The glycine decarboxylase multienzyme complex is composed of four component enzymes, the 100-kD P-protein (a pyridoxal 5-phosphate-dependent amino acid decarboxylase), the 14-kD H-protein (a lipoamide-containing carrier protein), the 45-kD T-protein (a tetrahydrofolate transferase), and the 59-kD L-protein (lipoamide dehydrogenase) (5). The Iipoamide cofactor of the H-protein carries reaction intermediates between the active sites of the other three component proteins. These proteins form a large molecular weight complex (4) within the matrix ofleaf mitochondria where, together with the enzyme serine hydroxymethyltransferase, it catalyzes the photorespiratory conversion of two molecules of glycine to one each of serine, NH3, and CO2 with the reduction of NAD+ to NADH. This complex is the site of photorespiratory CO2 and NH3 release. The glycine decarboxylase complex has been identified in mitochondria from plant and animal sources. Because of the rapid flux of metabolites through the photorespiratory pathway in the light, green leaf mitochondria contain large amounts of the enzyme (in peas the concentration within the mitochondrial matrix exceeds 150 mg/ml) (4). The amount of H-protein is controlled at the transcriptional level in a manner similar to the small subunit of ribulose bisphosphate carboxylase (2, 3) The H-protein cDNA was selected from an Arabidopsis thaliana library in Xgt II by heterologous selection with the cDNA clone for H-protein from peas (2).