Adaptation in Metapopulations: How Interaction Changes Evolution
270
Adaptation in Metapopulations: How Interaction Changes Evolution
270eBook
Available on Compatible NOOK devices, the free NOOK App and in My Digital Library.
Related collections and offers
Overview
With Adaptation in Metapopulations, Michael J. Wade explores a key component of this new understanding of evolution: interaction. Synthesizing decades of work in the lab and in the field in a book both empirically grounded and underpinned by a strong conceptual framework, Wade looks at the role of interaction across scales from gene selection to selection at the level of individuals, kin, and groups. In so doing, he integrates molecular and organismal biology to reveal the true complexities of evolutionary dynamics from genes to metapopulations.
Product Details
| ISBN-13: | 9780226129877 |
|---|---|
| Publisher: | University of Chicago Press |
| Publication date: | 05/03/2016 |
| Series: | Interspecific Interactions |
| Sold by: | Barnes & Noble |
| Format: | eBook |
| Pages: | 270 |
| File size: | 4 MB |
About the Author
Read an Excerpt
Adaptation in Metapopulations
How Interaction Changes Evolution
By Michael John Wade
The University of Chicago Press
Copyright © 2016 The University of ChicagoAll rights reserved.
ISBN: 978-0-226-12987-7
CHAPTER 1
Introduction
The central question guiding my research throughout my career has been this: How is the process of adaptation different if the members of a population live clustered in small groups instead of being homogenously distributed like grass on a lawn? The field is called "evolution in subdivided populations" or "adaptation in metapopulations." It has led me to investigate a diverse array of topics, including group selection, family selection, kin selection, and sexual selection, as well as speciation genetics, maternal and paternal genetic effects, and host-symbiont coevolution. In my lab, my students and I have approached these topics using a combination of theoretical, field, and laboratory studies and a diversity of living systems ranging from our laboratory model of flour beetles in the genus Tribolium to other animals, plants, and microbes. Through the generosity of the National Science Foundation Opportunities for Promoting Understanding through Synthesis (OPUS) Program, the Sabbatical Scholars Program at the National Evolutionary Synthesis Center, and the Sabbatical Leave Program at Indiana University, I have had the opportunity to write a conceptual, historical synthesis of the findings from these studies and the relationship between adaptation in metapopulations and broader questions in evolutionary genetics.
This first chapter is an overview of and introduction to concepts considered in greater depth in later chapters. I first provide a bit of personal background about my family and my schooling, since both influenced my career and shaped my research interests.
The organization of later chapters follows the flow from questions and concepts to laboratory experiments and to field studies. The majority of my mathematical models developed from the physical activity of moving flour beetles within and among families (kin selection) or within and among populations (group selection) and the synthetic activity of analyzing data. The dates on some theoretical publications precede those of the experimental works that inspired them, but only because the design, execution, and analysis of experiments is much slower (and generally more tedious) than working out the results of mathematical models.
New data are included in later chapters; data from experiments that went unpublished, owing to interruptions from teaching, administrative tasks, or lapses in funding, as well as to the dispersion of young collaborators away from my lab to career opportunities elsewhere. Published or not, results from these studies became part of lab lore and influenced our thinking and the planning of subsequent research.
Personal Background
In 1949, I was born in Evanston, Illinois, the oldest of eight children, into a fundamentalist, Irish Catholic family. I spent most of my waking hours in the swamps near our home in Westport, Connecticut, collecting frogs, toads, turtles, snakes, and butterflies at Lee's Pond, Willow Brook Cemetery, and the banks of the Saugatuck River. My parents encouraged our interest in nature by allowing us to keep anything we could catch and my dad built elaborate circus wagons, wheeled cages, for moving our menagerie in and out of the garage. My childhood ambition was to become not a scientist, but rather curator of reptiles at the Bronx Zoo, inspired by Marlin Perkins and his show, The Wild Kingdom.
Although neither of my parents had completed college, my mother emphasized our education above all else. (Long after we left home, my mom returned to college and completed her degree at St. Joseph's University of Pennsylvania in 1983 at the age of 61.) And, although she did not believe in evolution herself, she introduced us to it by reading us "dinosaur books" at bedtime. Each summer, we took family trips to Yale's Peabody Museum, where we spent our allowances on rubber dinosaurs and the army men to fight them.
In 1963, my father was appointed director of research for the Triangle Broadcast Center and we moved from Connecticut to Drexel Hill, Pennsylvania, a Catholic enclave near Philadelphia. I attended St. Joseph's Preparatory High School and my endless hours in the swamps were replaced by 5–6 hours of homework each night; a test every week in every subject; and daily quizzes in Latin and math. The Prep faculty had two of my best teachers, Mr. Earl Hart in honors mathematics and Stephen A. Garber, S. J., in honors chemistry. In 1967, I entered another Jesuit institution, Boston College, on scholarship. College was easy compared to high school and a year of advanced placement credit allowed me the freedom to experiment with majors in chemistry, English, sociology, mathematics, and biology. I spent my senior year writing plays, oil painting, and learning Boolean algebra as a "Scholar of the College" before graduating in 1971 with a double major in mathematics and biology. Somewhat unfocused, I applied to law school as well as to graduate schools in anthropology, linguistics, and evolution and ecology. With guidance from my biophysics professor, Dr. Donald J. Plocke, S. J., I applied to and was accepted by the Department of Theoretical Biology at the University of Chicago, where I was supported by a four-year National Institutes of Health (NIH) Graduate Training Fellowship (1971–1975). It was amazing to get paid to go to school, especially to a doctoral program uniquely suited to my combined interests in biology and math. The Chicago faculty applied mathematical models to everything from development to neuroscience to biological clocks. My background in natural history caused me to gravitate toward ecology and evolution. My dissertation was coadvised by Drs. Thomas Park, founder of the field of laboratory ecology and soon to be professor emeritus, and Montgomery Slatkin, a biomathematician and beginning assistant professor. Both were challenging and supportive mentors who guided my earliest ventures into group selection and the evolutionary genetics of metapopulations (chapters 3 and 4).
Interactions and Context
In graduate school, I learned for the first time that Darwinian evolution required variation, replication, and heredity and that any system whose units had those properties could evolve (Lewontin 1970b; later, Maynard Smith 1976) and that selection could operate simultaneously at more than one level and in more than one direction. This made it particularly difficult to determine a priori whether selection at one level was more efficient in producing evolutionary change than selection acting at another. Quantifying the relative efficacy of multilevel selection required an understanding of selection and heredity at each level. The goal of my research was to develop methods for this kind of multilevel quantification and to test the relative efficacy of the levels of selection.
In my graduate classes, single genes were the primary conveyors of heredity and evolutionary change equaled change in gene frequency. Interactions between genes or between genes and the environment were largely ignored. Individuals could be broken up into their component parts, genes and the environment. Reciprocally, these parts could be summed back together to re-create the individual. In a world without interactions, this one-gene-at-a-time evolution made a great deal of sense. But it also implied that the fitness differences at lower levels in the biological hierarchy determined those at the higher levels, because higher level units were merely aggregates of lower level entities. This view that groups were nothing more than the sum of their parts made the levels-of-selection problem a fairly trivial one and it eliminated group selection as a viable adaptive process. It also favored reductionism as the primary research methodology in evolution. Understanding the essence of complex living systems was equivalent to understanding processes at the lowest level of the biological hierarchy; the more you knew about the lowest level, the more you knew about everything else. Of course, my instructors in genetics, biochemistry, and biophysics believed the same thing, with particle physicists winning this race to the bottom of the hierarchy.
My instructors and textbooks asserted that Darwinian selection among individuals, and not among cells, groups, populations, species, or communities, was the strongest and most efficient kind of selection. Natural selection produced adaptations, which were responsible for the fit between an organism's morphology, physiology, and behavior and its environment. Natural selection was privileged in the constellation of evolutionary forces (which included mutation, migration, and random genetic drift) because of its crucial role in adaptation. Kimura's Neutral Theory of Evolution (1968) and the famous paper "Non-Darwinian Evolution," by King and Jukes (1969), both challenged that primacy. These papers were discussed, dissected and found wanting over and over again in lectures and in graduate seminars during my doctoral training. Absent critical evidence, it was relatively easy to tell an adaptive story about almost anything. With data, however, it was much harder to detect the expected signal of natural selection through the noise of mutation and random genetic drift.
War and Poker
To illustrate adaptive evolution as we were taught it, consider the card games war and poker. The received view was that Nature and the evolutionary processes that produced her are much more like the game of war than they are like poker. In war, high cards win and low cards lose, just as genes good for fitness increase in frequency by natural selection and those bad for fitness decrease. In war, good hands have high cards and poor hands have low cards. The quality of a hand equals the sum of its cards. To see the parallel with evolution, pretend that a hand is an individual. The fitness of an individual equals the sum of the effects of its genes and the environments it has experienced. The adaptive quality of an individual can be assessed from the sum of the fitness effects of its genes, just as a hand in war can be assessed from the sum of its high cards. The fittest individuals, favored by natural selection, are those with the best genes; the winner in war is the person with the best hand.
In contrast, if Nature were more like a game of poker, then individuals would not be the sum of their parts. A card in a hand of poker can be labeled good or bad only in the context of the other cards in the hand, the people playing those hands, and often the order in which the hands are dealt and played. An ace is a good card in some hands but it can be a bad card in others. In fact, an ace of one suit can be more valuable than an ace of another suit in the context of a flush. The same hand might win in one game but lose in another; win when held by one player but lose when played by another; or win early in a series of games, but lose later. Context is irrelevant in war, but it is the essence of poker.
A theory of gambling based on unchanging card values would work very well in describing war but poorly or not at all in describing poker. Poker games have emergent properties because interactions between cards, between hands, and between players confer importance to contextual variations in the value of a card. In war, unlike poker, there are no emergent properties. War lacks interactions; it is essentially a context-free game. Where war is mind-numbingly boring, poker can be endlessly fascinating, almost as fascinating as Nature.
Because higher levels of biological organization have emergent properties that are lacking at lower levels, Nature is more similar to poker than it is to war. The genes causing complex human genetic diseases are harder to find and map because genes behave more like the cards in poker than in war. A gene can be good in one background or environmental context, but harmful in another, so the same gene may appear both in healthy and in sick individuals. If upper levels in the biological hierarchy are not simple aggregates of the lower level entities, then selection at a higher level can be stronger than or more efficacious than selection at a lower level. As Feldman et al. (1983, p. 1009) concluded from their review of models of fertility selection based on interactions between mating males and females, "the simplest interactions between individuals in the process of selection can produce evolutionary conclusions not expected from individual fitness models." Unfortunately, we were taught to analyze Nature and adaptation using a genetical theory of war without interactions, not a theory of poker.
Multilevel Selection Theory in the 1970s
Multilevel selection theory was in its infancy when I began graduate school. The year before I started at Chicago, Dr. R. Levins (1970) defined a metapopulation as a population of populations. He emphasized that the processes of extinction and colonization of local populations (technically called demes) within a metapopulation were analogous to the births and deaths of individuals within a population. In his articles and graduate lectures, Levins emphasized that natural selection occurred within as well as among populations and that a theoretical and experimental framework capable of handling both kinds of selection at the same time was needed. It was the absence of a multilevel selection theory that caught my attention as a graduate student.
Also in 1970, Levins' colleague Dr. R. C. Lewontin (1970b) published an influential review entitled "The Units of Selection." In class, Lewontin lectured about levels of selection below the individual, particularly gametic selection (also known as meiotic drive), where one allele is favored over another when heterozygous individuals form sperm or eggs. A particularly interesting case occurred when males produced more Y-bearing sperm than X- bearing sperm: they produced primarily sons. Such a population could run out of females and go extinct. Lewontin stressed that group selection could be a means of limiting such gametic selection because populations with genetic tendencies to overproduce either kind of sperm would go extinct, while populations without such a tendency would persist. The importance of this example was unmistakable because it appeared on our written doctoral preliminary exam as a problem. Other than the genetics of gametic selection or segregation distortion, however, multilevel selection was rarely discussed in class.
The first class to allow me to develop a research project was Park's Field Ecology course, which was based in the vernal ponds of Chicago's Tinley Creek Forest Preserve. The array of ponds there seemed a natural example of a metapopulation. But my research did not become focused on evolution in metapopulations until my second year of graduate school, when I made the transition from field to laboratory work, using Tribolium, Park's flour beetle, as a model system — a transition I discuss in chapter 3.
Interactions and Metapopulations
The standard view of evolutionary theory misses the important fact that genetic adaptation to the internal environment of other genes occurs more rapidly than genetic adaptation to the external environment (Drown and Wade 2014). The "fit" between organisms and the external environment, which motivates the standard theory, is visually striking and can be seen by simply taking a walk in the woods. However, the fit between organism and external environment is no more striking than the complexity of the regulatory "fit" between genes that control development from fertilized egg to adult. ENCODE (Encyclopedia of DNA Elements) has found that, although the human genome consists of a mere 24,000 genes, it has 4,000,000 regulatory regions, which determine where and when each gene is expressed. This is more than 150 regulatory elements per gene. Unlike the external environment, the internal environment of other genes can coevolve, so that a good gene fit on a particular genetic background drags that internal environment along with it to the next generation. In this way, not only do the best genes become more common, but they also make the genetic environments in which they excel more common (Drown and Wade 2014).
Multilevel selection theory offers a more comprehensive account of the adaptive process for complex traits than does standard theory for species whose members are distributed spatially as many, more or less isolated, small populations. It allows us to understand macroevolutionary processes, like the origins of biodiversity and species richness, as the outcomes of microevolutionary processes involving interactions between genes, between genes and environments, between genes in different individuals (like mothers and offspring) and in different species. It is my conviction that the inclusion of interactions of all sorts into evolutionary theory is essential for uniting micro- and macroevolution in an understanding of Nature's enormous diversity of species.
The existence of interactions expands the domain of multilevel selection and, at the same time, constrains individual selection. To return to the card game analogy, the existence of interactions in poker expands the range of strategies and constrains the value of the "high cards only" strategy learned in war. Just as artificial selection helped Darwin think about natural selection acting on individuals, I find artificial group selection useful for thinking about multilevel selection. Consider a trait that affects both individuals and the groups to which they belong, such as the relationship between leaf area and seed yield in plants. At the individual level, larger plants with greater leaf area produce more seeds than smaller plants. At the group level, however, a large plant grows at the expense of its neighbors, shading them with its large leaves, stealing their water and nutrients with its large root system, thereby reducing their seed yield. The best individuals harm their neighbors, reducing average yield of the group. It is this kind of trade-off between an individual and its social partners that decades ago led plant breeders interested in maximizing yield to abandon simple individual selection in favor of group or stand selection. This same trade-off is currently leading animal breeders to adopt group selection as a means of maximizing yield while improving animal welfare at the same time, by breeding kinder, gentler chickens, pigs, and cows (see below). These examples from agriculture and animal breeding are representative of competition between individuals for scarce resources in Nature. This is the kind of competition that lies at the heart of Darwin's evolutionary logic, yet, in experimental systems, the genes affecting group productivity do not respond well or at all to individual selection.
(Continues...)
Excerpted from Adaptation in Metapopulations by Michael John Wade. Copyright © 2016 The University of Chicago. Excerpted by permission of The University of Chicago Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
Table of Contents
1 Introduction2 What Is Group Selection?
3 Group Selection in the 1970s
4 Career Beginnings and Science after the Thesis
5 Experimental Studies of Population Heritability
6 Population Ecology and Population Heritability
7 The Evolution of Sociality
8 Calibrating the Laboratory to Nature
9 Experimental Studies of Wright’s Shifting Balance Theory
10 Beyond the Shifting Balancing Theory
Acknowledgments
Reference List
Index