In microorganisms, menaquinone is an obligatory component of the electron-transfer pathway. It is derived from chorismate by seven enzymes in Escherichia coli. However, a bioinformatic analysis of whole genome sequences has suggested that some microorganisms, including pathogenic species such as Helicobacter pylori and Campylobacter jejuni, do not have orthologs of the men genes, even though they synthesize menaquinone. We deduced the outline of this alternative pathway in a nonpathogenic strain of Streptomyces by bioinformatic screening, gene knockouts, shotgun cloning with isolated mutants, and in vitro studies with recombinant enzymes. As humans and commensal intestinal bacteria, including lactobacilli, lack this pathway, it represents an attractive target for the development of chemotherapeutics.

We recently demonstrated that the futalosine pathway was operating in some bacteria for the biosynthesis of menaquinone and that futalosine was converted into dehypoxanthinyl futalosine (DHFL) by an MqnB of Thermus thermophilus. In this study, we found that aminodeoxyfutalosine, which has adenine instead of hypoxanthine in futalosine, was directly converted into DHFL by an MqnB of Helicobacter pylori. Therefore, this step is potentially an attractive target for the development of specific anti-H. pylori drugs.In prokaryotes, menaquinone (MK) is used for respiration. In Escherichia coli, MK is biosynthesized from chorismate by eight enzymes (4). However, we recently identified an alternative pathway (the futalosine pathway shown in Fig. 1) (2, 3, 5), which operates in bacteria, including some pathogens, such as Helicobacter pylori, Campylobacter jejuni, Chlamydia trachomatis, and Leptospira borgpetersenii. Disruption of this pathway in Streptomyces coelicolor has been shown to lead to the inhibition of bacteriostatic growth. We also showed that the Mqn genedisrupted S. coelicolor mutants required more than 100 g/ml menaquinone for their growth. Because we do not consume a diet containing this much menaquinone, the essential amount of MK is not supplied from our diet if the futalosine pathway is blocked. Moreover, humans and commensal intestinal bacteria, including lactobacilli, lack the futalosine pathway. Taking these results together, we think that the futalosine pathway is a potential attractive target for the development of specific anti-H. pylori drugs. However, the details of each of the biosynthetic steps in the futalosine pathway remain unclear. In this study, we investigated the second step of the futalosine pathway, which is the conversion of futalosine (FL) into dehypoxanthinyl futalosine (DHFL) by MqnB (futalosine hydrolase [EC 3.2.2.26]) (3), in Acidothermus cellulolyticus, H. pylori, and S. coelicolor. We found that aminodeoxyfutalosine (AFL), which has adenine instead of hypoxanthine in FL, was an intermediate in these microorganisms.To ascertain the details of each step in the futalosine pathway, we selected the second step catalyzed by MqnB, since FL, the substrate of MqnB, was the only compound that we could readily prepare from the culture broth of the MqnB-disrupted mutant (2). We first investigated the distribution of the futalosine pathway and the organization of the Mqn genes in microorganisms whose genomes had been completely sequenced.In almost all microorganisms, the Mqn genes were scattered throughout the genome. However, in A. cellulolyticus, a thermophilic actinobacterium, the Mqn genes, and other menaquinone-related genes, such as prenylation and methylation genes, were clustered in two loci (1). One cluster contained Acel_0105 and Acel_0106 genes, which encoded orthologs of MqnC and MqnB, respectively. The other was composed of 11 genes (Acel_0255 to Acel_0266) encoding MqnA (Acel_0261) and MqnC (Acel_0263) orthologs. To determine whether Acel_0106 did indeed encode MqnB, we prepare...