Potassium transport in a halophilic member of the bacteria domain: identification and characterization of the K+ uptake systems TrkH and TrkI from Halomonas elongata DSM 2581T

Abstract

The halophilic bacterium Halomonas elongata accumulates K+, glutamate, and the compatible solute ectoine as osmoprotectants. By functional complementation of Escherichia coli mutants defective in K+ uptake, we cloned three genes that are required for K+ uptake in H. elongata. Two adjacent genes, named trkA (1,374 bp) and trkH (1,449 bp), were identified on an 8.5-kb DNA fragment, while a third gene, called trkI (1,479 bp), located at a different site in the H. elongata chromosome, was found on a second 8.5-kb fragment. The potential protein expressed by trkA is similar to the cytoplasmic NAD+/NADH binding protein TrkA from E. coli, which is required for the activity of the Trk K+ uptake system. The deduced amino acid sequences of trkH and trkI showed significant identity to the transmembrane protein of Trk transporters. K+ transport experiments with DeltatrkH and DeltatrkI mutants of H. elongata revealed that TrkI exhibits a Km value of 1.12 mM, while the TrkH system has a half-saturation constant of 3.36 mM. Strain KB12, relying on TrkH alone, accumulated K+ with a lower Vmax and required a higher K+ concentration for growth in highly saline medium than the wild type. Strain KB15, expressing only TrkI, showed the same phenotype and the same K+ transport kinetics as the wild type, proving that TrkI is the main K+ transport system in H. elongata. In the absence of both transporters TrkH and TrkI, K+ accumulation was not detectable. K+ transport was also abolished in a trkA deletion mutant, indicating that TrkI and TrkH depend on one type of TrkA protein. Reverse transcriptase PCR experiments and Northern hybridization analyses of the trkAH locus revealed cotranscription of trkAH as well as a monocistronic transcript with only trkA.

FIG. 1.

Gene organization at the trkAH and trkI loci of H. elongata. The relevance of the trk genes for K⁺ uptake was determined by K⁺ transport experiments (see Fig. 4) and growth experiments with minimal medium under K⁺ limitation (see Fig. 3). +, K⁺ transport and growth like that of the wild type; ±, reduced K⁺ uptake and growth; −, no K⁺ uptake via TrkH and TrkI, no growth under K⁺ limitation in mineral salt medium. Open reading frame orf1 upstream of trkAH encodes a potential RNA methyltransferase; orf2 downstream of trkH as well as orf3 and orf4 adjacent to trkI are of unknown function. P, putative promoter sequence; S, putative stem-loop sequence. Sequence analysis did not reveal promoter or stem-loop sequences for trkAH.

FIG. 2.

(A) Growth of H. elongata wild-type (WT) and trk deletion mutants KB12 (ΔtrkI) and KB15 (ΔtrkH) on minimal medium at different salinities (0.51, 1.03, and 2.05 M NaCl) and low (0.1 mM) and high (5 mM) K⁺ concentrations. Growth was scored after 3 days (0.51 and 1.03 M NaCl) and 4 days (2.05 M NaCl) of incubation at 30°C. Mutant KB15 grew similarly to the wild type in low and high K⁺ at all salinities, while growth of KB12 accumulating K⁺ only via TrkH was hampered in low-K⁺ medium, especially at high salinity. (B) Growth of H. elongata wild type, trkA deletion mutant KB14, and KB12.2 (ΔtrkH ΔtrkI) on minimal medium at 1.03 M NaCl and different K⁺ concentrations (0.25 and 200 mM). For (B), growth was scored after 3 days' incubation at 30°C.

FIG. 3.

Kinetic analysis of K⁺ transport via TrkH and TrkI in H. elongata wild type and strains KB12 (ΔtrkI), KB15 (ΔtrkH), KB12.2 (ΔtrkH ΔtrkI) and KB14 (ΔtrkA). Osmolarity of growth medium and transport buffer was adjusted by 0.51 M NaCl. Transport was started by adding potassium to the assay. (A) Lowering the K⁺ concentration from 20 mM to 500 μM led to a decreasing transport rate. The K⁺ uptake data were fitted by nonlinear regression. The curve for K⁺ transport via TrkI into KB15 (•) and transport by the wild type (○) were fitted best by the Michaelis-Menten model and showed K_m values of 1.12 and 1.18 mM K⁺, respectively. The transport data for TrkH in strain KB12 (▪) showed a sigmoidal dependence of transport rate versus K⁺ concentration. The half-saturation constant was determined to be 3.36 mM. Deletion of trkI and trkH abolished transport activity in strain KB12.2 (▴), proving that transporters TrkH and TrkI are responsible for the observed uptake of K⁺. No transport activity was measured for the ΔtrkA deletion mutant KB14 (♦), proving that trkA encodes the potential NAD binding protein for both TrkH and TrkI. Error bars indicate standard deviations. (B) Hill plot based on transport data from TrkI (•, KB15) and TrkH (▪, KB12) to evaluate the Hill coefficient h, which was estimated to be 0.98 for TrkI and 2.1 for TrkH. Values of h greater than 1 indicate a deviation of the Michaelis-Menten kinetics and can be used as an index of enzyme cooperativity (see Discussion). Symbols are as defined in the legend to panel A.

FIG. 4.

Analysis of the transcriptional organization of the trkAH gene cluster by Northern hybridization experiments and RT-PCR. (A) Northern blot analysis of total RNA from H. elongata wild-type (WT) and the trkA deletion mutant KB14 using a trkA-specific RNA probe. Total RNA was isolated from cells grown in minimal medium containing 510 mM NaCl and 200 mM KCl. RNA was electrophoretically separated on agarose gels, blotted onto a nylon membrane, and hybridized with a DIG-labeled RNA probe. A single transcript of approximately 1.4 kb corresponding to the calculated size of a trkA transcript was detected by hybridization with the WT RNA, while no hybridization signal was detectable with the RNA of control strain KB14 (ΔtrkA). There were no transcripts detectable using a trkH-specific probe in similar experiments. (B) RT-PCR analysis of the trkAH locus proving the existence of a common trkAH transcript. cDNA was synthesized by using a reverse primer binding to trkH mRNA. trkAH cDNA was amplified by PCR with the help of forward primer 2trk2 binding at the 3′ end of trkA and reverse primer trkAup binding at the 5′ region of trkH. A 550-bp PCR product was separated by agarose gel electrophoresis (I), which matched the size of the calculated trkAH PCR product (554 bp). No DNA was amplified from purified total RNA prior to cDNA synthesis, proving that the trkAH product was indeed amplified from cDNA and not from contaminating chromosomal DNA (II). M, DNA size marker (bp).