By S. K. Wolfson, Jr., Jean A. Spencer, R. L. Sterkel, and H. G. Williams-Ashman Ben May Laboratory for Cancer Research and Departments of Medicine and Biochemistry, University of Chicago, Chicago, III.

The pyridine nucleotides were discovered and characterized by Otto Warburg,1 who devised the elegant and sensitive optical methods for measurement of their reduction by specific dehydrogenase systems that have received such wide application in biochemical analysis. In his classic study of the increased serum aldolase activity of tumor-bearing animals, Warburg2 was the first to show that mammalian blood serum contains a number of glycolytic enzymes including one that uses a pyridine nucleotide as a hydrogen acceptor, namely, lactic dehydrogenase (LD). Since that time two other diphosphopyridine nucleotide (DPN)-linked enzymes (malic3 and a-glycerophosphate4 dehydrogenases) have been demonstrated in mammalian extracellular fluids, and changes in serum LD and malic dehydrogenase levels in many disease states have been investigated extensively.5-10

Our interest in the oxidizing enzymes of blood serum arose from attempts to study enzymes participating in specific chemical functions of mammary tissue in relationship to carcinoma of the breast. The synthesis of lactose is a biochemical event peculiar to the mammary gland, and there is evidence that lactose is manufactured therein from blood glucose by a series of enzymatic reactions in which uridine diphosphoglycosyl coenzymes are reactants.11, 12 Attention was directed at first toward uridine diphosphoglucose (UDPG) pyrophosphorylase, which catalyzes the reversible synthesis of UDPG from glucose-l-phosphate and uridine triphosphate. The presence of this enzyme in mammary tissue13 and in erythrocytes14 is well established. The UDPG pyrophosphorylase can be measured by coupling with the action of phospho-glucomutase and glucose-6-phosphate dehydrogenase, whereby dihydrotri-phosphopyridine nucleotide (TPNH) is formed in stoichiometric amounts for every molecule of UDPG that is split upon the addition of inorganic pyrophosphate. With the aid of preparations of phosphoglucomutase and glucose-6-phosphate dehydrogenase that were devoid of UDPG pyrophosphorylase activity, it was possible to demonstrate small but significant amounts of the latter enzyme in fresh human blood serum (approximately 80 numiole substrate transformed per 1 ml. of serum per hour at 25° C. (Gotterer and Williams-Ashman, unpublished data). During the course of these experiments it was observed that serum alone catalyzed the reduction of triphosphopyridine nucleotide (TPN) by glucose-6-phosphate. Dehydrogenases specific for TPN had been found in silkworm blood16 and in the ascitic plasma of mice bearing the Ehrlich ascites tumor,18 but had not been described in mammalian blood serum Accordingly, a survey of TPN-linked enzymes in serum was undertaken.

* The studies described in this paper were supported by grants from the American Cancer Society, Inc., New York, N. Y., the Jane Coffin Childs Memorial Fund for Medical Research, New Haven, Conn., and the Public Health Service, Bethesda, Md.

TPN-linked enzymes in serum. Samples of serum without visible evidence of hemolysis obtained from normal adults did not reduce TPN in the absence of added substrates, but catalyzed the formation of TPNH upon the addition of glucose-6-phosphate, 6-phosphogluconate, and isocitrate. Methods for the determination of 6-phosphogluconic (PGD) and isocitric (ICD) dehydrogenases in serum were developed.17 Magnesium ions and cysteine are necessary for the optimal activity of the former enzyme, and manganous ions for ICD. In normal adults the activity of both enzymes is of the same order of magnitude, that is, in the range of SO to 250 mµmolesof substrate oxidized per 1 ml. of serum per hour at 25° C. On a molar basis, this represents the same order of activity as that of serum aldolase,18 glutamic pyruvate transaminase (GPT),19 and glutamic-oxalacetic transaminase,20 but is much less than that of LD,5 malic dehydrogenase,3 phosphohexoseisomerase,21 and 5-phosphoriboseisomer-ase.72 Serum glucose-6-phosphate dehydrogenase levels are about the same as those of serum PGD.17 The activity of serum ICD and PGD could not be correlated with age, race, sex, or total serum protein concentration in apparently healthy adults and was within the normal range in the sera of term pregnancies. Serum ICD levels were elevated, however, in the cord blood of newborn infants.

TPNH is barely oxidized by human serum under the conditions of the ICD and PGD test systems when the appropriate substrates are omitted from the reaction mixtures. Furthermore, TPNH was not oxidized by the serum of normal individuals even in the presence of either a-ketoglutarate plus ammonium chloride or of pyruvate plus bicarbonate (saturated with carbon dioxide) and manganous ions, indicating that glutamic dehydrogenase24 and "malic" enzyme26 are absent from serum. The stability of TPNH in the presence of serum is in marked contrast to the behavior of dihydrodiphosphopyridine nucleotide (DPNH), which may undergo considerable oxidation when added to serum, expecially if the pyruvate content of the latter is high.*

The rate of reduction of TPN in the ICD and PGD lest systems was found to be strictly proportional to the amount of serum added. The ICD and PGD activity of erythrocytes was found to be very much greater than that of serum.17 However, contact of erythrocytes with serum for many hours at room temperature did not increase the levels of serum ICD and PGD unless hemolysis had occurred.

Serum isocitric dehydrogenase in disease. An examination of the serum ICD activity of 250 patients26 revealed that the levels of this enzyme seldom fell outside the limits of the normal unless hepatic disease was present. Normal values were observed in subjects with a wide variety of infectious, pulmonary, gastrointestinal, cardiovascular, urological, and neoplastic diseases. Of particular interest was the finding (hat within the first 24 hours after the onset of symptoms, the levels of serum ICD in patients with large, acute myocardial infarctions were found invariably to fall within the normal range. In this connection it is of interest that the ICD activity of human liver and cardiac muscle, determined from specimens obtained within 10 hours post mortem, were of the same order of magnitude (400,000 to 700,000 mumoles substrate oxidized per gram of tissue per hour at 25° C).

* After this work had been completed, another TPN-specific enzyme, glutathione reductase, was reported to be present in human blood serum by Manso and Wroblewski.23 These authors observed that the activity of serum glutathione reductase was about ten times as great as serum ICD and PGD." Scrum glutathione reductase was found to be elevated in viral hepatitis, although not to such a great extent as serum ICD. Manso and Wroblewski also found abnormally high glutathione reductase levels in the sera of patients with carcino-mata. The presence of a very active glutathione reductase in malignant tumors has been reported.16

Figure 1 shows that within the first 10 days following the onset of jaundice, serum ICD levels were invariably in excess of the upper limit of the normal in paiients with acute viral hepatitis. Increases of serum ICD as great as 40 times the normal mean were found in this disease. Large elevations of this serum enzyme were also associated with hepatitis, complicating infectious mononucleosis. Serial determinations of serum ICD in viral hepatitis showed that the levels of this enzyme usually declined to normal around the third week following the onset of icterus. However, when the chronic form of the disease ensued, serum ICD activities remained markedly elevated for as long as 4 months.