Identification of GutQ from Escherichia coli as a D-arabinose 5-phosphate isomerase

Abstract

The glucitol operon (gutAEBDMRQ) of Escherichia coli encodes a phosphoenolpyruvate:sugar phosphotransferase system that metabolizes the hexitol D-glucitol (sorbitol). The functions for all but the last gene, gutQ, have been previously assigned. The high sequence similarity between GutQ and KdsD, a D-arabinose 5-phosphate isomerase (API) from the 3-deoxy-D-manno-octulosonate (KDO)-lipopolysaccharide (LPS) biosynthetic pathway, suggested a putative activity, but its role within the context of the gut operon remained unclear. Accordingly, the enzyme was cloned, overexpressed, and characterized. Recombinant GutQ was shown to indeed be a second copy of API from the E. coli K-12 genome with biochemical properties similar to those of KdsD, catalyzing the reversible aldol-ketol isomerization between D-ribulose 5-phosphate (Ru5P) and D-arabinose 5-phosphate (A5P). Genomic disruptions of each API gene were constructed in E. coli K-12. TCM11[(deltakdsD)] was capable of sustaining essential LPS synthesis at wild-type levels, indicating that GutQ functions as an API inside the cell. The gut operon remained inducible in TCM7[(deltagutQ)], suggesting that GutQ is not directly involved in d-glucitol catabolism. The conditional mutant TCM15[(deltagutQdeltakdsD)] was dependent on exogenous A5P both for LPS synthesis/growth and for upregulation of the gut operon. The phenotype was suppressed by complementation in trans with a plasmid encoding a functional copy of GutQ or by increasing the amount of A5P in the medium. As there is no obvious obligatory role for GutQ in the metabolism of d-glucitol and there is no readily apparent link between D-glucitol metabolism and LPS biosynthesis, it is suggested that A5P is not only a building block for KDO biosynthesis but also may be a regulatory molecule involved in expression of the gut operon.

FIG. 1.

The gut operon of E. coli K-12 MG1655. Genes that are highly conserved in both gram-negative and gram-positive bacterial gut operons are shaded dark gray while those that are specific to gram-negative bacteria are shaded light gray.

FIG. 2.

The effect of GutQ on LPS biosynthesis and d-glucitol utilization. (A) Silver-stained Tricine SDS-PAGE LPS gels of proteinase K-treated whole-cell lysates from BW30270, TCM7, and TCM11. Equal amounts of bacterial cells growing in minimal medium (0.2% glycerol) with (+) or without (−) d-glucitol (10 mM) were harvested in early log phase and processed as described in Materials and Methods. (B) Diauxic growth curves for BW30270 (□), TCM7 (○), and TCM3 (▵). Overnight cultures grown in M9 minimal medium supplemented with 1 μg/ml thiamine and 10 mM d-glucose were diluted into fresh medium with 2 mM d-glucose and 2 mM d-glucitol as dual carbon sources. Cell growth was monitored by measuring the turbidity at 600 nm.

FIG. 3.

Growth and LPS synthesis in the ΔAPI strain TCM15. (A) Growth curve of E. coli TCM15 in MOPS minimal medium with thiamine (1 μg/ml) and glycerol (0.1%) as sole carbon source. Sugar phosphates were supplemented in the medium with either 10 μM G6P (▵), 15 μM A5P (□), or both (○). (B) Correlation of LPS with concentration of A5P. A stationary-phase culture grown in MOPS minimal medium (0.2% glycerol, 5 μM A5P, 10 μM G6P) that had ceased dividing was diluted into fresh medium containing G6P (10 μM) and various concentrations of A5P (0.1, 1, 10, 50, and 100 μM) and shaken for 6 h. LPS samples were prepared from the same number of cells based on optical density at 600 nm and analyzed by Tricine SDS-PAGE and silver staining. (C and D) LPS Tricine SDS-PAGE (C) and qualitative RT-PCR transcript analysis (D) of gutD in samples prepared from wild-type BW30270 (lanes 1 and 2) and TCM15 (lanes 3 and 4). TCM15 was preinduced with 10 μM G6P and 5 μM A5P, pelleted, resuspended in fresh MOPS minimal medium (0.2% glycerol) with only A5P and 10 mM d-glucitol as indicated, and shaken for an additional 4 h before harvesting for analyses. Lane S, 0.1- to 1-kb DNA molecular size markers; lane Con, genomic DNA as template.