Fish face a trade-off between ‘eating big’ for growth efficiency and ‘eating small’ to retain aerobic capacity (2024)

Associated Data

Supplementary Materials

Data Availability Statement

Abstract

Feeding provides the necessary energy to fuel all fitness-related processes including activity, growth and reproduction. Nevertheless, prey consumption and digestive processes can have physical and physiological trade-offs with other critical functions, many of which are not clearly understood. Using an ambush predator, barramundi (Lates calcarifer), fed meals ranging 0.6–3.4% of body mass, we examined interrelations between meal size, growth efficiency and surplus aerobic metabolic capacity (aerobic scope, AS). Large meals required a greater absolute investment of energy to process (a larger so-called specific dynamic action, SDA), but the percentage of digestible meal energy required in the SDA response (SDA coefficient) decreased with increasing meal size. Combined with the findings that growth rate and growth efficiency also increased with food intake, our results demonstrate that it is energetically advantageous for fish to select large prey. However, following a large meal, SDA processes occupied up to 77% of the available AS, indicating that other oxygen-demanding activities like swimming may be compromised while large meals are processed. This trade-off between meal size and AS suggests that fishes like barramundi would benefit from regulating prey size based on imminent requirements and threats.

Keywords: aerobic scope, digestion, metabolic rate, meal size, prey selection, specific dynamic action

1. Background

Predators often have the opportunity to select their prey size in an ephemeral energy landscape, yet little is known of the attributes driving this selection. Studies on prey-size selectivity have mainly focused on trade-offs between prey size and predator gape size, or search and handling times (e.g. [1]). There has been some suggestion that post-consumptive processes such as digestion and metabolic rate may act as important regulators of prey selectivity [2,3], but these ideas remain insufficiently explored.

Aerobic scope (AS) is the difference between an animal's standard (resting) and maximal aerobic metabolic rate, and thus represents the capacity to simultaneously supply oxygen/energy to processes like locomotion, growth and reproduction. Specific dynamic action (SDA), the energy used in the digestion, absorption and assimilation of a meal, must also occur within the confines of AS. Although some animals may achieve a slightly higher maximum metabolic rate if exercising maximally during the SDA response, the processes are only partially additive and therefore SDA continues to compete for the available AS (see [4] and references within). While more food leads to faster growth, which is often advantageous because it takes juveniles beyond specific size ranges preferentially targeted by predators, the increased SDA associated with large meals may compromise AS to such an extent that predators select smaller-sized meals. Indeed, the SDA response following consumption of a satiation meal may leave predators vulnerable to predation themselves and also impair social interactions, if insufficient AS remains for escaping, fighting and recovering. In the absence of controlled experiments, it remains unclear how these conflicting requirements for rapid growth and surplus AS may regulate prey-size selection of predatory animals.

Here, we investigate these ideas in an ambush predator, the barramundi (Lates calcarifer), which often consumes large meals in single feeding events. We quantified the SDA responses of juvenile barramundi fed meals of different sizes, and placed these data in the context of their available AS. We then combined these data with growth rates and growth efficiencies measured over seven weeks to provide new insight into some of the physiological mechanisms that may underlie prey-size selection.

2. Material and methods

Detailed information on animals, equipment, analyses and the SDA response is presented in the electronic supplementary material (including figures S1 and S2). Briefly, metabolic rates and SDA variables of barramundi in seawater at 29°C were calculated from oxygen uptake rates () obtained using best practices [5] and an automated, intermittent-closed respirometry set-up described previously [6]. Unless otherwise stated, values are presented as means ± 2 s.e.

In March 2012, 24 post-absorptive barramundi (fasted for 44–48 h) were individually fed different rations of dry food pellets (see the electronic supplementary material) corresponding to 0.6–3.4% of fish body mass (M_b; 29.5 ± 1.8 g). The fish consumed their meals voluntarily as pellets were thrown (and counted) one at a time into the individual holding tanks over a 2 min period. Ten minutes after feeding, the fish were transferred into respirometry chambers using water-filled containers, then kept at normoxia and 29°C throughout experiments. measurements commenced immediately (5/5 min closed/flush cycles) and continued for approximately 42 h to capture the entire SDA response. No food was regurgitated using this protocol. All fish reached a stable baseline towards the end of the approximately 42 h period, which was used to calculate minimum oxygen uptake rate (; equivalent to standard aerobic metabolic rate) as the mean of the lowest 10% of the measurements (excluding outliers). A control, unfed group was also included to assess handling effects (n = 12; M_b = 32.3 ± 3.2 g).

Thereafter, individual fish were transferred to a tub containing 25 l of normoxic seawater at 29°C and challenged with a 3 min exhaustive exercise (chase) protocol. No fish could sustain burst swimming for more than 2 min, but chasing/stimulating was maintained to ensure physical exhaustion at the end of the protocol. Fish were placed back into respirometers within 8 s following cessation of exercise and the maximum oxygen uptake rate (; equivalent to maximum aerobic metabolic rate) was estimated as the highest post-exercise over 3 min (cf. fig. 1 in [4]). This allowed the SDA process to be evaluated in the context of available AS, calculated as minus .

Growth of another 20 barramundi, fed quantified meal sizes, was also recorded over a seven-week period (M_b,initial = 34.5 ± 3.2 g, M_b,final = 43.7 ± 4.2 g). Specific growth rate (SGR) was calculated as the percentage increase in M_b per day (see electronic supplementary material). Growth efficiency was calculated as total M_b gain (wet weight) divided by total food intake (dry weight).

In a follow-up experiment (June 2013), we investigated whether the of unfed barramundi differed from that of fed fish (i.e. if there was an additive metabolic effect of digestion and exercise). Fish were exercised either while fasted (n = 6; M_b = 15.2 ± 2.0 g) or approximately 3.5 h after being fed pellets corresponding to 2.0 ± 0.6% of fish M_b (n = 6; M_b = 17.4 ± 2.4 g), with 3.5 h post-feeding representing the average time to reach peak during digestion () (cf. electronic supplementary material, figure S2).

For calculations of SDA variables, the average measured in unfed fish in response to being introduced into the respirometers (i.e. the handling response) was first subtracted from the of fed fish, after which values were smoothed (5-point moving median) to account for brief periods of spontaneous activity. SDA duration was calculated as the time from feeding to when three values had fallen below + 2 s.d. was calculated as the mean of the highest five consecutive values. SDA was calculated from the amount of oxygen consumed above during the SDA process, converted to units of energy from the amount of digestible meal energy (18.44 kJ g⁻¹) and by assuming a caloric coefficient of 20.08 kJ (l O₂)⁻¹ [7]. The SDA coefficient was calculated as SDA divided by digestible meal energy multiplied by 100.

Statistical analyses were performed in R [8]. Relationships in the data were investigated using regression analyses, details of which are provided in figure captions.

3. Results

SDA increased predictably with meal size (F_1,22 = 187.20, p < 0.0001, r² = 0.895; figure1a), indicating that larger meals required more energy to digest. However, the SDA coefficient decreased exponentially with meal size (F_2,21 = 34.22, p < 0.0001, r² = 0.765; figure1b), revealing that the percentage of meal energy used in the SDA process decreased from greater than 25% to approximately 11% as meal size increased from approximately 0.5% to 1.5% of fish M_b, and then plateaued at approximately 11% with further increases in meal size. Growth rate (F_1,18 = 32.57, p < 0.0001, r² = 0.644; figure2a) and growth efficiency (F_1,18 = 11.03, p = 0.0038, r² = 0.380; figure2b) both increased linearly with food intake.

Both (F_1,22 = 29.54, p < 0.0001, r² = 0.573; figure1c) and SDA duration (F_1,22 = 7.106, p = 0.0141, r² = 0.244) were positively related to meal size. There was no difference between mean of fasted and digesting fish in our follow-up experiment (7.72 ± 0.82 versus 7.79 ± 0.80 mg O₂ min⁻¹ kg⁻¹, respectively; t = 0.125, d.f. = 10, p = 0.903), indicating that barramundi do not have an additive metabolic response to digestion and exercise. Consequently, the percentage of AS occupied at the peak of SDA (i.e. ) increased with meal size (F_1,22 = 16.44, p < 0.001, r² = 0.428; figure1d) and ranged from approximately 35% of AS after small meals to approximately 62% of AS after large meals, with values for individual fish ranging from 21 to 77%.

4. Discussion

This study identified a growth versus performance trade-off between meal size and AS in barramundi. Specifically, the improvements in growth rate and growth efficiency associated with large meals come at the cost of a significant reduction in surplus AS. Thus, the SDA processes following large meals leave little room for other important oxygen- and energy-demanding activities.

The inability of barramundi to increase during exercise after feeding suggests that the reduced AS will translate into reduced aerobic swimming capacity. This provides a metabolic mechanism behind the reduced swimming speeds observed for several fishes after ingestion of large meals [9–13]. The proposed trade-off between meal size and AS may have important consequences for fitness, as demonstrated by data for Atlantic silversides (Menidia menidia): high food consumption and rapid growth reportedly led to reductions in swimming performance, which in turn led to increased mortality through predation [13,14]. Thus, the benefits of fast growth were outweighed by the fitness costs associated with reduced AS.

Our results suggest that barramundi, and probably many other animals, would benefit from regulating their meal sizes based on circ*mstance. If habitat conditions are perceived to be safe, it would make sense to target relatively large prey in order to yield the associated growth efficiency. By contrast, if the habitat is perceived to be unsafe or uncertain, a better fitness strategy may be to eat mid-sized meals to minimize the SDA coefficient while retaining sufficient AS to endure social and predatory interactions. Environmental factors must also be taken into consideration. For example, temperature changes can impact the digestive capacity of ectotherms by reducing AS via unequal changes in and [15], suggesting that variable environments may demand changes in individual meal sizes to maintain AS at the cost of growth rate and efficiency.

Perceived predation risk is thought to influence foraging, growth rate and growth efficiency [16,17], and foraging decisions can be regulated based on body energy reserves [18]. We argue that bet-hedging on available AS through controlled food intake may contribute to such state-dependent decision-making.

Supplementary Material

Supplementary Material

Supplementary Material

Supplementary Material

Ethics

Experiments were conducted with ethical approval from the James Cook University Animal Ethics Committee in association with the Australian Institute of Marine Science (permit A1792).

Data accessibility

Data are available in the electronic supplementary material.

Authors' contributions

T.N. and T.D.C. designed and performed the experiments and wrote the paper. Both authors approved the final version of the manuscript and agree to be held accountable for the integrity of the work performed.

Competing interests

The authors have no competing interests.

Funding

T.N. was funded by the Graduate School of Science and Technology at Aarhus University, Denmark, and the Danish Council for Independent Research (grant DFF-4181-00297).

References