JEZ 796 THE JOURNAL OF EXPERIMENTAL ZOOLOGY 278:273–282 (1997) Protein Synthesis Under Conditions of Anoxia and Changing Workload in Ventricle Strips From Turtle Heart J.R. BAILEY* AND W.R. DRIEDZIC Department of Biology, Mount Allison University, Sackville, Canada, E0A 3C0 ABSTRACT An earlier study determined that protein synthesis in isolated perfused turtle (Trachemys [=Pseudemys] scripta elegans) hearts was three-fold lower under conditions of anoxia than under conditions of normoxia. However, the earlier study did not attempt to define the role of work in the isolated perfused preparation. In this study, the effects of varying workload, as defined by changing frequency of contraction, and anoxia on protein synthesis were examined. The ventricle strip preparation allows for comparison of multiple strips from a single heart, which aids in eliminating the variability found between individuals chosen from wild populations. Ventricle strips forced to contract at 24 contractions·min–1 under anoxic conditions failed more rapidly than strips forced to contract at 24 contrations·min–1 under normoxic conditions. Protein synthesis decreased by 32% when compared to normoxic controls. When stimulation was terminated after 2 hr of contraction, the rate of protein synthesis in strips under anoxic conditions was similar to that in strips under normoxic conditions. Also, returning strips to normoxic conditions after 2 hr of anoxia restored protein synthesis to the level of the normoxic controls. A significant correlation between pacing rate and protein synthesis was found under normoxic conditions but not under anoxic conditions when strips were paced at 12, 18, and 24 contractions·min–1. Protein synthesis increased by 30% at the 18 contractions·min–1 frequency and 45% at the 24 contractions·min–1 frequency over the rate at 12 contractions·min–1 frequency. Force-frequency studies revealed that under normoxic conditions force generation did not change until above 24 contractions·min–1, but under anoxic conditions there was a significant negative inotropic effect (20% decrease in force) at 24 contractions·min–1 and fell to 50% of initial at 36 contractions·min–1. These studies indicate that, in the turtle heart, anoxia per se is not the only determinant of protein synthesis but rather that work plays an important role in protein synthesis, as in the mammalian heart. J. Exp. Zool. 278:273–282, 1997. © 1997 Wiley-Liss, Inc. Protein synthesis has been well studied in the mammalian heart and is influenced by a variety of factors including oxygen availability and workload (Sugden and Fuller, ’91) such that under normoxic conditions workload causes an increase in protein synthesis. When mammalian hearts are exposed to severe hypoxic or anoxic conditions, decreases in both protein synthesis and degradation are observed (Jefferson et al., ’71; Fuller and Sugden, ’88). However, a problem encountered in experiments with mammals is that these hearts rapidly fail, and therefore, it is difficult to separate the effects of anoxia per se on metabolism from overall contractile failure effects. It is possible to alter cardiac metabolism chemically with blockers such as CO to induce chemical anoxia or to inhibit contractile events by the action of excess K+. These manipulations give data that, while useful, may not actually reflect metabolic events © 1997 WILEY-LISS, INC. in intact heart tissue. Studies of protein turnover in hearts of non-mammalian vertebrates are limited but are considered to be of value in elucidating fundamental principles of control. For example, in trout hearts, protein synthesis increases with power output (Houlihan et al., ’88), and in cod heart, RNA efficiency increases at higher temperatures (Foster et al., ’92). In previous work (Bailey and Driedzic, ’95, ’96), we have shown that the turtle heart can be used as a model system to study protein metabolism under anoxic conditions. The turtle heart is very resistant to anoxia, and isolated heart preparations continue to contract under total anoxia for *Correspondence to: Dr. John Bailey, Department of Biology, Mount Allison University, Sackville, NB, Canada EOA 3C0. Received 7 November 1996; Revision accepted 27 January 1997 274 J.R. BAILEY AND W.R. DRIEDZIC periods of up to 4 hr (Wasser et al., ’90). We found that protein synthesis is not affected by 1 hr of anoxia in the isolated perfused heart model (Bailey and Driedzic, ’95) but that 2 or 3 hr of anoxia produce a significant decrease in synthetic rates without a concomitant decrease in contractility (Bailey and Driedzic, ’96). Also, there is an apparent selective decrease in protein synthesis, with synthesis of proteins destined for the mitochondria being preferentially decreased (Bailey and Driedzic, ’96). This perfused heart model was probably working at near maximal workloads since input (atrial) pressure was very high compared to the in vivo condition and atrial pressure has been shown to be a determinant of cardiac work (Farrell et al.,’94; Jackson et al., ’95). However, we did not attempt to define the influence of work on protein synthesis in the turtle heart model system. In this study, we utilized a ventricle strip preparation which allows for multiple strips from the same heart, and thus, comparison of tissue responses from the same animal. This in turn allows for a greater insight into cellular events unaffected by individual differences, which are always encountered when using samples from wild populations. We report that there is a positive correlation between work and protein synthesis under normoxic but not under anoxic conditions. A short recovery period under normoxic conditions restores protein synthesis, and removing the work component under anoxic conditions results in a synthesis rate unchanged from the normoxic state. MATERIALS AND METHODS Animals Adult red-eared sliders (Trachemys [=Pseudemys] scripta elegans) of either sex and mean body mass of 683 ± 42 g were obtained from Carolina Biological Supply (Burlington, NC). Animals were kept in flow-through aquaria at ambient temperature (approximately 18°C) until use. They were fed a mixed diet of greens and pieces of fish ad lib. Aquaria were provided with basking platforms, allowing the animals to dry off periodically. Photoperiod was 12 hr light:12 hr dark. All experiments were carried out in spring and early summer seasons.Animals used in this study were obtained at three separate times and probably came from different wild populations. Media used Perfusion media consisted of a basic salt solution: 86.3 mM NaCl, 3.1 mM KCl, 2.3 mM CaCl2, 3.5 mM MgSO4 , 1.5 mM NaH2 PO4 , 43.5 mM NaHCO3, and 10 mM glucose, pH 7.8 (basic medium) at 20°C. Basic medium pH was adjusted to 7.8 after 30 min gassing with either 5% CO2: 95% O2 or 5% CO2:95% N 2. The basic medium was enriched by the addition of amino acids (Sigma #6407, physiologic acidic and neutrals; Sigma #6282, physiologic basics; Sigma Chemical Co., St. Louis, MO) to a final concentration of 0.25 mM, with the exception of cystine, which was 0.125 mM; proline, valine, and leucine, which were 0.5 mM; alanine and glycine, which were 1.0 mM; and phenylalanine, which was 1.25 mM (amino acid medium). In all cases, the experimental temperature was 20°C. Experimental protocol Animals were arbitrarily chosen from the holding tank and pithed and the plastron quickly removed. The heart was excised and placed in a beaker of ice-cold medium. The atria were dissected free, the arteries exiting the ventricle removed, and the ventricle bisected. In the first series of experiments, three strips were cut from each section of ventricle for a total of six strips, and in the remaining series of experiments, two strips were cut from each section of ventricle for a total of four strips. In all cases, each strip was visually inspected and any obvious connective tissue removed. Each strip was then mounted in a tissue bath between two platinum wire electrodes and attached to an isometric force transducer (Harvard Apparatus, Dover, MA), which was interfaced with a Biotronix BL 882 strip chart recorder. Strips were stimulated to contract with a Grass (Quincy, MA) model 5D square wave generator at stimulus strength of 50 V and 200 msec duration. Strips were stretched until maximal peak height was observed. Each strip was allowed to equilibrate under normoxic conditions in basic medium for 15 min before the onset of the experimental period. Series 1: force-frequency study Four strips were set up and allowed to equilibrate in basic medium for 15 min at 12 contractions·min–1 under normoxic conditions. Strips were then allowed to equilibrate for a further 18 min at 12 contractions·min–1 under either normoxic or anoxic conditions. Anoxic conditions were achieved by changing the gassing mixture from 5% CO2:95% O2 to 5% CO2:95% N2. The PO2 of the bathing medium was driven to less than 3 mm Hg after 5–10 min of vigorous gassing under PROTEIN SYNTHESIS IN ANOXIC VENTRICLE STRIPS these conditions. High-speed recordings were taken for the last 30 sec of this period for contractiontime measurements. Frequency was increased in steps of 6 contractions·min–1 at 3 min intervals up to a maximum of 36 contractions·min–1. During these increases in frequency, recordings were taken as before for contraction-time measurements. At the end of this period, strips were discarded. Series 2: performance and protein synthesis at set pacing rates Anoxia. The effects of a set pacing rate and anoxia on strips from the same heart were determined. Following the end of the equilibration period in basic medium, two strips were subjected to anoxic conditions and two strips to normoxic conditions. At T = 0 the medium was changed to the amino acid medium and the strips were forced to contract under either normoxic or anoxic conditions at a frequency of 24 contractions·min–1, with high-speed recordings being taken every 15 min. The initial reference point was taken to be T + 15 min to allow for readjustment to the mechanical disturbance following medium change. Previous work has shown that protein synthesis is significantly decreased under anoxic conditions only after 2 hr of anoxia in the isolated perfused heart preparation (Bailey and Driedzic, ’96); therefore, it was deemed necessary to have an equivalent equilibration time in the ventricle strip preparations. At T + 120 min, 20 µl of 1.0 mCi· min –1 L-[ring 2,6-3H]-phenylalanine (Amersham Canada, Oakville, ON) were added to each bath. A 100 µl sample of the medium was immediately taken and added to 10 ml of scintillation fluid (CytoScint; ICN Pharmaceuticals Canada, Montreal, PQ) for the determination of specific activity. Strips were then forced to contract for a further 60 min at 24 contractions·min–1. At the end of the experimental period, strips were taken down, rinsed in cold basic medium, blotted dry, and immediately processed for snalysis of phenylalanine uptake and protein and RNA content. Rest. A second set of experiments was carried out to separate the effects of work and anoxia on protein synthesis. Two strips were subjected to normoxic and two strips to anoxic conditions in amino acid medium for 2 hr at a frequency of 24 contractions·min–1. At T + 120 min the electrical stimulus was discontinued; contractions ceased and the ventricle strips entered a resting state. At T + 175 min electrical stimulation was restarted, frequency as before, and contractions recorded for 5 min to determine the viability of the 275 strips. The experiment was then terminated and the strips taken down. Recordings were taken, isotope was added, and the tissue treated as described in the anoxia experiment. Recovery. A third set of experiments was used to determine if the effects of anoxia on protein synthesis were due to acute regulation during this hour or due to a general deterioration of the protein synthetic mechanism. As before, two strips were forced to contract under anoxic conditions and two strips under normoxic conditions in amino acid medium for 2 hr at a frequency of 24 contractions·min–1. The gassing mixture for the anoxic medium was then changed to the normoxic gassing mixture (5% CO2:95% O2) at T + 120 min for the remainder of the experimental period. Strips were then forced to contract for a further 60 min at 24 contractions·min–1. At the end of this period, strips were taken down. As before, recordings were taken, isotope added, and tissue treated as described in the anoxia experiment. Series 3: protein synthesis with increasing workload (normoxia vs. anoxia) Three strips were set up and allowed to equilibrate for 15 min in basic medium. Each strip was paced at a different frequency, i.e., one strip at 12, one strip at 18, and one strip at 24 contractions·min–1, and forced to contract for 2 hr under either normoxic or anoxic conditions. The basic medium was replaced with amino acid medium at T = 0. In all cases, the initial reference point was taken to be T + 15 following the change in the medium to allow the strips to readjust to the mechanical disturbance. At T + 120 min radiolabeled phenylalanine was added to each bath. A 100 µl sample of the medium was immediately taken and added to 10 ml of scintillation fluid for the determination of specific activity. Strips were forced to contract for a further 1 hr. At the end of the experimental period, strips were taken down, rinsed in cold basic medium, blotted dry, and immediately processed for analysis of phenylalanine uptake and protein and RNA content. Whole tissue analysis All extraction procedures are based on the method of Fuller and Sugden (’88). Briefly, each tissue was homogenized in nine volumes of 0.56 M perchloric acid. Crude homogenates were centrifuged at 10,000g for 15 min. The supernatant was discarded, while the pellet was washed three times with 2 ml of 0.56 M perchloric acid. The pellet was then digested in 2.25 ml of 0.3 M NaOH 276 J.R. BAILEY AND W.R. DRIEDZIC at 37°C for 1 hr. Digestion was regarded as complete when there was no visual evidence of any solid material. Phenylalanine incorporation One hundred microliters of 100% (w/v, 1 g/ml) trichloroacetic acid (TCA) were added to a 1 ml aliquot of the digestate to precipitate proteins. This mixture was centrifuged at 10,000g for 15 min and the supernatant discarded. The protein pellet was washed with 1 ml diethyl ether and centrifuged and the supernatant discarded. The pellet was then dried in a 37°C water bath and digested in 200 µl of 1 M NaOH in a boiling water bath for approximately 30 min, then added to 10 ml of scintillation fluid and counted in a RackBeta scintillation counter. RNA determination Two hundred and fifty microliters of 2.5 M perchloric acid were added to a 1 ml aliquot of the digestate and the mixture centrifuged. The supernatant was reserved for RNA analysis using the orcinol method (Munro and Fleck, ’66). RNA from Torula yeast (Sigma #R6625) was used as a standard. Protein content The remaining 250 µl aliquot of the digestate was analyzed for protein content by a comercial method (Biorad D C Protein Assay #500-0116, Biorad, Richmond, CA). Data analysis Specific activity of the isotope-labeled perfusate was calculated by determining disintegrations per minute and dividing by concentration of phenylalanine in the medium. The amount of phenylalanine incorporated into protein pools was calculated from the extracellular specific activity. This is the standard method of data analysis and is based on the findings that extracellular phenylalanine rapidly equilibrates with the intracellular pool, which in turn equilibrates with phenylalanine tRNA (McKee et al., ’78; Sugden and Fuller, ’91). The rate of phenylalanine incorporation into total protein is expressed as a percentage of normoxic controls. Contractility was calculated as a percentage of the initial (T + 15 min) post-equilibration period peak height from the strip chart records. Time to peak tension and time to 50% relaxation were calculated from the strip chart records. Relative isometric work was calculated as a function of contraction frequency and relative force at each frequency from the force-frequency curve. Means between normoxic and anoxic conditions were compared using the Mann-Whitney U test. In all cases, P < 0.05 was considered significant. RESULTS Force-frequency study A force-frequency study was conducted in order to have a better understanding of the relative workloads at different frequencies. Although it is possible to calculate an estimate of the force exerted by the strips, force exerted is expressed in relative terms since the composition of the turtle ventricle is not uniform and the amount of spongy vs. compact tissue may vary from section to section (Brady and Dubkin, ’64). Strips subjected to normoxic conditions maintained contractility at approximately 90% of the initial level over the 18 and 24 contractions·min–1 frequencies, but force fell off rapidly above 24 contractions·min–1 (Fig. 1A). However, anoxic strips tended to lose contractility after 18 contractions·min–1 and fell off more rapidly than did the strips under normoxic conditions (Fig. 1A). There was a significant negative correlation between time to peak tension and increasing frequency under both normoxic and anoxic conditions (Fig. 1B). Also, there was a significant difference in time to peak tension between normoxic and anoxic conditions at the 12 and 18 contractions·min–1 frequencies but not the remaining frequencies (Fig. 1B). In this particular experiment, normoxic and anoxic strips were from different hearts. This may account for the observed difference in time to peak tension as these preparations may have developed different absolute levels of force. Time to 50% relaxation also displayed a significant negative correlation with increasing frequency in both conditions, but there was no difference between anoxic and normoxic conditions at any frequency (Fig. 1C). Protein synthesis and RNA and protein content In series 2, three independent experiments were conducted involving a variety of experimental perturbations with paired preparations. In no case was there a significant difference in protein content between the control strips run under normoxic conditions, mean value 88.2 ± 4.3 mg protein·g–1 tissue (N = 19), and the experimental srips run under anoxic conditions, mean value 89.0 ± 4.7 protein mg·g–1 tissue (N = 19). Similarly, there was no difference in RNA content between PROTEIN SYNTHESIS IN ANOXIC VENTRICLE STRIPS 277 control strips run under normoxic conditions, mean value 23.6 ± 1.5 µg RNA·mg–1 protein and experimental strips run under anoxic conditions, mean value 22.7 ± 2.0 µg RNA·mg–1 protein. In series 3 experiments (N = 11), there was no significant difference in protein content of strips subjected to normoxic (86.8 ± 2.0 mg·g–1 tissue) or anoxic (86.0 ± 2.0 mg·g–1 tissue) conditions. Also, RNA values did not differ significantly, mean value 24.1 ± 1.3 µg·mg–1 protein for normoxic conditions and 24.6 ± 1.8 µg·mg–1 protein for anoxic conditions. Protein synthesis rates were quite variable between experimental groups ranging from 0.37 ± 0.04 to 2.1 ± 0.2 nmol phenylalanine incorporation·mg–1 protein·hr for normoxic conditions and 0.25 ± 0.02 to 1.38 ± 0.2 nmol phenylalanine incorporation·mg–1 protein·hr for anoxic conditions. However, values were extremely consistent within experimental groups. Consequently, rates of protein synthesis are expressed on a relative basis in terms of the normoxic control for each experiment in series 2. In experiments involving increasing frequencies (series 3), protein synthesis is expressed relative to the lowest frequency, 12 contractions·min–1. Performance and protein synthesis at set pacing rates Under anoxic conditions the force of contraction exerted by the strips decreased by 50% over the 3 hr time period (Fig. 2A). Under anoxic conditions protein synthesis during the last hour was depressed by 30% when compared to normoxic conditions (Fig. 3, left bar). In an effort to assess if the decrease during the last hour was due to reduced work or the lack of oxygen per se, electrical stimulation was discontinued during the last hour. If the strips were not stimulated to contract for the third hour of either normoxia or anoxia, contractility was restored to the anoxic strips (Fig. 2B). When the strips were unpaced and, consequently, did zero isometric work, protein synthesis was unchanged from the normoxic state (Fig. 3, middle bar). Also, once a normoxic environment was restored, the strips regained contractility (Fig. 2C) and protein synthesis remained the same as under 3 hr of normoxic (Fig. 3, right bar). There were no significant differences in either time to peak tension or time to 50% relaxation when strips under experimental conditions were compared to paired controls (data not shown). Fig. 1. Force-frequency study of turtle ventricle strips. All values are means ± SEM, N = 12. Circles represent normoxic conditions and squares represent anoxic conditions. *P < 0.05 between normoxic and anoxic conditions. A: Force genera- tion with increasing frequency. Force is calculated as a percentage of the initial force. B: Time to peak tension with increasing frequency. C: Time to 50% relaxation with increasing frequency. 278 J.R. BAILEY AND W.R. DRIEDZIC Fig. 3. Protein synthesis in turtle ventricle strips under various conditions. All values are means ± SEM. N = 6 for anoxia at 24 bpm, N = 6 for anoxia at 0 bmp, and N = 7 for normoxia at 24 bpm. The three conditions relate to Figure 1A, 1B, and 1C, respectively. Anoxia at 24 bpm refers to strips that were forced to contract for 3 hr under anoxic conditions. Anoxia at 0 bpm refers to strips that were forced to contract for 2 hr under anoxic conditions and then were unpaced for a further 1 hr of anoxia. Normoxia at 24 bpm refers to strips that were forced to contract under anoxic conditions for 2 hr followed by 1 hr of normoxia. Protein synthesis rate is normalized as a percentage of paired normoxic controls. Protein synthesis at increased workload (normoxia vs. anoxia) Fig. 2. Force generation by turtle ventricle strips under conditions of normoxia and anoxia. 3H-Phenylalanine was added at times indicated by arrows. All values are means ± SEM. A: Strips paced at 24 contractions·min–1 for 3 hr. Circles represent normoxic conditions and squares represent anoxic conditions. N = 6 for each data point. B: Strips paced for 2 hr followed by 1 hr of rest under normoxic or anoxic conditions. Circles represent normoxic conditions and squares represent anoxic conditions. N = 6 for each data point. C: Strips paced at 24 contractions·min–1 for 3 hr. Circles represent strips paced under normoxic conditions and squares represent strips paced under anoxic conditions for 2 hr followed by 1 hr of normoxic conditions. N = 7 for each data point. The series 2 experiments suggest that oxygen availability alone is not the only determinant of protein synthesis. In order to further investigate the relationship between isometric work, anoxia, and protein synthesis, the effect of changing frequency was examined. Under normoxic conditions increasing frequency of contraction resulted in a slight decrease, approximately 10%, in the ability of the strips to maintain contractility when paced at 24 contractions·min–1 but contractility remained relatively unaffected at 12 and 18 contractions·min–1 (Fig. 4A). Under anoxic conditions the strips paced at 12 contractions·min–1 were able to maintain contractibility, but the strips paced at 18 and 24 contractions·min–1 showed an approximate 50% decrease in force after 3 hr (Fig. 4B). There was no significant change in either time to peak tension or time to 50% relaxation over the time period under either normoxic or anoxic conditions at any pace rate (data not shown). PROTEIN SYNTHESIS IN ANOXIC VENTRICLE STRIPS 279 tions·min–1 (Fig. 5). Under anoxic conditions, however, the relationship between frequency and protein synthesis was different in that there was no change in protein synthesis as a function of contraction frequency (Fig. 5). DISCUSSION Performance Force-frequency study Above 24 contractions·min–1 turtle heart ventricle strips display a significant negative inotropic isometric twitch response with increasing frequency. This response is not unique to turtle hearts as both fish and amphibian hearts show a similar response (Driedzic and Gesser, ’85, ’88; Morad and Cleeman, ’87). The reason for this decrease in twitch force remains unclear but is probably related to the poorly developed sacroplasmic reticulum in these hearts (Santer, ’85; Driedzic and Gesser, ’88). Consequently, for contraction, calcium must diffuse across the plasma membrane and then be pumped out against a concentration gradient before relaxation can occur. It is known that increasing calcium concentrations can pro- Fig. 4. Force generation of turtle ventricle strips under normoxic and anoxic conditions paced at different frequencies. 3H-Phenylalanine was added at times indicated by arrows. All values are means ± SEM. Force generated is expressed as a percentage of initial value. Squares represent 12 contractions·min–1, circles represent 18 contractions·min–1, and triangles represent 24 contractions·min–1. A: Normoxic conditions. N = 5 for each data point. B: Anoxic conditions. N = 6 for each data point. Both the 18 and 24 contractions·min–1 frequencies show a significant negative correlation with time. Under normoxic conditions protein synthesis increased with the increase in frequency. There was a significant (P < 0.05) linear correlation such that the ratio of protein synthesis at 24 contractions·min–1 was 45% higher than at 12 contrac- Fig. 5. Protein synthesis in turtle ventricle strips paced at different frequencies under normoxic and anoxic conditions. All values are means ± SEM. Circles represent normoxic conditions (N = 5) and squares represent anoxic conditions (N = 6). Protein synthesis is normalized as a percentage of the synthetic rate at 12 contractions·min–1. Synthesis under normoxic conditions shows a significant positive correlation with increasing frequency. 280 J.R. BAILEY AND W.R. DRIEDZIC tect against this loss of contractility at elevated frequencies (Driedzic and Gesser, ’85), which indicates that the limiting step is the inward diffusion of calcium. It is quite likely that a similar situation exists in the turtle heart since the response to increasing frequency is similar. The important finding here is that the force-frequency curve is steeper under anoxic conditions than under normoxic conditions. The reason for this is unclear, but it is possible that the inward Ca2+ flux is compromised under anoxia as a result of a depencence on anaerobic metabolism and a consequent fall in intracellular pH. Long-term anoxia In both normoxic and anoxic preparations, there is a negative inotropic effect associated with increasing frequency of contraction and length of the experimental period, but this effect is more pronounced under anoxic conditions. Strips were able to regain contractility and force generation very quickly once normoxic conditions had been restored, indicating that the 3 hr of anoxia did not result in a permanent loss of functionality. In addition, ventricle strips were able to regain contractility after 3 hr of anoxia when the work component was removed. Under anoxic conditions both the 18 and 24 contractions·min–1 frequencies resulted in an approximate 50% decrease in the ability of the strips to develop force over the experimental period (Figs. 2, 4). The energy for contraction under anoxic conditions is supplied by anaerobic glucose metabolism with a concomitant buildup of lactic acid (Reeves, ’63) and a lowering of intracellular pH (Wasser et al., ’90). These factors in turn could influence the ability of the strips to maintain contractility, although in isolated, perfused heart preparations anoxia alone is not sufficient to produce a decrease in cardiac output (Jackson et al., ’95; Bailey and Driedzic, ’96). This preparation’s specific response may relate to a better washout of H+ ions in perfused hearts than in ventricle strips. Anoxia and set pacing rates In the initial experiment, in which strips were stimulated at a frequency of 24 contractions·min–1, the rate of protein synthesis decreased by about 30% in association with a decrease in force development. However, protein synthesis did not decrease when the work component was removed, even though the strips were still in an anoxic environment. These data show that anoxia per se is not the sole determinant in the decreased rate of protein synthesis since the ventricle strips can main- tain protein synthesis under anoxic conditions when the work component is removed. In contrast, rat hearts are unable to maintain protein synthesis under anoxic conditions when the work component is removed by arresting these hearts with tetrodotoxin (Jefferson et al., ’71). When normoxic conditions were restored to the turtle ventricle strip preparation, both contractility and protein synthetic rates were restored to control levels. This indicates that the decrease in protein synthesis was due to an acute regulatory mechanism and not to a deterioration of the protein synthetic machinery during the period of anoxia. Unlike the isolated perfused heart preparation (Bailey and Driedzic, ’96), there was no drop in cellular RNA levels, so the decrease in protein synthesis cannot be attributed to an impairment at that gross a level. It is known that a fall in intracellular pH may adversely affect protein synthesis in mammalian hearts (Fuller et al., ’89; Sugden and Fuller, ’91) and that anoxia does result in a decrease in intracellular pH in working turtle hearts (Wasser et al., ’90; Jackson et al., ’95). We suggest that intracellular pH may be the control factor in protein synthesis in these experiments. It has long been known that the turtle heart can cycle large quantities of glucose through glycolysis to maintain energy production (Reeves, ’63) and that ATP levels remain relatively constant over at least 4 hr of anoxia (Wasser et al., ’90). Therefore, we do not view ATP itself as an important intracellular regulator under these conditions. If there is no mechanical work being done, the degree of acid production is probably reduced since energetic requirements would be significantly reduced when the strips are in a noncontractile state. The restoration of normoxic conditions could result in a rapid resetting of intracellular pH and, thus, a restoration of protein synthesis. These experiments indicate that there is a complex interaction between the variables of oxygen availability and mechanical work on protein synthesis in these preparations. Anoxia and increasing workload The rate of protein synthesis, as measured by phenylalanine incorporation, displays a positive correlation with frequency of contraction in normoxic strips. We are not able to quantitate work since there is not a common set point for normalization for all strips, but we can provide a qualitative estimate of work by taking into account the differences in frequency and the impact of contraction frequency alone on force development. PROTEIN SYNTHESIS IN ANOXIC VENTRICLE STRIPS These data are presented in Figure 6A and B. The 12 contractions·min–1 frequency is set to 100%, the 18 contractions·min–1 frequency is multiplied by 1.5, and the 24 contractions·min–1 frequency is multiplied by 2. Both of the latter are then adjusted for the decrease in force observed in the force-frequency study. This analysis reveals that the strips paced at the higher frequencies do more work than those paced at the lowest frequency (Fig. 6A). Therefore, we are confident in stating that under normoxic conditions protein synthesis 281 increases in concert with an increase in work. This coupling of work and protein synthesis also has been reported for rat heart (Jefferson et al., ’71) and for fish heart (Houlihan et al., ’88) and may be a common principle applicable to all vertebrates. Under anoxic conditions an increase in contraction frequency did not result in an increase in protein synthesis. The qualitative analysis of work suggests that strips paced at 18 and 24 contractions·min–1 under anoxic conditions do less work than the strips paced at 12 contractions·min–1 (Fig. 6B). If protein synthesis and work were coupled under anoxic conditions, there should be a reduction in protein synthesis under anoxia, but there was no change from the protein synthesis rate observed at the 12 contractions·min–1 frequency. These data indicate that work and protein synthesis become uncoupled under anoxia, and thus, control mechanisms may differ under the two conditions. In summary, in ventricle strips prepared from turtle hearts, work and protein synthesis are coupled under normoxic conditions. Under anoxic conditions protein synthesis is depressed and, moreover, protein synthesis and work become uncoupled. It is known that intracellular pH is depressed under anoxic conditions in turtle heart, and this may mediate the uncoupling of work and protein synthesis. At present, work to elucidate the effects of acidotic conditions on protein synthesis in turtle heart is being carried out in this laboratory. ACKNOWLEDGMENTS This study was supported by the Natural Sciences and Engineering Research Council of Canada and the New Brunswick Heart and Stroke Foundation. LITERATURE CITED Fig. 6. Theoretical relative isometric work of turtle ventricle strips under normoxic and anoxic conditions. Values are calculated from data in Figures 2 and 4. Squares refer to 12 contractions·min–1 , circles refer to 18 contractions·min–1, and triangles refer to 24 contractions·min–1. A: Normoxic conditions. B: Anoxic conditions. Bailey, J.R., and W.R. Driedzic (1995) Short-term anoxia does not impair protein turnover in isolated perfused turtle heart. J. Comp. Physiol. [B], 164:622–628. Bailey, J.R., and W.R. Driedzic (1996) Decreased total ventricular and mitochondrial protein synthesis during extended anoxia in turtle heart. Am. J. Physiol. 271:R1660–R1667 Brady, A.J., and C. Dubkin (1964) Coronary circulation in the turtle ventricle. Comp. Biochem. Physiol., 13:119–128. Driedzic, W.R., and H. Gesser (1985) Ca2+ protection from the negative inotropic effect of contraction frequency on teleost hearts. J. Comp. Physiol., 156:135–142. Driedzic, W.R., and H. Gesser (1988) Differences in force-frequency relationships and calcium dependency between elasmobranch and teleost hearts. J. Exp. Biol., 140:227–241. Farrell, A.P., C.E. Franklin, P.G. Arthur, H. Thorarensen, and K.L. Cousins (1994) Mechanical performance of an in situ 282 J.R. BAILEY AND W.R. DRIEDZIC perfused heart from the turtle Chrysemys scripta during normoxia and anoxia at 5°C and 15°C. J. Exp. Biol., 191:207–229. Foster, A.R., D.F. Houlihan, S.J. Hall, and L.J. Burren (1992) The effects of temperature acclimation on protein synthesis rates and nucleic acid content of juvenile cod (Gadus morhua L) Can. J. Zool., 70:2095–2102. Fuller, S.J., and P.H. Sugden (1988) Acute inhibition of rat heart protein synthesis in vitro during β-adrenergic stimulation or hypoxia. Am. J. Physiol., 255:E537–547. Fuller, S.J., C.J. Gaitanaki, and P.H. Sugden (1989) Effects of increasing extracellular pH on protein synthesis and protein degradation in the perfused working rat heart. Biochem. J., 259:173–179. Houlihan, D.F., C. Agnisola, A.R. Lyndon, C. Gray, and N.M. Hamilton (1988) Protein synthesis in a fish heart: Responses to increased power output. J. Exp. Biol., 137:565–587. Jackson, D.C., H. Shi, J.H. Singer, P.H. Hamm, and R.G. Lawler (1995) Effects of input pressure on an in vitro turtle heart during anoxia and acidosis: A 31P-NMR study. Am. J. Physiol., 268:R683–689. Jefferson, L.S., E.B. Wolpert, K.E. Giger, and H.E. Morgan (1971) Regulation of protein synthesis in heart muscle: III. Effect of anoxia on protein synthesis. J. Biol. Chem., 246:2171–2178. McKee, E.E., J.Y. Cheung, D.E. Rannels, and H.E. Morgan (1978) Measurement of the rate of portein synthesis and compartmentation of heart phenylalanine. J. Biol. Chem., 253:1030–1040. Morad, M., and L. Cleeman (1987) Role of Ca2+ channel in development of tension in heart muscle. J. Mol. Cell. Cardiol., 19:527–533. Munro, H.N., and A. Fleck (1966) The determination of nucleic acids. In: Methods of Biochemical Analysis, vol. 14. D. Glick, ed. Wiley, New York, pp. 113–176. Reeves, R.B. (1963) Energy cost of work in aerobic and anaerobic turtle heart muscle. Am. J. Physiol., 205:17–22. Santer, R.M. (1985) Morphology and innervation of the fish heart. Adv. Anat. Embryol. Cell Biol., 89:1–102. Sugden, P.H., and S.J. Fuller (1991) Regulation of protein turnover in skeletal and cardiac muscle. Biochem. J., 273:21–37. Wasser, J.S., K.C. Inman, E.A. Arendt, R.G. Lawler, and D.C. Jackson (1990) 31 P-NMR measurements of pHi and highenergy phophates in isolated turtle hearts during anoxia and acidosis. Am. J. Physiol., 259:R521–R530.