TECHNICAL NOTES 
Annals of Nuclear Medicine Vol. 9, No,1, 51-57, 1995 

Synthesis and preliminary evaluation of [1-11C]hexanoate as a PET tracer of fatty acid metabolism 

Kiichi ISHIWATA,* Kenji ISHII,* Koji OGAWA,** Toru SASAKI,* Hinako TOYAMA,* Shin-ichi ISHII,* 
Tadashi NOZAKI** and Michio SENDA* 

*Positron Medical Center, Tokyo Metropolitan Institute of Gerontology 
**Faculty of Hygienic Sciences, Kitasato University 

The potential of [1-11C]hexanoate (11C-HA) as a radiopharmaceutical assessing fatty acid metabolism of the myocardium and brain tissues by PET studies was evaluated. 11C-HA was synthesized by the Grignard reaction of pentylmagnesium bromide and 11CO2. 11C-HA, [1-14C]acetate and [3H]deoxyglucose were simultaneously injected i.v, into mice, and the tissue distribution of the three radionuclides was measured. In the heart, high uptake and rapid clearance of 11C and 14C was found. The brain uptake of 11C was twice as high as that of 14C, and both 11C and 14C decreased slowly compared to the heart. The level of 3H increased with time in both the heart and brain. In fasting conditions, the uptake of 11C by the heart was enhanced and the level of 3H decreased with time. The brain uptake of 11C and 3H was also enhanced. The fasting conditions did not affect the distribution of 14C. The radiation absorbed dose of 11C-HA was also estimated. 

Key words: [1-11C]hexanoate, heart, brain, oxidation, PET 
INTRODUCTION 

RADIOLABELED FATTY ACIDS have been used to determine the myocardial metabolism. 1-5 In positron emission tomography (PET), carboxyl-11C-labeled long straight-chain fatty acids are used as a probe to measure the rate of poxidation of fatty acids in the heart, in which high extraction and rapid clearance of the radioactivity was found. On the other hand, B-methylated fatty acids are expected to be resistant to B-oxidation and to be retained in the heart.6-10 In contrast to B-methyl fatty acids, a-methyl fatty acids are not resistance to B-oxidation. Therefore, no attention has been paid to the application of the a-methyl fatty acids as a tracer of fatty acid metabolism. 

It is considered that glucose is the only fuel in the brain and that fatty acids are not an energy source, but mitochondria in the brain have ability to oxidize fatty acids.11 The B-oxidation process of the fatty acids in the brain may play a role in regulating lipid metabolism. When radio-labeled long-chain fatty acids such as palmitate were injected into rats, a part of the radioactivity in the brain was incorporated into the lipid fraction and the rest became water-soluble materials probably via the B-oxidation process.12,13 On the other hand, a short-chain fatty acid octanoate was scarecely incorporated into the lipids and its major metabolites were glutamate and glutamine in the rat brain.14 
Recently, Ogawa et al, synthesized radiolabeled a-methyl fatty acids with carbon-chain length C2 to C16 by malonic ester synthesis.15,16 They found that a-[14C]methylated short-chain fatty acids and [3-14C]propionic acid were taken up by the brain and the level of radioactivity was retained for a while. The highest brain uptake was observed in C6 form, a-methyl hexanoic acid (MHA). When MHAs Iabeled at two different sites, a-methyl and carboxyl 14C-labeled MHAs, were administered to mice, the time-radioactivity curves for the two compounds in the brain were quite different. The level of radioactivity of the [methyl-14C]MHA in the brain increased for the first 20 min, then decreased gradually, whereas that of [1-14C]MHA decreased rapidly with time after the injection similar to straight-chain [1-14C]hexanoic acid. This contrast in the time-radioactivity profile corresponded closely to the difference between [3-14C]propionic acid and [1-14C]propionic acid. The level of the radioactivity of the [methyl-14C]MHA was even higher than that of the [3-14C]propionic acid. We considered that the short-chain fatty acids taken up by the brain were degraded through B-oxidation in the same way in the myocardium, then further metabolized to 14CO2. 

In this study, we synthesized straight-chain [1-11C]hexanoate (11C-HA), and investigated its potential as a radiopharmaceutical for assessing metabolism in the heart and brain by PET. a-Methyl fatty acids are degraded by B-oxidation to propionyl-CoA, which Is altered to succinyl-CoA by three successively enzymic reactions and enters the tricarboxylic acid cycle. On the other hand, straight-chain fatty acids become acetyl-CoA by the p-oxidation, which directly enters the tricarboxylic acid cycle. Fatty acids labeled with a carboxylic group are preferable for the kinetic analysis by PET, because the radioactivity taken up by the heart and brain could disappear depending upon the oxidative metabolism via the tricarboxylic acid cycle in the tissues. Since 2-deoxy-2-[18F]fluoro-D-glucose and [1-11C]acetate are currently used as metabolic radiopharmaceuticals for PET studies, the potential of 11C-HA was investigated compared with [3H]deoxyglucose (3H-DG) and [1-14C]acetate (14C-AA) in normally fed and fasting mice. 

MATERIALS AND METHODS 
Materials 
3H-Deoxy-D-glucose (specific activity of 392 GBq/mmol) and sodium [1-14C]acetate (specific activity of 2.11 GBq/mmol) were purchased from Amersham International plc (England). Pentylmagnesium bromide (2.0 M solution in diethylether) specially packed in Sure/SeaITM bottles was purchased from Aldrich Chemical Co. Inc. NEFA C-test Wako for measuring nonesterified free fatty acids (NEFA) in the plasma was purchase from Wako Chemical Industries Ltd. 

Synthesis of sodium [1-l1C]hexanoate (11C-HA) 
[11C]Carbon dioxide was produced as described previously.17,18 The [11C]CO2 trapped in the copper tube was transferred with a flow of 10 mL/min into 0.2 mL of 2.0 M pentylmagnesium bromide in diethylether at -40-50'C in a CH3OH bath cooled with liquid N2. After the carboxylation, 1 mL of 1.0 M HCI was added into the reaction mixture, which was extracted with 2 mL of diethylether. The ether phase was added into 1 mL of 7% (0.875 M) NaHCO3, then the solution was evaporated to dryness to remove radioactive and non-radioactive volatile materials. The residue was dissolved in physiological saline and the llC-HA solution was passed through a 0.22 um membrane filter. 

The radiochemical and chemical purity of the 11C-HA were analyzed by high-performance liquid chromatography (HPLC). HPLC conditions: column, eluent, flow rate, temperature and retention time of HA, were: Aminex Fermentation Monitoring column (7.8 mm x 150 mm, Bio-Lad), H2O, 1.0 mL/min, 65'C and 14.0 min; and Partisil 10 SAX (4.0 mm x 250 mm, Whatman), NaH2PO4, 2 mL/min, 40'C and 2.8 min. 10 mM 

Biodistribution studies 
Two groups of eight-week-old male ddY mice were used. The first group of mice was allowed free access to food and water (the control group), and the other group of was given only water for 1 day before the experiment (the fasting group). 

A set of control mice were intravenously injected with 11C-HA (8.0 MBq), then killed by cervical dislocation at 5, 15, 30, 60 and 90 min post injection. Blood was removed by heart puncture with a heparinized syringe and the organs were dissected. The samples were counted for 11C in a gamma-counter, and weighed. The tissue uptake of radioactivity was expressed as % injected dose per g tissue. Radiation dosimetry was estimated as described previously.19,20 

Another set of control mice and the fasting mice intravenously injected with a mixture of llC-HA (4.0 MBq), 3H-DG (1.5 kBq) and 14C-AA (1.5 kBq) were killed by cervical dislocation at l,3,5,10,15,30 and 60 min post injection. Blood was removed as described above, and centrifuged to obtain plasma. The heart and brain were dissected and divided into two portions. A part of the heart and cerebral cortex (about 50 to 100 mg) and 20 uL of plasma were dissolved in the tissue solubilizer and counted for 3H and 14C in a liquid scintillation counter after the decay of llC radioactivity. The rest of the tissues and plasma were counted for l1C. The tissue uptake of radioactivity was expressed as the % injected dose per g of tissue. Concentrations of glucose and NEFA in the plasma were measured in a Clinical Chemistry Analyzer (CL760, Shimadzu) and with a NEFA C-test Wako kit (a method using acyl-CoA synthetase and acyl-CoA oxidase), respectively. 

RESULTS 

llC-HA was synthesized by the Grignard reaction. The yield of carboxylation of pentylmagnesium bromide with 11CO2 was dependent upon the flow rate of l1CO2 and temperature. When 11CO2 was introduced into the pentylmagnesium bromide solution at -4O--50'C at 10 mL/ min, the slowest flow rate in our system, followed by HCl hydrolysis, 42.1 % (mean of 4 runs, ranging of 35.0% to 49.6%) of the total 11CO2 was reacted to the carboxylation. At -16'C and 0'C in the carboxylation, the yields were 30.2% and 17.6%, respectively. Over 94% of the radioactivity in the reaction mixture was detected as [llC]HA. Undesirable radioactive and nonradioactive materials, including pentane derived from the pentylmagnesium bromide, were removed by evaporation of the solution to dryness. The decay-corrected radiochemical yield was 27.9% (mean of 4 runs, ranging of 24.9% to 32.5%) based on the [11C]CO2 used. The HPLC analysis showed radio chemical purity greater than 98%. 

Tissue distribution of radioactivity after the injection of [11C]HA into the control mice is summarized in Tables l and 2. The level of radioactivity 5 min after the injection was highest in the kidneys, followed by the spleen, pancreas, lungs, liver, brain and heart. The levels in all tissues except for the pancreas decreased rapidly with time. Four % of the total radioactivity was excreted into urine within 90 min after the injection. 

Three metabolic radiotracers, 11C-HA, 3H-DG and 14C-AA, were given simultaneously to the control and fasting mice. After I day fasting, plasma glucose significantly decreased to 55% of the control, whereas plasma NEFA increased (Table 3). The time course of radioactivity in the plasma, heart and brain are summarized in Figures 1,2 and 3. The levels of the three nuclides in the plasma decreased with time (Fig. 1). The fasting delayed the clearance of all three nuclides. In the heart of the fasting mice (Fig. 2), the initial uptake of 11C-HA was enhanced, and was followed by rapid clearance of the radioactivity as observed in the control, whereas the radioactivity level of 14C-AA was not affected by the fasting. The uptake of 3H-DG decreased with time. In the control mice, the initial brain uptake of the 11C-HA was half of 3H-DG and twice as high as 14C-AA (Fig. 3). The 11C-radioactivity ratios of brain to plasma increased for the first 5 min, remained constant until 30 min and then decreased. Fasting conditions significantly enhanced the brain uptake of all three. The initial uptake during the first 5 min after the injection of 11C-HA and 14C-AA was 1.4-1.5 and 1.3-1.8 times higher than that in the control. The maximal brain uptake of 3H-DG at 10 min was 3.4 times higher in the fasting than in the control. The radiation-absorbed doses are sununarized in Table 4. Because of the short half-life of 11C and the significant amount of expired radioactivity as 11CO2, radiation-absorbed doses were lower than those of 18F-labeled compounds.20 

DISCUSSION 

This study has indicated the potential of 11C-HA as a metabolic tracer for PET studies. High uptake and rapid clearance of 11C-HA were observed in the myocardium, and these were enhanced by fasting. The myocardial uptake of 3H-DG in fasting was significantly decreased. This suggests that six carbon-chain HA is degraded by poxidation in the same way as long carbon-chain fatty acids, such as palmitic acid. In the myocardium, there are several groups of enzymes for B-oxidation, including short-, medium-, long- and very-long-chain acyl-CoA dehydrogenases, that act on straight-chain acyl-CoAs.21 The short- and medium-chain acyl-CoA dehydrogenases in the liver and heart act on hexanoyl-CoA.22-24 Deficiency of these enzymes is known.25,26 In the plasma, long-chain fatty acids (C16 and C18) are predominant com ponents, and short-chain fatty acids are minor.27-29 The myocardial extraction of fatty acids containing 12 to 18 carbon atoms tends to decrease with the increase in the number of carbon atoms in the saturated forms.28 In this work, the extraction of 11C-HA by the heart seemed to be lower than that of palmitate, because the heart/blood ratios for 11C-HA were lower than those for 14C-palmitate.6 The level of radioactivity 5 min after the injection of [1-14C]palmitate was 1.6 times higher than that of 14C-HA (unpublished date). The lower uptake of 11C-HA may be explained by the lower affinity of the fatty acid binding protein for the com pounds. 30-32 In fasting conditions, the blood clearance of 14C-AA was delayed as in case of 11C-HA, although the myocardial uptake and clearance of 14C-AA were not affected.33 Fasting may alter the concentrations of metabolic substrates other than glucose and NEFA in the plasma, as discussed below, and this may affect the blood clearance of 14C-AA. The brain took up 11C-HA, and the radioactivity gradually disappeared with time. It is known that glucose is the only metabolic fuel in the brain. In the case of extremely abnormal circumstance such as starvation, the major fuel source is shifted from glucose to the ketone bodies, acetoacetate, B-hydroxybutyrate and acetone.34-38 Recently, Hasselbalch et al. confirmed this phenomenon in humans exposed to 3.5 days of starvation, by measuring the glucose metabolism by PET with [18F]fluorodeoxyglucose and the arterio-venous (AV) difference of the several metabolic substrates.39 Although the AV difference in B-hydroxybutyrate was greater in the starvation than in the control, the AV differences in acetoacetate, pyruvate, lactate and free fatty acids were negligible. The mean AV difference for the free fatty acids seems greater, but not significantly, because of individual differences. They did not mention fatty acid utilization in the brain. On the other hand, Oldendorf investigated the brain uptake of short-chain monocarboxylic acids .40 He concluded that the transport of monocarboxylic acids across the blood-brain barrier was mediated by a carrier system. Brain uptake increased with the increase in the number of carbon atoms. The brain uptake of hexanoate, octanoate and decanoate was 76%, 94% and 88%, respectively, of 3H-labeled H2O, whereas that of acetate was 14% of 3H-H2O. We confirmed in this study that the brain uptake of 11C-HA was twice as high as that of 14C-AA. We also confirmed that the brain uptake of 14C-HA at 5 min after injection was ten times that of [1-14C]palmitate (unpublished data). When we considered the previous results for fatty acid metabolism in the brain,11-16 the clearance of 11C-HA from the brain probably indicates that 11C is washed out as 11CO2 after the oxidation of 11C-HA. That is, as in the case of B-oxidation in the heart, 11C-HA was metabolized to 11C-acetyl CoA in the brain, which was further degraded to 11CO2 Via the tricarboxylic acid cycle. This process is the same as the oxidative metabolism of 11C-AA in the heart.33,41-43 Otherwise, a part of the 11Cacetyl CoA is possibly utilized for de novo synthesis of amino acids and other compounds,44,45 which was demonstrated for the metabolism of [1-14C]octanoate in the rat brain.14 It is also considered that 11C-HA could not be incorporated into the lipids, as shown in the metabolism of [1-14C]octanate.14 The brain uptake value of 11C-HA (% injected dose/g tissue) suggests that the 11C-HA is taken up sufficiently to image its oxidation process by PET. In fasting conditions, the plasma glucose concentration decreased, but the brain uptake of 3H-DG was enhanced, suggesting that the total uptake of glucose by the brain was preserved. On the other hand, the concentration of NEFA increased and the brain uptake of 11C-HA was enhanced. It is possible that the 1 day fasting enhanced the use of a minor fuel source such as ketone bodies and other substrates. Short-chain fatty acids are minor components in the plasma,27-29 but, 11C-HA may be taken by the brain under low concentrations of glucose in the plasma, via the transport system for B-hydroxybutyrate or monocarboxylic acids discussed above. As another possibility, the delayed clearance of 11C-HA from the plasma resulted in the enhanced uptake. Further studies examining the utilization of 11C-HA in the brain are in progress. 

A method for preparing 11C-labeled short-chain fatty acids such as hexanoate is very convenient compared to that for 11C-labeled palmitate. Carboxylation of the Grignard reagents with 11CO2 is effective, and the reagents are easily removed after hydrolysis by evaporation. The llC-labeled products are soluble in physiological saline, whereas serum albumin or detergent is required to solubilize 11C-palmitate. An improved method with ion-exchange resin instead of ether extraction, which is suitable for routine production of short-chain fatty acids including 11C-acetate, will be described elsewhere. 

ACKNOWLEDGMENTS 

This work was supported by Grant-in-Aid for Scientific Research (C) No. 06807076 from the Ministry of Education, Science and Culture, Japan, and Life Science Foundation of Japan. The authors thank Masaki Shinoda for his technical assistance. 

REFERENCES 

l. Storch-Becker A, Kaiser KP, Feinendegen LE. Cardiac nuclear medicine: positron emission tomography in clinical medicine. Eur J Nucl Med 13: 648-652, 1988. 
2. Jacobson HG. Application of positron emission tomography in the heart. J Am Med Assoc 259: 2438-2445, 1988. 
3. Syrota A, Jehenson P. Complementarity of magnetic resonance spectroscopy, positron emission tomography and single photon emission tomography for the in vivo investigati on of human cardi ac metabolism and neurotransmission. Eur J Nucl Med 18: 897-923, 1991. 
4. Melon P. Schwaiger M. Imaging of metabolism and autonomic innervation of the heart by positron emission tomography. Eur J Nucl Med 19: 453-464, 1992. 
5. Saha GB, Go RT, Macintyre WJ. Radiopharmaceuticals for cardiovascular imaging. Nucl Med Biol 19: 1-20, 1992. 
6. Elmaleh DR, Livni E, Levy S. Varnum D, Strauss HW, Brownell GL. Comparison of l1C and 14C-labeled fatty acids and their ~methyl analogs. Int J Nucl Med Biol 10: 181-187, 1983. 
7. Yamamoto K, Som P, Brill AB, Yonekura Y, Srivastave SC. Meinken GE, et al. Dual tracer autoradiographic study of B-methyl-(1-14C) heptadecanoic acid and 15-p-(131I)iodophenyl-B-methylpentadecanoic acid in normotensive and hypertensive rats. J Nucl Med 27: 1178-1183, 1986. 
8. Abendschein DR, Fox KAA, Ambos HD, Sobel BE, Bergmann SR. Metabolism of beta-methyl[1-llC]heptadecanoic acid in canine myocardium. Nucl Med Biol 14: 579-585, 1987. 
9. Jones GS, Livni E, Strauss HW. Hanson RN, Elmaleh DR. Synthesis and biologic evaluation of 1-[11C]-3,3-dimethyl-heptadecanoic acid. J Nucl Med 29: 68-72, 1988. 
10. Elmaleh DR, Livni E. Alpert NM, Strauss HW, Buxton R, Fischman AJ. Myocardial extraction of 1-[11C]betamethylheptadecanoic acid. J Nucl Med 35: 496-503, 1994. 
11. Beattie DS, Basford RE. Brain mitochondria-III Fatty acid oxidation by bovine brain mitochondria. J Neurochem 12: 103-11l, 1965. 
12. Nariai T, DeGeorge JJ, Greig NH. Genka S, Rapoport SI, Purdon AD. Differences in rates of incorporation of intravenously injected radiolabeled fatty acids into phospholipids of intracerebrally implanted tumor and brain in awake rats. Clin Exp Metastasis 12: 213-225, 1994. 
l3. Freed LM, Wakabayashi S, Bell JM, Rapoport SI. Effect of inhibition of B-oxidation on incorporation of [U-14C]palmitate and [1-14C]arachidonate into brain lipids. Brain Res 645: 41-48, 1994. 
14. Cremer JE, Teal HM, Heath DF, Cavanagh JB. The influence of portocaval anastomosis on the metaboli sm of labeled octanoate, butyrate and leucine in rat brain. J Neurochem 28: 215-222, 1974. 
15. Ogawa K, Niishawa K, Sasaki M. Nozaki T. Malonic ester synthesis of various a-(11,l4C-methyl)-carboxylic acids and related compounds. J Label Compds Radiopharm 30: 417-419, 1991 (abstract). 1
6. Ogawa K. Nozaki T. Sasaki T, Ishiwata K, Senda M. Comparison of biodistribution in 2-methyl-fatty acids labeled at different positions. J Label Compds Radiopharm 35: 343-345, 1994 (abstract).
17. Ishiwata K, Ishii S, Sasaki T, Senda M, Nozaki T. A distillation method of preparing C-11 labeled acetate for routine clinical use. Appl Radiat Isot 44: 761-763, 1993. 
l8. Ishiwata K, Ishii S, Senda M. HPLC separation using physiological saline for quality control of radiopharmaceuticals used in PET studies. Appl Radiat lsot 44: 1119-1124, 1993. 
19. Ishiwata K. Ido T, Mejia AA, Ichihashi M, Mishima Y. Synthesis and radiation dosimetry of 4-boron0-2-[18F] fluoro-D,L-phenylalanine: A target compound for PET and boron neutron captuer therapy. Appl Radiat Isot 142: 325-328, l991. 
20. Mejia AA, Nakamura T, Itoh M, Hatazawa J, Ishiwata K, Ido T, et al. Absorbed dose estimates in positron emission tomography studies based on the administration of 18F-labeled radiopharmaceuticals. J Radiat Res 32: 243-261 , l991 . 
21. Schulz H. Beta oxidation of fatty acids. Biochim Biophys Acta 1081: 109-120, 1991 . 
22. Davidson B, Schulz H. Separation, properties, and regulation of acyl coenzyme A dehydrogenases from bovine heart and liver. Arch Biochem Biophys 213: 155-162, 1982. 
23. Finocchiaro G, Ito M, Tanaka K. Purification and properties of short chain acyl-CoA, medium chain acyl-CoA, and isovaleryl-CoA dehydrogenases from human liver. J Biol Chem 262: 7982-7989, 1987. 
24. VanhoveG,VanVeldhovenPP,Eyssen HJ,MannaertsGP. Mitochondrial short-chain acyl-CoA dehydrogenase of human liver and kidney can function as an oxidase. Biochem J 292: 23-30, 1993. 
25. Kclvraa S, Gregersen N. Christensen E. Hobolth N. In vitro fibroblast studies in a patient with C6-C10-dicarboxylic aciduria: evidence for a defect in general acyl-CoA dehydrogenase. Clin Chim Acta 126: 53-67, 1982. 
26. Turnbull DM, Bartlett K. Stevens DL, Alberti KGMM, Gibson GJ, Johnson MA, et al. Short-chain acyl-CoA dehydrogenase defficiency associated with a lipid-storage myopathy and secondary camitine defficiency. New Engl J Med 311 : 1232-1236, 1984. 
27. Chlouverakis C, Harris P. Composition of the free fatty acid fraction in the plasma of human arterial blood. Nature 188: l111-1112, 1960. 
28. Harris P, Chlouverakis C, G]oster J, Jones JH. Arteriovenous differences in the composition of plasma free fatty acids in various regions of the body. Clin Sci 22: 113-118, 1962. 
29. Yamaguchi M, Matsunaga R, Fukuda K, Nakamura M, Ohkura Y. Highly sensitive determination of free polyunsaturated, Iong-chain fatty acids in human serum by high-performance liquid chromotography with fluorescence detection. Anal Biochem 155: 256-261 , 1986. 
30. Mishkin S, Stein L, Gatmaitan Z, Arias IM. The binding of fatty acids to cytoplasmic proteins: Binding to Z protein in liver and other tissues of the rat. Biochem Biophys Res Commun 47: 997-l003, 1972. 
31. Rustow B, Hodi J, Kunze D, Reichmann G, Egger E. Specific binding of saturated and unsaturated fatty acids on the 'Z'-protein of rat liver cytosol. FEBS Lett 95: 225-228, l978. 
32. Sweetser DA, Heuckeroth RO, Gordon Jl. The metabolic significance of mammalian fatty-acid-binding proteins: aboundant proteins in search of a function. Ann Rev Nutr 7: 337-359, 1987. 
33. Brown MA, Myears DW, Bergman SR. Validity of estimates of myocardial oxidative metabolism with carbon-11 acetate and positron emission tomography despite altered patterns of substrate utilization. J Nucl Med 30: 187-193, 1989. 
34. Owen OE, Morgan AP, Kemp HG, Sullivan JM, Herrera MG, Cahill Jr GF. Brain Metabolism during fasting. J Clin Invest 46: 1589-1595, 1967. 
35. Hawkins RA, Williamson DH, Krebs HA. Ketone-body utilization by adult and suckling rat brain in vivo. Biochem J 122: 13-18, 1971. 
36. Ruderman NB, Ross PS, Berger M, Goodman MN. Regulation of glucose and ketone-body metabolism in brain of anaesthetized rats. Biochem J 138: 1-10, 1974. 
37. Gjedde A, Crone C. Induction processes in blood-brain transfer of ketone bodies during starvation. Am J Physiol229: 1165-1169, 1975. 
38. Hawkins RA, Biebuyck JF. Ketone bodies are selectively used by individual brain regions. Science 205: 325-327, l979. 
39. Hasselbalch SG, Knudsen GM, Jakobsen J, Hageman LP, Holm S, Paulson OB. Brain metabolism during short-term starvation in humans. J Cereb Blood Flow Metab 14: 125-131, 1994. 
40. Oldendorf WH. Carrier-mediated blood-brain barrier transport of short-chain monocarboxylic organic acids. Am J Physiol 224: 1450-1453, 1973. 
41. Armbrecht JJ, Buxton DB, Brucken RC, Phelps ME, Schelbert HR. Regional myocardial oxygen consumption determined noninvasively in humans with [1-llC]acetate and dynamic positron tomography. Circulation 80: 863-872, 1989.  
42. Henes CG, Bergmann SR, Walsh MN, Sobel BE, Geltman EM. Assessment of myocardial oxidative metabolic reserve with positron emission tomography and carbon-1l acetate. J Nucl Med 30: 1489-1499, 1989. 
43. Buck A, Wolpers HG, Hutchins GD. Savas V, Mangner TJ, Nguyen N, et al. Effect of carbon-11-acetate recirculation on estimates of myocardial oxigen consumption by PET. J Nucl Med 32: 1950-1957, 1991. 
44. Badar-Goffer RS, Bachelard HS, Morris PG. Cerebral metabolism of acetate and glucose studied by 13-C-n.m.r, spectroscopy. Biochem J 266: 133-139, 1990. 
45. Cerdan S, Kunnecke B, Seelig J. Cerebral metabolism of [1,2-13-C]acetate as detected by in vivo and in vitro 13C NMR. J Biol Chem 265: 12916-12926, 1990.