Introduction

Copper (Cu) is an essential element in all forms of life. There are two oxidation states of copper, Cu(I)/Cu+ (cuprous ion) and Cu(II)/Cu2+ (cupric ion). Cu+ prefers to bind to the thiol group in cysteine or the thioether group in methionine, while Cu2+ exhibits a high affinity for the secondary carboxyl group in aspartic/glutamic acid or the imidazole nitrogen group in histidine (Festa and Thiele 2011). Thus, copper ions readily form complexes with these amino acids or the peptides containing them. On the one hand, Cu acts as a cofactor of some enzymes due to its potential to either accept or donate an electron during the switch between Cu(I) and Cu(II). On the other hand, Cu can stabilize the conformations of proteins by binding to them (Festa and Thiele 2011). However, excess copper can be harmful to cells. Copper may generate reactive oxygen species (ROS), such as superoxide anions (O2−), nitric oxide (NO−), hydroxyl radicals (OH−), and hydrogen peroxide (H2O2), which can damage various molecules in cells. High levels of copper ions may also disrupt the normal conformations and functions of proteins by binding to them. Owing to the dual roles of this metal, all life forms evolved different mechanisms to maintain copper homeostasis, such as copper chelation, transport, and efflux (Festa and Thiele 2011; Xiao et al. 2004).

The primary understanding of copper metabolism comes from the study of baker’s yeast, Saccharomyces cerevisiae (Festa and Thiele 2011). Copper metabolic pathways that have been characterized in yeast share a high degree of conservation with those of mammalian systems (Zhou and Gitschier 1997). As a model organism, this microbe has advantages in multiple aspects: (a) it is the first eukaryote whose genome has been completely sequenced and well understood; (b) without long-stretch of non-coding DNA fragments, the yeast genome is relatively easy to manipulate; (c) this single-celled organism is the simplest eukaryote that shares many characteristics with multi-cellular creatures in terms of the cellular homeostasis of most transition metals; (d) it is convenient to manipulate yeast’s nutritional environment to study the biochemical pathways of the metabolism of copper and other metals. Due to these advantages, this eukaryote is widely employed to investigate the mechanism of copper metabolism and numerous breakthroughs have been made in this field (De Freitas et al. 2003). The key genes involved in copper metabolism have been identified (see Table 1). In this review, we will describe how copper is metabolized in Saccharomyces cerevisiae, including the processes of copper influx, utilization, detoxification and homeostasis (as shown in Fig. 1).

Table 1 Key genes involved in copper metabolism
Full size table
Fig. 1
figure 1

Schematic representation of copper metabolism in a yeast cell. Extracellular cupric ions (Cu2+) are reduced into cuprous ions (Cu+) by Fre1 or Fre2, prior to Ctr1-mediated incorporation. Cu2+ can be incorporated by Fet4. After incorporation by Ctr1, Cu+ ions bind to GSH in the cytosol. Then, GSH dispenses Cu+ to a series of chaperones, which deliver Cu+ to different utilization pathways: (1) Ccs1 escorts Cu+ to Sod1 in cytoplasm; (2) Atx1 passes Cu+ to Ccc2, which loads copper onto diverse secretory cuproproteins (CuPrs) at the Golgi body; (3) Pic2 or Mrs3 imports cuprous ions from cytosol into mitochondria, where they are loaded to CcO directly or via several chaperones including Cox17, Cox11, and Sco1/2. Excessive copper ions are sequestered by metallothioneins (Cup1 and Crs5) or separated in vacuoles. Vacuolar copper can be exported and mobilized by Ctr2. How Cu+ ions are transported into the vacuole and nucleus remains obscure. Genes involved in copper homeostasis are regulated by the copper sensing transcription factors Ace1 and Mac1: Cu+ promotes Ace1-mediated transactivation of copper scavenger genes (Cup1, Crs5, Sod1, etc.) and inhibits Mac1-mediated transactivation of genes responsible for copper uptake (Ctr1, Fre1, Fre2, etc.)

Full size image

Influx of copper ions

Most extracellular copper ions are in the form of Cu(II). These cupric ions (Cu2+) are reduced into cuprous ions (Cu+) by reductase Fre1 or Fre2 on the cell surface before entering the yeast cell (Dancis et al. 1990; Georgatsou and Alexandraki 1999; Yun et al. 2001). It was recently reported that Fre1 overexpression is sufficient to increase copper internalization and increase stress tolerance to H2O2 exposure (Berterame et al. 2018), showing a key role of this reductase in copper influx. Once being reduced, the univalent ions are transported by copper transporters (CTRs) located on plasma membranes. Under physiological conditions, transmembrane transportation of copper is mediated by two high-affinity CTRs, Ctr1 and Ctr3 (Dancis et al. 1994a, b; Knight et al. 1996; Luk et al. 2003; Pena et al. 2000; Yonkovich et al. 2002), which both form a homo-trimeric Cu+-selective ion-channel-like architecture to import the metal (Pena et al. 2000; Ren et al. 2019). However, in most laboratory-bred yeast strains, such as BY4742, the Ctr3 gene is disabled by the insertion of a Ty2 transposon (Knight et al. 1996). Thus, copper-dependent enzymes in yeast cell lines lacking the components of the Ctr1 transport system generally exhibit copper deficiency (Dancis et al. 1994a; Pena et al. 2000). Yeast strains without functional Ctr1 and Ctr3 genes demonstrate growth defects on non-fermentable media due to a lack of Cu+ for cytochrome c oxidase (CcO). Using this characteristic, Zhou et al. cloned the human Ctr1 gene, whose expression perfectly complements the growth defect of Ctr1∆ yeast and significantly increases the level of cellular copper (Zhou and Gitschier 1997). These mutant yeast strains defective in high-affinity copper transport have also been used to identify candidate copper importers in a wide range of organisms, including algae and land plants (Page et al. 2009; Senovilla et al. 2018). Besides Cu+, platinum is also imported by Ctr1 (Bodiga et al. 2018). It has been reported that reduced glutathione (GSH) is the first acceptor of copper after it enters into the cell, following which Cu+ can be delivered to the chaperones or metallothioneins (MTs) (Freedman et al. 1989).

In addition to the high-affinity copper transport systems, yeast has a “backup” mechanism for obtaining copper (Lee et al. 2002). When the environmental concentration of copper is high, low-affinity copper proteins on cell surfaces, such as iron transporter 4 (Fet4), can mediate the influx of cupric ions (Dix et al. 1994; Hassett et al. 2000; Li and Kaplan 1998). In general, these low-affinity transporters incorporate not only copper but also a variety of divalent metal ions. Saccharomyces cerevisiae Fet4, for example, can nonspecifically transport Cu2+, Fe2+ and other divalent metal ions into cells (Hassett et al. 2000; Portnoy et al. 2001).

Detoxification of copper ions

After incorporation, a large proportion of the copper ions go through a detoxification pathway to prevent the accumulation of toxic free copper ions in the cell. It has been reported that the concentration of free copper ions within the cytoplasm is as low as 10–18 M (about one molecule of free Cu+ per cell) (Rae et al. 1999). The excess cytosolic copper ions are sequestered by scavengers such as metallothioneins (MTs) (Hamer 1986; Krezel and Maret 2017; Niederwanger et al. 2017). MTs are a family of low molecular weight proteins rich in cysteine residues, which can bind excess ions of copper and other heavy metals in the cell (Fogel and Welch 1982; Winge et al. 1985). MTs are encoded by multiple genes, including CRS5, CUP1-1 and CUP1-2 (Culotta et al. 1994; Karin et al. 1984). Expression of these genes is responsible for mediating copper resistance in yeast (Butt et al. 1984; Katju et al. 2009). This is because these MTs bear multiple thiol groups organized as clusters and can store high amounts of nonexchangeable copper (Calderone et al. 2005; Jensen et al. 1996). GSH is another natural chelator of transition metals. It is composed of three amino acids—glutamate, cysteine, and glycine—among which cysteine can directly bind to copper via its thiol group. GSH is not encoded by a gene but is synthesized in two steps by γ-glutamylcysteine synthetase (GSH1) and GSH synthetase (GSH2) in the cell (Grant et al. 1996). GSH can also be assimilated from the extracellular environment by specific transporters (Bourbouloux et al. 2000; Miyake et al. 1998). Besides its function in dispensing Cu+ from Ctr1 to other cuproprotiens, GSH can also act as a copper reservoir by forming complexes with Cu(I) (Freedman et al. 1989). Due to its copper buffering function, GSH can lower copper sensitivity and increase yeast vitality (Zimdars et al. 2019). Cu(I) complexes of GSH have been found in organelles such as the nucleus (Carroll et al. 2004). These GSH-bound Cu+ can directly metalize MTs and SOD1 (Carroll et al. 2004; Ferreira et al. 1993). Surplus copper ions can also be separated in vacuoles by an unknown mechanism (Miner et al. 2019). When necessary, vacuolar copper can be mobilized by Ctr2, which is a low-affinity copper transporter located at the vacuolar membrane and provides copper for cytosolic metallochaperones from vacuoles (Rees et al. 2004).

Copper utilization pathway

Another portion of copper ions is targeted to different destinations by multiple copper chaperones for the following uses (Festa and Thiele 2011): (a) Ccs1 mediates the assembly of Cu+ to Sod1 in cytosome; (b) Atx1 targets Cu+ to the Golgi body for various cuproproteins; (c) Cox17 passes Cu+ to mitochondrial cytochrome c oxidase (CcO) (Luk et al. 2003); and (d) Cu+ can also be delivered to the nucleus via an unknown mechanism.

Cytosolic pathway

Superoxide dismutase 1 (Sod1 or Cu/Zn SOD) is the most important copper protein in cytoplasm (Slekar et al. 1996). Each molecule of Sod1 also combines with a zinc atom, but the way that it obtains the zinc ion remains unknown. In contrast, it is well known that copper is targeted to Sod1 by its specific copper chaperone Ccs1 (copper chaperon for Sod1, also named Lys7) (Culotta et al. 1997). Ccs1 can bind to the cell membrane and directly interact with Ctr1 to obtain Cu+ (Pope et al. 2013). Then, Ccs1 can associate with Sod1 to deliver Cu+ to the latter molecule (Lamb et al. 2000). On the other hand, it has been reported that a small fraction of Sod1 molecules can acquire free Cu+ directly from the cytoplasm to maintain Sod1 activity in the absence of Ccs1 (Rae et al. 1999). Our previous data are consistent with this concept since we found that supplementation with copper endows yeast that is lacking Ccs1 with Sod1 activity (Li et al. 2010). This Ccs1-independent activation of Sod1 likely involves GSH, as it does in mammals (Carroll et al. 2004). Recently, Winkler’s group discovered a multifunctional chaperoning role for Ccs1 during Sod1 activation: Ccs1 delivers Cu+ to an entry site at the Sod1·Ccs1 interface upon binding, and then to the Sod1 active site by a thermodynamically-driven affinity gradient. The Sod1·Ccs1 interaction also promotes high-affinity Zn2+ binding in Sod1; as a result, Sod1 is fully activated (Boyd et al. 2019; Fetherolf et al. 2017). It has been reported that a fraction of active Sod1, along with Ccs1, localizes at the mitochondrial intermembrane space (IMS) and nucleus, where it protects yeast cells against oxidative damage (Sturtz et al. 2001), mediates copper-sensing activation of Mac1 (Wood and Thiele 2009), and functions as a transcription factor to regulate the expression of oxidative stress-responsive genes (Tsang et al. 2014).

Golgi pathway

All copper-dependent enzymes from the secretory pathway are loaded with copper ions on the Golgi apparatus and thus activated. The copper-transporting P-type ATPases translocate cuprous ions from cytoplasm to the lumen of the Golgi body, making them available for enzyme assembly (Yuan et al. 1995). In yeast cells, this P-type ATPase is called Ccc2, which cannot directly obtain free cuprous ions from cytoplasm, but accepts them through a copper chaperone, Atx1 (antioxidant 1) (Banci et al. 2001; Huffman and O'Halloran 2000). Atx1 was originally identified as a multi-copy suppressor of oxidative damage in yeast lacking SOD (Lin et al. 1997). The crystal structures demonstrate that Atx1 proteins dimerize to bind Cu+ (Lee et al. 2017). Though it was conventionally assumed that Ctr1 passes Cu+ to Atx1 via direct interaction between both proteins (Xiao et al. 2004), it has been reported that in the presence of copper, GSH induces dimerization of Atx1 and formation of a complex containing two Atx1, two Cu(I), and two GSH (Miras et al. 2008). This indicates that Cu+ incorporated by Ctr1 associates with GSH, which further delivers the metal ion to chaperon Atx1. In Golgi compartments, the secretory cuproproteins are loaded with Cu+ before they are targeted to specific organelles or secreted out of the cell (Fu et al. 1995; Lin et al. 1997; Pufahl et al. 1997; Yuan et al. 1995). The multicopper oxidase Fet3, which is required for high-affinity iron uptake, is metalized via this route (Askwith et al. 1994; Yuan et al. 1995). Thus, this Golgi pathway of copper delivery is required for iron uptake (Lin et al. 1997).

Mitochondrial pathway

Cytochrome c oxidase (CcO) is the terminal oxidase in the mitochondrial respiratory chain, catalyzing the transfer of electrons to oxygen to form H2O (Kloeckener-Gruissem et al. 1987). According to our previous study as well as data from other groups, copper deficiency leads to an impairment of CcO enzymatic activity (Li et al. 2010). Two copper binding sites were housed in the core subunits of CcO: Cox1 and Cox2 (Marechal et al. 2012). Incorporation of Cu+ into both sites requires multiple chaperones. Sco1 or Sco2 locates on the inner membrane of mitochondria and passes copper atoms to the copper A (CuA) site of Cox2 (Beers et al. 1997; Glerum et al. 1996a, b). It was traditionally assumed that Cox17 shuttles between cytoplasm and mitochondrial lumen, delivering cytosolic cuprous ions to Sco1 or Sco2 (Balatri et al. 2003; Lode et al. 2002; Nittis et al. 2001; Rentzsch et al. 1999; Schulze and Rodel 1988, 1989). However, recent reports suggest that Cox17 is imported unfolded into mitochondrial IMS and receives Cu from the mitochondrial lumen, not from the cytosol (Banci et al. 2009; Cobine et al. 2006). Cox17 also delivers copper to Cox11, which acts as a copper donor to the copper B (CuB) site of Cox1 (Hensgens et al. 1984; Hiser et al. 2000; Horng et al. 2004; Tzagoloff et al. 1990). In yeast, Cox17 is of great importance for normal growth and metabolism, since the deletion of Cox17 leads to defects in the respiratory chain. However, it seems that copper can partially bypass Cox17 and be loaded to CcO, given that supplementation with copper in the culture medium can make Cox17 mutant cells viable (Giaever et al. 2002).

The finding of a labile copper pool within the mitochondrial matrix may explain the Cox17-independent metallation of CcO (Cobine et al. 2004, 2006; Dodani et al. 2011; Vest et al. 2016). These IMS copper ions can be used to metalize CcO and mitochondrial SOD1 (Cobine et al. 2006). Cobine’s group also found that the mitochondrial carrier proteins Pic2 and Mrs3 are implicated in copper importation into mitochondria: Pic2- or Mrs3-deficient yeast strains exhibit defects in mitochondrial copper uptake and copper-dependent growth phenotypes owing to impaired CcO activity. Deletion of both genes (pic2Δmrs3Δ) leads to a more severe respiratory growth defect (Vest et al. 2013, 2016). However, it is still unclear where Pic2 and Mrs3 get copper from, since there is virtually no free copper in the cell, and it has been demonstrated that copper is associated almost exclusively with chaperones or the copper-binding proteins/peptides (Rae et al. 1999). In addition, the identity of the matrix copper ligand is unknown, since GSH was eliminated by Cobine et al. Organic acids such as citrate and oxaloacetate may act as ligands of these Cu+ ions (Cobine et al. 2004).

Nuclear pathway

As discussed below and shown in Fig. 1, fluctuation of the intracellular Cu+ concentration regulates copper-sensing transcription factors Ace1 and Mac1. It has been reported that metallation of Ace1 and Mac1 occurs independently within the nucleus (Keller et al. 2005). This indicates the presence of a labile nuclear Cu+ pool (Carroll et al. 2004). However, it is unclear how copper ions are transported into the nucleus. On the other hand, over-expression of the Crs5 metallothionein in this subcellular compartment does not compromise Mac1 Cu-responsive regulation (Keller et al. 2005), which is indicative of the inability of this labile nuclear Cu+ pool to promote Mac1 metallation. Wood et al. reported that Sod1 and Ccs1 partially localize to the yeast nucleus, and both molecules are required for the activation of Mac1 (Wood and Thiele 2009). This indicates that copper chaperone Ccs1 may deliver copper into the nucleus and activate Sod1, whose activity is involved in copper deficiency-induced Mac1 DNA binding. Furthermore, they found that C. elegans Sod1, which is copper-metalized and activated independently of Ccs1, can also rescue Mac1 activation and metallation in a Ccs1-independent manner in Saccharomyces cerevisiae (Wood and Thiele 2009). Currently, the ways that Mac1 and Ace1 are metalized remain unclear. In addition, the nature or mechanism of Cu transport into the nucleus remains obscure, in contrast to the existing body of knowledge on the routing of Cu by cytosolic chaperones and transporters between the secretory compartment, vacuole and mitochondria.

Copper homeostasis in yeast

Due to the dual effects of copper ions in cells, it is crucial to maintain a relatively constant concentration of cellular copper. On the one hand, there should be enough copper ions in the cell to execute normal physiological metabolism; on the other hand, over-intake of copper ions should be prevented to avoid toxicity. Organisms from yeast to humans maintain the cellular homeostasis of copper mainly by regulating the gene expressions of the transport and detoxification systems (Winge et al. 1998).

Genes involved in copper homeostasis in yeast are regulated mostly through two transcription factors, Ace1 and Mac1, which are positively and negatively regulated by copper ions, respectively. When a cell is abundant in copper, Ace1 (also called Cup2) is loaded with cuprous ions and subsequently activated (Buchman et al. 1989; Thiele 1988). As a consequence, proteins responsible for copper detoxification, such as metallothioneins (encoded by CUP1-1, CUP1-2 and CRS5) and Sod1 are up-regulated to neutralize the excess metal ions and remove harmful free radicals (Mehta et al. 2018; Rae et al. 1999). Mac1 contains an amino-terminal DNA binding domain and a carboxyl-terminal activation domain and is localized to the nucleus (Serpe et al. 1999). A high concentration of copper ions can inactivate Mac1, whose activity is essential to the expression of copper transporter genes including Ctr1, Ctr3, Fre1, Fre2, and Fre7. Without Cu+ binding to Mac1, this transcription factor occupies copper responsive elements (CuREs) in the promoters of its target genes and promotes their expression, facilitating the uptake of copper ions (Labbe et al. 1997; Martins et al. 1998). As mentioned above, Sod1 activity is essential to Mac1-mediated copper sensing: Mac1 is transcriptionally inactive in mutants that lack Sod1 or its copper chaperone Ccs1 (Wood and Thiele 2009). In response to mild DNA damage, Mac1 is activated via reduction of cysteine residues in its trans-activation domain, while severe DNA damage induces reversible oxidation of these residues and a consequent repression of Mac1 trans-activity. Both Sod1 and checkpoint kinase Rad53 are required for an unknown aspect of Mac1-mediated Ctr1 expression in response to mild DNA damage. Increased Ctr1 proteins can enhance copper influx, which can ensure Sod1 activity and Rad53 signaling in response to DNA damage in yeast (Dong et al. 2013). Recently, Alexandraki’s group found that another checkpoint protein, Rad9, directly binds with Mac1 under non-DNA-damaging conditions, suppressing its DNA binding and trans-activation functions. On the other hand, Rad53 also localizes to the Ctr1 coding region in a Rad9-dependent manner (Gkouskou et al. 2020). These data suggest a connection between copper-responsive transcription and the DNA repair pathway. The underlying mechanism remains obscure.

Mobilization of stored cuprous ions is another aspect of copper homeostasis (Rees et al. 2004; Rees and Thiele 2007). According to our investigation, the metal-sensing transcription factors Mac1 and Aft1 cooperatively activate the expression of vacuolar copper transporter Ctr2 upon deficiency of cytosolic Cu+, making yeast resistant to copper starvation (Liu et al. 2012; Qi et al. 2012).

Regulation of the key factors in copper homeostasis can also occur at a post-translational level. It has been reported that Ctr1 proteins are stable under a low level of intracellular copper and unstable at a high concentration of copper ions (Ooi et al. 1996). The copper sensing function of Mac1 is required in the degradation of Ctr1 at an elevated copper level (Yonkovich et al. 2002). Recently, Bodiga’s group found that the protein level of Ctr1 may be positively regulated by mitochondrial cuprochaperone Sco1. In Sco1-deficient yeast strains, copper restriction failed to increase the Ctr1 protein level, though Ctr1 mRNA was upregulated (Bodiga et al. 2018). This indicates that mitochondrial utilization of copper, which is mediated by Sco1, may up-regulate Ctr1 at the protein level via an as-yet-unknown mechanism. More importantly, this Sco1-dependent Ctr1 expression and consequential copper influx have also been found in humans (Hlynialuk et al. 2015; Leary et al. 2007). On the other hand, the stability of Mac1 itself is also regulated by the concentration of copper ions; when the intracellular copper concentration goes up, Mac1 protein is prone to degradation (Zhu et al. 1998).

Taken together, yeast cells evolve into a delicate system to maintain copper homeostasis (Fig. 1). In the presence of a high level of copper ions, genes involved in copper detoxification are switched on, while proteins responsible for copper incorporation are down-regulated at either the transcriptional or post-translational level. (Jungmann et al. 1993). On the contrary, upon deficiency of this metal, copper sensing transcription factors will down-regulate genes for copper detoxification and enhance the expression of proteins involved in copper intake or mobilization (Georgatsou et al. 1997; Hassett and Kosman 1995; Labbe et al. 1997; Yamaguchi-Iwai et al. 1997).

Conclusions

In summary, Saccharomyces cerevisiae has evolved delicate mechanisms for reaching copper homeostasis. Membrane proteins Fre1 and Fre2 mediate the reduction of extracellular Cu2+ into Cu+, which can be incorporated by the high-affinity transporters Ctr1 or Ctr3. Cu2+ can be directly transported by Fet4 in a low-affinity manner. Recent studies suggest that glutathione (GSH) is the first acceptor of Cu+ after it enters into the cell. Due to the sequestration effect of GSH, MTs, and vacuoles, there are almost no free copper ions present in the cell. Cu+ incorporated by Ctr1/3 can be targeted to different pathways. In cytosol, superoxide dismutase Sod1 is metalized and activated by Cu+ via its chaperone Ccs1. This process has recently been elucidated by studies revealing that Sod1 and Ccs1 also reside in mitochondria and nuclei to scavenge free radicals and that both proteins are involved in Mac1-dependent transcription. Atx1 is a chaperone targeting Cu+ into the Golgi body. GSH induces dimerization of Atx1 and passes Cu+ to it via direct interaction. Then, P-type ATPase Ccc2 obtains Cu+ from Atx1 and pumps Cu+ into the Golgi body, where various cuproproteins, including Fet3, are loaded with Cu+. Copper chaperone Cox17 shuttles between cytoplasm and mitochondrial lumen, delivering Cu+ to Sco1/2 or Cox11 on the inner mitochondrial membrane. Cox11 and Sco1/2 present Cu+ to different subunits of cytochrome c oxidase (CcO) to activate this essential enzyme in the respiratory chain. It was recently reported that Pic2 and Mrs3 mediate copper importation into mitochondria to generate a labile copper pool within the mitochondrial matrix, which may account for Cox17-independent metallation of CcO. However, the details remain obscure. It is unclear how copper ions are imported into the nucleus and how they are sensed by the copper-responsive transcription factors Mac1 and Ace1. Ace1 can mediate the transcription of MT-encoding genes and Sod1 in response to increased cellular Cu+. Copper deficiency induces Mac1-dependent transcription of copper transport components including Fre1/2 and Ctr1/3. A high concentration of Cu+ inactivates Mac1 or induces degradation of Mac1 and Ctr1. Mac1 is also involved in the transactivation of Ctr2 to mobilize vacuolar copper upon copper deficiency. On the other hand, several new pieces of data reveal that Mac1-dependent transcription is not only involved in copper uptake but also DNA damage and checkpoints. In future, these unresolved issues should be explored to thoroughly unravel the mechanism of copper metabolism in yeast. This will undoubtedly be helpful in elucidating the corresponding mechanism in mammals and unveiling the mechanisms of copper-related human disease.