Aerobic scope falls to nil at P_crit and anaerobic ATP production increases below P_crit in the tidepool sculpin, Oligocottus maculosus

Abstract

The critical oxygen tension of whole-animal oxygen uptake rate, or P_crit, has historically been defined as the oxygen partial pressure (P_O2) at which aerobic scope falls to zero and further declines in P_O2 require substrate-level phosphorylation to meet shortfalls in aerobic ATP production, thereby time-limiting survival. Despite the inclusion of aerobic scope and anaerobic ATP production in the definition, little effort has been made to verify that P_crit measurements, the vast majority of which are obtained using respirometry in resting animals, actually reflect the predictions of zero aerobic scope and a transition to increasing reliance on anaerobic ATP production. To test these predictions, we compared aerobic scope and levels of whole-body lactate at oxygen partial pressures (P_O2s) bracketing P_crit obtained in resting fish during progressive hypoxia in the tidepool sculpin, Oligocottus maculosus. We found that aerobic scope falls to zero at P_crit and, in resting fish exposed to P_O2s < P_crit, whole-body lactate accumulated pointing to an increased reliance on anaerobic ATP production. These results support the interpretation of P_crit as a key oxygen threshold at which aerobic scope falls to nil and, below P_crit, survival is time-limited based on anaerobic metabolic capacity.

1. Introduction

In marine environments, global climate change is leading to increasing mean temperatures, greater frequency of temperature extremes, increasing size and severity of hypoxic zones and declines in total ocean oxygen content [1–3]. Understanding and predicting how changes in temperature and oxygen affect the distribution and abundance of marine organisms depends on understanding the physiological effects of temperature and oxygen on ecologically relevant performance. Several frameworks for projecting climate change impacts on marine organisms focus on temperature and oxygen impacts on aerobic metabolic performance [4–8].

Recent efforts to ground projections of changing ocean temperatures and oxygen on marine organism distributions through temperature and oxygen effects on organismal physiology have relied on the critical oxygen tension for standard metabolic rate (SMR), or P_crit, and its temperature sensitivity [4,6,7]. Following the classic framework developed by Fry [9], P_crit is defined as the oxygen partial pressure (P_O2) at which an organism's aerobic scope, the difference between maximum and standard oxygen uptake rates, is zero and, in the absence of metabolic rate suppression, exposure to further declines in P_O2 requires a proportional increase in anaerobic ATP production to meet standard metabolic energy demands [4,7,8,10–13].

Fry recommended P_crit be determined as the intercept of a maximum oxygen uptake rate (${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$)–P_O2 curve (i.e. a limiting oxygen level curve, [13]) and SMR to define P_crit based on an empirically determined threshold in aerobic scope [9]. Metabolic analyses of P_crit estimated under this definition support the Fry hypothesis [14]; however, the vast majority of P_crit measurements are made in resting fish exposed to progressive hypoxia [15], and recent analyses of best practices continue to advocate for estimation of P_crit in resting fish [16,17]. The discrepancy between the theoretical definition of P_crit and its usual method of empirical estimation [15] has led to controversy surrounding the physiological significance of this trait [10,18–23]. Critics claim P_crit measured in resting fish does not indicate the reduction of aerobic scope to nil or a concomitant transition to supplemental anaerobic metabolism in hypoxia. Although methodological variation contributes to the confusion around P_crit [16,17,20,21], a key problem is that no study has directly experimentally tested whether P_crit under resting conditions reflects the respiratory and metabolic predictions made by Fry's hypothesis.

The Fry hypothesis makes three predictions about the respiratory and metabolic significance of P_crit: (1) ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ is equal to the standard rate of oxygen uptake (${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Standard}}$) in an organism exposed to P_O2 = P_crit, (2) at least within a range of oxygen partial pressures (P_O2s) near P_crit, ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ will increase or decrease in direct proportion with an increase or decrease in P_O2 above or below P_crit and (3) end products of substrate-level phosphorylation, such as lactate, should not accumulate in organisms under standard conditions unless P_O2 < P_crit. Although some datasets from studies measuring P_crit under resting conditions are consistent with some of these predictions [9,24–29], others are not [30], and no study has directly experimentally tested all three.

We tested whether P_crit measured in resting fish indicates a threshold in aerobic scope and anaerobic ATP production in a tractable model, the tidepool sculpin Oligocottus maculosus. Oligocottus maculosus is a benthic, tidepool-specializing fish found in abundance along the northern Pacific coast of North America [31]. This species is regularly exposed to acute, dramatic variation in water P_O2 due to tidal emersion of its low-volume home pools, and as such remains relatively quiescent during progressive hypoxia exposure [32]. Following current recommendations for P_crit measurement [16,17], we measured normoxic ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$, ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Standard}}$, P_crit and ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ at P_O2s including and bracketing P_crit. We also measured whole-body lactate in quiescent fish held at P_O2s bracketing and including P_crit. Together our results demonstrate that in O. maculosus, consistent with the predictions from the Fry hypothesis [9], P_crit occurs due to an environmentally imposed limitation of maximum respiratory capacity to meet oxygen uptake demands at ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Standard}}$. As a consequence, P_crit is a threshold below which anaerobic substrate-level phosphorylation must be used to meet minimum ATP demands, likely limiting survival time.

2. Material and methods

(a) . Animal collection and housing

Tidepool sculpins (O. maculosus) used in respirometry experiments were collected in 2018 and 2019 from rocky intertidal pools on the west coast of Bowen Island (British Columbia, Canada, 49.3411°N, 123.4319°W) using minnow traps and dipnets. Fish used in lactate experiments were collected from the same location in 2020. Body mass data for all experiments are reported in electronic supplementary material, table S1. Animals were transported to The University of British Columbia (UBC) and housed in a recirculation system with aerated 35‰ artificial seawater (Instant Ocean, Blacksburg, VA, USA) at 12°C with a 12 L : 12 D photoperiod for at least one month prior to experimentation. Fish were fed ad libitum three times per week on a diet of Antarctic krill, spirulina-fed brine shrimp and blood worms. Fish collections were permitted under Department of Fisheries and Oceans Canada permits (XR 239 2017) and (XE 59 2019). All experimental procedures were approved by the UBC Animal Care Committee (A17-0293).

(b) . Experimental protocols

(i) . Respirometry: normoxic ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$, ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Standard}}$ and P_crit

Detailed respirometry protocols and calculations are given in the electronic supplementary material and electronic supplementary material, table S2 summarizes details recommended for reporting to enhance reproducibility in respirometry [33]. For the 2018 and 2019 collections, normoxic ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$, ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Standard}}$, and P_crit were first determined for each fish individually in a contiguous protocol. After 3 days of fasting, fish were chased to exhaustion in acclimation-tank water and immediately placed in a glass respirometer with a mass (g) to volume (ml) ratio of approximately 1 : 15 (electronic supplementary material, table S2). The respirometer was sealed immediately to obtain an estimate of ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$. After approximately 5 min of closed respirometry, intermittent-flow respirometry was used over the subsequent 48 h (2018) or 70 h (2019) period to estimate ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Standard}}$ followed by a closed-respirometer P_crit trial, in which fish oxygen consumption produced a progressive hypoxia exposure.

(ii) . Respirometry: ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ in hypoxia

In 2018, fish were recovered from normoxic ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$–${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Standard}}$–P_crit trials for a minimum of 3 days during which they were fed at least once followed by 3 days of fasting. To estimate ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ at P_crit, each fish collected in 2018 was chased to exhaustion in the same conditions as for normoxic ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ measurement and immediately placed in a respirometer at a P_O2 0.31 ± 0.09 kPa above each individual's estimated P_crit and allowed to consume oxygen to a P_O2 0.36 ± 0.15 kPa below P_crit (mean P_crit = 2.22 kPa). This method ensured the mean P_O2 experienced during the trial was equal to the target P_O2, i.e. P_crit.

In 2019, in addition to estimating normoxic ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$, ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Standard}}$, P_crit (mean = 2.55 kPa), and ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ at P_crit, ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ estimates were collected at the following three P_O2s bracketing each individual fish’ P_crit: 65% P_crit (mean = 1.66 kPa), P_crit + 6.5% air saturation (air sat.)) (mean = 3.88 kPa) and P_crit + 15.5% air sat (mean = 5.79 kPa). The ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ measurements at each of the five hypoxic P_O2s collected in 2019 were performed over a 3 day period. ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ at P_crit was collected on day 1. Fish were recovered overnight, and ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ at P_crit + 6.5% air sat. were made the morning of day 2, followed by ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ at P_crit + 15.5% air sat. in the afternoon of day 2 with a minimum recovery of 3.5 h. Measurements of ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ at 65% P_crit were made the morning of day 3 and fish were then recovered for a minimum of 3.5 h. The fish were then subjected to a second round of ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ at P_crit + 15.5% air sat. to test for training effects on ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ at this P_O2. There was a significant increase in the second ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ at P_crit + 15.5% air sat. measurement (t = −3.66, df = 8, p = 0.0064, mean difference = 3.2 µmol h⁻¹, electronic supplementary material, figure S4). The effect size of training, calculated as Cohen's d: $\overline{X_{\mathrm{D}}}\hspace{0.17em}/\hspace{0.17em}{\sigma }_{\mathrm{D}}$ where $\overline{X_{\mathrm{D}}}$ is the mean difference between the second and first ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ trials at P_crit + 15.5% AS, respectively, and ${\sigma }_{\mathrm{D}}$ is the standard deviation of the differences, was 1.22, which is considered an appreciable effect. We therefore calculated the mean of the two trials for each individual and used this value in subsequent statistical analysis of P_O2 effects on ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$.

(iii) . Whole-body lactate dynamics at or near P_crit

Fish from the 2020 collection were divided into five aquaria and acclimated overnight before hypoxia exposures the following morning. To start the experiment, P_O2 in four of the five aquaria was reduced over 45 min to one of the hypoxic P_O2s used in the 2019 hypoxic ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ experiment, where P_O2 = P_crit was based on the mean P_crit from the 2019 respirometry experiment. The fifth aquarium served as a normoxic control. P_crit in fish collected in 2020 for the lactate experiment did not differ from the 2019 fish (90% CI for mean P_crit of 2019 fish = [1.07 kPa, 4.15 kPa], mean P_crit of fish tested (N = 2) from 2020 collection = 2.95 kPa, electronic supplementary material, figure S9). Fish were held at the target P_O2s for 6 h, after which a lethal dose of benzocaine (250 mg l⁻¹, Sigma-Aldrich, USA) was unobtrusively added to each aquarium. Fish were fully anaesthetized within 1 min of exposure without showing signs of stress. Fish were frozen whole in liquid nitrogen and stored at −80°C until analysis of whole-body lactate content. Justification of euthanasia method and detailed description of lactate analytical methods are given in electronic supplementary material and methods.

3. Results

${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ declined progressively from normoxia to the lowest P_O2 assessed (linear mixed model (LMM) 2018: F_2,10 = 150, p = 3 × 10⁻⁸; LMM 0219: F_5,39.9 = 287, p < 2.2 × 10⁻¹⁶, figure 1, electronic supplementary material, table S5). ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ at P_crit either did not differ from (2018) or was significantly lower than ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Standard}}$ (2019, 32% lower than ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Standard}}$), with a further decline in ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ below P_crit or progressive increases in ${\dot{M}}_{{\mathrm{O}}_2,\mathrm{Max}}$ above P_crit (figure 1).

Figure 1.

Oxygen uptake rate (${\dot{M}}_{{\mathrm{O}}_2}$) versus oxygen partial pressure (P_O2) from the 2018 (N = 6 fish) and 2019 (N = 10 fish) experiments in the tidepool sculpin, Oligocottus maculosus. Within each year, ${\dot{M}}_{{\mathrm{O}}_2}$ was measured in each fish at each P_O2. Data are mean ± sem. Different letters indicate significant differences between means in the 2018 experiment (Greek letters) or 2019 experiment (Latin letters) based on Tukeys post hoc comparisons following detection of significant effects of oxygen level (linear mixed model (LMM) 2018: F_2,10 = 150, p = 3 × 10⁻⁸; LMM 2019: F_5,39.9 = 287, p < 2.2 × 10⁻¹⁶, electronic supplementary material, table S5).

In fish at rest, during 6 h of hypoxia exposure, whole-body [lactate] did not significantly differ from normoxic resting levels at P_O2s at or above P_crit (ordinary least-squares regression: F_4,47 = 39.4, p = 1.85 × 10⁻¹⁴, figure 2, electronic supplementary material, table S5). Whole-body [lactate] appeared to increase nearly exponentially when fish were held at P_O2s below P_crit.

Figure 2.

Whole-body lactate concentration versus P_O2. Data are mean ± sem. P_crit = mean P_crit obtained in 2019 (figure 1, see materials and methods). Different letters indicate significant differences between means based on Tukeys post hoc comparisons following detection of significant effects of oxygen level (ordinary least-squares model, F_4,47 = 39.4, p = 1.85 × 10⁻¹⁴, electronic supplementary material, table S5). Normoxia: N = 12, all other P_O2s: N = 10.

4. Discussion

Critiques of the P_crit concept and interpretation of P_crit's physiological significance have suggested that P_crit does not indicate a threshold in respiratory capacity to meet minimum aerobic metabolic demands and therefore contains little physiologically or ecologically useful information [18,34]. The responses to these critiques have focused on either methodological or theoretical considerations supporting P_crit as typically interpreted in the Fry framework [16,17,20,21]. Here, we find that P_crit in O. maculosus, measured using the widely implemented closed-respirometry style progressive hypoxia exposure in resting fish, captures an important threshold in respiratory capacity to meet minimum aerobic metabolic demands, tied to a specific environmental P_O2, below which survival is time limited by the capacity for substrate-level phosphorylation (i.e. anaerobic ATP production) and metabolic suppression. Consistent with predictions from the Fry hypothesis, P_crit in the tidepool sculpin appears to occur due to the hypoxic inhibition of respiratory capacity to meet standard O₂ requirements, i.e. absolute aerobic scope falls to nil at P_crit (figure 1). With any further decline in environmental P_O2 below P_crit, substrate-level phosphorylation increases to meet minimum (standard) ATP demand associated with fasted quiescence (figure 2). This imposes a time-limitation on survival dependent on the capacity for substrate-level phosphorylation to meet the shortfall in aerobically produced ATP and any capacity for metabolic suppression of ATP demands. Our empirical results, determined using currently recommended methodologies [16,17], support the theoretical arguments made in support of the physiological and ecological utility and significance of P_crit [20,21].

Our empirical support for the classic interpretation of the physiological significance of P_crit supports the use of P_crit in climate change-related modelling efforts, but several challenges remain. O. maculosus remains quiescent during acute progressive hypoxia exposure [32], but agitated activity during hypoxia exposure could confound analyses of P_crit in other species and should be taken into account. P_crit clearly responds to acclimation to various stressors [15,35–38] and evolves in species that have invaded permanently hypoxic or oxygen-variable environments [35,39–41], complicating a straight-forward use of P_crit data in projections of species responses to climate change using models which do not account for these processes [42]. Although unifying methodological approaches may result in support for the Fry conception of P_crit in species where previously there was not (e.g. [30]), more work incorporating factors known to impact respiratory and metabolic traits (e.g. acclimation, adaptation and ecology, [43–47]) is needed to establish broad support for P_crit's physiological significance and use in ecophysiological modelling.

Ethics

Fish were collected under Department of Fisheries and Oceans permits (XR 239 2017) and (XE 59 2019). All experimental procedures were approved by the UBC Animal Care Committee (A17-0293).

Data accessibility

All raw data and R code for this study are publicly available at the following FigShare link: https://doi.org/10.6084/m9.figshare.c.6211609.v1 [48]

Supplementary material is available online [49].

Authors' contributions

D.A.S.: conceptualization, data curation, formal analysis, investigation, methodology, visualization, writing—original draft, writing—review and editing; K.C.: investigation, writing—review and editing; J.G.R.: conceptualization, funding acquisition, supervision, writing—review and editing.

All authors gave final approval for publication and agreed to be held accountable for the work performed therein.

Conflict of interest declaration

We declare we have no competing interests.

Funding

Funding support was provided by a Natural Sciences and Engineering Research Council (NSERC) of Canada Discovery Grant to J.G.R. D.A.S. was supported by The University of British Columbia Graduate Four Year Fellowship.