We have examined the effect of a hemodialysis-induced 40% reduction in plasma amino acid concentrations on rates of muscle protein synthesis and breakdown in normal swine. Muscle protein kinetics were measured by tracer methodology using [2H5]phenylalanine and [1-13C]leucine and analysis of femoral arterial and venous samples and tissue biopsies. Net amino acid release by muscle was accelerated during dialysis. Phenylalanine utilization for muscle protein synthesis was reduced from the basal value of 45 ± 8 to 25 ± 6 nmol · min−1 · 100 ml leg−1 between 30 and 60 min after start of dialysis and was stimulated when amino acids were replaced while dialysis continued. Muscle protein breakdown was unchanged. The signal for changes in synthesis appeared to be changes in plasma amino acid concentrations, as intramuscular concentrations remained constant throughout. The changes in muscle protein synthesis were accompanied by a reduction or stimulation, respectively, in the guanine nucleotide exchange activity of eukaryotic initiation factor (eIF)2B following hypoaminoacidemia vs. amino acid replacement. We conclude that a reduction in plasma amino acid concentrations below the normal basal value signals an inhibition of muscle protein synthesis and that corresponding changes in eIF2B activity suggest a possible role in mediating the response.
- eukaryotic initiation factor 2B
- stable isotopes
over the past several years, it has become clear that an increase in the concentrations of plasma amino acids, particularly the essential amino acids, can stimulate the rate of muscle protein synthesis (4,8). However, the response of muscle protein synthesis to changes in concentrations in the opposite direction, i.e., reductions below the normal basal values, is less clear. Furthermore, there is limited information regarding the effect of amino acid concentrations on the rate of muscle protein breakdown (22). Results from studies performed in vitro suggest that muscle protein breakdown is responsive to changes in amino acid availability (e.g., Ref.16), but there have been few in vivo studies specifically addressing this issue (2). Therefore, the principal goal of this study was to examine the effect of a reduction in plasma amino acid concentrations below the normal postabsorptive levels on the rates of muscle protein synthesis and breakdown. We used hemodialysis to reduce amino acid concentration.
A second goal of our study was to attempt to identify the signal for changes in the rate of synthesis or breakdown when the plasma concentrations of amino acids drop. Whereas it is reasonable to presume that effects of amino acid availability would be mediated by changes in intracellular availability, our previous studies have shown remarkably small changes in intracellular concentrations despite significant changes in plasma concentrations (8, 22). However, in previous studies, intracellular samples were obtained in physiologically steady-state circumstances, usually after 2 h or more of continuous infusion of amino acids (e.g., Ref. 8). Thus, in this study, we sampled frequently from the femoral artery and vein, as well as from muscle tissue, as changes in concentration were induced by dialysis. We hoped that a more detailed characterization of the time course of changes in concentrations, synthesis, and breakdown would enable us to gain insight regarding the signal for changes in muscle protein synthesis and breakdown.
A third goal of our study was to investigate potential mechanisms by which a reduction in plasma amino acid concentrations might mediate a change in muscle protein synthesis. For this purpose, we chose to examine the guanine nucleotide exchange activity of a protein required for the initiation of mRNA translation, i.e., eukaryotic initiation factor (eIF)2B. The activity of eIF2B has been shown to be rate limiting for protein synthesis under a number of circumstances in which the availability of amino acids is limited (12).
Ten normal Yorkshire swine (K-bar live stock, Sabinal, TX), weighing 46.6 ± 12.8 (mean ± SD) kg, were used. This experimental protocol was approved by the Institutional Animal Care and Use Committee of the University of Texas Medical Branch (IACUC 00–03–017). The animals were fasted overnight before the study. The study was performed under general anesthesia of ∼2% isoflurane inhalation. Depth of anesthesia was monitored throughout the experiment by heart rate and blood pressure, and appropriate modifications in the rate of anesthesia administration were made to ensure a stable physiological condition throughout the experiment.
A Swan-Ganz catheter was inserted via the right external jugular vein, and the tip was advanced to the pulmonary artery for infusions of isotope tracers and amino acids. Body temperature was monitored using a Swan-Ganz catheter and kept around 37°C using a heating blanket. A double-lumen catheter was placed in the left external jugular vein for hemodialysis. A third catheter was inserted into the carotid artery for arterial blood sampling and blood pressure and heart rate monitoring. An indwelling catheter was placed in the femoral vein for venous blood sampling. A flow probe (Transonic Systems, Ithaca, NY) was placed on the femoral artery and connected to a flowmeter (Transonic Systems T106) to measure femoral arterial blood flow.
Two experiments were performed separately (Fig.1). At the start of both experiments, blood was taken to determine background enrichment. Thereafter, an infusion ofl-[ring-2H5]phenylalanine (0.10 μmol · kg−1 · min−1; prime, 4.0 μmol/kg) and l-[1-13C]leucine (0.16 μmol · kg−1 · min−1; prime, 9.6 μmol/kg) was given into the pulmonary artery (at −240 min; Fig. 1) and maintained until the end of the experiment. Hemodialysis was started at 0 min and continued for 4 h inexperiment 1 [dialysis only (DIA), n = 4], which was designed as “a time control” for experiment 2. In experiment 2 [dialysis + amino acid replacement (DIA + REP)], six animals received hemodialysis from 0 min for 4 h. However, after 2 h, a primed continuous infusion of a mixed amino acid solution was infused. The infusion rates of the individual amino acids are shown in Table1. The amino acid infusion was maintained for the remaining 2 h of the study to replace the amino acids lost in the dialysate determined from experiment 1 (in a pilot study). The amino acids contained an appropriate amount ofl-[ring-2H5]phenylalanine and l-[1-13C]leucine to prevent the blood enrichments from changing.
The replacement rates of amino acids were determined empirically. The four experiments in the DIA group were performed first. On the basis of the changes in arterial amino acid concentrations during dialysis in these experiments and dialysis extraction rates (determined from amino acid concentrations in the dialysate), the priming doses and infusion rates of the individual amino acids in the replacement mixture were determined. Specifically, changes in the individual amino acid concentrations in arterial plasma were found by subtracting arterial concentrations at the 120-min time point (i.e., after dialysis) from the concentrations at the zero time point (before start of dialysis). The priming doses for the replacement were calculated as concentration changes multiplied by amino acid distribution volume, which was assumed to be 200 ml/kg body wt. To determine the composition of the replacement mixture, the average ratio between amino acid concentrations in dialysate and in the arterial plasma was found, and this was multiplied by plasma amino acid concentrations at the zero time point (with the assumption that basal amino acid concentration would be maintained in the absence of hemodialysis) and dialysate flow rate (900 ml/min). The composition of the mixture was based on a 50-kg pig and an infusion rate of 100 ml/h. To prepare the mixture, individual amino acids (Ajinomoto, Tokyo, Japan) were mixed with double-distilled H2O. Because the solubility of tyrosine is poor, the dipeptide glycine-tyrosine was used. The dose of free glycine was adjusted accordingly. On the basis of the responses (increase in arterial amino acid concentrations) to the replacement mixture in the first three experiments in the DIA + REP group, modest modifications were made in the doses of amino acids.
A polymethyl methacrylate dialyzer (Filtryzer B3–1.6A; Toray, Tokyo, Japan) was used. The effective surface area of the dialyzer was 1.6 m2, and the blood volume of the dialyzer was 95 ml. The volume of dialyzer plus tubes was replaced with saline at the start of dialysis. The flow rate of blood through the dialyzer was 450 ml/min, and the dialysate flow rate was 900 ml/min. The dialysate composition was (in meq/l) 139–145 sodium, 2.0–3.0 potassium, 3.0–3.5 calcium, 1.0 magnesium, 4.0 acetate, and 107 chloride, and 2.0–2.5 g/l dextrose (NaturaLyte, Fresenius Medical Care, Lexington, MA, and Renasol BC-1-L, Minntech Renal Systems, Minneapolis, MN).
Arterial blood samples to determine total amino acid and electrolyte concentration and pH were taken throughout the experiment from the carotid artery. Blood samples to determine phenylalanine and leucine enrichments and concentrations were simultaneously taken from the carotid artery and femoral vein at frequent intervals and immediately placed into tubes containing 15% sulfosalicylic acid. Muscle samples were obtained by biopsies from the gracilis muscle at 0, 15, 30, 60, 90, 120, 150, 180, 210, and 240 min in experiment 1 and at −30, 0, 15, 60, 120, 135, 180, and 240 min inexperiment 2. Tissue biopsies were ∼1 g each. One-half of the biopsies was taken from each leg, and the biopsy sites were separated by ≥3 cm to ensure that samples would not be affected by a local response to a previous biopsy. A portion of the tissue was immediately frozen in liquid nitrogen and stored at −80°C until analysis. Another portion was processed for analysis of eIF2B activity, as described in Measurement of eIF2B activity. Femoral arterial blood flow rate was recorded at each sampling.
Measurement of eIF2B Activity
A portion of each sample of gracilis muscle was homogenized in 4 volumes of buffer consisting of (in mM) 45 HEPES (pH 7.4), 0.375 magnesium acetate, 0.075 EDTA, 95 potassium acetate, 2.03 digitonin, 3 microcystin, and 10% glycerol and for the measurement of eIF2B activity. The guanine nucleotide exchange activity of eIF2B in 100,000-g supernatant was measured by exchange of [3H]GDP bound to eIF2 for nonradioactively labeled GDP, as described previously (1).
Arterial and femoral venous blood for measurement of enrichments was precipitated with 15% sulfosalicylic acid, centrifuged, and separated. The supernatant was passed through a cation exchange column and eluted with 4 M NH4OH. The eluted solution was dried under vacuum with a speed-vac (Savant Instrument, Farmingdale, NY), and the dried amino acids were then reconstituted witht-butyldimethylsilyl and acetonitrile (50:50) and derivatized on a heating block for 1 h at 90°C. Muscle samples were weighed (∼30–40 mg), and the proteins were precipitated with 800 μl of 14% perchloric acid. The tissue was then homogenized and centrifuged, and the supernatant was collected, passed through a cation exchange column, and derivatized as detailed above. Enrichments of phenylalanine and leucine were measured by gas chromatography-mass spectrometry (Hewlett-Packard 5973) as thet-butyldimethylsilyl derivative (22). Concentrations of phenylalanine and leucine were calculated by an internal-standard method (6). Arterial blood pH was measured by a blood gas analyzer (Instrumentation Laboratory, Lexington, MA) immediately after sampling. Plasma samples separated from arterial blood were analyzed for amino acid concentrations by high-performance liquid chromatography (HPLC System 2690; Waters Alliance, Milford, MA) and for insulin concentration by a radioimmunoassay method (Porcine Insulin RIA Kit; Linco Research, St. Charles MO). Serum samples separated from arterial blood were analyzed for electrolyte (sodium, potassium, and chloride) concentrations by VITROS 250 Chemistry System (Ortho-Clinical Diagnostics, Raritan, NJ).
The three-compartment model of leg muscle amino acid kinetics has been described elsewhere (6). Briefly, net balance (NB), intracellular disappearance (FO,M), and intracellular appearance (FM,O) of the amino acids were calculated as follows where CA and CV are free phenylalanine or leucine concentrations in femoral arterial and venous blood, respectively; EA, EV, and EM are phenylalanine or leucine enrichments in femoral arterial and venous blood and muscle, respectively; and BF is femoral arterial blood flow rate. Averages for blood flow and blood and muscle amino acid concentrations and enrichments, which were calculated from individual samples drawn during −30–0 min (Basal), 30–60 min (DIA1), 90–120 min (DIA2), 150–180 min (DIA3), and 210–240 min (DIA4) respectively, were used for the model calculations. When phenylalanine is used as a tracer, intracellular disappearance (FO,M) represents utilization for muscle protein synthesis, and intracellular appearance (FM,O) represents appearance from protein breakdown, because phenylalanine is neither oxidized nor synthesized in muscle. In the case of leucine, which is oxidized in muscle, intracellular disappearance (FO,M) represents the rate of utilization for muscle protein synthesis plus oxidation. The ratio of leucine to phenylalanine in muscle protein roughly equals the ratio of intracellular appearance of leucine [FM,O(Leu)] to that of phenylalanine [FM,O(Phe)]. Utilization of leucine for protein synthesis [PS(Leu)] and oxidation of leucine [Ox(Leu)] were calculated as follows where FO,M(Phe) and FO,M(Leu) are intracellular disappearance of phenylalanine and leucine, respectively.
Data are expressed as means ± SE. Treatment response was assessed by ANOVA with repeated measures. If differences were detected (P < 0.05), then a Fisher's paired least significant difference post hoc test was performed to determine pairwise differences.
Physiological parameters are shown in Table2. In both experiments, arterial blood pressure fell throughout dialysis, and the extent of reduction was similar in both experiments. Leg blood flow fell significantly inexperiment 1 but not in experiment 2. Neither blood pressure nor leg blood flow was affected by the replacement of amino acids in experiment 2. Heart rate was constant throughout both experiments. The plasma pH and electrolytes are shown in Table 3. In both experiments, there was a small, but significant, increase in pH during dialysis. However, in experiment 2, the increase was only from a basal value of 7.426 ± 0.018 to a value of 7.471 ± 0.013 at the end of dialysis, and the pH during hemodialysis in experiment 2 was not different from that in hemodialysis plus amino acids. Whereas modest changes in sodium, potassium, and chloride were statistically significant at selected time points, there were no changes of physiological significance. In experiment 1, the basal insulin concentration was 2.2 ± 0.2 μU/ml, and there were no changes throughout dialysis. In experiment 2, amino acid replacement during dialysis significantly increased insulin concentrations (P < 0.01), but the peak value was only 5.0 ± 0.8 μU/ml at the end of the 2 h of replacement.
Amino Acid Concentrations
The time course for total amino acid concentration in arterial plasma is shown in Fig. 2. Inexperiment 1, within 30 min after the start of hemodialysis, total arterial plasma amino acid concentration decreased ∼65% (P < 0.01) and remained approximately at that level for the remainder of the experiment. In experiment 2, total plasma amino acid concentration decreased ∼50% during hemodialysis and returned to 94% of the basal value when amino acids were replaced (P < 0.01). Concentrations of phenylalanine and leucine in arterial blood were decreased by hemodialysis in a manner corresponding to the changes in total amino acid concentration, and the infusion of amino acids over the last 2 h of experiment 2 successfully restored the basal concentrations of each amino acid (Table 4). In contrast to the changes in plasma concentration, the intramuscular phenylalanine concentration did not fall at any time point after the start of hemodialysis (Table 4). Similarly, intramuscular leucine concentration did not fall over the first 2 h of dialysis in experiment 1, but by the end of dialysis in experiment 1, intramuscular leucine had dropped to 80% of the basal level (Table 4). In experiment 2, the intramuscular concentration of phenylalanine remained constant throughout the experiment, whereas intramuscular leucine concentration increased ∼15% during amino acid replacement (Table 4).
Phenylalanine and leucine kinetics across the leg.
In all experiments, isotopic enrichments of phenylalanine and leucine were essentially constant. The average enrichments for experiment 2 are shown in Fig. 3, demonstrating that the addition of tracer to the replacement amino acids successfully maintained the isotopic equilibrium.
Figure 4 shows phenylalanine and leucine kinetics across the leg in both experiments. Net balance of phenylalanine across the leg was decreased by hemodialysis inexperiment 1 (P < 0.01), indicating net muscle catabolism. Net leucine balance changed correspondingly, but the differences were not significant (P = 0.13 and 0.70, respectively). The net balance of both phenylalanine and leucine was decreased significantly (P < 0.01) by hemodialysis inexperiment 2 (P = 0.03). Net balance of phenylalanine remained low during the replacement of amino acids, but net leucine balance returned to the basal value.
Utilization of phenylalanine for protein synthesis was significantly reduced within 15 min of the start of dialysis and remained depressed throughout the remainder of the experiment (P < 0.01) (Fig.5). In contrast, the release of phenylalanine from protein breakdown was unchanged throughout the study (P = 0.65). Similarly, leucine utilization for protein synthesis decreased promptly and remained depressed throughout hemodialysis (P < 0.01), whereas leucine released from protein breakdown did not change throughout the study (P = 0.71) (Fig. 6). When amino acids were replaced during dialysis in experiment 2, utilization of phenylalanine for muscle protein synthesis was significantly increased (P < 0.01), whereas release of phenylalanine from protein breakdown was unchanged throughout the study (Fig. 5). Leucine utilization for protein synthesis was reduced ∼50% by hemodialysis and returned to 84% of the basal value during the amino acid replacement period in experiment 2(P < 0.01) (Fig. 6). As was the case for phenylalanine, release of leucine from protein breakdown was unchanged throughout the study (P = 0.20). Also, leucine oxidation was reduced by hemodialysis and returned to the basal value when amino acids were replaced (Fig. 6). The ratio of leucine oxidation to utilization of leucine for muscle protein synthesis did not change significantly regardless of the rate of leucine uptake and averaged ∼0.23.
We sought to define the mechanism by which hemodialysis-induced hypoaminoacidemia and subsequent amino acid replacement alters the synthesis rate of total mixed proteins in skeletal muscle by measuring the activity of eIF2B. The guanine nucleotide exchange activity of eIF2B can be estimated in whole muscle extract by use of an in vitro assay that measures over time the release of [3H]GDP bound to purified eIF2. With the use of this assay, it was determined that, 15 min after the commencement of hemodialysis, eIF2B activity in skeletal muscle was significantly reduced compared with basal (zero) time point (Fig. 7). This inhibition was sustained throughout the remainder of the dialysis time course. Thus the data suggest that reducing amino acid concentrations in the blood results in a reduction in the eIF2 · GTP binary complex, which presumably slows formation of the ternary initiation complex in skeletal muscle. During amino acid replacement, the guanine nucleotide exchange activity of eIF2B remained suppressed for the 1st h, but by the 2nd h of replacement it had rebounded and reached a value threefold higher than time zero.
A principal finding of this study was that a fall in plasma amino acid concentrations induced by hemodialysis inhibited muscle protein synthesis but did not affect muscle protein breakdown. These observations are consistent with the previous findings that whole body protein synthesis is reduced during hemodialysis in renal failure patients (13) and after hemodialysis in normal volunteers (14). However, these earlier studies did not isolate the effect of the decrease in amino acids per se, and they did not determine the response of muscle protein breakdown. We have shown in the present study that the reduction in muscle protein synthesis occurs rapidly at the onset of dialysis and parallels the drop in the concentration of plasma, rather than intracellular, amino acids. Finally, changes in synthesis generally corresponded to changes in the activity of the eukaryotic initiation factor eIF2B.
The role of amino acids as regulators of muscle protein synthesis has become well established over the past several years, but the effect of amino acids has been assessed in vivo entirely in experiments in which amino acid concentrations were increased either by infusion (e.g., Refs. 4, 8) or ingestion (e.g., Ref.20). However, the response to a reduction in amino acid concentrations is also important in understanding the regulation of muscle protein synthesis, as certain hormonal responses (i.e., increase in insulin) result in a fall in plasma amino acid concentrations below the basal value (23). However, it is difficult to experimentally produce an isolated reduction in amino acid concentrations. For example, restriction of dietary intake of protein does not affect basal amino acid concentrations (10). Therefore, in this study, we used hemodialysis to reduce amino acid concentrations below the normal postabsorptive value. It is well known that hemodialysis lowers amino acid concentrations (14). However, hemodialysis is normally performed in renal failure patients, and in that circumstance a number of other factors, such as pH, that could also affect muscle protein synthesis, are affected by hemodialysis. Therefore, we performed dialysis in normal pigs to minimize the overall physiological effect. Nonetheless, even in normal animals, dialysis could potentially elicit physiological responses that could affect muscle protein metabolism. For example, hemodialysis may give rise to an inflammatory response (5, 9). Therefore, in one group of animals, after 2 h of dialysis, we replaced the amino acids lost in the dialysate while continuing dialysis for another 2 h. With this approach, we were able to directly assess for the first time the effect of a reduction in amino acid concentrations below the basal level on muscle protein synthesis and breakdown. It is possible that the response to replacement of amino acids was not entirely due to the amino acid effect, because when the amino acids were replaced, the plasma insulin concentration also increased significantly. However, the insulin concentration rose to only 5 μU/ml, which is unlikely to have had much effect. Also, although insulin at higher concentrations (e.g., 77 μU/ml in Ref.7) has a direct stimulatory effect on muscle protein synthesis, an adequate availability of amino acids is required for insulin to exert such a stimulatory effect. It has been consistently shown that insulin fails to stimulate muscle protein synthesis when amino acid concentrations are allowed to fall (2,15).
We used two amino acid tracers so that our conclusions would not be entirely dependent on the assumption that changes in one essential amino acid are representative of all essential amino acids. The fact that the leucine balance data indicated a restoration of the basal value when amino acids were infused and the phenylalanine balance remained negative during the replacement is evidence that, during acutely changing circumstances, it may be questionable to assume that the metabolism of a single amino acid tracer accurately reflects protein metabolism within muscle. Nonetheless, the calculated rates of synthesis and breakdown showed parallel responses to the changes in amino acid concentrations, which supports the validity of the values calculated with each tracer. Furthermore, we were able to calculate the rate of leucine oxidation as the difference between total leucine disappearance and leucine incorporation into protein. It was previously published that, over a wide range of leucine intakes, the rate of whole body leucine oxidation (as determined by collection of13CO2 during [13C]leucine infusion) was a relatively constant function of uptake (∼19%) (23). Thus our estimation that ∼23% of muscle intracellular disappearance of leucine (FO,M) was oxidized over a range of uptake rates is consistent with previously published whole body data. Because our oxidation values were not directly measured but rather derived from the same tracer data from which the synthesis and breakdown rates were calculated, the agreement between our value of fractional oxidation of leucine and that of previous whole body studies in which leucine oxidation was measured directly supports the validity of our calculation of the rates of synthesis and breakdown.
This is the first report of the response of muscle protein breakdown to a reduction in plasma amino acid concentrations to values below the normal postabsorptive concentrations. Several lines of evidence support the notion that an increase in breakdown might have been expected. For example, in vitro experiments (e.g., Ref. 16) indicate that amino acids may have a suppressive effect on muscle protein breakdown. Also, the accelerated rate of breakdown following burn injury (11) or exercise (8) in human subjects is ameliorated by exogenous amino acids. On the other hand, increasing plasma amino acid concentrations have consistently failed to affect the basal rate of muscle protein breakdown (e.g., Ref. 20). A possible explanation for these apparently discrepant results is that, in a wide variety of circumstances, there is a close link between the rates of muscle protein synthesis and breakdown (22), and this link may supersede any direct influence of amino acid concentrations on muscle protein breakdown. Thus, in either anabolic (e.g., Ref. 8) or catabolic (e.g., Ref. 11) states, the extent of changes in muscle protein synthesis and breakdown differs with respect to each other, but the two factors generally change in the same direction. The mechanism responsible for the linking of muscle protein synthesis and breakdown is not clear or whether in fact they are mechanistically linked in all circumstances. An alternative possibility is that, unlike the process of synthesis, breakdown is principally responsive to changes in intracellular essential amino acid concentrations. In this case, a change in breakdown would not have been expected during dialysis because the intracellular concentrations remained essentially constant.
Our experiment was designed to define the time course of the response to changes in amino acid concentration to gain insight into the nature of the signaling system responsible for the changes in synthesis. We (22) have previously discussed the notion that the rates of amino acid transmembrane transport are regulated to maintain relatively constant concentrations of intramuscular essential amino acids even when changes in plasma concentrations occur. However, from previous experiments, it was unclear whether transient changes in intracellular concentrations occurred that then signaled changes in transport (and possibly synthesis) that restored normal intracellular concentrations, or, rather, whether the intracellular amino acid pool was regulated even more tightly, such that significant changes in concentrations normally did not occur. In the present study, we addressed this issue by sampling muscle tissue frequently and found that, even in the early (i.e., first 15 min) phase of response to hemodialysis, intracellular concentrations remained constant as plasma concentrations fell. This suggests that it was the change in plasma, as opposed to intracellular, amino acid concentrations that was the signal for the reduction in synthesis. The correspondence in the time course of the response of synthesis to the change in plasma amino acid concentrations further supports the regulatory role of the plasma, as opposed to intracellular, concentrations. The mechanism whereby changes in plasma concentrations could signal the intracellular synthetic process is unclear, but it has been shown that there is a leucine-specific binding protein on the plasma membrane of isolated hepatocytes that regulates intracellular protein metabolism (17). Perhaps a similar protein exists on muscle cell membranes that transmits signals to the intracellular compartment.
The delivery of amino acids to the leg is the product of concentration times flow, and, for reasons that are unclear, the leg blood flow dropped significantly during dialysis in experiment 1(dialysis only) but did not fall in experiment 2(replacement). Nonetheless, it is unlikely that the blood flow changes affected the responses. In experiment 2, the replacement of amino acids was started after 2 h of dialysis, meaning that during the first 2 h of dialysis the animals in the replacement group were treated the same as the animals receiving only dialysis. Even though there was a lower blood flow rate in experiment 1than in experiment 2 during that time, the falls in net balance were not statistically different, and in fact the falls inexperiment 2 were numerically greater than inexperiment 1. Thus it appears that changes in concentration, rather than delivery, were the principal signal for changes in muscle protein synthesis.
There are at least two potential mechanisms whereby a reduction in plasma amino acid concentrations could limit muscle protein synthesis. First, synthesis could simply be limited by amino acid availability, i.e., deficient charging of tRNA. Whereas charging of tRNA is generally considered to rarely limit muscle protein synthesis in the basal state (18), in vivo measurements of muscle tRNA charging in conditions of decreased amino acid availability have not been made. However, it has been shown that the true precursor enrichment for muscle protein synthesis is best reflected by the intracellular free amino acid pool rather than the plasma pool (3). Because there was no change in the availability of amino acids in the intracellular pool, a decrease in the charging of tRNA is not a likely explanation for the decrease in synthesis. An alternative explanation is that changes in eIFs were responsible for the decrease in synthesis.
Translation initiation is a complex, multistep process requiring more than a dozen eIFs (19). In this study, we focused on the activity of eIF2B, which is involved in the binding of initiator methionyl-(met)tRNA to the 40 S ribosomal subunit. We found that eIF2B activity was decreased when amino acid concentrations fell and returned to normal when amino acids were replaced. This response contrasts with the situation in which amino acids or insulin increases, where stimulation of muscle protein synthesis is not accompanied by changes in eIF2B activity (1). Thus, whereas binding of the initiator met-tRNA to the 40 S ribosomal subunit is essential for translation of mRNA, under normal physiological conditions this step is not likely rate limiting for the initiation of muscle protein synthesis (12). However, when amino acid availability is insufficient, this step is limited; therefore, other mechanisms that might normally stimulate muscle protein synthesis via the binding of mRNA to the 43 S preinitiation complex would be expected to be ineffective (12). In this regard, the activity of eIF2B seems to function primarily in a permissive role (12).
The immediate drop in eIF2B activity at the onset of dialysis occurred in a time frame that corresponded with the timing of the reduction in net balance and protein synthesis. This suggests that the mechanism responsible for the reduction in synthesis involved eIF2B. On the other hand, eIF2B activity returned to normal only in the 120-min sample during the replacement of amino acids, whereas the calculated rate of muscle protein synthesis increased by 60 min. However, the absence of precise overlap of the timing of responses does not necessarily preclude a mechanistic role for changes in eIF2B activity. We did not directly measure tracer incorporation into protein but rather calculated synthesis from tracer uptake from plasma. Whereas in a steady-state situation the two approaches yield the same values for synthesis (6), it is possible that, in the nonsteady state, as was the case in this study, changes in net amino acid uptake (and thus our calculated rate of synthesis) preceded actual changes in protein synthesis. The general agreement in the pattern of change of responses at least suggests the possibility that eIF2B activity played a role in controlling the changes in the rate of synthesis.
In summary, a reduction in plasma amino acid concentrations caused by hemodialysis reduced the rate of muscle protein synthesis without affecting the rate of muscle protein breakdown. The signal for the reduction in synthesis appeared to be the change in plasma concentrations, as intramuscular free concentrations remained essentially constant, whereas changes in synthesis corresponded to changes in plasma concentrations. Changes in synthesis generally corresponded to changes in the activity of eIF2B, suggesting a possible role of eIF2B activity in controlling the response of synthesis to the reduction in plasma amino acid concentrations.
We thank Lillian Traber, Dr. Donald J. Deyo, Donald Harper, and Dr. Kazunori Murakami for help in animal preparation, Elizabeth A. Buckner, Cynthia A. Schaeper, Angel D. Male, and James M. Ramirez for help in performing dialysis, and Julie M. Vargas (GC-MS) and Zhanpin Wu (HPLC) for analytical assistance.
This work was supported by National Institutes of Health grants AR-049038 (R. R. Wolfe) and DK-15658 (L. S. Jefferson), and Shriners Hospital for Children Grant 8490. T. G. Anthony was supported by an American Diabetes Association Mentor-Based Postdoctoral Fellowship.
Present address of H. Kobayashi: AminoScience Laboratories, Ajinomoto Co., Kawasaki, Japan 210-8681.
Address for reprint requests and other correspondence: R. R. Wolfe, UTMB/Shriners Burns, 815 Market St., Galveston, TX 77550 (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2003 the American Physiological Society