Uroporphyrinogen III is the committed intermediate common to heme and siroheme biosynthesis in E. coli. Uroporphyrinogen III decarboxylase is the first enzyme at the branch point which commits to heme synthesis. A hemin-permeable hemA mutant which could grow on 5-aminolevulinic acid (ALA) or hemin, was mutagenized to give a double mutant, 10L2-1. The second mutation which was identified as hemE because it was mapped to 90.1 min. by F' and Hfr mapping and P1 transduction, accumulated uroporphyrin and had no uroporphyrinogen decarboxylase activity. This mutation could be complemented with a plasmid harboring the hemE gene of Synechococcus. The complemented strain could grow on ALA and accumulated coproporphyrin and protoporphyrin but not uroporphyrin. The E. coli hemE gene was cloned by transducing 10L2-1 with an E. coli genomic library in lambda gt11. hemE with upstream regions of various sizes was cloned in front of a promoterless CAT gene. Good growth on chloramphenicol (25-75 micrograms/ml) depended on a promoter within 152 bp upstream of the hemE structural gene start of translation site. In addition, this construct could complement the hemE requirement of 10L2-1 as well as allow it to grow on chloramphenicol. Addition of hemin did not inhibit this growth and therefore it appears that it does not affect the hemE promoter. The hemE structural gene alone allowed good growth on 10 micrograms/ml but poor growth on 25 micrograms/ml chloramphenicol, suggesting that there is a weak promoter within hemE for a downstream ORF. Quantitation of CAT protein in these strains showed a weak promoter within hemE, a promoter 152 bp upstream of hemE and another promoter within 1.3 kb upstream of hemE. The 1.3 kb region contains an ORF 40 bp upstream of hemE, thus suggesting that hemE is part of an operon.