Temperature and O2, but not CO2, interact to affect aerobic performance of European sea bass (Dicentrarchus labrax)

Climate change causes warming, decreased O2, and increased CO2 in marine systems and responses of organisms will depend on interactive effects between these factors. We provide the first experimental assessment of the interactive effects of warming (14 to 22°C), reduced O2 (∼3 – 21 kPa O2), and increased CO2 (∼400 or ∼1000 µatm ambient CO2) on four indicators of aerobic performance (standard metabolic rate, SMR, maximum metabolic rate, MMR, aerobic scope, and hypoxia tolerance, O2crit), blood chemistry, and O2 transport (P50) of a marine fish, the European sea bass (Dicentrarchus labrax). Warming increased SMR and O2crit (i.e. reduced hypoxia tolerance) as well as MMR in normoxia but there was an interactive effect with O2 so that hypoxia caused larger reductions in MMR and aerobic scope at higher temperatures. Increasing CO2 had minimal effects on SMR, MMR and O2crit and did not show interactive effects with temperature or O2 for any measured variables. Aerobic performance was not linked to changes in blood chemistry or P50. Despite lack of effects of CO2 on aerobic performance, increased CO2 induced 30% mortality of fish exercised in low O2 at 22°C indicating important threshold effects independent of aerobic performance. Overall, our results show temperature and O2, but not CO2, interact to affect aerobic performance of sea bass, disagreeing with predictions of the oxygen- and capacity-limited thermal tolerance hypothesis.

experiments investigating effects of O2 or CO2 and temperature on aerobic 96 performance may not accurately reflect interactive effects caused by all three 97 environmental factors. 98 In this study we investigated how temperature, O2, and CO2 interact to affect 99 aerobic performance of European sea bass (Dicentrarchus labrax), a species 100 showing recent northward range expansions thought to be related to warming 101 (Pawson et al., 2007). Two separate populations exist (Souche et al., 2015) and, 102 although the physiological responses of this species to environmental change have 103 been regularly examined, to date only one study has used fish from the Atlantic 104 population. As such, our experiment had three aims: to assess how aerobic performance of sea bass from the Atlantic population 106 will respond to predicted future environmental changes; 107 ii. to determine whether combinations of hypoxia and increased CO2 interact 108 with temperature to affect aerobic scope, as predicted by the OCLTT exhibiting a Q10 temperature coefficient of 2.09 ( Figure 1). The best model indicated 118 that CO2 had a small effect on SMR -reducing SMR by 7.4 mgO2 kg -1 h -1 , ~10 % of 119 the smallest SMR, (95 % CI = -1.65 to 16.43 mgO2 kg -1 h -1 ) across all temperatures at 120 ~1000 µatm CO2 (Figure 1). However, the model including temperature but not CO2 121 had a ΔAICc <2 indicating that including the effect of CO2 in the best model did not 122 lead to a large improvement in model fit (Table S5). There was no evidence of an 123 interactive effect between increasing temperatures and increasing CO2.  The effect of temperature meant that for a given value of SMR O2crit would reduce as  caused by a significant reduction in pHe in sea bass exposed to ~1000 µatm CO2 at 233 18 o C (7.80 ± 0.03) when compared to fish at ambient CO2 levels (7.97 ± 0.04)

234
(pairwise comparisons of least square means, t = 3.242, df = 1, p = 0.026). We did 235 not find significant differences between pHe across all other treatment groups.

238
There were no significant differences in plasma lactate levels across all  The consistency in temperature of peak performance in sea bass from distinct sub- fish in our study was higher at a given temperature than for fish from Mediterranean 300 stocks (Claireaux & Lagardère, 1999). This may support the theory of metabolic cold 301 adaptation, that basal energy demand in fish from warmer environments will be lower 302 than in fish from cold environments when measured at the same temperature 303 (Krogh, 1916  given SMR. This suggests that temperature affects O2crit via another mechanism (or 334 mechanisms) independent of SMR. This is unlikely to be related to O2 transport 335 capacity of the blood as there were few consistent effects of temperature on Hct, Hb 336 or P50 of sea bass ( Figure 5, S1 and S2 CO2 has a greater impact on MMR at lower O2 levels (Figure 3 C, D, and E).

364
Decreased MMR when fish are exposed to acutely increased CO2 is usually thought  showing that aerobic scope of sea bass would be expected to remain constant 431 across an 8 o C temperature range at an O2 level of ~6 kPa (Figure 3 A and B). As

453
In summary, our research shows that aerobic scope of European sea bass will   Rates of oxygen consumption (ṀO2) were made as a proxy of metabolic rate 536 using an intermittent-flow respirometer system, details of which can be found in    (Table S3). We then measured extracellular pH  We conducted all statistical analysis in R v3.6.3 (R Core Team, 2020). Results

642
are reported as mean ± S.E unless otherwise stated. Sample sizes for respirometry 643 data can be seen in Table S4. The effects of temperature, O2, and CO2 on individual AICc (Burnham & Anderson, 1998;Hurvich & Tsai, 1989). Residual diagnostic plots of each GLMM were then assessed using package 'DHARMa' to confirm validity of 652 model fit (Hartig, 2020). Once the best supported model for each physiological 653 parameter was identified (see Table S5  Kruskal-Wallis test. Sample sizes for blood chemistry data can be seen in Table S4.