Fundamentals of Evolutionary Biology (Gompel) PDF

Observing, naming, describing and organising natural variation: From the 18th century on, naturalists have developed a system to describe and organise the living world; Initially a mere catalogue of God's creations, this system has grown into an explanatory framework implying de...

Observing, naming, describing and organising natural variation: From the 18th century on, naturalists have developed a system to describe and organise the living world; Initially a mere catalogue of God's creations, this system has grown into an explanatory framework implying descent with modifications; The description focused initially on the most visible manifestation of variation → morphological characters. From systematics to evolution: Less than a century after Linnaeus, Owen, Lamark, Haeckel, Darwin, Wallace and other acknowledge that species change over time; Their argumentation is based on the comparison and interpretation of morphological characters. Harnessing morphological variation: What is a morphological character? o Any three-dimensional material attribute (form) of an organism, or the organisation of the attributes (structure) that can be described or quantified: ▪ Body size or size of an organ; ▪ Shape of an organ, at any supra-cellular scale; ▪ Colour patterns, colour of an organ; ▪ Relative arrangement of organs, density; o Different from physiological character and behaviour. Harnessing morphological variation → characters that don't change (too much) define group. Morphological homology and the origin of characters:  As organisms evolve, their characters are modified. But when corresponding characters derived from the same ancestral character can be recognised among organisms, we refer to them as homologous characters; Morphological similarity is neither necessary nor sufficient to characterise homology: o The human arm and the whale pectoral fins are not very similar but are homologous; o The raptorial forelegs of Mantispids, Drynidae and mantises are not homologous: o The wing patterns of mimetic butterflies are not homologous to their models; Example of homologous characters → vertebrate forelimbs: o Quite variable among species, but always connected at the same level to the backbone; o General structure is well conserved among species (humerus + radius/ulna + hand): Example of homologous characters → vertebrate skulls: o Fossil shark, crocodile and human skull. Defining character homology (Wagner): Homology → the hypothesis that a similar trait shared by different species is derived from common ancestry; The criterion to assessing homology is the connection of a character to another one that can be recognised without ambiguity among different organisms; This works for morphological characters as well as for DNA and protein sequences: o For morphology → the reference is an unchanged and recognisable trait connected to the trait that we study, and used as a reference; o For sequences → blocks of conserved sequence flanking the sequence of interest are the reference; Morphology example blocks of conserved sequence flanking the sequence of interest are the reference; Morphology example → insect wings are homologous because they are all connected to the second and third thoracic segment (the thorax is recognisable in all insect and is always made of 3 segments bearing legs); Sequence example → sequence. Serial homology:  Serial homologues → repeated structures of a single organism that share a similar developmental origin; Classical examples → vertebrate limbs, insect appendages; Body segments have the same developmental origin (they result from the same developmental program deployed in different places. Organs connected to different segments (references) in similar ways are dimmed homologous, even if they look quite different. Harnessing morphological variation: Characters that differ are used to separate group; Example → the number of legs is one of the characters that define several classes of invertebrates: o Insects → 6; o Arachnids → 8; o Decapods → 10; Example → the morphology of mouthparts distinguishes birds (bill, no teeth0 from other vertebrates; Continuous characters variation and the limit of natural classifications. Patterns of morphological trait evolution: Ancestral → trait non present; Novelty → development of trait; Diversification → mutations happen: o Shape; o Trait properties (colour, texture); o Size; Loss → some traits are then loss if not suitable. Repeated/parallel/convergent evolution: Similar characters often appear independently in different lineages (they do not have a common origin); Example: o Resistance to toxic cardiac glycosides in toad-feeding reptiles; o Wings in birds, bats and pterosaurs; o Mimetic patterns of pigmentation; o White fur in polar bears and Kermod bears. Several outstanding questions in the face of these patterns of variation; How are these characters formed? o Which and how many genes, which biological processes? Trends? Principles? o Every evolutionary transition is a so-so story, a particular case. But can we identify commonalities in the underlying processes? Are their biases or biological constraints that shape morphological diversification?  If we were to rewind the tape of evolution and play it again, would we get the same output of diversity? o The case of repeated evolution offers a unique possibility to address this question, by comparing the processes that lead to the formation of similar traits. Are the same genes involved? Are the same principles at stake in the underlying mechanisms? Morphological differences find their origin in embryonic or post-embryonic development. Functionalism/adaptationism vs. Structuralism: There is a long tradition in evolutionary biology to explain changes by the functional requirements → feathered wings facilitate flight, we can easily understand how they have been selected, as they endow an animal with special powers; A functionalist perspective has powerful explanatory power in evolution → it is relatively easy to imagine evolutionary advantages to many traits. It is much more difficult to demonstrate the existence of these advantages. Most often, the advantage, the adaptation, is assumed rather than demonstrated; A functionalist perspective, however, carries two issues: o It easily calls for finalist or ad-hoc explanations; o It does not explain the ontology (the construction, the development) of traits; Structuralism is concerned with the underlying structure of traits, regardless of their possible function or adaptive advantage. Finalism is creeping in the adaptationist perspective:  A typical example from a recent paper in evolutionary biology → throughout their evolutionary history, organisms have evolved numerous complex morphological, physiological, and behavioural adaptations to increase their chances of survival and reproduction. Insects have evolved wings and flight, which allowed them to better disperse, beetles have grown horns to fight over females, and moths and butterflies have decorated their wings with bright circles of coloured scales to scare off predators. The way that most of these and other adaptations first evolved, however, is still largely unknown; Phrasing ideas in a finalist way seriously biases our representation of the evolutionary process: o Natural selection is the process through which the better endowed is at an advantage. This is very different from saying that organisms become better endowed to be at an advantage. An organism is better endowed for no reason, just by any chance and random mutations. If it is, it may survive better. The spandrels of San Marco, a blunt rebuttal of adaptationism: In a virulent essay, Stephen J. Gould and Richard Lewontin criticised the widespread trend of evolutionary biologists to always assume that all traits have a function and are under selection; Using the analogy of spandrels in cathedrals, which must exist as a by-product of the architecture of the building, they argued that many morphological traits exist not because they are under selection, but merely as by-products of a developmental program; They compared the adaptationism to a Panglossian program, in reference to Voltaire's book Candide: o "Everything is made for the best purpose. Our noses were made to carry spectacles, so we have spectacles. Legs were clearly intended for breeches, and we wear them". Yet evolutionary biologists, in their tendency to focus exclusively on immediate adaptation to local conditions, do tend to ignore architectural constraints and perform just such an inversion of explanation;  They also highlighted that with a focus on adaptation, evolutionary biologists have abandoned the underlying developmental programs governing trait formation. Richard Owen and the structuralist perspective of evolution: Structuralism, pioneered by Richard Owen (1849, The Nature of Limbs) is concerned with the underlying structure of traits and the "unity of type"; Owen considered limbs involved in locomotion and noted that: o Vertebrate and invertebrate limbs achieve similar functions with different structures; o Vertebrate limbs with different specialisation have the same structure; o Owen also makes a comparison to human tools achieving similar functions as vertebrate limbs (swimming, flying, burrowing, etc.) and reflects that they have different structures → a paddle, a plane, a shovel; o In contrast to the human devices, all vertebrate limbs have a common underlying pattern, with corresponding bones. Owen concludes that these commonalities cannot be explained by functional necessities; o Owen did not know the origin of these commonalities, but structuralism calls for the developmental origin of traits; o Owen referred to the structural commonalities as the "unity of type", the idea that apparently dissimilar structures across organisms, possibly with various functions, may share fundamental similarities in their organisation and construction. From morphology to genes, structuralism runs deep: Structuralists established morphological correspondence among distantly related organisms, but what about the underlying components? Before the advent of developmental genetics, it was easy to imagine that the eyes of cats, lions, tigers, and perhaps even humans were produced by very similar genetic instructions; But few imagined that gene sets that build organisms would be comparable over long evolutionary distances; The reckoning that genes governing development for a universal toolkit paved the emergence of evolutionary developmental biology. The homeobox and the birth of molecular structuralism: Homeobox → a DNA sequence, around 180 base pairs long, that regulates large-scale anatomical features in the early stages of embryonic development. Mutations in a homeobox may change large-scale anatomical features of the full-grown organism; William Bateson (1894) introduced the term "homeotic" to identify variation in which "something has been changed into the likeness of something else"; Such transformations emerge through spontaneous or induced mutations; Biologists have long thought that they revealed something important about the developmental specification, without knowing how;  In the 1980s, Drosophila geneticists revealed that genes responsible for these transformations were transcription factors with a conserved domain, encoded in the so-called homeobox; In insects, homeotic genes organised in two complexes all result from tandem duplications of an ancestral gene; Mutations in their sequences produce partial or complete transformations of one segment into another segment; These genes encode transcription factors characterised by a similar DNA- binding domain, the homeodomain (encoded in the homeobox); In vertebrates, homologous genes, organised in 4 complexes, also result in segment transformation when mutated. Conserved sequence and function of the Pax6 gene over long phylogenetic distances: The unique origin of eyes among bilaterial was long debated, because of their structural differences; The gene Pax6, a transcription factor, governs eye development in Drosophila; Its downregulation results in eye loss, its overexpression in ectopic eyes; But Pax6 sequence and function are conserved among Bilateria and its expression always prefigures eye formation; Eye structure has simply diverged over time, but the bilaterian eye has a unique origin; Pax6 is a good example of conserved development toolkit gene. Genes controlling cell division: Regulated cell growth is an essential part of embryonic development; Major progress to understand the control of the cell cycle were made in the 1980s studying the fission yeast (Schizosaccharomyces pombe); No one then assumed that the genes involved in yeast mitosis had anything to do with their counterpart in metazoans. The yeast system was used as an analogy to understand how cells divide; Until the group of Paul Nurse discovered that homologous genes were involved in controlling cell division across Eukaryotes. The genetic toolkit for development: The toolkit is composed of a small fraction of all genes → only a small subset of the entire complement of genes in the genome affects development in discrete ways (control of embryonic development); Most toolkit genes encode either transcription factors or components of signalling pathways → therefore, toolkit genes generally act, directly or indirectly, to control the expression of other genes; The spatial and temporal expression of toolkit genes is often closely correlated with the regions of the animal in which the genes function; Toolkit genes can be classified according to the phenotypes caused by their mutation → similar mutant phenotypes often reflect genes that function in a single developmental pathway. Distinct pathways exist for the generation of body axes, for example, and for the formation and identity of fields; Many toolkit genes are widely conserved among different animal phyla. Morphological differences find their origin in embryonic and post-embryonic development: There is little we can infer about the construction of a character once it is formed; Morphological differences appear during development, resulting from divergent cell behaviours, growth, or differentiation: Drosophila suzukii, a pest species with an enlarged ovipositor. Patterns of digit loss as first experimental evidence for the casual role of developmental program to shaping evolution: Tetrapods typically have 5 digits per limb, but this number is occasionally reduced in some species to 5, 3, 2, 1 or none; Digit loss always follows the same order → 1, 5, 2, 3 and 4; In Urodeles (salamanders, newts), however, the pattern of digit loss differs from all other Tetrapods → 5, 4, 3; Where is the difference coming from? Function (natural selection) or structure (developmental program)? Functionalists would suggest that urodeles use their digits in a fundamentally different way than other tetrapods (eu-tetrapods), and this explains the contrasting patterns of digit loss; Pere Alberch and Emily Gale had a different hypothesis → fundamental differences in the respective development programs of digit formation;  To test this possibility, they experimentally induced digit loss in frogs and in salamanders using the same experimental treatment, and examine whether it resulted in different patterns of digit loss; The treatment consisted in inhibiting cell proliferation at a critical stage of limb bud development, either in Ambystoma (salamander) or in Xenopus (frog); They found that the pattern of loss was different between Ambystoma and Xenopus and paralleled the patterns of evolutionary digit loss between urodeles and eu-tetrapods; Hence, they showed for the first time, that differences in a developmental program, rather than in a character function, was the main determinant of evolution. Evolutionary Developmental Biology (Evo-Devo), a modern version of comparative embryology: Understanding the origin of evolutionary changes in multicellular organisms → at the crossroads of traditional disciplines: o Developmental biology; o Genetics; o Evolution. For an evolutionary transition, which evo-devo questions? What cellular changes underlie the evolutionary diversification of morphology?  What genetic changes underlie the evolutionary diversification of morphology? From heredity to developmental biology: Upon the rediscovery of Mendel's laws, understanding of the process of inheritance became a central focus of Biology; Thomas Morgan and his students turned the abstract laws of Mendel into a very concrete problem, identifying chromosomes as the substrate of genetic information, and physically mapping the first genes; Their endeavour leaned extensively on the analysis of mutants, whose phenotypes informed not only about inheritance, but also about how an organism is built; The distinction between genotype and phenotype was first put forward by Wilhelm Johannsen in 1909; Between genotype and phenotype lies the entire process of development. Germ cells, soma, and evolution: Preformationists believed that a homunculus was curled up in the head of each sperm; The "cell theory", which emerged between 1820 and 1880 fuelled progress in developmental biology. It recognised cells as the basic units of life and acknowledge that new cells are formed by the division of pre-existing cells. Walther Flemming → discovered mitosis (1878-1882). Germ line → soma segregation: In each generation, a germ cell contributes to the zygote, which gives rise to both somatic cells and germ cells, but inheritance is through the germ cells only. How does the information contain in the fertilised egg segregate during development? The cells of a fully developed organism are different in shape and function; The first, the main question in developmental biology is → how does one cell become two different cells? Weismann's theory of nuclear determination (1880s);  The fundamental notion of asymmetric segregation of a determinant upon division was born; Yet, Weismann's theory appears to imply that the information is diluted as cells divide. Conflicting experimental results to test Weismann's hypothesis: Wilhelm Roux's experiment tends to support Weismann's idea; Hans Driesch's results contradict Weismann's hypothesis and first show the phenomenon of regulation. Model organisms to explore mechanisms of development: Model organisms offer definition and lack of variation essential due to understand and interpret developmental processes. Labelling cells to track them: By injecting a vital fluorescent marker → lipophilic dyes, fluorescent dextrans; By driving the expression of a reporter gene. Hijacking gene regulation to label cells → reporter constructs (allows the study of gene's function and localization of a gene product. The promoter reporter constructs allow a protein to be expressed under the control of a target gene): What we can do with reporter genes: Marker of transformation for microorganisms; Reveal gene expression patterns in multicellular organisms: o Enhancer traps; o Reporter constructs; New generation of fluorescent reporter genes → Green Fluorescent Protein (GFP), isolated from the jellyfish Aequorea victoria. GFP and other fluorescent proteins opened new perspectives for biologists: Seeing live cells; Co-localisation (used to determine if a protein is localizing to an organelle or other well defined cellular structure); Gene expression quantification. Immunochemistry → revealing gene products in context: In the early 1980s, the distribution of gene products, mRNA and proteins started to be revealed for the first time directly in the embryo; For mRNA → in situ hybridisation; For proteins → antibody staining. Graft and cell transplant, a powerful approach to reveal cell development potential, fate, and interactions: Development is progressive and the fates of cells become determined at different times;  Cells from a wild type donor transplanted in a pigmentation mutant result in tractable tissue in the adult; This trick is used to evaluate the degree of specification of cells at different developmental stages. We can observe and describe what happens: Anatomical changes; Cells changing shape, moving; Intracellular changes, including molecular changes. Life cycle and the origin of morphological characters: The life cycle can be considered a central unit in biology. The adult form needs not to be paramount (more important). In a sense, the life cycle is the organism; In Bilaterians, the basic life cycle consists of fertilisation, cleavage, gastrulation, germ layer formation, organogenesis, metamorphosis, adulthood, and senescence; Shapes visible in adults start to form during development, sometimes very early; At the cellular level, there is a succession of phases from egg cell to the adult organism → growth, patterning, migration, differentiation and morphogenesis. The broad phases of embryonic development: Growth → cells increase in number (proliferation) or in size, or both;  Patterning → a coordinate system. Cells look similar but accumulate invisible differences in gene expression based on their position in the embryo. These differences determinate their fate. The embryo acquires an invisible blueprint of the future developed organism; Cell migration and morphogenesis → relative movements and rearrangements of cells to each other. These movements contribute to giving a 3D shape to organs or the entire embryo;  Differentiation → cells change their form, secrete material, activate various enzymes, sometimes become mobile, or produce electrical and chemical signal; These different phases are not always separate in time. Rapid cleavage (growth) during early Bilaterian's development: The egg cell undergoes more or less rapid divisions resulting in the so-called blastula stage; The divisions are often polarised, defining axes (anteroposterior, dorso- ventral); The divisions are synchronous in some organisms, no in others; they are symmetrical in some organisms, not in others. Presumptive maps of early vertebrate embryos: Blastula cells have their general fates determined, as shown with the presumptive map of blastula in frogs; The fate maps of vertebrates are variations on a basic plan. Gastrulation, a signature of Bilaterian's development: From a single layer of cells at the blastoderm stage, morphogenetic cell movements produce three cell layers: o Ectoderm; o Mesoderm; o Endoderm; Subsequent interactions between the layers determine their future fate; A group of cells at the edge between the animal and the vegetal poles invaginated into the cavity of the blastula. After gastrulation:  Gastrulation in deuterostomes results in three germ layers → endoderm, mesoderm, ectoderm; The three germ layers undergo massive remodelling and morphogenesis, giving rise to different organs and derivates. The animal-vegetal axis is maternally determined in Xenopus: The maternal growth factor Vg-1 mRNA is distributed asymmetrically in the unfertilised amphibian egg and influences the orientation of the first division; The future dorsal side of the amphibian embryo develops opposite the site of sperm entry as a result of the redistribution of maternal dorsalising factors by cortical rotation; The animal-vegetal axis of the egg is related to the antero-posterior axis of the tadpole, as the head will form from the animal region, but the axes of the tadpole and the fertilised egg are not directly comparable. The discovery of cell induction → the Spemann-Mangold organiser (phenotype- based tool that can be used to browse the effect of your selection of genes on body parts): In 1924, Hans Spemann and Hilde Mangold carried out a revealing transplantation experiment in amphibian embryos. Mesoderm induction; The formation of the mesoderm results from interactions between cells of the animal and the vegetal poles. Neurulation: The nervous system of Xenopus is induced during gastrulation by the notochord; The ectoderm overlying the notochord rolls upwards on both sides of the dorsal midline to form the neural folds, which meet over the midline to form a tubular structure, the neural tube. Vertebrate segmentation → sequential somite formation:  Somites, derived from the paraxial mesodermal plate after gastrulation, are added sequentially from anterior to posterior in all vertebrates. The clock and wavefront model: Two opposite signalling gradients meet in the presomitic mesoderm, retinoic acid and FGF/Wnt; In parallel, cycles of gene expression sweep through the mesoderm from posterior to anterior; During each cycle a pair of new somites is specified. Differentiation:  Patterning of the embryo provides a blueprint, a positional canvas that will guide cell differentiation; The invisible patterns are later revealed in the form of phenotypes upon cell differentiation; Cell differentiation generally means a massive rearrangement of the cytoskeleton leading to cell shape changes; Cell differentiation can also involve changes in planar polarity, secretion of a matrix (collage, cuticle, cellulose), or the production of specific enzymes (toxins). The phases of early embryogenesis repeat themselves during organogenesis. Organogenesis almost always consists of four processes: Tissue growth, regulated in space and time; Patterning → the progressive, but invisible, assignment of identities to cells in different locations; Morphogenesis and/or migration → the relative reorganisation of cells in space (mesenchyme-epithelium transition, invagination);  Differentiation. Growth at later stages → limb bud in the shrimp Parahyale: Cell proliferation is a regulated process: Proliferation is defined in space and time, unlike for instance tumour growth; Cells in a tissue somehow know when they should stop growing;  In some cases, cells "count"the number of divisions they undergo, and stop at a fixed number → fixed lineage; In other cases, cells exploit gradients of a gene product to "read" their position in the gradient. the position changes with growth and regulate growth. Cell migration: Group migration → lateral line primordium; Individual cell migration → neural crests. Developing cells stick together, talk to each other, instruct each other. Similar cells recognise each other: Old dissociation experiments show that cells of similar origin have affinity for each other. Tissue cohesion; Different types of cell junctions are associated with different molecular complexes, anchored by transmembrane receptors. How do developing cells decide what to do next? Where are the instructions coming from? The starting point of all developmental processes → one cell becomes two cells, usually distinct from each other; These decisions are regulated through cell determination (inheriting information) and cell communication (signalling); Both processes lead to a particular transcriptional state that determines the next developmental steps. Asymmetric cell division; Cell determination happens through the asymmetric distribution of a particular determinant. Cells communicate through surface molecules: Cell-cell dialogue includes reception of diffusing ligands, interaction between transmembrane proteins, and direct exchange of molecules through gap junctions. Signal transduction: A receiving cell triggers a signalling cascade that results in changes in gene expression. T. H. Morgan → a prophetic perspective for developmental biology: Between the characters that are used by the geneticist and the genes that Mendel's theory postulates lie the whole field of embryonic development, where the properties implicit in the genes become explicit in the protoplasm (all the material inside a cell) of the cells. A paradox explicated: If, as generally implied by genetic work (although not often explicitly stated), all of the genes are active all the time; and if characters of the individual are determined by the genes, then why are not all the cells of the body exactly alike? The same paradox appears when we turn to the development of the egg into an embryo. Every cell comes to contain the same kind of genes. Why then, is it that some cells become muscle cells, some nerve cells, and other remain reproductive cells? A visionary statement: The answer to these questions seemed relatively simple at the end of the last century. The protoplasm of the egg is visibly different at different levels. The  fate of the cells in each region is determined, it was said, by the differences in the different protoplasmic regions of the egg. Such a view is consistent with the idea that the genes are all acting. But there is an alternative view that can not be ignored. It is conceivable that different batteries of genes come into action one after the other, as the embryo passes through its stages of development. The developmental capacity of nuclei taken from intestinal epithelium cells of feeding tadpoles: Is there a dilution of genetic information as a cell divides (during embryonic development)? DNA sequence is identical in most somatic cells; What, then, is the molecular correlate of different cell fates? RNA competition experiments → change in RNA composition as development proceeds and between cell types; Cell with the same DNA can have different RNA. Finding all genes involved in a developmental process → saturation mutagenesis: Cells have the same DNA but their RNA content changes as development proceeds. What are these genes, expresses sequentially, which presumably govern developmental process? To reveal these genes and their functions, biologists use mutations; To find all developmental genes, the logic is to mutagenize the genome of an organism, screen for mutants, map the corresponding gene, until we find no new genes. Or at least until we keep getting the same genes. This is called a saturation mutagenesis. Introducing perturbations to understand which genes control development. From mutant phenotype to gene expression: The invention of whole-mount in situ hybridisation shed a new line on the function of developmental genes only known by mutant phenotypes; e.g. Ultrabithorax (ubx) specifies the identity of the third thoracic segment. Segmentation genes are expressed in segmental patterns: Fushi tarazy expression pattern;  Visualising gene expression matters because it allows us to virtually connect what genes do, where they do it, and a final result (a phenotype). Spatial gene expression correlates with trait formation: The spatial expression of patterning developmental genes is often a surrogate of the morphological character that they contribute to form; Rather than focus on final physical forms, which are often arrested prematurely and disfigured in lethal mutants, the patterns in which key developmental genes are expressed at different developmental stages can be used as surrogates for the ultimate form. A word of caution → correlation vs causality: Let's consider two events, A and B. we observe that they happen at the same time. How can they relate to each other?  o A triggers B (A causes B) → e.g. Pax6 causes the formation of eyes; o B triggers A; o A and B are independent results of C; o A have no correlation with B; In developmental genetics: o Loss-of-function data suggests that A causes B, but other explanations are possible → if A is experimentally removed, B does not occur; o Gain-of-function data is most convincing → if A happens where or when it does not usually occur, then B also happens in this new time or place. Spurious correlations are commonplace: How do developmental genes control the formation of a developmental blueprint? We will examine the principles of embryonic patterning were unravelled through a classic example → the set-up of a body plan in Drosophila; Flies are segmented, as many Bilaterians are. Segments are formed very early on, during embryonic development. A genetic screen to identify genes controlling segmentation: A simple readout for Drosophila segment identity and number → denticle belt; These cuticle productions are stereotypical and segment specific;  In the late 1970s, Christiane Nüsslein-Volhard and Eric Wieschaus initiated a mutagenesis screen, looking for defects in the embryonic segmentation; Their screen was a saturation screen → they mutagenized the genome and looked for multiple mutant alleles of each kind. The reasoning was that they would thereby not miss any gene involved in segmenting the embryo. Types of mutations: Mutations in developmental genes are often embryonic lethal and can't be studied; Dominant lethal mutations can't be studied; Recessive lethal mutations can be maintained as heterozygous and studied in homozygous individuals until they stop to develop. Screen results → 15 loci falling into different classes of phenotypes affect segment number and segment polarity: All these genes are therefore involved in the patterning of the embryo → they somehow establish the blueprint of their body plan. Other mutants affect the antero-posterior axis formation: Bicoid mutants form posterior structures instead of a head and lose anterior segments; Note that bicoid phenotype is found in embryos whose mother is homozygous mutant for bicoid → bicoid gene product must be provided maternally. Polytene chromosomes to visualise deletions: With Drosophila, giant polytene chromosomes can be used to reveal relatively large deletions and chromosomal rearrangements, and can help decide if a mutation is a loss-of-function. Classical genetics to order the interactions of segmentation genes: A few concepts to define relationships at the genetic level: o Loss-of-function (amorphic) mutation → a mutation resulting in a non- functional gene;  o Gain-of-function → a mutation that confers new or enhanced gene function; o Hypomorphic mutation → a partial loss-of-function mutation; o Epistasis → means that the mutant phenotype caused by one gene can't be seen in the presence of mutations in another gene; These relationships can be tested directly by combining mutations: o This phenotype happens because B is triggered by A, but if A is altered (lof), then B is not going to develop and does not weight in phenotype. A genetic cascade for segmentation: The determination of epistatic interactions among segmentation genes resulted in a genetic cascade controlling the slicing of the embryo. From genes to cell identity → how do embryonic cells sort themselves into segments? The hierarchy of Nüsslein-Volhard and Wieschaus are abstract. they tell us that there is likely a temporal genetic specification, but how it relates to cell along the embryo is elusive; What's missing is the expression of the genes in space and time during embryonic development. A necessary detour by the early stages of embryogenesis: Cleavage → 13 rounds of synchronous rapid nuclear division in a syncytium; Nuclei migration to the membrane; Cellularisation. Axis patterning genes at the top of the segmentation cascades are expressed in gradients → different distribution of proteins along the body: Antero-posterior axis → bicoid; Dorso-ventral axis → dorsal. Putting faces on a genetic cascade → temporal sequence. Explaining pattern regularity with invoke Reaction-Diffusion models (Turing models): Modelling the Drosophila pair-rule pattern by reaction-diffusion → gap input and pattern control in a 4-morphogen system: o The uniform spacing of stripes, despite underlying peaks and troughs of gap gene expression, shows that pattern wavelength is relatively insensitive to parameter change, also a property of the model. A simple mathematical model to explain the formation of patterns: Reaction-diffusion models are predictive of actual forms, but not formally tested: The biological reality is else → an unexpected mechanism of pattern formation: Modular organisation of cis-regulatory elements: Short stretches (∼5–15 base pairs) of DNA capable of being bound by a transcription factor and influencing the expression of nearby genes; The notion of enhancer modularity in the late 1980s, with reporter constructs, largely with the work of Michael Levine's group. How genes are produced in specific places: The eukaryotic gene model, beyond the coding sequence (CDS): Transcriptional enhancers structure and function: Enhancers are distinct genomic regions that contain binding site sequences for transcription factors and that can upregulate the transcription of a target gene from its transcription start site. Converting a morphogen gradient into discrete expression domains through enhancers: A morphogen (transcription factor) activates expression of target genes; Target gene activation, and therefore the patterning output, depends on: o The concentration of morphogen; o The affinity of transcription factor binding sites in target genes; A gradient is also converted into discrete domains; In low affinity binding sites, there is a threshold where the sites stop being activated because the concentration of binding molecules is too low. The French flag model → explaining the conversion of a continuous gradient into discrete gene expression domains: The concentrations of morphogen at both ends are kept constant but are different from each other; The concentrations at any point of morphogen provide coordinates for the cells; If the cells respond to threshold concentrations of the morphogen (for example, above a particular concentration the cells develop as blue, whereas below this concentration they become white, and at yet another, lower concentration, they become red) the line of cells will develop as a French flag; If you include a high affinity binding site, the blue cells will increase since less morphogen concentration is needed to trigger the response. Pair-rule genes integrate positional information provided by maternal and gap genes: Conclusion on segmentation: Segmentation is a visible morphogenetic process determined by gene products acting before segments are visible; Segmentation gene product distribution prefigures the actual segment morphologies; Gradients of maternal information are converted into discrete domains of gene expression, step by step: o Maternal genes;  o Gap genes; o Pair-rule genes; o Segment polarity genes; The conversion happens at the level of cis-regulatory elements and depends on how they read the positional information; The interpretation of positional information ultimately depends on the number, affinity and strength of the transcription factor binding sites; Most of the segmentation genes are transcription factors. Dll expression and the specification of insect appendage: Distal-less (Dll) is a homeodomain (homeobox → DNA sequence that regulates large-scale anatomical features) transcription factor expressed during development in different tissues in all Metazoans; Distal-less has an ancient and conserved function in the specification of appendage development; In insects, Dll expression is one of the earliest events of appendage specification, long before the expressing cells remotely look like an appendage; Wing primordia separate from the early limb field shortly after the onset of Dll expression. Wing is a bilayer caused by folding of two layers during embryo development. A coordinate system determines Dll expression in the limb field: Appendage precursors appear at the intersection of 2 positional information established during the early embryogenesis: o Segmental expression of wingless, a segment-polarity gene; o A narrow strip of decapentaplegic (dpp) along the dorsoventral axis; The "appendage precursors" cell fate is repressed on most segments by Hox TFs; This results in Dll expression in 20-30 cells that form the appendage primordia. The appendage primordium, at the crossroad of genetic identities: The appendage primordium is determined by the intersection of dpp and wg expression; The appendage primordium inherits an antero-posterior axis from the segment polarity. Early Dll expression in the leg primordium results from the activity of an identified enhancer: This enhancer (Dll-304) was identified by screening the Dll locus with reporter constructs. How does the Dll leg enhancer specify localised, segment-specific expression? Early during embryogenesis, the segmented embryo is regionalised along the A/P axis by the expression of Hox genes;  The Dll-304 enhancer contains and activation and a repression domain → without repressing domain the active domain would be expressed everywhere: The activation domain responds to the wg and dpp pathways; The repression domain combines segmentation (En, Slp) and Hox inputs, 2 different kinds of positional information. Clockwork of an integrating device: A composite enhancer with a repression and an activation domain; Hox TFs promote enhancer repression on most segments; DMX repression is mediated in a compartment-specific manner. Hox co-factors and extended specificity of DNA binding: Hox genes have narrow DNA binding targets (6 bp), with very similar sequences; How then, does a hox target CRE respond differently to different homeobox TFs? o Co-factors, such as Extradenticle or Homothorax increase target size and specificity by acting only with certain hox TFs; Exd and Hth are DNA binding proteins that also interact with some, but not all of genes through specific protein-protein interactions: o They work like "specificity plugins". Setting a new axis with transcriptional tools: The activity of Dll-304 is transient and fades after a few hours; Dll expression is however maintained by the activity of 2 other CREs (cis- regulatory elements) taking over → DKO and LT;  DKO and LT decouple Dll expression in two cell populations (levels, timing) and together with other genes contribute to the birth of a third axis → proximo- distal. The genetic potential for wing formation exists in every segment but is repressed by homeotic genes (genes that regulate the development of anatomical structures) in T1 and abdominal segments: This is in contrast with the fate of the leg primordia that is repressed only on abdominal and head segments. The wing primordium, at the crossroad of genetic identities: Homeotic gene expression superimposed to the segmental blueprint determines which segments form wings;  Wing primordia are restricted to segments T2 and T3 by Hox genes repression. From axis formation to appendage patterning → the case of wings: After the onset of Dll expression, we are left with appendage primordia, each polarised by 3 axes. These are secondary fields of the developing embryo; Each secondary field will undergo 3 developmental stages: o Acquisition of an identity, imparted by selector genes and the hox ground plan; o Growth and patterning → positioning of morphological elements through a new coordinate system; o Differentiation; The unfolding of these steps is based on transcriptional mechanisms. Axes of the wing: The wing primordium is a single layer of cells, the wing is a bilayer of cells;  How does this happen? A fly wing is a very stable form, with very defined proportions and features; What determines the wing axes and controls the wing pattern? From a single layer to a double epithelium: The wing pouch (single layer) everts like a sock and flatten on itself. The wing primordium is selected during early embryogenesis: The wing primordia are singled out from the embryonic epidermis at the end of segmentation; Their selection depends on the positional blue print of A/P and D/V information available at the end of segmentation; This wing fate results from wing cells expressing the genes vestigial and scalloped. Vestigial is necessary for wing formation → in absence of vestigial function, the wing does not develop. Vestigial and the selection of a wing field: The axes are inherited from early embryonic gene expression (dpp, wg, en); These axis genes determine the expression of the dorsal appendage selector → vestigial. From genes to cells, from cells to morphogenesis → the wing model: The fly wing is a simple example of what a developmental process can produce; As in the case of the fly embryo, the wing likely results from interactions between genes and interactions between cells; Staring at the finished wing, however, does not tell us much about how it was built; To probe into the information of wings, two general approaches can be taken: o We can study the behaviour of cells in normal and mutant context;  o We analyse wing development in terms of gene expression and reconstruct the genetic instruction guide to make a wing. Are these axes real or arbitrarily defined by us? One way to answer the question is to examine the behaviour of cells during wing development → can we identify natural structures in how cell organise that correspond to axes? Wing development does not involve cell migration. It is just a field of cells that divide and push each other; Therefore, if we could mark a group of cells in the wing primordium and just look where they end up in the adult wing, we may learn something about wing patterning; There is a way to mark cells at defined developmental times and see where they end up in the adult wing → clonal analysis, invented by García-Bellido. Turning comparative embryology into developmental genetics → Antonio García- Bellido, with others like Ed Lewis, Eric Weischaus, Christian Nüsslein-Volhard or Peter Lawrence, has reformulated the classical questions of embryology in terms of cells nd their genetic specification. Clonal analysis: Cells can be marked genetically by making them homozygous for a recessive mutation as they grow; Mutate a cell and check where the mutation is shown during development: o The earlier the mutation is induced, the more cells are produced from the original one; Way to mark a cell and track its development. Clones and developmental time → clones induced early are fewer but bigger. Adding cell competition to the picture: A growth advantage can be conferred to the cells of the clone by exploiting the Minute mutation; This gives big clones, but also reveals rules of cell growth; Even if you give a growth advantage to some cells, the final product is the same because other cells will compensate. Cells communicate and regulate each other. Revealing a construction rule → compartment in development: In their early experiments, García-Bellido and his team observed M+ clones that followed straight lines and preferential growth patterns; In particular in the wings, cells seemed to never cross an invisible line dividing the wing in two compartments → anterior and posterior. Lessons from clonal analysis: Cell growth rate; Cell sorting → salt and pepper pattern; Growth boundaries → smooth clone edges; Growth direction; Fixed lineage. The discovery of compartments prompts two questions: What is the genetic determinant of compartments? What is the biological significance of these boundaries? A gene to give cells a compartmental identity: Clonal analysis in the context of engrailed mutation shows the necessity of this gene for cell sorting along the AP boundary; Engrailed is required in the posterior compartment only. From compartment to compartment selector gene:  There are 4 compartments in the wing (because it is a bilayer): o Anterior dorsal; o Anterior ventral; o Posterior dorsal; o Posterior ventral; The compartmental identity is defined by the expression of selector genes: o Engrailed → posterior; o Apterous → dorsal; The juxtaposition of compartments creates boundaries between cells of distinct genetic identities; This boundary is soon turned into a signalling centre; Engrailed is a transcription factor, its expression releases Hh → signalling molecule. Engrailed expression creates a boundary; Cells in close contact with engrailed expressing cells, express dpp: o Dpp concentration and engrailed presence/absence gives direction to cells where to locate; Antero-posterior and dorso-ventral boundaries work in a similar way. Boundaries as signalling centres: For instance, the AP boundary signal at short and long range, shaping the wing along the AP axis. Similar principles using alternative signalling pattern the DV boundary: Apterous, the selector gene for the dorsal compartment is responsible of setting the DV boundary as a signalling centre. From field identity to patterning → recycling the old morphogen trick: Once the AP boundary exists, dpp is produced there and diffuses, creating a symmetrical gradient; The dpp gradient is read by target genes responding to different concentrations, thereby creating different domains in the wing;  These domains refine and prefigure wing elements at determined positions, such as veins and sensory organs. Regulation of gene expression and embryonic development: Embryonic development is similar to the building of a house; When a house is built, different kinds of workers take action at different times, in a coordinated manner (architects, builders, wood workers, plumbers, designers, painters, etc); Embryonic development requires different kind of genes: o "Worker" genes; o "Selector or architect" genes that coordinate and regulate the formers; A set of architect and worker genes define the identity of the cell expressing them; Worker genes → cytoskeleton genes, enzymes affecting what cell do, any gene involved in differentiation. Activated by architect genes (TFs). Cell specification, cell identity, profiling RNAs: The notion of cell identity is an important concept in developmental biology;  The identity of a cell determines what a cell does, how much it divides, how it interacts with its neighbours, reads surrounding molecular signals, and also what it looks like; At a given time, the identity of a cell can be defined by all the genes it expresses → its transcriptome; Nowadays, we can profile the transcriptome of each cell of an embryo, using single-cell transcriptomics, and we can compare these transcriptomes; Single-cell sequencing → sequencing of a single cell material to observe all RNAs present (individual cell specific identity). Cell content is marked and compared with other cells to understand if the new sequenced cell has different identity. Single-cell transcriptomes reflect developmental trajectories: As development proceeds, the set of genes expressed by different cells diverge; Sister cells have more similar transcriptomes than cousin cells; A large set of co-expressed gene is a mark of a given cell-type, but this remains a somewhat arbitrary notion. Transcriptomes are the output of Gene Regulatory Networks: The gene regulatory networks that produce the progressive fragmentation of the embryo into domains have been formalised mostly in terms of regulatory elements by Erich Davidson → Davidsonian networks; It is however becoming clear that other levels of regulation play a major role → at the level of the core promoter. Revisiting Morgan's vision in the light of transcription: "But there is an alternative view that cannot be ignored. It is conceivable that different batteries of genes come into action one after the other, as the embryo passes through its stages of development"; This coordinated action of batteries of genes is primarily controlled in space and time at the level of transcription. A blueprint of gene expression turned into actual forms through the coordination of spatial and temporal information:  At the genetic level, development proceeds through the progressive refinement of positional information; This positional information exists in the form of gene expression patterns; At the beginning of development, the embryo can be seen as uniform a field of cells; Expression of early patterning genes parses the field into broad domains. Each domain becomes fragmented into smaller domains, and so on; This is reminiscent of a pixelated image that becomes crisper over time. Conclusion on patterning: All the developmental processes are happening step by step, with one or a few genes governing each step, but influencing hundreds of other genes; Developmental gene expression patterns are surrogates of physical pattern elements, much before these elements become visible; Together, you could summarise development in 3 big steps: o Growth: o Establishment of an invisible blueprint; o Cell differentiation (visible). Genomes, a new kind of zoo confirming the existence of a shared gene toolkit for animal development: The conservation of toolkit genes (genes aimed at creating the structure of the animal) in fruit flies and mice, based on few examples (Pax6), anticipated the findings of comparative genomics. Essential components of the gene toolkit predate the emergence of multicellular organisms: Choanoflagellates, the closest living relatives of animals are unicellular organisms that can occasionally aggregate into small multicellular forms; As the sister group of animals, their genome offers insight into the origin of the development gene toolkit; choanoflagellates possess many adhesion and communication molecules, such as cadherins, Notch, Delta, and homologs of the animal Toll-like receptor genes. They also possess signalling pathway genes and transcription factor families essential to animal development; this indicates that genes essential to animal development existed before the emergence of animals. Gene duplication, a key molecular step toward organism complexity: Genes of the developmental toolkit have likely been duplicated in last common ancestor of animals, or early during Metazoan evolution; The expansion of gene families during animal evolution can be traced by comparing the genes found in extant organisms and by mapping gene duplication events relative to animal phylogeny; tandem duplication → wrong alignment of genes during duplication + crossing-over. New gene inherited (2 "copies"); Orthologous → genes separated by speciation. Same gene on different species; Paralogous → genes separated by duplication. 2 copies on same organism. Large-scale duplications in vertebrate genomes: Syntetic regions between vertebrate chromosomes are signs of large-scale duplications; In humans, some genes have been lost post-duplication within the Hox cluster; Only a few genes are new of vertebrate or humans. We have more genes than a Drosophila mostly because we have more copies of the same gene; Genes may randomly be linked to regulatory genes and this link has been kept. Original function could have been completely different. Mechanisms of gene divergence: With gene duplication both cis-regulatory regions and the coding sequence are copied;  The resulting copies diverge with time/generations, through the accumulation of mutations; Mutations can alter the regulatory regions (where/when a gene is expressed), the coding sequence (what the gene product does), or both. Expansion of the toolkit correlates with increased complexity, not diversity: Comparative genomics reveals two periods of major genomic change during animal evolution: o At the transition to triploblastic bilaterians; o At the base of the vertebrate lineage; The changes correlate with the emergence of more complex animal forms; By contrast, the content of the genetic toolkit for development appears roughly equivalent among other morphologically disparate bilaterian phyla, suggesting that diversification does not require more toolkit genes. Evolution of metazoan Hox genes reflects the evolution of the development gene toolkit: A unique origin for animal appendages: Drosophila genetics tells us that the homeodomain transcription factor distal- less (Dll) is required for the onset of appendage development; Dll is expressed along the proximo-distal axis of developing limb-like structures of six coelomate phyla; This strongly suggests that the ancestor of the protostomes and the deuterostomes possessed body wall outgrowths prefiguring appendages. is the evolution of body plans driven by changes in the function of Hox genes? Rationale for the hypothesis → loss of Hox genes leads to loss of segment specialisation; Genetic manipulation of model organisms like Drosophila can reveal the potential of developmental systems to undergo particular types of morphological changes → for example, patterns of segmental specialisation may be altered by changing function of Hox genes. This approach alone, however, cannot identify the actual genetic changes that take place over macro-evolutionary timescales. Lewis' hypothesis (defeated) → the evolution of segment diversity is associated with the duplication and diversification of Hox genes. All arthropods share the same set of Hox genes: Is Hox expression conserved among arthropods? In 1997, Averof and Patel showed that changes in the expression pattern of the Hox genes (Ubx and Abd-A) in different crustaceans correlate well with the modification of their anterior thoracic limbs into feeding appendages. Crustacean appendage evolution associated with changes in Hox gene expression: In some Crustacean lineages some thoracic limbs, normally involved in locomotion, are modified into feeding appendages, and involved in the manipulation of food. They are called maxillipeds; Maxillipeds are morphologically and functionally more similar to the feeding appendages, reminiscent of homeotic transformations → a mutation transforms one organ into another (or appendage, etc.); To test whether these new segmental identities correlated with changes in Hox expression, Averof and Patel surveyed the expression of two Hox genes, Ubx and Abd-A, in 13 crustaceans; In crustaceans, these genes are thought to be involved in specifying the part of the trunk posterior to the mouthparts; In Triops, which belong to an order with no maxillipeds, Ubx and Abd-A are expressed in all thoracic segments;  In Mysidium, a crustacean with a single pair of maxillipeds, Ubx/Abd-A are shifted backwards (not expressed in T1); In lobster larva, which possess two pairs of maxillipeds, Ubx/Abd-A are further shifted backwards (not expressed in T1 and T2); Conclusion → there is a striking correlation between the anterior boundary of Ubx/Abd-A expression in the thorax of crustacean larva, and the presence and position of maxillipeds; In the light of the known function of these genes, to specify thoracic segment with legs, it is likely that the posterior shifts in expression may underlie the change of segmental identity. Lewis' hypothesis defeated → the evolution of segment diversity is not associated with the duplication and diversification of Hox genes. From correlation to causality → gain- and loss-of-function of Ubx in a crustacean species: Parhyale hawaiensis is a model species for crustaceans, with a pair of maxillipeds; Ubx is not expressed in T1 in Parhyale correlating with the presence of maxillipeds on this segment; Pavlopoulos et al. used transient, conditional misexpression of Ubx in Parhyale larvae; Ectopic Ubx expression leads to maxilliped-to-leg transformations;  Partial to complete maxilliped-to-leg transformation likely depends on Ubx dose; Conversely, knocking-down Ubx with RNAi results in leg-to-maxilliped transformations; Conclusion → Ubx expression correlates with leg vs maxilliped segmental identity in crustacean, and, Ubx is necessary and sufficient to repress maxilliped identity. Hox genes and the evolution of tetrapod axial identities: In vertebrates, changes in the number of vertebrae within regions of the vertebral column correlate with Hox expression patterns in the paraxial mesoderm; The transition from one type of vertebra to another corresponds to the anterior limits of expression of specific Hox genes;  For instance, Hoxc6 marks the transition from cervical to thoracic vertebrae but lies at a different axial position relative to the head; Boundary of some Hox genes varies among vertebrates but remain homologous (e.g. Hoxc6 → switch between cervical and thoracic vertebrae, position 12 in mice, 19 in chicks, and 22 in geese). Wing evolution in insects: Homeotic genes and the regulation and evolution of insect wing number: The insect fossil record suggests that, in the primitive state, wing-like structures formed on all of the trunk segments (around 325 mya); More recent fossilised mayfly nymphs had larger wing-like structures on thoracic segments;  The fossil record suggests that over time abdominal wing structures became reduced in size and ultimately disappeared, as did wing-like structures on the first thoracic segment and smaller winglets on the abdomen; The restriction of insect wings to the second and third thoracic segments suggests that the Hox genes sculpted the evolution of wing number and pattern; Genetic analysis has indeed established that specific Hox genes (Ubx, Scr) repress the initiation of wing development in particular segments. The evolution of body plan in insects → how many wings? The genetic potential for wing formation exists in every segment but has been repressed by homeotic genes in T1 and abdominal segments for more than 250 my. Sex-combs reduced (Scr) represses T1 dorsal appendages in insects → the down- regulation of Scr in segment T1 of various insects results in the formation of a wing- like structure. diversification of insect wing morphology → butterflies Following the crustacean logic, one would expect that wing identity changes between flies and butterflies result from changes in Hox expression; In Drosophila, Ubx represses wing fate on the third thoracic segment T3. Is Ubx not expressed in butterfly T3 segment? It is. Ubx expression is unchanged between flies and butterflies; Ubx is a transcription factor that imparts segmental identity by controlling the expression of a battery of downstream genes; The targets of Ubx must have changed, to some extent, between flies and butterflies. Body plan innovation in treehoppers through the evolution of an extra wing-like appendage: Treehoppers (Membracidae) are a group of hemimetabolous flying insects, with bizarre shape; The helmet is connected to the first thoracic segment T1; Just like a wing, the adult helmet is a bi-layer, plywood-like structure while the nymphal pronotum is a monolayer. Helmets are not paranotal expansion: Paranotal expansions are non-articulated cuticular projections of the thorax;  Dorsal fusion of wing serial homologs is not unique to treehoppers; The helmet and the wings share similar veins. Distal-less, a marker of appendage extremities, is expressed at the tip of the helmet: Parallel growth of wings and helmet primordia: Conclusion → the helmet is a T1 dorsal appendage, possibly a wing serial homolog. Hypotheses for the genetic basis of helmet evolution: 1. A change in Scr expression (repression in T1): a. Scr expression is detected in the developing helmet;  2. A change in Scr protein function: a. Can be tested by swapping the Scr sequence in Drosophila with the Scr sequence of treehopper to see if an extra appendage is developed; b. Membracid Scr is still capable of repressing wing development when misexpressed in wing fly discs; 3. Changes downstream of Scr: a. Membracids have escaped Scr-dependent mechanisms imposed by selective pressures that have operated for over 250 my. Conclusions: The membracid helmet is a novel T1 dorsal appendage and may be a wing serial homolog; The evolution of a T1 dorsal appendage has been prohibited by selective or genetic constraints for more than 250 my of insect evolution; After the "lock" was bypasses, this novelty, freed from any functional pressure, evolved "left to the free play of the various laws of growth" and experienced a remarkable and rapid diversification of its shape. Wing-like appendage diversification is not unheard of: Beetle elytra provide an independent example of wing serial homologs not used for flight and "left to the free play of the various laws of growth". Co-option of wing-patterning genes underlies the evolution of the treehopper helmet: To test the shared developmental origin of wing and helmets, fisher et al. compared the transcriptomes of different body parts during development; In treehoppers, helmet gene expression is most similar to that of wings; The authors concluded that the wing-patterning network was co-opted by the helmet. This is not very different from saying that the wing program has bypassed repression on T1. Vertebrate limb evolution → pelvic reduction in three-spined sticklebacks: Pelvic fins in fish correspond to our legs. They have been reduced in some freshwater population of sticklebacks after the last ice age; Shapiro et al. have used genetic mapping to identify genes underlying the difference between marine (normal fin) and freshwater (reduced pelvic fin) fish;  Pitx1 is the major QTL in the mapping; Pitx1 is a transcription factor known to determine limb formation; Pitx1 has changed expression during development between the two forms. Defining evolutionary novelty: Morphological character present in a group of species and either (I) lacking obvious homology to other traits in outgroup species, or (II) qualitatively different from pre-existing traits; There are many other definitions in the literature, for instance "a structure or pattern element, or even an entire body plan, that has a new adaptive function"; As with all definitions, there is a grey zone, primarily because the notion of novelty implies a clean break (before vs after), but the evolutionary process is continuous. Tinkering with sense organs:  Insects are covered with external mechanosensory organs; Sense organs emerge from regions of neural competence defined by the expression of proneural genes: o In Drosophila, on wing discs, small groups of epidermal cells (30-40) acquire the transient competence to become sensory organ precursor; o This competence is given by the so-called proneural genes, and is later restricted to a single precursor per group, the sensory organ precursor (SOP); o For bristles, the proneural genes involved are the transcription factors achaete and scute; o Other proneural genes determine the formation of other sense organs (atonal, amos); o Proneural genes and their function are generally conserved in invertebrates and vertebrates. From competence to bristle, achaete-scute genes regulate a set of target genes: achaete-scute genes regulate all steps of sensory organ development, from the sensory organ precursor, to the control of the lineage and to cell differentiation. Butterfly scales, an evolutionary innovation? It has been proposed that the sensory bristles of the insect peripheral nervous system and the wing scales of Lepidoptera are homologous structures; To determine if the developmental pathways leading to Drosophila sensory bristle and butterfly scale formation use similar genetic circuitry, Galant et al. examined the expression of achaete-scute homologues in the butterfly Precis coenia; They determined that scale and socket cell precursors expressed achaete- scute homologues, and that part of the lineage of these precursor (neuron/glial cell) is absent. More bristles turned into colour scales: Many beetle species are also covered with scales, which may be modified bristles;  Chances are, that these scales are also derived from external sensory bristles. Sex comb as a model of evolutionary innovation: Male-specific morphological structure; Evolved recently in the genus Drosophila; Diverse morphology in closely related taxa; Recent innovation: o Rapid divergence; o Convergent evolution; o Secondary loss; o Involved in copulation → used to cling on females' wings. The origin of a new developmental pathway: The origin of new sex-specific structures is driven by the evolution of doublesex regulation: Evolution of new sex-specific traits involves new doublesex expression domains: The origin of new sex-specific traits requires new doublesex (dsx) expression domains; These evolutionary novelties results from a reconfiguration of ancestral developmental program with (I) spatial rearrangements (density, rotation, morphogenesis) and (II) sex-specific modulation. Evolutionary origin of insect wings: Two hypotheses have been proposed for the origin of insect wings; o Wings evolved by modification of limb branches that were already present in multibranched ancestral appendages and probably functioned as gills;  o Wings arose as novel outgrowths of the body wall not directly related to any pre-existing limbs; Averof and Cohen examined the expression of crustacean homologues of two genes that have wing-specific functions in insects, pdm/nubbin and apterous; pdm/nubbin and apterous expression patterns support the hypothesis that insect wings evolved from gill-like appendages (epipods) that were already present in the aquatic ancestors of both crustaceans and insects. More novelty from ancestral epipods: Tracking the expression of pdm/nubbin in other appendages, Damen et al. examined pdm/nubbin and apterous expression in chelicerates; In spiders (terrestrial chelicerates), pdm/nubbin and apterous are expressed in successive segmental primordia that give rise to book lungs, lateral tubular tracheae, and spinnerets, novel structures that are used by spiders to breathe on land and to spin their webs. The butterfly eyespot, an iconic evolutionary novelty; Eyespot are colourful, conspicuous, and concentric markings that butterflies, mostly from the family Nymphalidae, display on the margins of their wings; Eyespots originated multiple times independently across the Lepidoptera. We are now focusing on a single gain in Nymphalidae. Evolution of the butterfly eyespot → Dll redeployment provides new positional landmarks: The transcription factor Distal-less (Dll) is expressed at the wing margin in all butterflies and all insects; In butterfly species with an eyespot on their wings, Distal-less expression has gained an additional component, a focus of expression prefiguring the adult eyespots; Brakefield et al. identified that the new component appears later in wing development and that Dll expression becomes restricted to scale precursors; With transplant experiments, they showed that Dll-expressing cells could induce an ectopic eyespot, which colours depended on the graft position. Dll foci works as a developmental organiser. Co-option of an ancestral circuit organises the new morphogenetic field of the eyespot: Dll is an early marker of the eyespot, but probably not the first step in its determination. The earliest gene remains unknown; The early marker gene activates a battery of genes in the spot focus, including the hedgehog and wingless signalling that pattern the spot colour rings. Nuances in the target genes of the eyespot regulatory circuit produces colour diversity: Slightly different use of the spot morphogenetic field produces diversity of the eyespot phenotype. Another iconic morphological novelty → beetle horns: Several beetle species in the family Scarabaeidae are equipped with male- exaggerated thoracic and cephalic weapons. Beetle horn's development is governed by appendage patterning genes:  Horned beetles' thoracic and cephalic weapons are body wall outgrowths expressing classical appendage patterning genes. The notochord, a key innovation at the origin of vertebrates: The notochord is a stiff, axial rod of cells that represents the functional precursor of the vertebral column in basal chordates; The dorsal-most region of the mesoderm, where the Spemann organiser arises, produces the notochord. It acts as an organiser for the early development of the CNS and adjacent axial mesoderm; The notochord was long seen as an innovation of chordates, the lineage leading to vertebrates; More recent comparison of gene expression and anatomical organisation suggest instead that the notochord may have much deeper evolutionary origin. Evolution of the notochord:  The specification of the notochord involves the transcription factor Brachyury, which was thought to have been co-opted; An alternative view emerged more recently, proposing that the notochord is homologous to the annelid axochord, a midventral longitudinal muscle closely associated with the nerve cord; The proposed homology is mainly based on the involvement of similar transcription factors (Brachyury, FoxA, FoxD, twist, SoxD and SoxE) and signalling molecules (noggin and hedgehog) in the formation of both structures. Conclusion on the evolution of the notochord: The notochord can still be considered an evolutionary novelty and a morphological signature of chordates. Its structural and inductive properties are unmatched among animals; Nevertheless, it may not be a complete de novo gain, but rather, the modification of an ancestral structure; The novelty may originate from new target genes downstream of Brachyury, rather than the co-option of Brachyury itself; This case illustrates the limits of defining a novelty and pinpointing its genetic origin. A signature of vertebrates → sensory placodes and neural crest:  The vertebrate head, with an array of sense organs and features involved in an active predatory lifestyle is a main novelty of vertebrates. These structures are derived from two embryonic cell populations, cranial placodes and the neural crest; Intriguingly, the genes specifying placodes and neural crest in the regions flanking the presumptive neural tube of vertebrates have similar expression patterns in more basal chordates. These are transcription factors of the Dlx, Msx, slug/snail, Pax2/5/8, and Pax3/7 families; Other markers of neural crest cells are not expressed at the lateral edge of the neural plate in Amphioxus (lancelets), but can be found in neural crest cells in lampreys. This suggests that the spatial coordinates of the neural crest populations predate their emergence. Re-evaluating the notion of evolutionary novelty: Initially defined an evolutionary novelty as a morphological character present in a group of species and either (I) lacking obvious homology to other traits in outgroup species, or (II) qualitatively different from pre-existing traits; Although some novelties look radically different from pre-existing structures, the genes that determine them may sign a homologous relationship: o Butterfly scales and insect bristles in general are determined by the same proneural genes and likely have a common ancestry;  o Membracidae's helmet, once seen as a complete novelty, is in fact leaning on the wing developmental program; o The notochord may have a deep homology to an annelid muscle; This blurs the notion of novelty even more; Superficially, a shared character in a group may look like an innovation because apparent equivalents are lacking in outgroups. But when thinking of the evolutionary process, it is far most likely that a novelty is an extremely modified version from a pre-existing trait or developmental program than a newly assembled structure or gene regulatory network. Novelty from recycling → evolution and tinkering: A common theme to all examples is the notion that novelty devolves from pre- existing genes and structures, rather than de novo creation of new structures; François Jacob, geneticist and molecular biologist, had likened this phenomenon to the work of a tinkerer; Contrasting with an engineer, whose creation follow a plan and use tailored tools, a tinkerer operates by improvising and recombining material at hand, in an opportunistic manner; At the molecular level, evolutionary tinkering means that old genes change their regulation, become duplicated, or are recycled in any possible way, resulting in new forms (or other traits); The emergence of new form through evolutionary tinkering happens, regardless of any biological function or adaptation; Recycling of old genes or regulatory networks happens frequently, simply because it is more likely to happen than de novo evolution of new genes. When a trait changes, what do evolutionary developmental biologists want to know? Changes to one or more loci? Which one(s)? Changes in the function of a protein or in its expression pattern? How many mutations does it take to change the function or expression of a gene?  Are these mutations new, or did they exist in the form of standing genetic variation? When evolution repeats itself, does it follow the same genetic paths? A change in gene expression does not imply that the genetic changes happened at the locus of this gene: Most evolutionary transitions examined so far in this lecture point to genes that have changed expression in line with a change of morphology; This does not mean that the genetic changes responsible for the new morphology have happened at these genes; The changes may be in cis of this gene, or in one or several other genes that control it; How do we find those genetic changes? Identifying loci underlying an evolutionary transition: There are two general approaches to find the locus or the loci responsible for an evolutionary change; Genetic mapping is an exhaustive inventory of the genetic changes associated with the new phenotype; it does not require previous knowledge on the trait formation;  Testing candidate genes, instead, is an educated guess, based on what we know of genes regulating the development of the trait of interest, with the assumption that these genes may have been directly modified during evolution. Genetic mapping: This approach aims at describing the genetic architecture of phenotypic variation; Genetic architecture → the number of loci involved in the variation of a trait, their relationships, and their relative contributions to this variation; It makes no sense to talk about the genetic architecture of a trait, only of variation. Examples of genetic mapping: QTL analysis → identification of the genetic architecture underlying variation in bony lateral plates in sticklebacks: o Introgressing genotypes (transfer of genetic material from one species into another → hybridization); o Correlating phenotype and allelic variation along the genome; Association study: o Focus on comparison with already known loci; o Check if there is a change in expression;  GWAS: o Check what alleles are associated with a trait; o Gene can be identified if the resolution of the analysis is high enough. Advantages and limitations of genetic mapping: A powerful method that indicates not only the number and identity of loci involved in the variation of a trait, but also their relative contribution; A strictly correlative method, but with very strong statistical support; Pitfalls exist. The gene cortex, once mapped as responsible for colour variation in Lepidoptera, has nothing to do with colour variation; Genetic mapping is only possible for: o Estimating variation within a population, or between species whose hybrids are fertile; o For species that can be maintained in the laboratory; Useful only when mapping resolution is high; o Sufficient depth of sequencing; o Sufficient recombination rate and sufficient linkage disequilibrium; The identification of a locus by genetic mapping does not give information on its mode of action. Candidate gene approach: This approach leans on existing knowledge on the development of a trait; The development of a character has been studied, for instance with mutagenesis, to identify the genes that control its information; If this character varies among species, it is likely that the genes that control its information have changed among species. First compare these genes, their expression, their sequences, among species. Morphological divergence often correlates with differences in expression during development: bmp-4 and the shape of the beak of Darwin's finches: o Surveying expression patterns of several growth factors known to be expressed during avian craniofacial development; bab-2 and the abdominal pigmentation of flies: o bric-à-brac2 was known to control abdominal pigmentation in Drosophila melanogaster. How to test whether a candidate gene is the locus of evolutionary change?  Heterologous transgenesis → the transformation of gene X in an animal representing the ancestral state results in a phenotype similar to that of the derived state; Association study. Advantages and limitations of candidate gene approach: A lead to locating genetic changes underlying phenotypic variation; Enables comparisons to be made at any phylogenetic distance; Depends on available knowledge about the development of the trait of interest; Often the only method of tracking genetic changes; Gene expression correlated with a trait, no matter how perfectly, is no proof of a causal link; The causative changes may have nothing to do with what is known of the development of a trait. For instance, a newly recruited repressor may repress a pre-existing developmental program; For polygenic traits (traits whose variation is caused by multiple loci), the candidate gene approach reveals, at best, one aspect of the story; Candidate gene approach gives no information on the degree of contribution of the gene to the total variation. From locus to sequence: Once a locus underlying the evolution of a trait has been identified, it is time to refine the analysis to understand the nature of the changes and how they occurred: o Coding or regulatory change? (What it does or where it works) o Single mutation or multiple mutations scattered across the locus? o New mutation or fixation of existing variation in normal populations? A framework to think of morphological evolution in terms of gene regulation: Britten and Davidson (1969) → moving the focus of animal development from protein function to the regulation of gene expression; King and Wilson (1975) → proteins of closely related species with divergent morphology are too similar to explain divergence. The invention of gene regulatory networks (1969): First conceptualisation of the idea that developmental programs are controlled by complex and modular networks of transcriptional regulation (also known as Gene Regulatory Networks). Britten and Davidson (1969) → observations: Cell states are defined by sets of genes that act coordinately (gene batteries): o There must be a coordinating mechanism; Genes active in a given territory are usually not contagious on the DNA molecule: o Distinct from bacterial operons → a single promoter reads and expresses a downstream of genes, because bacteria have more genes read with a single promoter; Genomes of "higher animals" are larger than those of simpler ones, thus potentially supporting a more complex developmental program:  o Increase in genome size allows "vastly increased complexity of regulation, rather than a vastly increased number of producer genes"; repetitive DNA sequences are broadly distributed across the animal genomes, expressed tissue specifically and interspersed with producer (structural) genes: o They could contain the regulatory program. 7 concepts: Gene → elementary functional unit of DNA (does not necessarily encode a protein): o Sequence feature; Producer gene → a region of the genome transcribed to yield a template RNA: o Coding locus; Receptor gene → a DNA sequence linked to a producer gene which causes transcription of the producer gene: o cis-regulatory sequence; Activator RNA → RNA molecules which form a sequence-specific complex with receptor genes linked to producer genes: o trans-regulatory factor; Integrator gene → gene whose function is the synthesis of an activator RNA: o trans-regulatory factor gene; Sensor gene → a sequence serving as a binding site for agents that induce the occurrence of specific patterns of activity in the genome (hormones, inducers): o cis-regulatory sequence for transduction pathway;  Gene battery → set of producer genes which is activated in response to a particular sensor gene. Model: A proposed model of genome-wide regulatory interactions substantiating Morgan's vision; Wrong: o Activator RNA forming helix with DNA (but protein mentioned); o No specific repressors (histones act as general inhibitors of transcription); o Linear cascade → no feedback loops; Correct: o Most of the rest, in spite of very scarce knowledge; Consequences on evolution: o Any evolutionary changes in the phenotype of an organism require, in addition to changes in the producer genes, consistent changes in the regulatory system. Not only must the changes be compatible with the interplay of regulatory processes in the adult, but also during the events of development and differentiation. At higher grades of organisation, evolution might indeed be considered principally in terms of changes in the regulatory system. First (indirect) evidence for regulatory evolution: Modern humans and chimpanzees differ extensively in their morphology in spite of their close phylogenetic proximity; How do these differences reflect in our genomes? In 1975, King and Wilson attempted to directly address this question by comparing many proteins of the two species; They marked that DNA or protein sequences can be used as a common "yardstick" to measure genetic distances; They compared protein similarities between human and chimpanzees in three different ways: o Directs sequence comparison (9 proteins); o Microcomplement fixation, which provides immunological distances linearly correlated with amino acid sequence difference (4 proteins); o Electrophoresis (44 proteins); They also compared the annealing temperatures of pure human DNA, pure chimpanzee DNA, and hybrid DNA. The differences between species are estimated at around 3%; Based on the protein comparisons, they estimated the genetic distance between human and chimps and compared it to within humans' genetic distances (Caucasian, black African, and Japanese populations): o This distance is 25-60 times greater than the distance within human populations; Compared to the genetic distance between sibling species in Drosophila or mammals, the human-chimp distance is extraordinarily small; There is a discrepancy between genetic distance and phenotypic differences between humans and chimps. Genetically divergent Drosophila are in the same genus and look alike. Humans and chimps are in different genera;  The intriguing result, is that all the biochemical methods agree in showing that the genetic distance between humans and chimpanzee is probably too small to account for their substantial organismal differences; Interpretation → if protein divergence does not align with phenotypic divergence between human and chimpanzee, where is the difference encoded? The regulatory framework developed by Jacob and Monod in bacteria, and later by Britten and Davidson for metazoans, offers an alternative hypothesis: o Small differences in the time of activation or in the level of activity of a single gene could in principle influence considerably the systems controlling embryonic development. The organismal differences between chimpanzees and humans would then result chiefly from genetic changes in a few regulatory systems, while amino acid substitutions in general would rarely be a key factor in major adaptive shifts. A lasting debate around the molecular nature of changes underlying morphological evolution: King and Wilson's hypothesis fuelled research on the type of DNA changes that govern changes of forms; Several case studies identified regulatory changes, other changes in the coding region of specific genes. The genetic basis of albinism in Astyanax cavefish;  Astyanax fish exist in two forms in Mexico → a surface and pigmented form found in rivers, and a blind, unpigmented cave form; The transition from surface to cave happened at least twice independently; How was the pigmentation lost? The gene Oca2 is associated with variation in pigmentation in Astyanax: o Genetic mapping indicates that the gene ocular and cutaneous albinism-2 (Oca2) explains 70-90% of the variance in pigmentation (both cave populations); o Polymorphism in the coding sequence of Oca2 suggests changes in the function of the gene; Independent deletions of Oca2 exons cause loss of function in separate populations: o To test the function of the different Oca2 alleles, Protas et al. expressed them in a melanocyte (pigment producing cell) cell line generated from an Oca2-deficient mouse; o The cave alleles do not rescue the pigmentation phenotype, the surface allele does; o Oca2 is not involved in any other process than the control of pigmentation. Coat colour evolution in rock pocket mice (Chaetodipus intermedius): Rock pocket mice exist in two morphs in the desert of Arizona and New Mexico, each morph is over-represented on the substrates where it is less visible; How did pale mice turn dark? An association study with the MC1R locus: o We have learnt from 100+ years of collecting mouse mutants that there are over 100 ways to turn a mouse pale; Polymorphism in MC1R coding sequence uniquely associated with dark mice: o MC1R coding sequence analysed from 69 mice; o 4 non-synonymous amino acid polymorphisms are unique to dark mice; o The distinct molecular basis for the same phenotype in two different populations provides strong evidence for convergent phenotypic evolution on a relatively short timescale; Mutations in the coding sequence of MC1R explain the overall change in coat colour: o In summary, the evolutionary change of colour is repeatedly linked to coding changes at the MC1R locus. Regulatory changes in the target of a Hox gene underlies evolution of pigmentation in flies: Pigmentation of the posterior male abdomen is a recently acquired trait in the Drosophila melanogaster lineage; Dark pigmentation results from the action of several metabolic genes that convert pigment precursors into dark pigment deposits. yellow is one of these genes; The expression pattern of these pigmentation genes determines which body part of a fly are dark and which are pale; Jeong et al. have shown that the Hox gene AbdB (Abdominal B) expressed in most posterior part of the fly abdomen, was co-opted by the regulatory regions of yellow in species with dark abdomens → AbdB activates yellow expression where it is present (posterior part of the abdomen0; This very regulatory link between AbdB and yellow was subsequently lost, resulting in species with pale abdomens; yellow codes for production of black pigment. Mutations in yellow kills production of black pigment and turn the individual yellow in colour. Duffy antigen and malaria resistance (change in regulatory sequence): The Duffy antigen function as a chemokine receptor for a number of pro- inflammatory cytokines, but also as a receptor for the malaria parasites Plasmodium vivax and Plasmodium knowlesi; The duffy gene is expressed in different tissues under the control of distinct enhancers; A single nucleotide variant in the red-blood cell enhancer abolishes duffy expression in this tissue, preventing Plasmodium to enter those cells. The pleiotropy → a major bias that influences the evolutionary path: Developmental genes, particularly those involved in patterning, are typically expressed in several tissues and at different stages → the mutation of their coding sequence is most often pleiotropic. Regulatory control of pattern formation, another evolutionary bias: Change in discrete elements is a dominant theme of morphological evolution; Changes in pattern elements typically involve changes in enhancers, the very elements that build patterns; Coding changes can lead to changes in discrete patterns, but it is more difficult. Conclusions → predicting the genetic basis of morphological evolution: Genes that govern embryonic/post-embryonic development are involved in many processes, mutations in their coding sequences are often pleiotropic; The level of pleiotropy determines the type of mutations that are permissible under natural selection; For these reasons, cis-regulatory evolution dominates morphological diversification,