In Salmonella typhimurium, thiamine is a required nutrient that is synthesized de novo. Labeling studies have demonstrated probable precursors for both the 4-amino-5-hydroxymethyl-2-methylpyrimidine pyrophosphate moiety and the 4-methyl-5-(-hydroxyethyl) thiazole monophosphate moiety. The isolation of thiamine auxotrophs with mutations in at least five different genetic loci is reported. The majority (22 of 25) of the mutants required only the thiazole moiety of thiamine to satisfy their growth requirement. Most (14 of 25) of the mutants were affected in the thi cluster at min 90 on the S. typhimurium genetic map. Data provided herein indicate that this cluster encodes an operon whose transcription is regulated by thiamine and suggest that thiamine pyrophosphate, or a molecule derived form it, is the effector molecule. Mutants with altered regulation of this operon were isolated, and we propose that they are defective in thiamine phosphate kinase, the product of the thiL gene.The discovery of the alternative pyrimidine biosynthetic (APB) pathway for the synthesis of the pyrimidine moiety of thiamine has raised an interesting question about how the cell coordinates two pathways which are apparently redundant for the synthesis of 4-amino-5-hydroxymethyl-2-methylpyrimidine pyrophosphate (HMP-PP) (13). To address the contribution of the APB pathway to cellular thiamine pools, it is necessary to understand both the biochemistry and the regulation of the other pathway(s) also involved in thiamine biosynthesis.Although thiamine pyrophosphate (TPP) is a cofactor for a number of well-characterized enzymes (e.g., transketolase, pyruvate dehydrogenase, and ␣-ketogluterate dehydrogenase), its synthesis and regulation remain poorly understood. Previous work had identified mutations of Salmonella typhimurium which were thought, on the basis of the nutritional requirement that they caused, to be in loci involved in thiamine biosynthesis. It is important for us to understand these loci both in the context of the work recently done with Escherichia coli (3, 34) and with respect to apb loci involved in alternative pathways for thiamine synthesis.Thiamine monophosphate (TMP) is formed by coupling two precursors, 4-methyl-5-(-hydroxyethyl)thiazole monophosphate (THZ-P) and HMP-PP (Fig. 1). The thiazole moiety is thought to be derived from cysteine, tyrosine, and 1-deoxy-Dthreo-2-pentulose (10, 12, 17), while the pyrimidine moiety (HMP) is thought to be derived from aminoimidazole ribotide (AIR), an intermediate in purine biosynthesis (16, 18) and a proposed intermediate in the APB pathway (13). In E. coli, five kinases involved in TPP formation have been identified, and their genes have been mapped; thiL (min 10) encodes TMP kinase (22); thiK (min 25) encodes thiamine kinase (22); thiM, thiN, and thiD (min 46) encode THZ kinase, HMP kinase, and HMP-P kinase, respectively (22,27,28). Clearly, in E. coli and S. typhimurium, the genetic loci involved in thiamine synthesis are map throughout the chromosome (2, 31). In E. coli, other ...