Thermal tolerance is a critical determinant of ectotherm distribution, which is likely to be influenced by future climate change. To predict such distributional changes, simple and comparable measures of heat tolerance are needed and these measures should ideally correlate with the characteristics of the species current thermal environments. A recent model (thermal tolerance landscapes—TTLs) uses the exponential relation between temperature and knockdown time to describe the thermal tolerance of ectotherms across time/temperature scales. Here, we established TTLs for 11 Drosophila species representing different thermal ecotypes by measuring knockdown time at 9–17 stressful temperatures (0.5°C intervals). These temperatures caused knockdown times ranging from <10 min to >12 hrs and all species displayed the expected exponential relation between temperature and knockdown time (average R2 = 0.98). Previous studies using TTLs have reported a trade‐off between tolerance to acute and chronic heat stress in ectotherms. The present study did not find evidence to support this trade‐off in drosophilids. Instead, we show how this “trade‐off” can arise as an analytical artefact caused by insufficient data collection and excessive data extrapolation. Dynamic assays represent an alternative method to describe heat tolerance of ectotherms, where animals are exposed to gradually increasing temperatures until knockdown. The comparability of static and dynamic assays has previously been questioned, but here we show that static and dynamic assays give comparable information on heat tolerance. Using the constants derived from static TTLs, we mathematically model the expected dynamic knockdown temperature and subsequently confirm this model by comparison to empirically obtained knockdown temperatures from all 11 species. Characterisation of heat tolerance in laboratory settings is an important tool in thermal biology, but more so if the measures correlate with the environmental gradients that characterise the fundamental niche of species. Here, we show that both static and dynamic assays were characterised by strong correlations to precipitation of the driest month and maximum temperature of the warmest month combined (R2 = 0.68–0.71). This demonstrates that both assay types offer simple measures of heat tolerance that are ecologically relevant for the tested drosophilids. A plain language summary is available for this article.

Temperature tolerance is critical for defining the fundamental niche of ectotherms and researchers classically use either static (exposure to a constant temperature) or dynamic (ramping temperature) assays to assess tolerance. The use of different methods complicates comparison between studies and here we present a mathematical model (and R-scripts) to reconcile thermal tolerance measures obtained from static and dynamic assays. Our model uses input data from several static or dynamic experiments and is based on the well-supported assumption that thermal injury accumulation rate increases exponentially with temperature (known as a thermal death time curve). The model also assumes thermal stress at different temperatures to be additive and using experiments with Drosophila melanogaster, we validate these central assumptions by demonstrating that heat injury attained at different heat stress intensities and durations is additive. In a separate experiment we demonstrate that our model can accurately describe injury accumulation during fluctuating temperature stress and further we validate the model by successfully converting literature data of ectotherm heat tolerance (both static and dynamic assays) to a single, comparable metric (the temperature tolerated for 1 h). The model presented here has many promising applications for the analysis of ectotherm thermal tolerance and we also discuss potential pitfalls that should be considered and avoided using this model.

Upper thermal limits (CTmax) are frequently used to parameterize the fundamental niche of ectothermic animals and to infer biogeographical distribution limits under current and future climate scenarios. However, there is considerable debate associated with the methodological, ecological and physiological definitions of CTmax. The recent (re)introduction of the thermal death time (TDT) model has reconciled some of these issues and now offers a solid mathematical foundation to model CTmax by considering both intensity and duration of thermal stress. Nevertheless, the physiological origin and boundaries of this temperature–duration model remain unexplored. Supported by empirical data, we here outline a reconciling framework that integrates the TDT model, which operates at stressful temperatures, with the classic thermal performance curve (TPC) that typically describes biological functions at permissive temperatures. Further, we discuss how the TDT model is founded on a balance between disruptive and regenerative biological processes that ultimately defines a critical boundary temperature (Tc) separating the TDT and TPC models. Collectively, this framework allows inclusion of both repair and accumulation of heat stress, and therefore also offers a consistent conceptual approach to understand the impact of high temperature under fluctuating thermal conditions. Further, this reconciling framework allows improved experimental designs to understand the physiological underpinnings and ecological consequences of ectotherm heat tolerance.

No abstract

Ectotherm thermal tolerance is critical to species distribution, but at present the physiological underpinnings of heat tolerance remain poorly understood. Mitochondrial function is perturbed at critically high temperatures in some ectotherms, including insects, suggesting that heat tolerance of these animals is linked to failure of oxidative phosphorylation (OXPHOS) and/or ATP production. To test this hypothesis we measured mitochondrial oxygen consumption rates in six Drosophila species with different heat tolerance using high-resolution respirometry. Using a substrate-uncoupler-inhibitor titration protocol we examined specific steps of the electron transport system to study how temperatures below, bracketing and above organismal heat limits affected mitochondrial function and substrate oxidation. At benign temperatures (19 and 30°C), complex I-supported respiration (CI-OXPHOS) was the most significant contributor to maximal OXPHOS. At higher temperatures (34, 38, 42 and 46°C), CI-OXPHOS decreased considerably, ultimately to very low levels at 42 and 46°C. The enzymatic catalytic capacity of complex I was intact across all temperatures and accordingly the decreased CI-OXPHOS is unlikely to be caused directly by hyperthermic denaturation/inactivation of complex I. Despite the reduction in CI-OXPHOS, maximal OXPHOS capacities were maintained in all species, through oxidation of alternative substrates; proline, succinate and, particularly, glycerol-3-phosphate, suggesting important mitochondrial flexibility at temperatures exceeding the organismal heat limit. Interestingly, this failure of CI-OXPHOS and compensatory oxidation of alternative substrates occurred at temperatures that tended to correlate with species heat tolerance, such that heat-tolerant species could defend “normal” mitochondrial function at higher temperatures than sensitive species. Future studies should investigate why CI-OXPHOS is perturbed and how this potentially affects ATP production rates.