Glutamate 1-semialdehyde aminotransferase (glutamate 1-semialdehyde 2,1-aminomutase; EC 5.4.3.8; GSA-AT) catalyzes the transfer of the amino group on carbon 2 of glutamate 1-semialdehyde (GSA) to the neighboring carbon 1 to form delta-aminolevulinic acid (ALA). To gain insight into the mechanism of this enzyme, possible intermediates were tested with purified enzyme and the reaction sequence was followed spectroscopically. While 4,5-dioxovaleric acid (DOVA) was efficiently converted to ALA by the pyridoxamine 5'-phosphate (PMP) form of the enzyme, 4,5-diaminovaleric acid (DAVA) was a substrate for the pyridoxal 5'-phosphate (PLP) form of GSA-AT. Thus, both substances are reaction intermediates. The purified enzyme showed an absorption spectrum with a peak around 338 nm. Addition of PLP led to increased absorption at 338 nm and a new peak around 438 nm. Incubation of the purified enzyme with PMP resulted in an additional absorption peak at 350 nm. The reaction of the PLP and PMP form of the enzyme with GSA allowed the detection of a series of peaks which varied in their intensities in a time-dependent manner. The most drastic changes to the spectrum that were observed during the reaction sequence were at 495 and 540 nm. Some of the detected absorption bands during GSA-AT catalysis were previously described for several other aminotransferases, indicating the relationship of the mechanisms. The reaction of the PMP form of the enzyme with DOVA resulted in a similar spectrum as described above, while the spectrum for the conversion of DAVA by the PLP form of the enzyme indicated a different mechanism. To understand the functional significance of a conserved lysyl residue found in all cloned GSA-ATs, lysine 265 of Escherichia coli GSA-AT was changed to arginine by oligonucleotide-directed mutagenesis. The mutant enzyme K265R was overexpressed, purified to apparent homogeneity, and analyzed for its structural, catalytic, and spectroscopic properties. The enzymatic activity of K265R was only 2% of the wild-type enzyme activity, while its dimeric structure was not influenced by the mutation. The enzyme activity was stimulated by the addition of exogenous amines such as ethanolamine and methylamine. The spectrum of purified K265R showed significant absorbance only at 280 nm, indicating the absence of bound cofactor. The addition of PLP to K265R in the presence of ethanolamine prompted the formation of a peak at 438 nm, which was subsequently converted into a new peak at 338 nm. Further addition of GSA led to the conversion of the 338-nm peak into a peak at 305 nm. The nature of the catalysis performed by the mutant enzyme is different from the activity of the wild-type enzyme. To test the loss of GSA-AT function in vivo, the hemL gene, encoding GSA-AT, was disrupted by insertion of the Tc(r) gene into its coding region. This E. coli strain has leaky ALA auxotrophy, indicating the presence of a compensatory pathway for ALA formation in E. coli. Transformation of the mutant strain with the wild-type gene restored normal growth, while transformation with the mutant gene encoding K265R resulted in drastically reduced growth but still differed from the control strain containing an empty plasmid.
