Measurement of Free Intracellular and Transfer RNA Amino Acid Specific Activity and Protein Synthesis in Rat Brain In Vivo

Transcription

1 Journal of Cerebral Blood Flow and Metabolism 10: Raven Press, Ltd., New York Measurement of Free ntraellular and Transfer RNA Amino Aid Speifi Ativity and Protein Synthesis in Rat Brain n Vivo Katharine M. Hargreaves-Wall, Jody L. Buiak, and William M. Pardridge Department of Mediine and Brain Researh nstitute, UCLA Shool of Mediine, Los Angeles, California, U.S.A. Summary: Brain protein synthesis was measured in anesthetized adult, male Sprague-Dawley rats by an in situ internal arotid arterial perfusion tehnique using [3H]leuine. The speifi ativity of free intraellular leuine and of trna leuine were determined by HPLC separation of phenylisothioyanate (PTC) derivatives of amino aids. The speifi ativity of the leuyl-trna pool rapidly equilibrated with the free intraellular leuine pool within 2 min. The speifi ativity of the trna and free leuine pools in brain reahed equilibrium by 10 min. Plasma amino aid speifi ativity, however, remained threefold higher than the speifi ativity of trna and free leuine pools. Estimates of protein synthesis were 0.62 ± 0.06 nmollmin/g and were onstant between 10 and 30 min of perfusion. The in situ perfusion model for protein synthesis desribed is a ontrolled system suited to measurements of protein synthesis in brain that an be applied to the study of brain metabolism under hanging physiologial onditions. Key Words: Bloodbrain barrier-amino aid transport-leuine. The quantitative measurement of protein synthesis in brain poses methodologial problems beause of the presene of the blood-brain barrier (BBB), whih limits the equilibration between radiolabeled amino aid in the irulating and the intraellular preursor pools. n animal models, autoradiographi methods have been employed involving ompartmental modeling that assumes the plasma amino aid pool is equivalent to both the free intraellular and the transfer RNA (trna) amino aid pool (Smith et a., 1984; Lestage et a., 1987; shiwata et a., 1988; Kirikae et a., 1988). Measurements of brain protein synthesis in humans is possible using positron emission tomography (PET) (Phelps et a., 1984; Erison et a., 1987), whih also employs ompartmental modeling and the assump- Reeived May 2, 1989; revised August 18, 1989; aepted August 28, Address orrespondene and reprint requests to Dr. W. M. Pardridge at Department of Mediine, Division of Endorinology, UCLA Shool of Mediine, Los Angeles, CA , U. S.A. Abbreviations used: BBB, blood-brain barrier; HPLC, highperformane liquid hromatography; PET, positron emission tomography; PTC, phenylisothioyanate; P.S., protein synthesis; S.A., speifi ativity; TCA, trihloroaeti aid; trna, transfer RNA. tion that the plasma, free intraellular, and trna amino aid pools in the brain are all equivalent. Despite the importane of verifying this assumption, there are few measurements of amino aid speifi ativities in the free intraellular and trna amino aid ompartments (Bodsh and Hossmann, 1983). As a result of methodologial obstales involved with the measurement of protein synthesis in brain, there is a need for a model to study brain protein synthesis in vivo under ontrolled onditions. The aim of this study, therefore, was twofold. The first goal was to expand on the original measurements of Bodsh and Hossmann (1983) and to develop methods for measuring the speifi ativity of amino aid in the free intraellular and trna pools using highperformane liquid hromatographi (HPLC) separation of derivatized amino aids. The seond goal was to estimate rates of protein synthesis in vivo by an in situ steady-state internal arotid artery perfusion tehnique. MATERALS AND METHODS Materials L-[3,4,5-3H(N)]leuine (143 Cilmmol) was purhased from NEN-Dupont (Boston, MA, U.S.A.). Phenyl- 162

2 BRAN PROTEN SYNTHESS 163 isothioyanate (PTC) and amino aid standards were obtained from Piere Chemial Company (Rokford, L, U.S.A.). Norleuine and all other hemials were purhased from Sigma Chemial Company (St. Louis, MO, U.S.A.). Sint-A was obtained from Pakard nstrument Company (Downers Grove, L, U.S.A.). Brain perfusion tehnique Male Sprague-Dawley rats weighing g were anesthetized with ketamine-hcl (Ketalar; 125 mg/kg) and xylazine (Rompum; 5 mg/kg) intraperitoneally. The right superior thyroid, ophthalmi, and pterygopalatine arteries were auterized and the right external arotid artery atheterized (PE-10 tubing) for retrograde infusion (Takasato et al., 1984). For perfusion periods of 5 min or longer, the left femoral artery was atheterized (PE-50 tubing) for withdrawal of blood (1 mumin) by a syringe pump (Harvard Biosiene, Boston, MA, U.S.A.). The right ommon arotid artery was ligated and the perfusion fluid introdued into the right external arotid artery using a peristalti pump (Harvard, Model 1210) at a flow rate of 1 mumin. The perfusion was terminated after min by deapitation; the right (ipsilateral) hemisphere was removed, weighed, and homogenized in 3 ml of ieold buffer ontaining 0.1 M KCl, 6 mm MgC2, 30 mm NaCl, and 1 mm Na aetate, ph 4.5, followed by soniation for 60 s on ie (Branson sonifier, Model 185). The perfusate was Krebs-Henseleit buffer (118 mm NaCl, 4.7 mm KCl, 1.2 mm MgS04, 1.2 mm KH2P04, 25 mm NaHC0 3, 2.5 mm CaC2) ontaining 10 mm gluose, 3% bovine albumin, /J-Cilml of [3H]leuine, 1 x plasma onentrations of 20 amino aids, and 30% freshly isolated rat red blood ells. The onentration of the amino aids added to the perfusate were asparti aid 38 /J-M, threonine 270 f.lm, serine 243 f.lm, asparagine 63 f.lm, glutamine 667 f.lm, proline 186 f.lm, glutamate 76 f.lm, glyine 408 /J-M, alanine 470 f.lm, valine 173 /J-M, ysteine 38 /J-M, methionine 47 /J-M, isoleuine 91 f.lm, leuine 161 /J-M, phenylalanine 54 f.lm, lysine 418 f.lm, histidine 63 /J-M, arginine 132 f.lm, tryptophan 69 f.lm, and tyrosine 83 f.lm, per Tolman et al. (1973). Red blood ells were isolated immediately before use. Approximately 10 ml of blood was withdrawn from the abdominal aorta of anesthetized rats using a syringe and an 18-gauge needle, and plaed in 2 ml of antioagulant buffer (1.32% D-gluose, 1.32% Na itrate, 0.44% itri aid). The blood was entrifuged for 3 min at 500 g and the red ells washed twie with two volumes of 0.9% NaC and twie with two volumes of Krebs-Henseleit buffer before being added to the perfusate. The perfusate was ontinuously oxygenated with 95% 02/5% CO2 and warmed to 37 C. solation of free intraellular and trna amino aid pools Brain free amino aids were reovered from 250 /J-l of homogenate by ethanol preipitation; 680 /J-l of absolute ethanol were added to the homogenate (final onentration 70% ethanol) and plaed at -20 C overnight. The mixture was entrifuged at 15,000 g for 20 min at 4 C, and the supernatant reovered and dried under vauum in a Speed-Va vauum entrifuge evaporator (Savant nstruments, n., Farmingdale, NY, U.S.A.). The amino aids were resuspended in 0.5 ml oupling buffer (50% aetonitrile, 25% pyridine, 10% triethylamine) and entrifuged at 1,000 g for 5 min to remove any insoluble residue; 0.1 ml was removed and dried under vauum with 1 nmol of norleuine amino aid standard. This was derivatized with 5 f.l of PTC in the presene of 100 /J-l oupling buffer for 10 min at room temperature, dried under vauum, and stored at -20 C until ready for HPLC analysis. Amino aids were reovered from the trna pool aording to the method in Fig. 1 and this was adapted from previous reports (Yang and Novelli, 1971; M Kee et al., BRAN PROTEN SYNTHESS METHODS --- Free 0.25 ml Homogenate 0.25 ml TCA Amino Aid < ---> nsoluble Pool 70% EtOH 3.0 ml 10% TCA Pool 15,000 g PTC 2.5 ml Phenol extration EtOH pptn of nulei aids FG. 1. Flow hart of methods employed to measure the amino aid speifi ativities in the free intraellular and in the trna pools. Nulei Aids 1.25 M NaCl pptn high MW nulei aids EtOH pptn of trna HPLC RNA PTC Na C0 (ph 10), 37'C 2 3 Et O H pptn of trna De-aylated Amino Aid J Cereb Blood Flow Metab. Vol. 10, No.2, 1990

3 164 K. M. HARGREAVES-WALL ET AL. 1978). To 2.5 ml of homogenate was added 1/5 volume of 2 M surose ontaining 2 mg/ml of bentonite. The mixture was shaken for 30 min at 4 C after addition of 6 ml of H20-saturated phenol and 3 ml of 0.15 MNaC ontaining 1 mm EDT A and 0.4 mg/ml of bentonite. The upper aqueous phase was reovered after entrifugation for 10 min at 15,000 g at 4 C, and 2.5 volumes of old absolute ethanol ontaining 12 gil of potassium aetate (ph = 6.5) were added to the aqueous phase and left to sit at -20 C overnight. After entrifugation for 10 min at 15,000 g at 4 C to ollet preipitated nulei aids, the supernatant was disarded and 2 ml of 50 mm N ac ontaining 10 mm MgC2, 1 mmedta, 10 mmna aetate (ph 4.5), and 0.4 mg/ml of bentonite plus 114 volume of 5 M NaCl were added to the residual pellet. This mixture stood on ie for 2 h, followed by entrifugation for 10 min at 10,000 g at 4 C to remove preipitated high moleular weight nulei aids (Fig. 1). The supernatant ontaining transfer RNA was added to 2.5 volumes of old absolute ethanol and left to sit at -20 C overnight. The preipitated trna was reovered by entrifugation for 10 min at 10,000 g and 4 C, and washed with 1 ml of old 70% ethanol, and then dissolved in 2 ml of 0.05 M Na2C0 (ph = 10) and inubated for 90 min at 3rC. The amino aids freed by deay 3 lation were separated from the trna by addition of 5 ml of absolute ethanol, and the deaylated trna was preipitated at -20 C for 4 h and entrifuged for 15 min at 10,000 g and 4 C. The resulting supernatant was dried under vauum and stored at -20 C. The samples were then delipidated by addition of 1 ml of 70% methanol and 1 ml of hloroform, vortexed, and entrifuged for 10 min at 1,000 g. The hloroform/methanol delipidation step was found to be essential and removed ethanol soluble lipids that omigrated with PTC-leuine in the HPLC system and that preluded measurement of trna leuine speifi ativities. The upper water-methanol layer was dried under vauum and resuspended in 200 fl.l of oupling buffer. Any insoluble residue was removed by entrifugation for 10 min at 1,000 g. A 0.1 nmol aliquot of norleuine internal standard was added to the supernatant, whih was then dried under vauum. Amino aids were derivatized using 5 fl.l of PTC in the presene of 100 fl.l of oupling buffer for 10 min at room temperature. The samples were dried under vauum and stored at -20 C until analyzed by HPLC. When 106 dpm of [3H]leuine was added to the homogenate of unperfused brain, no [3H]leuine was deteted in the final ethanol supernatant of the trna amino aids isolation, indiating no ontamination of the trna pool by the free intraellular pools. HPLC separation of amino aids Samples derivatized with PTC were resuspended in 0.01 M KP04 (ph 6.5) for HPLC analysis (Sholze, 1985). The injetion volume was 200 fl.l. The samples were separated on an Altex C18 (5 fl.m) 4.6 x 25 m olumn, using 0.01 M KP04 (ph = 6.5; solvent A) and 70% aetonitrile in 0.01 M KP04 (ph = 6.5; solvent B) at a flow rate of 1.5 mllmin. Detetion was at 254 nm (Hitahi variable wavelength spetrophotometer), and peaks were quantitated using a Perkin-Elmer integrator (Model LC-100). The shedule for brain free amino aids was 15% solvent B isoratially from 0-15 min; gradient of 15% solvent B to 60% solvent B from min; gradient of 60% solvent B to 100% solvent B from min; 100% solvent B isoratially from min; gradient of 100% solvent B to 15% solvent B from min; and 15% solvent B iso- ratially from min. The reovery of free intraellular leuine was determined in separate experiments by adding [3H]leuine to ontrol brain homogenate. The shedule for trna-derived amino aids was 20% solvent B isoratially from 0-10 min; gradient of 20% solvent B to 60% solvent B from min; gradient of 60% solvent B to 100% solvent B from min; 100% solvent B isoratially from min; gradient of 100% solvent B to 20% solvent B from min; and 20% solvent B isoratially for 5 min. Samples were olleted in 1.5 ml frations for sintillation ounting of ehjleuine. Reovery of [3H]leuine from the HPLC olumn was >97%, using [3H]leuine standards. The sensitivity of this method using an ultraviolet detetion system was approximately nmol. Amino aids were initially identified using amino aid standards and eluted at onsistent times throughout all analyses (Fig. 2). Norleuine (amino aid internal standard tested; Sigma Chemial Company) was shown to integrate equivalently with leuine standards. t was neessary to injet 80% of the hemisphere homogenate to reord measurable peaks of trna leuine by spetrophotometri detetion, owing to the very low pool size of trna amino aid (see the Results setion). solation of amino aid inorporated into protein norporation ofehjleuine into the protein pool (aidinsoluble pool) was determined by trihloroaeti aid (TCA) preipitation of the brain homogenate (Fig. 1); 250 fl.l of the initial homogenate was added to 2 ml of old 10% TCA and left to sit on ie for 15 min. The pellet obtained after entrifugation at 1,000 g for 15 min at 4 C was washed one with old O%TCA and solubilized in 0.5 ml of 2 N NaOH at 60 C. The entire sample was ounted in 10 ml of Sint-A and expressed as dpm/g of brain. Calulation of protein synthesis (P.S.) rates Speifi ativities of the amino aid pools are expressed as dpmlnmol of leuine. Rates of protein synthesis (nmol/min/g) were alulated aording to the following formula, using either the speifi ativity (S.A.) of the free intraellular or the trna leuine pool: P.S. = dpm in TCA-insoluble poollg of brain/min S.A. (free or trna pools) The ontribution of the brain plasma pool to the free intraellular pool was ignored sine the brain plasma volume is only about 1 % of the brain volume (Takasato et a., 1984). Statistial analysis was performed with Student's t test and statistial signifiane was judged at the p < 0.05 level. Analysis of metabolites Latate in brain was measured enzymatially as desribed by Lowry and Passonneau (1972). Animals used for the determination of metabolite levels at ontrol (zero time) and 10 min and 20 min perfusion times were killed by mirowave irradiation (Thermex, Model 4101; 2.7 s, 3.2 kw). The brain was removed, the right and left hemispheres isolated, weighed, and homogenized on ie in 3 ml of ie-old 1 N perhlori aid, and the supernatant was neutralized with KOH. J Cereb Blood Flow Metab, Vol. /0, No.2, 1990

4 BRAN PROTEN SYNTHESS a.> 20 u « a.> 20 u «Free leuine rleu w...j z \ la_f-.a. W, t-rna leuine wj v w ::;)...J wz 1 t : Fration / // A254 FG. 2. (Top) Reverse-phase HPLC separation of PTC amino aid derivatives in the free intraellular pool. The migration of leuine and the norleuine internal standards are shown at approximately 35% aetonitrile; 1.5 ml frations were olleted at the beginning of the aetonitrile gradient and the [3H] radioativity omigrated with the leuine peak. The leuine peak was positively identified in pilot studies measuring the omigration of [3H]leuine with a leuine amino aid standard. The peak migrating just prior to the leuine was found to be isoleuine using similar methods. Amino aids were deteted by measurement of absorption at 254 nm, and peaks were quantitated using a Perkin-Elmer integrator (see the Methods setion). (Bottom) The leuine and norleuine peaks for the trna fration are shown using a slightly different gradient than that used for measurement of free leuine speifi ativity. With this gradient shedule, leuine and norleuine eluted at approximately 30% aetonitrile. Chart reorder speed = 0.5 m/min. 40 TCA nsoluble DPM 19 x r=0.99 O L- -L o Minutes FG. 3. norporation of [3H]leuine into the trihloroaeti aid (TCA) insoluble pool of brain is plotted vs. the length of perfusion time. Data are mean ± SEM (n = 3-4 rats per point). perfusion. At perfusion times of 10, 20, and 30 min, the speifi ativity of the free intraellular leuine pool has stabilized, indiating that equilibration has ourred between the steady-state perfusion fluid and the ellular pools. The speifi ativity of the intraellular pool remains, however, at a level approximately threefold lower than that of the perfusate (Fig. 4). When equilibration of leuine in the brain has ourred, estimates of protein synthesis stabilize (Fig. 5). Estimates of protein synthesis at to, 20, and 30 min of perfusion were not statistially different and averaged 0.62 ± 0.06 nmollmin/g (mean ± SEM, n = 11) PERFUSATE NTRACELLULAR RESULTS The inorporation of eh]leuine into the TCAinsoluble protein pool was linear with time from 30 s to 30 min (Fig. 3). n Fig. 4, the speifi ativity of the intraellular leuine pool is ompared to that of the perfusate, whih was held onstant during the perfusion. The speifi ativity of the intraellular leuine pool rises rapidly during the first 2 min of Minutes FG. 4. Speifi ativity (dpm/nmol) in the perfusate and the free intraellular pool is plotted vs. perfusion time. Data are mean ± SEM (n = 3-4 rats per point). n the experiments desribed in Figs. 3-5, the perfusate ontained 50 /LCi/ml of [3Hjleuine and the perfusate speifi ativity was 688,514 ± 28,369 dpm/nmol (mean ± SEM, n = 7). J Cereb Blood Flow Metab, Vol. 10, No.2, 1990

5 166 K. M. HARGREAVES-WALL ET AL. Ol - E E - 3 (/) (/) Q)..::. -- >. (/) Q) a. 0.. CD 2 }- 10 mean ± 1 S.D Minutes FG. 5. Brain protein synthesis is plotted vs. perfusion time. Values at early time paints are artifatually high due to lak of equilibration of the free intraellular pool (Fig. 4). The protein synthesis values at 10, 20, and 30 min of perfusion are not statistially different and averaged 0.62 ± 0.06 nmoll min/g (mean ± SEM, n = 11). n a seond series of experiments, the speifi ativity of the trna leuine pool was measured. Measures of trna leuine speifi ativity indiate that the trna amino aid pool is in rapid equilibration with the free intraellular amino aid pool, sine the speifi ativities of eh]leuine in these two pools were not statistially different at 2-20 min of perfusion (Table O. To verify that trna pool sizes an be reliably measured following deapitation, the onentration of trna leuine in rat brain was also measured following mirowave irradiation TABLE 1. [3Hlleuine speifi ativity (dpmlnmol) in the free intraellular and trna pools vs. perfusion time Perfusion time (minutes) Free intraellular 162,475 ± 9, ,777 ± 44, ,354 ± 30, ,544 ± 41,806 trna 232,381 ± 43, ,685 ± 70, ,736 ± 74, ,806 Mean ± SEM (n = 3-4, exept trna at 20 min, where n = 1). There are no statistially signifiant differenes between the free intraellular and trna speifi ativities at any time point. The free intraellular speifi ativities at 10 and 20 min of perfusion are not statistially different. n these experiments, the perfusate ontained Cilml of [3Hlleuine and the perfusate speifi ativity was 1,335,321 ± 42,221 dpm/nmol (mean ± SEM, n = 4). prior to deapitation (see the Methods setion). As shown in Table 2, the leuine trna undergoes no measurable enzymati degradation between deapitation and preparation of the homogenate. The onentration of free intraellular leuine was 83 ± 5 nmollg (mean ± SEM, n = 20). The metaboli state of the brain was evaluated by measuring brain latate onentrations following 0, 10, and 20 min of perfusion. As shown in Table 3, latate was inreased two- and threefold, respetively, following 10 and 20 min of perfusion in the ipsilateral hemisphere. Brain latate was also inreased in the ontralateral or unperfused hemisphere (Table 3). DSCUSSON The present studies desribe an internal arotid artery perfusion tehnique for quantitating erebral protein synthesis in rat brain in vivo, based on diret HPLC measurements of amino aid speifi ativities in both the free intraellular and trna pools. These studies suggest the following onlusions: First, the trna leuine is in rapid equilibration with the free intraellular pool of leuine, suh that the speifi ativities of these two pools are not statistially different within 2 min of internal arotid artery perfusion (Table O. Seond, the speifi ativity of either the free intraellular or the trna leuine pool is substantially overestimated by the speifi ativity of the plasma pool of labeled amino aid (Fig. 4). Third, rates of leuine inorporation into rat brain proteins in vivo average 0.62 ± 0.06 nmol/minlg (Fig. 5) for the whole hemisphere under onditions of ketamine anesthesia and internal arotid artery perfusion. The rapid equilibration between the free intraellular and trna pools is not unexpeted, given the very high ativity of trna amino aylating enzymes in brain. For example, the V max/km ratio for the tryptophan trna aylating enzyme in brain is 900 f.lmollminl100 mg of protein with a high affinity Km of 2 f.lm for tryptophan (Liu et al., 1973). Assuming 100 mg of protein/g of brain, this V max/km value for the aylating enzyme is log orders greater than the V maxi Km ratio of amino aid transport into brain from blood (Pardridge, 1983). TABLE 2. Leuyl-tRNA onentrations in rat brain hemisphere vs. method of termination of metabolism Condition Mirowave irradiation Deapitation LEU-tRNA (nmol/g) 0.16 ± 0.01 (6) 0.17 ± 0.01 (13) Data are mean ± SEM (n given in parentheses). J Cereb Blood Flow Metab, Vol. 10, No.2, 1990

6 BRAN PROTEN SYNTHESS 167 TABLE 3. Latate onentrations in hemisphere ipsilateral and ontralateral to arotid artery perfusion Condition Control 10 min perfusion 20 min perfusion Right (ipsilateral) 1.2 ± ± ± 0.5 Left (ontralateral) l.l ± ± 0.2* 2. 9 ± 0. 3** Mean ± SEM (n = 4-6). Data are reported as flmol/g. * p < 0.005, **p < differene between right and left hemispheres. The finding of equal speifi ativities in the free intraellular and transfer RNA pools are in agreement with the observations of Smith et al. (1988), but ontrast with those of Keen et al. (1989). The latter group reports a twofold higher speifi ativity in the leuyl-transfer RNA pool as ompared to the free intraellular leuine. However, this study employed a single intravenous injetion of only 50!LCi of e4c]leuine and Keen et al. (1989) report the diffiulty in adequately labeling the leuyl transfer RNA pool using this tehnique. This diffiulty arises from the very low pool size of amino ayl transfer RNA in brain (Table 2). Given the very small pool size of amino ayl trna in rat brain, it an be alulated that quite large speifi ativities (dprnlnmol) must be generated in the experiment to yield measurable radioativities in the HPLC fration following separation of derivatized amino aids obtained from the trna pool. For example, a speifi ativity of 1,000 dprnlnmol in the trn A pool for a given amino aid would only yield <50 pm in the HPLC fration for an entire hemisphere sample. Consequently, in the present studies, statistially signifiant amounts of radioativity in the individual HPLC frations for the trna pool measurements were obtained by infusing large amounts of eh]-labeled amino aids, e.g., !LCi/ml in the perfusate (see the Methods setion). These onsiderations should be made when measuring trna speifi ativities in brain following intravenous injetion of labeled amino aid, sine very little of an intravenous bolus of amino aid is atually delivered to the brain, as opposed to the present studies employing internal arotid artery perfusion. Our estimates of the pool size of leuyl-trna in brain are intermediate between the values of approximately 0.02 nmol/g reported by Smith et al. (1988) and the value of 0.6 nmol/g measured by Keen et al. (1989). Apart from these reent measurements, there are few determinations of speifi amino ayl trnas in brain, but an analysis of the available literature suggests that the values shown in Table 2 are aurate. For example, the onentration of aylated leuine trna in mouse brain is 21 pmol/a26o unit of trna (Hughes and Johnson, 1977). Given one A260 unit = 45!Lg of trna (Davey and Manhester, 1969), and 350!Lg of trna/g of brain (Khasigov and Nikolaev, 1987), then these data suggest there are 0.16 nmol/g of brain of leuyltrna. n rat brain, the total trna is known to be 3.3 A260 units/g of brain, or 6 nmol/g (Maenpaa and Twari, 1983), given a moleular weight of 23,000 (Davey and Manhester, 1969). f the leuyl-trna is 5% of the total (Davey and Manhester, 1969), then this is equivalent to 0.3 nmol/g of brain of leuyl-trna, and if 50% of this is aylated (Johnson and Chou, 1973), then the onentration of aylated leuine trna in rat brain is 0.15 nmol/g of brain, whih equals the experimentally observed values reported in the present study (Table 2). The onentrations of leuine trna in rat brain annot be muh higher than these values sine it is known that the onentration of valyl-trna in rat liver is 0.6 nmol/g of liver (Airhart et a., 1974), and the rate of protein synthesis in liver (Mortimore et al., 1972) is more than tenfold greater than the rate of protein synthesis in brain (Pardridge, 1983). Our observation that the speifi ativity of leuine in blood is about threefold higher than the speifi ativity of leuine in the free intraellular pool is in agreement with the studies of Keen et al. (1989) and Smith et al. (1988). They report fourfold and twofold, respetively, higher speifi ativities in blood leuine as ompared to free intraellular leuine in brain. These observations demonstrate the rate-limiting role played by the blood-brain barrier (BBB) in amino aid transport from blood to brain. f barrier transport was not rate-limiting, then it would be expeted that there would be equal speifi ativities in the plasma and intraellular pools, whih has been expliitly assumed in previous ompartmental modeling studies. However, it is generally regarded that transport of amino aids aross the BBB is rate-limiting for overall uptake, owing to the muh greater surfae area of neuronal and glial membranes as ompared to the brain apillary endothelial or BBB membrane (Pardridge, 1983). Another fator that auses BBB transport to be ratelimiting is the high degree of saturation of the rat and human BBB neutral amino aid transporter by normal physiologi onentrations of neutral amino aid (Oldendorf, 1971; Hargreaves and Pardridge, 1988). Owing to a very low Km (very high affinity) of this transporter, the arrier is normally about 90% saturated under physiologi onditions. t is possible that the normal rate-limiting role of BBB transport may be irumvented by the intravenous infusion of very high onentrations of unlabeled amino aid along with the partiular labeled amino aid. J Cereb Blood Flow Metab. Vol. 10. No

7 168 K. M. HARGREAVES-WALL ET AL. This pratie may ause an equivalene of the plasma and free intraellular pool of the respetive amino aid, although the diret measurements of amino aid speifi ativity in the free intraellular pool or trna pool have not been performed in previous "amino aid loading" experiments (Dunlop et ai., 1975; Dienel et ai., 1980; Dwyer et ai., 1982). The amino aid loading paradigm to ause equilibration of amino aid in the plasma and free intraellular pools has worked suessfully in organs suh as liver (Mortimore et ai., 1972) or myoardium (MKee et ai., 1978), wherein amino aid transport is haraterized by very low affinity systems that an tolerate large inreases in plasma amino aid. However, in brain, the amino aid loading paradigm may ause a seletive saturation of the neutral amino aid transporter by the loaded amino aid, and this may alter the availability of other ompeting neutral amino aids in brain that are also inorporated into brain proteins along with the labeled amino aid. The rate of leuine inorporation into rat proteins in vivo in the present experiments of 0.62 ± 0.06 nmol/min/g is lower than the value reported for rat brain using ompartmental modeling approahes that assume equivalene of the plasma and free intraellular speifi ativities for a given amino aid (Smith et ai., 1984; Lestage et ai., 1987; Kirikae et ai., 1988). For example, the rate of valine inorporation into ortial gray strutures of the rat brain averages 1.4 nmol/min/g, and the valine inorporation rates into proteins are about one-half those of leuine (Kirikae et ai., 1988). n making these omparisons, several fators should be onsidered. First, the value of protein synthesis reported in Fig. 5 is for the whole hemisphere. Conversely, measurements of valine inorporation into gray vs. white matter determined by ompartmental approahes generally show a threefold higher rate of inorporation in the gray strutures as opposed to white matter (Kirikae et ai., 1988). Seond, ketamine/xylazine anesthesia was employed in the present study and anesthesia is known to inhibit brain protein synthesis by about 50% (Lestage et ai., 1987). n addition, in the present studies, the internal arotid artery perfusion of brain with oxygenated buffer ontaining 30% rat erythroytes was assoiated with modest inreases in brain latate onentrations (Table 3). However, the mild hypoxemia assoiated with the perfusion protool appears to ause no inhibition of brain protein synthesis, sine the rate of leuine inorporation into protein is linear for 30 min (Fig. 3). n summary, the present studies desribe methods for measurement of free intraellular and trna amino aid speifi ativities and an internal arotid artery perfusion tehnique for quantitating erebral protein synthesis in rat brain in vivo. The internal arotid artery perfusion approah allows for adequate labeling of individual amino ayl transfer RNA pools, and allows for measurement of erebral protein synthesis under onditions in whih the omposition of the erebral blood supply is arefully ontrolled. n agreement with Smith et ai. (1988), who used a onstant intravenous infusion method, the present studies show an equivalene of the speifi ativities of the free intraellular and trna pools. However, this equivalene may not hold in all experimental paradigms and it is important, when possible, to inorporate measurements of trna amino aid speifi ativities in the measurement of brain protein synthesis (Fig. O. The present studies also show that the speifi ativity of the free intraellular pool is about one-third of the amino aid speifi ativity in blood. The ratelimiting role played by the BBB interfae in the overall transport of amino aid from blood to brain intraellular spae is an important fator in ompartmental modeling studies that assume that the plasma and free intraellular pools are equivalent. Aknowledgment: Dawn Brown skillfully prepared the manusript. This work was supported by NH Grant RO-NS Dr. Hargreaves is the reipient of a Postdotoral Fellowship from the Medial Researh Counil of Canada. This work was presented at the 19th Annual Meeting of the Soiety for Neurosiene, November 1989, in Phoenix, Arizona. REFERENCES Airhart J, Vidrih A, Khairallah EA (1974) Compartmentation of free amino aids for protein synthesis in rat liver. Biohem J 140: Bodsh W, Hossmann KA (1983) A quantitative regional analysis of amino aids involved in rat brain protein synthesis by high performane liquid hromatography. J Neurohem 40: Davey PJ, Manhester KL (1969) solation of labelled aminoayl transfer RNA from musle: Studies of the entry of labelled amino aids into ayl transfer RNA linkage in situ and its ontrol by insulin. Biohim Biophys Ata 192:85-97 Dienel GA, Pulsinelli WA, Duffy TE (1980) Regional protein synthesis in rat brain following aute hemispheri ishemia. J Neurohem 35: Dunlop DS, van Elden W, Lajtha A (1975) A method for measuring brain protein synthesis rates in young and adult rats. J Neurohem 24: Dwyer BE, Donatoni P, Wasterlain CG (1982) A quantitative autoradiographi method for the measurement of loal rates of brain protein synthesis. Neurohem Res 7: Erison K, Blomqvist G, Bergstrom M, Eriksson L, Stone Elander S (1987) Appliation of a kineti model on the methionine aumulation in intraranial tumours studied with positron emission tomography. Ata Radiol 28: Hargreaves KM, Pardridge WM (1988) Neutral amino aid transport at the human blood-brain barrier. J Bioi Chem 263: J Cereb Blood Flow Metab, Vol. 10, No.2, 1990

8 BRAN PROTEN SYNTHESS 169 Hughes JV, Johnson TC (1977) The effets of hyperphenylalaninaemia on the onentrations of aminoayl-transfer ribonulei aid in vivo. Biohem J 162: shiwata K, Vaalburg W, Elsinga PH, Paans AMJ, Woldring MG (1988) Metaboli studies with L-[ 1-14C]tyrosine for the investigation of a kineti model to measure protein synthesis rates with PET. J Nul Med 29: Johnson TC, Chou L (1973) Level and amino aid aeptor ativity of mouse brain trna during neural development. J Neurohem 20: Keen RE, Barrio JR, Huang SC, Hawkins RA, Phelps ME (1989) n vivo erebral protein synthesis rates with leuyl-transfer RNA used as a preursor pool: determination of biohemial parameters to struture traer kineti models for positron emission tomography. J Cereb Blood Flow Metab 9: Khasigov PZ, Nikolaev A Y (1987) The role of isoaeptor transfer RNAs in regulation of age-related hanges in the rate of protein synthesis. Biohem nt 15: Kirikae M, Diksi M, Yamamoto YL (1988) The transfer oeffiients for L-valine and the rate of inorporation of L-[1-14Clvaline into proteins in normal adult rat brain. J Cereb Blood Flow Metab 8: Lestage P, Gonon M, Lepetit P, Vitte PA, Debilly G, Rossatto C, Leestre D, Bobillier P (1987) An in vivo kineti model with L-[35S1-methionine for the determination of loal erebral rates for methionine inorporation into protein in the rat. J Neurohem 48: Liu C-C, Chung C-H, Lee M-L (1973) Amino aid ativation in mammalian brain. Purifiation and haraterization of tryptophan-ativating enzyme from buffalo brain. Biohem J 135: Lowry OH, Passonneau JV (1972) A Flexible System of Enzymati Analysis, Aademi Press, New York Maenpaa H, Twari S (1983) A hange in isoaepting leuine transfer RNA speies in rat brain after prolonged ingestion of ethanol. Med Bioi 61: MKee EE, Cheung JY, Rannels DE, Morgan HE (1978) Measurement of the rate of protein synthesis and ompartmentation of heart phenylalanine. J Bioi Chem 253: Mortimore GE, Woodside KH, Henry JE (1972) Compartmentation of free valine and its relation to protein turnover in perfused rat liver. J Bioi Chem 247: Oldendorf WH (1971) Brain uptake of radiolabeled amino aids, amines, and hexoses after arterial injetion. Am J Physiol 221: Pardridge WM (1983) Brain metabolism: a perspetive from the blood-brain barrier. Physiol Rev 63: Phelps ME, Barrio JR, Huang S-C, Keen RE, Chugani H, Mazziotta JC (1984) Criteria for the traer kineti measurement of erebral protein synthesis in humans with positron emission tomography. Ann Neural 15(suppl):S92-S202 Sholze H (1985) Determination of phenylthioarbamyl amino aids by reversed-phase high-performane liquid hromatography. J Chromatogr 350: Smith CB, Crane AM, Kadekaro M, Agranoff BW, Sokoloff L (1984) Stimulation of protein synthesis and gluose utilization in the hypoglossal nuleus indued by axotomy. J Neurosi4: Smith CB, Deibler GE, Eng N, Shmidt K, Sokdoff L (1988) Measurement of loal erebral protein synthesis in vivo: influene of reyling of amino aids derived from protein degradation. Pro Natl Aad Si USA 85: Takasato Y, Rapoport S, Smith QR (1984) An in situ brain perfusion tehnique to study erebrovasular transport in the rat. Am J Physiol 247:H484-H493 Tolman EL, Shworer CM, Jefferson LS (1973) Effets of hypophysetomy on amino aid metabolism and gluoneogenesis in the perfused rat liver. J Bioi Chem 248: Yang W-K, Novelli GD (1971) Analysis of isoaepting trna's in mammalian tissues and ells. Methods Enzymol 20:44-55 J Cereb Blood Flow Metab, Vol. /0, No. 2, 1990