Download Evo-Devo, Devo-Evo, and Devgen

Biology and Philosophy 18: 347–352, 2003.
© 2003 Kluwer Academic Publishers. Printed in the Netherlands.
Evo-Devo, Devo-Evo, and Devgen-Popgen
SCOTT F. GILBERT
Department of Biology
Martin Research Laboratories
Swarthmore College
Swarthmore, PA 19081
U.S.A.
E-mail: [email protected]
In his guest editorial in Evolution and Development, “Evo-devo or Devo-evo
– Does It Matter?”, Brian Hall (2000) concluded that Evo-devo and Devo-evo
are not the same and that it mattered a great deal. According to Hall, Evo-devo
is continuous with the evolutionary embryology of the nineteenth century and
is a way of integrating proximate and ultimate causes for the origin of phenotypes. The interactions of evolutionary biology and developmental biology,
he believes, should produce emergent properties not found in either parent
discipline. Interactions tend to do that, whether inside the embryo or inside
academia. While Evo-devo is merely the expansion of developmental biology
into a new area, Hall views developmental evolutionary biology (Devo-evo)
as constituting something more revolutionary. According to Hall, Devo-evo
sees the neo-Darwinian view of evolution (the Modern Synthesis) as incomplete and in need of complementation by developmental biology. Moreover,
such a new synthesis “would not place primary emphasis on genes as the only
units of evolutionary change.”
I have been asked to comment on this matter, and I find that Hall and I
differ with respect to these definitions. I think that Hall’s definition of Devoevo has two parts, and that these parts do not necessarily have to go together.
One part of his definition of Devo-evo is that its practitioners claim that
the Modern Synthesis is insufficient to explain evolution and needs developmental biology to complete it. The second part of his definition is that
Devo-evo seeks to accomplish this completion by downplaying the roles of
genes in evolution. I think that the first part of the definition is consonant with
Evo-devo (sensu Gilbert et al. (1996) and probably Hall (2000), as well, if I
am reading his paper correctly); while the second portion of the definition is
a separate issue as to what should be included in the integrative project of
evo-devo.1
348
Definitions matter, and I think the differences between Evo-devo and
Devo-evo depend upon what we think is at stake and how we define
our terms. It’s been said that Evo-devo is the synthesis of evolutionary
biology and developmental biology. That is not entirely true. First, there are
other agents involved in the synthesis. Paleontology, morphometrics, QTL
mapping, ecology, life history strategy research, and functional morphology
certainly are players in this field, and they shuttle back and forth between
developmental and evolutionary biologies, making different types of unions
possible. But more importantly, not all parts of developmental biology and
not all parts of evolutionary biology are involved in these new unions. What is
happening, I believe, is a series of interactions occurring between population
genetic models of evolution and developmental genetic models of evolution. Both of these models emphasize genes. But in one case, evolution
depends upon the frequency of gene variants within a population. In the other
case, evolution depends on variations of gene expression between populations. As Ron Amundson has noted, the new union might by characterized
as “devgen-popgen”; the genes are not going away when these models are
synthesized.
To be sure, there are at least three groups who would like to see genes
play a less important role in evolutionary biology. The first is a group of
developmental biologists who are interested in generic control mechanisms
that may have preceded genetic control mechanisms in the evolution of
development. These people include Stuart Newman (see Newman 2003), and
they are interested in the evolutionary origins of development, itself. Their
interests complement those of evolutionary developmental biologists whose
work involves the modification of genetic regulatory networks, rather than
their origin. The second group (in which I would place myself) sees various
modular entities as mediating the actions of genes on development and evolution (e.g., Opitz and Gilbert 1993; Gilbert et al. 1996; Raff 1996). Thus, genes
are critically important for development and evolution, but since phenotypes
do not reduce directly to genotypes, and since modular organization allows
for heterochony, allometry, co-option, and divergence, these intermediate
modules (such as morphogenetic fields) must also be considered units of
evolution. These modules are usually the products of gene expression, but by
the interactions of their component gene products, they form a new level of
informational organization. Both of the above groups could be accommodated
into Evo-devo. A separate “Devo-evo” category need not separate them.
The third group of people who would like to see less importance given
to genes in evolutionary biology are those people constituting a field of
the philosophy of science called developmental systems theory (DST).
These philosophers (and their biological allies such as Brian Goodwin)
349
have expressed the belief that a more balanced view of evolution is
possible through non-genetic, interactionist, non-hierarchical developmental
approaches. Although I believe Hall is reacting to these views (see Robert
et al., 2001), I do not think that Evo-devo is likely to adopt the ideas of
Goodwin (1994), Gray (1992), Oyama (2000), or even Keller (2000). In fact,
Evo-devo is probably the gravest danger DST has had to face. As long as
evolutionary biology was almost completely dominated by neo-Darwinian
population biology, DST could point to its deficiencies and claim that
development could solve these problems, and that genes might not be necessary. Evo-devo, however, has taken these arguments from the DST people
(perhaps indirectly), in a sense co-opting them into mainstream evolutionary
biology. Evo-devo can say that it, too, doesn’t believe that the population
genetics of neo-Darwinism is sufficient to explain evolution; but residing
inside “Darwin’s Black Box” is neither God (Behe 1996) nor DST, but
developmental genetics (but see Hall, this issue).
Moreover, like DST, evo-devo also postulates that there is, to use Oyama’s
phrase, an “ontogeny of information.” Whether or not a particular sequence
of DNA is a gene depends upon the developmental history of the cell in which
it resides. Whether a protein such a GSK-3 is an enzyme or part of a signal
transduction cascade similarly depends upon the cellular context. Whether
reception of BMP4 causes a cell to undergo apoptosis, chondrogenesis, or
epidermal differentiation similarly depends upon the developmental history
of the responding cell (see Gilbert and Sarkar 2000).
DST has been valuable in pointing out where neo-Darwinism has not
provided complete explanations of phenotype origin, and where DNA does
not have instructive properties. However, DST has, generally, made the error
of not assigning instructive or permissive influences in the interactions. To
most of the developmental systems theorists, all the participants are on the
same informational level. In this way, the genome is just one other participant,
just as are cells or the environment. However, the specificity of the reaction
(that it is a jaw that forms and not an arm; that it is a salamander jaw that
forms and not a frog jaw) has to come from somewhere, and that is often
a property of the genome. Embryology has long recognized that in developmental interactions, there are usually instructive partners and permissive
partners (Holtzer 1968). Instructive partners provide specificity to the reaction, whereas permissive partners are necessary, but do not provide specificity.
If a notochord is placed near a chick neural tube, the neural tube region closest
to the notochord will be induced to express the genes of the floorplate neurons
(Yamada et al. 1993). Here, the notochord has acted instructively. It can
induce floorplate specification in any region of the neural tube. In other cases,
the notochord acts permissively (Edlund 2002). The instructions to specify a
350
pancreas probably occur before the interaction of the endodermal tube with
the notochord. But the notochord allows for the pancreatic differentiation
to occur, probably by providing non-specific growth factors or structural
support. Here, the notochord is not providing any specificity, it is just permitting a reaction that has been specified by something else to occur. Permissive
and instructive interactions occur at many levels in the embryo. When the
jaw-forming region of a frog embryo is transplanted into the jaw-forming
region of a salamander embryo, it is told to make a jaw; but the jaw becomes
a frog jaw (Spemann and Schotté 1932). Here, the genome is providing the
specificity for the reaction.
The gene has to be given its proper respect. Granted, it is not the unmoved
mover that the Human Genome rhetoricians or the extreme neo-Darwinists
would claim; but it does provide the inherited specificity of most important
embryonic interactions. Granted, there are modules that intercede between
the gene and the phenotype, and these have properties different from those
of the gene. In many instances whether a gene is mutant or wild-type
depends upon its context. There are even times where the environment gets to
provide the specificity of developmental interactions (Gilbert 2001; van der
Weele 1999; Goldschmidt 1940).2 The sex of a turtle depends on its incubation temperature, not on its chromosomes. Here, the environment plays the
instructive role whereas the genome is permissive. But these phenomena can
all be integrated into a developmental genetic approach to evolution (Gilbert
2001). Just because evolution and development have genetic bases does not
mean they have to be reduced to these bases. Only by mistaking developmental biology as reductionistic would we risk the error of getting rid of the
genes as explanations. The merging of evolution and development does not
mean that we have to throw away our developmental notions of genes or of
specificity.
I think that we can have the scientific program of Evo-devo (sensu Hall)
and use that scientific program to show that the neo-Darwinan Modern
Synthesis is incomplete (i.e., Devo-evo, sensu Hall). It may even be the case
that the population genetics model turns out to be placed within a developmental framework (Goldschmidt 1940). However, I would claim that one
must be more precise, and say that the population genetic model of evolution
is incomplete and must be supplemented with or complemented by (but not
replaced by) a developmental genetic model (Wagner 2000). But the developmental genetic program should not be reductionist, since we know that
genes act according to their contexts and that the environment can regulate
gene expression (Gilbert and Sarkar 2000). I think that the term “evolutionary
developmental biology” (“Evo-Devo”) is fine for both the scientific program
to explain evolution through changes in development and for the academic
351
program to change evolutionary biology. I do not think the two can be kept
separate.
Waddington (1975) used the term “diachronic biology” to denote the intersection of evolutionary biology, genetics, and development. He also used the
term “epigenetics” to describe the developmental mechanisms that integrated
genetics and epigenetic interactions to bring about the production of the
phenotype. However, I suspect that like many of his other neologisms, these
ones will not be widely used. “Evolutionary developmental biology (“evodevo”) seems a fit designation for the field, especially as it can encompass the
evolution of developmental mechanisms and the ecologically relevant aspects
of this synthesis (developmental plasticity, developmental symbioses, etc. –
see Gilbert and Bolker 2003). Indeed, this phrase is the title of Hall’s own
version of the synthesis (Hall, 1999) and was used by Hall to designate the
field in its first separate entry in an encyclopedia of science (Hall 1996).
Whatever its name, I suspect that Evo-devo will be successfully integrated
into evolutionary accounts of nature, and as part of normative “evolutionary
biology,” become called as such. But whatever it is called, it will be genealogy
writ enormous and a fundamental way of explaining biodiversity.
Notes
1 The controversy over whether evo-devo is separate from the population genetic framework
of the Modern Synthesis or if it fits into that framework has its precursor in the discussions over
whether genetic assimilation fits into the modern synthesis. I have suggested (Gilbert 1994)
that Schmalhausen presented his model as part of the population genetics framework, whereas
Waddington presented a similar model as complementary to and completing the population
genetics framework.
2 Compare DST scholar Cor van der Weele’s (1999) with Gilbert (2001) or with chapter 21
of my textbook (Gilbert 2000) to gain a perspective on the different ways of presenting similar
material.
References
Behe, M.J.: 1996, Darwin’s Black Box: The Biochemical Challenge to Evolution, Free Press,
New York.
Edlund, H.: 2002, ‘Pancreatic Organogenesis – Developmental Mechanisms and Implications
for Therapy’, Nature Reviews Genetics 3, 524–532.
Gilbert, S.F.: 1994, ‘Dobzhansky, Waddington and Schmalhausen: Embryology and the
Modern Synthesis’, in M.B. Adams (ed.), The Evolution of Theodosius Dobzhansky:
Essays on His Life and Thought in Russia and America, Princeton: Princeton University
Press, pp. 143–154.
Gilbert, S.F.: 2000, Developmental Biology (6th edn.), Sinauer Associates, Sunderland, MA.
352
Gilbert, S.F.: 2001, ‘Ecological Developmental Biology: Developmental Biology Meets the
Real World’, Developmental Biology 233, 1–12.
Gilbert, S.F., Opitz, J. and Raff, R.A.: 1996, ‘Resynthesizing Evolutionary and Developmental
Biology’, Developmental Biology 173, 357–372.
Gilbert, S.F. and Sarkar, S.: 2000, ‘Embracing Complexity: Organicism for the Twenty-First
Century’, Developmental Dynamics 219, 1–9.
Gilbert, S.F. and Bolker, J.A.: 2003, ‘Special Issue: Ecological Developmental Biology’,
Evolution and Development 5(1).
Goldschmidt, R.B.: 1940, The Material Basis of Evolution, Yale University Press, New Haven.
Goodwin, B.: 1994, How the Leopard Changed its Spots, Scribner, New York.
Gray, R.: 1992, ‘Death of the Gene: Developmental Systems Strike Back’, in P.E. Griffiths
(ed.), Trees of Life: Essays in the Philosophy of Biology, Kluwer, Dordrecht, pp. 165–209.
Hall, B.K.: 1996, ‘Evolutionary Developmental Biology’, McGraw-Hill Yearbook of Science
and Technology, McGraw-Hill, New York, pp. 110–112.
Hall, B.K.: 1999, Evolutionary Developmental Biology, Chapman and Hall, London.
Hall, B.K.: 2000, ‘Guest Editorial: Evo-Devo or Devo-Evo – Does it Matter?’, Evolution and
Development 2, 177–178.
Hall, B.K.: 2003, ‘Opening the Black Box between Genotype and Phenotype: Cells and cell
Condensations as Fundamental Units of Evolutionary Developmental Biology’, Biology
and Philosophy (this issue).
Holtzer, H.: 1968, ‘Induction of Chondrogenesis. A Concept in Terms of Mechanisms’, in
R. Glieschmajer and R.E. Billingham (eds),Epithelial-mesenchymal Interactions, William
and Wilkins, Baltimore, pp. 152–164.
Keller, E.F.: 2000, The Century of the Gene, Harvard University Press, Cambridge.
Newman, S.A.: 2003, ‘From Physics to Development: The Evolution of Morphogenetic
Mechanisms’, in G.B. Müller and S.A. Newman (eds), Origination of Organismal Form:
Beyond the Gene in Developmental and Evolutionary Biology, MIT Press, Cambridge,
MA, pp. 221–339.
Opitz, J.M. and Gilbert, S.F.: 1993, ‘Developmental Field Theory and the Molecular Analysis
of Morphogenesis: A Comment on Dr. Slavkin’s Observations’, American Journal of
Medical Genetics 47, 687–688.
Oyama, S.: 2000, The Ontogeny of Information: Developmental Systems and Evolution
(revised edn.), Duke University Press, Durham.
Raff, R.A.: 1996, The Shape of Life: Genes, Development, and the Evolution of Animal Form,
University of Chicago Press, Chicago.
Robert, J.S., Hall, B.K. and Olson, W.M.: 2001, ‘Bridging the Gap between Developmental
Systems Theory and Evolutionary Developmental Biology’, BioEssays 23, 954–962.
Spemann, H. and Schotté, O.: 1932, ‘Über xenoplastiche Transplantation als Mittel zur
Analyse der embryonalen Induktion’, Naturwissenschaften 20, 463–467.
van der Weele, C.: 1999, Images of Development: Environmental Causes in Ontogeny, SUNY
Press, NY, New York.
Waddington, C.H.: 1975, The Evolution of an Evolutionist, Cornell University Press, Ithaca.
Wagner, G.P.: 2000, ‘What is the Promise of Developmental Evolution? Part I: Why is
Developmental Biology Necessary to Explain Evolutionary Innovations?’, Journal of
Experimental Zoology (Molecular Development and Evolution) 288, 95–98.
Yamada, T. Pfaff, S.L., Edlund, T. and Jessell, T.M.: 1993, ‘Control of Cell Pattern in the
Neural Tube: Motor Neuron Induction by Diffusible Factors from Notochord and Floor
Plate’, Cell 73, 673–686.