Escherichia coli phnN, encoding ribose 1,5-bisphosphokinase activity (phosphoribosyl diphosphate forming): dual role in phosphonate degradation and NAD biosynthesis pathways

Abstract

An enzymatic pathway for synthesis of 5-phospho-D-ribosyl alpha-1-diphosphate (PRPP) without the participation of PRPP synthase was analyzed in Escherichia coli. This pathway was revealed by selection for suppression of the NAD requirement of strains with a deletion of the prs gene, the gene encoding PRPP synthase (B. Hove-Jensen, J. Bacteriol. 178:714-722, 1996). The new pathway requires three enzymes: phosphopentomutase, ribose 1-phosphokinase, and ribose 1,5-bisphosphokinase. The latter activity is encoded by phnN; the product of this gene is required for phosphonate degradation, but its enzymatic activity has not been determined previously. The reaction sequence is ribose 5-phosphate --> ribose 1-phosphate --> ribose 1,5-bisphosphate --> PRPP. Alternatively, the synthesis of ribose 1-phosphate in the first step, catalyzed by phosphopentomutase, can proceed via phosphorolysis of a nucleoside, as follows: guanosine + P(i) --> guanine + ribose 1-phosphate. The ribose 1,5-bisphosphokinase-catalyzed phosphorylation of ribose 1,5-bisphosphate is a novel reaction and represents the first assignment of a specific chemical reaction to a polypeptide required for cleavage of a carbon-phosphorus (C-P) bond by a C-P lyase. The phnN gene was manipulated in vitro to encode a variant of ribose 1,5-bisphosphokinase with a tail consisting of six histidine residues at the carboxy-terminal end. PhnN was purified almost to homogeneity and characterized. The enzyme accepted ATP but not GTP as a phosphoryl donor, and it used ribose 1,5-bisphosphate but not ribose, ribose 1-phosphate, or ribose 5-phosphate as a phosphoryl acceptor. The identity of the reaction product as PRPP was confirmed by coupling the ribose 1,5-bisphosphokinase activity to the activity of xanthine phosphoribosyltransferase in the presence of xanthine, which resulted in the formation of 5'-XMP, and by cochromatography of the reaction product with authentic PRPP.

FIG. 1.

NAD metabolism of E. coli. The de novo and salvage pathways are shown. Abbreviations: Asp, aspartate; DHAP, dihydroxyacetone phosphate; Gln, glutamine; Glu, glutamate; NA, nicotinate; NAm, nicotinamide; NAMN, nicotinate mononucleotide; NAAD; nicotinate adenine dinucleotide; NMN, nicotinamide mononucleotide; QA, quinolinate; Rib-5P, ribose 5-phosphate. Relevant enzyme-catalyzed reactions are indicated by gene designations (nadA, quinolinate synthase; nadC, quinolinate phosphoribosyltransferase; pncB, nicotinate phosphoribosyltransferase).

FIG. 2.

Identification of PRPP as the product of the PhnN-catalyzed reaction. Incubation and one-dimensional chromatography were performed as described in Materials and Methods. Abbreviations: Appl., application line; F, solvent front. (A) Incubation of PhnN with various ribose compounds and [γ-³²P]ATP. All of the incubation mixtures contained [γ-³²P]ATP. Lanes 1 and 6 to 9 contained PhnN, whereas PhnN was omitted from lanes 2 to 5. Other additions were as follows: lane 1, no ribose compound; lanes 2 and 6, ribose; lanes 3 and 7, ribose 1-phosphate; lanes 4 and 8, ribose 5-phosphate; and lanes 5 and 9, ribose 1,5-bisphosphate. The positions of ATP, PRPP, ADP, and P_i are indicated on the right. In the reaction mixture shown in lane 9 a small amount of ADP was formed in addition to PRPP, presumably due to the presence of traces of [β-³²P]ATP in the [γ-³²P]ATP preparation. (B) Consumption of ³²P-labeled PhnN reaction product (lanes 1 to 5) or PRPP synthase-generated [³²P]PRPP (lanes 6 to 10) by XPRTase. Production of labeled compounds and incubation with XPRTase were performed as described in Materials and Methods. The reaction mixture analyzed in each lane contained reaction cocktail with the following supplement(s): lanes 1 and 6, no supplement; lanes 2 and 7, xanthine; lanes 3 and 8, XPRTase; lanes 4 and 9, xanthine and XPRTase; lanes 5 and 10, xanthine, XPRTase, and unlabeled PRPP (5 mM). The positions of ATP, PRPP, and P_i are indicated on the right. Lanes 1 to 5 contained appreciable amounts of ATP, whereas no ATP was present in lanes 6 to 10. We were unable to identify conditions that allowed complete consumption of ATP by PhnN, whereas ATP was readily consumed by PRPP synthase. (C) Conversion of unlabeled PhnN reaction product or PRPP to 5′-[¹⁴C]XMP in the presence of XPRTase and [¹⁴C]xanthine. The assay conditions are described in Materials and Methods. The reaction mixture analyzed in each lane contained reaction cocktail consisting of [¹⁴C]xanthine with the following addition(s): lane 1, no addition; lane 2, XPRTase; lane 3, PhnN reaction product; lane 4, PRPP (synthesized by PRPP synthase); lane 5, XPRTase and PhnN reaction product; lane 6, XPRTase and PRPP (synthesized by PRPP synthase); lane 7, XPRTase and commercial PRPP (5 mM); lane 8, XPRTase and a sample of PhnN assay mixture to which no PhnN enzyme was added (the mixture contained ribose 1,5-bisphosphate and ATP); and lane 9, XPRTase and a sample of PRPP synthase assay to which no PRPP synthase was added (the mixture contained ribose 5-phosphate and ATP). The positions of xanthine and 5′-XMP are indicated on the right. Spots were visualized under a UV mineral lamp at 254 nm.

FIG. 3.

Purification and activity of histidine-tailed ribose 1,5-bisphosphokinase. Cells of strain HO1088 (Δprs)/pTR553 (specifying histidine-tailed PhnN) were grown in NZY medium, phnN gene expression was induced by IPTG, and ribose 1,5-bisphosphokinase was purified and assayed as described in Materials and Methods. In addition, a culture sample was removed before addition of IPTG. Furthermore, strain HO1088/pHO500 (specifying wild-type PhnN) was grown in parallel with strain HO1088/pTR553. Extracts and purified enzyme were subjected to gel electrophoresis in 17.5% polyacrylamide containing sodium dodecyl sulfate. Lane 1, crude extract (10 μg of protein) of HO1088/pTR553 cells before IPTG addition; lane 2, crude extract (10 μg) of HO1088/pTR553 cells after IPTG addition; lane 3, purified PhnN protein (1.3 μg); lane 4, molecular weight markers (positions indicated on the right, as follows: a, phosphorylase b [molecular weight, 97,400]; b, serum albumin [66,200]; c, ovalbumin [45,000]; d, carbonic anhydrase [31,000]; e, trypsin inhibitor [21,500]; f, lysozyme [14,400]); lane 5, crude extract (10 μg) of HO1088/pHO500 cells after IPTG addition. The arrow on the left indicates the position of the PhnN band. The ribose 1,5-bisphosphokinase activity of each fraction (in micromoles per minute per milligram of protein) is indicated at the bottom.

FIG. 4.

Proposed pathway for conversion of ribose 5-phosphate to the 5′-phosphoribosyl moiety of NAMN. Abbreviations: P, phosphate; RPK, ribose 1-phosphokinase. Other enzyme-catalyzed reactions are indicated by gene designations (deoB, phosphoribomutase; xapA, xanthosine phosphorylase; phnN, ribose 1,5-bisphosphokinase activity; nadC, quinolinate phosphoribosyltransferase; pncB, nicotinate phosphoribosyltransferase.)

FIG. 5.

Hypothetical reactions involved in the degradation of methylphosphonate by the C-P lyase pathway. An intermediate in the pathway is 5-phospho-α-1-(methylphosphono)ribose. This compound may be either demethylated to ribose 1,5-bisphosphate or phosphorylated to methylphosphono-PRPP. Ribose 1,5-bisphosphate or methylphosphono-PRPP may be phosphorylated or demethylated, which results in the formation of PRPP.