Content uploaded by Nann A Fangue
Author content
All content in this area was uploaded by Nann A Fangue
Content may be subject to copyright.
A preview of the PDF is not available
Countergradient Variation in Temperature Preference in Populations of Killifish Fundulus heteroclitus
Abstract
Behavioral thermoregulation can allow ectotherms to buffer the effects of changes in environmental temperature, and thus an organism's preferred temperature is thought to be under strong selection. However, this contention has seldom been tested. We used common killifish Fundulus heteroclitus from high-latitude (northern) and low-latitude (southern) populations to investigate intraspecific variation in thermal preference and its relationship to habitat temperature. We quantified the preferred temperatures of northern and southern killifish populations acclimated to three temperatures (5 degrees , 15 degrees , and 25 degrees C) to evaluate two alternative hypotheses for the evolution of differences in thermal preference among latitudinally separated populations: local thermal adaptation, which predicts that organisms from high latitudes should prefer lower temperatures than individuals from lower latitudes, versus countergradient variation, which predicts that high-latitude organisms should prefer higher temperatures to compensate for shorter growing seasons. All killifish selected their final thermal preferendum within 4 h. Southern killifish and killifish acclimated to warmer temperatures had greater variability in selected temperature. This increase in variability was the result of an increase in interindividual variation in preferred temperature rather than a reduction in the precision of temperature selection in these groups. Northern killifish preferred significantly higher temperatures than southern fish (30.6 degrees vs. 29.0 degrees C, respectively, when calculated on the basis of the temperature selected consistently for at least 30 min; 28.4 degrees vs. 26.5 degrees C, respectively, when calculated on the basis of the mean temperature occupied), regardless of acclimation temperature. These data are not consistent with local adaptation in thermal preference but instead can be better explained by countergradient variation in thermal preference in killifish.
... A previous study showed that the embryonic development of subtropical annual fish species was faster under higher temperatures (Arenzon et al., 2002). Furthermore, many fish populations are capable of adjusting their development according to local environmental conditions (Polačik et al., 2014(Polačik et al., , 2018, leading to inter-and intraspecific variation in metabolism, physiology and thermal tolerance along environmental gradients (Fangue et al., 2009;Polačik et al., 2018;Powers & Schulte, 1998). Since low-temperature limits diversity and distributions of the freshwater biota in mountain areas, many authors consider that wetland species at high altitudes are especially vulnerable to climate change (Funnell et al., 1998;Wissinger et al., 2016). ...
... We specifically assessed the effects of increasing temperatures and exposure to heat waves on two aspects of annual fish embryonic development: egg mortality and trajectory. We expected heat waves and higher temperatures to be associated with higher egg mortality and higher number of eggs in diapause II in the high-altitude species, based on the assumption that the thermal physiology of fish varies among populations distributed along altitudinal gradients (Fangue et al., 2009;McKenzie et al., 2013). ...
Climate change projections predict warmer temperatures and increased frequency of heat waves in many regions across the globe. These scenarios are expected to strongly impact temperature‐dependent life‐history traits of aquatic species, such as the embryonic stage of annual fishes inhabiting temporary freshwater wetlands. Here, we assessed the effects of different incubation temperatures (18, 24, 27 and 30°C) and exposure to heat waves (18–30°C for 6 h) on aspects of the embryonic development: mortality and trajectory (direct development or diapause entry) of two Neotropical species of annual fishes from high‐ ( Austrolebias nubium ) and low‐altitude ( A. cyaneus ) areas. Temperatures of 30°C resulted in 100% embryo mortality in both species, and development trajectory differed between species across temperatures. Embryo mortality in the high‐altitude species was higher at 27°C, while a higher number of embryos of the low‐altitude species entered diapause II at 18°C. Embryo mortality and development trajectory after exposure to heat waves were similar between species. In terms of implications for climate change scenarios, temperatures above 27°C seem critical for the embryo survival of both species studied, although they seem capable to endure short‐term events of heat waves. In summary, our results indicate that the persistence of populations of high‐altitude annual fish species may be subject to higher threat under warming scenarios, since embryos from this species were more sensitive to increasing temperatures than the low‐altitude species.
... Secondly, the counter-gradient relationship hypothesis predicts that individuals experiencing extreme temperatures should be more sensitive to adverse thermal conditions (Llewelyn et al., 2016), and therefore they select lower temperatures (i.e., lower Tpref) when they are given the opportunity. Hence, individuals from cold environments select higher Tpref to compensate for lower opportunities to find favourable temperatures within the thermal environment (Conover and Schultz, 1995;Fangue et al., 2009;Hodgson and Schwanz, 2019;Llewelyn et al., 2016;McElroy, 2014). Thirdly, the Bogert effect (Bogert, 1949) suggests that behavioural thermoregulation inhibits the plastic response of Tpref (Huey et al., 2003;Marais and Chown, 2008). ...
... Across all six populations, CT max values increased with acclimation temperature and were consistent with previous work on salmonids (Cech and Myrick, 1999;Cech, 2000, 2001;Chen et al., 2015). Unlike research on Brook Trout (S. fontinalis [Stitt et al., 2014]) or Killifish (Fundulus heteroclitus [Fangue et al., 2009]), we found weakly-significant and equivocal effects of latitude on CT max a result that mirrors cross-taxa metanalysis on freshwater ectotherms (Sunday et al., 2019). This result suggests that latitude may be a poor predictor of acute thermal tolerance, especially if local watershed characteristics (e.g. ...
Understanding interpopulation variation is important to predicting species responses to climate change. Recent research has revealed interpopulation variation among several species of Pacific salmonids; however, the environmental drivers of population differences remain elusive. We tested for local adaptation and countergradient variation by assessing interpopulation variation among six populations of fall-run Chinook Salmon from the western United States. Juvenile fish were reared at three temperatures (11, 16 and 20°C), and five physiological metrics were measured (routine and maximum metabolic rate, aerobic scope, growth rate and critical thermal maximum). We then tested associations between these physiological metrics and 15 environmental characteristics (e.g. rearing temperature, latitude, migration distance, etc.). Statistical associations between the five physiological metrics and 15 environmental characteristics supported our hypotheses of local adaptation. Notably, latitude was a poor predictor of population physiology. Instead, our results demonstrate that populations from warmer habitats exhibit higher thermal tolerance (i.e. critical thermal maxima), faster growth when warm acclimated and greater aerobic capacity at high temperatures. Additionally, populations with longer migrations exhibit higher metabolic capacity. However, overall metabolic capacity declined with warm acclimation, indicating that future climate change may reduce metabolic capacity, negatively affecting long-migrating populations. Linking physiological traits to environmental characteristics enables flexible, population-specific management of disparate populations in response to local conditions.
... A common acclimation period for temperate fish species is 3-4 weeks under normoxic conditions (e.g., Fangue et al. 2009); a new cardiac phenotype can begin to appear even after 8 h of 4 • C acclimation (Sutcliffe et al. 2020;Gilbert et al. 2022). Therefore, we assumed that a steady-state phenotype would emerge after a normoxic acclimation period of 4 weeks and be stable thereafter. ...
Phenotypic plasticity manifested after acclimatization is a very important source of biological variability among fish species. We hypothesized that hypoxic acclimation, besides potentially generating a temporary hypoxic respiratory phenotype, would also manifest as a continued benefit after re-acclimation to normoxia. Hence, we holistically characterized the respiratory phenotype of European sea bass (Dicentrarchus labrax Linnaeus, 1758) acclimated to normoxia with or without prior acclimation to hypoxia. Compared with the original normoxic phenotype, prior acclimation to hypoxia and return to normoxia produced a 27% higher absolute aerobic scope (AAS), a 24% higher citrate synthase activity in red muscle and a 28% lower excess post-exercise O2 consumption. Additional testing of hypoxia-acclimated fish under normoxia explored the specific effects of hypoxic acclimation. The hypoxic phenotype, when compared with the original normoxic phenotype, had a lower standard metabolic rate, a better hypoxia performance and a lower minimum PO2 for supporting 50% AAS. Given this respiratory malleability, general predictions for marine fish exploiting a more hypoxic future should better consider respiratory plasticity and prolonged effects of hypoxic exposures.
... lability, behavioral responses can play a role in facilitating or hindering adaptation to new thermal environments, depending on certain attributes of the species and environmental conditions (Sih et al., 2011;Tuomainen & Candolin, 2011;van Jaarsveld et al., 2021). Thermoregulatory behavior could be especially important in the context of climate change, as it allows animals to buffer the effects of temperature changes by choosing suitable microhabitats (Fangue et al., 2009;Huey et al., 2012;Kearney et al., 2009). Animals usually seek temperatures that coincide with their optimal physiological performance and growth, which tends to be determined by their thermal evolutionary history (Díaz et al., 2007;Jobling, 1981;Kellogg & Gift, 1983;Köhler et al., 2011;Pörtner & Farrell, 2008; but see Buckley et al., 2022;Huey & Bennett, 1987;Martin & Huey, 2008). ...
Given the threat of climate change to biodiversity, a growing number of studies are investigating the potential for organisms to adapt to rising temperatures. Earlier work has predicted that physiological adaptation to climate change will be accompanied by a shift in temperature preferences, but empirical evidence for this is lacking. Here, we test whether exposure to different thermal environments has led to changes in preferred temperatures in the wild. Our study takes advantage of a "natural experiment" in Iceland, where freshwater populations of threespine sticklebacks (Gasterosteus aculeatus) are found in waters warmed by geothermal activity year-round (warm habitats), adjacent to populations in ambient-temperature lakes (cold habitats). We used a shuttle-box approach to measure temperature preferences of wild-caught sticklebacks from three warm-cold population pairs. Our prediction was that fish from warm habitats would prefer higher water temperatures than those from cold habitats. We found no support for this, as fish from both warm and cold habitats had an average preferred temperature of 13°C. Thus, our results challenge the assumption that there will be a shift in ectotherm temperature preferences in response to climate change. In addition, since warm-habitat fish can persist at relatively high temperatures despite a lower-temperature preference, we suggest that preferred temperature alone may be a poor indicator of a population's adaptive potential to a novel thermal environment.
... We parameterized the model for the two subspecies of Atlantic killifish that live along the North American East Coast 33,34 (Fig. 2). The southern population is more heat tolerant (7-31 °C range) than the northern population (− 1.4 to 21 °C range) 33 , and the daily and seasonal temperature fluctuations vary widely between populations 33,35 . Thus, we calculated E values for a range of acclimation temperatures for northern and southern killifish based on data from an extensive killifish acclimation experiment 36 . ...
Species distribution models predict a poleward migration for marine ectotherms with ocean warming. However, a key limitation in current species distribution models (SDM) is that they do not account for population-specific heterogeneity in physiological responses to temperature change resulting from local adaptations and acclimatization. To address this gap, we developed a novel, Physiology Integrated BioClimate Model (PIBCM) that combines habitat-specific metabolic thermal physiological tolerance of a species into a bioclimate envelope model. Using a downscaling approach, we also established a fine-resolution coastal sea-surface temperature data set for 2050–2080, that showed a high degree of location-specific variability in future thermal regimes. Combining predicted temperature data with the PIBCM model, we estimated habitat distribution for a highly eurythermal intertidal minnow, the Atlantic killifish ( Fundulus heteroclitus ), a species that likely presents a best-case-scenario for coastal vertebrates. We show that the killifish northern boundary shifts southwards, while distinct habitat fragmentation occurs in the southern sub-population (due to migration of adjacent fish populations to the nearest metabolically optimal thermal habitat). When compared to current SDMs (e.g., AquaMaps), our results emphasize the need for thermal physiology integrated range shift models and indicate that habitat fragmentation for coastal fishes may reshape nursery habitats for many commercially and ecologically important species.
... For example, in the study by Healy and Schulte (2012a) of metabolic rate across a range of temperatures, thermal sensitivity was much more dependent on the temperatures being examined. Yet, Fangue et al. (2009b) found the thermal sensitivity of CT max remained somewhat consistent (Q 10 ≈1.1) across a 30°C temperature range. Lastly, repeated measurements in an individual for Q 10 would provide insight as to whether this trait is repeatable, although given past repeatability of metabolic rate and CT max , it is likely to be consistent (Drown et al., 2020;Healy and Schulte, 2012b;Morgan et al., 2018;Nespolo and Franco, 2007). ...
Physiology defines individual responses to global climate change and species distributions across environments. Physiological responses are driven by temperature on three timescales: acute, acclimatory and evolutionary. Acutely, passive temperature effects often dictate an expected two-fold increase in metabolic processes for every 10°C change in temperature (Q10). Yet, these acute responses often are mitigated through acclimation within an individual or evolutionary adaptation within populations over time. Natural selection can influence both responses and often reduces interindividual variation towards an optimum. However, this interindividual physiological variation is not well characterized. Here we quantified responses to a 16°C temperature difference in six physiological traits across nine thermally distinct Fundulus heteroclitus populations. These traits included whole animal metabolism (WAM), critical thermal maximum (CTmax), and substrate-specific cardiac metabolism measured in approximately 350 individuals. Traits exhibit high variation among both individuals and populations. Thermal sensitivity (Q10) was determined, specifically as the acclimated Q10, in which individuals were both acclimated and assayed at each temperature. The interindividual variation in Q10, is unexpectedly large: ranging from 0.6 to 5.4 for WAM. Thus, with a 16°C difference, metabolic rates are unchanged in some individuals, while in others are 15-fold higher. Furthermore, a significant portion of variation is related to habitat temperature. Warmer populations have a significantly lower Q10 for WAM and CTmax after acclimation. These data suggest that individual variation in thermal sensitivity reflects different physiological strategies to respond to temperature variation, providing many different adaptive responses to changing environments.
... In two latitudinally distinct populations of lake sturgeon, thermal history during early rearing and population source had a significant effect on growth and survival with the more southern population demonstrating a higher critical thermal maximum than the northern population at warmer (24°C vs 16°C) rearing temperatures. However, the northern population demonstrated an increased mass over time in comparison to the southern population at 16 and 20°C which may indicate latitudinal countergradient variation with northern fish growing faster at those temperatures to take advantage of a shorter growing season (Fangue et al., 2009). Population and rearing temperature also had a significant effect on cold adaptation in lake sturgeon with the southern population mounting a more vigorous cellular stress response compared to their northern counterparts when acutely exposed to a cold shock (Bugg et al., 2021b). ...
Sturgeon are found circumglobally throughout the northern hemisphere. Longevity, late age to sexual maturation and infrequent spawning are typical life history strategies that have supported their evolution over millennia but at the same time have made them vulnerable to extirpation or extinction when population sizes have been reduced as a result of overharvest, habitat degradation and/or pollution. As a consequence sturgeons are among the most at risk species as listed by the International Union for the Conservation of Nature. Aquaculture has been used as a conservation strategy to arrest declines or maintain current population levels for several sturgeon species. Here we describe the development of this conservation strategy in sturgeons, examining the importance of understanding genotype and environment in development of appropriate rearing strategies toward improved conservation practice for this imperiled group of fishes. Recent research indicates a significant effect of environment on phenotype, with early life history being particularly relevant; evidence suggests that repatriation should be the preferred approach where possible. We suggest that while aquaculture can be a valuable conservation and recovery tool, it is particularly effective when the impacts on phenotypic development during early life history are considered and combined with effective post-release monitoring and implementation of additional restoration efforts.
Thermal performance curves (TPCs) provide a framework for understanding the effects of temperature on ectotherm performance and fitness. TPCs are often used to test hypotheses regarding local adaptation to temperature or to develop predictions for how organisms will respond to climate warming. However, for aquatic organisms such as fishes, most TPCs have been estimated for adult life stages, and little is known about the shape of TPCs or the potential for thermal adaptation at sensitive embryonic life stages. To examine how latitudinal gradients shape TPCs at early life stages in fishes, we used two populations of Fundulus heteroclitus that have been shown to exhibit latitudinal variation along the thermal cline as adults. We exposed embryos from both northern and southern populations and their reciprocal crosses to eight different temperatures (15°C, 18°C, 21°C, 24°C, 27°C, 30°C, 33°C, and 36°C) until hatch and examined the effects of developmental temperature on embryonic and larval traits (shape of TPCs, heart rate, and body size). We found that the pure southern embryos had a right-shifted TPC (higher thermal optimum (Topt) for developmental rate, survival, and embryonic growth rate) whereas pure northern embryos had a vertically shifted TPC (higher maximum performance (Pmax) for developmental rate). Differences across larval traits and cross-type were also found, such that northern crosses hatched faster and hatched at a smaller size compared to the pure southern population. Overall, these observed differences in embryonic and larval traits are consistent with patterns of both local adaptation and countergradient variation.
Preferred temperatures of food-deprived (>24 hours post feeding) and fed (<2 hours post feeding) female marine threespine sticklebacks, Gasterosteus aculeatus, from the Bodega Harbor, California, population were determined using a laboratory, horizontal temperature gradient apparatus. Food-deprived fish preferred significantly cooler temperatures (mean = 15.9°C, final = 15.8°C) than fed fish (mean = 20.1°C, final = 19.0°C), with larger specimens preferring lower temperatures than smaller specimens. Food-deprived fish would probably benefit from this behavior by conserving energy (approximately 17% decrease in metabolic requirements at the lower temperature), whereas fed fish would probably benefit by decreasing gut transit times, facilitating energy acquisition.
We compare the thermoregulatory behavior of ornate box turtles (Terrapene ornata) from a northern population in Wisconsin with published information about a southern population in Kansas. In a laboratory thermal gradient, body temperature ( T b) of Wisconsin turtles (23.5 ± 1.3 C SE) averaged 6.3 C lower than reported for the same species from Kansas. T b of free-living turtles in Wisconsin (measured with temperature-sensitive transmitters) averaged 4.0 C lower than T b of Kansas turtles when the turtles were basking, moving, or seeking thermal protection in burrows from high temperatures. Using field measures of operative environmental temperature ( T e; measured with hollow copper models), we deduce (1) that the lower field T b of Wisconsin turtles reflects active selection of lower T b and is not simply a passive result of a lower environmental heat load; and (2) that behavioral selection by Wisconsin turtles of a lower and broader range of T b increases potential daily activity time.
Both the number of oil droplets within the egg and the external morphology of the chorion of Fundulus heteroclitus were found to vary extensively among populations along the US east coast. The number of oil droplets per egg varied clinally from a mean of 12.7 at Mount Desert Island, ME to 172.3 at Jacksonville, FL for individuals maintained in the laboratory under identical conditions. The pattern of variation in the external morphology of the chorion, as established by scanning electron microscopy, was that of a step cline. The chorion of eggs from Mount Desert Island, ME to Newark Bay, NJ was characterized by a surface sculptured with chorionic papillae approximately 0.5 μ in diameter. The chorionic filaments were smooth and conically modified at the base. Over the same area, the filament diameters ranged from 0.8-2.2 μ and the filament density ranged from 0-5 per 10,000 μ 2 of chorion surface area. In striking contrast, the chorion surface of eggs from Sandy Hook, NJ to Jacksonville, FL lacked papillae, but had granular, pebble-surfaced filaments attached directly to the chorion with no conical modifications at the base. Filament diameters ranged from 0.7-0.8 μ and filament densities ranged from 150-220 per 10,000 μ 2 among these southern populations. Samples from a location intermediate between Newark Bay and Sandy Hook, NJ had an intermediate filament density of 82 per 10,000 μ 2. This pattern of geographic variation in oil droplet number and chorion morphology was repeated within Chesapeake Bay and Delaware Bay. Within Chesapeake Bay, oil droplets varied clinally from 31.5 per egg at an upper bay locality to 102.0 per egg at a lower bay locality. Within Delaware Bay, oil droplets varied from 46.9 per egg at an upper bay locality to 102.7 per egg at a lower bay locality. The chorionic filament diameter and density at the most southern Chesapeake Bay locality was 0.6 μ and 200-210 per 10,000 μ 2 while the filament diameter and density from the most northern locality was 1.9-2.4 μ and 0-2 per 10,000 μ 2. Within Delaware Bay, the filament diameter and density ranged from 0.7 μ and 190 per 10,000 μ 2 at the lower bay locality to 1.9 μ and 0-2 per 10,000 μ 2 at the upper bay locality. On the basis of these characters, upper bay eggs were indistinguishable from eggs of coastal populations north of Newark Bay, NJ while lower bay eggs were indistinguishable from eggs of coastal populations south of Sandy Hook, NJ. We suggest that the present distribution of F. heteroclitus populations of different egg types may reflect the consequences of secondary intergradation, with zones of intergradation at northern New Jersey and within Chesapeake and Delaware bays.
We demonstrate the presence of significant differentiation in development rate, adult body length, and stomatic growth rate in the estuarine harpacticoid copepod Scottolana canadensis (Willey) collected from a broad range of latitudes (27@?-43@? N) and reared in the laboratory for several generations under the same conditions (15 g salts per 1000 g seawater, and 15@?, 20@?, 25@?, or 28@?C). The changing pattern of differential growth with increased temperature suggests local adaptation to maximize scope for growth under prevailing temperatures conditions; northern-derived individuals grew faster at low but not at high temperatures.
Hochachka P.W., Somero G.N. (2002) Biochemical adaptation: mechanism and process in physiological evolution. New York: Oxford University Press. 466 p.
1. Experiments were designed to reveal geographic differentiation, and its evolutionary significance, in early life-history characters of Fundulus heteroclitus. The inherited capacity for growth was studied for five populations, from South Carolina to Maine, USA, at three temperatures representative of field conditions. Two generations of newly hatched fish were reared and tested. The second generation of fish was produced from parents that were themselves reared in the laboratory, in order to control for maternally transmitted environmental effects. 2. Over the first three weeks post-hatch, individuals from northern populations grew more rapidly than those from southern populations at higher temperatures (21 and 28 degrees C) whereas there were generally no differences among populations at the lowest temperature (17 degrees C). 3. The results do not conform to thermal adaptation models, in which northern populations are expected to show higher growth rates than southern populations at cold temperatures, and vice versa at higher temperatures. Instead, the higher growth rate of northern fish is interpreted as an adaptation to a short growing season. This represents an example of countergradient variation, in that genetic influences favouring rapid growth at high latitudes oppose the environmental effect of a shorter growing season. The dine of growth rates is concordant with morphological, behavioural and genetic contrasts between subspecies of Fundulus heteroclitus.
Fishes exhibit a perplexing variety of temperature-preference relationships, ranging from positive to negative functions of acclimation temperature. However, these relationships tend to be consistent within species even across broad geographic ranges. We developed and tested an hypothesis that temperature-preference relationships are related to the amplitude of the annual thermal-cycle experienced by species during their recent evolutionary histories. Species experiencing annual cycles of relatively high amplitude (temperate species) were predicted to exhibit temperature-preference relationships that are positive functions of acclimation temperature. Those with low thermal amplitudes, such as cold stenotherms, or relatively high short-term amplitudes, such as tropical mesotherms, were predicted to exhibit relationships that are either independent or negative functions of acclimation temperature. This hypothesis was tested and strongly supported by a synthesis of studies from the literature. Of 42 species, for which data were available, 40 exhibited the class of temperature-preference relationships that we predicted on the basis of their thermal cycle. The concept of thermal guilds can be expanded by taking into consideration the capacity of fishes to adjust preferred temperatures through acclimation. This work shows the adaptive nature of temperature-preference behavior and may be useful for predicting general characteristics of temperature-preference relationships for the many species not yet examined in this regard.
In ectotherms, lower mean temperatures and shorter growing seasons at higher latitudes would be expected to cause a reduction in the annual growth rate of an individual. If slower growth reduces fitness, then organisms at higher latitudes may evolve compensatory responses for these climatic effects. Two such forms of local adaptation with increasing latitude are possible: (1) the capacity for growth may shift to a lower range of temperatures (i.e., temperature adaptation) or (2) maximum growth rate may evolve inversely with length of the growing season (i.e., countergradient variation). A third alternative is a mixed strategy involving both of the above. We hypothesized that the form of local adaptation may be affected by constraints that vary within vs. among species. We used common-environment experiments to compare reaction norms for growth in response to temperature among local population., of two contiguous, closely related fish species, the Atlantic silverside, Menidia menidia (L.), and the tidewater silverside, M. peninsulae (Goode and Bean), which together ha,e a range spanning much of the North American Atlantic coast. The common-environment experiments revealed countergradient variation: maximum growth rate increased with latitude both within and among species. However. growth reaction norms of the northern species were shifted to a lower range of temperatures than those of the southern species, indicating adaptation to temperature at the interspecific level. Hence, adaptation to temperature contributes to the interspecific variation, while countergradient variation contributes to both the intra- and interspecific differences.