A non-carboxylating pentose bisphosphate pathway in halophilic archaea

Abstract

Bacteria and Eucarya utilize the non-oxidative pentose phosphate pathway to direct the ribose moieties of nucleosides to central carbon metabolism. Many archaea do not possess this pathway, and instead, Thermococcales utilize a pentose bisphosphate pathway involving ribose-1,5-bisphosphate (R15P) isomerase and ribulose-1,5-bisphosphate (RuBP) carboxylase/oxygenase (Rubisco). Intriguingly, multiple genomes from halophilic archaea seem only to harbor R15P isomerase, and do not harbor Rubisco. In this study, we identify a previously unrecognized nucleoside degradation pathway in halophilic archaea, composed of guanosine phosphorylase, ATP-dependent ribose-1-phosphate kinase, R15P isomerase, RuBP phosphatase, ribulose-1-phosphate aldolase, and glycolaldehyde reductase. The pathway converts the ribose moiety of guanosine to dihydroxyacetone phosphate and ethylene glycol. Although the metabolic route from guanosine to RuBP via R15P is similar to that of the pentose bisphosphate pathway in Thermococcales, the downstream route does not utilize Rubisco and is unique to halophilic archaea.

Fig. 1. Schematic drawing of archaeal nucleoside metabolic pathways in Thermococcus and in halophilic archaea including Halobacterium.

a The classical pentose bisphosphate pathway identified in Thermococcus is predicted to be involved in the nucleoside degradation and/or the conversion of nucleosides to NMPs. The NMP shunt, composed of NMP phosphorylase, R15P isomerase, and Rubisco, degrades NMPs to 3-PGA. b The non-carboxylating pentose bisphosphate pathway proposed in this study is shown. Red arrows indicate reactions specific to the pathway identified in this study and absent in the classical pentose bisphosphate pathway. The locus tags encoding the enzymes whose recombinant proteins were examined in this study are shown. Locus tags with parenthesis are genes predicted to encode enzymes whose activities were detected in the cell-free extract from H. salinarum. Regarding VNG_6270G, both native and recombinant enzymes were investigated. NMP nucleoside-5’-monophosphate, R1P ribose-1-phosphate, R15P ribose-1,5-bisphosphate, RuBP ribulose-1,5-bisphosphate, 3-PGA 3-phosphoglycerate, Ru1P ribulose-1-phosphate, DHAP dihydroxyacetone phosphate.

Fig. 2. Enzyme activity analyses of Ht-R15P isomerase and Hx-RbsK proteins with HPLC.

a In the presence or absence of Ht-R15P isomerase enzyme, the generation of RuBP from R15P was investigated. Black, pink, blue, and green lines indicate the reaction product without the enzyme, that with the enzyme, 20 mM R15P standard compound, and 10 mM RuBP standard compound, respectively. b In the presence or absence of the Hx-RbsK protein, the generation of R15P from R1P was examined. Black, pink, blue, and green lines indicate the reaction product without the enzyme, that with the enzyme, 10 mM R1P standard compound, and 20 mM R15P standard compound, respectively. c Conversion of R1P with the Hx-RbsK and Ht-R15P isomerase proteins was examined. After the kinase reaction by Hx-RbsK, the isomerase reaction by Ht-R15P isomerase was carried out. The reaction product was analyzed by HPLC. Pink and black lines indicate reaction products with and without Ht-R15P isomerase in the second reaction, respectively. Compounds separated with a column were monitored with a differential refractive index detector.

Fig. 3. Schematic diagram of gene arrangement on the eight haloarchaeal genomes.

Color codes of relevant genes are indicated in the figure. White and gray-arrowed boxes are genes most likely forming operons with the focused five genes. In the gray-arrowed boxes, the gene length is not reflected to the width of each arrowed box. Black arrows indicate predicted operons.

Fig. 4. Enzyme activity analyses of Hl-Urdpase1, Hx-HAD hydrolase, and Hx-FucA proteins.

a Nucleoside phosphorylase activity of Hl-Urdpase1 was analyzed toward six nucleosides by quantifying the released nucleobases with HPLC. b Phosphatase activity of Hx-HAD hydrolase was investigated toward eleven sugar phosphates and pNPP by quantifying released phosphate with malachite green. G1P glucose-1-phosphate, G6P glucose-6-phosphate, G16P glucose-1,6-bisphosphate, F1P fructose-1-phosphate, F6P fructose-6-phosphate, F16P fructose-1,6-bisphosphate, R5P ribose-5-phosphate, Ru5P ribulose-5-phosphate, pNPP p-nitrophenylphosphate. c Aldolase activity of the Hx-FucA condensing DHAP and six aldehydes was examined by quantifying the residual DHAP after reactions with a coupling enzyme. The activities were calculated from n = 3 independent experiments. Error bars indicate standard deviations.

Fig. 5. Nucleoside degradation networks in halophilic archaea.

The results obtained in this study and the distribution of relevant genes imply the occurrence of three types of nucleoside degradation pathways in halophilic archaea. The three metabolic routes are with NMP phosphorylase and Rubisco (a), with Rubisco but without NMP phosphorylase (b), without Rubisco or NMP phosphorylase but with RuBP phosphatase and Ru1P aldolase (c).

Fig. 6. Relationship between the presence of the three nucleoside metabolic pathways and phylogenetic positions of halophilic archaea.

The phylogenetic tree was constructed using 16S rRNA gene sequences. A sequence from Thermoplasma volcanium (tvo) was used as an outgroup. The colors of the circles correspond to those of the nucleoside metabolic pathways in Fig. 5. Organisms shown in red and pink codes indicate the halophilic archaea shown in Fig. 3. The organisms in red codes show those whose proteins were actually examined in this study.