Cellular compartmentation of two species of delta-aminolevulinic acid synthetase in a facultative photohetero-trophic bacterium (Rps. spheroides Y.)

Regulation of 5-aminolevulinic acid synthesis in Rhodobacter sphaeroides 2.4.1: the genetic basis of mutant H-5 auxotrophy

Rhodobacter sphaeroides H-5 was isolated as a 5-aminolevulinic acid (ALA) auxotroph following treatment of wild-type cells with N-methyl-N-nitroso-N'-nitroguanidine (J. Lascelles and T. Altshuler, J. Bacteriol. 98:721-727, 1969). The existence in R. sphaeroides 2.4.1 of the genes hemA and hemT, each encoding the enzyme 5-aminolevulinic acid synthase (EC 2.3.1.37), raised questions as to the genetic basis for the ALA auxotrophy in mutant H-5. We therefore cloned both the hemA and hemT genes from mutant H-5. The hemA gene has been sequenced in its entirety and bears four base pair substitutions which encode three amino acid changes relative to the sequence of wild-type strain 2.4.1. Complementation analysis of an Escherichia coli ALA auxotroph has revealed that the loss of ALA synthase activity in the HemA mutant enzyme could be localized to two of the amino acid substitutions. On the other hand, the hemT gene from mutant H-5 was able to complement an E. coli mutant requiring ALA for growth. Complementation analyses were also carried out by introducing the cloned hemA or hemT gene of mutant H-5 or wild-type 2.4.1 in trans into H-5 and, in parallel, into our previously described HemA-HemT double mutant strain AT1 (E. L. Neidle and S. Kaplan, J. Bacteriol. 175:2304-2313, 1993). This analysis revealed that while the complementation pattern of mutant AT1 parallels that for the E. coli ALA auxotroph, mutant H-5 could only be complemented by the wild-type hemA gene. The ability of the hemT gene of either mutant H-5 or wild-type 2.4.1 to complement the ALA auxotrophy of mutant AT1 but not mutant H-5 was consistent with beta-galactosidase activities obtained with hemT-lacZ transcriptional fusions. We conclude that the ALA auxotrophy of mutant H-5 arises from (i) a nonfunctional HemA protein containing multiple missense substitutions and (ii) an inability of the normal hemT gene to be expressed in the mutant H-5 genetic background, i.e., an additional mutation of unknown origin is required for hemT expression. These studies bear directly on the regulation of the expression of the hemA and hemT genes of R. sphaeroides 2.4.1.

Expression of the Rhodobacter sphaeroides hemA and hemT genes, encoding two 5-aminolevulinic acid synthase isozymes

The nucleotide sequences of the Rhodobacter sphaeroides hemA and hemT genes, encoding 5-aminolevulinic acid (ALA) synthase isozymes, were determined. ALA synthase catalyzes the condensation of glycine and succinyl coenzyme A, the first and rate-limiting step in tetrapyrrole biosynthesis. The hemA and hemT structural gene sequences were 65% identical to each other, and the deduced HemA and HemT polypeptide sequences were 53% identical, with an additional 16% of aligned amino acids being similar. HemA and HemT were homologous to all characterized ALA synthases, including two human ALA synthase isozymes. In addition, they were evolutionarily related to 7-keto-8-aminopelargonic acid synthetase (BioF) and 2-amino-3-ketobutyrate coenzyme A ligase (Kbl), enzymes which catalyze similar reactions. Two hemA transcripts were identified, both expressed under photosynthetic conditions at levels approximately three times higher than those found under aerobic conditions. A single transcriptional start point was identified for both transcripts, and a consensus sequence at this location indicated that an Fnr-like protein may be involved in the transcriptional regulation of hemA. Transcription of hemT was not detected in wild-type cells under the physiological growth conditions tested. In a mutant strain in which the hemA gene had been inactivated, however, hemT was expressed. In this mutant, hemT transcripts were characterized by Northern (RNA) hybridization, primer extension, and ribonuclease protection techniques. A small open reading frame of unknown function was identified upstream of, and transcribed in the same direction as, hemA.

Cloning and characterization of the 5-aminolevulinate synthase gene(s) from Rhodobacter sphaeroides

The 5-aminolevulinate synthase gene (hemA) from Rhizobium meliloti was used to probe a genomic lambda bank derived from Rhodobacter sphaeroides DNA. Two phage clones were found to bear homology to the Rhizobium probe. Southern hybridization analysis of the two lambda phage clones, which we designated lambda Hem 10 and lambda Hem 12, showed that the homology to the Rhizobium hemA gene was localized to a 3.1-kb SalI fragment derived from lambda Hem 10 and a 7.0-kb SalI fragment derived from lambda Hem 12. Each of the SalI fragments was subsequently cloned into the multiple cloning site of pUC19 in both orientations relative to the lac promoter. Restriction analysis confirmed that each SalI fragment was unique. It was also shown from Southern hybridization analysis that the regions of homology within each of the R. sphaeroides restriction fragments and the Rhizobium probe were different. Further, we have tentatively concluded that each R. sphaeroides hemA gene shows a relatively low degree of homology to the other. Data obtained from in vitro transcription-translation studies in a homologous R. sphaeroides cell-free system, and complementation of hemA mutations of both Escherichia coli and R. sphaeroides by either of the putative hemA clones suggested the presence of a gene encoding 5-aminolevulinate synthase on each DNA sequence. The fact that 5-aminolevulinate synthase activity could be demonstrated in mutant strains complemented in trans with either cloned DNA fragment further supported this conclusion.

Labelling of chlorophylls and precursors by [2-14C]glycine and 2-[1-14C]oxoglutarate in Rhodopseudomonas spheroides and Zea mays. Resolution of the C5 and Shemin pathways of 5-aminolaevulinate biosynthesis by thin-layer radiochromatography

The C-5 of 5-aminolaevulinate, a tetrapyrrole precursor which accumulates when inhibitory laevulinate is present, is derived from either the C-2 of glycine by the 5-aminolaevulinate-synthase-mediated Shemin pathway or the C-1 of 2-oxoglutarate by the C5 pathway. Thin-layer-radiochromatographic procedures are described for determining whether [2-14C]glycine or 2-[1-14C]oxoglutarate labelled the macrocycle of bacteriochlorophyll a, in addition to or rather than the methyl ester or phytyl ester moieties of the side-chains. The method was also used for detecting whether the same substrates label the formaldehyde (C-5) or the succinate (C-1 to C-4) fragments, obtained by periodate cleavage of 5-aminolaevulinate. These methods therefore can readily distinguish between the Shemin and C5 pathways as was demonstrated by using Rhodopseudomonas spheroides and Zea mays (maize), respectively, as examples of each pathway. Both [2-14C]glycine and, to a lesser extent 2-[1-14C]oxoglutarate labelled the macrocycle of bacteriochlorophyll a formed during adaptation of respiring R. spheroides cells to photosynthetic (anaerobic, illuminated) conditions. This and earlier evidence suggested augmentation of the Shemin pathway by a minor C5 pathway contribution. The present studies revealed only Shemin pathway activity: with laevulinate present, [2-14C]glycine formed 5-[5-14C]aminolaevulinate as proved by H14CHO production during periodate cleavage. These methods were sufficiently sensitive also to detect the incorporation of 14CO2, from degradation of either substrate, into 5-aminolaevulinate via the Shemin pathway thus labelling the succinate fragment produced with periodate: this explains bacteriochlorophyll a labelling by 2-[1-14C]oxoglutarate and proves double labelling of 5-aminolaevulinate by [2-14C]glycine. The same techniques were applied to etiolated maize leaves exposed to aerobic illuminated conditions with laevulinate and either 2-[1-14C]oxoglutarate or [2-14C]glycine as substrates. Only the C5 pathway was detected: 2-[1-14C]oxoglutarate was converted to 5-[5-14C]aminolaevulinate, which yielded H14CHO on periodate cleavage. This is not inconsistent with our earlier 13C-NMR studies [Porra, R.J., Klein, O. and Wright, P. E. (1983) Eur. J. Biochem. 130, 509-516] showing that the C5 pathway formed all the 5-aminolaevulinate for chlorophyll biosynthesis.(ABSTRACT TRUNCATED AT 400 WORDS)