* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Assessment of the mathematical issues involved
Survey
Document related concepts
Ribosomally synthesized and post-translationally modified peptides wikipedia , lookup
Ancestral sequence reconstruction wikipedia , lookup
Protein–protein interaction wikipedia , lookup
Western blot wikipedia , lookup
Fatty acid synthesis wikipedia , lookup
Two-hybrid screening wikipedia , lookup
Metalloprotein wikipedia , lookup
Artificial gene synthesis wikipedia , lookup
Basal metabolic rate wikipedia , lookup
Point mutation wikipedia , lookup
Peptide synthesis wikipedia , lookup
Proteolysis wikipedia , lookup
Genetic code wikipedia , lookup
Biochemistry wikipedia , lookup
Transcript
Clinical Science (1993) 84, 177-183 (Printed in Great Britain) I77 Assessment of the mathematical issues involved in measuring the fractional synthesis rate of protein using the flooding dose technique David L. CHINKES, Judah ROSENBLATT and Robert R. WOLFE Metabolism Unit, Shriners Burns Institute, and University of Texas Medical Branch, Galveston, Texas, U.S.A. (Received 24 September 1991/20 July 1992; accepted 19 O c t o b e r 1992) 1. The fractional synthesis rate of protein is commonly measured by either the constant infusion method or the flooding dose method. The two methods often give different results. 2. An underlying assumption of the traditional flooding dose formula is that the protein synthesis rate is not stimulated by the flooding dose. A new formula for calculation of the fractional synthesis rate is derived with the alternative assumption that the protein synthesis rate is stimulated by an amount proportional to the change in the intracellular concentration of the infused amino acid. The alternative formula is: Fractional synthesis rate - EB( t, - EB( O) j: EFlCdt where E B and EF are the enrichments of bound and free amino acid, respectively (atom per cent excess), and C=l-(E,/E,), where EI is the enrichment of the infusate. This approach defines the lowest possible value for the fractional synthesis rate. The traditional equation gives a maximal value for the fractional synthesis rate. 3. When data from the literature are considered, the fractional synthesis rate of muscle protein as calculated by the constant infusion technique falls between the values of fractional synthesis rate calculated by the two flooding dose formulae when leucine is the tracer, suggesting that a flooding dose of leucine exerts a stimulatory effect on the rate of protein synthesis, but that the increase is not as great as the increase in the intracellular concentration of leucine. 4. The precision of the formula for the calculation of fractional synthesis rate is limited by the accuracy of the underlying assumptions regarding the effect of the flooding dose on the fractional synthesis rate. At present, the best approach would appear to be the use of both equations to calculate the upper and lower bounds of the true fractional synthesis rate. INTRODUCTION Difficulty in identifying the true precursor enrichment for the calculation of protein synthesis by the constant tracer infusion method limits the accuracy of the calculated values. The so-called ‘flooding dose’ technique was designed to overcome this limitation. The rationale of the method is to inject simultaneously boluses of both tracer and tracee amino acid, thereby ‘flooding’ the entire precursor pool (extracellular and intracellular components) to such an extent that the enrichment is the same everywhere. The fractional synthesis rate (FSR) of the specific protein is then determined by dividing the rate of increase in the product enrichment in a given amount of time by the average precursor enrichment over that time. There are several practical advantages to the flooding dose technique. Theoretically, it should eliminate uncertainty regarding precursor enrichment, since the precursor is at the same enrichment in all compartments. Furthermore, because of this flooding of the precursor pool, it should be possible to use only plasma enrichment measurements to quantify the true precursor enrichment. Another advantage is that it is possible to give enough tracer to enable sufficient tracer incorporation for accurate measurement in a short period of time. This is particularly important in the measurement of the FSR of proteins that turn over slowly, such as muscle, which may require 4 h or more of constant tracer infusion to achieve a comparable level of protein-bound amino acid enrichment. The ability to complete the determination in a short time is not only a logistical advantage (i.e. a shorter experimental time for subject and experimenter), but may enable specific questions to be answered that otherwise could not be addressed. For example, the response of protein synthesis to a hormone infusion may be short-lived (due to a counter-regulatory response), yet physiologically important. Because of Key words: amino acids, constant infusion, flooding dose, fractional synthesis rate, leucine, metabolism, modelling, muscle, protein degradation, protein synthesis, tracer. Abbreviations: APE, atom per cent excess; FSR, fractional synthesis rate. Correspondence: Dr Robert R. Wolfe, Metabolism Unit, Shriners Burns Institute, 610 Texas Avenue, Galveston, T X 77550, U.S.A. I78 D. L. Chinks et al. these practical advantages, the flooding dose technique has become widely popular over the past ten years. In particular, the recent modification of the technique for the use of stable isotopic tracers [l], instead of the original use of radioactive tracers [2], represents a potentially important new advance in methodology. Concomitant with the advantages of the flooding dose technique, there are potential problems stemming from the requisite assumptions. The most crucial assumption is that the bolus injection of an amount of amino acid well in excess of the total body free pools of that amino acid does not stimulate the rate of protein synthesis. The available physiological evidence pertaining to this assumption will be discussed below. However, if this assumption is not correct, then one could still conclude that the traditional formula serves as the upper bound for the true FSR. This is because all observed synthesis is ascribed to the rate pre-existent before the flooding dose. To obtain a lower bound for the true FSR, we will derive, for the first time, an alternative flooding dose formula with the assumption that the stimulation of the protein synthesis rate will be, at most, proportional to the increase in the intracellular concentration of the infused amino acid (see Fig. 1). For example, if the leucine intracellular concentration was doubled as a consequence of giving a leucine flooding dose, then we will assume that the protein synthesis rate would be at most doubled. The calculation of the upper and lower bounds of the true FSR has two advantages. First, if the bounds are sufficiently close, then one can conclude that the true FSR has been obtained. Second, when one compares the calculation of the upper and lower bounds of the FSR of two different physiological states, then if the bounds of the FSRs of the two states do not overlap, one can conclude that there is a difference in the FSRs, regardless of the uncertainties of the relationship between the precursor enrichment and the synthesis rate. We will also show that when stable-isotope flooding doses are given, the free amino acid enrichments in the traditional formula need to be expressed as atom per cent excess (APE) rather than as the tracer/tracee ratio. Since the free amino acid enrichment has been in some cases as high as 20 APE [l], the difference between the APE and the tracer/ tracee ratio is as great as 20%. ASSUMPTIONS NECESSARY FOR THE TRADITIONAL DERIVATION The usual equation used to calculate the FSR after giving a constant infusion or a flooding dose of labelled amino acid is: Normal li Before After Before After u- lntracellular precursor Synthesis rate concn. (mmol of free (g of protein h - ' k g - ' ) precursor amino acid/kg) Before After w Before After u intracellular precursor Synthesis rate concn. (mmol of free (g of protein h - ' kg-') precursor amino acid/kg) Fig. I. Differences in the assumptions of the two models used to estimate FSR from a flooding dose. In the traditional model (a), it is assumed that increasing the intracellular concentration of the amino acid being infused causes no change in the protein synthesis rate. The alternative model (b) assumes that the increase in protein synthesis rate is proportional to the increase in the intracellular concentration of the amino acid infused. The purpose of looking at these two models is to obtain upper and lower bounds for FSR, for presumably the true response is not outside of these two boundaries. where E , and E , are the enrichments of free and bound amino acid, respectively. (Enrichment in the case of radioisotopes is the specific activity and, in the case of stable isotopes, is APE throughout the paper, except where noted. Tracer is the labelled infused amino acid and tracee is the endogenous amino acid and the unlabelled infused amino acid. As we shall see, APE is the proper way to express enrichments in the specific case of measuring protein incorporation.) A correct derivation of an equation similar to eqn. (1) is given by Zilversmit [3]. Zilversmit's formula was not derived with the flooding dose technique in mind, so we will review in detail the assumptions necessary for the derivation of Zilversmit's formula to apply. Assumption (1) of the traditional derivation is that the rate of incorporation of amino acid (both labelled and unlabelled) is constant and at its normal value over the duration of the experiment. This is reasonable in the constant tracer infusion method. In the flooding dose method, however, a bolus of amino acid of 50mg/kg body weight or more is given. Assessing the physiological effects of Mathematical validity of the flooding dose method administering a large quantity of a specific amino acid on protein synthesis is a difficult task. There is a large body of literature, which has recently been reviewed in [4], indicating that leucine flooding doses promote protein synthesis in vitro when other amino acids are plentiful. One should be cautious, however, when extrapolating from these results to the situation in vivo, where the availability of other amino acids might be limiting. If stimulation of protein synthesis does occur from the addition of only leucine, then one would have to presume that there is a reservoir of the other amino acids needed to produce the protein and that the stimulation occurs either because leucine might be the amino acid which is in shortest supply or because of a regulatory effect of leucine. It is not clear whether leucine alone stimulates muscle protein synthesis. On the one hand, data in uitro suggest a stimulatory role for leucine [4]. O n the other hand, McNurlan et al. [5] found that a flooding dose of phenylalanine gave the same results as a flooding dose of leucine. They concluded that a leucine flooding dose did not stimulate protein synthesis in vivo, since studies in vitro have not indicated that phenylalanine flooding doses stimulate protein synthesis [ 6 ] . Also, the addition of a large amount of leucine did not affect muscle FSR calculated by using a flooding dose of phenylalanine [SJ. O n the other hand, studies in which a constant infusion of another amino acid was simultaneously given with a flooding dose of leucine have produced conflicting opinions as to whether protein synthesis was promoted [7, 81. However, interpretation of results from this type of experimental protocol is made difficult by the fact that the transport rate between plasma and tissue of many amino acids changes when a single amino acid is infused [9], thereby potentially confounding the results from the constant infusion technique when a flooding dose is simultaneously administered. Thus the extent to which a flooding dose of leucine stimulates muscle protein synthesis in humans is unclear. Assumption (2) is that the enrichment or specific activity of the free amino acid pool at any time t is equal to the rate that labelled amino acids are incorporated into protein at time t divided by the rate that amino acids (both labelled and unlabelled) are incorporated into bound protein at time t. Assumption (2) may be criticized in that there may be a significant delay between the time when amino acids are no longer ‘free’, i.e. are bound to transfer RNA, and the time when amino acids are ‘bound’, i.e. are measurable as bound proteins. The delay time in liver cells is usually around 1-2min [lo]. In other systems, the delay may be longer, and may involve other factors. Failure to account for the delay causes an underestimation of the true value of FSR when the flooding dose technique is used, so that the fact that there may be an important delay factor can not explain the differences in the computed muscle FSR values in humans using the I79 constant infusion and flooding dose methods. On the contrary, an important delay factor would only amplify the differences. Assumption ( 3 ) is that none of the protein into which tracer is incorporated is degraded before the experiment ends. This assumption might be of slight concern during a prolonged tracer infusion for slowly turning over pools, because there may be some recycling of tracer. This assumption is reasonable in the flooding dose situation for muscle because all sample collection is complete within 2 h. For tissues with faster turnover rates than muscle, the sampling interval is proportionally shorter as well. Furthermore, it should be emphasized that eqn. (1) is valid for tissues that satisfy assumption ( 3 ) even if other tissues do not satisfy assumption ( 3 ) . In other words, if, for example, liver tissue fails to meet this requirement, then calculations involving the calculation of FSR in muscle are not invalidated. Assumption (4) is that amino acids can enter the bound amino acid pool only by incorporation from the free amino acid pool. This presumably is not controversial. Assumption (5) is that the pool size of bound protein (both labelled and unlabelled) is constant and at its normal value throughout the experiment. Some consideration may be necessary in long studies in small, young animals, since rats have been shown to grow at a rate of 5%/day [ll]. It is also of interest to point out the assumptions that are not needed for this calculation. N o assumptions are made about irreversible loss of free amino acid by routes other than incorporation into protein, whether the rate of this loss changes over the course of the experiment, or what the interactions are between free amino acids and tissues not being sampled. Since the kinetics of the free amino acids Tracee kinetics n 4 €FS { Bound pool k- Bound pool u Tracer kinetics Fig. 2. Model of amino acids needed to derive the formula to estimate FSR by the flooding dose technique. The only assumptions needed about the free amino acid pool are that the enrichment is known (EF) and the incorporation rate (S) (derived solely from the free amino acid pool) is constant. D. L. Chinkes et al 180 play no role in the traditional calculation of FSR, it is perhaps simpler to think in terms of a single-pool model (shown in Fig. 2), where the single pool represents bound protein. We see in this model that the only information that we need to know about the free amino acid pool is the enrichment of the free amino acid pool and the rate at which it delivers tracee to the bound amino acid pool. Some misconceptions should be cleared up about the misleading notation in more recent derivations of the FSR formula. For instance, in the derivation in [ll], the starting assumption is that k,, the FSR, is equal to k,, the fractional degradation rate. Confusion arises because in compartmental modelling the usual definition of k , in this case would be the incorporation rate into protein divided by the pool size of free amino acid. What is meant (although not stated) instead in reference [11] is that k , is the incorporation rate into protein divided by the pool size of bound amino acid. In the case of muscle, the magnitude of difference in these two definitions is great, since there is a large difference in the bound and free amino acid pool sizes. DERIVATION OF THE TRADITIONAL FSR FORMULA In this section we will go through a derivation of the traditional FSR formula using the set of assumptions discussed above. There are two ways of expressing enrichment commonly used in stable-isotope studies. In flooding dose experiments, it is common to express enrichment as APE, which is defined as the number of mols of tracer divided by the number of mols of tracer and tracee combined. Enrichment can also be expressed as the tracer/tracee ratio, which is defined as the number of mols of tracer divided by the number of mols of tracee. The use of tracer/tracee ratios has been generally advocated over the use of APE because in most tracer applications the formulae are less complex when expressed in terms of tracer/tracee ratio [12]. In the specific case of measuring protein synthesis rates, however, expressing enrichment in terms of APE results in less complex formulae. Conversion of tracer/tracee ratio to APE can be easily accomplished by the formula: APE = (1 (tracer/tracee ratio) + tracer/tracee ratio) Therefore, the formulae discussed below can be used if the tracer/tracee ratio is measured, provided that the above conversion is performed first. If we combine assumptions (3) (no release of tracer from protein) and (4) (entrance to bound pool only via incorporation from free pool), we can conclude that the rate of increase in the pool size of bound labelled amino acid at time t [qB(t)] is equal to the rate at which that labelled amino acid is incorporated into the bound pool at time t [S,(t)], i.e. Assumption (2) (enrichment of free pool =rate of labelled incorporation of amino acids into bound pool divided by rate of incorporation of amino acids into bound pool) states that the relationship between the rate at which labelled amino acid is incorporated into the bound pool at time t , the rate that amino acid is incorporated into the bound pool at time t [ S ( t ) ] and the enrichment or specific activity of free amino acid at time t [&(t)] is: If we combine eqns. (2) and (3), we get: If we divide both sides of eqn. (4) by the pool size of bound amino acid (qB+QB) we get: Assumption (5) states that the pool size of bound amino acid ( q B + Q B ) is constant, so the left-hand side of eqn. (5) is equal to the rate of change of enrichment or specific activity of the pool size of amino acid bound to protein. By definition, FSR is equal to the rate of incorporation of amino acid divided by the pool size of bound amino acid. Therefore, eqn. (5) can be written as It is assumed that FSR is constant, which is true if assumptions (1) (incorporation rate is constant) and (5) (bound pool size is constant) are true, Hence, if eqn. (6) is integrated from time zero, we find: Solving for FSR yields: Mathematical validity of the flooding dose method 181 which is the traditional formula. Note that this derivation of FSR is valid for both the constant infusion method and the flooding dose method, and any other incorporation method, given that the discussed assumptions are true. CALCULATION OF FSR IF THE PROTEIN SYNTHESIS RATE IS STIMULATED BY AN AMOUNT PROPORTIONAL TO THE INCREASE IN THE INTRACELLULAR FREE AMINO ACID POOL SIZE Assumption (1) of the traditional calculation is that the incorporation of free amino acids into protein does not change over the course of the experiment. The uncertainties involved with this assumption were enumerated in a previous section. The specific issue which will be addressed in this section is how the calculation of FSR is affected if assumption (1) is changed to assumption (1'): the incorporation rate of amino acids into protein is stimulated by an amount proportional to the intracellular concentration of free amino acids (see Fig. 3). In other words, the rate of incorporation of the endogenous tracee would not be affected by the flooding dose, and any tracee or tracer from the flooding dose incorporated into protein would therefore represent an increased rate of incorporation. This alternative assumption would require that endogenous amino acids other than the amino acid being traced would be incorporated into the protein at an increased rate, i.e. that protein synthesis is stimulated. It is important to note that this model is based on the rate of incorporation of the endogenous tracee, not the total rate of incorporation of tracer and tracee. The traditional model is based on the rate of incorporation of exogenous and endogenous tracer and tracee. Stated mathematically, suppose that: where SBo,(t)is the rate of incorporation of the bolus at time t, s , , d ( t ) is the rate of incorporation of the endogenous free amino acid not from the bolus, and S ( t ) is the total incorporation rate of unlabelled free amino acid. The consequence of choosing assumption (1') rather than assumption (1) is that sE,d(t) is constant, not S(t). We wish to define FSR as &,,d/(qB QB), since we are interested in the FSR that would have occurred if we had not given a flooding dose. Another assumption [assumption (6)] that is needed is that E,, the ratio of the labelled bolus size to the total bolus size, is equal to the rate at which the labelled bolus is being incorporated into bound protein divided by the rate at which all of the bolus is being incorporated into protein. In other words, an injected labelled amino acid molecule has the + Total incorporation Endogenous Before After -+ Before After 'rate (S) Endogenous SE" 'incorporation rate when bolus given lntracellular Drecursor IncorDoration rate concn. (mmoi of free (minil of tracee and tracer amino precursor amino acid incorporated h - ' kg -I) acid/ kg) Total incorporation rate (S) Endogenous +bolus Endogenous I I Belore After incorporation rate when bolus given Before After u lntracellular precursor Incorporation rate concn. (mmol of free (mmol of tracee and tracer amino acid incorporated h - l kg - I ) precursor amino acid/ kg) Fig. 3. Detailed look a t the assumptions underlying the t w o models. In the traditional model (a), the total incorporation rate of amino acids is unchanged. In the new model (b), the rate that amino acids are incorporated (S) is divided into the rate at which amino acids from the bolus are incorporated (Seal) and the rate at which endogenous unlabelled amino acids are incorporated (SEnd). Finding the relative proportions of these two rates at any given time so that the endogenous incorporation rate can be found is accomplished using the enrichment of the infusate (El) and the ratio of labelled to total unlabelled amino acids at that given time [EF(t)]. Endogenous; 0, unlabelled bolus; N, labelled bolus. .. same odds of being incorporated into a particular protein as an unlabelled amino acid molecule injected at the same site. There should be no problems with this assumption. From this last assumption we have: (9) where SL is the rate of incorporation of the labelled bolus. The incorporation of endogenous amino acid is the difference between the total incorporation rate and the rate at which the bolus is being incorporated, i.e. SEnd =s(t) -s ~ 3 o d t ) (10) Substituting eqn. (9) and then eqn. (3) into eqn. (10) yields: D. L. Chinkes et al. I82 [ y] 7 =S(t) 1- *t Solving eqn. (1 1) for S(t) yields: S(t ) = * * SEnd ~~ By assumption (l'), S E n d is constant, so if we substitute eqn. ( 2) into eqn. ( 5 ) and go through the same reasoning sed to get the traditional equation, we get: 45 90 I20 Constant infusion Time that bound enrichment is sampled (min) (m) jb [#] dt where Fig. 4. Comparison of new (0) and old flooding dose formulae (with a flooding dose of 27% enriched [I,2-13C]leucine) at different sampling times with constant infusion results (using [U-"C]leucine) using data from a comparative study in muscle of anaesthetized postabsorptive dogs. *, indicates that the difference in the FSR computed by the old and new formulae is statistically different from zero (P<O.OI); 7. indicates that the FSR computed by the old formula is statistically different from the FSR computed by the old formula at either 90 or 120min (P<O.Ol). Data are from [El. If physiological evidence proves that protein synthesis is stimulated by the flooding dose in proportion to the increase in the free intracellular amino acid pool, then the traditional equation can be corrected by using eqn. (14). RESULTS A comparison in muscle of results obtained by using the two flooding dose formulae and the constant infusion technique in muscle tissue in anaesthetized postabsorptive dogs using a flooding dose of 50mg of 27% enriched [1,2-'3C]leucine/kg and a constant infusion of [U-'4C]leucine is shown in Fig. 4, using the data from [S]. Using the new formula based on assumption (l'), the results from the constant infusion study (1.8 fO.l%/day) were not statistically different from the results obtained in the flooding dose study (ranging from 2.0f0.2 to 2.2 f0.3%/day). The old flooding dose formula [based on assumption (l)] gave values significantly in excess of those obtained with the new flooding dose formula and the constant infusion technique ( P <0.01). The old formula applied to the samples taken at 45 min (FSR = 5.0 k0.43%/day) gave values which were statistically different ( P < 0.01) than were obtained by applying the old formula to the samples taken at 90 and 120min (3.410.4 and 3.5 f0.4"/,/day). When eqn. (13) is used in a comparative study in muscle in postabsorptive man (with a flooding dose of 50mg of 20% enriched ~-[l-'~C)leucine/kgand a constant infusion of ~-['~C]valine)[7], the average calculated FSR is 0.025 0.0027Jh as compared with 0.043kO.O02%/h as measured by the constant infusion method and 0.063 1O.OOS%/h as measured - 0.08 i 0.06 ..-.. E. e 0.04 2 r, 0.02 0.00 1 2 3 4 5 Subject no. 6 1 (m) Fig. 5. Comparison of new (H) and old flooding dose formulae with constant (with flooding dose of 20% enriched ~-[~["C]leucine) using data from a comparainfusion results (using ~-["C]valine, 0) tive study in muscle of postabsorptive man. Data are from [7]. by the traditional FSR formula. These values are all statistically different ( P <0.01). The results for the individual subjects are shown in Fig. 5. In all seven cases of this experiment, FSR as computed by the constant infusion method was between the values calculated from the two flooding dose formulae. The bound protein samples were measured 90min after the flooding dose was given. DISCUSSION In this paper we have for the first time correctly derived the traditional flooding dose equation for Mathematical validity of the flooding dose method the calculation of the FSR of a product when a stable-isotope bolus dose of tracer is administered, with assumptions appropriate for the application of the formula to the flooding dose technique, showing that the enrichments in the formula should be expressed as the APE (as they commonly are) rather than as the tracer/tracee ratio. In this traditional approach it is assumed that protein synthesis is not stimulated by the flooding dose. We have also derived a new, alternative equation. In this case, the alternative assumption is made that FSR is increased by a rate in proportion to the change in the intracellular concentration of the amino acid. Use of these two equations provides the upper and lower bounds around the true value of FSR calculated with a flooding dose in human and dog muscle. The choice of the most appropriate equation equation depends on the extent to which the flooding dose stimulates FSR. This latter question has not been completely resolved. In the case of the study in the dog, the new formula corresponds well with the constant infusion data, suggesting a stimulation of protein synthesis in accord with assumption (1’). The human data suggest that there is an increase in protein synthesis, although not in proportion with the increase in intracellular free leucine concentration. The bounds given by the two flooding dose formulae are reasonable, although they are not very ‘tight’. Taking into account a delay factor may cause the new formula to be a better predictor of the true value, but it can only cause the old formula to be a poorer predictor. It is not possible to account for the discrepancies between the results obtained with the traditional flooding dose FSR formula and those determined by the constant infusion method by errors in calculating FSR when using the constant infusion method. It was pointed out in the derivation of the traditional flooding dose FSR formula that the underlying assumptions and resulting formulae were identical for the two methods. The advantage of the constant infusion technique is that there is no doubt that the protein synthesis is not significantly affected by a constant infusion of tracer. The advantage of the flooding dose technique is that there is little doubt as to the proper precursor (free amino acid) enrichment. The true precursor enrichment (transfer RNA enrichment) is not readily measurable, and it is not certain what substitute for that value is optimal. I83 However, none of the candidates commonly used (i.e. intracellular and extracellular free amino acid, or a marker of intracellular enrichment such as GIketoisocaproate) result in calculated values that can explain the difference between the constant infusion technique and the traditional flooding flooding dose FSR calculation. A common convention to deal with the uncertainties in constant infusion studies is to state that the FSR lies between the bounds determined by the intracellular and extracellular enrichments. We suggest that the two flooding dose formulae described here can be used as bounds of the true FSR value in muscle in humans, keeping in mind that if there is a significant delay in the labelling of the transfer RNA pool, the values from both equations will be too high. ACKNOWLEDGMENTS This work was supported in part by grants from the National Institutes of Health (DK 33952, 38010) and a grant from the Shriner’s Hospital. REFERENCES I. Garlick PJ, Wernerman J, McNurlan MA et al. Measurement of the rate of protein synthesis in muscle of postabsorptive young men by injection of a ‘flooding dose’ of [I-i3C]leucine. Clin Sci 1989; 77: 329-36. 2. Henshaw EC, Hirsch CA, Morton BE, Hiatt HH. Control of protein synthesis in mammalian tissues through changes in ribosome activity. J Biol Chem 1971; 246: 436-46. 3. Zilvenmit DB. The design and analysis of isotope experiments. Am J Med 1960; 29: 8 3 2 4 . 4. May ME, Buse MG. Effects of branched-chain amino acids on protein turnover. Diabetes/Metab Rev 1989; 5 22745. 5. McNurlan MA, Fern EB, Garlick PJ. Failure of leucine t o stimulate protein synthesis in vivo. Biochem J 1982; 2M 831-8. 6. Manchester KL, Tyobeka EM. Influence of individual amino acids on incorporation of [i4C]leucine by rat liver ribosomes. J Nutr 1980; 110: 241-7. 7. Smith K, Barua JM, Watt PW, Scrimgeour CM, Rennie MI. Flooding dose of I-[l-’3C]leucine stimulates human muscle protein incorporation of continuously infused ~-[l-~~C]valine. Am J Physiol 1992; 262: E372-6. 8. Jahoor F, Zhang X-J, Baba H, Sakurai Y, Wolfe RR. Comparison of constant infusion and flooding dose techniques t o measure muscle protein synthesis rate. J Nutr 1992; 112: 878-87. 9 Cespedes C De, Thoene JG, Lowler K, Christensen HN. Evidence for inhibition of exodus of small neutral amino acids from non-brain tissues in hyperphenylalaninaemic rats. J lnher Metab Dis 1989; 12: 166-80. 10. Waterlow JC, Garlick PJ, Millward DJ. Protein turnover in mammalian tissues and in the whole body. Amsterdam: Elsevier-North Holland, 1978. ( I . Samarel AM. In vivo measurements of protein turnover during muscle growth and atrophy. FASEB 1. 1991; 5 2 0 2 H . 12. Rosenblatt J, Wolfe RR. Calculation of substrate flux using stable isotopes. Am J Physiol 1988; 254 E52631.