14

 

 ORIGINS  OF  EVOLUTIONARY  NOVELTY

 

We now understand that animal form is not so much a matter of the genes an animal has, but how they’re used during the process of development, of going from a single egg to the complete mature animal, the building of all of its body parts. So these genes are used in every animal, but the choreography is what differs. And it’s that different choreography between species that accounts for the endless forms of animals.          

                                                                                          S. Carroll

 

Apparently, most of evolutionary changes and phenotypic diversity in the animal kingdom result from modification of pre-existing traits. A large body of evidence, part of which will be reviewed in this chapter, indicates that evolutionary modification of animal phenotype is brought about by specific changes in developmental pathways. In a substantial number of cases it is demonstrated that signals for modifying developmental pathways originate in the central nervous system. Observational evidence on the occurrence of many evolutionary changes in nature, as well as evidence from induced evolutionary changes in experiments, shows that such changes occur within time frames that rule out  gradual evolution via accumulation of favorable mutations as a possible explanation. Such evolutionary changes, often of tremendous phenotypic magnitude, are not related to any specific changes in genes or gene regulatoy networks (GRNs), which functionally remain highly conserved across species and higher taxa. The fact that epigenetic manipulation of developmental pathways during ontogeny leads to development of alternative phenotypes, or even new phenotypes, also indicates that evolutionary changes in phenotype and the underlying developmental mechanisms are of epigenetic origin.

 

The Epigenetic Hypothesis of Evolution and its Predictions

By any objective consideration, the modern neoDarwinian theory has not succeeded in its long attempt to account for the mechanism of the emergence of the evolutionary change.

In Neural Control of Development (2004) I put forward an alternative epigenetic theory of heredity of metazoans. Almost all predictions of that theory have been validated and substantiated by adequate evidence presented in the first part of this book (chapters 1-7).

Substantial research I made on relevant observational and experimental evidence of evolutionary change, encouraged me to extend and apply the epigenetic theory of heredity in developing an epigenetic hypothesis of metazoan evolution.

Metazoan evolution is the observable result of changes occurring over time in the phenotype of animals. The elementary unit of the evolution of metazoans is the evolutionary change, which comprises both heritable genetic (gene mutations) changes at the molecular level, i.e. in proteins, and heritable epigenetic changes in the morphology, behavior and life history of animals.

My hypothesis of the epigenetic mechanism of evolution of metazoans is the following:

Evolution of metazoan morphology, behavior and life history, is result of heritable epigenetic changes in developmental pathways.

These epigenetic changes are essentially different from genetic changes (gene mutations) in an essential aspect. Genetic changes unavoidably and randomly arise from errors occurring in the process of DNA replication. In contrast, epigenetic changes in developmental pathways are anything but random. They arise as solutions to the problems of adaptation of the organism to the changed environment. These solutions result from the computational activity of neural circuits that perform the processing of the external/internal stimuli. This is suggested by two general facts:

Firstly, in distinction from gene mutations, which arise spontaneously and at a constant rate independently of the changes in environment, epigenetically determined changes (intragenerational developmental plasticity, transgenerational developmental plasticity, and evolutionary changes) depend on the environmental stimuli or cues. In distinction from gene mutations these changes are not rare events affecting isolated individuals in populations but, depending on the intensity  and nature of the stimuli, they may  simultaneously affect many, most or even all individuals in a population.

Secondly, epigenetic changes in developmental pathways are result of epigenetic changes in the nature of neural signals that activate developmental pathways or in the spatio-temporal pattern of secretion of the neural signals.

Thirdly, and in clear distinction from gene mutations that are mostly deleterious, epigenetic changes, as a rule, are adaptive.

In anticipation of the criticism on possible teleological implications of the concept I will remind the reader that it excludes any teleological agent in the meaning that while it is product of  neural computation it has nothing to do with the consciousness for it is a textbook knowledge that  most of neural computation is not related to consciousness.

The main predictions of the epigenetic theory of evolution developed in this work are:

1. Evolutionary changes in phenotype arise during the individual development as a result of epigenetic changes in developmental pathways occurring without, or independently of, changes in genes,

2. Evolutionary loss of phenotypic traits is epigenetically determined, and occurs without loss of genes.

3. Evolution of animal phenotype (morphology, physiology, behavior and life histories) is reversible.

4. Reversion to ancestral phenotypes is epigenetically determined and is not rerlated to reversion of any lost gene.

5. Sympatric speciation is possible because reproductive isolation can prezygotically evolve by neuro-cognitive mechanisms without geographic isolation of populations.

6. Metazoan speciation is an epigenetic phenomenon that takes place independently of the evolution of genes. Hence,

7. No correlation exists between the evolution and number of genes and the degree of complexity of metazoans.

The remaining part (chapters 14-20) of this book is devoted to verification and substantiation of  the above predictions of this epigenetic hypothesis of evolution.

The traditional taxonomic classification, determination of nodes in cladograms and cladogenesis in general, whose study has contributed so much to establishing relationships between taxa and elucidating patterns of evolution, is out of the scope of this work. My causal inquiry will overextend at the expense of these descriptive aspects of metazoan evolution. I will focus on the mechanisms of evolutionary change of morphological, behavioral and life history traits involving no changes in genes by tracing back the development of these traits as they occur in the course of ontogeny, from the early individual development to adulthood.

To the extent that the present knowledge allows it, I will consider  the evolution of metazoan morphology, behavior, and life histories in concrete examples, focusing on the changes in developmental mechanisms that made these changes possible. Description of examples of the phenotypic evolution will be followed by brief  explanations of the identified or presumptive mechanisms that induced the evolutionary changes.

There is no alternate way for an evolutionary change in metazoan morphology, behavior, or life history to arise but through a specific change in a developmental pathway. Hence, what controls changes in developmental pathways in metazoans also controls their evolution. Activation of a new/modified developmental pathways, in absence changes in genes, implies generation and utilization of new, epigenetic information. Epigenetic information enables metazans that with the same genes to evolve widely different morphologies.

For enabling the reader to himself assess the explanatory power of the neoDarwinian and epigenetic paradigms, I will use a comparative approach: the neoDarwinian rationalization of the evolutionary change will be followed by an epigenetic interpretation of the concrete examples of evolutionary change.

I will devote this chapter (chapter 14) to mechanisms of modification of existing structures, and chapters 15, 16, and 17 - to the loss/vestigialization of structures, evolutionary reversions, and the morphological novelties determined by the  neural crest, respectively.

 

The Nature of the Evolutionary Change  

 

An ideal world of a stable, unchanging environment would allow metazoans to almost perfectly adapt to the environment and establish a quasi-permanent equilibrium with it. However, our planet, far from a static entity, is a dynamical, ever-changing system. Hence, evolutionary pressure for evolving not only  intragenerational phenotypic plasticity but also inherited adaptations to changing conditions of living would have risen as soon as living systems emerged. Inherited phenotypic adaptations are known as evolutionary changes.

For a long time biologists have focused on the process of selection of evolutionary changes rather than on their origin and the mechanisms determining their emergence. Natural selection acts on, and even exists for the sake of, evolutionary changes. No selection will occur before the evolutionary change arises. In the beginning there is evolutionary change. Selection follows it:

 

All novel adaptive phenotypes must originate before they can be molded by selection, and they need not be altered under selection to be adaptive. (West-Eberhard, 2003b1)

 

The nature and origin of evolutionary changes in animal morphology never became the central object of the evolutionary inquiry and for a long time remained an almost terra incognita of evolutionary biology. The fact that the science of biology for such a long time focused on selection rather than on generation of evolutionary changes that have to be selected is anything but surprising. Only in the last few decades the substantial breakthroughs have been made in the molecular mechanisms of individual development, the ontogeny at a deep causal level, where all the evolutionary changes originate.

In their perpetual effort to heritably adapt to the changing environment, metazoans evolved numerous novelties [sensu West-Eberhard, “a trait that is new in composition or context of expression relative to established ancestral traits” (West-Eberhard, 2003f)] and modifications of their parts, traits and organs. Most of the new evolutionary adaptations arose in the form of modifications of existing structures. The prevailing opinion in the 20^th century biology has been that the majority of new adaptations involve “an already present feature” (Bock and Wahlert, 1998). The idea that most evolutionary novelties are modified versions of older structures (Campbell et al., 1999), still remains almost unchallenged. De novo adaptations are relatively rare phenomena in the living world.

By early 80es, Rudolf Raff and Thomas Kaufman expressed the revolutionary idea that evolution by DNA mutations “is largely uncoupled from morphological evolution” (Raff and Kaufman, 1983). Indeed, no example has ever been presented of a gene mutation leading to an adaptive novelty in metazoans and adequate solid evidence on adaptive evolutionary changes in existing structures of animals shows that they involve no changes in genes or genetic information (Matsuda 1987, West-Eberhard, 1989, 2003; Cabej, 1999a,b,c, 2001, 2004; Newman and Müller, 2000, etc. ), although gene mutations unavoidably occur and are selected over time. But if there is no genetic information involved in the emergence of evolutionary novelties, then the only remaining alternative is to assume that some form of epigenetic information (in the broad meaning of the word) is responsible for that change.

In the first part of this work I have presented adequate empirical evidence that the signals for  activating developmental pathways determining the normal development of animal tissues, organs, and morphology in general come from the central nervous system. In chapter 11 it was argued that intragenerational discrete changes in animal morphology, physiology, behavior, and life history also involve no changes in genes or genetic information but result from signal cascades starting in the nervous system in response to specific external/internal stimuli. The epigenetic information, generated by processing these stimuli in neural circuits, flows from the nervous system to the target organs/parts of the organism via specific signal cascades. In that same chapter I have also presented adequate observational and experimental evidence that inherited changes in morphology, i.e., transgenerational developmental plasticity, which, in principle, is indistinguishable from evolutionary changes, appear in the offspring of individuals that have experienced specific environmental (physical, chemical or social) stimuli and involve no changes in genes. These transgenerational changes as well are induced by signal cascades starting in the CNS in adaptive responses to external stimuli. In all the considered cases of transgenerational developmental plasticity, the CNS functions as generator of the de novo epigenetic information necessary for inducing inherited phenotypic changes.  

 

 

Interactions Organism-Environment in Evolution: the Causal Relationship

 

Reacting to external forces, agents or conditions is a universal property of all the material systems. These reactions are commonly considered to be effects of external factors on these systems. Effects of external agents on the anorganic system can be predicted by the properties of the agents.

Unlike anorganic sytems, the effect of an environmental agent or stimulus on a metazoan system cannot be predicted from the nature or quantity of external agents or stimuli: metazoan organisms of different species, often of the same species, can respond to the same environmental stimulus in different and sometimes opposite ways. This suggests that it is something intrinsic to the living metazoan that determines the phenotypic result. Metazoan systems react to external stimuli adaptively, surely unconsciously, but “to the best of their interest”.

To emphasize this essential distinction, let’s say that anorganic systems react to external forces and agents, whereas metazoans respond, i.e. react adaptively to external influences.

The adaptive response to external stimuli is a novel, essential and unique feature of living systems in general, unknown in anorganic systems. It tends to counteract, avoid, overcome, or compensate for, the harmful effects of the action of external agents. The considerable degree of the free will that metazoans, and living systems in general, exhibit in their adaptive reactions to environmental influences is the reason why their responses cannot be predicted by the nature or the intensity of the external influences.

For at least two centuries the causal basis of evolution has been a major topic of theoretical biology. While the role of environment in evolution has been recognized from all the parties involved in the debate, disagreements have always arisen, and still are, on the issue of the relative role of organisms and the environment in metazoan evolution. All the different views on the issue may roughly be grouped in two opposing classes. On the one hand, there is a school of thought holding that the environment is the driving force of the evolutionary change in living systems, and on the other, those believing that it is the living system itself, its inherent properties, that determine the evolutionary change, its nature and extent, with the environmental factors representing conditions that may favor or not the evolutionary change.

The issue is fundamentally related with the problem of causation in biology. The concept of the final cause (causa finalis from Latin causa - cause and finis – purpose) was introduced by Aristotle (384-322 BC) in conformity with his idea of the rule of order and finality in nature. He believed that the aim of biological studies is

 

not merely to discover facts (that things are so), but reveal causes (how and why they are so), and in particular reveal the final causes and the absence of chance in the works of nature. (Lloyd, 1999c)

 

Aristotle believed that causa finalis, the “end” of the change could be found within the end-product of the goal-oriented action of the causa finalis (Lloyd, 1999b) and that “natural objects have their “ends” within themselves” (Lloyd, 1999b). For Aristotle the purpose and cause of things is in things themselves. Accordingly, the end of the evolutionary change is the organism itself.

Here I will use the concept of the “end” of the evolutionary change in the teleonomical sense introduced by Mayr:

 

There is no conflict between causality and teleonomy, but that scientific biology has not found any evidence that would support teleology in the sense of various vitalistic or finalistic theories. (Mayr, 1961)

 

Aristotle’s concept implies that intrinsic, not external, causes determine the change. Darwin introduced the idea of the end of evolution as a product of natural selection, but even he was skeptical of the idea that it is the environment alone that determines evolutionary changes:

 

Naturalists continually refer to external conditions, such as climate, food, &c., as the only possible cause of variation. In one very limited sense, as we shall hereafter see, this may be true; but it is preposterous to attribute to mere external conditions, the structure, for instance, of the woodpecker, with its feet, tail, beak, and tongue, so admirably adapted to catch insects under the bark of trees. (Darwin, 1859a)

 

An important implication of the Aristotelian concept of causae finales is that if the study of the developmental mechanism of formation of a new phenotypic trait would lead to a complete knowledge of that mechanism it would automatically reveal the causa finalis. Numerous examples on the developmental plasticity reviewed by West Eberhard (2003) and others, as well as additional examples to be presented in this work, especially several cases of transgenerational (evolutionary) developmental plasticity, seem to prove that causa finalis is not an external cause but an information-generating process taking place within the animal organism:

 

The self-organized ontogenesis of brain structures constitutes a natural language, and all evolution had to do is use this language to write the particular text that defines us. (von der Malsburg, 2002b)

 

In his attempt to apply Aristotelian principles of proximal and final (ultimate) causes to the neoDarwinian doctrine, by 60es of the 20^th century Ernst Mayr came to the conclusion that the causal chains of any evolutionary change start in the distant past in the natural selection, which shaped all the genetic programs, whose change makes evolution of living beings possible.

Identification of causae finales in biology makes sense as far as they are helpful to understanding the underlying mechanisms of evolutionary change but not more. Natural selection has always been used, even by Darwin himself (see  Introduction of this work), ambiguously, sometimes implying the cause and sometimes the effect of the evolutionary change. In dealing with the problem of the causation and natural selection in evolution, however, it is necessary to avoid the ambiguity.

If natural selection will be conceived as a result of the struggle for life, of differential reproductive success, or as a result of the organism-environment interaction in a broad meaning of the word, by definition it cannot be a  cause, let alone the causa finalis. If we, after Mayr, will identify the causa finalis with the principle of natural selection, conceived as a cause, a principle which is external to the living organism, we would hardly add anything to our modest knowledge on the mechanisms of evolutionary change and to the Darwinian principle of natural selection. Any attempt to reconstruct the causal chain (the numerous untraceable steps of the action of natural selection over generations on the structure and functions of animals) of the evolutionary change is doomed to failure.

However, Mayr implies that the action of natural selection as a causal agent is materialized in the structure and functions of the animal. If this interpretation is correct, if the structure of metazoans embodies causes that acted on metazoans in the past, then developmental mechanisms that determine formation of these structures during the ontogeny may give us crucial clues on the mechanisms of evolutionary changes that occurred in the course of phylogeny. Hence, the study of developmental mechanisms could lead to identification of mechanisms that determined particular evolutionary changes, which beclouded in the evolutionary past, would otherwise be unidentifiable.

In any case, natural selection is a means for preserving the “useful” and eliminating the disadvantageous inherited changes. It determines the survival or extinction of a species but it tells nothing about the deeper cause, about “why” and  “how” the internally determined evolutionary change emerges. And the real problem in biology is why and how evolutionary changes are produced rather than how the changes are accumulated or eliminated under the action of natural selection. Each evolutionary change has its own specific causa finalis in the specific mechanisms, in developmental pathways that bring the change about during the ontogeny.

Using natural selection as an off the shelf answer for explaining any particular evolutionary change adds nothing to our still limited knowledge on the cause and mechanisms of evolution. Natural selection is not, and cannot be, a substitute for empirical identification of the mechanisms of evolutionary change. Restriction of the scientific inquiry to the action of natural selection, would prevent investigation of the underlying causes of the evolutionary changes, of the “raw material” on which natural selection has to act.

It is commonplace in modern biology to speak, sometimes in a Lyssenkoan style, of environmental stimuli (changes in photoperiod, temperature, humidity, social environment, intraspecific and interspecific competition, etc.) as being in “control” of, or even “regulating”, various functions of the organism. This concept implies recognition of the metazoan organism as a passive entity, destined to “obey” “instructions” from external agents. It implies that the environment is in possession, and capable of transmitting to metazoan organisms, of information on the morphological changes they have to accomplish in order to adapt to current changes in the environment. It is not difficult to show that metazoans are neither “regulated” nor “controlled” by environmental stimuli, by the environment in general. It is easy to prove the opposite, i.e. that metazoan organisms are not “changed by external environment” but rather they respond to changes in the environment, most of time adaptively and antientropically, by changing their behavior, physiology, and morphology at both the developmental and evolutionary levels. It is the living organism that generates these changes and it generates these changes epigenetically, not involving changes in genes; not only because changes in genes are inappropriate for inducing adaptive changes in morphology, behavior, and life history but “useful mutations” are extremely rare to help the survival of metazoans in times of survival emergency related to drastic changes in the environment.

In distinction from gene mutations, epigenetic changes are not random changes, but are adaptive responses to the changed environment. Solid evidence from nature and from experiments shows that living organisms are able to change, sometimes suddenly, developmental pathways and produce adaptive changes without changes in genes.

From the neoDarwinian point of view ultimate causes of evolutionary change have been selective processes that occurred in the past.

 

Where, then, is it legitimate to speak of purpose and purposiveness in nature, and where it is not? To this question we can now give a firm and unambiguous answer. An individual who – to use the language of the computer – has been “programmed” can act purposefully. Historical processes, however, cannot act purposefully. A bird that starts its migration, an insect that selects its host plant, an animal that avoids a predator, a male that displays to a female – they all act purposefully because they have been programmed to do so. (Mayr, 1961)

 

In relation to the cause of bird migration he adds:

 

The lack of food during the winter and the genetic disposition of the bird, are the ultimate causes. These are causes that have a history and that have been incorporated into the system through many thousands of generations of natural selection. It is evident that the functional biologist would be concerned with analysis of proximate causes, while the evolutionary biologist would be concerned with analysis of ultimate causes. (Mayr, 1961).

 

Migration of birds, like migration of all metazoans, invertebrate and vertebrate species, is an innate behavior. Now, almost half a century since Mayr’s publication we know that, while one or many genes are involved in expression of behaviors, all the attempts to prove that specific gene(s) may be responsible (i.e. both necessary and sufficient) for  any behavior have failed (see Animal Behavior is not Determined by Genes in chapter 9). Hence, the neoDarwinian point of view on accumulation of  “useful mutations” in the DNA that leads to new programs and characters is still far from being proven. To the contrary, adequate empirical evidence from the field of developmental plasticity, especially transgenerational plasticity, shows that changes in developmental pathways/programs often occur within one, two or several generations and affect all the individuals of a population, a fact that excludes gene mutations, existing genetic variability and the related action of natural selection in the evolution of developmental programs. Adequate empirical evidence also shows that changes in genes and accumulation of “useful gene mutations” have not been necessary for evolution of characters and speciation in metazoans.

In the case of migrating birds, environmental stimuli per se provide instructions neither for starting migration nor for the course the birds have to follow during the flight to the migration site. It is the hypothalamic clock that determines the timing of migration based on the environmental cues (which represent only environmental data from which the hypothalamus generates the information for timing the migration) and are other neural circuits that determine the position of the bird during the flight and the itinerary based on the processing of other sensory cues (visual, acoustic, geomagnetic, etc) and turn these cues into instructions (=information) for  staying- or correcting the course of migration. Thus, the causa finalis of the bird migration is in the processing activity of migration neural circuits, activation of identified complex neural circuits in the birds’ brain (Beason, 2005) rather than any imaginary accumulation of “useful gene mutations” and their selection over generations.

As for the proposition that the causa finales may be embodied in the genetic programs that living beings have acquired during their evolution, one should be reminded that all attempts to prove the existence of genetic programs have failed. Nevertheless, for the sake of argument, let’s suppose that the genome represents the “genetic program”. If this would indeed be the case, then it would be expected that somatic cells, which contain the genome, the full set of species-specific genes, would be able to develop into an adult organism of their kind. As we all know, somatic cells fail to do so and only the zygote (and the egg cell in parthenogenetic metazoans), which contains the same set of genes, succeed in developing into an adult organism. This fact  proves two things:

First, that the zygote and the egg cell of parthenogenetic organisms contains a developmental program that enables it to enter the process of individual development, and,

Second, that the genome contains no “genetic program” for individual development.

Indeed, for more than three decades we know that the program determining the development of the egg/zygote during the early stage of embryonic development is an epigenetic program, consisting of mRNAs, proteins, hormones, neurotransmitters, nutrients, etc., arranged in the eggcell/zygote in a strictly determined spatial order. The lack of this epigenetic program in somatic cells, which have all the genetic material the egg/zygote is in possession of, is the reason why somatic cells fail to enter the individual development and develop into an adult organism. With the benefit of the knowledge accumulated in about half a century since Mayr’s publication, we know that all behaviors such as migration of birds, considered to be purposeful are determined not by DNA, by any gene or any “genetic program”, but by information generated by a non-genetic, computational integration and processing of internal/external stimuli in neural circuits in the central nervous system (see sections Neural Mechanisms of Metazoan Migration and Light-dependent Magnetic Orientation, chapter 14).

In anticipation of the neoDarwinian counterargument that the activity of the neural circuits themselves is determined by genes and undemonstrated and undemonstrable “genetic programs”, recall that the properties of neural circuits, and behaviors they determine, can and do change not only in the course of evolution but even within the life of an individual, without changes in genes.

By definition, natural selection is the second stage in the process of the evolutionary change: the inherited, evolutionarily relevant change has to occur (by changing a developmental pathway or by a gene mutation) before the natural selection can act on it. It does not exists for its own sake but for the sake of the evolutionary change. There can be no selection without change: in the beginning there is the change.

The change in the developmental pathways precedes the action of natural selection and the generation of the evolutionary change is the first event of the causal chain. The ultimate cause in a causal chain is the first link in the chain, which is the emergence of the evolutionary change not the ensuing selection of the change. Natural selection, although necessary, is always a post factum process in relation to the emergence of the evolutionary change. Hence, it is logically inaccurate to believe that “many thousands of generations of natural selection” have been the causa finalis of actual phenotypes in metazoans; the inherited change is.

But any evolutionary change at a supracellular level is always manifestation of a corresponding change in the developmental pathway that takes place during ontogeny. This suggests that the study of the ontogeny of the extant organisms can give us important clues on the ultimate causes that might have acted long before in the course of the species phylogeny. For, as a rule, the developmental pathway that brought about the evolutionary change is still present and operational in the ontogeny of bearers of the evolutionary change, i.e., as long as evolutionary change is maintained.

NeoDarwinians neglect the fact that the evolutionary change that took place in the course of evolution would not exist if the developmental process that brought it about would not still be operational in the individual development of the bearer of the change. They miss the crucial fact that in the the structure and function of metazoans, as well as living systems in general, in distinction from anorganic systems, is inscribed the evolutionary history. Hence, in modern species and individuals, the ultimate cause, the evolutionary change in the developmental pathway, which preceded the natural selection, is still active in the ontogeny of each individual as long as these individuals express the evolved character.

The species ontogeny, thus, is a chronicle of the evolutionary history, often written in undeciphrable epigenetic codes. It may be argued, however, that the present ontogeny may not exactly reflect the original change in the developmental pathway. In a linguistic analogy, identification in the ontogeny of the initial change in the evolution of a structure presents to the biologist the difficulties a linguist encounters in attempts to reconstruct the ancient root of a modern word. Comparative developmental studies give sufficient clues to make that identification possible, just as comparative linguistic in etymological studies does.

Obviously, natural selection is not involved in the emergence of the new/modified developmental pathway. What natural selection can do, in exact terms, is to eliminate carriers of the change if less fit under existing conditions of the environment.

In their interaction with the environment such systems are influenced by the environment and at the same time influence it.

It is easy to prove that causae finales (ultimate causes) and causae efficientes (proximate causes) of material productions of human activity are external to these objects, and lie respectively on the human activity and human intention to produce them. But I find it impossible to believe that causae finales and causae efficientes of living systems are external to these systems (in an environmentally determined selection or any imaginary external force).

A basic two-centuries old question still is waiting for an unambiguous answer: Is the change in environment that adapts the organism or is the organism that tends to adapt itself to the changed environment? Is the organism a passive entity in the process of adaptation or is it actively acting to accommodate its structure, function, behavior and life history to the changing environment? After all, essential for the evolution of new adaptation is neither the source of matter nor the energy but the source of information. Is the source of information necessary for adapting the animal phenotype within the organism or is it external to the organism?

NeoDarwinian concepts of natural selection fail to plausibly explain the causal basis of evolutionary change. The  concept of natural selection as an external agent of evolution implies that. the cause of the evolutionary change is not in the living system, but outside it, in its environment. But, to attempt to find outside the living systems, the cause of what has taken place within these systems during the evolution, implies denying these systems what is one of their basic properties, the ability to adaptively respond to changes in the environment.

As pointed out earlier, the concept of natural selection as a result of the interaction of living beings with their environment, by definition is is excluded as a potential cause of evolutionary change and evolution in generally.

Vast empirical evidence (part of which is included in this work) shows that living organisms, in sharp contrast with anorganic systems, not simply react but adaptively respond to environmental agents that disturb their homeostasis, by adaptively changing their behavior, physiology, morphology, and life histories. These responses are teleonomic responses (sensu Mayr) enabled by the existence in living systems of control systems (see chapter 1) that actively act to restore the normal function and structure when they are disturbed by adverse actions of environmental agents. Living systems are actively adapting machines rather than passive receivers of external influences.

Adaptive responses are manifestations of the causal basis of evolution, of the ultimate cause, of an intrinsic drive, which intends to restore the disturbed homeostasis, by counteracting, resisting or avoiding harmful effects of the changes in environment. The causa finalis, this intrinsic (teleonomical, not teleological) purpose is actualized via the adaptability and evolvability, i.e. their ability of living systems to change their behavior, morphology and function in order to resist or avoid destabilizing effects of the changed environment. The causa finalis and adaptation are realized and embodied in the end-product of evolution, in the adaptively changed animal structure.

Crucial in evolving new characters in biological systems is not the matter or the energy, rather than the source of information, “instructions”, on how to produce the evolutionary change. Any evolutionary change requires and implies change (acquisition, but sometimes loss) in the informational content of the system. Hence, the final cause is to be identified with the source of information for the evolutionary change. Is the environment in possession of that information, and can it provide metazoans with specific information to adaptively change their behavior, morphology or physiology? The answer is clearly no. Metazoans are themselves able to generate and use the information necessary for erecting and evolving their structure.

What is the source of information in the case of incipient species where individuals and/or populations inhabiting the same niche suddenly are reproductively isolated from the rest of populations that are in possession of the same genome, or in a neoDarwinian term, the same “genetic program”? Certainly differences of some kind exist between the two groups of individuals that preferentially mate individuals within their own group, but the differences are not genetic (both groups or populations are in possession of the same genotype) and not related to the action of natural selection (see chapter 19 Epigenetics of Sympatric Speciation). Differences in mating preferences that lead to the reproductive isolation of two populations of the same genotype in sympatry indicate that the cause of the differences and of the resulting process of speciation is not genetic. We know what is the difference between the two sympatrically isolated populations (=incipient species); we know the neural basis of mate preferences and mating biases  determining group- or species-specific mating behavior; we know that the differences involve no changes in genes, that they are epigenetic differences related to sudden changes in properties of neural circuits, determining mating preferences (see Neural-cognitive Mechanisms of Sympatric Reproductive Isolation, etc. in chapter 19). 

The neoDarwinian distinction between the so-called “functional biology” (developmental biology) and “causal biology” (evolutionary biology), with the first answering the question how and the latter the question why, is artificial and erroneous. That distinction is based on the wrong premise that causa finalis (ultimate cause) of evolutionary changes is in the environment and natural selection (See earlier in this section). The artificial separation of biology into causal and functional has played a negative heuristic role. As  M.J. West-Eberhard has pointed out:

 

The answer to “why” an organism is, or behaves in a certain way can be answered either in terms of mechanisms (proximate causes) or in terms of selection and evolution (termed “ultimate causes” by Mayr, 1961). This distinction is designed to prevent confusion between levels of explanation in biology. But it was an easy step from this important point to the idea that the mechanisms of development have nothing directly to do with evolution or that they are the focus of a different research approach, one not primarily concerned with evolution and justifiably left aside by those primarily interested in selection and adaptation. (West-Eberhard, 2003a)

 

In fact, there is no other discipline in biology where both the “hows” and “whys” of biological processes unfold so clearly as in developmental biology. The study of individual development could provide answers to many fundamental hows and whys of biological phenomena. There is no other way to understand how a structure develops, but by identifying the developmental mechanism that leads to the development of that structure, and there is no other way to understand why that structure develops, but by identifying the function it performs.

.

Behavioral Prelude of Evolutionary Modifications of Animal Phenotype

 

As a rule, any evolutionary change triggered by drastic changes in environment, whether it is a modification, a loss of a feature, a reversal to an ancestral feature or a de novo morphology begins with an adaptive change in the behavior of animals. This is related to the fact that behavior is the most plastic of all the phenotypic traits in metazoans. It is the environmentally-imposed change in behavior and the accompanying stress state that sets the stage for the morphological adaptations. According to the “behavior evolves first” hypothesis, all the major morphological changes in evolution are preceded by changes in the behavior of animals, in response to changes in conditions of living. The idea is not new and has been attractive to majority of biologists during the last two centuries. Charles Darwin unambiguously pointed out the role of  behavioral changes in evolution:

 

Habit also has a decided influence, as in the period of flowering with plants when transported from one climate to another. In animals it has a more marked effect; for instance, I find in the domestic duck that the bones of the wing weigh less and the bones of the leg more, in proportion to the whole skeleton, than do the same bones in the wild-duck; and I presume that this change may be safely attributed to the domestic duck flying much less, and walking more, than its wild parent. (Darwin, 1959b)

 

In order to survive in adversely changing environments, animals have to adaptively change their behavior. Adaptive behavioral changes represent extemporaneous responses necessary for their survival but they can also help the species for “buying” the time necessary for evolving corresponding heritable changes in morphology, physiology, and life history.

Normally, under drastically changed conditions of living, animals learn to behave differently, adaptively. The animals’ learning is aided by the fact that in learning new behaviors animals quite often use pre-existing FAPs (fixed action patterns) and motor patternings. This logically raises a problem: is it possible for an animal to be in possession of such an apparently large number of FAPs and motor patternings it might need under different conditions of an ever changing environment? The difficulty is obvious but not as unsurmontable as it looks at first sight if one would bear in mind that:

Firstly, the survival in a severely changed environment does not necessarily require a perfect behavioral adaptation from the beginning; once the changed behavior makes the survival possible, perfection of the behavior may come later.

Secondly, radical changes in environment imply contrasting conditions of living, such as, e.g., aquatic-terrestrial environment, cold-warm weather, low-high degree of illumination, abundance-scarcity of food, presence-absence of predators, herbivorous-carnivorous diet, etc., which are not as numerous as their intermediate states would be. Reiterating, even imperfect behavioral adaptation to those alternative states  may enable the organism to survive, with perfection coming later.

Thirdly, there is evidence that often the same circuitry for a FAP or motor pattern can be modified to serve more than a single “purpose”, without changes in genes. So, e.g., in lampreys the same central motor program for a basic undulatory pattern is modulated to perform different forms of locomotion such as swimming, burrowing, and crawling. It is also experimentally demonstrated that the same neuronal circuit may produce different behavior patterns in response to application of different stimuli or hormones.

Fourthly, animals can adaptively modify their behavior by switching to new patterns of connections between the same neurons as it occurs with neurons that control lobster’s eating and digestion.

Fifthly, there is experimental evidence showing that the neural elements and connections for performing an ancestral behavior are conserved even in cases when the species has lost the specific part or organ used to perform the behavior.

From this common stage of behavioral adaptations to the drastically changed conditions of living starts the divergence of causal chains leading to four different evolutionary results, namely modification of existing phenotypes, loss of phenotypic traits, reversal of ancestral phenotypes and evolution of de novo phenotypes.

 

Epigenetic Determination of Modifications of Existing Phenotypes

 

Any animal structure has evolved for performing some function, which is the ultimate cause of the evolution of the structure. Under changed environmental conditions, sometimes metazoans may use the same organ for performing a behavior that is different from its original function: it is likely that the first fish species that ventured to explore the terra firma might have used its fins to support and carry its body weight for locomotion on dry land. There is reason to believe that a switch from the innate swimming behavior to a “learned” shambling, which was facilitated by the existence in the fish brain of a highy plastic neural circuit for different forms of locomotion, including crawling, enabled the early aquatic gambler to survive until its fins could evolve into true walking limbs. In all likelihood, the “learned” shambling preceded and facilitated the evolution of tetrapod limbs from swimming fins.

Arguably, new behaviors precede evolutionary modifications and if the behavioral adaptation is indeed the first step in the process of evolution suggests that evolutionary changes start with a neural mechanism. How these neural mechanisms of behavioral adaptation to the changed environment are related to the ensuing specific evolutionary changes in morphology still represents one of the great enigmas of modern biology. Difficult as an understanding of that relationship is, it must be pointed out that examples of neural determination of morphology and even of evolution of morphology and other phenotypic characters are not lacking.

As mentioned earlier, it is generally believed that most of evolutionary innovations arise by modification of existing structures and probably represent the most widespread mode of evolution of morphology in metazoans. De novo structures that are not homologous to any structure in ancestral species are rare events in metazoan evolution. Morphological novelties result from adaptive changes in developmental programs rather than gradual accumulation of useful gene mutations. At a large evolutionary scale this is monumentally manifested in the “Cambrian explosion”: about 540 million years ago, in the space of only several million years, an incredibly rapid evolution and diversification of the animal world occurred, with the appearance of ~100 phyla, including all of ~ 30 modern phyla, characterized by distinct major body plans (Baupläne). Abundant paleontological evidence also suggests that often new taxa appear so suddenly that cannot be accounted for by the action of “natural selection” on spontaneously arising useful mutations. Generally, such rapid diversifications could not be predicted by the contemporary neoDarwinian hypotheses of evolutionary gradualism.

The same conclusion can be drawn from observations at a smaller scale, i.e., from processes of speciation events. So, e.g., in ~one thousand years since the introduction of banana plants in Hawaiian islands several new species of moths of the genus Hedylepta, which still feed exclusively on banana plants, have evolved; five new species of cichlid fish evolved from a single original species in the small (three miles by five) lake Nabugabo within ~ 4,000 years since it was formed out of a sand spit at the margin of Lake Victoria in Africa. Here is another impressive example: in the small (half square mile) lake Bermin, West Cameroon, nine endemic species have evolved within a few thousand years. These, and many other similar examples, preclude involvement of any mechanism of gradual evolution via accumulation of “favorable mutations” and geographical isolation in the speciation process (random mating would eliminate most of the extremely rare favorable gene mutations, if they would ever occur, under sympatric conditions).

Numerous cases of polyphenisms, when individuals of the same genotype, in response to specific environmental stimuli, switch to alternative discrete morphologies, demonstrate the ability of animals to change their morphology without changing their genotype in an adaptive response to the changed conditions of life.

Metamorphosis is a widespread phenomenon in the animal kingdom demonstrating that no changes in genes are necessary even for radical morphological changes, affecting Baupläne of different classes of invertebrates and vertebrates. Premetamorphic anuran ampibians develop a fish Bauplan and insects a worm Bauplan before metamorphosing into the Baupläne of their own classes. It is interesting to observe that during the larval stages, these species along the ancestral morphology also display ancestral behaviors related to that morphology, which suggests that basic neural circuits for performing these ancestral  behaviors ( fish and worm behaviors, respectively) are also conserved after the loss of ancestral structures.

Although evolutionary modifications of animal morphology most of the time are triggered by environmental stimuli, the mechanisms that determine modifications, and the source of information necessary for their development, are of intrinsic origin. As argued in chapter 2, the external stimulus provides no information for the morphological change, as it is suggested by the fact that the modification it triggers is not predictable from the nature of the external stimulus: in response to the same environmental stimulus different organisms respond by developing different, and often opposite, morphological modifications.

How is the environmental stimulus related to the evolutionary modification of the morphology that it triggers? There is reason to believe that the change in behavior provides a crucial link in the causal chain of events extending between the environmental stimulus and the evolutionary modification of the morphology it stimulates. This epigenetic link between environment, behavior and morphology is visualized in many observations from nature and experiments. R. Denver, e.g., has shown that as tadpoles, desert amphibians live in temporary ponds that contain water for unpredictable periods of time. In the years of low precipitations these ponds dry up earlier. The input of the external stimulus (earlier receding water in the pond) and internal stimuli are integrated and processed in neural circuits in the tadpole’s brain (hypothalamus). This leads to a neurally determined stress response (habitat stress) in tadpoles and a series of identified changes in the behavior of tadpoles, besides changes imposed by the new conditions of living under threat of getting caught by the draught. The epigenetic link between the behavior and morphology in the case of tadpole metamorphosis is provided by secretion of the hypothalamic neurohormone CRH (corticotropin-releasing hormone), which is responsible for both the behavioral stress and for inducing metamorphosis (Denver, 1997b). Secretion of CRH, the principle vertebrate stress neurohormone, by the hypothalamus is the key element for the changed behavior, stress  response, and for speeding up the morphological transformation of the aquatic tadpole into an adult terrestrial amphibian organism.

Denver demonstrated that the same result of speeding up the transformation of the fish-like tadpole into an adult amphibian is also obtained in laboratory when the water level is experimentally lowered earlier. The lowering of the water level is a “drastic change in the environment”, which causes a stress condition and  the related changes in the behavior of tadpoles struggling to avoid the threat of draught.

 

The lowering of the level of the water in the environment makes the hypothalamus to produce more CRH, which stimulates pituitary to produce hormones that stimulate thyroid and adrenal glands whose products help organism to cope with the stress, in this case by losing their tail and beginning the growth of their limbs. (Denver, 1997b)

 

Evolutionary Modifications of  Morphology

 

Evolution of Anterior-posterior Axis in Vertebrates

 

The establishment of the A-P axis in vertebrates is considered to result from expression patterns of Hox (homeobox) genes. But the inescapable question is: Where the information, “instructions”, on where and when to express each of these Hox genes?

Reliable evidence shows that expression pattern of these Hox genes is extracellularly regulated by upstream signals ultimately originating in the embryonic CNS. In more concrete terms, expression pattern of Hox genes along the animal body, is determined by RA (retinoic acid). Experimental administration of RA, by modifying expression of Hox gene domains along the A/P axis, causes posterior/anterior transposition of body regions along the axis, thus leading to respecification of vertebral identities (Kessel and Gruss, 1991; Kessel, 1992; Burke et al., 1995) (figure 14.1).

The ability of RA to induce expression of Hox genes derives from the presence of RARE (retinoic acid response elements) in enhancers of these genes. RA is known to immediately regulate expression of all the known Hox genes (Conlon, 1995; Cupp et al., 1999).

 

Figure 14.1. The patterns of axial homeotic transformations seen in Hox and RAR loss-of-function mutants are similar to each other, and are complementary to those caused by retinoic acid (RA) excess and Hox overexpression. The mouse has seven cervical vertebrae (C1 to C7) in the wild type. Under conditions of RA excess, or Hox gene overexpression, vertebrae have a tendency to be transformed to posterior identities. A composite of each is shown, where the seventh cervical vertebra (C7) is transformed into the first thoracic (T1) and C6 to C7, etc. In RAR and Hox mutants, the tendency is in the opposite direction, where transformations to anterior vertebral identities are prevalent  (From Conlon, 1995).

 

A direct molecular link between RA, RAR and (receptor), and the activation of specific homeobox genes is also observed during embryogenesis. The availability of retinoid response elements in Hoxal, Hoxb1, and Hoxd4 genes suggests that retinoids act as an early developmental signal, possibly conditioning the posterior-to-anterior gradient in the gastrula and providing positional specification of the A/P axis in the developing vertebrate embryo. How other Hox genes located in more 5’-positions of the cluster are regulated by RA is not known. The Hoxa1 and Hoxb1 genes appear to be modulated directly via the RARE, and the protein products of these genes may activate the proximal 5’-genes in the clusters, that is Hoxa2 and Hoxb2, respectively. This scenario would allow for a sequential activation of Hox genes. It may also be that downstream targets of Hox genes, as well as other developmental genes regulated by RA, are involved in the regulation of the Hox genes located in more 5’-positions. Hox gene regulation is mediated partly by positional information supplied by retinoids. (Ross et al., 2000)

Evolution of the A-P axis in metazoans has been determined by the patterns of expression of RA (retinoic acid) along the animal body. No function-affecting changes have occurred in genes responsible for RA synthesis. As for the source of the RA along the axis, it is interesting to point out that

 

The total amount of RAs varied by 29-fold across different tissues with the lowest in the heart and the highest in the neural tube. (Maden et al., 1998)

 

The fact that evolution of the A-P axis in vertebrates is not related with mutations in genes responsible for RA synthesis and Hox genes, excludes the possibility of a feasible neoDarwinian explanation of that evolution. Epigenetic mechanisms of regulation of these genes have been involved in the evolution of A-P axis in vertebrates (see later on the nature of gene regulatory networks).

 

Evolution of Body Size in Manduca sexta

 

Attainment of species-specific body size in animals roughly coincides with the onset of sexual maturity. Mechanisms of control and regulation of body mass have only recently been thoroughly investigated. Different mechanisms seem to be operative at different steps of the evolutionary ladder. Studies carried out so far show that signal cascades for regulating this basic phenotypic character start in the nervous system.

Among invertebrates, several studies on regulation of body size have been carried out in insects such as  Drosophila and  Manduca sexta (Sphingidae). In 2003, H.F. Nijhout argued that the problem with the determination of body size is how an animal determines when to stop growing (Nijhout, 2003). Different animals use different external and internal cues for determining when to stop growing.

An experimental case of the evolutionary change in body weight is recorded under laboratory conditions on Manduca sexta. Within ~220 generations (30 years) the insect increased its body mass by 50% (D’Amico et al., 2001). Experimental studies have shown that the adult body size in Manduca sexta depends on a number of factors: the initial size of the last larval instar, the growth rate during that instar, the value of the critical weight, the time required for the clearance of JH during the last instar, and the timing of the photoperiodic gate for secretion of the neuropeptide PTTH (prothoracicotropic hormone) by secretory neurons in the insect brain. Evolutionary changes in the three last factors (growth rate, critical weight and secretion of PTTH) “almost completely account for the evolutionary increase in body size observed.” (D’Amico et al., 2001; Davidowitz et al., 2003; Davidowitz et al., 2004).

The difference between the adult body weight and the critical weight results from the fact that cessation of JH synthesis does not automatically lead to cessation of growth. Residual JH and JH mRNA still continue to stimulate growth and prevent secretion of the neuropeptide PTTH, which stimulates secretion of ecdysone by the prothoracic gland, thus arresting the larval growth (Davidowitz et al., 2003; Davidowitz et al., 2004). The reason for this interval to cessation of growth (ICG) is that larva, after achieving competence for PTTH synthesis, has to wait for the “photoperiodic gate”. The photoperiodic gate is 8 hours long, after which one third of larvae will become competent while the rest of them has to wait until next photoperiodic gate opens (D’Amico et al., 2001). 

There was no genetic variation for plasticity of critical weight and no indication exist on mutations having played any role in the evolutionary increase of the body weight of the laboratory strains of the tobacco hawkworm, Manduca sexta. Hence the  conclusion:

 

Plasticity of body size must thus be due to plasticity in this underlying endocrine control mechanism. (Davidowitz et al., 2004)

 

Evolutionary changes in the growth rate, critical weight, and PTTH delay time are responsible for 95% of the evolutionary increase in body mass  of M. sexta (D’Amico et al., 2001) (see also The Neural Control of the Body Mass in chapter 1).

A closer look on the developmental mechanisms determining growth rate, critical weight, and PTTH delay time may shed some light on the origin and nature of information for the evolutionary change.

 

The growth rate. A correlation exists between the patterns of secretion of IGFs (insulin-like peptides) and PTTH in the brain of insects and their growth rate during larval stages (Rulifson et al. 2002). The insulin/IGF (insulin growth factors) signaling pathway is the mediator of the function of the CNS in determining the growth rate in Drosophila melanogaster (Colombani et al. 2005). Five insulin-like peptide genes, homologous to human and mouse insulins, are identified in Drosophila (Rulifson et al. 2002) and an insulin supergene family of 37 genes is identified in the nematode worm, Cenorhabditis elegans (Leevers, 2001). Although these genes are expressed in other tissues, it seems that their expression in the insulin-producing neurons in the CNS during larval stages is of determining importance for larval growth and development in Drosophila. These insulin-like peptides are synthesized and secreted by two bilateral clusters of neurons in the pars intercerebralis of the protocerebrum: ablation of these secretory neurons leads to production of flies of normal body proportions but of smaller size, with wing size reduced to 61% and wing cell number reduced to 72 % of the normal condition.

 

Critical weight. There is a moment during the last instar of Manduca sexta when secretion of JH (juvenile hormone) suddenly stops and the activity of JH esterase increases so that hemolymph is cleared of JH, making thus secretion of ecdysteroids by the prothoracic gland possible. Secretion of ecdysteroids is regulated by a brain signal, the neuropeptide PTTH (prothoracicotropic hormone), whose synthesis starts during the first photoperiodic gate after the clearance of JH, 2.5 hours after onset of the photophase (Davidowitz et al. 2003). With secretion of ecdysteroids the insect stops feeding and growing. Suppression of JH secretion and of JH esterase activity coincides with the time when the insect attains the critical body mass, which is about 55% of the adult body mass. This suggests that a causal relationship exists between the attainment of the critical body size and the cascade of events leading to metamorphosis.

 

The assessment of this critical body size is made in the brain and is based on processing of signals sent by the insect’s stretch proprioceptive neurons that receive mechanical stimuli of increasing stretch as a result of increased body size (Gorbman and Davey, 1991).

 

By integrating and processing these stretch stimuli, neural circuits determine the time for sending signals (neuropeptides of the family of allatostatins) to corpora allata for suppressing the synthesis of JH.

 

PTTH (prothoracicotropic hormone) delay time. The growth of the fifth instar larvae continues after the termination of JH secretion until ecdysteroids from the prothoracic gland are released in hemolymph. Needless to say, the synthesis of ecdysteroids is cerebrally regulated by the neurohormone PTTH. Although at the time that the larva suppresses JH secretion and  it is competent for PTTH synthesis, it does not do so until the brain is stimulated by a photic cue, which is the first photophase after the interruption of JH secretion.

 

The fact that the brain starts secreting PTTH, not  at any time of the day but only during the photoperiodic gate, creates a time span between the suppression of JH secretion and the commencement of PTTH secretion. This interval between the termination of JH secretion (its secretion is also terminated by brain signals) by corpora allata and the beginning of PTTH secretion is known as PTTH delay time. During this interval until the PTTH secretion, larvae continue to grow. Evidently, the length of this interval influences the adult body size that the insect attains at the onset of metamorphosis.

A generalized diagram of the mechanisms involved in determination of  body size in insects is presented in figure 14.2.

 

 

 

Figure 14.2. Diagramatic representation of the neural control of body size in insects.

 

It is important to know what determines the timing of PTTH release in the insect’s brain. As early as 80 years ago, Wigglesworth found that insect proprioceptors may receive and transmit to the brain input on increasing stretch, and processing of this neural input leads to secretion in the brain of PTTH, which in turn stimulates secretion of ecdysone by the prothoracic gland (Gorbman and Davey, 1991; West-Eberhard, 2003a). This suggests that a set point for an upper limit of stretch must exist, beyond which the brain activates the ecdysone cascade and suppresses JH cascade, thus stoping further growth and determining the species specific adult body size.