Atlantic cod (Gadus morhua) were held either at seasonal ambient temperatures (−0.3 to 11°C) or at a relatively constant control temperature (8–11°C) to investigate aspects of protein synthesis during a period of compensatory growth. Protein synthesis rate, total RNA, and RNA-specific protein synthesis rate were determined in white muscle and liver when ambient temperatures were −0.3, 4.5, and 11°C in February, June, and July, respectively. To allow for comparisons between treatment temperatures, fish were also acutely transferred to a comparable assay temperature in February and June. Over the transition from 4.5 to 11°C (June to July), the ambient-held cod had a significant increase in size and a substantially higher growth rate relative to control-held fish over the same period, consistent with cold-induced compensatory growth. During the onset of this enhanced growth, in June when ambient temperature was ∼4.5°C, ambient-held fish elevated their capacity for protein synthesis in the white muscle and liver via elevation of the RNA content. When ambient temperature reached the same point as for the control fish (11°C), the rate of white muscle protein synthesis remained higher in the ambient-held vs. that in the control-held fish, a process facilitated by elevated RNA content and greater RNA-specific rate of protein synthesis. In the liver, all measured characteristics of protein synthesis were the same for ambient and control fish in July. The latter suggests that compensatory growth may be in part explained by improved efficiency of protein synthesis.
- efficiency of protein synthesis
- cold adaptation
many ectothermic organisms display enhanced growth rates after periods of reduced growth due to low temperature or to dietary limitation. This phenomenon, termed compensatory growth, has been documented in numerous fish species and studied most extensively in salmonids [4, 14 (and references therein), 16, 19]. The majority of work on compensatory growth in fish has dealt with behavioral responses and food availability, with the underlying biochemical processes of how increased net protein accretion is required for growth yet to be addressed.
Atlantic cod (Gadus morhua) is a relatively eurythermal fish, having a species temperature range of −0.5 to 20°C with regional differences in thermal ranges and preferences (3, 20). Recent work (18) has demonstrated compensatory growth in response to low temperature in juveniles of this species, making G. morhua an appropriate model for the study of the mechanisms involved in this response to low temperature.
Protein synthesis is an integral component of growth, and, as expected, the in vivo rate in ectotherms is sensitive to acute temperature changes, with the rate of synthesis decreasing with temperature. Within fish species, the capacity for protein synthesis is quite plastic with respect to temperature acclimation. Watt et al. (22) demonstrated that, when tested at similar temperatures, common carp (Cyprinus carpio) acclimated to 8°C have higher rates of muscle protein synthesis than those acclimated to 28°C. Similarly, the liver of the toadfish (Opsanus tau) displays a compensation of protein synthesis 14 days after the transfer from 20–22°C to 10°C (9). With respect to G. morhua, Foster et al. (6) found similar fractional rates of protein synthesis (%/day) in a number of tissues from fish acclimated to 5 or 15°C. The similarity in rates was attributed to an increase in RNA content at 5°C; thus the rate of protein synthesis may be conserved independent of temperature. These results indicate that G. morhua has some capacity for adaptation of protein synthesis (6). However, liver and white muscle synthesis rates were not included in the study by Foster et al. (6). Given the importance of liver in overall protein metabolism and given that white muscle is quantitatively the largest tissue mass in teleosts, making up over 40% of total wet mass in some species (21), these two tissues warrant investigation in the context of compensatory growth. Furthermore, in the experiment by Forster et al. (6), the temperatures used were substantially higher than the low range of tolerance (<0°C), during which adaptation of protein synthesis may be more critical for growth and survival.
The wide thermal tolerance, coupled with previous studies of protein metabolism in this species (5, 6, 11–13), make G. morhua an excellent model for the investigation of what role, if any, protein synthesis may play in the compensatory growth response. In this study, we assessed the rate of protein synthesis in the white muscle and liver using a “flooding” dose of [2,3-3H]phenylalanine in fish held at ambient temperatures (−0.3 to 11°C) and at relatively constant control temperatures (8.3–11°C). To our knowledge, this is the first time that this hypothesis, that protein synthetic capacity is enhanced after low seasonal temperatures experienced during winter, coincident with an expected period of compensatory growth, has been tested.
MATERIALS AND METHODS
Cultured Atlantic cod (G. morhua) hatched in March 2000 were obtained from the Aquaculture Research and Development Facility, Ocean Sciences Centre (St. John's, NF, Canada). In September 2001, fish were transferred to a covered, outdoor 25-m3 flow-through tank and maintained at 8–11°C (control acclimation temperature) for the duration of the experiments (Fig. 1A). In November 2001, a subgroup of fish was removed and transferred to a similar tank that received ambient seawater that decreased in temperature as winter progressed. Fish were fed ad libitum twice a day with commercial pelleted food (Shurgain). Experiments for the validation of the flooding dose method were conducted from October to December 2001. For temperature effects on protein synthesis, experiments took place from February to July 2002 (Fig. 1B).
Validation of Flooding Dose Method
It was anticipated that, to be able to measure protein synthetic rate in the low-protein turnover white muscle at subzero temperatures, a long incubation time would be required. As such, we developed a method for a flooding dose of radiolabeled phenylalanine based on the principles of Garlick et al. (8). We report the incorporation of radioactivity into protein alone, rather than determining the fractional rate of synthesis (%/day). Thus our unit of protein synthesis is nanomoles of phenylalanine incorporated per milligram of protein per hour. This rationale has been utilized in several other studies of protein synthesis (for example, Refs. 1, 7, 16).
Fish (∼150–1,000 g) were removed from the above-described control seawater tank, weighed, tagged, and transferred to indoor 1-m2 tanks flushed with seawater at the same temperature as the outdoor tank (8–11°C). Specimens were selected over the range of sizes for the population in the tank. After ∼18 h, during which time the fish were not fed, each fish was injected without anesthesia via caudal puncture with 1.0 ml/100 g of [2,3-3H]phenylalanine (Amersham International) injection solution. This solution contained 135 mM phenylalanine, including sufficient [2,3-3H]phenylalanine for a dosage of 100 μCi/ml, in a buffered solution of (in mM) 150 NaCl, 5 KCl, 5 NaPO4, 2 CaCl2, 10 NaHCO3, 2.0 Na2HPO4, 1.0 MgSO4, 5 d-glucose, and 5.0 HEPES at pH 7.6 (20°C). Immediately after the injection, specimens were returned to the indoor tanks (2–6 fish/tank). We randomly assigned incubation times to individual fish based on tag identifications.
After the predetermined incubation period (from 0.5 to 9 h), fish were killed by a blow to the head; the tissues were quickly dissected out, blotted dry on paper towels, and frozen in liquid nitrogen. White muscle samples were taken from the deep dorsolateral region between the first and second dorsal fin on the side opposite to the tag, and liver samples were dissected out from the left and ventral most located lobe where the base of this lobe joins the remainder of the liver. Samples were stored at −70°C until analyzed (within 4 mo).
Experimental Protocol for Long-Term Temperature Effects on Protein Synthesis
Figure 1 illustrates the temperature profiles of the two thermal regimes, the sample dates, and fish weight. The rates of protein synthesis for white muscle and liver were determined in both groups when ambient seawater temperature was approximately −0.3°C, 4.5°C, and 11°C. Two or three fish from each experimental group were killed after isotope injection, as described above, at 3, 6, or 9 h postinjection (total of 7–9 fish sampled/experimental group for all incubation time points combined).
To allow for direct comparisons between the thermal regimes, fish were acutely transferred overnight to comparable temperatures when ambient conditions were markedly different from those in the control tank. For the first sample period, from late February to early March, fish were weighed and then put in 1-m2 tanks with running seawater either at the same temperature as their respective long-term acclimation temperature or at an intermediate temperature, average of 4.3°C (range of 4.0–4.6°C). During the second sample period, early June, fish from the ambient group were transferred to the control seawater and vice versa with ambient and control indoor tank temperatures being 4.5°C (range of 4.1–4.7°C) and 8.3°C (range of 7.9–8.7°C), respectively. For the final sample period, mid-July, the ambient and control groups were at the same temperature; thus acute transfer experiments were not conducted.
Tissue Preparation and Scintillation Counting
Samples were homogenized with a Polytron in 9 or 4 vol of ice-cold 6% perchloric acid (PCA) for white muscle and liver, respectively. After a 10-min incubation on ice, homogenates were thoroughly mixed and a 1-ml aliquot was transferred to a 1.5-ml microcentrifuge tube and centrifuged in an Eppendorf 5414 benchtop centrifuge (fixed speed of 15,000 rpm, 15,600 g) for 5 min. Supernatant was removed and frozen at −20°C for analysis of the free-pool phenylalanine. Care was taken to not include any lipid or precipitate from the bottom of the tube in the supernatant.
For white muscle samples, 1 ml of 6% PCA was added to the tube; the precipitated pellet was mechanically resuspended, vortex mixed to “wash” the pellet, allowed to settle for 10 min, and then centrifuged as above. Afterward, the supernatant was discarded. We completed this washing procedure a total of three times, after which 1 ml of 0.3 M NaOH was added and the tube was incubated at 37°C until the protein pellet dissolved (1–4 h). The dissolved protein was frozen at −20°C until analysis.
The very high lipid content of liver samples required additional processing compared with white muscle. After we removed a subsample of supernatant, the remaining supernatant and the liquid lipid phase were removed by aspiration. The solid lipid fraction was then removed from the side of the tube with a cotton swab. The precipitated pellet at the bottom of the centrifuge tube was washed and centrifuged as described above. The supernatant and any remaining liquid lipid phase were removed by aspiration, and again the side of the tube was swabbed clean of solid lipids. This washing procedure was completed three times, removing all visible lipid contamination. The remaining protein pellet was dissolved as described above.
Aliquots of the original PCA extraction supernatants and dissolved protein fractions were mixed with 10 ml of Ecolume scintillation cocktail and counted on a Packard Minaxi tricarb 4000 liquid scintillation counter to determine the [2,3-3H]phenylalanine content of the free and protein-bound phenylalanine pools, respectively.
Free-pool phenylalanine content was measured in the PCA extraction supernatant by a Sigma Diagnostics fluorometric assay kit. Of note, PCA has substantial inhibitory effects on this assay, requiring standards to be made up in an equivalent PCA concentration as samples. The protein content of the NaOH solubilized protein fraction was determined with the Bio-Rad Dc kit method, using bovine serum albumin as a standard.
We extracted total RNA using TRIzol reagent (Invitrogen), a commercially available modification of the single-step RNA isolation method (2). Briefly, ∼100 mg of tissue were homogenized in 1 ml of TRIzol. The samples were incubated at room temperature for 5 min and then centrifuged at 12 000 g for 10 min at 4°C. Chloroform (0.2 ml) was added to the supernatant, and the samples were mixed vigorously by hand for 15 s. The samples were then incubated at room temperature for 3 min and centrifuged at 12,000 g for 15 min at 4°C. The top layer was transferred to a fresh tube, and the RNA was precipitated via addition of 0.25 ml of isopropanol followed by 0.25 ml of high-salt solution (0.8 M sodium citrate, 1.2 M sodium chloride). The samples were incubated at room temperature for 10 min followed by centrifugation at 12,000 g for 10 min at 4°C. The RNA pellet was washed with 1 ml of 75% ethanol and centrifuged at 7,500 g for 5 min at 4°C. The pellet was air dried for ∼10 min, and the RNA was dissolved in 0.1 ml of RNase-free water by pipetting and incubation at 60°C for 10 min. Total RNA concentration (μg/ml) was determined by subtracting the absorbance at 320 nm from the absorbance at 260 nm and then normalized to tissue weight.
Statistics and Analysis
For the validation experiments, mean tissue phenylalanine content and specific activity over incubation time were compared by one-way ANOVA, and incorporation of radioactivity into protein was assessed by linear regression. We made comparisons between the temperature acclimation groups using a generalized linear model with a Bonferroni post hoc correction. When necessary, data were log transformed to satisfy the assumption of equal variance. In all cases, P < 0.05 was considered significant.
Validation of Flooding Dose Method
Data for the validation of the flooding dose technique for white muscle and liver are illustrated in Fig. 2. White muscle phenylalanine content was significantly elevated 2 h postinjection; afterward, levels did not differ (Fig. 2A). There were no significant differences in the specific activity of phenylalanine throughout the time course of the experiment (Fig. 2B). In liver, the phenylalanine content was highly variable (Fig. 2D), possibly because of differing amounts of blood contamination. The specific activity of phenylalanine in liver was relatively constant for the entire time course, with the exception of the 3-h time point, which was lower than the 0.5- and 1.5-h values but similar to the specific activities found at 6 and 9 h (Fig. 2E). There was no significant trend of decreasing specific activity in liver with time. As such, in this experiment, true stabilization of the white muscle and liver phenylalanine pools occurred well after 1 h postinjection but was relatively stable for up to 9 h.
To determine the incorporation of free phenylalanine into the protein pool, we used incubation times ranging from 0.5 to 9 h. In white muscle (Fig. 2C) and liver (Fig. 2F), the incorporation of phenylalanine into protein was significant and linear with time and the intercept was not different from the origin. Thus the method as described fulfills the assumptions of stable phenylalanine-specific radioactivity as well as linearity for the incorporation of the label into the protein pool for both tissues. As shown by the regression analysis, the rate of incorporation was 0.022 and 0.49 nmol of phenylalanine incorporated·mg protein−1·h −1 for white muscle and liver, respectively.
Impact of Long-Term Temperature Effects on Protein Synthesis
Fish from both groups were of similar size during the initial sample in February (Fig. 1B). From February to June, when ambient temperatures rose from ∼0 to 4.5°C, the control fish, held at much higher temperatures, grew significantly larger than the ambient temperature fish, with specific growth rates of 0.23 and −0.03%/day, respectively [calculated as in Purchase and Brown (18)]. As ambient temperature rose from 4.5 to 11°C, the ambient temperature fish displayed a significant increase in size from June and a substantial increase in growth rate (0.59%/day), whereas the control fish had a very low growth rate (0.05%/day) and did not have a significant change in mass. Of note, all fish sampled were juveniles, showing no appreciable development of the gonad in either experimental group.
Rate of protein synthesis.
The capacity for protein synthesis in white muscle was consistently elevated in the ambient temperature fish compared with the control fish (Fig. 3A) when measured at the same assay temperature. In February, when ambient temperature was lowest (−0.3°C) and assays were run at their respective acclimation temperatures, ambient-held fish had substantially lower rates of synthesis in the white muscle compared with the control temperature group (8.3°C). In fish from both groups that had been acutely transferred overnight to the same temperature (4.3°C), protein synthesis in the ambient temperature fish was equivalent to the that in the control group at 8.3°C, whereas rates in the control fish dropped to levels not different from ambient-acclimated fish at 0.3°C.
In June, when temperatures were 4.5 and 8.5°C for the ambient and control groups, respectively, white muscle protein synthesis was similar between groups when assayed at their respective acclimation temperatures (Fig. 3B). Acute transfer to the other temperature resulted in rate changes as seen in February. The only significant difference was between ambient temperature fish transferred to 8.5°C and control temperature fish transferred to 4.5°C. When both acclimation groups were at the same holding and test temperature (July at ∼11°C), the rate of protein synthesis in white muscle of fish exposed to ambient seasonal temperature was more than double (0.053 compared with 0.022 nmol of phenylalanine incorporated·mg protein−1·h−1) that of the control group maintained at high (8–11°C) temperature.
Unlike white muscle, no significant compensation for protein synthesis in liver of the ambient temperature fish occurred in February or July, the extreme seasonal comparison points used (Fig. 3B). In June, the rate of protein synthesis in fish at ambient temperature (4.5°C) was the same as the control temperature group at 8.5°C, suggesting some degree of thermal compensation in liver. The low rate observed in control fish transferred to 4.5°C bordered on significance, P = 0.057, compared with ambient-held fish measured at 4.5°C. The only significant difference for liver protein synthesis in June was between the fish that were acutely transferred to the other long-term treatment temperatures, i.e., fish acclimated at ambient (4.5°C) and tested at 8.5°C compared with control fish acclimated to 8.5°C and tested at 4.5°C.
RNA and protein contents.
In the February and June sample periods, acute transfer to a different water temperature did not significantly affect the protein and RNA content (P > 0.05) for white muscle and liver within each respective acclimation group. Thus, for each sample period, the animals from both test temperatures were pooled by their respective acclimation groups.
Aside from a small, but significant, difference in February, the protein content of white muscle was similar between acclimation groups and sampling periods, with a slight trend of decreasing protein content with time (Table 1). Conversely, the RNA content of white muscle was elevated in the ambient-acclimated fish compared with the control fish at all sample periods (Table 1). Similarly, the RNA-to-protein ratio was also higher in ambient-acclimated fish at all sample periods (Table 1).
The protein content of liver was similar between acclimation groups at all sample periods (Table 1). However, consistent with an enhancement in protein synthetic rates, the RNA content and RNA-to-protein ratio were elevated in the liver of ambient-acclimated fish during the June sample period.
RNA-specific rate of protein synthesis.
In white muscle, acute transfer to higher or lower temperature usually resulted in a similar directional change in RNA-specific rates of protein synthesis. More specifically, in February, RNA-specific rates of protein synthesis in ambient-held fish were higher at 4.3°C than at −0.3°C and lower in control fish at 4.3°C than at 8.3°C. In June, transfer of control fish from 8.5 to 4.5°C resulted in a decrease; however, when assayed at the same temperature in February (4.5°C) or June (4.5 or 8.5°C), there were no differences in RNA-specific rates of protein synthesis between treatment groups.
A major exception was in July, when ambient temperature fish had an RNA-specific rate of protein synthesis that was almost twice as high as that of the control group, with values of 11 and 6 nmol phenylalanine incorporated·ng RNA−1·h−1 for the ambient and control groups, respectively.
In liver, acute temperature change usually had no impact on RNA-specific rates of protein synthesis. The one exception was in June, when the transfer of control fish from 8.5 to 4.5°C resulted in a substantial decrease (Fig. 3D). However, there was no significant difference in liver between groups when assayed at the same temperature in either February, June, or July.
The experimental design utilized in the present study allows for direct comparisons of protein synthesis rates between fish acclimated at distinct temperatures. The acute transfer overnight to comparable temperatures did not result in significant changes in RNA content of tissues, indicating that changes in the rates of synthesis are not the result of short-term and rapid regulation in response to an ∼4°C temperature change. Other potential confounding influences are explored below.
The tanks that fish were held in for temperature acclimation were similar, with seawater coming from the same source and with water for the control treatment being heated when required. Although the tanks were covered, sufficient ambient light entered to allow fish exposure to the natural photoperiod. In addition, both treatment groups were fed twice daily to satiation on the same diet; therefore, it seems unlikely that our results are due to natural cues such as photoperiod, to some unknown factor in the seawater, or to insufficient food availability. However, growth rate was influenced by temperature, with the control group being larger than the ambient temperature fish. It is important to appreciate that temperature will likely have an impact on the level of voluntary food intake. Indeed, this is one anticipated component of the general compensatory growth response. Possible differences in food consumption do not detract from the overall conclusions of this study (see below) that protein synthesis is increased in ambient-held fish after the winter period.
Protein synthesis is negatively scaled with size in fish (10), and in June and July the ambient temperature fish were significantly smaller than the control group. Thus the potential for confounding effects of size on protein synthesis may exist. Within the control group, mean fish mass in June and July had increased by over 100 g compared with the February sampling period, similar to the difference between the control- and ambient-acclimated groups in June and July (Fig. 1B). Yet, when protein synthesis for the control fish was assayed at the respective acclimation temperature in February and June (8.3 and 8.5°C), there was no difference between fish of different sizes. Furthermore, the variance for protein synthesis between fish within a given treatment is generally large and likely to overshadow scaling relationships, especially in white muscle. Given the above, it is unlikely that our results are due to confounding effects of body size.
Finally, it is important to note that stress hormones, namely cortisol and catecholamines, are known to be effectors of protein metabolism in fish (see Ref. 15 for review of cortisol effects). Our experimental fish were injected without anesthetic to be consistent with other protocols for protein synthesis in ectotherms. Circulating levels of the above hormones can be elevated under this procedure. We did not measure cortisol or catecholamines, and the possibility that the observed results have been influenced by elevated circulating stress hormones cannot be dismissed. Furthermore, we cannot eliminate the possibility that levels of stress hormones have been differentially elevated by different temperature treatments. The below results with respect to protein synthesis should be interpreted with some degree of caution with the above-outlined caveats in mind.
Elevated Protein Synthesis and Compensatory Growth
A previous experiment showed that Atlantic cod displayed compensatory growth with respect to temperature (18). In the present study, from February to June, the ambient temperature fish did not display any appreciable growth, whereas the control group significantly increased in size. From June to July, the ambient group not only increased in size but also displayed a growth rate an order of magnitude higher than the control group (0.59 and 0.05%/day, respectively). These results are very consistent with past observations on compensatory growth in fish.
During the same time frame, June to July, the rate of protein synthesis in white muscle was elevated in the ambient-held fish and in July was more than double that in the control temperature group (Fig. 3A). The substantially higher rate of white muscle protein synthesis in the ambient temperature fish during July reveals that compensatory growth is facilitated not only by an increase in the RNA content but also by enhanced efficiency of protein synthesis, represented by the increase in the RNA-specific rate of protein synthesis. This increased efficiency may be a key component to compensatory growth.
The Role of RNA Levels in Cold Adaptation of Protein Synthesis
A substantial increase in the rate of white muscle protein synthesis, when assayed at a common temperature, was observed in ambient temperature-acclimated fish relative to the control group in February when temperatures were below 0°C. This strongly supports the hypothesis of cold adaptation of protein synthesis in Atlantic cod during periods of low temperature. Furthermore, similar to the results of Foster et al. (6) for other tissues, the similarity between the rate of white muscle protein synthesis in ambient-acclimated fish at 4.5°C and control fish assayed at 8.5°C also implies enhanced capacity at lower temperature. For fish at the same test temperature, a consistent trend of higher protein synthesis coincident with higher RNA content was found for the ambient-acclimated vs. the control temperature fish. Also, as in Foster et al. (6), elevated RNA is at least partially involved in the adaptation of protein synthesis to lower temperature within a species.
In February and July, there was no enhanced liver protein synthetic capacity or elevated RNA content in the ambient-acclimated fish. Perhaps the substantially higher constitutive capacity for protein synthesis in the liver, a full order of magnitude higher than in white muscle, masks or negates any need for increased capacity in the ambient group at these sampling periods. Curiously, as with white muscle, in June the rate of protein synthesis in liver was equivalent in fish from the ambient and control groups when assayed at 4.5 and 8.5°C, respectively. Consistent with enhanced liver protein synthesis in June, the ambient temperature fish had higher RNA content and RNA-to-protein ratio than the control fish. Taken as a whole, this supports the notion that liver protein synthetic capacity is elevated in the June-sampled ambient-acclimated fish. If protein synthesis is enhanced in the liver during the period of increasing temperature, it is unclear what role this enhancement plays.
We have found that Atlantic cod can cold adapt protein synthesis to counter the effects of temperature on synthesis rates. This response is tissue specific, with only white muscle displaying an incomplete compensation of protein synthesis at very low temperatures (−0.3°C); at intermediate temperatures (4.5°C), however, both white muscle and liver show complete compensation. Furthermore, the results in July when acclimation temperatures were equivalent suggest that compensatory growth may be in part facilitated by enhanced protein synthesis via increased RNA content and improved efficiency of protein synthesis in white muscle.
This work was funded by AquaNet, a Network of Centres of Excellence (awarded to Drs. J. S. Ballantyne, T. W. Moon, K. V. Ewart, and W. R. Driedzic) and the Natural Sciences and Engineering Research Council of Canada (awarded to W. R. Driedzic).
The authors thank Connie Short for technical assistance and the staff of the Aquaculture Research Development Facility for animal husbandry.
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 © 2005 the American Physiological Society