Blackwell Science, LtdOxford, UKMMIMolecular Microbiology0950-382XBlackwell Publishing Ltd, 2004? 2004552588600Original ArticleS. cerevisiae Khalp is an intracellular Na +/H+ antiporterL. Maresova and H. Sychrova
Molecular Microbiology (2005) 55(2), 588–600
doi:10.1111/j.1365-2958.2004.04410.x
Physiological characterization of Saccharomyces cerevisiae kha1 deletion mutants Lydie Maresova* and Hana Sychrova Department of Membrane Transport, Institute of Physiology, Academy of Sciences CR, 142 20 Prague 4, Czech Republic. Summary Maintenance of intracellular K+ homeostasis is one of the crucial requisites for the survival of yeast cells. In Saccharomyces cerevisiae, the high K+ content corresponds to a steady state between simultaneous influx and efflux across the plasma membrane. One of the transporters formerly believed to extrude K+ from the yeast cells (besides Ena1-4p and Nha1p) was named Kha1p and presumed as a putative plasma membrane K+/H+ antiporter. We prepared kha1 and tok1-kha1 deletion strains in the B31 and MAB 2d background. Both the strains contain the ena1-4 and nha1 deletions; that means they lack the main active sodium and potassium efflux systems. MAB 2d has additional trk1 and trk2 deletions, i.e. is impaired in active K+ uptake as well. We performed a large physiological study with these strains to specify the phenotype of kha1 deletion. In our experiments, no difference in K+ content or efflux was observed in strains lacking the KHA1 gene compared with control strains. Two main phenotype manifestations of the kha1 deletion were growth defect on high external pH and hygromycin sensitivity. The correlation between these phenotypes and the kha1 deletion was confirmed by plasmid complementation. Fluorescence microscopy of green fluorescent protein (GFP)-tagged Kha1p showed that this antiporter is localized preferentially intracellularly (in contrast to the plasma membrane Na+/H+ antiporter Nha1p). Based on these findings, Kha1p is probably not localized in plasma membrane and does not mediate efflux of alkali metal cations from cells, but is important for the regulation of intracelular cation homeostasis and optimal pH control, similarly as the Nhx1p. Introduction Intracellular cations and anions participate in many physAccepted 27 September, 2004. *For correspondence. E-mail
[email protected]; Tel. (+420) 296 442 194; Fax (+420) 296 442 488.
© 2004 Blackwell Publishing Ltd
iological functions of yeast cells. In Saccharomyces cerevisiae, K+ is the most abundant intracellular cation and yeast cells spend energy to accumulate the required amount against large electrochemical gradient. In contrast, high Na+ concentration is toxic for most cells, including yeast. The intracellular K+/Na+ ratio plays a crucial role in cation homeostasis, because K+ is necessary for many cellular functions and capable of preventing most Na+ inhibitory effects (Gomez et al., 1996). One of the strategies to keep high intracellular K+/Na+ ratio is strict discrimination among these cations at the level of influx (for a review, see Rodriguez-Navarro, 2000; Sychrova, 2004). There are two homologous proteins, Trk1p and Trk2p, that mediate highly selective active K+ uptake (Ko and Gaber, 1991; Ramos et al., 1994). Cells carrying mutations in both TRK1 and TRK2 are not able to grow at low K+ concentration in the media. Another potassium-specific transport system was named Tok1p (Ketchum et al., 1995). This voltage-dependent K+ channel displays strong outwards rectification, but under certain conditions is able to mediate K+ uptake as well (Bertl et al., 1993; Fairman et al., 1999). Besides Trk1p, Trk2p and Tok1p, yeast cells possess several systems that mediate import or export of alkali metal cations with lower selectivity. A non-selective cation channel, NSC1, characterized by whole-cell patch clamp analysis (Bihler et al., 1998), is able to facilitate the uptake of a wide range of cations including K+ (Bihler et al., 2002). Although the complete S. cerevisiae genome has been known for almost 10 years, the gene(s) encoding the NSC1 system have not been found yet. In contrast to K+, the amount of Na+ in cells under nonstressed conditions is very low. But upon salt stress, Na+ enters cells following its electrochemical gradient and must be efficiently extruded to prevent toxic effects. The main system eliminating Na+ from S. cerevisiae cells is a plasma membrane P-type ATPase encoded by the ENA/ PMR2 locus, consisting of a tandem array of nearly identical genes. The number of gene copies depends on yeast strain (Haro et al., 1991; Wieland et al., 1995). Furthermore, three distinct genes encoding putative alkali metal cation/H+ antiporters have been identified by the yeast genome-sequencing project. First of them, the Nha1p, was cloned by selection for increased NaCl tolerance from a multicopy genomic library (Prior et al., 1996). The expression of the ENA1 or the NHA1 gene in an ena1-4D nha1D strain showed complementary function of
S. cerevisiae Khalp is an intracellular Na+/H + antiporter 589 both systems in the maintenance of optimal intracellular Na+ and K+ concentration. The Ena1 ATPase is responsible for cell growth on high concentrations of NaCl and KCl at high external pH values, whereas the Nha1 antiporter is necessary at acidic pH (Bañuelos et al., 1998). Both proteins transport several alkali metal cations, and no sodium or potassium efflux was measured in a strain with ena1-4 and nha1 deletions in buffer pH 5.5 or 8.0 (Bañuelos et al., 1998; Kinclova et al., 2001), confirming an important role of these two transport systems in the export of alkali metal cations. Another Na+/H+ antiporter, Nhx1p, was found to be localized in prevacuolar compartments, equivalent to late endosomes in animal cells (Nass and Rao, 1998). Nhx1p mediates intracellular sequestration of Na+ cations, and thus participates in Na+ tolerance of yeast cells (Nass et al., 1997; Nass and Rao, 1999). Recently, the role of Nhx1p in intracellular potassium homeostasis was also demonstrated (Fukuda et al., 2004). The Nhx1 protein sequence shows high homology to mammalian sodium/ proton exchangers of the NHE family. The last predicted alkali metal cation/H+ exchanger, encoded by YJL094c, with homology to a Na+/H+ antiporter from Enrerococcus hirae (NapA; Waser et al., 1992) and to a K+ efflux system of Escherichia coli (KefC; Munro et al., 1991), was characterized by comparison of a deletion yeast strain to a wild type. Based on the assumption that its primary function is the K+/H+ antiport across plasma membrane, this protein was named Kha1p (Ramirez et al., 1998). Kha1p is a 97 kDa protein (873 amino acids) with 12 putative transmembrane domains and very low level of expression. On average, there is about 0.3 mRNA copies (Jansen and Gerstein, 2000) and 172 molecules of the Kha1 protein (Ghaemmaghami et al., 2003) per cell. The work of Ramirez et al. (1998) is the only study dedicated to the function of Kha1 protein itself. According to their results, the kha1 deletion strain should have shorter duplication time than wild type, faster DNA replication, higher membrane potential, higher intracellular pH and K+ concentration. The mutant strain should also grow faster than wild type in media with either high salt concentration or a high pH value. Some additional information about possible roles of Kha1p in cells is to be found in the results of systematic deletion projects and genome-wide microarray studies. The kha1D mutant strains had no detectable phenotype in a variety of systematic tests assaying heat and cold sensitivity, mating, cytoskeleton and mitochondrial morphology, pseudohyphal growth and sensitivity to KCl or H2O2 (Entian et al., 1999). The only two mutant phenotypes observed in this study were growth defect on acetate, but not on other carbon sources, and slightly enhanced NaCl tolerance. Differently, growth defect on all © 2004 Blackwell Publishing Ltd, Molecular Microbiology, 55, 588–600
tested non-fermentable carbon sources (ethanol, glycerol, lactate) for kha1D mutants was observed in another study (Steinmetz et al., 2002). None of the kha1D phenotypes (observed by Ramirez et al., Entian et al. or Steinmetz et al.) has been confirmed by plasmid complementation. Also in the SGD microarray database, it is not easy to find conditions specific for changes in expression of the KHA1 gene. In most experiments, the KHA1 expression was without significant changes – for example, glucose limitation experiments (DeRisi et al., 1997; Chu et al., 1998), expression during the diauxic shift (DeRisi et al., 1997), response to alpha-factor (Roberts et al., 2000), varying zinc levels (Lyons et al., 2000), or alkali/acidic colony phase switching (Palkova et al., 2002). The KHA1 gene was shown to be partially repressed during sporulation (Chu et al., 1998), induced in cells entering stationary phase (Gasch et al., 2000), moderately induced by histone depletion (Wyrick et al., 1999) and slightly induced by salt stress (Posas et al., 2000). In this work, we present the characteristics of kha1D strains in the background of various multiple mutations in genes encoding alkali-metal-cation transporters. According to our results, the Kha1 protein does not mediate potassium efflux from cells and perhaps is not localized in plasma membrane at all. We believe this protein is localized in an intracellular organelle membrane, participates in alkali-metal-cation homeostasis regulation and is important for cell growth in alkalic pH environment. Results Based on the supposition that Kha1p is a plasma mambrane K+ exporter, we were first trying to find a kha1 mutant phenotype related to extracellular K+ concentration, because we knew that deletion of genes encoding other K+ efflux systems (ENA1-4, NHA1) increases the sensitivity to extracellular K+ (Bañuelos et al., 1998). We compared the W303 strain with a kha1::URA3 mutant (KMW 40-1 strain), but we did not observe any difference between the kha1 disruption strain and the parental strain growing at various concentrations of KCl (data not shown). Then we tried to observe some of the phenotypes described for kha1 mutants in Ramirez et al. (1998). We measured growth curves of W303 and the kha1::URA3 mutant in standard YPD media with 20 mg ml-1 uracil added (to compensate for the difference between ura– wild type and kha1::URA3 deletion mutant). In these conditions, the kha1::URA3 mutant did not grow faster than the control strain (Fig. 1). We believe that the shorter duplication time of the mutant observed by Ramirez et al. (1998) was caused by the difference in uracil metabolism. When we measured the growth curves in YPD media without the
590 L. Maresova and H. Sychrova
Fig. 1. Comparison of growth rates of the kha1::URA3 mutant (KMW 40-1, ) and wild type (W303, ) in standard YPD medium with 20 mg ml-1 uracil. Liquid media were inoculated to OD600 = 0.01, incubated in a rotation shaker at 30∞C and OD600 was measured in various time intervals. The experiment was repeated three times; the scatter was
