By EUGENE P. KENNEDY and SYLVIA WAGNER SMITH (From the Ben May Laboratory for Cancer Research and the Department of Biochemistry, University of Chicago, Chicago, Illinois)

In a study of the synthetic processes of the Ehrlich ascites tumor of the mouse carried out in this laboratory, isolated washed tumor cells were incubated in the presence of inorganic orthophosphate labeled with Pw. The phosphorus compounds of the cell were then fractionated by the method of Schneider (1), and it was found that the so called "phosphoprotein" residue obtained in the Schneider procedure had a specific activity about 30 times higher than that of the phospholipide or nucleic acid fractions. Similar high values for the rate of turnover of "phosphoprotein" in the Yoshida sarcoma have been reported by Araki el al. (2). Experiments by Davidson et al. (3) had previously shown that the specific activity of the "phosphoprotein" of the livers of normal animals 2 hours after injection with P32 is about 4 to 6 times higher than that of the other acid-insoluble fractions. Results indicating a high rate of turnover of the phosphorus of the "phosphoprotein" fraction in intact animals have also been reported by Johnson and Albert (4).

Friedkin and Lehninger (5) and Williams-Ashman and Kennedy (6) have also described a high rate of turnover of the "phosphoprotein" of cell-free particulate enzyme systems of normal liver and of the Flexner-Jobling tumor.

The studies cited are of considerable interest in that they indicate a potentially important active metabolic rdle of the "phosphoproteins" of normal and malignant tissues. However, it has been pointed out by Friedkin and Lehninger (5) as well as by Davidson et al. (3) and others (7, 8) that the "phosphoprotein" fraction of Schneider's procedure is in no sense a pure chemical entity. It is merely the insoluble residue obtained after successive extractions of the acid-soluble, lipide, and nucleic acid fractions. It may be contaminated with radioactive inorganic phosphate or other components of high specific activity which are tenaciously bound to such precipitates. If one uses the procedure of Schmidt and Thann-hauser (9) rather than that of Schneider to obtain the "phosphoprotein" phosphorus, the situation is not much more satisfactory since, here also, the "phosphoprotein" activity may in fact be attributed to the presence of small amounts of inorganic phosphate or of alkali-labile phosphate of high specific activity.

* This investigation was supported by a grant from the American Cancer Society as recommended by the Committee on Growth of the National Research Council.

The extremely high activity of the "phosphoprotein" residue of the Ehr-lich ascites tumor, as well as the results of experiments of other workers with normal tissues, made it imperative that the active components of this fraction be identified. The work of Lipmann (10, 11) has demonstrated that the phosphorus of casein and of vitellinic acid may be isolated in part at least as phosphoserine. More recently de Verdier (12) has reported the isolation of minute amounts of phosphothreonine from casein. If the radioactivity of the "phosphoprotein" fraction of Ehrlich ascites tumor is really caused by a rapid turnover of phosphorylated amino acid residues, then it should be possible to isolate radioactive phosphorylated amino acids from this fraction after partial hydrolysis in hydrochloric acid. The present paper reports the isolation of phosphoserine of very high specific activity from hydrolysates of Ehrlich ascites tumor "phosphoprotein" by chromatographic methods specially designed for the separation, identification, and isolation of small amounts of phosphorylated amino acids. No significant amounts of other radioactive phosphorylated amino acids could be detected. Preliminary experiments have yielded evidence of the presence of phosphoserine in hydrolysates of the phosphoprotein fraction of isolated mitochondria from normal rat liver incubated in vitro with labeled ortho-phosphate under conditions designed for vigorous oxidative phosphorylation (5).

Materials And Methods

The inoculation, growth, and harvesting of the Ehrlich ascites tumor in mice were carried out exactly as described by Kun, Talalay, and Williams-Ashman (13).

Inorganic phosphate labeled with Pw was obtained from the United States Atomic Energy Commission through the Oak Ridge National Laboratory and was purified before use as described previously (14).

Phospho-l-threonine, phospho-l-hydroxyproline, phospho-l-tyrosine, phospho-dl-serine, and phospho-l-serine were synthesized by a method based on that of Plimmer (15). These compounds were subjected to a brief acid hydrolysis and purified by passage over columns of Dowex 1 chloride resin before isolation as the barium salts.

P32 measurements were made from aliquots dried in aluminum cups and counted in a gas flow counter under conditions of negligible self-absorption.

A method similar to that of Moore and Stein (16) was used to detect free and phosphorylated amino acids in fractions derived from chromatography on ion exchange resins.

Phosphorus was determined by the method of Gomori (17).

Experimental

Incorporation of P32 into Ehrlich Ascites Tumor Cells-The ascites fluid from twenty to thirty mice was pooled for each experiment. Hemorrhagic ascites fluid was discarded. The cells were separated from the ascitic plasma by centrifugation at 0° and were washed twice with the following medium: NaCl 0.125 m, KC1 0.005 m, MgCl2 0.001 m, NaHCO, 0.025 m. After washing, the cells were suspended in an amount of the medium equal to about twice the original volume of the ascites fluid. Glucose (0.2 per cent) and Na2HP32O4 (0.1 µM of P per ml. with specific activity of 10 µc. per µM of P) were added, and the cells were shaken for 2 hours under an atmosphere of 95 per cent 02 + 5 per cent CO2. The temperature was 37°.

At the end of the incubation, one-tenth of a volume of 100 per cent (weight per volume) trichloroacetic acid was added to stop the reaction. The trichloroacetic acid-insoluble precipitate was then fractionated according to the method of Schneider (1). The final "phosphoprotein" residue obtained by the Schneider method was washed three times with acetone and dried in a vacuum desiccator overnight. The "phosphoprotein" so obtained contained 0.2 per cent phosphorus.

The results of such an experiment are presented in Table I. The very high specific activity of the "phosphoprotein" fraction in comparison to that of the other acid-insoluble fractions is particularly noteworthy. The incorporation of P32 into these fractions is dependent upon an energy supply, as shown by the anaerobic experiment in which incorporation is very small when glucose is omitted. Aerobically, the incorporation reactions are not dependent upon an added substrate, because of the vigorous aerobic endogenous metabolism displayed by these tumor cells (13).

Acid Hydrolysis of "Phosphoprotein" Fraction-The crude "phosphoprotein" residue made radioactive by incubation of Ehrlich ascites tumor cells in vitro in the presence of P32 as described above was hydrolyzed for 10 hours in 10 volumes of 2 n HC1 at 100°. Approximately 2 gm. of "phosphoprotein" were used in each experiment. At the end of the hydrolysis, during which the protein was only partially hydrolyzed, the hydrolysate was clarified by centrifugation and chilled in an ice bath. Concentrated ammonia was added to adjust the pH to about 9. Inorganic phosphate was then removed as magnesium ammonium phosphate. Phosphorus and radioactivity measurements were made on the inorganic and esterified phosphorus fractions of the hydrolysate. For the results of two such experiments see Table II.