Glutathione reductase from human erythrocytes. Isolation of the enzyme and sequence analysis of the redox-active peptide.

Abstract

1. From 90 l erythrocytes 200 mg glutathione reductase was purified in four steps with an overall yield of 40%. The specific activity rose from 3.71 U glutathione reductase per g hemoglobin in the erythrocytes to 235 U/mg protein in the isolated enzyme. The purification procedure which was completed in 5 days, included (a) denaturation of hemoglobin by a mixture of butan‐1‐ol and chloroform [Scott, E. M. (1976) Prep. Biochem. 6, 147–152], (b) precipitation of glutathione reductase with acetone, (c) extraction of the precipitate and fractionation of the extract with ammonium sulfate, (d) affinity chromatography on adenosine‐2′,5′‐bisphosphate–Sepharose 4B.2. The 280/460 nm absorbance ratio was found to be 5.9 for the purified enzyme; all other studies on physical and chemical properties of the enzyme confirmed the results of Worthington and Rosemeyer [Eur. J. Biochem. 67, 231–238 (1976) and earlier papers].3. The new purification procedure provided crystals suitable for X‐ray diffraction analysis at high resolution. Furthermore sufficient amounts for sequencing the enzyme were isolated. As a first step in the determination of the protein's primary structure, CNBr‐produced fragments of the carboxymethylated enzyme were fractionated.4. A chain segment containing the catalytic disulfide of the native protein was isolated. Its sequence was determined to be . An identical sequence is present in yeast glutathione reductase, and lipoamide dehydrogenases contain an active site peptide with very similar sequence [Williams, C. H. Jr(1976) The Enzymes 13, 89–173]. Thus the long‐recognized mechanistic similarities between glutathione reductases and lipoamide dehydrogenases are reinforced by sequence similarities around the two redox‐active cysteine residues. The sequenced regions are, however, too short to allow speculations on phylogenetic relationships within this category of proteins.