Retrograde regulation of multidrug resistance in Saccharomyces cerevisiae

Gene 354 (2005) 15 – 21 www.elsevier.com/locate/gene

Review

Retrograde regulation of multidrug resistance in Saccharomyces cerevisiae W.S. Moye-Rowley* Department of Physiology and Biophysics, 6-530 Bowen Science Building, 51 Newton Road, University of Iowa, Iowa City, 52242, United States Received 21 December 2004; received in revised form 14 February 2005; accepted 23 March 2005 Available online 17 May 2005 Received by R. Butow

Abstract Communication between the mitochondria and the nucleus is essential to ensure correct metabolic coordination of the cell. Signaling pathways leading from the mitochondria to the nucleus are referred to as retrograde signaling and were first discovered in the yeast Saccharomyces cerevisiae. Cells that lack their mitochondrial genome (U0 cells) trigger expression of the nuclear CIT2 gene in order to ensure adequate amino acid biosynthesis. More recently, it has been found that a different set of genes involved in multidrug resistance in S. cerevisiae is strongly induced in U0 cells. During a search for negative regulators of the ATP-binding cassette (ABC) transporter-encoding gene PDR5, it was observed that U0 mutants exhibited dramatic up-regulation of the transcript of this gene. This induction was due to the post-translational activation of a direct upstream regulator of PDR5 that was designated Pdr3p. Loss of the LGE1 gene led to a block in U0mediated induction of PDR5 expression. Lge1p has been observed by others to be involved in histone H2B ubiquitination along with the ubiquitin-conjugating enzyme Rad6p and the ubiquitin ligase Bre1p. Our studies provide evidence that Lge1p has another function unique from H2B ubiquitination that is required for retrograde regulation of PDR5 transcription. We have also found that the Pdr pathway regulates expression of several genes involved in sphingolipid biosynthesis. These findings suggest that the physiological role of the PDR genes might be to regulate membrane homeostasis and U0-triggered changes in this parameter may be the signal controlling PDR gene expression. D 2005 Published by Elsevier B.V. Keywords: Mitochondria; Transcription; PDR3; PDR5; ATP-binding cassette transporter

1. Introduction A defining characteristic of eukaryotic cells is the presence of multiple genomes within a common cytoplasm. These organelle-associated genomes include the DNA within the mitochondria and the chloroplasts. While it is believed that these organellar genomes have evolved from free-living organisms, their DNA content has been adjusted over time to encode only a small number of the total proteins that will ultimately make up the functional organelle. While several hundred different proteins are

Abbreviations: PDRE, Pdr1p/Pdr3p response element; ABC, ATPbinding cassette; Pdr, Pleiotropic drug resistance; IPC, Inositol-phosphoceramide; MIPC, Mannose – inositol-phosphoceramide; M(IP)2C, Mannose – (inositol-P)2-phosphoceramide; LCB, Long chain base; LCBP, Long chain base phosphate; DHS, Dihydrosphingosine; PHS, Phytosphingosine. * Tel.: +1 319 335 7874; fax: +1 319 335 7330. E-mail address: [email protected]. 0378-1119/$ - see front matter D 2005 Published by Elsevier B.V. doi:10.1016/j.gene.2005.03.019

required to synthesize a fully operational mitochondrion, only 13 of these are encoded by the mitochondrial genome (see (Reichert and Neupert, 2004) for a recent review). Coordinating the expression of the nuclear and mitochondrial genomes is critical for proper metabolism and physiology. The regulatory connection from the mitochondria to the nucleus in cells is referred to as retrograde regulation (reviewed in (Butow and Avadhani, 2004)). Retrograde regulation was first genetically defined in the yeast Saccharomyces cerevisiae in studies of nuclear gene expression of cells that lack their mitochondrial DNA (U0 cells) (Parikh et al., 1987). Expression of the nuclear CIT2 gene was found to be strongly induced at the transcriptional level in order to maintain levels of the tricarboxylic acid cycle intermediate a-ketoglutarate, the crucial precursor for the amino acid glutamate (Small et al., 1995). Extensive genetic analyses of the retrograde regulation of CIT2 identified two basic region-helix/loop-helix transcription

16

W.S. Moye-Rowley / Gene 354 (2005) 15 – 21

factors designated Rtg1p and Rtg3p as the key transcriptional activators of CIT2 expression in response to loss of the mitochondrial genome (Liao and Butow, 1993; Jia et al., 1997). More recent experiments have focused on elaborating the signaling pathway connecting the mitochondria with activation of Rtg1p/Rtg3p function and this impressive progress was recently reviewed (Butow and Avadhani, 2004). Since many different metabolic processes interface with the mitochondria, U0 cells require transcriptional reprogramming beyond activation of CIT2 expression in order to remain viable. Several years ago (Hallstrom and MoyeRowley, 2000), our laboratory found that U0 cells strongly induced expression of multidrug resistance genes along with the previously described activation of CIT2 transcription. Recent progress in the understanding of this new retrograde regulatory pathway will be discussed here.

2. Multidrug resistance in S. cerevisiae Early genetic studies on drug resistant mutants led to the unexpected finding that a large number of mutant alleles in a locus designated PDR1 led to the simultaneous acquisition of resistance to a number of different, functionally unrelated drugs (Rank and Bech-Hansen, 1973). These different drugs included a number of mitochondrial poisons as well as the cytoplasmic translation elongation inhibitor cycloheximide. Genetic analyses of these drug hyper-resistant PDR1 alleles indicated that the mutant forms of this gene were likely gain-of-function hypermorphs as elevated drug resistance could be seen even in the presence of a wild-type copy of the gene (Rank and Bech-Hansen, 1973). Cloning and characterization of the PDR1 locus demonstrated that this gene encoded a Zn2Cys6 zinc cluster-containing transcriptional regulatory protein (Balzi et al., 1987). A combination of genetic analyses and DNA sequencing led to the discovery that a protein sharing a high degree of sequence similarity with Pdr1p was present near the centromere of chromosome II and that similar drug hyper-resistant forms of this protein could be selected (Subik et al., 1986; Delaveau et al., 1992). This homologue was designated Pdr3p. The knowledge that both PDR1 and PDR3 encoded transcription factors argued that neither of these genes was directly responsible for the observed effects on drug resistance and that target genes under the influence of these proteins were likely the effectors of multidrug resistance. A screen of a high-copy-number plasmid library for genes that conferred multidrug resistance when present in elevated copy number identified the PDR5 locus (Leppert et al., 1990). Molecular characterization of PDR5 determined that this gene encodes an ATP-binding cassette (ABC) transporter protein that was transcriptionally induced in cells carrying hyperactive gain-of-function alleles of PDR1 (Meyers et al., 1992) and PDR3 (Nourani et al., 1997).

Mutational analyses and DNA binding experiments established that both Pdr1p and Pdr3p recognized three elements in the PDR5 promoter that have been designated Pdr1p/ Pdr3p response elements (PDREs) (Katzmann et al., 1994, 1996). These PDREs are typically found in the promoters of all genes known to be regulated by Pdr1p and/or Pdr3p.

3. Retrograde control of PDR5 expression Analyses of the gain-of-function forms of Pdr1p and Pdr3p led to the finding that these mutant alleles represented single amino acid substitution mutations clustered in discrete regions of each factor (Carvajal et al., 1997; Nourani et al., 1997; Simonics et al., 2000). Coupled with the known dominant or semi-dominant behavior of these mutants, these findings supported the hypothesis that these factors might normally be maintained in a state of repressed activity that is relieved by these mutant alleles. To test this hypothesis, we carried out a search for negatively acting regulatory genes using a transposon mutagenesis strategy (Burns et al., 1994). Transposons were inserted at random into the genome of a wild-type S. cerevisiae strain containing an integrated PDR5-lacZ reporter gene. Cycloheximide hyper-resistant colonies were selected and clones that also exhibited high level expression of PDR5-lacZ were retained. Genetic analysis and cloning of the genes inactivated by transposon insertion led to the finding that disruption of the FZO1 or OXA1 genes elicited overproduction of PDR5 (Hallstrom and Moye-Rowley, 2000). Fzo1p is required for normal mitochondrial fusion and loss of this protein causes cells to become U0 (Hermann et al., 1998; Rapaport et al., 1998). Oxa1p is an important assembly factor that is required for faithful production of the mitochondrial inner membrane complexes present in the cytochrome c oxidase and Fo ATPase subcomplex (Altamura et al., 1996). Mutants lacking OXA1 are nuclear petites but segregate U0 cells at a high frequency (unpublished data). Further analyses demonstrated generation of U0 cells by ethidium bromide treatment of a wild-type strain was sufficient to lead to both a robust multidrug resistant phenotype and PDR5 overexpression. To examine the molecular basis for this observed induction of PDR5 in U0 cells, genetic studies were performed in which either the PDR1 or PDR3 gene was removed individually from U0 cells. Loss of PDR3 was sufficient to block the increased PDR5 transcription that would normally occur in U0 cells while PDR1 could be eliminated without decreasing PDR5 induction. Importantly, loss of RTG1 only slightly diminished PDR5 activation in U0 cells, demonstrating that the retrograde pathway connecting multidrug resistance to the mitochondria is distinct from that linking CIT2 expression to mitochondrial status. A feature found in the PDR3 gene but not PDR1 is the presence of an autoregulatory loop controlling PDR3 transcription (Delahodde et al., 1995). Two PDREs are

W.S. Moye-Rowley / Gene 354 (2005) 15 – 21

present in the PDR3 promoter and are required for this gene to be induced in U0 cells. Replacement of the PDR3 promoter with the PDR1 regulatory region still supported higher levels of cycloheximide resistance in U0 cells than in wild-type, consistent with the idea that a post-translational input to Pdr3p is still active even when the PDR3 autoregulatory loop is not functioning ((Hallstrom and Moye-Rowley, 2000), our unpublished data). Together, these data indicate that a signal emerges from U0 mitochondria, post-translationally activates Pdr3p which in turn induces expression from its own promoter as well as from target genes like PDR5 (Fig. 1). Assay of a battery of different mutations in nuclear proteins required for normal mitochondrial function demonstrated that only mutants lacking normal Fo ATPase subcomplex function triggered PDR5 induction while other petite mutants, including F1 ATPase subcomplex defective strains, did not (Zhang and Moye-Rowley, 2001). Given that U0 cells activated PDR3 transcription, we carried out a genetic screen searching for transposongenerated mutations that blocked the increase in cycloheximide resistance that would normally occur in these cells. From this screen, we recovered transposon insertion

17

mutations in the LGE1 gene that strongly reduced cycloheximide tolerance of the resulting integrants (Zhang et al., 2005). These same insertion mutants also failed to normally induce either PDR3 or PDR5 expression and provide evidence that Lge1p is required for normal retrograde signaling to PDR3. The LGE1 gene was first described as a locus that influences control of cell size in S. cerevisiae (Jorgensen et al., 2002). Mutants lacking Lge1p exhibited a delay in Start progression and had a larger cell size than isogenic wildtype cells. Later work demonstrated that Lge1p was required, along with the E2 ubiquitin-conjugating enzyme called Rad6p and an E3 ubiquitin ligase designated Bre1p, to attach ubiquitin to lysine 123 in histone H2B (Hwang et al., 2003). This modification is required to recruit the Set1p histone methylase which then methylates lysine 4 of histone H3 (Briggs et al., 2001). Loss of Lge1p, Bre1p or Rad6p completely blocks ubiquitination of H2B lysine 123 but histone H3 lysine 4 methylation is only partially reduced in lge1D cells while no methylation can be seen in either bre1D or rad6D strains. The differential requirement for Bre1p/Rad6p and Lge1p in terms of effect on histone H3 methylation suggests that

Drugs

Pdr5p

ρ0 Mitochondria

PDR5

Pdr3p Lge1p

PDR3

Fig. 1. Retrograde control of PDR gene expression. Interactions and functions of several genes and proteins that are involved in the response of the Pdr pathway to loss of the mitochondrial genome are diagrammed. An unknown signal from U0 mitochondria regulates Pdr3p at a post-translational level. This activated Pdr3p binds to the two Pdr1p/Pdr3p response elements (PDREs: small black boxes) in the PDR3 promoter leading to an increase in Pdr3p levels. Lge1p acts at the level of the PDR3 promoter to allow efficient gene activation. Pdr3p also induces transcription of PDR5 leading to elevated Pdr5p in the plasma membrane. Pdr5p acts as an ATP-dependent drug efflux pump to stimulate drug export from cells.

18

W.S. Moye-Rowley / Gene 354 (2005) 15 – 21

the phenotypes of these mutants will not be identical. This was found to be the case as only mutants lacking LGE1 exhibited a defect in U0-mediated induction of PDR5 transcription (Zhang et al., 2005). The non-equivalent effect of lge1D, bre1D or rad6D mutants on PDR5 function was also evaluated by cycloheximide resistance, Northern blotting and PDR5-lacZ reporter assays in both U+ and U0 genetic backgrounds. Only loss of LGE1 from U0 cells led to a defect in all these measurements of PDR5 function while bre1D or rad6D mutants in a background exhibited only minor effects on PDR5-dependent processes. The global transcriptional effects of lge1D, bre1D or rad6D mutants were also compared in U+ and U0 genetic backgrounds by DNA microarray experiments (Zhang et al., 2005). As expected, based on the selectivity of the defect in PDR5 expression described above, only lge1D U0 cells exhibited depression in Pdr3p-dependent U0 induction of gene expression. However, in U+ cells, the situation was much more complex. Loss of Rad6p had the most pronounced effect on gene expression with nearly 250 genes exhibiting detectable alterations in expression with most of these genes being induced upon loss of the RAD6 gene. Bre1p and Lge1p also exerted their effects mainly at the level of repression, consistent with a role for these proteins along with Rad6p, in gene silencing (Huang et al., 1997). Rad6p and Lge1p shared more positive target genes than either did with Bre1p. This suggests that Rad6p and Lge1p functions in terms of activation of gene expression are more closely related than to Bre1p. Together, these data suggest that Lge1p is required for normal transcriptional induction of the PDR3 promoter in U0 cells. The effect of Lge1p can be localized to the PDR3 promoter as a PDR3-lacZ fusion gene containing only 600 bp of PDR3 promoter sequence still required LGE1 function for normal induction in U0 cells (Zhang et al., 2005). The precise function of Lge1p is still unknown as far as its role in positive regulation of gene expression but using a heterologously regulated form of PDR3 (PDR3 gene placed under control of a copper-inducible transcription factor (Ace1p (Zhou and Thiele, 1993)), the requirement for Lge1p function in U0 cells could be bypassed. Interestingly, this was not true in U+ cells as copper induction of PDR3 expression in LGE1 cells drove nearly 300% more PDR5 expression than in lge1D cells. This suggests that Lge1p may act to permit PDR3 to assume an active conformation in U+ cells, allowing its effective induction in response to signals from U0 mitochondria. Further work is required to validate this hypothesis.

4. Sphingolipid biosynthesis interfaces with the Pdr pathway While the defining feature of the genes in the Pdr pathway has always been their effect on drug resistance, the true physiological role of this pathway remains in

question. Cycloheximide, a drug that is often used to assess the activity of the Pdr pathway is a natural product and seems unlikely to represent the relevant substrate of the PDR genes, although this possibility remains. Recent experiments have linked Pdr function with the sphingolipid biosynthetic pathway in S. cerevisiae and suggest that the role of the Pdr pathway may be to coordinate the synthesis and transport of these important lipid constituents of the plasma membrane. Three major sphingolipid species are produced in S. cerevisiae: IPC (inositol-phosphoceramide), MIPC (mannose – inositol-phosphoceramide) and M(IP)2C (mannose – (inositol-P)2-ceramide) with the last form representing 70% of the total sphingolipids in this yeast (reviewed in (Dickson and Lester, 2002)). Unless stated otherwise, this discussion will focus on sphingolipids and their metabolites in S. cerevisiae only. These glycosylated lipids are produced in a multistep biosynthetic pathway that is nearly identical to that of mammalian cells. Briefly, condensation of serine and palmitoyl-CoA by the enzyme serine-palmitoyl transferase initiates the de novo synthesis of sphingolipids. Early intermediate in the sphingolipid biosynthetic pathway are referred to as long chain bases (LCBs) which are designated dihydrosphingosine (DHS) or phytosphingosine (PHS) in S. cerevisiae. DHS and/or PHS may be phosphorylated to the corresponding long chain base phosphate (LCBP) that can be degraded by the enzyme Dpl1p (Saba et al., 1997). DHS and PHS are also substrates for the enzyme ceramide synthase that produces the ceramide for use in production of the more complex sphingolipids. Both forms of LCBs and ceramides are thought to act as signaling molecules in regulation of cell proliferation (See (Mathias et al., 1998; Cuvillier, 2002) for reviews). These lipids have both a structural role in the cell and act as signaling intermediates, making the control of their biogenesis critical in ensuring normal physiology. The initial hint that a connection existed between the Pdr pathway and sphingolipid biosynthesis came from a DNA microarray analysis of genes that were transcriptionally induced in cells containing hyperactive alleles of Pdr1p or Pdr3p (DeRisi et al., 2000). The IPT1 gene, encoding the last step in sphingolipid biosynthesis (Dickson et al., 1997), exhibited increased mRNA levels in the presence of this elevated Pdr pathway activity. This finding, coupled with the observation that a PDRE was located in the IPT1 promoter region, led to a closer examination of the interaction of the Pdr pathway with control of IPT1 transcription (Hallstrom et al., 2001). Northern blotting and reporter gene assays confirmed that IPT1 was induced by increases in the activity of either Pdr1p or Pdr3p. The single PDRE in the IPT1 promoter was required for this transcriptional regulation and was necessary for induction in U0 cells. The effects of changes in Pdr pathway activity were also assessed for the influence on sphingolipid pathway syn-

W.S. Moye-Rowley / Gene 354 (2005) 15 – 21

thesis. M(IP)2C levels were found to increase in U0 cells in a Pdr3p-dependent fashion, consistent with elevated Pdr3p activity leading to an increase in Ipt1p enzyme levels. Mutant strains lacking IPT1 had complex effects on drug resistance. Cycloheximide resistance was found to increase in ipt1D cells while oligomycin tolerance decreased. These findings are not consistent with a simple model in which defects in plasma membrane sphingolipids lead to a common effect on membrane transporters but rather suggest that the response of membrane proteins to the altered lipid environment depends on the protein. The clear Pdr-dependent influence on IPT1 expression led to a more extensive inspection of other genes in the sphingolipid biosynthetic pathway that might also be Pdrregulated. Three candidates were found: LCB2 (encodes a subunit of serine palmitoyl transferase (Nagiec et al., 1994)), SUR2 (hydroxylase (Haak et al., 1997)) and LAC1 (encodes one of a pair of homologous proteins required for ceramide synthase activity (Barz and Walter, 1999)). LAC1 was found to exhibit the largest response to increased Pdr pathway activity and its promoter was analyzed in some detail (Kolaczkowski et al., 2004). While LAC1 was also induced in U0 cells, detection of a phenotypic consequence of this regulation was problematic. One reason for this difficulty is the presence of the homologous LAG1 gene that may be adjusted either at the level of the protein or the mRNA for changes in LAC1 expression. The other complication is the existence of a highly regulated buffering system of enzymes and transporters that act to ensure correct regulation of LCBs and ceramide. This system includes the enzyme Dpl1p (Saba et al., 1997), ceramidases (Mao et al., 2000a,b) and integral membrane proteins that can be used to transport LCBs out of the cell (Kihara and Igarashi, 2002). RSB1 encodes one of these LCB efflux proteins and is highly regulated by the Pdr pathway as shown by several different microarray experiments (DeRisi et al., 2000; Devaux et al., 2002). These interactions are summarized in Fig. 2. RSB1 was originally identified as a locus that conferred resistance to PHS when present in dpl1D cells on a highcopy-number plasmid (Kihara and Igarashi, 2002). Biochemical analyses demonstrated that overproduction of this protein led to an increase in LCB efflux from cells. More recent experiments have provided the striking finding that pdr5D cells induce RSB1 gene expression in a strictly Pdr1p-dependent fashion (Kihara and Igarashi, 2004). These same strains that lack Pdr5p exhibit increased PHS resistance and increased LCB efflux as would be expected from elevated Rsb1p levels. Pdr5p has been shown to be involved in the distribution of phospholipids across the plasma membrane (Kean et al., 1997) and acts to increase levels of phosphatidylethanolamine and phosphatidylcholine in the outer leaflet (Decottignies et al., 1998). These findings have led to the hypothesis that the Pdr pathway may coordinately regulate LCB and phos-

19

PM

Rsb1p M(IP) C 2 IPT1

Serine + Palmitoyl-CoA

LCB2

SUR2 DHS

LAC1 PHS

Ceramide

ER

Fig. 2. Pdr pathway interfaces with sphingolipid biosynthesis. A simplified version of the sphingolipid biosynthetic pathway is shown with the genes thought to be targets of Pdr pathway regulation listed. Rsb1p is believed to export long chain bases (LCBs) across the plasma membrane and out of cells. RSB1 is strongly transcriptionally induced in U0 cells, correlating with an increase in resistance to LCB challenge ((Devaux et al., 2002), our unpublished data). Biosynthesis of sphingolipids up to the ceramide intermediate is believed to occur via endoplasmic reticulum-localized proteins (Huh et al., 2003) while the later steps occur on other membrane compartments (Levine et al., 2000).

pholipid transport to ensure proper membrane structure (Kihara and Igarashi, 2004).

5. Conclusions Mitochondria that have lost their organellar genome likely represent an extreme in retrograde signaling. Clearly, major nutritional challenges emerge when the functions of the mitochondria are severely compromised as would occur in U0 cells. While elegant physiology has emerged to explain why U0 cells induce expression of CIT2 to ensure appropriate amino acid biosynthesis (Butow and Avadhani, 2004), the rationale for triggering PDR gene expression remains elusive. However, at least two models seem possible. First, U0 cells may induce expression of multidrug transporters like Pdr5p to allow export of toxic intermediates that accumulate in the presence of non-functioning mitochondria. It is possible that the presence of fully reduced electron transport components, as would be formed in U0 cells, may produce oxidation products in the cell that must be removed to remain viable. Second, disturbances in the mitochondria may lead to imbalances in a crucial cellular process, like membrane biosynthesis, that are compensated for by induction of PDR gene products. The induction of RSB1 transcription in pdr5D cells provides support for the idea that PDR gene activity is linked to membrane features such as lipid composition (Kihara and Igarashi, 2004). While the existence of a Pdr5p – Rsb1p regulatory circuit is a provocative possibility, the inverse relationship

20

W.S. Moye-Rowley / Gene 354 (2005) 15 – 21

between PDR5 and RSB1 expression is difficult to rationalize given what is known of the regulation of these two genes. Both PDR5 and RSB1 are induced under all circumstances known to activate PDR gene expression such as U0 cells (Devaux et al., 2002) or hyperactive mutant forms of Pdr1p or Pdr3p (DeRisi et al., 2000). Together, these data suggest that PDR5 and RSB1 would be expected to be co-regulated although the complete absence of Pdr5p might trigger a new signaling pathway leading to RSB1 gene activation. Further investigation of the nature of pdr5D-induced RSB1 transcription will shed light on the regulatory circuitry here. The finding that multidrug resistance genes are induced in U0 cells has also been reported in mammalian cells. Hepatoma cells that have been cured of their mitochondrial genome by growth in the presence of ethidium bromide exhibit elevated expression of P-glycoprotein and enhanced adriamycin resistance (Pillay et al., 1998). DNA microarray experiments have not yet detected changes in multidrug resistance gene expression, although only a single cell line was converted to U0 via ethidium bromide treatment (Delsite et al., 2002). A potential problem with the use of ethidium bromide to remove the organellar genome is the intrinsic nuclear mutagenicity of this compound (Turner and Denny, 1996). It is difficult to rule out the presence of multiple mutations that might complicate interpretation of the microarray experiment. Since mitochondrial –nuclear communication is likely to be an ancient adaptation of eukaryotic cells, it seems probable that features of the transcriptional circuitry found in S. cerevisiae will be conserved in mammalian cells. Further examination of the transcriptional profiles of U0 mammalian cells will be instructive in testing this hypothesis.

Acknowledgement Work on multidrug resistance in my laboratory is supported by NIH GM49825. I thank Dr. Sneh Lata Panwar for a critical reading of this manuscript.

References Altamura, N., et al., 1996. The Saccharomyces cerevisiae OXA1 gene is required for the correct assembly of cytochrome c oxidase and oligomycin-sensitive ATP synthase. FEBS Lett. 382, 111 – 115. Balzi, E., et al., 1987. The multidrug resistance gene PDR1 from Saccharomyces cerevisiae. J. Biol. Chem. 262, 16871 – 16879. Barz, W.P., Walter, P., 1999. Two endoplasmic reticulum (ER) membrane proteins that facilitate ER-to-Golgi transport of glycosylphosphatidylinositol-anchored proteins. Mol. Biol. Cell 10, 1043 – 1059. Briggs, S.D., et al., 2001. Histone H3 lysine 4 methylation is mediated by Set1 and required for cell growth and rDNA silencing in Saccharomyces cerevisiae. Genes Dev. 15 (24), 3286 – 3295. Burns, N., et al., 1994. Large-scale analysis of gene expression, protein localization, and gene disruption in Saccharomyces cerevisiae. Genes Dev. 8, 1087 – 1105.

Butow, R.A., Avadhani, N.G., 2004. Mitochondrial signaling: the retrograde response. Mol. Cell 14 (1), 1 – 15. Carvajal, E., et al., 1997. Molecular and phenotypic characterization of yeast PDR1 mutants that show hyperactive transcription of various ABC multidrug transporter genes. Mol. Gen. Genet. 256, 406 – 415. Cuvillier, O., 2002. Sphingosine in apoptosis signaling. Biochim. Biophys. Acta 1585, 153 – 162. Decottignies, A., et al., 1998. ATPase and multidrug transport activities of the overexpressed yeast ABC protein Yor1p. J. Biol. Chem. 273, 12612 – 12622. Delahodde, A., et al., 1995. Positive autoregulation of the yeast transcription factor Pdr3p, involved in the control of the drug resistance phenomenon. Mol. Cell. Biol. 15, 4043 – 4051. Delaveau, T., et al., 1992. Sequence of a 12.7 kb segment of yeast chromosome II identifies a PDR-like gene and several new open reading frames. Yeast 8, 761 – 768. Delsite, R., et al., 2002. Nuclear genes involved in mitochondria-to-nucleus communication in breast cancer cells. Mol. Cancer 1 (1), 6. DeRisi, J., et al., 2000. Genome microarray analysis of transcriptional activation in multidrug resistance yeast mutants. FEBS Lett. 470, 156 – 160. Devaux, F., et al., 2002. Genome-wide studies on the nuclear PDR3controlled response to mitochondrial dysfunction in yeast. FEBS Lett. 515, 25 – 28. Dickson, R.C., Lester, R.L., 2002. Sphingolipid functions in Saccharomyces cerevisiae. Biochim. Biophys. Acta 1583, 13 – 25. Dickson, R.C., et al., 1997. Synthesis of mannose – (inositol-P)2-ceramide, the major sphingolipid in Saccharomyces cerevisiae, requires the IPT1 (YDR072c) gene. J. Biol. Chem. 272, 29620 – 29625. Haak, D., et al., 1997. Hydroxylation of Saccharomyces cerevisiae ceramides requires Sur2p and Scs7p. J. Biol. Chem. 272, 29704 – 29710. Hallstrom, T.C., Moye-Rowley, W.S., 2000. Multiple signals from dysfunctional mitochondria activate the pleiotropic drug resistance pathway in Saccharomyces cerevisae. J. Biol. Chem. 275, 37347 – 37356. Hallstrom, T.C., et al., 2001. Coordinate control of sphingolipid biosynthesis and multidrug resistance in Saccharomyces cerevisiae. J. Biol. Chem. 276, 23674 – 23680. Hermann, G.J., et al., 1998. Mitochondrial fusion in yeast requires the transmembrane GTPase Fzo1p. J. Cell Biol. 143, 359 – 373. Huang, H., et al., 1997. The ubiquitin-conjugating enzyme Rad6 (Ubc2) is required for silencing in Saccharomyces cerevisiae. Mol. Cell. Biol. 17 (11), 6693 – 6699. Huh, W.K., et al., 2003. Global analysis of protein localization in budding yeast. Nature 425 (6959), 686 – 691. Hwang, W.W., et al., 2003. A conserved RING finger protein required for histone H2B monoubiquitination and cell size control. Mol. Cell 11 (1), 261 – 266. Jia, Y., et al., 1997. A basic helix – loop – helix – leucine zipper transcription complex in yeast functions in a signaling pathway from mitochondria to the nucleus. Mol. Cell. Biol. 17, 1110 – 1117. Jorgensen, P., et al., 2002. Systematic identification of pathways that couple cell growth and division in yeast. Science 297, 395 – 400. Katzmann, D.J., et al., 1994. Transcriptional control of the yeast PDR5 gene by the PDR3 gene product. Mol. Cell. Biol. 14, 4653 – 4661. Katzmann, D.J., et al., 1996. Multiple Pdr1p/Pdr3p binding sites are essential for normal expression of the ATP binding cassette transporter protein-encoding gene PDR5. J. Biol. Chem. 271, 23049 – 23054. Kean, L.S., et al., 1997. Plasma membrane translocation of fluorescentlabeled phosphatidylethanolamine is controlled by transcription regulators, PDR1 and PDR3. J. Cell Biol. 138, 255 – 270. Kihara, A., Igarashi, Y., 2002. Identification and characterization of a Saccharomyces cerevisiae gene, RSB1, involved in sphingoid longchain base release. J. Biol. Chem. 277, 30048 – 30054. Kihara, A., Igarashi, Y., 2004. Crosstalk between sphingolipids and glycerophospholipids in the establishment of plasma membrane asymmetry. Mol. Biol. Cell 15 (11), 4949 – 4959.

W.S. Moye-Rowley / Gene 354 (2005) 15 – 21 Kolaczkowski, M., et al., 2004. Differential regulation of ceramide synthase components LAC1 and LAG1 in Saccharomyces cerevisiae. Eukaryot. Cell 3 (4), 880 – 892. Leppert, G., et al., 1990. Cloning by gene amplification of two loci conferring multiple drug resistance in Saccharomyces. Genetics 125, 13 – 20. Levine, T.P., et al., 2000. Inositol phosphorylceramide synthase is located in the Golgi apparatus of Saccharomyces cerevisiae. Mol. Biol. Cell 11 (7), 2267 – 2281. Liao, X., Butow, R.A., 1993. RTG1 and RTG2: two yeast genes required for a novel path of communication from the mitochondria to the nucleus. Cell 72, 61 – 71. Mao, C., et al., 2000a. Cloning of an alkaline ceramidase from Saccharomyces cerevisiae: an enzyme with reverse (CoA-independent) ceramide synthase activity. J. Biol. Chem. 275, 6876 – 6884. Mao, C., et al., 2000b. Cloning and characterization of a Saccharomyces cerevisiae alkaline ceramidase with specificity for dihydroceramide. J. Biol. Chem. 275, 31369 – 31378. Mathias, S., et al., 1998. Signal transduction of stress via ceramide. Biochem. J. 335, 465 – 480. Meyers, S., et al., 1992. Interaction of the yeast pleiotropic drug resistance genes PDR1 and PDR5. Curr. Genet. 21, 431 – 436. Nagiec, M.M., et al., 1994. The LCB2 gene of Saccharomyces and the related LCB1 gene encode subunits of serine palmitoyltransferase, the initial enzyme in sphingolipid synthesis. Proc. Natl. Acad. Sci. U. S. A. 91, 7899 – 7902. Nourani, A., et al., 1997. Clustered amino acid substitutions in the yeast transcription regulator Pdr3p increase pleiotropic drug resistance and identify a new central regulatory domain. Mol. Gen. Genet. 256, 397 – 405. Parikh, V.S., et al., 1987. The mitochondrial genotype can influence nuclear gene expression in yeast. Science 235, 576 – 580.

21

Pillay, V., et al., 1998. Upregulation of P-glycoprotein in rat hepatoma rho(o) cells: implications for drug – DNA interactions. J. Cell. Biochem. 69 (4), 463 – 469. Rank, G.H., Bech-Hansen, N.T., 1973. Single nuclear gene inherited cross resistance and collateral sensitivity to 17 inhibitors of mitochondrial function in Saccharomyces cerevisiae. Mol. Gen. Genet. 126, 93 – 102. Rapaport, D., et al., 1998. Fzo1p is a mitochondrial outer membrane protein essential for the biogenesis of functional mitochondria in Saccharomyces cerevisiae. J. Biol. Chem. 273, 20150 – 20155. Reichert, A.S., Neupert, W., 2004. Mitochondriomics or what makes us breathe. Trends Genet. 20 (11), 555 – 562. Saba, J.D., et al., 1997. The BST1 gene of Saccharomyces cerevisiae is the sphingosine-1-phosphate lyase. J. Biol. Chem. 272, 26087 – 26090. Simonics, T., et al., 2000. Isolation and molecular characterization of the carboxy-terminal pdr3 mutants in Saccharomyces cerevisiae. Curr. Genet. 38, 248 – 255. Small, W.C., et al., 1995. Enzymatic and metabolic studies on retrograde regulation mutants of yeast. Biochemistry 34 (16), 5569 – 5576. Subik, J., et al., 1986. Genetic mapping of nuclear mucidin resistance mutations in S. cerevisiae. Curr. Genet. 10, 665 – 670. Turner, P.R., Denny, W.A., 1996. The mutagenic properties of DNA minorgroove binding ligands. Mutat. Res. 355 (1-2), 141 – 169. Zhang, X., Moye-Rowley, W.S., 2001. Saccharomyces cerevisiae multidrug resistance gene expression inversely correlates with the status of the Fo component of the mitochondrial ATPase. J. Biol. Chem. 276, 47844 – 47852. Zhang, X., et al., 2005. Transcriptional regulation by Lge1p requires a function independent of its role in H2B ubiquitination. J. Biol. Chem. 280 (4), 2759 – 2770. Zhou, P., Thiele, D.J., 1993. Copper and gene regulation in yeast. BioFactors 4, 105 – 115.