Does selection for stress resistance lower metabolic rate?

Ecology, April, 1997 by Minou Djawdan, Michael R. Rose, Timothy J. Bradley

INTRODUCTION

Environments frequently undergo fluctuations that can place organisms under stress. Such conditions include exposure to toxins, disease, and adverse environmental conditions such as drought or cold. Even the actions of the organism itself, such as migration, reproduction, or hatching, can expose it to stress. Stress is thus a ubiquitous hazard, one that has recently come into greater prominence as an area of research (e.g., Hoffmann and Parsons 1991).

The phenomenon of stress is viewed somewhat differently in different ecological specializations. For evolutionary ecologists, stress is important because it influences the survival of organisms and thus helps determine fitness. Indeed, stress has been defined by Koehn and Bayne (1989) as "any environmental change that acts to reduce the fitness of an organism." Similarly, the differential survival of organisms in various habitats can influence the geographical range of a species, which is of interest to biogeographers. Physiological ecologists have been interested in determining the specific physiological mechanisms used by organisms to resist stress. A considerable literature in physiological ecology has arisen concerning both the specific mechanisms used for resisting stress and the differentiation of stress resistant mechanisms across the geographical range of a species. Examples include differential amino acid concentration during osmotic stress in the copepod Tigriopus californicus (Burton et al. 1979) and differences in desiccation resistance across a broad geographic range in Australian populations of Drosophila melanogaster and D. simulans (Parsons 1970, McKenzie and Parsons 1974).

Important recent developments in the study of stress have involved laboratory selection for increased stress resistance (Hoffmann and Parsons 1989b, Bennett et al. 1990, Rose et al. 1992). It has proven relatively easy to select populations for increased resistance to a variety of stresses, e.g., desiccation, starvation, and extreme temperatures. This success has led to a search for the mechanistic basis of stress resistance, particularly in Drosophila (e.g., Service 1987, Hoffmann and Parsons 1989a, b).

One proposed mechanism of stress resistance involves reduction in metabolic rate (Hoffmann and Parsons 1989a). The advantages of a reduction in metabolic rate are clear in organisms that exhibit hibernation or torpor, or in plants that are deciduous in response to drought or periods of cold (reviewed in Hoffmann and Parsons 1991). Reductions in metabolic rate may be of advantage under other circumstances as well, however, particularly in marginal habitats where food, energy, or water resources may be limiting. The hypothesis that a population's resistance to stress is correlated with a lower metabolic rate and that such resistance may be of ecological significance has been tested in many different types of organisms (McNab and Morrison 1963, Hinds and MacMillen 1985, Diehl et al. 1986, McNab 1986, Lighton and Bartholomew 1988).

Mechanistic aspects of the short-term evolution of stress resistance can also be pursued in laboratory selection experiments. Selection for increases in desiccation resistance in derivatives of isofemale lines of D. melanogaster was found to be associated with decreases in mass-specific metabolic rate in the selected lines compared to controls (Hoffmann and Parsons 1989a, b). These authors found that lines selected for desiccation resistance also showed increased resistance to starvation, heat, ethanol, acetic acid vapors, and gamma radiation. They suggested that the common factor leading to resistance to multiple stresses in these lines is a reduction in mass-specific metabolic rate. The hypothesis that reduced metabolic rate is a central physiological change, which promotes resistance to a variety of stresses, derives from studies involving Drosophila (Hoffmann and Parsons 1989a, b). In the course of our studies of aging, particularly the correlation of postponed aging with increased stress resistance in D. melanogaster (Service et al. 1985, Rose et al. 1990, 1992), we have generated a number of replicate lines selected for three distinct forms of stress. While Hoffmann and Parsons (1989a, b) used isofemale lines to obtain populations for selection, our lines have all been generated from outbred populations of D. melanogaster and have been maintained at large population sizes. Each selection treatment was applied to five independent replicate lines, such that five stress-selected and five control populations were produced. This replicated design was applied to populations that were selected for postponed aging, desiccation, or starvation. We are therefore able to test the metabolic responses of flies to the stresses for which they were selected, as well as other stresses. This provides a new approach to the question of the universality of a lowering of metabolic rate in resisting one and/or multiple stresses. With these questions in mind, we undertook the following study of the effects of stress selection on metabolic rate in D. melanogaster.