Glutamate in the Walker 256 Carcinosarcoma in Vivo1

Home , Glutaminase

[CANCER RESEARCH 50, 4839-4844, AuguÂ§t15.1990] Short-Term Metabolic Fate of L-[13N]Glutamate in the Walker 256 Carcinosarcoma in Vivo1

Sabina Filc-DeRicco, Alan S. Gelbard,2 Arthur J. L. Cooper, Karen C. Rosenspire,3 and Edward Nieves4

Nuclear Medicine Research Laboratory, Memorial Sloan-Ketlering Cancer Center [S. F-D., A. S. G., K. C. R., E. N.J, and the Departments of Biochemistry' and Neurology, Cornell University Medical College [A. J. L. C.], New York, New York 1002 1

ABSTRACT nitrogen derived from ammonia and amino acids in the brain (8, 9) and liver (10-12). Other laboratories have used HPLC to In vivo studies with L-[13N)glutamate in the Walker 256 carcinosar- study the fate of uN-metabolites in the myocardial septum (13, coma implanted under the renal capsule of female Sprague-Dawley rats 14) and we have previously described the metabolic fate of [I3N]- demonstrate that uptake of glutamate and the rate of incorporation of ammonia and L-[a/w/We-uN]glutamine in glutaminase-sensitive the nitrogen label from this amino acid into metabolites is slower in the and -resistant murine tumors (15). Since tracer studies with "N tumor than in nontumorous kidney tissue. Glutamate dehydrogenase, glutaminase, and alanine aminotransferase activities are significantly have provided much useful information on nitrogen turnover lower within the tumor than within the adjoining kidney. However, the in these tissues in vivo, we have adopted these procedures to tumor expresses high levels of aspartate aminotransferase, attesting to study nitrogen metabolism in an intact rat tumor model in vivo. the importance of this enzyme in the metabolism of glutamate. Indeed, Accordingly, we chose to study L-['3N]glutamate metabolism in high performance liquid Chromatographie analysis showed that the prin an implanted Walker 256 carcinosarcoma. The rationale for cipal metabolic fate of label derived from L-|13N)glutamate in the tumor the choice of the 13N-amino acid and the tumor model system is incorporation into aspartate. Measurement of specific activity ratios of is given below. glutamate to aspartate shows that the transfer of nitrogen from glutamate In vitro studies of tumor mitochondria and of tumor tissue to aspartate is rapid and that equilibration of label among components of grown in culture indicate that exogenously supplied glutamine the aspartate aminotransferase reaction is attained within minutes after is a major energy source for neoplasms (16-18). The major tumor uptake. Analyses of the nontumorous portion of the implanted kidney also showed that aspartate is the major recipient of glutamate pathway for glutamine metabolism involves conversion to glu nitrogen. However, high performance liquid Chromatographie analyses of tamate as a first step. Moreover, since glutamate can be readily labeled with I3N by an immobilized enzyme procedure, we chose deproteinized tissue revealed that glutamine and ammonia are also sig nificant "Vlabt'li'd metabolites formed from L-[13N|glutamate within the to investigate the short-term tumor metabolism of this amino kidney. Proportionately lower amounts of these labeled metabolites were acid nitrogen. To facilitate this study, we chose a rat tumor found in the tumor. model system that provides for: (a) efficient delivery of labeled glutamate; and (/>) ready quantitation of metabolites entering and leaving the implant region without perturbing or occluding INTRODUCTION blood flow. Amino acids are avidly taken up by tumors for various cellular In vitro studies performed on Ehrlich ascites carcinoma cells metabolic processes. When these compounds are labeled with by Kovacevic et al. (19, 20) demonstrated that aspartate was an a positron-emitting radionuclide such as I3N (f./,9.96 min), they important metabolite formed from the oxidation of glutamine can be used for imaging neoplasms either by positron emission and that its rate of utilization is much slower than its rate of tomography or by two dimensional scanning (1-5). Assessment synthesis. This present report details the in vivo short-term metabolic fate of label derived from L-[uN]glutamate and also with external scanning devices of changes in the concentration of label in tumors after treatment with chemotherapeutic agents emphasizes the significance of label transfer from glutamate to has been used for the design of treatment regimens for several aspartate via aspartate aminotransferase in the in situ Walker patients at Memorial Sloan-Kettering Cancer Center (6). 256 carcinosarcoma. Aspartate plays a pivotal role in supplying In addition to their value as imaging agents, l3N-labeled nitrogen for purine and pyrimidine synthesis and in the transfer compounds offer unique advantages as biochemical tracers for of reducing equivalents across the mitochondrial membrane via the study of short-term nitrogen metabolism. Nitrogen-contain the malate-aspartate shuttle. ing compounds can be labeled with high specific activity and Since a portion of the kidney was left intact and not affected can be administered in tracer doses without perturbing physio by tumor tissue growth, we also report on the short-term metabolic fate of L-[13N]glutamate in this tissue. Although logical processes. Labeled metabolites can then be rapidly sep arated by HPLC5 and quantitated by an on-line flowthrough aspartate is the major nitrogen-containing metabolite formed radioactivity detector system (7). Our laboratory has used this from glutamate, the rate of glutamate metabolism is greater protocol to determine the short-term in vivo metabolic fate of and proportionately more label is present in glutamine and ammonia in this tissue than in the tumor. Reflecting the dis Received 1/4/90; revised 4/24/90. parate metabolic requirements, marked differences were found The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in in several enzyme activities affecting glutamate metabolism as accordance with 18 U.S.C. Section 1734 solely to indicate this fact. well as in the uptake and fate of L-['3N]glutamate between the ' This work was supported in part by NIH Grants CA-34603 and DK-16739 and by Department of Energy Grant DE-FG-02-86-ER-60407. Presented in part tumor and normal kidney. at the 34th Annual Meeting of the Society of Nuclear Medicine, June 2-5. 1987, Toronto, Ontario, Canada. 2To whom requests for reprints should be addressed, at Nuclear Medicine MATERIALS AND METHODS Research Laboratory, Memorial Sloan-Kettering Cancer Center. 1275 York Avenue, New York, NY 10021. Biochemicals and Reagents. o-Phthaldialdehyde, L-lactic dehydrogen- 'Present address: PET/Cyclotron Facility, University of Michigan, Ann Ar ases from various sources (with specific activities of 500-1000 units/ bor, MI 48108. mg), L-aspartate, L-alanine, pyridoxal 5'-phosphate, a-ketoglutarate, 4 Present address: Department of Neoplastic Diseases, Mount Sinai Medical Center, New York, NY 10029. triethanolamine-HCl, EDTA-Na2H2-H2O, ammonium acetate, and 9The abbreviation used is: HPLC, high performance liquid chromatography. ADP were purchased from Sigma Chemical Company (St. Louis, MO). 4839

NADH was obtained from Calbiochem (San Diego, Ã‡A).Glutamate of renal implanted tumor, nontumorous kidney, and blood were sepa dehydrogenase was purchased from Boehringer Mannheim (Indianap rately analyzed for metabolites of L-['3N]glutamate by cation exchange olis, IN) and cyanogen bromide-activated Sepharose was obtained from chromatography (Partisil SCX column, 250 x 4.6 mm) using a HPLC Pharmacia (Piscataway, NJ). on-line flowthrough radioactivity detector system (7). The SCX column i.-|13N|Glutamate Preparation. ["NJAmmonia was prepared in the was eluted with 5 mM potassium phosphate-H,PO4 (pH 2.55, at a flow Memorial Sloan-Kettering Cancer Center cyclotron (Model CS-15; rate of 1 ml/min) for 8 min and subsequently eluted with 20 mM Cyclotron Corp., Berkeley, CA) by the 16O(p,Â«)"N reaction on water potassium phosphate-H,POj (pH 3.5, at a flow rate of 1 ml/min) for and subsequent reduction of '\ labeled nitrates and nitrites with the remainder of the elution time (10). The column was connected Devarda's alloy (21). L-["N]Glutamate was prepared by passing a directly to a radioactivity monitor/analyzer (Ramona D; IN/US Service mixture of [13N]ammonia, a-ketoglutarate, and NADH through a col Corp., Fairfield, NJ) by a 6-port valve. Data received by the detector umn on which glutamate dehydrogenase had been immobilized to a were decay corrected, plotted, analyzed, and stored by a separate solid support of CNBr-activated Sepharose (22). computer program (ISOMESS, Straubenhardt. West Germany). Me Experimental Animals and Tumor Implants. Walker 256 carcinosar- tabolite elution profiles were compared with the profiles of known coma was obtained from the National Cancer Institute (Bethesda, MD). nonradioactive standards. Female Sprague-Dawley rats (HarÃanSprague-Dawley, Madison, WI) Analysis of Unlabeled Amino Acids and Urea Â¡nTissues.Frozen tissue weighing 250-375 g were anesthetized with an i.m. injection of 4-6 samples were pulverized to a fine frozen powder in a cryostat. Known mg/100 g body weight of ketamine HC1. To assure the effectiveness of amounts of Â»-aminoadipate were added, as an internal standard, to the the anesthesia, the animals were supplemented with ether inhalation. pulverized tissue immediately before deproteinization with 3 M per Ketamine was selected as the optimal anesthetic because of its minimal chloric acid (8). The tissues were then processed as described previously effects on the cardiovascular system (23, 24). and the soluble amino acid components were separated via reverse A suspension of Walker 256 carcinosarcoma cells was prepared by phase HPLC as their precolumn o-phthaldialdehyde derivatives (10). digestion of minced tumor with 0.05% trypsin and 0.02% EDTA in The reverse phase chromatography system consisted of an ODS Hanks' balanced salt solution that lacked CA2* and Mg2* for 15 min column (Spherex 3 dÂ«, 100 x 4.6 mm; Phenomenex, Palo Alto, CA) at 37"C and by subsequent processing using mild mechanical procedures which was eluted with a microprocessor-controlled gradient mixture of (25). The cells were then washed twice in complete Hanks' balanced two buffers over a period of 25 min at a rate of 1.5 ml/min. Buffer A salt solution and resuspended in the same medium to a final volume of consisted of acetonitrile:12.5 mM sodium phosphate, pH 7.2 (3:97); less than 1.5 ml. To obtain renal implants, tumor cells (lOVanimal) buffer B consisted of acetonitrile:12.5 mM sodium phosphate, pH 7.2 were injected under the left renal capsule of anesthetized female rats (50:50). The initial setting was 95% buffer A and 5% buffer B. At 25 according to the method tirsi described by Cullino and Grantham (26). min, the final eluant was 45% buffer A and 55% buffer B. o-Phthaldi- In some cases, where it was necessary to obtain s.c. implants for enzyme aldehyde-derivatized amino acids eluting from the column were de activity analysis, a tumor cell suspension (obtained by the process tected with a fluorescence monitor (Kratos model 980). A detailed described above) was injected s.c. under the right flank. description of this buffer system and its applicability has been published Ten to 14 days postimplantation, 1.5 ml of L-["N]glutamate (10- by Graser et al. (29). Unlabeled urea concentrations in renal artery and 100 mCi, with a specific activity of ~500 mCi/^mol) were infused into vein blood were measured according to the method described by the left renal artery through a 30-gauge catheter at a rate of 4 ml/min Kerscher and Ziegenhorn (30). for 20 s. At periods of 1, 3, and 5 min after the start of injection, Determination of Tissue Enzyme Levels. Portions of frozen s.c. portions of tumor and the nontumorous remainder of the kidney were Walker 256 implants, renal implants, and nontumorous kidney tissue were individually homogenized with a 5-fold volume of ice-cold saline quickly separated by dissection, frozen in liquid nitrogen, and homog enized in a 5-fold volume of ice-cold 1% (w/v) picric acid with a glass- in glass-to-glass tissue grinders. Total (i.e., cytosolic plus mitochon to-glass tissue grinder. The homogenized tissue was then centrifuged at dria!) aspartate aminotransferase, alanine aminotransferase, glutamate 12,000 x g for 30 s and the supernatant was analyzed for 13N-labeled dehydrogenase, and glutaminase activities were determined on a mini metabolites by cation exchange HPLC. Other portions of frozen tissues mum of 3 samples of each tissue homogenate. All reagents and solutions were stored at -70Â°C and analyzed, at a later date, for amino acid were prepared as described (30-34). The oxidation rate of NADH was content and enzyme activities. Blood samples from the renal artery and determined by noting the change in absorbance at 340 nm with an HP 8452 A diode array spectrophotometer (Hewlett Packard) (E = 6.32 x vein were collected over a period of 15 s just prior to freezing of the IO3). tissues and were homogenized with approximately 100 mg of sulfosal- icylic acid (deproteinizing agent) per 1 ml of blood. Radioactivity Statistical Analysis. Data are expressed as mean Â±SE. Statistical significance was determined by the two-tailed Mann-Whitney U test. content of tissues is expressed as percentage of dose per g; in a few cases, relative concentration is additionally reported for the 5-min data The ratios of specific activities of glutamate to aspartate, glutamate to and was determined after injection of a bolus (0.2 ml) of L-["N]- alanine, and glutamate to glutamine were determined from the term (*X/*Y)/(X/Y), where *X/*Y is the ratio of radioactivity and X/Y is glutamate. Tissues were quickly dissected at the appropriate time, the ratio of concentration of these components. *X/* Y and X/ Y were counted in an 1KB gamma counter, and weighed. Relative concentra tion is di-finn! as the fraction of dose found in a tissue specimen divided determined in two separate groups of rats. Therefore, in order to determine the standard error of the mean (a) of the ratio of specific by the fraction of body weight contained in that specimen (27). Since the experimental protocol requires L-["N]glutamate to be activity (A), we assume that the errors are propagated randomly. The equation used for this type of error propagation is rapidly injected into the renal artery, it was necessary to establish that blood flow to the region was not impeded in any way and that the o2 = 62 animals were not acidotic. Perfusion studies utilizing "Tc-labeled A2~ B2 c2_ C2 human serum albumin revealed that the injected material passed through the implant region prior to systemic entry. Blood clearance where B and C are the means of the ratio of radioactivity (*X/* Y) and (washout) studies performed with '"Xe demonstrated that flow to the the ratio of concentration (X/Y), respectively; b and c are the respective implant region was not impeded by the 30-gauge catheter needle. standard errors of the mean. Analysis of blood gases and pH in a number of experimental animals showed no evidence of acidosis. In some cases, the blood volume of tissues was determined by comparison of the radioactivity in the organ RESULTS AND DISCUSSION of interest versus that in the blood after "CO inhalation (28). Blood To evaluate the degree of L1Nlabel transfer from glutamate, content (w/w) expressed as a percentage (n = 2) in nontumorous kidney, contralateral kidney, renal implanted tumor, and s.c.-implanted tumor it was necessary to determine the pool sizes of unlabeled amino were (4.1., 3.3), (3.5, 2.4), (2.0, 3.3), and (0.91, 0.96), respectively. acids that are involved in glutamate metabolism. The concen HPLC of l3N-Metabolites. Aliquots (20 M')of deproteinized extracts tration values reported in Table 1 and 2 are within range of 4840

Table 1 Percentage of label in metabolites in nontumorous kidney at various time points after L-["N]glutamate injection these studies, some important conclusions can nevertheless be deduced. (jimol/g wet The proportion of radioactivity in aspartate at all times was min0.07 min1.06 min3.12 wt)NQ*0.54 appreciable (Table 1). Aspartate aminotransferase is of high Urea + 0.04" Â±0.31 + 0.34 activity in most organs. Indeed, the components of the reaction Asp 14.1 Â±5.3 20.8 Â±1.5 18.2 Â±2.3 Â±0.07 in liver (11) and brain (8) are in thermodynamic equilibrium. 80.0 + 6.7 60.6 + 2.6 Glu 60.4 Â±2.8 1.96 Â±0.13 In the liver, exchange of labeled nitrogen between aspartate and Gin 2.44+ 1.33 6.33 Â±0.76 9.15 Â±0.74 0.51 Â±0.05 Ala 0.14 + 0.08 0.80 Â±0.29 1.12 Â±0.24 0.59 Â±0.05 glutamate is very rapid (order of seconds) (11, 12). In the Ammonia 0.41 Â±0.13 8.16 Â±2.75 4.81 + 1.05 NQ present work, at 3 and 5 min postinjection of L-[13N]glutamate, ArgSuc ND 0.15 Â±0.06 0.1 2 Â±0.07 NQ Arg1 ND3 0.43 Â±0.155 0.49 + 0.16ConcentrationNQ the ratio of specific activities of glutamate to aspartate in the kidney are not significantly different from 1.0, suggesting that nitrogen is fully equilibrated between the two amino acids by 3 Ratios of radioactivity min (Table 1). Even at 1 min the ratio is not far from 1.0; min7.50 min2.97 min3.83 deviation from 1.0 may reflect the fact that a considerable portion of the recirculating counts at 1 min are still in glutamate Glu/Asp Â±2.47 + 0.32 + 0.60 Glu/Gln 55.5 Â±26.7 9.95 Â±0.94 6.89 Â±0.61 (Table 3) or that at 1 min equilibration of label between cytosol GiÃ¹/Ala1 447 + 206318.4Ratios59.0 Â±10.65 77.9+ and mitochondria is not complete [see discussion by School- activitiesGlu/Asp of specific werth and LaNoue (38) and by Cooper et al. (10)]. Thus, as shown for the rat liver, exchange of label among components min2.06 min0.82 min1.05 of the aspartate aminotransferase reaction is also very rapid + 0.73 Â±0.14 + 0.21 and that the overall reaction is probably in thermodynamic Glu/Gln 14.7 Â±7.3 2.64 + 0.39 1.83 Â±0.26 GiÃ¹/Ala1 141 Â±663 18.6 Â±3.75 24.6 + 6.2 equilibrium in the kidney. Aspartate appears to be synthesized 1Mean Â±SE. and utilized within the kidney cell; very little aspartate is taken *NQ, not quantitated; ND, not detected; ArgSuc, argininosuccinate. up or released by the kidney (Table 3). In contrast to the aspartate aminotransferase reaction, the Table 2 Percentage of label in metabolites in Walker 256 renal implant at alanine aminotransferase reaction in the rat kidney is not in various time points after L-f'^/VJglutamate injection thermodynamic equilibrium. Thus the ratio of specific activities of glutamate to alanine is far greater than 1.0 and appears to (jimol/g wet plateau at a value close to 20 by 3 min (Table 1). Equilibration min0.33 min1.01 wt)NQ"0.46 min0.01 of label between glutamate and alanine appears to occur within Urea Â±0.01Â° + 0.08 Â±0.31 Asp 7.37 Â±3.02 8.64 Â±0.84 17.8 + 2.4 + 0.06 minutes as distinct from the aspartate aminotransferase reac Glu 89.6 Â±3.9 82.6+ 1.8 73.2 + 4.3 3.37 Â±0.14 tion in which equilibration is more rapid. Undoubtedly, the Gin 0.57 + 0.26 4.45 + 0.98 4.62 Â±1.66 0.42 Â±0.05 slower exchange reflects the inherently lower activity of alanine Ala ND 0.88 Â±0.37 1.33 Â±0.34 1.34 + 0.09 Ammonia 0.20 Â±0.10 0.91 + 0.39 0.48 Â±0.18 NQ aminotransferase compared to aspartate aminotransferase ArgSuc ND 0.40 Â±0.20 0.22 Â±0.11 ND within this tissue (Table 4). H1 33 5Ratios 45 5Concentration

of radioactivity Table 3 Percentage of label in metabolites in blood at various times after L-f'NJglutamate injection 3 min 5 min ConcentrationUreaAspGluGinAlaAmmoniaArgSucArgaUreaAspGluGinAlaAmmoniaArgSucArgn1In renal artery blood Glu/Asp 20.3+ 10.8 9.9 Â±1.2 4.7+ 1.0 230 + 78 22.3 Â±5.9 20.3+10.1 wciwt)6.15i iiMi g Glu/Gln min0.67,min 3 min 5 GiÃ¹/Ala >100 82.3 + 24.4 81.4 Â±27.4 2.6Â°0.00,0.150.50 7.38, 11.3 22.2 Â± Â±0.530.04 Ratios of specific activities Â±0.0696.5, 0.00,0.16 0.16 +0.010.11 5.51.05,1.8696.4 67.6, 64.8 46.2 Â± +0.010.24 3 min 5 min 15.6+1.10.95,0.4111.6,10.3 Â±0.030.19 +0.03NQÂ»NQNQ6Concentration(//mol/g Glu/Asp 2.62+1.41 1.28 + 0.19 0.60 Â±0.14 Â±1.70.00.0.0111.4,11.4 11.6 0.140.00.0.03 0.00.0.22 0.44 + Glu/Gln 27.7 + 9.7 8.87 Â±2.45 2.44 Â±1.53 GiÃ¹/Ala >IOO 32.7 Â±9.7 32.3 Â±10.9 Â±0.580.00,0.020.53,0.31 1.08 0.51+0.07224In0.00,0.45 â€¢¿MeanÂ±SE. * NQ, not quantitated; ND, not detected; ArgSuc, argininosuccinate. blood1 renal vein wetwt)4.86 min0.04,0.02min 3 min 5 min' 5 values that have been previously reported by other investigators for rat kidney (35, 36) and for a type of tumor similar to the 2.3)0.02, 2.20 12.6 Â±1.7 (17.4 Â± +0.500.03 5)96. 0. 13 0.44 0.37 + 0. 11 (0.47 Â±0.1 Â±0.010.10 one studied in the present investigation (37). 4.4)2.04.13.51, 8 1.7 71.5 36. 1 Â±3.2 (49.8 + Â±0.010.19 L-|'3N|Glutamate Metabolism in Kidney. The relative concen 1.2)'1.37,0.0014.5 28.4 Â±0.9 (39.2 Â± +0.020.19 (14.9+1.4)0.03,6.78 10.8+1.04 +0.03NQNQNQ4 tration of I3N in the nontumorous kidney and blood at 5 min )'0.04,0.030. 12 0.23 1.98 + 0.08 (2.7 + 0. 1 are 19.6 Â±4.2 and 0.95 Â±0.30 (n = 4), respectively. The blood 0.5)0.08,4.01 0.40 1.77 Â±0.35 (2.4 Â± content in kidney is ~3.5%. Thus the amount of label present 3.8)'215 2.95 6.75 Â±2.74 (9.3 Â± 5U< in the kidney contributed by the blood is negligible (~0.2%). " Mean Â±SE. Nevertheless, some exchange of label between blood and kidney " NQ, not quantitated in this tissue. (and presumably between kidney and urine) must have occurred. c At 5 min it was found that the kidney was donating more label to the venous blood than it was receiving in the arterial blood. The excess was calculated to be Such an exchange is exemplified by the steady rise of percentage 38%. Therefore, the percentage values in this column were multiplied by 1.38 of label recovered in urea within the kidney (Table 1). Although (shown in parentheses) to facilitate comparison of "N-labeled metabolites in venous blood with those in arterial blood. exchange of label between kidney and recirculating metabolites d Significantly different from the corresponding values of 13N-labeled metabo complicates interpretation of the pathways of nitrogen flux in lites in the arterial blood at P < 0.01. 4841

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1990 American Association for Cancer Research. GLUTAMATE METABOLISM IN WALKER 256 CARCINOSARCOMA Table 4 Enzyme activities in tissues at 25'C wt"Walker t'nits/g wet

256 renal implant Â±1.3Â°-''22.9 Â±0.15''n Â±0.82"7.93 Â±0.07" tumor extract = 4 n = 4 Walker 256 s.c. tumor Â±0.7' 0.39 Â±0.12'3.27 Â±0.77' 0.82 Â±0.07' extract n = 4 n = 5 Nontumorous kidney 79.7 Â±3.9AlaAT0.69 Â±1.21GluDH9.73 136 Â±13Glnase0.67 6.89 Â±0.44 extractAspAT*20.4 a A unit is defined as 1 Â¿imolNAD formed/min; 25*C. ' AspAT. aspartate aminotransferase: AlaAT. alanine aminotransferase: GluDH, glutamate dehydrogenase: Glnase, glutaminase. ' Mean Â±SE. rf/* < 0.01 between kidney and renal implanted tumor. ' Not significant between renal implanted and s.c. tumor.

The steady increase of radioactivity in glutamine deserves mine whereas the glutamine synthetase-enriched perivenous some comment. Although the specific activity of glutamine hepatocytes excrete glutamine. The net portal vein-hepatic vein synthetase in rat kidney as a whole is relatively low, most of difference for glutamine will depend strongly on the interplay the enzyme is localized within a small volume of cells located between the two cellular processes. See Ref. 42 for a recent in the proximal convoluted tubule of the kidney (39). Thus, the review. Finally, as others have suggested (43, 44), the data in slow rise in percentage of radioactivity in glutamine (Table 1) Table 3 demonstrate that the kidney is a net source of circulat may reflect labeled glutamate entering this cellular compart ing arginine. ment. Some labeled glutamine may arise from labeled glutamine L-l'-'NJGlutamate Metabolism in the Renal-implanted Walker in the blood, but the net output of labeled glutamine from the 256 Tumor. Because the relative concentrations of I3N in the kidney (Table 4, 5 min data; see below) strongly suggests that tumor and blood are 4.9 Â±0.7 and 0.95 Â±0.30 (n = 4), the labeled glutamine was synthesized within the kidney. respectively, at 5 min and the blood content of the tumor is The kidney excretes ammonia into the urine; it is generally <3%, the amount of label present in the tumor due to blood is supposed that glutamine and possibly glutamate are sources of negligible (<0.6%). The lower concentration of radioactivity in urinary ammonium (40). Our data bear on this point. In two the tumor (2.2 Â±0.7% dose/g; relative concentration, 4.9 Â± sham-operated animals, urine was collected and analyzed ~5 1.4) compared with the kidney (8.5 Â±1.8% dose/g; relative min after injection of L-["N]glutamate. Very little label was concentration, 19.6 Â±4.2) may be due to a slower penetration found in urea (>1%) of the total reflecting low radioactivity in of blood-borne glutamate into the tissue, slower blood flow, urea within the kidney over the first three minutes (Table 1). lower relative blood volume, or a combination of these factors. In one animal 75% of the radioactivity recovered in the urine Nevertheless, despite the relatively low uptake of label, a few was in ammonia and 25% was in glutamate. In the second points concerning glutamate metabolism in the tumor are evi animal, ~20% of the recovered activity was in ammonia and dent. ~80% was in glutamate. As noted in the experimental section, The relative rate of appearance of label into glutamine is ketamine was chosen as the anesthetic because of its minimal slower in the tumor than in the kidney (Table 2). This slower effect on the cardiovascular system. However, we cannot rule uptake may reflect low glutamine synthetase activity in tumor out the possibility that the i.v. volume was altered in the or a slow uptake of labeled glutamine from the blood. Perhaps ketamine-anesthetized animals and this affected the distribution the most important finding is that the major metabolic fate of of the label from venous blood and urine. Nevertheless these glutamate nitrogen in the tumor relates to the aspartate ami experiments show that kidney glutamate is indeed a source of notransferase reaction. Thus, as was found for the nontumorous urinary ammonia. This ammonia can theoretically arise directly kidney, aspartate is very rapidly labeled once L-["N]glutamate from glutamate (glutamate dehydrogenase) or indirectly via enters the tumor (Table 2). Again, the ratios of specific activities aspartate (through the purine nucleotide cycle). The latter seems of glutamate/aspartate are remarkably close to 1.0 at all times; less likely as no detectable radioactivity (<0.05%) was found in the slightly higher value at 1 min may reflect the fact that the kidney homogenates in the fraction eluting from the HPLC equilibration of label between cytosol and mitochondrial pools column that contains AMP (Ã©lÃ»tesjustbefore urea). of glutamate may not yet have occurred at this time. The specific A few additional points concerning kidney nitrogen metabo activity (units/g wet weight) of "total" aspartate aminotrans lism are evident from Tables 1 and 3. Not surprisingly, the data ferase in the tumor is 25% ofthat exhibited by the nontumorous show considerable label in urea. In agreement with previous kidney (Table 4), but the activity is nevertheless substantial. In suggestions (41), the present work also indicates that kidney contrast, the activity of glutamate dehydrogenase in the tumor contributes ammonia to the systemic blood (Table 3). The is only 6-7% that Â¡nthe kidney. The activities of alanine normal rat kidney is generally thought to utilize glutamine, in aminotransferase and glutaminase in tumor are also markedly part, as an energy source (40). Indeed, the arteriovenous data lower than those demonstrated in kidney. (Table 3, Column 5) suggest a trend toward glutamine removal The present work relates to the notion that glutamine is a by the kidney. However, at 5 min, a significant net output of major energy source of tumors. On conversion of glutamine to labeled glutamine was noted (Table 3, Column 4). These find glutamate, five carbon units enter the tricarboxylic acid cycle ings may be explained by assuming that: (a) glutamine is taken as rt-ketoglutarate. This can be accomplished either by trans- up from the blood in a population of cells that is distinct from amination or via the glutamate dehydrogenase reaction (45). a population of cells that contain glutamine synthetase; and (b) Greenhouse and Lehninger have investigated glutamate metab that labeled glutamine is released from these glutamine synthe- olism in tumor cells. The authors have presented strong evi tase-rich cells. The situation may be analogous to the "gluta dence that the malate-aspartate shuttle is important for the mine cycle" proposed for the liver by HÃ¤ussinger. In the liver transport of reducing equivalents across the mitochondrial the urea cycle-enriched periportal hepatocytes take up gluta membrane in various isolated tumor cells (46). Indeed, it was 4842

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1990 American Association for Cancer Research. GLUTAMATE METABOLISM IN WALKER 256 CARC1NOSARCOMA metabolic fate of [13N]ammonia, L-('3N|alanine. L-|13N|glutamate and L- estimated that on the average about one-third of the respiratory |omiWe-'3N|glutamine in normal rat liver in viro. In: P. B. Soeters. J. H. P. ATP in tumor cells is generated by electron flow originating Wilson, A. J. Meijer and E. Holm (eds.). pp. 11-25. Advances in Ammonia from cytosolic NADH via the malate-aspartate shuttle (47). Metabolism and Hepatic Encephalopathy. Amsterdam: Elsevier Science Pub Crucial components of the malate-aspartate shuttle include lishers. 1988. 13. Keen, R. E., Krivokapich, J., Phelps. M. E., Shine, K. I., and Barrio, J. R. cytosolic and mitochondrial aspartate aminotransferase. Mo- Nitrogen-13 flux from L-[N-13jglutamate in the isolated rabbit heart: effect readith and Lehninger (17), using isolated mitochondria, have of substrates and transaminase inhibition. Biochim. Biophys. Acta, 884: 531- also shown that glutamate is converted to Â«-ketoglutarate al 544. 1986. 14, Krivokapich. J., Barrio, J. R., Phelps, M. E., Watanabe, C. R.. Keen, R. E., most entirely via the aspartate aminotransferase reaction; glu Padgett. H. C.. Douglas. A., and Shine. K. I. Kinetic characterization of I3NHj and [N-13]glutamine metabolism in rabbit heart. Am. J. Physiol.. 246: tamate dehydrogenase was not important in this process. The H267-H273, 1984. results of our studies complement the work of Lehninger and 15. Rosenspire. K. C., Gelbard. A. S., Cooper, A. J. L., Schmid, F. A., and his colleagues and further emphasize the importance of aspar Roberts. J. |"N]Ammonia and L-[am;Vfr-"N]glutamine metabolism in glutaminase-sensitive and glutaminase-resistant murine tumors. Biochim. Bio tate aminotransferase in the energy metabolism of intact tumors phys. Acta, 843: 37-48. 1985. in vivo. Thus our work directly demonstrates that glutamate is 16. Moreadith. R. W., and Lehninger. A. L. Purification, kinetic behavior, and converted to Â«-ketoglutarate predominantly via the aspartate regulation of NAD(P)+ malic enzyme of tumor mitochondria. J. Biol. Chem.. 259:6222-6227, 1984. aminotransferase reaction in the Walker 256 carcinosarcoma 17. Moreadith, R. W., and Lehninger, A. L. The pathways of glutamate and in vivo; little glutamate is converted to Â«-ketoglutarate via the glutamine oxidation by tumor cell mitochondria. Role of mitochondrial glutamate dehydrogenase reaction. This reliance on aspartate NAD(Pr dependent malic enzyme. J. Biol. Chem.. 259: 6215-6221, 1984. 18. Reitzer, L. J.. Wice. B. M.. and Kennell. D. Evidence that glutamine. not aminotransferase may have therapeutic implications. It has sugar, is the major energy source for cultured HeLa cells. J. Biol. Chem.. been shown that following i.p. injection of ÃŸ-methylene-DL- 254: 2669-2676. 1979. 19. Kovacevic. Z., Popovic, J., Brkljac. O., and Lelas, S. Interaction of metabo aspartate, aspartate aminotransferase is inhibited acutely in lism of aspartate and inosine and energy state of malignant cells. Biochem. mouse liver and kidney by 20 and 40%, respectively (48). The J., 2*7:47-51, 1987. mice exhibited no obvious untoward effect. Since the tumor 20. Kovacevic. Z., Jerance. D.. and Brkljac, O. The role of glutamine oxidation and the purine nucleotide cycle for adaptation of tumour energetics to the cells rely so heavily on aspartate aminotransferase for their transition from the anerobic to the aerobic state. Biochem. J., 252: 381-386. energy requirements, it is possible that even a partial block of 1988. this activity will be deleterious to tumor cell viability. The 21. Gelbard. A. S.. Clarke, L. P.. McDonald, J. M.. Monahan. W. G., Tilbury. R. S., Kuo. T. Y. T.. and Laughlin. J. S. Enzymatic synthesis and organ possibility of designing inhibitors of aspartate aminotransferase distribution studies with UN labeled L-glutamine and L-glutamic acid. Ra as tumoricidal agents or as adjuncts to conventional treatment diology. 116: 127-132. 1975. warrants further study. 22. Gelbard. A. S.. Benua. R. S.. Reiman. R. E., McDonald. J. M.. Vomero. J. J., and Laughlin. J. S. Imaging of the human heart after administration of u-(N-13)glutamate. J. NucÃ.Med., 21: 988-991. 1980. 23. Chan. R. C., Babbs, C. F.. Vetter. R. J.. and LÃ¡mar.C. H. Abnormal response ACKNOWLEDGMENTS of tumor vasculature to vasoactive drugs. J. Nati. CÃ¡ncerInst.. 72: 145-150, 1984. The authors thank the staff of the Memorial Sloan-Kettering Cancer 24. Oikawa, M.. Milne, E. N. C., Whitmore, E., Gilday, D., and Oliver, C. A Center Cyclotron Core Facility for the production of 13N. morphological and quantitative study of tumor blood flow. Part I. During growth and immune rejection. Cancer (Phila.). 35: 385-398, 1975. 25. Penning, J. J., and LeVan, J. H. A modified enzymatic technique for REFERENCES production of cell suspensions of murine fibrosarcoma. J. Nati. Cancer Inst., 66:85-87. 1981. 1. Reiman. R. E., Huvos, A. G.. Benua. R. S., Rosen. G.. Gelbard, A. S.. and 26. Cullino. P. M., and Grantham. F. H. Studies on the exchange of fluids Laughlin. J. S. Quotient imaging with N-13-L-glutamate in osteogenic sar between host and tumor. I. A method of growing "tissue-isolated" tumors in coma: correlation with tumor viability. Cancer (Phila.). 48; 1976-1981.1981. laboratory animals. J. Nati. Cancer Inst., 27: 679-693, 1961. 2. Reiman, R. E.. Rosen. G.. Gelbard. A. S.. Benua. R. S.. and Laughlin. J. S. 27. Woodard. H. Q.. Bigler, R. E.. Freed. B., and Russ. G. Expression of tissue Imaging of primary Ewing sarcoma with N-13 i.-glulamate. Radiology. 142: isotope distribution. J. NucÃ.Med., 16: 958-959, 1975. 495-500. 1982. 28. Clark. J. C.. and Buckingham. P. P. Short-Lived Radioactive Gases for 3. Reiman, R. E., Benua. R. S.. Gelbard. A. S., Allen. J. C., Vomero, J. J.. and Clinical Use, pp. 227-231. Boston: Butterworth. 1975. Laughlin, J. S. Imaging of brain tumors after administration of N-13 i - 29. Graser, T. A., Godei, H. G.. Albers, S., FÃ¶ldi,P., and FÃ¼rst,P. An ultra glutamate. J. Nuel. Med., 23: 682-687, 1982. rapid and sensitive high-performance liquid Chromatographie method for 4. Sordillo, P. P., Benua, R. S., Gelbard. A. S., Bading, J. R., Reiman, R. E., determination of tissue and plasma free amino acids. Anal. Biochem., 151: Magill, G. It., and Laughlin. J. S. Imaging of tumors and organs with N-13 142-152, 1985. labeled L-methionine. Am. J. Physiol. Imaging. /: 193-200. 1986. 30. Kerscher. L.. and Ziegenhorn. J. Amino acid-related compounds. Methods 5. Knapp. W. H.. Helus. F., Sinn, H.. Ostertag. H.. Georgi. P.. Brandeis. W- Enzymatic Anal., 7:444-453. 1984. E.. and Braun. A. N-13 L-Glutamate uptake in malignancy: its relationship 31. Rej, R.. and Horder. M. Aspartate aminotransferase. Methods Enzymatic to blood now. J. NucÃ.Med., 25: 989-997. 1984. Anal.. 3: 416-433, 1983. 6. Reiman. R. E., Rosen, G., Gelbard, A. S.. Benua, R. S.. Yeh, S. D. J., and 32. Horder. M., and Rej. R. Alanine aminotransferase. Methods Enzymatic Laughlin, J. S. Diagnostic demands in clinical and experimental oncology: Anal.. 3: 444-456. 1983. application of substrates labeled with positron-emitting radionuclides. In: W. 33. Schmidt. E., and Schmidt, F. W. Glutamate dehydrogenase. Methods Enzy H. Knapp and K. Vyska (eds.). Current Topics in Tumor Cell Physiology matic Anal., 3: 216-227, 1983. and Positron-Emission Tomography, pp. 73-85. Berlin: Springer-Verlag, 34. Kvamme, E., Torgner, I. A. A., and Svenneby, G. Glutaminase from mam 1984. malian tissues. Methods Enzymol., 113: 241-256, 1985. 7. Nieves, E., Rosenspire. K. C.. Filc-DeRicco. S.. Gelbard. A. S.. and Cooper, 35. Herbert. J. D.. Coulson. R. A., and Hernandez. T. Free amino acids in the A. J. L. A high performance liquid Chromatographie on-line flow-through caiman and rat. Comp. Biochem. Physiol., 17: 583-598. 1966. radioactivity detector system for analyzing amino acids and metabolites 36. Parry. D. M., Hall. B., and Brosnan. J. T. Renal metabolite concentrations labeled with nitrogen-13. J. Chromatogr.. 383: 325-337. 1986. and the activities of glutaminase and glutamate dehydrogenase during recov 8. Cooper, A. J. L., McDonald. J. M., Gelbard, A. S., Gledhill. R. F., and ery from metabolic acidosis in the rat. Can. J. Biochem.. 59: 871-876. 1981. Duffy. T. E. The metabolic fate of 13N-labeled ammonia in rat brain. J. Biol. 37. Pine, M. J., Kim, U., and Ip. C. Free amino acid pools of rodent mammary Chem., 254: 4982-4992. 1979. tumors. J. Nati. Cancer Inst.. 69: 729-735, 1982. 9. Cooper, A. J. L.. Mora, S. N., Cruz, N. F., and Gelbard, A. S. Cerebral 38. Schoolwerth, A. C., and LaNoue, K. F. The role of microcompartmentalion ammonia metabolism in hyperammonemic rats. J. Neurochem.. 44: 1716- in the regulation of glutamate metabolism by rat kidney mitochondria. J. 1723. 1985. Biol. Chem.. 255: 3403-3411,1980. 10. Cooper, A. J. L., Nieves, E., Coleman. A. E., Filc-DeRicco, S., and Gelbard, 39. Burch. H. B.. Choi. S.. McCarthy, W. Z., Wong, P. Y., and Lowry, O. H. A. S. Short-term metabolic fate of [13N]ammonia in rat liver in vivo. J. Biol. The location of glutamine synthetase within the rat and rabbit ncphron. Chem., 262: 1073-1080. 1987. Biochem. Biophys. Res. Commun.. Â«2:498-505, 1978. 11. Cooper, A. J. L., Nieves. E.. Rosenspire, K. C.. Filc-DeRicco, S.. Gelbard, 40. Pitts, R. F. Renal production and excretion of ammonia. Am. J. Med., 36: A. S., and Brusilow. S. W. Short-term metabolic fate of N-13-labeled gluta 720-742, 1964. mate, alanine and glutamine (amide) in rat liver. J. Biol. Chem., 263: 12268- 41. Nash, T. P., Jr.. and Benedict. S. R. The ammonia content of the blood and 12273, 1988. its bearing on the mechanism of acid neutralization in the animal organism. 12. Cooper. A. J. L., Nieves. E.. Filc-DeRicco, S.. and Gelbard, A. S. Short-term J. Biol. Chem.. 48: 463-488. 1921. 4843

42. HÃ¤ussinger, D. Glutamine metabolism in the liver: overview and current York: John Wiley & Sons, Inc., 1985. concepts Metabolism, 38: (Suppl 1): 14-17, 1989. 46- Greenhouse, W. V. V., and Lehninger, A. L. Occurrence of the malate- 43. Windmueller, H. G., and Spaeth, A. E. Source and fate of circulating aspartate shuttle in various tumor types. Cancer Res., 36: 1392-1396. citrullme.citrulline AmAm.Jl-nysiol.. J Phvsiol 241-Â¿Â«/.h4M E473-E480 L48U. 1981iyÂ»l. 47. Greenhouse,reduced nicotinamide W. V. V., adenine and Lehninger, dinucleotide A. L. shuttle Magnitude activity of in malate-aspartate intact respiring 44. Brosnan.J. T. The role of the kidney in ammo acid metabolism and nutrition. tumor cells Cancer Res Â¿7-4173-4181 1977 Can. J. Physiol. Pharmacol., 65: 2355-2362, 1987. 48 Cooper, A. J. L., Fitzpatrick, S. M., Ginos, J. Z., Kaufman, C., and Dowd, 45. Cooper, A. J. L., and Meister. A. Metabolic significance of transamination. p. E. Inhibition of glutamate-aspartate transaminase by (3-methylene-DL- Â¡n:P.Christen and D. E. Metzler (eds.), Transaminases, pp. 533-563. New aspartate. Biochem. Pharmacol., 32: 679-689, 1983.

4844

Sabina Filc-DeRicco, Alan S. Gelbard, Arthur J. L. Cooper, et al.

Cancer Res 1990;50:4839-4844.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/50/16/4839

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/50/16/4839. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.