Nic1p, a relative of bacterial transition metal permeases in Schizosaccharomyces pombe, provides nickel ion for urease biosynthesis.

The Schizosaccharomyces pombe genome sequencing project identified an open reading frame (O74869 and O74912, named Nic1p in the present study) with significant similarity to members of a family of bacterial transition metal permeases. These uptake systems transport Ni(2+) ion with extremely high affinity across the bacterial cytoplasmic membrane, but they differ in selectivity toward divalent transition metal cations. An S. pombe mutant harboring an interrupted nic1 allele (nic1-1) was strongly impaired in (63)Ni(2+) uptake in the presence of a high molar ratio of Mg(2+) relative to Ni(2+), conditions that reflect the natural situation. Under these conditions, the nic1-1 mutant contained only background activities of the nickel-dependent cytoplasmic enzyme urease and could not catabolize urea. Among a series of divalent transition metal cations tested (Cd(2+), Co(2+), Cu(2+), Mn(2+), and Zn(2+)), only Co(2+) caused considerable inhibition of Nic1p-mediated Ni(2+) uptake. On the other hand, experiments with (57)Co(2+) (at nm concentrations) did not show significant differences in Co(2+) uptake between the nic1-1 mutant and the parental strain. Our data suggest that Nic1p acts as a plasma-membrane nickel transporter in fission yeast, a finding that invites searches for isologous counterparts in higher eukaryotes.

Nickel-dependent enzymes catalyze key reactions in energy and nitrogen metabolism in both prokaryotes and eukaryotes (for reviews, see Refs. 1 and 2). The most widespread nickel metalloenzyme is urease (EC 3.5.1.5), which has been identified in prokaryotes, fungi, algae, plants, and invertebrates. Ureases allow their hosts to utilize urea as a source of nitrogen. They are also important virulence factors in bacteria and fungi (reviewed in Refs. 3 and 4). Urease-mediated hydrolysis of urea yielding ammonium ion and carbamate is the common mechanism of biological urea degradation, although an alternative mechanism is known. In baker's yeast Saccharomyces cerevisiae, other yeasts, and certain green algae, the breakdown of urea is mediated by a biotin-and ATP-dependent carboxylation to give 1-carboxyurea (allophanate) and subsequent hydrolysis to ammonia and carbon dioxide (5).
The fission yeast Schizosaccharomyces pombe contains a urease and is able to grow on urea as the sole source of nitrogen.
The soluble, cytoplasmic urease has been purified to homogeneity, and its kinetic properties have been determined (6). Its structural gene (ure1 ϩ ) has been cloned and sequenced, and a knockout mutation was shown to inhibit the growth of S. pombe on agar plates containing 10 mM urea as the nitrogen source (7). S. pombe urease is neither controlled by nitrogen repression nor by urea induction (6).
High affinity nickel transport, a prerequisite for the biosynthesis of nickel-containing metalloenzymes, and the underlying uptake mechanisms have been investigated in a number of prokaryotes (reviewed in Refs. 8 and 9). Comparable transporters, however, have not yet been reported for eukaryotes. Trace amounts of Ni 2ϩ ion are sufficient for maximal urease activity of S. pombe, and this activity was not stimulated by the addition of Ni 2ϩ to the medium (6). This result suggested that an uptake system with a very high affinity for Ni 2ϩ operates in fission yeast. The S. pombe genome sequencing project identified an open reading frame (Nic1p) that displays similarity to a family of bacterial transition metal permeases (9). In the present report we show that interruption of the respective gene strongly affected urease activity under conditions of nickel limitation and prevented the growth of S. pombe on urea. We demonstrate that Nic1p is a high affinity nickel permease. This is the first example of this kind of transporter in a eukaryotic organism. We also present evidence that nonspecific metal uptake systems allow S. pombe to transport Ni 2ϩ ion under certain conditions.

MATERIALS AND METHODS
Organisms, Media, and Growth Conditions-S. pombe var. pombe 972 h Ϫs Lindner (wild type, DSM 70576) was obtained from the Deutsche Sammlung fü r Mikroorganismen und Zellkulturen GmbH (Braunschweig, Germany). S. pombe strain FY254 (ATCC 201402) (h Ϫ can1-1 leu1-32 ade6-M210 ura4-D18) was a gift of Susan Forsburg (The Salk Institute, La Jolla, CA) and was used as the parental strain for gene interruption. S. pombe strains were grown in YES (0.5% w/v yeast extract, 3% w/v D-glucose) medium or in Edinburgh minimal medium (EMM) (10) at 30°C. Under standard conditions, adenine, L-leucine, and uracil were added as supplements to both media to give final concentrations of 100 g/ml. Cell densities of cultures in YES medium were calculated upon measuring the optical density (A 595 ) in a spectrophotometer and establishing the ratio between A 595 and the cell concentration. One A 595 unit corresponded to 2.2 ϫ 10 7 colony-forming units/ml. Growth on urea as the nitrogen source was monitored on agar plates containing a modified EMM medium. The aforementioned supplements were added to final concentrations of 10 g/ml, and urea (10 mM) was supplied in place of ammonium salt as the nitrogen source. Plasmids containing S. pombe DNA were propagated in Escherichia coli strains DH5␣FЈ (Life Technologies, Inc.) and XL1-Blue (Stratagene, Amsterdam) and the Dam Ϫ strain GM2163 (New England Biolabs, * This work was funded by grant Ei 374/1-2 from the Deutsche Forschungsgemeinschaft (to T. E.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Schwalbach, Germany). Recombinant E. coli strains were grown in LB medium supplemented with appropriate antibiotics.
Gene Interruption-The nic1 ϩ gene was cloned following amplification using total DNA of S. pombe wild-type strain 972 h Ϫ as the template. Cells (10 ml) grown overnight in YES medium were harvested, washed in SCE buffer (900 mM sorbitol, 50 mM sodium citrate, 10 mM EDTA, pH 7.5), and resuspended in 350 l SCE buffer containing 2-mercaptoethanol (0.8% w/v) and lyticase (Sigma) (1 mg/ml). After 30 min at 37°C, the cells were pelleted and lysed by vigorous shaking after the addition of 250 l of lysis buffer (10 mM Tris hydrochloride, pH 8.0, 100 mM NaCl, 2% w/v Triton X-100, 1% w/v SDS), 300 mg of acidwashed glass beads (Sigma), and 250 l of phenol/chloroform/2-pentanol (25:24:1 v/v). After the addition of 200 l of TE buffer (10 mM Tris hydrochloride, pH 8.0, 1 mM EDTA) and centrifugation, nucleic acids in the aqueous phase were precipitated with ethanol. The pellet was washed with 70% (w/v) ethanol, dried under vacuum, resuspended in 50 l of distilled water containing 5 g of RNase A, and stored at 4°C. 1 l of this sample was used for PCR (30 cycles of 20 s at 94°C, 30 s at 50°C, 2 min at 72°C) in the presence of primer A (5Ј-GCGCCTCGAGATAAA-CACGTTTAAGCATCACC-3Ј), primer B (5Ј-CGAAAAAGAGCTCG-GCATTAAACATACC-3Ј) (Fig. 1), and 2.5 units of Taq DNA polymerase (Pan Systems, Aidenbach, Germany). The purified PCR product was incubated with restriction endonucleases XhoI and SacI and ligated into vector pBluescript II SKϩ (Stratagene). For interruption of nic1 ϩ , this plasmid was isolated from E. coli GM2163. 246 base pair of nic1 were deleted by incubation with BclI and SpeI. The fragment was replaced by a 1.8-kilobase BamHI/XbaI fragment harboring the S. pombe ura4 ϩ marker gene (a gift of Jü rg Kohli, Universitä t Bern, Switzerland). The disrupted nic1 gene (nic1-1) was amplified by PCR (30 cycles of 20 s at 94°C, 30 s at 45°C, 2.5 min at 72°C) using primer A, primer P4 (5Ј-CGGGAGCTCTCAAACCTTAGAATCCACTGTATCG-3Ј) (Fig. 1) and Taq DNA polymerase. Approximately 20 g of the purified PCR product was used to transform S. pombe FY254 by the method of Bä hler et al. (11). A 20-ml culture of FY254 in YES was grown to a density of approximately 10 7 cells/ml. The cells were washed twice with distilled water and once with LiAc/TE (100 mM lithium acetate, 10 mM Tris-hydrochloride, 1 mM EDTA, pH 7.5) and resuspended in 100 l of LiAc/TE. 20 g each of carrier DNA (sheared salmon sperm DNA, Stratagene), and the PCR product was added. After 10 min at room temperature, 260 l of polyethylene glycol 4000 (40% w/v in LiAc/TE) was added, and the mixture was incubated at 30°C for 1 h. Finally, 43 l of dimethyl sulfoxide was added followed by heating to 42°C for 5 min. The cells were washed and resuspended in 500 l of distilled water. 250 l was plated on uracil-free EMM agar. Colonies of ura4 ϩcontaining transformants appeared after 4 days and were purified by streaking on uracil-free agar plates. The disruption of nic1 ϩ in recombinants was verified by PCR analysis (primer combinations P1 (5Ј-GGGCATATGTCTGAATATGTTAAACC-3Ј)/P4 and P1/B) and Southern blotting (Fig. 1). For the latter purpose, a PCR product obtained with primers P1 and P4 and the nic1 ϩ allele as the template was labeled with digoxigenin-11-dUTP (Roche Molecular Biochemicals) and used as the probe.
Nickel Accumulation in Growing Cells-Nickel accumulation in S. pombe strains was analyzed by a modification of the method previously described for the analysis of recombinant E. coli (12,13,14). S. pombe cells were grown to an optical density (A 595 ) of approximately 1.5 in YES medium containing 63 NiCl 2 (24.4 TBq/mol) and the indicated supplements. Cells were harvested, washed twice with 50 mM Tris hydrochloride, pH 7.5, and concentrated 10-fold. The radioactivity of 50 l of the cell suspension was quantitated by liquid scintillation counting in a Canberra-Packard 1600 TR counter using Zinsser Aquasafe 300 Plus as scintillation mixture. Nickel accumulation is expressed as pmol/10 9 cells.
Nickel Transport-S. pombe strains were grown in 10 ml of YES medium to an optical density (A 595 ) of approximately 5 absorbance units, harvested, washed twice, and resuspended in 10 ml of transport buffer (20 mM MES/NaOH, pH 6.2, 10 mM MgCl 2 , 2% w/v D-glucose) and equilibrated at 30°C in a water bath shaker for 5 min. 63 NiCl 2 (24.4 TBq/mol) was added to a final concentration of 10 nM. Samples (200 l) were taken at the indicated time points, passed through Whatman glass microfiber filters (GF/C), and washed twice with 3.5 ml of glucose-free transport buffer. The radioactivity of filter-bound cells was analyzed by liquid scintillation counting. Nickel transport is expressed as pmol/10 9 cells.
Urease Assay-Cells were grown in 50 ml of YES medium overnight in the presence of the indicated supplements, harvested, washed, and resuspended in 2 ml potassium phosphate buffer (20 mM, pH 7.5). After two passages through a French pressure cell, the crude extracts were separated by ultracentrifugation (59,000 ϫ g, 30 min, 4°C). 20 l of the supernatant (containing approximately 300 g of protein) was added to 880 l of potassium phosphate buffer (20 mM, pH 7.5), and the mixture was equilibrated at 37°C. The reaction was started by the addition of 100 l of a freshly prepared urea solution (100 mM). Urease activity was determined spectrophotometrically by quantitating the rate of ammonium ion released from urea. For this purpose, ammonium ion was converted into indophenol (15). Protein was estimated by a modification Hydropathy profile (9) and amino acid sequence alignments (Fig. 2) revealed that Nic1p is closely related to the bacterial counterparts. The aforementioned sequence motifs are fully conserved. The putative transmembrane helices IV and V of Nic1p are linked by a hydrophilic loop (residues 164 to 208) containing 12 potentially charged residues.
Interruption of the nic1 ϩ Gene-nic1 ϩ is located on chromosome III of S. pombe between the long terminal repeat of a Tf2-type retrotransposon and an open reading frame of un-known function. The strategy for gene interruption is illustrated in Fig. 1. 246 base pairs of nic1 ϩ were deleted and replaced by an approximately 1.8-kilobase ura4 ϩ marker gene (17). Using 20 g of the amplified construct for transformation into S. pombe FY254, approximately 1,000 transformants were obtained on uracil-free EMM agar plates. Two transformants were chosen and shown by PCR (Fig. 1B) and Southern blotting (Fig. 1C) to contain the disrupted nic1 allele. Both mutants grew normally in mineral medium in the presence of ammonium salt as the nitrogen source as well as in complex medium, indicating that nic1 ϩ is dispensable under both conditions.
Nickel Uptake-Based on the similarity to the bacterial nickel permeases, we suspected that Nic1p plays a role in nickel transport into S. pombe cells. To investigate this hypothesis, nickel accumulation of growing cells of the S. pombe nic1-1 mutant was compared with metal uptake by the parental strain under various conditions (Table I). In the presence of 5 M 63 NiCl 2 in complex medium, both strains accumulated high amounts of Ni 2ϩ ion. The addition of magnesium salt to the medium resulted in a 20-fold decrease in nickel accumulation. At a low Ni 2ϩ concentration (100 nM), the nic1-1 mutation caused a moderately reduced Ni 2ϩ accumulation when the medium was not supplemented with Mg 2ϩ ion. In the presence of 10 mM Mg 2ϩ , however, a strong effect was observed. Although 10 9 cells of the parental strain accumulated 48 pmol of nickel, metal accumulation of the nic1-1 mutant decreased to 4 pmol of nickel/10 9 cells (Table I) confirmed the assumption that Nic1p acts as a high affinity nickel transporter in S. pombe. This conclusion was further substantiated by uptake assays with resting cells in buffer. While the nic1-1 mutant was unable to transport 63 Ni 2ϩ over the 50-min test period, significant transport was observed for the parental strain (Fig. 4).
Our data are compatible with the view that S. pombe is able to transport Ni 2ϩ ion by nonspecific Mg 2ϩ uptake systems. A similar situation has been reported for S. cerevisiae (18). Lesions in the two S. cerevisiae genes ALR1 and ALR2 were found to produce a magnesium-deficient phenotype while conferring increased resistance to certain metal ions including the divalent ions of the transition metals copper, manganese, nickel, and zinc. Alr1p and Alr2p belong to the CorA family of membrane transporters, the most widespread type of nonspecific Mg 2ϩ uptake system in bacteria and archaea (19). Alr1p-and Alr2p-like proteins are also encoded in the genome of S. pombe. The respective open reading frames (O13779 and O13657) together with Alr1p and Alr2p contain large N-terminal extensions compared with their prokaryotic counterparts and represent a CorA subfamily (19).
At very low Ni 2ϩ concentrations and high molar ratios of Mg 2ϩ to Ni 2ϩ , Ni 2ϩ uptake of S. pombe was dependent on Nic1p. This result indicated that nonspecific systems contribute little to Ni 2ϩ uptake under conditions that reflect the situation in the natural environment.
Selectivity of Nic1p-To test the specificity of Nic1p, the effect of cadmium, cobalt, copper, manganese, and zinc ions on  nickel accumulation of S. pombe FY254 was investigated. For this purpose, the cells were grown in YES medium containing 100 nM 63 NiCl 2 and 10 mM MgCl 2 . Under these conditions, high level nickel accumulation is dependent on Nic1p. The competing metal ions were added to final concentrations of 1 M. With the exception of Co 2ϩ , none of the metal ions caused significant inhibition of nickel accumulation (data not shown). The effect of Co 2ϩ was analyzed in more detail. Fig. 5 illustrates that Co 2ϩ was an inhibitor. At a 50-fold excess, Co 2ϩ ion abolished 63 Ni 2ϩ accumulation in S. pombe. We then addressed the question of whether Nic1p is capable of transporting Co 2ϩ ion. The nic1-1 mutant and its parental strain FY254 were grown in YES medium supplemented with 57 CoCl 2 at concentrations between 100 and 500 nM in the presence of 10 mM MgCl 2 . Both strains were able to accumulate cobalt, and no obvious difference was found under any conditions tested (data not shown). This result suggested that Nic1p is not the main mediator of Co 2ϩ uptake. The observed Co 2ϩ accumulation could be due to the activity of a nonspecific transition metal transporter of the Nramp (natural resistance-associated macrophage protein) family. Three Nramp-like proteins (Smf1p, Smf2p, and Smf3p) have been identified in S. cerevisiae, and a homologous open reading frame (Q10177) is also encoded in the genome of S. pombe (see Refs. 20 and 21 for recent reviews).
Physiological Role of Nic1p-To elucidate the physiological consequences of the nic1-1 mutation, we first monitored growth on mineral agar plates containing 10 mM urea as the nitrogen source (not shown). Although S. pombe FY254 formed colonies similar to those observed on ammonium-containing plates after 3 to 4 days, the nic1-1 mutant failed to grow on urea under standard conditions. Growth of the mutant was partially restored by adding nickel salt at M concentrations to the medium. At Ni 2ϩ concentrations above 500 M, growth of both strains as well as the wild-type strain 972 h Ϫ was completely inhibited. S. pombe harbors a putative metallothionein and has recently been shown to produce a phytochelatin synthase (22). Metallothioneins and phytochelatins mainly mediate resistance toward cadmium, copper, and zinc and apparently do not allow S. pombe to escape nickel toxicity under the conditions tested.
We next investigated the role of Nic1p in urea metabolism by quantitative urease assays (Table II). Upon growth of the cells in standard YES medium, soluble extracts of the nic1-1 mutant and its parental strain contained urease-specific activities of approximately 200 to 250 milliunits/mg of protein. These values were in good agreement with those published previously (6). The addition of NiCl 2 (5 M) to the growth medium resulted in a slightly increased urease activity of strain FY254. Surprisingly, high concentrations of MgCl 2 (20 mM) had almost no effect on urease activity of both strains. Under comparable conditions, efficient Ni 2ϩ uptake was dependent on Nic1p (Table I, Figs. 3 and 4). We hypothesized that low level Ni 2ϩ uptake, which could be due to a nonspecific Nramp-like transporter, is sufficient for maximal urease activity under these conditions. Therefore, a nickel-complexing agent (NTA), which had proved to efficiently inhibit nonspecific nickel uptake in bacteria (13), was added to the growth medium. 50 M NTA led to a small but significant decrease of urease activity in the absence of Nic1p. The addition of a combination of NTA and Mg 2ϩ , however, gave a strong response. Although the parental strain was almost unaffected, urease activity in the mutant was below the threshold of the assay. Since the natural habitats of S. pombe contain nutrients with strong metal-complexing capacity and since the molar ratio of Mg 2ϩ to Ni 2ϩ is generally very high, the latter growth conditions reflect the situation in the environment.
Our results identified Nic1p as an important auxiliary factor for urease activity in S. pombe, a finding that may be of general significance for the analysis of urease biosynthesis in eukaryotes. Additional urease accessory genes in fission yeast have been tentatively mapped (23). The genome sequencing project revealed homologs of the bacterial urease operon proteins UreD and UreF, which are essential for urease metallocenter assembly (4). An isolog with high similarity to the bacterial UreG, a GTPase that is important for nickel incorporation into urease apoprotein (24), is not obvious from the S. pombe genome sequence.
Although at present we have only indirect data on the localization of Nic1p, our results strongly suggest that it represents a plasma-membrane transporter. Protein-sorting signals are not obvious from the primary structure, and recent work on the Nramp homolog Smf1p of S. cerevisiae has demonstrated that sorting is a mechanism of post-translational activity control of plasma-membrane transporters in yeast (25,26).
Compared with its bacterial relatives, Nic1p has a unique specificity. Ni 2ϩ transport by HoxN of Ralstonia eutropha, for instance, is not inhibited by Co 2ϩ , and HoxN does not transport Co 2ϩ ion (14). On the other hand, Ni 2ϩ uptake by NhlF of Rhodococcus rhodochrous is specifically inhibited by Co 2ϩ , and this permease is able to transport Co 2ϩ ion (14). Nic1p seems to be a third type of nickel permease, since Co 2ϩ was an inhibitor but, if at all, only a weak substrate for transport. Understanding the molecular basis of the differences in ion selectivity is a challenging problem.