Analysis of the arabinose-5-phosphate isomerase of Bacteroides fragilis provides insight into regulation of single-domain arabinose phosphate isomerases

Abstract

Arabinose-5-phosphate isomerases (APIs) catalyze the interconversion of d-ribulose-5-phosphate and D-arabinose-5-phosphate, the first step in the biosynthesis of 3-deoxy-D-manno-octulosonic acid (Kdo), an essential component of the lipopolysaccharide in Gram-negative bacteria. Classical APIs, such as Escherichia coli KdsD, contain a sugar isomerase domain and a tandem cystathionine beta-synthase domain. Despite substantial effort, little is known about structure-function relationships in these APIs. We recently reported an API containing only a sugar isomerase domain. This protein, c3406 from E. coli CFT073, has no known physiological function. In this study, we investigated a putative single-domain API from the anaerobic Gram-negative bacterium Bacteroides fragilis. This putative API (UniProt ID Q5LIW1) is the only protein encoded by the B. fragilis genome with significant identity to any known API, suggesting that it is responsible for lipopolysaccharide biosynthesis in B. fragilis. We tested this hypothesis by preparing recombinant Q5LIW1 protein (here referred to by the UniProt ID Q5LIW1), characterizing its API activity in vitro, and demonstrating that the gene encoding Q5LIW1 (GenBank ID YP_209877.1) was able to complement an API-deficient E. coli strain. We demonstrated that Q5LIW1 is inhibited by cytidine 5'-monophospho-3-deoxy-D-manno-2-octulosonic acid, the final product of the Kdo biosynthesis pathway, with a Ki of 1.91 μM. These results support the assertion that Q5LIW1 is the API that supports lipopolysaccharide biosynthesis in B. fragilis and is subject to feedback regulation by CMP-Kdo. The sugar isomerase domain of E. coli KdsD, lacking the two cystathionine beta-synthase domains, demonstrated API activity and was further characterized. These results suggest that Q5LIW1 may be a suitable system to study API structure-function relationships.

FIG 1

Standard curve from native molecular mass determination by gel filtration chromatography (V_e is the elution volume, and V_o is the void volume).

FIG 2

Effects of divalent metals on the activity of Q5LIW1. The enzyme was incubated as isolated (no additive), with EDTA, or with a divalent metal ion as described in Materials and Methods before being assayed for activity. The activity of the EDTA-incubated sample was assigned a value of 100%.

FIG 3

pH rate profile of Q5LIW1. The rate of A5P isomerization to Ru5P was measured at 37°C in a series of different pH environments. Full experimental details can be found in Materials and Methods.

FIG 4

Complementation of an A5P auxotroph on agar plates. (A) Agar plate containing MOPS medium supplemented with 15 μM A5P, 10 μM G6P, 0.1 mg/ml ampicillin, and 0.05 mg/ml kanamycin. (B) Agar plate containing MOPS medium supplemented with 0.1 mg/ml ampicillin and 0.05 mg/ml kanamycin. In both panels, the wedges were streaked with E. coli TCM15 harboring pT7-7 (vector control, wedge 1), E. coli TCM15 harboring pT7-7-c3406 (c3406 control, wedge 2), and E. coli TCM15 harboring pT7-7-Q5LIW1 (Q5LIW1, wedge 3).

FIG 5

Complementation of an A5P auxotroph on LB agar plates. (A) Agar plate containing LB medium supplemented with 15 μM A5P, 10 μM G6P, 0.1 mg/ml ampicillin, and 0.05 mg/ml kanamycin. (B) Agar plate containing LB medium supplemented with 0.1 mg/ml ampicillin and 0.05 mg/ml kanamycin. In both panels, the wedges were streaked with: E. coli TCM15 harboring pT7-7 (vector control, wedge 1), E. coli TCM15 harboring pT7-7-yrbH (KdsD wild-type control, wedge 2), and E. coli TCM15 harboring pT7-7-KdsDΔ2CBS (KdsD truncated to contain only the SIS domain, wedge 3).

FIG 6

(A) Nonlinear regression fit of the effect of CMP-Kdo on Q5LIW1 API activity. Rates in nM/s were measured at CMP-Kdo concentrations ranging from 0 to 68.51 μM and various A5P concentrations (0.25 mM and 0.5 mM). (B) Hill plot of CMP-Kdo binding to Q5LIW1, showing a 1:1 binding.