e), suggesting that high Cu²⁺ may activate calcineurin-dependent pathways which are detrimental to cell survival under high Cu²⁺. This phenotype also paralleled the tolerance to H₂O₂ of cnb1Δ cells (Popa et al. 2010).

The yeast mutants with defects in the low-affinity Cu²⁺ uptake exhibit attenuated Cu²⁺-dependent [Ca²⁺]_cyt elevations

Copper is a redox-very active metal, and in yeast treated with copper, reactive oxygen species are generated via Fenton/Haber–Weiss pathways both in the reaction medium and inside the cell (Liang and Zhou 2007). As such, the question which arose was whether Cu²⁺ was more active in generating a [Ca²⁺]_cyt pulse from outside or after entering the cytosol. To find an answer to this question, yeast mutants with defects in Cu²⁺ transport across the plasma membrane under high concentration conditions were further investigated. Under higher-than-normal concentrations, three low-affinity transporters were reported to transport Cu²⁺ across plasma membrane: Fet4, Smf1, and Pho84 (Cohen et al. 2000; Hassett et al. 2000; Jensen et al. 2003). Under normal to low concentrations, Cu²⁺ is transported across the yeast plasma membrane by the high-affinity transporter Ctr1p. CTR1 transcription is strongly regulated by Cu²⁺ availability, and it is completely abolished at Cu²⁺ as low as 10 μM (Dancis et al. 1994). As CTR1 is not active under Cu²⁺ surplus and to avoid interference of with the low-affinity transporters, the CTR1 transcription was inhibited by growing the cells in media containing 10 μM CuCl₂, a concentration which was completely non-toxic to the yeast cells.

Cells defective in Fet4, Smf1, and Pho84 transporters are more tolerant to high Cu²⁺ (data not shown) apparently due to decreased Cu²⁺ accumulation (Cohen et al. 2000; Hassett et al. 2000; Jensen et al. 2003). The Cu²⁺ uptake of knock-out mutants fet4Δ, smf1Δ, and pho84Δ was recorded in the first 10 min of Cu²⁺ exposure (time corresponding to the average duration of the Cu²⁺-dependent [Ca²⁺]_cyt elevation), and it was found that in all three mutants, Cu²⁺ accumulated less than in the wild-type cells (Fig. 3a).

Fig. 3

Effect of mutations affecting low-affinity Cu²⁺ uptake upon [Ca²⁺]_cyt elevations in response to high Cu²⁺. a Cu²⁺ accumulation. Exponentially growing yeast cells were shifted to SD medium supplemented with 1 mM CuCl₂ and were incubated with shaking (200 rpm) at 28 °C for 5 or 10 min before cells were harvested for determination of Cu²⁺ accumulation, as described in “Materials and methods” section. Values are mean ± SD of three determinations made on two distinct experiments (n = 6). *p < 0.05, **p < 0.01, significantly lower than the control strain. b–e Cu²⁺-induced mobilization of [Ca²⁺]_cyt in yeast cells defective in low-affinity Cu²⁺ transport across plasma membrane. The yeast cells expressing coelenterazine-reconstituted cytAEQ were exposed to Cu²⁺ directly in the luminometer tube. The arrows indicate the moment when 1 mM CuCl₂ was added from a sterile stock. b Wild-type strain. c Null-mutant fet4Δ strain. d Null-mutant smf1Δ strain. e Null-mutant pho84Δ strain. Strains were pre-grown in SD-Ura supplemented with 10 μM CuCl₂ for inhibition of the high-affinity transporter Ctr1. The luminescence traces represent the mean ± SD from independent experiments performed on three different days

To see if the lower Cu²⁺ accumulation modifies the Cu²⁺-induced [Ca²⁺]_cyt elevations, the Ca²⁺-dependent luminescence in cells expressing transgenic aequorin in response to Cu²⁺ was recorded. It was noted that fet4Δ cells, having the lowest level of Cu²⁺ accumulation in the first 10 min of Cu²⁺ exposure, also exhibited lower Cu²⁺-dependent [Ca²⁺]_cyt elevations (Fig. 3b). The smf1Δ and pho84Δ cells also accumulated less Cu²⁺, but the Cu²⁺-dependent [Ca²⁺]_cyt elevations were stronger compared to fet4Δ cells, but still weaker than in the wild-type cells (Fig. 3c, d). This observation suggested that Cu²⁺ ions must enter the cell to exert their full action on inducing [Ca²⁺]_cyt elevation.

The yeast yap1Δ knock-out mutant is tolerant to Cu²⁺

The different tolerance to Cu²⁺ of WT, cch1Δ, mid1Δ, yvc1Δ, and cnb1Δ cells paralleled the tolerance of these mutants to H₂O₂ (Popa et al. 2010), suggesting that Cu²⁺ might induce [Ca²⁺]_cyt indirectly, by generating reactive oxygen species. In yeast, Cu²⁺ is a potent superoxide (ubiquitous in all the aerobic cells) scavenger, easily accepting one electron from this anion radical to generate H₂O₂ (Lapinskas et al. 1995). If this were the case, mutants with altered tolerance to H₂O₂ might behave differently when exposed to high Cu²⁺. We therefore tested the tolerance to Cu²⁺ of a H₂O₂-hypersensitive mutant, yap1Δ. This mutant lacks Yap1, the extensively studied transcription factor which regulates the response to oxidative and Cd²⁺ stress (Wu et al. 1993; Kuge and Jones 1994); cells lacking a functional Yap1 are hypersensitive to both H₂O₂ and Cd²⁺. Surprisingly, it was found that the yap1Δ cells were slightly more tolerant to Cu²⁺ than the wild-type cells (Fig. 4a), an observation which further supported the idea that Ca²⁺-mediated response to high Cu²⁺ follows a different mechanism from that involved in the tolerance to Cd²⁺. When recording the [Ca²⁺]_cyt in yap1Δ cells expressing functional aequorin, it was seen that exposure to high Cu²⁺ induced a sudden and sharp peak of Ca²⁺-dependent luminescence which lasted only a few seconds before the Ca²⁺ levels were restored to normal (Fig. 4b). This observation suggested that sudden and short pulses of [Ca²⁺]_cyt may be benefic, orienting the cell toward adaptation pathways. In yap1Δ cells, the oxidative state is altered due to low expression of the genes regulated by Yap1; thus, the Cu²⁺-related oxidative stress would add up to generate an overall shock transduced in the sharp [Ca²⁺]_cyt elevations. It was shown before that strong and brief [Ca²⁺]_cyt pulses can be correlated with adaptation to external stimuli (Ruta et al. 2014), explaining why YAP1 deletion results in gained tolerance to high Cu²⁺. In yap1Δ cells, the Cu²⁺-induced [Ca²⁺]_cyt elevation depended mainly on the Ca²⁺ internal stores, since exposing the cells to Cu²⁺ in Ca²⁺-free resuspension buffer did not result in significant attenuation of Ca²⁺-dependent cell luminescence (Fig. 4c). This observation suggested that the oxidative state inside the yap1Δ boosted the Cu²⁺ action by triggering the Ca²⁺ release from the internal stores. In this line of evidence, the Ca²⁺-dependent luminescence of aequorin-expressing yap1Δ was considerably attenuated by cell pre-incubation with N-acetyl-cysteine (NAC), a cell-permeant antioxidant which elevates the intracellular reduced glutathione levels (Fig. 4d).

Fig. 4

Effect of Cu²⁺ on yap1Δ cells. a YAP1 deletion results in Cu²⁺ tolerance. Exponentially growing cells from the isogenic strains WT and yap1Δ were 10-fold serially diluted (from 10⁷ cells/mL, left) and were spotted (approximately 4 μL) on SD plates supplemented with the indicated concentrations of CuCl₂. Cells were photographed after 3-day incubation at 28 °C. The experiments were repeated three times, and the results were similar. One representative set of plates is shown. b–d Changes in the [Ca²⁺]_cyt in yap1Δ cells upon exposure to Cu²⁺. The yap1Δ cells expressing coelenterazine-reconstituted aequorin were exposed to 1 mM Cu²⁺ directly in the luminometer tube, as described in the “Materials and methods” section. The arrow indicates the time when Cu²⁺ (in the form of CuCl₂ stock solution) was added. b No pre-treatment. c Cells pre-grown in SD supplemented with Ca²⁺ chelator BAPTA (1 mM) and finally resuspended in Ca²⁺-free buffer. d Cells pre-grown in SD supplemented with the cell-permeant antioxidant N-acetyl-cysteine (NAC, 5 mM). The luminescence traces represent the mean ± SD from independent experiments performed on three different days

The yeast mutants fet3Δ and ftr1Δ exhibit increased tolerance to H₂O₂

As the study progressed, it became clear that there is a connection between the Ca²⁺-mediated response to high Cu²⁺ and the oxidative state of the cell. Apart from the low-affinity transporters Fet4, Smf1, and Pho84, Cu²⁺ is also transported by the Ftr1/Fet3 complex, in which Fet3 oxidizes the deleterious Cu¹⁺ to the less toxic Cu²⁺ at the cell surface, the Cu²⁺ ions being subsequently transported into the cell by the Ftr1 (Shi et al. 2003; Stoj and Kosman 2003). As the Ftr1/Fet3 complex is induced by high Cu²⁺ (Gross et al. 2000), the behavior of cells defective in either of the two components was investigated further.

First, we tested the effect of exogenous oxidative stress generated by H₂O₂ exposure upon the growth of mutants with defective Cu²⁺ transport across the plasma membrane. It was found that while fet3Δ, smf1Δ, and pho84Δ showed similar sensitivity to exogenous H₂O₂ as the wild type, fet3Δ, and ftr1Δ were clearly more tolerant (Fig. 5a). The mutants’ behavior flipped when the medium was supplemented with the cell-permeant antioxidant N-acetyl-cysteine (NAC, Fig. 5a), indicating that a more reduced environment is deleterious to the cells defective in the Fet3/Ftr1 system. It is largely believed that the growth defects of fet3Δ and ftr1Δ mutants are the result of the inability to efficiently oxidize Cu¹⁺ to Cu²⁺ at the cell surface (Shi et al. 2003), a defect which can be either alleviated by exogenous oxidants (in this case H₂O₂) or augmented by reducing agents (in this case NAC). In this line of evidence, it was noted that the sensitivity to high Cu²⁺ of fet3Δ and ftr1Δ was mitigated by the presence of exogenous H₂O₂ (Fig. 5b). Numerous studies indicate that copper toxicity is not given by the amount accumulated, but rather by the amount of osmotically free ions, the Cu⁺¹ ions being the most deleterious (Shi et al. 2003; Stoj and Kosman 2003). Under such circumstances, the mitigation of copper toxicity by H₂O₂ in fet3Δ and ftr1Δ mutants may be, at least in part, the result of H₂O₂-related Cu⁺¹ oxidation at the cell surface.

Fig. 5

H₂O₂ alleviates the Cu²⁺ sensitivity of yeast cells defective in the complex Fet3/Ftr1. Exponentially growing cells from the isogenic wild-type and knock-out strains were 10-fold serially diluted (from 10⁷ cells/mL, left) and were spotted (approximately 4 μL) on SD plates supplemented with the indicated chemicals. Cells were photographed after 3-day incubation at 28 °C. The experiments were repeated three times and the results were similar. One representative set of plates is shown. a Effect of exogenous oxidative stress (H₂O₂, 4 mM), antioxidants (N-acetyl-cysteine, NAC, 10 mM), or Ca²⁺ depletion (EGTA, 15 mM) on yeast mutants defective in Cu²⁺ transport across plasma membrane. b Effect of exogenous H₂O₂ on the Cu²⁺ sensitivity of yeast cells defective in Fet3/Ftr1 complex

Cu²⁺ and H₂O₂ synergistically induce [Ca²⁺]_cyt elevations in cells defective in the Fet3/Ftr1 system

The tolerance to Cu²⁺ gained by the H₂O₂-sensitive yap1Δ cells, as well as the tolerance to H₂O₂ gained by the Cu²⁺-sensitive fet3Δ or ftr1Δ cells, suggested a synergism between Cu²⁺ and H₂O₂ in signaling the stress conditions.

We further investigated the cumulative effect of H₂O₂ and Cu²⁺ on inducing [Ca²⁺]_cyt elevations in cells defective in the Fet3/Ftr1 system. The fet3Δ cells accumulated more Cu²⁺ (Fig. 6a) than the wild-type cells, resulting in sudden and more transient elevations in [Ca²⁺]_cyt, with maxima reached after approximately 20 s from Cu²⁺ exposure; a more rapid restoration of the low levels of [Ca²⁺]_cyt was also noticed in fet3Δ cells (Fig. 6b, left). On the other hand, the ftr1Δ cells accumulated less Cu²⁺ than the fet3Δ cells within the first 10 min of Cu²⁺ exposure (Fig. 6a), causing a broadening of the [Ca²⁺]_cyt trace (Fig. 6c, left). It was difficult to assess that the role of the Cu²⁺-dependent [Ca²⁺]_cyt elevation was under high Cu²⁺ in fet3Δ and ftr1Δ mutants. Knocking out the YVC1 gene under fet3Δ and ftr1Δ background did not significantly alter the cell behavior under Cu²⁺ stress (data not shown); nevertheless, the fet3Δ and ftr1Δ cells were sensitive to the calcium chelator EGTA (Fig. 5c, right), indicating that these mutations required available Ca²⁺ for growth and probably for signaling of the stress conditions.

Fig. 6

Cu²⁺-induced mobilization of [Ca²⁺]_cyt in cells defective in the complex Fet3/Ftr1 is augmented by H₂O₂. a Cu²⁺ accumulation. Exponentially growing yeast cells were shifted to SD medium supplemented with 1 mM CuCl₂ and were incubated with shaking (200 rpm) at 28 °C for 5 or 10 min before cells were harvested for determination of Cu²⁺ accumulation, as described in “Materials and methods” section. Values are mean ± SD of three determinations made on two distinct experiments (n = 6). **p < 0.01, ***p < 0.001, significantly lower than the control strain. b, c The fet3Δ or ftr1Δ cells expressing coelenterazine-reconstituted aequorin were exposed to millimolar Cu²⁺, H₂O₂, or (Cu²⁺ + H₂O₂) directly in the luminometer tube as described in the “Materials and methods” section (final concentration, 1 mM for both chemicals. Traces showing Ca²⁺-dependent aequorin luminescence of b fet3Δ and c ftr1Δ. The arrow indicates the time when the chemicals were added. The luminescence traces represent the mean ± SD from independent experiments performed on three different days

The alleviation of copper toxicity by H₂O₂ in the fet3Δ and ftr1Δ mutants (Fig. 5b) may be result of Cu⁺¹ oxidation at the cell surface to the less toxic Cu²⁺ form. Nevertheless, besides oxidizing the cuprous ions, H₂O₂ also had the ability to induce [Ca²⁺]_cyt waves in fet3Δ (Fig. 6b, middle) and ftr1Δ (Fig. 6c, middle). Moreover, H₂O₂ and Cu²⁺ showed a synergistic effect on inducing stronger and sharper [Ca²⁺]_cyt pulses in both fet3Δ (Fig. 6b, right) and ftr1Δ (Fig. 6c, right), thus increasing the cell chances of survival.

Conclusions

In a previous study (Ruta et al. 2014), it was demonstrated that yeast cells use Ca²⁺ to signal the presence of high concentrations of Cd²⁺ in the environment. In the present study, we investigated the Ca²⁺-mediated response to high Cu²⁺ and we found that Cd²⁺ and Cu²⁺ interact with yeast cells differently, yielding different Ca²⁺-mediated responses. While Cd²⁺ induced sudden [Ca²⁺]_cyt elevations as a result of direct interaction with cell surface, the Cu²⁺ effect upon cell was more likely indirect. It was determined that contrary to Cd²⁺, the cell exposure to high Cu²⁺ determined broad and prolonged [Ca²⁺]_cyt elevations which lasted approximately 600 s in wild-type cells. Clearly, Cch1 played a role in Ca²⁺ entry under Cu²⁺ stress, but mainly to potentiate the release of Ca²⁺ from the vacuole via the Yvc1 channel. As in the case of other stress conditions (Pinontoan et al. 2002; Ene et al. 2015; Rigamonti et al. 2015),  Cch1 activation by Cu²⁺ was not significantly dependent on Mid1, as it had been reported for Cch1 activation by H₂O₂ (Popa et al. 2010). Since Cu²⁺ can generate reactive oxygen species (H₂O₂, for instance, by reacting with the superoxide anion radical), it is highly probable that the effect of Cu²⁺ on Ca²⁺ release is indirect, through generation of reactive oxygen species which trigger the Ca²⁺ release into the cytosol. In this line of evidence, other divalent cations (Co²⁺, Ni²⁺, Zn²⁺) failed to elicit Ca²⁺-dependent responses (Ruta et al. 2014). On the other hand, other redox-active metals such as Mn²⁺ or Fe³⁺ were also inactive in inducing [Ca²⁺]_cyt waves (Ruta et al. 2014; data not shown), probably because these metals are less redox reactive than the Cu²⁺/Cu⁺¹ couple is under aerobic conditions (Shi et al. 2003).

Generation of broad [Ca²⁺]_cyt elevations did not improve the tolerance to high Cu²⁺; the longer Ca²⁺ lingered in the cytosol, the more sensitive the cells became, such as was the case of mid1Δ cells (Fig. 2c, e). The elevations of [Ca²⁺]_cyt in response to high Cu²⁺ probably resulted in calcineurin activation, which in turn activated processes deleterious for cell survival under high Cu²⁺. This would explain why cells defective in calcineurin activity, such as the cnb1Δ cells, gain tolerance to high Cu²⁺. In this line of evidence, the absence of [Ca²⁺]_cyt elevation in yvc1Δ cells as response to high Cu²⁺ (Fig. 2d) keeps the calcineurin inactive and the yvc1Δ cells Cu²⁺-tolerant (Fig. 2e). It was shown that calcineurin activity is reduced by Nmt1-dependent myristoylation of the regulatory subunit Cnb1 in response to submaximal Ca²⁺ signals in order to prevent constitutive phosphatase activity (Connolly and Kingsbury 2012). This would explain why cch1Δ and mid1Δ mutants would have opposite tolerance to high Cu²⁺: in cch1Δ, the low-intensity Ca²⁺ signal may render calcineurin inactive and the cells slightly more tolerant to Cu²⁺. In this way, the regulation of Cu²⁺ tolerance appears as a complex and finely tunable process depending both on the broadness and on the intensity of the Ca²⁺ signal. In either direction, the adaptation to high Cu²⁺ was favored when the oxidative state of the cell shifted to pro-oxidant (for instance, by deleting YAP1 gene or by adding H₂O₂ to the growth medium). In this case, the cells gained tolerance to high Cu²⁺ not only by limiting the amount of available Cu¹⁺, but also by potentiating the cell ability to respond to high Cu²⁺ through rapid and sharp [Ca²⁺]_cyt pulses, leading the cells toward the adaptation alternative.