Transcript Functional asymmetry

Evolution's middle floor:
functional morphology and the
origin of complexity
Graham Budd
Budd G.E. (2006) On the origin and evolution of major
morphological characters. Biol. Rev. 81(4).
What is structuralism?
Idea being that form, in both genotype
and phenotype, can’t just freely vary,
but is governed by certain discoverable
laws.
Structuralists would normally claim that
these extra laws are required to
understand fully how organisms evolve.
General Themes
• E.g. constraint. Why have some features
not changed over hundreds of millions of
years?
'A developmental constraint is a bias on
the production of variant phenotypes or a limitation on
phenotypic variability caused by the structure, character,
composition, or dynamics of the developmental system.'
Maynard Smith et al. 1985
Bauplan – an idea that is “misguided and
dispensable”?
(G. C. Williams 1992)
“The text-books would have you believe that all vertebrates have hemoglobin, but
there is a diverse group of vertebrates, the Antarctic fish family Chaenichthyidae,
in which hemoglobin is entirely lacking. This is merely one of many possible
examples of generalizations for which exceptions must be recognised”
G. C. Williams 1992, p. 88.
Flavours of structuralism

Self-organisers (e.g. Goodwin, Muller,
Seilacher etc)
Functionalists (e.g. me, Schwenk, Galis etc)

As opposed to:



Geneists
Populationists
Self-organisation
Self-organization explains flocking behaviour on
different levels
Local interactions cause flocking
behavior
1.
2.
Svim in the same direction as the neighbors
Don´t collide - modulate speed and direction
Patterns that arise through self-organization
Peculiarities of self-organizing systems
1. Many parts
2. Many interactions
3. Few and simple rules
4. No plan or directives from “above“
5. Emergence - novel properties arise in the system
6. Often modelled with cellular automata (Conway‘s Game of
life)
More stripes, same distance
Developmental
symbiosis
Squidbacteria
Nutritional polyphenisms - social insects
Functional rules
Cichlid jaws
Comparative anatomy
Frazzetta on system evolution:
“like making improvements to an
engine while the engine is running”
Why is there a problem?
Both the
developmental
genome and the
developing and adult
phenotype are
complex.
Complexity is hard to
define, but here is
seen as a measure of
internal structure that
provides resistance to
evolutionary change.
Genotype-Phenotype mapping
•Polyphenisms show that the same genotype can generate many different
morphologies depending on environment
•Yet we also know that different genotypes can generate the same morphology
(phenogenetic drift; genetic code redundancy etc).
•The evolutionary relationship between the two is thus likely to be complex!
Complex system change


Complex systems typically fail if changes are
randomly made to them because of their
integrated nature, even if locally the changes
”work”. In the phenotype, this can be seen
in terms of functional morphology, in the
genotype, in terms of pleiotropy
Partial recognition of this fact in the
genotype has led to an emphasis on
promoter evolution etc.
Modularity and interconnection


Generally considered that high degrees of
modularity (= many quasi-autonomous
units) makes evolution “easier” as
pleiotropic effects are reduced.
The opposite is interconnectedness: the idea
of the phylotypic stage was that it was a
developmental period of low modularity, as
genes had maximal global effect at this
period.
Developmental constraint: the
zootype?
Origins of modularity

But the origins of modularity itself are
highly controversial:


An effect of self-organisation, either in genetic
interactions or morphology?
Or imposed, perhaps by adaptive selection?
What does Cambrian evolution tell
us about evolution?
NOT like
this!!
- but rather,
functional
adaptation
Reminder of stem groups
Animals are…

Complex integrated organisms that appear to
show increasing specialisation and
compartmentalisation through time.

Study of how they evolve in the Cambrian
allow us to extract more general principles…
Organisms can be seen as
functional networks
How do we build integrated
structures?
QuickTime™ and a
decompressor
are needed to see this picture.
Fundamentals of morphological
evolution (?)

Redundancy

Preadaptation

Functional asymmetry

Least constraint
The many faces of redundancy

Multiple ways of performing the same
functions will enable functional shifts, so in
that sense complex systems may be more
evolvable than simple ones.

Not just one way of doing this though!
Types of redundancy
Amplification
Parcellation
Types of redundancy
{ {
{
Functional complex
Functional
degeneracy
Functional
take-over
Preadaptation: being in the right
place at the right time

Almost every novelty involves a functional
shift (classical example: middle ear bones of
mammals used to be jaw bones of reptiles)

But not everything is placed well to make
make such a shift.
Functional asymmetry


Even in a tightly
integrated system with
little redundancy, there
may be mutually
interacting components
that do not have equal
dependency on each other
E.g. what came first,
muscles or segments
(chicken and egg)?
Functional asymmetry


Can be represented on a diagram by arrows
showing interactions between components
Interactions between components are textured:
they have both direction and strength
Direction of
dependence
Least constraint

In any system, some components will be less
functionally constrained than others.

In theory it should be these components that
evolve first.
Example: bilaterian systems
Classical bilaterian features such as segmentation,
BVS, metanephridia and the coelom are not
independent, but relate to each other in an
asymmetric and partly definitional sort of way.
For example, a large, fully segmented animal requires
a BVS in order to transport oxygen etc around the
body, not rely on internal mixing.


QuickTime™ and a
decompressor
are needed to see this picture.
System dependencies
Segmentation
Coelom
Metanephridia
BVS
Implied order of acquisition: coelom; BVS; then segmentation and
metanephrida
Least constrained components can
evolve to allow redundancy to
develop…
L
P
L
P
…and therefore for new
dependencies to develop
L
Example: mammalian jaw
Burden is higher in the centre!
Preliminary conclusions



Fossil stem groups show that morphological
evolution is in fact governed by certain principles.
Note that in this view, “burden” is an evolutionary
property that can evolve in both directions,
although change in highly burdened characters
requires preparation in terms of shifting of
constraint as outlined before.
Almost all of these types of change apply to genetic
changes too (cf kernels etc!)
Towards a theoretical biology

Ecology, organismal structure (and
physiology) and genes form three distinct but
interlinked networks.

Can one understand how all three relate to
each other??
ecology
Evolution’s middle floor…
genes
Genetics of micro- and macroevolution
Macroevolutionary
innovations
(limbs, livers, etc)
????
“genetics of
macroevolution”?
?
Population variation
Population genetics
“problem of variation”
Body-patterning genes?
The problem:
Matching up
evolution in
morphology and
in the genome so
that functionality
is not
compromised
(Frazzetta etc)

Fig. 2. Examples of putative GRN kernels
E. H. Davidson et al., Science 311, 796 -800 (2006)
Published by AAAS
Once more!
QuickTime™ and a
decompressor
are needed to see this picture.
Relationship between developmental
genotype and morphology


Dev. Gen. best seen as a physical positioning
system: timing less important than eventual
positioning of various morphological elements.
Changes in the dev. gen. can thus be correlated
with positioning of morphological elements,
and thus the effect on the functional map of the
organism can be seen
QuickTime™ and a
decompressor
are needed to see this picture.
QuickTime™ and a
decompressor
are needed to see this picture.
Functional evaluation of
homeotic mutation
The developmental hierarchy?
How it works in fruit flies
High-level developmental changes




…tend NOT to be allowed by the structuralist
rules governing morphological evolution.
But they have clearly occurred anyway!
Suggests that the conditions have to be right
(just as in morphological evolution on its own)
i.e. out of the whole system, the phenotype, or
at least downstream genetic components, is the
least constrained component.
QuickTime™ and a
decompressor
are needed to see this picture.
Budd 2006
When might this take place?
• E.g. “genetic assimilation” of Waddington:
• Environmentally prompted changes can
become ”fixed” in the genome by
selection…
Developmental plasticity
Could take place at other levels…
Budd 1999, Bioessays
Genotype-Phenotype revisited (I)

Genotype and phenotype together form a
functional complex. Genes make phenotype,
but selection acts on phenotype, not genotype.
Hence the structuralist rules governing
evolution of phenotype (strictly morphology
here) take precedence over genetic change; it
provides a set of important boundary
conditions that genetic changes cannot cross.
Genotype-Phenotype revisited (II)

Phenotypic change can be sourced either from
subtle down-stream changes or even from
environmental change. If the latter is
consistent, then there is enough time for the
directive genotype to react to stabilise and
canalise the resulting changes.
Genotype-Phenotype revisited (III)
In this view, adaptive specialisation imposes
modularity on morphology.
 Morphology acts as a template around which genetic
structures are built.
 In a primary sense then, developmental genes are at
the whim of morphology, not the other way round, =
“genes are not important in evolution” sort of
conclusion.
However, one can still be highly adaptive in this view,
unlike the extremist structuralists.

General “conclusions”




Fossils are essential for reconstructing morphological
evolution - the framework within which all genetic change must
take place.
Ecology channels selective pressures onto
morphology - and only through this filter onto genes.
Morphological evolution is by far, and bizarrely, the
least understood component of the evolutionary
building we have erected in the last 150 years…