ORIGINAL ARTICLE 
Annals of Nuclear Mediclne Vol. 11, No. 2, 115-122, 1997 
Carbon-11 labeled ethionine and propionine as tumor detecting agents 
Kiichi ISHIWATA,* Chiaki KASAHARA,* Kentaro HATANO,** Shin-ichi ISHII* and Michio SENDA* 
*Positron Medical Center, Tokyo Metropolitan Institute of Gerontology **Cyclotron Research Center, Iwate Medical University 
To develop 18F-fluoroalkyl derivatives of methionine (MET) as a tumor detecting agent by mean of clinical PET, a pilot study assessing the potential of their parent compounds, 11C-labeled ethionine (11C ETH) and propionine (11C PRO) was performed. 11C-ETH and 11C-PRO were prepared by the reaction of L-homocysteine thiolactone and corresponding 11C-alkyl iodides. After i,v. injection of a mixture of 3H-MET, 14C-ETH and 11C-PRO into mice bearing FM3A mammary carcinoma, the highest FM3A uptake was found in 14C-ETH, followed by 3H-MET and 11C-PRO, while the FM3A-to-brain and FM3A-to-muscle ratios were nearly the same for all three compounds. The FM3A uptake of 14C-ETH and 11C-PRO were nearly equal or slightly higher than the liver uptake. In the pancreas, liver, FM3A and brain tissues, incorporation of 14C-ETH into acid-precipitable materials was much lower than that of 3H-MET, whereas no incorporation of 11C-PRO was found. Brain uptake of all three compounds was significantly reduced by carrier MET-loading (5 min p.i.) or by cycloheximide treatment to inhibit protein synthesis (60 min p.i.), whereas the FM3A uptake was not affected. Incorporation of 14C-ETH into acid-precipitable materials was inhibited by the cycloheximide. The results suggest that 11C-labeled ETH has a similar potential for tumor detection by PET as 11C-MET, and that 11C-PRO has similar properties to those of other artificial amino acids. The development of 18F-fluoroalkyl derivatives of MET is of interest as the next step. 
Key words: [11C]ethionine, [11C]propionine, tumor detection, PET 
INTRODUCTION 
POSITRONN EMISSION TOMOGRAPHY (PET) with appropriate positron-emitting amino acids is now an established in vivo means of measuring amino acid metabolism in tissues such as the brain and tumors. Among the many amino acids prepared,1 L-[methyl-11C]methionine (11C-MET) is widely used for clinical PET studies.2-10 Indeed, it is considered that 11C-MET is a useful radiopharmaceutical for tumor diagnosis because of its simple preparation in 11C-labeling as well as favorable biological properties of MET in vivo. Rat tumor uptake of 11C-MET was higher than that of other amino acids,11 resulting in tumor detection with high contrast to surrounding normal tissue such as brain and lung. From the biological point of view, the label of 11C-MET is not only incorporated into proteins but also into non-protein materials such as lipids and RNA fractions.12,13 The pathway to transmethylation processes possibly indicates advantageous properties in 11C-MET compared with other amino acids. On the other hand, 18F-labeled amino acids have practical benefits for tumor diagnosis in clinical PET studies. Because a large amount can be produced with 18F-anion, single production of the tracer is sufficient for several patients. It is also possible that the radiopharmaceuticals with a longer half-life can be delivered to a distant PET clinic from the production site. The MET analogs which contain the 18F-labeled alkyl group used in the transalkylation process in vivo are therefore of interest for tumor detection. Methyl-18F-labeled MET [2-amino-4-(fluoromethylthio)butylic acid], is a candidate compound; but is reported to be unstable.14 Fluoroalkyl derivatives of MET may be another candidate. It is reported that ethionine [2-amino-4-(ethylthio)butylic acid], an S-ethyl analog of MET, is incorporated into proteins and that the ethyl group is also used as an ethyl donor via the transethylation process as in the case of the methyl group of MET.15,16 
The aim of this study is to prepare 11C-labcled ethionine (11C-ETH) and propionine L2-amino-4-(propylthio)butylic acid, 11C-PRO] as parent compounds of 18F-fluoroalkyl derivatives of MET and to assess their potential for tumor detection. The labeled compounds were prepared by the reaction of L-homocysteine thiolactone and correspond-ing 11C- or 14C-labeled alkyl iodides. In tumor bearing mice, tumor uptake of L-[methyl-3H]methionine (3H-MET), L-[ethyl-14C]ethionine (14C-ETH) and L-[propyl-11C]propionine (11C-PRO) and their incorporation into macromolecular materials were measured. Biological properties of ETH have been reviewed,15,16 and distribution of L-[1,2-ethyl-14C]ethionine in normal rats over three days after the tracer injection was investigated by mean of whole-body autoradiography17 and tissue sampling,18 but the tumor accumulation of the compound in animals is unknown. On the other hand, no information is available for PRO. 
MATERIALS AND METHODS 
3H-MET (specific activity of 2.59 GBq/umol) and [1-14C]ethyl iodide (specific activity of 1.85 MBq/umol) were purchased from American Radiolabeled Chemicals Inc. (St. Louis, MO, USA). Methylmagnesium bromide (3 M solution in diethyl ether) ethylmagnesium bromide (1 M solution in THF). LiAlH4 (1.0 M solution in THF) and THF, which were specially packed in Sure/SealTM bottles, were purchased from Aldrich Japan Inc. (Tokyo), L-homocysteine thiolactone was from Sigma-Aldrich Japan (Tokyo). L-methionine, L-ethionine and cycloheximide were from Wako Pure Chemical Industries, Ltd. (Tokyo) and Soluen-350 was from Packard Instrument Company. Inc. (Meriden, USA). Male C3H/He mice were supplied by Tokyo Laboratory Animals Co., Ltd. (Tokyo). 
Synthesis of 11C-ETH, l4C-ETH and 11C-PRO 
Carbon-11 labeled ethyl and propyl iodides were prepared by the method of Langstrom et al.19 Briefly, 0.15 mL THF solution of 1 M methylmagnesium or ethylmagnesium bromide was carbonated with 11CO2. To the reaction mixture 0.05 mL of 1 M LiAlH4 was added. After dryness of THF, 0.5 mL of HI was added and the solution was heated. The 11C-ethyl iodide or 11C-propyl iodide generated was transferred with a N2 flow into acetone or ethanol at -40degC. To the solution, aqueous L-homocysteine thiolactone solution and NaOH solution were added. The final 0.5 mL reaction mixture in 75%, 50% or 25% of acetone or ethanol solution contained 2.5 mg of L-homocysteine thiolactone and 0.1 mmol NaOH. The mixture was heated at 60degC for 5 min. One-mL of 0.1 M HCl was then added to the reaction mixture, and the solution was subjected to HPLC separation with a Megapak SIL C18 column (10 mm i.d. x 250 mm, Japan Spectroscopic Co. Ltd., Tokyo). The mobile phase for the separation of 11C-ETH was physiological saline. 11C-ETH and a byproduct 11C-MET were eluted at 5.9-6.9 min and 2.9-3.3 min, respectively, at a flow rate of 10 mL/min. In the case of 11C-PRO the mobile phase was a mixture of CH3OH and 5 mM HCO2NH4 (1/9, v/v), and 11C-MET and 11C-PRO were eluted at 2.5-2.7 min and 6.5-7.5 min, respectively, at a flow rate of 10 mL/min. The 11C-PRO fraction was collected and evaporated to dryness. The 11C-PRO was dissolved in physiological saline followed by membrane filtration and used for the animal study. 
Carbon-14 labeled ETH was synthesized by the reaction of L-homocysteine thiolactone and 14C-ethyl iodide in 2 mL of 50% aqueous acetone at 60degC for 5 min. After adding HCl to the reaction mixture, unreacted 14C-ethyl iodide and acetone were removed with a N2 flow at 60degC. The 14C-ETH was separated under the same HPLC conditions as in the case of 11C-ETH. The radiochemical yield was 67%. The radiochemical and enantiomeric purity analyzed as described below were > 99% and 90%, respectively. 
Radiochemical purity was analyzed by HPLC on a Crestpak C18S-10 column (4.6 mm i.d. x 150 mm, Japan Spectroscopic Co. Ltd.). The retention times for 11C-MET and 11C-ETH were 3.0 min and 5.4 min, respectively, with 10 mM HCO2NH4 as the elution solution at the flow rate of 1 mL/min. 11C-PRO was eluted at 9.8 min with a mixture of CH3OH and 10 mM HCO2NH4(5/95, v/v) at a flow rate of 1 mL/min. 
Enantiomeric purity was analyzed on a Crownpak CR (+) column (4.0 mm i.d. x 150 mm, Daicel Chemical Industries, Tokyo). The elution solution was 0.01 N HClO4, pH 2.0 and a flow rate was O.8 mL/min. The retention times were: D- and L-enantiomers of MET, 3.4 min and 5. I min; D- and L-enantiomers of ETH, 5.4 min and 9.2 min; and D- and L-enantiomers of PRO, 11.4 min and 20.9 min. 
Synthesis of PRO 
PRO was synthesized by the reaction of 10 mg (65 p:mol) of L-homocysteine thiolactone and 60 /uL (620 ,umol) of propyl iodide in 1 mL of 50% aqueous acetone at 60degC for 5 min. After adding HCl, the reaction mixture was evaporated to dryness, and the residue was dissolved in 2 mL of water. The solution was loaded onto a YMC-pack ODS-A column (20 mm i,d. x 150 mm. SH-342-5, S-5 120 A, YMC Co. Ltd., Kyoto), and eluted with a mixture of CH3OH and 5 mM HCO2NH4 (1/9, v/v) at a flow rate of 10 mL/min. The PRO fraction was collected and evaporated to dryness. After this purification procedure was repeated once, the PRO fraction was lyophilized to remove concomitant HCO2NH4. 
PRO was identified by 1H-NMR spectroscopy. 1H-NMR spectra were recorded with a JNM-EX90A spectrophotometer (JEOL, Tokyo) on 89.45 MHz. Chemical shifts (6) referred to the external TMS standard were expressed as ppm. Samples were dissolved in D20 containing a few drops of DCl. 1H-NMR data: PRO delta 0.89 (t, J = 7.1 Hz, 3H), 1.42-1.74 (m, 2H), 2.12-2.31 (m, 2H), 2.44-2.77 (m, 4H), 4.19 (t, J = 6.1 Hz, 1H); ETH delta 1.18 (t, J = 7.4 Hz, 3H), 2.14-2.80 (m, 6H), 4.21 (t, J = 6.4 Hz, 1H); and MET delta 2.09 (s, 3H), 2.13-2.77 (m, 4H), 4.23 (t, J = 6.2 Hz, 1H). 
Tissue distribution study 
FM3A mammary carcinoma-bearing mice weighing 24.3+-1.1 g were prepared as described previously.20 A mixture of 3H-MET (150 kBq/0.058 nmol), 14C-ETH (30 kBq/16 nmol) and 11C-PRO (5 MBq) was injected intravenously into the mice. They were killed by cervical dislocation 5, 15, 30 and 60 min post injection. Blood was removed by heart puncture with a heparinized syringe. Plasma was separated by centrifugation. Brain, liver, pancreas, muscle and FM3A tissues were dissected. The tissue uptake of the three radionuclides was expressed as the % injected dose per g of tissue (%ID/g), as described 12,13 previously. 
The second group of mice were injected with a mixture of three amino acids together with carrier MET (25 umol/ animal), and killed 5 min post injection. 
The third group of mice were given intraperitoneally cycloheximide (100 mg/kg body weight) dissolved in physiological saline.12,13 Thirty minutes later, the mixture of labeled amino acids was injected into the mice, which were killed 60 min post injection. The tissue uptake of three radionuclides was measured as described above. 
The animal studies were approved by the Animal Care and Use Committee of Tokyo Metropolitan Institute of Gerontology. 
Incorporation of 3H-MET, 14C-ETH and 11C PRO into acid-precipitable materials 
The incorporation of 3H- and 14C-radioactivity into acid-precipitable materials was measured in samples of brain, tumor, pancreas and liver tissues and plasma as described previously.12,13 To measure incorporation of 11C-radioactivity, only 11C-PRO (5 MBq) was injected intravenously into another group of mice (n = 3). The incorporation of 11C-radioactivity into acid-precipitable materials was measured as described previously.12,13 
RESULTS 
Figure 1 summarizes the results of the radiochemical synthesis. The radiochemical yields of 11C-ETH and 11C-PRO increased with the increasing water contents. The yields were higher in the aqueous ethanol than in the aqueous acetone. Radiochemical purity was > 99%. The enantiomeric purity of the three amino acids also de-creased with the decreasing water content, and was slightly higher in the aqueous acetone than in the aqueous ethanol. 
The tissue distribution of 3H-MET, 14C-ETH and 11C-PRO in the tumor bearing mice is summarized in Table 1. In the plasma, the levels of the 3H- and 14C-radioactivity decreased for the first 30 min, after which the level of the 3H increased, but the level of the 14C remained constant. The level of 11C-radioactivity decreased with time. The pancreas showed the highest uptake of 3H-MET and 14C-ETH, whereas the highest uptake of 11C-PRO was found in the kidneys. The liver uptake of 14C-ETH and 11C-PRO was significantly lower than that of 3H-MET. The uptake of all three tracers is higher in the tumor than in the brain and muscles. In these three tissues, the highest uptake was found in 14C-ETH, followed by 3H-MET and 11C-PRO, but the tumor to brain and tumor to muscle uptake ratios were similar for all three tracers, and only the tumor to brain and tumor to plasma ratios for 11C-PRO at 60 min were larger than that for 3H-MET (Table 2). The tumor to liver ratios for 14C-ETH and 11C-PRO were nearly unity. 
The effects of carrier MET-loading and cycloheximide-treatment on the tissue distribution of the three tracers are summarized in Figs. 2 and 3, respectively. Only the brain uptake of all three tracers was significantly decreased by the co-injection of carrier MET. Following the cycloheximide treatment, the uptake of all three tracers significantly decreased in the brain and pancreas. In the liver, the uptake of 3H-MET and 14C-ETH was enhanced by cycloheximide, but that of 11C-PRO decreased. Neither treatment affected the tumor uptake of any of the three tracers. 
Figure 4 shows the incorporation of 3H-MET and 14C-ETH into the acid-precipitable materials at 60 min after injection. Pretreatment with cycloheximide greatly reduced the acid-precipitable fraction of 3H-MET in the brain, tumor, pancreas and plasma. This fraction of 14C-ETH was lower than that of 3H-MET, but it decreased after the pretreatment with cycloheximide. In the liver, the cycloheximide slightly reduced the acid-precipitable fraction of 3H-MET, and enhanced that of 14C-ETH. No incorporation of 11C-PRO into the acid-precipitable fraction was found in the brain, tumor or plasma. 
DISCUSSION 
In the reaction of L-homocysteine thiolactone and 11C-methyl iodide, D-enantiomer of 11C-MET was produced.21 Similar results were also found in the synthesis of 11C-ETH and 11C-PRO. Aqueous acetone was preferable in order to maintain high enantiomeric purity. We found that the D-form of 11C-ETH reached 43% in the DMF solution (data not shown). In HPLC separation, physiological saline was successfully used as an eluting solution to separate 11C-ETH as in the case of 11C-MET,22 but, this was not applicable to the separation of 11C-PRO (retention time, 17-19.5 min). 
In the tumor-bearing mice, the pattern of distribution of 14C-ETH in the normal organs was similar to that in normal rats described previously.16,17 As for the tumor uptake of 14C-ETH and 11C-PRO, several interesting characteristics as methionine analogs were found. Although the tumor uptake increased in the order of 11C-PRO, 3H-MET and 14C-ETH, the tumor to brain and tumor to muscle uptake ratios were similar for all three tracers. The radioactivity levels of 14C-ETH and 11C-PRO in the brain and tumor may depend on the corresponding protein-free radioactivity in the plasma. An advantage of 14C-ETH and 11C-PRO is the low uptake by the liver and pancreas compared with 3H-MET: the order of the liver to tumor uptake ratio, 11C-PRO (>=1), 14C-ETH (=1) and 3H-MET (1>>), suggesting practicability for tumor imaging in the abdominal region. Since 11C-PRO did not follow the metabolic pathways of methionine as described below, it was cleared from the liver and other organs and excreted into urine via the kidneys, but the 11C-PRO was retained in the amino acid pool of tumor tissues. A similar characteristic was also found in other artificial amino acids.11 
The reduced brain uptake of the three tracers caused by carrier methionine loading suggests these three amino acids are taken by the same transport system across the blood-brain barrier. On the other hand, the tumor uptake was not affected. These results also suggest that the contrast between the tumor and the surrounding normal brain tissue is enhanced by the methionine-loading. 
It has been reported that the 14C-ETH was incorporated into proteins and other macromolecules via the transethylation process.15,16 As shown in Fig. 4, we confirmed the incorporation of 14C-ETH into acid-precipitable materials, but the fraction was small compared with that of 3H-MET. The cycloheximide treatment, which inhibited protein synthesis in vivo (Fig. 4 and references 12 and 13), reduced the brain uptake of all three tracers but not the tumor uptake (Fig. 3). A similar conflicting response to the cycloheximide treatment between the brain and tumor is also observed for other natural amino acids including L-12 13 leucine12,13 and L-tyrosine.23 Furthermore, the cycloheximide-treatment significantly reduced the incorporation of 14C-ETH into acid-precipitable materials in the pancreas and tumor. In the liver the cycloheximide-treatment enhanced the incorporation, suggesting the incorporation of the radioactivity into non-protein materials such as phospholipids, as shown in the case of 3H-MET.23 These facts demonstrate that the 14C-ETH was incorporated not only into proteins but also into non-protein materials via the transmethylation process. On the other hand, 11C-PRO was not incorporated into the acid-precipitable materials. 
It is pointed out that ETH induces liver cancer as chronic biological effects, although it shows signs of similar metabolism to that of MET and inhibits tumor growth.16 Nevertheless, the addition of extra methionine completely counteracts the carcinogenic activity and every other biochemical and morphologic effect of the ETH,16 In PET studies, the estimated injected dose of 11C-ETH is very low in inducing even acute biological effects16 because of the high specific activity of the 11C. In addition, the PET studies are usually limited within a few times. 11C-ETH would therefore be used for the tumor diagnosis under careful supervision in which subjects are given a sufficient amount of the natural amino acid containing MET immediately after the PET studies. 
The present study shows that 14C-ETH but not 11C-PRO has similar in vivo characteristics to 3H-MET, suggesting that 11C-labeled ETH possibly has a similar potential for tumor detection by PET as 11C-MET. 11C-PRO showed signs of similar properties to other artificial amino acids11 reflecting the amino acid transport. Both 11C-ETH and 11C-PRO may be of interest for tumor imaging in the abdominal region from the point of view of the tumor-to-organ uptake ratios. 18F-fluoroalkyl derivatives of MET are therefore of interest as a next step in the development of 18F-labeled amino acids for tumor detection in clinical PET studies. In view of the similar van der Waals radiuses of fluorine and hydrogen, a 18F-fluoroethyl derivative of ETH [18F-2-amino-4-(fluoroethylthio)butylic acid] is an ETH analog, as in the case of 18F-2-deoxy-2-fluoro-D-glucose as a 2-deoxy-D-glucose analog. On the other hand, when taking account the electronegativity of fluorine, methyl-18F-labeled MET [18F-2-amino-4-(fluoromethylthio)butylic acid] and ethyl-18F-labeled ETH may correspond to ETH and PRO, respectively, as in the case of 18F-5-fluoro-2'-deoxyuridine as a thymidine analog. 
ACKNOWLEDGMENTS 
This work was supported by Grants-in-Aid for Scientific Research on Priority Areas Nos. 07274271 and 08266270 from the Ministry of Education, Science, Sports and Culture. Japan. 
REFERENCES 
1. Vaalburg W, Coenen HH, Crouzel C, Elsinga PH, Langstr6m B, Lemaire C, et al. Amino acids for the measurement of protein synthesis in vivo by PET. Nucl Med Biol 19: 227-237, 1992. 
2. Bustany P, Chatel M, Derlon JM, Darcel F, Sgouropoulos P, Soussaline F, et al. Brain tumor protein synthesis and histological grade: a study by positron emission tomography (PET) with C11-L-methionine. J Neuro-Oncol 3: 397-404, 1986. 
3. Schober O. Duden C, Meyer G-J. Muller JA, Hundeshagen H. Non selective transport of [11C-methyl]-L- and D-methionine into a malignant glioma. Eur J Nucl Med 13: 103-105, 1987. 
4. Bergstr6m M, Ericson K, Hagenfeldt L, Mosskin M, von Holst H, Nor6n G, et al. PET study of methionine accumulation in glioma and normal brain tissue: competition with branched chain amino acids. J Comput Assist Tomogr 11: 208-213, 1987. 
5. Lilja A, Lundqvist H, Olsson Y, Sphnnare B, Gullberg P, Langstrom B. Positron emission tomography and computed tomography in differential diagnosis between recurrent or residual glioma and treatment-induced brain lesions. Acta Rediol 30: 121-128, 1989. 
6. Hatazawa J, Ishiwata K, Itoh M, Kameyama M, Kubota K, Ido T, et al. Quantitative evaluation of L-[methyl-C-11]methionine uptake in tumor using positron emission tomography. J Nucl Med 30: 1809-1813, 1989. 
7. O'Tuama L, Phillips PC, Smith QR, Uno Y, Dannals RF, Wilson AA, et al. L-Methionine uptake by human cerebral cortex: maturation from infancy to old age. JNucl Med 32: 16-22, 1991. 
8. Daemen BJG. Zwertbroek R, Elsinga PH, Paans AMJ, Doorenbos H. Vaalburg W. PET studies with L-[1-11C]tyrosine, L-[methyl-11C]methionine and 18F-fluoro-deoxyglucose in prolactinomas in relation to bromocryptine treatment. Eur J Nucl Med 18: 453-460, 1991. 
9. Leskinen-Kallio S, N~gren K. Lehikoinen P. Routsalainen U, Teras M, Joensuu H. Carbon-11-methionine and PET as an effective method to image head and neck cancer. J Nucl Med 33: 691-695, 1992. 
10. Ogawa T, Shishido F, Kanno I, Inugami A, Fujita H, Murakami M, et al. Cerebral glioma; evaluation with methionine PET. Radiolo 186: 45-53, 1993. 
11 . Kubota K, Yamada K, Fukuda H, Endo S, Ito M, Abe Y, et al. Tumor detection with carbon-11-labelled amino acids. Eur J Nucl Med 9: 136-140, 1984. 
12. Ishiwata K, Kubota K, Murakami M, Kubota R, Senda M. A comparative study on protein incorporation of L-[methyl-3H]methionine, L-[1-14C] leucine and L-2- [18F] fluorotyrosi ne in tumor bearing mice. Nucl Med Biol 20: 895-899, 1993. 
13. Ishiwata K, Kubota K, Murakami M, Kubota R, Sasaki T, Ishii S, et al. Re-evaluation of amino acid PET studies: can the protein synthesis rates in brain and tumor tissues be measured in vivo? J Nucl Med 34: 1936-1943, 1993. 
14. Janzen AF, Wang PMC, Lemire AE. Fluorination of methionine and methionylglycine derivatives with xenon difluoride. J Fluorine Chem 22: 557-559, 1983. 
15. Alix J-H. Molecular aspects of the in vivo and in vitro effects of ethionine, an analog of methionine. Microbiol Rev 46: 281-295, 1982. 
16. Farber E. Ethionine carcinogenesis. Adv Cancer Res 7: 383-474, 1982
17. Durand S, Benard P. Whole-body autoradiographic study of the distribution of radioactivity in rats given L-[1,2-ethyl-14C ethionine. Toxicol Appl Pharmacol 53: 204-210, 1980. 
18. Durand S, Lloveras J, Thouvenot JP, Douste-Blazy L. Comparative diffusion of L-14C-ethionine and 14C-methionine in rat tissue. C R Seances Soc Biol Fil 5169: 1622-1626, 1975. (in French) 
19. Langstrom B, Antoni G, Gullberg P, Halldin C, Nagren K, Rimland A, et al. The synthesis of L-11C-labelled ethyl, propyl, butyl and isobutyl iodides and examples of alkylation reactions. Appl Radiat Isot 37: 1141-1145, 1986. 
20. Kubota K. Ishiwata K. Tada M, Yamada S. Kubota R. Iwata R, et al. Investigation of metastatic potential with N-[18F]fluoroacetyl-D-glucosamine. Nucl Med Biol 19: 747-752, 1992. 
21. Ishiwata K, Ido T, Vaalburg W. Increased amount of D-enantiomer dependent on alkaline concentration in the synthesis of L-[methyl-11C]methionine. Appl Rediat Isot 39: 311-314, 1988. 
22. Ishiwata K, Ishii S, Senda M. HPLC separation using physiological saline for quality control of radiopharmaceuticals used in PET studies. Appl Rediat Isot 44:1119-1124,1993. 
23. Ishiwata K. Enomoto K, Sasaki T, Elsinga PH, Senda M, Okazumi S, et al. A feasibility study on L-[1-11C]tyrosine and L-[methyl-llC]methionine to assess liver protein syn-thesis by PET. J Nucl Med 37: 279-285, 1996.