INTRODUCTION

Manganese is an important trace mineral element for all organisms. In plants and cyanobacteria, Mn²⁺ is an absolute requirement for light-induced oxygen evolution. In several microorganisms, high affinity manganese transport systems have been well documented in terms of kinetic parameters and substrate specificity (1). We have recently identified three structural genes for a manganese transporter protein complex in the cyanobacterium Synechocystis sp. PCC 6803 (hereafter called Synechocystis 6803). These genes are organized in an operon, mntCAB, which could complement BP13, a random mutant strain impaired in the activity of the manganese cluster of photosystem II (PSII)¹ (2). Our analysis indicated that mntCAB encodes the components of an ABC (TP inding assette)-type transporter system, the first such protein complex for manganese identified in any organism. The lesion in the BP13 cells is a single amino acid substitution in MntA, a soluble protein containing an ATP binding motif. Compared with the wild-type cells, the growth rate of the BP13 cells was significantly reduced in a manganese-deficient medium and could be restored to a normal level by the addition of micromolar amounts of manganese in the incubation medium. These results indicated that either the mutation in the mntA gene in the BP13 strain does not completely inhibit the function of the MntABC transporter system, or manganese uptake can be mediated by additional transporters that function under higher levels of manganese.

The cyanobacterium Synechocystis 6803 is a highly attractive organism for genetic manipulations. It is naturally transformable with exogenous DNA, and the introduced DNA molecules undergo homologous double-reciprocal recombination leading to directed gene replacements (3). In this study, we describe the inactivation of the genes in the mntCAB cluster using such an approach. The phenotypes of the resultant mutant strains were similar to those of BP13, indicating the presence of additional transport systems for manganese in Synechocystis 6803, which function during cellular growth in the presence of relatively high concentrations of manganese. We have also studied the kinetics and the specificity of manganese uptake in these bacterial cells.

EXPERIMENTAL PROCEDURES

The bacterial strains and plasmids are described in Table I. The Synechocystis 6803 strains were grown at 30 °C under 60 microeinsteins·m² s¹ of white light in the BG11 medium (4), for which MnCl₂ was sterilized separately from other components and added to a final concentration of 5 µM. The solid BG11 medium was supplemented with 1.5% (w/v) agar. In the manganese uptake assays, we used the BG11 medium without any added manganese (BG11-Mn). To avoid potential manganese contaminations, divalent cation salts of 99.99% grade (Aldrich) and deionized water were used for the preparation of the BG11-Mn medium. In addition, citric acid was treated with the chelating resin Chelex 100 (Sigma), and ferric ammonium citrate was replaced by iron(III) nitrate (99.99% grade, Aldrich). All glassware for growth of cyanobacteria in manganese-deficient medium were cleaned with 4 N HCl. Growth of cyanobacterial cells was monitored by the measurement of light scattering at 730 nm on a DW2000 spectrophotometer (SLM-Aminco).

Table I. Bacterial strains and plasmids Strain or plasmid Characteristic Source or reference Synechocystis 6803 strains  WT Wild type Pasteur Culture Collection, Paris  AK mntA::Kmr This study  BK  mntB::Kmr This study  CK  mntC::Kmr This study E. coli strain  TG1 supE hsd5 thi (lac-proAB) F[traD36 proAB+ lacq lacZM15] 6 Plasmids  pSL844 4-kb HindIII-BamHI fragment carrying the mntC, mntA and part of mntB genes of Synechocystis 6803 cloned in pUC119 Laboratory collection  pAK Derivative of pSL844, mntA::Kmr replacing mntA This study  pBK Derivative of pSL844, mntB::Kmr replacing mntB This study  pCK Derivative of pSL844, mntC::Kmr replacing mntC This study  pSL791 Derivative of pUC119 plasmid carrying 1.1-kb Kmr gene cartridge from Tn5 Laboratory collection

For Mn²⁺ uptake experiments, Synechocystis 6803 cells, usually maintained on solid BG11 medium, were inoculated into liquid BG11-Mn medium and grown for 30-36 h. These cells were then harvested by centrifugation, washed, and resuspended in fresh BG11-Mn with the addition of different concentrations of MnCl₂ where indicated. After 20-24 h of incubation, the cells were again washed and resuspended in fresh BG11-Mn medium to a final concentration of approximately 4 × 10⁸ cells/ml. ⁵⁴Mn²⁺ (as MnCl₂, DuPont NEN) was added at ~0.4 µCi/ml (0.4 µCi corresponded to ~0.05 nmol of Mn²⁺). After specific time intervals, 50-µl samples were removed, diluted 100-fold in 10 mM MnCl₂ solution, and immediately filtered through nitrocellulose membrane filters (BA85, Schleicher & Schuell). The filters were suspended in scintillation mixture and counted on an LS 5000 TD scintillation counter (Beckman Instruments). The manganese uptake experiments were performed at 28 °C under 30 microeinsteins·m² s¹ of light. The samples were preincubated under such conditions for at least 20 min. 3-(3,4-Dichlorophenyl)-1,1-dimethylurea (DCMU) and salts of divalent cations were added immediately before the addition of ⁵⁴ Mn²⁺ to the incubation medium. Inhibition constants (K_i) for different cations were determined from Dixon plots (5). Each experiment was repeated three to five times.

Synechocystis 6803 cells were transformed to kanamycin resistance (Km^r) as described in Ref. 3. Southern hybridization and other routine DNA manipulations were performed essentially as described in Ref. 6. DNA sequencing was carried out by the dideoxynucleotide chain termination procedure using custom-made oligonucleotides as primers (6).

Enzymes for recombinant DNA work were from New England Biolabs Inc.; modified T7 polymerase (Sequenase) was from U. S. Biochemical Corp.; [-³²P]dCTP for radioactive labeling of DNA fragments and [-³⁵S]dATP for DNA sequencing were from Amersham Corp.

RESULTS

Using gene replacement techniques, we have created the Synechocystis 6803 mutant strains AK, BK, and CK, in which the mntA, mntB and mntC genes encoding the components of an ABC transporter system for manganese were inactivated, respectively (Table I). Toward this goal, the pSL844 plasmid carrying the mntC, mntA, and part of the mntB genes was partially digested with Sau3AI, ligated to a 1.1-kb BamHI-digested (Km^r) gene cartridge from pSL791, and used for the transformation of Escherichia coli strain TG1. From the transformed cells, the plasmids pBK and pCK (with deletions in mntB and mntC genes, respectively, Fig. 1) were isolated, and the exact locations of the Km^r-gene cartridge were determined by sequencing the plasmid DNA molecules. The plasmid pAK (with an insertion in the mntA gene, Fig. 1) was obtained by ligating a 1.1-kb XbaI-digested Km^r-gene cartridge and XbaI-digested pSL844. These plasmids were used for the transformation of wild-type Synechocystis 6803 strain to kanamycin resistance, and the transformants were examined by Southern hybridization analysis to confirm the disruption of each of the genes in the mntCAB cluster. Chromosomal DNA from wild-type and mutant strains was digested with HindIII and used for Southern blotting. A ³²P-labeled DNA fragment containing part of the mntCAB operon hybridized to a 6.5-kb band in the wild-type sample (Fig. 2). Only one band of expected size was observed in each mutant strain, indicating that the Km^r-gene cartridge was inserted in the genome of each mutant strain, and each of the mutations was completely segregated.

All of these mutants had significantly reduced growth rates in a manganese-deficient medium, whereas the growth rate of wild-type cells was not affected significantly (Table II). The addition of 5 µM manganese to this medium increased the growth rates of mutant cells to near normal levels. These data were comparable with earlier results obtained for the BP13 mutant strain (2) and indicated that (i) the products of the mntA, mntB, and mntC genes are indeed involved in manganese transport; (ii) the ABC transporter encoded by the mntCAB operon effectively functions under low levels of manganese and the BG11-Mn medium contained enough residual manganese for the normal growth of wild-type cells; and (iii) manganese can be transported by another transport system in Synechocystis 6803 cells grown in the presence of micromolar concentrations of manganese.

Table II. Growth properties of Synechocystis 6803 wild-type and mutant strains Strain Doubling time in h BG11 BG11-Mn WTa 7.5 8.2 AK 8.5 39  BK 8.7 40.5  CK 8.5 39 a  WT, wild type.

We have investigated the effects of different extracellular manganese concentrations on the rates of ⁵⁴Mn²⁺ accumulation in the Synechocystis 6803 wild-type and CK mutant strains. As shown in Fig. 3, the CK cells grown under manganese starvation conditions had very slow ⁵⁴Mn²⁺ uptake, while the rate of uptake in the cells grown in the medium with 0.5 µM added manganese was significantly higher. In contrast, the wild-type cells grown in a manganese-deficient medium had about a 2-fold higher rate of ⁵⁴Mn²⁺ accumulation than in the medium supplemented with manganese. ⁵⁴Mn²⁺ uptake by the wild-type cells grown in the presence of 0.5-5 µM manganese and by the CK cells grown under a concentration of 2 µM manganese were comparable. Taking together, these data suggested that trace levels of manganese can be transported by two distinctly different systems in these cyanobacterial cells. Among them, the MntABC transporter system is induced by growth in the presence of low concentrations of manganese, whereas higher manganese concentrations repress this system and induce a second manganese transport system. At 0 °C, ⁵⁴Mn²⁺ accumulation was inhibited in cells grown with or without manganese (see Fig. 6), indicating that both systems mediate active transport of manganese.

Regulation of ⁵⁴Mn²⁺ uptake by the wild-type (A) and CK (B) cells, by manganese in the growth media.

Effects of low temperature (0 °C, ), 10 µM DCMU (), and darkness () on ⁵⁴Mn²⁺ uptake by wild-type cells grown without manganese (A) and by CK cells grown in a 1 µM Mn²⁺-containing medium (B).

The initial rate of Mn²⁺ uptake in the wild-type and CK mutant cells grown with or without manganese exhibited biphasic saturation kinetics (Fig. 4A). This result suggested that in Synechocystis 6803, manganese can also be accumulated by a low affinity system. The wild-type and CK cells grown in 2 µM Mn²⁺ showed the same saturation kinetics, while under manganese starvation conditions the saturation rate of Mn²⁺ uptake was somewhat lower in the wild-type strain and significantly lower in the CK strain. In the wild-type cells grown in a manganese-deficient medium, two phases of saturation kinetics were distinctly visible. Moreover, up to a concentration of 7 µM Mn²⁺ (first phase), the rate of Mn²⁺ uptake followed simple Michaelis-Menten kinetics, with a K_m of 1-3 µM and a V_max of 3-8 pmol/min·10⁸ cells (Fig. 4B).

We have studied the effects of several divalent cations on ⁵⁴Mn²⁺ accumulation by Synechocystis 6803 cells. As shown in Fig. 5, Cd²⁺, and the micronutrients Co²⁺ and Zn²⁺ did not affect manganese uptake in the wild-type and CK cells grown in 1 µM Mn²⁺. In contrast, in the wild-type cells grown in a manganese-deficient medium, these cations competitively inhibited manganese accumulation, with K_i = 4-8 µM for Cd²⁺ and 8-15 µM for Co²⁺ and Zn²⁺. Cu²⁺, another micronutrient element, stimulated ⁵⁴Mn²⁺ accumulation by both strains independent of their growth conditions. Addition of 10 µM Fe³⁺ did not inhibit manganese accumulation (data not shown). The macronutrient elements, Mg²⁺ and Ca²⁺, also did not show any significant effect on manganese uptake. Mg²⁺ slightly inhibited and Ca²⁺ slightly increased ⁵⁴Mn²⁺ accumulation for the wild-type cells grown in a manganese-deficient medium and for the CK cells grown in 1 µM Mn²⁺ (data not shown). However, such effects were only observed when the ratio of magnesium or calcium to manganese was as high as 10⁵: 1. 

Effects of different cations on the accumulation of manganese by wild-type cells grown in a manganese-deficient medium (A), and by wild-type (B) and CK (C) cells grown in a 1 µM Mn²⁺ containing medium.

Since manganese is an important element for photosynthetic O₂ evolution, we examined Mn²⁺ uptake rates when photosynthetic activity of the cyanobacterial cells was inhibited. Manganese accumulation was significantly reduced in the darkness in the cells grown with or without manganese (Fig. 6). DCMU, a specific inhibitor of PSII, also inhibited manganese uptake. We have also found that in PSII-deficient mutant strains of Synechocystis 6803, the rate of manganese accumulation was reduced in a manner similar to that in wild-type cells kept in the dark (data not shown). These results suggest that an optimal accumulation of manganese in Synechocystis 6803 depends on its photosynthetic electron transport activity.

DISCUSSION

We have recently described an ABC transporter system for manganese in the cyanobacterium Synechocystis 6803 (2). In the present study, inactivation of genes encoding components of this multisubunit complex has allowed us to demonstrate the presence of a second manganese transport system in these bacterial cells. Based on ⁵⁴Mn²⁺ uptake experiments we have determined that these systems function under different growth conditions. The ABC transporter system was induced in the manganese deficient medium, while the second system, the molecular nature of which has not been established yet, was induced by micromolar levels of manganese. Both systems accumulated trace levels of ⁵⁴Mn²⁺, and such transport was inhibited under low temperature conditions. Taken together, these data indicated that in Synechocystis 6803 cells there are two high affinity active transport systems for manganese that are regulated by different extracellular manganese concentrations.

The saturation kinetics for manganese transport were biphasic, suggesting that in Synechocystis 6803 the transport of different concentrations of manganese occurs through systems with different affinities. Previous reports on manganese transport in other microorganisms have also indicated that manganese can be a low affinity alternative substrate for magnesium uptake systems in E. coli (7) and Salmonella typhimurium (8). For the wild-type cells grown under manganese starvation conditions, we determined that in the first phase of the biphasic saturation profile, the rate of Mn²⁺ uptake follows simple Michaelis-Menten kinetics with K_m of 1-3 µM and V_max of 3-8 pmol/min·10⁸ cells. These parameters defined the characteristics of the ABC-type transporter, MntABC. The presence of such high affinity manganese transport systems has also been described in a number of bacterial species. For example, under low manganese concentrations, manganese uptake in E. coli (9, 10), Bacillus subtilis (11, 12), and Rhodobacter capsulata (13) followed Michaelis-Menten kinetics with a K_m of 0.2-2.4 µM and V_max of 0.02-1.3 µmol/min·g. The K_m was somewhat higher in Staphylococcus aureus (between 6 and 8 µM) (14) while the V_max was significantly higher in Lactobacillus plantarum (23.8 µmol/min·g at K_m = 0.2 µM) (15).

We have found that in Synechocystis 6803, the ABC transporter for manganese was competitively inhibited by Cd²⁺, Co²⁺, and Zn²⁺. In contrast, the manganese uptake system induced by micromolar levels of manganese was highly specific and was not inhibited by any tested cation. These cations are also known to inhibit high affinity manganese transport systems in other bacteria. Cadmium competitively inhibits manganese uptake in B. subtilis with a K_i of 3.4 µM (12), in L. plantarum with a K_i of 0.9 µM (15), and in S. aureus with a K_i of 10 µM (14). Moreover, in B. subtilis, it has been demonstrated that the transport of manganese and cadmium occurs through the same system (12). In E. coli, Cd²⁺ as well as Zn²⁺ are noncompetitive inhibitors of manganese uptake, while Co²⁺ is a competitive inhibitor (K_i = 20-34 µM) (9, 10). Small effects of Co²⁺ on manganese uptake have also been observed in R. capsulata (13). Zn²⁺ slightly inhibits manganese accumulation in B. subtilis (12). Thus, for most bacterial species, Cd²⁺ is a highly specific competitive inhibitor of manganese uptake, and perhaps it is an alternative substrate for most such high affinity manganese transport systems. Co²⁺ and Zn²⁺ may be other possible substrates for these systems in some species.

In all of the bacterial systems studied so far, the macronutrient elements Ca²⁺ and Mg²⁺ do not specifically inhibit high affinity manganese transport. The present study indicates that this is true for the cyanobacterium, Synechocystis 6803. The reason for the enhancement of manganese uptake by Ca²⁺ and Cu²⁺ is currently unknown. However, it is noteworthy that Cu²⁺ increases the rate of manganese uptake by E. coli cells also (10). It has been suggested that copper may bind nonspecifically to the cell surface displacing manganese and raising in such a way the free concentration of manganese available for active transport.

Cheniae and Martin (16) have demonstrated that for the cyanobacterium Anacystis nidulans and the green alga Scenedesmus, two photosynthetic microorganisms, rates of manganese uptake are decreased in cells incubated in darkness and are stimulated by a relatively low intensity of light. DCMU, a specific inhibitor of PSII, also has an inhibitory effect on manganese accumulation in these organisms. We have obtained similar results for Synechocystis 6803. In our experiments, manganese uptake was significantly lower in the darkness. In comparison, the effect of the addition of 10 µM DCMU was less dramatic, although such levels of this inhibitor completely abolished PSII activity. However, we have also found that in PSII-deficient mutant strains of Synechocystis 6803, manganese accumulation was significantly decreased. In our experiments, DCMU was added immediately prior to the measurements of uptake and, perhaps, the cells still had enough residual energy produced by photosynthesis so that Mn2+ transport could occur. Our data indicate that in Synechocystis 6803, manganese uptake depends on active photosynthesis, as well as photosynthetic oxygen evolution depends on manganese accumulation.

In conclusion, we have shown the presence of two high affinity active transport systems for manganese in the cyanobacterium Synechocystis 6803. The first system (the ABC transporter) was induced in manganese-deficient medium and was specifically inhibited by some cations, while the second one was induced in micromolar concentrations of manganese and was highly specific. Further study will be concentrated on the search and study of additional structural and regulatory genes involved in manganese transport in Synechocystis 6803.