A 1950s CLASSIC OF THERMAL ADAPTATION TO COLD

Abstract

It is unusual to re-visit a paper a half-century after first reading it, as I have done with the classic publication of Per Scholander, Raymond Hock, Vladimir Walters and Lawrence Irving published in The Biological Bulletin in 1950 (Scholander et al., 1950). My first reading of this paper, and others by Scholander and colleagues on temperature adaptation in aquatic and terrestrial ectotherms (Scholander et al., 1953; Scholander et al., 1957), occurred around 1962, early in my doctoral studies when I trekked to Antarctica to study cold adaptation in fish. These papers provided much of the conceptual and empirical foundation for the work I anticipated doing with notothenioid fishes. My interests in polar birds and mammals at that time were largely confined to their roles as photographic subjects. However, my reading of the 1950 Biological Bulletin paper turned out to be highly important in broadening my scientific horizons and helping me to formulate a perspective on evolutionary adaptation to temperature that spanned the full range of biological (body) temperatures and the adaptive patterns observed among all taxa, whether endothermic or ectothermic. A fresh reading of this paper by scientists of my generation and, even more, a first reading by scientists whose parents may not have been born when the paper first appeared, still has much to offer in teaching us about thermal biology and how scientific research in comparative and evolutionary physiology has evolved.The studies of thermal biology by Scholander and colleagues that were initiated shortly after the end of World War II and conducted with major logistical support from the US military helped in important ways to define the evolutionary strategies used by endothermic homeotherms – mammals and birds – to cope with differences in ambient temperature. Importantly, the focus of these initial studies was on evolutionary adaptation, not phenotypic acclimation or acclimatization to cold, phenomena that later took center stage in analysis of thermal biology in endotherms. The 1950 paper was one in a series from this group that elucidated the roles of what they termed three possible ‘avenues for cold adaptation’ in mammals and birds: variations in the body-to-air thermal gradient, insulation level and basal metabolic rate. In other words, how were heat production and heat exchange altered to allow mammals and birds to maintain a stable and high body temperature? The previous papers by these authors, plus earlier literature cited in the 1950 paper, led to the tentative conclusion that adaptive changes in body temperature to reduce the body-to-air thermal gradient and, thereby, reduce heat flux between animal and environment, are unlikely to play a role in cold adaptation in active adult mammals and birds. To quote the 1950 paper, ‘…there are no signs so far that body temperature of mammals and birds is adaptive to the different climates on earth.’ This well-buttressed conclusion leads to another – and yet unanswered – set of questions: What is the basis for selection of temperatures near 37–40°C for avian and mammalian core body temperatures? Is there something special about this particular range of temperatures? If so, what is it?The remaining two ‘avenues for cold adaptation’ had also been partially explored in earlier studies by Scholander and colleagues. It was clear from their work and that of others that insulation played a critical role. As stated in the 1950 paper, ‘…we established, by insulation measurements, the general rule that arctic mammals have warmer furs than do tropical mammals.’ However, this adaptive mechanism, while of importance, could not at that time be regarded as offering the full story. The final ‘avenue for cold adaptation’, the role of evolved temperature-adaptive differences in basal metabolic rate, remained a possible mechanism of cold adaptation, with animals from cold polar environments possibly exhibiting higher metabolic rates per unit mass than animals from warmer temperate locations. Most existing basal metabolic rate data were from studies on temperate species; there was not an adequate diversity of environments represented in these data sets to fully evaluate whether basal metabolic rates reflected evolutionary thermal history. To resolve this issue, it was necessary to conduct additional studies of mammals and birds from thermal environments that were as different as possible in terms of maximal temperature. Thus the dual focus – Arctic versus tropical – of the studies presented in the 1950 paper. The metabolic rate measurements made with endemic species from Point Barrow, Alaska, and Barro Colorado Island in the Central American tropics essentially put the final nail in the coffin of this potential mechanism of adaptation to cold. The metabolic rate measurements presented in the 1950 paper were made with animals spanning a wide range of sizes and rate measurements were made as closely as possible under ‘resting’ or ‘basal’ conditions and in the absence of acute cold stress. Appropriately, the data gathered by Scholander et al. were analyzed in the context of Francis Benedict's famous ‘mouse to elephant’ relationship of metabolic scaling (Benedict, 1938). The results of these experiments seemed unambiguous: ‘…we may state as a tentative generalization that the basal metabolic rate of terrestrial mammals from tropics to arctic is fundamentally determined by a size relation according to the formula Cal./day = 70 kg3/4 and is phylogenetically nonadaptive to external temperature conditions. Equally nonadaptive is the body temperature, and the phylogenetic adaptation to cold therefore rests entirely upon the plasticity of the factors which determine the heat loss, mainly the fur insulation.’ Thus, natural selection seemed not to have packed more metabolic capacity into a gram of an arctic mammal than into a gram of a similar-sized tropical mammal. At least this seemed to be the case when analyses involved normothermic adult animals, such as those used by Scholander et al. However, as the caveat in the above quotation (‘tentative generalization’) seems to hint, later work showed that there was more to the story.As is the case of many foundational papers, observations and comments made in the 1950 paper seem prescient when examined in the context of what has subsequently been discovered. One such comment is the brief mention that a small mammal, the arctic weasel, showed an extraordinarily high rate of heat production when exposed to cold; this species' metabolic rate was well above the regression line of the ‘mouse to elephant’ curve. Scholander et al. remarked that it seemed ‘somewhat odd’ that the weasel didn't simply increase its insulation when exposed to cold. This ‘oddity’ and similar observations on other small, cold-stressed mammals helped to pave the way for the discovery of the thermogenic role of brown adipose tissue (BAT) (Cannon and Nedergaard, 2004; Smith, 1961). The discovery of this specialized heat-generating tissue casts metabolic adaptation to cold by mammals (no avian equivalent of BAT is known) in a new light. The importance of heat generation by BAT is arguably the most significant extension of our understanding of thermal biology in mammals made since the 1950 paper was published. I'd like to think that the ‘oddity’ noted by Scholander and colleagues puzzled readers enough to catalyze a deeper look at non-basal metabolic capacities, namely those of BAT, the one type of thermogenic tissue known in mammals.A second easily missed point in the 1950 paper is one that is not explicitly indicated by the paper's title. In addition to addressing the adaptations required to withstand cold, Scholander et al. pay heed to the challenges faced by tropical mammals and birds. Thus they state, ‘It seems then that the problem for tropical mammals is neither overheating nor cooling, but, actually, both.’ Moreover, the authors emphasize (presciently, as things are turning out) that, ‘Many parts of the tropics are so hot and humid that a few degrees’ rise in the temperature would mean death for mammals and birds because they cannot adapt to it by raising their body temperatures.' My re-examination of this paper over 50 years since first reading it was done in concert with the appearance in 2010 of articles by Sherwood and Huber and McMichael and Dear that dealt with this threat to low latitude mammals and birds, a danger that would have seemed completely ‘academic’ back in 1950 when concerns about global warming – an expression that seems to have been introduced to the scientific literature only in 1975 in a classic paper by Wallace Broecker in Science (Broecker, 1975) – were still well off in the future (Sherwood and Huber, 2010; McMichael and Dear, 2010). In their analysis of the capacities of birds and mammals to cope with the combination of rising temperatures and humidity, these authors build on (but, regrettably, do not cite) the types of studies found in the 1950 classic by Scholander et al. The 2010 papers point out that, as the need for effective evaporative cooling becomes more critical in a warming world, capacities to evaporate water from the body surfaces becomes increasingly challenging. The basic physics of the situation shows that, when wet bulb temperatures (TW) exceed 35°C, evaporative dissipation of metabolic heat by mammals and birds ceases to be possible. How close are we to this situation of ‘melt down’, the exclusion of habitats from occupancy by endothermic homeotherms due to TW values >35°C? Currently, TW never exceeds 31°C, even in the hottest climates (Sherwood and Huber, 2010). However, TW is predicted to rise with global mean temperature at a ratio of 3°C for every 4°C of global warming. The models used to predict rates of global warming offer a range of estimates and are typically limited for extrapolating trends beyond the current century. However, as Sherwood and Huber emphasize, using models based on realistic assumptions about release of greenhouse gases and extending these models beyond 2100, within two or three centuries the temperatures of some regions may increase by ~7°C and, therefore, have combinations of heat and humidity that result in TW near 35°C. Because such a large fraction of the human population resides at low latitudes, Sherwood and Huber conjecture that global warming of 11–12°C would lead to intolerable values of TW in the regions where most of the human population is currently found (Sherwood and Huber, 2010). Moreover, these low latitude regions are zones where the greatest biodiversity is commonly found. The early studies of Scholander and colleagues help to put into sharp relief the challenges faced by endothermic homeotherms in confronting rising temperatures and humidity. Birds and mammals are ‘stuck’, evolutionarily, with high mass-specific metabolic (heat generation) rates. It seems inconceivable that they can evolve a physiological solution that would allow them to cope with the threats posed by TW values greater than 35°C.In summary, reading (or re-reading) this classic paper offers a number of rewards. First, in terms of its primary question about fundamental evolutionary strategies for cold adaptation, we can come to appreciate how the principal ‘avenue’ taken by birds and mammals for maintaining high and stable body temperatures regardless of ambient temperature was elucidated. Second, we find a nice example of how an experimental ‘oddity’ can lead to further work that, in this case, helped open up the broad field of BAT physiology. Third, and certainly without the intention of the authors, the paper has provided a valuable context for evaluating the challenges posed to endothermic homeotherms by global climate change. And, lastly, there are rewards from (re)visiting this work that deal less with specific scientific discoveries than with the ways in which doing science have evolved. These lessons seem especially valuable for younger scientists for whom the practice of reading literature that is more than a year or two old may not exist. In our (usually futile) obsession to stay on top of the current literature, the classics of the past – the formative publications that built the foundation for our field – are too often neglected or just downright forgotten entirely. As Sydney Brenner recently commented, ‘…most scientists are too busy working in the present and thinking anxiously about the future and have no time to view their work in the context of what has gone before’ (Brenner, 2012). Disregarding the past is a shame, for several reasons. One is that, by neglecting the foundational literature of one's field, one is apt to lack an appreciation of how the shaping principles (‘paradigms’) of one's discipline have originated and subsequently evolved. Another benefit of examining the older literature is to get a sense of how different it was to do science at a time when major questions were first being addressed experimentally. For comparative physiology, the first two or three decades after World War II were an era of both intellectual and geographical explorations of exciting new territory. These were halcyon days when opening up entirely new lines of study could be done with simple instruments, minimal needs for costly reagents, and, therefore, relatively tiny budgets (with the exception of travel costs for exploring distant and exotic lands). And, as the working and writing styles of early, intrepid explorers of comparative physiology such as Per Scholander and his collaborators, Knut Schmidt-Nielsen, and others suggest, doing research was probably a lot more fun and adventurous back then. Writing could be a bit more colorful too, compared with the prose found in the space-limited pages of contemporary journals. Can you think of a recent paper in which there occurs something equivalent to a statement that the sample size was low because the sled dogs ate one-third of the fish that were caught (Scholander et al., 1957)? In these incredibly busy times, it is worthwhile to pause, read some of the classics and contemplate what life was like back in the days when the ratio of novelty of discovery to grant dollars expended was arguably relatively high compared with the present. Classic papers such as those published by Scholander and colleagues over a half-century ago give their readers a clear sense of not only how the foundational discoveries of our field were made, but also of the adventure that was associated with these early intellectual and geographical explorations. Having been his faculty colleague for two decades, I think I can safely say that when Per Scholander chose the title for his autobiography, Enjoying a Life in Science (Scholander, 1990), he was at once giving an honest summary of a great career and challenging his readers to follow an example that, while increasingly tough to meet, should still be our goal.